Synthesis and biological properties of aryl methyl sulfones

Article abstract of DOI:10.1016/j.bmc.2018.06.038

A novel group of aryl methyl sulfones based on nonsteroidal anti-inflammatory compounds exhibiting a methyl sulfone instead of the acetic or propionic acid group was designed, synthesized and evaluated in vitro for inhibition against the human cyclooxygenase of COX-1 and COX-2 isoenzymes and in vivo for anti-inflammatory activity using the carrageenan induced rat paw edema model in rats. Also, in vitro chemosensitivity and in vivo analgesic and intestinal side effects were determined for defining the therapeutic and safety profile. Molecular modeling assisted the design of compounds and the interpretation of the experimental results. Biological assay results showed that methyl sulfone compounds 2 and 7 were the most potent COX inhibitors of this series and best than the corresponding carboxylic acids (methyl sulfone 2: IC₅₀ COX-1 = 0.04 and COX-2 = 0.10 μM, and naproxen: IC₅₀ COX-1 = 11.3 and COX-2 = 3.36 μM). Interestingly, the inhibitory activity of compound 2 represents a significant improvement compared to that of the parent carboxylic compound, naproxen. Further support to the results were gained by the docking studies which suggested the ability of compound 2 and 7 to bind into COX enzyme with low binding free energies. The improvement of the activity of some sulfones compared to the carboxylic analogues would be performed through a change of the binding mode or mechanism compared to the standard binding mode displayed by ibuprofen, as disclosed by molecular modeling studies. So, this study paves the way for further attention in investigating the participation of these new compounds in the pain inhibitory mechanisms. The most promising compounds 2 and 7 possess a therapeutical profile that enables their chemical scaffolds to be utilized for development of new NSAIDs.

Full text of DOI:10.1016/j.bmc.2018.06.038

Accepted Manuscript
Synthesis and biological properties of aryl methyl sulfones
Lorena Navarro, Gloria Rosell, Silvia Sánchez, Núria Boixareu, Klaus Pors,
Ramon Pouplana, Josep M. Campanera, M. Dolors Pujol
PII:
S0968-0896(18)30914-3
https://doi.org/10.1016/j.bmc.2018.06.038
BMC 14430
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
11 May 2018
26 June 2018
28 June 2018
Please cite this article as: Navarro, L., Rosell, G., Sánchez, S., Boixareu, N., Pors, K., Pouplana, R., Campanera,
J.M., Pujol, M.D., Synthesis and biological properties of aryl methyl sulfones, Bioorganic & Medicinal
Chemistry (2018), doi: https://doi.org/10.1016/j.bmc.2018.06.038
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 and biological properties of aryl methyl sulfones
¥
€
€
‡
‡
Lorena Navarro, Gloria Rosell, Silvia Sánchez, Núria Boixareu, Klaus Pors, Ramon
§
§
‡
Pouplana, Josep M. Campanera, * and M. Dolors Pujol *
‡Laboratori de Química Farmacèutica (Unitat Associada al CSIC). Facultat de Farmàcia. Universitat de
Barcelona. Av. Diagonal 643. E-08028-Barcelona (Spain)
§
Departament de Farmàcia i Tecnologia Farmacèutica, i Fisicoquímica, Facultat de Farmàcia i Ciències de
l'Alimentació, Universitat de Barcelona, Av. Diagonal 643, E-08028, Barcelona (Catalonia, Spain)
¥
Unitat de Farmacologia, Departament Patologiai Terapèutica Experimental. Facultat de Medicina. Universitat de
Barcelona. L’Hospitalet de Llobregat. E-08907-Barcelona (Spain)
€
Institute of Cancer Therapeutics, School of Pharmacy and Medical Sciences, Faculty of Life Sciences, University of
Bradford, BD7 1DP, West Yorkshire, U.K.
*To whom correspondence should be addressed. For M.D.P.: E-mail: mdpujol@ub.edu, phone, +34-93-4024534;
a
fax, +34-93-4035941. Abbreviations: COX-1, cyclooxigenase-1; COX-2, cyclooxigenase-2; NSAIDs, Nonsteroidal
anti-inflammatory drugs; PG, prostaglandin; PGH , prostaglandin G ; SAR, structure-activity relation-ships; QSAR,
2
2
quantitative structure-activity relation-ships.
ABSTRACT: A novel group of aryl methyl sulfones based on nonsteroidal anti-inflammatory
compounds exhibiting a methyl sulfone instead of the acetic or propionic acid group was
designed, synthesized and evaluated in vitro for inhibition against the human cyclooxygenase of
COX-1 and COX-2 isoenzymes and in vivo for anti-inflammatory activity using the carrageenan
induced rat paw edema model in rats. Also, in vitro chemosensitivity and in vivo analgesic and
intestinal side effects were determined for defining the therapeutic and safety profile. Molecular
modeling assisted the design of compounds and the interpretation of the experimental results.
Biological assay results showed that methyl sulfone compounds 2 and 7 were the most potent
COX inhibitors of this series and best than the corresponding carboxylic acids (methyl sulfone 2:

IC COX-1 = 0.04 and COX-2 = 0.10 M, and naproxen: IC COX-1 = 11.3 and COX-2 = 3.36
5
0
50
M). Interestingly, the inhibitory activity of compound 2 represents a significant improvement
compared to that of the parent carboxylic compound, naproxen. Further support to the results
were gained by the docking studies which suggested the ability of compound 2 and 7 to bind into
COX enzyme with low binding free energies.
The improvement of the activity of some sulfones compared to the carboxylic analogues would
be performed through a change of the binding mode or mechanism compared to the standard
binding mode displayed by ibuprofen, as disclosed by molecular modeling studies. So, this study
paves the way for further attention in investigating the participation of these new compounds in
the pain inhibitory mechanisms. The most promising compounds 2 and 7 possess a therapeutical
profile that enables their chemical scaffolds to be utilized for development of new NSAIDs.
Keywords: anti-inflammatory, analgesic, NSAIDs, methyl sulfones, COX-inhibitors, binding
free energy estimation
1
. Introduction
Non-steroidal anti-inflammatory drugs (NSAIDs) are the most prescribed pharmaceutical
compounds in the world to alleviate inflammation and pains associated with several pathological
conditions and are often the initial treatment for common inflammation. NSAIDs are a
heterogeneous group of various chemical structures with variable benefit/risk profile. Usually the
use of NSAIDs is associated with several adverse effects, including gastrointestinal damage. The
clinically used NSAIDs exert their therapeutical effects by inhibition of the biosynthesis of
prostaglandins (PGs) and thromboxanes (TX). In general, the biosynthesis involves the
conversion of arachidonic acid to prostaglandin G (PGH ), a reaction catalyzed by the
2
2
sequential action of cyclooxygenase (COX) (Figure 1). Most NSAIDs inhibit COX-1 and COX-2

