Novel positive allosteric modulators of A2B adenosine receptor acting as bone mineralisation promoters

Article abstract of DOI:10.1080/14756366.2020.1862103

Small-molecules acting as positive allosteric modulators (PAMs) of the A_2B adenosine receptor (A_2B AR) could potentially represent a novel therapeutic strategy for pathological conditions characterised by altered bone homeostasis, including osteoporosis. We investigated a library of compounds (4-13) exhibiting different degrees of chemical similarity with three indole derivatives (1-3), which have been recently identified by us as PAMs of the A_2B AR able to promote mesenchymal stem cell differentiation and bone formation. Evaluation of mineralisation activity of 4-13 in the presence and in the absence of the agonist BAY60-6583 allowed the identification of lead compounds with therapeutic potential as anti-osteoporosis agents. Further biological characterisation of one of the most performing compounds, the benzofurane derivative 9, confirmed that such a molecule behaves as PAM of the A_2B AR.

Full text of DOI:10.1080/14756366.2020.1862103

Journal of Enzyme Inhibition and Medicinal Chemistry
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/ienz20
Novel positive allosteric modulators of A2B
adenosine receptor acting as bone mineralisation
promoters
Elisabetta Barresi , Chiara Giacomelli , Laura Marchetti , Emma Baglini ,
Silvia Salerno , Giovanni Greco , Federico Da Settimo , Claudia Martini ,
Maria Letizia Trincavelli & Sabrina Taliani
To cite this article: Elisabetta Barresi , Chiara Giacomelli , Laura Marchetti , Emma Baglini ,
Silvia Salerno , Giovanni Greco , Federico Da Settimo , Claudia Martini , Maria Letizia Trincavelli
& Sabrina Taliani (2021) Novel positive allosteric modulators of A2B adenosine receptor acting
as bone mineralisation promoters, Journal of Enzyme Inhibition and Medicinal Chemistry, 36:1,
286-294, DOI: 10.1080/14756366.2020.1862103
To link to this article: https://doi.org/10.1080/14756366.2020.1862103
© 2020 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group.
Published online: 17 Dec 2020.
Article views: 48
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ienz20

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
2021, VOL. 36, NO. 1, 286–294
https://doi.org/10.1080/14756366.2020.1862103
BRIEF REPORT
Novel positive allosteric modulators of A2B adenosine receptor acting as bone
mineralisation promoters
a
a
ꢀ
ꢀ
Elisabetta Barresi
, Chiara Giacomelli
, Laura Marchettia , Emma Baglinia , Silvia Salernoa
,
Giovanni Grecob , Federico Da Settimoa , Claudia Martinia , Maria Letizia Trincavellia
Sabrina Taliania
and
b
aDepartment of Pharmacy, University of Pisa, Pisa, Italy; Department of Pharmacy, University of Naples “Federico II”, Naples, Italy
ABSTRACT
ARTICLE HISTORY
Received 9 November 2020
Revised 27 November 2020
Accepted 1 December 2020
Small-molecules acting as positive allosteric modulators (PAMs) of the A2B adenosine receptor (A2B AR)
could potentially represent a novel therapeutic strategy for pathological conditions characterised by
altered bone homeostasis, including osteoporosis. We investigated a library of compounds (4-13) exhibit-
ing different degrees of chemical similarity with three indole derivatives (1-3), which have been recently
identified by us as PAMs of the A2B AR able to promote mesenchymal stem cell differentiation and bone
formation. Evaluation of mineralisation activity of 4-13 in the presence and in the absence of the agonist
BAY60-6583 allowed the identification of lead compounds with therapeutic potential as anti-osteoporosis
agents. Further biological characterisation of one of the most performing compounds, the benzofurane
derivative 9, confirmed that such a molecule behaves as PAM of the A2B AR.
KEYWORDS
A2B adenosine receptor;
allosteric modulators;
mesenchymal stem cell;
bone formation;
BAY60-6583
Introduction
cAMP, calcium or potassium channels, phospholipase C, phospho-
lipase D and the mitogen-activated protein kinases9–12. The modu-
lation of these pathways significantly affects the mesenchymal
stem/stromal cell (MSC) biology by controlling their fates13.
MSCs are multipotent cells able to differentiate into osteoblasts
or adipocytes. MSCs express various AR subtypes based on the dif-
ferentiation lineage and stages13,14. Human bone marrow-derived
MSCs (BM-MSCs) are the main sources for newly osteoblasts in
bone. The extracellular adenosine acts as an autocrine/paracrine
hormone by promoting the differentiation of MSCs to mature
osteoblasts which results from activation of the A2B AR15–17. The
role of A2B AR in osteoblast differentiation has been confirmed by
experiments demonstrating that the deletion of this receptor in a
mouse model produces a defective bone formation18. In addition,
the A2B AR negatively controls the MSCs differentiation to
adipocyte19.
Osteoporosis is a widespread systemic and chronic disorder char-
acterised by decreased bone mass and micro-architectural deteri-
oration, which predisposes to low-energy fractures. It affects the
health of many adults, with an incidence that increases with age
and varies among genders, being the postmenopausal women at
particularly high risk1,2
.
Several approaches have been proposed for the management
of osteoporosis, in terms of prevention and/or treatment, all final-
ised to the achievement of normal peak bone mass, including an
adequate dietary regimen, calcium and vitamin D supplementa-
tion. Among the pharmacological agents, bisphosphonates, calci-
tonin, oestrogens, inhibitors of receptor activator of nuclear factor
jB ligand, parathyroid hormone and parathyroid hormone–related
peptide should be mentioned, all possessing one or more draw-
backs in terms of efficacy, safety, bioavailability, drug–drug inter-
In 2013 and 2014, we disclosed three indole derivatives (1, 2
and 3, Table 1) acting as positive allosteric modulators (PAMs) of
the A2B AR20–22. None of these compounds bind with appreciable
affinity to the orthosteric site of each of the ARs. Moreover, they
increase efficacy of adenosine or synthetic agonists without affect-
ing their potency. To our knowledge, compounds 1, 2 and 3 are
actions and costs3–5
.
