Synthesis and receptor docking studies of N-substituted indole-2-carboxylic acid esters as a search for COX-2 selective enzyme inhibitors

Article abstract of DOI:10.1016/S0223-5234(01)01258-2

A series of N-substituted indole-2-carboxylic acid esters have been prepared by replacing the benzoyl group of indomethacin with a benzyl and a phenyl group. The carbocyclic acid side chain was extended via creating an ester structure by using several dialkylaminoalkyl groups. The receptor docking studies were performed to investigate the docking mode of each compound by using dock 4.0. All the compounds were shown to be docked at the site where intact flurbiprofen was embedded for COX-1 and s-58 (1-phenylsulphonamide-3-trifluoromethyl-5-para-bromophenylpyrazole) for COX-2. It was predicted that N-phenyl-indole-2-carboxylic acid piperazine ester 22 can be a fairly strong COX-2 selective compound which was compared to the others. Other predicted COX-2 selective compounds included are N-H indole-2-carboxylic acid diethyl 30 and piperazine 34 esters. In view of these findings, compounds 22, 30 and 34 were chosen for the in vitro biological assays.

Full text of DOI:10.1016/S0223-5234(01)01258-2

Eur. J. Med. Chem. 36 (2001) 747–770
´
© 2001 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved
PII: S0223-5234(01)01258-2/FLA
Laboratory note
Synthesis and receptor docking studies of N-substituted indole-2-carboxylic
acid esters as a search for COX-2 selective enzyme inhibitors
Sureyya Olgena,*, Eiichi Akahob, Dogu Nebioglua
aDepartment of Pharmaceutical Chemistry, Faculty of Pharmacy, University of Ankara, T-06100 Tandogan Ankara, Turkey
bFaculty of Pharmaceutical Sciences, Kobe Gakuin University, 518 Arise, Ikawadani-cho, Nishiku, Kobe 651-2180, Japan
Received 29 May 2001; accepted 11 July 2001
Abstract – A series of N-substituted indole-2-carboxylic acid esters have been prepared by replacing the benzoyl group of
indomethacin with a benzyl and a phenyl group. The carbocyclic acid side chain was extended via creating an ester structure by
using several dialkylaminoalkyl groups. The receptor docking studies were performed to investigate the docking mode of each
compound by using DOCK 4.0. All the compounds were shown to be docked at the site where intact flurbiprofen was embedded for
COX-1 and s-58 (1-phenylsulphonamide-3-trifluoromethyl-5-para-bromophenylpyrazole) for COX-2. It was predicted that N-
phenyl-indole-2-carboxylic acid piperazine ester 22 can be a fairly strong COX-2 selective compound which was compared to the
others. Other predicted COX-2 selective compounds included are NꢀH indole-2-carboxylic acid diethyl 30 and piperazine 34 esters.
´
In view of these findings, compounds 22, 30 and 34 were chosen for the in vitro biological assays. © 2001 Editions scientifiques et
me´dicales Elsevier SAS
docking / indole esters / COX-2 selectivity / COX-1 and COX-2 inhibition
1. Introduction
COX-2 probably act by binding to a pocket in en-
zyme that is present in COX-2 but not in COX-1 [2,
3]. Preclinical and clinical studies suggest that COX-2
inhibitors are highly promising agents for the treat-
ment of pain and inflammation, and for the preven-
tion of cancer [1]. As a result of this critical finding, a
substantial discovery effort has been underway in the
pharmaceutical industry to identify selective and
orally active COX-2 inhibitors, because they may
provide the desired anti-inflammatory and analgesic
profiles without the deleterious side effects commonly
associated with the existing NSAIDs [4, 5].
Recently a number of selective inhibitors of COX-2
were shown to possess anti-inflammatory activity with
little or no gastric side effects [6–8]. As a complement
to this work, Black et al. [9] thought that it should be
possible to modify a conventional, nonselective
NSAID to obtain COX-2 selectivity, and thus take
advantage of a structural class with a well-established
safety profile. They synthesised a series of COX-2
selective inhibitors based on the nonselective NSAID
The discovery of a second isoform of cyclooxyge-
nase (cyclooxygenase-2), which is expressed in inflam-
matory cells and the central nervous system, but not
in the gastric mucosa, offers the possibility of devel-
oping anti-inflammatory and analgesic agents that
lack the gastrointestinal side effects of currently avail-
able nonsteroidal anti-inflammatory drugs [1].
Both isoforms of the enzyme cyclooxygenase
(COX-1 and COX-2), responsible for prostaglandin
synthesis, have enabled us to develop drugs capable of
sparing the gastric mucosa. The inducible COX-2
enzyme is responsible for some aspects of pain and
inflammation in arthritis while the constitutive COX-
1 enzyme appears responsible for most of the gastro-
protective prostaglandin synthesis in the stomach and
duodenum. Drugs selective in their inhibition for
Abbreviations: NSAIDs, nonsteroidal antiinflammatuvar drugs.
* Correspondence and reprints
E-mail address: olgen@pharmacy.ankara.edu.tr (S. Olgen).

748
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table I. Structural formulas of compounds 5–34.
indomethacin. In earlier studies, it was found that the
COX-2 enzyme might have a larger active site than
COX-1 [1]. Based on this hypothesis, they thought
that it may be possible to increase the size of the
indomethacin nucleus to produce a compound that
would still fit into the COX-2 active site but not into
the COX-1 active site, thus generating the desired
selectivity. Replacement of the 4-chlorobenzoyl group
in indomethacin with a 2,4,6-trichlorobenzoyl group
led to the formation of a reasonably COX-2 selective
compound L-748,780 [9] (figure 1). To pursue this
method, a number of indole acetic acid analogues
from Merck’s sample collection were examined, re-
sulting in the identification of benzoyl indole deriva-
tives 1 and 2 as highly selective COX-2 inhibitors
(figure 1). Several alkyl-substituted propanoic acids 3
of indomethacin were prepared by same authors, and
it was found that the one with an alkyl-substituted
side chain is the most promising (figure 1).
Compound Type of salt
R
R1
R2
5
HCl
HCl
HCl
HCl
CH2Ph
CH2Ph
CH2Ph
CH2Ph
H
H
H
H
CH3
C2H5
pyrrole
pyrimidine
6
7
8
9
HCl
CH2Ph CH3 CH3
10
11
12
13
14
17
18
19
20
21
22
23
24
25
26
27
29
30
31
32
33
34
CH3I
CH3I
CH3I
CH3I
CH3I
HCl
HCl
HCl
HCl
HCl
CH2Ph
CH2Ph
CH2Ph
CH2Ph
H
H
H
H
CH3
C2H5
pyrrole
pyrimidine
CH2Ph CH3 CH3
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
H
H
H
H
H
CH3
C2H5
pyrrole
pyrimidine
CH3 CH3
HCl
H
H
H
H
H
piperazine
CH3
C2H5
pyrrole
pyrimidine
CH3I
CH3I
CH3I
CH3I
CH3I
HCl
CH3I
HCl
HCl
CH3 CH3
H
H
H
H
C2H5
C2H5
pyrrole
pyrimidine
H
H
H
H
HCl
HCl
CH3 CH3
H piperazine
H
Recently, Kalgutkar et al. [10] synthesised some
ester 4 and amide 5 derivatives of indomethacin. They
found that large alkyl, aryl aralkyl and heterocyclic
esters or amides of indomethacin exhibit high potency
and selectivity (figure 1).
These recent findings of indomethacin derivatives
with a high degree of selectivity for COX-2 have led
us to design potent and selective inhibitors 6 (see all
structures of the compounds in table I) based on these
structures (figure 1).
Docking simulations are widely used for screening
of compound libraries to identify new drug leads,
employing a simple model for rapid testing of thou-
sands of compounds. Docking simulations are also
useful for lead enhancement, using more detailed
models to analyse the atomic interactions between
inhibitors and target macromolecules [11].
Fig. 1. Structures of designed compounds as COX-2 selective
inhibitors.

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
749
Undoubtedly, a detailed three-dimensional (3D)
picture of interactions between, synthesised in-
domethacin derivatives, COX-1 and COX-2 enzyme
active sites would help us to determine the mechanism
of activity.
In this study, the synthesised indole derivatives
were docked in the cyclooxygenase channel using an
in-house DOCK 4.0 program, in order to determine
differences in binding modes of compounds in the
cyclooxygenase channel.
water molecules and ions were removed before the
docking, The docking results were evaluated based on
the method described in Ref. [13].
3. Results and discussion
3.1. Chemistry
The protection of the NH group of indoles is impor-
tant in synthetic indole chemistry and several methods
for the protection of indole nitrogen have been devel-
oped [14–16]. Among these, the benzyl group is found
to be the most stable protecting group. In this study we
have synthesised a 1-benzyl indole-2-carboxylic acid
derivative by using the method of Murakami et al. [17].
The N-benzyl indole-2-carboxylic acid was prepared in
good yields from the corresponding sodium salt of
indole-2-carboxylic acid with benzyl chloride in
dimethylformamide (figure 2).
2. Docking study
The docking program used was DOCK 4.0 devel-
oped by Kuntz et al. [12]. INSIGHT II (Molecular
Simulation Incorporated) was used for visualisation
and molecular modelling of the compounds. The 3D
coordinates of COX-1 and COX-2 were obtained
through the Internet at http://pdb.protein.osaka-
u.ac.jp/pdb, the Research Collaboratory for Struc-
tural Bioinformatics (RCSB). Upon acquiring the
COX-1 and COX-2 coordinates through the Internet,
Indole containing aryl substituents at the nitrogen are
not easily accessible by previous synthetic methods [18].
In the 1970s, Ullmann’s reaction (copper-catalysed con-
densation using aryl halides) proved to be an efficient
Figure 2. Synthesis of N-benzyl-indole-2-carboxylic acid eters. Reagents: (a) HCl gase/MeOH, rt; (b) BrCH2Ph, NaH, DMF, rt; (c)
10% NaOH/MeOH, reflux; (d) 1, 1%-carbonyldiimidazole/DMF; (e) HCl gase/anhydrous diethylether, rt; (f) Mel/anhydrous
diethylether, rt.

