Coumarins and adenosine receptors: New perceptions in structure–affinity relationships

Article abstract of DOI:10.1111/cbdd.13075

Adenosine receptor (AR) subtypes are involved in several physiological and pharmacological processes. Ligands that are able to selectively modulate one receptor subtype can delay or slow down the progression of diverse diseases. In this context, our research group focused its investigation into the discovery and development of novel, potent and selective AR ligands based on coumarin scaffold. Therefore, a series of 3-phenylcarboxamidocoumarins were synthesized and their affinity for the human AR subtypes was screened by radioligand binding assays for A₁, A_2A and A₃ receptors and for A_2B by adenylyl cyclase assay. Compound 26 was found to be the most remarkable, with a hA₁/hA₃ and hA_2A/hA₃ selectivity of 42, for the A₃ AR (K_i?=?2.4?μm). Receptor-driven molecular modelling studies have provided valuable information on the binding/selectivity data of compound 26 and for the following optimization process. Moreover, compound 26 presents drug-like properties according to the general guidelines linked to the concept.

Full text of DOI:10.1111/cbdd.13075

PROFESSOR FERNANDA BORGES (Orcid ID : 0000-0003-1050-2402)
Article type : Research Article
Coumarins and adenosine receptors: new perceptions in structure-affinity
relationships
André Fonsecaa, b, Maria João Matos*a, Santiago Vilara,b, Sonja Kachlerc, Karl-Norbert Klotzc
, Eugenio Uriarteb, d and Fernanda Borges*a
aCIQUP/Department of Chemistry and Biochemistry, Faculty of Sciences, University of Porto, 4169-
007, Porto, Portugal
bDepartamento Quimica Orgánica, Facultade de Farmacia, Universidad Santiago de Compostela,
15782, Santiago de Compostela, Spain
cPharmacology and Toxicology Institute, University of Würzburg, 97078, Würzburg, Germany
dInstituto de Ciencias Químicas Aplicadas, Universidad Autónoma de Chile, 7500912, Santiago de
Chile, Chile.
*Corresponding Authors: Fernanda Borges, CIQUP/Department of Chemistry and Biochemistry,
Faculty of Sciences, University of Porto, 4169-007, Porto, Portugal. E-mail: fborges@fc.up.pt
Maria João Matos, Organic Chemistry Department, Faculty of Pharmacy, University of Santiago de
Compostela, 15782, Santiago de Compostela, Spain.Email: mariacmatos@gmail.com
Keywords: Coumarins; Adenosine Receptors; Carboxamidocoumarin;
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/cbdd.13075
This article is protected by copyright. All rights reserved.

Abstract
Adenosine receptors (ARs) subtypes are involved in several physiological and pharmacological
processes. Ligands able to selectively modulate one receptor subtype can delay or slow down the
progression of diverse diseases. In this context, our research group focused its investigation into the
discovery and development of novel, potent and selective ARs ligands based on coumarin scaffold.
Therefore, a series 3-phenylcarboxamidocoumarins were synthesised and their affinity for the
human ARs subtypes was screened by radioligand binding assays for A1, A2A and A3 receptors and for
A2B by adenylyl cyclase assay. Compound 26 was found to be the most remarkable, with a hA1/hA3
and hA2A/hA3 selectivity of 42, for the A3 AR (Ki = 2.4 µM). Receptor-driven molecular modelling
studies have provided valuable information on the binding/selectivity data of compound 26 and for
the following optimization process. Moreover, compound 26 present drug-like properties according
to the general guidelines linked to the concept.
Introduction
Adenosine is an endogenous purine nucleoside with an important role in a diversity of biochemical
processes, namely in energy transfer (adenosine triphosphate or ATP) and in cellular signalling (cyclic
AMP). In the 1920’s it was demonstrated for the first time adenosine biologic action in the
cardiovascular system.1 In addition to its clinical role as anti-arrhythmic agent, adenosine has been
implicated in diverse pharmacological areas and throughout the years adenosine signaling pathways
have often been used in drug design and development projects, with adenosine itself or its
derivatives being used clinically since the 1940s.2
Extracellular adenosine is a signaling molecule that can activate adenosine receptors (ARs).2 To this
date four distinct and widely expressed human AR subtypes – A1, A2A, A2B and A3 – have been
discovered, each one implicated in a number of physiological and pathological processes. The
structure, function, and basis for classification of ARs has been extensively reviewed.3 In particular,
activation of ARs can induce inhibition (A1AR and A3AR) or activation (A2AAR and A2BAR) of adenylyl
cyclase an enzyme that catalyses the conversion of ATP into cyclic adenosine monophosphate
(cAMP). Activation or blockade of ARs is responsible for a wide range of effects in numerous organ
systems and therefore the regulation of ARs can have many potential therapeutic applications.4
Pharmacological modulation of AR pathways open a new window for drug treatment of a variety of
pathologies, such as asthma, neurodegenerative disorders, cancer, inflammatory and ischaemic
related diseases.5–9
This article is protected by copyright. All rights reserved.

Coumarins are naturally occurring heterocyclic compounds with an in-depth history in Medicinal
Chemistry,10 as they have been exploited in quite a lot of projects aiming to find, for instance, anti-
cancer, antioxidant, anti-inflammatory, antimicrobial and antiviral agents.11 Although benzopyran is
considered a privileged structure, few studies have been addressed towards its application in the
discovery of new ARs ligands. In this context, a number of coumarin based derivatives, in particular
3-arylcoumarins, have been reported by our group as inspiring ligands (Figure 1). 12–15
Preliminary structure-activity relationship studies indicated that the nature of the substituents
located in the coumarin ring, the presence or absence of a spacer between the pyrone ring and an
aryl or alkyl side chain can modulate their affinity and selectivity, in particular towards A3AR. To gain
insight over the structural requirements needed for the development of a potent and selective AR
coumarin-based ligand a series of 6-substituted coumarin derivatives was synthesised, characterised
and pharmacologically evaluated. In addition, as amide group has also been proposed as operative
either in coumarins or in chromones, a isomeric system, this type of spacer was chosen for the
present study..16 The research was accomplished by receptor-driven molecular modelling studies.
Methods and materials
Materials and instruments
All starting materials and reagents were obtained from commercial suppliers and were used without
further purification. Melting points (mp) were determined using a Reichert Kofler thermopan or a
1
13
Büchi 510 apparatus and were not corrected. H (250 MHz) and C (63 MHz) NMR spectra were
recorded on a Bruker AMX spectrometer, using DMSO-d6 or CDCl3 as solvents. Chemical shifts (δ)
and coupling constants (J) were expressed in ppm and in Hz, respectively. TMS was used as internal
standard. The notations for multiplicity patterns were: s (singlet), d (doublet), dd (double doublet), t
(triplet), dt (double triplet) and m (multiplet). Mass spectrometry data was acquired with a Hewlett-
Packard-5972-MSD spectrometer. Silica gel (Merck 60, 230–400 mesh) was used for flash
chromatography (FC). Analytical thin layer chromatography (TLC) was performed on plates
precoated with silica gel (Merck 60 F254, 0.25 mm). Organic solutions were dried over anhydrous
Na2SO4 he s ents e e e te n t e t ( h t ).
Synthesis
General procedure for the synthesis of coumarin 3-phenylcarboxamides (8-19):
This article is protected by copyright. All rights reserved.

