Evaluation of heteroatom-rich derivatives as antitubercular agents with InhA inhibition properties

Article abstract of DOI:10.1007/s00044-017-2064-x

Two series of heterocyclic compounds derived from 3-acetyl-4-hydroxy-6-methyl-2H-pyran-2-one (DHA) and 2-acetylbutyrolactone have been synthesized and characterized. The compounds were evaluated for their activities against Mycobacterium tuberculosis stra

Full text of DOI:10.1007/s00044-017-2064-x

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-017-2064-x
ORIGINAL RESEARCH
Evaluation of heteroatom-rich derivatives as antitubercular agents
with InhA inhibition properties
Bachar Rébat Moulkrere1,2 Beatrice S. Orena3 Giorgia Mori3
●
●
●
Nathalie Saffon-Merceron4 Frédéric Rodriguez5,6 Christian Lherbet7
●
●
●
Nadji Belkheiri1 Mohamed Amari1,8 Pascal Hoffmann5,6 Mokhtar Fodili1
●
●
●
Received: 21 April 2017 / Accepted: 5 September 2017
© Springer Science+Business Media, LLC 2017
Abstract Two series of heterocyclic compounds derived
from 3-acetyl-4-hydroxy-6-methyl-2H-pyran-2-one (DHA)
and 2-acetylbutyrolactone have been synthesized and
characterized. The compounds were evaluated for their
activities against Mycobacterium tuberculosis strain, and as
inhibitors of InhA, a key enzyme involved in the type II
fatty acid biosynthesis pathway of M. tuberculosis. Among
the tested compounds, one DHA derivative, compound 2,
showed promising activity against both mycobacteria and
InhA. Docking studies were also carried out and give some
new structure-activity trends compatible with current
structural knowledge.
●
●
●
Keywords Tuberculosis Enoyl-ACP reductase InhA
●
●
Pyran-2-one Pyrazole Molecular docking
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s00044-017-2064-x) contains supplementary
material, which is available to authorized users.
Introduction
* Pascal Hoffmann
pascal.hoffmann@univ-tlse3.fr
Tuberculosis (TB), an infectious disease caused by the
bacillus Mycobacterium tuberculosis (M.tb), remains a
major global health problem (WHO 2016). According to the
World Health Organization (WHO), in 2015, it was esti-
mated that 10.4 million people were affected by TB and 1.4
million died from the disease, with an additional 0.4 million
deaths among HIV-positive people (WHO 2016). M.
tuberculosis multidrug resistant (MDR) (Bemer-Melchior
et al. 2000; Jain and Mondal 2008) and extensively drug
resistant MT (XDR) (Jain and Mondal 2008) strains are one
of the major challenge to TB treatment, with resistances to
antibiotics as isoniazid, rifampicin (for MDR-TB) and
fluoroquinolone kanamicyn, amikacin, and capreomycin
(for XDR-TB). Under these circumstances, there is an
urgent need of new drugs to combat the spread of this
disease (Fig. 1).
* Mokhtar Fodili
fodilimokhtar@hotmail.com
1
Université Ziane Achour, Laboratoire de Chimie Organique &
Substances Naturelles Djelfa, Djelfa, Algeria
2
Faculté des Sciences Exactes, des Sciences de la nature et de la vie
Université Mohamed Khider, Biskra, Algeria
3
Department of Biology and Biotechnology Lazzaro Spallanzani
University of Pavia, Pavia, Italy
4
Institut de Chimie de Toulouse, ICT FR 2599, Université de
Toulouse 118 Route de Narbonne, 31062 Toulouse cedex 9,
Toulouse, France
5
Center National de la Recherche Scientifique (CNRS) 118 Route
de Narbonne, 31062 Toulouse cedex 09, Toulouse, France
6
Laboratoire de Synthèse et Physicochimie de Molécules d’Intérêt
Biologique (SPCMIB) Université Paul Sabatier Toulouse 3, 118
Route de Narbonne, 31062 Toulouse Cedex 09, Toulouse, France
Isoniazid (INH) is one of the main front-line drug to fight
M.tb growth, which inhibits enoyl ACP reductase protein
called InhA, an enzyme involved in the synthesis of the
mycobacterial cell wall (Rozwarski et al. 1998). Isoniazid is
a prodrug activated by KatG, a catalase-peroxidase enzyme;
7
ITAV USR 3505 Centre Pierre Potier 1 Place Pierre Potier, 31106
Toulouse cedex 1, Toulouse, France
8
Faculté de Chimie—USTHB—BP32, El-Alia 16111 Bab Ezzouar,
Algiers, Algeria

Med Chem Res
Fig. 1 Examples of potential antitubercular drugs bearing DHA or
pyrazole moieties
the encoding gene is the most frequently mutated in case of
INH resistance. To work around this problem of resistance,
several groups have sought to directly inhibit InhA without
this preactivation step. Several families of inhibitors with
different scaffolds, i.e., diaryl ethers (Sullivan et al 2006),
pyrrolidine carboxamides (Lu et al. 2009), piperazine
indoleformamides (Kuo et al. 2003; Chollet et al. 2015),
arylamides (He et al. 2007), methyl thiazoles (Shirude et al.
̌
2013), rhodanine (Seefeld et al. 2003), thiadiazoles (Sink
et al. 2015), were discovered and showed good potencies
against InhA. Recently, a new class of direct InhA inhibi-
tors, the 4-hydroxy-2-pyridones was identified using phe-
notypic high-throughput whole-cell screening. The lead
compound, NITD-916 (compound A), showed good inhi-
bition of InhA and good efficacy in vivo in mouse models of
M. tuberculosis infection (Manjunatha et al. 2015).
Dehydroacetic acid (3-acetyl-4-hydroxy-6-methyl-2H-
pyran-2-one, DHA), that has similar structural parameters to
NITD-916, is commonly used as a preservative, or as
antimicrobial (Kunigoshi 1960, compound B) or antifungal
agents in cosmetic and pharmaceutical products, and it is
also used as a versatile starting molecule for the synthesis of
numerous pharmacophores with a variety of biological
activities (Gupta et al. 2014). For example, a number of
DHA derivatives have been readily prepared by its con-
version into other heterocyclic systems, such as benzodia-
(or thia)-zepines or pyrimidines through a condensation
with a variety of nucleophiles like o-phenylenediamine
derivatives (Hammal et al. 2009). Pyrazoles, such as com-
pounds C and D, are also of interest and are already known
as potential antitubercular drugs (Manetti et al. 2006; Cas-
tagnolo et al. 2009; Ntie-Kang et al. 2014).
Fig. 2 Structures of compounds synthesized from DHA (1–7) or from
2-acetylbutyrolactone (8–10) evaluated for their activity against
Mycobacterium tuberculosis
starting compounds (Fig. 2) were prepared and evaluated
for their activity against the InhA enzyme and M. tuber-
culosis. These heteroatom-rich derivatives could interact
easily with key residues in the binding site of InhA, like
Tyr158 (Chollet et al. 2015). Compound 2, prepared from
condensation of dehydroacetic acid and 4-chlor-
obenzaldehyde, was further subjected to docking studies to
investigate and assess possible binding modes of this ligand
in the binding cavity of InhA enzyme.
Material and methods
Chemistry
All chemicals were obtained from Aldrich or Acros
Organics and used without further purification. Nuclear
magnetic resonance spectra were recorded on a Bruker AC
300 spectrometer (1H, 13C NMR; solvent residue signals
were used for calibration of spectral data). Mass
In this study, a series of DHA derivatives or derivatives
prepared either from DHA or from 2-acetylbutyrolactone as

