Methyl-thiazoles: A novel mode of inhibition with the potential to develop novel inhibitors targeting InhA in mycobacterium tuberculosis

Article abstract of DOI:10.1021/jm4012033

InhA is a well validated Mycobacterium tuberculosis (Mtb) target as evidenced by the clinical success of isoniazid. Translating enzyme inhibition to bacterial cidality by targeting the fatty acid substrate site of InhA remains a daunting challenge. The recent disclosure of a methyl-thiazole series demonstrates that bacterial cidality can be achieved with potent enzyme inhibition and appropriate physicochemical properties. In this study, we report the molecular mode of action of a lead methyl-thiazole, along with analogues with improved CYP inhibition profile. We have identified a novel mechanism of InhA inhibition characterized by a hitherto unreported "Y158-out" inhibitor-bound conformation of the protein that accommodates a neutrally charged "warhead". An additional novel hydrophilic interaction with protein residue M98 allows the incorporation of favorable physicochemical properties for cellular activity. Notably, the methyl-thiazole prefers the NADH-bound form of the enzyme with a K_d of ～13.7 nM, as against the NAD⁺-bound form of the enzyme.

Full text of DOI:10.1021/jm4012033

Article
pubs.acs.org/jmc
Methyl-Thiazoles: A Novel Mode of Inhibition with the Potential to
Develop Novel Inhibitors Targeting InhA in Mycobacterium
tuberculosis
Pravin S. Shirude,*,† Prashanti Madhavapeddi,‡ Maruti Naik,† Kannan Murugan,† Vikas Shinde,†
Radha Nandishaiah,‡ Jyothi Bhat,‡ Anupriya Kumar,† Shahul Hameed,† Geoffrey Holdgate,#
Gareth Davies,# Helen McMiken,# Naina Hegde,‡ Anisha Ambady,‡ Janani Venkatraman,‡
Manoranjan Panda,† Balachandra Bandodkar,† Vasan K. Sambandamurthy,‡ and Jon A. Read*,#
‡
†Department of Medicinal Chemistry, Department of Biosciences, AstraZeneca India Pvt. Ltd., Bellary Road, Hebbal,
Bangalore-560024, India
#Discovery Sciences, AstraZeneca, Alderley Park, Macclesfield SK10 4TG, U.K.
S
* Supporting Information
ABSTRACT: InhA is a well validated Mycobacterium tuberculosis (Mtb) target as
evidenced by the clinical success of isoniazid. Translating enzyme inhibition to
bacterial cidality by targeting the fatty acid substrate site of InhA remains a daunting
challenge. The recent disclosure of a methyl-thiazole series demonstrates that
bacterial cidality can be achieved with potent enzyme inhibition and appropriate
physicochemical properties. In this study, we report the molecular mode of action of
a lead methyl-thiazole, along with analogues with improved CYP inhibition profile.
We have identified a novel mechanism of InhA inhibition characterized by a hitherto
unreported “Y158-out” inhibitor-bound conformation of the protein that
accommodates a neutrally charged “warhead”. An additional novel hydrophilic
interaction with protein residue M98 allows the incorporation of favorable
physicochemical properties for cellular activity. Notably, the methyl-thiazole prefers
the NADH-bound form of the enzyme with a Kd of ∼13.7 nM, as against the NAD+-bound form of the enzyme.
enzyme oxidizes INH to an acyl radical which then forms a
covalent adduct (INH-NAD) with nicotinamide adenine
dinucleotide (NAD).6 The active drug is the INH-NAD
covalent adduct (1) and one of the resistance mechanisms to
INH is via a specific mutation in the KatG gene.6 A number of
additional drugs such as ethionamide and propionamide also
target InhA via an adduct with cofactors.8 Additionally, the
existence of Mtb clinical isolates that harbor mutations in the
inhA structural gene or inhA promoter region that confer
resistance to INH have been reported. These observations
imply that identifying direct inhibitors of InhA would have
tremendous clinical value in combating TB, not least due to the
likelihood of being devoid of cross-resistance with current
therapies that target InhA via a pro-drug mechanism.7
Multiple molecular modes of action have been attempted to
target InhA. The early success of INH in inhibiting InhA relied
on targeting the NAD adduct formation which competes
kinetically with the cofactor NADH. A similar mode of action
has been attempted with boronate NAD adducts with limited
success.9 Mimicking the molecular mode of action of INH
could be achieved either using a new pro-drug or by discovering
INTRODUCTION
■
Tuberculosis (TB) continues to be a major global cause of
morbidity and mortality due to the infectious pathogen
Mycobacterium tuberculosis (Mtb). The emergence of Mtb
strains resistant to first line and second line TB drugs adds to
the challenge in global efforts to control this infection.1 The
bacterial fatty acid biosynthesis pathway represents a validated
and yet relatively unexploited target for drug discovery.2 Fatty
acids are essential for bacterial growth, however, they cannot be
scavenged from the host and must be synthesized de novo.3,4 In
Mtb, Enoyl-acyl carrier protein reductase (ACPER), known as
InhA, is encoded by the inhA gene as an essential NADH
dependent enzyme in the mycolic acid biosynthetic pathway.4,5
Mycolic acids are linked to the cell wall and form a waxy
protective coating around the bacterial cell, which serves as a
permeability barrier. The bacterial fatty acid biosynthetic
pathway (FAS-II) is fundamentally distinct from the multi-
enzyme FAS-I complex found in mammals. This combination
results in a molecular target which is both essential in Mtb as
well as sufficiently different from human enzymes to be an
attractive target for small molecule drug discovery.
InhA is a clinically validated target based on the success of
isoniazid (INH) in treating TB patients.5−7 INH is a pro-drug
and is activated by KatG, a catalase−peroxidase enzyme. This
Received: June 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Chemical structures of key InhA inhibitors. 1, isoniazid-NAD adduct (INH-NAD); 2, pyridomycin; 3, triclosan; 4, PT70; 5,
benzimidazolo-pyrrolidinone; 6, aminoquinolines; 7, methyl thiazole.
a
a small molecule which would mimic 1. The latter option poses
significant issues such as the molecular weight of 1, which is in
excess of 800 Da. If analogues targeting this pathway required a
similar molecular weight, this would likely limit opportunities
for oral absorption.10 The natural product 2 (pyridomycin), is
reported to kill Mtb by targeting InhA.11 Compound 2 does
not show reduced activity in cell lines which are resistant to
INH via mutation in KatG. Interestingly, it has been established
that 2 competes with cofactor.11 Although the complex
structure of 2 with InhA has not yet been published, this
result implies that 2 may act by binding in the same binding site
as cofactor and 1.
Alternative approaches for direct inhibition of InhA include
allosteric modulators of InhA binding outside the active site or
inhibitors that bind within the substrate binding site, thereby
competing for substrate or product. To date, no allosteric
inhibitors of InhA have been reported. Substrate competitive
compounds, however, such as 3 (triclosan), have been known
for some time. Compound 3, a highly effective broad-spectrum
antibacterial agent, targets FabI, the nonmycobacterial equiv-
alent of InhA.2 3 binds directly to the NAD+ binary complex of
the enzyme and therefore does not require KatG activation for
antibacterial activity.12,13 3 is a submicromolar (0.2 μM)
inhibitor of InhA with very weak antimycobacterial activity
(minimal inhibitory concentration (MIC) of 64 μM). The
discovery of highly ligand efficient analogues of 3 such as 4
(PT70)14 strengthens this approach to target InhA. However,
translating enzyme inhibition to bacterial cidality while
retaining the physicochemical properties required for achieving
optimal bioavailability remains a significant challenge. Our
efforts within AstraZeneca have led to the identification of a
number of different scaffolds in this pathway including 5
(benzimidazolo-pyrrolidinone) and 6 (aminoquinolines), which
lack Mtb MIC despite good enzyme inhibition (Figure 1 and
Table 1). This led us to conclude that there is disconnect
between enzyme inhibition and cellular potency.