1
,3
isoforms.
The constitutive COX-1 is responsible for the synthesis of cytoprotective
prostaglandins in the gastrointestinal tract, and for the pro-aggregator thromboxane in blood
4
platelets. In the inflammation process COX-2 is responsible for production of prostaglandins,
considered mediators of inflammation (Figure 1).
Figure 1. Inflammation biochemical pathways
It has recently been reported that classical NSAIDs such as celecoxib possess preventive effects
5
6
against colorectal cancer (CRC) and Alzheimer’s disease. Moreover, these compounds
7
7c
continue to be used as remedies for rheumatic and autoimmune anti-inflammatory diseases. In
general, NSAIDs (Figure 2) are potent anti-oxidants that exert both anti-inflammatory and
8
-10
9a
antitumor activity.
In regard the latter, ibuprofen inhibits tumor growth and liver metastasis.
While long-term use of acetaminophen enhances the development of leukemia regular treatment
1
1
with NSAIDs correlates with a reduced risk of lymphoma and leukemia development. The
chronic use of NSAIDs inhibit the growth of adenomatous polyps of patients with familial
12
adenomatous polyposis and reduce the risk of CRC. Selective COX-2 inhibitors showed a safe
profile in the gastrointestinal tract, but recent studies suggest that the long-term treatment by
selective COX-2 inhibitors is limited because of cardiovascular thrombotic events related to the

1
3
aggregatory properties of these drugs. Moreover, COX-2 makes a significant contribution to
the production of inflammatory PGs while the inhibition of COX-2 attenuates the expression of
14
inflammatory mediators such as TNF-, iNOS and IL-1. NSAID compounds are often
associated with the presence of gastrointestinal side effects and new agents with diminished off-
target effects are highly desirable.
Figure 2. Classical NSAIDs
1
5
In the present study based on the previous work related to COX-1/COX-2 inhibitors and the
properties of NSAIDs, we designed a series of seven new compounds derived from classic and
commercial NSAIDs in order to study their potential therapeutic properties (Figure 3).
Specifically, the purpose of the study was to evaluate whether the replacement of the acetic or
propionic acid group by a methyl sulfone group, a more lipophilic group with low acidic
properties, could lead to high COX activity and reduced risk of side-effects, thereby result in
better anti-inflammatory and safety profile than the parent classical NSAIDs scaffolds.
Here we report on the synthesis of a series of sulfones as a new class of NSAIDs and report on
their biological properties by comparing them with the respective carboxylic acids chosen as
models.

Figure 3. Design strategy of methyl sulfones
Compounds 1-5 derive from the direct substitution of alkylcarboxylic group by methyl sulfone of
the classical NSAIDs fenoprofen, naproxen, ibuprofen, metiazinic acid and indomethacin
respectively, and were chosen due to their structural variability and anti-inflammatory profile.
Two additional methyl sulfones were also studied, compounds 6 and 7, which were derived from
not only a direct substitution of the alkyl carboxylic acid of ketoprofen but also changes at other
positions of the scaffold.
2
. Results and discussion
2
.1. Synthesis and biological evaluation of methyl sulfones
The methods used for the synthesis of aryl sulfones from different starting materials were readily
available and enabled the synthesis of target compounds in a conventional and safe manner. The
general synthetic routes are illustrated in Schemes 1-7.
The methyl sulfone 1 was prepared from 1-bromo-3-methylsulfonylbenzene (9) and phenol
16
under cross-coupling conditions, which was obtained from 3-bromothioanisole (8) by oxidation
with m-CPBA (Scheme 1).

Scheme 1. Synthesis of compound 1
2
-Bromo-6-methoxynaphthalene (10) was transformed to the methylthionaphthalene derivative
1 using butyl lithium for the Br/Li interchange, followed by addition of dimethyl disulfide
1
(DMDS). Methyl sulfone 2 was obtained by oxidation of 11 with m-CPBA (Scheme 2).
Scheme 2. Synthesis of compound 2
The methyl sulfone 3 was obtained in a three-step procedure: thioanisole was subjected to
Friedel Crafts acylation, followed by oxidation of the methylthio intermediate 14 to the
corresponding methyl sulfone 15 with m-CPBA using the optimized conditions described above
(
Scheme 3). The final step involved Clemmensen conditions using Zn-HgCl to obtain target
2
1
7
compound 3.
Scheme 3. Synthesis of compound 3
The phenothiazine 4 was prepared in 3-step transformation from the commercial phenothiazine
6 (Scheme 4). N-Acylation of the phenothiazine under classical conditions followed by
1
methylsulfonation at the C-2 position and later alkylation with methyl iodide at the position 10 of
the heterocycle led to the desired methyl sulfone in acceptable overall yield (19%).

Scheme 4. Synthesis of compound 4
The 3-iodoindole 20 was prepared by iodination of the commercial indole 19 using molecular
iodine and potassium hydroxide to afford the resulting 3-iodoindole 20, which was converted
into methylmercaptane 21 with dimethyldisulfide (DMDS) via a lithium intermediate in
excellent yields. Then regioselective oxidation of 21 with m-CPBA provided the methyl sulfone
2
2, which was reacted with 4-chlorobenzoyl chloride in DMF to afford the N-acylated target
compound 5 (Scheme 5).
Scheme 5. Synthesis of compound 5
The methyl sulfone 6 was prepared by oxidation of the methylthio derivative to the
1
8
corresponding sulfone using m-CPBA (Scheme 6). The intermediate benzophenone (24) was
obtained in one step by acylation of commercial thioanisole (12) with 4-chlorobenzoyl chloride
(23) under classical conditions.
Scheme 6. Synthesis of compound 6

