New inha inhibitors based on expanded triclosan and di-triclosan analogues to develop a new treatment for tuberculosis

Article abstract of DOI:10.3390/ph14040361

The emergence of multidrug-resistant (MDR) and extensively drug-resistant (XDR) tuberculosis (TB) has reinforced the need for the development of new anti-TB drugs. The first line drug isoniazid inhibits InhA. This is a prodrug requiring activation by the enzyme KatG. Mutations in KatG have largely contributed to clinical isoniazid resistance. We aimed to design new ‘direct’ InhA inhibitors that obviate the need for activation by KatG, circumventing pre-existing resistance. In silico molecular modelling was used as part of a rational structure-based drug-design approach involving inspection of protein crystal structures of InhA:inhibitor complexes, including the broad spectrum antibiotic triclosan (TCS). One crystal structure exhibited the unusual presence of two triclosan molecules within the Mycobacterium tuberculosis InhA binding site. This became the basis of a strategy for the synthesis of novel inhibitors. A series of new, flexible ligands were designed and synthesised, expanding on the triclosan structure. Low Minimum Inhibitory Concentrations (MICs) were obtained for benzylphenyl compounds (12, 43 and 44) and di-triclosan derivative (39), against Mycobacterium bovis BCG although these may also be inhibiting other enzymes. The ether linked di-triclosan derivative (38) displayed excellent in vitro isolated enzyme inhibition results comparable with triclosan, but at a higher MIC (125 μg mL^?1 ). These compounds offer good opportunities as leads for further optimisation.

Full text of DOI:10.3390/ph14040361

pharmaceuticals
Article
New InhA Inhibitors Based on Expanded Triclosan and
Di-Triclosan Analogues to Develop a New Treatment for
Tuberculosis
Sarentha Chetty 1,†, Tom Armstrong 1, Shalu Sharma Kharkwal 1,†, William C. Drewe 1 , Cristina I. De Matteis 2,
Dimitrios Evangelopoulos 3,† , Sanjib Bhakta 3 and Neil R. Thomas 1,
*
1
Biodiscovery Institute, School of Chemistry, University of Nottingham, University Park,
Nottingham NG7 2RD, UK; Sarentha.Chetty@wits.ac.za (S.C.); tom.armstrong@nottingham.ac.uk (T.A.);
shalu.sharma02@gmail.com (S.S.K.); Will.Drewe@lhasalimited.org (W.C.D.)
School of Pharmacy, University of Nottingham, University Park, Nottingham NG7 2RD, UK;
Pazcm@exmail.nottingham.ac.uk
Department of Biological Sciences, Birkbeck, Institute of Structural and Molecular Biology, University of
London, London WC1E 7HX, UK; d.evangelopoulos@ucl.ac.uk (D.E.); s.bhakta@bbk.ac.uk (S.B.)
Correspondence: neil.thomas@nottingham.ac.uk
2
3
*
†
Current Addresses: S.C. Pharmacology Division, Department of Pharmacy and Pharmacology, Faculty of
Health Sciences, University of the Witwatersrand, Johannesburg 1864, South Africa; D.E. Department of
Microbial Diseases, UCL Eastman Dental Institute, University College London, London NW3 2PF, UK;
S.S.K. Agenus, Lexington, MA 02421, USA.
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀꢁꢂꢃꢄꢅꢆ
Abstract: The emergence of multidrug-resistant (MDR) and extensively drug-resistant (XDR) tu-
berculosis (TB) has reinforced the need for the development of new anti-TB drugs. The first line
drug isoniazid inhibits InhA. This is a prodrug requiring activation by the enzyme KatG. Mutations
in KatG have largely contributed to clinical isoniazid resistance. We aimed to design new ‘direct’
InhA inhibitors that obviate the need for activation by KatG, circumventing pre-existing resistance.
In silico molecular modelling was used as part of a rational structure-based drug-design approach
involving inspection of protein crystal structures of InhA:inhibitor complexes, including the broad
spectrum antibiotic triclosan (TCS). One crystal structure exhibited the unusual presence of two
triclosan molecules within the Mycobacterium tuberculosis InhA binding site. This became the basis of
a strategy for the synthesis of novel inhibitors. A series of new, flexible ligands were designed and
synthesised, expanding on the triclosan structure. Low Minimum Inhibitory Concentrations (MICs)
Citation: Chetty, S.; Armstrong, T.;
Sharma Kharkwal, S.; Drewe, W.C.;
De Matteis, C.I.; Evangelopoulos, D.;
Bhakta, S.; Thomas, N.R. New InhA
Inhibitors Based on Expanded
Triclosan and Di-Triclosan Analogues
to Develop a New Treatment for
Tuberculosis. Pharmaceuticals 2021, 14,
361. https://doi.org/10.3390/
ph14040361
were obtained for benzylphenyl compounds (12, 43 and 44) and di-triclosan derivative (39), against
Academic Editor: Paweł Kafarski
Mycobacterium bovis BCG although these may also be inhibiting other enzymes. The ether linked
di-triclosan derivative (38) displayed excellent in vitro isolated enzyme inhibition results comparable
with triclosan, but at a higher MIC (125
leads for further optimisation.
Received: 11 March 2021
Accepted: 10 April 2021
Published: 14 April 2021
µ
g mL−1). These compounds offer good opportunities as
Keywords: tuberculosis; structure-based drug-design; molecular modelling; InhA; triazole; triclosan;
isoniazid; Mycobacterium tuberculosis; Mycobacterium bovis BCG
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
1. Introduction
Tuberculosis (TB) has been a major cause of morbidity and mortality for centuries. It
is one of the top 10 causes of death worldwide and is the number one killer amongst the
infectious diseases [1,2]. A deadly synergy exists between HIV and TB [2], and in recent
years, HIV has significantly contributed to the increase in the number of people infected
with and dying from TB [3].
Drug susceptible TB can be successfully treated using a three or four drug combination.
Current directly observed treatment (short course) (DOTs) consists of a two-month course
of first-line drugs: isoniazid (INH), rifampicin, pyrazinamide and/or ethambutol followed
Copyright:
© 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Pharmaceuticals 2021, 14, 361. https://doi.org/10.3390/ph14040361
https://www.mdpi.com/journal/pharmaceuticals

Pharmaceuticals 2021, 14, 361
2 of 20
by a four-month course of isoniazid and rifampicin [
4
,5]. Over the past five decades,
varying degrees of resistance to existing TB drugs have emerged. Multidrug-resistant
tuberculosis (MDR-TB), characterized as resistant to rifampicin and isoniazid is now
widespread and extensively drug resistant tuberculosis (XDR-TB) defined as MDR-TB
together with resistance to second-line drugs such as quinolones and aminoglycosides is
increasing [
1,
2]. It is estimated that one third of the world’s population is already infected
with the latent form of the disease, which carries a lifelong risk of activation [1,2
]. The
high infection rates and growing resistance means the hunt for new drugs has become a
global priority.
Mycolic acids are an essential component of the mycobacterial cell wall and are
synthesized by the fatty acid biosynthetic pathway. Fatty acid biosynthesis in Mycobacterium
tuberculosis (Mtb) occurs via two discrete pathways (FASI and FASII). In the ubiquitous
FASI pathway, fatty acids of lengths C16 and C24/26 are biosynthesized from acetyl CoA
and malonyl CoA. Elongation of these fatty acids into full length mycolic acids occurs in
the FASII cycle. The steps in the FASI pathway are carried out by one large, multifunctional
fatty acid synthase enzyme. This enzyme is also found in eukaryotes and advanced
prokaryotes. The enzymes in the FASII pathway, however, are encoded by discrete genes
which are not present in eukaryotes, thereby making the enzymes in this pathway clinically
validated drug targets [6–8].
Isoniazid was first synthesized in 1952 and is one of the most effective drugs in the
current TB arsenal [8,9]. It is a prodrug and its activation requires oxidation by the enzyme
KatG, resulting in an unstable acyl radical intermediate or an acyl anion [10–12]. This, in
turn, forms a covalent adduct with nicotinamide adenosine dinucleotide (NAD+). The
INH-NAD adduct then binds to the active site of the enoyl reductase (ENR) InhA, its main
target, and exerts its inhibitory action [13,14]. Prevention of fatty acyl substrate reduction
stops full length mycolate production, resulting in cell lysis [15,16].
The majority of INH-resistance is attributed to point mutations in the KatG enzyme
rather than InhA [
8
,12
,16]. The S315 mutation of KatG accounts for 50–95% of INH-
resistant clinical isolates [17
,18]. This mutation does not prevent the natural catalase
activity and mutant strains retain fitness and virulence [18]. INH resistance can also occur
through mutations in the InhA active site (S94A, I47T, and I21V) preventing INH-NAD
binding. Mutations in the InhA promoter region also result in overexpression of the protein
operon [19]. While these InhA mutations do account for some resistance, KatG mutations
are responsible for the vast majority of clinically observed INH-resistance [18].
The development of an inhibitor that acts directly on InhA without the requirement
for prior activation would bypass KatG-mediated resistance. Triclosan (TCS) is a broad-
spectrum antimicrobial that until recently has been found in several commercial products
including toothpaste and mouthwashes. The withdrawal of TCS was due to its potential to
disrupt endocrine function and in particular reproductive hormones [20]. These inherent
problems could be present in the development of new TCS derivatives which will need
to be evaluated for this activity. TCS has been shown to directly inhibit InhA at low
concentrations, with higher concentrations resulting in disruption of bacterial protein
synthesis [
lipophilic thereby lowering its oral bioavailability and thus limiting its potential as an
anti-tuberculosis drug [22 24]. TCS also has a lower potency in comparison to INH thereby
7,21,22]. The chlorine atoms on the TCS structure, however, make it highly
–
requiring significantly higher concentrations to elicit its effect [22].
SAR and Design of Compounds
The TCS:InhA crystal structure (PDB:1P45) reveals the key protein interactions [25].
The A-ring engages in л-stacking interactions with both the nicotinamide group on the
cofactor and Phe149. The hydroxyl group forms a H-bonding network with the 20-OH
group on the ribose group of the NADH cofactor and Tyr158. An additional H-bond is
made between the ether linkage and the ribose 20-hydroxyl group [26]. These essential
interactions are summarised in Figure 1b.