Table 1. IC50 and MIC of Key InhA Inhibitors
1
2
3
4
5
6
7
Ki (nM)
0.75
NA
NA
6
NA
0.2
NA
NA
1
NA
NA
InhA IC50
(μM)
0.022
0.02
0.003
b
MIC (μM)
0.05
0.57
64
NA
>64
>64
0.19
a
For compounds 1−4: values are from literature. NA: not available.
b
MIC for INH.
cellular potency (MIC) can be achieved with potent enzyme
inhibition (IC50). The methyl thiazole scaffold (7) (Figure 1)
has been reported15,16 as a direct InhA inhibitor where potent
enzyme inhibition (InhA IC50 = 0.003 μM) translates into
cellular potency (Mtb MIC = 0.19 μM) (Table 1). We have
elucidated the molecular mode of action of this series and the
properties that drive cellular potency. In addition, we report
new analogues of the methyl thiazole series which act through a
similar mode of inhibition to the reported methyl thiazoles
while showing improved cytochrome P450 (CYP) inhibition
and safety profile.
RESULTS AND DISCUSSION
■
To understand the molecular mode of action of 7, we profiled
this compound with various techniques. These included
biophysical characterization to determine the form of InhA to
which the compound bound and the crystal structure
determination of InhA in complex with 7 to gain a more
detailed insight into the mode of action. Finally, selection
mutagenesis was employed to confirm that InhA is indeed the
target of these compounds. The binding affinity of 7 to InhA
was shown to be dependent upon the presence of the added
cofactor and its oxidation state (Figure 2, Table 2). Using
isothermal titration calorimetry (ITC) and thermal melting, no
binding was detected to apo InhA by 7 under the conditions
used, while 7 showed weak binding to the NAD+-bound form
(9.3 μM) and tight binding to the NADH-bound form (13.7
nM). The ITC data was used primarily to determine Kd and
Most recently in 2011, the publication of the methyl thiazole
series by GlaxoSmithKline has shown that mycobacterial
B
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
as observed on addition of either the oxidized (NAD+) or
reduced (NADH) cofactor. Titration of 7 with apo enzyme or
the E.NAD+ complex (enzyme−NAD+ complex) resulted in no
significant increase in melting temperature. However, a
dramatic increase in the melting temperature of InhA is
observed on the titration of 7 to the E.NADH complex (Table
2, Supporting Information Figure S1).
The ITC binding studies suggest that 7 follows uncompe-
titive kinetics with respect to both oxidized and reduced
cofactor. Binding therefore occurs only after the cofactor has
bound to the free enzyme under the assay conditions.
Compound 7 binds with an approximately 1000-fold higher
affinity to the E.NADH complex. The stoichiometry measured
for 7 binding to the E.NADH complex was 0.86
0.002.
Binding of 7 can therefore be described by Scheme 1 below,
Scheme 1. Proposed Mode of Inhibition for Compound 7
Figure 2. Isothermal titration calorimetry (ITC) data for 7 binding to
InhA. Titrations of 7 with apo InhA (blue) and with NADH-bound
(black) and NAD+-bound (red) are shown.
wherein the higher affinity binding to the E.NADH compared
to the E.NAD+ form suggests a mode of inhibition that may be
different to the stacking interactions typical of 3 and 4 with
E.NAD+. Previous kinetic studies have revealed that substrate
binds in random order,18 suggesting that 7 could inhibit InhA
by binding to E.NADH to form a ternary complex or by
binding to E.NADH.S (where S refers to long chain trans 2-
enoyl-acyl carrier protein (ACP) substrates) to form a
quaternary complex, however, as there was no substrate present
in the ITC experimental setup, we cannot distinguish between
these two modes of inhibition from the ITC data alone.
These data suggest that of the two binary complex options, 7
binds to the binary E.NADH form to make a ternary complex.
The thermal melting data are consistent with this hypothesis.
Interestingly, estimates of intracellular concentrations of
NADH and NAD+ levels in Mycobacteria are similar and well
above Km for each cofactor species.19 As Km (NADH) is
approximately 100-fold tighter than Km (NAD+), it could be
hypothesized that binding to E.NADH would be an efficient
way to inhibit a binary complex form of InhA as the vast
majority of the enzyme would be in this form, assuming the
estimates of intracellular cofactor concentrations are valid and
that other effects such as local concentrations, are not evident.
The crystal structure of InhA in complex with 7 is shown in
Figure 3. Consistent with the biophysical measurements, 7
binds to the E.NAD complex and occupies the substrate
determine the preference of the compound for E:NAD+,
E.NADH, or apo InhA. There may be multiple other reasons
for the change in enthalpy/entropy such as changes in protein
dynamics, changes in contribution of water molecules, entropy/
enthalpy compensation, entropy/enthalpy transduction, and
differential conformational changes. We have not tried to
interpret these values in this context.
The binding affinity of 7 to the NADH-bound form of the
enzyme, determined from surface plasmon resonance (SPR)
studies is 11.5 nM (Supporting Information), agreed well with
that from ITC (13.7 nM). In addition, binding affinities for
NAD+ and NADH to InhA were determined by SPR under the
same conditions as described for 7, with Kd values being 434
μM (95% confidence intervals 263−606 μM) and 5.6 μM (95%
confidence intervals 5.3−5.8 μM), respectively. The binding
affinity for NADH was also measured by ITC, with good
agreement to the SPR derived value, with Kd = (2.4 0.3) μM.
Compound binding to the NADH form fits to a “one binding
site per monomer” model in both ITC and SPR. It is difficult to
know whether the protein is a monomer or tetramer from ITC
and SPR experiments. However, the results are consistent
between the two techniques. Further, the mode of inhibition
was investigated using binding studies of 7 in thermal melting
experiments utilizing the differential scanning fluorimetry.17
The data shows a modest increase in thermal stability of InhA
a
Table 2. Biophysical Measurements of 7
ITC
SPR
thermal melting
ΔTm (°C)
(cofactor
bound form)
InhA
form
Tm (°C)
(apoenzyme)
Kd (nM)
ΔG (kJ/mol)
ΔH (kJ/mol) −TΔS (kJ/mol)
Kd (nM)
Tm (°C)
ΔTm (°C)
Apo
NAD+
NB
n/a
n/a
n/a
NA
NA
54.2 0.22
55.2 0.29
n/a
9300 1100
−28.7 0.3
−58.2 7.1
−29.5 7.1
+1.0
55.0 0.003
(at 50
μM of 7)
−0.2
NADH
13.7 3.8
−44.9 0.7
−39.3 0.2
5.4 0.7
11.5 2.8
56.5 0.11
+2.3
69.3 0.20(at
50 μM of 7)
+12.8
a
NB: no binding. n/a: not applicable. NA: not available.
C
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Substrate site binding regions of InhA. Protein complexes of
7 (PDB ID 4bpq), 4 (PDB ID 2x23),20 and C16 fatty acyl substrate
(PDB ID 1bvr)21 are overlaid. The figure highlights three regions
within the substrate binding site: site I, the catalytic site; site II, the
hydrophobic pocket; site III, also termed the size-limited region.14
Proteins atoms have been removed for clarity, except for M98 from the
InhA−NADH−7 complex which is drawn with carbon atoms in
yellow. 7 is shown as sticks with green carbon atoms, 4 is shown with
cyan carbon atoms, and C16 fatty acyl substrate is shown with orange
carbon atoms. NAD is shown as sticks with pink carbon atoms. Both
the phenolic ring of 4 and the methyl thiazole ring of 7 occupy a
similar positioning, allowing π-stacking with the nicotinamide ring of
the cofactor NAD.
Figure 3. X-ray crystal structure (PDB ID 4bpq) of 7 bound to InhA
of Mtb. Carbon atoms for 7 are shown in green. The protein backbone
cartoon is represented in yellow. Selected atoms for the InhA side
chains including the catalytic residue Y158 and M98 are shown as
sticks. NADH is shown as sticks colored with cyan carbon atoms.