The increasing knowledge about the numerous signalling path-
ways involved in bone and osteocytes, as well as osteoblasts and
osteoclasts turnover, has encouraged research projects aimed to
investigate innovative therapeutic options for the treatment of
osteoporosis.
still the only PAMs of A2B AR reported in literature20,21
.
In this line, emerging evidences support the role of extracellu-
lar adenosine in bone homeostasis6–8. Adenosine is ubiquitous
and regulates several physiological and pathological conditions
through the activation of specific G-protein coupled receptors,
namely the A1, A2A, A2B and A3 adenosine receptors (ARs)9. The
binding of adenosine to its receptors causes the activation of dif-
Subsequent studies on 2, chosen as the most promising of the
three, according to in vitro experiments20,21, revealed that this
compound promotes MSC differentiation and bone formation by
inducing the differentiation of BM-MSCs to osteoblasts16.
These data suggest that the use of small molecules able to
produce a positive A2B AR allosteric modulation could potentially
ferent intracellular signalling pathways such as those involving represent a novel therapeutic strategy not only for osteoporosis
CONTACT Sabrina Taliani
sabrina.taliani@unipi.it; Maria Letizia Trincavelli
maria.trincavelli@unipi.it
Department of Pharmacy, University of Pisa, Via
Bonanno Pisano, 6, Pisa 56126, Italy
ꢀ
These two authors contributed equally to this work.
ß 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
287
20,21
Table 1. Structures of the reference indoles 1-3
and novel compounds 4–13.
1
2
3
4
5
6
8
9
7
12
10
11
13

288
E. BARRESI ET AL.
but also for all the pathological conditions characterised by an commercially available (Ambinter); its structure was confirmed by
excessive bone destruction and/or a decrease in bone formation 1H-NMR and 13 C-NMR.
(e.g. fracture mal-union, osteogenesis imperfecta, rheumatoid arth-
ritis and multiple myeloma), as well as in pathology characterised
N-Benzyl-2-(1-methyl-1H-pyrrol-2-yl)-glyoxylamide (11)
1H-NMR (DMSO-d6, ppm): 3.92 (s, 3H); 4.39 (s, 2H); 6.19-6.21 (m,
1H); 7.26–7.36 (m, 7H); 9.22 (bs, 1H). 13 C-NMR (DMSO-d6, ppm):
37.60; 42.41; 109.54; 124.75; 127.39; 127.68; 127.76; 128.80; 134.77;
139.34; 164.74; 178.74.
by
pain, CIP)23.
These results prompted a deep investigation to identify new
a high bone fragility (e.g. congenital insensitivity to
lead compounds, structurally related to 1, 2 and 3, endowed with
the ability to allosterically enhance A2B AR agonists and to stimu-
late matrix mineralisation when used alone or in the presence of
the orthosteric agonist BAY60-6583. In this respect, the present
study was aimed at gaining information about the pharmaco-
phoric requirements of this novel class of mineralisation agents
and hopefully identifying new compounds with improved activity
as compared to 2. Thus, we selected ten compounds, whose struc-
tures are reported in Table 1, exhibiting different degrees of
chemical similarity with the reference indole derivatives 1, 2
and 3.
General procedure for the synthesis of N-(indol-3-yl)glyoxylamides
4-7 and 3-(phenylglyoxylyl)indoles 12-13
Sodium hydride (60% dispersion in mineral oil, 44 mg, 1.10 mmol)
was portionwise added to an ice-cooled solution of 16,1724 or 26
(1.10 mmol) in 5 ml of anhydrous DMF. The resulting mixture was
stirred at room temperature for 1 h. Then, the appropriate halide
((2-bromoethyl)benzene for 4, 5 and 12 or methyl iodide for 6, 7
and 13, 1.10 mmol) was added and stirring was continued over-
night. The reaction mixture was diluted with water and ice and
the formed precipitate was collected by filtration and purified by
The library includes novel indole-based derivatives featuring
modified substitution patterns at 1- and/or 3-positions, as well as
compounds in which the indole scaffold was replaced by different
aromatic heterocycles. Specifically, compounds 4-7 feature side recrystallization from ethanol (for 4-7) or by flash chromatography
chains at 3-position of the indole nucleus corresponding to those (for 12-13, petroleum ether 60–80 ꢃC/ethyl acetate in ratio 8:2 as
of
2 (N-phenylglyoxylylamide) or 1 (N-benzylglyoxylylamide), eluting system).
whereas compounds 12 and 13 feature the phenylglyoxylyl side
chain present in 3. In compounds 4-7 and 12-13 the indole
nucleus was maintained and was decorated with a phenylethyl (4,
5 and 12) or a methyl (6, 7 and 13) moiety at 1-position in place
of the benzyl group of 1-3; instead, in derivatives 8-11 the indole
was replaced by different heterocycles such as benzothiophene
(8), benzofurane (9), thiophene (10) and N-methylpyrrole (11).
With the exception of the commercially available 11, the above
compounds were easily prepared by means of experimental proto-
cols modified according to reported procedures or set up ex novo
in order to obtain the desired products with high yield and pur-
ity degree.
