Tricyclic 4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imidazole derivatives as microsomal prostaglandin E2 synthase-1 (mPGES-1) inhibitors: SAR and in vivo efficacy in hyperalgesia pain model

Article abstract of DOI:10.1016/j.bmcl.2017.03.068

A series of substituted tricyclic 4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imidazole derivatives have been synthesized and their mPGES-1 biological activity has been disclosed in detail. Structure-activity relationship (SAR) optimization provided inhibitors with excellent mPGES-1 potency and low to moderate PGE₂ release A549 cell potency. Among the mPGES-1 inhibitors studied, 7, 9 and 11l provided excellent selectivity over COX-2 (>200-fold) and >70-fold selectivity for COX-1 except 11l, which exhibited dual mPGES-1/COX-1 activity. Furthermore, the above tested mPGES-1 inhibitors demonstrated good metabolic stability in liver microsomes, high plasma protein binding (PPB) and no significant inhibition observed in clinically relevant CYP isoforms. Besides, selected mPGES-1 tool compounds 9 and 11l provided good in vivo pharmacokinetic profile and oral bioavailability (%F?=?33 and 85). Additionally, the representative mPGES-1 tool compounds 9 and 11l revealed moderate in vivo efficacy in the LPS-induced thermal hyperalgesia guinea pig pain model.

Full text of DOI:10.1016/j.bmcl.2017.03.068

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Tricyclic 4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imidazole
derivatives as microsomal prostaglandin E2 synthase-1 (mPGES-1)
inhibitors: SAR and in vivo efficacy in hyperalgesia pain model
Nagarajan Muthukaman a, Macchindra Tambe a, Mahamadhanif Shaikh a, Dnyandeo Pisal a,
Sanjay Deshmukh a, Shital Tondlekar a, Neelam Sarode a, Lakshminarayana Narayana a,
Jitendra M. Gajera a, Vidya G. Kattige b, Srinivasa Honnegowda b, Vikas Karande b,
Abhay Kulkarni b, Dayanidhi Behera c, Satyawan B. Jadhav c, Girish S. Gudi c,
Neelima Khairatkar-Joshi b, Laxmikant A. Gharat a,
⇑
a Chemical Research, Glenmark Pharmaceuticals Limited, Glenmark Research Center, Navi Mumbai, Maharashtra 400709, India
b Biological Research, Glenmark Pharmaceuticals Limited, Glenmark Research Center, Navi Mumbai, Maharashtra 400709, India
c Drug Metabolism and Pharmacokinetics, Glenmark Pharmaceuticals Limited, Glenmark Research Center, Navi Mumbai, Maharashtra 400709, India
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of substituted tricyclic 4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imidazole derivatives have
been synthesized and their mPGES-1 biological activity has been disclosed in detail. Structure-activity
relationship (SAR) optimization provided inhibitors with excellent mPGES-1 potency and low to moder-
ate PGE2 release A549 cell potency. Among the mPGES-1 inhibitors studied, 7, 9 and 11l provided excel-
lent selectivity over COX-2 (>200-fold) and >70-fold selectivity for COX-1 except 11l, which exhibited
dual mPGES-1/COX-1 activity. Furthermore, the above tested mPGES-1 inhibitors demonstrated good
metabolic stability in liver microsomes, high plasma protein binding (PPB) and no significant inhibition
observed in clinically relevant CYP isoforms. Besides, selected mPGES-1 tool compounds 9 and 11l pro-
vided good in vivo pharmacokinetic profile and oral bioavailability (%F = 33 and 85). Additionally, the rep-
resentative mPGES-1 tool compounds 9 and 11l revealed moderate in vivo efficacy in the LPS-induced
thermal hyperalgesia guinea pig pain model.
Received 7 December 2016
Revised 8 March 2017
Accepted 23 March 2017
Available online xxxx
Keywords:
Arachidonic acid
Rheumatoid arthritis
mPGES-1 inhibitor
PGE2
Aryl imidazole
Hyperalgesia
Ó 2017 Elsevier Ltd. All rights reserved.
Microsomal prostaglandin E synthase-1 (mPGES-1) is the termi-
nal enzyme in the biosynthesis of PGE2. Membrane-bound and
secretory phospholipase A2 (PLA2) isoforms convert phospholipids
(PL) to arachidonic acid (AA) in the first step. Next, the COXs con-
vert AA into the unstable intermediate, PGH2. Finally, terminal
PGESs isomerize PGH2 into Prostaglandin E2 (PGE2)1,2 PGH2 is the
precursor for several structurally related PGs, which are formed
by the action of specific prostaglandin synthases.3 The PGs synthe-
sized by this pathway include the afore-mentioned PGE2, as well as
thermore, mPGES-1 expression is low in most normal tissues,
although abundant and constitutive expression is detected in a
limited number of organs, such as the lung, kidney and reproduc-
tive organs.3b
PGE2 plays a significant role in the inflammatory response and
is involved in several chronic inflammatory disease conditions,
such as inflammation, fever, rheumatoid arthritis, cardiovascular
diseases, cancer, bone disorders, periodontitis and pain.6
Prostaglandin E-synthase (PGES) exist in different isoforms and
co-expression studies have demonstrated preferential functional
coupling between COX and PGES isoenzymes. Moreover, mPGES-
1 is a member of Membrane-Associated Proteins in Eicosanoid
and Glutathione metabolism (MAPEG) superfamily and is function-
ally coupled with COX-2. There are three terminal synthases
responsible for PGE2 biosynthesis: one cytosolic isoform cPGES,
and two membrane associated enzymes, mPGES-1 and mPGES-2.
cPGES and mPGES-2 are constitutively expressed and are likely
to be involved in the production of PGE2 responsible for normal
prostaglandin D2 (PGD2), prostaglandin F2 (PGF2 ), prostaglandin
a
a
I2 (PGI2, also known as prostacyclin) and thromboxane A2 (TXA2). It
is recognized that inhibition of COX-2 activity affects the synthesis
of all prostanoids down-stream of PGH2, whereas selective target-
ing of mPGES-1 would only reduce PGE2 production and it should
be noted that shunting towards other PG has been observed.4,5 Fur-
⇑
Corresponding author.
