Novel analgesic/anti-inflammatory agents: 1,5-Diarylpyrrole nitrooxyalkyl ethers and related compounds as cyclooxygenase-2 inhibiting nitric oxide donors

Article abstract of DOI:10.1021/jm301370e

A series of 3-substituted 1,5-diarylpyrroles bearing a nitrooxyalkyl side chain linked to different spacers were designed. New classes of pyrrole-derived nitrooxyalkyl inverse esters, carbonates, and ethers (7-10) as COX-2 selective inhibitors and NO donors were synthesized and are herein reported. By taking into account the metabolic conversion of nitrooxyalkyl ethers (9, 10) into corresponding alcohols, derivatives 17 and 18 were also studied. Nitrooxy derivatives showed NO-dependent vasorelaxing properties, while most of the compounds proved to be very potent and selective COX-2 inhibitors in in vitro experimental models. Further in vivo studies on compounds 9a,c and 17a highlighted good anti-inflammatory and antinociceptive activities. Compound 9c was able to inhibit glycosaminoglycan (GAG) release induced by interleukin-1β (IL-1β), showing cartilage protective properties. Finally, molecular modeling and ¹H- and ¹³C-NMR studies performed on compounds 6c,d, 9c, and 10b allowed the right conformation of nitrooxyalkyl ester and ether side chain of these molecules within the COX-2 active site to be assessed.

Full text of DOI:10.1021/jm301370e

Subscriber access provided by NORTH CAROLINA STATE UNIV
Article
Novel Analgesic/Anti-inflammatory Agents: 1,5-
Diarylpyrrole Nitro-oxyalkyl Ethers and Related Compounds
as Cyclooxygenase-2 Inhibiting Nitric Oxide Donors
Maurizio Anzini, Angela Di Capua, Salvatore Valenti, Simone Brogi, Michele Rovini, Germano
Giuliani, Andrea Cappelli, Salvatore Vomero, Luisa Chiasserini, Alessandro Sega, Giovanna
Poce, Gianluca Giorgi, Vincenzo Calderone, Alma Martelli, Lara Testai, Lidia Sautebin,
Antonietta Rossi, simona pace, Carla Ghelardini, Lorenzo Di Cesare Mannelli, Veronica Benetti,
Antonio Giordani, Paola Anzellotti, Melania Dovizio, Paola Patrignani, and Mariangela Biava
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm301370e • Publication Date (Web): 27 Mar 2013
Downloaded from http://pubs.acs.org on April 1, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Neuroscienze ed Imaging
Dovizio, Melania; University of Chieti,
Patrignani, Paola; Università degli Studi di Chieti, Dipartimento di
Neuroscienze ed Imaging
Biava, Mariangela; Università di Roma "La Sapienza", Chimica e Tecnologie
del Farmaco
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Novel Analgesic/Antiꢀinflammatory Agents: 1,5ꢀ
Diarylpyrrole Nitroꢀoxyalkyl Ethers and Related
Compounds as Cyclooxygenaseꢀ2 Inhibiting Nitric
Oxide Donors
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a*
a
a
a
a
Maurizio Anzini, Angela Di Capua, Salvatore Valenti, Simone Brogi, Michele Rovini,
a
a
a
a
a
Germano Giuliani, Andrea Cappelli, Salvatore Vomero, Luisa Chiasserini, Alessandro Sega,
b
c
d
d
d
Giovanna Poce, Gianluca Giorgi, Vincenzo Calderone, Alma Martelli, Lara Testai, Lidia
e
e
e
f
f
Sautebin, Antonietta Rossi, Simona Pace, Carla Ghelardini, Lorenzo Di Cesare Mannelli,
g
h
i
i
i
Veronica Benetti, Antonio Giordani, Paola Anzellotti, Melania Dovizio, Paola Patrignani,
b
and Mariangela Biava
a
Dipartimento Farmaco Chimico Tecnologico, Università degli Studi di Siena, Via A. De Gasperi
b
2
, Iꢀ53100 Siena, Italy; Dipartimento di Studi di Chimica e Tecnologie del Farmaco, Università
c
degli Studi di Roma "La Sapienza”, Piazzale Aldo Moro 5, Iꢀ00185 Roma, Italy; Dipartimento
di Chimica, Università degli Studi di Siena, Via A. De Gasperi 2, Iꢀ53100 Siena, Italy;
d
Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Università degli
e
Studi di Pisa, Via Bonanno 6, Iꢀ56126 Pisa, Italy; Dipartimento di Farmacologia Sperimentale,
Università degli Studi di Napoli “Federico II”, Via D. Montesano 49, Iꢀ80131 Napoli, Italy;
f
Dipartimento di Farmacologia, Università degli Studi di Firenze, Viale G. Pieraccini 6, Iꢀ50139
ACS Paragon Plus Environment

