Design, synthesis and evaluation of small molecule reactive oxygen species generators as selective Mycobacterium tuberculosis inhibitors

Article abstract of DOI:10.1039/c2cc35343a

Here, we report 5-hydroxy-1,2,3,4,4a,9a-hexahydro-1,4-ethano-9,10- anthraquinone (13), a small molecule generating reactive oxygen species (ROS) in pH 7.4 buffer under ambient aerobic conditions that has selective and potent Mycobacterium tuberculosis gro

Full text of DOI:10.1039/c2cc35343a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 10325–10327
www.rsc.org/chemcomm
COMMUNICATION
Design, synthesis and evaluation of small molecule reactive oxygen
species generators as selective Mycobacterium tuberculosis inhibitorsw
Allimuthu T. Dharmaraja,a Mallika Alvala,b Dharmarajan Sriram,b Perumal Yogeeswarib
and Harinath Chakrapani*a
Received 24th July 2012, Accepted 20th August 2012
DOI: 10.1039/c2cc35343a
Here, we report 5-hydroxy-1,2,3,4,4a,9a-hexahydro-1,4-ethano-
9,10-anthraquinone (13), a small molecule generating reactive
oxygen species (ROS) in pH 7.4 buffer under ambient aerobic
conditions that has selective and potent Mycobacterium tuberculosis
growth inhibitory activity.
size ‘‘n’’ and functional groups X and Y could perturb keto–enol
equilibria and as a consequence, ROS yields, allowing us to study
the effects of varying ROS levels on Mtb growth inhibitory
activity. Finally, 1,4-quinones such as C are candidates for
bioreduction to regenerate B, which is capable of generating ROS.
In order to test our hypothesis, 2,3-dihydro-1,4-benzoquinones
1–12 were synthesized (ESIw) from commercially available starting
materials by a [4+2] cycloaddition between a quinone and
appropriate equivalents of a 1,3-dieneÀ(Fig. 1).
Tuberculosis (TB) remains a leading cause of morbidity and
mortality in much of the world. Mycobacterium tuberculosis
(Mtb), the etiological agent of this infectious disease, is a
highly adaptive pathogen and an extremely difficult pathogen
to kill. The alarming rise in co-infection with HIV as well as
extensively drug resistant (XDR) and multi-drug resistant
(MDR) strains1 necessitates new chemotherapeutic strategies.2
Maintenance of redox homeostasis is integral to the sustenance of
metabolic processes and mycobacterial persistence. Any significant
alterations in oxidative or reductive species might cause break-
down of this balance and lead to cell death.3 For example, elevated
ꢀ
The ability of 1–12 to generate O2 (Fig. 2a) and H2O2
(Fig. 2b) under ambient aerobic conditions in pH 7.4 buffer
was estimated using reported assays.9,10 The juglone derivative
4 generated the highest levels of ROS amongst 1–12. Based on
this result, we prepared a closely related analogue 13 by
hydrogenation of 4 (Fig. 1). This compound was found to be
superior to 4 in ROS generation for 30 min (Fig. 2a and b).
Anti-microbial effects of ROS are primarily mediated
through induction of DNA damage.4 Fe(II)-mediated nuclease
activity of 1–13 was estimated using a pBR322 supercoiled
plasmid cleavage assay and nick induction was observed in
nearly all compounds tested (Fig. 2c). The compound’s nuclease
activity correlated well with its ROS generating ability, indicating
that the observed DNA damaging effects were ROS-mediated
(Pearson’s correlation coefficient R = 0.9475, P-value o 0.0001,
Fig. S2, ESIw).
In order to test our hypothesis that ROS generators can inhibit
Mtb growth, a reported assay was used to determine minimum
inhibitory concentrations (MICs) for 1–13 against Mycobacterium
tuberculosis H37Rv (Table 1, entries 1–13).11 Amongst these, four
compounds: 3, 4, 6 and 13 were found to be good Mtb growth
inhibitors with MICs o 10 mg mLÀ1. Compounds 5 and 12 did
not show any detectable inhibitory activity at 100 mg mLÀ1
(Table 1, entries 5 and 12). MIC of the best ROS generator 13
was superior to those of clinically used first-line TB drugs
ethambutol and pyrazinamide (Table 1, entries 16–17) and presents
opportunities for further development as a TB drug candidate.
