Identification and characterisation of small-molecule inhibitors of Rv3097c-encoded lipase (LipY) of Mycobacterium tuberculosis that selectively inhibit growth of bacilli in hypoxia

Article abstract of DOI:10.1016/j.ijantimicag.2013.03.007

The mycobacterial Rv3097c-encoded lipase LipY is considered as a true lipase involved in the hydrolysis of triacylglycerol stored in lipid inclusion bodies for the survival of dormant mycobacteria. To date, orlistat is the only known LipY inhibitor. In view of the important emerging role of this enzyme, a search for small-molecule inhibitors of LipY was made, leading to the identification of some new compounds (8a-8d, 8f, 8h and 8i) with potent inhibitory activities against recombinant LipY, with no cytotoxicity [50% inhibitory concentration (CC₅₀) ≥500 μg/mL]. The compounds 6a, 8c and 8f potently inhibited (>90%) the growth of Mycobacterium tuberculosis H₃₇Rv grown under hypoxia (oxygen-depleted condition) but had no effect on aerobically grown bacilli, suggesting that these new small molecules are highly selective towards the growth inhibition of hypoxic cultures of M. tuberculosis and hence provide new leads for combating latent tuberculosis.

Full text of DOI:10.1016/j.ijantimicag.2013.03.007

International Journal of Antimicrobial Agents 42 (2013) 27–35
Contents lists available at SciVerse ScienceDirect
International Journal of Antimicrobial Agents
journal homepage: http://www.elsevier.com/locate/ijantimicag
Identification and characterisation of small-molecule inhibitors of
Rv3097c-encoded lipase (LipY) of Mycobacterium tuberculosis that
selectively inhibit growth of bacilli in hypoxiaଝ
Anil K. Saxenaa,∗, Kuldeep K. Roya, Supriya Singha, S.P. Vishnoia, Anil Kumarb,
Vivek Kr. Kashyapb, Laurent Kremerc, Ranjana Srivastavab, Brahm S. Srivastavab,∗∗
a Division of Medicinal and Process Chemistry, CSIR Central Drug Research Institute, Lucknow 226 001, India
b Microbiology Division, CSIR Central Drug Research Institute, Lucknow 226 001, India
c Laboratoire de Dynamique des Interactions Membranaires Normales et Pathologiques, Université de Montpellier II et I, CNRS, UMR 5235, case 107, Place
Eugéne Bataillon, 34095 Montpellier Cedex 05, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 October 2012
Accepted 17 March 2013
The mycobacterial Rv3097c-encoded lipase LipY is considered as a true lipase involved in the hydrolysis of
triacylglycerol stored in lipid inclusion bodies for the survival of dormant mycobacteria. To date, orlistat is
the only known LipY inhibitor. In view of the important emerging role of this enzyme, a search for small-
molecule inhibitors of LipY was made, leading to the identification of some new compounds (8a–8d, 8f,
8h and 8i) with potent inhibitory activities against recombinant LipY, with no cytotoxicity [50% inhibitory
concentration (CC50) ≥500 ␮g/mL]. The compounds 6a, 8c and 8f potently inhibited (>90%) the growth of
Mycobacterium tuberculosis H37Rv grown under hypoxia (oxygen-depleted condition) but had no effect
on aerobically grown bacilli, suggesting that these new small molecules are highly selective towards the
growth inhibition of hypoxic cultures of M. tuberculosis and hence provide new leads for combating latent
tuberculosis.
Keywords:
Mycobacterium
Rv3097c
Lipase
Inhibitor
Hypoxia
Latent TB
© 2013 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
1. Introduction
Determination of the complete genome sequence of the most
widely known H37Rv strain of M. tuberculosis has opened new per-
Tuberculosis (TB) is a globally challenging disease primarily
caused by Mycobacterium tuberculosis. Despite the availability of
highly efficacious treatment(s) for decades, TB remains a major
global health problem with an estimated 9 million new cases and 2
million deaths every year [1]. In addition, over 2 billion people har-
bour latent TB infection, thus representing an enormous reservoir
of M. tuberculosis that can subsequently progress to active TB. Effec-
tive treatment of TB can be achieved by combination of three to four
drugs [2]. However, the spread of TB remains unchecked for a num-
ber of reasons, including the emergence of multidrug-resistant TB
(MDR-TB) and extensively drug-resistant TB (XDR-TB), latent TB
infection, and its synergy with HIV/AIDS [1]. This highlights the
need for new therapies that act via a novel mechanism of action
to shorten the treatment duration of active drug-sensitive TB, of
MDR-TB and XDR-TB, and of persistent and latent bacilli [1].
spectives for developing novel drugs to treat TB [3]. Comparative
sequence analysis of the M. tuberculosis genome has revealed the
presence of nearly 250 genes encoding lipolytic enzymes involved
in lipid metabolism, a major pathway in Mycobacterium spp. [4].
The complex cell wall of mycobacteria contains unusual lipids that
are essential for survival and virulence [5,6]. In addition, the cyto-
sol of mycobacteria contains organites (lipid inclusion bodies) that
store important lipids [triacylglycerol (TAG)] for long-term sur-
vival during the dormancy phase [7]. Utilisation of the stored TAG
requires its hydrolysis and the release of fatty acids for catabolism
by ␤-oxidation.
