Synthesis, COX-1/2 inhibition activities and molecular docking study of isothiazolopyridine derivatives

Article abstract of DOI:10.1016/j.bmc.2016.10.036

One of the main challenges for nowadays medicine is drugs selectivity. In COX-1 and COX-2, the active sites are composed of the same group of amino acids with the exception of the only one residue in position 523, in COX-1 is an isoleucine, while in COX-2

Full text of DOI:10.1016/j.bmc.2016.10.036

Accepted Manuscript
Synthesis, COX-1/2 inhibition activities and molecular docking study of iso-
thiazolopyridine derivatives
Piotr Świątek, Malgorzata Strzelecka, Rafal Urniaż, Katarzyna Gębczak,
Tomasz Gębarowski, Kazimierz Gąsiorowski, Wieslaw Malinka
PII:
S0968-0896(16)31092-6
DOI:
Reference:
http://dx.doi.org/10.1016/j.bmc.2016.10.036
BMC 13363
Bioorganic & Medicinal Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
29 July 2016
25 October 2016
29 October 2016
Please cite this article as: Świątek, P., Strzelecka, M., Urniaż, R., Gębczak, K., Gębarowski, T., Gąsiorowski, K.,
Malinka, W., Synthesis, COX-1/2 inhibition activities and molecular docking study of isothiazolopyridine
derivatives, Bioorganic & Medicinal Chemistry (2016), doi: http://dx.doi.org/10.1016/j.bmc.2016.10.036
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

SYNTHESIS, COX-1/2 INHIBITION ACTIVITIES AND MOLECULAR DOCKING
STUDY OF ISOTHIAZOLOPYRIDINE DERIVATIVES
Piotr Świątek a *, Małgorzata Strzelecka a, Rafał Urniaż b, Katarzyna Gębczak c, Tomasz
Gębarowski c, Kazimierz Gąsiorowski c, Wiesław Malinka a
a Department of Chemistry of Drugs, Wrocław Medical University, Borowska 211,
50-556 Wrocław, Poland
b Department of Medicine, Institute for Medical Research, University of Cambridge,
Cambridge Biomedical Campus, Addenbrooke's Hospital, Hills Rd, CB2 0SP, Cambridge,
UK
c Department of Basic Medical Science, Wrocław Medical University Borowska 211,
50-556 Wrocław, Poland
ABSTRACT
One of the main challenges for nowadays medicine is drugs selectivity. In COX-1 and
COX-2, the active sites are composed of the same group of amino acids with the exception of
the only one residue in position 523, in COX-1 is an isoleucine, while in COX-2 is a valine.
Here, we presented a series of isothiazolopyridine/benzisothiazole derivatives substituted
differently into an isothiazole ring, which were synthesized and investigated for their
potencies to inhibit COX-1 and COX-2 enzymes by colorimetric inhibitor screening assay.
All the tested compounds inhibited the activity of COX-1, the effect on COX-2 activity was
differential. The mode of binding was characterized by a molecular docking study.
Comparing biological activity of the investigated compounds, it was observed that
compounds sharing the most similar position to flurbiprofen and meloxicam, representing the
two main enzyme subdomains, achieved higher biological activity than others. It is directly
related to the fit to the enzyme’s active site, which prevents too early dissociation of the
compounds.
1. INTRODUCTION
Cyclooxygenase (COX) converts arachidonic acid to prostanoids, which include
prostaglandins (PGs), prostacyclin, and thromboxane.1 Prostanoids are important mediators
that regulate diverse functions in the cardiovascular, gastrointestinal, urogenital, and nervous

systems, as well as playing crucial roles in immunity and inflammation.1 There are several
cyclooxygenase isoforms: COX-1, a constitutive form expressed in almost all tissues; COX-2,
which is predominantly induced and constitutively expressed in a number of tissues (renal
medulla, prostate, brain, and endothelium);1 and COX-3, a COX-1-derived protein,
abundantly found in the cerebral cortex and heart.2 COX-1 is a constitutive, housekeeping
enzyme that is ubiquitously expressed and is responsible for prostanoid production in most
tissues.1 In contrast, COX-2 is an inducible enzyme, expressed at sites of inflammation,
infection, and cancer, that generates prostanoids important for driving disease pathogenesis.1,3
However, COX-2 is also constitutively expressed in kidney, gastrointestinal tract, brain, and
thymus, without any signs of inflammation.4 Constitutive COX-2 expression plays a major
role in homeostatic function, in development and maintenance of physiological function of
those organs.1
COX-1 and COX-2 are the therapeutic targets for anti-inflammatory drugs, including
ibuprofen, naproxen, diclofenac and piroxicam, as well as newer COX-2 selective inhibitors
(coxibs); collectively these drugs are known as nonsteroidal anti-inflammatory drugs
(NSAIDs). NSAIDs are among the most commonly taken as the first-line anti-inflammatory
therapy.
It is well known that benzothiazole/benzisothiazole, benzoxazole and pyridoisothiazole
derivatives possess interesting anti-inflammatory and analgesic activity.5-12 Compound A, for
example, shown in Figure 1 exhibited higher antinociceptive activity than aspirin and
moderate anti-inflammatory activity in comparison with indomethacin.7 Moreover, the
benzothiazole derivatives B were reported to interact with COX-2 enzyme more efficiently
than rofecoxib, with anti-inflammatory activity as well.8 Additionally, compound C revealed
a better inhibitory profile of induction of edema in the carrageenan-induced paw edema than
indomethacin.9 Furthermore, 4-arylpiperazine derivatives of pyridoisothiazole showed high
antinociceptive activity in the writing test, in particular, compounds 22, 23, 24 and 25, which
were able to reduce the number of abdominal constrictions by more than 50% at the doses of
1.8-7.8 mg/kg.11,12
OH
H
S
N
Cl
N
N
O
N
O
O
R=H, -NO2
R1=H,-C6H5
O
R
N
R1
A
B

Cl
CH3
N
O
Am
N
(CH2)n
S
S
H3C
N
Cl
O
N
H
N
N
N
22 n=3, Am=
23 n=1, Am=
N
N
F
S
C
N
N
N
24 n=1, Am=
25 n=1, Am=
N
H
CH3
Figure 1. Structures of reference compounds
Considering the above, we designed a group of amide and acid derivatives of isothiazolo[4,5-
b]pyridine, which can be considered as 7-aza analogues of the benzisothiazole system. The
presence of a strong basic and electron-deficient pyridine ring in the isothiazolopyridine
moiety makes these systems different from the better known benzoxazole and benzothiazole
rings. This study was focused on exploration of the central alkanyl chain length and
branching. In order to verify the pharmacological significance of the size and aromaticity of
the isothiazolo[4,5-b]pyridine moiety, a benzisothiazole derivative was synthesized.
The compounds were investigated for their potencies to inhibit COX-1 and COX-2 enzymes
by colorimetric inhibitor screening assay, and the mode of binding was characterized by a
molecular docking study
2. RESULTS AND DISCUSSION
2.1 CHEMISTRY
All the new compounds were prepared according to the synthetic pathways outlined in
Scheme 1. 4,6-Dimethylisothiazolo[5,4-b]pyridin-3(2H)-one 113 or commercially available
1,2-benzisothiazol-3(2H)-one 2 were used as a starting materials. Compounds 21-25 were
obtained earlier.11,12,14
Preparation of the propanamide derivatives (8-11) involves Michael-type addition of
isothiazolopyridine 1 or benzisothiazole 2 to corresponding acrylamide derivatives (3a-c).
Compounds 3a-c were synthesized by N-acylation of corresponding secondary amines (N-
arylpiperazines or 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline) with 3-bromopropionyl
chloride to give a mixture of acrylamides 3a-c and 3-bromopropanamides 26a-c which were
separated by CC (Scheme 2). 3-Bromopropanamide derivatives proved to be unstable under

the reaction conditions, and heating in acetonitrile in the presence of potassium carbonate led
to dehydrohalogenation of bromopropanamides to give α,β-unsaturated amides (3a-c).
Synthesis of the acetamide derivatives (12 and 13) was started by obtaining 1-
chloroacetyl-4-(3-chlorophenyl)piperazine (4) and 1-(2-bromobutanoyl)-4-phenylpiperazine
(5), respectively. N-acylation of N-arylpiperazines with corresponding chloroacetyl chloride
or 2-bromobutyryl bromide led to 4 or 5 without any by-product. Then, the treatment of
isothiazolopyridine 1 with intermediates 4 or 5 in acetonitrile in the presence of potassium
carbonate provided target compounds (12 and 13).
Hydrolysis of esters 7a, 6b and 7b in the presence of an acid catalyst afforded the
corresponding acid derivatives (14, 15 and 16). The experimental details for the preparation
of esters 6b and 7b were recently described.15 Ethyl 2-[(4,6-dimethylisothiazolo[5,4-
b]pyridin-3-yl)oxy]propanoate (7a) was obtained by reaction of isothiazolopyridine 1 with
ethyl 2-bromopropanoate to give a mixture of 2-N- (6a) and 3-O-alkylated (7a) isomers
which were separated by CC. The yield of 2-N-alkylated isomer was low, and this compound
is not described here.
The isomeric butanamides 17-20 were prepared after treatment of 2-N- or 3-O-
butanoic acid of isothiazolopyridine (15 or 16) with ethyl chloroformate in the presence of
trimethylamine, and in the subsequent step, the appropriate amine derivative [1-(3-
chlorophenyl)piperazine or 1-phenylpiperazine] was added.
Final compounds were characterized by a sharp melting point (m.p.), correct elemental (C, H,
N) analyses, IR and 1H NMR spectra. The structures of the isomeric isothiazolopyridines (14-
1
20) were assigned on the basis of H NMR. In the 2-N-substituted isomer 15, 17 and 19 the
signal of the methylene protons adjacent to the 2-N-nitrogen of the isothiazolopyridine 1 was
recorded within 3.81-3.97 ppm. The 3-O-substitution (14, 16, 18, 20) produced downfield
shift of these protons in the range 4.46-4.59 ppm. The above spectral data agree with those
previously reported for related isomeric 2-N- and 3-O-substituted isothiazolopyridines.11