750
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Figure 3. Synthesis of N-phenyl-indole-2-carboxylic acid esters. Reagents: (a) BrPh, anhydrous K2CO3, DMF, 154°C; (b)
1,1%-carbonyldiimidazole/DMF, rt; (c) HCl gase/anhydrous diethylether, rt; (d) Mel/anhydrous diethylether, rt.
Figure 4. Synthesis of indole-2-carboxylic acid esters. Reagents: (a) 1,1%-carbonyldiimidazole, DMF, rt; (b) HCl gase/anhydrous
diethylether, rt; (c) Mel/anhydrous diethylether, rt, in dark.
method for the synthesis of N-arylazoles and was used
with success for the N-arylation of indole [19, 20]. In
this work bromobenzene was chosen as the arylating
agents for the synthesis of N-phenyl-indole-2-carboxylic
acid derivatives. In this arylation, good yields of N-
arylindole-2-carboxylic acids were obtained (figure 3).
The catalyst used for the arylations was copper(II)
oxide. It was reported that N%,N-dimethylformamide
was the best solvent [21]. Therefore N,N-dimethylfor-
mamide was used as the reaction solvent to obtain good
yield.
N-Phenyl or N-benzyl indole-2-carboxylic acid esters
(table I) were prepared by the reaction of N-phenyl or
N-benzyl indole-2-carboxylic acid with the coupling
reagent 1,1%-carbonylbis(1H-imidazole) [22]. Ester syn-
thesis by using this reagent was utilised because it
provided an efficient way to synthesise and isolate the
desired esters in high yield and purity (figures 2–4).

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
751
and for the ꢀCH3ꢀ proton a doublet at 1.20–1.35 ppm.
The results of elemental analysis and mass spectroscopic
analysis also confirmed the structures deduced.
Spectroscopic methods and microanalysis confirmed
the structure of the compounds mentioned. The purity
of the compounds was checked by TLC on silica gel HF
254+366 nm and they have sharp melting points (m.p.).
The UV spectra of the compounds show main intensive
absorption bands between 291 and 301 nm. In the IR
spectra of all the compounds, carbonyl-stretching bands
belonging to ester were seen in addition to the carbonyl
band at 1705–1725 cm−1. In the 1H-NMR spectra,
benzylic protons at the indole ring were seen as a singlet
at 5.8–5.9 ppm. In all the quaternary ammonium com-
pounds, which have methyl halogen salts, methyl pro-
tons were seen as singlets at 2.9–3.2 ppm. The
derivatives containing methyl substitution at the neigh-
bouring protons to the ester carbonyl atom showed for
the ꢀCHꢀ proton a quartet between 5.3 and 5.5 ppm
3.2. Computational receptor docking studies
In the current discussion of this paper two different
types of enzymes, COX-1 which should not be inhibited
by a proposed compound, and COX-2, which should be
inhibited by the proposed compound, are involved.
Therefore, an ideal docking result of the proposed com-
pound should be such that ‘a proposed compound
docked more firmly or inhibited more strongly the en-
zyme COX-2 than the enzyme COX-1.’ Secondly, in
order to perform a computer docking study an appro-
priate enzyme should be obtained. Protein Data Bank
Figure 5.

752
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Figure 6.
COX-1 complex and treated as a so-called model in-
hibitor for the docking study.
For the COX-2 complex pdb code 6cox was selected
and used for the docking study. The motif of the
complex is shown in Figure 6. Three types of ligand were
clarified as HEM (protoporphyrin IX Fe [Heme], NAG
(URL: http://www.rcsb.org/pdb/index.html) is one of
the relevant Internet web sites where one can obtain
over 12 000 protein structures. For COX-1 the one with
pdb code 1cqe was used. The pdb file 1cqe is a complex
of COX-1 with four ligands. The motif of the complex is
shown in figure 5. In the COX-1 portion of the complex,
five sheets, ten strands, 32 helices, 45 beta turns, seven
gamma turns, one betabulge, five betahairpins, and five
disulphide bridges are identified (Figure 5). Four types
(N-acetyl-D-glucosamine), and s58 (1-phenylsulphona-
mide-3-trifluoromethyl-5-rabromophenylpyrazole).
Since s58 seems to play a role of inhibitor, the vicinity
where this s58 is situated should be an active site.
Therefore, s58 was taken out of the pdb file of this
COX-2 complex and treated as the so-called model
inhibitor for the docking study.
The next question is how one can judge from the
computational docking result whether ‘a proposed com-
pound docked more firmly or inhibited more strongly
of ligands are clarified as NAG–NAG (N-acetyl-D-glu-
cosamine), HEM (protoporphyrin IX containing Fe
[Heme], BOG (b-octylglucoside), and FLP (flurbipro-
fen). Since flurbiprofen is a known nonsteroidal anti-
inflammatory drug (NSAID), the vicinity where this
flurbiprofen is situated should be an active site. There-
fore, flurbiprofen was taken out of the pdb file of this

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table II. Docking results of indomethacin derivatives.
753
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond (distance
A)
Energy
Energy
H-bond (distance
A)
Energy
(kcal mol−1
01. −31.01
02. −30.90
03 −30.59
)
(kcal mol−1
)
(kcal mol−1
01. −35.95
02. −35.88
03. −35.53
)
(kcal mol−1
)
5
6
17.N-2 with O-1 of −20.36
Leu A321 (2.60)
18.N-2 with O-1 of −3.67
Leu A321 (2.60)
21.N-2 with O-2 of
Ser A499 (2.60)
23.N-2 with O-1 of 444.65
Phe A350 (2.75)
11.83
01. −4.38
02. −4.22
03. −4.18
15.N-1 with O-1 of −2.59
Met A491 (2.69)
15.N-2 with H-1 of −2.59
Met A494 (1.73)
15.N-2 with O-2 of −2.59
Ser A490 (0.37)
01. −11.15
02. −1014
03. −6.15
1.N-2 with O-1 of −11.15
Leu A321 (2.97)
5.N-2 with N-1 of
Leu A321 (2.82)
9.N-2 with H-1 of
Tyr A324 (2.23)
−4.59
−1.90
12.N-2 with O-1 of −3.22
Pro A497 (2.72)
12.N-2 with O-1 of −3.22
Ala A496 (2.03)
13.N-2 with O-1 of −2.77
Ile A492 (2.83)
13.N-2 with O-1 of −2.77
Met A491 (2.01)
7
01. −4.32
02. −4.26
03. −4.16
1. N-2 with O-1 of −4.32
Ile A492 (2.20)
1. O-1 with H-1 of −4.32
Ser A499 (2.37)
01. −26.57
02. −22.44
03. −21.86
15. N-2 with O-1
of Tyr A324 (2.81)
15. N-2 with H-1
of Tyr A324 (1.94)
19. N-2 with H-1
of Arg A89 (1.67)
19. N-2 with H-1
of Arg A89 (2.50)
14. N-2 with H-1
of Tyr A324 (2.25)
34.69
34.69
71.69
71.69
31.68
1. O-2 with N of
Pro A497 (2.25)
−4.32
20.N-2 with O-3 of −2.6
Glu A493 (2.19)
20.N-2 with H-1 of −2.6
Arg A89 (0.82)
20.N-1 with O-1
Tyr A324 (2.70)
−2.6
3. N-2 with O-1 Ile −4.16
A492 (2.33)
3. N-2 with O-1
Met A494 (2.83)
−4.16
8
9
01. −3.22
02. −2.75
03. −2.46
9. N-1 with O-1 of −1.09
Ser A499 (2.57)
9. N-2 with O-1 of −1.09
Gly A495 (2.20)
8. N-2 with O-1 of −1.1
Ser A322 (2.78)
12.N-2 with O-1 of −0.92
Val A318 (2.59)
01. −8.56
02. −5.21
03. −4.69
6. N-2 with H-1 of −1.00
Tyr A324 (2.36)
7. N-2 with O-1 of −0.45
Met A491 (2.51)
9. N-2 with H-1 of
His A58 (2.16)
2.45
01. −3.13 02.
−3.10 03. −3.09 Gln A161 (2.96)
23.N-2 with O-1 of −1.96
23.N-2 with O-1 of −1.96
01. −1.61
02. −0.88
03. −0.72
18.N-2 with O-1 of
Leu A321 (2.68)
18.N-2 with O-2 of
Ser A499 (2.86)
12.N-2 with H-1 of
His A58 (2.14)
33.33
33.33
13.28
Leu A321 (2.59)
23.N-2 with O-2 of −1.96
Ser A322 (2.42)