N-Phenyl-2-(1-phenylethyl-1H-indol-3-yl)glyoxylamide (4)
Yield: 61%; mp 139–140 ꢃC; 1H-NMR (DMSO-d6, ppm): 3.14 (t, 2H,
J ¼ 7.4 Hz); 4.59 (t, 2H, J ¼ 7.4 Hz); 7.19-7.21 (m, 2H); 7.26-7.32 (m,
4H); 7.34-7.41 (m, 4H); 7.71-7.74 (m, 1H); 7.87 (d, 2H, J ¼ 8.0 Hz);
8.31-8.33 (m, 1H); 8.75 (s, 1H); 10.67 (s, 1H). 13 C-NMR (DMSO-d6,
ppm): 36.04; 48.28; 111.43; 111.84; 120.69; 121.95; 123.53; 124.11;
124.71; 127.00; 127.20; 128.84; 129.20; 129.33; 136.74; 138.40;
138.53; 141.59; 162.74; 182.05.
N-Benzyl-2-(1-phenylethyl-1H-indol-3-yl)glyoxylamide (5)
Yield: 67%; mp 83–85 ꢃC; 1H-NMR (DMSO-d6, ppm): 3.12 (t, 2H,
J ¼ 7.4 Hz); 4.42 (d, 2H, J ¼ 6.4 Hz); 4.56 (t, 2H, J ¼ 7.4 Hz); 7.18–7.37
(m, 12H); 7.69–7.71 (m, 1H); 8.26–8.28 (m, 1H); 8.75 (s, 1H); 9.29 (t,
1H, J ¼ 6.4 Hz). 13 C-NMR (DMSO-d6, ppm): 36.01; 42.53; 48.23;
111.64; 111.72; 122.01; 123.37; 123.97; 126.99; 127.26; 127.35;
127.82; 128.78; 128.83; 129.31; 136.60; 138.41; 139.48; 141.54;
163.92; 182.06.
The present paper describes the evaluation of compounds
4–13 as novel bone mineralisation agents and the resulting struc-
ture-activity relationships.
Materials and methods
Chemistry
Uncorrected melting points were determined using a Reichert
Kofler hot-stage apparatus. NMR spectra were obtained on a
Bruker AVANCE 400. The coupling constants (J) are given in Hertz.
Magnesium sulphate was used as the drying agent. Evaporations
were made in vacuo (rotating evaporator). Analytical TLC have
been carried out on Merck 0.2 mm precoated silica gel aluminium
sheets (60 F-254). Silica gel 60 (230–400 mesh) was used for col-
umn chromatography. Purity of the target inhibitors 4–13 was
determined, using a Shimadzu LC-20AD SP liquid chromatograph
equipped with a DDA Detector (k ¼ 254 nm) using a column C18
(250 mm ꢁ 4.6 mm, 5 mm, Shim-pack); the mobile phase, delivered
at isocratic flow, consisted of acetonitrile (60-70%) and water (40-
30%); flow rate 1 ml/min. All the compounds showed percent pur-
ity ꢂ 95%.
N-Phenyl-2-(1-methyl-1H-indol-3-yl)glyoxylamide (6)
Yield: 82%; mp 167–169 ꢃC, lit ref25 mp 175–177 ꢃC.
N-Benzyl-2-(1-methyl-1H-indol-3-yl)glyoxylamide (7)
Yield: 71%; mp 147–149 ꢃC, lit ref25 mp 147–148 ꢃC.
1-(1-Phenylethyl-1H-indol-3-yl)-2-phenyl-1,2-ethanedione (12)
Yield: 71%; mp 101–103 ꢃC; 1H-NMR (DMSO-d6, ppm): 3.08 (t, 2H,
J ¼ 7.2 Hz); 4.52 (t, 2H, J ¼ 7.2 Hz); 7.10-7.22 (m, 5H); 7.34–7.40 (m,
2H); 7.60–7.64 (m, 2H); 7.74–7.88 (m, 2H); 7.93 (d, 2H, J ¼ 7.2 Hz);
8.07 (s, 1H); 8.24 (d, 1H, J ¼ 6.8 Hz). 13 C-NMR (DMSO-d6, ppm):
35.82; 48.27; 111.94; 111.99; 121.88; 123.65; 124.38; 125.94; 126.94;
Reagents, starting materials and solvents were purchased from
commercial suppliers and used as received. The N-phenyl-indol-3-
ylglyoxylamide 16, N-benzyl-indol-3-ylglyoxylamide 17 were pre- 128.75; 129.25; 129.63; 130.15; 133.31; 135.19; 137.26; 138.30;
pared according to a reported procedure24. Compound 11 is 140.85; 188.58; 194.36.

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
289
1-(1-Methyl-1H-indol-3-yl)-2-phenyl-1,2-ethanedione (13)
Adenosine receptor binding assay
Yield: 72%; mp 88–90 ꢃC; lit ref26 mp 91–92 ꢃC.
[3H]DPCPX and [3H]NECA were obtained from DuPont-NEN
(Boston, MA). ADA (Cat. N. 10102105001) was from Sigma Aldrich
(Italy). All other reagents were from commercial sources and of
the highest commercially available grade. CHO cells stably
expressing human A1, A2A, A3 and A2B ARs were kindly supplied
by Prof. K. N. Klotz, Wurzburg University, Germany. MSCs (Cat. N.
SCC034) were purchased by Sigma Aldrich (Milan, Italy).
General procedure for the synthesis of 2-(benzothien-3-yl)glyoxyla-
mide 8 and 2-(aryl-2-yl)glyoxylamides 9-10
An excess of thionyl chloride (0.28 ml, 3.86 mmol) was added at
0 ꢃC to a suspension of the acids 18-20 (1.93 mmol) in dry toluene
(2 ml). The mixture was refluxed for 3 h and then the excess of thi-
onyl chloride was distilled off under reduced pressure and the
residue was washed three times with dry toluene. The oily residue
obtained was dissolved in 10 ml of anhydrous THF, cooled at 0 ꢃC
and added with triethylamine (0.29 ml, 2.12 mmol). Then, a solu-
tion of the appropriate amine (benzylamine for 8 and 10, 4-chlor-
obenzylamine for 9, 1.93 mmol) in 2 ml of dry THF was added. The
reaction mixture was left under stirring at room temperature for
24–48 h (TLC analysis; petroleum ether 60–80 ꢃC/ethyl acetate in
ratio 7:3 as eluting system). After filtering off the triethylamine
hydrochloride, the solution was concentrated to dryness. The resi-
due was triturated with diluted hydrochloric acid and then with
saturated sodium hydrogen carbonate aqueous solution, washed
with water and collected to give a crude product, which was puri-
fied by recrystallization from the appropriate solvent.