Med Chem Res
spectrometry (MS) data were obtained on a ThermoQuest
TSQ 7000 spectrometer, high-resolution mass spectra
(HRMS) were recorded on a ThermoFinnigan MAT 95 XL
spectrometer using chemical ionization (CI; CH4 or NH3)
and electrospray ionization (ESI) methods. Melting points
were measured on a Mettler Toledo MP50 melting point
system. IUPAC names of the compounds were generated
using Accelrys Draw 4.2.
g) in ethyl acetate (7.5 mL) at 5 bars hydrogen-pressure in a
stainless steel reactor at room temperature for 24 h. The
catalyst was then filtrated, and the solvent was evaporated to
dryness. The oily residue was then purified by column
chromatography on silica gel using dichloromethane as
eluent (Yield, 90%). Compound 3 was obtained by con-
densation of reduced-DHA compound 3′ (1.70 g, 0.01 mol)
and N,N-dimethylformamide dimethylacetal (1.78 g, 0.015
mol) in the presence of a catalytic amount of triethylamine
(around 200 µL) in ethanol (30 mL). The reaction was
heated to reflux under magnetic stirring for 6 h. After
cooling, the solvent was evaporated to dryness under
reduced pressure, and the oily residue was triturated in ethyl
ether to give a white precipitate, which was filtered to give
4-Hydroxy-6-methyl-3-[(Z)-3-phenylprop-2-enoyl]pyran-2-
one (1)
Compound 1 was prepared by condensation of an equimolar
solution of dehydroacetic acid (DHA, 1.68 g, 0.01 mol) and
benzaldehyde (1.06 g, 0.01 mol) in the presence of piper-
idine (0.28 g, 0.0033 mol) in chloroform (30 mL). The
reaction mixture is heated to reflux for 4 h. After cooling,
the solvent was evaporated to dryness under reduced pres-
sure, and the oily residue was triturated in ethanol to give a
precipitate, which was filtered to give pyranone 1. Yield
63%; M.p. 88–90 °C; IR (cm−1, neat): 3083, 2932, 1720,
compound 3. Yield 58%; M.p. 149–151 °C; IR (cm−1
,
neat): 3131, 3027, 2980, 2932, 1675, 1606, 1538, 1445; 1H-
NMR (CDCl3, 300 MHz) δ (ppm) = 7.96 (d, 1H, J = 12.3,
CH=CH), 6.44 (d, 1H, J = 12.3 Hz, CH=CH), 4.47 (m,
1H, O-CH), 3.22 (s, 3H, NCH3), 3.02 (s, 3H, NCH3), 2.50
(m, 2H, CH2), 1.40 (d, 3H, J = 6.3 Hz, CH3); 13C-NMR
(CDCl3, 75 MHz) δ (ppm) = 195.7, 184.0, 166.6, 155.8,
90.7, 70.0, 45.8, 41.6, 37.7, 20.7; MS/ESI, m/z 226 (MH+)
(Bouaziz et al. 2012).
1
1623, 1577, 1517, 1448; H-NMR (CDCl3, 300 MHz) δ
(ppm) = 8.29 (d, 1H, J = 15.9 Hz, = CH-C=O), 7.92 (d,
1H, J = 15.6 Hz, = CH-C6H5), 7.69 (m, 2H, C6H5),
7.40–7.42 (m, 3H, C6H5), 5.93 (s, 1H, CH-C(OH)), 2.26 (s,
3H, CH3); 13C-NMR (CDCl3, 75 MHz) δ (ppm) = 192.6,
183.1, 168.1, 145.7, 134.8, 130.9, 129.1, 128.9, 123.5,
102.9, 101.9, 20.6. MS/DCI (NH3), m/z 257 (MH+), 274
(MNH4+) (Kinger et al. 2014).
3-(2-Carbamothioylhydrazino)-5-(4-hydroxy-6-methyl-2-
oxo-pyran-3-yl)-3,4-dihydropyrazole-2-carbothioamide (4)
3-[(Dimethylamino)prop-2-enoyl]-4-hydroxy-6-methyl-
pyran-2-one (compound 4′) was prepared by condensation
of an equimolar solution of DHA (1.68 g, 0.01 mol) and N,
N-dimethylformamide dimethylacetal (1.19 g, 0.01 mol) in
the presence of glacial acetic acid (3 mL, 0.05 mol) in
dichloromethane (30 mL). The reaction mixture is heated to
reflux for 6 h. After cooling, the solvent was concentrated
under reduced pressure, and the yellow precipitate was fil-
tered without further purification (yield, 65%; mp, 170–172
°C; m/z 224 (MH+)). Compound 4 was then prepared by
heating to reflux in ethanol (30 mL) an equimolar solution
of compound 4′ (2.23 g, 0.01 mol) with thiosemicarbazide
(0.91 g, 0.01 mol) for 3 h. After cooling, the solvent was
evaporated to dryness under reduced pressure, and the oily
residue was triturated in ethanol to give a precipitate, which
was filtered to give compound 4. Yield 37%; M.p. 247–248
°C; IR (cm−1, neat): 3356, 3296, 3242, 3163, 1724, 1703,
1602, 1532, 1489. 1H-NMR (d6-DMSO, 300 MHz) δ (ppm)
= 11.56 (br. s, 1H, OH), 8.78 (s, 1H, NH), 8.19 (s, 2H,
NH2), 7.85 (br. s, 1H, NH), 7.59 (br. s, 1H, NH), 6.31 (s,
1H, NH), 5.89 (s, 1H, CH-C(OH), 5.57 (dd, 1H, J = 6.9 Hz,
3.3 Hz, CH), 3.60 (dd, 1H, J = 19.8 Hz, 10.2 Hz, CH2),
3.26 (dd, 1H, J = 19.5 Hz and 3.0 Hz, CH2),2.24 (s, 3H,
CH3); 13C-NMR (d6-DMSO, 75 MHz) δ (ppm) = 19.5, 39.7
(in DMSO signal) = 182.5, 174.9, 170.2, 165.2, 161.2,
155.7, 100.6, 93.1, 72.4; MS/DCI (NH3), m/z 343 (MH+).
3-[(Z)-3-(4-chlorophenyl)prop-2-enoyl]-4-hydroxy-6-
methyl-pyran-2-one (2)
Compound 2 was synthesized from DHA (1.68 g, 0.01
mol), 4-chlorobenzaldehyde (1.42 g, 0.01 mol) and piper-
idine (0.28 g, 0.0033 mol) in refluxing chloroform (30 mL)
for 2 h using a similar procedure described for compound 1.
Yield 84%; M.p. 148–150 °C; IR (cm−1, neat) 3101, 3024,
1
2997, 1722, 1627, 1489. H-NMR (CDCl3, 300 MHz) δ
(ppm) = 8.28 (d, 1H, J = 15.6 Hz, = CH-C=O), 7.89 (d, J
= 15.6 Hz, 1H, = CH-C6H5), 7.62 (d, 2H, J = 8.4 Hz,
C6H5), 7.39 (d, 2H, J = 8.4 Hz, C6H5), 5.97 (s, 1H, CH-C
(OH)), 2.29 (s, 3H, CH3); 13C-NMR (CDCl3) δ (ppm) =
192.6, 183.1, 168.8, 144.6, 137.0, 133.2, 130.3, 129.2,
123.6, 102.4, 20.7; MS/ESI, m/z 291 (MH+) (Kinger et al.
2014).
5-[(E)-3-(dimethylamino)prop-2-enoyl]-4-hydroxy-2-
methyl-2,3-dihydropyran-6-one (3)
5-Acetyl-4-hydroxy-2-methyl-2,3-dihydropyran-6-one
(compound 3’) was synthesized by catalytic hydrogenation
of DHA (4.2 g, 0.025 mol) on 10% palladium carbon (0.25