O
R2
X
Am
R2
X
N
N
CH3
N
O
O
R
O
O
Z
Cl/Br
z
O
N
N
4, 5
b / c
3a-c
O
N
R
NH
N
S
a
R1
S
S
R1
H3C
Am
1 X=N, R1=R2=-CH3
2 X=C, R1=R2=H
12 Z=-CH2-, R=3-Cl
Cl
8 X=N, R1=R2=-CH3, Am=
9 X=N, R1=R2=-CH3, Am=
10 X=N, R1=R2=-CH3, Am=
11 X=C, R1=R2=H, Am=
13 Z=-CH(C2H5)-, R=H
N
N
N
N
O
Cl/Br
d / e
z
O
O
O
N
Cl
O
O
CH3
N
N
CH3
CH3
O
O
N
O
Z
CH3
O
OH
f
Z
N
N
+
CH3
S
O
S
S
H3C
N
H3C
N
H3C
N
CH3
O
14
7a Z=-CH(CH3)-
7b Z=-(CH2)3-
6a Z=-CH(CH3)-
6b Z=-(CH2)3-
f
f
CH3
O
O
CH3
O
OH
C
N
N
S
H3C
N
S
N
H3C
OH
15
16
O
g
h
g
h
CH3
N
O
O
CH3
N
N
O
N
S
H3C
N
N
N
N
S
R
H3C
R
O
17 R=3-Cl
19 R=H
18 R=3-Cl
20 R=H
Scheme 1. Reagent and conditions: (a) toluene, reflux; (b) EtONa, anh. ethanol, reflux / (c)
K2CO3, acetonitrile, reflux; (d) K2CO3, acetonitrile, reflux / (e) EtONa, DMF, reflux; (f) conc.
HCl, reflux; (g) Et3N, ethyl chloroformate, dichloromethane, 0°C; (h) 1-(3-
chlorophenyl)piperazine/1-phenylpiperazine, room temp.
H-Am
O
N(C2H5)3
Br
Cl
Diethyl ether
Am:
Cl
Am
N
N
N
Br
Am
+
O
N
O
26a-c
3a-c
O
O
K2CO3
N
acetonitrile
Scheme 2. Synthesis of compounds 3a-c and 26a-c.

2.2 IN VITRO CYCLOOXYGENASE (COX) INHIBITION ASSAY
Figure 2 shows the impact of piroxicam, the standard inhibitor of cyclooxygenases, on COX-
1 and COX-2 activities at 2 min of incubation, as recommended for this test kit.
COX-1
COX-2
1.25
1.25
control
*
1.00
0.75
0.50
0.25
0.00
control
1.00
0.75
0.50
0.25
0.00
*
**
*
*
**
**
**
*
**
**
**
**
***
***
***
1
2
3
4
5
6
7
8
9 10
1
2
3
4
5
6
7 8 9 10
time [min]
time [min]
Figure 2. Impact of the model inhibitor piroxicam on COX-1 and COX-2 activity during the
incubation period from 1 min to 10 min. Results obtained in samples with the tested drug
were compared to those in the control cultures (the enzyme incubated without piroxicam; E0),
and the ratios of E/E0 are given in Figure 2. Statistical significance was calculated with the t
test (*p<0.05; **p0.01; ***p<0.001).
As can be seen in Figure 2, COX-1 activity markedly decreased within the first 6 min
of incubation with piroxicam, even by 60-65% at 1 and 2 min. However, in the longer period,
i.e. at 7-10 min, the inhibitory effect of the drug became weak and activity of the enzyme was
even higher (although not statistically significant) than that in the control samples (the
enzyme incubated without the tested drug). On the other hand, the decrease of COX-2 activity
by piroxicam was significant within the whole period of incubation; when compared to
relative control samples the (without the drug) the activity of COX-2 was decreased by 57%
at 2 min and by 40% at 10 min.
The influence of the tested compounds on COX-1 and COX-2 activities at 2 min of
incubation (fast impact) and at 10 min (persisting effect) was compared to the relative control
samples (the enzyme activity without tested compounds) and is shown in Figures 3 and 4 as
E/E0 ratios. Statistical significance of the results was estimated with the paired t test.
Figure 3 presents the results of the COX-1 (A) and the COX-2 (B) activities estimated
at 2 min of incubation with the tested compounds.

A
B
COX-2 activity at 2 min of incubation
1.50
**
1.25
1.00
0.75
0.50
0.25
0.00
**
control
*
*
*
*
*
*
*
*
**
**
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Px
*
tested compounds
Figure 3. Activity of the COX-1 (A) and COX-2 (B) enzymes at 2 min of incubation with the
tested compounds. Results were compared to relative control samples (without the tested
compounds; E0) and expressed as E/E0 ratios. Statistical significance was calculated with the t
test (*p<0.05; **p<0.01; ***p<0.001). The effect of the model cyclooxygenase inhibitor
piroxicam (Px) was also estimated and is shown in the figure.
All the tested compounds inhibited the activity of COX-1 at 2 min after their addition,
and the effect was the strongest in the cases of compounds 9, 13-16, 21 and 24 (inhibition of
the enzyme by 25-40%).
Figure 3B reveals diverse effects of the tested compounds on COX-2. The enzyme’s
activity was strongly inhibited by compounds 11, 13-15, 21, 23 (by 20-25%), and, in contrast,
was elevated in samples containing compounds 17 and 20 (by 7% and by 19%, respectively,
when compared to the control samples).
The influence of the tested compounds on activity of both cyclooxygenases as
estimated at 10 min of incubation is shown in Figure 4.

A
B
COX-1 activity at 10 min of incubation
COX-2 activity at 10 min of incubation
***
1.50
1.25
1.00
0.75
0.50
0.25
0.00
1.50
1.25
1.00
0.75
0.50
0.25
0.00
**
***
*
*
*
*
*
*
*
*
control
*
control
*
*
*
*
*
*
**
*
**
**
*
*
**
**
***
**
**
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Px
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Px
tested compounds
tested compounds
Figure 4. Activity of the COX-1 (A) and COX-2 (B) enzymes at 10 min of incubation with
the tested compounds. Results were compared to relative control samples (without tested
compounds; E0) and expressed as E/E0 ratios. Statistical significance was calculated with the t
test (*p<0.05; **p<0.01; ***p<0.001). The effect of the model cyclooxygenase inhibitor
piroxicam (Px) was also estimated and is shown in the figure.
As shown in Figure 4A, the activity of COX-1 was strongly inhibited during
incubation with compounds 9, 13, 24 and 25 (by 18-30%). On the other hand, compounds 20
and 23 caused higher activity of the enzyme (3-7% higher than in the control). It should be
noted that piroxicam enhanced COX-1 activity by 11% during 10 min of incubation.
Figure 4B demonstrates the diverse impact of the tested compounds on COX-2 activity
at 10 min of incubation; both the inhibitory effect of compounds 15, 16, 21 and 25 (12-21%)
and the stimulatory result of the enzyme’s activity by compounds 17, 18, 20 and 22 (15-25%)
were estimated. The inhibitory effect was the strongest in the case of compound 11 (38%).
For comparison, the inhibitory effect of piroxicam was 40%.
Two-way analysis of variance confirmed that the tested compounds differed in their
impact on cyclooxygenase activity (F=19.78, df=35, p<10-4), and the impact on COX-1 and
on COX-2 activity was markedly diverse (F=86.72, df=1, p<10-4).
We calculated IC50 values (i.e. the concentration of tested compounds [µM] which can
exert 50% inhibition of the enzyme activity), separately with COX-1 and COX-2 activity
estimations at 2 min of incubation with the tested compounds. Selectivity of the compounds
to COX-1 or to COX-2 was assessed by calculation of the IC50 ratios. The IC50 values were
not calculated with those tested compounds which exerted very low inhibitory activity at 2
min of incubation. The results of the calculation are given in Table 1.