Page 3 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
g
h
Firenze, Italy; Innovative Biotech, Via di Oratoio, 56121 Pisa, Italy; Rottapharm Madaus , Via
i
Valosa di Sopra 7, Iꢀ20052 Monza, Italy; Dipartimento di Neuroscienze ed Imaging, Università
degli Studi di Chieti, “G. d’Annunzio” and CeSI, Via dei Vestini 31, Iꢀ66100 Chieti, Italy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS CINODs; COX-2 inhibitors; Anti-inflammatory agents; Analgesic agents; Pyrrole
derivatives.
ABSTRACT A series of 3ꢀsubstituted 1,5ꢀdiarylpyrroles bearing a nitroꢀoxyalkyl side chain
linked to different spacers were designed. New classes of pyrroleꢀderived nitroꢀoxyalkylꢀinverse
esters, ꢀcarbonate and ꢀethers (7-10), as COXꢀ2 selective inhibitors and NO donors were
synthesized and are herein reported. By taking into account the metabolic conversion of nitroꢀ
oxyalkyl ethers (9,10) into corresponding alcohols, derivatives 17 and 18 were also studied.
Nitroꢀoxy derivatives showed NOꢀdependent vasorelaxing properties while most of the
compounds proved to be very potent and selective COXꢀ2 inhibitors in in vitro experimental
models. Further in vivo studies on compounds 9a,c and 17a, highlighted good antiꢀinflammatory
and antiꢀnociceptive activities. Compound 9c was able to inhibit glycosaminoglycans (GAG)
release induced by interleukinꢀ1β (ILꢀ1β), showing cartilage protective properties. Finally,
1
13
molecular modeling, Hꢀ and CꢀNMR studies performed on compounds 6c,d, 9c and 10b
allowed the right conformation of nitroꢀoxyalkyl ester and ꢀether side chain of these molecules
within the COXꢀ2 active site to be assessed.
Introduction
Traditional nonꢀsteroidal antiꢀinflammatory drugs (tNSAIDs) represent the most widely
prescribed, efficacious and costꢀeffective pharmacological treatment of rheumatologic and
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
inflammatory disorders. In addition, this class of drugs is widely used to treat mild to moderate
2
pain. Their therapeutic effects are mediated by the inhibition of cyclooxygenase (COX)ꢀ2 which
is the most important COXꢀisoform contributing to prostanoid generation at inflammatory sites
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
and spinal cord. Selective COXꢀ2 inhibitors (coxibs) have been developed with the aim to
produce an efficacy comparable to tNSAIDs while reducing gastrointestinal (GI) related adverse
events which are mainly due to the inhibition of COXꢀ1ꢀderived cytoprotective prostanoids in the
4,5
GI tract. However, the use of tNSAIDs and coxibs is associated with an increased risk of
6
thrombotic and renal adverse events. This is due to their inhibitory effects on the biosynthesis of
vascular prostacyclin (PGI2), a powerful platelet inhibitor and vasodilator, which is mainly
7
derived from the haemodynamic shear induced by COXꢀ2. Recently, Yu et al showed that
vascular COXꢀ2 deletion ends in the reduction of the expression of endothelial nitric oxide (NO)
8
synthase and consequent release of NO. Similarly to prostacyclin, NO is endowed with
9
important cardioprotective properties, such as vasodilation and inhibition of platelet function.
Suppression of PGI2 formation due to inhibition of vascular COXꢀ2 is sufficient to account for
1
0
the cardiovascular (CV) hazard from NSAIDs (traditional and selective for COXꢀ2), but it may
8
be increased by secondary mechanisms such as suppression of NO production.
These findings justify the need to develop NSAIDs endowed with NOꢀreleasing properties in
order to mitigate their CV, and GI, hazard. Indeed, the novel class of antiꢀinflammatory agents,
named COXꢀinhibiting nitric oxideꢀdonors (CINODs), developed by linking a NOꢀreleasing
moiety to a tNSAID, has been shown to have a more favourable clinical profile than the parent
11,12
tNSAIDs in randomized clinical trials.
This is plausibly due to the property of CINODs to
1
3
release NO which may replace the functions of inhibited prostanoids by the tNSAIDs.
ACS Paragon Plus Environment

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Naproxcinod, 4ꢀ(nitroxy)butylꢀ(S)ꢀ2ꢀ(6ꢀmethoxyꢀ2ꢀnaphtyl)ꢀpropanoate is the first drug of the
1
4
CINODs the class of which is in advanced state of development. In randomized clinical trials,
naproxcinod showed an improved safety profile with respect to blood pressure (BP) and GI tract
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15,16
compared to naproxen.
Naproxcinod is the ester of naproxen with 4ꢀ(nitroxy)butanol
(
NOBA) and undergoes hydrolysis, mainly in the GI tract, leading to the release of naproxen and
the organic nitrate NOBA. The analgesic and antiꢀinflammatory activities of naproxcinod are
dependent on released naproxen while NOBA is the main source of NO (for more details, see
Supporting Information, SI). A common pathway for NO generation among naproxcinod, NOꢀ
1
7,18
flurbiprofen (3, a naproxcinod congener) and glyceryltrinitrate (GTN) was recognized.
NOBA, as well as GTN, is a relatively fast NOꢀreleasing compound and the CV effects are
detected for a maximum of 3 hours, after the administration with naproxcinod. On the contrary,
the analgesic and antiꢀinflammatory effects of naproxcinod are more extended because they
depend on the pharmacokinetics of naproxen which is characterized by a long halfꢀlife (i.e. 17
1
9
hrs).
Our research was focused on the development of a new class of coxibs based on the
2
0ꢀ22
diarylpyrrole scaffold
and subsequently to the search of NOꢀreleasing compounds endowed
with strong analgesic and antiꢀinflammatory properties. For the sake of completeness, here we
report the synthesis and the biological evaluation of a series of 1,5ꢀdiarylpyrrole nitroꢀoxyalkyl
congeners (7-10) as a further transformation of the very recently reported nitroꢀoxyalkyl esters
2
3,24
(6).
In the title compounds different linkers between the 1,5ꢀdiarylpyrrole scaffold and the
NO releasing moiety were introduced. Thus, O NOꢀalkylꢀ inverse ester (7), ꢀcarbonate (8), and ꢀ
2
ethers (9ꢀ10) were prepared and evaluated in vitro and in vivo in order to disclose new chemical
entities useful in the longꢀterm treatment of osteoarthritis (OA) with reduced GI and mainly CV
adverse effects (Chart 2). In particular, nitroꢀoxyalkyl ethers 9 and 10 were designed with the aim
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to prevent the hydrolysis of the sideꢀchain, leading to compounds with the inhibitory activity
towards COXꢀ2 and NO releasing property in the same molecule. Conversely, with regards to the
series of nitroꢀoxyalkyl esters 6, inhibitory activity towards COXꢀ2 and the NOꢀreleasing
property were retained in two different parts of the molecule. These esters, however, showed
almost the same liability to esterase previously described for naproxcinod and NOꢀflurbiprofen.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chart 1. Structures of Reference Compounds.
Compd 6a 6b 6c 6d 6e 6f
X
n
H
2
H 3ꢀF 3ꢀF 4ꢀF 4ꢀF
3
2
3
2
3
ACS Paragon Plus Environment

