Benzo[d]isothiazole 1,1-dioxide derivatives as dual functional inhibitors of 5-lipoxygenase and microsomal prostaglandin E2 synthase-1

Article abstract of DOI:10.1016/j.bmcl.2014.04.006

A series of 6-nitro-3-(m-tolylamino) benzo[d]isothiazole 1,1-dioxide analogues were synthesized and evaluated for their inhibition activity against 5-lipoxygenase (5-LOX) and microsomal prostaglandin E₂ synthase (mPGES-1). These compounds can inhibit both enzymes with IC₅₀ values ranging from 0.15 to 23.6 μM. One of the most potential compounds, 3g, inhibits 5-LOX and mPGES-1 with IC₅₀ values of 0.6 μM, 2.1 μM, respectively.

Full text of DOI:10.1016/j.bmcl.2014.04.006

Bioorganic & Medicinal Chemistry Letters 24 (2014) 2764–2767
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Benzo[d]isothiazole 1,1-dioxide derivatives as dual functional
inhibitors of 5-lipoxygenase and microsomal prostaglandin
E2 synthase-1
a
c
a
a
Erchang Shang a, Yiran Wu a,b, Pei Liu a, Ying Liu a,b, , Wei Zhu , Xiaobing Deng , Chong He , Shan He ,
⇑
Cong Li a, Luhua Lai a,b,c
a BNLMS, State Key Laboratory for Structural Chemistry of Unstable and Stable Species, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
b Center for Quantitative Biology, Peking University, Beijing 100871, China
c Peking-Tsinghua Center for Life Sciences, Peking University, Beijing 100871, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 6-nitro-3-(m-tolylamino) benzo[d]isothiazole 1,1-dioxide analogues were synthesized and
evaluated for their inhibition activity against 5-lipoxygenase (5-LOX) and microsomal prostaglandin E2
synthase (mPGES-1). These compounds can inhibit both enzymes with IC50 values ranging from 0.15
Received 4 December 2013
Revised 20 March 2014
Accepted 3 April 2014
Available online 13 April 2014
to 23.6
lM. One of the most potential compounds, 3g, inhibits 5-LOX and mPGES-1 with IC50 values of
0.6 M, 2.1
l
l
M, respectively.
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Benzo[d]isothiazole 1,1-dioxide
5-Lipoxygenase
Microsomal prostaglandin E2 synthase
Dual functional inhibitor
For inflammation and related disorders, leukotrienes and pros-
taglandins are two classes of important mediators produced from
arachidonic acid (AA).1 Leukotrienes are potent mediators of
inflammatory and allergic reactions that are generated by the 5-
lipoxygenase (5-LOX) pathway.2 5-LOX is a non-heme iron dioxy-
genase and catalyzes the transformation from AA to leuktriene A4
(LTA4), which is further mediated to other proinflammatory medi-
ators such as leukotriene B4 (LTB4) and leukotriene C4 (LTC4)3,4
Thus 5-LOX inhibition represents an attractive strategy for thera-
peutic intervention. Though many 5-LOX inhibitors have been
reported, currently only one compound with micromolar activity,
zileuton, is on market for treating asthma.5
Prostaglandins are generated by the cyclooxygenase (COX)
pathway. Among the prostaglandins, prostaglandin E2 (PGE2) is
the most common inflammatory mediator and plays crucial roles
in various biological events, such as neuronal functions, vascular
hypertension, fever, pain hypersensitivity, and tumor genesis.6,7
PGE2 is synthesized from COX product prostaglandin H2 (PGH2)
by terminal PGE2 synthases (PGES). Three isoforms of PGES have
been identified: cytosolic prostaglandin E2 synthase (cPGES) and
two microsomal prostaglandin E2 synthases (mPGES-1 and
mPGES-2). mPGES-1 is an inducible enzyme and responsible for
the elevated levels of PGE2 in inflammatory stimuli.8 In recent
years, mPGES-1 is identified as a novel anti-inflammatory thera-
peutic target.7
Traditional nonsteroidal anti-inflammatory drugs (NSAIDs,
which inhibit both two isoforms of COX, COX-1 and COX-2) and
COX-2 selective inhibitors (coxibs) are widely used to treat corre-
sponding disorders. However, the use of NSAIDs is associated with
gastrointestinal side effect and coxibs exert cardiovascular side
effects.9,10 According to the simulation result of the AA metabolic
network dynamics, inhibiting multiple targets in the network
may be effective in controlling the inflammation mediators and
reduce the side effects.11 Licofelone, a compound which can inhibit
both 5-LOX and COX pathways, successfully completed phase III
trials for treating osteoarthritis.12 Furthermore, our recent study
on the effects of inhibitor combinations on the AA network verified
that the 5-LOX and mPGES-1 is a potential optimal dual target
combination.13
Several 5-LOX and mPGES-1 dual functional inhibitors have
been reported, such as pirinixic acid derivatives, arylpyrrolizines,
modified acidic nonsteroidal anti-inflammatory drugs and
5-hydroxyindole-3-carboxylates.14–17 However, there is no effec-
tive 5-LOX and mPGES-1 inhibitor used in clinic yet. Development
⇑
Corresponding author. Tel.: +86 10 62751490; fax: +86 10 62751725.