Refined (2fo fc) electron density contoured at 1.0σ for 7 is represented
as wire mesh. Some atoms of the active site covering loop have been
removed for clarity.
binding site. The methyl-thiazole group engages with the
nicotinamide and ribose groups of the NAD cofactor, and this
interaction can be seen more clearly in Figure 5d. These
specific interactions with the cofactor are in common with
known substrate competitive InhA inhibitors. The site defined
by ligand groups interacting with the nicotinamide ring and
ribose hydroxyl in substrate competitive inhibitor complexes
will be termed “site I” subsequently in this article. The thiazole
ring “N” acts as a hydrogen bond acceptor to the hydroxyl
group of the ribose of the cofactor NADH. The thiazole ring
itself π-stacks with the nicotinamide ring of NADH. Uniquely
among known InhA substrate competitive inhibitors, the
methyl group of the methyl thiazole forces Y158 to maintain
a “flipped out” conformation by preventing the residue from
adopting an active “flipped in” conformation as seen in all
previously reported substrate competitive InhA inhibitors. The
geometry of the tetrahedral carbon adjacent to the thiazole ring
allows the remainder of the molecule to curl away from a path
approximately parallel with NADH and avoid steric clashes with
the protein, namely the main chain of G96.
A unique H-bond donor−acceptor pair is formed between a
thiadiazole ring nitrogen and the amine “NH” between the
thiadiazole and pyrazole rings with the M98 backbone amide
“NH” and carbonyl “O”, respectively. This interaction is also
previously unreported in crystal structures of InhA with
inhibitors. The orientation of this donor−acceptor hydrogen
bond pairing provides an excellent vector for the introduction
of a linking pyrazole which allows the difluoro-phenyl ring to
access a hydrophobic pocket. The hydrophobic pocket
accommodates the alkyl chain of the InhA substrate during
the InhA catalytic cycle (Figure 4). The thiadiazole, pyrazole,
and phenyl rings of 7 wrap around the side chain of M103.
Finally, the active site loop of InhA (residues 198−206), a helix
which may be disordered or ordered in crystal structures of
InhA dependent on interactions with ligand, makes nonpolar
contacts between M199 and the methyl-thiazole sulfur, I202
and the phenyl ring. The loop is fully ordered in subunits A−D,
and density is weaker in subunits E and F. The crystal structure
of the complex between InhA and 7 correlates well with
biophysical analysis. The structure is consistent with the
compound binding to a binary complex of InhA. At 2.0 Å, the
resolution is not sufficient to determine whether the cofactor is
present in a reduced or oxidized state. From this point onward,
NAD will be used to denote cofactor when either the actual
redox state is not confirmed or the statement refers equally to
NAD+ or NADH. However, the crystals were soaked in reduced
NADH cofactor concomitant with the introduction of 7. As it
had previously been shown that cofactor can be soaked out of
and into this crystal form (unpublished results), it can be
assumed that the cofactor in this crystal is in the reduced
(NADH) form.
The InhA substrate binding site can be divided into three key
regions that substrate-competitive inhibitors occupy and which
are highlighted in Figure 4. In 323 and 4,20 site I contains
phenolic oxygen which is often linked to a ring system which
stacks with the nicotinamide group. In other ligands, such as
the pyrrolidine carboxamides,22 the ring at site I is not aromatic
and the oxygen engaging with the ribose at site I need not be
directly attached to a ring as demonstrated by the piperazine-
indole-formamide compound GENZ10850 (PDB ID 1p44)13
or the natural fatty acid substrate. Prior to the identification of
7, all substrate-competitive inhibitors with known binding
mode possessed an oxygen atom that interacts with the ribose
ring and the hydroxyl of Y158. This interaction with Y158 will
be discussed further below. Site II is a hydrophobic pocket that
accommodates the long alkyl chains of the substrate.21 The
pocket is flexible due to significant movements in the side
chains of a few key residues such as Y158 and F149 and large
movements in the substrate-binding loop (amino acids 196−
208). Flexibility within the hydrophobic pocket is exploited in
InhA substrate-competitive inhibitors. Notably 3, a moderately
potent inhibitor of InhA, does not exploit this pocket.23 While
extending into the hydrophobic pocket provides benefit by
increasing potency, it also has the effect of significantly
increasing lipophilicity. A third and relatively unexplored
region, site III is located to the top right of Figure 4. This
region presents the opportunity for hydrophilic interactions
with M98 as well as the placement of hydrophilic groups (in
this case a hydroxyl group) adjacent to the phosphate groups of
D
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 5. InhA catalytic site I interactions: (a,b) diagrammatic representation of chemical structures and (c,d) stick representations of X-ray crystal
structures representing two classes of catalytic site interactions. Substrate competitive inhibitors have been categorized into two types by their
interactions at site I according to the conformation of tyrosine residue 158; “Y158-in” (a,c) as demonstrated by 420 and a previously unreported
“Y158-out” conformation (b,d) as demonstrated by the complex of 7 with InhA.
Table 3. Methyl-thiazoles Are Potent Inhibitors of InhA and Inhibit the Growth of Replicating Mtb in Broth
NAD. These unexplored features of the InhA binding pocket
present a significant opportunity to modulate the physicochem-
ical properties of InhA binding molecules.
interactions with the nicotinamide ring of NAD (exceptions are
pyrrolidine carboxamides22 and piperazine indole forma-
mides13). In Figure 5c,d, NAD is represented with pink carbon
atoms. Compound 4 is represented with cyan carbon atoms
and 7 with light-green carbon atoms. Y158 carbon atoms are in
blue or green.
Triclosan analogues such as 4 have been reported to bind
preferentially to the NAD+ form of InhA.20 Compound 3 and
by analogy 4, mimics the enolate anion of the substrate12 of
ACPER and is probably present in the complex as a phenolic
anion24 (calculated pKa = 7.9) at physiological pH.14 This likely
contributes to affinity by complementing the charge on the
In the “Y158-in” binding mode, the ligand has an atom which
is capable of simultaneously forming a polar interaction with
both a ribose hydroxyl and the Y158 phenolic hydroxyl. The
“Y158-out” binding mode retains the interaction of the ligand
with the ribose hydroxyl group; however, Y158 adopts an
alternative χ1 torsion angle similar to the apo form of the
enzyme, preventing the tyrosine phenolic oxygen from
interacting with the ligand. Both Y158-in and Y158-out
inhibitors often have aromatic rings which can form π-stacking
E
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
oxidized NAD+ cofactor. In addition, the phenolic oxygen
allows triclosan or PT70 to make a bidentate interaction with
the hydroxyl of Y158, which adopts the Y158-in conformation,
and an NAD+ ribose hydroxyl. While the charge complemen-
tarity may contribute to the potency of these compounds, it
may be a factor in the poor permeability and bioavailability of
such compounds.
An overlay of the complex structures of InhA with
compounds 4 and 7 is shown in Figure 4. The binding mode
of the methyl-thiazole at site I is novel with respect to the
binding motif around the nicotinamide and adjacent ribose
groups of the NAD cofactor and the interactions with Y158 of
the protein and is shown more clearly in Figure 5. In the
complex of 7 with InhA, there is a single hydrogen bond
between the acceptor “N” on the thiazole ring and the ribose of
NAD. The methyl group of 7 prevents Y158 from rotating
inward, and Y158 remains in the “Y158-out” conformation as
seen in the binary form of the InhA:NAD complex with 7
(Figure 5) and the apo structure of InhA25 (PDB ID 1eny). All
previously identified substrate competitive inhibitors have a
binding mode similar to 3 and 4, i.e., Y158-in.26 In contrast to a
charged Y158-in inhibitor, an inhibitor which binds to the
“Y158-out” conformation of the E.NADH form of InhA could
accept a hydrogen bond from the ribose hydroxyl but would be
uncharged.
Keeping these interactions and considerations in mind, we
designed and synthesized new analogues in order to understand
structure−activity relationships (SAR), as shown in Table 3. All
new analogues were racemic. Compound 8 shows similar
enzyme inhibition as that of its chiral analogue 7, however, the
MIC is 2-fold weaker (0.5 μM) than 7. In 9, moving the “N” of
the pyrazole ring as compared to 7 resulted in weaker IC50 and
MIC. Introduction of an additional “N” in the pyrazole ring, to
yield a triazole in 10, showed much better enzyme inhibition
(0.03 μM) and cellular potency (<0.39 μM). Pyrazole amines
are known to inhibit CYP enzymes. We therefore decided to
investigate bioisosteric replacements for the triazole ring in 10
as well as changing the pyrazole ring to the open amide in 11.