Human A1 adenosine receptors
Competition binding experiments were performed as previously
reported29. Briefly, membranes of A1 CHO cells were incubated at
25 ꢃC for 180 min in 500 mL of buffer containing [3H]DPCPX (3 nM)
and different concentrations of the compounds. R-PIA (50 mM) was
used for the determination of non-specific binding. The dissoci-
ation constant (Kd) of [3H]DPCPX in A1 CHO cell membranes
was 3 nM.
Human A2A adenosine receptors
Competition binding experiments were performed as previously
reported29. Briefly, membranes of A2A CHO cells were incubated at
25 ꢃC for 90 min in 500 mL of buffer in the presence of [3H]NECA
(30 nM) and six different concentrations of the newly synthesised
compounds. NECA (100 mM) was used for the determination of
non-specific binding. The dissociation constant (Kd) of [3H]NECA in
A2A CHO cell membranes was 30 nM.
N-Benzyl-2-(benzothien-3-yl)glyoxylamide (8)
Yield: 75%; mp 87–89 ꢃC (petroleum ether 60–80 ꢃC); 1H-NMR
(DMSO-d6, ppm): 4.49 (d, 2H, J ¼ 6.0 Hz); 7.27–7.30 (m, 1H); 7.33-
7.37 (m, 4H); 7.51 (t, 1H, J ¼ 7.2 Hz); 7.57 (t, 1H, J ¼ 7.2 Hz); 8.15 (d,
1H, J ¼ 8.0 Hz); 8.63 (d, 1H, J ¼ 8.0 Hz); 9.34 (s, 1H); 9.52 (bs, 1H).
13 C-NMR (DMSO-d6, ppm): 42.65; 123.59; 124.76; 126.19; 126.63;
127.49; 127.89; 128.86; 130.21; 136.93; 139.06; 139.63; 146.24;
164.14; 183.92.
Human A3 adenosine receptors
Competition binding experiments were performed as previously
reported30. Briefly, cell membranes of A3 CHO cells were incu-
bated at 25 ꢃC for 180 min in 500 mL of buffer in the presence of
[3H]NECA (30 nM) and six different concentrations of the newly
synthesised compounds. NECA (100 mM) was used for the deter-
mination of non-specific binding.
N-(4-Chlorobenzyl)-2-(benzofuran-2-yl)glyoxylamide (9)
Yield: 64%; mp 137–139 ꢃC (ethyl acetate); 1H-NMR (DMSO-d6,
ppm): 4.45 (d, 2H, J ¼ 6.4 Hz); 7.36-7.42 (m, 5H); 7.59-7.63 (m, 1H);
7.76 (d, 1H, J ¼ 8.4 Hz); 7.93 (d, 1H, J ¼ 7.6 Hz); 8.35 (s, 1H); 9.61 (t,
1H, J ¼ 6.4 Hz). 13 C-NMR (DMSO-d6, ppm): 42.05; 112.75; 121.32;
124.81; 125.07; 127.25; 128.78; 129.79; 130.21; 132.09; 137.96;
149.64; 155.98; 161.83; 177.48.
Measurement of cyclic AMP levels in A2B AR CHO
Intracellular cyclic AMP (cAMP) levels were measured using a com-
petitive protein binding method, as previously reported31 A2B AR
CHO cells were seed at a density of 24 000 cell/cm2 in a 24 multi-
well plate. After 48 h, the medium was removed and the cells
were incubated at 37 ꢃC for 15 min with 500 mL of Dulbecco’s
Modified Eagle Medium (DMEM) in the presence of ADA (1 U/mL)
and the phosphodiesterase inhibitor Ro20-1724 (20 mM).
N-Benzyl-2-(thien-2-yl)glyoxylamide (10)
Yield: 65%; mp 95.6–95.8 ꢃC, lit. ref27 mp 93–94 ꢃC (petroleum
ether 60–80 ꢃC).
The agonist efficacy profile of the compounds at ARs was eval-
uated by assessing their ability to modulate intracellular cAMP lev-
els in the absence of standard agonists. The antagonism profile of
the new compounds was evaluated by assessing their ability to
inhibit 100 nM NECA-mediated accumulation of cAMP. The new
compounds were incubated 10 min before the addiction of the
agonists. Then, cells were incubated in the reaction medium for
15 min at 37 ꢃC. The reaction was stopped by the removal of the
medium and the addition of 0.4 N HCl. After 30 min, lysates were
neutralised with 4 N KOH and the suspension was centrifuged at
800 g for 5 min. For the determination of cAMP production, bovine
adrenal cAMP-binding protein was incubated with [3H]cAMP
(2 nM) at 0 ꢃC for 150 min in a total volume of 300 mL. Bound
radioactivity was separated by rapid filtration. The radioactivity
was measured by liquid scintillation spectrometry.