Pharmaceuticals 2021, 14, 361
3 of 20
Figure 1.
(
a
) Representation of TCS (red) bound to the active site of InhA. (Active site residues (green)
and NAD+(magenta) (derived from PDB.2B35). (
b) The constituents of the TCS structure important
for activity as identified by Sivaraman et al. [26] are highlighted. (c) Structure of inhibitor 8PP.
SAR studies by Sivaraman et al. revealed that for the E. coli enoyl reductase FabI, the
two Cl substituents on the B-ring are not essential for activity as their removal causes a
7-fold decrease in the Ki value from 7 pM for TCS to 1 pM for 5-chloro-2-phenoxyphenol
(CPP). This compound has also been characterized as a slow, tight-binding inhibitor with a
significant delay in the onset of inhibition, probably due to a large conformational change
in the FabI/NAD/TCS (or CPP) ternary complex. Both X-ray crystal and docked structures
indicate that the chlorine at the ortho position on the B ring of TCS is in direct contact
with the NAD+ phosphates, whilst simulations of the FabI/NAD+/CPP ternary complex
indicate a significant change in the position of the NAD+ [26]. The A-ring Cl group is
important for van der Waals interactions as well as decreasing the pKa of the phenol in
ring A from 9.12 to 7.8. Its removal therefore significantly affects the binding affinity to
FabI [26,27]. The binding of CPP to InhA has not been reported, but Sullivan et al. demon-
strated that potent inhibition of InhA could be achieved through replacement of the A-ring
Cl atom with aliphatic chains of varying lengths, whilst only having an unsubstituted
phenyl as the B-ring. The best analogue from the diphenyl ether series of compounds
featured an octyl chain (8PP) (Figure 1) [28]. The chain was shown to engage the substrate
binding loop, matching the arrangement seen by Rozwarski et al. with their C16 substrate
mimic [28,29]. In the quest for new inhibitors, other research groups have since explored dif-
ferent modifications of the core TCS structure [23,25,26,28,30–47]. Rodriguez et al. explored
combining two TCS molecules and synthesised a novel macrocyclic InhA inhibitor [48]. A
recent review by Vosátka et al. provides a comprehensive summary of efforts to date [22].
More recently groups have employed computational techniques including virtual screen-
ing, density functional theory (DFT) calculations and QSAR to identify potential InhA
inhibitors [23,49].
Interestingly, the X-ray crystallography work (PDB:1P45) conducted by Kuo et al. [25
]
revealed the presence of two TCS molecules within the InhA active site. This was unique
to Mycobacterial InhA and not found in the crystal structures of enoyl reductases from any
other organisms solved to date. This occurrence was presumably due to InhA having a
longer substrate binding loop to accommodate a larger substrate. One molecule occupies
the same space and is in the same orientation as that of all the other TCS-ENR complexes.
The second TCS molecule lies adjacent to the first, with its A- and B-rings aligned with
those of the first TCS, engaging the substrate binding loop and promoting loop ordering
(Figure 2) [25].

Pharmaceuticals 2021, 14, 361
4 of 20
a
b
Figure 2.
(
a
) A representation of the interaction of a C16 fatty acyl substrate (yellow) with the InhA
active site residues (green) and the cofactor NAD+ (magenta), H-bonds (blue) (from PDB.1BVR) [29].
(b) A view of the key interactions of two molecules of TCS within the active site. TCS molecules
(red), the active site residues (green) and NAD+ (magenta) (PDB.1P45). The two TCS molecules are
separated by a distance of 4.6 Å between the two carbons attached to the B-ring para-chlorine atoms.
We reasoned that a di-triclosan molecule would be able to preferentially fit in the
InhA active site and have reduced affinity for the enoyl reductases of other organisms,
thereby having the potential to be selective for InhA. This was the basis of our drug design
approach which aimed to bridge the distance of 4.6 Å between the 40-carbon atoms of the
two TCS molecules by a symmetrical and flexible ether linkage. We hypothesised that
either a longer more flexible chain or a well-chosen rigid linker (see also reference [47] for a
series of rotational restricted compounds based on the 1,5-triazole group) may allow the
enlarged molecule to adopt a conformation similar to that of the fatty acylCoA substrate or
the two TCS molecules within the loop. By retaining the most essential features of TCS that
give it antibacterial activity, we generated three scaffolds (Figure 3).
Cl
X
A
or
A =
R'(B)
OH
B
=
R
O
Y
N
N
R'
N
X= Cl, F, Br
R= OMe, OH
Y= O, NH
Cl
Cl
TCS
Figure 3. Overall target summary.

Pharmaceuticals 2021, 14, 361
5 of 20
The Cl atom on the A-ring was either retained or substituted for by another halogen
(F, Br). The hydroxyl substituent on the A-ring was replaced with a methoxy group. It was
envisaged that the hydroxylated derivatives would also be tested but attempts to selectively
demethylate this were unsuccessful. We retained the inclusion of the methoxy compounds
as this functional group could still act as a H-bond acceptor with the co-factor and the active
site residues. The B-ring was either retained as an aryl group or replaced by a heterocyclic
ring (Figure 3). For the benzyl phenyl ether and aniline derivatives, the B-ring was kept
similar to TCS, but both the chloride atoms were removed, and the para-Cl substituent was
replaced with a methoxy substituent. The ether linkage was maintained or replaced with
an amine linkage and further extended by one carbon unit. The reason for choosing these
linkers was their ease of synthesis by a nucleophilic substitution reaction and to provide
rotational flexibility which would allow the molecules to explore more of the binding site
occupied by the second TCS and form stronger interactions with the flexible binding loop.
The hypothesis was that the amine linkage should have the same stabilizing activity to that
of the ether linkage in the TCS molecule in the TCS: InhA complex. The replacement of
the ether group for an amine group could also affect the H-bonding as a secondary amine
can function as both a H-bond donor and acceptor. For the triazole linked derivatives,
the B-ring was replaced with a 1,4-triazole ring. The basis of the different scaffold was to
provide opportunity to easily extend the scaffold further by one or two rings. It was hoped
that a larger molecule would mimic the second TCS and form more interactions within
the loop, and cause loop ordering. The triazole is also a bio-isostere of benzene and could
adopt a similar orientation to that of the B-ring in TCS. It was anticipated that one of the
triazole nitrogens could also form an additional H-bond with the cofactor. It was envisaged
that the additional aromatic ring would increase the interactions with the cofactor and
other hydrophobic residues in this region.
We chose to incorporate different halogens as their different properties with regards
to electronegativity and size determine their impact on the structure-activity relationship
(SAR), binding to the receptor and cellular uptake. Sivaraman et al. demonstrated that
even subtle changes in the steric and electronic properties of the ring can have a major
effect on the binding potential [26,27].
A series of benzylphenyl aniline, triazole linked biaryl anilines and di-triclosan ana-
logues were developed to find suitable new lead compounds and optimise them further
following biological and SAR analyses. The long-term goal is to obtain inhibitors with
good selectivity for InhA whilst being mindful of the requirements for compounds to be
orally active and be prepared at low cost given the funding constraints of the healthcare
systems of many of those affected by TB.
2. Results and Discussion
2.1. Drug-Like Properties
Assessments of the compounds from the three series of compounds for drug-like prop-
erties (Table 1) revealed all comply with Lipinski’s rule of five except compound 31 which
is non-compliant with regards to both MW and ClogP [50] and 38 with regards to ClogP.
These compounds were found to be soluble at 100 and 50 µM concentrations respectively,
as used in the screening procedure. Calculations of topological polar surface area (TPSA)
using a web-based molecular descriptors calculator (http://www.molinspiration.com,
accessed on 25 March 2019) revealed that all compounds designed had a TPSA < 140 Å
which is indicative of a reasonable potential to penetrate membranes.
2.2. Docking Studies
Docking of the designed analogues into the active site gave a prediction of the binding
mode. Overall, all molecules fitted well into the active site with the di-triclosan analogues
having the best fitness scores followed by the triazole linked analogues, and then the benzyl
phenyl ether or aniline analogues. This was not surprising as the di-triclosan molecules,
having larger surface areas, made more favourable van der Waals and hydrophobic in-