E-mail address: Laxmikant.gharat@glenmarkpharma.com (L.A. Gharat).
http://dx.doi.org/10.1016/j.bmcl.2017.03.068
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

2
N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
physiological reactions.3b,7 On the other hand mPGES-1 is an
inducible isoform, and its expression is dramatically increased in
response to a pro-inflammatory stimulus. mPGES-1 knockout
(KO) experiments in mice demonstrate the link between mPGES-
1 and reduced inflammation, pain and fever response in animal
models.8c Also, additional knockout studies in mice has shown no
abnormalities in thrombogenesis, blood pressure and renal
function when the mice were fed a normal salt diet. In the same
study, selective inhibition, knockout or mutation of COX-2 was
shown to accelerate thrombogenesis and elevated blood pressure
in mice.9
The therapeutic interventions developed to inflammation so far
include either the inhibition of COX-1 and COX-2 enzymes by non-
steroidal anti-inflammatory drugs (NSAIDs). Further, COX-2 inhibi-
tors can induce serious cardiovascular risks, plausibly due to
upstream blockade of the pathway. However, selective inhibition
of terminal mPGES-1 will block the production of PGE2 in the dis-
ease state while sparing other arachidonic acid metabolites, prob-
ably without affecting the balance between prostaglandins PGI2
and TXA2 in order to have a safer cardiovascular profile.1–3,9 Both
COX-1/2 have severe gastrointestinal side-effects and pronounced
effects on the renal function that limit their use in long-term treat-
ments as required in chronic inflammation processes,10 thus
prompting to explore new enzymatic targets in this pathway. In
this perspective, selective inhibition of mPGES-1 has emerged as
an alternative promising strategy to develop effective and safer
agents to treat inflammation devoid of side effects of COX
inhibitors.3b,9,11 Recent study highlighted that, prolonged
inhibition of mPGES-1 in dogs did not affect renal function which
further support the development of mPGES-1 inhibitors as
therapeutic agent.12
Recently, our group has published potent mPGES-1 inhibitors
(VII and VIII) with an excellent efficacy (both ex vivo and in vivo
efficacy) and good oral bioavailability13,14 in addition to several
reported mPGES-1 inhibitors,7,15 such as MF-63 (I),8a,16 AZ-4284
(II),17,18 AF-3442 and AF-3485 (III, IV),19 compound V,20 PF-
04693627 (VI)21 as exemplified in Fig. 1. Although many potent
mPGES-1 inhibitors were developed in the past decade, only a
few inhibitors were published with oral in vivo efficacy data.8,13–16
Furthermore, mPGES-1 inhibitors from Eli Lilly (LY3023703)22
and our group (GRC27864)23 has entered human clinical trials for
inflammatory pain. As part of an ongoing drug discovery program,
we were required to design structurally distinct mPGES-1 inhibi-
tors for SAR optimization and to identify compounds for the
pre-clinical development. We observed that most of the recently
published mPGES-1 inhibitors were having non-acidic core for
the potential activity as exemplified in Fig. 1.24 Therefore, in our
design, inspired from the moderate potency and non-developmen-
tal activity of known compound IX25 (mPGES-1 IC50: 56 nM) prod-
ded us to design novel scaffold from it. We applied the ring closure
and scaffold-hopping26 approach (also called lead hopping) on
compound IX to design novel and conformationally rigid tricyclic
3,4-dihydrochromeno[3,4-d]imidazole X, which envisioned to pro-
vide enhanced mPGES-1 potency over structurally similar com-
pounds like V, MF-63 (I) and IX as displayed in Figs. 1 and 2.
Here we report the synthesis and pharmacological characteriza-
tion of series of 4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imida-
zole derivatives bearing aryl group at C(2)-position and halogens,
aryl-, aryl-alkynes at C(7)-position (compounds 4a–d, 5–9, 11a–
v, 16a–f, 20a–c, 21a–b and 32a–c) as depicted in Scheme 1–3.27
The synthesis of compounds 4a–d and 5–9 were initiated from
the commercially available halogen substituted 2-hydroxyace-
tophenones 1a–c, which reacted with acetone in the presence of
pyrrolidine at reflux to provide 2,2-dimethylchroman-4-one
derivatives 2a–c. Compounds 2a–c was further reacted with
isopentylnitrite28 in the presence of conc. HCl to afford compounds
3a–c, respectively. Compounds 4a–d were obtained by condensing
2,2-dimethylchroman-3,4-dione derivatives 3a–c with 2,6-dibro-
mobenzaldehyde and or 2-chloro-6-fluorobenzaldehyde in the
presence of NH4OAc in acetic acid under heating conditions in
53–66% yield (Scheme 1). Further, Suzuki coupling of compound
4c–d with various para- and meta-substituted arylboronic acid in
the presence of Pd(Ph3P)4 afforded 4,4-dimethyl-3,4-dihy-
drochromeno[3,4-d]imidazole derivatives 5–9 in 45–70% yield
(Scheme 1).
The above synthesized compounds were tested for biological
activity before generating more SAR in order to understand scaf-
fold feasibility as mPGES-1 inhibitors. The inhibitory activity29 of
halogenated 3,4-dihydrochromeno[3,4-d]-imidazoles 4a and 4b
with 2,6-dibromophenyl at C(2)-position (Fig. 2) gave moderate
to poor enzyme potency with IC50s of 325.2 and 709.5 nM, respec-
Z
D
O
2'
N
X
6'
Y
1
N
Cl
Scaffold hopping
HN
N
N
3
C
N
8
A
Z
B
O
NH
4
ring closure to
get novelty and
improve potency.
Br
F
R
7
(IX)
mPGES-1 IC50 = 56 nM
(X)
Fig. 2. General structure X: R = Cl, Br and I, substituted aryl and substituted aryl-
alkynes; X, Y = combination of Cl, F and CN; Z = C and or N.
O
Cl
Cl
NC
H
N
N
HN
N
H
N
O
N
H
S
O
CF3
O
N
H
N
R
F
O
O
NC
SO2NH2
MF-63 (I)
Merck
AF-3442 (R = ethyl) - (III)
AF-3485 (R = hydroxyethyl) - (IV)
Angelina
AZ- 4284 (II)
AstraZeneca
(V)
Dainippon Sumitomo
F
O
Cl
O
O
N
O
F
N
NH
O
N
N
F3C
N
HN
NH
Cl
H
N
N
H
N
O
Cl
OH
F3C
Cl
Me
O
PF-04693627 (VI)
Pfizer
(VII)
Glenmark pharma
(VIII)
Glenmark pharma
Fig. 1. Structure of reported mPGES-1 inhibitors.