Table 1. IC50 values (SD) calculated for COX-1 and COX-2 activities at 2 min of incubation
with the tested compounds [mean (SD); n=3].
IC50 [µM] ( SD)
COX-2
Ratio:
Compound
COX-2/COX-1
COX-1
241.9 (10.4)
206.6 (2.2)
497.3 (9.7)
330.6 (5.3)
374.6 (5.0)
153.6 (3.9)
229.1 (3.8)
235.2 (6.1)
248.4 (1.5)
722.3 (7.2)
818.2 (26.5)
349.2 (10.3)
-
388.0 (8.8)
368.2 (4.2)
303.4 (16.0)
129.9 (27.7)
605.1 (22.4)
343.6 (9.4)
326.9 (22.8)
302.3 (4.9)
269.0 (26.5)
-
1.60
1.78
0.61
0.39
1.61
2.24
1.43
1.28
1.08
-
8
9
10
11
12
13
14
15
16
17
-
-
18
459.0 (7.4)
-
1.31
-
19
20
2114
2211
2312
2412
2512
Piroxicam
210.5 (3.0)
815.6 (39.1)
-
222.6 (24.3)
-
1.06
-
358.8 (7.8)
-
-
156.4 (8.4)
218.9 (2.5)
254.6 (6.5)
-
494.3 (14.9)
102.8 (2.2)
2.26
0.40
Data on the ratios of COX-2 to COX-1 activity included in Table 1 clearly indicate
those compounds which exerted significant COX-1 selectivity (13 and 25) and COX-2
preferences (10 and 11). However, compounds 16 and 21 exhibited almost equally strong
inhibitory action on COX-1 and COX-2 activities.
These compounds should be further studied in terms of their selectivity and balanced
inhibitory action on cyclooxygenases as candidates for new anti-inflammatory drugs.

2.3 STRUCTURE-BASED STUDIES
2.3.1 MOLECULAR MODELING STUDIES
In the present study 18 new inhibitors were tested. Every compound was synthesized
and biological activity was determined as described in the materials and methods section.
Additionally, a molecular docking study was calculated to evaluate the mode of binding for
the compounds. The docking results and biological activity expressed as pIC50 are presented
in Table 2. The full list of results is available in Table S1 in the supporting materials.
Currently, almost 70 crystal structures of cyclooxygenases are deposited in the Protein
Data Bank. For most of them the high-resolution crystal structure co-crystallized with
ligand(s) is available. It allows extensive investigation of the binding mode for potential
inhibitors.
Figure 5. Illustration of four characteristic subdomains of the COX enzyme ligand binding
domain, indicated by red boxes A, B, C and D.
Here, the authors specified in the COX ligand binding domain four characteristic
subdomains of the enzyme (Figure 5, red boxes A, B, C and D). The subdomains represent
pockets where the native substrate (arachidonic acid) as well as the enzyme inhibitors may
potentially bind. Two of the inhibitors (flurbiprofen and meloxicam) representing high
biological activity (pIC50 = 5.7 – 8 for flurbiprofen16 and COX-1 pIC50 = 6; COX-2 pIC50 =
6.82 for meloxicam17) were shown to characterize the subdomains. Subdomain A represents
the mode of binding of flurbiprofen, subdomain B represents the mode of binding of
meloxicam, subdomain C represents an entrance region of the enzyme binding domain, and
subdomain D represents the position of the residue in position 523: the isoleucine in COX-1
and the valine in COX-2. Position 523 is the main cluster of differentiation between COX-1

and COX-2 active sites, being the main target for selective compounds. The smaller Val523
residue in COX-2 allows access to a hydrophobic side-pocket in the enzyme which Ile523
sterically hinders.18 In terms of their molecular biology, COX-1 and COX-2 are of similar
molecular weight, approximately 70 and 72 kDa, respectively. The enzymes have near-
identical catalytic sites although their amino acid sequence homology is only 65%, which
implies a slightly different mode of binding for the same compounds under COX-1 and COX-
2 binging conditions, as described in the example of meloxicam elsewhere.18-20 To correlate
mode of binding with biological activity, the compounds were docked to the appropriate
crystal structures and evaluated by scoring function.
Table 2. Selected docking results sorted by biological activity for COX-1 and COX-2.
COX-1
COX-2
MolDock
Score
MolDock
Score
Compound
pIC50
Compound pIC50
3.86
3.82
3.76
3.52
-94.8466
-91.3921
-81.7048
-106.809
3.89
3.69
3.6
-248.91
-167.797
-206.268
-215.663
13
11
2114
2114
2512
11
13
2512
2.28
2.3.2 GENERAL MODE OF BINDING
As reported previously, induced fit and dissociation times are the most important
components reflecting biological activity in the case of enzyme inhibition.17,24 In general, this
observation explains the behavior of the presented compounds and their biological activity.
Due to high homology between COX-1 and COX-2 binding domains for almost all of the
compounds, observed interactions with both binding domains are quite similar, which is also
reflected when average activities COX-1 pIC50 = 3.58 and COX-2 pIC50 = 3.37 are compared.
In general, the bulky compounds (8-13, 17-20, 22-25) bind to the enzyme by the
phenylpiperazine rings to internal part of the COX-1 and COX-2 enzyme (Figure 6, red box
A) and the isothiazolopyridine/benzisothiazole rings directed to the opposite site (Figure 6,
red box C). This type of binding in the example of compound 13 (Figure 6, blue compound) is
observed for almost all of the bulky compounds, except compound 11 under COX-2 binding
conditions.

Figure 6. Docking poses of 13 (light blue) and 11 (yellow) under COX-1 and COX-2 binding
domain conditions (top panel) and ligand maps representing interactions by hydrogen bonds
(blue dot line) and steric interactions (red dot line) (bottom panel). Residues colored cyan; all
hydrogens were hidden. A, B, C, D – binding subdomains; details in the text.
As presented in Table 2, when compounds are compared by biological activity the
most potent compounds are 13 in the group of COX-1 and 11 in the group of COX-2. As can
be observed, both compounds bind by the phenylpiperazine rings to the internal part of the
COX-1 and COX-2 enzyme (Figure 6, red box A). In both cases, the phenylpiperazine rings
are in a similar position as represented by flurbiprofen (supporting materials, Figure S1, red
compound). The isothiazolopyridine/benzisothiazole rings are directed to the opposite site
(referring to the phenylpiperazine ring) representing an entrance region of the enzyme binding
domain (Figure 6, red box C). Compound 13 occupies almost the same position under COX-1
and COX-2 binding conditions opposite to 11, which is similar to position 13 under COX-1
and quite different in the case of COX-2. It can be observed that the benzisothiazole ring of
11 occupies subdomain B, which represents the meloxicam mode of binding (supporting
materials, Figure S1, green compound). An explanation of this behavior is the difference
between COX-1 and COX-2 binding domains in position 523. The smaller valine residue in
COX-2 (Figure 6, red box D) compared to isoleucine in COX-1 (Figure 6, red box D) allows
access to a hydrophobic side-pocket in the enzyme.18 Due to higher obtained activity of 13
(pIC50 = 3.86) compared to 11 (pIC50 = 3.52) under COX-1 owing to the ethyl group in
position X’ it sterically interacts with the meloxicam-like binding domain (Figure 6, red box
B). Also, the ethyl group is important for 13 under COX-2 (pIC50 = 3.6), although the 11

benzisothiazole ring system goes deeper into the domain, which increases activity (pIC50 =
3.89) and prevents too early dissociation. This small pocket in the COX-2 active site admits
compound 11 to bend and access by the benzisothiazole rings a similar domain as the
meloxicam benzothiazine rings (Figure 6, red box B). Additionally, compounds 13 and 11
interact with the binding domain by hydrogen bonds. Compound 13 interacts with COX-1 by
arginine 120 (Arg 120) and serine 353 (Ser 353) and with COX-2 by tyrosine 355 (Tyr 355),
arginine 120 (Arg 120) and serine 353 (Ser 353). Compound 11 interacts with COX-1 by
arginine 120 (Arg 120) and with COX-2 by serine 353 (Ser 353) and tyrosine 355 (Tyr 355),
as shown in Figure 6, bottom panel.
For the other compounds, the biological activity is a direct result of the dissociation
time from the binding domain, which is a composition of two components: the length of the
linker and steric fit to the external domain more dissimilar between enzyme variants. The
general correlation between binding mode and observed biological activity might be
explained by considering the group of compounds 17, 20 and 22. Compounds 17, 20 and 22
are presented in Figure 7 as green, violet and red, respectively. The compounds are
biologically active under COX-1 and inactive under COX-2 binding conditions. The general
mode of binding for both groups is almost the same, with the difference that the compounds
under COX-1 may occupy an external hydrophobic domain which is lost under COX-2. Even
though the compounds under COX-2 penetrate the domain deeper, their steric anchor is less
powerful and the compounds dissociate faster. This observation correlates with biological
activity pIC50 = 3.28, 3.43, 2.81 for 22, 17, 20, respectively.