Med Chem Res
4-Hydroxy-6-methyl-3-[5-methyl-3-(2-thienylmethyl)
imidazol-4-yl]pyran-2-one (5)
(1.08 g, 0.01 mol) in ethanol (30 mL) under magnetic stirring.
After cooling, the yellow precipitate was filtered, and
recrystallized in ethanol to give pure phenylhydrazone 6′
(yield, 85%; mp, 200 °C). A solution of 6′ (1.29 g) was
heated to reflux in acetic acid (50 mL) under magnetic stirring
for 3 h. After removing the solvent under reduced pressure,
the product was recrystallized in acetonitrile to afford pure
diketo compound 7′ as a yellow powder (yield, 73%; M.p.
101 °C), which exists as two tautomeric forms, as revealed by
1H-NMR (CDCl3, 300 MHz) δ (ppm) = 2.13 (s, CH3), 2.35
(s, CH3), 2.47 (s, CH3), 3.87 (s, CH2), 5.72 (s, CH), 6.80–7.25
(m, H-arom). Compound 6 was prepared by heating to reflux
under magnetic stirring a solution of 2-amino-4-methyl-1,3-
benzothiazole (1.64 g, 0.01 mol) and diacetoacetyl pyrazolone
7′ (2.58 g, 0.01 mol) in n-butanol (40 mL) for 3 h. After
cooling, the solution was then concentrated under reduced
pressure to a volume of approximately 20 mL, and the orange
precipitate was filtered. Yield 70%; M.p. 201–202 °C; IR
(cm−1, neat): 2916, 1621, 1594, 1551, 1523, 1496, 1463. 1H-
NMR (CDCl3, 300 MHz) δ (ppm) = 13.16 (s, 1H, OH), 7.85
(dd, 2H, J = 8.7 Hz, 1.2 Hz, H-arom), 7.66 (dd, 1H, J = 7.8
Hz and 1.2 Hz, H-arom), 7.45 (t, 2H, J = 8.4 Hz, H-arom),
7.16–7.28 (m, 3H, H-arom), 5.59 (s, 1H, C=CH), 2.69 (s, 3H,
CH3), 2.64 (s, 3H, CH3), 2.51 (s, 3H, CH3); 13C-NMR
(CDCl3, 75 MHz) δ (ppm) = 186.1, 159.7, 158.2, 150.7,
146.6, 137.5, 131.6, 131.4, 129.0, 127.0, 126.4, 123.9, 120.7,
118.3, 99.1, 22.9, 18.3, 15.9; MS/EI, m/z 404 (M+).
Ketimine 5′was prepared by reaction of dehydroacetic acid
(DHA, 1.68 g, 0.01 mol) with 2-thienylmethanamine (1.13
g, 0.01 mol) in ethanol (40 mL) for 6 h at room temperature
under magnetic stirring. The reaction mixture was then
concentrated under reduced pressure to a volume of about
10 mL, and the pale yellow precipitated enamine 1 was
collected by filtration, and washed with cold ethanol. Yield
68%; M.p. 112 °C. FT-IR (cm−1, neat): 3105, 3082, 1691,
1659, 1575, 1468, 1390, 1361, 1324, 1062, 998, 928, 728.
1H-NMR (CDCl3, 300 MHz) δ (ppm) = 14.56 (br.s, 1H,
-NH), 7.28 (dd, J = 1.3, 5.0 Hz, 1H, -CH-S), 7.00 (m, 2H,
H-βthio), 5.67 (s, 1H, DHA-H), 4.84 (d, 2H, J = 5.4 Hz,
-CH2-), 2.71 (s, 3H, -CH3), 2.11 (s, 3H, -CH3); 13C-NMR
(CDCl3, 75 MHz) δ (ppm) = 184.75, 176.04, 163.63,
162.91, 137.17, 127.30, 126.55, 126.00, 107.31, 96.89,
42.83, 19.80, 18.25; MS (TOF MS ES+) m/z 264.0698
(MH+) (calc. mass for C13H13NO3S + H+: 264.0694). A
solution of ketamine 5′ (2.63 g, 0.01 mol), toluenesulfo-
nylmethyl isocyanide (TosMIC, 1.95 g, 0.01 mol), bismuth
(III) trifluoromethanesulfonate (66 mg, 0.1 mmol) and tert-
butylamine (3.65 g, 0.05 mol) in methanol (40 mL) was stir-
red at room temperature for 48 h. The solution was then
evaporated to dryness under reduced pressure and the crude
product was purified by column chromatography on silica gel
using a mixture of ethyl acetate and methanol (9:1) as eluent
to afford 2.11 g (Yield: 70%) of compound 5. M.p. 227–228 °
C; FT-IR (cm−1, neat): 3101, 3087, 1666, 1655, 1626, 1537,
1479, 1434, 1414, 1294, 1270, 1226, 1207, 994, 904, 778,
737. Fully attribution of imidazole 5 in solution was per-
formed using 1D proton and carbon experiments, and 2D
COZY, HSQC, HMBC, NOESY experiments. These
experiments were recorded on a BRUKER AVANCE 500
(4E)-4-(2-Ethylsulfanyl-7-methyl-[1,3,4]thiadiazolo[3,2-a]
pyrimidin-5-ylidene)-5-methyl-2-phenyl-pyrazol-3-one (7)
Compound 7 was prepared by heating to reflux a solution of
2-amino-5-ethylsulfanyl-1,3,4-thiadiazole (1.61 g, 0.01 mol)
and diacetoacetyl pyrazolone 7′ (2.58 g, 0.01 mol) in n-
butanol (40 mL) for 2 h. The solution was then concentrated
under reduced pressure to a volume of approximately 15
mL, and the precipitate was filtered. Recrystallization from
ethanol gave pure pyrazolone derivative 7. Yield 70%; M.p.
238–240 °C; IR (cm−1, neat): 2988, 2937, 1638, 1584,
1
MHz equipped with a 5 mm TCI cryoprobe. H-NMR (D6-
DMSO, 500 MHz) δ (ppm) = 8.52 (s, 1H, N2 -CH-), 7.55
(dd, 1H, J = 1.28, 6.36 Hz, -CH-S), 7.17(m, 1H, H-γ thio),
7.05 (dd, 1H, J = 3.50, 8.61 Hz, H-βthio), 5.78 (q, 1H, J =
0.9 Hz, -CH-), 5.53 (s, 2H, -CH2-), 3.60 (s, 1H, -OH), 2.24 (s,
3H, -CH3),2.09 (d, 3H, J = 0.9 Hz, -CH3); 13C-NMR (D6-
DMSO, 125 MHz) δ (ppm) = 173.52 (-C-), 162.82 (-C=O),
160.33 (-C-), 138.60 (-C-), 133.17 (-C-), 129.43 (-C-), 128.03
(-CH-), 127.81 (-CH-), 127.44 (-CH-), 124.53 (-C-), 104.99
(-CH-), 90.14 (-C-), 43.99 (-CH2-), 19.71 (-CH3), 10.52
(-CH3), MS (TOF MS ES+) m/z 303.0799 (MH+) (calc. mass
for C15H14N2O3S + H+: 303.0803) (Fodili et al. 2015).
1
1507, 1489, 1467. H-NMR (CDCl3, 300 MHz) δ (ppm) =
8.48 (s, 1H, H-pyr), 8.04 (dd, 2H, J = 3.3 Hz, 1.2 Hz, H-
arom), 7.35 (m, 2H, H-arom), 7.10 (t, 1H, J = 7.2 Hz, H-
arom), 3.35 (q, 2H, J = 7.2 Hz, CH3-CH2-S), 2.48 (s, 3H,
CH3), 2.38 (s, 3H, CH3); 1.47 (t, 3H, J = 7.2 Hz, CH3-CH2-
S); 3C-NMR (CDCl3, 75 MHz) δ (ppm) = 161.5, 160.9,
160.5, 147.9, 146.3, 139.5, 128.4, 123.6, 119.1, 111.9, 85.6,
28.4, 24.0, 19.8, 13.9; MS/EI, m/z 383 (M+).
1-(5-Hydroxy-3-methyl-1-phenyl-pyrazol-4-yl)-3-[(4-
methyl-1,3-benzothiazol-2-yl)amino]but-2-en-1-one (6)
4-(2-Hydroxyethyl)-5-methyl-3-oxo-1H-pyrazole-2-
carbothioamide (8)
Compound 6′ was obtained by heating to reflux for 1 h a
solution of DHA (1.68 g, 0.01 mol) and phenylhydrazine
2-Acetylbutyrolactone (1.28 g, 0.01 mol) and thiosemi-
carbazide (0.91 g, 0.01 mol) were heated in refluxed

