Elaboration of a benzofuran scaffold and evaluation of binding affinity and inhibition of Escherichia coli DsbA: A fragment-based drug design approach to novel antivirulence compounds

Article abstract of DOI:10.1016/j.bmc.2021.116315

Bacterial thiol-disulfide oxidoreductase DsbA is essential for bacterial virulence factor assembly and has been identified as a viable antivirulence target. Herein, we report a structure-based elaboration of a benzofuran hit that bound to the active site groove of Escherichia coli DsbA. Substituted phenyl groups were installed at the 5- and 6-position of the benzofuran using Suzuki-Miyaura coupling. HSQC NMR titration experiments showed dissociation constants of this series in the high μM to low mM range and X-ray crystallography produced three co-structures, showing binding in the hydrophobic groove, comparable with that of the previously reported benzofurans. The 6-(m-methoxy)phenyl analogue (2b), which showed a promising binding pose, was chosen for elaboration from the C-2 position. The 2,6-disubstituted analogues bound to the hydrophobic region of the binding groove and the C-2 groups extended into the more polar, previously un-probed, region of the binding groove. Biochemical analysis of the 2,6-disubsituted analogues showed they inhibited DsbA oxidation activity in vitro. The results indicate the potential to develop the elaborated benzofuran series into a novel class of antivirulence compounds.

Full text of DOI:10.1016/j.bmc.2021.116315

Bioorg. Med. Chem. 45 (2021) 116315
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Elaboration of a benzofuran scaffold and evaluation of binding affinity and
inhibition of Escherichia coli DsbA: A fragment-based drug design approach
to novel antivirulence compounds
,
,
Luke F. Duncana 1, Geqing Wangb 1, Olga V. Ilyichovac, Rabeb Dhouibd, Makrina Totsikad,
Martin J. Scanlonc , Begona Heras *, Belinda M. Abbotta
,
e
b
,
,*
˜
a Department of Chemistry and Physics, La Trobe Institute for Molecular Science, La Trobe University, Melbourne, VIC 3086, Australia
b Department of Biochemistry and Genetics, La Trobe Institute for Molecular Science, La Trobe University, Melbourne, VIC 3086, Australia
c Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Parade, Parkville, VIC 3052, Australia
d Centre for Immunology and Infection Control, School of Biomedical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia
e ARC Training Centre for Fragment Based Design, Monash Institute of Pharmaceutical Sciences, Monash University, 381 Royal Parade, Parkville, VIC 3052, Australia
A R T I C L E I N F O
A B S T R A C T
Keywords:
Bacterial thiol-disulfide oxidoreductase DsbA is essential for bacterial virulence factor assembly and has been
identified as a viable antivirulence target. Herein, we report a structure-based elaboration of a benzofuran hit
that bound to the active site groove of Escherichia coli DsbA. Substituted phenyl groups were installed at the 5-
and 6-position of the benzofuran using Suzuki-Miyaura coupling. HSQC NMR titration experiments showed
dissociation constants of this series in the high µM to low mM range and X-ray crystallography produced three co-
structures, showing binding in the hydrophobic groove, comparable with that of the previously reported ben-
zofurans. The 6-(m-methoxy)phenyl analogue (2b), which showed a promising binding pose, was chosen for
elaboration from the C-2 position. The 2,6-disubstituted analogues bound to the hydrophobic region of the
binding groove and the C-2 groups extended into the more polar, previously un-probed, region of the binding
groove. Biochemical analysis of the 2,6-disubsituted analogues showed they inhibited DsbA oxidation activity in
vitro. The results indicate the potential to develop the elaborated benzofuran series into a novel class of anti-
virulence compounds.
Fragment-based drug discovery
Benzofuran
Oxidoreductase
DsbA
Enzyme inhibition
Antivirulence
1. Introduction
antibiotics, including a large repertoire of unexplored virulence targets,
a reduced selective pressure on pathogenic bacteria to develop resis-
Bacterial infections are becoming a serious threat to global health.
The rapid emergence and spread of antibiotic resistant bacteria coupled
with a drying pipeline of new antibiotics is threatening our ability to
control drug resistant infections. If no action is taken, we are heading for
a post-antibiotic era where common infections are no longer treatable.1
In the US alone, 2.8 million people acquire antibiotic-resistant infections
annually with more than 35,000 deaths claimed.2 With the reduction in
efficacy of current therapeutics, new classes of antibiotics with novel
mechanisms of action are urgently needed. One strategy to prevent or
reduce bacterial resistance is to develop antimicrobials that target
virulence rather than viability.3ꢀ 5 This new paradigm for targeting
bacterial infections offers a number of advantages over conventional
tance and preservation of the host microbiota.3
Pathogenic bacteria rely on virulence factors to colonize the host,
establish infections and cause disease.6 Examples of virulence factors
include proteins that make up fimbrial and non-fimbrial adhesins,
secreted toxins, secretion systems and motility organelles.4 Common
among many of these virulence proteins is the requirement of enzymatic
oxidative folding to achieve their functional conformation.7,8 This task is
undertaken by the thiol-disulfide oxidoreductase DsbA, a thioredoxin-
like protein with a redox active CXXC motif that introduces disulfide
bonds to cysteine pairs in unfolded substrates.7,9 As DsbA folds a broad
range of virulence factors, lack of functional DsbA has pleiotropic effects
on multiple virulence-associated phenotypes and therefore represents an
* Corresponding authors.
E-mail addresses: B.Heras@latrobe.edu.au (B. Heras), B.Abbott@latrobe.edu.au (B.M. Abbott).
1
Both authors contributed equally to this work.
https://doi.org/10.1016/j.bmc.2021.116315
Received 20 May 2021; Received in revised form 15 July 2021; Accepted 16 July 2021
Available online 22 July 2021
0968-0896/Crown Copyright © 2021 Published by Elsevier Ltd. All rights reserved.

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
attractive approach to inhibit bacterial virulence.10 The prototypical
Escherichia coli (Ec) DsbA, the best characterised DsbA enzyme, consists
of a thioredoxin domain and an inserted α-helical domain. A notable
feature of EcDsbA is the presence of a hydrophobic groove adjacent to
the CXXC active site, which serves as the binding site for redox partner
DsbB and substrates (Fig. 1A).11,12 During the disulfide bond catalysis,
the active site cysteines of DsbA become reduced when the substrate is
oxidised and the membrane-bound partner enzyme DsbB restores DsbA
to its oxidised state. As the hydrophobic groove is critical for redox in-
teractions, a range of peptide and small molecule inhibitors have been
developed to disrupt DsbA activity by targeting the hydrophobic
groove.13ꢀ 16
Scheme 1. C-5 and C-6 substitution of acids 1a-c via Suzuki-Miyaura cross-
coupling (isolated yields shown in brackets). Reagents and conditions: (a) R-B
(OH)2, CsF, Pd(dppf)Cl2, CH3CN:H2O (4:1), 90 ◦C, 16–24 h.
In recent decades fragment-based drug discovery (FBDD) has
emerged as a powerful alternative method to conventional high-
throughput screening approaches for drug lead discovery and develop-
ment.17 Due to the small size (<300 Da) and lack of complexity of the
compounds typically found in fragment screening libraries, their struc-
tures are less likely to contain motifs that will sterically hinder target
product could not be isolated from the crude reaction mixture. To aid
purification, an alternate two-step synthetic route using methyl ester 3a
or 3b was employed (Scheme 2).
Aryl bromide 3a was coupled to phenylboronic acid via Suzuki-
Miyaura coupling and the cross-coupled product 4a was able to be
binding, considerably increasing the likelihood of hit identification.18
A
fragment-based screening approach combining X-ray crystallography
and NMR, was previously used to identify the first non-peptide EcDsbA
inhibitors.14 This work underscores the utility of FBDD to tackle chal-
lenging targets like DsbA, where the active site comprises a shallow and
extended hydrophobic groove involved in protein-protein interactions.
More recently, we reported a new class of EcDsbA inhibitors identified
from a fragment library.19 Using X-ray crystallography we were able to
show that benzofuran derivatives bound to the hydrophobic groove of
EcDsbA (Fig. 1B). Furthermore, using HSQC NMR titration we deter-
mined that 6-phenoxy and 6-benzyl analogues were the strongest
binders.19 Herein we report the chemical synthesis of 2-, 5- and 6-subsi-
tuted benzofuran derivatives and their biophysical characterisation and
in vitro analysis of EcDsbA inhibition. We have generated, guided by X-
ray crystallography and NMR spectroscopy, benzofurans exhibiting
improved in vitro inhibition of EcDsbA.
2. Results and discussion
2.1. Chemical synthesis
For optimisation of the binding capacity of the previously identified
benzofuran hits, we explored substitutions at the 5- and 6-position as
possible sites for elaboration (Fig. 1B). Precursors 1a-c, which were
synthesised as previously described,19 were substituted at C-5 and C-6
via Suzuki-Miyaura coupling with boronic acids (Scheme 1). Com-
pounds 2a-c were isolated in low to moderate yields after a difficult
purification using chromatography and recrystallisation. Similarly, C-5
substituted 2d was isolated in low yield after purification by RP-HPLC.
Phenylboronic acid was also coupled to 1a but the cross-coupled
Scheme 2. C-6 substitution of methyl esters 3a and 3b via Suzuki-Miyaura
cross-coupling (isolated yields shown in brackets). Reagents and conditions: (a)
R-Ph(BOH)2, CsF, Pd(dppf)Cl2, CH3CN:H2O (4:1), 90–120 ◦C, 0.25–28 h. (b) 2
M NaOH, CH3OH:CH2Cl2 (9:1), r.t, 16 h.
Fig. 1. Crystal structures of EcDsbA in apo and
compound-bound forms. (A) Crystal structure of
oxidised EcDsbA (PDB ID: 1FVK). The Cys30-Pro31-
His32-Cys33 motif and cis-Pro loop (Val150-cis-
Pro151) of the active site are presented as orange
sticks. The hydrophobic groove adjacent to the active
site is shaded in yellow. (B) Crystal structure of
EcDsbA in complex with the best benzofuran analogue
from the previous series (PDB ID: 6POI).19 Growth
vectors used for compound elaboration in this study
are indicated by dashed arrows of the 2-,5- and 6-po-
sitions of the benzofuran core. The active site is
shaded in orange.
2

