Synthesis of (E)-8-(3-chlorostyryl)caffeine analogues leading to 9-deazaxanthine derivatives as dual A2A antagonists/MAO-B inhibitors

Article abstract of DOI:10.1021/jm301686s

A systematic modification of the caffeinyl core and substituents of the reference compound (E)-8-(3-chlorostyryl)caffeine led to the 9-deazaxanthine derivative (E)-6-(4-chlorostyryl)-1,3,5,-trimethyl-1H-pyrrolo[3,2-d]pyrimidine- 2,4-(3H,5H)-dione (17f), w

Full text of DOI:10.1021/jm301686s

Subscriber access provided by FORDHAM UNIVERSITY
Article
Synthesis of (E)-8-(3-chlorostyryl)caffeine analogs leading to 9-
deazaxanthine derivatives as dual A2A antagonists /MAO-B inhibitors.
Silvia Rivara, Giovanni Piersanti, Francesca Bartoccini, Giuseppe Diamantini, Daniele Pala, Teresa Riccioni,
Maria Antonietta Stasi, Walter Cabri, Franco Borsini, Marco Mor, Giorgio Tarzia, and Patrizia Minetti
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm301686s • Publication Date (Web): 02 Jan 2013
Downloaded from http://pubs.acs.org on January 16, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis of (E)-8-(3-chlorostyryl)caffeine analogs leading to 9-
deazaxanthine derivatives as dual A antagonists /MAO-B inhibitors
2A
a
b
b
b
a
Silvia Rivara, * Giovanni Piersanti, * Francesca Bartoccini, Giuseppe Diamantini, Daniele Pala,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
c
c
c
c
a
Teresa Riccioni, Maria Antonietta Stasi, Walter Cabri, Franco Borsini, Marco Mor, Giorgio
b
c
Tarzia and Patrizia Minetti*
a
Dipartimento di Farmacia, Università degli Studi di Parma, Viale G.P. Usberti 27 A, I-43124
Parma, Italy.
b
Department of Biomolecular Sciences, University of Urbino, Piazza Rinascimento 6, I-61029
Urbino (PU), Italy.
c
Sigma-Tau Industrie Farmaceutiche Riunite S.p.A., Via Pontina Km 30,400, I-00040 Pomezia,
Italy.
Reference authors: Giovanni Piersanti, Silvia Rivara, Patrizia Minetti
Silvia Rivara
Giovanni Piersanti
Dipartimento di Farmacia
Università degli Studi di Parma
Viale G.P. Usberti 27/A
Iꢀ43124 Parma (Italy)
Phone: +390521905061
Fax: +390521905006
Dipartimento di Scienze Biomolecolari
Università degli Studi di Urbino
Piazza Rinascimento 6
Iꢀ61029 Urbino (PU), Italy
Phone: +390722303313
Fax: +390722303320
Eꢀmail: giovanni.piersanti@uniurb.it
Eꢀmail: silvia.rivara@unipr.it
Patrizia Minetti
SigmaꢀTau Industrie Farmaceutiche Riunite S.p.A.
Via Pontina Km 30,400
Iꢀ00040 Pomezia, Italy.
Tel: +390691393906
Fax: +390691393638
Eꢀmail. Patrizia.Minetti@sigmaꢀtau.it
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
A systematic modification of the caffeinyl core and substituents of the reference compound (E)ꢀ8ꢀ
3ꢀchlorostyryl)caffeine led to the 9ꢀdeazaxanthine derivative (E)ꢀ6ꢀ(4ꢀchlorostyryl)ꢀ1,3,5,ꢀ
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
trimethylꢀ1Hꢀpyrrolo[3,2ꢀd]pyrimidineꢀ2,4ꢀ(3H,5H)ꢀdione (17f) which acts as a dual antagonist of
human A /MAOꢀB inhibitor (K (A ) = 260 nM; IC (MAOꢀB) = 200 nM; IC (MAOꢀA) = 10 ꢀM)
2a
i
2A
50
50
and dose dependently counteracts haloperidolꢀinduced catalepsy in mice from 30 mg/kg by the oral
route. The compound is the best balanced A2A antagonist /MAOꢀB inhibitor reported to date and it
could be considered as a new lead in the field of antiꢀParkinson’s agents. A number of analogs of
1
7f were synthesized and qualitative SARs are discussed. Two analogs of 17f, namely 18b and 19a,
inhibit MAOꢀB with IC50 of 68 nM and 48 nM, respectively, being from five to seven folds more
potent than the prototypical MAOꢀB inhibitor deprenyl (IC50 = 334 nM).
2
ACS Paragon Plus Environment

Page 3 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
The complexity of the interactions among the several neurotransmitters acting within the central
nervous system (CNS) and the difficulty in obtaining acceptable therapeutic options for the
treatment of diseases such as depression, Alzheimer’s (AD) and Parkinson’s (PD) diseases has led
to a number of attempts to discover compounds acting simultaneously on more than one of the
neurotransmitters known to play a role in these disorders, particularly when their respective
receptors are localized in identical or vicinal areas of the brain or when synergic actions among
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
them have been shown. In the case of PD the development of selective antagonists of the A2A
adenosine receptor (A2AR) initially led to the discovery of a number of promising agents, such as
2
3
4
5
(E)ꢀ8ꢀ(3ꢀchlorostyryl)caffeine (CSC) istradefylline (KW6002), KFꢀ17837, DMPX and DPMTX,
6
7
8
9
various derivative of 3,7ꢀdimethylxanthines, MSXꢀ2 and its prodrugs, SCH 58261, ZM241385,
1
0
11
CGS 15943 and ST1535 (Figure 1). In particular, the therapeutic potential of KW6002 for the
treatment of PD has been extensively investigated through a number of clinical trials, which showed
that this compounds significantly improved motor impairment and reduced “off” time when coꢀ
12
administered with levodopa. The therapeutic application of two irreversible inhibitors of
1
3
monoamine oxidase B (MAOꢀB), (R)ꢀdeprenyl and rasagiline, for alleviating the symptoms of PD,
1
4
15
the ageꢀrelated increase in the expression of MAOꢀB, its cerebral distribution and the dual
1
6
activity of CSC as A2A antagonist and MAOꢀB inhibitor led to the proposal of jointly downꢀ
regulating A2a and MAOꢀB as a novel approach for treating PD and to the design of several
1
7
A2A/MAOꢀB dualꢀacting agents.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. A2A receptor antagonists.
Studies on dualꢀacting A2A antagonists/MAOꢀB inhibitors are limited to compounds structurally
1
8
related to caffeine and 8ꢀstyrylxanthine and the subject has been recently reviewed. CSC is a
i
reference dualꢀacting agent with good A2A affinity (K = 36 nM on receptors expressed on rat brain
6
striatal membranes) and MAOꢀB inhibitory potency (K = 235 nM in tests performed on human
i
16b
liver mitochondria), effective in vivo in reversing the biochemical and behavioral modifications
19
of 6ꢀhydroxydopamineꢀlesioned rats. We decided therefore to examine whether it could be
possible to obtain new dual A2A antagonists/MAOꢀB inhibitors with higher and more balanced
potencies at the two targets. In order to select a suitable scaffold we run a preliminary exploratory
modification of the caffeinyl core of CSC to obtain derivatives belonging to the different chemical
4
ACS Paragon Plus Environment