À
ꢀ
levels of reactive oxygen species (ROS) such as superoxide O2
,
hydrogen peroxide H2O2 and hydroxyl radical ꢀOH can cause
irreparable damage to biomacromolecules including DNA leading
to oxidative stress and might severely affect Mtb survival
(Scheme 1).4 A growing body of evidence points to normal human
cells’ ability to better tolerate alterations in redox homeostasis in
comparison with cancer cells5,6 and forms the basis of an emerging
paradigm for new redox-directed cancer chemotherapeutics.7
Thus, small molecules capable of perturbing mycobacterial redox
homeostasis through generation of ROS might have selective Mtb
growth inhibitory potential.4
2,3-Dihydro-1,4-benzoquinones (A) were considered as
potential ROS generators (Scheme 2). Conversion of A to its
tautomer 1,4-dihydroxyarene (B) might occur at physiological
pH; the diol B can undergo aerobic oxidation to give C with
À
8
ꢀ
concomitant production of O2 and H2O2. Variation of ring
a Department of Chemistry, Indian Institute of Science Education and
Research (IISER) Pune, Pune 411 008, Maharashtra, India.
E-mail: harinath@iiserpune.ac.in; Fax: +91-20-2589-9790;
Tel: +91-20-2590-8090
b Department of Pharmacy, Birla Institute of Technology & Science
Pilani, Hyderabad Campus, Jawahar Nagar, Hyderabad 500 078,
India
w Electronic supplementary information (ESI) available: Compound
characterization data and assay procedures with relevant plots. CCDC
893408. For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2cc35343a
Scheme 1 Reactive oxygen species (ROS) generated through reduction
of oxygen can damage biomacromolecules including DNA.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 10325–10327 10325

View Article Online
Compound 14, the exclÀusive organic product during incubation
ꢀ
of 13 did not generate O2 or H2O2 in aerobic buffer (Tables S1
and S3, ESIw) and neither did it induce significant DNA damage
(Fig. 2c). An increased intracellular fluorescence in the DCFH-
DA-based assay conducted in the presence of Mycobacterium
smegmatis MC2155 (Msm) was observed (Fig. S4, ESIw).
A quinone with a reduction potential in the range of À0.7 to
À1.1 V would be predicted to be a candidate for reduction by
enzymes such as NADPH:cytochrome P450 reductase15 and CV
analysis of 14 gave a reduction potential of À0.94 V (Fig. S9,
ESIw). Together, these data are consistent with a bioreductive
mode of activation of 14 to produce oxidative species
(Scheme 1). Quinone 14 inhibited Mtb growth with a MIC of
1.56 mg mLÀ1 (Table 1, entry 14) supporting our hypothesis that
direct generation of ROS by 13 (Fig. 1) in conjunction with
intracellular production of oxidative species by redox cycling by
14 could result in inhibition of Mtb growth (Scheme 2).
The ROS generator 13 was not a significant inhibitor
(IC50 > 25 mM) of normal human kidney (HEK) cells (Fig. S10,
ESIw). This result is consistent with anti-cancer oxidative stress
inducers which have shown selective toxicity towards cancers but
spared normal cells.5,6 The lead compound 13 (at 100 mg mLÀ1) did
not significantly inhibit several fast-growing Gram-positive
and Gram-negative bacterial strains tested (Table S8, ESIw).
The microbe selectivity of 13 parallels that of an artemisinin–
mycobactin drug conjugate which has shown high Mtb inhibitory
potency.17
Scheme 2 Design of compounds that can perturb redox homeostasis
by generating ROS.
Upon incubation of 13 at pH 7.4, HPLC analysis showed
complete disappearance of 13 with concomitant formation of 14
at comparable rates in a nearly quantitative HPLC yield (Fig. 3a
and b).12 A first order increase of H2O2 for 12 h was observed with
a maximal H2O2 yield of 72% (Fig. 3c). These results are consistent
with a tentative mechanism based on enolization of 2,3-dihydro-
1,4-naphthoquinones to produce a 1,4-diolate followed by reaction
À
ꢀ
with oxygen to sequentially produce O2 and H2O2 (Scheme S1,
ESIw).13,14 The presence of a proximal hydroxyl group appeared to
diminish carbonyl p-character (IR spectroscopy, Fig. S6, ESIw);
and enhanced electronegativity on the carbonyl oxygen (computa-
tional analysis, Fig. S7, ESIw). Together, these structural effects,
likely due to intramolecular H-bonding (Fig. S8, ESIw), could
promote enolization of C-4 carbonyl of 4 (or 13) leading to
enhanced ROS generation by these juglone derivatives in
comparison with other analogues without a proximal hydroxyl
group including the benzylated derivatives 8 and 9 (Fig. 1).
Fig. 1 Compounds 1–14.