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
isolated via chromatography albeit in low yield. Following this, 4a was
hydrolysed to the desired acid 5a. Similarly, m-carboxyphenylboronic
acid was coupled to 3a, however chromatography could not separate the
homo-coupled boronic acid by-product from cross-coupled 4b. Subse-
quent ester hydrolysis of impure 4b allowed for isolation of the desired
acid 5b by chromatography in 14% yield over two steps. The 6-iodoben-
zofuran 3b was used for synthesis of the p-methoxyphenyl and m-tri-
fluoromethylphenyl derivatives 5c and 5d. The increased reactivity of
the aryl iodide coupling partner resulted in superior yields and,
furthermore, reaction times were dramatically reduced by heating the
reaction by microwave irradiation.
With the aim to elaborate the benzofuran scaffold from the 2-position
(Fig. 1B), compound 3a was brominated using bromine in acetic acid
(Scheme 3). The desired 2,6-dibromo 6a formed in a favourable 1:0.4
ratio with regioisomer 6b according to 1H NMR. The regioisomers were
inseparable by chromatography but recrystallisation from isopropanol
exclusively afforded the desired 2,6-dibromo isomer 6a in a 52% yield.
To investigate the reactivity of the 2-position versus the 6-position, 6a
was coupled to an equimolar amount of phenylboronic acid via Suzuki-
Miyaura coupling which resulted in a mixture of regioisomers 7a and 7b
as well as di-substituted 7c. These compounds were separated via RP-
HPLC. While the preferred formation of 7b indicated there was
slightly higher reactivity at the 2-position, it was not significant enough
to exploit. The 2-phenyl-6-bromo isomer was hydrolysed to acid 8a in
quantitative yield.
Scheme 4. Formation of 2,6-disubsituted benzofurans via Suzuki-Miyaura
cross-coupling (isolated yields shown in brackets). Reagents and conditions: (a)
Br2, CH3COOH, r.t, 3 h; (b) R-PhB(OH)2, CsF, Pd(dppf)Cl2, CH3CN:H2O (4:1),
80–120 ◦C MW, 15–60 m; (c) 2 M NaOH, CH2Cl2:CH3OH (9:1), r.t, 16 h.
only a minor amount of a 2,6-disubstituted by-product was observed,
which could be separated via chromatography to afford 10a in a 55%
yield. Repeating the reaction using four equivalents of m-methox-
yphenyl boronic acid led to the disubstituted product 10b in good yield,
which was hydrolysed to acid 11. The 2-bromo position of 10a was then
substituted with p-methoxyphenylboronic acid and m-carbox-
yphenylboronic acid by Suzuki-Miyaura coupling to afford 12a and 12b
To improve selectivity of the coupling reaction, the bromination and
subsequent substitution was repeated on the more reactive 6-iodobenzo-
furan 3b (Scheme 4). As with compound 3a, bromination afforded a
mixture of regioisomers 9a and 9b which required recrystallisation from
isopropanol to isolate the desired isomer 9a in a 57% yield. With the
appropriate coupling partner in hand, the increased reactivity of the aryl
iodide over the bromide could be exploited to selectively install an ar-
omatic group at C-6. Compound 9a was coupled with one equivalent of
m-methoxyphenylboronic acid in an attempt to avoid di-substitution.
The reaction was heated to 80 ◦C under microwave irradiation, which
allowed a shorter reaction time of 1 h. The cross-coupling proceeded
with far superior selectivity at the more reactive 6-iodo position and
◦
in fair to high yields. The reactions were heated to 120 C under mi-
crowave irradiation for 15 min. The methyl esters of 12a and 12b were
hydrolysed using sodium hydroxide in methanol:DCM (1:9) to give acids
13a and 13b respectively in good to high yields.
2.2. Structure-guided elaboration of benzofuran as DsbA inhibitors
To guide chemical elaboration of the benzofuran series as DsbA in-
hibitors, we developed a fast and robust crystallography protocol for
compound soaking, data acquisition and structure determination.14
Synthesised compounds were tested for their ability to bind DsbA by
three ligand-detected 1D NMR methods: saturation-transfer difference
(STD)20
,
Carr-Purcell-Meiboom-Gill (CPMG)21 and water-ligand
observed via gradient spectroscopy (waterLOGSY)22 (Table 1, assess-
ment criteria has been detailed in section 4.5). We and others have
shown that compounds validated using multiple ligand-detected NMR
experiments, have a higher likelihood of yielding the co-crystal struc-
tures.14,23 Therefore, these experiments provided a benchmark to pri-
oritise compounds for crystal soaking and binding affinity
determination. As exemplified in Fig. 2A, the binding of 5a to DsbA was
validated in all three ligand-detected NMR experiments. Comparison of
1
the 1D H NMR spectrum of 5a and the STD spectrum of 5a + DsbA
showed magnetization transfer from DsbA to the aromatic resonances of
5a. The waterLOGSY NMR spectrum showed that the aromatic reso-
nances of 5a became less negative in the presence of DsbA compared to
5a alone. The CPMG NMR spectrum (τ = 350 ms) of 5a showed atten-
uation of its aromatic resonances in the presence of DsbA.
In previous work we introduced various aryl and alkyl groups with
methylene, ether and amino bridges at C-6 of the benzofuran scaffold.19
X-ray crystallography and NMR spectroscopy showed that the binding
mode and affinity were dependent on the size and substitution of the
group. Noteworthy was the installation of a p-anisidine moiety which
resulted in the flipping and shift of the benzofuran core which inter-
Scheme 3. Suzuki-Miyaura coupling of phenylboronic acid to 2,6-dibromo
compound 6a and subsequent ester hydrolysis (isolated yields shown in
brackets). Reagents and conditions: (a) Br2, CH3COOH, r.t, 3 h; (b) R-PhB(OH)2,
CsF, Pd(dppf)Cl2, CH3CN:H2O (4:1), 90 ◦C, 24 h. (c) 2 M NaOH, CH3OH:CH2Cl2
(9:1), r.t, 16 h.
rupted the important π-π stacking interactions with His32 of DsbA and
compromised binding affinity.19 This outcome suggests that a large
3

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
Table 1
Biophysical and biochemical characterisation of benzofuran analogues.
I.D
Structure
Crystal Structure
obtained
wLOGSY
Pass
CPMG
Pass
STD
Pass
Dose response HSQC KD
Ligand efficiency (kcal molꢀ 1
(µM)
HACꢀ 1
0.21
)
2a
Yes
Yes
No
830 ± 250
2b
2c
Pass
Pass
Pass
Pass
Pass
Pass
1100 ± 300
2300 ± 800
0.19
0.17
2d
5a
No
Pass
Pass
Pass
Pass
Pass
Pass
820 ± 140
0.22
0.21
Yes
1137 ± 178
5b
No
Fail
Pass
Fail
≫ 1000
≫0.19
5c
5d
No
No
Pass
Pass
Pass
Pass
Fail
460 ± 70
403 ± 69
0.22
0.20
Pass
11
Yes
Yes
Pass
Pass
Fail
Fail
531 ± 281
274 ± 34
0.15
0.17
13a
Fail
Pass
(continued on next page)
4

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
Table 1 (continued)
I.D
Structure
Crystal Structure
obtained
wLOGSY
CPMG
STD
Pass
Dose response HSQC KD
Ligand efficiency (kcal molꢀ 1
(µM)
HACꢀ 1
)
13b
Yes
Pass
Pass
978 ± 99
0.14
supplementary Fig. S2. The co-structure revealed that the benzofuran
core formed -stacking interactions with His32 and the carboxylic acid
π
was oriented towards the more polar region of the binding groove,
which resembles previously reported co-structure of aniline ana-
logues.19 Due to the reduced size of the phenyl group relative to aniline
at C-6 of the benzofuran, the RHS pocket of the hydrophobic binding
groove was only partially occupied (Fig. 2B). Based on this observation,
a series of functional groups were introduced in the meta and para po-
sitions of 6-phenyl in 2a-c and 5b-d to occupy this pocket. Most of these
compounds passed all three ligand-detected NMR experiments, except
5b only showed binding in the CPMG experiment and 5c exhibited
binding in both wLOGSY and CPMG experiments but not in the STD
experiment (Table 1). Among these analogues, only 2a and 2b produced
good electron density in the resulting crystal structures. The co-structure
(i.e. crystal structure of the protein-compound complex) of 2b confirmed
the design strategy and uncovered the high complementarity between
the 6-(m-methoxy)phenyl group and the pocket (Fig. 3A). It is inter-
esting to note that the side chain of M171, which defines one end of the
hydrophobic groove, could re-orientate to accommodate the bound
molecules. The m-hydroxy 2a analogue produced a co-structure
comprising two ligands bound to the hydrophobic groove in a head-to-
tail manner (Supplementary Fig. S3). This is possibly a crystallography
artefact due to the high concentration of ligands used on soaking and has
been observed in other reported DsbA-compound complexes.14 To
interrogate the 5-position of the benzofuran a phenyl group was intro-
duced in 2d. Although 2d passed all three ligand-detected NMR exper-
iments, no co-crystal structure was obtained to support further
elaboration from this position.
Fig. 2. Overview of biophysical and structural characterisation of
benzofuran analogues. (A) Binding of 5a to EcDsbA was evaluated by a series
of ligand-detected NMR spectroscopy including STD, waterLOGSY and CPMG.
In an attempt to expand into the left-hand-side (LHS) of the hydro-
phobic groove, a phenyl group was installed at the C-2 position of the
benzofuran core in compound 7b. The lack of an aromatic substituent at
C-6 resulted in a shift of the benzofuran scaffold further into the hy-
(B) Crystal structure of 5a bound to EcDsbA. Simulated annealing omit σA
-
weighted mFo-DFc electron density map for 5a is contoured at 2.5σ and shown
as blue mesh. Residues within 4 Å of 5a are presented as grey sticks. Residues
that were perturbed in the HSQC spectra but not within 4 Å of 5a are presented
as green sticks. (C) A portion of the 1H-15N HSQC spectra of 15N-labelled
EcDsbA showing chemical shift perturbation (CSP) upon the titration of
increasing concentrations of 5a (0, 0.0625, 0.125, 0.25, 0.5 and 1 mM). The
direction of the CSP is indicated by black arrows. Insert shows the equilibrium
dissociation constant (KD) determination of 5a by measurement of CSP as a
function of compound concentration. Full HSQC spectra are shown in supple-
mentary Fig. S1.
drophobic end of the binding groove, possibly weakening the π-stacking
interaction with His32 (Fig. 3A). Nevertheless, the co-structure vali-
dated the 2-position as an appropriate vector to grow the scaffold into
more polar region of the binding groove. As shown earlier, the co-
structure of 2b highlighted the favourable binding pose of the com-
pound and favourable interactions of the (m-methoxy)phenyl group at
the C6-position of the benzofuran, therefore we decided to use this
pharmacophore in the next round of elaboration. To further explore the
polar region in the LHS of the groove, methoxy and carboxy groups were
introduced in the meta and para positions of the 2-phenyl group in
compounds 11, 13a and 13b. EcDsbA crystals soaked with these com-
pounds all diffracted between 1.99 and 2.30 Å resolution. Data collec-
tion and refinement statistics are listed in the supplementary Table S1.
2Fo-Fc electron density maps and omit maps of these compounds are
shown in supplementary Fig. S2. The co-crystal structures of 11, 13a and
13b bound to EcDsbA (Fig. 3B-D) showed that these compounds adopted
moiety cannot be accommodated in the right-hand-side (RHS) pocket of
the hydrophobic groove lined by residues Gln35, Leu40, Thr168,
Met171 and Phe174 (Fig. 1C). In this work, to further optimise the in-
teractions in the RHS pocket, we substituted the benzofuran moiety with
a phenyl group to generate compound 5a. A crystal structure of EcDsbA
in complex with 5a was determined at 1.83 Å resolution. Data collection
and refinement statistics are listed in the supplementary Table S1. 2Fo-Fc
electron density maps and omit maps of these compounds are shown in
5