Figure 7. Docking poses of 22 (green), 17 (violet) and 20 (red) under COX-1 and COX-2
binding domain conditions (left panel) and the same docking positions from the binding
domain entrance perspective (right panel). Residues colored cyan; all hydrogens were hidden.
A, B, C, D – binding subdomains; details in the text.
2.3.3 COMPOUNDS’ SELECTIVITY
The small compounds (14-16, 21) bind to the central part of the catalytic domain. As
presented in the example of 21 (Figure 8, magenta), it shares almost the same position as
flurbiprofen (Figure S1 in supporting materials) and its selectivity is slight (21, Δ pIC50 =
0.13). Higher selectivity might be observed for compound 11 in the case of COX-2 (Δ pIC50 =
0.29) and 13 in the case of COX-1 (Δ pIC50 = 0.34). However, the highest selectivity is
observed for compound 25 (Δ pIC50 = 1.48), due to compound 25 binding in a slightly
different manner compared to 21 (Figure 8, magenta). It resembles compound 11’s mode of
binding under COX-2 binding conditions. In the case of 25 the isoleucine 523 residue in
COX-1 sterically stabilizes the compound in the binding domain, and this steric promoter is
lost when valine 523 is under the COX-2 binding domain (Figure 8, ligand map with red
dotted line). Compound 25 stabilized by isoleucine stays longer in the binding domain, which
is directly reflected in biological activity.
Figure 8. Docking poses of 21 (light blue) and 25 (yellow) under COX-1 and COX-2 binding
domain conditions (top panel) and ligand maps representing interactions by hydrogen bonds
(blue dotted line) and steric interactions (red dotted line) (bottom panel). Residues colored
cyan; all hydrogens were hidden. A, B, C, D – binding subdomains; details in the text.

Additionally, the Schrödinger atom-based QSAR, CoMFA and CoMSIA models were
determined, but they did not provide any further insight for the discussion.
3. CONCLUSION
Here, we have presented the previously unexplored isothiazolopyridine scaffold as a
COX inhibitor and its phenylpiperazine derivatives connected by different linkers.
All the tested compounds inhibited the activity of COX-1 at 2 min after addition of the
tested compounds. This effect was the strongest in the cases of compounds 9, 13-16, 21 and
24 (inhibition of the enzyme by 25-40%). The COX-2 enzyme’s activity was strongly
inhibited by compounds 11, 13-15, 21 and 23 (by 20-25%).
Activity of COX-1 was strongly inhibited during 10 min incubation with compounds
9, 13, 24 and 25 (by 18-30%). On the other hand, compounds 20 and 23 caused higher
activity of the enzyme (by 3-7% as high as in the relative control). It should be noted that
piroxicam enhanced COX-1 activity by 11% during 10 min of incubation. The diverse impact
of the tested compounds on the COX-2 activity at 10 min of incubation was observed. The
inhibitory effect of compounds 15, 16, 21 and 25 (by 12-21%) and the stimulatory effect of
the enzyme’s activity by compounds 17, 18, 20 and 22 (by 15-25%) were noted. The
inhibitory effect was the strongest in the case of compound 11 (by 38%). For comparison, the
inhibitory effect of piroxicam was 40%.
Data on the ratios of COX-2 to COX-1 activity clearly indicate those compound which
showed significant COX-1- (13 and 25) and COX-2- (10 and 11) selectivity. However,
compounds 16 and 21 exhibited almost equally strong inhibitory action on COX-1 and COX-
2 activities.
These compounds should be further studied in terms of their selectivity or balanced
inhibitory action on cyclooxygenases as candidates for new anti-inflammatory drugs.
The smallest acetic acid derivative of the proposed scaffold (21) showed a similar
mode of binding as flurbiprofen from the crystal structure and similar activity as piroxicam.
4. EXPERIMENTAL SECTION
4.1. Chemistry
All chemicals used were purchased from commercial suppliers. Dry solvents were obtained
according to the standard procedure. Progress of the reaction was monitored by TLC on silica
gel 60 F254-coated TLC plates (Fluka Chemie GmbH) and visualized by UV light at 254 nm.

Flash column chromatographic purifications were performed using Sigma-Aldrich 60A silica
gel 230–400 mesh. The proton nuclear magnetic resonance (1H NMR) spectra were recorded
on a Bruker 300 MHz NMR spectrometer in d-chloroform (CDCl3) or d6-dimethylsulfoxide
(DMSO-d6) and tetramethylsilane (TMS) was used as an internal reference. Chemical shifts
are given in ppm units, and coupling constant values are given in Hz. Infrared (IR) spectra
were run on a Perkin-Elmer Spectrum Two UATR FT-IR spectrometer, and frequencies are
reported in cm-1. The samples were applied as solids. Elemental analyses for carbon, nitrogen
and hydrogen were carried out on a Carlo Erba NA 1500 analyser and were within ±0.4 % of
the theoretical value.
4.1.1. General method for preparation of 2-(3-oxo-3-substituted-propyl)-4,6-
dimethylisothiazolo[5,4-b]pyridin-3(2H)-one/1,2-benzisothiazol-3(2H)-one derivatives (8-
11) (Scheme 1)
4.1.1.1. Acrylamide (3a-c) and 3-bromopropanamide (26a-c) derivatives (Scheme 2)
5 mmol of the appropriate amine derivative [1-(3-chlorophenyl)piperazine to obtain 3a and
26a, 1-phenylpiperazine to obtain 3b and 26b, 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline
to obtain 3c and 26c] was dissolved in 30 mL of diethyl ether with addition of 5 mmol
triethylamine and stirred slowly at room temperature for 10 minutes. Then 5 mmol of 3-
bromopropionyl chloride in 10 mL of diethyl ether was slowly instilled and stirring was
continued for another 1 hours. After this time diethyl ether was evaporated in a vacuum, 10
mL of water and 60 mL of chloroform were added, and the mixture was separated in a funnel
into two parts. The chloroform part was dried with magnesium sulfate and evaporated in a
vacuum. The resulting residue was chromatographed, and obtained compounds were used in
the subsequent reactions without further purification.
4.1.1.1.1. 1-acryloyl-4-(3-chlorophenyl)piperazine (3a)
Anal. C13H15ClN2O (m.w. 250.72); CC (ethyl acetate, Rf=0.57), m.p. 69-71˚C, FT-IR (UATR,
1
selected lines) νmax/cm-1: 1655 (C=O), 1680 (C=C), 3085 (=C-H). H NMR (CDCl3) δ: 3.18-
3.22 (t, 4H, 2xCH2-piperazine, J=5.1), 3.72-3.84 (m, 4H, 2xCH2-piperazine), 5.72 (dd, 1H, CH2=CH,
J=2.1 and J=10.5 Hz), 6.30 (dd, 1H, CH2=CH, J=1.8 and J=16.8 Hz), 6.55 (dd, 1H, CH2=CH,
J=10.5 and J=16.8 Hz), 6.88-7.31 (m, 5H, ArH).
4.1.1.1.2. 1-acryloy-4-phenylpiperazine (3b)
Anal. C13H16N2O (m.w. 216.29); CC (ethyl acetate, Rf=0.53), m.p. 63-65˚C, FT-IR (UATR,
selected lines) νmax/cm-1: 1650 (C=O), 1675 (C=C), 3080 (=C-H).1H NMR (CDCl3) δ: 3.18-