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
Cl
X
Y
CHO
O
O
O
F
Y
HN
HN
O
R-B(OH)2
d
N
X
c
N
a
b
R1
O
R1
1a
1b
1c
OH
R1
O
R1
O
R
O
1
: R1 = Cl
: R = Cl
2a: R1 = Cl
2b: R1 = Br
2c: R1 = I
3a
3b
3c
4a: R1 = Cl; X, Y = Br
4b: R1 = Br; X, Y = Br
4c: R1 = Br; X = Cl; Y = F
4d: R1 = I; X = Cl; Y = F
5: R = 4Cl-phenyl
6: R = 4CF3-phenyl
7: R = 3CF3-phenyl
8: R = 4CF3O-phenyl
9: R = 3CF3O-phenyl
: R1 = Br
: R1 = Br
: R1 = I
1
: R = I
Si
e, f
Cl
Cl
F
HN
F
HN
N
N
g
O
O
R
10
11a-v
(Table 2; R = Substituted aryl
and heteroaryl groups)
Scheme 1. Reagents and conditions: (a) acetone, pyrrolidine, MeOH, reflux, 4–6 h, 50–65%; (b) isopentyl nitrite, ethanol, conc. HCl, rt, 12–16 h, 57–70%; (c) 2,6-Di-Br-
benzaldehyde and or 2Cl-6F-benzaldehyde, acetic acid, NH4OAc, 100 °C, 6–8 h, 53–66%; (d) R-B(OH)2, K2CO3, Pd(Ph3P)4, DMSO, 90 °C, 8–10 h, 45–70%; (e) CuI, DMSO,
PdCl2(PPh3)2, TEA, 70 °C, 8 h, 60%; (f) TBAF, THF, rt, 65%; (g) PdCl2(PPh3)2, TBAF, R-Br and or R-I (R = substituted phenyl, heteroaryl), 90 °C, 4–8 h, 12–65%.
Br
Br
CHO
Br
O
O
O
O
Br
O
O
Br
HN
Br
HN
Si
a
N
N
O
d
b
c
I
O
O
2c
12
R
14
15a-f
(R = substituted aryl)
13
(not isolated)
Cl
CHO
N
c
e
X
17 (X = Cl)
18
(X = F)
N
N
NC
Cl
Cl
CN
HN
X
HN
X
HN
N
N
N
d
O
O
O
R
R
16a-f
(R = Substituted aryl, Table 3)
20a-c (X = Cl, see Table 3 for R)
21a-b (X = F, see Table 3 for R)
19a (X = Cl)
19b (X = F)
Scheme 2. Reagents and conditions: (a) CuI, DMSO, PdCl2(PPh3)2, TEA, 70 °C, 6 h, then TBAF, THF, rt, 62%; (b) ethanol, isopentyl nitrite, conc. HCl, 0 °C to rt, 60% (crude); (c)
NH4OAc, acetic acid, 100 °C, 29–38%; (d) PdCl2(PPh3)2, TBAF, R-Br and or R-I (R = substituted phenyl), 90 °C, 4–6 h, 19–42%; (e) CuCN, N-methyl-2-pyrrolidinone or DMF, 100–
120 °C, 9–28%.
tively. Further, halogenated (Br- and I-) compounds 4c and 4d with
2-chloro-6-fluorophenyl linked at C(2)-position provided lower %
inhibition at 10 mM test concentration. Concurrently, we replaced
the bromine and/or iodine in compounds 4c–d with substituted
phenyl afforded compounds 5–9, which furnished moderate to
potent mPGES-1 inhibitors. Among them, meta-substituted phenyl
compounds 7 (IC50 = 92.94 nM) and 9 (IC50 = 56.89 nM) exhibited
improved mPGES-1 enzyme potency in comparison to para-substi-
tuted phenyl derivatives 5, 6 and 8, as well as halogen analogs 4a
and 4b as disclosed in Table 1 (Similar structure-activity relation-
ship was disclosed for compounds V and VII which unveiled
enhanced enzyme potency over their halogen-substituted
analogs).20,13 In order to understand our scaffold further as
mPGES-1 inhibitors, enzyme potent compounds 7 and 9 were fur-
ther assessed for cell-based activity.29,30 Therefore, microsomes of
IL-1b stimulated human A549 epithelial lung cells expressing
mPGES-1 were pre-incubated, and the amount of PGE2 release
was measured using HTRFÒ (Homogeneous Time Resolved Fluores-
cence) assay kit. Compounds 7 and 9 revealed moderate PGE2
release with an IC50s of 616.1 and 839.7 nM, respectively (Foot
note in Table 1). Next, introduction of heterocyclic groups and aryl
amides at C(7)-position of tricyclic core (Fig. 2) revealed dimin-
ished mPGES-1 potency (see supporting info for the trivial SAR
and enzyme potency details as in Table 7).27
Subsequently, we next explored the possibility of improving the
mPGES-1 potency by attaching the substituted aryl-alkyne group
at C(7)-position of 2-(2-chloro-6-fluorophenyl)-4,4-dimethyl-3,4-
dihydrochromeno[3,4-d]imidazole as shown in Scheme-1. Iodo
compound 4d was subjected to Sonogashira coupling with
trimethylsilylacetylene, followed by trimethylsilyl (TMS) cleavage
to provide key alkyne intermediate 10. Palladium-catalyzed
cross-coupling of alkyne compound 10 with commercially avail-
able substituted aryl, heteroaryl bromides and or iodides provided
various aryl-alkyne appended 3,4-dihydrochromeno[3,4-d]
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

4
N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
O
O
O
O
O
OH
O
OH
O
OH
OH
OH
OMe
OEt
OEt
c
a
b
d
e
Cl
N
Cl
N
Cl
N
OMe
Cl
N
OMe
Cl
N
OMe
Cl
N
OMe
27
25
22
23
24
26
Cl
O
O
O
O
O
O
F
HN
O
O
N
O
g
f
j
h
i
I
N
28
N
N
N
O
N
R
R
R
29
30a: R = 2CF3-Ph
31a: R = 2CF3-Ph
32a: R = 2CF3-Ph
30b: R = 2Cl, 5CF3-Ph
30c: R = 2F, 6CF3-Ph
31b: R = 2Cl, 5CF3-Ph
31c: R = 2F, 6CF3-Ph
32b: R = 2Cl, 5CF3-Ph
32c: R = 2F, 6CF3-Ph
Scheme 3. Reagents and conditions: (a) CDI, potassium 3-ethoxy-3-oxopropanoate, MgCl2, TEA, THF, acetonitrile, 0 °C-rt, 72%; (b) NaBH4, THF, 0 °C, 77%; (c) CH3MgCl, THF,
0 °C, 60%; (d) PCC, DCM, rt, 5 h, 63%; (e) 48% HBr in acetic acid, 80 °C, 2 h, then POCl3, 100 °C, 5 h, 85% (single flask); (f) 4 M HCl, dioxane, rt, 5 h, then NaI, THF, 100 °C, 2 h, 58%;
(g) CuI, DMSO, TEA, trimethylsilylacetylene, PdCl2(PPh3)2, 90–100 °C, 12 h; then TBAF, rt, 4 h, 40% for two steps; (h) R-I, PdCl2(PPh3)2, CuI, TEA, DMSO, 100 °C, 4 h, 28–46%;
(i) ethanol, isopentyl nitrite, conc. HCl, rt, 6 h, 80–86%; (j) NH4OAc, 2-chloro-6-fluorobenzaldehyde, acetic acid, 90 °C, 12 h, 27–52%.