Med Chem Res
methanol (40 mL) for 6 h. After cooling, the white pre-
cipitate formed was filtered and washed with cold methanol
to give compound 8. Yield 80 %; M.p. 172–173 °C. IR
(cm−1, neat): 3362, 3290, 3126, 2944, 2880, 1651, 1600,
1569, 1518, 1435. 1H-NMR (d6-DMSO, 300 MHz) δ (ppm)
= 11.59 (br. s, 1H, NH), 10.39 (br. s, 1H, NH2), 9.61 (br. s,
1H, NH2), 4.51 (br. s, 1H, NH), 3.41 (t, 2H, J = 6.9 Hz,
CH2), 2.28 (t, 2H, J = 6.6 Hz, CH2), 2.16 (s, 3H, CH3); 13C-
NMR (d6-DMSO, 75 MHz) δ (ppm) = 175.8, 162.9, 150.9,
100.7, 59.8, 25.3, 10.7; MS/ESI, m/z 202 (MH+) (Zaoui
et al. 2013).
constant at 5% v/v in a final volume of 1 mL for all kinetic
reactions. Kinetic assays were performed using trans-2-
dodecenoyl-coenzyme A (DDCoA) and wild type InhA
as previously described (Matviiuk et al. 2014). Briefly,
reactions were performed at 25 °C in an aqueous buffer (30
mM PIPES and 150 mM NaCl pH 6.8) containing addi-
tionally 250 µM cofactor (NADH), 50 µM substrate
(DDCoA) and the tested compound (at 50 µM or 10 µM).
Reactions were initiated by the addition of InhA (100 nM
final), and NADH oxidation was followed at 340 nm. The
inhibitory activity of each derivative was expressed as the
percentage inhibition of InhA activity (initial velocity of the
reaction) with respect to the control reaction without
inhibitor.
4-(2-Hydroxyethyl)-5-methyl-2-(4-phenylthiazol-2-yl)-1H-
pyrazol-3-one (9)
To a suspension of compound 8 (2.01 g, 0.01 mol) and
sodium acetate (0.82 g, 0.01 mol) in ethanol (50 mL) 2-
bromoacetophenone (1.99 g, 0.01 mol) was added, and the
mixture was brought to reflux for 3 h. After cooling, the
white precipitate was filtered, washed with water, and dried
under vacuum. Yield 75%; M.p. 93–94 °C; IR (cm−1, neat):
In vitro assays: growth conditions and minimum inhibitory
concentration (MIC) determination
M. tuberculosis H37Rv was used as the reference strain.
The strains were grown at 37 °C in Middlebrook 7H9 broth
(Difco), supplemented with 0.05% Tween 80, or on solid
Middlebrook 7H11 medium (Difco) supplemented with
oleic acid-albumin-dextrose-catalase (OADC). MICs
for the new compounds were determined by means of the
micro-broth dilution method. Dilutions of M. tuberculosis
wild-type (about 105–106 cfu/ml) were streaked onto
7H11 solid medium containing a range of drug concentra-
tions (0.25 μg/mL to 40 μg/mL). Plates were incubated at
37 °C for about 21 days and the growth was visually
evaluated. The lowest drug dilution at which visible growth
failed to occur was taken as the MIC value. Results were
expressed as the average of at least three independent
determinations.
1
3299, 2927, 2861, 1626, 1582, 1537, 1505. H-NMR (d6-
DMSO, 300 MHz) δ (ppm) = 7.99 (d, 2H, J = 7.2 Hz,
C6H5), 7.74 (s, 1H, H-thiazol), 7.45 (m, 2H, C6H5), 7.34
(m, 1H, C6H5), 4.67 (br. s, 1H, NH), 3.49 (t, 2H, J = 6.7
Hz, CH2), 2.39 (t, 2H, J = 6.8 Hz, CH2), 2.22 (s, 3H,
CH3);13C-NMR (d6-DMSO, 75 MHz) δ (ppm) = 150.6,
148.9, 133.9, 128.7, 128.0, 125.9, 108.5, 101.5, 59.9, 25.6,
11.0; MS/ESI, m/z 302 (MH+).
4-(2-Hydroxyethyl)-5-methyl-2-(4-methylthiazol-2-yl)-1H-
pyrazol-3-one (10)
To a suspension of compound 8 (2.01 g, 0.01 mol) and
sodium acetate (0.82 g, 0.01 mol) in ethanol (50 mL) bro-
moacetone (1.37 g, 0.01 mol) was added, and the mixture
was brought to reflux for 3 h. After cooling, the white
precipitate was filtered, washed with water, and dried under
vacuum. Yield 70%; M.p. 226–227 °C; IR (cm−1, neat):
3225, 3096, 2938, 1634, 1599, 1535, 1522, 1492. 1H-NMR
(d6-DMSO, 300 MHz) δ (ppm) = 11.76 (br. s, 1H, NH),
6.88 (d, 1H, J = 1.2 Hz, C=CH), 4.64 (br. s, 1H, NH), 3.46
(t, 2H, J = 6.9 Hz, O-CH2), 2.34 (m, 5H, C=C-CH3 and
CH2), 2.16 (s, 3H, CH3); 13C-NMR (d6-DMSO, 75 MHz) δ
(ppm) = 159.9, 152.8, 150.0, 146.8, 108.5, 101.2, 60.0,
25.6, 16.9, 10.9. MS/ESI, m/z 240 (MH+).
Molecular docking
Molecular graphics were performed with the UCSF Chi-
mera package (Pettersen et al. 2004). Chimera is developed
by the Ressource for Biocomputing, Visualization, and
Informatics at the University of California, San Francisco
(supported by the NIGMS P41-GM103311).The protein
structures used in this paper were obtained from the Protein
Data Bank (Berman et al. 2000) (http://www.pdb.org) and
were structurally aligned with structure 1BVR (Rozwarski
et al. 1999) (chain A) set as reference and using UCSF
Chimera/Matchmaker program (Meng et al. 2006). The
protein structures, in this reference space, were prepared
(structure checks, rotamers, hydrogenation) using Biovia
Discovery Studio 2016 (http://accelrys.com/) and UCSF
Chimera. The new compounds were sketched using Che-
mAxon Marvin 16, (http://www.chemaxon.com). All
ligands (extracted from protein structures or new) were
checked (hybridization, hydrogenation, some geometry
Biological assays
InhA activity inhibition
Stock solutions of all compounds were prepared in DMSO
such that the final concentration of this co-solvent was