Compound 11 with the open amide is a new lead with good
enzyme inhibition and cellular potency. The crystal structure of
InhA in complex with 11 (Figure 6) revealed similar
interactions to that of 7 (Figure 3). The methyl thiazole and
thiadiazole groups form identical hydrogen bonding inter-
actions with the ribose hydroxyl of NAD and the backbone
amide of M98, respectively. The major difference in binding
occurs at the difluorophenyl group. In the complex of 7 with
InhA, the difluorophenyl group penetrates into the hydro-
phobic pocket of InhA. In the complex of 11 with InhA, the
conformation of the difluorophenyl group is restricted by the
amide linkage between the thiadiazole and the difluorophenyl
group. This orients the difluorophenyl group toward solvent
rather than toward the hydrophobic pocket. The active site
loop which is ordered only in subunits A and D has minimal
changes with respect to the conformation in complex with 7.
The most notable change is a rotation of the χ1 torsion angle in
I202, which is required to accommodate the phenyl ring. These
suboptimal interactions (i.e., loss of interaction with the
hydrophobic pocket) are reflected in the weaker binding
affinity of 11 for InhA vs that of compound 7. Compounds 12
and 13 are bioisosteric replacements of the pyrazole in 8, where
the pyrazole ring is replaced with basic pyridine and pyrimidine
rings, respectively. Both 12 and 13 showed good enzyme
potency, however, they showed weaker cellular potency as
Figure 6. Overlay of X-ray crystal structures of InhA in complex with 7
(green) (PDB ID 4bpq) and 11 (yellow) (PDB ID 4bpr). Compounds
7 and 11 are shown as sticks. Protein backbone is shown as a cartoon
with selected residues surrounding the substrate binding site shown as
sticks. NAD is shown as sticks with magenta carbon atoms. Hydrogen
bonds between the compounds and InhA and NAD are shown as
yellow dotted lines.
compared to 8. Substitution of a cyclopropyl group on the
pyrazole ring in 14 resulted in poor enzyme inhibition and thus
weaker cellular activity. Introduction of a methyl oxazolidinone
in place of the pyrazole ring in 15 retained enzyme inhibition
but lost cellular potency, possibly due to permeability or efflux
issues. Compound 16, where the pyrazole amine was replaced
with a pyrazole amide, lost both enzyme and cellular potency.
This may be due to the additional carbonyl group, which in
combination with the pyrazole and difluorophenyl rings
prevents an optimal conformation of the compound for
binding to InhA.
We also profiled compounds 10 and 11 in preliminary
DMPK (drug metabolism and pharmacokinetics) and safety
assays and compared them with 7 (Table 4). The solubility for
10 and 11 were lower than 7, which can be attributed to the
amide group in 11. The chiral isomers of 10 and 11 may
possess better solubility as observed for 7 in comparison with
its racemic analogue, 8 (data not shown). The plasma protein
binding (PPB) for 10 was much better with 14% free, while 7
and 11 were similar with values between 2% and 4%. The
clearance (CL) based on predicted percent liver blood flow
(LBF) for 7 and 10 was in the moderate to high range for
human microsomes and rat hepatocytes, respectively, however,
11 was much improved with clearance based on percent LBF of
4.5 and 7.1 for human microsomes and rat hepatocytes,
respectively. The permeability measured by using a Caco-2
assay suggested that these compounds are highly permeable
with no efflux. The CYP inhibition data showed significant
inhibition of CYP 3A4 for 7 (1.9 μM), while 10 and 11 showed
much improved profiles with CYP 3A4 inhibition 48 and >50
μM, respectively. The mammalian cytotoxicity (MMIC) against
the A549 cell line was much cleaner for 10 and 11 (>100 μM),
suggesting that compounds in the series are selective for
bacteria and inactive on eukaryotes, while 7 showed MMIC of
78 μM. Compounds 10 and 11 were tested in 98 off-target
assays were active in fewer than five targets at 30 μM (Table 4).
On the basis of the observed moderate to high clearance for 10
and moderate MIC for 11, these compounds were not tested
for in vivo pharmacokinetic and efficacy study. Further
medicinal chemistry efforts are required to optimize pharma-
F
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 4. Properties of 10 and 11 in Comparison with 7
7
10
11
InhA IC50 (μM)
0.003
0.195
78
0.03
0.2
Mtb MIC (μM)
<0.391
3.75
MMIC (μM)
>100
>100
aqueous solubility (μM)
human CL % LBFpred (microsomal)
rat CL % LBFpred (hepatocytes)
PPB (% free)
208
69
26
34.9
35.2
3.4
48
4.5
38
14.3
7.1
1.9
27/17
human Caco-2 A−B/B−A
Cyp3A4 IC50 (μM)
35/21
1.9
28/27
48
>50
a
b
secondary pharmacology hits
ND
AT, A1, and A3
A1, PDE4, 5HT2B
a
b
Additional information in Supporting Information Table S1 ND: not determined.
cokinetic properties and desirable cellular potency into a set of
molecules to enable in vivo efficacy testing.
to the NADH-bound form of the enzyme as opposed to the
NAD+ bound form of the enzyme which may well reflect the
charge complementarity between the site I group and NADH
(i.e., both neutral). Additionally, the current study indicates
that novel hydrophilic interactions with the protein at site III
lead to favorable physicochemical properties, resulting in
cellular activity. On the basis of this mechanism, we have
synthesized new analogues with potent enzyme inhibition,
attractive cellular potency, improved CYP inhibition, and safety
profile as compared to the known methyl-thiazole lead
molecule, 7. Furthermore, we believe that aspects of the
binding mode and molecular mode of inhibition of this series
that we have reported are transferable to other series, opening
up new avenues of medicinal chemistry against this important
and clinically validated TB target.
Having established the mechanism of inhibition using
biophysical studies, we embarked on confirming the molecular
target for methyl-thiazoles by generation of spontaneous
resistant mutants to the parent compound. Single-step mutants
arose at a frequency of 3 × 10−9 following exposure of Mtb cells
to increasing concentrations of 7 (Table 5). These resistant
Table 5. MIC Values for Selected Compounds against Wild-
Type and Resistant Strains of Mtb
mutation
compd
unit
wild-type
M103I
M103L
G96V
isoniazid
rifampicin
moxifloxacin
7
μg/mL
μg/mL
μg/mL
μM
0.03
0.007
0.12
0.39
1.56
0.06
0.007
0.12
0.03
0.007
0.12
0.06
0.01
0.12
>25
EXPERIMENTAL SECTION
■
>25
>25
All anhydrous solvents, reagent grade solvents for chromatography and
starting materials were purchased from either Sigma Aldrich Chemical
Co. or Fisher Scientific. Water was distilled and purified through a
Milli-Q water system (Millipore Corp., Bedford, MA). General
methods of purification of compounds involved the use of silica
cartridges purchased from Grace Purification Systems. The reactions
were monitored by TLC on precoated Merck 60 F254 silica gel plates
and visualized using UV light (254 nm). All compounds were analyzed
10
μM
>25
>25
>25
mutants displayed a greater than 64-fold increase in MIC to the
parent compound in comparison to the MIC against wild-type
Mtb H37Rv. There was no change in MIC to other standard
drugs acting through different mechanism of action, indicating
the specificity of this genetic mutation. Sequencing of the inhA
gene from these mutants revealed a single amino acid
substitution in each case resulting in three different mutations
G96V, M103L or M103I (Table 5). This mutation data
provides strong genetic evidence of target engagement by 7 and
highlights the genetic basis for resistance. In addition to the
correlation with biophysical data, the functional impact of the
mutations identified by mutant selection is neatly explained by
the crystal structure of 7 in complex with InhA (see Figure 3).
The mutation of G96 to any other residue, including valine,
would result in a clash with the linker region adjacent to the
methyl-thiazole warhead. As the thiadiazole, pyrazole, and
phenyl rings pack closely around M103, it is not surprising that
a change of this residue to a branched leucine or isoleucine
would alter the ability of the methyl-thiazole compounds to
bind to InhA.