Pharmaceuticals 2021, 14, 361
6 of 20
teractions with the residues and cofactor within the active site and hence gave higher
scores. A summary of the docking scores for the individual compound series together
with interactions with active site residues is provided in the Supplementary Information
(
Tables S1–S4; Figures S1–S4). The docking results suggest that the halogen in the para
position on ring A, could have a significant effect on binding. Fitness scores were very
similar for the benzyl phenyl and triazole ether-linked diaryl series with the trend being
Br > Cl > F. For the benzylphenyl aniline and triazole-linked diaryl aniline analogues it
was observed that Cl > F.
2.3. Synthesis
The three scaffolds for the target molecules were synthesized according to the general
routes as follows: The benzyl phenyl analogues were prepared via a nucleophilic substitu-
tion reaction, the triazole-linked diaryl derivatives via a copper (I) catalysed alkyne-azide
cycloaddition (CuAAC) ‘click’ reaction and the di-triclosan target molecules via a Buchwald
coupling reaction to generate the methylated analogues.
The synthesis of the benzylphenyl ether compounds 1, 2, 3 with the ether linkage is
described in Scheme 1. Moderate yields from 23–47% were achieved.
X
X
X = Cl
F
1
2
3
O
a
Br
O
O
OH
O
Scheme 1. Synthesis of benzylphenylether derivatives. Reagents and conditions: (
DMF, 0 C to r.t., 4–6 h, X = Cl (29%), F (47%), Br (23%).
a) PMBCl, NaH,
◦
The benzylated aniline compounds were synthesized as detailed in Schemes 2 and 3
.
The commercially available starting materials for the chlorinated and fluorinated deriva-
tives were different; requiring slightly different synthetic routes. The 2-methoxy-4-
chloroaniline was generated from an amine starting material
protected with a Boc group. Methylation of the free hydroxyl group followed by removal
of the Boc group generated the desired 2-methoxy-4-chloroaniline ( , Scheme 2). Nucle-
4. The free amine was initially
7
ophilic substitution with p-methoxybenzyl chloride (PMBCl) gave the final compound
8 in
12% yield.
Cl
Cl
Cl
Cl
Cl
O
a
c
b
d
HN
O
OH
O
OH
NH2
NHBoc
NH2
NHBoc
4
7
5
6
O
8
Scheme 2. Synthesis of the chlorinated N-benzylphenyl aniline derivative
8
. Reagents and conditions:
◦
◦
(a
) Boc2O, NaHCO3, H2O, 0 C to r.t., 56%; (
b
) MeI, K2CO3, acetone, 60 C, 6.5 h, 82%; (
c) TFA, DCM,
◦
◦
0 C to r.t., 2 h, 85%; (d) PMBCl, NEt3, DCM, 0 C to r.t., 5.5 h, 12%.

Pharmaceuticals 2021, 14, 361
7 of 20
The fluorinated derivative 12, was prepared from 4-fluoro-6-nitrophenol (
9) as de-
scribed in Scheme 3. The phenol group was first converted to its methyl ether followed
by a reduction of the nitro group to generate amine 11. Amine 11 was then subjected
to nucleophilic substitution to generate target molecule 12 in 14% yield. In addition, a
nucleophilic substitution reaction with 3 equivalents of PMBCl yielded an N, N-substituted
amine 13 in 36% yield. The synthesis of this compound has been reported previously [51].
F
F
F
F
O
c
a
b
HN
O
OH
O
NO2
NO2
NH2
9
10
11
O
d
F
12
O
O
N
O
13
Scheme 3. Synthesis of the fluorinated N-benzylphenylaniline derivatives 12 and 13. Reagents and
conditions: (
a
) MeI, K2CO3, 67–69 ◦C, 6 h; 81%; (
b
) NH4HCO2, 10% Pd/C, MeOH, r.t.; overnight,
◦
◦
100%; (
c) PMBCl, K2CO3, DMF, 55 C, 4 h, 14%; (
d
) PMBCl (3 equiv.), K2CO3, DMF, 55 C, 4 h, 36%.
The next set of target compounds (triazole-linked diaryl target molecules) was synthe-
sised using a CuAAC ‘click’ chemistry strategy, as shown in Scheme 4. The triazole-linked
target molecules were synthesised by subjecting the respective aniline or phenol to a nu-
cleophilic substitution with propargyl bromide to generate the terminal alkynes: 14–18.
The azide 19 was formed by diazotisation of the commercially available aniline, followed
by the CuAAC to generate the target molecules 20 (74%), 21 (77%), 22 (83%), 23 (84%), 24
(77%), respectively, in excellent yields.
For the di-triclosan derivatives, the synthesis (Scheme 5) started with an initial Buch-
wald coupling followed by a nucleophilic substitution reaction to join symmetrical halves
to generate di-triclosan analogue 31 in a 79% yield. The same strategy as outlined in
Scheme 5 was applied to obtain unsymmetrical compound 39 in a 67% yield (Scheme 6).

Pharmaceuticals 2021, 14, 361
8 of 20
Scheme 4. Synthesis of the triazole-linked diaryl compounds via ‘click’ chemistry. Reagents and
conditions: (
a
) Propargyl bromide, K2CO3, DMF, 55 ◦C to r.t., 4 h◦, Y = O, X = Cl (75%), F (75%), Br
(71%), Y = NH, X = Cl (7%), F (30%) ; (b) H2SO4, NaNO2, H2O, 0 C to r.t., 0.5 h followed by NaN3,
H2O, r.t., 3 h, 82%; (c
) tBuOH, sodium ascorbate, CuSO4, H2O, r.t., overnight, Y = O, X = Cl (74%), F
(77%), Br (83%), Y = NH, X = Cl (84%), F (77%).
Scheme 5. Synthesis of the di-triclosan dimethyl ether 31. Reagents and conditions: (a) BH3, THF, r.t.,
overnight, 91%; (
toluene, r.t., 3 days, 41%; (
(f) NaH, THF, r.t. to 78 C, 18 h, 79%.
b
) TBDMSCl, imidazole, DCM, r.t, overnight, 94% (
c
) CuI, Cs2CO3, naphthoic acid,
◦
◦
d
) TBAF, THF, 0 C to r.t., 4.5 h, 79%; (
e) MsCl, Et3N, 0 C to r.t., 4 h, 65%;
◦

Pharmaceuticals 2021, 14, 361
9 of 20
Attempts to demethylate derivative 31 were unsuccessful, therefore another approach
was adopted. The hydroxylated symmetrical di-triclosan derivative 38 followed a similar
synthetic route (Scheme 7), but had different starting materials, demethylation and repro-
tection of the resulting alcohol with a MOM protecting group followed by reduction of the
aldehyde to generate the alcohol 35. A nucleophilic reaction followed by deprotection gen-
erated the final hydroxylated di-triclosan derivative 38 in a 44% yield. The three di-triclosan
derivatives were thus successfully synthesized in moderate yields.
Cl
O
O
O
O
HO
O
O
O
a
Cl
Cl
39
30
◦
Scheme 6. Synthesis of di-triclosan derivative 39. Reagents and conditions: (a) NaH, DMF, r.t. to 0 C
to r.t., 4.5 h, 67%.
Scheme 7. Synthesis of di-triclosan derivative 38. Reagents and conditions: (a
) K2CO3, DMF, 130 ◦C,
18 h, 89%,◦(
b
) AcOH, 47% HBraq, 110 ◦C, 18 h, 29%, (
c
) DIPEA, MOMCl, r.t., 18 h, 92%, (
d
) NaBH4,
◦
◦
MeOH, 0 C to r.t., 4 h, 86%, (
e) Et3N, MsCl, 0 C to rt, 4 h, 56%, (
f
) NaH, DMF, 0 C, 1 h, 45%,
◦
(g) 6 M HCl, MeOH, 70 C 2 h, 44%.
The dibenzylether 40 was obtained in two steps from iodobenzoic acid by reduction
with BH3 followed by an SN1 reaction with 4-iodobenzyl bromide to give 40 in 71% yield.
This intermediate was produced en route to finding a suitable method to generate the
di-triclosan derivatives. We thought it might be useful to test against Mtb. This compound
has been previously synthesized for other purposes [52]. Ketone 41 was subjected to a