To a solution of coumarin-3-carboxylic acid (5, 6 or 7, 1.0 mmol) in dichloromethane (DCM, 5 mL) 1-
ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC, 1.1 mmol) and 4-dimethylaminopyridine
(DMAP, 1.1 mmol) were added. The mixture was kept in an argon flux at 0 ºC for five minutes.
Shortly after, the aromatic amine (1.0 mmol) with the pretended substitution pattern was added in
small portions. The reaction mixture was stirred at room temperature for 4 hours. The obtained
precipitate was filtered and purified by column chromatography (hexane/ethyl acetate 9:1) or by
recrystallization with ethanol to give the desired compounds.
The synthesis of the precursors (3-7) and compounds (9, 12, 15, 21, 24, 27, 30, 36, 39, 42, 45 and 51)
has been previously described.17
1
N-(2’-Hydroxyphenyl)coumarin-3-carboxamide (8) Yield: 54%; mp: 269-270 ºC. H NMR (DMSO-d6) δ
(ppm), J (Hz): 6.82-6-92 (m, 3H, H-8, H-4’, H-5’), 7 46-7.57 (m, 2H, H-6, H-7), 7.76 (d, 1H, H-5, J=7.4),
8.04 (d, 1H, H-3’, J=7.1), 8.39 (d, 1H, H-6’, J=7.1), 9.05 (s, 1H, H-4), 10.23 (s, 1H, OH), 11.11 (s, 1H,
NH). 13C NMR (DMSO-d6) δ (ppm): 161.2, 154.1, 148.5, 146.8, 134.6, 130.6, 126.7, 125.4, 124.4,
120.0, 119.3, 118.8, 116.4, 114.7, 101.4, 98.2. MS m/z (%): 281 (M+, 81), 174 (38), 173 (100), 101
(40), 89 (34).
N-(2’-Hydroxyphenyl)-6-methoxycoumarin-3-carboxamide (10) Yield: 47%; mp: 232-233 ºC. 1H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, OCH3), 6.75-6.82 (m, 1H, H-4’), 6 86-6.95 (m, 2H, H-3’, H-5’),
7.33 (dd, 1H, H-7, J=2.9, 9.1), 7.46 (d, 1H, H-8, J=9.1), 7.58 (d, 1H, H-5, J=2.9), 8.36 (dd, 1H, H-6’,
J=1.3, 7.8), 8.96 (s, 1H, H-4), 10.17 (s, 1H, OH), 11.10 (s, 1H, NH). 13C NMR (DMSO-d6) δ (ppm): 161.2,
158.9, 156.1, 148.6, 148.3, 146.7, 126.6, 124.4, 122.4, 120.0, 119.3, 119.7, 118.8, 117.4, 114.7,
111.9, 55.9. MS m/z (%): 312 (M+, 32), 311 (100), 204 (62), 203 (96), 119 (60), 65 (18).
1
N-(2’-Methylphenyl)coumarin-3-carboxamide (11) Yield: 56%; mp: 212-213 ºC. H NMR (CDCl3) δ
(ppm), J (Hz): 2.42 (s, 3H, CH3), 7.10 (td, 1H, H-4’, J=1.2, 7.4), 7.22-7.29 (m, 2H, H-7, H-8), 7.38-7.47
(m, 2H, H-6, H-5’), 7 67-7.77 (m, 2H, H-5, H-3’), 8 25 ( , 1H, H-6’, J=8.2), 9.04 (s, 1H, H-4), 10.79 (s,
13
1H, NH). C NMR (CDCl3) δ (ppm): 164.3, 158.2, 148.8, 148.8, 134.2, 130.3, 129.8, 128.4, 126.6,
125.3, 124.8, 121.7, 118.6, 116.6, 116.5, 99.9, 18.0. MS m/z (%): 279 (M+, 71), 261 (29), 173 (100),
106 (70), 101, (37), 89 (35), 77 (10), 63 (11).
This article is protected by copyright. All rights reserved.

1
N-(2’-Methylphenyl)-6-methoxycoumarin-3-carboxamide (13) Yield: 71%; mp: 186-187 ºC. H NMR
(CDCl3) δ (ppm), J (Hz): 2.39 (s, 3H, CH3), 3.86 (s, 3H, OCH3), 7.02- 7.12 (m, 2H, H-4’, H-5’), 7 18- 7.25
(m, 2H, H-3’, H-7), 7.27 (d, 1H, H-5, J=2.9), 7.36 (d, 1H, H-8, J=9.1), 8.22 (d, 1H, H-6’, J=8.1), 8.96 (s,
1H, H-4), 10.82 (s, 1H, NH). 13C NMR (CDCl3) δ (ppm): 162.1, 159.2, 158.8, 148.9, 136.0, 135.9, 130.3,
128.7, 126.5, 125.0, 122.7, 121.7, 118.9, 117.6, 115.5, 110.5, 55.8, 17.9. MS m/z (%): 309 (M+, 86),
291 (51), 281 (29), 203 (100), 119 (49), 106 (72), 77 (18), 65 (15).
N-(2’-Methoxyphenyl)coumarin-3-carboxamide (14) Yield: 64%; mp: 239-240 ºC. 1H NMR (CDCl3) δ
(ppm), J (Hz): 3.97 (s, 3H, OCH3), 6.92-7.14 (m, 3H, H-8, H-3’, H-6’), 7 41 (m, 2H, H-6, H-4’), 7 71 (m,
13
2H, H-7, H-5’), 8 54 ( , 1H, H-5, J=7.6), 9.00 (s, 1H, H-4), 11.28 (s, 1H, NH). C NMR (CDCl3) δ (ppm):
172.1, 150.2, 148.4, 144.5, 140.9, 134.0, 129.7, 127.4, 125.2, 124.4, 120.8, 120.5, 118.3, 116.6,
111.0, 110.1, 55.9. MS m/z (%): 295 (M+, 79), 264 (27), 187 (10), 173 (100), 122 (17), 101 (30), 89
(23).
N-(2’-Methoxyphenyl)-6-methoxycoumarin-3-carboxamide (16) Yield: 78%; mp: 193-194 ºC. 1H NMR
(CDCl3) δ (ppm), J (Hz): 3.85, 3.94 (s, 6H, 2 x OCH3), 6.91 (td, 1H, H-4’, J=1.5, 7.9), 6.99 (dd, 1H, H-3’,
J=1.5, 7.9), 7.03- 7.12 (m, 2H, H-5, H-5’), 7 22 ( , 1H, H-7, J=3.0, 9.1), 7.33 (d, 1H, H-8, J=9.1), 8.51
(dd, 1H, H-6’, J=1.7, 7.5), 8.91 (s, 1H, H-4), 11.32 (s, 1H, NH). 13C NMR (CDCl3) δ (ppm): 156.5, 149.0,
148.2, 134.5, 131.9, 127.5, 124.39, 122.5, 120.7, 120.4, 119.0, 118.9, 117.6, 110.5, 110.0, 106.8,
63.4, 55.8. MS m/z (%): 326 (M+, 30), 325 (82), 294 (28), 204 (30), 203 (100), 119 (53), 65 (18).
1
N-(2’-Chlorophenyl)coumarin-3-carboxamide (17) Yield: 68%; mp: 220-221 ºC. H NMR (CDCl3) δ
(ppm), J (Hz): 7.10 (dt, 1H, H-7, J=1.5, 7.7), 7.26-7.47 (m, 4H, H-6, H-4’, H-5’, H-6’), 7.67-7.76 (m, 2H,
H-8, H-3’), 8 56 ( , 1H, H-5, J=1.5, 8.7), 9.02 (s, 1H, H-4), 11.34 (s, 1H, NH). 13C NMR (CDCl3) δ (ppm):
159.4, 154.4, 150.7, 149.0, 134.3, 129.8, 129.2, 127.3, 125.3, 125.1, 122.1, 118.4, 117.7, 116.6,
114.9, 99.9. MS m/z (%): 299 (M+, 37), 264 (100), 173 (99), 145 (9), 101 (51), 89 (47), 75 (13), 63 (22).
1
N-(2’-Chlorophenyl)-6-methylcoumarin-3-carboxamide (18) Yield: 61%; mp: 203-204 ºC. H NMR
(DMSO-d6) δ (ppm), J (Hz): 2.35 (s, 3H, CH3), 7.14 (td, 1H, H-4’, J=1.5, 6.2), 7.32-7.45 (m, 2H, H-8, H-
5’), 7 50-7.60 (m, 2H, H-6, H-3’), 7 79 (s, 1H, H-5), 8.48 (dd, 1H, H-6’, J=1.5, 6.8), 8.95 (s, 1H, H-4),
11.24 (s, 1H, NH). 13C NMR (DMSO-d6) δ (ppm): 171.4, 168.9, 159.7, 152.5, 149.2, 136.3, 134.9,
This article is protected by copyright. All rights reserved.