In recent years, mycobacterial lipases have received consider-
able attention [8]. Deb et al. [9] cloned the putative lipase genes of
M. tuberculosis and expressed them in Escherichia coli. They com-
pared the enzyme activity of recombinant enzymes in vitro and
suggested that the newly identified Rv3097c-encoded enzyme LipY
was active as a true lipase [10]. Upregulation of LipY was observed
ex vivo in infected murine macrophages and in M. tuberculosis
growing in the lungs of infected mice [11,12]. Under conditions
of nutrient starvation and oxygen-depleted hypoxic conditions,
extensive LipY-induced breakdown of TAG was observed [9,13],
emphasising the role of LipY in survival of mycobacteria in dormant
ଝ
Central Drug Research Institute (CDRI) communication no.: 8431.
Corresponding author. Tel.: +91 522 2624 273; fax: +91 522 2623 938.
Corresponding author. Tel.: +91 522 2620 908.
E-mail addresses: anilsak@gmail.com (A.K. Saxena), drbrahm@gmail.com
∗
∗∗
(B.S. Srivastava).
0924-8579/$ – see front matter © 2013 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.
http://dx.doi.org/10.1016/j.ijantimicag.2013.03.007

28
A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
conditions. A strong humoral immune response against LipY
observed in clinical TB suggested the importance of the Rv3097c
antigen in immunopathogenicity of M. tuberculosis [10]. Interest-
ingly, extensive LipY-induced hydrolysis of TAG in Mycobacterium
bovis bacillus Calmette-Guerin (BCG) reduced the efficacy of the
BCG vaccine to protect against infection with M. tuberculosis, and
it was found that loss of efficacy of the vaccine correlated with
suppression of host immune responses [14]. A striking observa-
tion was made that synthesis of a unique phenolic glycolipid in
some strains of M. tuberculosis conferred a hypervirulent phe-
notype by inhibiting the innate immune response of the host
[15].
Cumulative results suggested that LipY is most likely the major
lipase involved in the hydrolysis of stored TAG. The N-terminal
domain of LipY is homologous to proteins belonging to the PE-
PGRS family, whereas the C-terminal resembles the conserved
hormone-sensitive lipase family. Pair-wise alignment of the amino
acid sequence of LipY with 35 representative lipases from all of
the eight reported families of bacterial lipases indicated that LipY
shared only 12–23% global amino acid identity with known bacte-
rial lipases [9]. To date, there is no knowledge of any compound
that can potently inhibit the mycobacterial active lipase encoded
by Rv3097c.
Orlistat, a US Food and Drug Administration (FDA)-approved
anti-obesity drug [16–18], has recently been shown to inhibit
and disrupt cell wall formation in several mycobacterial species
through inhibition of the essential M. tuberculosis lipase Rv3802c.
The three-dimensional (3D) structure of M. tuberculosis lipase
Rv3802c is yet to be elucidated, although a moderate-resolution
crystal structure of MSMEG 6394 [Protein Data Bank (PDB) ID: 3aja]
reported by Crellin et al. [19] is considered as a close homologue of
Rv3802c. West et al. [20] recently reported several lead compounds
exhibiting nanomolar inhibitory activity against the mycobacterial
lipase Rv3802c.
In view of the emerging role of Rv3097c-encoded lipase from
M. tuberculosis in latent TB, here we report identification of a
new class of Rv3097c-encoded lipase inhibitors. These new LipY
inhibitors selectively inhibited the growth of M. tuberculosis grow-
ing in hypoxia but not of aerobically growing bacilli and thus
represent an interesting scaffold for identification of candidate
molecules for the treatment of dormant bacilli.
2.3. Cloning and expression of the lipY gene
The lipY gene was amplified by PCR using M. tuberculosis
H37Rv genomic DNA and the following primers: upstream primer
pSDLipY-up 5ꢀ-AAA GGA TCC GTG TCT TAT GTT GTT GCG TTG-3ꢀ;
and downstream primer pSDLipY 5ꢀ-AAG GAT CCG GCG GCG ATA
CCG AGT TGC TG-3ꢀ [primers contain BamHI sites (underlined)].
The PCR-amplified gene was cloned at the BamHI site into pSD26.
The vector pSD26 is a M. smegmatis expression vector contain-
ing the mycobacterial inducible acetamidase promoter that allows
translational fusion with the 6× His tag [21]. The recombinant
plasmid was electroporated into M. smegmatis mc2155 strain and
then hygromycin-resistant colonies were selected. The presence of
the insert (1314 bp) was confirmed in transformants by restriction
digestion and PCR amplification.