Page 7 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chart 2. Structures of 1,5ꢀDiarylpyrrole Nitroꢀoxyalkyl Congeners 7-10.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compd 9a 9b 9c
9d
10a 10b 10c
3ꢀF 4ꢀF
X
H 3ꢀF 4ꢀF 3,4ꢀF2
H
Chemistry
As is sketched in Scheme 1, hydroxyethyl derivative 11a was the common starting material for
the synthesis of most of the title compounds and it was prepared in gram scale following the
2
0,21
previouslyꢀreported procedure.
The condensation of alcohol 11a with 4ꢀnitroxy butanoic acid
(
see SI) in the presence of EDC and DMAP gave nitroꢀoxyalkyl inverse ester 7. A different
activation of 11a was requested for the synthesis of compounds 8 and 13. So, compound 11a was
condensed with CDI (1,1’ꢀcarbonyldiimidazole) to give imidazo carboxylate 14 which by
reaction with hydroxypropan nitrate (see SI) in dry pyridine and in the presence of 1,8ꢀ
diazabicyclo[5.4.0]undecꢀ7ꢀene (DBU) afforded nitroꢀoxyalkyl carbonate 8. The transformation
of 11a into the corresponding mesyl chloride 12 and further reaction with (Bu N)NO in toluene
4
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
at reflux gave compound 13. This can be regarded as the simplest among the nitroesters
designed. The synthesis of nitroꢀoxyalkyl ethers 9a-d and 10a-c (Scheme 2) started with the
alkylation of alcohols 11a-d with the proper hydroxyalkyl bromide in its tetrahydropyranyl
protected form (see SI) to obtain compounds 15a-d and 16a-c which were successively
deprotected to give hydroxyalkyl ethers 17a-d and 18a-c. These ethers were then mesylated to
give 19a-d and 20a-c and transformed into the expected nitroꢀoxyalkyl ethers 9a-d and 10a-c
following the same procedure used for the preparation of compound 13.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of Compounds 7, 8, and 13.
Reagents and conditions: (i) MsCl, DMAP, DIPEA, CH Cl , 3h, r.t.; (ii) (Bu N)NO , toluene,
2
2
4
3
reflux; (iii) EDC, DIMAP, CH Cl r.t., 3h; (iv) CDI, dry Pyr, r.t., 2 h; (v) dry Pyr, DBU, r.t., 3 h.
2
2
ACS Paragon Plus Environment

Page 9 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 2. Synthesis of Compounds 9-11, and 15-20.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (i) (Bu N)Br, NaOH 50%, 70°C, 20h; (ii) PPTS, MeOH, 55 °C, 20h;
4
(iii) MsCl, DMAP, DIPEA, r. t., 3h; (iv) (Bu N)NO , toluene, reflux, 1h.
4
3
Results and Discussion
An in vitro cell culture (J774 murine macrophage) assay was performed to evaluate the title
compounds’ inhibitory potency and selectivity on both COXꢀisoforms. Results showed that, in
2
2ꢀ24
very recently reported 1,5ꢀdiarylpyrrole derivatives 6,
ester moiety into a nitroꢀoxyalkyl ether group surprisingly led to very active compounds (9 and
0). Some of these were endowed with interesting NOꢀdonating properties along with good and
the transformation of the nitroꢀoxyalkyl
1
selective COXꢀ2 inhibitory activity ranging from low nanomolar to micromolar values (Table 1).
On the basis of the structureꢀactivity relationships (SAR), nitroꢀoxyalkyl ethers 9a-d and 10a-c
showed, on the whole, a better COXꢀ2 inhibitory activity with respect to the corresponding
hydroxyalkyl derivatives 17a-d and 18a-c. This result could be due to an electronic interaction of
the nitrate group or other interactions of this moiety at the inner hydrophobic channel of the
enzyme. On the whole, compounds 9a-d and 10a-c appear equipotent to the aboveꢀcited nitroꢀ
oxyalkyl esters 6a-f. In particular, the inhibitory activity of nitroꢀoxyethyl derivatives 9a-c seems
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to be independent of the presence, and to some extent, of the position of the fluorine atom but not
from the number of substituents. 3’,4’ꢀdifluoro derivative 9d is much less active (0.920 ꢁM) than
the corresponding unꢀsubstituted or monoꢀhalogenated derivatives, 9a (0.017 ꢁM) or 9b,c (0.027
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢁ
M and 0.014 ꢁM), respectively. A different behavior may be found in nitroꢀoxypropyl ethers
10a-c. Only nonꢀhalogenated derivative 10a still retains a potent COXꢀ2 inhibitory activity
0.015 ꢁM). As for 3’ꢀ and 4’ꢀfluorinated derivatives 10b and 10c, their inhibitory activity has
(
partially decreased. This demonstrates that the presence of the halogen in 4’ position (10c) is not
fully compatible with the lengthening of the spacer between the O NOꢀgroup and the ethereal
2
oxygen atom of the sideꢀchain. This finding was not confirmed when 3’ꢀ and 4’ꢀfluorinated nitroꢀ
oxypropyl esters 6e and 6f were taken into account. In fact, both compounds showed a nanomolar
COXꢀ2 inhibitory activity, compound 6d being the most active (0.0073 ꢁM) within the series. In
this case, the three carbon atoms spacer, the presence of the carbonyl group and the 3’ꢀposition of
the fluorine atom in the N ꢀphenyl ring seem to be the optimal combination for a profitable
1
interaction with the active site of the isoenzyme. Furthermore, among hydroxyethylꢀ and
hydroxypropyl ethers 17a-c and 18a-c, which could be regarded as the metabolites of nitroesters
9
and 10, respectively, compounds 17a-c showed an efficacious and selective COXꢀ2 inhibitory
activity spanning from 0.027 ꢁM of nonꢀhalogenated compound 17a to 0.046 ꢁM and 0.089 ꢁM
of 3’ꢀF and 4’ꢀF derivative, respectively. In this small series of compounds the presence and the
position of the fluorine atom, although not appearing to be detrimental to the retaining of the
COXꢀ2 inhibitory activity, certainly decreases to some extent the activity of unꢀsubstituted
alcohol 17a. On the contrary, within the series of hydroxylꢀpropyl ethers 18a-c, the absence of
fluorine atom gives rise to the inactive compound 18a. The presence of the halogen at 3’ꢀposition
of the N ꢀphenyl ring of the pyrrole nucleus in connection with the hydroxylꢀpropyl chain seems
1
to be decisive for the strong COXꢀ2 inhibitory activity. This inhibitory activity drastically
ACS Paragon Plus Environment