Page 5 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
classes reported in Figure 2. In particular, we evaluated the role of the nitrogen, carbonyl group and
imidazole ring in the purine system by preparing compounds 4a,b and the pyrimidinedione
derivative 6. Bioisosteric replacement of the purine nitrogen atom in position 9 led to 9ꢀ
deazaxanthines (17a-y, 18a-d, 19a-d, 20a,b, 21-23) and to their conformationally constrained
tricyclic derivatives (14a-c). Preliminary screening of a few prototypes for their A2A/MAOꢀB dualꢀ
activity allowed us to identify the 9ꢀdeazaxanthine as the most promising scaffold which was
optimized by modifying the steric, electronic and lipophilic properties of the 8ꢀsubstituent to give
compound 17a-y, 18a-d, 19a-d, 20a,b, 21-23 (Figure 2, Schemes 1ꢀ8).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Chemical scaffolds of newly synthesized compounds.
Chemistry
2
0
The imidazopyridinones 4a,b were obtained by direct condensation of 1a,b with transꢀcinnamic
acid or cinnamaldehyde to yield 2a,b that were oxidized to 3a,b with mꢀchloroperoxybenzoic acid
2
1
22
23
(mꢀCPBA). Rearrangement of 3a,b with acetic anhydride or trifluoroacetic anhydride followed
by N-alkylation with methyl iodide gave 4a,b in satisfactory yields (Scheme 1).
a
Scheme 1. Synthesis of imidazopyridinone derivatives.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (i) For 2a: (E)ꢀPhCH=CHCO
MeOH, AcOH, (E)ꢀPhCH=CHCHO, rt, 1 h, 40%; (ii) For 3a: CH
68%; For 3b CHCl , mꢀCPBA, rt, 30 min, 88%; (iii) (CH CO) O, 140 °C, 2 h, 56%. (iv) DMF,
CO , MeI, rt, 60 h, 28%; (v) DMF, (CF CO) O, rt, 16 h, 47%. (vi) NaOH 6 N, MeI, 50 °C, 12 h,
5%;
2
H, POCl
3
, 140 °C, 5 h, 44%; For 2b:
2
Cl
2
, MeOH, mꢀCPBA, rt, 1 h,
3
3
2
K
5
2
3
3
2
2
4
Amide coupling of the scarcely reactive secondary enamine 5 with trans-cinnamic acid was
obtained using 1ꢀethylꢀ3ꢀ(3ꢀdimethylaminopropyl) carbodiimide (EDCI) as a coupling agent and
afforded the desired amide (6) in modest yield. (Scheme 2).
a
Scheme 2. Synthesis of pyrimidines derivative 6.
a
2
Reagents and conditions: (i) MeOH, EDCI, PhCH=CHCO H, rt, 20 h, 37%.
1
,3,5ꢀTrimethylꢀ1Hꢀpyrimido[5,4ꢀb]indoleꢀ2,4(3H,5H)ꢀdione derivatives 14a-c were prepared as
described in Scheme 3. Compound 14a and 14c were obtained respectively from the commercially
6
ACS Paragon Plus Environment

Page 7 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
available 7a and 9c. Compound 14b was obtained from 7b which was synthesized by Suzuki
2
5
reaction of 4ꢀbromobenzaldehyde with 3ꢀchlorophenylboronic acid. Formation of ethyl 2ꢀ
azidocinnamate (8a,b) by condensation of arylaldehyde (7a,b) with ethyl azidoacetate in the
presence of sodium ethoxide at 0 °C and subsequently pyrolysis of 8a,b in pꢀxylene a 150 °C gave
the corresponding ethyl indoleꢀ2ꢀcarboxylate (9a,b). Reaction of indole derivatives 9a-c and
benzenediazonium chloride provided azo compounds 10a-c. Reduction with tin and hydrochloric
acid, followed by treatment with ethoxycarbonyl chloride, furnished carbamates 12a-c. Cyclization
with methylamine and Nꢀmethylation with methyl iodide provided 14a-c (Scheme 3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Scheme 3. Synthesis of 1,3,5ꢀtrimethylꢀ1Hꢀpyrimido[5,4ꢀb]indoleꢀ2,4(3H,5H)ꢀdione derivatives.
a
3 2 2
Reagents and conditions: (i) EtOH, N CH CO Et, EtONa, 0 °C, 2h, 55ꢀ58%; (ii) pꢀxylene, 150
°
C, 2 h, 78ꢀ98%; (iii) 1) C H NH , HCl 6 N, NaNO , 0 °C, 15’. 2) DMF, Na CO 2 N, 0 °C, 1 h,
6 5 2 2 2 3
3
0ꢀ80%; (iv) (CH ) CHOH, conc. HCl, Sn, 85 °C, 2 h, 46ꢀ69%; (v) Xylene, ClCOOEt, 140 °C, 2 h,
3 2
40ꢀ83%; (vi) (CH ) CHOH, CH NH 40% wt in H O, 85 °C, 24 h, 15ꢀ52%; (vii) DMF, NaH, MeI,
3
2
3
2
2
rt, 24 h, 46ꢀ86%.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1,3,5ꢀTrimethylꢀ6ꢀstyrylꢀ1Hꢀpyrrolo[3,2ꢀd]pyrimidineꢀ2,4(3H,5H)diones 17a-w were obtained as
2
6
described in Schemes 4 and 5. Compound 15a was synthesized as previously reported and the
introduction of the styryl group was not without problems. Synthesis of styryl derivatives with
electronꢀdonating substituents was achieved by palladium crossꢀcoupling reaction under standard
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
6
reaction condition, whereas the procedure failed in the case of styryl containing electronꢀ
withdrawing substituents. The vinylation of 15a with vinyl boronate ester gave compound 16 which
was used for Heck coupling with aryl bromides containing the electron withdrawing groups.
Compounds 17a-n were synthesized by path (ii), whereas compounds 17o-t were prepared by path
(
iii) in Scheme 4.
a
Scheme 4. Synthesis of 8ꢀstyrylꢀ9ꢀdeazaxanthine derivatives.
1
2
3
1
2
3
Compounds
7a
7b
7c
7d
X
C
C
C
C
C
C
C
C
C
C
Y
C
C
C
C
C
C
C
C
C
C
R
R
R
Compounds
17k
X
C
C
C
N
C
C
C
C
C
C
Y
C
C
C
C
C
C
C
C
C
N
R
R
R
1
1
1
1
H
H
H
H
H
H
H
H
H
H
Cl
H
OMe
H
H
H
H
F
H
H
OMe
H
H
H
H
H
H
H
Cl
ꢀOCH
2
H
Cl
Oꢀ
H
H
H
H
COMe
SO Me
CN
H
17l
17m
17n
17o
17p
17q
17r
17s
17t
CF
H
3
NH
2
17e
17f
17g
17h
17i
COMe
CN
H
H
H
H
H
H
H
Cl
CF
OMe
OnPr
Br
3
2
17j
H
H
a
Reagents and conditions: (i) DME, H
ii) DMF, KOAc, TBAB, molecular sieves, styrene derivatives, Pd(OAc)
DMF, aryl bromide, TEA, P(oꢀtol) , Pd(OAc) , 80 °C, 2 h, 30ꢀ84%.
2
O, CH
2
=CHBpin, TEA, PdCl
2
(PPh
3 2
) , 80 °C, 16 h, 67%;
(
2
, 80 °C, 20 h, 20ꢀ86%; iii)
3
2
8
ACS Paragon Plus Environment