Med Chem Res
optimizations, 3D sketching) and were merged in SDF
libraries using Discovery Studio. Molecular modeling stu-
dies (Sousa et al. 2013) were carried with Molegro Virtual
Docker 6 software (http://www.clcbio.com) using the chain
A of structure 1P45 (Kuo et al. 2003) (1P45a), the chain E
of structure 4R9R (Manjunatha et al. 2015) (4R9Re), the
chain E of structure 4R9S (Manjunatha et al. 2015) (4R9Se)
as targets. A search space volume of 17 Å radius centered in
the binding pocket was used. According to structural study,
19 residues were defined as flexible during the docking:
GLY96, PHE97, GLY102, ASP148, PHE149, ALA154,
ALA157, TRP160, MET161, ALA164, GLY192, ARG195,
ALA198, ALA201, GLN214, LEU217, LEU218, GLY221
and ASN231. The ligands were set flexible during the
docking, and two different docking protocols were used.
The NAD (the reduced form NADH for 4R9R/4R9S
structures, the oxidized form NAD+ for 1P45) was used as
cofactor (Molegro feature) during the docking calculations.
The protocol P1 (P1-OPT) is based on flexible docking at
protein level (softened potentials and minimization of pro-
tein residues and backbone) and Moldock optimizer used as
searching algorithm. Docking process uses the MolDock
(Thomsen and Christensen 2006) function (Moldock [grid]
with a resolution of 0.3 Å) for scoring and Moldock opti-
mizer (6000 iteration steps, other parameters let as default,
20 independent runs) for searches. Final minimization was
parameterized using 2000 steps for lateral chains and pro-
tein backbone; other parameters were let with default
values. No water molecules (entropy penalty) were taken in
account in the study. MolDock and Rerank (Thomsen and
Christensen 2006) scores were calculated post-docking and
post-minimization.
The protocol P2 (P2-GPU) is based on rigid docking at
protein level and a GPU (Nvidia Tesla CUDA hardware,
http://www.nvidia.com) screening algorithm. Docking pro-
cess uses the MolDock function (Moldock [grid] with a
resolution of 0.3 Å) for scoring and CUDA optimizer
(MVD, 3000 iteration steps, other parameters let as default,
40 independent runs) for searches. Post-docking, all the
poses were re-ranked using MolDock and Rerank scoring
systems (without a post-minimization step).
The clustering parameters (RMSD threshold 2 Å, other
parameters let as default) were the same for P1 and P2
protocols. The same post-docking filtering protocol was
also used. The poses issued from the calculations (20 poses
for P1, 40 for P2) were filtered according to the following
sequential rules: {A} the GPU, MolDock, Rerank, con-
tributions terms (Interaction, Protein, Cofactor) issued from
calculation were selected to give a table of poses/scores for
P1 and P2 protocols. {B} For P1 and P2 (independently) the
table was ranked according to MolDock scores. {C} For all
types of scores/contributions, the values were distributed
into ranges (as a histogram) of 5 units (giving a scale color
in tables). {D} Each pose of the table was compared to
others and the poses with the same conformation were
tagged. {E} Each pose of the table was manually inspected
for the interactions of ligand with NAD and Tyr158 groups.
The poses were tagged if a typical network of hydrogen
bonds involving both NAD (i.e., nicotinamide and ribose,
eventually phosphate) and Tyr158 was possible. {F} The
first four best MolDock scores issued from P1 and P2 were
retained by compound, giving two sets of 4 poses per
compound.
Results and discussion
Chemistry
Compounds 1 and 2 were easily prepared from condensa-
tion of an equimolar solution of dehydroacetic acid and
benzaldehyde or 4-chlorobenzaldehyde in the presence of
piperidine in refluxing chloroform (Scheme 1). Despite of
1
high J values (15.6 – 15.9 Hz) measured in their H-NMR
spectra, it was possible from the NOESY spectra to identify
the geometric isomerism of the double bond as cis in both
compounds, because of the strong crosspeaks between the
two ethylenic protons.
Compound 3 was prepared by condensation of dihydro-
DHA 3′, obtained by catalytic hydrogenation of DHA on
palladium carbon, with N,N-dimethylformamide dimethy-
lacetal in the presence of a catalytic amount of trimethyla-
mine (Scheme 2). Compound 3 was fully characterized by
standard spectral analyzes and its structure was confirmed
through X-ray crystallographic analysis (Fig. 3 and Table 1
for selected geometric parameters).
Compound 4 was prepared from dimethylaminoprope-
noyl derivative 4′, easily prepared from condensation of
DHA with N,N-dimethylformamide dimethylacetal under
acidic conditions (Scheme 3). By heating an equimolar
solution of compound 4′ and thiosemicarbazide in ethanol,
an unexpected double addition of thiosemicarbazide with
concomitant cyclization occurred, leading to pyrazole 4. As
the analysis of the 13C and 1H-NMR spectra proved
inconclusive for an unambiguous structure determination,
Scheme 1 Synthesis of DHA derivatives 1-2

Med Chem Res
Scheme 2 Synthesis of DHA
derivative 3
compound 4 was recrystallized from a slow evaporation of
an N,N-dimethylformamide solution to afford crystals of
sufficient quality for X-ray structure determination. The
compound was found to crystallize in the triclinic P1 space
group, with two solvent (DMF) molecules and one mole-
cule of water in the asymmetric unit. The three-dimensional
structure of 4 is shown in Fig. 3 and selected geometric
parameters are summarized in Table 1.
Fig. 3 X-ray structure of compounds 3, 4, and 7 (only the hydrogen
atoms bonded to carbons are shown). Color code: C: gray, H: white, N:
blue, O: red, S: yellow. See Table 1 for crystal data (the cif files can be
obtained free of charge from the Cambridge Crystallographic Data
Center via www.ccdc.cam.ac.uk/data_request/cif)
Imidazole derivative bearing both a thiophenyl group and
the DHA moiety (compound 5) was synthesized by reaction
of the corresponding dehydroacetic-substituted acetimide
5′, prepared by Schiff-base condensation of DHA with 2-
thienylmethanamine, with tosylmethyl isocyanide (Tos-
MIC) reagent in the presence of tert-butylamine and cata-
lytic amounts of bismuth(III) triflate, as already reported
(Scheme 4) (Fodili et al. 2015). The mechanism of the
reaction was thought to go through the formation of an
imidazolium intermediate, followed by an unusual intra-
molecular rearrangement and subsequent aromatization to
the imidazole ring, leading to the 1,5-disubstituted-5-methyl
imidazole compound 5. The structure of the latter had been
unequivocally determined by a combination of one- and
two-dimensional NMR methods and X-ray crystallography.
Both compounds 6 and 7 were synthesized from com-
pound 6′, easily prepared by heating an equimolar solution
of DHA and phenylhydrazine in ethanol. When hydrazone
6′ was refluxed in acetic acid, it underwent a rearrangement
involving a nitrogen nucleophilic attack at the C-2 lactone
carbonyl, with ring opening to yield diketopyrazolinone 7′
(Scheme 5). This latter compound contains a β-tricarbonyl
system, which exists as two tautomeric forms in chloroform,
probably dienol and monoenol tautomers, as revealed by
1H-NMR. As expected, the corresponding eneamine com-
pound 6 was obtained when pyrazolinone 7′ was allowed to
react with 2-amino-4-methyl-1,3-benzothiazole. On the
other hand, when 2-amino-5-ethylsulfanyl-1,3,4-thiadiazole
was used as nucleophile in the same conditions, compound
7 bearing a condensed thiadiazole pyrimidine heterocyclic
system was formed. The structure of compound 7 was
deduced from NMR studies and confirmed by X-ray crys-
tallographic analysis (Fig. 3, Table 1).
base condensation of acetylbutyrolactone with thiosemi-
carbazide, followed by an intramolecular cyclization lead-
ing to the butyrolactone ring opening, finally yielding the
pyrazole 8. The latter compound was then reacted with 2-
bromoacetone or 2-bromoacetophenone under basic condi-
tions to afford thiazole compounds 9 and 10, respectively.
Biological activity
The activity of the synthesized compounds was evaluated
by determining both M. tuberculosis (H37Rv) growth
inhibition and InhA enzyme inhibition, and compared to
that of well-known isoniazid (INH), GEQ and Triclosan
(TCL) (Fig. 4), used as references (Table 2) (Chollet et al.
2015 and references herein).
Minimum inhibitory concentrations (MIC) were deter-
mined for each product and reference compounds (Table 2).
Among all the evaluated compounds, compounds bearing a
DHA moiety showed activities between 17 and 44 μM
against M.tb H37Rv. In this series, compound 2 bearing a
chlorine on the phenyl ring was found the most active, and
could reasonably serve as the lead molecule for drug
development by varying the aryl moiety. In the pyrazole
series, compound 8 was the most active with a MIC of 12
μM and it could also be used as a platform for derivatiza-
tion. This compound possesses different modifiable posi-
tions i.e., the alcohol, the amine and the thioamide on the
pyrazole ring. Indeed, two compounds 9 and 10 both syn-
thesized from pyrazolone 8 were also evaluated. Compound
10 with a methyl group on the pyrazone showed a loss of
antitubercular activity while compound 9 with a phenyl
group was found to be more valuable, with a MIC of 16 μM.
To confirm that InhA enzyme could be the target,
enzyme inhibition studies were carried out. The compounds
Pyrazole 8 was obtained in one step by reacting com-
mercially available 2-acetylbutyrolactone with thiosemi-
carbazide in refluxing methanol (Scheme 6). The
mechanism of formation of this pyrazole compound is a
process that comprises two consecutive steps, i.e., a Schiff-