Thioanisole 12 was reacted with furan-3-yl carboxylic acid chloride under Friedel Crafts
conditions using TiCl as a catalyst. The resulting ketone 26 was oxidized with m-CPBA to
4
afford the methyl sulfone 7 in high yield (Scheme 7).
Scheme 7. Synthesis of compound 7
2
.2. Pharmcology
Biological Results. Synthesized methyl sulfones 1-7 were assayed in vitro cyclooxygenase
inhibition, in vivo anti-inflammatory activity, in vivo analgesic activity, in vivo gastrointestinal
side effects and the chemosensitivity in human cells.
In Vitro Cyclooxygenase 1 and Cyclooxygenase 2 Inhibition. The methyl sulfones and the
reference compounds were tested at the same concentrations (10 M, 1 M, and 1 nM).
In vitro COX-1/COX-2 isoenzyme inhibition studies [19] showed that compounds 2, 3, 5, 6 and
7
were more potent inhibitors of COX-1 compared with COX-2 (Table 1) while 1 was shown to
be slightly more selective for COX-2 (Table 1, entry 3). Compounds 1, 2 and 5 exhibited a
prominent COX inhibition (IC50 < 1 M, Table 1, entries 1, 3 and 11). Moreover, compounds 1
and 2 are the only methyl sulfones of these series that improve the potency of COX-2 inhibition
in regard to their carboxylate analogues by 93- and 33- fold, respectively. The methyl sulfone 5
was equipotent and non-selective to its carboxylate analogue, indomethacin (Table 1, entries 10
and 11).

Compound 6 was less potent than ketoprofen when measured against both COX-1 and COX-2
isoenzymes. Also the methyl sulfone 3 showed less activity than the parent NSAIDs ibuprofen.
In contrast, sulfone 1 was equipotent with fenoprofen in regard to COX-1 but almost 100-fold
more superior against COX-2. Methyl sulfones 2 and 7 showed interesting inhibitory COX-1 and
COX-2 activities but little differential isoform selectivity. The phenothiazine 4 presents an
inhibiting activity against COX-1 (4.5-fold) and COX-2 (5.8-fold) and is superior to the
metiazinic acid. The pyrrole 7 was designed according to the molecular modelling results of the
different interactions between several NSAIDs drugs and COX-1 and COX-2 isoforms and it is
related to ketoprofen. Compound 2, an analogue of naproxen, was the most active novel
compound with COX-1 and COX-2 inhibitory activity higher than any of the known NSAIDs in
this study apart from inhibition of COX-2 by celecoxib.
Only the methyl sulfone 1 was shown to be selective for COX-2 compared to COX-1 with an
activity ratio of 17.5 (table 1, entry 13); this observation might be important as 1 could exhibit
relatively low propensity for causing gastric damage.
Table 1. In vitro Anti-inflammatory Activity
COX-1 (M) COX-2 (M) Selectivity COX-2
Entry
Compound
IC50
IC50
(COX-1/COX-2)
1
2
4.89±0.97
2.73±0.44
0.28±0.05
17.47
Fenoprofen*
Naproxen*
26.3±1.67
0.10±0.02
3.36±0.87
0.10
0.4
0
.04±0.001
3
4
11.30±1.08
3.36

7
.02±0.88
3.20
22.04±1.23
0.32
5
1
5
6
7
Ibuprofen
1.50
2.10
2.40
Ibuprofen*
3.35±0.94
1.36±0.12
6.23±0.85
1.40±076
8
9
1.79±0.09
0.76
0.60
Metiazinic acid*
10.45±1.10
0
.39±0.01
0.76±0.02
0.51
1
0
1
1
1
2
Indomethacin*
0.10±0.05
0.90±0.12
0.11
0.47
5
.29±0.78
11.37±1.02
1
1
3
4
0.18±0.07
0.11±0.09
0.71±0.12
0.18±0.07
0.25
0.61
Ketoprofen*
2
1
0
5
1
1
1
5
6
7
Celecoxib
Celecoxib
22.90
6.86
0.0507
0.10
404
69
Celecoxib*
7.10±0.79
0.11±0.02
65

Results are expressed as the mean (n = 3) of the % inhibition of PGH production by test
2
compounds with respect to control samples. In vitro COX-2 selectivity index (IC COX-1 /IC
5
0
50
COX-2). *Test in vitro realized by us in this work.
The comparison of COX-2 binding affinities of methyl sulfones versus carboxylic compounds
revealed that five compounds, 1, 2, 4, 5 and 7, exhibit an improvement in the binding affinity
(Figure 4). However the most surprising result may be the high inhibition of COX-1 showed by
sulfone 2, which was considered the most active of this series. The above-mentioned compound
has a 283-fold higher activity against COX-1 and 34 against COX-2 than naproxen (the parent
compound) and also a higher activity than any of the NSAIDs in this study apart from ketoprofen
and celecoxib (Figure 4). Furthermore, the pyrrole derivative 7 showed major activity against
COX-1 (63 times) and COX-2 (5 times) than the reference naproxen.
In general, for methyl sulfones, the more rigid compounds have a better anti-inflammatory
activity than the more flexible ones (Table 1, compounds 2, 4 and 5 versus 1, 3 and 6), while in
arylacetic and arylpropionic acids it is not so clear (Table 1, entries 4, 9 and 11).
●
1
●
4
●
2
●
3
5
●
●
●
7
6
0
5
10
15
IC50 (Sulfones),
20
25
30
M
Figure 4. Experimental COX-2 binding affinities (IC ) of the sulfones derivatives compared to carboxylate
50
analogues. Sulfone derivatives below the dotted line (in blue) are less active than the corresponding carboxylate
analogues whereas the compounds located upper the dotted line behave oppositely. Compounds that derive from the
substitution of the arylacetic or arylpropionic acid group by a methyl sulfone group in a classical NSAID are colored

in black (compounds 1-5), whereas in grey those sulfone compounds that indirectly derive from ketoprofen
compounds 6 and 7). As can be seen, compound 2 is the most active.
(
In Vivo Anti-inflammatory Activity. The % inhibition of inflammation oral dose was
determined using a carrageenan induced paw edema protocol in rats and the ibuprofen was used
2
1
as a reference compound for the in vivo assays. These compounds were tested to reduce the
formation of edema in a dose of 70 mg/Kg and time-dependent manner. The edema volume was
determined at 3 and 4 h after the carrageenan injection. Compounds 3 and 6, analogues of
ibuprofen and ketoprofen, respectively showed low anti-inflammatory activity (Table 2, entries 3
and 6) attributable to problems of solubility very marked for the sulfone 3 analog of ibuprofen.
Compounds 1, 2, 5 and 7 reduced inflammation between 27.8 and 29.9% at 3 h post-carrageenan
and 25.4 to 34.6% after 4 h. The naproxen analogue 2 and the indomethacin derivative 5
displayed an interesting anti-inflammatory profile, showing 29.4 and 29.9% inhibition by the 3 h
but reached 34.6 and 32.5% inhibition of inflammation at 4 h after carrageenan respectively.
Both compounds demonstrated inhibition of the carrageenan paw edema with sustained activity
after 4 h, which was markedly better than ibuprofen.
Table 2. In vivo Anti-inflammatory Activity
3
h
4 h
Entry
1
Compound
swelling
% inhibition Swelling
% inhibition
25.4
1
a) 28.1 ± 2.5**
b) 38.9 ± 6
27.8
29.4
9.6
a) 34.0 ± 3.9**
b) 45.6±6.3
2
3
4
2
3
4
a) 33.1 ± 3.6 ***
b) 46.9 ± 3.6
a) 35.2 ± 3.5***
b) 53.8 ±3.6
34.6
9.5
a) 32.3 ± 2.8 **
b) 35.7 ± 6.9
a) 38.4 ± 3.2 *
b) 42.5 ± 5.2
a) 45.6 ± 6.3
a) 39.6 ± 5.3 #
NA
NA