Page 11 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
decreases when the fluorine atom is shifted from position 3’to 4’. As for O NOꢀalkylꢀ inverse
2
ester (7), and ꢀcarbonate (8), these compounds were found to be devoid of both activities being
unable to release NO in an efficient way, lacking at the same time the COXꢀ2 inhibitory activity,
unlike it happens for compounds 6, 9, and 10. This different behavior may be attributed to the
diversity of the functional group inserted as linker between the pyrrole scaffold of compounds 7
and 8 and the nitroester moiety. As a result, its sideꢀchains do not seem to have the right
stereoelectronic requisites for a profitable interaction with the isoenzyme. Compounds 9a-d and
10a-c were further evaluated to assess their efficacy and potency in determining NOꢀvasorelaxing
responses in a suitable experimental model of vascular smooth muscle and, in particular, in
endotheliumꢀdenuded rat aortic rings. In this experimental approach, five of the seven nitrooxyꢀ
alkyl derivatives exhibited vasorelaxing effects. In particular, compounds 9a-d and 10a evoked a
concentrationꢀdependent relaxation of the vascular smooth muscle with potency levels ranging
around 10 ꢁM and efficacy parameters ranging from 43% to 65% (Table 2). Only compounds
10b and 10c showed negligible levels of vasorelaxing efficacy (<20%). The experiments carried
out in the presence of guanylateꢀcyclase (GC) inhibitor 1Hꢀ[1,2,4]oxadiazoloꢀ[4,3ꢀa]quinoxalinꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1ꢀone (ODQ), confirmed that the vasorelaxing effects were due to the release of NO. The effect
of ODQ in fact, significantly antagonized the vasodilator responses evoked by nitrooxyꢀ
derivatives 9a-d and 10a (Table 2). Previous experiments also showed that corresponding
hydroxyalkyl derivatives 17a-d and 18a-c were devoid of significant vasorelaxing effects (data
not shown). This result, further indicated that such a pharmacodynamic feature is due to the
nitrooxy group and to its probable biotransformation to NO. Consistent with the data emerged in
2
3,24
previous studies,
the present experimental results indicate some relevant structural aspects,
capable of influencing the vasorelaxing effects (i.e., the NOꢀreleasing process). In particular, it
can be observed again that the length of the alkyl chain bearing the nitroꢀoxy group significantly
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
affects the pharmacological behavior. Indeed, the nitroꢀoxyethyl ethers (twoꢀcarbon alkyl chain)
exhibit higher levels of vasorelaxing effects than those shown by the corresponding nitroꢀ
oxypropyl ethers (threeꢀcarbon alkyl chain). This is clearly evident from the direct comparison of
the couples of analogues 9a vs 10a, 9b vs 10b, and 9c vs 10c (Table 2). It should also be noted
that glyceryl trinitrate (GTN), selected as reference drug endowed with rapid NOꢀreleasing
properties, showed strong vasorelaxing effects, with higher levels of vasorelaxing potency (Table
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2). Such an experimental evidence suggests that the compounds herein reported ensure a NO
release slower than GTN, thus acting as a “modulated” NOꢀdonor. Thus, we have identified
novel hybrid compounds endowed with selective COXꢀ2 inhibitory activity, important for
analgesic and antiꢀinflammatory efficacy, and slow NOꢀreleasing properties which may translate
2
5
into prolonged protective effects for the cardiovascular system.
Table 1. In Vitro COXꢀ1 and COXꢀ2 Inhibitory Activity (J774 murine macrophage assay) of
Compounds 6, 9, 10, 17, and 18.
COX-
COX-1
COX-2
Cpd
X
n
1/COX
a
a
IC (µM) IC (µM)
5
0
50
b
-2 (SI)
c
6
a
H
H
2
3
2
>10
>10
>10
0.0430
0.0420
0.0190
>232.6
> 238.1
> 526.3
c
6
b
c
6
c
3ꢀF
ACS Paragon Plus Environment