Page 9 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Phenol derivatives 17u-w were obtained by demethylation with BBr of the corresponding
3
methoxy derivatives 17c,h,m (compound 17m was not tested) (Scheme 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Scheme 5. Synthesis of hydroxystyryl derivatives.
a
2 2 3 2 2
Reagents and conditions: (i) CH Cl , BBr in CH Cl , 0 °C, 1 h, 27ꢀ90%.
26
Ethynyl derivatives 18a-c were synthesized by Sonogashira reaction as previously reported
(
Scheme 6).
a
Scheme 6. Synthesis of alkynyl derivatives.
a
Reagents and conditions: (i) Dioxane, TEA, CuI, (PPh ) PdCl , alkynyl derivatives, 100 °C, 20
3
2
2
h, 54ꢀ62%.
The transitionꢀmetalꢀcatalyzed formation of carbonꢀheteroatom bonds via crossꢀcoupling
reactions plays an important role in the preparation of numerous products of pharmaceutical
2
7
interest, allowing the introduction of various nitrogenꢀ and oxygenꢀfunctions (amine, amide, urea,
carbamate, alcohol, phenol, thiol) onto aromatic or heteroaromatic cycles. The bromo derivative
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
15a failed to give the desired Nꢀ or Oꢀarylated products under variously modified reaction
conditions (solvent, base, palladium and copper precursor, ancillary coligand), whereas the more
reactive iodo derivative 15b, prepared from 15a by copperꢀcatalyzed halogen exchange reaction
2
8
using potassium iodide and N,N’ꢀdimethylethylendiamine ligand (Scheme 7), underwent copper or
palladiumꢀcatalyzed coupling with different heteroꢀnucleophiles. Ether derivatives 19a-c could be
obtained in acceptable yield by copperꢀcatalyzed reaction using 1,10ꢀphenanthroline as a ligand at
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
10 °C for prolonged time. The same reaction condition did not work with phenols as substrates and
the diarylethers 20a,b were obtained by using ferric chloride as a catalyst in DMF in the presence of
2
9
2
2 3
,2,6,6ꢀtetramethylꢀ3,5ꢀheptanedione and Cs CO at 130 °C for 72 hours. Copper (I) catalysts
3
0
(bromide or iodide) allowed coupling of amides and triazoles to give 21 and 22 in good yield.
Similar reaction conditions permitted also a CꢀC bond formation by 2ꢀCꢀH activation of the
benzoxazole system to give compound 23 (Scheme 7).
a
Scheme 7. Synthesis of 8ꢀfunctionalized deazaxanthine derivatives.
O
N
N
Br
O
N
15a
O
N
O
N
i
R1
Cl
N
N
O
N
N
N
O
N
O
R2
ii
vi
O
N
O
N
I
19a-c
23
1
2
O
19a, R = H R = Cl
19b, R = CF R = Cl
9c, R = R = -CH-(CH) -CH-
1
2
15b
3
1
2
1
iii
v
2
O
N
O
R1
N
N
N
N
iv
N
N
N
O
O
O
N
22
Cl
20a,b
O
1
2
0a, R = Cl
N
N
1
20b, R = Ph
NH
Cl
O
N
O
2
1
a
Reagents and conditions: (i) CuI, KI, N,N’ꢀdimethylethylendiamine, nꢀBuOH, 120 °C, 36 h, 72%;
10
ACS Paragon Plus Environment

Page 11 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(ii) Toluene, benzylalcohol derivatives, CuI, 1,10ꢀ phenanthroline, Cs CO , 100 °C, 5 h, 38ꢀ43%;
2
3
(iii) DMF, phenol derivatives, FeCl , 2,2,6,6ꢀtetramethylꢀ3,5ꢀheptanedione, Cs CO ,130 °C, 72 h,
3
2
3
3
7ꢀ43%; (iv) dioxane, K CO , CuI, transꢀN,N’ꢀdimethyl cyclohexanꢀ1,2ꢀdiamine, 4ꢀ
2 3
ClC
6
H
4
CONH
2
, 100 °C, 20 h, 35%; (v) DMSO/H
2
3
O 9:1, NaN , 4ꢀClC
6
H
4 4 2
C≡CH, CuSO ꢁH O,
sodium ascorbate, LꢀProline, 60 °C, 18 h, 39%; (vi) DMF, 6ꢀchlorobenzo[d]oxazole, Pd(OAc)
Bu) , Cs CO , CuBr, 150 °C, 2 h, 39%.
2
, P(tꢀ
3
2
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The reaction conditions reported above were employed for the synthesis of different Nꢀalkylated
deazaxanthines (17x,y, 18d, 19d). In particular, compounds 17x,y were synthesized by Heck
coupling starting from 15c and 15d, respectively, compound 18d was synthesized by Sonogashira
coupling starting from 15d, while compound 19d was obtained by Oꢀarylation of 15e (Scheme 8).
a
Scheme 8. Synthesis of 8ꢀfunctionalized deazaxanthine derivatives.
O
R3
R1
N
N
X
iii
i
O
N
O
R3
O
N
R2 15c-e
R1
O
N
N
1
2
3
N
15c, X = Br R = R = Me R = Et
N
1
2
3
O
1
1
5d, X = Br R = R = Et R = Me
Cl
1
2
3
Cl
N
5e, X = I R = R = Et R = Me
O
R2
1
7x,y
2
19d
ii
1
3
1
1
7x, R = R = Me R = Et
1
2
3
O
7y, R = R = Et R = Me
N
N
Cl
O
N
18d
a
Reagents and conditions: (i) DMF, KOAc, TBAB, molecular sieves, styrene derivatives, Pd(OAc)
0 °C, 20 h, 43ꢀ45%; (ii) Dioxane, TEA, CuI, (PPh PdCl , alkynyl derivatives, 100 °C, 20 h, 45%;
iii) Toluene, benzylalcohol derivatives, CuI, 1,10ꢀ phenanthroline, Cs CO , 100 °C, 5 h, 40%.
2
,
8
(
3
)
2
2
2
3
8ꢀ(pꢀChlorostyryl)xanthine (25) was synthesized by a novel Pd/Cuꢀcatalyzed dehydrogenative Heck
coupling that allows direct alkenylation of caffeine (24) with 4ꢀchlorostyrene (Scheme 9) which
31
eliminates the need of preparing the heteroaryl halides.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 9. Synthesis of 8ꢀfunctionalized xanthine derivative.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (i) 4ꢀClC
6 4 2 2 2 2 3 2
H CH=CH , Cu(OAc) , Pd(OAc) , Ag CO , tꢀBuCO H, 90
°C, 72 h, 60%.
Results and discussion
The compounds were tested for their binding affinity for human A2A (hA2A) receptors and inhibitory
potency on human MAOꢀB (hMAOꢀB) (Tables 1ꢀ3). Selectivity versus MAOꢀA was also evaluated
for the compounds displaying MAOꢀB IC50 values lower than 1400 nM. Investigation on the
structural determinants for the interaction with A2A and MAOꢀB was initially focused on the
pyrimidinedione moiety of the dechloroꢀCSC. Replacement by 1ꢀmethylꢀ2ꢀpyridinone ring led to
compound 4a, which was inactive at both targets, and the same result was obtained with its
regioisomer 4b (Table 1). Deletion of the imidazole ring of the xanthine scaffold and insertion of an
amide group on the pyrimidinedione ring (6) to maintain the coplanarity of the styryl substituent
was also detrimental for the interaction with both targets. A more conservative modification,
yielding the 9ꢀdeazaxanthine 17a was tolerated by both targets, with only a limited decrease of
potency (Table 1). The chlorine atom in position meta slightly decreased A2A binding affinity,
compared to the unsubstituted derivative 17b, and had a negligible effect on MAOꢀB inhibitory
3
2
2
potency. 8ꢀStyrylꢀ9ꢀdeazaxanthines have already been reported as A receptor ligands, but this is
the first time that their MAOꢀB inhibitory potency is described. Crystal structures of the A2A
receptor in complex with caffeine and XAC (Nꢀ(2ꢀaminoethyl)ꢀ2ꢀ[4ꢀ(2,3,6,7ꢀtetrahydroꢀ2,6ꢀdioxoꢀ
3
3
1
,3ꢀdipropylꢀ1Hꢀpurinꢀ8ꢀyl)phenoxy]ꢀacetamide) (where no water molecules were resolved)
12
ACS Paragon Plus Environment