130.2, 129.5, 128.0, 125.5, 121.8, 118.5, 118.0, 116.8, 115.6, 20.6. MS m/z (%): 313 (M+, 22), 279
(32), 278 (93), 187 (100), 115 (25), 103 (20), 77 (15).
1
N-(2’-Chlorophenyl)-6-methoxycoumarin-3-carboxamide (19) Yield: 73%; mp: 200-201 ºC. H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, OCH3), 7.15 (td, 1H, H-4’, J=1.4, 7.8), 7.31-7.42 (m, 2H, H-7, H-
5’), 7 48 ( , 1H, H-8, J=9.0), 7.54 (dd, 1H, H-3’, J=1.3, 7.8), 7.61 (d, 1H, H-5, J=3.0), 8.49 (dd, 1H, H-6’,
13
J=1.4, 8.4), 9.01 (s, 1H, H-4), 11.29 (s, 1H, NH). C NMR (DMSO-d6) δ (ppm): 166.6, 159.8, 156.2,
154.9, 149.0, 148.7, 138.0, 134.8, 129.6, 128.3, 128.1, 122.9, 121.5, 118.4, 117.8, 112.1, 56.0. MS
m/z (%): 329 (M+, 45), 295 (54), 294 (99), 204 (37), 205 (100), 119 (60).
1
N-(2’-Bromophenyl)coumarin-3-carboxamide (20) Yield: 32%; mp: 218-219 ºC. H NMR (CDCl3) δ
(ppm), J (Hz): 7.04 (t, 1H, H-4’, J=7.6), 7.33-7.47 (m, 3H, H-6, H-8, H-5’), 7 60-7.76 (m, 3H, H-5, H-7, H-
13
3’), 8.51 (d, 1H, H-6’, J=8.0), 9.01 (s, 1H, H-4), 11.21 (s, 1H, NH). C NMR (CDCl3) δ (ppm): 160.9,
160.0, 154.4, 149.1, 136.1, 134.4, 132.6, 129.8, 128.0, 125.6, 125.3, 122.7, 118.5, 118.4, 116.7,
114.4. MS m/z (%): 345 (M+, 25), 343 (24), 265 (58), 264 (99), 173 (100), 101 (59), 89 (66), 63 (33)
1
N-(2’-Bromophenyl)-6-methoxycoumarin-3-carboxamide (22) Yield: 36%; mp: 203-204 ºC. H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, OCH3), 7.09 (td, 1H, H-4’, J=1.5, 8.0), 7.33-7.45 (m, 2H, H-7, H-
5’), 7 49 ( , 1H, H-8, J=9.1), 7.61 (d, 1H, H-5, J= 2.9), 7.69 (dd, 1H, H-3’, J=1.4, 8.0), 8.43 (dd, 1H, H-6’,
13
J= 1.5, 8.3), 9.02 (s, 1H, H-4), 11.16 (s, 1H, NH). C NMR (DMSO-d6) δ (ppm): 171.9, 170.1, 166.9,
156.3, 149.1, 148.7, 148.0, 142.8, 136.6, 128.5, 124.4, 122.9, 122.6, 122.4, 117.6, 112.3, 56.0. MS
m/z (%): 375 (M+, 16), 295 (44), 294 (98), 204 (23), 203 (100), 187 (55), 119 (24).
1
N-(3’-Hydroxyphenyl)coumarin-3-carboxamide (23) Yield: 51%; mp: 283-284 ºC. H NMR (DMSO-d6)
δ (ppm), J (Hz): 6.53 (dd, 1H, H-4’, J=2.0, 8.1), 6.97-7.19 (m, 2H, H-8, H-5’), 7.31 (t, 1H, H-2’, J=2.0),
7.41-7.58 (m, 2H, H-6, H-7), 7.72-7.80 (m, 1H, H-5), 8.00 (dd, 1H, H-6’, J=2.0, 7.8), 8.88 (s, 1H, H-4),
13
9.55 (s, 1H, OH), 10.57 (s, 1H, NH). C NMR (DMSO-d6) δ (ppm): 161.3, 160.8, 156.7, 149.1, 147.0,
134.9, 129.9, 126.9, 125.1, 122.6, 119.8, 119.7, 119.0, 117.2, 114.1, 100.4. MS m/z (%): 281 (M+, 60),
253 (21), 173 (100), 101 (30), 89 (24).
This article is protected by copyright. All rights reserved.

N-(3’-Hydroxyphenyl)-6-methoxycoumarin-3-carboxamide (25) Yield: 57%; mp: 232-233 ºC 1H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, CH3), 6.51 (dd, 1H, H-4’, J=2.3, 8.0), 7.11 (t, 1H, H-5’, J=8.0),
7.26-7.37 (m, 2H, H-7, H-6’), 7 48 ( , 1H, H-8, J=9.1), 7.55 (t, 1H, H-2’, J=2.3), 8.83 (s, 1H, H-4), 9.55 (s,
13
1H, OH), 10.59 (s, 1H, NH). C NMR (DMSO-d6) δ (ppm): 161.6, 160.0, 157.8, 153.1, 147.4, 130.0,
125.7, 122.5, 120.0, 117.6, 111.7, 110.9, 108.9, 107.0, 106.7, 104.4, 56.6. MS m/z (%): 312 (M+, 32),
311 (100), 204 (62), 203 (96), 119 (60), 80 (18).
1
N-(3’-Methylphenyl)coumarin-3-carboxamide (26) Yield: 34%; mp: 206-207 ºC. H NMR (CDCl3) δ
(ppm), J (Hz): 2.38 (s, 3H, CH3), 6.98 (d, 1H, H-4’, J=7.4), 7.24 (dd, 1H, H-5, J=1.3, 7.4), 7.38-7.46 (m,
2H, H-6, H-7), 7.50-7.59 (m, 2H, H-8, H-6’), 7.66-7.75 (m, 2H, H-2’, H-5’), 9.02 (s, 1H, H-4), 10.79 (s,
13
1H, NH). C NMR (CDCl3) δ (ppm): 161.1, 159.1, 154.3, 148.8, 138.8, 134.2, 129.8, 128.8, 125.5,
125.4, 121.4, 117.5, 116.6, 103.6, 101.6, 98.4, 21.4. MS m/z (%): 279 (M+, 75), 173 (100), 101 (29), 89
(24).
1
N-(3’-Methylphenyl)-6-methoxycoumarin-3-carboxamide (28) Yield: 73%; mp: 186-187 ºC H NMR
(CDCl3) δ (ppm), J (Hz): 2.34 (s, 3H, CH3), 3.87 (s, 3H, OCH3), 7.07 (d, 1H, H-5, J=2.9), 7.17-7.28 (m, 2H,
H-2’, H-5’), 7 34 ( , 1H, H-8, J=9.1), 7.47-7.55 (m, 2H, H-7, H-6’), 8 92 (s, 1H, H-4), 10.81 (s, 1H, NH).
13C NMR (CDCl3) δ (ppm): 162.1, 159.2, 158.8, 148.6, 136.0, 135.9, 130.3, 128.8, 126.6, 125.0, 122.8,
121.7, 118.9, 117.7, 115.53, 110.52, 55.8, 18.0. MS m/z (%): 310 (M+, 22), 309 (86), 291 (51), 281
(30), 203 (100), 119 (49), 106 (72), 77 (17).
N-(3’-Methoxyphenyl)coumarin-3-carboxamide (29) Yield: 47%; mp: 189-190 ºC. 1H NMR (CDCl3) δ
(ppm), J (Hz): 3.85 (s, 3H, OCH3), 6.70-6.76 (m, 1H, H-4’), 7.21-7.32 (m, 2H, H-8, H-2’), 7 38-7.52 (m,
3H, H-6, H-7, H-5’), 7.67-7.77 (m, 2H, H-5, H-6’), 9.02 (s, 1H, H-4), 10.85 (s, 1H, NH). 13C NMR (CDCl3)
δ (ppm): 161.6, 160.0, 159.1, 154.3, 148.8, 138.7, 134.2, 129.8, 129.6, 125.4, 118.6, 118.5, 116.6,
112.7, 110.8, 109.8, 55.2. MS m/z (%): 295 (M+, 68), 267 (35), 187 (10), 173 (100), 101 (32), 89 (21).
N-(3’-Methoxyphenyl)-6-methoxycoumarin-3-carboxamide (31) Yield: 67%; mp: 193-194 ºC. 1H NMR
(CDCl3) δ (ppm), J (Hz): 3.81, 3.85 (s, 6H, 2 x OCH3), 6.68 (dt, 1H, H-4’, J=2.2, 7.0), 7.06 (d, 1H, H-5,
J=2.8), 7.18-7.28 (m, 3H, H-7, H-5’, H-6’), 7 34 ( , 1H, H-8, J=9.1), 7.44 (t, 1H, H-2’, J=2.2), 8.92 (s, 1H,
13
H-4), 10.87 (s, 1H, NH). C NMR (CDCl3) δ (ppm): 161.8, 160.0, 159.2, 156.6, 149.0, 148.6, 138.7,
This article is protected by copyright. All rights reserved.