Med Chem Res
Table 1 Crystal data and structure refinement for compounds 3, 4 and 7
3
4
7
CCDC number
Empirical formula
Formula weight
Temperature
CCDC-1538417
C11H15NO4
CCDC-1538418
C17H28.50N8O5.25S2
493.10
CCDC-1538419
C18H17N5OS2
383.48
225.24
193(2) K
193(2) K
193(2) K
Wavelength
0.71073 Å
0.71073 Å
0.71073 Å
Crystal system
Space group
Monoclinic
Triclinic
Monoclinic
P21/n
P-1
P21/n
Unit cell dimensions
a = 8.3279(5) Å
α = 90°
a = 9.294(5) Å
α = 72.418(2)°
b = 11.5558(7) Å
β = 73.147(2)°
c = 12.9935(8) Å
γ = 70.448(2)°
1225.68(13) Å3
2
a = 13.6784(7) Å
α = 90°
b = 12.5350(9) Å
β = 95.907°
b = 6.9492(4) Å
β = 103.079(3)°
c = 19.0795(10) Å
γ = 90°
c = 10.4853(7) Å
γ = 90
1088.75(13) Å3
Volume
1766.54(17) Å3
Z
4
4
Density (calculated)
Absorption coefficient
F(000)
1.374 mg/m3
0.105 mm−1
1.336 mg/m3
0.262 mm−1
1.442 mg/m3
0.320 mm−1
480
521
800
Crystal size
0.200 × 0.200 × 0.200 mm3
5.184–26.367°
−9 ⩽ h ⩽10, −15 ⩽ k ⩽ 15,
−13 ⩽ l ⩽ 13
8263
0.200 × 0.060 × 0.040 mm3
5.111–27.870°
−12 ⩽ h ⩽ 11, −14 ⩽ k ⩽ 15,
−17 ⩽ l ⩽ 17
19,115
0.700 × 0.100 × 0.100 mm3
3.298–27.876°
−17 ⩽ h ⩽ 17, −6 ⩽ k ⩽ 9,
−24 ⩽ l ⩽ 25
17,407
Theta range for data coll.
Index ranges
Reflections collected
Independent reflections
Completeness to theta = 25.242°
Refinement method
2209 [R(int) = 0.0795]
99.1%
Full-matrix least-squares on F2
5752 [R(int) = 0.0489]
98.6%
Full-matrix least-squares on F2
4189 [R(int) = 0.0542]
99.7%
Full-matrix least-squares on F2
Data/restraints/parameters
Goodness-of-fit on F2
2209/0/152
5752/3/337
4189/0/238
1.086
1.013
1.029
Final R indices [I > 2sigma(I)]
R indices (all data)
R1 = 0.0635, wR2 = 0.1438
R1 = 0.1274, wR2 = 0.1691
0.375 and −0.242 e Å−3
R1 = 0.0484, wR2 = 0.0976
R1 = 0.0886, wR2 = 0.1128
0.372 and −0.270 e Å−3
R1 = 0.0462, wR2 = 0.1007
R1 = 0.0758, wR2 = 0.1153
0.525 and −0.419 e Å−3
Largest difference peak and hole
Scheme 3 Synthesis of compound 4
were tested for their capacity to inhibit the reduction of the
double bond of DDCoA, a C12-substrate, by NADH in the
presence of InhA. Among the inhibitors, five compounds 5,
6, 8, 9, and 10 were totally inactive at 50 μM against the
InhA enzyme. The lack of inhibition of InhA for
compounds 8 and 9 in contrast with the low MICs suggests
that the target is clearly not InhA. Compounds 1, 3, 4, and 7
showed poor activity against InhA. It is only compound 2
that showed the highest potency against the InhA enzyme. It

Med Chem Res
Scheme 4 Synthesis of
imidazole derivative 5
Scheme 5 Synthesis of
derivatives 6 and 7
was shown to be three times more potent than compound 1
without the para-chloro substitution.
molecular docking was performed in an attempt to explore
interactions of compound 2 with InhA binding site.
As DHA derivative, this compound shows strong simi-
larities with hydroxy-2-pyridones. In the case of this last
class (4-hydroxy-2-pyridones) of molecules, two com-
pounds are known to be co-crystallized with NAD and
InhA, including the lead compound NITD-916. Conse-
quently, the two corresponding PDB entries 4R9S (Chain E,
ligand 3KY/NITD-916) and 4R9R (chain E, ligand 3KX)
were used to dock compound 2 (Manjunatha et al. 2015).
Using the in-house software, we also found that the SBL
loop conformations of 4R9Re and 4R9Se were similar to
that of PDB entry 1P45 (Kuo et al. 2003). According to this
result, the PDB structure 1P45 (chain A) was added to the
receptor’s list used to dock the compound 2. The three
protein structures all show an opened binding site (major
portal). In the case of 1P45, two molecules of triclosan
(TCL) are co-crystallized; one of the TCL molecules shows
the conformation and the hydrogen bonding network (NAD,
Tyr158) typical of co-crystalized compounds based on tri-
closan core.
Docking studies
Performing docking experiments in the case of InhA
structures is a difficult task for several reasons. The protein
is characterized by a flexible binding site showing two
variable entries (major portal and minor portal) at the two
sides of the binding cavity (Rozwarski et al. 1999; Kumar
et al. 2016) and structural transitions involving the substrate
binding loop (SBL) at the side (residues 195–210) of the
active site (Li et al. 2014; Lai et al. 2015). Direct inhibitors
show rapid reversible association with protein and NAD
cofactor, but some compounds are characterized by a longer
residence time in binding site; in this case, induced-fit
structural transitions involving a slow reordering of the SBL
loop are expected (Luckner et al. 2010; Merget and Sotriffer
2015). This context makes difficult the choice of structures
for in silico studies, particularly molecular docking studies.
Particularly if a given compound is not closely structurally
related (chemical structure) to a co-crystalized inhibitor,
there is more than one structure that can be potentially used
to carry in silico experiments. Knowing these limitations,
Two different and independent docking protocols (P1-
OPT, P2-GPU, described in experimental section) were
used and found to give similar results in terms of pose

Med Chem Res
Scheme 6 Synthesis of derivatives 9 and 10
Table 2 Compounds 1–10 tested as inhibitory agents of M.
tuberculosis H37Rv strain and biological activities against InhA
enzyme
Cpd.
MIC
% Inhibitiona
(µg/mL)/(μM)
InhA at 50 μM
1
10 / 39
5 / 17
19
60
26
26
0
2
3
10 / 44
4
10 / 29
---b
5
Fig. 4 Structures of GEQ and Triclosan (TCL) used as references in
this study
6
> 10 / > 25
> 10 / > 26
2.5 / 12
0
7
13
0
8
conformations and docking scores. One conformation’s type
(named A2+) is linked to the best scores (molecular
docking data table, supplementary information). Figs. 5 and 6
show two views (lateral, upper) of this ligand’s conforma-
tion, including TCL and 3KX counterparts from 1P45 and
4R9R structures. Fig. 7 shows the hydrogen bond network
in this pose (1P45 structure). The cyclic keto interacts with
Tyr158 (cyclic ether groups are shown in front of the pyr-
idone’s cycle, towards the observer) and the chlorophenyl
group of compound 2 is matching the position of cyclo-
hexane group of 3KX.
The A2+ poses correspond always to strong poses: the
GPU and MolDock and Rerank values are the best of each
run. We noticed also that these poses display conformity
properties: the interaction with Tyr158 and (at least one,
better two) the interactions with NAD. The best strength/
conformity descriptors are generally obtained in docking
simulations that reproduce crystallographic conformations
with minimal RMSD values.
We noticed that alternative conformations could exist,
but they do not correspond to these best strength/conformity
characteristics. For example, the pyridone’s cycle can be
tilted; in this case (A1+ pose’s type) the hydroxyl group is
interacting with Tyr158 and the keto/ether groups are
behind the pyridone’s cycle. Inversion of poses can be
found for P1-OPT protocol. In this case, the chlorophenyl
group is not matching the position of cyclohexane group (of
9
5 / 16
0
10
> 10 / > 42
10 / 34.5
20 / 50
0
TCL
GEQ
INH
a
92
89
---
0.05 / 0.4
Experiments performed in triplicate
Precipitated in the medium
b
3KX) but that of the dichlorophenyl group of TCL. More-
over, the hydroxyl group of Tyr158 makes a hydrogen bond
with the hydroxyl group of compound 2. The other poses do
not show the typical Tyr158/NAD combined hydrogen
bond interaction network.
The conformation of co-crystallized 3KX and 3KY is
closely related to TCL (Fig. 6). Here we found that the best
strength/conformity poses of compound 2 correspond to
those (A2+ type) showing the interaction of the keto group
of the pyridone cycle (rather than central keto group) with
Tyr158, in place of the hydroxyl group of TCL and 3KX.
The orientation of compound 2 is inverted (chlorine is
oriented towards minor portal of InhA) relatively to 3KX
(chlorine is oriented towards major portal, triclosan’s
chlorines are oriented towards the two portals). These
results suggest the possibility of an interesting binding
mode for compound 2. We also noticed that some mono- or