Table 1
Table 2
SAR of halogens and substituted aryl on A-ring.
SAR of substituted aryl alkyne on A-ring.
Entry
Compd
mPGES-1% inhibitiona
mPGES-1
IC50 (nM)
Cl
b,c
@ 1 mM
@ 10 mM
F
N
1
2
3
4
5
6
4a
4b
4c
4d
5
6
7
8
9
73.98
68.56
40.80
44.30
76.04
81.37
91.41
70.08
97.93
88.98
91.97
94.09
84.21
94.80
83.31
94.62
86.89
91.77
325.2
709.5
–
–
–
191.4
92.94
–
NH
O
R
11a-v
d
7
Compd
R
mPGES-1
IC50 (nM)
A549, 2% FBS PGE2
8
a,b
a,b
d
IC50 (nM)
9
56.89
11a
11b
11c
11d
11e
11f
11g
11h
11i
Phenyl
83.02
64.68
39.16
38.24
19.85
15.61
29.31
61.81
35.55
9.30
–
–
–
–
–: Not determined.
2Cl-Phenyl
3Cl-Phenyl
4Cl-Phenyl
2CF3O-Phenyl
2CF3-Phenyl
a
% Inhibition values are means of two experiments in duplicates.
MF-63 (1) was used as a positive control in this experiment and for MF-63 in-
b
house enzyme and A549 cell IC50s are 1.9 nM and 56 nM. For the reported potency
of MF-63, see Ref. 16a.
578.0
692.6
–
–
c
IC50 values are derived from graphs plotted with data from a minimum of two
3CF3-Phenyl
4CF3-Phenyl
experiment in duplicates.
d
A549 whole cell IC50s of 7 and 9 are 616.1 and 839.7 nM respectively.
2, 3-Di-Cl-Phenyl
2, 5-Di-Cl-Phenyl
2, 6-Di-Cl-Phenyl
2F, 5Cl-Phenyl
2CF3, 4Cl-Phenyl
2CF3, 5F-Phenyl
2CF3, 6F-Phenyl
3CF3, 4F-Phenyl
2F, 5CF3-Phenyl
2Cl, 4CF3-Phenyl
5CF3-Pyridin-2-yl
3F-Pyridin-4-yl
Pyrimidin-5-yl
5CH3-Thiazole
–
11j
11k
11l
1416.0
–
11.22
36.28
7.76
imidazole derivatives 11a–v, as shown in Table 2.31 Also, we
focused our attention on the d-ring modification (Fig. 2) to improve
the mPGES-1 enzyme potency (Schemes 2 and 3). Accordingly, 7-
ethynyl-2,2-dimethylchroman-3,4-dione 13 was prepared by cou-
pling of 7-iodochromanone 2c with trimethylsilylacetylene using
Sonogashira conditions and further TMS-cleavage using tetrabuty-
lammonium fluoride (TBAF), followed by oxidation using isopentyl
nitrite28 as described earlier. Compound 13 was further condensed
with 2,6-dibromobenzaldehyde in the presence of ammonium
acetate (NH4OAc) to afford alkyne derivative 14. A direct conver-
sion of dibromo group to dicyano functionality in compound 14
did not work satisfactorily.32 Therefore, compound 14 was
first coupled with substituted aryl iodides and or bromides using
palladium-catalyzed coupling conditions yielded compounds
15a–f, respectively. Then, the dibromo group in compound
15a–f was individually converted to dicyanophenyl derivatives
16a–f using Cu(I)CN in DMF16a as displayed in Scheme 2.
Furthermore, intermediate 13 was condensed independently
with 3,5-dichloroisonicotinaldehyde (17) and 3-chloro-5-
fluoroisonicotinaldehyde (18)33 in the presence of NH4OAc to
afford compounds 19a and 19b in 29–38% yield. Then, compounds
19a and 19b were subjected to Sonogashira coupling with substi-
tuted aryl iodides or bromides to provide compounds 20a–c and
21a–b, correspondingly (Scheme 2 and Table 3).
838.7
2828.0
705.7
691.5
1252.0
461.0
–
–
–
–
–
11m
11n
11o
11p
11q
11r
c11s
11t
21.02
12.28
29.65
25.04
28.20
320.0
640.8
796.7
464.8
11u
11v
–: not determined.
a
MF-63 (1) was used as a positive control in this experiment and the in-house
enzyme and A549 cell IC50s are 1.9 nM and 56 nM. For the reported potency of MF-
63, see Ref. 16a.
IC50 values are derived from graphs plotted with data from a minimum of two
experiment in duplicates.
b
c
Check Ref. 29 for details.
Scheme 3 describes the preparation of tricyclic chromenoimida-
zoles 32a–c with nitrogen incorporated in the A-ring (Fig. 2). Com-
mercially available 6-chloro-2-methoxynicotinic acid 22 was
converted into nicotinoyl acetate 23 using ethyl potassium malo-
nate and magnesium chloride (MgCl2) in the presence of 1,10-car-
bonyldiimidazole (CDI) in 72% yield.34 The carbonyl group in
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
5
Table 3
SAR of D- and A-ring modification.
Y
NC
Cl
Z
Z
N
N
NH
NH
X
O
O
R
R
20a-c (X = C; Y = N; Z = Cl)
21a-b (X = C; Y = N; Z = F)
32a-c (X = N; Y = C; Z = F)
16a-f (Z = CN)
Compd
R
Z
mPGES-1 IC50 (nM)bor% inh.@1/10 mMa
A549, 2% FBS PGE2
IC50 (nM)b
16a
16b
16c
16d
16e
16f
20a
20b
20c
21a
21b
32a
32b
32c
Phenyl
2CF3-Phenyl
CN
CN
CN
CN
CN
CN
Cl
Cl
Cl
F
24.76
15.90
11.77
12.72
12.65
3.89
81.28
33.06
31.50
73.89
169.7
407.0
65%/63%
75%/76%
984.0
297.6
266.7
2320.0
835.9
1196.0
–
–
–
–
–
–
–
–
2CF3O-Phenyl
2,5-Di-Cl-Phenyl
2F, 6CF3-Phenyl
2Cl, 5CF3-Phenyl
2CF3-Phenyl
2Cl, 5CF3-Phenyl
5F, 2CF3-Phenyl
2Cl, 4CF3-Phenyl
2CF3-Phenyl
F
F
F
F
2CF3-Phenyl
2Cl, 5CF3-Phenyl
2F, 6CF3-Phenyl
–: Not determined;
a
% Inhibition values are means of two experiments in duplicates.