1
for purity by HPLC and characterized by H NMR using Bruker 300
MHz NMR and/or Bruker 400 MHz NMR spectrometers. Chemical
shifts are reported in ppm (δ) relative to the residual solvent peak in
the corresponding spectra; chloroform δ 7.26, methanol δ 3.31,
DMSO-d6 δ 3.33, and coupling constants (J) are reported in hertz
(Hz) (where s = singlet, bs = broad singlet, d = doublet, dd = double
doublet, bd = broad doublet, ddd = double doublet of doublet, t =
triplet, tt = triple triplet, q = quartet, m = multiplet) and analyzed
using ACD NMR data processing software. Mass spectra values are
reported as m/z.
All reactions were conducted under nitrogen unless otherwise
noted. Solvents were removed in vacuo on a rotary evaporator. All final
compounds were purified by reverse phase HPLC or otherwise stated
with >95% purity.
Abbreviations: NMP = N-methyl pyrrolidine, HCl = hydrochloric
acid, DMF = N,N-dimethylformamide, DCM = dichloromethane,
NaH = sodium hydride, ES = electrospray ionization, HRMS = high
resolution mass spectrometry, RT = room temperature, RB = round-
bottom; h = hour.
(1S)-1-(5-{ [1-(2,6-Difluorobenzyl)-1H-pyrazol-3-yl]amino}-
1,3,4-thiadiazol-2-yl)-1-(4-methyl-1,3-thiazol-2-yl)ethanol (7).
In a 100 mL single-necked RB flask was taken 1-(2,6-difluoroben-
zyl)-3-isocyanato-1H-pyrazole(200 mg, 8.50 mmol) and 2-hydroxy-2-
(4-methylthiazol-2-yl)propanehydrazide (172 mg, 8.50 mmol).
Ethanol (10 mL) was added to get a clear solution. The reaction
mass was heated under reflux for 2 h. The desired compound was
CONCLUSION
■
In conclusion, we report the molecular mode of action of the
known Mtb InhA methyl-thiazoles. We have identified for the
first time a mechanism of InhA inhibition which shows a
hitherto unknown neutrally charged “warhead” being accom-
modated in the “Y158-out” conformation at site I of the InhA
protein. Notably, these compounds show preferential binding
G
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
confirmed by LCMS. Reaction mixture was evaporated to dryness. The
residue was cooled in ice-bath and added concd sulphuric acid (5 mL)
cautiously. The dark color reaction mixture was stirred at RT for 5 h.
After the completion of the reaction, the reaction mass was diluted
with ice water and neutralized with aqueous ammonia. The aqueous
phase was extracted with ethyl acetate and washed with water and
brine solution. The organic layer was separated, dried over sodium
sulfate, and evaporated at high vacuum to give the crude compound
which was purified by column chromatography using methanol/DCM.
This compound was subjected for chiral reverse phase HPLC.
methylthiazol-2-yl)propanehydrazide (287 mg, 1.43 mmol) and stirred
for 30 min at RT. The reaction mixture was concentrated under
vacuum, and the resultant crude compound was purified by column
chromatography to get N-(1-(2,6-difluorobenzyl)-1H-1,2,4-triazol-3-
yl)-2-(2-hydroxy-2-(4-methylthiazol-2-yl)propanoyl) hydrazine carbo-
thioamide (300 mg, 46.4%) as a solid. [ES + MS] m/z = 454 (M + 1).
In a 25 mL of RB flask, N-(1-(2,6-difluorobenzyl)-1H-1,2,4-triazol-3-
yl)-2-(2-hydroxy-2-(4-methylthiazol-2-yl)propanoyl)-
hydrazinecarbothioamide (300 mg, 0.66 mmol) was added followed by
concd sulfuric acid (2 mL) and stirred for 30 min. The reaction
mixture was neutralized carefully with aqueous ammonia solution at 0
°C. The reaction mixture was extracted with DCM, and the combined
organic layer was washed with water and brine solution, dried over
sodium sulfate, and concentrated under vacuum. The crude compound
was purified by column chromatography and then triturated with
diethyl ether to get 1-(5-(1-(2,6-difluorobenzyl)-1H-1,2,4-triazol-3-
ylamino)-1,3,4-thiadiazol-2-yl)-1-(4-methylthiazol-2-yl)ethanol (50.0
mg, 17.36%) as a solid. 1H NMR (300 MHz, DMSO-d6) δ ppm
11.56 (br s, 1H), 8.55 (s, 1H), 7.43−7.55 (m, 1H), 7.36 (s, 1H), 7.23
(d, J = 0.9 Hz, 1H), 7.15 (t, J = 8.0 Hz, 2H), 5.39 (s, 2H), 2.29 (s,
3H), 1.98 (s, 3H). [ES + MS] m/z = 436 (M + 1). HRMS calculated
for C17H15F2N7OS2, 436.082; found, 436.08.
Analytical chiral conditions: CHIRALPAK-IA; dimension, 30 mm ×
250 mm; 5 μ; mobile phase A, hexane; mobile phase B, 1:1
methanol:2-propanol; additive 0.1% diethylamine; flow rate 40 mL/
min; temperature 40 °C; outlet pressure 100 bar; detection 280 nm;
loading 100 mg/injn; concentration 50 mg/mL; diluent 4:1 methanol/
1
THF. Optical rotation +118. H NMR (300 MHz, DMSO-d6) δ ppm
10.87 (br s, 1 H), 7.71 (d, J = 2.26 Hz, 1 H), 7.35−7.57 (m, 1 H), 7.29
(s, 1 H), 7.24 (d, J = 0.94 Hz, 1 H), 7.11 (t, J = 7.91 Hz, 2 H), 5.94 (d,
J = 2.26 Hz, 1 H), 5.27 (s, 2 H), 2.24−2.39 (m, 3 H), 1.98 (s, 3 H),
1.04 (d, J = 6.03 Hz, 1 H). [ES + MS] m/z 436.1 (M + 1). HRMS
calculated for C18H16F2N6OS2, 435.0879; found, 435.0875.
1-(5-(1-(2,6-Difluorobenzyl)-1H-pyrazol-3-ylamino)-1,3,4-
thiadiazol-2-yl)-1-(4-methylthiazol-2-yl)ethanol (8). In a 100 mL
single-necked RB flask was taken 1-(2,6-difluorobenzyl)-3-isocyanato-
1H-pyrazole (200 mg, 8.50 mmol) and 2-hydroxy-2-(4-methylthiazol-
2-yl)propanehydrazide (172 mg, 8.50 mmol). Ethanol (10 mL) was
added to get a clear solution. The reaction mass was heated under
reflux for 2 h. The desired compound was confirmed by LCMS.
Reaction mixture was evaporated to dryness. The residue was cooled
in ice-bath and concd sulphuric acid (5 mL) was added cautiously. The
dark color reaction mixture was stirred at RT for 5 h. After the
completion of the reaction, the reaction mass was diluted with ice
water and neutralized with aqueous ammonia. The aqueous phase was
extracted with ethyl acetate and washed with water and brine solution.