E-mail address: liuying@pku.edu.cn (Y. Liu).
http://dx.doi.org/10.1016/j.bmcl.2014.04.006
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

E. Shang et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2764–2767
2765
Table 1
of novel highly efficient 5-LOX and mPGES-1 dual functional inhib-
5-LOX and mPGES-1 inhibitory activity
itors is highly demanded. In previous research, we identified a ser-
ies of novel inhibitors for 5-LOX through virtual screening, and two
of them appeared to be potential dual functional inhibitors of
5-LOX and mPGES-1. The two inhibitors provided new scaffolds
to design multi-target anti-inflammatory drugs.18 One of the dual
functional inhibitors is 6-nitro-3-(m-tolylamino) benzo[d]isothiaz-
O
O
N
O2N
S
N
H
R
c
Compd
R
In vitro IC50 (lM)
ole 1,1-dioxide (JMC-7) (IC50 values are 1.9
lM for 5-LOX and
5-LOX
mPGES-1
6.7 M for mPGES-1). No application has been reported for this
l
Zileutona
MF63b
1.1 0.5
–
–
compound. Herein we synthesized a series of JMC-7 analogues
and evaluated their inhibition activity toward 5-LOX and mPGES-1.
It is highly difficult to improve the potency of a dual-functional
ligand to both targets simultaneously. To identify key interactions
between JMC-7 and the two targets, we docked the compound into
the substrate binding sites of 5-LOX and mPGES-1. The crystal
structure of 5-LOX (PDB entry: 3O8Y19) and a model of active con-
formation of mPGES-1 built by our group were used.20 Molecular
docking with flexible ligands and rigid receptor was performed
0.0030 0.0001
6.7 0.2
JMC-7
3a
1.9 0.1
8.6 0.7
>50
Me
23.6 4.9
5.7 1.2
3b
Cl
Br
3c
3.4 0.5
7.3 1.5
Me
O
3d
12.3 3.8
16.9 3.5
Me
O
3e
3f
23.5 6.5
>50
>50
O
Me
>50
Br
3g
0.6 0.3
2.1 0.4
Me
3h
3i
>50
>50
>50
0.25 0.05
Me
Me
Me
3j
0.06 0.02
0.24 0.02
>50
Me
Me
3k
8.8 1.3
O
Me
3l
0.7 0.9
>50
>50
3m
0.15 0.06
N
3n
3o
>50
>50
>50
Br
Figure 1. Key interactions of JMC-7 (magenta) with 5-LOX (A, yellow) and mPGES-1
(B, cyan; letters following residue names stand for chain ID) (predicted by
molecular docking).
14.9 0.8
O
O
a
b
c
Positive control of 5-LOX IC50 values.
Positive control of mPGES-1.
Data are given as mean SE of n P 3 determinations.