b) 38.9 ± 6
b) 44.8 ± 5.4
5
6
7
8
5
a) 32.9 ± 4.1 ***
b) 46.9 ± 3.6
29.9
a) 36.3 ± 5.1***
b) 53.8 ± 3.6
a) 50.0 ± 5.3*
b) 61.7 ± 6.4
a) 45.7 ± 4.5**
b) 61.7 ± 6.4
a) 34.4 ± 6.9 **
b) 45.1 ± 5.0
32.5
18.9
25.9
23.9
6
7
a) 43.2 ± 4.8 *
b) 50.7 ± 5.5
14.9
a) 36.7 ± 6.2**
b) 50.7 ± 5.5
27.6
30.3
Ibuprofen
a) 27.1 ± 3.6***
b) 38.9 ± 6.0
a) Swelling for the dose of 70 mg/Kg. b) Swelling control. NA no activity attributable due to low solubility in
water.
Values significantly differ from controls as indicated: *P<0.05; **P<0.01; ***P<0.001; # does not differ
significantly according to unpaired one-tailed Student's t-test.
The data represent the mean ±S.E.M. of 6 animals (p<0.05). The rat paw edema occurred at 3 and 4 h post-
carrageenan administration.
The methoxynaphthalen of naproxen and the N-acylindole of indomethacin (compounds 2 and 5)
confer greater activity both in vitro and in vivo than the substituted phenyl groups of fenoprofen,
ibuprofen or ketoprofen (compounds 1, 3 and 6). Compound 7 containing the pyrrole ring
showed high activity, much more in vivo than in vitro but inferior to naproxen. The in vivo anti-
inflammatory activity of the phenothiazine 4 could not be determined due to problems of poor
solubility in water.
Analgesic Activity
In general, analgesic drugs relieve symptomatic pain only without affecting its cause. The methyl
sulfones 2 and 7, which presented an interesting anti-inflammatory activity, were selected for the
evaluation of analgesia. The analgesic effect of methyl sulfones 2 and 7 was investigated using
acetic acid induced writhing method (chemical model of nociception). In the present study,
analgesic activity was detected by measuring a decrease in the frequency of writhing and the

analgesic activity was correlated with the ability of the compounds to neutralize the pain
sensation. The percentage inhibition of writhing by an analgesic compound is calculated
according to the following formula:
Average withers in control group - Average withers in treated group ×100
1
00% Inhibition = Average withers in control group
Table 3. In vivo Analgesic Activity
Entry
Compound
Dose
mg/Kg)
0.0
Abdominal
writhing
21.70±1.12
n = 7
% Inhibition
0.00
(
1
2
3
4
5
6
7
8
9
Saline solution
2
10
20
70
10
20
70
10
20
11.00±1.91
n = 4
49.34±8.81
56.25±6.03
60.09±7.00
45.89±7.60
50.49±3.93
64.88±5.43
48.01±5.25
52,98±5.82
2
9.50±1.31
n = 6
2
8.66±1.52
n = 6
7
11.75±1.65
n = 4
7
10.75±0.85
n = 4
7
7.62±1.17
n = 16
Ibuprofen
Ibuprofen
11.28
n = 4
10.20±1.87
n = 6

10
Ibuprofen
70
7.23±1.21
n = 6
66.72±3.44
n = number of mice used for every experiment. A dose of 10, 20 and 70 mg/Kg of every compound was
administered. The abdominal writhing and % inhibition are expressed as mean values ± S.E.M. of at least
four mice. Each test is indicative of statistically significant differences (P<0.05; one way ANOVA followed
by Student-Newman-Keuls test)
Our results revealed significantly reduction of writhing with regards to the control study for 2
and 7, reduction of the frequency of writhing approximately 45 and 64%, respectively (Table 3).
The experimental data evidence obtained in the present study indicates that the methyl sulfones 2
and 7 possess time dependent analgesic activity in the tested experimental animals, with both
compounds able to reduce contractions by 60 % at a dose of 70 mg/Kg.
In vivo gastrointestinal side effects
Gastrointestinal damage by NSAIDs is a commonly observed problem especially for those
people with chronic diseases that must take NSAIDs over longer periods of time. Here, the study
was carried out to determine the effect of methyl sulfones on the gastrointestinal section in rats.
Gastric ulcers were not observed for any of two evaluated compounds and the treated animals
did not suffer from any toxic side effects apart from minor hyperemia associated with treatment
of compound 2 (Table 4, entry 1).
Table 4. In vivo gastrointestinal Activity
Entry
Compound
Rat weight (g)
Volume
Gastric ulcers
administrated
(
mL)
1
2
3
2
2
2
170.0
170.5
155.9
1.70
1.70
1.56
*
0
0