129.6, 122.8, 118.9, 118.5, 117.7, 112.7, 110.7, 110.5, 105.8, 55.8, 55.2. MS m/z (%): 325 (M+, 53),
204 (100), 173 (94), 122 (30).
N-(3’-Chlorophenyl)coumarin-3-carboxamide (32) Yield: 60%; mp: 217-218 ºC. 1H NMR (DMSO-d6) δ
(ppm), J (Hz): 7.10 (dt, 1H, H-4’, J=1.5, 7.8), 7.37-7.49 (m, 2H, H-6, H-8), 7.53-7.59 (m, 2H, H-7, H-5’),
7.78 (t, 1H, H-2’, J=1.5), 7.96 (s, 1H, H-5), 8.01 (dd, 1H, H-6’, J=1.5, 7.8), 8.89 (s, 1H, H-4), 10.74 (s, 1H,
13
NH). C NMR (DMSO-d6) δ (ppm): 164.4, 160.9, 159.4, 147.7, 133.4, 132.7, 130.8, 130.5, 128.4,
125.5, 120.0, 119.6, 118.6, 116.4, 113.4, 101.0. MS m/z (%): 299 (M+, 67), 173 (100), 101 (36), 89
(30), 63 (13).
1
N-(3’-Chlorophenyl)-6-methylcoumarin-3-carboxamide (33) Yield: 61%; mp: 238-239 ºC H NMR
(DMSO-d6) δ (ppm), J (Hz): 2.36 (s, 1H, CH3), 7.17 (dt, 1H, H-4’, J=1.8, 8.0), 7.32-7.44 (m, 2H, H-8, H-
5’), 7 50-7.60 (m, 2H, H-7, H-6’), 7 75 ( , 1H, H-5, J=1.3), 7.93 (t, 1H, H-2’, J=1.8), 8.79 (s, 1H, H-4),
10.73 (s, 1H, NH). 13C NMR (DMSO-d6) δ (ppm): 174.7, 166.3, 160.5, 152.2, 149.2, 147.7, 135.5,
134.9, 133.4, 130.8, 129.9, 120.9, 119.6, 118.6, 118.2, 108.6, 20.4. MS m/z (%): 315 (M+1, 60), 313
(M+, 93), 188 (77), 187 (100), 115 (70), 103 (63), 89 (16), 77 (53), 63 (19).
1
N-(3’-Chlorophenyl)-6-methoxycoumarin-3-carboxamide (34) Yield: 48%; mp: 222-223 ºC H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, OCH3), 7.17 (dt, 1H, H-4’, J =1.9, 7.8), 7.30-7-38 (m, 2H, H-5, H-
7), 7.47 (d, 1H, H-8, J=9.1), 7.50-7.57 (m, 2H, H-5’, H-6’), 7 94 (t, 1H, H-2’, J=1.9), 8.83 (s, 1H, H-4),
10.76 (s, 1H, NH). 13C NMR (DMSO-d6) δ (ppm): 162.0, 160.5, 156.2, 149.3, 147.6, 139.5, 133.4,
130.8, 128.6, 127.7, 122.5, 119.6, 118.6, 117.6, 112.0, 103.6, 55.9. MS m/z (%): 331 (M+2, 15), 329
(M+, 45), 295 (54), 294 (99), 204 (37), 203 (100), 119 (60).
1
N-(3’-Bromophenyl)coumarin-3-carboxamide (35) Yield: 33%; mp: 232-233 ºC. H NMR (CDCl3) δ
(ppm), J (Hz): 7.21 (d, 1H, H-8, J=7.8), 7.26-7.31 (m, 1H, H-7), 7.40-7.47 (m, 2H, H-6, H-5’), 7 62 (dt,
1H, H-6’, J=1.8, 7.7), 7.68-7.76 (m, 2H, H-4’, H-5), 8.03 (d, 1H, H-2’, J=1.8), 9.02 (s, 1H, H-4), 10.88 (s,
13
1H, NH). C NMR (CDCl3) δ (ppm): 161.6, 160.9, 156.2, 154.4, 145.4, 144.0, 143.6, 141.2, 139.5,
138.8, 132.6, 129.9, 125.7, 125.5, 118.9, 106.1. MS m/z (%): 345 (M+2, 58), 343 (M+, 58), 174 (36),
173 (100), 101 (46), 90 (17), 89 (51), 63 (30).
This article is protected by copyright. All rights reserved.

1
N-(3’-Bromophenyl)-6-methoxycoumarin-3-carboxamide (37) Yield: 53%; mp: 234-235 ºC. H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.81 (s, 3H, OCH3), 7.28-7.33 (m, 2H, H-5’, H-7), 7.36 (d, 1H, H-5, J=3.0),
7.47 (d, 1H, H-8, J=9.0), 7.54-7.63 (m, 2H, H-4’, H-6’), 8 07 (s, 1H, H-2’), 8.82 (s, 1H, H-4), 10.74 (s, 1H,
NH). 13C NMR (DMSO-d6) δ (ppm): 161.5, 160.6, 159.1, 155.0, 145.9, 145.2, 143.6, 141.2, 138.8,
132.6, 129.9, 125.7, 125.5, 120.7, 118.9, 101.4, 56.0. MS m/z (%): 375 (M+1, 23), 374 (M+, 61), 203
(100), 187 (42), 119 (33).
1
N-(4’-Hydroxyphenyl)coumarin-3-carboxamide (38) Yield: 42%; mp: 261-262 ºC. H NMR (DMSO-d6)
δ (ppm), J (Hz): 6.75 (d, 2H, H-3’, H-5’, J=7.5), 7.41- 7.58 (m, 4H, H-7, H-8, H-2’, H-6’), 7 76 (t, 1H, H-6,
J=6.8), 8.01 (d, 1H, H-5, J=6.8), 8.89 (s, 1H, H-4), 9.37 (s, 1H, OH), 10.44 (s, 1H, NH). 13C NMR (DMSO-
d6) δ (ppm): 164.0, 161.0, 159.1, 152.0, 151.5, 144.2, 139.0, 135.1, 130.1, 126.4, 121.1, 120.3, 107.4.
MS m/z (%): 281 (M+, 59), 173 (100), 101 (17), 89 (20)
N-(4’-Hydroxyphenyl)-6-methoxycoumarin-3-carboxamide (40) Yield: 59%; mp: 242-243 ºC 1H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, OCH3), 7.23-7.60 (m, 5H, H-5, H-7, H-8, H-3’, H-5’), 7 80 ( , 2H,
13
H-2’, H-6’, J=8.4), 8.86 (s, 1H, H-4), 9.01 (s, 1H, OH), 10.76 (s, 1H, NH). C NMR (DMSO-d6) δ (ppm):
161.2, 159.0, 156.2, 148.6, 148.3, 146.8, 126.6, 124.5, 122.5, 120.1, 119.3, 119.2, 118.9, 117.5,
114.7, 112.0, 56.0. MS m/z (%): 312 (M+, 21), 311 (86), 204 (22), 203 (100), 119 (23).
1
N-(4’-Methylphenyl)coumarin-3-carboxamide (41) Yield: 88%; mp: 236-237 ºC H NMR (CDCl3) δ
(ppm), J (Hz): 2.27 (s, 3H, CH3), 7.17 (d, 2H, H-3’, H-5’, J=8.4), 7.45 (m, 1H, H-6), 7.53 (d, 1H, H-8, J=
8.3), 7.59 (d, 2H, H-2’, H-6’, J=8.4), 7.75 (m, 1H, H-7), 7.99 (dd, 1H, H-5, J=1.7, 8.0), 8.89 (s, 1H, H-4),
13
10.57 (s, 1H, NH). C NMR (CDCl3) δ (ppm): 160.4, 160.2, 159.8, 157.2, 147.5, 135.6, 134.4, 133.5,
130.4, 129.6, 122.5, 120.0, 118.7, 116.4, 20.6. MS m/z (%): 279 (M+, 54), 173 (100), 137 (53), 101
(19), 84 (21), 66 (22).
1
N-(4’-Methylphenyl)-6-methoxycoumarin-3-carboxamide (43) Yield: 61%; mp: 186-187 ºC H NMR
(CDCl3) δ (ppm), J (Hz): 2.24 (s, 3H, CH3), 3.72 (s, 3H, OCH3), 7.10-7.21 (m, 3H, H-5, H-2’, H-6’), 7 41 ( ,
1H, H-8, J=8.6), 7.52-7.61 (m, 3H, H-7, H-3’, H-5’), 8 79 (s, 1H, H-4), 10.56 (s, 1H, NH). 13C NMR (CDCl3)
This article is protected by copyright. All rights reserved.