3.21 (t, 4H, 2xCH2-piperazine, J=5.1), 3.72-3.85 (m, 4H, 2xCH2-piperazine), 5.71 (dd, 1H, CH2=CH,
J=1.8 and J=10.5 Hz), 6.29 (dd, 1H, CH2=CH, J=1.8 and J=16.8 Hz), 6.56 (dd, 1H, CH2=CH,
J=10.5 and J=16.8 Hz), 6.77-7.16 (m, 4H, ArH).
4.1.1.1.3. 2-acryloyl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline (3c)
Anal. C14H17NO3 (m.w. 247.30); CC (ethyl acetate, Rf=0.43), m.p. 99-101˚C, FT-IR (UATR,
1
selected lines) νmax/cm-1: 1655 (C=O), 1680 (C=C), 3060 (=C-H). H NMR (CDCl3) δ: 2.79-
2.85 (m, 2H, CH2-tetrahydroisoquinoline), 3.74-3.78 (m, 2H, CH2-tetrahydroisoquinoline), 3.85 (s, 6H,
2xCH3O), 4.65 (s, 0.75H, CH2-tetrahydroisoquinoline), 4.72 (s, 1.25H, CH2-tetrahydroisoquinoline), 5.70
(dd, 1H, CH2=CH, J=1.8 and J=10.5 Hz), 6.29 (dd, 1H, CH2=CH, J=1.8 and J=16.8 Hz),
6.57-6.69 (m, 3H, CH2=CH + 2xArH).
4.1.1.1.4. 1-(3-bromopropanoyl)-4-(3-chlorophenyl)piperazine (26a)
Anal. C13H16BrClN2O (m.w. 331.64); CC (ethyl acetate, Rf=0.75), m.p. 78-80˚C, FT-IR
1
(UATR, selected lines) νmax/cm-1: 1650 (C=O). H NMR (CDCl3) δ: 2.97 (t, 2H, CH2CO,
J=7.2 Hz), 3.17-3.23 (m, 4H, 2xCH2-piperazine), 3.63-3.83 (m, 6H, 2xCH2-piperazine + BrCH2),
6.81-7.23 (m, 4H, ArH).
4.1.1.1.5. 1-(3-bromopropanoyl)-4-phenylpiperazine (26b)
Anal. C13H17BrN2O (m.w. 297.20); CC (ethyl acetate, Rf=0.72), m.p. 73-76˚C, FT-IR (UATR,
1
selected lines) νmax/cm-1: 1660 (C=O). H NMR (CDCl3) δ: 2.95 (t, 2H, CH2CO, J=7.2 Hz),
3.15-3.21 (m, 4H, 2xCH2-piperazine), 3.64-3.86 (m, 6H, 2xCH2-piperazine + BrCH2), 6.89-7.30 (m,
5H, ArH).
4.1.1.1.6. 2-(3-bromopropanoyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline (26c)
Anal. C14H18BrNO3 (m.w. 327.9); CC (ethyl acetate, Rf=0.62), m.p. 80-84˚C, FT-IR (UATR,
selected lines) νmax/cm-1: 1645 (C=O).1H NMR (CDCl3) δ: 2.77-3.02 (m, 4H, CH2-
tetrahydroisoquinoline + CH2CO), 3.64-3.71 (m, 2H, BrCH2), 3.81-3.86 (m, 8H, CH2-tetrahydroisoquinoline
+ 2xCH3O), 4.55 (s, 0.75H, CH2-tetrahydroisoquinoline), 4.66 (s, 1.25H, CH2-tetrahydroisoquinoline), 6.58-
6.62 (m, 2H, ArH).
4.1.1.2. Dehydrohalogenation of 3-bromopropanamide derivatives to the corresponding
acrylamides (Scheme 2)
A mixture of 5 mmol of corresponding 3-bromopropanamides and 5 mmol of anhydrous
potassium carbonate in 30 mL of acetonitrile was refluxed with stirring for 10 h. The hot
reaction mixture was filtered and evaporated to dryness to give appropriate acrylamides.

4.1.1.3.
2-(3-oxo-3-substituted-propyl)-4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-
one/1,2-benzisothiazol-3(2H)-one derivatives (8-11) (Scheme 1)
A mixture of 4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-one 1 or 1,2-benzisothiazol-3(2H)-
one 2 (3 mmol) with 3 mmol of corresponding acrylamide derivatives in 30 mL of toluene
was refluxed with stirring for 8 h. After this time the mixture was evaporated to dryness. The
product was isolated from the resulting residue by column chromatography (ethyl acetate) and
purified by crystallization from appropriate solvents.
4.1.1.3.1.
2-{3-oxo-3-[4-(3-chlorophenyl)piperazin-1-yl]propyl}-4,6-
dimethylisothiazolo[5,4-b]pyridin-3(2H)-one (8)
Anal. C21H23ClN4O2S (m.w. 430.96); CC (ethyl acetate, Rf=0.37), 67% yield, m.p. 141-143˚C
1
(methanol). FT-IR (UATR, selected lines) νmax/cm-1: 1670 (3-C=O), 1640 (amide C=O). H
NMR (CDCl3) δ: 2.58 (s, 3H, CH3-pyridine), 2.73 (s, 3H, CH3-pyridine), 2.84 (t, 2H, CH2CON,
J=6.6 Hz), 3.13-3.16 (m, 4H, 2xCH2-piperazine), 3.59 (t, 2H, CH2-piperazine, J=5.1 Hz), 3.78 (t, 2H,
CH2-piperazine, J=5.1 Hz), 4.20 (t, 2H, Nisothiazole-CH2, J=6.6 Hz), 6.74-7.20 (m, 5H, Hβ-pyridine
+
4ArH). 13C NMR (CDCl3) δ: 17.55 (CH3), 24.05 (CH3), 32.54 (CH2), 40.18 (C-piperazine), 41.35
(C-piperazine), 45.01 (CH2), 49.12 (C-piperazine), 49.35 (C-piperazine), 114.73 (C-pyridine), 115.41
(C-Ar), 116.64 (C-Ar), 117.46 (C-Ar), 118.84 (C-Ar), 120.70 (C-Ar), 123.10 (C-Ar), 127.44 (C-
pyridine), 137.35 (C-pyridine), 142.26 (C-pyridine), 151.46 (C-pyridine), 164.55 (CO), 168.47 (CO).
Anal. Calcd: C, 58.52; H, 5.39; N, 13.00. Found: C, 58.70; H, 5.55; N, 12.86.
4.1.1.3.2.
2-[3-oxo-3-(4-phenylpiperazin-1-yl)propyl]-4,6-dimethylisothiazolo[5,4-
b]pyridin-3(2H)-one (9)
Anal. C21H24N4O2S (m.w. 396.52); CC (ethyl acetate, Rf=0,33), 45% yield, m.p. 124-126˚C
1
(ethanol). FT-IR (UATR, selected lines) νmax/cm-1: 1670 (3-C=O), 1640 (amide C=O). H
NMR (CDCl3) δ: 2.58 (s, 3H, CH3-pyridine), 2.73 (s, 3H, CH3-pyridine), 2.85 (t, 2H, CH2CON,
J=6.6 Hz), 3.14-3.18 (m, 4H, 2xCH2-piperazine), 3.64 (t, 2H, CH2-piperazine, J=5.1 Hz), 3.82 (t, 2H,
CH2-piperazine, J=5.1 Hz), 4.20 (t, 2H, Nisothiazole-CH2, J=6.6 Hz), 6.92-7.31 (m, 6H, Hβ-pyridine
5ArH). Anal. Calcd: C, 63.61; H, 6.10; N, 14.13. Found: C, 63.50; H, 6.42; N, 14.07.
+
4.1.1.3.3.
2-[3-oxo-3-(6,7-dimethoxy-3,4-dihydroisoquinolin-2(1H)-yl)propyl]-4,6-
dimethylisothiazolo[5,4-b]pyridin-3(2H)-one (10)
Anal. C22H25N3O4S (m.w. 427.53); CC (ethyl acetate, Rf=0.16), 49% yield, m.p. 144-145˚C
1
(methanol). FT-IR (UATR, selected lines) νmax/cm-1: 1670 (3-C=O), 1640 (amide C=O). H
NMR (CDCl3) δ: 2.59-2.79 (m, 8H, 2xCH3-pyridine + CH2-tetrahydroisoquinoline), 2.87 (q, 2H,
CH2CON, J=6,3 Hz), 3.61-3.65 (t, 1H, CH2-tetrahydroisoquinoline, J=6.0 Hz), 3.80-3.85 (m, 7H,
CH2-tetrahydroisoquinoline + 2xCH3O), 4.20 (q, 2H, Nisothiazole-CH2, J=6.6 Hz), 4,49 (s, 1H, CH2-