Page 13 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
showed that the N9 of the xanthine nucleus does not directly interact with residues of the binding
site (Figure 3A). On the other hand, superposition of the crystal structures of the A ꢀXAC complex
2A
34
with the recently solved A2AꢀZM241385 complex, where several water molecules are present,
showed that N9 of XAC is roughly superposed to the N4 of ZM241385, forming a hydrogen bond
with a conserved water molecule belonging to a cluster located inside the binding site (Figure 3B).
Lack of interaction with this cluster may explain the limited loss of binding affinity of the 9ꢀdeaza
derivative 17a compared to CSC.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. (A) Superposition of the crystal structures of A2A receptor in complex with caffeine
(green carbons, PDB code: 3RFM) and XAC (yellow carbons, PDB code: 3REY). (B)
Superposition of the crystal structures of A2A receptor in complex with ZM241385 (orange carbons,
PDB code: 4EIY) and XAC (yellow carbons). Water molecules are represented with red spheres. (C)
Best docking poses for compounds 17f (orange carbons) and 17q (cyan carbons) within the A2A
receptor structure built from 3REY. XAC molecule is depicted with yellow carbons.
No crystal structure is available for either xanthine or 9ꢀdeazaxanthine derivatives within the MAOꢀ
3
5
B catalytic site. We therefore docked CSC into the MAOꢀB crystal structure using Glide software
and Figure 4A depicts the best docking solution. CSC occupies both the substrate and the entrance
cavities, with the xanthine moiety near the FAD cofactor and the 6ꢀcarbonyl group interacting with
Tyr435 (not shown in Figure 4A). The purine core fits between Tyr398 and Leu171 on one side and
Tyr435 and Gln206 on the other one. This accommodation is very similar to that described for some
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
6
8ꢀbenzyloxycaffeine analogs. The 3ꢀchlorostyryl protrudes into the entrance cavity in s-cis
conformation, favored by the lower steric hindrance of N9 compared to the N7ꢀmethyl group. In
this pose, the mꢀchlorine atom occupies the same region as the mꢀfluorine of safinamide and mꢀ
3
7
chlorine of some coumarin inhibitors (Figure 4B) in their crystallographic complexes. The
nitrogen atom in position 9 of CSC does not make polar interactions with the amino acids in the
catalytic site. The best docking solution obtained for 17a was strictly superimposable on that
described for CSC (data not shown). To test the reliability of these poses, we constrained the s-cis
geometry of the styryl group in the 9ꢀdeaza series into a pyrimido[4,5ꢀb]indole nucleus (14a-c).
This tricyclic ring could be docked within the MAOꢀB active site with the metaꢀchlorine of 14b
occupying the same position as that of 3ꢀfluoro of safinamide (Figure 4A).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. (A) Best docking poses for CSC (yellow carbons) and 14b (orange carbons) within the
MAOꢀB active site. The coꢀcrystallized safinamide is depicted with green transparent carbons and
the FAD cofactor in ballꢀandꢀsticks with white carbons. (B) Chemical structures of MAOꢀB
inhibitors safinamide and coumarin derivative.
17c
Consistently with what had been observed in the xanthine series, removal of the mꢀchlorine (14a)
led to a decrease of inhibitory potency for MAOꢀB. Limited MAOꢀB inhibitory potency was also
observed for compound 14c, having a more flexible benzyloxy substituent in position 8.
14
ACS Paragon Plus Environment