2.4. Purification of recombinant protein
Mycobacterium smegmatis [pSD26::rv3097c] was grown in
100 mL of Middlebrook 7H9 (MB7H9) complete medium (Difco)
with hygromycin (Sigma, St Louis, MO) (50 ␮g/mL) at 37 ◦C to an
optical density at 600 nm of 0.6. The culture was divided into two
aliquots. One aliquot was induced with 0.2% acetamide (Sigma)
for 7 h at 37 ◦C, whilst the other formed an uninduced control
and received an equal volume of sterilised Milli-Q water. Cells
were harvested by centrifugation at 6000 rpm for 10 min at 4 ◦C,
washed with 20 mM Tris–HCl (pH 7.5) and re-suspended in 15 mL of
20 mM Tris–HCl (pH 7.5) containing phenylmethanesulfonyl fluo-
ride (PMSF). Cells were disrupted by sonication in ice for 10–15 min
and the lysate was centrifuged at 20 000 × g at 4 ◦C for 20 min to
separate the cytoplasmic fraction from the cell wall. It is known
that LipY is partitioned in the pellet [10]. Recombinant LipY (rLipY)
was purified using Ni-nitrilotriacetic (NTA) beads under denaturing
conditions by solubilising the inclusion body in 8 M urea, which did
not inhibit the activity of LipY. Briefly, the pellet was re-suspended
in 10 mL of binding buffer [20 mM Tris–HCl (pH 7.5), 200 mM NaCl,
0.1 mM PMSF and 8 M urea] and was incubated at 37 ◦C for 1 h
followed by centrifugation at 12 000 rpm for 20 min at 4 ◦C. The
supernatant was loaded on the Ni-NTA agarose column (3 mL bed
volume), pre-equilibrated with 15 mL of the binding buffer. The
flow through was passed five times through the Ni-NTA agarose
column. Unbound proteins were washed out by passing 15 mL of
binding buffer. The column with bound proteins was washed with
washing buffer (the binding buffer with 2, 5, 10 and 30 mM imid-
azole, 10 mL each). rLipY was eluted in the presence of 400 mM
imidazole and its molecular weight was determined by sodium
dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE).
2. Materials and methods
2.1. Chemistry
The synthesis of title compounds (Table 1) is outlined in Fig. 1.
Reaction of compound 1 with diethyl ethoxymethylenemalonate
(2) under reflux produced compound 3, which was then refluxed to
produce compound 4 (Scheme 1). Treatment of 4 with methanolic
ammonia under reflux yielded amide 5, which upon reaction with
3-phenylpropanal produced compound 6a, whereas reaction of
compound 4 with hydrazine hydrate produced hydrazide 7, which
upon reaction with benzaldehyde(s) produced compounds 8a–k
(Scheme 2).
2.5. Mycobacterial LipY inhibition assay
Lipase activity was assayed according to the method reported
by Mishra et al. [10], with some modifications, using p-nitrophenyl
stearate as substrate. Orlistat was used as a positive control. The
assay was performed in 96-well plates (in 200 ␮L of reaction
mixture) containing 50 ␮L (0.5 ␮g) of protein in elution buffer,
150 ␮L of sodium phosphate buffer saline [100 mM sodium phos-
phate buffer (pH 8.0), 200 mM NaCl] and a final concentration
of 0.5 mM p-nitrophenyl stearate (20 mM p-nitrophenyl stearate
stock solution prepared in acetonitrile). Test compounds were
dissolved in 10% dimethyl sulphoxide (DMSO). The mixture was
incubated for 1 h with intermittent shaking at 37 ◦C and the
release of p-nitrophenol was measured spectrophotometrically at
405 nm. The blank reaction was set with protein sample substi-
tuted by elution buffer. Assays were performed in triplicate and
protein activity was expressed as the average ‘nM of p-nitrophenol
2.2. Bacterial and plasmid strains
Mycobacterium tuberculosis H37Rv (TMC-702) was used as the
template for PCR amplification of the lipY gene. Mycobacterium
smegmatis mc2155 strain was received from Central JALMA Insti-
tute for Leprosy and other Mycobacterial Diseases (India). Vector
pSD26 was the expression vector for LipY in the M. smegmatis host
strain.
released/mg of protein/min’. The Michaelis–Menten constant (Km
)

A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
29
Table 1
Structures and in vitro inhibitory activities (IC50 values) against Mycobacterium tuberculosis LipY of the evaluated compounds (6a, 8a–k).
Compound
Chemical structure
XP GlideScore
LipY IC50 (mean S.E.M.) (␮M)
Vero cell CC50 (␮g/mL)
OH
O
F
Cl
N
6a
−5.30
25.00 1.31
>500
>500
N
Cl
OH
O
O
F
N
N
8a
8b
8c
−8.68
−6.64
−5.64
7.75 0.21
12.50 0.32
12.50 0.17
N
H
Cl
N
OH
F
N
H
137.89
>500
OH
Cl
N
OH
OH
O
O
O
F
N
N
N
N
H
Cl
N
OH
OH
OH
F
8d
8e
−6.63
−5.42
9.25 0.17
>500
N/D
N
H
Cl
N
O
OH
F
>100
N
H
Cl
N
O
O
OH
O
O
8f
−8.73
−5.67
−5.58
8.25 0.43
18.00 0.11
9.25 0.21
>500
N/D
N/D
F
N
N
N
H
Cl
N
O
O
OH
F
8g
8h
N
H
Cl
N
O
O
O
OH
O
F
N
N
H
Cl
N
Cl
OH
O
O
8i
8j
−9.26
−4.99
5.13 0.20
>500
F
N
N
N
H
Cl
N
OH
Cl
N
OH
F
>100
N/D
N
H
Cl
N

30
A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
Table 1 (Continued)
Compound
Chemical structure
XP GlideScore
LipY IC50 (mean S.E.M.) (␮M)
Vero cell CC50 (␮g/mL)
N
OH
O
F
N
8k
−5.37
>50
N/D
N
H
Cl
O
N
O
C11H23
H
N
O
Orlistat
<1.50
N/D
O
C6H13
O
IC50, concentration of inhibitor that caused 50% inhibition of the activity at 500 ␮M substrate concentration; CC50, % inhibitory concentration; S.E.M., standard error of the
mean; N/D, not determined.