δ (ppm): 162.1, 159.2, 158.8, 148.6, 136.0, 135.9, 130.3, 128.8, 126.6, 125.0, 122.8, 121.7, 118.9,
117.7, 115.5, 110.5, 55.80, 18.0. MS m/z (%): 309 (M+, 86), 291 (51), 281 (29), 203 (100), 119 (49),
106 (72).
1
N-(4’-Methoxyphenyl)coumarin-3-carboxamide (44) Yield: 74%; mp: 219-220 ºC H NMR (CDCl3) δ
(ppm), J (Hz): 3.96 (s, 3H, OCH3), 6.96 (d, 2H, H-3’, H-5’, J=8.3), 7.32-7.43 (m, 2H, H-6, H-8), 7.59-7.66
13
(m, 3H, H-7, H-2’, H-6’), 7 89 (m, 1H, H-5), 8.76 (s, 1H, H-4), 10.88 (s, 1H, NH). C NMR (CDCl3) δ
(ppm): 163.7, 160.5, 159.9, 158.3, 146.1, 134.2, 134.0, 133.0, 131.4, 129.9, 120.5, 120.1, 118.7,
109.8, 55.9. MS m/z (%): 295 (M+, 66), 187 (22), 173 (100), 101 (43), 89 (18).
1
N-(4’-Methoxyphenyl)-6-methoxycoumarin-3-carboxamide (46) Yield: 59%; mp: 201-202 ºC H NMR
(CDCl3) δ (ppm), J (Hz): 3.80, 3.85 (s, 6H, 2 x OCH3), 6.87-6.95 (m, 3H, H-5, H-3’, H-5’), 7 29 ( , 1H, H-
13
8, J=9.0), 7.44-7.58 (m, 3H, H-7, H-2’, H-6’), 8 34 (s, 1H, H-4), 10.46 (s, 1H, NH). C NMR (CDCl3) δ
(ppm): 162.9, 159.8, 159.3, 153.0, 148.6, 148.4, 130.1, 128.7, 120.0, 119.2, 118.4, 114.4, 112.4,
110.0, 105.1, 102.1, 55.8, 55.3. MS m/z (%): 325 (M+, 61), 203 (100), 173 (90), 108 (21).
1
N-(4’-Chlorophenyl)coumarin-3-carboxamide (47) Yield: 41%; mp: 264-265 ºC H NMR (DMSO-d6) δ
(ppm), J (Hz): 7.46 (d, 2H, H-3’, H-5’, J=7.5), 7.52-7.58 (m, 2H, H-6, H-8), 7.76 (d, 2H, H-2’, H-6’, J=7.5),
7.80-7.83 (m, 1H, H-7), 7.97 (d, 1H, H-5, J=8.3), 8.90 (s, 1H, H-4), 10.72 (s, 1H, NH). 13C NMR (DMSO-
d6) δ (ppm): 163.1, 160.8, 159.2, 140.7, 139.7, 136.7, 132.7, 130.0, 129.4, 120.3, 119.7, 111.5, 110.9,
110.3, 102.0, 101.9. MS m/z (%): 301 (M+2, 28), 299 (M, 59), 173 (100), 101 (33), 89 (29), 63 (15).
1
N-(4’-Chlorophenyl)-6-methylcoumarin-3-carboxamide (48) Yield: 42%; mp: 216-217 ºC H NMR
(DMSO-d6) δ (ppm), J (Hz): 2.36 (s, 3H, CH3), 7.35-7.45 (m, 3H, H-8, H-3’, H-5’), 7 57 ( , 1H, H-7,
J=2.0, 8.6), 7.69-7.79 (m, 3H, H-5, H-2’, H-6’), 8 79 (s, 1H, H-4), 10.70 (s, 1H, NH). 13C NMR (DMSO-d6)
δ (ppm): 165.7, 160.3, 152.2, 147.7, 137.4, 135.53, 134.9, 132.1, 130.0, 122.0, 119.9, 118.6, 116.2,
107.8, 20.5 MS m/z (%): 315 (M+2, 40), 313 (M+, 92), 188 (42), 187 (100), 115 (36), 103 (29), 77 (23).
1
N-(4’-Chlorophenyl)-6-methoxycoumarin-3-carboxamide (49) Yield: 49%; mp: 216-217 ºC H NMR
(DMSO-d6) δ (ppm), J (Hz): 3.80 (s, 3H, OCH3), 7.28-7.40 (m, 2H, H-3’, H-5’), 7 46 ( , 1H, H-8, J=9.1),
This article is protected by copyright. All rights reserved.

13
7.53 (d, 1H, H-5, J=2.8), 7.66-7.88 (m, 3H, H-7, H-2’, H-6’), 8.82 (s, 1H, H-4), 10.71 (s, 1H, NH). C
NMR (DMSO-d6) δ (ppm): 162.9, 160.9, 155.8, 149.1, 148.7, 139.5, 131.2, 131.1, 128.0, 126.3, 122.7,
119.1, 118.2, 117.8, 113.7, 100.4, 56.0. MS m/z (%): 329 (M+, 41), 295 (56), 294 (97), 204 (30), 203
(100), 119 (66), 76 (25).
1
N-(4’-Bromophenyl)coumarin-3-carboxamide (50) Yield: 34%; mp: 247-248 ºC H NMR (DMSO-d6) δ
(ppm), J (Hz): 7.44-7.49 (m, 1H, H-6), 7.52-7.57 (m, 3H, H-8, H-3’, H-5’), 7.72 (d, 2H, H-2’, H-6’, J=8.6),
13
7.71-7.77 (m, 1H, H-7), 8.00 (dd, 1H, H-5, J=1.5, 7.9), 8.89 (s, 1H, H-4), 10.70 (s, 1H, NH). C NMR
(DMSO-d6) δ (ppm): 168.6, 167.3, 148.0, 147.2, 145.4, 139.7, 136.2, 132.0, 129.7, 128.7, 117.4,
117.0, 111.1, 101.4. MS m/z (%): 345 (M+, 25), 343 (25), 173 (100), 101 (15), 89 (15).
1
N-(4’-Bromophenyl)-6-methoxycoumarin-3-carboxamide (52) Yield: 42%; mp: 269-270 ºC H NMR
(DMSO-d6) δ (ppm), J (Hz): 7.56-7.69 (m, 4H, H-7, H-8, H-3’, H-5’), 7 72 ( , 2H, H-2’, H-6’, J=8.4), 7.78
13
(d, 1H, H-5, J=1.2), 8.64 (s, 1H, H-4), 10.49 (s, 1H, NH). C NMR (DMSO-d6) δ (ppm): 163.4, 161.2,
158.9, 148.8, 144.4, 140.0, 136.6, 130.1, 129.7, 129.5, 128.4, 117.8, 100.6, 99.6. MS m/z (%): 376
(M+2, 27), 374 (69), 173 (100), 101 (22).
Pharmacology
The affinity of compounds 8–52 for the human AR subtypes hA1, hA2A, hA3, was determined with
radioligand competition experiments in Chinese hamster ovary (CHO) cells that were stably
transfected with the individual receptor subtypes. The radioligands used were 1.0 nM (2R,3R,4S,5R)-
2-(2-chloro-6-cyclopentylamino-purin-9-yl)-5-hydroxymethyl-tetrahydro-3,4-diol([3H]CCPA) for hA1,
10.0 nM (1-(6-amino-9H-purin-9-yl)-1-deoxy-N-ethyl-b-D-ribofuronamide) ([3H]NECA) for hA2A, and
1.0 nM 2-(1-hexynyl)-N6-methyladenosine [3H] ([3H]HEMADO) for hA3 receptors. The results were
expressed as Ki values (dissociation constants), which were calculated with the program GraphPad.
Due to the lack of a suitable radioligand for hA2B receptors, the potency of antagonists at the hA2B
receptor (expressed on CHO cells) was determined by inhibition of NECA-stimulated adenylyl cyclase
activity. The 50% inhibitory concentration (IC50) for inhibition of cAMP (cyclic adenosine
monophosphate) production was determined and converted to a Ki value using the Cheng and
Prusoff equation. The Ki values (Table 1) are reported as geometric means of three independent
experiments, with each tested concentration measured in duplicate. As an interval estimate for the
This article is protected by copyright. All rights reserved.