Pharmaceuticals 2021, 14, 361
10 of 20
nucleophilic substitution reaction followed by a Baeyer-Villiger oxidation which converted
the ketone 43 to an ester 44 (Scheme 8).
Cl
O
Cl
O
O
Cl
O
Cl
a
b
O
O
+
O
S
O
OH
O
42
41
O
O
O
O
S
S
O
O
44
43
Scheme 8. Synthesis of the benzylphenyl compounds via a Baeyer-Villiger oxidation. Reagents and
◦
conditions: (a) K2CO3, DMF, 55 C to r.t., 4 h, 51%; (b) mCPBA, CHCl3, r.t., overnight, 68%.
2.4. Inhibition Studies with Purified InhA
Initial screening was carried out on all the TCS analogues and some of the inter-
mediates at 100
µM, with the exception of compounds 1, 3, 20 and 38 (50 µM) and 22
(25 M) due to their limited solubility (Table 1 with structures in Figure 4). The percentage
µ
inhibition results revealed that those compounds with good inhibition (>50% against InhA)
were compounds 8, 21, and 38, moderate inhibition (20–50%) was observed with 3, 13, 20,
23, 24, 32, 39, 40, 43 and poor inhibition (<20%) with 2 and 44. Compounds 1, 22 and 12
were inactive.
Table 1. Summary of the physical properties of the compounds reported in this paper and their percentage inhibition
of InhA, MIC, GOLD fitness scores and drug like properties for the benzyl phenyl analogues, triazole linked derivatives
and di-triclosan analogues as well as intermediate compounds. Molecular weight (MW) Total polar surface area (TPSA),
Hydrogen bond donors (HBD) and Hydrogen bond acceptors (HBA) are also reported.
%
MIC
(µg mL
Cpd
MW
ClogP
TPSA
HBD
HBA
Fitness Scores
−1
)
Inhibition
TCS
INH
8PP
1
2
3
289.54
137.14
298.42
278.73
262.28
323.18
277.75
261.29
381.44
345.78
329.32
390.23
344.80
328.34
511.39
483.34
384.85
450.05
354.80
370.81
5.53
29.46
68.01
29.46
27.70
27.70
27.70
30.50
30.50
21.71
58.42
58.42
58.42
61.21
61.21
46.17
68.16
27.70
9.23
1
2
1
0
0
0
1
1
0
0
0
0
1
1
0
2
0
0
0
0
2
4
2
3
4
3
3
4
5
6
7
6
6
7
5
5
4
1
5
6
-
92
N/A
-
NI
18
25
63
NI
43
36
56
NI
45
23
42
100
27
30
35
14
20 [48]
0.05
1.9 [28]
>50
50
>50
>50
7.8
>125
>50
>50
>50
>50
>125
>125
125
−0.67
N/A
-
7.53
4.44 ± 0.27
3.70 ± 0.62
4.68 ± 0.52
4.27 ± 0.56
3.76 ± 0.60
5.85 ± 0.80
4.03 ± 0.82
3.29 ± 1.05
4.26 ± 0.94
3.86 ± 0.96
3.34 ± 1.03
8.21 ± 0.50
7.88 ± 0.54
5.40 ± 0.48
6.01 ± 0.45
3.50 ± 0.28
2.92 ± 0.26
57.22
53.86
58.66
60.28
55.62
69.25
75.59
71.55
75.80
76.22
72.63
90.79
86.50
72.92
59.03
68.54
62.44
8
12
13
20
21
22
23
24
31
38
39
40
43
44
7.8
>50
7.8
69.68
78.92
15.6
All the compounds were tested at concentrations of 100
(No Inhibition).
µ
M, with the exception of
1
,
3,
20 and 39 and TCS (50 µM) and 22 (25 µM). NI

Pharmaceuticals 2021, 14, 361
11 of 20
Figure 4. Compounds that were evaluated for biological activity together with compounds 45
and 46 from our previous study on a related series of triazole-linked di-triclosan analogues (47) and
compound 8PP that is one of the most potent triclosan derivatives reported to date (IC50 5.0 ± 0.3 nM
;
(M. tuberculosis H37RV)) [28].
2.5. Minimum Inhibitory Concentration (MIC)
The triclosan analogues showed moderate to good inhibition of mycobacterial growth.
The MIC data suggested that 12, 39, 43 and 44 has good inhibition properties, but in vitro
assays revealed that 39 and 43 had only moderate inhibition of InhA, with compound 12
not inhibiting the enzyme. Likewise, 44 had reasonable inhibition properties against whole
cells, but the in vitro assays revealed that this compound had poor inhibition properties
against InhA. There is a possibility that these compounds inhibit other enzymes in the micro-
organism given their MIC values. At high concentrations TCS has been reported to act as a
mitochondrial uncoupling agent, cell membrane disruptor and to cause perturbations to
lipid and protein biosynthesis.
2.6. Correlation of the Docking and Isolated Enzyme Results
The docking data for the hydroxylated di-triclosan derivative 38 correlated well with
the inhibition data. Compound 38 displayed 100% inhibition at just 50 µM and the docked
structure predicts a hydrogen bond as well as a л-л stacking with the cofactor. An additional
л-л stacking interaction was predicted to occur with Phe 149 as well as favourable van der
Waals interactions with residues Leu 218 and Ile 215 (Figure 5).

Pharmaceuticals 2021, 14, 361
12 of 20
Figure 5. A view of the docking of 38 into the InhA active site. The active site residues (blue), NAD+
(magenta) and ligand (green).
2.7. Structure Activity Relationships for the Compounds Evaluated
The mycobacterial cell envelope is known to be a significant barrier to drugs therefore
it is interesting to observe that compounds given in Table 1 with a ClogP range of
−
0.67 to
7.88, all have some activity in the whole cell assay. The largest compound tested here is
still considerably smaller than rifampicin (MW 822.95; 6 HBD; 16 HBA; CLogP 3.71; TPSA
220.15) which is used in the clinical treatment of tuberculosis and is active against the latent
form of the disease.
The compounds that have methyl ethers on the A- and B-rings (1–3, 8, 12, 20–24, 31)
all show poor activity against whole M. bovis, even when active against InhA directly.
Their level of activity against the isolated InhA is lower than those triclosan analogues
that have a free hydroxyl group on the ‘A’-ring such as 38 and 8PP. This may be due to
the methoxy group on the ‘A-ring’ being unable to behave as an efficient hydrogen bond
acceptor with Tyr158 in the same way that the hydroxyl group on one of the triclosan
molecules. Compound 39 has the methoxy ether and is only a moderate inhibitor of InhA
but has one of the lowest MIC values of the compounds tested suggesting it may have other
targets in mycobacteria. Extending this compound to give the full ether linked di-triclosan
structure 38 whilst improving the activity against the isolated enzyme when compared
with its dimethyl ether 31 is detrimental to its activity against whole cells. The compounds
in which the B-ring of triclosan is replaced by a triazole group, whilst exhibiting moderate
activity against the isolated enzyme, have poor activity against the whole cells suggesting
that the introduction of this group is impacting on their uptake. The removal of the methyl
ethers from compound 31 significantly improves its activity against both InhA and M. bovis.
A number of points that can be gleaned from examining the structure activity rela-
tionships for closely related compounds. Whilst compound 12 exhibits a low MIC value
(
7.8 µg mL−1), it does not inhibit the isolated InhA. Replacing the chlorine with a fluorine
gives compound . This change increases the MIC to >50 mM, but does generate a com-
pound that inhibits InhA (63% at 40 M). Given the smaller size of the fluorine, the change
8
µ
in activity may be due to the increase in basicity of the amine in this compound compared
with the chlorine rather than any steric effects as compound 12 is predicted to bind tighter
to InhA than 8.
Compounds 43 and 44 differ in the nature of the substituent ortho to the phenolic ether,
with the mesomerically electron donating ester substituent which would increase the HBA
ability of the phenolic ether giving slightly better binding to InhA, but a higher MIC, whilst
these properties are reversed with the electron withdrawing keto-substituent, suggesting
that groups that retaining an electron withdrawing group in this position is preferred.