Med Chem Res
Fig. 5 InhA clipped binding site (PDB ID 1P45, chain A (Kuo et al.
2003), light green) and Tyr158 residue (green) showing the super-
imposition of crystallographic triclosan (pink, TCL 400 residue, TCL
450 not shown), crystallographic NAD (light pink) and the best
docking poses (type A2+) of compound 2 issued from P1-OPT
(white) and P2-GPU (salmon) calculation protocols (color figure
online)
Fig. 7 Conformation of type A2+ issued from a P1-OPT calculation
in structure 1P45 and the corresponding putative hydrogen bond net-
work between compound 2 and NAD (bottom) and Tyr158 (top). The
interatomic distances are given in Å
Another difficulty in calculations, involving direct inhi-
bitors and InhA, is related to the presence of the cofactor.
Pyridomycin is known to fill the binding cavity indepen-
dently of cofactor (Hartkoorn et al. 2014) but in the case of
most inhibitors, the bound NAD+/NADH masks a sig-
nificant part of binding site and the co-crystallized NAD is
generally involved in interactions with the co-crystalized
inhibitors. In other terms, depending on each compound, the
most favorable conformation for
intermolecular
protein–ligand interactions may not be the most favorable
conformation for intermolecular cofactor–ligand interac-
tions, because some interactions must be done with the
cofactor. This brings in a new approach for calculations: for
a given compound, we speculate that the search of the best
in silico affinity must be directed by the best balance
between the protein–ligand and the cofactor–ligand inter-
molecular contributions. Even if these contributions are not
directly given by many docking softwares, we believe that
this kind of information may be important for analyzing
results of calculations involving a ligand–cofactor–protein
triplet. The docking data table (supplementary information)
shows three contributions used to calculate the final Mol-
Dock score: the total intermolecular interaction energy
(Interaction, as computed by the software) and the
cofactor–ligand (Cofactor) and ligand–protein (Protein)
interaction terms. We noticed that the strong (bolded in
table) poses of compound 2 correspond to the best Inter-
action and Cofactor terms values. This tendency is less clear
for protocol P1-OPT; an explanation could be that the
softened potentials used in this protocol are known to
increase the rate of false positives (Sinko et al. 2013). The
best poses for ‘pure’ protein–ligand interactions term do not
correspond to the strong poses. Careful pose’s inspections,
Fig. 6 InhA clipped binding site (PDB ID 4R9R, chain E (Manjunatha
et al. 2015), gray) and Tyr158 residue (gray) showing the super-
imposition of crystallographic ligand 3KX (hydroxy-2-pyridone
NITD-564, gold), crystallographic NAD (light pink, thin sticks) and
the typical best conformation of compound 2 issued from calculations
(green, type A2+) protocols. The TCL 400 residue from aligned PDB
ID 1P45 (Kuo et al. 2003), chain A (pink) is added for comparison
(color figure online)
bi-chlorated compounds (pyrrolidine carboxamides deriva-
tives) show chlorine groups towards the minor portal of
InhA (He et al. 2006) even if the overall conformation of
molecules (compound 2 vs. 641, 744 and 468 co-
crystallized ligands) is not the same.

Med Chem Res
evaluation of new GEQ derivatives as inhibitors of InhA enzyme
and Mycobacterium tuberculosis Growth. Eur J Med Chem
101:218–235
shows clearly that these conformations do not meet
strength/conformity criteria. This is also an interesting result
and by first calculation evidence for this balance between
ligand–protein and ligand–cofactor interactions we assume
that this paradigm could be another brick to be added in
order to improve ligand design. Working is in progress in
our laboratory about the generalization of a such system for
InhA.
Fodili M, Nedjar-Kolli B, Vedrenne M, Saffon-Merceron N, Lherbet
C, Hoffmann P (2015) The first example of an unusual rearran-
gement in the van Leusen imidazole synthesis. Chem Heterocycl
Compd 51:940–943
García-Sosa AT, Hetenyi C, Maran U (2010) Drug efficiency indices
for improvement of molecular docking scoring functions. J Comp
Chem 31:174–184
Gupta GK, Mittal A, Kumar V (2014) DHA: an excellent source of
bioactive heterocycles. Lett Org Chem 2014 11:273–286
Hammal L, Bentarzi Y, Nedjar-Kolli B, Hoffmann P (2009) Synthesis
of Tetracyclic Pyran[4,3-b]-6H-Imidazo[1,2-a][1,5]Benzodiaze-
pines. Heterocycl Commun 15:209–216
Conclusion
Hartkoorn RC, Pojer F, Read JA, Gingell H, Neres J, Horlacher OP P,
Altmann K-H, Cole ST (2014) Pyridomycin bridges the NADH-
and substrate-binding pockets of the enoyl reductase InhA. Nat
Chem Biol 10:96–98
He X, Alian A, Stroud RM, Ortiz de Montellano PR (2006) Pyrroli-
dine carboxamides as a novel class of inhibitors of enoyl acyl
carrier protein reductase from Mycobacterium tuberculosis. J
Med Chem 49:6308–6323
He X, Alian A, Ortiz de Montellano PR (2007) Inhibition of the
Mycobacterium tuberculosis enoyl acyl carrier protein reductase
InhA by arylamides. Bioorg Med Chem 15:6649–6658
Jain A, Mondal R (2008) Extensively drug-resistant tuberculosis:
current challenges and threats. FEMS Immunol Med Microbiol
53:145–150
Kumar V, Sobhia ME (2016) Molecular dynamics-based investigation
of InhA substrate binding loop for diverse biological activity of
direct InhA inhibitors. J Biomol Struct Dyn 34(11):2434–2452
Kinger M, Gupta VD, Singh J, Arora AK, Jaswal VS (2014) Reaction
of heteroaryl hydrazines with 3-cinnamoyl-4-hydroxy-6-methyl-
2H-pyran-2-ones: Synthesis of some novel pyrazolines and their
iodine(III) mediated conversion to corresponding pyrazoles.
Asian J Chem 26:8084–8086
In conclusion, seven compounds synthesized from DHA
and three from 2-acetylbutyrolactone were evaluated as
antitubercular drugs against M. tuberculosis H37Rv strain.
Among them, compounds 2, 8 and 9 were the most active
with MICs around 15 μM. Further studies showed that
DHA derivatives, especially compound 2, inhibited InhA
enzyme. Molecular docking study of compound 2 were
carried out and crucial interactions were revealed into the
substrate binding site of InhA protein. Particularly, the best
poses of compound 2 show a binding mode with chlorine
oriented towards minor portal; in these cases, the typical
hydrogen bonding network (NAD/Tyr158) is also repro-
duced. This is an interesting result, which could give
insights for structure-activity trends. Overall, our results
show that these highly heteroatom-rich compounds are
interesting starting points for further optimization.
Compliance with ethical standards
Kunigoshi
U
(1960) Synthesis of dehydroacetic acid iso-
nicotinoylhydrazone potassium salt. Its in vitro antituberculous
effect on a strain of virulent human-type tubercle bacteria. Che-
motherapy 8:84–5
Conflict of interest No author had any financial or personal rela-
tionship with any person or organization that could inappropriately
influence the work.
Kuo MR, Morbidoni HR, Alland D, Sneddon SF, Gourlie BB, Sta-
veski MM, Leonard M, Gregory JS, Janjigian AD, Yee C, Musser
JM, Kreiswirth B, Iwamoto H, Perozzo R, Jacobs WR, Sac-
chettini JC, Fidock DA (2003) Targeting tuberculosis and malaria
through inhibition of enoyl reductase: compound activity and
structural data. J Biol Chem 278:20851–20859
References
Lai CT, Li HJ, Yu W, Shah S, Bommineni GR, Perrone V, Garcia-
Diaz M, Tonge PJ, Simmerling C (2015) Rational modulation
of the induced-fit conformational change for slow-onset inhibition
in mycobacterium tuberculosis InhA. Biochemistry 54:
4683–4691
Li HJ, Lai CT, Pan P, Yu W, Liu N, Bommineni GR, Garcia-Diaz M,
Simmerling C, Tonge PJ (2014) A structural and energetic model
for the slow-onset inhibition of the Mycobacterium tuberculosis
enoyl-ACP reductase InhA. ACS Chem Biol 9:986–993
Lu X-Y, Chen Y-D, Jiang Y-J, You Q-D (2009) Discovery of potential
new InhA direct inhibitors based on pharmacophore and 3D-
QSAR analysis followed by in silico screening. Eur J Med Chem
44:3718–3730
Bemer-Melchior P, Bryskier A, Drugeon HB (2000) Comparison of
the in vitro activities of rifapentine and rifampicin against
Mycobacterium tuberculosis complex. J Antimicrob Chemother
46:571–576
Berman H, Westbrook J, Feng Z, Gilliland G, Bhat T, Weissig H,
Shindyalov I, Bourne P (2000) The protein data bank. Nucl Acids
Res 28:235–242
Bouaziz O, Abboub F, Bayou-khier N, Fodili M, Hoffmann P, Amari
M (2012) Reactivity of the dehydroacetic acid hydrogenated at
C5-C6: Synthesis of the substituted pyrano-1,5-benzodiazepines
and the enaminone structure. C R Chimie 15:774–778
Castagnolo D, Manetti F, Radi M, Bechi B, Pagano M, De Logu A,
Meleddu R, Saddi M, Botta M (2009) Synthesis, biological
evaluation, and SAR study of novel pyrazole analogs as inhibitors
of Mycobacterium tuberculosis: part 2. Synthesis of rigid pyr-
azolones. Bioorg Med Chem 17:5716–5721
Luckner SR, Liu N, am Ende CW, Tonge PJ, Kisker C (2010) A slow,
tight binding inhibitor of InhA, the enoyl-acyl carrier protein
reductase from Mycobacterium tuberculosis.
J Biol Chem
585:14330–14337
Chollet A, Mori G, Menendez C, Rodriguez F, Fabing I, Pasca MR,
Madacki J, Korduláková J, Constant P, Quémard A, Bernardes-
Génisson V, Lherbet C, Baltas M (2015) Design, synthesis and
Manetti F, Magnani M, Castagnolo D, Passalacqua L, Botta M, Corelli
F, Saddi M, Deidda D, De Logu A (2006) Ligand-based virtual