IC50 values are derived from graphs plotted with data from a minimum of two experiments in duplicates. Positive control data is same as in the Tables 1 and 2.
b
compound 23 was reduced to alcohol 24 using sodium borohy-
dride reduction, followed by addition of an excess methylmagne-
sium chloride (Grignard reaction) to afford compound 25. The
secondary alcohol in compound 25 was oxidized using pyridinium
chlorochromate (PCC) to ketone 26, followed by one-pot methyl
ether cleavage and cyclization using 48% HBr in acetic acid and fur-
ther reaction with POCl3 (single flask conversion) afforded 7-
chloro-2,2-dimethyl-2H-pyrano[2,3-b]pyridin-4(3H)-one 27. Com-
pound 27 was further converted to iodo derivative 28 using NaI,
followed by coupling with trimethylsilylacetylene using Sono-
gashira condition and further trimethylsilyl(TMS) cleavage to
(Table 2). In another variation, dichloro-substituted phenyl alkyne
derivatives (11i–k) tested, 11j and 11k (IC50s = 9.3 and 11.2 nM)
displayed 8-fold improvement in the enzyme potency over 11a.
Another set of di-substituted phenyl-alkyne analogs having bulky
trifluoromethyl in combination with chloro- and fluoro-derived
compounds 11l–r exemplified excellent mPGES-1 potency
(<36 nM IC50
,
Table 2). Among this, 4-chloro-2-trifluo-
romethylphenyl alkyne derivative 11m (IC50: 7.76 nM) demon-
strated greater than 10-fold mPGES-1 potency and 6-fluoro-2-
trifluoromethylphenyl alkyne derivative 11o (IC50: 12.2 nM) pro-
vided 7-fold higher potency related to un-substituted phenyl
alkyne derivative 11a (IC50 = 83 nM). Furthermore, in order to
reduce the high lipophilicity of aryl alkyne analogs (Table 2), six-
and five-membered heterocycle alkynes 11s–v were introduced31
and tested for its mPGES-1 activity. Among the alkynes 11s–v
tested, substituted pyridine-alkyne derivatives 11s and 11t dis-
played >5-fold drop in mPGES-1 potency over similar derivative
11h, whereas pyrimidine (11u) and thiazole (11v) derived alkyne
analogs revealed substantial loss of mPGES-1 potency in compar-
ison with high potent analogs 11j and 11m, suggesting a trend
towards incompatibility of polar groups at C(7)-position of 4,4-
dimethyl-3,4-dihydrochromeno[3,4-d]imidazole core (X, Fig. 2).
Subsequently, some of the potent mPGES-1 inhibitors (11e, 11f
and 11j–q) were further evaluated for PGE2 formation (mPGES-1
biomarker) in human A549 epithelial lung carcinoma cell lines.
All the tested analogs as shown in Table 2 unveiled only moderate
PGE2 release and the IC50s in the cellular assays were significantly
lower in comparison to their in vitro enzyme potency.29,30
afford
7-ethynyl-2,2-dimethyl-2H-pyrano[2,3-b]pyridin-4(3H)-
one 29. Palladium-catalyzed, Sonogashira coupling of alkyne 29
with various substituted aryl bromides gave aryl-alkyne deriva-
tives 30a–c in moderate yield. Compounds 30a–c were indepen-
dently oxidized using isopentylnitrite28 to yield 3,4-diketo
derivatives 31a–c, followed by condensation with 2-chloro-6-fluo-
robenzaldehyde, as described previously yielded compounds 32a–
c, respectively.
Having accomplished moderate enzyme potency for aryl linked
4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imidazole
derivatives
5–9 (Table 1), we further evaluated various aryl-alkyne linked
compounds 11a–v for increased mPGES-1 potency (Table 2). While
simple phenyl-alkyne linked compound 11a (IC50: 83.02 nM)
retained enzyme potency similar to aryl linked analog 7, ortho-,
meta- and para-substituted chlorophenyl-alkyne derivatives
(11b–d) showed onefold higher intrinsic mPGES-1 potency in
comparison with 7 having IC50 of 92.9 nM, respectively. Next,
ortho-substituted phenylalkyne analogs such as, 2-(trifluo-
romethoxyphenyl)ethynyl linked compound 11e (IC50: 19.8 nM)
Therefore, in order to improve the physicochemical properties
and the cellular potency of aryl-alkyne analogs, we introduced
2,6-dicyanophenyl, 3,5-dichloropyridyl, 3-chloro-5-fluoropyridyl
groups as D-ring and incorporated nitrogen in the A-ring as shown
in Fig. 2, Table 3 and these alkyne analogs were evaluated for
mPGES-1 in vitro potency (Schemes 2 and 3).29 Simple phenyl-
alkyne derivative 16a (IC50: 24.76 nM and A549 IC50: 984 nM)
and 2-trifluoromethylphenylethynyl linked compound 11f (IC50
:
15.6 nM) offered 4-fold higher potency over phenyl alkyne 11a.
Further, among the m- and p-trifluoromethylphenylethynyl deriva-
tives tested, 11g (IC50: 29.3 nM) retained strong mPGES-1 inhibi-
tory potency, whereas 11h potency was dropped to 61.8 nM IC50
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

6
N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Table 4
In vitro PGE2 release activity, COX-selectivity and in vitro ADME profiles of selected compounds.
Compd
Cell A549, 2%
COX-1 IC50
COX-2 IC50
Metabolic stability in liver
microsomes (% remaining)c
CYP inhibitions at 10 mM
PPB
FBS PGE2, IC50 (nM)a
(mM)b
(mM)b
concentrations (%)d
(% bound)e
H/R/GP
1A2
2D6
3A4
2C9
2C19
H/G.Pig
7
9
11l
16b
616.1
839.7
838.7
297.6
8.35
6.75
0.06
–
>10.0f
>10.0g
9.88
–
89/99/93
84/83/80
76/44/73
81/–/100
–
56
NI
26.6
29
37
8.4
31
24
35
–
–
NI
33.9
–
45.4
33.3
32
27.2
17/26
–
>99.5/>99.5
>95.5/>99.5
–
–: not determined; NI: no inhibition.
a
MF-63 (1) was used as a positive control in this experiments and the in-house A549 cell IC50 = 56 nM. For the literature reported potency of MF-63, see Ref. 16a. IC50
values represent the concentration to inhibit 50% of PGE2 relative to vehicle control and derived from graphs plotted with data from a minimum of two experiments in
duplicates.
b
IC50 values are derived from graphs plotted with data from a minimum of two experiments in duplicates.
Percentage of test compound remaining after 60 min incubation with liver microsomes (human, rat and guinea Pig) at 37 °C. MS experiment was conducted in triplicates
c
(see supporting info (SI) for details).
d
Cytochrome P450 (CYP)% inhibition as compared to control (no inhibitor) and conducted in triplicates (see SI for details).
Plasma protein binding (PPB) was determined using equilibrium dialysis method. 10 mM concentrations were used (see SI for details).