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
Fig. 3. Crystal structure-guided elaboration of benzofuran analogues as EcDsbA inhibitors. (A) Structure overlay of the 2b(yellow)-EcDsbA complex and the
7b (blue) -EcDsbA complex. Residues within 4 Å of 2b are presented as purple sticks and residues within 4 Å of 7b are presented as grey sticks. (B-D) Crystal
structures of 11 (B), 13a (C) and 13b (D) bound to EcDsbA. Residues within 4 Å of the binders are presented as grey sticks. The sulfur atoms of the active site disulfide
bond are presented as orange spheres. Residues within 4 Å of compounds are shown as grey sticks. Chemical structures of compounds are shown in the upper right
part of each figure.
a similar binding pose in the hydrophobic groove with the benzofuran
core forming -stacking interactions with His32 and Phe174 of DsbA. As
and on the flexible cis-Pro loop, which is required for substrate recog-
π
nition and binding.12,26 It should be noted that although the co-crystal
intended, the 2-(m-methoxy)phenyl group of 11, 2-(p-methoxy)phenyl
group of 13a and the 2-(m-carboxy)phenyl group of 13b were found to
extend to the more polar region of the binding groove. Although no
polar interactions formed between the compounds and the protein, the
derivatised phenyl groups seem to be stabilised by a cluster of hydro-
phobic residues including Pro163, Phe36 and cis-Pro151. Their binding
modes are in consensus with previously reported DsbA inhibitors such as
phenylthiazole, diaryl ether and phenylthiophene14,15,19,24 and high-
light that DsbA is highly capable of binding hydrophobic molecules
without engaging many specific polar interactions. Among these inter-
acting residues, Phe36, His32 and cis-Pro151 are highly conserved
across DsbA homologues15 and the latter two are also critical for DsbA
redox activity,25 suggesting that these compounds may exert an inhib-
itory activity on DsbA.
structures showed π-stacking of the benzofuran analogues with His32,
this residue is not observed in the HSQC spectrum of apo EcDsbA and
therefore its CSP cannot be measured.
Ligands that induced appreciable CSP (>0.04 ppm) of at least three
residues were selected for dose-response titration over a concentration
gradient to calculate the equilibrium dissociation constant (KD)
(Table 1). Intriguingly, the additional interactions formed by the m-
methoxy group of 2b in the hydrophobic pocket did not translate to
improved binding affinity. 2b and 5a exhibited identical binding affin-
ity, 1100 ± 200 µM and 1100 ± 300 µM, respectively. At the meta po-
sition of the 6-phenyl group, hydroxy (2a) substitution produced similar
affinity (830 ± 250 µM) as 5a and 2b; trifluoromethyl substitution (5d)
improved the affinity to 403 ± 69 µM; whilst cyano (2c) and carboxy
substitution (5b) dramatically reduced the binding affinity to above 2
mM. At the para position of the 6-phenyl group, methoxy (5c) enhanced
the binding affinity to 460 ± 70 µM. Elaboration at the C5 position of the
benzofuran core produced compound 2d, which displayed similar
binding affinity as C6-substituted 5a. Among these analogues, 6-(p-
methoxy)phenyl derivative 5c and 6-(m-trifluoro)phenyl derivative 5d
showed the highest affinities but unfortunately failed to produce inter-
pretable electron density in the crystal structures. For this reason, 6-(m-
methoxy)phenyl derivative 2b, which had a KD of 1100 ± 300 µM but
displayed clear electron density in the co-structure, was chosen for
elaboration from the C-2 position.
2.3. Biophysical evalution of benzofuran analogues
Binding of benzofuran analogues to DsbA was further characterised
using 2D 1H-15N heteronuclear single quantum coherence (HSQC)
spectroscopy by monitoring the chemical shift perturbation (CSP) of
cross peaks upon addition of 1 mM compound. These experiments
allowed mapping the binding site onto DsbA for compounds like 2c, 5b,
5c and 5d, for which co-crystal structures could not be determined. The
CSP patterns showed that for all compounds tested, the most dramatic
changes were observed for cross peaks assigned to Gln164, Ser169,
Phe174 and Asn170. (Fig. 4) The majority of the perturbed residues
were located around the hydrophobic groove, involved in DsbB binding,
Among the three 2,6-disubstituted analogues, 13a produced the
highest binding affinity of 274 ± 34 µM, while 11 and 13b produced the
binding affinity of 531 ± 281 µM and 978 ± 99 µM, respectively. (Fig. 4)
6

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
Fig. 4. NMR characterisation of elaborated benzofuran analogues 11 (A), 13a (B) and 13b (C) binding to DsbA. Panel (A-C): Left: A portion of the 1H-15N HSQC
spectra of 15N-labelled EcDsbA showing chemical shift perturbation (CSP) upon the titration of increasing concentration of 11, 13a and 13b (0 (red), 0.0625 mM
(blue), 0.125 mM (green), 0.25 mM (yellow), 0.5 mM (red) and 1 mM (blue)). Right: the equilibrium dissociation constant (KD) determination of 11, 13a and 13b by
measurement of CSP as a function of compound concentration. The KD value is shown as KD ± error of fit.
It should be noted, precipitation of 11 was observed at 0.5 and 1 mM,
therefore only CSP at concentrations < 0.5 mM were used for KD
determination. Similarly, 13a was not soluble at 1 mM, therefore only
CSP at concentrations < 1 mM were used for KD determination. The
affinity difference of these compounds was supported by LIGPLOT
analysis of their co-structures, which showed that 13a makes the highest
number (n = 47) of interactions with DsbA while 13b makes the lowest
number (n = 34) of contacts with DsbA. The LIGPLOT analysis of the
three compounds is shown in supplementary Fig. S4, lists of detailed
interactions are presented in Table S2.
2.4. Biological evalution of benzofuran analogues
The binding of 2,6-disubtituted benzofuran analogues to the DsbA
hydrophobic groove and their improved binding affinities relative to the
parent compounds14 prompted us to investigate the ability of com-
pounds 13a, 13b and 11 to inhibit DsbA redox activity using a standard
peptide oxidation assay (POA). As illustrated in Fig. 5, compound 13a
inhibited DsbA activity in vitro in a concentration-dependent manner,
producing an IC50 of 140 ± 13 µM. Its inhibitory activity correlates well
with its NMR KD value (274 ± 34 µM). Compounds 11 and 13b however
were found to quench fluorescence signals in the assay. As fluorescence
Fig. 5. Concentration-dependent inhibition of EcDsbA oxidation activity
signal correlates with DsbA activity, this quenching effect resulted in
by 13a determined using a peptide oxidation assay. IC50 is shown as mean
artificially enhanced inhibition (Supplementary Fig. S5) and therefore it
± standard error of mean (SEM).
was not possible to reliably determine IC50 for these two compounds. It
is likely that 11 and 13b formed assay interfering aggregates.42 The
7

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
limited solubility and IC50 of these compounds also prevented the
detection of in vivo activity in E. coli swimming motility assays, which
only used 0.6% DMSO (Supplementary Fig. S6). Further medicinal
chemistry effort should focus on improving both binding affinity and
solubility of elaborated compounds. Nevertheless, these results indicate
that benzofuran analogues can inhibit DsbA oxidase activity by dis-
rupting the binding of the substrate peptide to the hydrophobic groove
of DsbA. It also highlights the significant challenge of developing in-
hibitors to target a protein site that favours hydrophobic interactions
and the importance of evaluating biological activities at the early stage
of drug discovery.27
binary buffer system described for RP-HPLC over 60 min with a flow rate
of 2 mL/min, unless otherwise stated. The purity of biologically tested
compounds was > 95% in all cases, unless specified otherwise. Micro-
wave assisted reactions were performed using a Milestone StartSYNTH
system. NMR spectra were recorded on Bruker AV-400 and AV-500
spectrometers at 400.13 and 500.02 MHz respectively, for 1H nuclei
and at 100.62 and 125.74 MHz respectively, for 13C nuclei at 300 K. For
1H NMR the residual CDCl3 peak (7.26 ppm) or DMSO‑d6 peak (2.50
ppm) were used as internal standards. Similarly, 13C NMR spectra were
referenced to the residual solvent; the central peak of the CDCl3 ‘triplet’
(77.0 ppm) or DMSO‑d6 ‘heptet’ (40.0 ppm). Chemical shifts were re-
ported as δ values in parts per million (ppm). All coupling constants J are
measured in Hertz. The following abbreviations have been used for
reporting spectral data: s, singlet; d, doublet; dd, doublet of doublets; t,
triplet; q, quartet; m, multiplet; and br, broad.
3. Conclusions
The dependence of bacterial virulence proteins on DsbA-mediated
disulfide bond formation makes this thiol-disulfide oxidoreductase an
attractive target for the development of antivirulence therapeutics. For
the development of novel inhibitors of EcDsbA, a structure-based ligand
design strategy was undertaken starting from 2-(6-bromobenozfuran-3-
yl)acetic acid. The fragment hit was first elaborated from the 5- and 6-
position using Suzuki-Miyaura coupling to install a small series of
phenyl groups. A combination of ligand detected NMR methods and X-
ray crystallography identified 2b as an ideal analogue for further elab-
oration from C-2 to access a more polar region of the binding groove. X-
ray crystal structures of 11, 13a and 13b confirmed C-2 as a valid
growth vector to access this region of the binding pocket. Further
evaluation by HSQC revealed improved KD values for C-2 analogues
relative to the parent compounds (274 ± 34 µM (13b)). Importantly,
biochemical assays showed inhibitory activity of C-2 analogues against
DsbA and suggest the potential to develop benzofuran analogues into
antivirulence compounds.
4.2. Chemical syntheses
4.2.1. General procedures
General Procedure A1: Preparation of 5- and 6-aryl analogues
via Suzuki-Miyaura cross-coupling. A solution of acetonitrile:water
(4:1) was degassed by bubbling a stream of nitrogen through the solu-
tion for one hour. The aryl halide, boronic acid (6 eq.), caesium fluoride
(3 eq.) and Pd(dppf)Cl2 (5 mol%) were added to the degassed solution
◦
(25 mL/mmol) and the suspension was heated to 90 C under an at-
mosphere of nitrogen. After 16–24 h the reaction mixture was parti-
tioned between ethyl acetate and brine. The organic layer was washed
with brine (x2), dried over magnesium sulfate then concentrated to the
crude residue. In cases where the aryl halide or boronic acid contained a
carboxylic acid the organic phase was washed with a saturated solution
of sodium hydrogen carbonate (x3). The combined aqueous phases were
acidified with concentrated HCl to pH 2, extracted with ethyl acetate
(x3) then worked up as above to afford the crude compound.
4. Materials and methods
General Procedure A2: Microwave assisted Suzuki-Miyaura
cross-coupling. Aryl halide (1 eq.), boronic acid (1.2–6 eq.) and
caesium fluoride (3 eq.) were dissolved in degassed acetonitrile:water
(4:1) (20 mL/mmol) in a Milestone microwave vessel. Pd(dppf)Cl2 was
added then the vessel was flushed with nitrogen and quickly sealed. The
reaction was irradiated at 90–120 ◦C for 15–60 min. The reaction was
cooled to room temperature then partitioned between ethyl acetate and
brine. The organic layer was washed with brine (x2), dried over mag-
nesium sulfate then concentrated under reduce pressure.
4.1. General
All commercial materials were used as received without further
purification, unless otherwise specified. Purification of solvents and
reagents, if required, was carried out by procedures described by Chai
and Armarego.28 Boronic acids were sourced commercially from
Advanced Molecular Technology. Moisture sensitive reactions were
performed under an atmosphere of nitrogen with all reactions carried
out at room temperature, unless otherwise noted. Glassware was oven-
dried and cooled under nitrogen prior to use. Analytical Thin Layer
Chromatography (TLC) was performed on Merck Kieselgel 60 F254
aluminium backed plates and visualised using a 254 nm UV lamp. Flash
chromatography was performed on silica gel (Davisil® LC60Å 40–63
µm). Melting points were determined on a Reichert ‘Thermopan’ mi-
croscope hot stage apparatus and values were corrected by a 12% in-
crease after calibration against known reference samples. Low-
resolution electrospray ionisation (ESI) mass spectra were recorded on
a Bruker Daltronics Esquire 6000 Ion Trap mass spectrometer in meth-
anol or acetonitrile (0.1% formic acid for positive mode) at 300 ◦C, a 40
eV cone voltage, with a scan rate of 5500 m/z/s. High-resolution elec-
trospray ionisation (ESI) mass spectrometry was carried out using an
Agilent Technologies Accurate Mass Q-TOF LC-MS 6530 using Auto-
sampler 1260 Infinity II in positive mode. The samples were analysed
using a flow rate of 1 mL/min, a mass range of 100–1000 m/z and a scan
rate of 10,000 m/z/second. Analytical RP-HPLC was performed on a
Shimadzu LC-20AB Prominence Liquid Chromatography system fitted
with a Phenomenex® Jupiter C18 300 Å column (250 mm × 4.6 mm, 10
General Procedure B: Base mediated ester hydrolysis. Esters
were hydrolysed according to the method of Theodorou and co-
workers.29 Esters were dissolved in a mixture of methanol:dichloro-
methane (1:9) and treated with a 2 M methanolic solution of sodium
hydroxide (1–5 eq.) to give a final 0.1 M concentration of ester. The
solution was stirred for 3–20 h at which time a cloudy suspension had
formed. The suspension was concentrated under reduced pressure and
the residue was dissolved in water. The solution was acidified with
concentrated HCl to pH 1 then the carboxylic acids were extracted with
dichloromethane or ethyl acetate (×3). The combined organic layers
were dried over magnesium sulfate then concentrated under reduced
pressure. Unless otherwise stated, no purification was required.
4.2.2. 2-(6-(3-Hydroxyphenyl)benzofuran-3-yl)acetic acid (2a)
Compound 1a (100 mg, 0.392 mmol) was coupled to m-hydrox-
yphenylboronic acid (325 mg, 2.35 mmol) according to General Pro-
cedure A1 for 24 h. The crude residue was purified by flash
chromatography (20% ethyl acetate, 1% acetic acid in hexanes) to afford
the product as a beige solid (55.1 mg, 52%), mp 181–184 ◦C. δH (500
MHz, DMSO‑d6) 9.51 (1H, br s, COOH), 7.89 (1H, s, ArH), 7.73 (1H, s,
ArH), 7.63 (1H, d, J 8.0, ArH), 7.48 (1H, d, J 8.5, ArH), 7.24 (1H, t, J 7.8,
ArH), 7.12 (1H, d, J 7.0, ArH), 7.05 (1H, s, ArH), 7.05 (1H, d, J 8.0,
ArH), 3.67 (2H, s, CH2). δC (100 MHz, DMSO‑d6) 172.4, 158.3, 155.6,
144.4, 142.2, 137.6, 130.4, 127.6, 122.1, 120.9, 118.2, 114.8, 114.7,
μ
m), using a buffered binary system; solvent A: 0.1% trifluoroacetic
acid; solvent B: acetonitrile. Gradient elution was performed using a
gradient of 90% solvent A to 90% solvent B over 20 min with a flow rate
of 1 mL/min, monitored at 254 nm. Semi-preparative RP-HPLC was
performed using a Phenomenex Jupiter C18 column with the same
8