Scheme 1: Synthesis of the intermediate compound 4a
F
O
O
O
F
i
O
O
O
Cl
N
H
O
+
Cl
NH2
O
O
3
1
2
ii
OH
N
O
F
O
Cl
4
a (i) Reflux, 3–5 h; (ii) Dowtherm, reflux at 250 °C.
Scheme 2: Synthesis of the title compounds (6a, 8a-k)a
OH
N
O
OH
O
F
NH2
F
iii
N
H
O
Cl
Cl
N
7
4
ii
i
OH
N
O
OH
N
O
F
N
Ar
F
N
H
NH2
Cl
Cl
5
8a–k
ii
OH
O
F
N
Cl
N
6a
a (i) Methanolic ammonia, reflux; (ii) aldehyde, acetic acid, ethanol, reflux; (iii) hydrazine hydrate, reflux.
Fig. 1. Outline for synthesis of title compounds.

A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
31
and maximum enzyme reaction rate (Vmax) of the purified rLipY
lipase enzyme were 317 ␮M and 514.54 nM/mg/min, respectively.
rLipY inhibitory activity was assayed at varying concentrations
(100–1.56 ␮g/mL) to determine the IC50 value, which was defined
as the concentration of inhibitor that caused 50% inhibition of the
activity at 500 ␮M substrate concentration.
oxygen in the medium and the cells were sensitive to metronida-
zole (MTZ) and resistant to isoniazid (INH). Compounds to be tested
were injected (100 ␮L/tube) at different concentrations and incu-
bated for 6 days at 37 ◦C under static conditions. Then, 350 ␮L of
0.02% resazurin was added to each tube and was incubated for
48 h at 37 ◦C. After this incubation, 200 ␮L aliquots from each tube
were transferred into a 96-well microplate and fluorescence was
determined.
2.6. Cytotoxicity assay
During these measurements, three controls were performed: (i)
without any drug; (ii) with MTZ; and (iii) with INH at different
concentrations [25]. Background fluorescence from medium (M)
and drug wells was subtracted. Percentage reduction in viability
was determined as: 1 − (test well fluorescence/mean fluorescence
of triplicate B wells) × 100 as described previously by Collins and
Franzblau [26].
The compounds were tested for cytotoxicity in an in vitro model
with Vero monkey kidney cells using the resazurin assay [22]. Vero
cells (ATCC CCL-81) were seeded overnight at 3 × 104 cells/well
in 96-well plates at 37 ◦C in RPMI supplemented with 10% heat-
inactivated fetal bovine serum and 5% CO2. Cells were exposed
to dilutions of experimental and control drugs in triplicate for 2 h
with each compound at a range of concentrations from 100 ␮g/mL
to 1.56 ␮g/mL. Rifampicin was used as a control at the same
concentration. Each well had 100 ␮L of the test material in seri-
ally descending concentrations. After 72 h of incubation, 10 ␮L of
resazurin indicator solution (0.1%) was added and incubation was
continued for 4–5 h. Any colour change from purple to pink or
colourless was recorded as positive. Fluorescence of each sample
was measured with an excitation wavelength at 530 nm and an
emission wavelength at 590 nm using a BMG POLARStar Galaxy
microplate reader (BMG Labtech, Cary, NC). CC50 values (50%
inhibitory concentrations) were calculated by plotting fluorescence
values using Microsoft Excel (Microsoft Corp., Redmond, WA).
2.8. Computational studies
The homology model of M. tuberculosis LipY protein was
developed using the Prime [27,28] module considering its clos-
est structural homologue triglyceride lipase from Staphylococcus
aureus (PDB ID: 3d7r) [29], sharing 22.6% sequence identity and
41.6% sequence similarity. The 3D structures of ligands were
sketched and prepared in Maestro using the optimised potentials
for liquid simulations–all atom (OPLS-AA)-2005 force field. Molec-
ular docking studies were carried out using Glide within a grid
generated using the O␥ atom of the Ser309 residue specifying an
˚
active site radius of 10 A. The predicted binding poses were ranked
using the XP GlideScore and the top-ranked poses were used for
further analysis.
2.7. Resazurin microtitre assay
Drug susceptibility testing using resazurin was performed under
aerobic and hypoxic conditions as described [23] for M. tuber-
culosis. Viable cells convert non-fluorescent resazurin dye to red
fluorescent resorufin dye in response to chemical reduction result-
ing from cell growth. Continued cell growth maintains a reduced
environment, whilst inhibition of growth maintains an oxidised
environment. Reduction related to growth causes the redox indi-
cator to change from the oxidised form (non-fluorescent, purple
colour) to the reduced form (fluorescent, red colour).
3. Results
3.1. Cloning, expression and purification of Rv3097c-encoded
LipY
Expression of recombinant Rv3097c-encoded LipY protein was
achieved in the acetamide-inducible vector pSD26 in a M. smegma-
tis host strain. rLipY protein was purified using Ni-NTA beads under
denaturing conditions by solubilising the inclusion body in 8 M
urea, which did not inhibit the activity of LipY. SDS-PAGE revealed a
single protein band of 47 kDa (Fig. 2). The purified protein was used
for characterisation of its enzymatic properties and for inhibitor
screening.