4
5
6
7
8
9
1
7
180.4
181.2
175.2
200.6
185.1
180.3
189.0
1.80
1.81
1.75
2.00
1.85
1.80
1.89
0
0
0
0
0
0
0
7
7
carboxymethylcellulose
carboxymethylcellulose
Saline solution
Saline Solution
0
*
Light hyperemia. A lot of 3 rates for every experiment has been used. A 70 mg/Kg of compound to tester was
administered in suspension on carboxymethylcellulose at 0.5%.
A desirable anti-inflammatory treatment would maintain efficacy and minimize the gastric side
effects. One of the aims of this work was to limit the undesired gastric effects of classical
NSAIDs, theoretically related to the COX-1 inhibition. The decrease in GI effects can be
attributed to the difference in acidity, so while ibuprofen has a pK ~ 4.8, the aryl methyl
a
22
sulfones have pK ~ 28-30 in aqueous solution.
a
Chemosensitivity
To evaluate if the most promising compounds 2, 5 and 7 possessed inherent cytotoxic properties
they were assessed in the human HT29 colon adenocarcinoma cell line using the MTT assay
(
Table 5). No IC values were obtained below 200 µM indicating no antiproliferative activity
5
0
under the conditions investigated. Taken together with the data obtained from the in vivo
gastrointestinal investigation it suggest a promising starting point for further optimization of
novel sulfones possessing good anti-inflammatory activity and lacking off-target toxicity.
Table 5. Growth inhibition of aryl methyl sulfones analogues against HT29 cancer cells
Entry
Compound
HT29 [µM]

a
1
2
3
4
2
>200
5
>200
>200
7
oxaliplatin
0.92 ± 0.25
a
IC values for all compounds are the mean ± SD of at least three independent assays
5
0
2
.3. Molecular Modeling
The carboxylic acids bind more strongly than methyl sulfone analogues while keeping the
same binding mode
Molecular modeling can provide valuable information about the ligand-receptor binding mode
and its associated relative free energies of binding of a series of inhibitor compounds. As
validation for the simulation protocol adopted by the present work, we initially reproduced the
crystallographic binding mode of ibuprofen in the COX-2 receptor (PDB code 4PH9). The
superposition of the crystallographic binding mode of ibuprofen in COX-2 and the final structure
of the corresponding modeled protein-ligand complex of compound 3 in the molecular dynamics
showed satisfactory results (Figure 5).

Heme group
Tyr355
Arg120
Arg513
Figure 5. Superposition of experimental and simulated ibuprofen and its sulfone analogue in the binding
site of mouse COX-2. Secondary structure elements and the bold ball & sticks representation correspond
to crystal structure of ibuprofen bound to COX-2 (pdb code 4PH9). Thin balls & sticks representation
correspond to ibuprofen (red spheres) and its methyl sulfone analogue (compound 3, blue spheres)
Likewise, the replacement of the carboxylate group by the methyl sulfone moiety in the
ibuprofen skeleton also yielded a stable complex structure (Figure 5), whose structural integrity
is maintained through the polar interaction between the sulfone group and both Arg120 and
Tyr355 residues. Therefore, like ibuprofen, its sulfone analogue also blocks the entrance of the
substrate to the catalytic site of COX-2 as Figure 5 shows. However the fluctuation of the
sulfone analogue ligand at the interior of the binding cavity is significantly higher than that of
the carboxylate analogue (see Figure S1). From the energetic point of view, the comparison of
the binding free energies of these two inhibitors also confirmed the significantly lower affinity of
the sulfone analogue 3 compared with its carboxylic analogue by a difference of 9.6 kcal/mol
according to MM-PBSA, 5.0 kcal/mol according to MM-GBSA and 15.4 computed with MM-

3
DRISM (entry 7 and 8 in Table 6 and Table S1). Experimentally, the same trend was observed
with a 15-fold decrease in affinity for ibuprofen (IC = 1.5 M, entry 7 in Table 2) when
5
0
compared with its sulfone analogue 3 (IC = 22.05 M, entry 5 in Table 2). In other words, if
5
0
the binding mode is maintained when comparing carboxylate and sulfone analogues, a decrease
of affinity is expected since the sulfone group acts as a weaker binder (less polar) to Arg120 at
the entrance of the binding site compared to the carboxylate group. Similarly the replacement of
the carboxylic acid group by the methyl sulfone in a bulky 1,5-biaryl pyrrole EP1 receptor
2
3
antagonist significantly reduced the binding affinity, in line with the present results.
Table 6. Experimental and calculated binding free energies (kcal/mol) at different binding modes for all
studied sulfone derivatives and reference inhibitors ibuprofen and celecoxib
Binding
mode
a
c
Entry
1
Compound
ΔGbind,exp
MD simulations
ΔGbind,MM-PBSA
b
2
-9.5
1
2
3
1
2 x 120 ns
2 x 60 ns
2 x 60 ns
1 x 120 ns
-14.7 ± 2.7
-13.6 ± 2.2
-11.8 ± 2.8
-15.8 ± 2.5
2
3
1
7
-8.9
-8.4
2
3
1
2
3
1
1
2
1 x 60 ns
1 x 60 ns
1 x 60 ns
1 x 60 ns
1 x 60 ns
1 x 60 ns
1 x 60 ns
1 x 60 ns
-15.6 ± 2.4
-12.3 ± 2.8
-12.9 ± 2.8
-14.9 ± 2.8
-11.0 ± 2.7
-22.5 ± 3.0
-17.6 ± 2.9
-15.7 ± 2.3
4
5
6
5
4
-8.3
-7.8

7
6
3
-6.7
-6.3
-
1
1
1
2
1 x 60 ns
1 x 60 ns
1 x 60 ns
1 x 60 ns
-15.7 ± 3.0
-13.5 ± 2.4
-23.1 ± 3.1
-24.9 ± 3.0
8
9
Ibuprofen
Celecoxib
10
-
a
b)
Computed using ΔGbind,exp = − R·T·ln(1/IC ), Mode 1 = ibuprofen-like binding mode, PDB code 4PH9,
50
Mode 2 = celecoxib-like binding mode, PDB code 3LN1 and mode 3 = diclofenac-like binding modem
c
PDB code 1PXX. Number of simulations x extension of the simulation.
Molecular modeling of the COX-2 inhibitory action of Methyl Sulfones analogues
The comparison of the experimental versus simulated binding free energies for all sulfone
analogues is displayed in Figure 6 for MM-PBSA (Table S1, Figure S2 and S3 for the
corresponding MM-GBSA and MM-3DRISM results, respectively). Two different groups are
differentiated and confirmed in all three methodologies. The binding free energies of less active
inhibitors, compounds 3, 5 and 6 (analogues of ibuprofen, indomethacin and ketoprofen,
respectively), based on the ibuprofen-like binding mode correlated with experimental binding
affinity whereas, on the contrary, no correlation was found for the most active inhibitors,
compounds 1, 2, 4 and 7 (analogues of fenoprofen, naproxen, metiazinic acid and ketoprofen).
Compounds 3, 5 and 6 (hereafter named as correlated compounds) would adopt the ibuprofen-
like binding mode (mode 1) like the corresponding parent compounds ibuprofen (PDB code
4
PH9) and indomethacin (PDB code 4COX. This mode is characterized by hydrogen bonding
interactions of the oxygen atoms of the sulfone group with Arg120 and Tyr355 and also
transitory hydrogen bond contacts between hydroxyl group of Ser530 and the ketone group of
compounds 5 and 6. Furthermore, the ibuprofen-like binding mode involves favorable
hydrophobic contacts with Ala527, Val523, Val349, Leu352, Ser353, Ala527 and Leu531 at the
entrance of the binding channel for all these inhibitors and with Phe518 at the rear part of the
binding site as shown by the per-residue decomposition of the binding free energy in Table S2