Pharmaceuticals 2021, 14, 361
13 of 20
3. Materials and Methods
3.1. Molecular Docking and Design studies
The 2H7I.PDB [33] crystal structure (resolution 1.62 Å), containing InhA in complex
with its respective ligand (1-cyclohexyl-5-oxo-N-phenylpyrrolidine-3-carboxamide) was
retrieved from the Protein Data Bank (PDB). Prior to docking, the protein structure was
prepared using the SYBYL X [53] structure preparation tool. The Amber force field FF99
was used to assign atom types. Gasteiger-Hückel charges were assigned to the ligand and
cofactor. The cofactor was kept rigid during protein minimization. Protein minimization
was done in three consecutive steps. The first step was hydrogen atom minimization
keeping the NAD+ and heavy atoms rigid, followed by side-chain minimization keeping
the NAD+ and backbone rigid. The last step was the whole protein minimization keeping
the NAD+ rigid. Minimization was carried out with a maximum of 1000 iterations, gradient
of 0.05 kcalmol−1, no initial optimization and using the Tripos force field.
Molecular docking studies were performed using Genetic Optimization for Ligand
Docking (GOLD) version 3.1. (CCDC, Cambridge, UK) [54,55]. Crystallographic waters
were removed from the minimized complex structure and the ligand was extracted from
the active site. The protein (with cofactor) was loaded with the corresponding ligand as
a reference point into GOLD. The active site was defined by a 10 Å radius around the
reference ligand.
The designed molecules were sketched in SYBYL X [53], assigned Gasteiger Hückel
charges and minimized with a maximum of 5000 iterations, no initial optimization and
using the Tripos forcefield. Docking was carried out using the default speed in GOLD. The
binding site was defined by x (43.4601), y (51.7939), z ( 82.8350) coordinates established in
−
SYBYL using the reference ligand from pdb.2H7I. Fitness scores were determined using
GOLDSCORE. For simplicity, the top scoring pose of each designed molecule was assessed.
ACD Labs [50] version 12 was used to calculate the ClogP of all the designed com-
pounds and TPSA was calculated using the Molinspiration Cheminformatics server [56].
3.2. Synthesis
3.2.1. Materials and General Methods
All reagents were supplied by either Sigma-Aldrich (part of Merck Life Science UK
Limited, Gillingham, Dorset, UK), Alfa Aesar (Avocado, Lancashire, UK), or Fisher Sci-
entific UK Ltd. (Loughborough, UK). Flash chromatography was performed on silica gel
(35–70 µm, 60 Å (Fluorochem, Glossop, UK)). Elution was achieved under forced flow with
the indicated reagent grade solvent mixtures. These are given in v/v ratios. All fractions
containing the desired product were pooled together, with solvent removal under reduced
pressure and dried under high vacuum to give the product. Melting points were obtained
using a SMP3 melting point apparatus (Stuart Scientific, Stone, UK) and are uncorrected.
Infrared spectra were recorded on the Tensor 27 FT-IR instrument (Bruker, Coventry, UK)
using diluted solutions in spectroscopic grade chloroform or within a KBr disc or in solid
1
state using a IR 200 FT-IR instrument (Nicolet, ThermoFisher, Loughborough, UK) H-
NMR spectra and 13C-NMR spectra were recorded on an AV400 spectrometer (Bruker,
Coventry, UK) at 400 MHz and 100 MHz, respectively, and were carried out at ambient
temperature and measured using CDCl3 (δH = 7.26), D2O (δH = 4.87), d4-MeOH (δH = 3.31
or d6-DMSO (δH = 2.50) as solvent internal reference. H-NMR spectra are reported as
)
1
follows: chemical shifts in ppm on a
δ scale; multiplicity (app. = apparent, b = broad,
s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet); coupling constant J reported
in Hz and quoted to the nearest 0.1 Hz. 13C-NMR spectra were recorded on a Bruker AV400
spectrometer at 100 MHz, at ambient temperature and were measured using d3-CDCl3
(
δ = 77.0), d -MeOH (δH = 49.0), or d6-DMSO (δ = 39.5) as a solvent internal reference
c c
4
and proton decoupled. 13C NMR spectra are reported in ppm on the
δ
scale. 19F NMR
spectra were recorded on the Bruker AV400 spectrometer at 376.5 MHz. High performance
liquid chromatography (HPLC) was performed using Agilent 1200 and Agilent 1100 series
systems (Agilent, Santa Clara, CA, USA). The columns used were the Agilent Eclipse XDB-

Pharmaceuticals 2021, 14, 361
14 of 20
C18 column (150 × 4.6 mm
,
5 µm) or an Agilent Eclipse XDB-C18 column (150 × 9.4 mm
,
5 µm). Mass spectra were recorded on a Micromass LCT spectrometer (Micromass Ltd,
Manchester, UK) using electrospray ionization (ESI) at 3000 eV, Micromass Autospec (EI),
or a Bruker MicroTOF (ESI). Elemental analysis was recorded on a CE-440 Elemental Ana-
lyzer (Exeter Analytical, Coventry, UK). Analytical information on individual synthesized
compounds is given in the Supplementary Information.
3.2.2. General Procedure for Library Synthesis
The General Procedures (A-E) listed below were used as indicated in the Supplemen-
tary Information Experimental Section to prepare the compounds evaluated in this paper.
•
General Procedure for Synthesis of the Benzylphenyl ether Compounds (A)
The general method was carried from an adapted literature procedure [57]. To a
suspension of NaH (60% dispersion in mineral oil) (1.5 eq) in DMF (1.6 mL/1.0 mmol)
◦
cooled to 0 C and flushed with nitrogen, a solution of the alcohol (1 eq) in DMF was added
slowly via a syringe. The suspension was stirred at room temperature for 20 min and then
cooled to 0 ◦C, followed by the dropwise addition of p-methoxybenzyl chloride (PMBCl)
(3 eq) via a syringe. The pale-yellow solution was stirred for 4–6 h at room temperature,
followed by neutralisation with saturated NH4Cl(aq) solution. The solvent was removed by
evaporation and the residue was dissolved in DCM. The organic layer was washed with
water (2
×
50 mL) and brine (2
×
50 mL) and dried over MgSO4, before the solvent was
removed in vacuo to give the crude product.
•
General Procedure for the Synthesis of the Propargyl Analogues (B)
The general method was carried out from an adapted literature procedure [58]. Under
an inert atmosphere of nitrogen, the required aniline/alcohol (1 eq) was dissolved in
anhydrous DMF (0.5 mmol/1 mL), followed by the addition of K2CO3 (2 eq), before
◦
heating to between 53–55 C for 30 min. The reaction mixture was allowed to cool to
room temperature and p-methoxybenzyl alcohol/propargyl bromide (1 eq) was added.
The solution was stirred at room temperature for 4–6 h, monitoring for completion using
TLC and then poured into ice-cold water. The aqueous layer was extracted with EtOAc
(
3 × 30 mL) and the combined organic phases were dried over MgSO4 before removal of
the solvent in vacuo.
•
General Procedure for the Synthesis of the Azide (C)
This was adapted from a literature procedure [59]. To a suspension of p-anisidine/4-
◦
aminophenol (1 eq) in 4 M H2SO4(aq) (30 mL) at 0 C, a solution of NaNO2 (1.5 eq) in water
(0.5 mL/mol), was added. The cloudy suspension was left stirring at 0 ◦C for 15 min open
to air till all the solids had disappeared. A solution of NaN3 (1.5 eq) in water (0.5 mL/mol)
was added very slowly with gas evolution. The brown solution was left stirring at room
temperature for 3 h, followed by extraction with Et2O (3
×
50 mL). The combined organic
phases were dried over MgSO4, before removal of the solvent in vacuo to give crude brown
oil. Note: Azides should be handled with CAUTION as potentially EXPLOSIVE. Store
below room temperature (18 ◦C) and protect from light.
•
General Procedure for the Synthesis of the Triazole-Linked Biaryl Anilines and Ethers
via Click Chemistry (D)
This was adapted from a literature procedure [60]. The propargyl analogue (1 eq),
sodium ascorbate (0.5 eq) and CuSO4 (0.05 eq) were added to a solution of azide (1.5 eq) in
tBuOH (1.0 mol/5 mL) and water (1 mol/5 mL). The cream suspension was stirred at room
temperature overnight. The brown slurry was diluted with water (10 mL) and extracted
with EtOAc (3
solvent was removed in vacuo.
×
10 mL). The combined organic layers were dried over MgSO4 before the
•
Purification by Precipitation (E)

Pharmaceuticals 2021, 14, 361
15 of 20
Flash column chromatography was inadequate to obtain pure benzylphenyl ether and
benzylphenyl aniline analogues; therefore, an additional purification (cold precipitation)
was carried out. The required compound for purification was dissolved in the minimum
amount of EtOAc required for complete dissolution, followed by the dropwise addition of
hexane until precipitation took place. The precipitate was filtered and washed with the
minimum amount of cold hexane to remove any soluble impurities, followed by drying of
the residual precipitate under high vacuum to give the desired product.
3.3. InhA Enzyme Assay
The InhA:pET15a construct was kindly provided by Prof. Peter Tonge [61] (Stony
Brook University, Stony Brook, NY, USA). This construct had the InhA gene attached
downstream to the 6-His coding sequence. The protein was induced with 1 mM IPTG in
a Luria Bertani (LB) broth (containing ampicillin 100 µg/mL), over-expressed in E. coli
BL21-DE3 cells and grown at 30 ◦C for 6h until the OD600 reached 0.6. This was followed
by extraction from the cell pellet and purification using a Ni-affinity His-Trap 5 mL column
(GE Healthcare, Chicago, IL, USA). The column was washed with 150 mL binding buffer
[Tris HCl (20 mM), NaCl (300 nM), imidazole (500 mM), pH 7.5] and the protein was eluted
with elution buffer [Tris HCl (20 mM, NaCl, (300 nM), imidazole (500 mM), pH 7.5] in a
0–100% gradient of the elution buffer. The protein was analysed with SDS-PAGE and then
exchanged to a storage buffer (PIPES, 30 mM, pH 6.8) and concentrated to 10 mg/mL using
a 10 kDa Amicon centrifugal filter (Millipore, Watford, UK). Aliquots were then mixed
with 10% glycerol and stored at
−
80◦C. The molecular weight (29.806 Da) and pI (6.10)
and extinction coefficient was determined using the ExPasy translate tool. The extinction
co-efficient 37.5 mM−1cm−1 at λ280 was used in calculation of the protein concentration.
In vitro enzyme inhibition studies were carried out where the compounds were
screened at a concentration of 100 µM using the thermostated UV-Vis. spectrophotometer
(Varian Cary Bio 100, Agilent, Santa Clara, CA, USA) at 340 nm as described above. To
overcome the problem of aqueous solubility of some of the compounds, the compound
concentration was reduced to 50 or 25 µM.
The assay reaction conditions were as follows for a 400
PIPES (30 mM), Inhibitor (varying concentrations), OCoA (400
NADH (100 M) and InhA (150 nM). All compounds were dissolved in DMSO. DMSO was
µL volume: water (to volume),
µM) (see SI for synthesis),
µ
found to cause inhibition of the enzyme and therefore its concentration in the final assay
reaction was 1% v/v. The vehicle control consisted of water (to volume), PIPES (30 mM),
DMSO (1%), OCoA (400 µM), NADH (100 µM) and InhA (150 nM).
To obtain the percentage of enzyme inhibition the initial velocity (v) for the first minute
was established and this was compared to the velocity of the blank (v0). The initial velocity
was calculated from the gradient of the absorbance vs time plots for each of the inhibitors
and blanks. Inhibitory activities of the compounds were calculated as a percentage using
the formula (1 − v/v0) × 100%, where v/v0 is the residual activity of the enzyme.
3.4. Minimum Inhibitory Concentration Determination
The MIC was calculated for some of the compounds. This was achieved using a Spot
Culture Growth assay with M. bovis BCG, the vaccine strain for TB and a good model
for drug screening [62
of Middlebrook 7H10 (BD Diagnostics, Franklin Lakes, NJ, USA) and placed in six-well
microplate. 5 L of the test compounds in a solution of DMSO at a concentration of 100
and 50 M was added to the microplate. INH was used as a reference standard and was
,63]. Glycerol (0.2%) and 10% OADC was added to a 5 mL aliquot
µ
µM
µ
tested in the following concentrations 1, 0.5, 0.2, 0.1, 0.05, and 0 g/L. 5 µL of a mid-log
phase (106 cfu/mL) M. bovis BCG culture was added to each well. Results were obtained
following an incubation period of two weeks at 37 ◦C. The MIC was determined as the
lower concentration where there was no visually growth in the well.