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
114.3, 109.6, 29.7. LRMS (ESI): m/z found 266.9 (Mꢀ H)ꢀ , HRMS (ESI)
80 ◦C for one hour. The crude brown solid was purified by flash chro-
matography (2.5% ethyl acetate in hexanes) to afford the titled com-
did not ionise, C16H11O4ꢀ required 267.0663.
◦
pound as an off white solid (60.9 mg, 65%), mp 154–156 C. δH (500
4.2.3. 2-(6-(3-Methoxyphenyl)benzofuran-3-yl)acetic acid (2b)
Compound 1a (100 mg, 0.392 mmol) was coupled to m-methox-
yphenylboronic acid (358 mg, 2.35 mmol) were coupled according to
MHz, CDCl3) 7.65 (1H, s, ArH), 7.64 (1H, s, ArH), 7.59 (1H, d, J 8.1,
ArH), 7.57 (2H, d, J 8.9, ArH), 7.47 (1H, d, J 8.1, ArH), 7.00 (2H, d, J
8.9, ArH), 3.86 (3H, s, OCH3), 3.75 (3H, s, OCH3), 3.73 (2H, s, CH2). δC
(125 MHz, CDCl3) 171.1, 159.1, 155.9, 143.1, 138.0, 133.8, 128.4,
126.3, 122.0, 119.7, 114.3, 113.0, 109.6, 55.3, 52.1, 29.6. LRMS (ESI):
m/z found 297.2 (M+H)+, C18H17O4+ required 297.1121.
◦
General Procedure A at 90 C for 24 h. The crude oil was purified by
flash chromatography (1–20% ethyl acetate, 1% acetic acid in hexanes)
to afford the titled compound as a light brown solid (47.8 mg, 43%), mp
132–133 ◦C. δH (400 MHz, CDCl3) 7.69 (1H, s, ArH), 7.66 (1H, s, ArH),
7.61 (1H, d, J 10, ArH), 7.51 (1H, d, J 10, ArH), 7.37 (1H, t, J 9.8, ArH),
7.21(1H, d, J 9.8, ArH), 7.16 (1H, s, ArH), 6.91 (1H, d, J 10.5, ArH), 3.88
(3H, s, OCH3), 3.78 (2H, br s, CH2). δC (100 MHz, CDCl3) 160.0, 155.8,
143.7, 142.7, 138.3, 129.8, 126.8, 122.5, 119.9, 119.7, 113.2, 112.7,
112.4, 110.2, 107.6, 55.3, 29.7. One quaternary carbon not observed.
LRMS (ESI): m/z found 280.9 (Mꢀ H)ꢀ . HRMS (ESI): m/z found
281.0850 (Mꢀ H)ꢀ , C17H13O4ꢀ required 281.0819.
4.2.8. Methyl 2-(6-(3-(trifluoromethyl)phenyl)benzofuran-3-yl)acetate
(4d)
Compound 3b (200 mg, 0.633 mmol) was coupled to m-fluo-
rophenylboronic acid (132 mg) according to General Procedure A2 at
120 ◦C for 15 min. The crude brown oil was purified by flash chroma-
tography (2.5–5% ethyl acetate in hexanes) which afforded the titled
compound as a white solid (153 mg, 72%), mp 75–77 ◦C. δH (500 MHz,
CDCl3) 7.88 (1H, s, ArH), 7.80 (1H, d, J 7.6, ArH), 7.70 (1H, s, ArH),
7.69 (1H, s, ArH), 7.65 (1H, d, J 8.1, ArH), 6.61 (1H, d, J 7.8, ArH), 7.57
(1H, t, J 7.7, ArH), 7.51 (1H, d, J 8.1, ArH), 3.76 (3H, s, OCH3), 3.75
(2H, s, CH2). δC (125 MHz, CDCl3) 171.0 (Cq), 155.8 (Cq), 143.8, 142.0,
136.7, 131.2 (Cq, q, J 32.0), 130.6, 129.3, 127.5 (Cq), 124.2 (Cq, d, J
272.4), 124.1 (q, J 3.8), 123.8 (q, J 3.7), 122.2, 120.1, 113.1 (Cq), 110.2,
52.2 (OCH3), 29.5 (CH2). δF (376 MHz, CDCl3) ꢀ 62.6. LRMS (ESI): m/z
found 335.2 (M+H)+, C18H14F3O3+ required 335.1.
4.2.4. 2-(6-(3-Cyanophenyl)benzofuran-3-yl)acetic acid (2c)
Compound 2b (300 mg, 0.949 mmol) was coupled to m-cyanophe-
nylboronic acid (837 mg, 5.69 mmol) according to General Procedure
◦
A1 at 90 C for 16 h. The crude material was purified by flash chro-
matography (2.5% methanol in dichloromethane). The fractions con-
taining the product were concentrated under reduced pressure and
recrystallised from toluene then from aqueous isopropanol to afford the
titled compound as white needles (37.1 mg, 14%), mp 159–161 ◦C. δH
(500 MHz, CDCl3) 7.90 (1H, s, ArH), 7.85 (1H, d, J 7.9, ArH) 7.71 (1H, s,
ArH), 7.68 (1H, s, ArH), 7.66 (1H, d, J 8.2, ArH), 7.64 (1H, d, J 7.8,
ArH), 7.56 (1H, t, J 9.7, ArH), 7.47 (1H, d, J 8.1, ArH), 3.80 (2H, s, CH2).
δC (100 MHz, CDCl3) 175.6, 155.8, 144.2, 142.4, 135.9, 131.7, 130.9,
130.6, 129.7, 127.6, 122.2, 120.3, 118.8, 113.1, 112.4, 110.3, 29.3.
LRMS (ESI): m/z found 300.1 (M + Na)+. HRMS (ESI): m/z found
276.0689 (Mꢀ H)ꢀ , C17H10NO3ꢀ required 276.0666.
4.2.9. 2-(6-Phenylbenzofuran-3-yl)acetic acid (5a)
Ester 4a (20.0 mg, 75.1 µM) was hydrolysed using 2 M methanolic
sodium hydroxide (1.5 eq, 57.8 µL) according to General Procedure B.
The resulting suspension was worked up after 20 h to afford the titled
product as a white solid (13.2 mg, 80%), mp 166–167 ◦C. δH (500 MHz,
CDCl3) 7.70 (1H, s, ArH), 7.68 (1H, s, ArH), 7.64–7.62 (3H, m, ArH),
7.52 (1H, dd, J 8.0, 1.5, ArH), 7.47–7.44 (2H, m, ArH), 7.36 (1H, t, J 7.5,
ArH), 3.79 (1H, s, CH2). δC (100 MHz, CDCl3) 175.7 (Cq), 155.9 (Cq),
143.6, 141.2 (Cq), 138.5 (Cq), 128.8, 127.4, 127.2, 126.6 (Cq), 122.4,
199.7, 112.4 (Cq), 110.1, 29.4 (CH2). LRMS (ESI): m/z found 250.9
4.2.5. 2-(5-Phenylbenzofuran-3-yl)acetic acid (2d)
Compound 1c (100 mg, 0.392 mmol) was coupled to phenylboronic
acid (287 mg) according to General Procedure A1 at 90 C for 16 h.
(Mꢀ H)ꢀ . HRMS (ESI): m/z found 251.0732 (Mꢀ H)ꢀ , C16H11O3
ꢀ
◦
Purification by flash chromatography (20% ethyl acetate, 1% acetic acid
in hexanes) gave a crude brown solid. Further purification by semi-
preparative RP-HPLC isolated the titled compound as a white solid
(5.0 mg, 5%), mp 123–125 ◦C. δH (400 MHz, CDCl3) 7.73 (1H, s, ArH),
7.68 (1H, s, ArH), 7.62–7.59 (2H, m, ArH), 7.54 (2H, d, J 1.2, ArH), 7.5
(2H, t, J 7.6 ArH), 7.35 (1H, t, ArH), 3.80 (2H, s, CH2). δC (100 MHz,
CDCl3) 176.0 (Cq), 154.8 (Cq), 143.7, 141.5 (Cq), 136.6 (Cq), 128.7,
127.9 (Cq),127.5, 127.0, 124.4, 118.1, 112.6 (Cq), 111.7, 29.3 (CH2).
LRMS (ESI): m/z found 251.0 (Mꢀ H)ꢀ . HRMS (ESI): m/z found
251.0770 (Mꢀ H)ꢀ , C16H11O3ꢀ required 251.0714.
required 251.0714.
4.2.10. 2-(6-(3-Carboxyphenyl)benzofuran-3-yl)acetic acid (5b)
Compound 3a (200 mg, 0.743 mmol) was coupled to m-carbox-
yphenylboronic (185 mg, 1.11 mmol) as per General Procedure A1 at
90 ◦C for 24 h. Flash chromatography (1.25% methanol in dichloro-
methane) isolated a mixture of the coupled product and an inseparable
by-product in a 1:1 ratio. Semi-preparative RP-HPLC improved the ratio
of cross-coupled product and a by-product to a ratio of 2:1. The mixture
was hydrolysed according to General Procedure B to give the crude acid.
Purification by flash chromatography (1.25–2.5% methanol, 1% acetic
acid in dichloromethane) afforded the titled compound as a white solid
4.2.6. Methyl 2-(6-phenylbenzofuran-3-yl)acetate (4a)
◦
Compound 3a (200 mg, 0.743 mmol) was coupled to phenylboronic
acid (136 mg, 1.12 mmol) according to General Procedure A1 at 60 ◦C
for 28 h. The crude compound was purified by flash chromatography
(1–10% ethyl acetate in hexanes) which isolated the starting material
along with the cross-coupled compound. The two compounds were
separated by semi-preparatory RP-HPLC to afford the titled compound
as a white powder (40 mg, 20%), mp 106–108 ◦C. δH (500 MHz, CDCl3)
7.70 (1H, s, ArH), 7.66 (1H, s, ArH), 7.65–7.61 (3H, m, ArH), 7.25 (1H,
dd, J 8.5, 1.5, ArH), 7.47–7.44 (2H, m, ArH), 7.36 (1H, t, J 5.5, ArH),
3.75 (3H, s, OCH3), 3.74 (2H, s, CH2). δC (125 MHz, CDCl3) 171.1 (Cq),
155.9 (Cq), 143.4, 141.2 (Cq), 138.3 (Cq), 128.8, 127.4, 127.2, 126.8
(Cq), 122.3, 119.8, 113.0 (Cq), 110.1, 52.2 (OCH3), 29.6 (CH2). LRMS
(ESI): m/z found 267.1 (M+H)+, C17H15O3+ required 267.1.
(30 mg, 14%), mp 129–132 C. δH (400 MHz, DMSO‑d6) 8.22 (1H, s,
ArH), 7.98–7.89 (4H, m, ArH), 7.69 (1H, d, J 8.0, ArH), 7.61–7.57 (2H,
m, ArH), 3.72 (2H, s, CH2). δC (100 MHz, DMSO‑d6) 172.4 (Cq), 167.8
(Cq), 155.8 (Cq), 144.9, 141.11 (Cq), 136.5 (Cq), 132.0 (Cq), 132.0,
129.9, 128.6, 128.1 (Cq), 128.0, 122.3, 121.3, 114.6 (Cq), 110.1, 29.5
(CH2). LRMS (ESI): m/z found 295.3 (Mꢀ H)ꢀ . HRMS m/z found
295.0643 (Mꢀ H)ꢀ , C17H11O5ꢀ required 295.0612.
4.2.11. 2-(6-(4-Methoxyphenyl)benzofuran-3-yl)acetic acid (5c)
Ester 4c (57.4 mg, 194 µmol) was hydrolysed using 2 M methanolic
sodium hydroxide (3 eq., 29.1 µL) according to General Procedure B.
The resulting turbid solution was worked up after 16 h to afford the
titled compound as an off white solid (49.2 mg, 90%), mp 220–222 ◦C.
δH (400 MHz, DMSO‑d6) 7.90 (1H, s, ArH), 7.79 (1H, s, ArH), 7.68–7.62
(3H, m, ArH), 7.53 (1H, d, J 8.2, ArH), 7.03 (2H, d, J 8.7, ArH), 3.80 (3H,
s, OCH3), 3.71 (2H, s, CH2). δC (100 MHz, DMSO‑d6) 172.4, 159.3,
155.8, 144.2, 137.2, 133.1, 128.5, 127.0, 121.8, 120.9, 114.9, 114.5,
4.2.7. Methyl 2-(6-(4-methoxyphenyl)benzofuran-3-yl)acetate (4c)
Compound 3b (100 mg, 317 µmol) was coupled to p-methox-
yphenylboronic acid (1.90 mmol) according to General Procedure A2 at
9