1-(1H-Indol-3-yl)-2-phenyl-1,2-ethanedione (26)
Thionyl chloride (0.95 ml, 13.1 mmol) and phenylglyoxylic acid 24
(2.00 g, 13.1 mmol) in dry CH2Cl2 (15 ml) were added dropwise to
a cooled solution of DMAP (1.628 g, 13.1 mmol) in the same solv-
ent (15 ml). The resulting mixture was allowed to warm at room
temperature and stirred for 3 h. DMAP (1.628 g, 13.1 mmol) and
indole (14) (1.523 g, 13.1 mmol) in anhydrous CH2Cl2 (5 ml) were
then added to the cooled mixture. The reaction mixture was kept
refluxing for 2 h, added with water (20 ml) and extracted with
CH2Cl2 (3 ꢁ 30 ml). The combined organic phases were dried
(Na2SO4), filtered and evaporated to dryness. The crude product
was finally purified by flash chromatography (petroleum ether 60-
80 ꢃC/ethyl acetate in ratio 8:2 as eluting system). Yield: 65%; mp
191–192 ꢃC, lit ref28 mp 194–195 ꢃC.

290
E. BARRESI ET AL.
Cell cultures
errors of the means (SEM). Differences were considered statistically
significant for p values of <0.05.
Human bone marrow derived MSCs were cultured in normal
growth medium (MSCGM, Sigma Aldrich), plated (5.000 cells/cm2)
in 75-cm2 flasks and incubated at 37 ꢃC in 5% CO2 and 95% air.
The medium was changed to remove non adherent cells every 3
to 4 days and the cells were used at passages 2 to 6.
Results and discussion
Chemistry
The synthesis of compounds 4-7 was carried out following our
previously procedure reported for 6 and 7, with slight modifica-
tions25 (Scheme 1). Acylation of indole 14 with oxalyl chloride, in
anhydrous diethyl ether, at room temperature, yielded the indol-
3-ylglyoxyl chloride 15, which was directly allowed to react over-
night with the appropriate amine (aniline or benzylamine), in the
presence of triethylamine in dry toluene solution at room tem-
perature, to obtain the amides 16, 17 (Scheme 1)24. Treatment of
16 and 17 with sodium hydride and subsequent addition of the
appropriate halide (2-bromoethyl)benzene (for 4 and 5) or methyl
iodide (for 6 and 7) in dry DMF yielded the target derivatives 4-7,
finally purified by recrystallization from ethanol.
Target compounds 8-10 were prepared according to the
experimental procedure outlined in Scheme 2. The acid chlorides
21-23 were prepared by treating the commercially available glyox-
ylic acids 18-20 with an excess of thionyl chloride in refluxing
toluene and used in the following reaction without further purifi-
cation. The condensation of 21-23 with the appropriate amine
(benzylamine for 8 and 10, 4-chlorobenzylamine for 9) in dry THF
solution, in the presence of triethylamine, yielded the target deriv-
atives 8-10, finally purified by recrystallization from the appropri-
ate solvent.
Quantification of mineralisation
MSCs were seeded at a density of 3300 cells/cm2 and maintained
in growth medium for 2 days. Then, medium was replaced with
Mesenchymal Stem Cell Osteogenic Differentiation Medium (Cat.
N. C-28013, Sigma Aldrich) and treated in the absence (control) or
in the presence of BAY60-6583 (5 nM) alone or in combination
with the new compounds. Treatments were repeated every 3 days
and the mineralisation was quantified after 15 days of treatment.
The rate of mineralisation was quantified using alizarin red stain-
ing as previously reported32. Briefly, cells were washed with PBS,
fixed (4% formaldehyde in PBS) for 15 min and washed three
times with PBS. Then, Alizarin Red S (1:100 in distilled water,
adjusted to pH 4.2 and filtered) was added and incubated for
20 min, at the end cells were washed (five times) in 50% ethanol
and air dried. For quantification, cells were destained overnight in
10% (w/v) cetylpyridinium chloride at room temperature and the
absorbance was read using a spectrophotometer (Victor Wallac 2,
Perkin Elmer) at 562 nm.
Scheme 3 reports the synthesis of compounds 12 and 13.
Phenylglyoxylic acid 24 was reacted with thionyl chloride in dry
Statistical analysis
GraphPad Prism (version 6.00) was used for data analysis and dichloromethane, in the presence of DMAP, for 3 h at room tem-
graphic presentation. Data are reported as the means standard perature to provide the 2-oxo-2-phenylacetyl chloride 25 that was
Scheme 1. Synthesis of target compounds 4–7. Scheme 2. Synthesis of target compounds 8–10. Scheme 3. Synthesis of target compounds 12–13.

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
291
Scheme 2. Synthesis of target compounds 4–7. Scheme 2. Synthesis of target compounds 8–10. Scheme 3. Synthesis of target compounds 12–13.
Scheme 3. Synthesis of target compounds 4–7. Scheme 2. Synthesis of target compounds 8–10. Scheme 3. Synthesis of target compounds 12–13.
then treated with 14 at 50 ꢃC for 2 h, using DMAP as a base, to stimulate matrix mineralisation when used in the presence of the
yield the desired a-diketo derivative 26. Treatment of 26 with
orthosteric agonist BAY60-6583 or alone (Figure 1).
sodium hydride and subsequent addition of the appropriate hal-
To evaluate the mineralisation activity of compounds 4-13,
ide ((2-bromoethyl)benzene for 12 or methyl iodide for 13) in dry they were first assayed in the presence of BAY60-6583 at 5 nM
DMF yielded the target derivatives 12 and 13, finally purified by
flash chromatography.