754
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table II. (Continued)
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
Energy
H-bond
(distance A)
Energy
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
23.O-2 with H-1 of −1.96
Tyr A324 (2.36)
16.N-2 with O-1 of
Ser A499 (2.63)
19.43
14.N-2 with O-1 of −2.58
Ile A492 (0.79)
14.N-2 with O-1 of −2.58
Met A491 (2.81)
14.N-2 with H-1 of −2.58
Ala A496 (1.38)
6.N-2 with O-1 of
Val A318 (2.27)
6.N-2 with O-1of
Ser A322 (2.24)
−2.9
−2.9
Note. (1) Energy represents the total binding energy between a ligand and a protein. (2) Details of circled numbers in H-bond
column are explained in figures. 4 and 9. (3) Front numbers in the first energy column and in the H-bond column represent the
docking result numbers which are in the order of the lowest energy first.
the enzyme COX-2 than the enzyme COX-1.’ Akaho et
al. proposed a method [13], and the modified version by
the current authors are shown as follows:
(e.g. one hydrogen bond) with the lowest binding
energy. If this docking result belongs to COX-2 then
one can say that a compound is COX-2 selective, or
it is docked more firmly or inhibited more strongly
with enzyme COX-2 than the enzyme COX-1, visa
versa.
1. After docking of a proposed compound against
COX-1 and COX-2, examine the top 25 docking
results which are arranged in the order of the one
with the lowest energy first, and pick up for COX-1
and COX-2, respectively, to three docking results
with negative binding energy (eliminate the one with
positive binding energy) in the order to the one with
more hydrogen bonds first creating List A.
2. Pick up from the List A the one with the highest
number of hydrogen bonds (e.g. two hydrogen
bonds) with the lowest binding energy. If this dock-
ing result belongs to COX-2 then one can say that a
compound is COX-2 selective, or it is docked more
firmly or inhibited more strongly with the enzyme
COX-2 than the enzyme COX-1, visa versa.
3. If both COX-1 and COX-2 docking results show the
same in terms of the number of hydrogen bonds and
the energy, then pick up the one whose number of
hydrogen bonds is the highest (e.g. two hydrogen
bonds) with the next lowest binding energy. If this
docking result belongs to COX-2 then one can say
that a compound is COX-2 selective, or it is docked
more firmly or inhibited more strongly with the
enzyme COX-2 than the enzyme COX-1, visa versa.
4. As the next priority selection step, pick the one
which has next highest number of hydrogen bonds
5. If both COX-1 and COX-2 docking results show the
same in terms of the number of hydrogen bonds and
the energy, then pick up the one whose number of
hydrogen bonds is next (e.g. one hydrogen bond)
with the next lowest binding energy. If this docking
result belongs to COX-2 then one can say that a
compound is COX-2 selective, or it is docked more
firmly or inhibited more strongly with the enzyme
COX-2 than the enzyme COX-1, visa versa.
With this criteria in mind docking between in-
domethacin derivatives and COX-1 and COX-2 were
performed. The protein structures of COX-1 and COX-
2 were obtained from the Protein Data Bank previously
mentioned. Based on the above criteria, dock results of
synthesised indomethacin derivatives were evaluated and
are shown in tables 2–5. Atoms responsible for H-bonds
are shown in figures 7–11. After docking, compound 22
formed three H-bonds against COX-2, while its binding
energy was −17.49 kcal mol−1. Since its binding energies
against COX-1 were positive it was eliminated from the
evaluation list. Therefore, it was predicted that com-
pound 22 was a COX-2 selective NSAID. The first
location of the H-bond formed was between the piper-

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table III. Docking results of indomethacin derivatives.
755
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond (distance A) Energy
(kcal mol−1
Energy
H-bond (distance A) Energy
(kcal mol−1
)
(kcal mol−1
01. −2.52
02. −2.51
03. −2.47
)
)
(kcal mol−1
01. −29.89
02. −28.95
03. −28.64
)
29
2.N-2 with H-1 of
Leu A500 (2.37)
2.N-2 with H-1 of
Ser A499 (2.09)
−2.51
17.O-2 with H-1 of
Tyr A324 (2.21)
−23.60
−2.51
−2.51
−2.51
−2.52
−2.52
−2.52
−1.6
17.N-2 with H-1 of
Arg A89 (2.44)
−23.60
325.32
325.32
445.94
445.94
2.N-2 with O-1 of
Gly A495 (2.29)
2.N-2 with O-1 of
Ala A496 (1.11)
1.N-2 with O-1 of
Gly A495 (2.38)
1.N-2 with O-1 of
Ala A496 (1.13)
1.N-2 with H-1 of
Leu A500 (2.13)
15.N-2 with O-1 of
Leu A321 (1.46)
15.O-1 with H-1 of
Tyr A324 (2.33)
24.N-2 with O-1 of
Gly A488 (2.67)
24.O-1 with H-1 of
Phe A487 (2.24)
25.N-2 with O-1 of
Leu A321 (2.57)
25.N-2 with O-1 of
Gln A161 (2.70)
−1.6
10
11
01. 0.95
02. 1.02
03. 1.21
16.N-1 with O-1 of
Tyr A324 (2.42)
16.O-1 with H-1 of
Tyr A324 (2.03)
7.N-1 with O-1 of
Tyr A324 (2.42)
18.N-1 with O-1 of
Ser A499 (2.25)
1.62
1.62
1.36
1.68
01. −23.66
02. −23.64
03. −23.37
13.O-2 with H-1 of
Arg A89 (2.44)
14.O-2 with H-1 of
Arg A89 (2.44)
15.O-2 with H-1 of
Arg A89 (2.44)
2.57
5.50
7.96
01. 0.64
02. 0.55
03. 0.60
15. N-1 with O-1 of
Gly A495 (2.98)
15. O-1 with H-1 of
Ala A496 (2.38)
3. O-1 with H-1 of
Tyr A324 (1.78)
0.95
0.95
0.6
01. −35.13
02. −34.56
03. −31.85
24. O-2 with H-1 of
Arg A89 (2.29)
29.46
6.N-1 with O-1 of
Met A491 (2.62)
0.67
30
23
01. 3.34
02. 3.66
03. 3.73
20.N-1 with O-1 of
Ser A499 (2.74)
21.N-1 with O-1 of
Ile A492 (2.71)
4.95
5.02
01. −22.73
02. −22.01
03. −21.90
10.O-2 with H-1 of
Asp A484 (2.41)
12.O-2 with H-1 of
Asp A484 (2.41)
13.O-2 with H-1 of
Asp A484 (2.41)
−17.76
−17.07
−17.04
01. 3.43
02. 3.57
03. 3.60
13.N-1 with S of
Met A491 (3.68)
13.N-1 with O-1 of
Met A491 (2.89)
14.N-1 with S of
Met A491 (3.68)
14.N-1 with O-1 of
Met A491 (2.89)
25.N-1 with O-1 of
Ile A492 (2.89)
4.40
4.40
4.44
4.44
5.14
01. −23.05
02. −22.85
03. −18.72
no hydrogen bond

756
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table III. (Continued)
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
Energy
H-bond
(distance A)
Energy
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
25.N-1 with O-1 of
Met A491 (2.60)
5.14
24
01. 2.31
02. 2.92
03. 3.11
2.O-1 with H-1 of
Tyr A324 (2.25)
4.O-2 with H-1 of
Tyr A324 (2.25)
7.O-1 with H-1 of
Tyr A324 (2.25)
2.92
3.20
3.33
01. −7.75
02. −5.00
03. 1.36
no hydrogen bond
12
25
01. −1.99
02. −1.89
19.N-1 with O-1 of
Tyr A324 (1.32)
19.O-1 with H-1 of
His A59 (2.80)
0.57
0.57
01. −30.65
02. −30.30
25.O-1 with H-1 of
His A58 (2.42)
947.68
03. −1.05
01. −0.74
03. −29.35
01. −25.01
1.N-1 with O-1 of
Met A491 (2.88)
1.O-2 with H-1 of
Met A494 (2.46)
1.O-2 with H-1 of
Gly A495 (1.98)
4.N-1 with O-1 of
Met A491 (2.88)
4.O-2 with H-1 of
Met A494 (2.46)
4.O-2 with H-1 of
Gly A495 (1.98)
6.N-1 with O-1 of
Met A491 (2.88)
6.O-2 with H-1 of
Met A494 (2.46)
6.O-2 with H-1 of
Gly A495 (1.98)
−0.74
−0.74
−0.74
−0.29
−0.29
−0.29
−0.17
−0.17
−0.17
02. −0.47
03. −0.38
02. −24.95
03. −22.20
no hydrogen bond
13
01. −1.31
02. −1.29
03. −0.96
23.O-2 with H-1 of
Gly A495 (2.01)
23.O-2 with H-1 of
Met A494 (1.06)
23.N-1 with O-1 of
Met A491 (1.72)
11.N-1 with O-1 of
Tyr A354 (2.90)
11.N-1 with O-1 of
Ser A499 (2.86)
15.O-1 with H-1 of
Tyr A324 (1.05)
0.55
0.55
0.55
0.33
0.33
0.46
01. −22.88
02. −20.83
03. −19.58
no hydrogen bond
26
01. −1.80
02. −1.53
03. −1.09
21.O-1 with H-1 of
Ala A496 (2.34)
21.O-2 with H-1 of
Ala A496 (2.47)
21.O-2 with H-1 of
Gly A495 (1.49)
0.97
0.97
0.97
01. −22.08
02. −8.11
03. −4.51
no hydrogen bond