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
109.2, 55.7, 29.5. LRMS (ESI): m/z found 280.93 (Mꢀ H)ꢀ . HRMS (ESI):
m/z found 281.0846 (Mꢀ H)ꢀ , C17H13O4ꢀ required 281.0819.
methanolic sodium hydroxide (3 eq., 46 µL) according to General Pro-
cedure B to afford the titled compound as a white solid (6.1 mg, 100%).
δH (400 MHz, CDCl3) 7.79 (2H, s, ArH), 7.68 (1H, s, ArH), 7.52–7.38
(5H, m, ArH), 3.88 (2H, s, CH2), 2.17 (3H, s, OCH3). δC (100 MHz,
CDCl3) 175.8 (Cq), 154.2 (Cq), 153.8 (Cq), 129.7 (Cq), 129.2, 128.9,
128.8, 127.4, 126.3, 129.7, 118.0 (Cq), 114.7, 107.9 (Cq), 30.91 (CH2).
4.2.12. 2-(6-(3-(Trifluoromethyl)phenyl)benzofuran-3-yl)acetic acid (5d)
Ester 4d (152 mg, 0.455 mmol) was hydrolysed using 2 M meth-
anolic sodium hydroxide (3 eq., 682 µL) according General Procedure B
to afford the titled compound as a white solid (132 mg, 91%), mp
123–125 ◦C. δH (400 MHz, CDCl3) 7.87 (1H, s, ArH), 7.80 (1H, d, J 7.5,
ArH), 7.71 (1H, s, ArH), 7.69 (1H, s, ArH), 7.65 (1H, d, J 8.1, ArH), 7.61
(1H, d, J 7.9, ArH), 7.57 (1H, t, J 7.6, ArH), 7.51 (1H, d, J 8.1, ArH), 3.80
(2H, s, CH2). δC (100 MHz, CDCl3) 176.6 (Cq), 155.8 (Cq), 144.0, 141.9
(Cq), 136.8 (Cq), 131.2 (Cq, q, J 32.0), 130.6, 129.3, 127.3, 124.2 (Cq, d,
J 270.4), 124.1 (q, 3.7), 123.9 (q, J 3.7), 122.3, 120.1, 112.3 (Cq), 110.3,
29.4 (CH2). δF (376 MHz, CDCl3) ꢀ 63.6. LRMS (ESI): m/z found 318.9
4.2.16. 2-(2,6-Diphenylbenzofuran-3-yl)acetic acid (8b)
Compound 7c (2.6 mg, 0.0076 mmol) was hydrolysed using 2 M
methanolic sodium hydroxide (5 eq., 16 µL) according General Pro-
cedure B to afford the titled compound as a white solid (2.5 mg, 100%).
δH (400 MHz, CDCl3) 7.84 (2H, d, J 1.4, ArH), 7.74 (1H, s, ArH),
7.68–7.64 (3H, m, ArH), 7.55–7.41 (6H, m, ArH), 7.36 (1H, t, J 7.4,
ArH), 3.95 (2H, s, CH2). δC (100 MHz, CDCl3) 175.3 (Cq), 154.6 (Cq),
153.7 (Cq), 141.3 (Cq), 138.6 (Cq), 130.2 (Cq), 128.93, 128.91, 128.88,
128.84, 127.40, 127.39, 127.2, 122.6 (Cq), 199.8, 109.7, 107.9 (Cq),
53.4, 29.7 (CH2).
ꢀ
(Mꢀ H)ꢀ . HRMS (ESI): m/z found 319.0617 (Mꢀ H)ꢀ , C17H10F3O3
required 319.0588.
4.2.13. Methyl 2-(2,6-dimbromobenzofuran-3-yl)acetate (6a)
A solution of 3a (100 mg, 0.371 mmol) in acetic acid (3.23 mL) was
treated with a solution of bromine (19.0 µL) in acetic acid (808 µL) and
the brown solution stirred at room temperature for 3 h. The solution was
poured onto cracked ice and the solid that precipitated was collected by
vacuum filtration. The crude off white solid was recrystallised from
isopropanol to afford the titled compound as a white solid (67.2 mg,
52%), mp 120–122 ◦C. δH (500 MHz, CDCl3) 7.62 (1H, s, ArH), 7.39 (1H,
d, J 8.0, ArH), 7.37 (1H, d, J 8.5, ArH), 3.72 (3H, s, OCH3), 3.65 (2H, s,
CH2). δC (125 MHz, CDCl3) 175.2 (Cq), 155.5 (Cq), 132.5, 129.2 (Cq),
127.6 (Cq), 120.5, 120.3, 112.5 (Cq), 88.2 (Cq), 30.0, 25.2. LRMS (ESI):
m/z found 368.52 (M[79Br,79Br] + Na)+, 370.93 (M[79Br,81Br] + Na)+,
372.92 (M[81Br,81Br] + Na)+. HRMS (ESI): m/z found 368.8744 (M
4.2.17. Methyl 2-(2-bromo-6-iodobenzofuran-3-yl)acetate (9a)
Compound 3b (1.06 g, 3.35 mmol) was dissolved in acetic acid (20
mL) and treated with bromine (189 µL, 3.68 mmol). The reaction stirred
for three hours then was poured onto ice. The resulting precipitate was
collected by vacuum filtration then recrystallised from isopropanol to
give the titled compound as an off white solid (751 mg, 57%), mp
131–133 ◦C. δH (400 MHz, CDCl3) 7.80 (1H, s, ArH), 7.56 (1H, d J, J 8.2,
ArH), 7.25 (1H, d J, J 8.3, ArH), 3.72 (3H, s, OCH3), 3.64 (2H, s, CH2). δC
(100 MHz, CDCl3) 169.8 (Cq), 155.5 (Cq), 132.5, 128.8 (Cq), 127.8 (Cq),
120.6, 120.2, 113.1 (Cq), 88.0 (Cq), 52.3 (OCH3), 30.2 (CH2). LRMS
(ESI): m/z found 394.92 (M[79Br] + H)+, 396.92 (M[81Br] + H)+. HRMS
m/z found 394.8798 (M[79Br] + H)+, 396.8760 (M[81Br] + H)+
,
79Br,79Br] + Na)+, 370.8725 (M[79Br,81Br] + Na)+, 372.8700 (M
C11H9[79Br]IO3
required 394.8774, C11H9[81Br]IO3
required
+
+
[
[
81Br,81Br]
+
Na)+, C11H8[79,79Br]2NaO3
required 368.8732,
396.8754.
+
C
11H8[79,81Br]2NaO3
required 370.8712, C11H8[81,81Br]2NaO3
+
+
required 372.8691.
4.2.18. Methyl 2-(2-bromo-6-(3-methoxyphenyl)benzofuran-3-yl)acetate
(10a)
4.2.14. Suzuki coupling of 6a to phenylboronic acid to give 7a-c
Compound 9a (600 mg, 0.506 mmol) was coupled to m-methox-
yphenylboronic acid (254 mg) according to General Procedure A2 at
80 ◦C for 1 h. Purification by flash chromatography (2.5% ethyl acetate
in hexanes) afforded the titled compound as an oily light brown semi
solid which was lyophilised to a white powder (315 mg, 55%), mp
50–52 ◦C. δH (500 MHz, CDCl3) 7.65 (1H, s, ArH). 7.55 (1H, d, J 8.2,
ArH), 7.50 (1H, d, J 8.2, ArH), 7.37 (1H, t, J 8.0, ArH), 7.20 (1H, d, J 7.7,
ArH), 7.14 (1H, s, ArH), 7.92 (1H, d, J 8.2, ArH), 6.90 (3H, s, OCH3),
3.74 (3H, s, OCH3), 3.70 (2H, s, CH2). δC (100 MHz, CDCl3) 170.1,
160.0, 155.8, 142.4, 138.2, 129.9, 128.6, 127.4, 123.0, 119.8, 119.2,
113.1, 113.0, 112.7, 109.6, 55.3, 52.3, 30.3. LRMS (ESI): m/z found
Dibromo 6a (43 mg, 0.1224 mmol) was coupled to phenylboronic
acid according to General Procedure A1 at 90 ◦C for 24 h, after which
time analytical RP-HPLC indicated unreacted starting material and three
new products. The reaction mixture was cooled to room temperature
and treated with brine then extracted with ethyl acetate (x3). The
combined organic phase was dried (MgSO4) then concentrated under
reduced pressure to a brown residue. The crude material, which
comprised of starting material and three products, was separated by
semi-preparative RP-HPLC. The fractions containing 6a and 7a-c were
lyophilised. Unreacted starting material 6a eluted at 19.7 min (10 mg,
23% recovery). Methyl 2-(2-bromo-6-phenylbenzofuran-3-yl)acetate (7a)
eluted at 21.2 min (2.0 mg, 1%). δH (500 MHz, CDCl3) 7.65 (1H, dd, J
1.4, 0.6, ArH), 7.62–7.61 (1H, m, ArH), 6.61–7.60 (1H, m, ArH), 7.55
(1H, dd, J 8.13, 0.6, ArH), 7.51 (1H, dd, J 8.2, 1.5, ArH), 7.46 (2H, m,
ArH), 7.36 (1H, tt, 7.4, 1.2, ArH), 3.74 (3H, s, OCH3), 3.70 (2H, s, CH2).
Methyl 2-(6-bromo-2-phenylbenzofuran-3-yl)acetate (7b) eluted at 22.1
min (6.4 mg, 15%). δH (500 MHz, CDCl3) 7.82–7.80 (2H, m, ArH), 7.68
(1H, dd, J 1.6, 0.4, ArH), 7.52–7.49 (2H, m, ArH), 7.48 (1H, dd, J 8.4,
0.35, ArH), 7.43 (1H, tt, J 7.4, 1.29, ArH), 7.40 (1H, dd, J 8.29, 1.65,
ArH), 3.86 (2H, s, CH2), 3.75 (3H, s, OCH3). Regioisomers 7a and 7b
were distinguishable from on another due to the chemical shift of the
methylene singlet (see supplementary Fig. S5). Methyl 2-(2,6-diphe-
nylbenzofuran-3-yl)acetate (7c) eluted at 23.1 min (2.6 mg, 1%). δH (500
MHz, CDCl3) 7.87–7.85 (2H, m, ArH), 7.74 (1H, d, J 0.9, ArH),
7.68–7.65 (3H, m, ArH), 7.55–7.50 (3H, m, ArH), 7.49–7.45 (2H, m,
ArH), 7.43 (1H, tt, J 7.4, 1.2, ArH), 7.36 (1H, tt, J 7.3, 1.2, ArH), 3.92
(2H, s, CH2), 3.77 (3H, s, OCH3).
375.1 (M[79Br] + H)+, 377.0 (M[81Br] + H)+, C18H16
[
79Br]O4+ required
375.0, C18H16
[
81Br]O4+ required 377.0.
4.2.19. Methyl 2-(2,6-bis(3-methoxyphenyl)benzofuran-3-yl)acetate
(10b)
Compound 9a (100 mg, 0.253 mmol) was coupled to m-methox-
yphenylboronic acid (156 mg) according to General Procedure A2 at
◦
80 C for 50 min. Purification by flash chromatography (5–10% ethyl
acetate in hexanes) afforded the titled compound as a light brown oil
(77.6 mg, 76%), mp 91–91 ◦C. δH (400 MHz, CDCl3) 7.73 (1H, s, ArH),
6.67 (1H, d, J 8.1, ArH), 7.54 (1H, d, J 8.1, ArH), 7.45–7.37 (4H, m,
ArH), 7.26–7.24 (1H, m, ArH), 7.20 (1H, s, ArH), 7.00–6.97 (1H, m,
ArH), 6.92 (1H, d, J 8.2), 3.92 (2H, s, CH2), 3.91 (3H, s, OCH3), 3.76
(3H, s, OCH3). δC (100 MHz, CDCl3) 171.3, 156.0, 159.9, 154.5, 153.4,
142.8, 138.3, 131.5, 129.9, 129.8, 129.2, 122.5, 119.9, 119.8, 119.8,
115.1, 113.1, 112.6, 112.4, 109.7, 108.8, 55.4, 55.3, 52.3, 30.7. LRMS
(ESI): m/z found 425.1 (M + Na)+, C25H22NaO5+ required 425.1.
4.2.15. 2-(6-Bromo-2-phenylbenzofuran-3-yl)acetic acid (8a)
4.2.20. 2-(2,6-Bis(3-methoxyphenyl)benzofuran-3-yl)acetic acid (11)
Ester 10b (77.0 mg, 0.191 mmol) was hydrolysed using 2 M
Compound 7b (6.4 mg, 0.019 mmol) was hydrolysed using 2 M
10