Pharmaceuticals 2021, 14, 361
16 of 20
3.5. MIC Determination for Di-Triclosan Derivative 39
The measurement of the MIC99 was performed in 96-well flat bottom, black polystyrene
microtiter plates (Greiner Bio-One, Kremsmünster, Austria) in a final volume of 200 L.
µ
Two-fold drug dilutions in neat DMSO were performed and added to the microtiter plate at
a final concentration of 1% v/v DMSO. DMSO (1% v/v) was used as a positive control and
rifampicin as a negative control. The inoculum of M. bovis BCG culture was standardised at
OD600 0.05 in Middlebrook 7H9 medium (Difco labs, Franklin Lakes, NJ, USA) and added
◦
to the plate. Plates were incubated without shaking at 37 C, 5% CO2 for 7 days. Following
incubation, 42
µL of resazurin (0.02% v/v in dH2O) was added to each well and incubated
for a further 24 h. Fluorescence was measured using a Polarstar omega plate reader (BMG
LABTECH SARL, Ortenberg, Germany) using
λ
ex
544,
λ
em
590 to determine the MIC value.
4. Conclusions
The rational design and synthesis of a series of ligands (benzylphenyl, 1,4-triazole-
linked diaryl, ether-linked di-triclosan) were based on TCS which is a known inhibitor
of InhA. All molecules synthesized complied with Lipinski’s rule of 5 regarding good
drug-like properties and oral availability except the di-triclosan analogue 31 which violates
two of the rules. Compounds 13, 38, 39 and 40 which all have a ClogP > 5, should still have
some oral availability as they only violate one of rules. All molecules fitted well into the
active site when docked in silico with the di-triclosan analogues having the best fitness
scores, followed by the triazole-linked analogues and the benzyl phenyl ether or aniline
analogues. The isolated enzyme assays indicate that benzylphenyl aniline derivatives (
8,
13), 1,4-triazole-linked diaryl aniline analogues 21 23 and di-triclosan derivatives 31 and
,
38 display promise as possible leads. This inhibitory activity, however, requires further
investigation as it does not correlate well with the whole cell activity suggesting that
the compounds that are active against whole cells may have targets other than InhA.
Preliminary biological screening reveals that there is evidence that some compounds in
each group display moderate inhibitory activity. Compounds 12
, 39, 43 and 44 had good
MICs when tested against the whole cell bacteria. The differences observed between the
MIC’s measured on whole cells and isolated enzyme assays suggest that compounds 12
and 44 may inhibit enzymes other than InhA in mycobacteria.
, 39
In our recently published complementary study of di-triclosan analogues that incorpo-
rated a 1,4- or 1,5-triazole as a more rigid linker between the groups that should occupy the
two triclosan binding sites in InhA [47], we identified two compounds, 45 which exhibited
100% inhibition of InhA at [50
µM] (MIC > 80 µM) and 46 which gave 11% inhibition at
50 M (MIC 12.9 M). These results also reflect a lack of correlation between the results
µ
µ
from the isolated enzyme assays and the whole cell MIC values observed. Replacing the
B-ring of the second TCS with a smaller structure as seen in compounds 39 and 46 does
lead to a lower MIC value when compared with the ‘complete’ di-triclosans 38 and 45
respectively. Given that these structures show comparable activity to TCS in the whole cell
assays, it would be worth exploring which essential target(s) in Mycobacterium tuberculosis
in addition to InhA, they may be inhibiting.
,
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10
.3390/ph14040361/s1, Table S1: Fitness scores and rank list (high to low) for designed compounds
and intermediates; Table S2: Summary of the interactions of the benzylphenyl analogues with the
InhA active site.; Table S3: Summary of the interactions of the triazole linked analogues with the InhA
active site.; Table S4: Summary of the interactions of the di-triclosan analogues with the InhA active
site. Figure S1: Compounds tested. Figure S2: Interactions with the active site with benzylphenyl
derivative 2. Figure S3: A view of the docking of 8 into the InhA active site. The active site residues
(blue), NAD+ (magenta) and ligand (green). Figure S4: Docking of 23 into the InhA active site.
Figure S5: A view of the docking of 31 into the InhA active site. The active site residues (blue),
NAD+ (magenta) and ligand (green). Scheme S1: Synthesis of Octenyl CoA. Experimental details of
compounds synthesised in this paper together with associated analytical data and NMR spectra.

Pharmaceuticals 2021, 14, 361
17 of 20
Author Contributions: S.C. and N.R.T. conceptualized the project. S.C. designed, synthesized the
majority of the compounds and conducted molecular docking and isolated enzyme assays. T.A.
synthesized several of the compounds and conducted molecular docking and isolated enzyme assays.
W.C.D. and S.C. synthesized the substrate for the isolated enzyme assays. S.S.K. expressed and
purified the recombinant InhA for the isolated enzyme assays. D.E. and S.B. conducted the M.I.C.
studies. Supervision of the project: N.R.T. and C.I.D.M. at Nottingham; S.B. at Birkbeck. S.C. prepared
the original draft of the paper. S.C., T.A., S.S.K., W.C.D., C.I.D.M., D.E., S.B. and N.R.T. reviewed
and edited the final draft of the paper. All authors have read and agreed to the published version of
the manuscript.
Funding: We acknowledge the BBSRC, MRC and Wellcome Trust for the funding of this project. Tom
Armstrong is supported by the Wellcome Trust Antimicrobials and Antimicrobial Resistance (AAMR)
doctoral training programme Birmingham/Nottingham [203974/Z/17/A]. Dr William C. Drewe by
MRC grant [G0801741].
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: We acknowledge the University of Nottingham for providing facilities and PhD
funding for Sarentha Chetty and Shalu Sharma Kharkwal. The MIC evaluation for compound 39 was
conducted by Alice Lanne, University of Birmingham, UK. The InhA:pET15a construct was kindly
provided by Peter Tonge (University of Stony Brook, Stony Brook, NY, USA).
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or
in the decision to publish the results.
Abbreviations
CuAAC Copper(I) Azide-Alkyne Cycloaddition
ENR
XDR
INH
GOLD
HBA
HBD
InhA
MDR
Mtb
Enoyl reductase
Extensively drug-resistant
Isoniazid
Genetic Optimization for Ligand Docking
Hydrogen bond acceptors
Hydrogen bond donors
M. tuberculosis enoyl reductase
Multidrug-resistance
Mycobacterium tuberculosis
Nicotinamide adenosine dinucleotide
Room temperature
NAD+
r.t.
SAR
TPSA
TCS
Structure-activity relationship
Topological polar surface area
Triclosan
TB
Tuberculosis
References
1.
Global Tuberculosis Report 2019. Available online: https://www.who.int/tb/publications/global_report/en/ (accessed on
12 June 2019).
2.
3.
4.
5.
Tuberculosis. Available online: https://www.who.int/news-room/fact-sheets/detail/tuberculosis (accessed on 10 February 2020).
HIV/AIDS. Available online: www.who.int/news-room/fact-sheets/detail/hiv-aids (accessed on 10 February 2020).
Grange, J.M. Mycobacteria and Human Disease, 2nd ed.; Arnold-Hodder Headline Group: London, UK, 1996; pp. 159–187.
Finch, R.G.; Greenwood, D.; Ragnar Norby, S.; Whitley, R.J. Antibiotics and Chemotherapy, Anti-Infective Agents and Their Use in
Treatment, 8th ed.; Elsevier Science Ltd.: Edinburgh, UK, 2003; 427p.
6.
Bhatt, A.; Molle, V.; Besra, G.S.; Jacobs, W.R.; Kremer, L. The Mycobacterium tuberculosis FAS-II condensing enzymes: Their role
in mycolic acid biosynthesis, acid-fastness, pathogenesis and in future drug development. Mol. Microbiol. 2007, 64, 1442–1454.
[CrossRef] [PubMed]