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
757
Table III. (Continued)
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
Energy
H-bond
(distance A)
Energy
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
9.N-1 with O-1 of
Met A491 (2.91)
−0.43
14
27
01. −2.87
02. −2.87
03. −2.69
23.O-1 with H-1 of
Tyr A89 (1.88)
23.O-2 with H-1 of
Arg A324 (2.33)
21.O-2 with H-1 of
Gly A495 (2.11)
−1.80
−1.80
−1.82
01. −15.98
02. −15.93
03. −15.76
no hydrogen bond
no hydrogen bond
01. −2.20
02. −2.15
03. −1.97
20.N-1 with O-1 of
Met A491 (2.67)
23.O-2 with H-1 of
Trp A356 (2.29)
25.N-1 with O-1 of
Ser A499 (2.90)
−1.77
−1.72
−1.59
01. −23.70
02. −22.59
03. −22.42
Note. (1) Energy represents the total binding energy between a ligand and a protein. (2) Details of circled numbers in H-bond
column are explained in figures. 5, 7, and 9. (3) Front numbers in the first energy column and in the H-bond column represent the
docking result numbers which are in order of the lowest energy first.
compounds suggest that not all carboxylate-containing
NSAIDs will be converted into COX-2 inhibitors by
esterification.
In our studies, molecular docking results indicate the
binding capability of the compounds to the enzyme
active site and the results, which we have obtained from
the activity tests, show COX-2 enzyme inhibitory capac-
ity. Therefore, as it is mentioned in general, it is not
always possible to obtain positive correlation between
activity and docking results.
azine hydrogen of compound 22 and the carbonyl
,
oxygen of Phe A487 with its distance being 1.93 A
(figure 13). The second location of the H-bond formed
was between the carbonyl oxygen of compound 22 and
the amine hydrogen of Phe A487 with its distance being
,
2.22 A. It was also found that compounds 30 (figure 14)
and 34 (figure 15) had more H-bonds against COX-2
than COX-1 and predicted to be a COX-2 selective
NSAID (figure 12).
3.3. Biological assays
4. Experimental protocols
Compounds 22, 30 and 34 were tested for the inhibi-
tion of purified human COX-2 and ovine COX-1 by
using thin layer chromatography (TLC) assay [23].
These compounds were chosen for the biological assays
because of their significant binding capability to the
COX-2 enzyme, according to molecular docking results.
Biological activity tests were applied by comparing the
COX-2 inhibitory effects of these compounds with in-
domethacin. Unfortunately, none of them showed any
inhibitory effect concentration up to 50 mM. The results
are shown in table VI. Although the present studies
indicate that the compounds containing large alkyl, aryl,
aryalkyl, and heterocyclic esters and amides exhibit high
potency and selectivity [23], the IC50 values of our test
Indole-2-carboxylic acid, 1,1%-carbonyldiimidazole,
2-hydroxyethylpiperazine from Fluka; benzyl bro-
mide, bromobenzene, anhydrous CaCl2, anhydrous
K2CO3, NaOH, isopropanol, hexanes, ether,
methanol, acetone, ethylacetate, hydrogen chloride,
ethanol from Merck and dimethylaminoethanol, 2-
hydroxyethylpyrolidine, dimethylamino-2-propanol,
2-hydroxyethylpiperidin, 2-hydroxyethylpiperazine,
dimethylamine, potassium bromide, sodium hydride
were purchased from Aldrich.
Melting points were recorded in a Buchi SMP 20
melting point apparatus. H-NMR spectra were con-
sistent with molecular structures and were recorded in
1

758
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table IV. Docking results of indomethacin derivatives.
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
Energy
H-bond (distance A) Energy
(kcal mol−1
H-bond (distance A) Energy
(kcal mol−1
)
(kcal mol−1
)
)
(kcal mol−1
01. −36.52
02. −36.49
03 −36.30
)
17
01. −2.10
02. −1.65
03 −1.63
25.N-1 with O-1 of
Met A491 (2.19)
25.N-2 with O-2 of
Ser A490 (2.19)
−0.19
25.N-2 with O-1 of
Tyr A324 (2.56)
25.O-2 with H-1 of
Tyr A324 (2.48)
21.N-2 with H-1 of
His A58 (2.50)
78.43
−0.19
−0.19
−0.19
−0.19
−2.10
−2.10
−2.10
−1.05
−1.05
78.43
25.N-2 with S of
Met A494 (2.68)
25.O-2 with H-1 of
Met A494 (2.25)
25.O-2 with H-1 of
Gly A495 (2.31)
1.N-2 with O-1 of
Gly A495 (2.64)
1.N-2 with O-1 of
Met A494 (2.93)
1.N-2 with O-1 of
Met A491 (2.88)
19.N-2 with O-1 of
Met A491 (1.55)
19.N-2 with S of
Met A491 (2.87)
−16.14
18
01. −3.67
02 −3.38
03. −3.31
13.N-2 with O-2 of
Ser A322 (2.53)
13.N-2 with O-1 of
Leu A321 (2.76)
13.N-2 with O-1 of
Ser A322 (2.66)
17.N-2 with O-1 of
Met A491 (2.56)
17.N-2 with O-1 of
Ile A492 (2.02)
17.N-1 with O-1 of
Tyr A324 (2.59)
8.N-2 with O-1 of
Met A494 (1.46)
8.N-1 with O-1 of
Gly A495 (2.74)
−2.44
−2.44
−2.44
−2.18
−2.18
−2.18
−2.80
−2.80
01. −25.89
02. −25.22
03. −23.69
22.N-2 with O-1 of
Tyr A324 (2.58)
22.N-2 with H-1 of
Arg A482 (1.97)
22.N-2 with H-1 of
Tyr A324 (1.82)
23.O-2 with N of
Pro A483 (2.52)
23.O-1 with N-2 of
His A58 (2.96)
32.38
32.38
32.38
76.41
76.41
−9.50
17.N-2 with H-1 of
Tyr A324 (2.50)
19
01. −2.80
02. −2.62
03. −2.59
3.N-2 with O-1 of
Ser A499 (2.55)
3.N-2 with O-1 of
Gly A495 (1.80)
3.N-1 with S of
Met A491 (3.24)
3.N-1 with O-1 of
Met A491 (2.94)
19.N-2 with S of
Met A491 (2.38)
19.N-2 with O-1 of
Leu A353 (2.20)
19.N-2 with H-1 of
Trp A356 (2.43)
−2.59
−2.59
−2.59
−2.59
−1.82
−1.82
−1.82
01. −32.66
02. −32.23
03. −26.50
7.N-2 with H-1 of
Arg A89 (2.08)
7.O-2 with H-1 of
Arg A89 (2.46)
1.O-2 with H-1 of
Arg A89 (2.46)
3.O-2 with H-1 of
Arg A89 (2.46)
−20.68
−20.68
−32.66
−26.50

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
759
Table IV. (Continued)
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
Energy
H-bond
(distance A)
Energy
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
19.N-1 with O-1 of
Met A491 (2.17)
12.N-2 with O-1 of
Val A318 (0.90)
12.N-2 with O-1 of
Ser A322 (2.70)
−1.82
−2.06
−2.06
−2.06
12.N-1 with O-1 of
Tyr A324 (2.73)
20
01. −3.10
02. −3.04
03. −2/61
1.N-1 with O-1 of
Tyr A354 (2.83)
1.N-2 with O-1 of
Tyr A354 (1.60)
1.N-2 with O-1 of
Tyr A317 (2.31)
24.N-2 with O-1
Phe A498 (2.78)
24.N-2 with O-2
Ser A499 (1.51)
24.N-2 with O-1 of
Ser A499 (2.97)
13.N-2 with O-1 of
Tyr A324 (1.73)
13.N-2 with H-1 of
Arg A89 (2.35)
−3.10
−3.10
−3.10
−1.13
−1.13
−1.13
−1.74
−1.74
01. −30.25
02. −29.43
03. −7.82
20.N-2 with O-1 of
Tyr A354 (2.61)
20.N-2 with O-1 of
Ser A499 (2.95)
7.N-2 with H-1 of
His A58 (2.13)
66.53
66.53
7.21
11.N-2 with O-1 of
Ser A499 (2.71)
47.22
21
01. −2.84
02. −2.82
03. −2.77
22.N-1 with O-1 of
Met A491 (2.60)
22.N-2 with O-1 of
Ile A492 (2.81)
22.N-2 with O-1 of
Glu A489 (2.31)
10.N-2 with O-1 of
Ala A496 (2.71)
10.N-2 with O-1 of
Gly A495 (2.65)
11.N-2 with O-1 of
Ile A492 (2.96)
−2.21
−2.21
−2.21
−2.56
−2.56
−2.52
−2.52
01. −31.67
02. −31.22
03. −31.20
6.O-1 with H-1 of
Arg A89 (2.34)
10.N-2 with H-1 of
His A58 (2.24)
21N-2 with O-1 of
Tyr A324 (2.84)
−25.73
−20.28
23.47
11.N-2 with O-2 of
Ser A90 (2.37)
22
01. 0.01
02. 0.12
03. 0.57
2.N-2 with O-1 of
Met A491 (3.00)
2.N-2 with O-1 of
Gly A495 (2.77)
2.N-2 with O-1 of
Met A494 (2.65)
2.H-1 with O-1 of
Ala A496 (2.25)
2.H-1 with N of
Ser A499 (2.11)
15.N-2 with O-2 of
Ser A322 (2.61)
0.12
0.12
0.12
0.12
0.12
1.41
01. −24.34
02. −20.71
03. −20.39
4.H-1 with O-1 of
Phe A487 (1.93)
4.H-1 with N-6 of
Phe A487 (2.22)
4.N-3 with H-1 of
Phe A487 (2.17)
18.H-1 with N of
Phe A350 (2.30)
18.H-1 with O-1 of
Phe A487 (1.82)
18.N-3 with H-1 of
Phe A487 (2.17)
−17.49
−17.49
−17.49
−3.33
−3.33
−3.33