and the representation of the ibuprofen-like binding mode in Figure 7. The higher affinity of
compound 5 over 3 and 6 can be attributed to the most favorable hydrophobic interaction energy
(see Evdw term in Table S3) at the entrance of the cavity for the most bulky compound 5
compared to the smaller ones (3 and 6). It is also important to note that the role of the bridged
water molecule between Ser530 and Tyr385 (see Figure 7) at the rear part of the cavity provides
remarkable structural stability to the binding site. This highly stable water molecule with long
lifetimes is also present in the crystal structure of ibuprofen bound to COX-2 (PDB code
2
4
4
PH9).
●
Mode 1, ibuprofen−like
Mode 2, celecoxib−like
●
3
●
●
2
7
●
6
●
1
●
4
5
●
−10
−9
−8
bind (Experimental), kcal/mol
−7
−6
G
Figure 6. Correlation between COX-2 experimental and simulated MM-PBSA binding free energies
ΔGbind) for all methyl sulfone derivative inhibitors. Black circles correspond to the binding free energies
simulated at binding mode 1 (ibuprofen-like) whereas the empty blue squares represent binding mode 2
celecoxib-like). Compounds 3, 5 and 6 correlate to the experimental values and compounds 1, 2, 4 and 7
(
(
identify the uncorrelated compounds.
The uncorrelated binding free energies computed at the binding mode 1 (ibuprofen-like) for the
most active inhibitors prompted us to explore other binding modes for these compounds (1, 2, 4

and 7, hereafter named as uncorrelated compounds), like that found for celecoxib (pdb code
3
LN1, mode 2) and diclofenac (pdb code 1PXX, mode 3).
Leu352
Val349
Ser353
Tyr385
Phe518
Ser530
Leu531
Tyr355
Val523
Ala527
Arg120
Arg513
Figure 7. Superposition of sulfone analogue structures that its calculated binding free energies correlate
to the experimental values in the binding mode 1 (ibuprofen-like), compounds 3, 5 and 6. Ligands are
shown in balls & sticks with stick and carbon atom colors indicating each compound (3/marine blue,
5
/yellow and 6/orange). Compound 4 (purple) is also added to the representation. The 10 most important
residues in the stabilization of such binding mode are shown in green sticks see also Table S2.
Additionally Ser530, Tyr385 and Arg513 are also shown. Most important hydrogen bonds are also
indicated with a dotted line.
Celecoxib-like binding mode is characterized by the polar interaction between the sulfone group
and Arg513 instead of Arg120 and also hydrophobic contacts with Phe518, Ile517 and Ala516 in
addition to the same hydrophobic residues listed for ibuprofen-like binding mode at the entrance
of the binding channel except Leu531 which is located far away from the cavity (see Figure 8
and Table S4). Likewise, transitory hydrogen bonds between the ketone group of compound 2
and Ser530 are observed. Meanwhile, the diclofenac-like binding mode orients the sulfone group
towards Ser530 at the rear part of the binding site. Even though any of the new explored binding
modes provided a significant enhancement of the binding affinity, celecoxib-like binding mode
emerges as a competitive binding mode to ibuprofen-like binding mode for the four uncorrelated

compounds. The computed energy difference between both binding modes ranges only from -0.2
to -2.0 kcal/mol computed at MM-PBSA level for these inhibitors, which corroborate its similar
stability. It is remarkable that three out of four of these compounds (1, 2 and 4) are precisely
those that the substitution of the carboxylate for a sulfone group causes an increment of their
activity, unlike to the correlated compounds (3 and 6), see Figure 4. On the other hand, notice
that compounds 5 and 7 do not alter their activities compared to the carboxylate analogue
compounds. In the case of the bulky indomethacin (compound 5) the conservation of the activity
after the substitution of the acidic group by a methyl sulfone would indicate the prominence of
the hydrophobic skeleton in the magnitude of the binding affinity. Taken into account that this
substitution, while keeping the same binding mode, yields a reduction in affinity (as seen in the
ibuprofen example above), it is a clear indication that, indeed, these uncorrelated sulfone
compounds change their binding mode or binding mechanism in full agreement with the
simulated data presented here. However, unfortunately the new explored binding modes are not
able to reproduce the final ranking, which could be indicative of the drawbacks of the present
2
5
computational methodologies to estimate accurate binding free energies or more probably that
other mechanisms are in place for these tight-binding sulfone compounds. One well-known but
poorly explained alternative mechanism for tight binding inhibitors that could explain the
behavior of the top binders consists of a time-dependence mechanism of binding. This
2
6
mechanism has been described for celecoxib in COX-2 and for flurbiprofen in COX-1, which
2
7
seems to involve the stabilization of a competing conformational substrate of the enzyme.

Leu352
Val349
Ile517
Ser353
Tyr385
Phe518
Ala516
Tyr355
Ser530
Arg513
Ala527
Val523
Arg120
Figure 8. Superposition of sulfone analogue structures in the binding mode 2 (celecoxib-like) for
compounds 1, 2 and 7. Ligands are shown in balls & sticks with stick and carbon atom color indicating
each compound (1/orange, 2/brown and 7/deep blue). The most important residues in the stabilization of
such binding mode are shown in green sticks. Additionally Ser530, Tyr385 and Arg120 are also shown.
Most important hydrogen bonds are also indicated with a dotted line.
To sum up, the ibuprofen-like binding mode is expected for inhibitors that are less active than
the carboxylate analogues whereas a different mechanism or binding mode is predicted for more
active inhibitors than the carboxylate analogues. So, on the one hand, for the less active sulfone
compounds the ibuprofen-like binding mode is maintained, and as a consequence of it, the
hydrophobic interactions at the entrance of the binding channel dominate and determine the
ranking of affinities of the methyl sulfone analogues. On the other hand, the unexpected high
affinity of the most active sulfone analogues might be induced by a time dependent character of
the protein-ligand binding.
3
. Conclusion
We have synthesized and evaluated a series of methyl sulfones possessing anti-inflammatory
properties. Results showed that the tested compounds (1-7) exhibited promising anti-