2.7.1. Assay under aerobic conditions
For aerobic cultures, the assay was performed in 96-well
microplates. Initial drug dilutions were prepared in either DMSO
or sterile deionised water, and subsequent two-fold serial dilutions
were performed in 0.1 mL of MB7H9-S/Dubos-S medium (Difco)
supplemented with 0.05% glycerol (Sigma) (without Tween 80)
in the microplates. Approximately 5 × 105 CFU of M. tuberculosis
3.2. Target-based biological screening and structure-based drug
design (SBDD)
H37Rv was added per well in a volume of 0.1 mL. Control wells
contained bacteria only (B), medium only (M) or drug only (to
detect autofluorescence of the drug) and were used to calculate
percentage inhibition of viability. Plates were incubated at 37 ◦C
for 6 days. Thereafter, 20 ␮L of 0.02% resazurin was added. The
wells were observed after 48 h for colour change from blue to pink.
The fluorescent signal was monitored using 530–560 nm excitation
wavelength and 590 nm emission wavelength in bottom-reading
mode.
Target-based in vitro screening of an in-house proprietary col-
lection of compounds against purified rLipY enzyme led to the
identification of an acyl-imine class of new molecule (6a) exhibit-
ing moderate inhibition of mycobacterial Rv3097c-encoded rLipY,
with an IC50 of 25 ␮M (Fig. 3). In the follow-up stage, design of a
new series of compounds considering the hit 6a was done using the
SBDD technique. The first theoretical model of Rv3097c-encoded
LipY was modelled based on the triglyceride lipase from S. aureus
(PDB ID: 3d7r). The local and global C␣-root mean square devi-
ations (RMSD) of the developed model and the template were
1.02 and 5.53, respectively. The developed LipY model contains
a canonical ␣/␤ hydrolase domain with an eight-stranded par-
allel ␤-sheet (␤1–␤8) surrounded by five ␣-helices (H3–H5, H7
and H8) and is comprised of a catalytic triad formed by Ser309,
Asp383 and His413 residues similar to the other serine esterases
(Fig. 4). The Ser309 residue was located at the bend of the tight
turn connecting the ␤5 strand and H5 helix, a feature known as
the ‘nucleophilic elbow’ that is conserved among all members of
2.7.2. Assay under hypoxia
Several 3 mL culture aliquots of M. tuberculosis containing
1.5 × 106 CFU/mL (A595 ∼ 0.003) were injected into 5 mL uncoated
Vacutainer® tubes and kept static at 37 ◦C to allow self-generation
of hypoxia. The medium contained the oxygen indicator methy-
lene blue (1.5 mg/L). Decolourisation of the medium suggested
depletion of dissolved oxygen as an indicator of hypoxia/anoxia.
As reported previously [24,25], a viable non-replicating persistent
culture could be obtained after 22 days with less than 1% dissolved

32
A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
the interface cavity (Fig. 5) explained well the observed inhibitory
activity of the compounds against mycobacterial rLipY (Table 1).
This was also supported by the quantitative structure–activity rela-
tionship (QSAR) studies between the observed rLipY inhibitory
activity (pIC50) and the computed physicochemical parameters for
hydrophobicity (ꢀꢁ), sterics (ꢀMR) and electronic (ꢀꢂ, ꢀF and
ꢀR), which revealed that the electronic parameter followed by
hydrophobicity influence the rLipY inhibitory activity variations
(Eq. (1)).
pIC50 = 0.517( 0.148)˙F + 4.709
N = 9, R = 0.796, R2 = 0.634, SEE = 0.133
(1)
3.3. In vitro cytotoxicity
Vero monkey kidney cells were used for evaluation of the in vitro
cytotoxicity of synthesised title compounds (Table 1) using the
resazurin assay, where the maximum concentration for the test
compounds was 500 ␮g/mL. The CC50 value, which is defined as the
concentration causing 50% reduction in the cell number compared
with that for the untreated controls, was determined for the tested
compounds. The observed CC50 values for all the tested compounds
are summarised in Table 1.
Fig. 2. Sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE)
(12.5%) analysis of recombinant Rv3097c (LipY) protein. Lane M, pre-stained protein
molecular weight ladder (MBI Fermentas, Amherst NY); lane 1, purified recombinant
LipY protein.