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
methanolic sodium hydroxide (3 eq., 287 µL) according to General
Procedure B. Purification by flash chromatography (5% methanol in
dichloromethane) afforded the titled compound as an off-white semi-
4.2.24. 3-(3-(Carboxymethyl)-6-(3-methoxyphenyl)benzofuran-2-yl)
benzoic acid (13b)
Ester 12b (87.2 mg, 0.209 mmol) was hydrolysed using 2 M meth-
anolic sodium hydroxide (5 eq., 524 µL) according to General Procedure
B. The reaction was worked up after 16 h to afford the titled compound
as a white solid (76.6 mg, 91%), mp 244–246 ◦C. δH (400 MHz,
DMSO‑d6) 8.39 (1H, s, ArH), 8.06 (1H, d, J 7.7, ArH), 8.01 (1H, d, J 8.0,
ArH), 7.99 (1H, s, ArH), 7.75 (1H, d, J, 8.3, ArH), 7.70 (1H, t, J 7.8,
ArH), 7.65 (1H, d, J 8.2, ArH), 7.39 (1H, t, J 7.8, ArH), 7.32 (1H, d, J 7.7,
ArH), 7.29 (1H, s, ArH), 6.94 (1H, d, J 7.9, ArH), 3.94 (2H, s, CH2), 3.84
(3H, s, OCH3). δC (100 MHz, DMSO‑d6) 172.1 (Cq), 167.3 (Cq), 160.3
(Cq), 154.4 (Cq), 151.6 (Cq), 142.1 (Cq), 138.1 (Cq), 132.1 (Cq), 131.0,
130.7 (Cq), 130.5, 130.0, 129.9, 129.7 (Cq), 127.7 122.8, 121.0, 119.8,
113.6, 112.9, 111.3 (Cq), 109.7, 55.6 (OCH3), 30.5 (CH2). LRMS (ESI):
m/z found 400.9 (Mꢀ H)ꢀ . HRMS (ESI): m/z found 401.1077 (Mꢀ H)ꢀ ,
◦
solid (40.4 mg, 54%), 137–139 C. δH (500 MHz, CDCl3) 7.73 (1H, s,
ArH), 7.66 (1H, d, J 8.1, ArH), 7.53 (1H, d, J 8.1, ArH), 7.42–7.41 (3H,
m, ArH), 7.38 (1H, t, J 7.9, ArH), 7.25 (1H, d, J 7.6, ArH), 7.19 (1H, s,
ArH), 7.00–6.96 (1H, m, ArH), 6.92 (1H, d, J 8.2, ArH), 3.94 (2H, s,
CH2), 3.89 (6H, s, OCH3). δC (125 MHz, CDCl3) 176.8 (Cq), 160.0 (Cq),
159.9 (Cq), 154.4 (Cq), 153.6 (Cq), 142.8 (Cq), 138.4 (Cq), 131.3 (Cq),
129.9, 129.8, 129.0, 122.6, 119.9, 119.8, 119.8, 115.1, 113.1, 1126,
112.5, 109.8, 108.1 (Cq), 55.4 (OCH3), 55.3 (OCH3), 30.6 (CH2). LRMS
+
(ESI): m/z found 389.2 (M+H)+. HRMS (ESI) did not ionise, C24H21O5
required 389.1.
4.2.21. Methyl 2-(6-(3-methoxyphenyl)-2-(4-methoxyphenyl)benzofuran-
3-yl)acetate (12a)
C
24H17O6ꢀ required 401.1031.
Compound 10a (94.0 mg, 0.251 mmol) was coupled to m-methox-
yphenylboronic acid (76.1 mg) according to General Procedure A2 at
4.3. Expression and Purification of Unlabelled and 15N-Labelled EcDsbA
◦
80 C for 50 min. Purification by flash chromatography (5–10% ethyl
Unlabelled and 15N-labelled EcDsbA were prepared as previously
described.14 Briefly, both unlabelled and 15N-labelled EcDsbA were
expressed by autoinduction using appropriate media. After harvesting
the cells, periplasmic proteins were extracted by resuspending cell pellet
with lysis buffer (20 mM Tris-HCl, pH 8.0, 25 mM NaCl, 4 mg/mL
colistin sulfate) and incubating for 18–24 h with gentle stirring at 4 ◦C.
After centrifugation, EcDsbA was purified from the supernatant using a
three-step purification method, including hydrophobic interaction
chromatography, anion exchange chromatography, and size exclusion
chromatography. DsbA was oxidised with copper-phenanthroline and
buffer exchanged to the final storage buffer (20 mM HEPES, pH 7.0).
acetate in hexanes) afforded the titled compound as an oily light brown
semi-solid which was lyophilised to an off-white solid (60.8 mg, 60%),
mp 109–111 ◦C. δH (400 MHz, CDCl3) 7.81 (2H, d, J 9.0, ArH), 7.72 (1H,
s, ArH), 7.64 (1H, d, J 8.1, ArH), 7.52 (1H, d, J 8.1, ArH), 7.38 (1H, t, J
7.9, ArH), 7.26–7.24 (1H, m, ArH), 7.20 (1H, s, ArH), 7.05 (2H, d, J 8.9,
ArH), 6.91 (1H, d, J 8.2, ArH), 3.89 (3H, s, OCH3), 3.88 (3H, s, OCH3),
3.88 (2H, s, CH2), 3.76 (3H, s, OCH3). δC (100 MHz, CDCl3) 171.4,
160.1, 160.0, 154.3, 153.6, 142.9, 137.8, 129.8, 129.4, 128.8, 122.9,
122.4, 119.8, 119.5, 114.3, 113.1, 112.5, 109.5, 107.2, 55.3, 55.3, 52.3,
30.6. LRMS (ESI): m/z found 403.2 (M+H)+, C25H23O5+ required 403.2.
4.2.22. 3-(3-(2-Methoxy-2-oxoethyl)-6-(3-methoxyphenyl)benzofuran-2-
yl)benzoic acid (12b)
4.4. Protein crystallisation and structure determination
Compound 10a (44 mg, 0.267 mmol) was coupled to m-carbox-
yphenylboronic acid (66.3 mg) according to General Procedure A2 at
EcDsbA was crystallised as previously described.14 Briefly, 1 µL of 30
mg/mL EcDsbA was mixed with an equal volume of crystallisation buffer
(11–13% PEG8000, 5–7.5% glycerol, 1 mM CuCl2, 100 mM, sodium
cacodylate pH 6.1) and equilibrated against 500 µL of reservoir buffer at
20 ◦C using hanging drop vapour diffusion. For compound soaking,
crystals were transferred into 2 µL drops of 24% PEG8000, 22% glycerol,
and 100 mM sodium cacodylate (pH 6.1) containing a compound of
interest at the final concentration of 10 mM (5% DMSO) and were
incubated for 3–6 h. Crystals were mounted on loops and flash-cooled in
liquid nitrogen.
◦
120 C for 15 min. Purification by flash chromatography (2.5% meth-
anol in dichloromethane) afforded the titled compound as an off-white
solid (87.4 mg, 79%), mp 60–62 ◦C. δH (400 MHz, CDCl3) 8.65 (1H, s,
ArH), 8.18 (1H, d, J 7.8, ArH), 8.13 (1H, d, J 7.8, ArH), 7.76 (1H, s,
ArH), 7.69 (1H, d, J 8.1, ArH), 7.65 (1H, t, J 7.8, ArH), 7.56 (1H, d, J 8.1,
ArH), 7.39 (1H, t, J 7.9, ArH), 7.26 (1H, d, J 7.6, ArH), 7.20 (1H, s, ArH),
6.93 (1H, d, J 8.2, ArH), 3.95 (2H, s, CH2), 3.90 (3H, s, OCH3), 3.80 (3H,
s, OCH3). δC (100 MHz, CDCl3) 171.4 (Cq), 171.0 (Cq), 160.0 (Cq), 154.6
(Cq), 152.0 (Cq), 142.7 (Cq), 138.7 (Cq), 132.2, 130.9 (Cq), 130.2, 130.0
(Cq), 129.8, 129.1, 129.0 (Cq), 128.9, 122.8, 120.0, 119.9, 113.1, 112.7,
109.8, 109.6 (Cq), 55.3 (OCH3), 52.4 (OCH3), 30.6 (CH2). LRMS (ESI):
m/z found 417.1 (M+H)+, C25H21O6+ required 417.1.
Datasets were collected at the Australian synchrotron on MX1 and
MX2 beamlines using the Blue-Ice software (CA, USA).30 MX1 beamline
was equipped with an ADSC Quantum 210r detector and MX2 with an
ADSC Quantum 315r detector. All datasets were indexed and integrated
with iMOSFLM31 or XDS32 and scaled using AIMLESS.33,34 Phasing was
performed by molecular replacement with Phaser35 using the previously
solved structure of EcDsbA as a search model (PDB code 1FVK).36 The
final structure was obtained after several rounds of manual model
building using Coot37 and refinement in phenix.refine.38 Data collection
and refinement statistics are summarised in Table S1. Structures, factors
and coordinates have been deposited in the Protein Data Bank (PDB;
http://www.pdb.org) under the accession codes of 7L76, 7L7C, 7LHP,
6XSP, 6XSQ and 6XT3.
4.2.23. 2-(6-(3-Methoxyphenyl)-2-(4-methoxyphenyl)benzofuran-3-yl)
acetic acid (13a)
Compound 12a (60 mg, 0.149 mmol) was hydrolysed using 2 M
methanolic sodium hydroxide (3 eq., 224. µL) according to General
Procedure B. The titled compound was afforded as a pale brown oil that
solidified under reduced pressure to a light brown solid (30.4 mg, 53%),
mp 150–153 ◦C. δH (500 MHz, CDCl3) 7.79 (2H, d, J 8.9, ArH), 7.71 (1H,
s, ArH), 7.64 (1H, d, J 8.1, ArH), 7.52 (1H, d, J 8.1, ArH), 7.38 (1H, t, J
7.9, ArH), 7.24 (1H, d, J 7.6, ArH), 7.19 (1H, s, ArH), 7.04 (2H, d, J 8.9,
ArH), 6.91 (1H, d, J 8.2, ArH), 3.90 (2H, s, CH2), 3.89 (3H, s, OCH3),
3.87 (3H, s, OCH3). δC (125 MHz, CDCl3) 199.1 (Cq), 160.2 (Cq), 160.0
(Cq), 154.3 (Cq), 153.9 (Cq), 142.9 (Cq), 137.9 (Cq), 129.8, 129.2 (Cq),
128.8, 122.7 (Cq), 122.5, 119.9, 119.5, 114.3, 113.1, 112.6, 109.6,
106.5 (Cq), 55.4 (OCH3), 55.3 (OCH3), 30.5 (CH2). LRMS (ESI): m/z
4.5. NMR spectroscopy
For ligand-detected NMR spectroscopy, two samples (compound
alone and compound + protein) were prepared for each compound.
Unlabelled EcDsbA was prepared at 10 µM in a buffer of 50 mM sodium
phosphate, pH 6.8, 25 mM NaCl, 100 µM 4,4-dimethyl 4-silapentane-1-
sulfonic acid (DSS), and 10% 2H2O. Compound was added to the protein
sample to achieve final concentration of 500 µM (2% 2H6-DMSO). 1H 1D
spectrum was acquired for each compound at 500 µM as a reference
found 411.1 (M + Na)+. HRMS (ESI) did not ionise, C24H20NaO5
+
required 411.1.
11