tetrahydroisoquinoline), 4.67 (s, 1H, CH2-tetrahydroisoquinoline), 6.44-6.93 (m, 3H, Hβ-pyridine + 2ArH).
Anal. Calcd: C, 61.81; H, 5.89; N, 9.83. Found: C, 62.10; H, 6.12; N, 9.64.
4.1.1.3.4. 2-{3-oxo-3-[4-(3-chlorophenyl)piperazin-1-yl]propyl}-1,2-benzisothiazol-3(2H)-
one (11)
Anal. C20H22ClN3O2S (m.w. 401.91); CC (ethyl acetate, Rf=0.41), 32% yield, m.p. 131-133˚C
1
(ethanol). FT-IR (UATR, selected lines) νmax/cm-1: 1710 (3-C=O), 1640 (amide C=O). H
NMR (CDCl3) δ: 2.85 (t, 2H, CH2CON, J=6.3 Hz), 3.09-3.15 (m, 4H, 2xCH2-piperazine), 3.58 (t,
2H, CH2-piperazine, J=5.4 Hz), 3.77 (t, 2H, CH2-piperazine, J=5.4 Hz), 4.22 (t, 2H, Nisothiazole-CH2,
J=6.3 Hz), 6.73-7.99 (m, 8H, ArH). Anal. Calcd: C, 59.76; H, 5.03; N, 10.46. Found: C,
60.14; H, 5.24; N, 10.39.
4.1.2. Method for the synthesis of 2-{2-oxo-2-[4-(3-chlorophenyl)piperazin-1-yl]ethyl}-
4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-one (12) (Scheme 1)
4.1.2.1. 1-chloroacetyl-4-(3-chlorophenyl)piperazine (4)
Compound 4 was prepared according to the procedure described in 4.1.1.1., starting from 1-
(3-chlorophenyl)piperazine and equimolar amounts of chloroacetyl chloride. The oily residue
was chromatographed (CC, ethyl acetate).
Anal. C12H14Cl2N2O (m.w. 273.16); FT-IR (UATR, selected lines) νmax/cm-1: 1655 (C=O). 1H
NMR (CDCl3) δ: 3.18-3.27 (m, 4H, 2xCH2-piperazine), 3.71 (t, 2H, CH2-piperazine, J=5.1 Hz), 3.81
(t, 2H, CH2-piperazine, J=5.1 Hz), 4.11 (s, 2H, CH2CO), 6.92-7.33 (m, 4H, ArH).
4.1.2.2. 2-{2-oxo-2-[4-(3-chlorophenyl)piperazin-1-yl]ethyl}-4,6-dimethylisothiazolo[5,4-
b]pyridin-3(2H)-one (12)
To the solution of sodium ethoxide, prepared from 0.23 g of Na and 50 mL of anh. ethanol, 2
mmol of 4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-one 1 and 3 mmol of 1-chloroacetyl-4-
(3-chlorophenyl)piperazine 4 were added. The mixture was refluxed with stirring for 5 h.
After this time the mixture was cooled, then the precipitated product was filtered off, washed
with water and crystallized from methanol.
Anal. C20H21ClN4O2S (m.w. 416.93); 47% yield, m.p. 144-146˚C. FT-IR (UATR, selected
lines) νmax/cm-1: 1650 (3-C=O and amide C=O). 1H NMR (CDCl3) δ: 2.67 (s, 3H, CH3-pyridine),
2.79 (s, 3H, CH3-pyridine), 3.22-3.26 (m, 4H, 2xCH2-piperazine), 3.75 (t, 2H, CH2-piperazine, J=5.1
Hz), 3.82 (t, 2H, CH2-piperazine, J=5.1 Hz), 4.70 (s, 2H, Nisothiazole-CH2), 6.84-7.18 (m, 5H, Hβ-
pyridine + 4ArH). Anal. Calcd: C, 57.62; H, 5.08; N, 13.44. Found: C, 57.24; H, 5.17; N, 13.25.

4.1.3. Method for the synthesis of 2-[1-oxo-1-(4-phenylpiperazin-1-yl)butan-2-yl]-4,6-
dimethylisothiazolo[5,4-b]pyridin-3(2H)-one (13) (Scheme 1)
4.1.3.1. 1-(2-bromobutanoyl)-4-phenylpiperazine (5)
Compound 5 was prepared according to the procedure described in 4.1.1.1., starting from 1-
phenylpiperazine and equimolar amounts of 2-bromobutyryl bromide. The oily residue was
chromatographed (CC, ethyl acetate, Rf=0.86).
1
Anal. C14H19BrN2O (m.w. 311.22); FT-IR (UATR, selected lines) νmax/cm-1: 1660 (C=O). H
NMR (CDCl3) δ: 1.02-1.08 (m, 3H, CH3CH2), 2.02-2.24 (m, 2H, CH3CH2), 3.25-3.28 (m, 4H,
2xCH2-piperazine), 3.70-3.97 (m, 4H, 2xCH2-piperazine), 4.35 (t, 1H, CH2CH(Br)CO, J=7.2 Hz),
6.91-7.33 (m, 5H, ArH).
4.1.3.2.
2-[1-oxo-1-(4-phenylpiperazin-1-yl)butan-2-yl]-4,6-dimethylisothiazolo[5,4-
b]pyridin-3(2H)-one (13)
A mixture of 5 mmol of 4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-one 1, 5 mmol of
anhydrous potassium carbonate and 5.5 mmol of 1-(2-bromobutanoyl)-4-phenylpiperazine 5
in acetonitrile (30 mL) was refluxed for 4 h. The hot reaction mixture was filtered and
evaporated to dryness. The product was isolated from the resulting residue by column
chromatography (ethyl acetate/cyclohexane/chloroform 2:2:1) and purified by crystallization
from methanol.
Anal. C22H26N4O2S (m.w. 410.54); 44% yield, m.p. 121-124˚C. FT-IR (UATR, selected lines)
1
νmax/cm-1: 1650 (3-C=O and amide C=O). H NMR (CDCl3) δ: 0.96 (t, 3H, CH3CH2, J=7.5
Hz), 2.03-2.09 (m, 2H, CH3CH2), 2.67 (s, 3H, CH3-pyridine), 2.80 (s, 3H, CH3-pyridine), 3.17-3.26
(m, 4H, 2xCH2-piperazine), 3.79-3.96 (m, 4H, 2xCH2-piperazine), 5.67 (t, 1H, CH2CH(Br)CO, J=7.5
Hz), 6.93-7.27 (m, 6H, Hβ-pyridine + 5ArH). Anal. Calcd: C, 64.36; H, 6.38; N, 13.65. Found:
C, 64.43; H, 6.59; N, 13.48.
4.1.4.
Method
for
synthesis
of
2-[(4,6-dimethylisothiazolo[5,4-b]pyridin-3-
yl)oxy]propanoic acid (14) (Scheme 1)
4.1.4.1. Ethyl 2-[(4,6-dimethylisothiazolo[5,4-b]pyridin-3-yl)oxy]propanoate (7a) and its
2-N-substituted isomer (6a)

The mixture of 4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-one 1 (5 mmol), potassium
carbonate (5.5 mmol) and ethyl 2-bromopropanoate (5.5 mmol) in 30 mL of acetonitrile was
heated to reflux and stirred for 3 h. The hot reaction mixture was filtered and evaporated to
dryness. The residue was chromatograhed (CC; ethyl acetate/cyclohexane/chloroform 3:2:1)
to give fractions containing 3-O-substituted isomer 7a (Rf=0.79) and 2-N-substituted isomer
6a (Rf=0.65, not described).
Anal. C13H16N2O3S (m.w. 280.35); 45% yield, m.p. 38-39˚C. 1H NMR (CDCl3) δ: 1.26 (t, 3H,
OCH2CH3, J=7.2Hz), 1.68 (d, 3H, CH(CH3)CO, J=6.9), 2.64 (s, 3H, CH3-pyridine), 2.73 (s, 3H,
CH3-pyridine), 4.19-4.26 (m, 2H, OCH2CH3), 5.46 (q, 1H, CH(CH3)CO, J=7.2Hz), 6.96 (s, 1H,
Hβ-pyridine). Anal. Calcd: C, 55.70; H, 5.75; N, 9.99. Found: C, 55.32; H, 5.70; N, 8.99.
4.1.4.2. 2-[(4,6-dimethylisothiazolo[5,4-b]pyridin-3-yl)oxy]propanoic acid (14)
Compound 7a (1 mmol) was dispersed in 30 mL of concentrated HCl (37%) and refluxed for
1 h. The mixture was cooled, filtered and the precipitate was dissolved in 30 mL of aqueous
sodium bicarbonate (5%). The solution was filtered and acidified with hydrochloric acid (2N).
The precipitate was filtered, washed with water, dried and crystallized from ethanol-water.
Anal. C11H12N2O3S (m.w. 252,29); 87% yield, m.p. 178-180˚C. FT-IR (UATR, selected lines)
1
νmax/cm-1: 3000-2500 br(OH), 1710 (C=O acid). H NMR (DMSO-d6) δ: 1.58 (d, 3H,
CH(CH3)CO, J=6.9), 2.57 (s, 3H, CH3-pyridine), 2.66 (s, 3H, CH3-pyridine), 5.31 (q, 1H,
CH(CH3)CO, J=6.9Hz), 7.20 (s, 1H, Hβ-pyridine). Anal. Calcd: C, 52.37; H, 4.79; N, 11.10.
Found: 52.48; H, 4.59; N, 10.72
4.1.5. General method for preparation of 2-[4-oxo-4-substituted-butyl]-4,6-
dimethylisothiazolo[5,4-b]pyridin-3(2H)-one and 3-[4-oxo-4-substituted-butoxy]-4,6-
dimethylisothiazolo[5,4-b]pyridine derivatives (17-20) (Scheme 1)
4.1.5.1. 4-(4,6-dimethyl-3-oxo-2,3-dihydroisothiazolo[5,4-b]pyridin-2-yl)butanoic acid
(15) and 4-[(4,6-dimethylisothiazolo[5,4-b]pyridin-3-yl)oxy]butanoic acid (16)
Compounds 15 and 16 were prepared by acid-catalyzed hydrolysis of ethyl 4-(4,6-dimethyl-3-
oxo-2,3-dihydroisothiazolo[5,4-b]pyridin-2-yl)butanoate 6b or its 3-O-substituted isomer 7b
according to the procedure described in 4.1.4.1.. The product was crystallized from ethanol-
water.
4.1.5.1.1. 4-(4,6-dimethyl-3-oxo-2,3-dihydroisothiazolo[5,4-b]pyridin-2-yl)butanoic acid
(15)