Pharmaceuticals 2021, 14, 361
18 of 20
7.
Heath, R.J.; White, S.W.; Rock, C.O. Lipid biosynthesis as a target for antibacterial agents. Prog. Lipid Res. 2001, 40, 467–497.
[CrossRef]
8.
9.
Janin, Y.L. Antituberculosis drugs: Ten years of research. Bioorg. Med. Chem. 2007, 15, 2479–2513. [CrossRef] [PubMed]
Cole, S.T.; Brosch, R.; Parkhill, J.; Garnier, T.; Churcher, C.; Harris, D. Deciphering the biology of Mycobacterium tuberculosis from
the complete genome sequence. Nature 1998, 393, 537–544. [CrossRef]
10. Rozwarski, D.A.; Grant, G.A.; Barton, D.H.; Jacobs, W.R., Jr.; Sacchettini, J.C. Modification of the NADH of the isoniazid target
(InhA) from Mycobacterium tuberculosis. Science 1998, 279, 98–102. [CrossRef]
11. Zhao, X.; Yu, H.; Yu, S.; Wang, F.; Sacchettini, J.C.; Magliozzo, R.S. Hydrogen peroxide-mediated isoniazid activation catalyzed by
Mycobacterium tuberculosis catalase-peroxidase (KatG) and its S315T mutant. Biochemistry 2006, 45, 4131–4140. [CrossRef]
12. Timmins, G.S.; Deretic, V. Mechanisms of action of isoniazid. Mol. Microbiol. 2006, 62, 1220–1227. [CrossRef]
13. Quemard, A.; Sacchettini, J.C.; Dessen, A.; Vilcheze, C.; Bittman, R.; Jacobs, W.R., Jr.; Blanchard, J.S. Enzymatic characterization of
the target for isoniazid in Mycobacterium tuberculosis. Biochemistry 1995, 34, 8235–8241. [CrossRef]
14. Banerjee, A.; Dubnau, E.; Quemard, A.; Balasubramanian, V.; Um, K.S.; Wilson, T.; Collins, D.; De Lisle, G.; Jacobs, W.R. inhA, a
gene encoding a target for isoniazid and ethionamide in Mycobacterium tuberculosis. Sciecnce 1994, 263, 227–230. [CrossRef]
15. Vilchèze, C.; Morbidoni, H.R.; Weisbrod, T.R.; Iwamoto, H.; Kuo, M.; Sacchettini, J.C.; Jacobs, W.R. Inactivation of the inhA-
encoded fatty acid synthase II (FASII) enoyl-acyl carrier protein reductase induces accumulation of the FASI end products and
cell lysis of Mycobacterium smegmatis. J. Bacteriol. 2000, 182, 4059–4067. [CrossRef]
16. Cole, S.T.; Davies, K.E.; Mcmurray, D.N.; Jacobs, W.R. Tuberculosis and the Tubercle Bacillus, 1st ed.; ASM Press, American Society
for Microbiology: Washington, DC, USA, 2005.
17. Zhang, Y.; Heym, B.; Allen, B.; Young, D.; Cole, S. The catalase-peroxidase gene and isoniazid resistance of Mycobacterium
tuberculosis. Nature 1992, 358, 591–593. [CrossRef]
18. Zhang, Y.; Yew, W.-W. Mechanisms of drug resistance in Mycobacterium tuberculosis: Update 2015. Int. J. Tuberc. Lung Dis. 2015, 19,
1276–1289. [CrossRef] [PubMed]
19. Dias, M.V.B.; Vasconcelos, I.B.; Prado, A.M.X.; Fadel, V.; Basso, L.A.; de Azevedo, W.F., Jr.; Santos, D.S. Crystallographic
studies on the binding of isonicotinyl-NAD adduct to wild-type and isoniazid resistant 2-trans-enoyl-ACP (CoA) reductase from
Mycobacterium tuberculosis. J. Struct. Biol. 2007, 159, 369–380. [CrossRef]
20. Rožman, K.; Sosicˇ, I.; Fernandez, R.; Young, R.J.; Mendoza, A.; Gobec, S.; Encinas, L. A new ‘golden age’ for the antitubercular
target InhA. Drug Discov. Today 2017, 22, 492–502. [CrossRef] [PubMed]
21. Heath, R.J.; Rubin, J.R.; Holland, D.R.; Zhang, E.; Snow, M.E.; Rock, C.O. Mechanism of triclosan inhibition of bacterial fatty acid
synthesis. J. Biol. Chem. 1999, 274, 11110–11114. [CrossRef] [PubMed]
22. Vosatka, R.; Kratky, M.; Vinsova, J. Triclosan and its derivatives as antimycobacterial active agents. Eur. J. Pharm. Sci. 2018, 114,
318–331. [CrossRef] [PubMed]
23. Rajkhowa, S.; Jha, A.N.; Deka, R.C. Anti-tubercular drug development: Computational strategies to identify potential compounds.
J. Mol. Graph. Model. 2015, 62, 56–68. [CrossRef] [PubMed]
24. Wang, L.Q.; Falany, C.N.; James, M.O. Triclosan as a substrate and inhibitor of 3’-phosphoadenosine 5’-phosphosulfate-
sulfotransferase and UDP-glucuronosyl transferase in human liver fractions. Drug Metab. Dispos. 2004, 32, 1162–1169. [CrossRef]
[PubMed]
25. Kuo, M.R.; Morbidoni, H.R.; Alland, D.; Sneddon, S.F.; Gourlie, B.B.; Staveski, M.M. Targeting tuberculosis and malaria through
inhibition of Enoyl reductase: Compound activity and structural data. J. Biol. Chem. 2003, 278, 20851–20859. [CrossRef]
26. Sivaraman, S.; Sullivan, T.J.; Johnson, F.; Novichenok, P.; Cui, G.; Simmerling, A.C.; Tonge, P.J. Inhibition of the Bacterial Enoyl
Reductase FabI by Triclosan: A Structure
509–518. [CrossRef]
−
Reactivity Analysis of FabI Inhibition by Triclosan Analogues. J. Med. Chem. 2004, 47,
27. Sivaraman, S.; Zwahlen, J.; Bell, A.F.; Hedstrom, L.; Tonge, P.J. Structure-activity studies of the inhibition of FabI, the enoyl
reductase from Escherichia coli, by triclosan: Kinetic analysis of mutant FabIs. Biochemistry 2003, 42, 4406–4413. [CrossRef]
28. Sullivan, T.J.; Truglio, J.J.; Boyne, M.E.; Novichenok, P.; Zhang, X.; Stratton, C.F.; Li, H.-J.; Kaur, T.; Amin, A.; Johnson, F.; et al.
High Affinity InhA Inhibitors with Activity against Drug-Resistant Strains of Mycobacterium tuberculosis. ACS Chem. Biol. 2006, 1,
43–53. [CrossRef] [PubMed]
29. Rozwarski, D.A.; Vilchèze, C.; Sugantino, M.; Bittman, R.; Sacchettini, J.C. Crystal Structure of the Mycobacterium tuberculosis
Enoyl-ACP Reductase, InhA, in Complex with NAD+ and a C16 Fatty Acyl Substrate. J. Biol. Chem. 1999, 274, 15582–15589.
[CrossRef] [PubMed]
30. Ende, C.W.A.; Knudson, S.E.; Liu, N.; Childs, J.; Sullivan, T.J.; Boyne, M.; Xu, H.; Gegina, Y.; Knudson, D.L.; Johnson, F.; et al.
Synthesis and in vitro antimycobacterial activity of B-ring modified diaryl ether InhA inhibitors. Bioorganic Med. Chem. Lett. 2008
18, 3029–3033. [CrossRef] [PubMed]
,
31. Freundlich, J.S.; Wang, F.; Vilcheze, C.; Gulten, G.; Langley, R.; Schiehser, G.A. Triclosan derivatives: Towards potent inhibitors of
drug-sensitive and drug-resistant Mycobacterium tuberculosis. ChemMedChem 2009, 4, 241–248. [CrossRef]
32. Luckner, S.R.; Liu, N.; Am Ende, C.W.; Tonge, P.J.; Kisker, C. A slow, tight binding inhibitor of InhA, the enoyl-acyl carrier protein
reductase from Mycobacterium tuberculosis. J. Biol. Chem. 2010, 285, 14330–14337. [CrossRef]
33. He, X.; Alian, A.; Stroud, R.; Ortiz de Montellano, P.R. Pyrrolidine carboxamides as a novel class of inhibitors of enoyl acyl carrier
protein reductase from Mycobacterium tuberculosis. J. Med. Chem. 2006, 49, 6308–6323. [CrossRef] [PubMed]