760
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table IV. (Continued)
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
Energy
H-bond
(distance A)
Energy
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
(kcal mol−1
)
15.N-2 with O-1 of
Ser A322 (2.18)
15.H-5 with N of
Gln A319 (2.48)
15.N-3 with H-1 of
Ser A322 (1.24)
15.N-3 with H-1 of
Gly A323 (1.96)
5.N-2 with O-1 of
Leu A321 (2.02)
5.N-2 with O-2 of
Ser A322 (2.95)
1.41
1.41
1.41
1.41
0.99
0.99
1.N-2 with H-1 of
His A58 (2.50)
1.N-3 with H-1 of
Phe A487 (2.12)
−24.34
−24.34
Note. (1) Energy represents the total binding energy between a ligand and a protein. (2) Details of circled numbers in H-bond
column are explained in figures 6 and 9. (3) Front numbers in the first energy column and in the H-bond column represent the
docking result numbers which are in order of the lowest energy first.
a Bruker AC 400 spectrometer. Elemental analyses
were performed in a LECO-932 CHNS-O Elemental
Analyser. The IR values were determined with a Pye
Unicam 1025 spectrophotometer. The UV spectral
analyses were measured in a Shimadzu 240 B. High
resolution mass spectra were run in a Fisions
instrument, VG Platform II LC MS spectrometer.
mL of MeOH and was added to 50 mL of 10% NaOH
solution. The reaction mixture was stirred at 65 °C for
2 h. The reaction mixture was cooled down to r.t. and
neutralised by AcOH to give a white precipitate. Crys-
tallisation with MeOH–water gave white crystals (13.0 g,
91.60%): m.p. 191 °C.
4.1.4. 1-Benzyl indole-2-dialkylaminoalkyl carboxylate
hydrogen chloride salts (5–9)
4.1. Chemistry
Compound 3 (1.0 g, 3.9 mmol) was dissolved in 10.0
mL of DMF. 1,1%-Carbonyldiimidazole (0.6 g, 3.9 mmol)
was added in portions. The solution was stirred in
ambient temperature for 1 h. The dialkylaminoalcohols
(3.9 mmol) was added and heated at 50–60 °C for 4–8
h. At the end of the reaction, the reaction mixture was
cooled to r.t. and was neutralised with saturated NaOH
solution. The oily residue was extracted with ether and
was washed with 5% NaCl solution and then water. The
organic phase was dried under anhydrous CaCl2 and
filtered. The filtrate was evaporated to dryness and the
residue was dissolved in anhydrous diethylether, which
was saturated with HCl gas to give the desired final
compounds.
4.1.1. Methyl indole-2-carboxylate (1)
Indole-2-carboxylic acid (25.0 g, 0.15 mol) was dis-
solved in 10% HCl in MeOH and refluxed at 65 °C for
1 h. The solvent was evaporated under vacuum and
neutralised by K2CO3. Crystallisation by ethanol gave
pale white crystals (18.0 g, 66.29%): m.p. 152 °C.
4.1.2. 1-Benzyl methyl indole-2-carboxylate (2)
Compound 1 (17.5 g, 0.10 mol) and 4.4 g (0.18 mol)
50% NaH were dissolved in 10.0 mL of DMF and stirred
at room temperature (r.t.) 30 min. Benzyl bromide (12.1
mL, 0.10 mol) was added dropwise and stirred at r.t. for
48 h. Then the reaction mixture was poured into ice-wa-
ter. Neutralisation with acetic acid gave an oily com-
pound which was crystallised by ethanol–water to gave
pale yellow crystals (16.0 g, 60.30%): m.p. 83 °C.
5: Yield 0.6500 g, 46.62%; m.p. 128 °C; MS; m/z: 322
1
[M+], H-NMR (DMSO-d6): l 2.85 (s, 6H, N(CH3)2),
3.40 (t, 2H, CH2N(CH3)2), 4.65 (t, 2H, COOCH2), 5.85
(s, 2H, CH2Ph), 7.05–7.75 (m, 10H, aromatic protons),
12.85 (s, 1H, NH). IR vmax (KBr, cm−1) 1715. Anal.
(C20H22N2O2·HCl) C, H, N, O.
4.1.3. 1-Benzyl indole-2-carboxylic acid (3)
Compound 2 (15.0 g, 0.56 mol) was dissolved in 50

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table V. Docking results of indomethacin derivatives.
761
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
Energy
H-bond
(distance A)
Energy
(kcal mol−1
01. −1.83
02. −1.82
03. −1.82
)
(kcal mol−1
)
(kcal mol−1
01. −31.43
02. −30.70
03. −28.36
)
(kcal mol−1
)
31
5.N-2 with O-1 of
Met A494 (2.96)
5.N-2 with O-2 of
Ser A490 (2.72)
5.N-2 with O-1 of
Met A491 (2.90)
5.N-1 with O-1 of
Met A491 (1.80)
5.N-1 with O-1 of
Ile A492 (2.13)
5.N-1 with H-1 of
Ala A496 (2.12)
1.N-1 with O-1 of
Ile A492 (2.13)
1.N-1 with O-1 of
Met A491 (1.80)
1.N-2 with O-1 of
Met A494 (2.52)
1.N-1 with H-1 of
Ala A496 (2.12)
11.N-2 with O-1 of
Ile A492 (2.65)
−1.78
−1.78
−1.78
−1.78
−1.78
−1.78
−1.83
−1.83
−1.83
−1.83
−1.59
−1.59
−1.59
25.N-2 with O-1 of
Leu A321 (2.36)
25.N-2 with O-1 of
Gln A161 (2.82)
1.O-2 with H-1 of
His A58 (2.47)
645.36
645.36
−31.43
−30.70
2.O-2 with H-1 of
His A58 (2.47)
11.N-2 with O-1 of
Met A491 (1.01)
11.N-2 with H-1 of
Gly A495 (1.69)
32
01. −1.50
02. −1.43
03. −1.38
25. N-2 with H-2 of
Tyr A354 (2.26)
25. N-2 with O-1 of
Phe A350 (1.18)
25. N-2 with O-1 of
Glu A349 (2.64)
1.N-1 with O-1 of
Met A491 (2.80)
1.N-2 with S of
Met A491 (3.65)
6.N-2 with O-1 of
Tyr A354 (1.93)
6.N-2 with O-1 of
Tyr A317 (2.85)
0.48
0.48
01. −29.77
02. −27.98
03. −27.19
5. O-1 with H-1 of
Phe A487 (2.45)
6.N-2 with H-1 of
His A58 (2.40)
8. O-1 with H-1 of
Phe A487 (2.45)
−26.89
−25.68
−25.21
0.48
−1.50
−1.50
−1.30
−1.30
33
01. −2.00
02. −1.97
03. −1.92
8.N-2 with O-2 of
Tyr A354 (2.80)
8.N-2 with O-1 of
Leu A353 (2.90)
8.N-1 with S of
Met A491 (3.63)
3.N-2 with S of
Met A491 (3.67)
3.N-2 with O-1 of
Met A491 (2.74)
9.N-2 with O-1 of
Gly A495 (2.12)
−1.76
−1.76
−1.76
−1.92
−1.92
−1.73
01. −27.53
02. −27.25
03. −27.25
25.N-2 with S of
Met A491 (2.53)
25.N-2 with H-1 of
Trp A356 (1.72)
21.N-2 with H-1 of
Tyr A324 (2.49)
23.O-1 with H-1 of
Phe A487 (2.47)
574.93
574.93
−12.96
2.38