O
O
O
O
O
O
O2N
S
O2N
S
O2N
S
b
a
N
NH
N
N
H
R
O
Cl
with the program AutoDock 4.00.21 Lamarckian genetic algorithm
was used with following parameters: number of individuals in
population, 300; maximum number of energy evaluations,
1
2
3a~3o
Scheme 1. Regents and conditions: (a) SOCl2, DMF (cat.), dioxane, 110 °C, 1.5 h; (b)
RNH2, NEt3, THF, rt, 4ꢀ6 h.

2766
E. Shang et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2764–2767
25,000,000; maximum of generations, 27,000, number of runs, 256.
Docked conformations of the 256 runs were clustered with a rms
tolerance of 2.0 Å, and the lowest energy conformation from the
largest cluster was taken as docking result.
5-LOX efficiently, indicating that flexible aryl group is not favored
at the hydrophobic moiety. Compound 3h with o-methyl group
loses the inhibition activity to 5-LOX. Among the compounds with
two-ring hydrophobic moiety, 3g (with 1-naphthyl) and 3l (with 3-
biphenyl) afford 3-folds increase in potency, while 3m, 3n, and 3o
lose the activities. Thus larger multiple ring groups at this position
may increase the interaction between the compound and 5-LOX.
The IC50 values of the compounds to inhibit mPGES-1 are also
listed in Table 1. Protein expression and enzyme activity were
measured using our reported protocols.20 Assessment of PGE2 con-
version from PGH2 was measured using the PGE2 EIA kit (Cayman
Chemical) for enzyme activity. A known mPGES-1 inhibitor, MF63,
was used as the positive control. Compound with smaller hydro-
phobic group (phenyl, 3a) has decreased inhibitory activity (IC50
Key interactions predicted by molecular docking are shown in
Figure 1. For 5-LOX, JMC-7 forms hydrogen bonds using the nitro
group to the main-chain nitrogen atoms of Lys423 and Ala424
and using the amino nitrogen atom to the hydroxyl group of
Tyr181. The aryl rings of JMC-7 form hydrophobic interactions
with the residues inside the cavity. For mPGES-1, hydrogen bonds
form between the nitro group and Arg126, and the amino nitrogen
atom and Glu77. There is also hydrophobic interaction between
the aryl rings of JMC-7 and residues inside the cavity. Thus both
the 6-nitro group and the amino nitro atom of JMC-7 are crucial
for specific binding of the compound to the targets. The aryl rings
of JMC-7 participate in forming hydrophobic interaction. As there
is extra space close to the m-tolyl group in both pockets, we opti-
mized the compound at this moiety.
23.6 lM). For the meta-position, smaller substitution groups (halo-
gen atom and methoxy) has similar activity to the parent com-
pound with methyl at this position the activity of the compounds
(3b, IC50 5.7 lM; 3c, IC50 7.3 lM; 3k, 8.8 lM), while larger substi-
Analogues of JMC-7 with m-tolyl replaced by different hydro-
phobic groups were synthesized. The synthetic route was shown
in Scheme 1.22 Commercially available 6-nitrosaccharin (1) was
initially reacted with sulfuryl chloride to generate the intermediate
3-chloro-6-nitrobenzo[d]isothiazole 1,1-dioxide (2). Subsequent
displacement of the chloride from this intermediate by a variety
of aromatic amines gave the target compounds (3aꢀ3o) in moder-
ate to high yields. 1H NMR, 13C NMR and HRMS data of all the syn-
thesized compounds were in full agreement with the proposed
structures (see Supporting Information).
All synthesized compounds were tested for inhibitory activities
to 5-LOX in vitro. Protein expression and enzyme activity determi-
nation were performed using the same method as in our previous
study.18 The enzyme assay measures the fluorescent signal from
the oxidation of an indicator, 2,7-dichlorodihydrofluorescein diac-
etate, to the highly fluorescent 20,70-dichlorofluorescein by the 5-
LOX enzyme reaction product.23 Zileuton was used as a positive
control. The IC50 values of all the compounds are shown in Table 1.
Most of JMC-7 derivatives preserve 5-LOX inhibitory activities.