Page 15 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Unfortunately, the tricyclic compounds were devoid of binding affinity for the A2A receptor. For
this reason we decided to abandon the exploration of pyrimido[4,5ꢀb]indole scaffold and went back
to the 9ꢀdeazaxanthine series. 8ꢀStyrylꢀ9ꢀdeazaxanthines did not strictly follow the same structureꢀ
activity relationships (SARs) as 8ꢀstyrylxanthines, lacking the positive effect of the mꢀchlorine
group. This might be attributed to different equilibria between s-cis and s-trans conformations, and
prompted us to further explore the SARs for this class of compounds. We thus focused on the role
of substituents in position 1, 3, 7 and 8. Substituents with different size, shape, lipophilicity and
electronic properties were introduced in meta or para position of the 8ꢀstyryl radical. Meta
substituents did not lead to an increase of binding affinity for the A2A receptor. Indeed, a chlorine
(17a) or a methoxy group (17c) slightly reduced binding affinity, while trifluoromethyl (17d),
hydroxyl (17u), acetyl (17o) or cyano (17p) substitution led to inactive compounds. Tests
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
performed on rat A receptors demonstrated that a mꢀchlorine doubled the binding affinity and that
2
a mꢀmethoxy and a mꢀtrifluoromethyl were tolerated by the corresponding 8ꢀstyrylxanthine analogs,
2
highlighting a different SAR profile. Metaꢀsubstituted compounds carrying lipophilic groups had
the highest MAOꢀB inhibitory potencies, with the best results obtained for the methoxy (17c) and
the trifluoromethyl (17d) derivatives. Hydrophilic substituents were not tolerated (17u, 17p).
Compound 17c was the best metaꢀsubstituted derivative, endowed with balanced potencies at the
two targets, even if with limited MAOꢀB/MAOꢀA selectivity. On the 8ꢀstyrylxanthine series
electronꢀwithdrawing groups showed a more pronounced effect. Indeed, mꢀchlorine and mꢀ
trifluoromethyl increased inhibitory potencies about ten folds, compared to the four folds observed
17c
for the 8ꢀstyrylꢀ9ꢀdeazaxanthine analogs.
A larger exploration was performed on the para position, but only the acetyl substituent (17q) led to
a modest increase of A2A binding affinity compared to the unsubstituted compound 17b. Lipophilic
substituents produced a limited decrease of binding affinity. Indeed, fluorine (17e), chlorine (17f),
trifluoromethyl (17g) and methoxy (17h) groups were tolerated, as well as the longer nꢀpropyloxy
substituent (17i), while the bromine derivative 17j completely lost A2A binding affinity. Hydrophilic
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
methylsolfonyl (17r), cyano (17s) and hydroxyl (17v) groups were not tolerated. Lipophilic,
electronꢀwithdrawing groups, such as halogens (17e, 17f and 17j) and trifluoromethyl (17g),
produced the greatest increase in MAOꢀB inhibitory potency. A pꢀchlorine and a pꢀtrifluoromethyl
substituent increased inhibitory potencies in the 8ꢀstyrylxanthine series as well, while fluorine had a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17c
negligible effect. Summarizing, different compounds with good potencies at both targets were
obtained within the 8ꢀ(paraꢀsubstitutedꢀstyryl)ꢀ9ꢀdeazaxanthines series, such as the acetyl (17q),
fluorine (17e), trifluoromethyl (17g) and, in particular, the chlorine derivative 17f. Compound 17f
i
(ST3564) is characterized by the best combination of A2A binding affinity (K = 260 nM) and MAOꢀ
B inhibitory potency (IC50 = 200 nM) and it is selective versus MAOꢀA (IC50 = 10000 nM). It is
more potent on MAOꢀB than CSC and the corresponding (E)ꢀ8ꢀ(4ꢀchlorostyryl)caffeine (25), that
we prepared and tested in our experimental conditions.
Other structural modifications were attempted on the 8ꢀstyrylꢀ9ꢀdeazaxanthine scaffold. An oꢀ
hydroxyl group (17w) reduced the potency on both targets, as already observed for the same
substituent in meta (17u) and para (17v) positions. The dichloroꢀderivative 17k was the most potent
MAOꢀB inhibitor in the 8ꢀstyrylꢀ9ꢀdeazaxanthine series (IC50 = 133 nM), but it did not bind to A2A
receptors. A similar behavior was shown by compound 17l, with a methylenedioxy portion bridging
positions meta and para, which was tolerated at the MAOꢀB binding site only. The lack of A2A
binding affinity of 17l represents another difference from the xanthine series. Indeed, 8ꢀ(3,4ꢀ
dimethoxystyryl)caffeine had
K
i
=
18 nM and 1ꢀpropargylꢀ3,7ꢀdimethylꢀ8ꢀ[3’,4’ꢀ
6
(methylenedioxy)styryl]xanthine had K
i
= 35 nM on rat striatum A2A receptors. Replacement of the
benzene ring in the styryl portion with a more hydrophilic and basic ring was detrimental for MAOꢀ
B inhibitory potency, as both the 4ꢀpyridyl (17t) and 3ꢀaminoꢀ2ꢀpyridyl (17n) derivatives were
inactive. Compound 17n showed modest affinity for A2A receptors, about five times lower than that
of its benzene analog 17b.
16
ACS Paragon Plus Environment

Page 17 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We prepared (E)ꢀ8ꢀ(4ꢀchlorostyryl)caffeine (25) to compare mꢀ and pꢀchlorineꢀsubstituted
compounds. For both xanthine and 9ꢀdeazaxanthine series, para substitution resulted in higher hA2A
binding affinity and hMAOꢀB inhibitory potency than those observed for meta substitution.
As already stated, SARs for our 8ꢀstyrylꢀ9ꢀdeazaxanthine derivatives are different from those
reported on rat A2A receptors for 8ꢀstyrylꢀxanthines, where meta substitution and meta-para
disubstitution were favorable, contrary to what we observed on human receptor. On the other hand,
para substitution with a chlorine atom gave higher hA2A binding affinity in both series (17f and 25
for 8ꢀstyrylꢀ9ꢀdeazaxanthine and 8ꢀstyrylꢀxanthine, respectively). To investigate if different binding
modes could be supposed for the two series, we docked the most potent paraꢀsubstituted 8ꢀstyrylꢀ9ꢀ
deazaxanthine derivatives 17f and 17q into the binding site of the A2A receptor, coꢀcrystallized with
the xanthine derivative XAC. In their best docking poses the two compounds showed an
accommodation inside the binding site very close to that of XAC, undertaking interactions with the
same amino acids (Figure 3C), and 8ꢀstyrylꢀxanthines docked in the same pose as well (data not
shown). Interestingly, the acetyl group of 17q had its carbonyl superposed to the amide group of
XAC. Therefore, no significant differences can be seen among the poses of the two classes, with
the exception of the already cited possibility to undertake interactions with the cluster of water
molecules within the binding site, and differences with SARs reported for 8ꢀstyrylꢀxanthines should
be ascribed to different receptor origins.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The styryl substituent of (E)ꢀ8ꢀstyrylxanthines is known to undergo lightꢀinduced isomerization to
1
6b,38
the less potent Z isomer.
We therefore investigated if it was possible to replace the 8ꢀstyryl
group with substituents lacking the double bond, while maintaining the ability to interact with both
3
9
targets. The benzyloxy group is typical of numerous MAOꢀB inhibitors, such as safinamide,
37
coumarin derivatives and also xanthine inhibitors.
We therefore prepared the 8ꢀ(pꢀ
chlorobenzyloxy)ꢀ9ꢀdeazacaffeine 19a, as the pꢀchlorine derivative 17f was the best one in the
styryl series. Compound 19a showed the highest MAOꢀB inhibitory potency (IC50 = 48 nM) in the
whole series, about four folds higher than that of the styryl analog 17f and with a good selectivity
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
versus MAOꢀA (Table 2). Unfortunately, it did not interact with the A2A receptor. The presence of
an additional mꢀtrifluoromethyl group (19b) led to a ten fold decrease of MAOꢀB inhibitory potency
with no effect on A2A binding affinity, while an increase in the size of the substituent in the βꢀ
naphthylmethyloxy derivative 19c led to a completely inactive compound. Shortening the 8ꢀ
methyloxy linker to an oxygen atom, as in the mꢀchlorophenyloxy (20a) and mꢀbiphenyloxy (20b)
derivatives, was also detrimental for the ability to interact with both targets. On the other hand,
replacement of the styryl double bond with a triple one led to the mꢀchlorophenylꢀ and pꢀ
chlorophenylꢀethynyl derivatives 18a and 18b having MAOꢀB inhibitory potencies higher than the
corresponding styryl analogs. However, neither one showed affinity for the A2A receptor.
Differently, the 1ꢀpropargylꢀ3,7ꢀdimethylxanthine carrying an 8ꢀphenylethynyl substituent
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
38
displayed K = 300 nM on A receptors of rat striatum. Insertion of a methylene spacer between
i
2A
the phenyl ring and the triple bond led to compound 18c which was inactive at both targets.
Replacement of the double bond with an amide group (21) or insertion of a 6ꢀ
chlorobenzo[d]oxazolꢀ2ꢀyl substituent (23) in position 8 was not tolerated by either targets. When a
1
,2,3ꢀtriazolyl ring replaced the ethylene linker (22) a good MAOꢀB inhibitory potency was
maintained.
SARs for A2A xanthine antagonists showed that longer alkyl substituents on the nitrogen atoms led
2
,38
to higher A2A binding affinity than the corresponding methyl analogs.
Therefore we inserted
ethyl groups in positions 1, 3, or 7 of the 9ꢀdeazaxanthine nucleus. Contrary to expectations, ethyl
groups had a negative effect on potency, irrespective of the nature of the substituent in position 8.
Inactive compounds at both targets were obtained with 8ꢀ(pꢀchlorostyryl) (17y, 17x), 8ꢀ(pꢀ
chlorophenylethynyl) (18d) and 8ꢀ(pꢀchlorobenzyloxy) (19d) substituents. This further confirms the
different SAR profile observed for xanthine and 9ꢀdeazaxanthine derivatives.
Compound 17f, having balanced A2A /MAOꢀB activity, was tested against a panel of fiftyꢀthree
40
receptors, ion channels and transporters. At the concentration of 10 ꢀM it produced 59%
displacement of specific ligand at the A receptor, 60% at the A receptor and 91% at the A
1
3
2A
18
ACS Paragon Plus Environment