COX-2% inhibition values are 2.65%@1 mM and 27.45%@ 10 mM.
COX-2% inhibition values are 2.37%@1 mM and 24.21%@ 10 mM.
e
f
g
having 2,6-dicyanophenyl as D-ring revealed approximately four-
fold higher mPGES-1 potency and moderate cell potency in com-
parison to structurally similar analog 11a. Furthermore, ortho-
substituted phenyl alkyne derivatives 16b (IC50: 15.9 nM and
tion at 1 mM test concentration). Albeit many compounds divulged
good mPGES-1 enzyme potency in the alkyne series (<30 nM),
A549 cellular potency was not significantly improved as antici-
pated (Tables 2 and 3) with all the projected D-ring modifications.
As part of the lead optimization strategy, selected potent com-
pounds were further screened for COX enzymes selectivity and
in vitro ADME profiles (Table 4). Most of the tested compounds
(7, 9 and 11l) exhibited >150-fold selectivity for COX-2 inhibition
A549 IC50: 297.6 nM) and 16c (IC50: 11.77 nM and A549 IC50
:
266.7 nM) exemplified good enzyme and enhanced cell potency
in comparison with 11f and 11e (A549 cell IC50s are 692.6 and
578 nM), which were structurally similar except for the D-ring
modification. Among the di-substituted phenyl alkyne derivatives
tested, analogs 16d–f demonstrated excellent enzyme potency
(<13 nM). However, their A549 cellular potency (>800 nM) was
significantly less in comparison with other dicyanophenyl as D-
ring analogs 16b and 16c (<300 nM) and the reason for the cell
potency discrepancy was not readily explainable (Table 3). Simi-
larly, 3,5-dichloropyridyl (as D-ring) derived alkyne compounds
20a–c exhibited slight drop in mPGES-1 potency in relation to sim-
ilar alkyne derivatives 11f and 11n (enzyme IC50s are 15.6 and
21 nM), respectively. In another variation, 3-chloro-5-fluoropyridyl
(as D-ring) derived alkyne analogs 21a (IC50: 73.89 nM) and 21b
(IC50: 169.7 nM) displayed roughly 3 to 11-fold lower mPGES-1
potency compared to 11r and 11f having identical substitution at
C(7)-position of A-ring. Next, the impact of nitrogen on the A-ring
of tricyclic 4,4-dimethyl-3,4-dihydrochromeno[3,4-d]imidazole
(Table 3) was assessed for the mPGES-1 potency. Thus, 2-trifluo-
romethylphenyl alkyne derivative 32a (IC50: 407 nM) provided ꢀ
37-fold lesser potency than the corresponding non-pyridyl analog
11f. In contrast to compounds having nitrogen at D-ring (20a–c
and 21a–b), nitrogen inserted at A-ring derivatives 32b and 32c
revealed complete loss of potency (only 65–75% mPGES-1 inhibi-
Fig. 3. Analgesic effects of 9 and 11l in the guinea pig hyperalgesia pain model. Data
are expressed as percentage of hyperalgesia, with the naive group (injected
intraplantarly with saline) as 0% and the vehicle-treated LPS-injected group as 100%
(Results are shown as mean SEM, n = 8 animals per dose group).
Table 5
Oral PK of 9 and 11l in rat and guinea pig.
PK parametersa
Compd: 9a,b,c
Compd: 11la,b
Rat i.v dose (5 mg/kg)
Rat oral dose (10 mg/kg)
G.Pig oral dose (100 mg/kg)
Rat IV dose (5 mg/kg)
Rat oral dose (10 mg/kg)
Cmax (ng/mL)
AUC (ng h/mL)
Tmax (h)
Bioavailability (%F)
T1/2 (h)
–
–
–
–
826 125
12,562 1100
4
33
11.2 1.95
–
–
803 201
13,427 2576
24
–
ND
–
–
–
–
–
1176 0.94
12,380 491
4
85
6.77 0.87
–
–
9.87 0.41
3.63 0.60
3.11 0.64
5.50 0.16
11.14 0.39
5.30 0.10
CL (mL/min/kg)
Vz (L/kg)
–
–: not applicable; ND: Not determined.
a
Cmax, AUC0–24, Tmax, T1/2, bioavailability (%F), Clearance (CL) and volume of distribution (Vz) were determined in male Sprague-Dawley rats.
Vehicle for rat Oral dosing – 0.5% methylcellulose MC suspension; Vehicle for rat i.v – 20% NMP + 20% Ethanol + 60% PEG 200). The data represented is mean SD (n = 3)
Vehicle for guinea pig Oral dosing – 0.5% methylcellulose suspension. The data represented is mean SD (n = 3).
b
c
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
7
Table 6
Plasma and CSF concentration of 9 and 11l at PD time point.
Compd
Dose
(mg/kg, po, od)
% Hyperalgesia inhibition
Plasma concentration
@PD time point (nM)a
CSF concentration
@ PD time point (nM)a
9
11l
200
100
38
26
7541
13,831
9.22
2.63
a
Concentrations are means of n = 8 animals per dose group. Study protocol is provided in the SI.
and >70-fold selectivity for COX-1 inhibition except 11l, which
revealed dual mPGES-1 (IC50: 36.2 nM) and COX-1 (IC50: 60 nM)
activity. The in vitro metabolic stability of compounds 7, 9, 11l
and 16b were assessed in human, rat and guinea pig liver micro-
somes using the standard procedure.29 Briefly, these compounds
were incubated at a concentration of 1.0 mM with 1.0 mg/mL pro-
tein at 37 °C for 60 min, then samples analyzed using LC/MS/MS
and further, the analysis of% remaining as disclosed in Table 4.
All the compounds were found to be metabolically stable except
11l which showed moderate stability in rat liver microsome. Com-
pounds 7, 9, 11l and 16b were further evaluated for the inhibition
of clinically relevant CYP isoforms (1A2, 2D6, 3A4, 2C9, 2C19) in 1
and 10 mM concentrations and no significant inhibition was
observed. Further, compounds 9 and 11l were found to be highly
protein binding as illustrated in Table 4, which plausibly explains
the poor cellular potency.35
Earlier, our group had studied several cell potent mPGES-1 inhi-
bitors (A549 cell IC50s: <20 nM) in the in vivo guinea pig animal
model which demonstrated excellent efficacy.13,14,23,36–38 Simi-
larly, in order to understand the correlation between PGE2 release
cell potency and in vivo efficacy, compounds 9 and 11l with poor
A549 cellular potency39,40 were further selected for in vivo oral
pharmacokinetics (PK) and efficacy study.38 Selected compounds
9 and 11l were evaluated in rat and guinea pig for oral pharma-
cokinetic studies29 and the results are disclosed in Table 5. The
gesia response is not only mediated by peripheral PGE2, but also
significantly involves centrally mediated PGE2.41,42
In summary, we have described a novel class of tricyclic 2-(2-
chloro-6-fluorophenyl)-4,4-dimethyl-3,4-dihydrochromeno[3,4-d]
imidazoles with aryl- and aryl alkyne-substituted compounds as
mPGES-1 inhibitor with excellent enzyme potency and moderate
to poor cellular potency. Careful modification on the D-ring
afforded several mPGES-1 inhibitors with improved cell potency.