Anal. C12H14N2O3S (m.w. 266.32); 87% yield, m.p. 141-143˚C. FT-IR (UATR, selected lines)
νmax/cm-1: 3100-2400 br(OH), 1720 (C=O acid), 1650 (3-C=O). 1H NMR (DMSO-d6) δ: 1.83-
1.87 (m, 2H, CH2CH2CH2), 2.26 (t, 2H, CH2CO, J=7.2), 2.53 (s, 3H, CH3-pyridine), 2.63 (s, 3H,
CH3-pyridine), 3.81 (t, 2H, Nisothiazole-CH2, J=6.9 Hz), 7.18 (s, 1H, Hβ-pyridine). 13C NMR (DMSO-
d6) δ: 17.70 (CH3), 24.57 (CH3), 30.72 (CH2), 41.72 (CH2), 45.28 (CH2), 114.55 (C-pyridine),
123.09 (C-pyridine), 127.44 (C-pyridine), 137.35 (C-pyridine), 142.26 (C-pyridine), 164.76 (CO), 176.42
(COOH). Anal. Calcd: C, 54.12; H, 5.30; N, 10.52. Found: C, 54.44; H, 5.47; N, 10.74.
4.1.5.1.2. 4-[(4,6-dimethylisothiazolo[5,4-b]pyridin-3-yl)oxy]butanoic acid (16)
Anal. C12H14N2O3S (m.w. 266.32); 90% yield, m.p. 193-195˚C. FT-IR (UATR, selected lines)
1
νmax/cm-1: 3200-2500 br(OH), 1710 (C=O acid). H NMR (DMSO-d6) δ: 2.01-2.05 (m, 2H,
CH2CH2CH2), 2.42 (t, 2H, CH2CO, J=7.2), 2.56 (s, 3H, CH3-pyridine), 2.62 (s, 3H, CH3-pyridine),
4.46 (t, 2H, Nisothiazole-CH2, J=6.3 Hz), 7.17 (s, 1H, Hβ-pyridine). Anal. Calcd: C, 54.12; H, 5.31;
N, 10.52. Found: C, 54.42; H, 5.26; N, 10.47.
4.1.5.2.
2-[4-oxo-4-substituted-butyl]-4,6-dimethylisothiazolo[5,4-b]pyridin-3(2H)-one
and 3-[4-oxo-4-substituted-butoxy]-4,6-dimethylisothiazolo[5,4-b]pyridine derivatives
(17-20)
1 mmol of 4-(4,6-dimethyl-3-oxo-2,3-dihydroisothiazolo[5,4-b]pyridin-2-yl)butanoic acid 15
or its 3-O-substituted isomer 16 in 30 mL of dichloromethane at 0°C (ice-bath) was treated
with triethylamine (0.05 mL) and 1 mmol of ethyl chloroformate. After stirring the reaction
mixture at 0°C for 15 min, 1.1 mmol of an appropriate amine derivative [1-(3-
chlorophenyl)piperazine to obtain 17 and 18, 1-phenylpiperazine to obtain 19 and 20] was
added to this solution. The final mixture was stirred at 0-25°C for 20 h. After this time the
mixture was evaporated to dryness. The residue was chromatograhed and crystallized from
appropriate solvents.
4.1.4.2.1.
2-{4-oxo-4-[4-(3-chlorophenyl)piperazin-1-yl]butyl}-4,6-
dimethylisothiazolo[5,4-b]pyridin-3(2H)-one (17)
Anal. C22H25ClN4O2S (m.w. 444.99); CC (ethyl acetate, Rf=0.25), 36% yield, m.p. 92-94˚C
1
(ethanol). FT-IR (UATR, selected lines) νmax/cm-1: 1640 (3-C=O and amide C=O). H NMR
(CDCl3) δ: 2.11-2.15 (m, 2H, CH2CH2CH2), 2.43 (t, 2H, CH2CO, J=7.2), 2.58 (s, 3H, CH3-
pyridine), 2.71 (s, 3H, CH3-pyridine), 3.12-3.15 (m, 4H, 2xCH2-piperazine), 3.56 (t, 2H, CH2-piperazine
,
J=6.9 Hz), 3.75 (t, 2H, CH2-piperazine, J=6.9 Hz), 3.95 (t, 2H, Nisothiazole-CH2, J=6.6 Hz), 6.75-
7.19 (m, 5H, Hβ-pyridine + 4ArH). Anal. Calcd: C, 59.38; H, 5.66; N, 12.59. Found: : C, 59.62;
H, 5.86; N, 12.66

4.1.4.2.2.
3-{4-oxo-4-[4-(3-chlorophenyl)piperazin-1-yl]butoxy}-4,6-
dimethylisothiazolo[5,4-b]pyridine (18)
Anal. C22H25ClN4O2S (m.w. 444.99); CC (ethyl acetate, Rf=0.54), 29% yield, m.p. 118-120˚C
1
(ethanol). FT-IR (UATR, selected lines) νmax/cm-1: 1630 (amide C=O). H NMR (CDCl3) δ:
2.23-2.27 (m, 2H, CH2CH2CH2), 2.57 (t, 2H, CH2CO, J=7.2), 2.63 (s, 3H, CH3-pyridine), 2.66
(s, 3H, CH3-pyridine), 3.14-3.17 (m, 4H, 2xCH2-piperazine), 3.61 (t, 2H, CH2-piperazine, J=5.1 Hz),
3.77 (t, 2H, CH2-piperazine, J=5.1 Hz), 4.58 (t, 2H, Nisothiazole-CH2, J=6.3 Hz), 6.75-7.20 (m, 5H,
Hβ-pyridine + 4ArH). Anal. Calcd: C, 59.38; H, 5.66; N, 12.59. Found: C, 59.72; H, 5.86; N,
12.59.
4.1.4.2.3.
2-[4-oxo-4-(4-phenylpiperazin-1-yl)butyl]-4,6-dimethylisothiazolo[5,4-
b]pyridin-3(2H)-one (19)
Anal. C22H26N4O2S (m.w. 410.54); CC (ethyl acetate, Rf=0.26), 32% yield, m.p. 81-83˚C
1
(ethanol). FT-IR (UATR, selected lines) νmax/cm-1: 1640 (3-C=O and amide C=O). H NMR
(CDCl3) δ: 2.12-2.17 (m, 2H, CH2CH2CH2), 2.45 (t, 2H, CH2CO, J=7.2), 2.60 (s, 3H, CH3-
pyridine), 2.72 (s, 3H, CH3-pyridine), 3.17-3.21 (m, 4H, 2xCH2-piperazine), 3.63 (t, 2H, CH2-piperazine
,
J=6.6 Hz), 3.82 (t, 2H, CH2-piperazine, J=6.6 Hz), 3.97 (t, 2H, Nisothiazole-CH2, J=6.6 Hz), 6.94-
7.30 (m, 6H, Hβ-pyridine + 5ArH). Anal. Calcd: C, 64.36; H, 6.38; N, 13.65. Found: C, 64.31;
H, 6.57; N, 13.51.
4.1.4.2.4.
3-[4-oxo-4-(4-phenylpiperazin-1-yl)butoxy]-4,6-dimethylisothiazolo[5,4-
b]pyridine (20)
Anal. C22H26N4O2S (m.w. 410.54); CC (ethyl acetate/cyclohexane 2:2, Rf=0.23), 44% yield,
m.p. 98-100˚C (ethanol). FT-IR (UATR, selected lines) νmax/cm-1: 1670 (3-C=O), 1630
1
(amide C=O). H NMR (CDCl3) δ: 2.24-2.28 (m, 2H, CH2CH2CH2), 2.58 (t, 2H, CH2CO,
J=7.2), 2.64 (s, 3H, CH3-pyridine), 2.67 (s, 3H, CH3-pyridine), 3.15-3.19 (m, 4H, 2xCH2-piperazine),
3.66 (t, 2H, CH2-piperazine, J=4.8 Hz), 3.82 (t, 2H, CH2-piperazine, J=4.8 Hz), 4.59 (t, 2H, Nisothiazole-
13
CH2, J=6.3 Hz), 6.94-7.32 (m, 6H, Hβ-pyridine + 5ArH). C NMR (CDCl3) δ: 19.06 (CH3),
24.34 (CH3), 29.67 (CH2), 41.36 (C-piperazine), 41.46 (C-piperazine), 45.01 (CH2), 49.78 (C-
piperazine), 49.89 (C-piperazine), 67.66 (CH2), 116.30 (C-pyridine), 117.02 (C-pyridine), 121.48 (C-Ar),
129.37 (C-Ar), 145.88 (C-pyridine), 160.53 (C-pyridine), 161.85 (C-pyridine), 170.48 (CO), 176.56
(CO). Anal. Calcd: C, 64.36; H, 6.39; N, 13.65. Found: C, 64.03; H, 6.44; N, 13.35.
4.2 In vitro cyclooxygenase inhibitory activity
COX Colorimetric Inhibitor Screening Assay Kit (Cayman Chemical Company, Ann
Arbor, MI)