dissociation constants, 95% confidence intervals are given in parentheses. Details for
pharmacological experiments are described in a previous work.14
Adenosine receptor homology models
Homology models of the hA3 were previously developed by our group.15,16 Briefly, MOE software18
was used for the construction of the models and the hA3 sequence was aligned to our template, the
crystallized hA2A AR (PDB code: 3EML).19 The alignment was based on previous studies related to
adenosine homology modeling carried out by Katritch et al..20 The geometry of the hA3 model was
assessed with the Protein Geometry module.18 Ability to discriminate ligands from decoys was also
evaluated through ROC curves (area greater than 0.80).15,16,20 A more detailed description of the
published homology models can be found in our previous studies.15,16,21
Molecular docking
Molecular docking simulations in the hA2A and the hA3 were performed with Glide from the
Schrödinger software.22 For the hA3 docking the homology model previously described was used,
whereas the crystal structure 3EML (PDB code)19 was used for the docking in the hA2A. Protein
structures were pre-processed with the Protein Preparation Wizard workflow included in
Schrödinger.22 This process includes the assignation of bond orders, addition of cap termini,
optimization of protonation states of the residues, and optimization of the hydrogen-bond protein
network, among others. Ligands were prepared with the LigPrep module. No water molecules were
included in the simulations. The compounds were docked to the proteins with Glide SP scoring
function (standard precision).22 Binding modes described for graphical purposes were selected using
parameters such as Emodel as well as number of similar poses generated through the calculations.
Theoretical evaluation of drug-like properties
The drug-like properties of the compounds under study were calculated using the Molinspiration
property program. In this program, cLogP and topological polar surface area (TPSA) were calculated
as a sum of fragment-based contributions and correction factors. The calculation of molecule
volume has been performed by fitting the sum of f gment nt but ns t ‘ e ’ th ee mens n
(3D) volume for a training set of about 12 000 compounds, mostly drug-like molecules.23
This article is protected by copyright. All rights reserved.

Results and discussion
Chemistry
Coumarin carboxamides 8-52 were efficiently synthesised following the strategy shown in Scheme 1.
Generally, the compounds were obtained by an amidation reaction occurring between a coumarin-
3-carboxylic acid (5, 6 and 7) and the appropriate substituted amines using EDC as coupling reagent.
Compounds 8-52 were obtained in moderate to high yields (32% to 88%). As only coumarin-3-
carboxylic acid (5) was commercially available, 6-methylcoumarin-3-carboxylic (6) and 6-
methoxycoumarin-3-carboxylic (7) acids were prepared by a Knoevenagel condensation of 5-
methylsalicylaldehyde (1) and 5-methoxysalicylaldehyde (2), respectively, with diethyl malonate in
ethanol using piperidine as catalyst and subsequent basic hydrolysis of the corresponding ester
derivatives (compounds 3 and 4, respectively). The overall reaction yield was 89% for the methyl
substituted compound and 86% for its methoxy counterpart.
Pharmacology
The affinity of the coumarin carboxamides (compounds 8-52) for the human AR subtypes hA1, hA2A,
hA3, which were expressed in Chinese hamster ovary (CHO) cells, was determined in radioligand
competition experiments. In these assays, the competition with the following agonist radioligands:
(i) [3H]CCPA at hA1AR, (ii) [3H]NECA at hA2A and [3H]HEMADO at hA3 receptors was measured. The
resulting binding affinity data expressed as Ki (dissociation constant) is reported in Table 1. The data
regarding A2B AR was not included as none of the tested compounds revealed a measurable affinity
(Ki > 30 µM).
Structure‐affinity relationship studies
In an effort for validate coumarin as a privileged structure for the design of AR ligands our research
group acquired so far relevant data about scaffold recognized decorations (Figure 1). However, at
the present step it was found important to perform a systematic study to attain a reliable structure-
activity-relationship (SAR). Herein we present the first studies on coumarin decoration using
carboxamide as a spacer and different substituents at position 6 of aromatic coumarin ring and at
the exocyclic aromatic ring (Figure 2). The significance of the presence of a substituent at position 6
of the coumarin core was studied by the introduction of methyl or methoxy groups, as this position
was denoted as relevant in some previous publications of our research group.14,15 In addition, the
This article is protected by copyright. All rights reserved.

effect our studies were focused on the effect of electron-donating and electron-withdrawing groups
located at ortho, meta and para positions of the exocyclic aryl ring.
Generally, no relevant affinity data in any of ARs subtypes (Ki > 100 µM) was attained for the
majority of the coumarin derivatives under study. However, it must be stressed that the n results are
of the utmost importance to improve our understanding on the effect of spacers and substituents on
coumarin scaffold towards ARs subtypes. Accordingly, it can be concluded that the presence of
electron-withdrawing substituents in the exocyclic aromatic ring (17-22, 32-37, 47-52, Table 1) have
a negative outcome in the systems with absence or presence of substituents at the position 6 of
coumarin ring.
In spite of these results, some interesting data was attained with electron-donating substituents
allowing to draw some insights in the significance of 3-carboxamidocoumarin scaffold for the design
of AR ligands. Although the introduction of a methoxy group in aromatic ring located on the side
chain (14-16, 29-31, 44-46, Table 1), even in the presence or absence of substituents at the position
6 of coumarin ring substituents, was found to be not beneficial relevant data was attained by its
replacement by methyl substituent (11-13, 26-38, 41-43, Table 1). In fact, compound 26 (hA3 Ki = 2.4
µM) in which the group is located in meta position, display a noticeable activity and selectivity
towards hA3, which is weakened if the group is moved for ortho (compound 13, hA3 Ki = 45.4 µM) or
para position (compound 43, hA3 Ki >100 µM). However, the affinity is dramatically reduced in the
coumarins having methoxy or methyl substituents at the position 6.
For coumarins with a hydroxyl function in the exocyclic aromatic ring the data attained was not so
ruled. Compound 8, without any substituent in the coumarin core and a hydroxyl group located in
ortho position, behave as a selective A3AR ligand (hA3 Ki = 31.5 µM) Interestingly, it was noticed that
the introduction of a methyl group at position 6 of the coumarin led to a loss of activity (compound
9) whereas a methoxy group lead to a 30-fold decrease of the selectivity towards A3AR (compound
10, hA1/hA3 Ki = 1.21). In the case of compounds 25 and 40, with a hydroxyl group in the meta and
para position and a methoxy at the position 6 of coumarin, respectively, a loss of selectivity was
observed. Compound 25 has an A1 AR binding affinity of 41 µM and a A3 AR Ki of 22 µM, while
compound 40 has poor selectivity (hA1 Ki = 22.3 µM, hA2A Ki = 28.3 µM and hA3 Ki = 24.2 µM). The
same tendency was observed for compounds 10 (hA1 Ki = 39.5 µM, hA2A Ki = 38.0 µM and hA3 Ki =
32.7 µM) and 38 (hA1 Ki > 30 µM, hA2A Ki > 30 µM and hA3 Ki > 30 µM).
This article is protected by copyright. All rights reserved.

Theoretical drug-like properties
To have a prediction of the drug-like properties of the most promising compounds some
physicochemical parameters were calculated using the tool Molinspiration (Table 2).23 These
parameters include molecular weight (MW), number of heavy atoms (N), partition coefficient
(clogP), topological polar surface area (tPSA in Å2), number of hydrogen bond acceptors (HBA),
number of hydrogen bond donors (HBD), and number of rotatable bonds (nrotb) and volume (Å3)
(Table 2). For the coumarins under study no violation of Lipinski's rule of five (MW, log P, number of
hydrogen donors and acceptors) were found. Moreover, the TPSA values, described as a predictive
indicator of the drug capacity of membrane penetration, are encouraging for pursuing a drug-like
lead. Consequently, the data represented in Table 2 provides a preliminary indication that these
type of compounds have drug-like properties.
Molecular docking simulations
Molecular docking simulations in the hA2A and the hA3 were carried out to study adenosine
selectivity and provide some insights in the relationship between the molecular structure and the
protein affinity. The crystal structure 3EML for the hA2A and a homology model for the hA3 were
used in the simulations. Ligands were docked with Glide SP.20 In previous studies15 our docking
protocol was validated and a root mean square deviation (RMSD) of 0.69 and 1.92 between the co-
crystallized and docking conformations for the ligands ZM241385 and T4E inside the hA2A pocket was
obtained.19,24
Our studies have been initially focused on compound 26 as a significant selective binding affinity
against the hA3 (Ki = 2.4 µM, Table 1) has been attained. Molecular docking in the hA3 yielded a pose
for compound 26 in which the coumarin ring is oriented towards the bottom of the pocket whereas
the 3’-methylphenyl group is located towards the extracellular area (see Figure 3a). The oxygen in
the pyrone ring establishes a hydrogen bond with the amide group of the residue Asn250.
Interactions with this residue have been already described in the literature for the different
adenosine receptors.15,16,25 Moreover, the benzene ring in the coumarin nucleus of compound 26
est b shes π-π st k ng nte t ns th the es ue 243 In t n, the nt but ns f the
different residues to the binding of the ligand were also measured (see Figure 3b). The residue
contribution score was calculated as the addition of van der Waals and Coulomb energies. The key
contribution of some residues in the recognition of the ligand, such as Phe168, Leu246, Met177,
Ile268, Trp243, Asn250 and Ala69 are shown in Figure 3b. However, other type of interactions, such
This article is protected by copyright. All rights reserved.