Page 13 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
c
6
d
3ꢀF
4ꢀF
4ꢀF
H
3
2
3
2
2
2
1.1
>10
>10
>10
>10
>10
>10
>10
2.9
0.0073
0.0290
0.0372
0.0170
0.0270
0.0140
0.9200
0.0150
0.0230
0.1900
0.0270
0.0460
0.0890
0.4200
8.9900
0.2400
0.9400
150.7
> 344.8
> 268.8
> 588.2
> 357.1
> 714.3
> 10.9
> 666.7
126.1
c
6e
c
6
f
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
a
9b
3ꢀF
4ꢀF
9
c
9d 3,4ꢀF2 2
1
0a
0b
10c
7a
7b
7c
H
3
3
3
2
2
2
1
3ꢀF
4ꢀF
H
>10
>10
>10
>10
>10
>10
3.7
> 52.6
> 370.4
> 217.4
> 112.3
> 23.8
> 1.11.
15.4
1
1
3ꢀF
4ꢀF
1
1
7d 3,4ꢀF2 2
18a
8b
8c
H
3
3
3
1
3ꢀF
4ꢀF
1
>10
>10.6
a
Results are expressed as the mean (n = 3 experiments) of the % inhibition of PGE production
2
by test compounds with respect to control samples. The IC values were calculated by GraphPad
5
0
Instat program; data fit was obtained by means of the sigmoidal dose–response equation (variable
b
slope) (GraphPad software). In vitro COXꢀ2 Selectivity Index [IC50 (COXꢀ1) /IC (COXꢀ
50
c
2
)]. See ref. 24.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
On the basis of their very encouraging COXꢀ2 inhibitory activity evidenced in the in vitro tests,
compounds 9a-d and 17a-d, and 18a were then selected and submitted to further
pharmacological tests to assess their in vivo antiꢀinflammatory and antiꢀnociceptive activities.
For this purpose compound effects were evaluated both in a chemical visceral pain model,
induced by the intraperitoneal injection of acetic acid (Writhing test), and in the carrageenanꢀ
induced inflammatory pain model (Tables 3 and 4). In Table 3 the analgesic effect of
compounds: 9aꢀd, 10a, 17aꢀd, and 18a compared to celecoxib (as reference drug), evaluated as
the number of abdominal constrictions induced by intraperitoneal acetic acid, is reported. For
each molecule a doseꢀresponse curve has been performed. All compounds exhibited an analgesic
effect. The minimal dose able to revert the painful condition was of 10 mg/kg po for 9a, 9c, and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ1
17a, 20 mg/kg for 9d, 10a, 17b,d, 18a and 40 mg kg for 9b, and 17c. Celecoxib was able to
increase the pain threshold starting from 3 mg/kg.
Table 2. Evaluation of Efficacy and Potency in Determining NOꢀDependent Vasorelaxing
Responses of Compounds 9a-d, 10a-c and Gliceryl Trinitrate (GTN).
a
b
Compd
X
H
n
2
3
2
3
Emax
pIC50
c
6a
65 ± 2 5.76 ± 0.08
c
6b
H
44 ± 8
≤ 5
c
6c
3ꢀF
3ꢀF
69 ± 4 6.48 ± 0.06
c
6d
39 ± 1
≤ 5
ACS Paragon Plus Environment

Page 15 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
c
6
e
4ꢀF
4ꢀF
H
2
3
2
2
2
2
3
3
3
58 ± 5 5.47 ± 0.07
c
6f
41 ± 2
≤ 5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
a
65 ± 3 5.22 ± 0.03
60 ± 4 5.32 ± 0.05
9
b
3ꢀF
4ꢀF
3,4ꢀF2
H
9
c
49 ± 4
43 ± 6
48 ± 5
n.a.
≤ 5
≤ 5
≤ 5
ꢀ
9
d
10a
1
0b
3ꢀF
4ꢀF
10c
n.a.
ꢀ
GTN
93 ± 2 6.90 ± 0.07
a
Emax represents the vasorelaxing efficacy, expressed as a % of the vasoconstriction induced by
b
the preꢀadministration of KCl. The parameter of potency is expressed as pIC , representing the ꢀ
5
0
c
Log of the molar concentration capable of inducing a vasorelaxing effect = 50% of Emax. See ref.
4.
2
Moreover, these compounds were assayed in rats to assess their effects on pain threshold
alteration and oedema induced by carrageenan. Four hours after the intraꢀplantar administration
of carrageenan in the paw, all inflammatory signs were observed (paw swelling hyperaemia and
hyperalgesia). The paw pressure test was used to measure pain. All molecules showed a
statistically significant capability to block the painful condition caused by the strong
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inflammatory agent in the range between 30 and 120 min after administration (Table 4). Thirty
minutes after a 20 mg/kg po administration, all compounds showed a good activity against
carrageenanꢀinduced hyperalgesia. This result was comparable to that obtained with celecoxib
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(10 mg/kg po). Moreover, with the exception of 9b and 9d, for all compounds, at the higher
tested dose, the persistence of the antiꢀhyperalgesic activity after the injection was longer than 60
min. Compounds 9a, 9c, 10a, 17c and 17d exhibited their efficacy up to 120 min after treatment.
In the carrageenanꢀinduced inflammatory pain model the measure of the paw volume allows to
discriminate the analgesic property from the antiꢀinflammatory one. In Table 4 is shown a
significant paw oedema decrease 60 min after the administration of all the investigated
compounds. These data strongly suggest the antiꢀinflammatory effect as base of the analgesic
mechanism of these molecules. Finally, the efficacy of compounds 9c and 17c were evaluated
also in the rat osteoarthritis model induced by the intraꢀarticular injection of monoiodoacetate
(MIA). The injection of MIA in the knee joint induces necrosis of condrocytes with decrease of
2
6
cartilage thickness and osteolysis. Kobayashi et al. showed that MIA is able to disorganize
27
condrocytes and to promote cartilage erosion. These alterations are comparable with joint
2
8ꢀ30
ꢀ1
damages typical of humans affected by osteoarthritis.
20 mg kg compound 9c or 17c were
p.o. administered twice daily for 13 days starting from the day of MIA injection to evaluate a
preventive effect. On the day 14, when pain as well as the degenerative articular process are
overt, the pain threshold was measured by Paw pressure test (Table 5). The repeated treatment
with 9c or 17c significantly prevented MIAꢀdependent hyperalgesia (Before treatment). A further
compound administration (after treatment, 30 min) did not induce an additive acute antiꢀ
hyperalgesic effect suggesting that a repeated treatment with the present antiꢀinflammatory
2
8ꢀ31
compounds could be able to prevent the joint damage and, consequently, pain.
ACS Paragon Plus Environment