This method allows one to estimate the peroxidase activity of COX by colorimetric
monitoring of occurrence of the oxidized form of N,N,N',N'-tetramethyl-p-phenylenediamine
(TMPD) at 590 nm. TMPD is a substrate for most enzymes with peroxidase activity, and high
throughput microplate assays using TMPD allow the rapid screening of a wide range of
therapeutics that inhibit COX activity in vitro.
The test is based on the oxidation of TMPD during the reduction of PGG2 (prostaglandin G2)
to PGH2, which is reflected by a change in color, measured spectrophotometrically (Victor2
microspectrophotometer, PerkinElmer Waltham, MA, USA). The assay uses Tris-HCl buffer
(0.1 M assay buffer, pH 8.0), a solution of heme in dimethylsulfoxide (DMSO), enzymes
(COX-1, COX-2), arachidonic acid (100 μM), KOH (0.1 M) and a solution of TMPD. The
assay mixture contains: 150 µL of assay buffer, 10 µL of heme, and 10 µL of COX-1 or
COX-2. To the wells for determination of 100% enzyme activity (each COX sample was
assayed in triplicate) were added 10 µL of the substances used as solvents (methanol, ethanol,
DMSO). To the other wells were added 10 μL of tested inhibitors at appropriate
concentrations. To all wells were added 20 µL of TMPD. The reaction was initiated by the
addition of arachidonic acid.
The effect of tested inhibitors on COX-1 and COX-2 enzyme activity was measured by
assaying the rate of TMPD oxidation within 2 minutes in a spectrophotometer at 590 nm.
We determined the activity factor at 2 min of incubation with the tested compounds in
comparison to the initial activity of the enzyme. This enabled the calculation of IC50 values
(concentrations at which 50% inhibition of enzyme activity occurred).
Statistical analysis
Statistical significance of the results was calculated with the paired t-test and with Two-way
analysis of variance ANOVA, following the routine statistical methods.
4.3 Molecular docking
To date, in the Protein Data Bank (PDB) there are deposited 67 x-ray crystal structures of
cyclooxygenases (COX), 27 of cyclooxygenase-1 (COX-1) and 40 of cyclooxygenase-2
(COX-2). Most of them were co-crystallized with ligands. In the absence of human enzymes
the high-resolution and complete x-ray structures of murine cyclooxygenase crystallized for
Cox-1 (PDB ID: 4O1Z) and Cox-2 (PDB ID: 4M11) variants co-crystallized with the same

ligand (meloxicam) were chosen for molecular docking. The sequence identity of the
structures with human variants is very high (~95%) and the residues comprising the active
sites are fully conserved, as described elsewhere.17 In the preparation step, the protein models
were verified in terms of chemical structure and the type of bonds. All cofactors were
removed. To unify coordinates the structures were aligned using a multiple structural
alignment algorithm (MUSTANG) implemented in Yasara 11.6.16.33. The program
constructs a multiple alignment using the spatial information of the α-carbon atoms of the
protein in the set. The ligands were built de novo in Spartan’10. The energies were minimized
by the Hartree−Fock ab initio algorithm with the 6-31G* basis set. The compounds were
exported as mol2 files and docked into the crystal structure. The docking study was
performed in Molegro Virtual Docker 6.0, which was successfully applied by other
investigators.21-23 Comparisons of docking positions with crystal structure are in the
supporting materials (Figure S3, Table S2). If not marked, the default settings were applied;
the calculations were parameterized as follows: blind docking to whole binding domain,
MolDock Score [GRID] searching and scoring algorithm, grid resolution 0.3 A., search area:
x: 63.25 y: 23.03 z: 206.04, radius 15, number of runs: 1000, number of iterations: 1500, max
population size 50, max number of poses return: 10.
Acknowledgements
The study was supported in part by the PL-Grid (Polish High Performance Computing
Infrastructure). The authors greatly appreciate Prof. Krzysztof Jozwiak, Medical University of
Lublin, Poland for access to the appropriate software.
References
1. Kirkby, N.S.; Chan, M.V.; Zaiss, A.K.; Garcia-Vaz, E.; Jiao, J.; Berglund, L.M.; Verdu,
E.F.; Ahmetaj-Shala, B.; Wallace, J.L.; Herschman, H.R.; Gomez, M.F.; Mitchell, J.A. Proc.
Natl. Acad. Sci. U.S.A. 2016, 113, 434.
2. Chandrasekharan, N.V.; Dai, H.; Roos, K.L.; Evanson, N.K.; Tomsik, J.; Elton T.S.;
Simmons, D.L. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 13926.
3. Nandakishore, R.; Yalavarthi, P.R.; Kiran, Y.R.; Rajapranathi, M. Curr. Drug Discov.
Technol. 2014, 11, 127.

4. Warner, T.D.; Giulliano, F.; Vojnovic, I.; Bukasa, A.; Mitchell, J.A.; Vane, J.R. Proc.
Natl. Acad. Sci. U.S.A 1999, 96, 7563.
5. Unlu, S.; Onkol, T.; Dundar, Y.; Okcelik, B.; Kupeli, E.; Yesilada, E.; Noyanalpan, N.;
Sahin, M.F. Arch. Pharm. 2003, 336, 353.
6. Banoglu, E.; Okcelik, B.; Kupeli, E.; Unlu, S.; Yesilada, E.; Amat, M.; Caturia, J.F.; Sahin,
M.F. Arch. Pharm. 2003, 336, 251.
7. Gulcan, H.O.; Kupeli, E.; Unlu, S.; Yesilada, E.; Sahin, M.F. Arch. Pharm. 2003, 336, 477.
8. Raghavendra, N.M.; Jyothsna, A.; Rao, A.V.; Subrahmanyam, C.V.S. Bioorg. Med. Chem.
Lett. 2012, 22, 820.
9. Eleftheriou, P.; Geronikaki, A.; Hadjipavlou-Litina, D.; Vicini, P.; Filz, O.; Filimonov, D.;
Poroikov, V.; Chaudhaery, S.S.; Roy, K.K.; Saxena, A.K. Eur. J. Med. Chem. 2014, 62, 197.
10. Keri, R.S.; Patil, M.R.; Patil, S.A.; Budagumpi, S. Eur. J. Med. Chem. 2015, 89, 207.
11. Malinka, W.; Sieklucka-Dziuba, M.; Rajtar, G.; Zgodzinski, W.; Kleinrok, Z. Pharmazie
2000, 55, 416.
12. Malinka, W.; Swiatek, P.; Filipek, B.; Sapa, J.; Jezierska, A.; Koll, A. Farmaco 2005, 60,
961.
13. Zawisza, T.; Malinka, W. Farmaco 1985, 40, 124.
14. Malinka, W.; Ryng, S.; Sieklucka-Dziuba, M.; Rajtar, G.; Gowniak, A.; Kleinrok, Z.
Farmaco 1998, 53, 504.
15. Krzyzak, E.; Sliwinska, M.; Malinka, W. J. Fluoresc. 2015, 25, 277.
16. Picot, D.; Loll, P.J.; Garavito, R.M. Nature, 1994, 367, 243.
17. Xu, S.; Hermanson, D.J.; Banerjee, S.; Ghebreselasie, K.; Clayton, G.M.; Garavito, R.M.;
Marnett, L.J. J. Biol. Chem. 2014, 289, 6799.
18. Gierse, J.K.; McDonald, J.J.; Hauser, S.D.; Rangwala, S.H.; Koboldt, C.M.; Seibert, K. J.
Biol. Chem. 1996, 271, 15810.
19. Dhanjal, J.K.; Sreenidhi, A.K.; Bafna, K.; Katiyar, S.P.; Goyal, S.; Grover, A.; Sundar, D.
In PLoS One; 2015; Vol. 10(8), e0134691, doi 10.1371, published online.

20. Blobaum, A.L.; Xu, S.; Rowlinson, S.W.; Duggan, K.C.; Banerjee, S.; Kudalkar, S.N.;
Birmingham, W.R.; Ghebreselasie, K.; Marnett, L.J. J. Biol. Chem. 2015, 290, 12793.
21. Gogoi, D.; Bezbaruah, R.L.; Bordoloi, M.; Sarmah, R.; Bora, T.C. Bioinformation 2012,
8, 812.
22. Palkar, M.B.; Singhai, A.S.; Ronad, P.M.; Vishwanathswamy, A.H.; Boreddy, T.S.;
Veerapur, V.P.; Shaikh, M.S.; Rane, R.A.; Karpoormath, R. Bioorg. Med. Chem. 2014, 22,
2855.
23. Urniaz, R.D.; Jozwiak, K. J. Chem. Inf. Model. 2013, 53, 1406.
24. Rowlinson, S.W.; Kiefer, J.R.; Prusakiewicz, J.J.; Pawlitz, J.L.; Kozak, K.R.; Kalgutkar,
A.S.; Stallings, W.C.; Kurumbail, R.G.; Marnett, L.J. J. Biol. Chem. 2003, 278, 45763.