The organic layer was separated, dried over sodium sulfate, and
evaporated at high vacuum to give the crude compound which was
2-(2,6-Difluorophenyl)-N-(5-(1-hydroxy-1-(4-methylthiazol-
2-yl)ethyl)-1,3,4-thiadiazol-2-yl)acetamide(11). In a 10 mL RB
flask, 2-(2,6-difluorophenyl)acetic acid (300 mg, 1.74 mmol) was
heated with thionyl chloride (1.5 mL, 17.43 mmol) at 60 °C for 30
min. Then the reaction mixture was evaporated under vacuum. The
resultant crude compound was dissolved in acetone (5 mL) and
potassium thiocyanate (186 mg, 1.92 mmol) was added. The reaction
mixture was stirred for 2 h. Then 2-hydroxy-2-(4-methylthiazol-2-
yl)propanehydrazide (351 mg, 1.74 mmol) was added and stirred for
another 30 min at RT. The formation of desired product was
confirmed by LCMS. The reaction mixture was concentrated under
vacuum to get crude 2-(2,6-difluorophenyl)-N-(2-(2-hydroxy-2-(4-
methylthiazol-2-yl)propanoyl)hydrazinecarbonothioyl)acetamide (210
mg, 29.1%) as a gum. [ES + MS] m/z = 415 (M + 1). In a 25 mL RB
flask, 2-(2,6-difluorophenyl)-N-(2-(2-hydroxy-2-(4-methylthiazol-2-
yl)propanoyl)hydrazinecarbonothioyl)acetamide (200 mg, 0.48
mmol) was added followed by concd sulfuric acid (2 mL) and stirred
for 30 min. The reaction mixture was neutralized carefully with
aqueous ammonia solution at 0 °C. The reaction mixture was extracted
with DCM, and the combined organic layer was washed with water
and brine solution, dried over sodium sulfate, and concentrated under
vacuum. The crude compound was purified by column chromatog-
raphy and then triturated with diethyl ether to get 2-(2,6-
difluorophenyl)-N-(5-(1-hydroxy-1-(4-methylthiazol-2-yl)ethyl)-1,3,4-
1
purified by column chromatography using methanol/DCM. H NMR
(300 MHz, DMSO-d6) δ ppm 10.88 (br s, 1 H), 7.71 (d, J = 2.07 Hz, 1
H),7.36−7.55 (m, 1 H), 5.26 (s, 2 H), 7.29 (s, 1 H), 7.24 (s, 1 H),
7.11 (t, J = 7.91 Hz, 2 H), 5.93 (d, J = 2.07 Hz, 1 H), 2.30 (s, 3 H),
1.98 (s, 3 H). [ES + MS] m/z 436.1 (M + H). HRMS calculated for
C18H16F2N6OS2, 435.0879; found, 435.0875.
1-(5-(1-(2,6-Difluorobenzyl)-1H-pyrazol-4-ylamino)-1,3,4-
thiadiazol-2-yl)-1-(4-methylthiazol-2-yl)ethanol (9). To a sol-
ution of 1-(2,6-difluorobenzyl)-4-isothiocyanato-1H-pyrazole (450 mg,
1.79 mmol) in DCM (10 mL) was added 2-hydroxy-2-(4-
methylthiazol-2-yl)propanehydrazide (360 mg, 1.79 mmol) and stirred
for 30 min at RT. The colorless solid was filtrated under vacuum and
dried to get N-(1-(2,6-difluorobenzyl)-1H-pyrazol-4-yl)-2-(2-hydroxy-
2-(4-methylthiazol-2-yl)propanoyl)hydrazinecarbothioamide (450 mg,
55.5%) as a solid. [ES + MS] m/z = 453 (M + 1). In a 10 mL of RB
flask, N-(1-(2,6-difluorobenzyl)-1H-pyrazol-4-yl)-2-(2-hydroxy-2-(4-
methylthiazol-2-yl)propanoyl)hydrazinecarbothioamide (200 mg,
0.44 mmol) was added followed by concd sulfuric acid (2 mL) and
stirred for 30 min. The reaction mixture was neutralized carefully with
aqueous ammonia solution at 0 °C. The reaction mixture was extracted
with DCM, and the combined organic layer was washed with water
and brine solution, dried over sodium sulfate, and concentrated under
vacuum. The crude compound was purified by column chromatog-
raphy to get 1-(5-(1-(2,6-difluorobenzyl)-1H-pyrazol-4-ylamino)-
1,3,4-thiadiazol-2-yl)-1-(4-methylthiazol-2-yl)ethanol (40.0 mg,
1
thiadiazol-2-yl)acetamide (90 mg, 47.0%) as a solid. H NMR (300
MHz, DMSO-d6) δ ppm 12.91 (s, 1H), 7.50 (s, 1H), 7.35−7.47 (m,
1H), 7.23 (d, J = 0.9 Hz, 1H), 7.06−7.18 (m, 2H), 3.91 (s, 2H), 2.30
(s, 3H), 2.02 (s, 3H). [ES + MS] m/z = 397 (M + 1). HRMS
calculated for C16H14F2N4O2S2, 397.0598; found, 397.0605.
1-(5-(6-(2,6-Difluorobenzyl)pyridin-2-ylamino)-1,3,4-thiadia-
zol-2-yl)-1-(4-methylthiazol-2-yl)ethanol (12). In a 100 mL
single-necked flask equipped with an air condenser connected to
nitrogen source was taken 2-(2,6-difluorobenzyl)-6-isothiocynatopyr-
idine (200 mg, 0.763 mmol). DCM (20 mL) was added to get a clear
solution. Then 2-hydroxy-2-(4-methylthiazol-2-yl)propanehydrazide
(180 mg, 0.90 mmol) was added and stirred at RT for 3 h. After
the completion of the reaction, the volatiles were evaporated to get the
solid, which was taken in concd sulfuric acid (10 mL) and stirred at
RT for 1 h. After completion of the reaction, it was neutralized with
aqueous ammonia solution. The aqueous layer was extracted with
ethyl acetate and washed with water and brine solution. The organic
layer was separated, dried over sodium sulfate, and evaporated to get
the crude compound which was purified by column chromatography
using methanol/DCM to get 100 mg (27.63%) of the final compound.
1H NMR (300 MHz, DMSO-d6) δ ppm 11.50 (s, 1 H), 7.60−7.70 (m,
1
20.83%) as a solid. H NMR (300 MHz, DMSO-d6) δ ppm 10.06
(s, 1H), 7.99 (s, 1H), 7.41−7.54 (m, 1H), 7.38 (s, 1H), 7.21 (d, J = 0.9
Hz, 1H), 7.15 (t, J = 8.0 Hz, 2H), 5.35 (s, 2H), 2.30 (s, 3H), 1.96 (s,
3H). [ES + MS] m/z = 435 (M + 1). HRMS calculated for
C18H16F2N6OS2 435.0867; found: 435.086.
1-(5-(1-(2,6-Difluorobenzyl)-1H-1,2,4-triazol-3-ylamino)-
1,3,4-thiadiazol-2-yl)-1-(4-methylthiazol-2-yl)ethanol (10). To a
solution of 1-(2,6-difluorobenzyl)-3-isothiocyanato-1H-1,2,4-triazole
(360 mg, 1.43 mmol) in DCM (10 mL) was added 2-hydroxy-2-(4-
1 H), 7.25−7.30 (m, 1H), 7.15−7.20 (m, 2H), 6.95−7.05 (m, 2H),
6.80−6.90 (m, 2H), 4.00 (s, 2H), 2.20 (s, 3H), 1.95 (s, 3H). [ES +
H
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
MS] m/z = 446 (M + 1). HRMS calculated for C20H17F2N5OS2,
446.0915; found, 446.09.
H), 2.31 (s, 3 H), 1.93 (s, 3 H). [ES + MS] m/z = 468 (M + 1).
HRMS calculated for C19H19F2N5O3S2, 468.097; found, 468.0973.
1-(2,6-Difluorobenzyl)-N-(5-(1-hydroxy-1-(4-methylthiazol-
2-yl)ethyl)-1,3,4-thiadiazol-2-yl)-1H-pyrazole-3-carboxamide
(16). To 1-(2,6-difluorobenzyl)-1H-pyrazole-3-carbonyl isothiocyanate
(100 mg, 0.36 mmol) and 2-hydroxy-2-(4-methylthiazol-2-yl)-
propanehydrazide (72 mg, 0.36 mmol) was added DCM (5 mL)
and then stirred for 30 min. The reaction mixture was concentrated
under vacuum to get 1-(2,6-difluorobenzyl)-N-(2-(2-hydroxy-2-(4-
methylthiazol-2-yl)propanoyl)hydrazinecarbonothioyl)-1H-pyrazole-3-
carboxamide (170 mg, 99%) as a solid. [ES + MS] m/z = 481 (M + 1).