Page 17 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 3. DoseꢀResponse Curves of Compounds 9a-d, 10a, 17a-d, and 18a in the Acetic Acid
Writhing Test.
n.
mice
36
10
8
Dose per os
Treatment
SALINE
n. writhes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
-1
mg kg
32.4 ± 1.9
21.2 ± 3.0^
15.5± 3.6*
32.5 ± 3.7
30.3 ± 2.1
25.3 ± 3.6*
28.5 ± 3.2
19.1 ± 2.7*
15.2 ± 3.3*
27.3 ± 2.5
24.8 ± 3.0*
25.6 ± 2.3*
29.4 ± 3.0
21.2 ± 3.0*
26.8 ± 3.2^
21.2 ± 3.0*
15.5 ± 3.6*
29.4 ± 3.0
26.8 ± 3.2^
9
9
a
a
10
20
10
20
40
3
9b
8
9b
8
9b
8
9
9
c
c
8
8
10
20
10
20
40
10
20
40
10
20
10
20
9c
8
9
9
9
d
d
d
7
7
7
10a
9
10a
10
8
1
1
1
0a
7a
7a
8
8
17b
8
17b
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
7b
8
7
40
10
20
40
10
20
40
10
20
40
1
17.3 ± 3.5*
31.3 ± 2.7
27.9 ± 3.6
18.2 ± 3.1*
32.3 ± 2.8
26.3 ± 3.4^
17.3 ± 2.2*
29.7 ± 3.4
22.3 ± 2.5*
21.5 ± 2.3*
25.6 ± 3.1*
15.4 ± 2.5*
11.3 ± 2.9*
17c
1
1
7c
7c
8
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
17d
9
1
1
7d
7d
8
8
1
1
8a
8a
9
9
18a
7
Celecoxib
Celecoxib
Celecoxib
10
11
15
3
10
Each value represents the mean of at least 7 mice. ^ P< 0.05; * P<0.01 in comparison with CMC
treated group.
ACS Paragon Plus Environment

Page 19 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. Effect of Compounds 9a-d, 10a, 17a-d, and 18a on Hyperalgesia and Oedema Induced
by Carrageenan in the rat Pawꢀpressure Test.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Paw pressure in rats (g)
After Treatment
Paw volume (ml)
Pre-
Treatment
intra-
Treatment
Before
30
60
120
Before
60
per os
Treatment
min
min min Treatment min
plantar
SALINE
SALINE
SALINE
9a 20
62.6 ± 2.4
31.4 ± 3.4
61.5 60.2 62.9 1.46 ± 0.05 1.42 ±
34.8 33.9 31.4 1.47 ± 0.07 2.48 ±
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
CARRAG
33.9 ± 3.1 54.2± 55.3± 48.7 1.31 ± 0.09 1.45±
9b 20
34.6 ± 3.0
31.6 ± 2.7
34.1 ± 2.7
32.0 ± 3.5
32.7 ± 3.7
35.2 ± 3.1
32.9 ± 3.3
31.8 ± 3.2
32.7 ± 3.9
33.5 ± 3.9
32.5 ± 3.4
30.5 ± 3.5
31.6 ± 2.8
34.1 ± 3.3
33.8 ± 3.8
31.7 ± 2.7
33.5 ± 2.6
43.9 40.7 35.2 1.58 ± 0.08 2.16 ±
47.3 41.0 34.3 1.55 ± 0.09 2.08 ±
55.1 52.6 48.3 1.49 ± 0.08 1.87 ±
56.8 49.2 43.8 1.53 ± 0.08 1.92 ±
44.6 46.9 38.8 1.54 ± 0.06 2.19 ±
49.5 42.5 35.3 1.48 ± 0.07 2.10 ±
53.7 51.6 45.6 1.53 ± 0.06 1.80 ±
56.6 49.2 37.8 1.26 ± 0.05 1.34±
46.9 48.6 37.2 1.55 ± 0.08 2.46 ±
51.4 45.3 39.5 1.49 ± 0.09 2.12 ±
46.8 38.7 38.3 1.57 ± 0.07 2.01 ±
57.2 58.3 47.6 1.55 ± 0.09 1.95 ±
43.6 46.7 41.4 1.56 ± 0.08 2.16±
53.2 55.2 49.3 1.58 ± 0.09 1.96 ±
52.9 50.3 46.6 1.47 ± 0.05 1.86 ±
44.3 41.6 40.7 1.59 ± 0.08 2.35±
52.9 48.3 39.8 1.50 ± 0.05 1.45 ±
9b 40
9c 10
9c 20
9d 20
9d 40
10a 20
17a 20
17b 20
17b 40
17c 20
17c 40
17d 20
17d 40
18a 20
CELECOX 3
CELECOX 10
There were 4 rats per group. ^ P< 0.05; * P<0.01 versus the carrageenan/salineꢀtreated group.
CARRAG= carrageenan.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 5. Effect of Repeated Treatment with Compounds 9c and 17c on Hyperalgesia Induced by
Osteoarthritis in the Paw Pressure Test.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Paw pressure in rats (g)
Pre-Treatment
Intra-articular
saline
Treatment
Per os
Before
Treatment
After Treatment
(30 min)
CMC
61.2 ± 3.0
58.7 ± 2.9
a
MIA
CMC
9c (20)
17c (20)
32.6 ± 2.9
43.5 ± 2.9*
49.7 ± 3.5*
30.4 ± 3.1
MIA
MIA
45.1 ± 3.7*
51.2 ± 3.9*
a
There were 4 rats per group.^ P< 0.05; * P<0.01 versus the MIA treated rat. MIA= monosodium
iodoacetate.
Study of cartilage catabolism in culture has received considerable attention over the past three
3
2
decades because of the pivotal role that this process plays in cartilage degeneration in arthritis.
On realizing that such degeneration is driven by cytokines, cartilage organ culture became a
standard technique for monitoring the effects of cytokines on chondrocytes metabolism. Under
such conditions proteoglycans degradation was found to be an early event, with the
glycosaminoglycans (GAG)ꢀrich degradation products being readily released from the tissue into
3
3
the culture medium. Therefore, following the assessment of their antiꢀinflammatory and
antinociceptive activity nitroꢀoxyethyl ether derivative 9c and its metabolite 17c were selected
and tested to determine their cartilage protective properties. The test consisted of quantitation of
GAG in the form of aggrecan fragments released from bovine articular cartilage of the
ACS Paragon Plus Environment