as hydrophobic interactions, could also be important to explain the different affinity between the
compound and the protein. In fact, compound 26 placed the coumarin ring in a deep hydrophobic
e n the 3’-methylphenyl is inserted also in a hydrophobic region (see Figure 3a with the
hydrophobic/hydrophilic surface calculated in the hA3 pocket).
The introduction of a hydroxyl group at ortho position of the phenyl group (compound 8) lead to a
dramatic decrease of the hA3 binding affinity (Ki = 31.5 µM, Table 1), compared to compound 26. The
docking pose for compound 8 e e h gen b n th the es ue Asn250 n π-π st k ng
nte t ns th the Phe168 A se n h gen b n s ete te bet een the 2’-hydroxyl group
in the phenyl exocyclic ring and the residue Gln167 of the second extracellular loop (see Figure 3c).
This fact could be a key factor to explain the hA3 selectivity shown by compound 8, since the other
adenosine receptors do not present a glutamine residue (Gln) at the same position. However, the
coumarin ring is placed in the hA3 in a shallower hydrophobic region compared to compound 26 that
could be responsible for the decrease of hA3 activity. Docking studies have also been accomplished
for compound 10 and from the pose extracted it was verified that the compound can establish a
h gen b n th the es ue Asn250 n π-π st k ng nte t ns th the Phe168 Yet, the
binding mode did not yield a hydrogen bond with the residue Gln167. However, the
hydrophobic/hydrophilic surface of the receptor can accommodate the methoxy substituent in the
hydrophobic area at the bottom of the cavity (see Figure 3d).
As compound 10 also exhibited a moderate affinity against hA1 and hA2A additional computational
studies were performed. Since the crystal structure of hA2A is available docking simulations were
carried out to explain the increment in the affinity for hA2A compared to the other compounds in the
series. The simulations showed that compound 10 has a similar binding mode inside the hA2A (see
Figure 4a).
he g n est b shes h gen b n th the es ue Asn253 M e e , π-π st k ng
interactions have been detected between the benzene ring in the coumarin nucleus and the
imidazole of the residue His250. The 6-methoxy substituent was found to be well accommodated in
a hydrophobic region, a pose that could be accountable for the observed increase on the hA2A
affinity compared to compounds 8 and 9 M e e , the 2’-hydroxyl substituent in the phenyl ring is
oriented towards a hydrophilic area. The replacement in that position of a hydroxyl group by
hydrophobic substituents, such as methyl or methoxy groups in compounds 13 and 16, is not
suitable for the interaction with the hA2A with the consequent loss of affinity. Compound 26 showed
a similar pose as compound 10. However, the lack of the 6-methoxy substituent in the deep
h h b eg n ng th the s t n f the 3’-methyl group of the phenyl ring in an area not
This article is protected by copyright. All rights reserved.

well defined as hydrophobic could be the reason for the decreased hA2A affinity (see Figure 4b) of
compounds 13 and 16, is not be suitable for the interaction with the hA2A with the consequent loss
of affinity.
Conclusions
In this study, evidences for the validation of 3-phenylcarboxamidocoumarin as a scaffold for the
development of AR ligands have been acquired. Due to coumarin synthetic accessibility and
decoration capability a small library has been successfully attained and a concise SAR study
performed. Although the majority of the compounds were not active for any of AR subtypes the data
can be useful for validate the ligand requirements. In general, a loss of activity/selectivity was found
for coumarins substituted in 6-position and with electron- withdrawing substituents in the aryl
exocyclic ring. From the study interesting remarks must be highlighted: compound 26 displayed a
relevant affinity and selectivity for hA3 (Ki = 2.4 µM), turning it in the most stimulating compound of
the series. Its hA3 selectivity was elucidated by docking experiments that indicate that the coumarin
ng s ente t s the b tt m f the ket he e s the 3’-methylphenyl group is located
towards the extracellular area and that the oxygen in the pyrone ring establishes a hydrogen bond
with the amide group of the residue Asn250.
Taking into account that A1 and A3 AR ligands are beneficial for the treatment of disorders of the
nervous system, such as chronic pain, neurodegeneration and brain injury, compound 26 can be
considered an interesting starting point for further studies aiming the development of effective and
selective coumarin-based AR ligands.
In summary, the data suggests that for this type of compounds the presence of substituents at
position 6 can be detrimental for the AR affinity constituting per se an important tool for building
upon the coumarin hits.
Acknowledgments
The authors would like to thank Fundação para a Ciência e Tecnologia (FCT) – QUI/UI0081/2013,
POCI-01-0145-FEDER-006980 for the financial support. Thanks are due to FCT, POPH and QREN for A.
Fonseca (SFRH/BD/80831/2011) and M. J. Matos (SFRH/BPD/95345/2013) doctoral and post-
doctoral grants. S. Vilar s th nks “Ange es A ñ g m, P n G eg e In est g ón,
Innovación e Crecemento 2011-2015 (I2C)” n “Eu e n S Fun (ESF)”
This article is protected by copyright. All rights reserved.

Conflict of interest
Auth s e e n nfl t f nte est
Figure 1. General structure of coumarin based derivatives described as AR ligands.12–15
Figure 2. Rational design followed in the present study.
Figure 3. a) Hypothetical binding mode extracted from molecular docking for compound 26 in the
hA3. Hydrogen bond with the residue Asn250 is shown in yellow dashes. Hydrophobic areas in the
pocket are represented in yellow and hydrophilic regions in red color. b) Residue contributions to
the binding between hA3 and compound 26 (sum of van der Waals and Coulomb energies). c)
Pose extracted for compound
8 in the hA3. Hydrogen bonds with the residues Asn250 and Gln167
are shown in yellow dashes. d) Hypothetical binding mode calculated for compound 10 in the hA3
(hydrophobic surface in yellow and hydrophilic in red color).
Figure 4. a) Hypothetical binding mode calculated with molecular docking for compound 10 in the
hA2A. Hydrogen bonds are represented in yellow dashes, hydrophobic surface in green color and
hydrophilic surface in magenta. b) Pose calculated with docking for the compound 26 in the hA2A.
References
1.
2.
3.
Drury, N.; Szent-Györgyi, J. Physiol. 1929, 68, 213–237.
Chen, J. F.; Eltzschig, H. K.; Fredholm, B. B. Nat. Rev. Drug Discov. 2013, 12, 265–286.
Fredholm, B. B.; Jzerman, A. P.; Jacobson, K. A.; Klotz, K.-N.; Linden, J. Pharmacol. Rev. 2001,
53, 527–552.
4.
5.
Bours, M. J. L.; Swennen, E. L. R.; Di Virgilio, F.; Cronstein, B. N.; Dagnelie, P. C. Pharmacol.
Ther. 2006, 112, 358–404.
Fredholm, B. B. Exp. Cell Res. 2010, 316, 1284–1288.
This article is protected by copyright. All rights reserved.