concentration (Figure 1(A)). Such a concentration was selected
based on the reported potency of BAY60-6583 on MSCs (EC50
¼
6.0 nM)16. Under these experimental conditions it can be safely
assumed that the orthosteric site is mainly occupied by BAY60-
6583 rather than adenosine, owing to the much higher potency
and concentration of the synthetic agonist compared with the
natural transmitter present in the culture medium at physiological
concentrations33. The mineralisation activity was expressed as per-
Effects of 4-13 on MSCs differentiation to osteoblasts
The A2B AR plays a pivotal role in MSCs differentiation to osteo-
blasts leading to the increase of mineralisation process. Herein, we
investigated whether a series of newly synthetised compounds (4-
13), potentially able to allosterically modulate the A2B AR, could centage with respect to 100% assigned to the control

292
E. BARRESI ET AL.
+ BAY 60-6583
5 nM
(A)
(B)
200
200
150
100
50
###
###
#
***
150
*
*
***
100
50
L
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
9 1
µ
1
CTRL
0-6583 5 nM
BAY 6
CTR
2 1
4 1
5 1
6 1
7 1
8 1
9 1
2 1
4 1
5
8 1
10 1
11 1
12 1
13 1
10 1
Y 60-6583 5 nM
A
B
Figure 1. Effects of the new compounds on mineralisation mediated by MSCs-derived osteoblast. MSCs were cultured for 15 days in osteogenic medium with com-
pounds at 1 lM concentration in the presence (A) or in the absence (B) of the A2B AR agonist BAY60-6583 (5nM). After treatments, cells were stained with Alizarin
Red and mineralisation was evaluated as reported in Experimental section. Data are expressed as percentage of mineralisation versus the CTRL set to 100% and repre-
ꢀ
ꢀꢀ
ꢀꢀꢀ
p < 0.001 vs. CTRL. #p < 0.05;
sent the mean SEM of at least two independent experiments. Each experiment was performed in duplicate. p < 0.05; p < 0.01;
###p < 0.001 vs. BAY60-6583 or NECA.
Table 2. Effects of 2 and 9 on cAMP production in CHO cells expressing human
The compounds able to modify the mineralisation induced by
BAY60-6583 in the above described assay (4, 5, 8, 9 and 10) were
subsequently evaluated for their mineralisation activity in absence
of BAY60-6583 in order to assess their therapeutic potential
(Figure 1(B)). It has been demonstrated that the presence of
endogenous adenosine in the medium affects MSCs differenti-
ation16; thus, under these experimental conditions, the tested
compounds alone could potentiate the adenosine activity medi-
ated by the A2B AR activation, acting as PAMs. In this assay, those
compounds displaying a mineralisation activity higher than that of
the control (set to 100%) were considered active. Three com-
pounds revealed to be more active than the control, specifically 4
significantly enhanced the mineralisation with respect to the con-
trol (130.0 6.8%, p < 0.01), while compounds 5 (128.7 3.6%) and
9 (111.5 4.6%) displayed a mineralisation activity higher com-
pared with the control although not significant. These products
might be considered potential anti-osteoporosis agents as they
are able to increase the bone mineralisation in experimental con-
ditions close to the physio-pathological ones. The inactivity of 8
and 10 might be ascribed to their low affinity for the allosteric
site of the A2B AR. However, as mentioned before, we cannot rule
out that factors others than the pharmacodynamic ones can inter-
fere with the ability of 8 and 10 to act as A2B AR allosteric
enhancers of adenosine.
The obtained results prompted us to focalise our attention on
the most promising compounds able to enhance osteoblast differ-
entiation. However, unlike 4 and 5 that are structurally analogues
of our already reported PAMs20,21, 9 features a peculiar chemical
structure that could make it a potential lead compound for the
development of novel PAMs. In this respect, a more accurate char-
acterisation of 9 was carried out.
When assayed in radioligand binding assays, 9 displayed no
appreciable binding affinity for A1, A2A and A3 ARs (Ki > 10
000 nM). Compound 9 was then evaluated for its ability to modu-
late the increase in cAMP levels, either alone or in the presence of
an EC50 concentration of the agonist NECA in human A2B AR-
transfected CHO cells (Table 2). When tested alone at a 10 mM
concentration, 9 did not significantly increase cAMP levels, dem-
onstrating a lack of A2B AR agonist activity. Interestingly, 9 poten-
tiated the effects of NECA, to a higher extent with respect to the
A2B AR.a
% cAMP production
Cpd
(vs NECA 100 nM set to 100%)b
hA2B AR
Alone
6.2 1.3
2.7 1.3
þ agonist
ꢀꢀꢀ
ꢀꢀꢀ
2
9
150.6 15.7
184.0 11.7
aThe effect of each compound (10 mM) on cAMP production was evaluated in
CHO cells expressing human A2B AR (see biological section). Each compound
was tested alone or in the presence of the agonist NECA (100 nM).
bData are expressed as percentage of cAMP production versus agonist maximal
effect set to 100%. All data represent the mean SEM of at least three inde-
pendent experiments.
ꢀꢀꢀ
p < 0.001 vs agonist alone.
(measurements in absence of BAY60-6583). The activity of the par-
ent compound
2
in the presence of 5 nM BAY60-6583
(171.9 0.8% p < 0.001 vs BAY60-6583) served as a benchmark.
MSCs, cultured under osteogenic conditions, underwent a spon-
taneous, time-dependent, mineralisation16. Accordingly, cells treat-
ment with BAY60-6583 induced a significant increase in the
mineralisation process of MSCs (129.5 1.9% p < 0.001 vs control;
Figure 1(A)). In this preliminary test, some new derivatives potenti-
ated BAY60-6583 activity (Figure 1(A)), thanks to their hypothes-
ised properties of positive allosteric modulators of A2B AR: namely
4 (179.2 9.6%. p < 0.001 vs BAY60-6583), 5 (152.3 4.3%), 8
(144.4 4-9%), 9 (156.8 7.6%, p < 0.05 vs BAY60-6583) and 10
(150.6 2.6%). Among these, only derivatives 4 and 9 significantly
ameliorated BAY60-6583 activity, while for compounds 5, 8 and
10 the increase is not statistically significant. It is worth noting
that 4 is more active than the reference compound 2.