Furthermore, the representative tool compounds 9 and 11l exhib-
ited good selectivity over COX enzymes and favorable in vitro and
in vivo pharmacokinetic properties and also demonstrated moder-
ate efficacy in the LPS-induced thermal hyperalgesia pain model
(acute study) in comparison to diclofenac (approved COX-1 inhibi-
tor and pain drug). Therefore, further SAR optimization and core
scaffold alteration is warranted to afford compounds with
improved cellular potency, favorable plasma protein binding and
acceptable lipophilicity for further therapeutic utility as mPGES-1
inhibitors.
Acknowledgements
We thank the scientists from the analytical support group for
their help in compound characterization. We also thank Drs. Sra-
van Mandadi, Vishwottam Kandikere and Sanjib Das for the review
of this manuscript.
rat oral PK profile of
9 (10 mg/kg) revealed adequate Cmax
(826 ng/mL), AUC0-inf (12562 ng h/mL) and moderate bioavailabil-
ity (33%), whereas 11l exemplified enhanced Cmax, (1176 ng/mL)
and oral bioavailability (85%) with an AUC0-inf (12,380 ng h/mL)
comparable to 9. Besides, both compounds (9 and 11l) exhibited
delayed absorption (Tmax = 4 and 4 h), low in vivo clearance (3.6
and 11.1 mL/min/kg, respectively), volume of distribution (3.11
and 5.3 L/kg, respectively) and adequate half-life (T1/2) in rat. The
guinea pig oral PK of 9 (100 mg/kg) was comparable to its
10 mg/kg rat PK study and revealed slow absorption (Tmax = 24 h,
respectively). Overall, both compounds exhibited favorable phar-
macokinetic profile thus suitable for in vivo efficacy study.
Having established favorable oral pharmacokinetics in rat and
guinea pig, the selected compounds 9 and 11l were further evalu-
ated in the LPS-induced hyperalgesia guinea pig pain model8,21 to
assess analgesic effects (Fig. 3, Table 6). Injection of LPS into the
plantar caused a significant increase in thermal hyperalgesic
response compared to saline injected animals. The clinically
approved pain drug, diclofenac inhibited 80% of the hyperalgesic
response at the dose of 10 mg/kg when administered orally at
1 h before LPS injection, whereas tool compounds 9 (200 mg/kg)
and 11l (100 mg/kg) revealed only 38% and 26% inhibition of
hyperalgesic response relative to vehicle treated animal. The
plasma concentration at pharmacodynamics (PD) time point for
the compounds 9 and 11l were found to be 7.5 mM and 13.8 mM,
respectively. Despite having adequate plasma PD concentrations,
the cerebrospinal fluid (CSF) concentrations of 9 and 11l at PD time
point were found to be very low (Table 6). Therefore, moderate
efficacy observed with compounds 9 and 11l might be due to sev-
eral factors including low CSF concentration, poor cellular potency
and unspecific high plasma protein binding,35 as thermal hyperal-
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.03.
068.
References
1. (a) Koeberle A, Werz O. Curr Med Chem. 2009;16:4274;
(b) Samuelsson B, Morganstern R, Jakobsson P-J. Pharmacol Rev. 2007;59:207.
2. (a) Ek M, Engblom D, Saha S, Blomqvist A, Jakobsson P, Ericsson-Dahlstrand A.
Nature. 2001;410:430;
(b) Jachak SM. Curr Opin Invest Drugs. 2007;8:411.
3. (a) Cha YI, Solnica-Krezel L, Dubois RN. Dev Biol. 2006;289:263–272;
(b) Chang HH, Meuillet EJ. Future Med Chem. 1909;2011:3.
4. (a) Grosser T, Fries S, FitzGerald GA. J Clin Invest. 2006;116:4;
(b) Wang M, Song WL, Cheng Y, Fitzgerald GA. J Intern Med. 2008;263:500.
5. Marnett L. J Annu Rev Pharmacol Toxicol. 2009;49:265.
6. Kamei D, Murakami M, Nakatani Y, Ishikawa Y, Ishii T, Kudo I. J Biol Chem.
2003;278:19396.
7. Friesen RW, Mancini JA. J Med Chem. 2008;51:4059.
8. (a) Xu D, Rowland SE, Clark P, et al. J Pharmacol Exp Ther. 2008;326:754;
(b) Leclerc P, Pawelzik SC, Idborg H, et al. Prostaglandins Other Lipid Mediat.
2013;102–103:1;
(c) Uematsu S, Matsumoto M, Takeda K, Akira S. J Immunol. 2002;168:5811.
9. (a) Cheng Y, Wang M, Yu Y, Lawson J, Funk CD, FitzGerald GA. J Clin Invest.
2006;116:1391;
(b) Korotkova M, Jakobsson P-J. Basic Clin Pharmacol Toxicol. 2014;114:64.
10. Mukherjee D, Nissen SE, Topol EJ. J Am Med Assoc. 2001;286:954–959.
11. Trebino CE, Stock JL, Gibbons CP, et al. Proc Natl Acad Sci USA. 2003;100:9044.
12. Salazar F, Vazquez ML, Masferrer JL, et al. Am J Physiol Renal Physiol. 2014;306:
F68.
13. Banerjee A, Pawar MY, Patil S, et al. Bioorg Med Chem Lett. 2014;24:4838.
14. Muthukaman N, Deshmukh S, Sarode N, et al. Bioorg Med Chem Lett.
2016;26:5977.
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068

8
N. Muthukaman et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
15. (a) Koeberle A, Laufer SA, Werz O. J Med Chem. 2016;59:5970;
(b) Chen Y, Liu H, Xu S, Wang T, Li W. MedChemComm. 2015;6:2081;
(c) Kuklish SL, Antonysamy S, Bhattachar SN, et al. Bioorg Med Chem Lett.
2016;26:4824;
30. A549 cell assay evaluates the compound’s permeability and provides crucial
information about its downstream effects. mPGES-1 overexpression upon IL-1b
stimulation and increased release of pro-inflammatory mediator PGE2 occurs
in A549 cells. Hence, this is a standard assay to study the anti-inflammatory
effects of mPGES-1 inhibitors. For more details see: Sweeney FJ, Wachtmann
JD, Eskra JD, et al. Mol Cell Endocrinol. 2003;205:151.