Compound 3a with phenyl group results in a little loss of potency
tuent i-propyl (3i) or t-butyl (3j) reduce their activities.
Compounds with methoxyl at the para-position (3d) or both meta-
and para-positions (3e) are inactive, as well as methyl at the ortho-
position (3h) and flexible aryl group (3f) at this moiety. In this test,
the 1-naphthyl derivative (3g) is also 3-fold more active than JMC-
7, while 5-quinolinyl derivative (3m) possesses 45-fold increasing
activity. The rest compounds with two-ring hydrophobic moiety
(3n and 3o) lose the activities.
According to the inhibitory activities listed above, several com-
pounds maintain comparable activity for inhibiting both enzymes.
The optimized moiety affects the activity of the compounds to both
5-LOX and mPGES-1. To increase hydrophobic interaction, substi-
tution group with proper size on phenyl is needed, and the meta-
position is validated. Flexible aryl groups at this moiety decrease
the inhibitory activities to both targets. Compounds with two-ring
hydrophobic moiety may also be preferred. The strongest 5-LOX
inhibitor is 3l, however shows no inhibitory activity to mPGES-1.
Also the strongest mPGES-1 inhibitor 3m is inactive in inhibiting
5-LOX. The most balanced inhibitor, 3g, has approximately 3-folder
higher activity than JMC-7 to both 5-LOX (IC50 0.6
lM) and
(IC50 8.6
l
M). This may due to size decreasing of the hydrophobic
mPGES-1 (IC50 2.1
in Figure 2.
lM). The dose-effect curves of 3g are shown
group. Replacement of methyl group with bromine produces 3-((3-
bromophenyl) amino)-6-nitrobenzo[d]isothiazole 1,1-dioxide (3c),
which possesses similar inhibition activity (IC50 3.4 lM) to JMC-7.
The replacement of m-methyl with other large substituent, such as
i-propyl (3i), t-butyl (3j), and methoxy (3k), show significant
increasing of inhibition to 5-LOX, while 3j is the strongest
To analyze the interactions between 3g and the targets, it was
docked also into the inhibitor binding pockets. The docking results
show that 3g employs the similar binding modes as JMC-7 does in
the both targets. The interactions of 3g to the targets were illus-
trated using LIGPLOT figures (Figure 3).24 All the hydrogen bonds
between 3g and 5-LOX remained. For mPGES-1, the hydrogen bond
formed with the amino nitro atom disappeared because of a slight
position shift of 3g, while the other two remained. Moreover, com-
pared with the parent compound JMC-7, 3g has stronger hydro-
phobic interaction to the pockets of 5-LOX and mPGES-1, the
ortho-position may also be potential. Therefore, hydrophobic inter-
action is essential for the activity enhancing of 3g. This may
(0.06 lM). These results further confirm hydrophobic substituent
in the meta-position enhance the interaction of the compounds
with 5-LOX. However, the only exception is the chloride analogue
3b, which is inactive. The introduction of methoxyl group into the
para-position (3d) and dimethoxyl group into the meta- and para-
positions (3e) reduces the activities of the compounds. In addition,
substitution of m-tolyl by m-bromobenzyl (3f) cannot inhibit
Figure 2. Dose-effect curves of compound 3g. (A) Inhibiting 5-LOX; (B) inhibiting mPGES-1.

E. Shang et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2764–2767
2767
Figure 3. Interactions of compound 3g with 5-LOX (A) and mPGES-1 (B) (generated using program LIGPLOT24).
provide useful guidance for the design of highly active 5-LOX and
References and notes
mPGES-1 dual functional inhibitors. Further studies are under
investigation.
1. Funk, C. D. Science 2001, 294, 1871.
2. Steinhilber, D. Curr. Med. Chem. 1999, 6, 71.
3. Rådmark, O.; Samuelsson, B. J. Lipid Res. 2009, 50, S40.
4. Rådmark, O.; Werz, O.; Steinhilber, D.; Samuelsson, B. Trends Biochem. Sci. 2007,
32, 332.