Page 19 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
receptor. Compound 17f appears therefore not selective for the A2A subtype, having also affinity for
the A and A subtypes. However, interaction with the A receptor could be a positive element, due
1
3
1
to the ability of A
1
antagonists to facilitate dopamine release and to counteract PDꢀrelated
4
1
symptoms. Compound 17f did not significantly bind to other receptors or ion channels and
transporters at 10 ꢀM, with the exception of NK and 5ꢀHT2B receptors and norepinephrine
transporter, where it showed 75%, 72% and 57% displacement of specific ligands, respectively
Supplementary Table S2). General behavioral response and antagonism of haloperidolꢀinduced
catalepsy were evaluated in vivo for compound 17f. In the Irwin test, compound 17f induced striatal
biting and licking) and limbic (sniffing) stereotypes, after 90 minutes post treatment at the doses of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
(
(
30 and 100 mg/kg. The dose of 100 mg/kg presented licking behavior after 240 minutes post
treatment. No stereotypes were observed at the dose of 10 mg/kg. All the animals showed catalepsy
before compound administration. Compound 17f significantly antagonized haloperidolꢀinduced
catalepsy at doses of 30 and 100 mg/kg (Figure 5, p<0.05), while it was inactive at a dose of 10
mg/kg.
1
1
20
00
80
60
40
20
0
*
*
halop. + 0
halop. + 10 halop. + 30 halop. + 100
Figure 5. Doseꢀresponse curve of compound 17f on haloperidolꢀinduced catalepsy in mice (AUC).
Columns represent mean + S.E.M. of 8 mice. Haloperidol (2 mg/kg) was administered i.p. 2.5 h
before compound 17f. AUC was calculated throughout three hours recording. One way Anova:
F3,28=4.9; p<0.01; Dunnett’s test: * p<0.05 vs 0 mg/kg.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Conclusions
Structural modulation of CSC allowed us to obtain 9ꢀdeazaxanthine derivatives acting as A2A
antagonists/MAOꢀB inhibitors. Chemical exploration in position 8 of the 9ꢀdeazaxanthine scaffold
provided the pꢀchlorostyryl derivative 17f characterized by balanced potencies at the two targets,
and by reversing haloperidolꢀinduced catalepsy in mice. Replacement of the 8ꢀstyryl portion of CSC
with either a benzyloxy (19a) or a phenylalkynyl (18b) substituent led to compounds with
remarkable MAOꢀB inhibitory potencies, higher than those of CSC and deprenyl. Structural
requirements for A2A binding were stringent and could be met by 8ꢀstyryl derivatives only. SARs
for 9ꢀdeazaxanthines were different from those reported for xanthines. The observed variance may
be due either to differences in the structure of the two classes or in the biological tests that were
used to characterize the compounds. Indeed, data for xanthines were mainly obtained from rat
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
A /A2A receptors and baboon or rat MAOꢀB, while our data come from human proteins.
Experimental Section
Chemistry
General methods. All reactions were run in air except when differently noted. Column
chromatography purifications were performed in flash conditions using Merck 230ꢀ400 Mesh silica
gel. Analytical thin layer chromatography (TLC) was carried out on Merck silica gel plates (silica
1
13
gel 60 F254), that were visualized by exposure to ultraviolet light. H NMR and C NMR spectra
were recorded on a Bruker Avance 200 spectrometer, using CDCl or DMSOꢀd as solvent.
3
6
Chemical shifts (δ scale) are reported in parts per million (ppm) relative to the central peak of the
solvent. Coupling constants (J values) are given in Hertz (Hz). EIꢀMS spectra (70 eV) were taken
on a Fison Trio 1000, molecular ions (M+ or Mꢀ) are given. ESIꢀMS spectra were taken on a
20
ACS Paragon Plus Environment