762
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
Table V. (Continued)
Indomethacin Against COX-1
derivatives
Against COX-2
Energy
H-bond
(distance A)
Energy
(kcal mol−1
Energy
(kcal mol−1
H-bond
(distance A)
Energy
(kcal mol−1
)
(kcal mol−1
)
)
)
9.N-2 with O-1 of
Met A494 (2.58)
−1.73
34
01. 1.43
6.N-1 with O-1 of
Ser A499 (2.17)
6.N-2 with O-1 of
Ser A499 (2.35)
6.N-2 with O-1 of
Ala A496 (2.46)
6.H-1 with N-6 of
Phe A498 (1.72)
8.N-2 with O-1 of
Tyr A324 (1.06)
8.O-2 with H-1 of
Tyr A324 (2.32)
8.N-3 with H-1 of
Arg A89 (2.25)
12.N-2 with H-1 of
Ala A496 (1.45)
12.N-2 with O-1 of
Ile A492 (1.07)
12.H-1 with N-3 of
Arg A89 (1.84)
1.60
1.60
1.60
1.60
1.63
1.63
1.63
1.81
1.81
1.81
01. −29.09
02. −28.31
03. −28.06
1.O-2 with H-1 of
His A58 (2.50)
1.O-1 with H-1 of
Phe A487 (2.47)
4.O-2 with H-1 of
His A58 (2.50)
4.O-1 with H-1 of
Phe A487 (2.47)
5.O-2 with H-1 of
His A58 (2.50)
5.O-1 with H-1 of
Phe A487 (2.47)
−29.09
−29.09
−25.61
−25.61
02. 1.48
03. 1.49
Note: (1) Energy represents the total binding energy between a ligand and a protein. (2) Details of circled numbers in H-bond
column are explained in figures 8 and 9. (3) Front numbers in the first energy column and in the H-bond column represent the
docking result numbers which are in order of the lowest energy first.
9: Yield 0.5212 g, 35.20%; m.p. 124–125 °C; MS;
m/z: 336 [M+], 1H-NMR (DMSO-d6): l 1.30 (d, 3H,
COOCH(CH3), 2.80 (s, 6H, N(CH3)2), 3.40 (t, 2H,
CH2N(CH3)2), 5.40 (m, 1H, COOCH(CH3), 5.85 (s, 2H,
CH2Ph), 7.05–7.80 (m, 10H, aromatic protons), 9.12 (s,
1H, NH). IR vmax (KBr, cm−1) 1715. Anal. (C21H24-
N2O2·HCl·0.1H2O) C, H, N, O.
6: Yield 0.9686 g, 63.05%; m.p. 144–145 °C; MS;
m/z: 350 [M+], 1H-NMR (DMSO-d6): l 1.30 (s, 6H,
N(CH2CH3)2), 3.15 (q, 4H, CH2N(CH2CH3)2), 3.40 (t,
4H, CH2N(CH2CH3)2), 4.75 (t, 2H, COOCH2), 5.80 (s,
2H, CH2Ph), 6.90–7.80 (m, 10H, aromatic protons),
12.30 (s, 1H, NH). IR vmax (KBr, cm−1) 1715. Anal.
(C22H26N2O2·HCl) C, H, N, O.
7: Yield 0.5529 g, 36.18%; m.p. 136–138 °C; MS;
4.1.5. 1-Benzyl indole-2-dialkylaminoalkyl carboxylate
methyl iodide salts (10–14)
1
m/z: 348 [M+], H-NMR (DMSO-d6): l 0.70–5.10 (m,
12H, pyrrolidine and CH2-pyrrolidine), 5.80 (s, 2H,
CH2Ph), 6.90–7.80 (m, 10H, aromatic protons), 12.45
The intermediate esters were synthesised as described
for compounds 5–9. The solution of esters in ether was
treated with alkyl iodide (1.20 mmol) at r.t. and was
stirred overnight with care to protect from light. The
quaternary methyl iodide compounds were obtained as a
thick white precipitate.
(s, 1H, NH). IR vmax (KBr, cm−1
(C22H24N2O2·HCl) C, H, N, O.
) 1725. Anal.
8: Yield 0.5669 g, 35.79%; m.p. 154 °C; MS; m/z: 362
1
[M+], H-NMR (DMSO-d6): l 0.70–3.60 (m, 12H, pipe-
ridine and CH2-piperidine), 4.90 (t, 2H, COOCH2), 5.80
(s, 2H, CH2Ph), 7.05–7.75 (m, 10H, aromatic protons),
12.20 (s, 1H, NH). IR vmax (KBr, cm−1) 1720. Anal.
(C23H26N2O2·HCl) C, H, N, O.
10: Yield 0.6208 g, 33.68%; m.p. 219–221 °C; MS;
m/z: 322 [M+], 1H-NMR (DMSO-d6): l 3.20 (s, 9H,
N(CH3)3), 3.80 (t, 2H, CH2N(CH3)3), 4.75 (t, 2H,
COOCH2), 5.85 (s, 2H, CH2Ph), 7.05–7.75 (m, 10H,

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
763
aromatic protons), 12.85 (s, 1H, NH). IR vmax (KBr,
cm−1) 1720. Anal. (C20H22N2O2·CH3I) C, H, N, O.
11: Yield 0.8165 g, 41.79%; m.p. 200–201 °C; MS;
m/z: 350 [M+], 1H-NMR (DMSO-d6): l 1.30 (s, 6H,
N(CH2CH3)2), 3.20 (s, 3H, N+ꢀCH3), 3.60 (q, 4H,
N(CH2CH3)2), 4.00 (t, 2H, CH2N(CH2CH3)2), 4.70 (t,
2H, COOCH2), 5.80 (s, 2H, CH2Ph), 6.80–7.90 (m,
10H, aromatic protons). IR vmax (KBr, cm−1) 1720.
Anal. (C22H26N2O2·CH3I) C, H, N, O.
12: Yield 0.6594 g, 33.87%; m.p. 230–231 °C; MS;
1
m/z: 348 [M+], H-NMR (DMSO-d6): l 2.20–2.45 (t,
4H, pyrrolidine), 3.20 (s, 3H, N+ꢀ CH3), 3.45–3.60 (t,
4H, pyrrolidine potons), 4.00 (t, 2H, CH2-pyrrolidine),
4.80 (t, 2H, COOCH2), 5.80 (s, 2H, CH2Ph), 6.90–7.80
(m, 10H, aromatic protons). IR vmax (KBr, cm−1) 1705.
Anal. (C22H24N2O2·CH3I) C, H, N, O.
Figure 8. Indomethacin derivatives (compounds 10-14) studied
for docking mode and atoms responsible for H-bonding with
COX. Note: Details of atoms with superscripts 1 and 2 are
indicated in Table III.
13: Yield 0.5293 g, 26.43%; m.p. 234–236 °C; MS;
1
m/z: 362 [M+], H-NMR (DMSO-d6): l 1.60–2.10 (t,
4H, piperidine protons), 3.20 (s, 3H, N+ꢀCH3), 3.60 (t,
4H, piperidine protons), 4.00–4.12 (t, 4H, piperidine
protons and CH2-piperidine), 4.80 (t, 2H, COOCH2),
5.80 (s, 2H, CH2Ph), 7.05–7.75 (m, 10H, aromatic pro-
tons). IR vmax (KBr, cm−1) 1715. Anal. (C23H26N2O2·
CH3I) C, H, N, O.
14: Yield 0.7752 g, 40.82%; m.p. 212 °C; MS; m/z:
336 [M+], 1H-NMR (DMSO-d6):
l 1.35 (d, 3H,
COOCH(CH3), 3.15 (s, 9H, N(CH3)3), 3.40 (t, 2H,
CH2N(CH3)3), 5.50 (m, 1H, COOCH(CH3), 5.85 (s, 2H,
CH2Ph), 7.05–7.75 (m, 10H, aromatic protons). IR vmax
(KBr, cm−1) 1705. Anal. (C21H24N2O2·CH3I) C, H, N,
O.
Figure 7. Indomethacin derivatives (compounds 5-9) studied
for docking mode and atoms responsible for H-bonding with
COX. Note: Details of atoms with superscripts 1 and 2 are
indicated in Table II.

764
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
4.1.6. 1-Phenyl indole-2-carboxylic acid (15)
Indole-2-carboxylic acid (8.0 g, 0.05 mol) was dis-
solved in 10.0 mL of DMF and then 7.0 g of anhydrous
K2CO3, 0.25 g CuO and 5.2 mL (0.05 mol) bromoben-
zene were added. The reaction mixture was refluxed for
24 h at 154 °C. At the end of reaction, the mixture was
cooled and added into ice-water. The water layer was
washed with CHCl3 (3×100 mL). The water layer was
acidified with concentrated HCl and let to stand
overnight. The precipitated N-phenyl indole-2-car-
Figure 10. Indomethacin derivatives (compounds 23-27 and
29, 30) studied for docking mode and atoms responsible for
H-bonding with COX. Note: Details of atoms with super-
scripts 1 and 2 are indicated in Table III.
boxylic acid was filtered off and purified by crystallisa-
tion from MeOH–H2O (4.95 g, 42.03%): m.p. 176 °C.
4.1.7. 1-Phenyl indole-2-dialkylaminoalkyl carboxylate
hydrogen chloride salts (17–22)
Compounds 17–22 were synthesised as described for
compounds 5–9. Compound 15 (0.4 g, 1.69 mmol) and
0.274 g (1.69 mmol) 1,1%-carbonyldiimidazole were used.
17: Yield 0.3229 g, 55.53%; m.p. 178–180 °C; MS;
m/z: 308 [M+], 1H-NMR (DMSO-d6): l 2.80 (s, 6H,
N(CH3)2), 3.30 (t, 2H, CH2N(CH3)2), 4.50 (t, 2H,
COOCH2), 7.05–7.80 (m, 10H, aromatic protons). IR
vmax (KBr, cm−1) 1725. Anal. (C19H20N2O2·HCl) C, H,
N, O.
Figure 9. Indomethacin derivatives (compounds 17-22) studied
for docking mode and atoms responsible for H-bonding with
COX. Note: Details of atoms with superscripts 1 and 2 are
indicated in Table IV.