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
spectrum. All ligand-detected NMR experiments (STD, CPMG, and
waterLOGSY) were acquired at 298 K on a Bruker Avance III 600 MHz
NMR spectrometer equipped with a 5 mm TXI CryoProbe. STD spectra
were acquired with 3 s of saturation at ꢀ 1 ppm (on-resonance) and 33.3
ppm (off-resonance). Compounds that gave an average difference in
signal intensity > 3% between the on- and off-resonance were identified
as hits. WaterLOGSY spectra were acquired for compound alone and
compound + protein with a 0.52 s acquisition time, 3 s relaxation delay,
and 3 s NOE (Nuclear Overhauser Effect) mixing time. Compounds that
gave an average difference in signal intensity > 70% between compound
alone and compound + protein were identified as hits. CPMG spectra for
compound alone and compound + protein were acquired with a con-
stant spin echo delay of 1 ms and spin-lock period of 350 ms. Com-
pounds that gave an average difference in signal intensity > 30%
between compound alone and compound + protein were identified as
hits. The data was analysed using Topspin3.5 (Bruker) and DSS was used
to reference the spectra.
analysed and plotted using GraphPad Prism 8.
4.7. Motility assays
Swimming motility of E. coli JCB816 was assessed as previously
described.41 Briefly, 2 µL of four independent liquid overnight JCB816
cultures were inoculated onto the surface of LB semi-solid (0.3% w/v)
agar containing 0.6% dimethyl sulfoxide (DMSO) or DsbA inhibitors 11,
13a or 13b at various concentrations (5
μM – 1 mM). Plates were
incubated at 37 ◦C, and the diameter of bacterial outward growth was
measured in millimetres at 6 and 9 h post inoculation. The mean motility
zone diameter was calculated from four replicates tested under each
condition, and group means were compared by one-way ANOVA (sta-
tistical significance set at p < 0.05). Bacterial motility in the presence of
different DsbA inhibitor concentrations was plotted as % motility over
concentration (M) by measuring the diameter of the motility zone in
plates containing 0.005, 0.05, 0.1, or 1 mM inhibitors and dividing by
the diameter of the same strain swimming in DMSO-containing media
(carrier control).
For the 2D 1H-15N-HSQC titration experiment, 100 µM of 15N-
labelled EcDsbA in a buffer of 50 mM HEPES, pH 6.8, 50 mM NaCl, 2%
2H6-DMSO, and 10% 2H2O, was titrated with increasing concentrations
of compounds (0, 0.625, 0.125, 0.25, 0.5, and 1 mM). Data were ac-
quired on a Bruker 600 MHz spectrometer equipped with a 5 mm TXI
CryoProbe and a Bruker 700 MHz spectrometer equipped with a 5 mm
TXI CryoProbe. Data was processed by Topspin3.5 (Bruker) and ana-
lysed by Sparky (T. D. Goddard and D. G. Kneller, SPARKY 3, University
of California, San Francisco (UCSF)). Weighted chemical shift pertur-
bations (CSP) that were observed upon addition of compounds to
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
EcDsbA were calculated using equation 1 39
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
:
Acknowledgments
CSP = ΔδH2 + (0.2 × ΔδN )2 ,
This research was supported by National Health and Medical
Research Council (NHMRC) project (grants 1099151 and 1144046) and
the Australian research council (Future Fellowship (FT130100580) and
Discovery Project (DP190101613)). Dr Luke Duncan would like to
acknowledge La Trobe University as a recipient of La Trobe University
Postgraduate Research Scholarship. Dr Makrina Totsika would like to
acknowledge support from Queensland University of Technology as a
recipient of a Vice-Chancellor’s Research Fellowship. We would also like
to acknowledge the La Trobe University-Comprehensive Proteomics
Platform and the Monash Fragment Platform (MFP) for providing
infrastructure and expertise. This research was undertaken on the MX1
and MX2 beamlines at the Australian Synchrotron, part of Australian
Nuclear Science and Technology Organisation (ANSTO). We are very
grateful for the ongoing support and many helpful chemical discussions
provided by Dr Les Deady.
where ΔδH and ΔδN denote the change in chemical shift of amide proton
and nitrogen resonances upon addition of the compound. Equilibrium
dissociation constants (KD) were determined by fitting the plot of CSP
against compound concentrations using one site with ligand depletion
model (equation 2):
[
]
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
ꢀ
)
CSPmax
×
(KD + Pt + Lt) ꢀ
(KD + Pt + Lt)2 ꢀ 4 × Pt × Lt
CSP =
,
2 × Pt
where CSP is the measured CSP at a given concentration, CSPmax is the
CSP observed at the saturating concentration of compound, and Pt and Lt
are the total protein and compound concentrations. The ligand effi-
ciency (LE) is calculated using equation 3 40
:
ΔG
HAC
RTln(KD)
LE = ꢀ
= ꢀ
Appendix A. Supplementary material
HAC
where R is gas constant (1.9858775 cal Kꢀ 1 molꢀ 1), T is temperature
(298 K), KD is equilibrium dissociation constant, and HAC is the number
of heavy (non-hydrogen) atoms.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmc.2021.116315.
References
4.6. Peptide oxidation assay
1
2
3
World Health Organization, Antimicrobial resistance: global report on surveillance;
2014.
The ability of compounds to inhibit the activity of EcDsbA was
evaluated using a peptide oxidation assay as previously described.40
Inhibition of DsbA activity is assessed as a reduction in the rate of
fluorescence signal increase in the presence of compounds. A standard
World Health Organization, 2019 antibacterial agents in clinical development: an
analysis of the antibacterial clinical development pipeline; 2019.
Totsika M. Disarming pathogens: benefits and challenges of antimicrobials that target
bacterial virulence instead of growth and viability. Future Med Chem. 2017;9:
267–269.
50
μ
L reaction mixture consists of 0.5
16 substrate
μ
M EcDsbA, 1 mM oxidised
4
5
6
7
8
Heras B, Scanlon MJ, Martin JL. Targeting virulence not viability in the search for
future antibacterials. Br J Clin Pharmacol. 2015;79:208–215.
Dickey SW, Cheung GYC, Otto M. Different drugs for bad bugs: antivirulence
strategies in the age of antibiotic resistance. Nat Rev Drug Discov. 2017;16:457–471.
Casadevall A, Pirofski LA. Host-pathogen interactions: redefining the basic concepts
of virulence and pathogenicity. Infect Immun. 1999;67:3703–3713.
Heras B, Shouldice SR, Totsika M, Scanlon MJ, Schembri MA, Martin JL. DSB
proteins and bacterial pathogenicity. Nat Rev Microbiol. 2009;7:215–225.
Landeta C, Boyd D, Beckwith J. Disulfide bond formation in prokaryotes. Nat
Microbiol. 2018;3:270–280.
glutathione,
μM
peptide (DOTA/Eu(III)-
CQQGFDGTQNSCK-MCA) and compounds in a buffer of 50 mM MES, pH
5.5, 50 mM NaCl, 2 mM EDTA, 2% DMSO. Assays were performed in a
PerkinElmer 384-well white opaque microplate (OptiPlate-384). The
time-resolved fluorescence of substrate peptide oxidation (excitation λ
= 340 nm and emission λ = 615 nm) was measured using a CLARIOstar
plate reader (BMG Labtech) fitted with TR-FRET module. Data was
12