In a 25 mL of RB flask, 1-(2,6-difluorobenzyl)-N-(2-(2-hydroxy-2-(4-
methylthiazol-2-yl)propanoyl)hydrazinecarbonothioyl)-1H-pyrazole-3-
carboxamide (310 mg, 0.65 mmol) was added followed by concd
sulfuric acid (2 mL) and stirred for 30 min. The reaction mixture was
neutralized carefully with aqueous ammonia solution at 0 °C. The
reaction mixture was extracted with DCM, and the combined organic
layer was washed with water and brine solution, dried over sodium
sulfate, and concentrated. The crude was purified by column
chromatography and then triturated with methanol:diethyl ether
mixture to get 1-(2,6-difluorobenzyl)-N-(5-(1-hydroxy-1-(4-methyl-
thiazol-2-yl)ethyl)-1,3,4-thiadiazol-2-yl)-1H-pyrazole-3-carboxamide
1-(5-(4-(2,6-Difluorobenzyl)pyrimidin-2-ylamino)-1,3,4-thia-
diazol-2-yl)-1-(4-methyl thiazol-2-yl)ethanol (13). 4-(2,6-Difluor-
obenzyl)-2-isocyanatopyrimidine (200 mg, 0.809 mmol) was taken in
a 100 mL single-necked flask equipped with an air condenser
connected to nitrogen source. DCM (10 mL) was added to get a clear
solution, followed by 2-hydroxy-2-(4-methylthiazol-2-yl)-
propanehydrazide (162.9 mg, 0.809 mmol). The reaction mass was
stirred at RT for 2 h. After completion of the reaction, the volatiles
were evaporated to get the residue N-(4-(2,6-difluorobenzyl)-
pyrimidin-2-yl)-2-(2-hydroxy-2-(4-methylthiazol-2-yl)propanoyl) hy-
drazine carbothioamide. The above residue (300 mg, 0.64 mmol)
was taken in concd sulfuric acid (5 mL) and stirred at RT for 2 h. After
the completion of the reaction, it was neutralized with aqueous
ammonia and the aqueous was extracted with ethyl acetate and washed
with water and brine solution. The organic layer was dried over
sodium sulfate and evaporated to get the crude compound, which was
purified by column chromatography using methanol/DCM as eluent
1
to get 110 mg (30.45%) of the desired compound. H NMR (300
MHz, DMSO-d6) δ ppm 12.00 (s, 1H), 8.50(s, 1H), 7.30−7.45 (m, 2
H), 7.25 (s, 1 H), 7.05−7.15 (m, 2 H), 6.90−7.00 (m, 1 H), 4.10 (s,
2H), 2.30 (s, 3 H), 2.04 (s, 3 H). [ES + MS] m/z = 447 (M + 1).
HRMS calculated for C19H16F2N6OS2, 447.0867; found, 447.0887.
1-(5-((5-Cyclopropyl-1-(2,6-difluorobenzyl)-1H-pyrazol-3-
yl)amino)-1,3,4-thiadiazol-2-yl)-1-(4-methylthiazol-2-yl)-
ethanol (14). 5-Cyclopropyl-1-(2,6-difluorobenzyl)-3-isothiocyanato-
1H-pyrazole (200 mg, 0.69 mmol) was taken in a 100 mL single-
necked flask equipped with an air condenser connected to nitrogen
source. DCM (10 mL) was added to get a clear solution followed by 2-
hydroxy-2-(4-methylthiazol-2-yl)propanehydrazide (138 mg, 0.69
mmol). The reaction mass was stirred at RT for 2 h. After completion
of the reaction, the volatiles were evaporated to get the residue (N-(5-
cyclopropyl-1-(2,6-difluorobenzyl)-1H-pyrazol-3-yl)-2-(2-hydroxy-2-
(4-methylthiazol-2-yl)propanoyl)hydrazinecarbothioamide). The
above residue (300 mg, 0.61 mmol) was taken in concd sulfuric acid
(5 mL) and stirred at RT for 2 h. After completion of the reaction, it
was neutralized, solid obtained, filtered, washed with water, and dried.
The compound was purified by column chromatography using
1
(150 mg, 50.3%) as a solid. H NMR (300 MHz, DMSO-d6) δ ppm
7.98 (d, J = 2.3 Hz, 1H), 7.44−7.56 (m, 2H), 7.12−7.25 (m, 3H), 7.05
(d, J = 2.5 Hz, 1H), 5.51 (s, 2H), 2.30 (s, 3H), 2.02 (s, 3H). [ES +
MS] m/z = 463 (M + 1). HRMS calculated for C19H16F2N6O2S2,
463.08169; found, 463.0803
ASSOCIATED CONTENT
* Supporting Information
■
S
Details of the synthesis of all compounds, details of structure
determination, and details of biological assays. This material is
available free of charge via the Internet at http://pubs.acs.org.
Accession Codes
The atomic coordinates and structure factors for InhA in
complex with 7 and 11 have been deposited in the Protein Data
Bank (pdb accession codes 4bpq and 4bpr, respectively) along
with structure factors and detailed experimental statistics.
1
methanol/DCM as eluent. H NMR (300 MHz, DMSO-d6) δ ppm
AUTHOR INFORMATION
Corresponding Authors
*Phone: +91 080 23621212. Fax: +91 080 23621214. E-mail:
pravin.shirude@astrazeneca.com.
7.31−7.54 (m, 2 H), 7.24 (s, 1 H), 7.06 (t, J = 7.91 Hz, 2 H), 5.12−
5.20 (m, 2 H), 2.32 (s, 3 H), 1.86−2.04 (m, 3 H), 1.72 (td, J = 8.76,
4.14 Hz, 1 H), 0.69−0.85 (m, 2 H), 0.44−0.60 (m, 2 H). [ES + MS]
m/z = 475 (M + 1). HRMS calculated for C21H20F2N6OS2, 475.118;
found, 475.1181.
■
*Phone: +44 1625 040376. E-mail: jon.read@astrazeneca.com.
3-(2,6-Difluorobenzyl)-5-((5-(1-hydroxy-1-(4-methylthiazol-
2-yl)ethyl)-1,3,4-thiadiazol-2-ylamino)methyl)oxazolidin-2-
one (15). In a 50 mL round-bottomed flask, 3-(2,6-difluorobenzyl)-5-
(isothiocyanatomethyl)oxazolidin-2-one (100 mg, 0.35 mmol) and 2-
hydroxy-2-(4-methylthiazol-2-yl)propanehydrazide (70.8 mg, 0.35
mmol) were taken in DCM (8 mL) under nitrogen. The resulting
clear solution was stirred for 50 min. After completion of the reaction,
the volatiles were evaporated to dryness to get N-((3-(2,6-
difluorobenzyl)-2-oxooxazolidin-5-yl)methyl)-2-(2-hydroxy-2-(4-
methylthiazol-2-yl)propanoyl)hydrazinecarbothioamide (110 mg,
64.4%) as a solid. This solid was taken in concd H2SO4 (5 mL) and
stirred at RT for 3 h. While stirring, the suspension turned to clear
solution. LCMS showed the required mass after 25 min. Reaction was
neutralized slowly with aqueous ammonia at 0 °C. The solution was
diluted with water and extracted with DCM. The organic layer was
separated and washed with water and brine solution. The organic layer
was separated, dried over sodium sulfate, and concentrated. The crude
compound was purified by reverse phase purification system to get
product 3-(2,6-difluorobenzyl)-5-((5-(1-hydroxy-1-(4-methylthiazol-2-
yl)ethyl)-1,3,4-thiadiazol-2-ylamino)methyl)oxazolidin-2-one (140 mg,
72.7%) as a solid. 1H NMR (300 MHz, DMSO-d6) δ ppm 7.94 (t, J =
5.75 Hz, 1 H), 7.35−7.51 (m, 1 H), 7.28 (s, 1 H), 7.21 (dd, J = 4.05,
1.04 Hz, 1 H), 7.11 (q, J = 8.29 Hz, 2 H), 4.67 (dd, J = 8.10, 5.65 Hz, 1
H), 4.42 (d, J = 7.72 Hz, 2 H), 3.42−3.57 (m, 3 H), 3.14−3.24 (m, 1
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the analytical support provided by S. Rudrapatna and
analytical team. Our sincere thanks to the Microbiology, Target
Biology, DMPK, and Safety team for technical support in
various assays. We would like to thank our colleagues in the
screening, protein supply, and characterization groups in
Discovery Sciences, AstraZeneca. We also thank V. Balasu-
bramanian, R. Tommasi, M. Perros, V. Hosagrahara, P. Iyer,
and S. Narayanan for their constant inspiration, support, and
encouragement.