The activity of compound 9 suggests that the indole scaffold
together with a N-arylalkyl substituent (benzyl or 2-phenylethyl) at
1-position are probably not key pharmacophore elements to dis-
play positive allosteric modulation of the A2B AR. Of course, such
a conclusion should be supported by further experiments demon-
strating the precise nature of the events occurring at the molecu-
lar levels. The activity measured in the mineralisation assay
reflects, in fact, not only pharmacodynamic factors but also phar-
macokinetic phenomena, such as stability to metabolism, which
cannot be taken into account separately.

JOURNAL OF ENZYME INHIBITION AND MEDICINAL CHEMISTRY
293
CHO A2B
CHO A2B
(A)
(B)
200
150
100
50
NECA
NECA+9 650 nM
NECA+9 6,5 µM
200
180
160
140
120
100
80
EC50= 636.2 65.3 nM
0
−10
−9
−8
−7
−6
−5
−4
−10
−9
−8
−7
−6
−5
log[NECA]
Figure 2. Effects of compound 9 as positive allosteric modulator of the A2B AR. A) Effects of 9 on NECA-mediated cAMP accumulation in human A2B AR-transfected
CHO cells. CHO cells were treated with an EC50 NECA concentration (100nM) in the absence or presence of different concentrations of the tested compound
(1nM–10 lM). B) Effects of 9 on agonist Emax. CHO cells were treated with different NECA concentration (1 nM–10 lM) in the absence or presence of 9 (650 nM or
6.5lM). After 15 min incubation, the reaction was stopped and the intracellular cAMP levels were quantified. The data are expressed as the percentage of cAMP/well
versus the maximal NECA effect, which was set to 100% and represent the mean SEM of at least three different experiments. Each experiment was performed
in duplicate.
parent compound 2, suggesting that this compound may behave
as PAM of the A2B AR.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Then, the potency of 9 in modulating the activity of A2B AR
agonists was determined by assessing the effects of different con-
centrations (ranging from 1 nM to 10 mM) on cAMP accumulation
induced by an EC50 concentration of the agonist NECA (100 nM).
The resulting concentration–response curves (Figure 2(A)) indi-
cated that 9 exhibited a positive allosteric modulation of A2B AR
with an EC50 636.2 65.3 nM. Subsequently, the effects of 9 on
the potency and efficacy of NECA were evaluated by assessing
agonist concentration–response curves in the absence or presence
Funding
This work was supported by a grant funded by the Italian Ministry
of Education [Project of National Research Interest PRIN 2015,
2015E8EMCM_007]. The Italian Ministry had no further role in the
study design, analysis or interpretation of the data, the writing of
the report or the decision to submit the paper for publication.
of two different concentrations of
9 (650 nM and 6.5 mM).
Compound 9 significantly enhanced the efficacy of NECA in stimu-
lating cAMP accumulation without affecting agonist potency
(Figure 2(B)).
ORCID
Elisabetta Barresi
Chiara Giacomelli
Laura Marchetti
Emma Baglini
Silvia Salerno
Giovanni Greco
http://orcid.org/0000-0002-9814-7195
http://orcid.org/0000-0002-6244-602X
http://orcid.org/0000-0002-2110-9481
http://orcid.org/0000-0002-8378-4052
http://orcid.org/0000-0002-6072-4698
http://orcid.org/0000-0002-4632-4620
Conclusions
A small library of molecules (4-13) exhibiting different degrees of
chemical similarity with the recently identified indole-based posi-
tive allosteric modulators (PAMs) of the A2B AR 1–3 was investi-
gated for their potential as bone mineralisation agents.
Compounds 4 and 9 turned out to stimulate matrix mineralisation
in MSCs either in the presence or in the absence of the agonist
BAY60-6583. The structure-activity relationships of 4–13 suggest
that the indole scaffold and a N-arylalkyl substituent (benzyl or 2-
phenylethyl) at 1-position are probably not key pharmacophore
elements to display positive allosteric modulation of the A2B AR.
The biological properties of the benzofurane derivative 9, featur-
ing a peculiar chemical structure with respect to the reference
indoles 1–3, were more accurately characterised. The obtained
results confirmed that such a molecule behaves as PAM of the A2B
AR, making it a lead structure for the development of novel com-
pounds potentially acting as anti-osteoporosis agents.
Federico Da Settimo
Claudia Martini http://orcid.org/0000-0001-9379-3027
Maria Letizia Trincavelli http://orcid.org/0000-0001-8124-977X
http://orcid.org/0000-0001-8675-939X
http://orcid.org/0000-0002-7897-7917
Sabrina Taliani
References
1. Lane JM, Russell L, Khan SN. Osteoporosis. Clin Orthop Relat
Res 2000;372:139–50.
2. Lamichhane AP. Osteoporosis - an update. J Nepal Med
Assoc 2005;44:60–6.
3. Roush K. Prevention and treatment of osteoporosis in post-
menopausal women: a review. Am J Nurs 2011;111:26–35.
4. Francis RM. Non-response to osteoporosis treatment. J Br
Menopause Soc 2004;10:76–80.
5. Salari SP, Abdollahi M, Larijani B. Current, new and future
treatments of osteoporosis. Rheumatol Int 2011;31:289–300.
6. Strazzulla LC, Cronstein BN. Regulation of bone and cartilage
by adenosine signaling. Purinergic Signal 2016;12:583–93.
Author contributions
The manuscript was written through the contributions of all
authors. All authors have given approval to the final version of
the manuscript.

294
E. BARRESI ET AL.
7. Cheng X, Cheng G, Xing X, et al. Controlled release of 20. Taliani S, Trincavelli ML, Cosimelli B, et al. Modulation of
adenosine from core-shell nanofibers to promote bone
regeneration through STAT3 signaling pathway. J Control
Release 2020;319:234–45.
A2B adenosine receptor by 1-benzyl-3-ketoindole deriva-
tives. Eur J Med Chem 2013;69:331–7.