3.4. Inhibition of mycobacterial viability in aerobic and anaerobic
conditions
the ␣/␤ hydrolase superfamily, and is the nucleophile that attacks
the substrate to form a covalent acyl–enzyme intermediate. The
active site of LipY was comprised of four subsites, namely the cat-
alytic triad, an exposed wide pocket, a short deep pocket and an
extended hydrophobic channel (Fig. 5). Molecular docking of 6a
into the active site of the LipY model using Glide [30] revealed a
quinoline moiety being accommodated in the extended hydropho-
bic channel, and a 4-hydroxyl group formed two bonds with Trp339
and Asp414 residues, whereas the terminal phenyl ring linked to
amidic NH was accommodated in the short deep pocket (Fig. 5). The
acyl-imine linker was located in a highly polar region, where the
hydrazine linker (Fig. 3) was recognised to be a better substitute
through docking studies. Therefore, we embarked on the synthesis
and biological evaluation of the series of hydrazides 8a–8k, mak-
ing major variations in the benzylidene fragment. The in vitro rLipY
inhibitory activities and the XP GlideScore of these hydrazides are
summarised in Table 1. The majority of these hydrazides exhib-
ited improved rLipY inhibitory activity compared with 6a, thus
validating better suitability of the hydrazine than the acyl-imine
for binding with LipY. Compound 8i (IC50 = 5.13 ␮M) exhibited
the highest rLipY inhibitory activity, whilst another four com-
pounds (8a, 8d, 8f and 8h) also exhibited promising inhibition
of rLipY (IC50 = 7.75–9.25 ␮M). The nature of groups present on
the phenyl ring of the benzylidene in terms of hydrophobic and
electronic interactions with the residues present on the rim of
Selected compounds (6a, 8a–d, 8f–i) with rLipY IC50 values
≤25 ␮M were evaluated for their potential to inhibit the viability of
aerobically grown and anaerobic hypoxic cultures of M. tuberculosis
H37Rv at two concentrations (50 ␮g/mL and 100 ␮g/mL) using the
resazurin reduction assay. Compounds 6a, 8c and 8f caused >90%
viability inhibition of hypoxic cultures of bacilli, whereas com-
pounds 8a, 8b and 8i caused 28–53% inhibition at 50 ␮g/mL. These
compounds did not show any growth inhibition of aerobic cultures
up to 100 ␮g/mL (Table 2). In this study, the observed results with
the two controls, namely INH (a known inhibitor of aerobic culture)
and MTZ (a known inhibitor of anaerobic culture), were similar to
the reported results [20,25]. Importantly, compounds 6a, 8c and
8f exhibited promising growth-inhibitory activity of the dormant
bacilli grown under hypoxia but not of the bacilli grown under
aerobic condition, hence these new molecules can be considered
as selective inhibitors of the growth/replication of dormant bacilli
under hypoxia targeting LipY.
As the comparison of the relative enzyme inhibitory activities
in relation to inhibition of whole-cell mycobacterial viability did
not correlate, and the least active compound 6a in the enzyme
assay was almost equally effective as compound 8f, which was also
not the most active compound (8i) in the enzyme assay (Table 2),
it appeared that the rate-limiting factor for whole-cell activity
may be transport of the compound/drug. Hence, log P values of six
Fig. 3. Chemical structure and LipY inhibitory activity of the preliminary hit 6a, and the structural formula of the designed new prototype. tPSA, total polar surface area;
MW, molecular weight; IC50, concentration of inhibitor that caused 50% inhibition of the activity at 500 ␮M substrate concentration.

A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
33
Fig. 4. (A) Schematic representation and annotation of the Mycobacterium tuberculosis LipY protein model. The secondary structural elements are coloured based on the
position with respect to the N- (blue) and C- (red) terminals. (B) Schematic representation of the LipY protein after rotation by 90◦. (C) Top view of the active site of M.
tuberculosis LipY, depicting the orientations of the active site amino acids. The secondary structural elements are shown in schematic representation, coloured grey, and
labelled. Residues forming the catalytic triad are shown as green-coloured carbon, and the other conserved residues are shown as magenta-coloured carbon. (D) Surface
representation of the M. tuberculosis LipY protein, depicting the active site residues and the depth of the binding pocket. (For interpretation of the references to colour in this
figure legend, the reader is referred to the web version of this article.)
Table 2
Inhibition of viability of Mycobacterium tuberculosis H37Rv by selected compounds in aerobic and anaerobic (hypoxia) conditions.
Compound
Concentration (␮g/mL)
% Survival (mean S.D.)
Aerobic
log Pa
[log P − log P)0]2b
Hypoxia
6a
100
50
25
95.83 1.87
96.23 1.11
N/D
5.23 1.44
6.19 1.24
62.09 1.40
4.17
0.10
8a
8b
8c
100
50
94.57 0.89
95.06 0.66
41.37 1.61
47.06 1.37
4.33
3.39
3.39
0.23
0.21
0.21
100
50
96.67 0.23
98.12 0.39
46.72 1.39
54.09 1.89
100
50
25
97.45 2.83
97.67 2.54
N/D
5.02 3.51
7.65 3.87
60.4 1.34
100
50
25
97.07 1.52
97.51 0.43
N/D
4.06 0.85
5.52 4.06
56.6 1.15
3.52
4.50
0.11
0.42
8f
100
50
97.07 1.32
97.92 0.50
64.72 1.62
72.31 3.48
8i
INH
MTZ
1.6
0.8
4.11 2.35
4.59 2.42
88.21 1.36
93.84 1.24
128
64
92.5 1.25
91.77 1.16
7.54 2.17
14.09 0.71
S.D., standard deviation; N/D, not determined. INH, isoniazid; MTZ, metronidazole.
a
Log P values were calculated using ChemDraw software (http://www.cambridgesoft.com).
(log P)0 = 3.85 calculated from Eq. (1).
b

34
A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
Fig. 5. (A) Perspective view of the docked orientations of 12 characterised new compounds in the binding site of modelled LipY protein. (B–D) Binding pose of compounds
8a (B) 8f (C) and 8i (D), depicting important direct or indirect interactions of active compounds into the active site of the LipY model.
compounds (6a, 8a–c, 8f, 8i) were calculated using ChemDraw soft-
ware (http://www.cambridgesoft.com) and were correlated with
the whole-cell activity log [p/(100 − p)], where p denotes the per-
centage inhibition/killing (i.e. 100 − %survival).