L.F. Duncan et al.
Bioorganic & Medicinal Chemistry 45 (2021) 116315
9
Shouldice SR, Heras B, Walden PM, Totsika M, Schembri MA, Martin JL. Structure
and Function of DsbA, a Key Bacterial Oxidative Folding Catalyst. Antioxid Redox
Signal. 2011;14:1729–1760.
27 Verderosa AD, Dhouib R, Hong Y, Anderson TK, Heras B, Totsika M. A high-
throughput cell-based assay pipeline for the preclinical development of bacterial
DsbA inhibitors as antivirulence therapeutics. Sci Rep. 2021;11:1569–1581.
28 Armarego WLF, Chai CLL. Chapter 4 - Purification of Organic Chemicals. In:
Armarego WLF, Chai CLL, eds. Purification of Laboratory Chemicals (Fifth Edition).
Burlington: Butterworth-Heinemann; 2003:80–388.
10 Smith RP, Paxman JJ, Scanlon MJ, Heras B. Targeting Bacterial Dsb Proteins for the
Development of Anti-Virulence Agents. Molecules. 2016;21:811–825.
11 Martin JL, Bardwell JCA, Kuriyan J. Crystal structure of the DsbA protein required
for disulphide bond formation in vivo. Nature. 1993;365:464–468.
12 Inaba K, Murakami S, Suzuki M, et al. Crystal structure of the DsbB-DsbA complex
reveals a mechanism of disulfide bond generation. Cell. 2006;127:789–801.
13 Duprez W, Premkumar L, Halili MA, et al. Peptide inhibitors of the Escherichia coli
DsbA oxidative machinery essential for bacterial virulence. J Med Chem. 2015;58:
577–587.
29 Theodorou V, Skobridis K, Tzakos AG, Ragoussis V. A simple method for the alkaline
hydrolysis of esters. Tetrahedron Lett. 2007;48:8230–8233.
30 McPhillips TM, McPhillips SE, Chiu HJ, et al. Blu-Ice and the Distributed Control
System: software for data acquisition and instrument control at macromolecular
crystallography beamlines. J Synchrotron Radiat. 2002;9:401–406.
31 Battye TGG, Kontogiannis L, Johnson O, Powell HR, Leslie AGW. iMOSFLM: A new
graphical interface for diffraction-image processing with MOSFLM. Acta Crystallogr,
Sect D: Biol Crystallogr. 2011;67:271–281.
14 Adams LA, Sharma P, Mohanty B, et al. Application of Fragment-Based Screening to
the Design of Inhibitors of Escherichia coli DsbA. Angew Chem, Int Ed. 2015;54:
2179–2184.
32 Kabsch W. Software XDS for image rotation, recognition and crystal symmetry
assignment. Acta Crystallogr, Sect D: Biol Crystallogr. 2010;66:125–132.
33 Evans PR, Murshudov GN. How good are my data and what is the resolution? Acta
Crystallogr, Sect D: Biol Crystallogr. 2013;69:1204–1214.
15 Totsika M, Vagenas D, Paxman JJ, et al. Inhibition of Diverse DsbA Enzymes in Multi-
DsbA Encoding Pathogens. Antioxid Redox Signal. 2018;29:653–666.
16 Bardwell JCA, McGovern K, Beckwith J. Identification of a protein required for
disulfide bond formation in vivo. Cell. 1991;67:581–589.
34 Bailey S. The CCP4 suite: programs for protein crystallography. Acta Crystallogr, Sect
D: Biol Crystallogr. 1994;D50:760–763.
17 Lamoree B, Hubbard RE. Current perspectives in fragment-based lead discovery
(FBLD). Essays Biochem. 2017;61:453–464.
35 McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni LC, Read RJ. Phaser
crystallographic software. J Appl Crystallogr. 2007;40:658–674.
18 Kirsch P, Hartman AM, Hirsch AKH, Empting M. Concepts and Core Principles of
Fragment-Based Drug Design. Molecules. 2019;24:4309–4330.
36 Guddat LW, Bardwell JCA, Glockshuber R, Huber-Wunderlich M, Zander T,
Martin JL. Structural analysis of three His32 mutants of DsbA: support for an
electrostatic role of His32 in DsbA stability. Protein Sci. 1997;6:1893–1900.
37 Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta
Crystallogr, Sect D: Biol Crystallogr. 2004;D60:2126–2132.
19 Duncan LF, Wang G, Ilyichova OV, Scanlon MJ, Heras B, Abbott BM. The fragment-
based development of a benzofuran hit as a new class of Escherichia coli DsbA
inhibitors. Molecules. 2019;24:3756–3782.
20 Mayer M, Meyer B. Characterization of Ligand Binding by Saturation Transfer
Difference NMR Spectroscopy. Angew Chem, Int Ed. 1999;38:1784–1788.
21 Carr HY, Purcell EM. Effects of Diffusion on Free Precession in Nuclear Magnetic
Resonance Experiments. Phys Rev. 1954;94:630–638.
38 Adams PD, Afonine PV, Bunkoczi G, et al. PHENIX: a comprehensive Python-based
system for macromolecular structure solution. Acta Crystallogr, Sect D: Biol
Crystallogr. 2010;66:213–221.
22 Dalvit C, Fogliatto G, Stewart A, Veronesi M, Stockman B. WaterLOGSY as a method
for primary NMR screening: Practical aspects and range of applicability. J Biomol
NMR. 2001;21:349–359.
39 Ziarek JJ, Peterson FC, Lytle BL, Volkman BF. Binding site identification and
structure determination of protein-ligand complexes by NMR: A semiautomated
approach. Methods Enzymol. 2011;493:241–275.
23 Hubbard RE, Murray JB. Chapter twenty - Experiences in Fragment-Based Lead
Discovery. In: Kuo LC, ed. Methods Enzymol. Academic Press; 2011:509–531.
24 Bentley MR, Ilyichova OV, Wang G, et al. Rapid Elaboration of Fragments into Leads
by X-ray Crystallographic Screening of Parallel Chemical Libraries (REFiLX). J Med
Chem. 2020;63:6863–6875.
40 Kurth F, Rimmer K, Premkumar L, et al. Comparative sequence, structure and redox
analyses of Klebsiella pneumoniae DsbA show that anti-virulence target DsbA
enzymes fall into distinct classes. PloS One. 2013;8(11):e80210.
41 Totsika M, Heras B, Wurpel DJ, Schembri MA. Characterization of Two Homologous
Disulfide Bond Systems Involved in Virulence Factor Biogenesis in Uropathogenic
Escherichia coli CFT073. J Bacteriol. 2009;191:3901.
25 Kadokura H, Tian H, Zander T, Bardwell JC, Beckwith J. Snapshots of DsbA in action:
detection of proteins in the process of oxidative folding. Science. 2004;303:534–537.
26 Paxman JJ, Borg NA, Horne J, et al. The structure of the bacterial oxidoreductase
enzyme DsbA in complex with a peptide reveals a basis for substrate specificity in the
catalytic cycle of DsbA enzymes. J Biol Chem. 2009;284:17835–17845.
42 Auld Douglas, Inglese James, Dahlin Jayme. In: Markossian S, Grossman A,
Brimacombe K, et al., eds. Assay Guidance Manual. Bethesda (MD): Eli Lilly &
Company and the National Center for Advancing Translational Sciences; 2004. https:
//www.ncbi.nlm.nih.gov/books/NBK442297/.
13