inflammatory activity compared to the parent drugs, with marked decreases in the ulcerogenic
side effects. The peripheral analgesic effect of these methyl sulfones may also be mediated via
inhibition of cyclooxygenase isoforms. Compounds 2 and 7 exert significant anti-inflammatory
and anti-nociceptive effects in tested mice while they are not toxic at high doses. These results
justify the study of methyl sulfones as a new class of NSAIDs that exhibit an interesting anti-
inflammatory activity profile. Finally, the results of the in vivo anti-inflammatory activity by the
standard acute carrageenan-induced paw edema method in rats, revealed remarkable activity for
compounds 1, 2, 5 and 7. Molecular modeling studies found a significant decrease of the COX-2
binding affinities of the sulfone derivatives compared to their carboxylate analogues while
keeping the same binding mode, as a consequence of that, the most active compounds (1, 2, 4
and 7) would accomplish their inhibitory activities through a different binding site or mechanism
than the less active compounds (3, 5 and 6) whose binding site would correspond to that found
for ibuprofen (4PH9, polar moiety of the inhibitor compound oriented to the mouth of the
binding cavity). Such increment of the inhibitory activity would be performed through a change
of the binding mode or mechanism compared to the standard ibuprofen binding site.
Unfortunately, the inhibitor structure does not provide any evidence of rationalization of such
behavior.
Compound 2 the most relevant methyl sulfone of these series exhibited superior inhibition of
COX-1 (IC = 0.04 M) and COX-2 (IC = 0.10 M) compared with that of naproxen (COX-1
5
0
50
(
IC = 11.30 M) and COX-2 (IC = 3.36 M) in vitro and better reduction of inflammation
5
0
50
than ibuprofen in vivo. It also demonstrated a good oral absorption and long duration of anti-
inflammatory effect (34.6% of inhibition after 4 h of administration of a 70 mg/Kg dose in rat).
No detectable toxicity against HT29 human cells, interesting analgesic activity and non
ulcerogenic effect under the tested conditions were other properties of compound 2, which
showed a promising pharmacological profile. The results reveal that the replacement of

carboxylic acid by methyl sulfone in classical NSAIDs may be a promising approach to design
new COX inhibitors with anti-inflammatory and analgesic properties and reduced
gastrointestinal side effects.
4
. Experimental Section
4
.1. Chemistry
Microwave assisted reactions were carried using a CEM Discover LabMate instrument and the
temperature vessel was measured by an IR sensor. The reactions were monitored by thin-layer
chromatography (TLC) analysis using silica gel (60 F254, Merck) plates. Compounds were
visualized by UV irradiation. Column chromatography was performed with silica gel 60 (230-
4
00 mesh, 0.040-0.063 mm) and automatic column chromatography was performed with a
CombiFlash R system with UV-vis (PN 68-5230-008) detector and RediSep R 4 and 12 g silica
f
f
gel column. Melting points (mp) were obtained on a MFB-595010M Gallenkamp apparatus with
digital thermometer in open capillary tubes and are reported without correction. IR spectra were
1
13
obtained using FTIR Perkin-Elmer 1600 Infrared Spectrophotometer. H and C NMR spectra
were recorded on a Varian Gemini-400 (100 MHz). Chemical shifts are reported in parts per
million (ppm) relative to the central peak of the solvent: CDCl (, 7.26 (H) and 77.00 (C)),
3
CD OD (, 3.31 (H) and 49.45 (C)), DMSO-d (, 2.49 (H) and 39.51 (C)) as internal standards.
3
6
The following abbreviations are used for the proton spectra multiplicities: s, singlet, d, doublet, t,
triplet, q, quadruplet, m, multiplet. Coupling constants (J) are reported in Hertz (Hz).
The purities of all isolated compounds were determined by electrospray ionization (ESI-HMRS)-
mass spectra on a LC/MSD-TOF (2006) (Agilent technologies) by the «Section of spectrometry
of masses » at the University of Barcelona. The purity of the compounds proved to be >95%. All
reagents were of high quality or were purified before use. Organic solvents were of analytical
grade or were purified by standard procedures.

4
.2. 3-Phenoxy-1-methylsulfonylbenzene (1). Aryl halide 9 (0.080 g, 0.34 mmol), phenol (0.048
g, 0.5 mmol), Cs CO (0.300 g, 0.6 mmol), ACNH and CuBr in catalyst amount (0.1%) in DMF
2
3
(
10 mL) were added to the microwave tube. The mixture was reacted in a microwave oven
CEM Discover LabMate) at 100 ºC (external temperature) for 1 hour and was monitored by
(
TLC. After completion of the reaction, it was cooled to room temperature; H O (10 mL) was
2
added and then the crude of reaction was extracted with ethyl ether (3 x 15 mL). The combined
organic extracts were washed with water (3 x 15 mL) and dried over Na SO . After removal of
2
4
the solvent, the residue was purified by automatic column chromatography CombiFlash R with
f
UV-vis detector using 12 g silica gel column and eluting with a stepwise gradient
(Hexane:EtOAc, 1:0  0:1) to give a yellow oil. 98% (83 mg) yield; IR (ATR diamond, cm-1)
1
υ: 1447, 1301, 1231, 1142, 753, 683; H NMR (400 MHz, CDCl ) δ: 3.04 (s, 3H, CH ), 7.04 (d,
3
3
J = 8.4Hz, 2H, H-2’, H-6’), 7.20 (t, J = 7.6Hz, 1H, H-4’), 7.25 (d, J = 8Hz, 1H, H-4), 7.40 (t, J =
8
6
.4Hz, 2H, H-3’, H-5’), 7.52 (t, J = 8.4Hz, 1H, H-5), 7.53 (s, 1H, H-2), 7.64 (d, J = 8Hz, 1H, H-
1
3
); C NMR (100 MHz, CDCl ) δ: 44.7 (CH ), 116.9 (CH), 120.0 (2CH), 121.7 (CH), 123.6
3
3
(CH), 125.5 (CH), 130.5 (2CH), 131.2 (CH), 142.4 (Cq), 155.9 (Cq), 158.8 (Cq); HRMS (ESI):
+
calcd for C H O S [M+H] : 249.0585 found 249.0592.
1
3
13
3
4
.3. General procedure of oxidation with m-CPBA
To a solution of the sulfone (1 mmol) in dichloromethane (20-25 mL) was added m-CPBA (2.5
mmol) at 0 ºC under argon atmosphere. The crude mixture was stirred at rt for 2 h before it was
basified with 2N NaOH and extracted with dichloromethane (3x20 mL). Organic layers were
dried (Na SO ), filtered and concentrated under vacuum.
2
4