Pharmaceuticals 2021, 14, 361
19 of 20
34. Cinu, T.A.; Sidhartha, S.K.; Indira, B.; Varadaraj, B.G.; Vishnu, P.S.; Shenoy, G.G. Design, synthesis and evaluation of antitubercular
activity of Triclosan analogues. Arab. J. Chem. 2019, 12, 3316–3323. [CrossRef]
35. Kar, S.S.; Bhat, G.V.; Rao, P.P.; Shenoy, V.P.; Bairy, I.; Shenoy, G.G. Rational design and synthesis of novel diphenyl ether derivatives
as antitubercular agents. Drug Des. Dev. Ther. 2016, 10, 2299–2310.
36. Khade, A.B.; Boshoff, H.I.M.; Arora, K.; Vandana, K.E.; Verma, R.; Shenoy, G.G. Design, synthesis, evaluation, and molecular
dynamic simulation of triclosan mimic diphenyl ether derivatives as antitubercular and antibacterial agents. Struct. Chem. 2020
31, 983–998. [CrossRef]
,
37. Verma, R.; Boshoff, H.I.; Arora, K.; Bairy, I.; Tiwari, M.; Bhat, V.G.; Shenoy, G.G.; Bhat, G.V. Synthesis, antitubercular evaluation,
molecular docking and molecular dynamics studies of 4,6-disubstituted-2-oxo-dihydropyridine-3-carbonitriles. J. Mol. Struct.
2019, 1197, 117–133. [CrossRef]
38. Verma, R.; Boshoff, H.I.; Arora, K.; Bairy, I.; Tiwari, M.; Varadaraj, B.G.; Shenoy, G.G. Synthesis, evaluation, molecular docking,
and molecular dynamics studies of novel N-(4-[pyridin-2-yloxy]benzyl)arylamine derivatives as potential antitubercular agents.
Drug Dev. Res. 2020, 81, 315–328. [CrossRef] [PubMed]
39. Boyne, M.E.; Sullivan, T.J.; amEnde, C.W.; Lu, H.; Gruppo, V.; Heaslip, D. Targeting fatty acid biosynthesis for the development of
novel chemotherapeutics against Mycobacterium tuberculosis: Evaluation of A-ring-modified diphenyl ethers as high-affinity InhA
inhibitors. Antimicrob. Agents Chemother. 2007, 51, 3562–3567. [CrossRef] [PubMed]
40. Kini, S.G.; Bhat, A.; Pan, Z.; Dayan, F.E. Synthesis and antitubercular activity of heterocycle substituted diphenyl ether derivatives.
J. Enzyme Inhib. Med. Chem. 2010, 25, 730–736. [CrossRef]
41. Kini, S.G.; Bhat, A.R.; Bryant, B.; Williamson, J.S.; Dayan, F.E. Synthesis, antitubercular activity and docking study of novel cyclic
azole substituted diphenyl ether derivatives. Eur. J. Med. Chem. 2009, 44, 492–500. [CrossRef] [PubMed]
42. Yang, Y.; Wang, Z.; Yang, J.; Yang, T.; Pi, W.; Ang, W.; Lin, Y.; Liu, Y.; Li, Z.; Luo, Y.; et al. Design, synthesis and evaluation of
novel molecules with a diphenyl ether nucleus as potential antitubercular agents. Bioorganic Med. Chem. Lett. 2012, 22, 954–957.
[CrossRef] [PubMed]
43. Pan, P.; Knudson, S.E.; Bommineni, G.R.; Li, H.J.; Lai, C.T.; Liu, N. Time-dependent diaryl ether inhibitors of InhA: Structure-
activity relationship studies of enzyme inhibition, antibacterial activity, and in vivo efficacy. ChemMedChem 2014, 9, 776–791.
[CrossRef]
44. Pan, P.; Tonge, P.J. Targeting InhA, the FASII Enoyl-ACP Reductase: SAR Studies on Novel Inhibitor Scaffolds. Curr. Top. Med.
Chem. 2012, 12, 672–693. [CrossRef] [PubMed]
45. Roberto Chiarelli, L.; Mori, G.; Esposito, M.; Silvia Orena, B.; Rosalia Pasca, M. Biological evaluation of potent triclosan-derived
inhibitors of the enoyl-acyl carrier protein reductase InhA in drug-sensitive and drug-resistant strains of Mycobacterium tuberculosis.
ChemMedChem 2014, 9, 2528–2537.
46. Upadhayaya, R.S.; Kulkarni, G.M.; Vasireddy, N.R.; Vandavasi, J.K.; Dixit, S.S.; Sharma, V.; Chattopadhyaya, J. Design, synthesis
and biological evaluation of novel triazole, urea and thiourea derivatives of quinoline against Mycobacterium tuberculosis. Bioorganic
Med. Chem. 2009, 17, 4681–4692. [CrossRef]
47. Armstrong, T.; Lamont, M.; Lanne, A.; Alderwick, L.J.; Thomas, N.R. Inhibition of Mycobacterium tuberculosis InhA: Design,
synthesis and evaluation of new di-triclosan derivatives. Bioorganic Med. Chem. 2020, 28, 115744. [CrossRef] [PubMed]
48. Rodriguez, F.; Saffon, N.; Sammartino, J.C.; Degiacomi, G.; Pasca, M.R.; Lherbet, C. First triclosan-based macrocyclic inhibitors of
InhA enzyme. Bioorganic Chem. 2020, 95, 103498. [CrossRef] [PubMed]
49. Owono, L.C.O.; Ntie-Kang, F.; Keita, M.; Megnassan, E.; Frecer, V.; Miertuš, S. Virtually Designed Triclosan-Based Inhibitors of
Enoyl-Acyl Carrier Protein Reductase of Mycobacterium tuberculosis and of Plasmodium falciparum. Mol. Inform. 2015, 34, 292–307.
[CrossRef]
50. Available online: www.acdlabs.com (accessed on 10 August 2011).
51. Wang, J.; Li, P.; Shen, Q.; Song, G. Concise synthesis of aromatic tertiary amines via a double Petasis–borono Mannich reaction of
aromatic amines, formaldehyde, and organoboronic acids. Tetrahedron Lett. 2014, 55, 3888–3891. [CrossRef]
52. Cheng, Y.-J.; Liu, M.S.; Zhang, Y.; Niu, Y.; Huang, F.; Ka, J.-W.; Yip, H.-L.; Tian, Y.; Jen, A.K.-Y. Thermally Cross-Linkable
Hole-Transporting Materials on Conducting Polymer: Synthesis, Characterization, and Applications for Polymer Light-Emitting
Devices. Chem. Mater. 2008, 20, 413–422. [CrossRef]
53. SYBYL 8.1, SYBYL X. Available online: http://www.tripos.co.uk (accessed on 4 April 2021).
54. GOLD 1.1 ed. Available online: http://www.ccducam.ac.uk/ (accessed on 4 April 2021).
55. Jones, G.; Willett, P.; Glen, R.C.; Leach, A.R.; Taylor, R. Development and validation of a genetic algorithm for flexible docking. J.
Mol. Biol. 1997, 267, 727–748. [CrossRef] [PubMed]
56. Molinspiration Cheminformatics Server. Available online: http://www.molinspiration.com (accessed on 20 February 2020).
57. Thomas, N.R.; Yang, Y.; Drew, W.C.; Lees, K.J. Biomolecular Labelling Using Multifunctional Biotin Analogues. United Kingdom
Patent WO 2010/106347 A2, 22 March 2010.
58. Pal, M.; Parasuraman, K.; Yeleswarapu, K.R. Palladium-Catalyzed Cleavage of O/N-Propargyl Protecting Groups in Aqueous
Media under a Copper-Free Condition1. Org. Lett. 2003, 5, 349–352. [CrossRef]
59. Spletstoser, J.T.; Flaherty, P.T.; Himes, R.H.; Georg, G.I. Synthesis and Anti-Tubulin Activity of a 3-(4-Azidophenyl)-3-
dephenylpaclitaxel Photoaffinity Probe. J. Med. Chem. 2004, 47, 6459–6465. [CrossRef]

Pharmaceuticals 2021, 14, 361
20 of 20
60. Drewe, W.C.; Neidle, S. Click chemistry assembly of G-quadruplex ligands incorporating a diarylurea scaffold and triazole
linkers. Chem. Commun. 2008, 5295–5297. [CrossRef]
61. Parikh, S.; Moynihan, D.P.; Xiao, G.; Tonge, P.J. Roles of tyrosine 158 and lysine 165 in the catalytic mechanism of InhA, the
enoyl-ACP reductase from Mycobacterium tuberculosis. Biochemistry 1999, 38, 13623–13634. [CrossRef]
62. Danquah, C.A.; Maitra, A.; Gibbons, S.; Faull, J.; Bhakta, S. HT-SPOTi: A rapid, gold standard drug susceptibility test (DST), to
detect antibiotic resistance profile as well as to evaluate novel chemical entities for new anti-infective drug discovery. Curr. Protoc.
Microbiol. 2015, 2015, 17–18.
63. Evangelopoulos, D.; Bhakta, S. Rapid methods for testing inhibitors of mycobacterial growth. Methods Mol. Biol. 2010, 642,
193–201. [PubMed]