À
Fig. 2 (a) Superoxide (O2 ꢀ) generated during incubation of 1–13 (5 mM) in pH 7.4 buffer at 37 1C for 30 min was estimated by a luminol-based
chemiluminescence assay. Hypoxanthine/xanthine oxidase was used as a positive control in this assay. (b) H2O2 generated during incubation of 1–13
(5 mM) in pH 7.4 buffer at 37 1C for 30 min was estimated by a xylenol orange-based colorimetric assay; (c) nuclease activity of 1–14: reaction mixtures
(20 mL) contained 100 ng of form I DNA in pH 8.0 phosphate buffer (100 mM) and compound (50 mM) were incubated at 37 1C for 6 h: lane 1, untreated
control; lane 2, 1; lane 3, 2; lane 4, 3; lane 5, 4; lane 6, 5; lane 7, 6; lane 8, 7; lane 9, 8; lane 10, 9; lane 11, 10; lane 12, 11; lane 13, 12; lane 14, 13; lane 15, 14.
Nicked DNA is represented by II. A relative increase in nuclease activity is defined as the ratio of form II (cleaved) to form I (uncleaved) in comparison
with untreated control (lane 1); (d) comparison of MICs determined against Mycobacterium tuberculosis and extracellular H2O2 yields during incubation
of that compound with Mycobacterium smegmatis for 1 h and Spearman’s correlation coefficient was found to be r = À0.91, P-value o 10À5
.
c
This journal is The Royal Society of Chemistry 2012
10326 Chem. Commun., 2012, 48, 10325–10327

View Article Online
Table 1 Extracellular ROS generation, Mycobacterium tuberculosis
growth inhibitory activity and calculated partition coefficients (clogPs)
MICa
MIC
(mM)
Extracellular
H2O2 (%)
b
Entry Compd
(mg mLÀ1
)
clogPc
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
25
111.0
52.4
13.0
3.0
>350
23.1
75.6
72.5
143.5
208.0
66.4
>350
3.0
3.6
4.2
5.8
6.8
1.9
4.3
2.5
4.2
2.6
1.0d
2.6
2.2d
11.0
3.8
—
2.43
2.99
2.76
3.32
2.72
3.28
4.29
4.85
0.43
1.93
0.99
3.04
3.81
4.52
À0.67
0.12
À0.68
12.5
3.13
0.76
>100
6.25
25
25
25
50
12.5
>100
0.76
1.56
0.05
1.56
Fig. 3 (a) HPLC traces of incubation of 13 in pH 7.4 buffer for 12 h show
complete conversion of 13 to 14; (b) time courses of disappearance of 13
(k13 = 0.24 Æ 0.01 hÀ1) and appearance of 14 (k14 was 0.28 Æ 0.04 hÀ1
)
based on HPLC analysis; (c) time course of H2O2 generated during
incubation of 13 (kHP = 0.7 hÀ1) in pH 7.4 buffer.
9
10
11
12
13
14
15
16
17
of a tuberculosis drug design strategy based on compounds
generating reactive oxygen species in order to perturb redox
homeostasis.
6.1
This work was supported in part by Department of Bio-
technology, India (BT/05/IYBA/2011). ATD acknowledges a
research fellowship from Council for Scientific and Industrial
Research (CSIR).
Isoniazid
Ethambutol
Pyrazinamide 6.25
0.37
7.64
50.8
—
—
a
Minimum inhibitory concentration (MIC) is the minimum concen-
tration of the compound required to inhibit 99% of bacterial growth
and was found against Mycobacterium tuberculosis H37Rv strain.
Notes and references
b
Extracellular H2O2 measured during incubation of 1–14 (50 mM)
1 R. S. Wallis, W. Jakubiec, M. Mitton-Fry, L. Ladutko, S. Campbell,
D. Paige, A. Silvia and P. F. Miller, PLoS One, 2012, 7, e30479.
2 L. G. Dover and G. D. Coxon, J. Med. Chem., 2011, 54, 6157–6165.
3 A. Kumar, A. Farhana, L. Guidry, V. Saini, M. Hondalus and
A. J. C. Steyn, Expert Rev. Mol. Med., 2011, 13, 1 (e39).
4 F. C. Fang, Nat. Rev. Microbiol., 2004, 2, 820–832.
5 L. Raj, T. Ide, A. U. Gurkar, M. Foley, M. Schenone, X. Li,
N. J. Tolliday, T. R. Golub, S. A. Carr, A. F. Shamji, A. M. Stern,
A. Mandinova, S. L. Schreiber and S. W. Lee, Nature, 2011, 475,
231–234.