(d) Bahia MS, Katare YK, Silakari O, Vyas B, Silakari P. Med Res Rev.
2014;34:825;
(e) Khurana P, Jachak SM. RSC Adv. 2016;6:28343.
31. Compound 11s (Table 2) was synthesized by coupling of commercially
available 2-chloro-5-(trifluoromethyl)pyridine with intermediate 10 as
described in Scheme 1.
32. The direct conversion of dibromo to dicyano functionality in compound 14
yielded * 15% yield and required tedious and multiple column chromatography
purifications due to mono- and dicyano derivatives. The scale-up of this
conversion was also troublesome due to multiple product formation.
33. Chloro-5-fluoro isonicotinaldehyde 18 (Scheme 2) was prepared using the
similar procedure as described and for details, see: Schlosser M, Bobbio C. Eur J
Org Chem. 2002;24:4174.
16. (a) Cote B, Boulet L, Brideau C, et al. Bioorg Med Chem Lett. 2007;17:6816;
(b) Giroux A, Boulet L, Brideau C, et al. Bioorg Med Chem Lett. 2009;19:5837.
17. Bylund J, Annas A, Hellgren D, Bjurstrom S, Andersson H, Syanhagen A. Drug
Metab Dispos. 2013;41:634.
18. Norberg JK, Sells E, Chang H-H, Alla SR, Zhang S, Meuillet EJ. Pharm Pat Anal.
2013;265:12.
19. (a) Bruno A, Di Francesco L, Coletta I, et al. Biochem Pharmacol. 2010;79:974;
(b) Finetti F, Terzuoli E, Bocci E, et al. PLoS One. 2012;7:1.
20. (a) Shiro T, Takahashi H, Kakiguchi K, et al. Bioorg Med Chem Lett. 2013;22:285;
(b) Shiro T, Kakiguchi K, Takahashi H, Nagata H, Tobe M. Bioorg Med Chem.
2013;21:2868.
34. Tsuzuki Y, Tomita K, Shibamori K-I, Sato Y, Kashimoto S, Chiba K. J Med Chem.
2004;47:2097.
21. (a) Walker DP, Arhancet GB, Lu H-F, et al. Bioorg Med Chem Lett. 2013;23:1120;
(b) Arhancet GB, Walker DP, Metz S, et al. Bioorg Med Chem Lett. 2013;23:1114.
35. (a) The correlation between poor cell potency and high unspecific plasma
protein binding (PPB) were cited in many articles. For the details, see: Riendeau
R, Aspiotis R, Ethier D, et al. Bioorg Med Chem Lett. 2005;15:3352(b) Also see
Refs. 16a,24b.
22. (a)
A
Single Dose Study of LY3023703 in Healthy Participants.
http://clinicaltrials.gov/ct2/show/NCT01632579.;
(b) Schiffler MA, Antonysamy S, Bhattachar SN, et al. J Med Chem. 2016;59:194;
36. Gharat LA, Banerjee A, Pawar MY, Khairatkar-Joshi N, Kattige VG. Substituted
Bicyclic Heteroaryl Compounds as mPGES-1 Inhibitors. PCT Intl Appl WO/2013/
038308, 2013.
(c) Chandrasekhar S, Harvey AK, Yu X-P, et al.
2016;356:635;
J Pharmacol Exp Ther.
(d) Jin Y, Smith CL, Hu L, et al. Clin Pharmacol Ther. 2016;99:274.
23. The structure of GRC27864 has not been disclosed yet. See: GRC27864 First in
37. Gharat LA, Muthukaman N, Narayana L, Khairatkar-Joshi N, Kattige VG.
Tricyclic Compounds as mPGES-1 Inhibitors. PCT Intl Appl WO/2012/055995,
2012.
Man,
Single
Ascending
Dose
Study
in
Healthy
Volunteers.
http://clinicaltrials.gov/ct2/show/NCT02179645.
38. In vivo efficacy data (oral) of high cell potent compounds from the Refs.
23,36,37 were not disclosed in the public domain and will be published in due
course.
39. (a) Compound 16b was not selected for further profiling in spite of better cell
potency, as it had nucleophilic 2,6-dicyanophenyl at C(2)-position of
chromenoimidazole core. Furthermore, DEREK Toxicity Alert program (in
silico predictive model) of 16b revealed potential mutagenicity and
nephrotoxicity. For the related references, see Sanderson DM, Earnshaw CG.
Hum Exp Toxicol. 1991;10:261–273;
24. After the discovery of MF-63, majority of the mPGES-1 inhibitors published in
the literature were having non-acidic core (hinge-binding region) for the
potential activity and devoid of carboxylic acid. These are called non-acidic
mPGES-1 inhibitors. The development of carboxylic acid containing mPGES-1
inhibitors were suffered from significant right shift and variable selectivity
with other PG synthases in the cellular assay due to high protein binding and
structural similarities with PGH2. For more details, see Ref. 7.
25. (a) No developmental activity has been reported to compound IX to date. The
de-bromo derivative IX exhibited 11-fold lower mPGES-1 potency.;
(b) Wu TYH, Juteau H, Ducharme Y, et al. Bioorg Med Chem Lett. 2010;20:6978.
26. (a) Sun H, Tawa G, Wallqvist A. Drug Discovery Today. 2012;17:310;
(b) Schneider G, Schneider P, Renner S. QSAR Comb Sci. 2006;25:1162.
27. Gharat, L. A.; Gajera, J. M.; Narayana, L.; Khairatkar-Joshi, N.; Kattige, V. G.
Tricyclic compounds as mPGES-1 inhibitors. PCT Intl. Appl. WO/2012/110860,
2012.
(b) Ridings JE, Barratt M, Cary RD, et al. Toxicology. 1996;106:267–279;
(c) Cariello NF, Wilson JD, Britt BH, Wedd DJ, Burlinson B, Gombar V.
Mutagenesis. 2002;17:321–329.
40. MF-63 (in vivo EDmax = ꢀ30 mg/kg) and MK-7285 (in vivo EDmax = 14 mg/kg)
having 2,6-dicyanophenyl at 2-position of imidazole core were not moved to
clinical development due to various unknown reasons. For details see Refs.
3b,8a,16a.
28. For the oxidation of chroman-4-one to chroman-3,4-dione, see: Vinot N, Maitte
P. J Heterocycl Chem. 1989;26:1013.
41. Yaksh TL, Kokotos G, Svensson CI, et al. Pharmacol Exp Ther. 2006;316:466.
42. Futaki DN, Harada M, Sugimoto M, et al. J Pharm Pharmacol. 2009;61:607.
29. See Supporting information for the in vitro and in vivo (pharmacology and
DMPK) experimental details.
Please cite this article in press as: Muthukaman N., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.03.068