Page 21 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
metacarpophalangeal joints of the feet. After being cultured according to the procedure reported
3
4
by Homandberg et al. (see SI) fragments were challenged for 48h with ILꢀ1β (30 ng/mL) and
quantitation of GAG release was assessed by means of most widely used Farndale’s colorimetric
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
35
method. Results are shown in Figure 1. Compounds 9c and 17c were able to inhibit GAG
release induced by ILꢀ1β in a concentrationꢀdependent manner, (IC50 of 1.5 and 33 ꢁM,
respectively). Treatment with compounds 9c, 17c or ILꢀ1β did not affect the viability of cell
recovery from cartilage. Our results show that compound 9c can contribute to the reduction of ILꢀ
1βꢀdependent cartilage catabolism both by the inhibition of prostanoid biosynthesis and by NO
releasing properties. Neverthless, the release of NO by 9c seems to play a major role. This
assumption is based on the fact that compound 17c which was unable to release NO, was found
to be 22ꢀfold less potent than the parent compound, although its activity in inhibiting COXꢀ2
activity in murine J774 cells was comparable to 9c.
Figure 1. Effect of compounds 9c and 17c on glycosaminoglycans (GAG) released from bovine
articular cartilage challenged for 48h with ILꢀ1β (30 ng/mL). Results are expressed as per cent of
control (ILꢀ1β alone) (mean ± SEM, n = 3); EC (9c and 17c concentration that reduced by 50%
5
0
ILꢀ1βꢀinduced GAG release) and its 95% confidence intervals (CI) are shown.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
With the aim of assessing COXꢀ2 selectivity, a Human Whole Blood (HWB) assay was carried
out on selected compounds 9c and 17c (nitroꢀoxy ether and hydroxyethyl ether, respectively). In
particular, the assay was performed to predict the actual extent of isozyme inhibition achievable
in vivo by circulating drug levels to consider the amount of variables potentially able to affect
drugꢀenzyme interaction. As shown in Figure 2 (panels A and B) the concentrationꢀresponse
curves for inhibition of COXꢀ1 and COXꢀ2 in human whole blood elicited by 9c and 17c,
respectively, were studied. Nitroꢀoxy ether derivative 9c and its NOꢀfree metabolite 17c inhibited
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
LPSꢀinduced whole blood PGE generation (COXꢀ2 assay) in a concentrationꢀdependent fashion
2
with IC values (Table 6) significantly comparable. The steepness of the curves by calculating
5
0
the Hill slope values was also studied. 9c Hill slope value was significantly comparable to 1.0,
which is typical for a standard sigmoid concentrationꢀresponse curve where response (Y) is a
function of the logarithm of compound concentration (X). The Hill slope value of 17c was
3
6
slightly, but significantly, <1.0 (Table 6). As has been previously reported, NSAIDs which
inhibit COX activity with a Hill slope value of 1.0 are freely reversible and exhibit simple
competitive inhibition. Since the administration of an analgesic dose of NSAIDs is associated
3
7
with a 80% inhibition of whole blood COXꢀ2 activity, the IC values for these compounds was
8
0
calculated. As shown in Table 6, compounds 9c and 17c showed comparable IC80 values in
inhibiting COXꢀ2. Taken together, these results suggest that nitroꢀoxyethyl ether derivative 9c
and its metabolite 17c have a comparable and potent inhibitory effect on monocyte COXꢀ2
activity of HWB. Furthermore, 9c and 17c inhibited platelet COXꢀ1 activity in a concentrationꢀ
dependent fashion (Figure 2, panels A and B); however, 17c was slightly, but significantly, more
potent in the inhibition of COXꢀ1 rather than 9c (both at IC and IC ) (Table 6). Interestingly,
5
0
80
both compounds inhibited COXꢀ1 activity with Hill slope values >1.0 (Table 6), normally a
ACS Paragon Plus Environment

Page 23 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
feature of steeper curves. A timeꢀdependent COX inhibiting effect was observed by NSAIDs
3
6
with Hill slope values >1.0. Selectivity of 9c and 17c towards COXꢀ2 was determined as IC
50
and IC ratios for the inhibition of whole blood COXꢀ1 and COXꢀ2. As shown in Table 6, both
8
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
compounds resulted 50ꢀfold more potent towards COXꢀ2 than COXꢀ1 at IC . A lower COXꢀ2
5
0
selectivity was found at IC80.
In conclusion, 9c and its metabolite 17c are potent and selective inhibitors of COXꢀ2. Although
further studies should be performed, our data may suggest that 9c and 17c inhibited COXꢀ2, but
not COXꢀ1, in a timeꢀdependent fashion.
Altogether our results offer proofꢀofꢀconcept that compounds endowed with dual
pharmacological activities, i.e. selective COXꢀ2 inhibition and NOꢀreleasing properties, show
analgesic and antiꢀinflammatory effects in experimental rodent models in vivo. Our results
evidence that COXꢀ2 inhibition is the central mechanism for obtaining analgesic and antiꢀ
inflammatory effects. Similarly, in randomized clinical trials naproxcinod has been shown to
3
8
cause comparable relief of the signs and symptoms of OA. However, we showed in vitro that
the NOꢀreleasing properties of the novel compound 9c introduced an additional protective effect
by preventing cartilage matrix degradation as compared to compound 17c which was unable to
release NO
.
Indeed, NO, at low concentrations, has been previously reported to play protective
39
roles in the joint.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Effects of Compounds 9c and 17c on in vitro Platelet COXꢀ1 and Monocyte COXꢀ2
Activity in Human Whole Blood. Results are expressed as average percentage of inhibition (N =
3ꢀ6, mean ± SEM).
ACS Paragon Plus Environment