6.
Gessi, S.; Cattabriga, E.; Avitabile, A.; Gafa, R.; Lanza, G.; Cavazzini, L.; Bianchi, N.; Gambari,
R.; Feo, C.; Liboni, A.; Gullini, S.; Leung, E.; Mac-Lennan, S.; Borea, P. A. Clin. Cancer Res. 2004,
10, 5895–5901.
7.
Jacobson, K. A.; Gao, Z. Nat. Rev. Drug Discov. 2006, 5, 247–264.
8.
Fonslow, B. R.; Stein, B. D.; Webb, K. J.; Xu, T.; Choi, J.; Kyu, S.; Iii, J. R. Y. 2013, 10, 54–56.
Press, N. J.; Fozard, J. R. Expert Opin. Ther. Pat. 2010, 20, 987–1005.
9.
10.
Borges, F.; Roleira, F.; Milhazes, N.; Santana, L.; Uriarte, E. Curr. Med. Chem. 2005, 12, 887–
916.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Borges, F.; Roleira, F. M. F.; Milhazes, N. J. da S. P.; Uriarte, E.; Santana, L. Front. Med. Chem.
2009, 4, 23–85.
Matos, M. J.; Vilar, S.; Kachler, S.; Celeiro, M.; Vazquez-Rodriguez, S.; Santana, L.; Uriarte, E.;
Hripcsak, G.; Borges, F.; Klotz, K.-N. Bioorg. Chem. 2015, 61, 1–6.
Vazquez-Rodriguez, S.; Matos, M. J.; Santana, L.; Uriarte, E.; Borges, F.; Kachler, S.; Klotz, K. N.
J. Pharm. Pharmacol. 2013, 65, 697–703.
Matos, M. J.; Hogger, V.; Gaspar, A.; Kachler, S.; Borges, F.; Uriarte, E.; Santana, L.; Klotz, K.-N.
J. Pharm. Pharmacol. 2013, 65, 1590–1597.
Matos, M. J.; Vilar, S.; Kachler, S.; Fonseca, A.; Santana, L.; Uriarte, E.; Borges, F.; Tatonetti, N.
P.; Klotz, K.-N. ChemMedChem 2014, 9, 2245–2253.
Cagide, F.; Gaspar, A.; Reis, J.; Chavarria, D.; Vilar, S.; Hripcsak, G.; Uriarte, E.; Kachler, S.;
Klotz, K. N.; Borges, F. RSC Adv. 2015, 5, 78572–78585.
Fonseca, A.; Matos, M. J.; Reis, J.; Duarte, Y.; Santana, L.; Uriarte, E.; Borges, F. RSC Adv. 2016,
6, 49764–49768.
MOE, version 2011.10; Chemical Computing Group, Inc.: Available at:
http://www.chemcomp.com. (Accessed: Jan 2012).
Veli-Pekka, J.; Mark, G.; Michael, H.; Vadim, C.; Ellen, Y.; Robert, L.; Adriaan, P. I.; Raymond, C.
S. Science. 2008, 322, 1211–1217.
Katritch, V.; Kufareva, I.; Abagyan, R. Neuropharmacology 2011, 60, 108–115.
This article is protected by copyright. All rights reserved.

21.
22.
Matos, M. J.; Vilar, S.; Kachler, S.; Vazquez-Rodriguez, S.; Varela, C.; Delogu, G.; Hripcsak, G.;
Santana, L.; Uriarte, E.; Klotz, K.-N.; Borges, F. Med. Chem. Commun. 2016, 7, 845–852.
Schrödinger suite 2016-3, Schrödinger, LLC, New York, USA, 2016. Available at:
http://www.schrodinger.com/ (Accessed: August 2016).
23.
24.
Lipinski, C.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Deliv. Rev. 2012, 64, 4–17.
Congreve, M.; Andrews, S. P.; Doré, A. S.; Hollenstein, K.; Hurrell, E.; Langmead, C. J.; Mason,
J. S.; Ng, I. W.; Tehan, B.; Zhukov, A.; Weir, M.; Marshall, F. H. J. Med. Chem. 2012, 55, 1898–
1903.
25.
26.
Kolb, P.; Phan, K.; Gao, Z. G.; Marko, A. C.; Sali, A.; Jacobson, K. A. PLoS One 2012, 7, e49910.
Klotz, K.-N.; Hessling, J.; Hegler, J.; Owman, C.; Kull, B.; Fredholm, B. B.; Lohse, M. J. Naunyn.
Schmiedebergs. Arch. Pharmacol. 1997, 357, 1–9.
Scheme 1. Synthetic strategy followed for the preparation of coumarin 3-phenylcarboxamides.
Reagents and conditions: a) diethyl malonate, EtOH, piperidine, reflux, overnight. b) NaOH (0.5%
aq./EtOH), reflux, 2 h. c) EDC, DMAP, DCM, corresponding amine, 0 ºC to r.t., 4 h
This article is protected by copyright. All rights reserved.

Table 1. Binding affinity (Ki in µM and 95% confidence intervals in parentheses) of
coumarins 8-52 in radioligand binding assays at human A1, A2A, A3 AR subtypes.
Selectivity
hA1
hA2A
hA3
Compound
R
R1
Ki (µM)
Ki (µM)
Ki (µM)
hA1/hA3 hA2A/hA3
31.5
(2.39 – 4.16)
> 30
8
H
2-OH
2-OH
2-OH
> 100
> 100
> 30
>3.17
_
>0.95
_.
917
10
CH3
> 100
39.5
38.0
32.7
OCH3
1.21
1.16
(34.3 – 45.4)
(34.5 -41.9)
(27.6 – 38.8)
11
H
2-CH3
2-CH3
> 100
> 100
> 100
> 100
> 100
> 100
_
_
_
_
1217
CH3
45.4
13
OCH3
2-CH3
> 100
> 100
>2.20
>2.20
(38.7 – 53.3)
14
1517
16
H
CH3
OCH3
H
2-OCH3
2-OCH3
2-OCH3
2-Cl
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
> 100
17
> 100
18
CH3
OCH3
H
2-Cl
> 100
19
2-Cl
> 100
20
2-Br
> 100
2117
22
CH3
OCH3
H
2-Br
> 100
2-Br
> 100
23
3-OH
3-OH
> 100
2417
CH3
> 100
41.0
22.0
25
OCH3
3-OH
> 100
1.86
>4.54
(15.3 – 31.7)
(29.0 – 58.1)
2.4
26
H
3-CH3
3-CH3
> 100
> 100
> 100
> 100
>41.67
_
>41.67
_
(1.81 – 3.19)
> 100
2717
CH3
This article is protected by copyright. All rights reserved.

28
29
OCH3
H
3-CH3
3-OCH3
3-OCH3
3-OCH3
3-Cl
> 100
> 100
> 100
> 100
> 30
> 100
> 100
> 100
> 100
> 30
> 100
> 100
> 100
> 100
> 30
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
3017
31
CH3
OCH3
H
32
33
CH3
OCH3
H
3-Cl
> 30
> 30
> 30
34
3-Cl
> 100
> 30
> 100
> 30
> 100
> 100
> 10
35
3-Br
3617
37
CH3
OCH3
H
3-Br
> 10
> 10
3-Br
> 100
> 30
> 100
> 30
> 100
> 30
38
4-OH
4-OH
3917
CH3
> 100
> 100
> 100
22.3
(21.5 – 23.2)
> 100
28.3
(25.2 – 31.8)
> 100
24.2
(22.2 – 26.4)
> 100
40
OCH3
4-OH
0.79
1.17
41
4217
43
H
CH3
OCH3
H
4-CH3
4-CH3
_
_
_
_
_
_
_
_
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 30
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 30
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 100
> 30
4-CH3
44
4-OCH3
4517
CH3
4-OCH3
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
46
47
OCH3
H
4-OCH3
4-Cl
48
CH3
OCH3
H
4-Cl
49
4-Cl
50
4-Br
4-Br
4-Br
5117
52
CH3
OCH3
> 100
> 100
> 100
> 100
> 100
> 100
6.77
86.40
theophylline26
-
0.08
1.2
(4.07 -11.30)
(73.60 – 101.30)
This article is protected by copyright. All rights reserved.

Table 2. Drug-like properties of the most promising coumarins.
Compd
8
MW
N
clogP
tPSA (Å2)
HBA
HBD
nrotb
Vol (Å3)
281.27 21
311.29 23
295.29 22
311.29 23
279.30 21
311.29 23
2.56
79.54
5
2
3
239.4
10
13
25
26
40
2.60
2.84
2.36
3.26
2.39
88.77
68.54
88.77
59.31
88.77
6
5
6
4
6
2
1
2
1
2
4
4
3
2
3
264.9
256.9
264.94
247.94
264.94
Figure 1. General structure of coumarin based derivatives described as AR ligands.12–15
This article is protected by copyright. All rights reserved.

Figure 2. Rational design followed in the present study.
Figure 3. a) Hypothetical binding mode extracted from molecular docking for compound 26 in the
hA3. Hydrogen bond with the residue Asn250 is shown in yellow dashes. Hydrophobic areas in the
pocket are represented in yellow and hydrophilic regions in red color. b) Residue contributions to
the binding between hA3 and compound 26 (sum of van der Waals and Coulomb energies). c)
Pose extracted for compound
8 in the hA3. Hydrogen bonds with the residues Asn250 and Gln167
are shown in yellow dashes. d) Hypothetical binding mode calculated for compound 10 in the hA3
(hydrophobic surface in yellow and hydrophilic in red color).
This article is protected by copyright. All rights reserved.

Figure 4. a) Hypothetical binding mode calculated with molecular docking for compound 10 in the
hA2A. Hydrogen bonds are represented in yellow dashes, hydrophobic surface in green color and
hydrophilic surface in magenta. b) Pose calculated with docking for the compound 26 in the hA2A.
This article is protected by copyright. All rights reserved.