Page 21 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Waters Micromass ZQ instrument, only molecular ions (M + 1 or M ꢀ 1) are given. IR spectra were
ꢀ1
obtained on a Nicolet Avatar 360 FTꢀIR spectrometer, absorbance are reported in cm . Melting
points were determined on a Buchi SMPꢀ510 capillary melting point apparatus and are uncorrected.
Elemental analyses were performed on a Carlo Erba analyzer and the results are within ± 0.4 of the
theoretical values (C,H,N). Purity of tested compounds was greater than 95%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General Procedure A for the Heck Coupling of 8-Halogen-9-deazaxanthine and styrene
derivatives (17a-n,x,y). A flameꢀdried Schlenk tube was charged with potassium acetate (73 mg,
0
.74 mmol), tetrabutylammonium bromide (119 mg, 0.37 mmol), powder 3Å molecular sieves (74
mg) and dry DMF (0.4 mL), the mixture was stirred for 15 minutes. The appropriate 6ꢀBromoꢀ
,3,5ꢀalkylꢀ1Hꢀpyrrolo[3,2ꢀd]pyrimidineꢀ2,4(3H,5H)ꢀdione (15a,c,d) (0.37 mmol) and the
opportune styrene (0.74 mmol) was successively added and the suspension was stirred for another
1
1
5 minutes before addition of Pd(OAc)
0 hours and then cooled at room temperature, diluted with CH
Cl (3 x 10 mL). The solvent was evaporated under reduced pressure and the
residue obtained was purified by flash chromatography (cyclohexane ethyl acetate 7:3).
2
(4 mg, 0.019 mmol). The mixture was stirred at 80 °C for
2
2
Cl (3 mL), filtered over Celite
2
and washed with CH
2
2
For
a
representative example. (E)-6-(3-Chlorostyryl)-1,3,5-trimethyl-1H-pyrrolo[3,2-
+
d]pyrimidine-2,4(3H,5H)-dione (17a): orange solid; (67 mg, 55%); MS (ESI): 330ꢀ332 [M+H] ;
1
H NMR (CDCl ): δ 3.42 (s, 3H), 3.49 (s, 3H), 4.11 (s, 3H), 6.19 (s, 1H), 6.90 (d, 1H, J = 16.2 Hz),
3
13
7
.08 (d, 1H, J = 16.2 Hz), 7.29ꢀ7.37 (m, 3H), 7.50 (s, 1H); C NMR (CDCl
3
): δ 27.9, 31.7, 31.8,
9
0.9, 111.4, 116.1, 124.9, 126.3, 128.5, 130.1, 131.4, 134.9, 135.6, 138.1, 139.5, 151.5, 155.8;
ꢀ1
FTIR (nujol, cm ): 1647, 1685, 2854, 2924; mp: decomposition with color change starting from
2
00 °C (ethanol); Anal. (C17 ) C, H, N.
3 2
H16ClN O
General Procedure B for the Heck Coupling of 8-vinyl-9-deazaxanthine and aryl bromide
(17o-t). A flameꢀdried Schlenk tube was charged with 1,3,5ꢀtrimethylꢀ6ꢀvinylꢀ1Hꢀpyrrolo[3,2ꢀ
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
d]pyrimidineꢀ2,4(3H,5H)ꢀdione (16) (80 mg, 0.36 mmol), the opportune aryl bromide (0.73 mmol),
TEA (0.24 mL, 1.69 mmol), P(oꢀtol) (22 mg, 0.072 mmol), Pd(OAc) (8 mg, 0.036 mmol) and dry
3
2
DMF (3.6 mL). The mixture was stirred at 80 °C for 2 hours, under N
temperature, diluted with CH Cl (15 mL) and saturated aqueous sodium chloride solution (15mL).
The phases were separated and the aqueous phase was extracted with further CH Cl (15 mL). The
combined organic phases were dried over anhydrous Na SO , the solvent was evaporated under
2
and then cooled at room
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
2
4
reduced pressure and the residue obtained was purified by flash chromatography (gradient from
cyclohexane/ ethyl acetate 7:3 to ethyl acetate/ methanol 9:1).
For
a
representative example. (E)-6-(4-Acetylstyryl)-1,3,5-trimethyl-1H-pyrrolo[3,2-
+
1
d]pyrimidine-2,4(3H,5H)-dione (17q): yellow solid; (73 mg, 60%); MS (EI): 337 (M) ; H NMR
(
CDCl ): δ 2.62 (s, 3H), 3.42 (s, 3H), 3.49 (s, 3H), 4.12 (s, 3H), 6.23 (s, 1H), 7.14 (s, 2H), 7.58 (d, J
3
1
=
8.5 Hz, 2H),7.98 (d, J = 8.5 Hz, 2H); H NMR (Acetoneꢀd ): δ 2.59 (s, 3H), 3.29 (s, 3H), 3.45 (s,
6
3
H), 4.15 (s, 3H), 6.61 (s, 1H), 7.40 (d, J = 16.2 Hz, 1H), 7.54 (d, J = 16.2 Hz, 1H), 7.79 (d, J = 8.5
1
3
Hz, 2H), 8.01 (d, J = 8.5 Hz, 2H),; C NMR (CDCl
126.6, 129.0, 131.5, 135.6, 136.7, 139.4, 140.7, 151.5, 155.8, 197.1; FTIR (nujol, cm ): 1693,
674, 1655; mp: decomposition with color change starting from 250 °C (ethanol); Anal.
C H N O ) C, H, N.
3
): δ 26.5, 27.8, 31.6, 31.8, 91.2, 111.6, 117.2,
ꢀ1
1
(
1
9
19
3
3
Pharmacology
Inhibitory potency on Monoamine Oxidase
Monoamine oxidase (MAO) activity was evaluated by a commercial kit (MAOꢀGlo Assay,
Promega). The kit provides a homogeneous luminescent method for measuring MAO activity and
the effects on it of test compounds. The assay is performed by incubating the MAO enzyme with a
luminogenic MAO substrate, which is a derivative of beetle luciferin ((4S)ꢀ4,5ꢀdihydroꢀ2ꢀ(6ꢀ
hydroxybenzothiazolyl)ꢀ4ꢀthiazolecarboxylic acid) converted by MAO to methyl ester luciferin.
22
ACS Paragon Plus Environment

Page 23 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
After the MAO reaction has been performed, the “Luciferin Detection Reagent” is added to
simultaneously stop the MAO reaction, convert the methyl ester derivative to luciferin and produce
light. The amount of light produced is directly proportional to MAO activity. The experiments was
performed in duplicate in 96ꢀwell plates at room temperature. In each well, the reaction mixture was
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
m
composed of 12.5 ꢂL of MAO substrate at its K values, corresponding to a final concentration of
40 ꢂM for reaction with MAOꢀA and 4 ꢂM for reaction with MAOꢀB, 12.5 ꢂL of test compounds
and 25 ꢂL of either human MAOꢀA or MAOꢀB (Sigma) at a final concentration of 1 ꢂg of
protein/well. After one h of incubation, 50 ꢂL/well of “Luciferin Detection Reagent” was added to
each well and, after additional 20 min of incubation to generate and stabilize the luminescent signal,
the plate was read at the luminometer (Wallac Victor). The net signal from reactions of substrate
and MAO enzyme in the absence of inhibitors represents the positive control (total MAO activity),
whereas the signal from reaction of substrate and inhibitors without MAO enzyme represents the
negative control. Compounds were first examined at a fixed concentration of 10 and 1 ꢂmol/L.
Then, if active at this concentration, a concentrationꢀresponse curve was performed. Results were
expressed as percent of control values and, in the case of multiple concentrations, as IC50
(concentration causing a half maximal inhibition of control values). Four experiments were
performed in duplicates.
A2A receptor binding affinity
Membranes from human embryonic kidney (HEK) 293 cells, stably transfected with the human
adenosine A2A receptor gene, were used in the radioligand binding experiments. Competition
binding experiments were performed incubating membranes (5ꢀ10 ꢂg of protein/sample) with a
3
single concentration of the A2A antagonist [ H]ZM241385 (Biotrend, Cologne, Germany) (2
nmol/L), in the presence of fixed concentrations of test compounds in 96ꢀwell filter plates
(
MultiScreen system, cat. MAFBN0B10, Millipore, Billerica, MA, USA) for one h at 4 °C in a total
volume of 200 ꢂL/well of appropriate buffer (50 mmol/L TrisꢀHCl, pH 7.4, 10 mmol/L MgCl ).
23
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Some compounds were retested at increasing concentrations ranging from 0.01 nmol/L to 10
ꢂmol/L. Nonspecific binding was determined in the presence of 10 ꢂmol/L cold ZM241385 (Tocris,
Ellisville, MO, USA). At the end of incubation, bound and free radioligands were separated by
filtering the 96ꢀwell filter plates using a Millipore filtration apparatus (MultiscreenHTS vacuum
manifold). Filter plates were then washed several times with iceꢀcold buffer (50 mmol/L TrisꢀHCl,
pH 7.4) and filterꢀbound radioactivity measured using a MicroBeta counter (PerkinElmer) after
addition of 30 ꢂL/well of OptiPhase SuperMix scintillation cocktail (PerkinElmer). Four
experiments were performed in duplicates. IC50 values were determined by nonlinear fitting
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
i
strategies with the program GraphPad Prism. The K values were calculated from the IC50 values in
4
2
accordance with the ChengꢀPrusoff equation.
In-vivo experiments
All experiments were conducted in accordance with the guidelines for care and use of experimental
animals of the European Communities Directive (86/609 EEC; 27.01.1992, No.116) and approved
by the company veterinarian and the Italian Ministry of Health. Experiments were performed
according to a randomized schedule. Compound 17f (ST3564) was suspended in a solution
containing 0.1% Tween 80 in 0.5 carboxymethylcellulose CMC (medium viscosity) (SigmaꢀAldrich,
Milan), that was used as vehicle. The drug was administered in a volume of 10 mL/kg in mice.
Male CDꢀ1 mice (Charles River, Calco, Italy), 5ꢀ6 wk old, were kept for 1 wk at 22 + 2, 22 + 2 °C,
at 50 + 15 % relative humidity, with 15ꢀ20 airꢀvolume/h changes, 12ꢀh light/dark cycle (lights on at
0
7:00 hours). Animals were groupꢀhoused (n=10) in Makrolon R cages (42.5 cm x 26.6 cm x 16
high cm) with standard food pellets and water ad libitum.
Irwin test
The test was performed at 60 and 240 min after oral 10, 30 and 100 mg/kg administration of
compound 17f. Mortality was recorded during the 24 hr postꢀdose injection. General parameters
(salivation, lacrimation, diarrhea, ptosis, tremors, convulsions, piloerection, Straub tail,
24
ACS Paragon Plus Environment