Page 25 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 6. IC and IC Values, Hill Slopes and COXꢀ2 Selectivity of Compounds 9c and 17c in
5
0
80
Human Whole Blood (HWB) Assays.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compd
9c
17c
COX2 COX1 COX2 COX1
0.64
(0.43ꢀ (26.16ꢀ (0.19ꢀ (17.33ꢀ
.96) 36.55) 0.83) 23.10)
31.24
0.40
20.00
IC (ꢁM)
5
0
0
1.10
1.85
0.79
2.24
Hill Slope
(0.73ꢀ (1.37ꢀ (0.59ꢀ (1.72ꢀ
1
.46)
2.33)
0.98)
2.77)
IC COXꢀ1/
5
0
a
50.00
50.00
IC COXꢀ2
5
0
IC (ꢁM)
2.26
29.33
Selectivities of compounds towards COXꢀ2 were determined as IC and IC ratios for HWBAꢀ
63.3
2.31
37.14
8
0
IC COXꢀ1/
8
0
a
16.07
IC COXꢀ2
8
0
a
5
0
80
COXꢀ1 and COXꢀ2. For IC values and Hill slopes, 95% confidence intervals are shown in
5
0
round brackets.
Molecular Modeling Simulations
To better understand the inhibitorꢀenzyme interactions, and to improve our previously proposed
24
mode of binding of NOꢀdonors and COXꢀ2 inhibitors 6, compounds 6d and 10b were
submitted to rigorous ab-initio calculations along with molecular docking and molecular
dynamics simulations. In particular, compounds 6d and 10b were built starting from the xꢀray
crystallographic data of 6c (Figure S1 in the SI), and then optimized by means of ab-initio
4
0
Restricted HartreeꢀFock (RHF) method taking into account the generic Nuclear Overhauser
4
1
1
Effect (NOE) constraints resulting from HꢀNMR solution studies (see SI). The partial atomic
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
40
point charges of compounds 6d and 10b were calculated by the RHF method and used in the
following molecular docking calculations. The binary complexes of compounds 6d and 10b, with
isozyme COXꢀ2 (PDB entry: 6cox), obtained from flexible docking simulations, performed by
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
42
means of Autodock 4.2, were subjected to an extensive molecular dynamics study aimed at
investigating: i) the binding mode profile of new derivatives; ii) the stability of the previously
2
4
published binding mode of nitroꢀoxyalkyl esters 6c and 6d; iii) and the main intermolecular
interactions based on a large timeꢀscale simulation. During 5ꢀns of molecular dynamics
simulations, the complexes showed a geometric stable profile according to which the binding
2
4
mode of compound 6d did not substantially diverge from that one previously reported,
confirming the significance of the docking results obtained on that occasion. This is the reason
why in the present study only the main interactions between the lateral side chain bearing the ꢀ
ONO moiety of the 1,5ꢀdiarypyrrole derivatives 6d and 10b and the COXꢀ2 active site were
2
taken into account and discussed. During the whole molecular dynamics, compound 6d by means
of the two oxygen atoms of the carboxypropyl group can engage, permanently, a double
hydrogen bond interaction with ARG 120 and TYR 355. Thus, nitroester group is allowed to lie
near to SER 119 favouring, in this way, an additional hydrogen bond interaction of one of the
oxygen atoms with the hydroxyl group of such amino acid (Figure 3).
ACS Paragon Plus Environment

Page 27 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Snapshot of the molecular dynamics simulation of Nitroꢀoxypropyl ester 6d into the
COXꢀ2 active site. The main residues and the inhibitor are represented as licorice. Hydrogen
bonds are represented as blue dotted lines. For the sake of clarity, only the lateral side chain
interacting residues are displayed.
It is worth noting that, when a 5ꢀns trajectory from a molecular dynamics simulation (MD) of the
complex of nitroꢀoxypropyl ether 10b with the active site of COXꢀ2 is analyzed, the hydrogen
bond interaction between one of the oxygen atoms of the nitro group and SER 119 is completely
lost in favor of a different orientation of the lateral side chain which seems to be no more
extended than it appears in compound 6d (Figure 5). In addition, the hydrogen bond between
ARG 120 and the ethereal oxygen of compound 10b is not permanently present during the MD
simulation. (Figure 4).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 58
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Snapshot of the molecular dynamics simulation of Nitrooxypropyl ether 10b into the
COXꢀ2 cyclooxygenase site. The main residues and the inhibitor are represented as licorice.
Hydrogen bonds are represented as blue dotted lines. For the sake of clarity, only the lateral side
chain interacting residues are displayed.
On the whole, according to the molecular dynamics simulations, it appears that the conformation
of the lateral side chain of the inhibitors along with the hydrogen bond interaction with the SER
119 have to be considered responsible for the different COXꢀ2 inhibitory activity of compounds
6d and 10b, the IC50 of which, being 7.3 and 23.0 nM, respectively. On the other hand,
compound 9d shows a different binding mode with respect to compounds 6d and 10b. In
particular, compound 9d is not able to form any hydrogen bond with SER 119 and the polar
contact to the ARG 120 is actually infrequent during the molecular dynamics simulation studies.
These differences, in agreement with the biological data are probably due to the stereoelectronic
effects induced by the introduction of 3’,4’ꢀdifluoro substituents coupled to a short lateral side
chain linker.
ACS Paragon Plus Environment

Page 29 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
13
H- and C-NMR Studies Performed on Compounds 6c,d, 9c, and 10b.
NMR analysis has been done with the aim of determining both dynamics and conformations of
compounds 6c,d, 9c, and 10b the atoms of which have been numbered according to Figure 5. It is
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
known that C R (R = 1/T ), relaxation rates, are almost exclusively determined by dipolar
1
1
1
interactions with directly bonded or nearby protons, thus allowing suitable delineations of the
4
3,44
molecular dynamics.
The motional features of aromatic carbons were firstly considered since
they have fewer degrees of freedom than sideꢀchain carbons. For ring B of compounds 6c, 6d,
and 10b, the relaxation rate of C was faster than those of the other carbon atoms, suggesting
1
4
that C – C is the main rotation axis for this aromatic ring. The relaxation rates of C , C , and
1
1
14
12
15
C were very similar, so that an anisotropic model consisting of rotational reorientation around
1
6
4
5
the main rotation axis with some degree of internal motion could be applied (eq. [1] with A =
2
2
2
2
4
0
.25(3cos α ꢀ 1) , B = 3(sin αcos α), and C = 0.75(sin α)).
1
13
Figure 5. Numbering of atoms of compounds selected for Hꢀ and CꢀNMR studies.
ACS Paragon Plus Environment