ABBREVIATIONS USED
■
TB, tuberculosis; Mtb, Mycobacterium tuberculosis; ACPER, acyl
carrier protein enoyl reductase; ACP, acyl carrier protein; INH,
isoniazid; inhA, structural gene of acyl carrier protein enoyl
reductase in Mycobacterium tuberculosis; InhA, Mycobacterium
tuberculosis acyl carrier protein enoyl reductase; KatG,
Mycobacterium tuberculosis catalase-peroxidase; NAD, nicotina-
I
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
mide adenine dinucleotide; NAD+, oxidized nicotinamide
adenine dinucleotide; NADH, reduced nicotinamide adenine
dinucleotide; MIC, minimal inhibitory concentration; CYP,
cytochrome P450; ITC, iso-thermal titration calorimetry; SPR,
surface plasmon resonance; SAR, structure−activity relation-
ships; DMPK, drug metabolism and pharmacokinetics; PPB,
plasma protein binding; CL, clearance; LBF, liver blood flow;
MMIC, mammalian cytotoxicity
́
(17) Ledz, P. S.; Sledz, P.; Lang, S.; Stubbs, C. J.; Abell, C. High-
́
́
Throughput Interrogation of Ligand Binding Mode Using a
Fluorescence-Based Assay. Angew. Chem., Int. Ed. 2012, 51, 7680−
7683.
(18) Parikh, S. L.; 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.
̀ ́
(19) Vilcheze, C.; Weisbrod, T. R.; Chen, B.; Kremer, L.; Hazbon, M.
H.; Wang, F.; Alland, D.; Sacchettini, J. C.; Jacobs, W. R., Jr. Altered
NADH/NAD+ ratio mediates coresistance to isoniazid and ethiona-
mide in mycobacteria. Antimicrob. Agents Chemother. 2005, 49, 708−
720.
(20) 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.
(21) Rozwarski, D. A.; Vilcheze, 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.
(22) 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.
(23) Sullivan, T. J.; Truglio, J. J.; Boyne, M. E.; Novichenok, P.;
Zhang, X.; Stratton, C.; Li, H. J.; Kaur, T.; Amin, A.; Johnson, F.;
Slayden, R. A.; Kisker, C.; Tonge, P. J. High Affinity InhA Inhibitors
with Activity against Drug-Resistant Strains of Mycobacterium
tuberculosis. ACS Chem. Biol. 2006, 1, 43−53.
(24) Chang, A.; Schiebel, J.; Yu, W.; Bommineni, G. R.; Pan, P.;
Baxter, M. V.; Khanna, A.; Sotriffer, C. A.; Kisker, C.; Tonge, P. J.
Rational Optimization of Drug-Target Residence Time: Insights from
Inhibitor Binding to the Staphylococcus aureus FabI Enzyme-Product
Complex. Biochemistry 2013, 52, 4217−4228.
REFERENCES
■
(1) Global Tuberculosis Control: Surveillance, Planning, Financing;
World Health Organization: Geneva, Switzerland, 2011.
(2) Lu, H.; Tonge, P. J. Inhibitors of FabI, an Enzyme Drug Target in
the Bacterial Fatty Acid Biosynthesis Pathway. Acc. Chem. Res. 2008,
41, 11−20.
(3) McDevitt, D.; Rosenberg, M. Exploiting genomics to discover
new antibiotics. Trends Microbiol. 2001, 9, 611−617.
(4) Payne, D. J.; Warren, P. V.; Holmes, D. J.; Ji, Y.; Lonsdale, J. T.
Bacterial fatty-acid biosynthesis: A genomics-driven target for
antibacterial drug discovery. Drug Discovery Today 2001, 6, 537−544.
(5) Lamichhane, G. Novel targets in M. tuberculosis: search for new
drugs. Trends Mol. Med. 2011, 17, 25−33.
(6) Rawat, R.; Whitty, A.; Tonge, P. J. The isoniazid-NAD adduct is a
slow, tight-binding inhibitor of InhA, the Mycobacterium tuberculosis
enoyl reductase: adduct affinity and drug resistance. Proc. Natl. Acad.
Sci. U. S. A. 2003, 100, 13881−13886.
(7) Vilcheze, C.; Jacobs, W. R. The Mechanism of Isoniazid Killing:
Clarity Through the Scope of Genetics. Annu. Rev. Microbiol. 2007, 61,
35−50.
(8) Banerjee, A.; Dubnau, E.; Quemard, A.; Balasubramanian, V.;
Um, K. S.; Wilson, T.; Collins, D.; de Lisle, G.; Jacobs, W. R., Jr. inhA,
a gene encoding a target for isoniazid and ethionamide in
Mycobacterium tuberculosis. Science 1994, 14, 227−230.
(25) Dessen, A.; Quemard, A.; Blanchard, J. S.; Jacobs, W. R., Jr.;
Sacchettini, J. C. Crystal structure and function of the isoniazid target
of Mycobacterium tuberculosis. Science 1995, 267, 1638−1641.
(26) Schiebel, J.; Chang, A.; Lu, H.; Baxter, M. V.; Tonge, P. J.;
Kisker, C. Staphylococcus aureus FabI: Inhibition, Substrate Recog-
nition and Potential Implications for In Vivo Essentiality. Structure
2012, 9, 802−813.
(9) Davis, M. C.; Franzblau, S. G.; Martin, A. R. Syntheses and
evaluation of benzodiazaborine compounds against M. tuberculosis
H37Rv in vitro. Bioorg. Med. Chem. Lett. 1998, 8, 843−846.
(10) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv. Drug
Delivery Rev. 2001, 46, 3−26.
(11) Hartkoorn, R. C.; Sala, C.; Neres, J.; Pojer, F.; Magnet, S.;
Mukherjee, R.; Uplekar, S.; Boy-Rottger, S.; Altmann, K. H.; Cole, S.
̈
T. Towards a new tuberculosis drug: pyridomycinnature’s isoniazid.
EMBO Mol. Med. 2012, 4, 1032−1042.
(12) Rafferty, J. B.; Simon, J. W.; Baldock, C.; Artymiuk, P. J.; Baker,
P. J.; Stuitje, A. R.; Slabas, A. R.; Rice, D. W. Common themes in redox
chemistry emerge from the X-ray structure of oilseed rape (Brassica
napus) enoyl acyl carrier protein reductase. Structure 1995, 3, 927−
938.
(13) Kuo, M. R.; Morbidoni, H. R.; Alland, D.; Sneddon, S. F.;
Gourlie, B. B.; Staveski, M. M.; Leonard, M.; Gregory, J. S.; Janjigian,
A. D.; Yee, C.; Musser, J. M.; Kreiswirth, B.; Iwamoto, H.; Perozzo, R.;
Jacobs, W. R.; Sacchettini, J. C.; Fidock, D. A. Targeting tuberculosis
and malaria through inhibition of enoyl reductase: compound activity
and structural data. J. Biol. Chem. 2003, 278, 20851−20859.
(14) 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.
(15) Ballell Pages, L.; Castro Pichel, J.; Fernandez Menendez, R.;
Fernandez Velando, E. P.; Gonzalez Del Valle, S.; Mendoza Losana,
A.; Wolfendale, M. J. (Pyrazol-3-yl)-1,3,4-thiadiazole-2-amine and
(Pyrazol-3-yl)-1,3,4-thiazole-2-amine Compounds. PCT publication
No. WO 2010/118852 A1, 2010.
(16) Castro Pichel, J.; Fernandez Menendez, R.; Fernandez Velando,
E. P.; Gonzalez Del Valle, S.; Mallo-Rubio, A. 3-Amino-Pyrazole
Derivatives Useful Against Tuberculosis. PCT publication No. WO
2012/049161 A1, 2012.
J
dx.doi.org/10.1021/jm4012033 | J. Med. Chem. XXXX, XXX, XXX−XXX