4
.4. Methylsulfonyl-6-methoxynaphthalene (2). Following general procedure of oxidation with
m-CPBA, the sulfone 2 was obtained from 11 as a yellow solid with 87% (227 mg) yield; mp:
1
1
32-134 ºC (dichloromethane); IR (ATR diamond, cm-1) υ: 1594, 1259, 1216, 1145; H NMR
(
400 MHz, CDCl ) δ: 3.11 (s, 3H, SCH ), 3.97 (s, 3H, OCH ), 7.20 (s, 1H, H-5), 7.28 (d, J = 9
3 3 3
1
3
Hz, 1H, H-7), 7.87–7.90 (m, 3H, H-3, H-4, H-8), 8.43 (s, 1H, H-1); C NMR (100 MHz, CDCl )
3
δ: 45.0(CH ), 55.9 (CH ), 106.2 (CH), 121.1 (CH), 123.2 (CH), 127.8 (Cq), 128.5 (CH), 129.1
3
3
(CH), 131.2 (CH), 135.3 (Cq), 137.4 (Cq), 160.5 (Cq); HRMS (ESI): calcd for C H O S
12 13 3
+
[
M+H] : 237.0585 found 237.0573.
4
.5. 4-Isobutylphenyl methyl sulfone (3). A mixture of zinc (3.76 g, 57.51 mmol), mercury (II)
chloride (0.38 g, 1.44 mmol) and HCl concd (1 mL) in H O (3 mL) was stirred at room
2
temperature for 5 min. Then, the crude mixture was filtered and the solid residue was treated
with H O (0.5 mL), HCl concd (1.3 mL), toluene (1 mL) and the ketone 15 (2.9 g, 12.82 mmol).
2
The reaction mixture was heated under reflux of water during 5 h. Then, the mixture was diluted
with water (20 mL) and extracted with diethyl ether (3× 20 mL). The combined organic layers
were washed with H O (20 mL) and dried with Na SO . The drying agent was filtered off and
2
2
4
the filtrate was evaporated to dryness to afford a brown oil. The crude reaction was purified by
automatic column chromatography CombiFlash R with UV-vis detector using 12 g silica gel
f
column and eluting with stepwise gradient (Hexane:EtOAc, 1:0  0:1) to give a yellow oil. 22%
1
(
260 mg) yield; IR (ATR diamond, cm-1) υ: 1596, 1257; H NMR (400 MHz, CDCl ) δ: 1.08 (d,
3
J = 6 Hz, 6H, CH (x2)), 2.02 (sept, J = 6 Hz, 1H, CH-((CH ) ), 2.61 (d, J = 6 Hz, 2H, CH -Ar),
3
3 2
2
1
3
2
.65 (s, 3H, SO CH3), 7.24 (d, J = 9 Hz, 2H, H-3,H-5), 7.38 (d, J = 9 Hz, 2H, H-2,H-6); C
2
NMR (100 MHz, CDCl ) δ: 16.7 (2CH ), 22.7 (CH ), 30.1 (CH), 45.2 (CH ), 127.3 (2CH), 130.0
3
3
3
2
+
(
2CH), 135.3 (Cq), 139.2 (Cq); HRMS (ESI): calcd for C H O S [M+H] : 213.0949 found
11 17 2
2
13.0954.

4
.6. 10-Methyl-2-methylsulfonyl-10H-phenothiazine (4). To a stirred solution of methyl sulfone
8 (0.096 g, 0.35 mmol) in 7 mL of DMF cooled at 0 °C was added a solution of NaH 60%
1
(0.014 g, 0.35 mmol) in 3 mL of DMF. Next, iodomethane (0.024 mL, 0.39 mmol) was added to
the crude mixture and stirred at room temperature for 24 h before the crude reaction was diluted
with ice-cold H O and extracted with diethyl ether and H O (2:1). The organic layers were dried
2
2
with Na SO , filtered off and concentrated under reduced pressure. The crude reaction was
2
4
purified by automatic column chromatography CombiFlash R with UV-vis detector using 12 g
f
silica gel column and eluting with stepwise gradient (Hexane:EtOAc, 1:0  0.8:0.2) to give a
green solid with 71% (72 mg) yield; mp: 130-132 ºC (hexane/ethyl acetate); IR (ATR diamond,
1
cm-1) υ: 1562, 1292; H NMR (400 MHz, CDCl ) δ: 2.99 (s, 3H, NCH ), 3.38 (s, 3H, SO CH ),
3
3
2
3
6
.83 (d, J = 6 Hz, 2H, H-4, H-6), 6.97 (t, J =6 Hz, 1H, H-7), 7.09 (d, J = 6 Hz, 1H, H-9), 7.18 (t,
1
3
J = 6 Hz, 1H, H-8), 7.60 (s, 1H, H-1), 7.67 (d, J = 6 Hz, 1H, H-3); C NMR (100 MHz, CDCl )
3
δ: 35.8 (CH ), 44.8 (CH ), 113.7 (CH), 114.9 (CH), 122.1 (CH), 123.7 (Cq), 124.7 (CH), 125.9
3
3
(Cq), 127.28 (CH), 127.30 (CH), 128.0 (CH), 133.6 (Cq), 144.1 (Cq), 150.4 (Cq); HRMS (ESI):
+
calcd for C H NO S [M+H] : 292.0466 found 292.0458.
1
4
14
2 2
4
.7. 2-Methyl-3-sulfonyl-5-methoxy-1-(p-chlorobenzoyl)indole (5). To a stirred solution of
methyl sulfone 22 (0.099 g, 0.41 mmol) in 7 mL of ice-cold DMF was added a solution of NaH
0% (0.024 g, 0.60 mmol) in 3 mL of DMF. After 30 min of stirring at 0 °C, 4-chlorobenzoyl
6
chloride (0.053 mL, 0.41 mmol) was added and the reaction was allowed to warm up to room
temperature and stirred for 24 h. Next, the solution was diluted with H O (20 mL) and extracted
2
with diethyl ether (3x20 mL). The combined organic layers were washed with H O (20 mL) and
2
dried with Na SO . The drying agent was filtered off and the filtrate was evaporated to dryness.
2
4
The crude reaction was purified by automatic column chromatography CombiFlash R with UV-
f