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
765
18: Yield 0.2403 g, 38.22%; m.p. 167–168 °C; MS;
m/z: 336 [M+], 1H-NMR (DMSO-d6): l 1.45 (t, 6H,
N(CH2CH3)2), 3.20 (q, 4H, N(CH2CH3)2), 3.35 (t, 2H,
CH2N(CH2CH3)2), 4.75 (t, 2H, COOCH2), 7.05–7.80
(m, 10H, aromatic protons). IR vmax (KBr, cm−1) 1715.
Anal. (C21H24N2O2·HCl) C, H, N, O.
19: Yield 0.2351 g, 37.59%; m.p. 175 °C; MS; m/z:
334 [M+], 1H-NMR (DMSO-d6): l 2.10–2.20 (t, 4H,
pyrrolidine protons), 2.80 and 3.40 (t, 4H, pyrrolidine
protons), 3.85 (t, 2H, CH2-pyrrolidine), 4.75 (t, 2H,
COOCH2), 6.90–7.80 (m, 10H, aromatic protons), 12.70
(s, 1H, NH). IR vmax (KBr, cm−1
(C21H22N2O2·HCl) C, H, N, O.
) 1710. Anal.
20: Yield 0.2664 g, 41.05%; m.p. 182–184 °C; MS;
1
m/z: 348 [M+], H-NMR (DMSO-d6): l 1.20–3.63 (m,
10H, piperidine protons), 3.45 (t, 2H, CH2-piperidine),
4.75 (t, 2H, COOCH2), 7.00–7.80 (m, 10H, aromatic
protons), 12.60 (s, 1H, NH). IR vmax (KBr, cm−1) 1710.
Anal. (C22H24N2O2·HCl·0.1H2O) C, H, N, O.
21: Yield 0.2155 g, 35.61%; m.p. 185–187 °C; MS;
m/z: 322 [M+], 1H-NMR (DMSO-d6): l 1.20 (d, 3H,
COOCH(CH3), 2.75 (s, 6H, N(CH3)2), 3.35 (t, 2H,
CH2N(CH3)2), 5.30 (m, 1H, COOCH(CH3), 7.05–7.80
Figure 12. Amino acid residues involved in hydrogen bonding:
superscripts indicate the atom responsible for hydrogen bond-
ing with ligand molecules.
(m, 10H, aromatic protons), 10.40 (s, 1H, NH). IR vmax
(KBr, cm−1) 1710. Anal. (C20H22N2O2·HCl) C, H, N, O.
22: Yield 0.2802 g, 39.35%; m.p. 171–172 °C; MS;
m/z: 349 [M+], 1H-NMR (DMSO-d6): l 2.10 (t, 4H,
piperazine potons), 3.00–3.30 (t, 4H, piperazine po-
Figure 11. Indomethacin derivatives (compounds 31-34) stud-
ied for docking mode and atoms responsible for H-bonding
with COX. Note: Details of atoms with superscripts 1 and 2
are indicated in Table V.

766
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
24: Yield 0.374 g, 46.37%; m.p. 197–198 °C; MS; m/z:
tons), 3.80 (t, 4H, CH2-piperazine), 4.65 (t, 2H,
COOCH2), 7.05–7.85 (m, 10H, aromatic protons). IR
vmax (KBr, cm−1) 1710. Anal. (C21H23 N3O2·2HCl) C, H,
N, O.
336 [M+], 1H-NMR (DMSO-d6):
l 1.20 (t, 6H,
N(CH2CH3)2), 3.00 (s, 3H, N+ꢀCH3), 3.35 (q, 4H,
N(CH2CH3)2), 3.65 (t, 2H, CH2N(CH2CH3)2), 4.60 (t,
2H, COOCH2), 7.15–7.85 (m, 10H, aromatic protons).
IR vmax (KBr, cm−1) 1725. Anal. (C21H24N2O2·CH3I) C,
H, N, O.
4.1.8. 1-Benzyl indole-2-dialkylaminoalkyl carboxylate
methyl iodide salts (23–27)
Compounds 23–27 were synthesised as described for
compounds 10–14.
25: Yield 0.314 g, 39.10%; m.p. 202–203 °C; MS; m/z:
334 [M+], 1H-NMR (DMSO-d6): l 2.20 (m, 4H,
pyrrolidine protons), 3.20 (s, 3H, N+ꢀCH3), 3.65 (t, 4H,
pyrrolidine protons), 3.80 (t, 2H, CH2-pyrrolidine), 4.70
(t, 2H, COOCH2), 7.05–7.85 (m, 10H, aromatic pro-
tons). IR vmax (KBr, cm−1) 1720. Anal. (C21H22N2O2·
CH3I) C, H, N, O.
23: Yield 0.2688 g, 35.40%; m.p. 208 °C; MS; m/z:
308 [M+], 1H-NMR (DMSO-d6):
l 3.10 (s, 9H,
N(CH3)3), 3.70 (t, 2H, CH2N(CH3)2), 4.60 (t, 2H,
COOCH2), 7.05–7.80 (m, 10H, aromatic protons). IR
vmax (KBr, cm−1) 1715. Anal. (C19H20N2O2·CH3I) C, H,
N, O.
26: Yield 0.3142 g, 38.00%; m.p. 213–215 °C; MS;
1
m/z: 348 [M+], H-NMR (DMSO-d6): l 1.65–1.78 (m,
Figure 13.

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
767
Figure 14.
6H, piperidine protons), 1.85–1.98 (m, 4H, piperidine
protons), 3.25 (s, 3H, N+ꢀCH3), 3.40 (t, 2H,
CH2ꢀpiperidine), 4.65 (t, 2H, COOCH2), 7.00–7.80 (m,
10H, aromatic protons). IR vmax (KBr, cm−1) 1720.
Anal. (C22H24N2O2·CH3I) C, H, N, O.
27: Yield 0.3699 g, 47.25%; m.p. 223–225 °C; MS;
m/z: 322 [M+], 1H-NMR (DMSO-d6): l 1.30 (d, 3H,
COOCH(CH3), 3.15 (s, 9H, N(CH3)3), 3.50 (t, 2H,
CH2N(CH3)3), 5.40 (m, 1H, COOCH(CH3), 7.05–7.80
(m, 10H, aromatic protons). IR vmax (KBr, cm−1) 1715.
Anal. (C20H23N2O2·CH3I) C, H, N, O.
4.1.9. NꢀH Indole-2-dialkylaminoalkyl carboxylate
hydrogen chloride salts (29–33)
Compound 29 was synthesised as described for com-
pounds 5–9. Indole-2-carboxylic acid (1.0 g, 6.2 mmol)
and 1.1 g (6.2 mmol) 1,1’-carbonyldiimidazole were
used.
29: Yield 1.1631 g, 56.25%; m.p. 178 °C; MS; m/z:
260 [M+], 1H-NMR (DMSO-d6):
l 1.35 (t, 6H,
N(CH2CH3)2), 3.20 (q, 4H, CH2N(CH2CH3)2), 3.35 (t,
2H, CH2N(CH2CH3)2), 4.55 (t, 2H, COOCH2), 7.05–
7.70 (m, 5H, aromatic protons), 12.00 (s, 1H, NH). IR

768
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
1H-NMR (DMSO-d6): l 1.30 (d, 3H, COOCH(CH3),
vmax (KBr, cm−1
) 1720. Anal. (C15H20N2O2·2HCl·
2.80 (s, 6H, N(CH3)2), 3.40 (t, 2H, CH2N(CH3)2), 5.40
(m, 1H, COOCH(CH3), 7.05–7.80 (m, 5H, aromatic
protons), 9.12 (s, 1H, NH). IR vmax (KBr, cm−1) 1715.
Anal. (C14H18N2O2·2HCl·0.2H2O) C, H, N, O.
0.1H2O) C, H, N, O.
31: Yield 1.3781 g, 67.05%; MS; m/z: 258 [M+],
1H-NMR (DMSO-d6): l 0.70–5.10 (m, 10H, pyrrolidine
protons and CH2-pyrrolidine), 4.65 (t, 2H, COOCH2),
6.90–7.80 (m, 5H, aromatic protons), 12.45 (s, 1H,
NH). IR vmax (KBr, cm−1) 1720. Anal. (C15H18N2O2·
2HCl) C, H, N, O.
34: Yield 1.4925 g, 62.85%; MS; m/z: 273 [M+],
1H-NMR (DMSO-d6): l 2.10 (t, 4H, piperazine pro-
tons), 3.00–3.30 (t, 4H, piperazine protons), 3.80 (t, 2H,
CH2-piperazine), 4.65 (t, 2H, COOCH2), 7.05–7.85 (m,
5H, aromatic protons). IR vmax (KBr, cm−1) 1710. Anal.
(C15H19N3O2·3HCl) C, H, N, O.
32: Yield 1.4077 g, 65.71%; MS; m/z: 272 [M+],
1H-NMR (DMSO-d6): l 0.70–3.60 (m, 12H, piperidine
protons and CH2-piperidine), 4.90 (t, 2H, COOCH2),
7.05–7.75 (m, 5H, aromatic protons), 12.20 (s, 1H,
NH). IR vmax (KBr, cm−1) 1715. Anal. (C16H20N2O2·
2HCl) C, H, N, O.
4.1.10. NꢀH Indole-2-diethylaminoethyl carboxylate
methyl iodide salt (30)
Compound 30 was synthesised as described for com-
pounds 10–14.
33: Yield 1.2890 g, 65.07%; MS; m/z: 246 [M+],
Figure 15.