6 J. S. Bair, R. Palchaudhuri and P. J. Hergenrother, J. Am. Chem.
Soc., 2010, 132, 5469–5478.
7 G. T. Wondrak, Antioxid. Redox Signaling, 2009, 11, 3015–3069.
8 J. M. Campos-Martin, G. Blanco-Brieva and J. L. G. Fierro,
Angew. Chem., Int. Ed., 2006, 45, 6962–6984.
9 H. Trung Pham, C. Marquetty, C. Pasquier and J. Hakim, Anal.
Biochem., 1984, 142, 467–472.
10 C. Gay, J. Collins and J. M. Gebicki, Anal. Biochem., 1999, 273,
149–155.
11 S. R. Malwal, D. Sriram, P. Yogeeswari, V. B. Konkimalla and
H. Chakrapani, J. Med. Chem., 2012, 55, 553–557.
12 J. A. Valderrama, O. Espinoza, M. F. Gonzalez, R. A. Tapia,
J. A. Rodriguez, C. Theoduloz and G. Schmeda-Hirschmann,
Tetrahedron, 2006, 62, 2631–2638.
13 K. U. Ingold, T. Paul, M. J. Young and L. Doiron, J. Am. Chem.
Soc., 1997, 119, 12364–12365.
14 J. Adamus, J. Gebicki, I. Ciebiada, E. Korczak and A. Denys,
J. Med. Chem., 1998, 41, 2932–2933.
15 M. H. Livertoux, P. Lagrange and A. Minn, Brain Res., 1996, 725,
207–216.
16 R. Singh, U. Manjunatha, H. I. Boshoff, Y. H. Ha, P. Niyomrattanakit,
R. Ledwidge, C. S. Dowd, I. Y. Lee, P. Kim, L. Zhang, S. Kang,
T. H. Keller, J. Jiricek and C. E. Barry, 3rd, Science, 2008, 322,
1392–1395.
in the presence of Msm at 37 1C for 1 h estimated by an Amplex red-
based fluorescence assay. Calculated values using Chembiodraw
c
d
Ultra. Calculated based on 2 mol H2O2 from 1 mol of compound.
At the outset our goal was two-fold: one, to design, synthe-
size and study compounds that were capable of generating
ROS; and two, to identify compounds with a range of ROS
generation profiles by structural modifications to study the
effects of varying ROS on Mtb growth. Having established
that compounds 1–13 were ROS generators in buffer, we
estimated ROS generated by 1–13 in the presence of cultured
Mycobacterium smegmatis MC2155 (Msm) and extracellular
H2O2 was estimated using an Amplex Red-based fluorescence
assay (Table 1). A good range of ROS in the presence of
mycobacteria (1–11%) was observed with 13 being the best
H2O2 generator (Table 1, entry 13). Statistical analysis
between the compound’s extracellular H2O2 yield in the
aforementioned assay and the compound’s measurable MIC
against Mtb revealed a strong negative correlation (Fig. 2d)
between these variables (Spearman’s rank correlation
coefficient r = À0.91, P-value o 10À5). This result is indicative
of ROS generating ability of the compound being important for
inhibitory effects but without detailed mechanistic investiga-
tions, a conclusive connection cannot be established. However,
previous studies show that the exquisite susceptibility of Mtb to
isoniazid, a clinically used TB drug that in part acts through
induction of oxidative stress, is attributable to inactivation of
oxyR, a regulatory system that activates >20 anti-oxidant
genes in response to ROS;18 while oxyR is present in most
actinobacteria, paradoxically, it is an inactive pseudogene in
Mtb suggesting that a coherent response to induction of
oxidative stress might be impaired.19 Precursors of reactive
species such as nitric oxide and sulfur dioxide have shown
Mtb inhibitory activity.11,16 Here, we demonstrate the feasibility
17 M. J. Miller, A. J. Walz, H. Zhu, C. Wu, G. Moraski,
U. Mollmann, E. M. Tristani, A. L. Crumbliss, M. T. Ferdig,
¨
L. Checkley, R. L. Edwards and H. I. Boshoff, J. Am. Chem. Soc.,
2011, 133, 2076–2079.
18 C. D. den Hengst and M. J. Buttner, Biochim. Biophys. Acta, 2008,
1780, 1201–1216.
´
19 E. Pagan-Ramos, S. S. Master, C. L. Pritchett, R. Reimschuessel,
M. Trucksis, G. S. Timmins and V. Deretic, J. Bacteriol., 2006,
188, 2674–2680.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 10325–10327 10327