5. Pergola, C.; Werz, O. Expert Opin. Ther. Pat. 2010, 20, 355.
6. Murakami, M.; Kudo, I. Prog. Lipid Res. 2004, 43, 3.
7. Friesen, R. W.; Mancini, J. A. J. Med. Chem. 2008, 51, 4059.
8. Sampey, A. V.; Monrad, S.; Crofford, L. J. Arthritis Res. Ther. 2005, 7, 114.
9. Rainsford, K. D. Subcell. Biochem. 2007, 3.
In conclusion, we synthesized a series of analogues of 6-nitro-3-
(m-tolylamino) benzo[d]isothiazole 1,1-dioxide (JMC-7), as dual
functional inhibitors of 5-LOX and mPGES-1. Through this optimi-
zation, we obtained several new dual functional inhibitors to 5-LOX
and mPGES-1. One of the compounds, 3-(naphthalen-1-ylamino)-
6-nitrobenzo[d]isothiazole 1,1-dioxide (3g), showed improved
inhibition activities to both 5-LOX and mPGES-1. These compounds
provide a new starting point for novel anti-inflammatory drug
development.
10. McGettigan, P.; Henry, D. JAMA 2006, 296, 1633.
11. Yang, K.; Bai, H.; Ouyang, Q.; Lai, L.; Tang, C. Mol. Syst. Biol. 2008, 4.
12. Steinhilber, D.; Hofmann, B. Basic Clin. Pharmacol. Toxicol. 2013.
13. He, C.; Wu, Y.; Lai, Y.; Cai, Z.; Liu, Y.; Lai, L. Mol. BioSyst. 2012, 8, 1585.
14. Koeberle, A.; Zettl, H.; Greiner, C.; Wurglics, M.; Schubert-Zsilavecz, M.; Werz,
O. J. Med. Chem. 2008, 51, 8068.
Acknowledgments
15. Liedtke, A. J.; Keck, P. R.; Lehmann, F.; Koeberle, A.; Werz, O.; Laufer, S. A. J. Med.
Chem. 2009, 52, 4968.
16. Elkady, M.; Niess, R.; Schaible, A. M.; Bauer, J.; Luderer, S.; Ambrosi, G.; Werz,
O.; Laufer, S. A. J. Med. Chem. 2012, 55, 8958.
This research was supported, in part, by the Ministry of Science
and Technology of China (Grant numbers: 2012AA020308,
2009CB918503) and the National Science Foundation of China
(Grant numbers: 20873003, 11021463).
17. Koeberle, A.; Haberl, E.-M.; Rossi, A.; Pergola, C.; Dehm, F.; Northoff, H.;
Troschuetz, R.; Sautebin, L.; Werz, O. Bioorg. Med. Chem. 2009, 17, 7924.
18. Wu, Y.; He, C.; Gao, Y.; He, S.; Liu, Y.; Lai, L. J. Med. Chem. 2012, 55, 2597.
19. Newcomer, M. E.; Gilbert, N. C.; Bartlett, S. G.; Waight, M. T.; Neau, D. B.;
Boeglin, W. E.; Brash, A. R. Science 2011, 331, 217.
20. He, S.; Li, C.; Liu, Y.; Lai, L. J. Med. Chem. 2013, 56, 3296.
21. Autodock, Version 4; The Scripps Research Institute: La Jolla, CA. 2007.
22. Haffner, C. D.; Thomson, S. A.; Guo, Y.; Schaller, L. T.; Boggs, S.; Dickerson, S.;
Gobel, J.; Gillie, D.; Condreay, J. P. Bioorg. Med. Chem. Lett. 2010, 20, 6983.
23. Pufahl, R. A.; Kasten, T. P.; Hills, R.; Gierse, J. K.; Reitz, B. A.; Weinberg, R. A.;
Masferrer, J. L. Anal. Biochem. 2007, 364, 204.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2014.04.006.
These data include MOL files and InChiKeys of the most important
compounds described in this article.
24. Wallace, A. C.; Laskowski, R. A.; Thornton, J. M. Protein Eng. 1995, 8, 127.