4. Discussion
The mycobacterial Rv3097c-encoded lipase LipY is considered
as a true lipase, essential for the survival of dormant mycobac-
terium, and hence is an attractive target for the development of
drugs against latent TB. To date, the anti-obesity drug orlistat is the
only known inhibitor of this attractive target LipY.
log[p/(100 − p)] = −38.768( 18.006) log P
+ 5.042( 2.304)(log P)2 + 72.919
(2)
Biological screening of in-house compounds against purified
rLipY afforded an interesting acyl-imine class of compound 6a
(IC50 = 25 ␮M). Further synthesis and screening of 11 hydrazides
(8a–k) against this enzyme suggested some promising compounds
with better inhibitory activities than 6a. Five compounds (8a, 8d, 8f,
8h and 8i) have shown potent inhibition of rLipY, with IC50 values
in the range of 5.13–9.25 ␮M, where 8i (IC50 = 5.13 ␮M) showed the
highest rLipY inhibitory activity and approximately five-fold better
activity than 6a (IC50 = 25 ␮M) (Table 1). This suggested the better
suitability of the hydrazide linker compared with the acyl-imine.
Most of these compounds, except 8b (CC50 = 137.89 ␮g/mL), lacked
any cytotoxicity up to 500 ␮g/mL, suggesting that the observed
antimycobacterial activities are not due to any general cytotoxicity.
Molecular docking in combination with QSAR studies on these
compounds has revealed the particular importance of groups
N = 6, R = 0.830, R2 = 0.689, SEE = 0.520
The obtained parabolic correlation (Eq. (2)) with optimum
(log P)0 value of 3.85 well explained the observed whole-cell
activity data and further supported the hypothesis that trans-
port/penetration of the compound is the rate-limiting step for
whole-cell activity. Since the most active compound 8i with a
log P value of 4.50 has the highest deviation [log P − (log P)0 = 0.65]
from the optimum (log P)0, it may not be available within the
cell for binding with the lipase enzyme owing to lack of or
poor penetration. On the other hand, compounds 6a (log P = 4.17)
and 8f (log P = 3.52) with minimum deviations of 0.32 and 0.33,
respectively, are effective against dormant bacilli in the whole-cell
assay.

A.K. Saxena et al. / International Journal of Antimicrobial Agents 42 (2013) 27–35
35
present on the benzylidene group in terms of hydrophobic and
electronic interactions with the residues present on the rim of the
interface cavity (Fig. 5). The electronic parameter ꢀF contributes
positively and describes ca. 64% of the variance in enzyme activity.
Furthermore, compounds 6a, 8c and 8f exhibited >90% inhibi-
tion of viability of the dormant bacilli even at a concentration of
50 ␮g/mL, whilst 8a, 8b and 8i caused 28–53% inhibition of hypoxic
cultures. Interestingly, these compounds did not show any growth-
inhibitory effect on aerobic cultures up to 100 ␮g/mL and hence
these compounds may be considered as selective inhibitors of
the growth/replication of dormant bacilli under hypoxia targeting
Rv3097c-encoded lipase LipY. The observed parabolic relationship
between the whole-cell activity and log P with an optimum (log P)0
value of 3.85 (Eq. (2)) indicates that the rate-limiting factor for the
whole-cell activity is cell wall transport/penetration.
One of the major impediments to TB control is the ability of
M. tuberculosis to enclose in lung granuloma where the bacilli live
in a low-oxygen environment with slow metabolism and enter an
almost non-replicating state. The energy requirement is met by
hydrolysing TAG. Hence, the role of Rv3097c lipase appears very
important. Bacilli in granuloma become refractory to frontline anti-
TB drugs [2,9,10,13]. The lead compounds and data presented here
become significant because the compounds appear to inhibit LipY
lipase. However, it would have been interesting if the level of TAG
in the cell in the presence of the inhibitors was measured.
In conclusion, the present study reports the discovery of some
new small molecules as potent mycobacterial lipase inhibitors
exhibiting low micromolar inhibitory activities against mycobac-
terial rLipY. Some compounds selectively inhibited mycobacterial
growth in hypoxia condition and are better than MTZ. There-
fore, these compounds represent interesting scaffolds to medicinal
chemists for identification of promising candidate molecules for
the treatment of latent TB.
[8] Côtes K, Bakala N’goma JC, Dhouib R, Douchet I, Maurin D, Carrière F, et al.
Lipolytic enzymes in Mycobacterium tuberculosis. Appl Microbiol Biotechnol
2008;78:741–9.
[9] Deb C, Daniel J, Sirakova TD, Abomoelak B, Dubey VS, Kolattukudy PE. A novel
lipase belonging to the hormone-sensitive lipase family induced under starva-
tion to utilize stored triacylglycerol in Mycobacterium tuberculosis. J Biol Chem
2006;281:3866–75.
[10] Mishra KC, de Chastellier C, Narayana Y, Bifani P, Brown AK, Besra GS,
et al. Functional role of the PE domain and immunogenicity of the Mycobac-
terium tuberculosis triacylglycerol hydrolase LipY. Infect Immun 2008;76:
127–40.
[11] Srivastava V, Rouanet C, Srivastava R, Ramalingam B, Locht C, Srivastava
BS. Macrophage-specific Mycobacterium tuberculosis genes: identification by
green fluorescent protein and kanamycin resistance selection. Microbiology
2007;153:659–66.
[12] Srivastava V, Jain A, Srivastava BS, Srivastava R. Selection of genes of Mycobac-
terium tuberculosis upregulated during residence in lungs of infected mice.
Tuberculosis (Edinb) 2008;88:171–7.
[13] Low KL, Rao PS, Shui G, Bendt AK, Pethe K, Dick T, et al. Triacylglycerol utilization
is required for regrowth of in vitro hypoxic nonreplicating Mycobacterium bovis
bacillus Calmette-Guerin. J Bacteriol 2009;191:5037–43.