21. Trincavelli ML, Giacomelli C, Daniele S, et al. Allosteric mod-
ulators of human A2B adenosine receptor. Biochim Biophys
Acta - Gen Subj 2014;1840:1194–203.
22. Taliani SD, Settimo F, Martini C, et al. G. Exploiting the
indole scaffold to design compounds binding to different
pharmacological targets. Molecules 2020;25:2331.
23. Zhang Y, Haga N. Skeletal complications in congenital
insensitivity to pain with anhidrosis: a case series of 14
patients and review of articles published in Japanese. J
Orthop Sci 2014;19:827–31.
8. Shih YR, Liu M, Kwon SK, et al. Dysregulation of ectonucleo-
tidase-mediated extracellular adenosine during postmeno-
pausal bone loss. Sci Adv 2019;5:eaax1387
9. Fredholm BB, IJzerman AP, Jacobson KA, et al. International
Union of Basic and Clinical Pharmacology. LXXXI.
Nomenclature and Classification of Adenosine receptors-an
update. Pharmacol Rev 2011;63:1–34.
10. Langemeijer EV, Verzijl D, Dekker SJ, IJzerman AP. Functional
selectivity of adenosine A1 receptor ligands? Purinergic
Signal 2013;9:91–100.
24. Da Settimo A, Primofiore G, Da Settimo F, et al. Structure-
activity relationships, and molecular modeling studies of N-
(Indol-3-ylglyoxylyl)Benzylamine derivatives acting at the
benzodiazepine receptor. J Med Chem 1996;39:5083–91.
25. Thompson MJ, Borsenberger V, Louth JC, et al. Design, syn-
thesis, and structure - activity relationship of indole-3-glyox-
ylamide libraries possessing highly potent activity in a cell
line model of prion disease. J Med Chem 2009;52:7503–11.
26. Katritzky AR, Zhang D, Kirichenko K. Synthesis of heteroaryl
1,2-diketones. J Org Chem 2005;70:3271–4.
27. Zhang YK, Wang B. Synthesis of a-ketoamides from b-ketoni-
triles and primary amines: a catalyst-free oxidative decyana-
tion-amidation reaction. Eur J Org Chem 2019;2019:5732–5.
28. Ivonin SP, Lapandin AV. Conversions of unsymmetrical ben-
zoins in basic media. ARKIVOC 2004; 2005:4–9.
29. Graziano A, Giovannoni MP, Cilibrizzi A, et al. C. New pyrazo-
lo[10,50:1,6]pyrimido[4,5-d]pyridazin-4 (3H)-ones fluoroderiva-
tives as human A1 adenosine receptor ligands. Acta Chimica
Slovenica 2012;59:648–55.
30. Cosimelli B, Greco G, Laneri S, et al. 4-Amino-6-alkyloxy-2-
alkylthiopyrimidine derivatives as novel non-nucleoside ago-
nists for the adenosine A1 receptor. Chem Biol Drug Des
2016;88:724–9.
11. Giacomelli C, Daniele S, Romei C, et al. The A2B adenosine
receptor modulates the epithelial- mesenchymal transition
through the balance of CAMP/PKA and MAPK/ERK pathway
activation in human epithelial lung cells. Front Pharmacol
2018; 9:54
12. Zezula J, Freissmuth M. The A 2A-adenosine receptor: A
GPCR with unique features? Br J Pharmacol 2008;153:
S184–S190.
13. Gharibi B, Abraham AA, Ham J, Evans BAJ. Adenosine recep-
tor subtype expression and activation influence the differen-
tiation of mesenchymal stem cells to osteoblasts and
adipocytes. J Bone Miner Res 2011;26:2112–24.
14. Costa MA, Barbosa A, Neto E, et al. On the role of subtype
selective adenosine receptor agonists during proliferation
and osteogenic differentiation of human primary bone mar-
row stromal cells. J Cell Physiol 2011;226:1353–66.
15. Shih YR, Hwang Y, Phadke A, et al. Calcium phosphate-bear-
ing matrices induce osteogenic differentiation of stem cells
through adenosine signaling. Proc Natl Acad Sci U S A 2014;
111:990–5.
16. Trincavelli ML, Daniele S, Giacomelli C, et al. Osteoblast dif-
ferentiation and survival: a role for A2B adenosine receptor
allosteric modulators. Biochim Biophys Acta - Mol Cell Res
2014;1843:2957–66.
31. Nordstedt C, Fredholm BB. A modification of a protein-bind-
ing method for rapid quantification of CAMP in cell-culture
supernatants and body fluid. Anal Biochem 1990;189:231–4.
32. Giacomelli C, Natali L, Nisi M, et al. Negative effects of a
high tumour necrosis factor-a concentration on human gin-
gival mesenchymal stem cell trophism: the use of natural
compounds as modulatory agents. Stem Cell Res Ther 2018;
9:135.
17. He W, Mazumder A, Wilder T, Cronstein BN. Adenosine regu-
lates bone metabolism via A1, A2A, and A2B receptors in
bone marrow cells from normal humans and patients with
multiple myeloma. Faseb J 2013;27:3446–54.
18. Carroll SH, Wigner NA, Kulkarni N, et al. A2B Adenosine
receptor promotes mesenchymal stem cell differentiation to
osteoblasts and bone formation in vivo. J Biol Chem 2012;
287:15718–27.
33. Jacobson KA, Tosh DK, Jain S, Gao ZG. Historical and current
adenosine receptor agonists in preclinical and clinical devel-
opment. Front Cell Neurosci 2019;13:124.
19. Eisenstein A, Carroll SH, Johnston-Cox H, et al. An adenosine
€
receptor-Kruppel-like factor 4 protein axis inhibits adipogen-
esis. J Biol Chem 2014;289:21071–81.