S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
769
Koboldt C.M., Seibert K., A single amino acid difference
between cycloxygenase-1 (COX-1) and -2 (COX-2) reverses the
selectivity of COX-2 specific inhibitors, J. Biol. Chem. 271
(1996) 15810–15814.
Table VI. Structural formulas of compounds 5–34.
[5] Prasit P., Riendeau D., Selective cylooxygenase-2 inhibitors,
Annu. Rep. Med. Chem. 32 (1997) 211–220.
[6] Bertenshaw R., Talley J.J., Rogier D.J., Graneto M.J., Koboldt
C.M., Zang Y., Conformationally restricted 1,5-diarylpyrazoles
are selective COX-2 inhibitors, Bioorg. Med. Chem. Lett. 6
(1996) 2827–2830.
Compound
IC50 (mM)
[7] Li J.J., Norton M.B., Reinhard E.J., Anderson G.D., Gregory
S.A., Isakson P.C., Koboldt C.M., Masferrer J.L., Perkins
W.E., Seibert K., Zang Y., Zweifel B.S., Reitz D.B., Novel
terphenyls as selective cyclooxygenase-2 inhibitors and orally
active anti-inflammatory agents, J. Med. Chem. 39 (1996)
1846–1856.
COX-1
COX-2
NS-398
SC-299
Indomethacin
22
30
34
0.12
0.060
0.75
50.0
50.0
50.0
\66
\66
0.05
ND
ND
ND
[8] Penning T.D., Talley J.J., Bertenshaw S.R., Carter J.S., Collins
P.W., Docter S., Graneto M.J., Lee L.F., Malecha J.M.,
Miyashiro J.M., Rogier D.J., Rogers R.S., Yu S.S., Anderson
G.D., Burton E.G., Cogburn J.N., Gregory S.A., Koboldt
C.M., Perkins W.E., Seibert K., Veenhuizen A.W., Zang Y.Y.,
Isakson P.C., Synthesis and biological evaluation of 1,5-di-
arylpyrazole class of cyclooxygenase-2 inhibitors: identification
of 4-(5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazole-1-yl-
(benzenesulfonamide (SC-58635, Celecoxib), J. Med. Chem. 40
(1997) 1347–1365.
ND=Not determined.
30: Yield 1.1544 g, 46.25%; m.p. 204–205 °C; MS;
m/z: 260 [M+], 1H-NMR (DMSO-d6): l 1.10 (t, 6H,
N(CH2CH3)2), 3.14 (s, 3H, N+ꢀCH3), 3.50 (q, 4H,
CH2N(CH2CH3)2), 3.75 (t, 2H, CH2N(CH2CH3)2), 4.70
(t, 2H, COOCH2), 7.05–7.30 (m, 5H, aromatic protons).
IR vmax (KBr, cm−1) 1710. Anal. (C15H20N2O2·CH3I) C,
H, N, O.
[9] Black W.C., Bayly C., Belley M., Chan C.-C., Charleson S.,
Denis D., Gauthier J.Y., Gordon R., Guay D., Kargman S.,
Lau C.K., Leblanc Y., Mancini J., Ouellet M., Percival D., Roy
P., Skorey K., Tagari P., Vickers P., Wong E., Xu L., Prasit P.,
From indomethacin to a selective COX-2 inhibitor: develop-
ment of indolealkanoic acids as potent and selective cyclooxy-
genase-2 inhibitors, Bioorg. Med. Chem. Lett.
725–730.
6 (1996)
4.2. Enzyme assays
[10] Kalgutkar A.S., Marnett A.B., Crews B.C., Remmel R.P.,
Marnett L.J., Ester and amide derivatives of the nonsteroidal
antiinflammatory drug, indomethacin, as selective cyclooxyge-
nase inhibitors, J. Med. Chem. 43 (2000) 2860–2870.
For enzymologic method and chemicals see Ref. [23].
[11] Olson A.J., Goodsell D.S., Automated docking and the search
for HIV protease inhibitors, SAR QSAR Environ. Res. 8
(1998) 273–285.
Acknowledgements
[12] Kuntz I.D., Blaney J.M., Oatley S.J., Langridge R., Ferrin
T.E., A geometric approach to macromolecule–ligand interac-
tions, J. Mol. Biol. 161 (1982) 269–288.
This work was partially supported by a grant from
Turkish Scientific and Technical Research Institute
(SBAG-AYD-108). This work was also supported in
part by the High-Technology Research Grant and grant-
in-aid for International Scientific Research both flfrom
the Ministry of Education in Japan. We thank Ms Akiko
Miura for her help in molecular docking studies. The
authors thank Dr Larry Marnett for biological assays.
[13] Akaho E., Fujikawa C., Runion H., Hill C., Nakaho H., A
study on binding modes of nonsteroidal anti-inflammatory
drugs to COX-1 and COX-2 as obtained by DOCK 4.0, J. Chem.
Software 4 (1999) 147–162.
[14] Sundberg R.J., Russell H.F., Syntheses with N-protected 2-
thioindoles, J. Org. Chem. 38 (1973) 3324–3330.
[15] Uhle F.C., Harris L.S., The synthesis and cyclization of (-
methylamino-(-(4-carboxy-3-indole)-propionic acid, J. Am.
Chem. Soc. 79 (1957) 102.
[16] Hutchison A.J., Kishi Y., Stereospecific total synthesis of DL
austamide, J. Am. Chem. Soc. 101 (1979) 6786–6788.
-
References
[17] Murakami Y., Watanabe T., Kobayashi A., Yokoyama Y., A
novel method for the debenzylation of protected 9-indole nitro-
gen, Synthesis (1984) 738–740.
[1] Marnett L.J., Kalgutkar A.M., Design of selective inhibitors of
cyclooxygenase-2 as nonulcerogenic anti-inflammatory agents,
Curr. Opin. Chem. Biol. 2 (1998) 482–490.
[18] (a) Sumpter W.C., Miller F.M., Heterocyclic Compounds with
Indole and Carbazole Systems, The Chemistry of Heterocyclic
Compounds, Interscience, New York, 1954;
[2] Needleman P., Isakson P.C., The discovery and function of
COX-2, J. Rheumatol. Suppl. 49 (1997) 6–8.
[3] Jackson L.M., Hawkey C.J., Cox-2 selective nonsteroidal anti-
(b) Houlian W.J., Indols, Parts I and II, The Chemistry of
Heterocyclic Compounds, Wiley–Interscience, New York,
1972.
inflammatory drugs, Drugs 59 (2000) 1207–1216.
[4] Gierse J.K., McDonald J.J., Hauser S.C., Rangwala S.H.,

770
S. Olgen et al. / European Journal of Medicinal Chemistry 36 (2001) 747–770
[19] Khan M.A., Polya J.B., Syntheses of heterocyclic compounds.
2. N-arylazoles by Ullmann condensation, J. Chem. Soc. 1
(1970) 85.
[22] Venuti M.C., Young J.M., Maloney P.J., Johnson D., Mc-
Greevy K., Synthesis and biological evaluation of (-(N,N,N-tri-
alkylammonium)alkyl esters and thioesters of carboxylic acid
nonsteroidal antiinflammatory agents, Pharm. Res. 6 (1989)
867–873.
[20] Unangst P.C., Conner D.T., Stabler S.R., Weikert R.J., Synthe-
sis of novel 1-phenyl-1H-indole-2-carboxylic acids. I. Utiliza-
tion of Ullman and Dieckmann reactions for the preparation of
3-hydroxy, 3-alkoxy, and 3-alkyl derivatives, Rec. Chem. Prog.
24 (1987) 811–815.
[23] Kalgutkar A.S., Marnett A.B., Crewws B.C., Remmel R.P.,
Marnett L.J., Ester and amide derivatives of the nonstero-
idal antiinflammatory drug, indomethacin, as selective
cyclooxygenase inhibitors, J. Med. Chem. 43 (2000) 2860–
2870.
[21] Khan M.A., Rocha E.K., Arylindoles.1. Synthesis of some
N-arylindoles, Chem. Pharm. Bull. 25 (1977) 3110–3114.