Med Chem Res
screening, parallel solution-phase and microwave-assisted
(2003) Indole naphthyridinones as inhibitors of bacterial
synthesis as tools to identify and synthesize new inhibitors of
mycobacterium tuberculosis. ChemMedChem 1:973–989
Manjunatha UH, Rao SPS, Kondreddi RR, Noble CG, Camacho LR,
Tan BH, Ng SH, Ng PS, Ma NL, Lakshminarayana SB, Herve M,
Barnes SW, Yu W, Kuhen K, Blasco F, Beer D, Walker JR,
Tonge PJ, Glynne R, Smith PW, Diagana TT (2015) Direct
inhibitors of InhA are active against Mycobacterium tuberculosis.
Sci Transl Med 7:269ra3
Matviiuk T, Mori G, Lherbet C, Rodriguez F, Pasca MR, Gorichko M,
Guidetti B, Voitenko Z, Baltas M (2014) Synthesis of 3-heteryl
substituted pyrrolidine-2,5-diones via catalytic Michael reaction
and evaluation of their inhibitory activity against InhA and
Mycobacterium tuberculosis. Eur J Med Chem 71:46–52
Meng EC, Pettersen EF, Couch GS, Huang CC, Ferrin TE (2006)
Tools for integrated sequence-structure analysis with UCSF
Chimera. BMC Bioinform 7:339–349
enoyl-ACP reductases FabI and FabK.
46:1627–1635
J
Med Chem
Shirude PS, Madhavapeddi P, Naik M, Murugan K, Shinde V, Nan-
dishaiah R, Bhat J, Kumar A, Hameed S, Holdgate G, Davies G,
McMiken H, Hegde N, Ambady A, Venkatraman J, Panda M,
Bandodkar B, Sambandamurthy VK, Read JA (2013) Methyl-
thiazoles: a novel mode of inhibition with the potential to develop
novel inhibitors targeting InhA in Mycobacterium tuberculosis. J
Med Chem 56:8533–8542
̌
̌
̌
Sink R, Sosic I, Zivec M, Fernandez-Menendez R, Turk S, Pajk S,
Alvarez-Gomez D, Lopez-Roman EM, Gonzales-Cortez C,
Rullas-Triconado J, Angulo-Barturen I, Barros D, Ballell-Pages
L, Young RJ, Encinas L, Gobec S (2015) Design, Synthesis and
evaluation of new thiadiazole-based direct inhibitors of enoyl acyl
carrier protein reductase (InhA) for the treatment of tuberculosis.
J Med Chem 58:613–624
Merget CA, Sotriffer CA (2015) Slow-Onset Inhibition of Myco-
bacterium tuberculosis InhA: revealing molecular determinants of
residence time by MD simulations. PLoS One 10:1–25
Sinko W, Lindert S, McCammon JA (2013) Accounting for receptor
flexibility and enhanced sampling methods in computer-aided
drug design. Chem Biol Drug Des 81:41–49
Ntie-Kang F, Kannan S, Wichapong K, Owono Owono LC, Sippl W,
Megnassan E (2014) Binding of pyrazole-based inhibitors to
Mycobacterium tuberculosis pantothenate synthetase: docking
and MM-GB(PB)SA analysis. Mol Biosyst 10:223–239
Sousa SF, Ribeiro AJM, Coimbra JTS, Neves RPP, Martins SA,
NSHN Moorthy, Fernandes PA, Ramos MJ (2013) Protein-ligand
docking in the new millennium-a retrospective of 10 years in the
field. Curr Med Chem 20:2296–2314
Pettersen, EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM,
Meng EC, Ferrin TE, (2004) UCSF Chimera–a visualization
system for exploratory research and analysis. J Comput Chem
25:1605–1612. http://www.cgl.ucsf.edu/chimera
Rozwarski DA, Grant GA, Barton DHR, Jacobs Jr WR, Sacchettini JC
(1998) Modification of the NADH of the isoniazid target (InhA)
from Mycobacterium tuberculosis. Science 279:98–102
Sullivan TJ, Truglio JJ, Boyne ME, Novichenok P, Zhang X, Stratton
CF, Li H-J, Kaur T, Amin A, Johnson F, Slayden RA, Kisker C,
Tonge PJ (2006) High affinity InhA inhibitors with activity
against drug-resistant strains of Mycobacterium tuberculosis.
ACS Chem Biol 1:43–53
Thomsen R, Christensen MH (2006) MolDock: a new technique
for high-accuracy molecular docking.
J Med Chem 49:
Rozwarski DA, Vilchèze C, Sugantino M, Bittman R, Sacchettini JC
(1999) Crystal structure of the Mycobacterium tuberculosis
enoyl-ACP reductase, InhA, in complex with NAD+ and a C16
fatty acyl substrate. J Biol Chem 274:15582–15589
3315–3321
World Health Organization (WHO) Global Tuberculosis Report 2016.
Geneva. Available at: http://www.who.int/tb/publications/global_
report/en/
Seefeld MA, Miller WH, Newlander KA, Burgess WJ, DeWolf WE,
Elkins PA, Head MS, Jakas DR, Janson CA, Keller PM,
Manley PJ, Moore TD, Payne DJ, Pearson S, Polizzi BJ,
Qiu X, Rittenhouse SF, Uzinskas IN, Wallis NG, Huffman WF
Zaoui A, Hammal I, Bennamane N, Merabtene S, Nedjar-Kolli B
(2013) A catalytic method for the synthesis of pyrazolone deri-
vatives using heteropolyacids and study of the anti-bacterial
activity. Int J pharm Chem Biol Sci 3:732–737