[14] Singh VK, Srivastava V, Singh V, Rastogi N, Roy R, Shaw AK, et al. Overexpression
of Rv3097c in Mycobacterium bovis BCG abolished the efficacy of BCG vac-
cine to protect against Mycobacterium tuberculosis infection in mice. Vaccine
2011;29:4754–60.
[15] Reed MB, Domenech P, Manca C, Su H, Barczak AK, Kreiswirth BN, et al. A gly-
colipid of hypervirulent tuberculosis strains that inhibits the innate immune
response. Nature 2004;431:84–7.
[16] Hadváry P, Lengsfeld H, Wolfer H. Inhibition of pancreatic lipase in vitro by the
covalent inhibitor tetrahydrolipstatin. Biochem J 1988;256:357–61.
[17] Hadváry P, Sidler W, Meister W, Vetter W, Wolfer H. The lipase inhibitor
tetrahydrolipstatin binds covalently to the putative active site serine of pan-
creatic lipase. J Biol Chem 1991;266:2021–7.
[18] Pemble 4th CW, Johnson LC, Kridel SJ, Lowther WT. Crystal structure of the
thioesterase domain of human fatty acid synthase inhibited by orlistat. Nat
Struct Mol Biol 2007;14:704–9.
[19] CrellinPK, VivianJP, ScobleJ, ChowFM, WestNP, BrammananthR, etal. Tetrahy-
drolipstatin inhibition functional analyses, and three-dimensional structure of
a lipase essential for mycobacterial viability. J Biol Chem 2010;285:30050–60.
[20] West NP, Cergol KM, Xue M, Randall EJ, Britton WJ, Payne RJ. Inhibitors of an
essential mycobacterial cell wall lipase (Rv3802c) as tuberculosis drug leads.
Chem Commun (Camb) 2011;47:5166–8.
[21] Daugelat S, Kowall J, Mattow J, Bumann D, Winter R, Hurwitz R, et al. The RD1
proteins of Mycobacterium tuberculosis: expression in Mycobacterium smegma-
tis and biochemical characterization. Microbes Infect 2003;5:1082–95.
[22] Anoopkumar-Dukie S, Carey JB, Conere T, O’Sullivan E, van Pelt FN, All-
shire A. Resazurin assay of radiation response in cultured cells. Br J Radiol
2005;78:945–7.
Funding: No funding sources.
Competing interests: None declared.
Ethical approval: Not required.
[23] Taneja NK, Tyagi JS. Resazurin reduction assay for screening of anti-tubercular
compounds against dormant and actively growing Mycobacterium tubercu-
losis, Mycobacterium bovis BCG and Mycobacterium smegmatis. J Antimicrob
Chemother 2007;60:288–93.
[24] Saxena A, Srivastava V, Srivastava R, Srivastava BS. Identification of genes
of Mycobacterium tuberculosis upregulated during anaerobic persistence
by fluorescence and kanamycin resistance selection. Tuberculosis (Edinb)
2008;88:518–25.
[25] Wayne LJ, Smarek HA. Metronidazole is bactericidal to dormant cells of
Mycobacterium tuberculosis. Antimicrob Agents Chemother 1994;38:2054–8.
[26] Collins L, Franzblau SG. Microplate alamar blue assay versus BACTEC 460
system for high-throughput screening of compounds against Mycobac-
terium tuberculosis and Mycobacterium avium. Antimicrob Agents Chemother
1997;41:1004–9.
[27] Prime, version 2.1. New York, NY: Schrödinger, LCC; 2009.
[28] Schrödinger, version 8.0. New York, NY: Schrödinger, LCC; 2009.
[29] Battaile KP, Lam R, Chan T, Romanov V, Lam K, Soloveychik M, et al.
References
[1] World Health Organization. WHO report 2011. Global tuberculosis control.
Geneva, Switzerland: WHO; 2011.
[2] Ginsberg AM. Tuberculosis drug development: progress, challenges, and the
road ahead. Tuberculosis (Edinb) 2010;90:162–7.
[3] Cole ST, Brosch R, Parkhill J, Garnier T, Churcher C, Harris D, et al. Deciphering
the biology of Mycobacterium tuberculosis from the complete genome sequence.
Nature 1998;393:537–44.
[4] Camus JC, Pryor MJ, Médigue C, Cole ST. Re-annotation of the genome sequence
of Mycobacterium tuberculosis H37Rv. Microbiology 2002;148:2967–73.
[5] Daffé M, Draper P. The envelope layers of mycobacteria with reference to their
pathogenicity. Adv Microb Physiol 1998;39:131–203.
[6] Guenin-Macé L, Siméone R, Demangel C. Lipids of pathogenic mycobacteria:
contributions to virulence and host immune suppression. Transbound Emerg
Dis 2009;56:255–68.
Crystal structure of
a putative esterase from Staphylococcus aureus.
[7] Neyrolles O, Hernández-Pando R, Pietri-Rouxel F, Fornès P, Tailleux L, Barrios-
http://dx.doi.org/10.2210/pdb3d7r/pdb
Payán JA, et al. Is adipose tissue
persistence? PLoS ONE 2006;1:e43.
a place for Mycobacterium tuberculosis
[30] Glide, version 5.0. New York, NY: Schrödinger, LCC; 2008.