Page 25 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
aggressiveness and stereotypes) and specific observations (loss of reflexes, catalepsy, motor activity,
hot plate, body temperature and acute death) were recorded.
Catalepsy
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Animals were gently placed by their forepaws on a small metal bar at a height of 4.5 cm. Catalepsy
was induced by haloperidol (2 mg/kg) injected intraperitoneally 2.5 hours before oral administration
compound 17f (10, 30, 100 mg/kg) or vehicle. At time 0 minutes, successful induction of catalepsy
in all animals was checked before compound administration, then, catalepsy was scored every 60
min for three hours. The catalepsy was measured as the time necessary for the animal to step down
with at least one forepaw with a cut off time for each animal of 60 seconds; after this time the
mouse was gently removed from the wire. Catalepsy was recorded using a videoꢀcamera and by an
observer who was unaware of the treatment. Area under curve (AUC) throughout three hours was
calculated and oneꢀway Anova followed by Dunnett’s test. Basal time was not included because this
timeꢀpoint was used only to check that catalepsy was successfully induced in all animals.
Molecular modeling
Molecular modeling was performed using the Schrodinger software suite. Receptor and ligand
4
3
44
structures were prepared in Maestro 9.2 and refined using Macromodel 9.9. Docking studies
35
were carried out with Glide 5.7 using the SP scoring function. Default settings were used, unless
stated otherwise.
Docking studies into MAO-B
37
The crystal structure of human MAOꢀB in complex with safinamide (PDB code: 2V5Z) was
selected for docking studies. After the correction of valences of the FAD cofactor and of the coꢀ
crystallized ligand, hydrogen atoms were added to the structure and protonation states for ionizable
side chains were chosen to be consistent with physiological pH. The allꢀatoms receptor structure
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
5
was submitted to an energy minimization procedure using the MMFFs force field to a
ꢀ1
ꢀ1
convergence threshold of 0.05 kJ mol Å , holding all heavy atoms fixed. Starting structures of
compound 14b and CSC were minimized with the MMFFs force field to a convergence threshold of
ꢀ1
ꢀ1
0
.05 kJ mol Å . Glide grids were centered on the coꢀcrystallized safinamide, setting the enclosing
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
box and bounding box dimensions to 30 Å and 10 Å, respectively. Twenty docking poses were
collected for each ligand. The topꢀranked poses according to the GlideScore value obtained for
compound 14b and CSC were merged into the MAOꢀB crystal structure and the resulting
complexes were energyꢀminimized using the MMFFs force field to a convergence threshold of 5 kJ
ꢀ1
ꢀ1
mol Å , keeping the protein backbone fixed.
Docking studies into A2A receptor
33
The crystal structure of human A2A receptor in complex with XAC (PDB: 3REY) was retrieved
from the Protein Data Bank and processed using the Protein Preparation Wizard. After the
4
6
reconstruction of the missing Lys150ꢀGln157 sequence with Prime 3.0, the protonation states for
all ionizable side chains were assigned consistent with physiological pH. The overall hydrogen
bonding network was optimized by adjusting the tautomerization states of histidine residues and by
sampling the orientation of hydroxy and thiol groups, as well as the side chain amides of asparagine
and glutamine residues. Protein Cꢀ and Nꢀtermini were capped with acetyl and methylamino groups,
respectively. The allꢀatoms receptor structure was submitted to a restrained minimization procedure
4
7
to the RMSD of 0.3 Å applying the OPLS2005 force field. A loop refinement procedure was then
46
performed with Prime on the Pro149ꢀGly158 sequence to optimize the geometry of the newlyꢀ
inserted residues. Ten different conformations were generated for the Pro149ꢀGly158 loop and the
structure having the lowest energy was selected. Initial conformations of compounds 17f and 17q
ꢀ1
ꢀ1
were energyꢀminimized applying a convergence threshold of 0.05 kJ mol Å using the OPLS2005
47
force field. Glide grids were centered on the coꢀcrystallized XAC, setting the dimensions of
enclosing and bounding boxes to 30 Å and 10 Å, respectively. Twenty docking poses were retained
26
ACS Paragon Plus Environment

Page 27 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
for each ligand and subsequently ranked according to their GlideScore value. The topꢀranked poses
for compounds 17f and 17q were selected and are depicted in Figure 3C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information:
Experimental details and characterizations for compounds 2a,b, 3a,b, 4a,b, 6, 8a,b, 9a,b, 10a-c,
1
1a-c, 12a-c, 13a-c, 14a-c, 17b-p,r-w, 18a-d, 19a-d, 20a,b, 21-23 and 25; selectivity profile of
compound 17f. This material is available free of charge via the Internet at http://pubs.acs.org.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
Table 1. A2A adenosine receptor affinities (K ) and MAOꢀB and MAOꢀA inhibitory potencies (IC ) for the synthesized compounds.
i
50
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
a
Ki hA2A (nM)
IC50 hMAO-B (nM) IC50 hMAO-A (nM)
(95%
1
2
3
Compound
R
R
R
(95% confidence
interval)
(95% confidence
interval)
confidence
interval)
3
34.0
Deprenyl
(260.7ꢀ427.9)
1
2.4
Clorgyline
(11.6ꢀ13.2)
2
04.2
587.6
(260.7ꢀ1324.0)
CSC
(
90.2ꢀ462.5)
KW-6002
47.4
>10000
(15.7ꢀ143.3)
2
8
ACS Paragon Plus Environment

Page 29 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
4a
4b
6
N
N
N
>10000
>10000
>10000
>10000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
N
>10000
>1000
N
O
O
N
O
N
N
O
N
1
4a
7630
>10000
(
5610ꢀ10360)
14b
1390
3110
>1000
(
670ꢀ2870)
(530ꢀ18400)
2
9
ACS Paragon Plus Environment