1,3-Dialkyl-substituted tetrahydropyrimido[1,2-f]purine-2,4-diones as multiple target drugs for the potential treatment of neurodegenerative diseases

Article abstract of DOI:10.1016/j.bmc.2013.09.044

Adenosine receptors and monoamine oxidases are drug targets for neurodegenerative diseases such as Parkinson's and Alzheimer's disease. In the present study we prepared a library of 55 mostly novel tetrahydropyrimido[2,1-f] purinediones with various substituents in the 1- and 3-position (1,3-dimethyl, 1,3-diethyl, 1,3-dipropyl, 1-methyl-3-propargyl) and broad variation in the 9-position. A synthetic strategy to obtain 3-propargyl-substituted tetrahydropyrimido[2,1-f]purinedione derivatives was developed. The new compounds were evaluated for their interaction with all four adenosine receptor subtypes and for their ability to inhibit monoamine oxidases (MAO). Introduction of mono- or di-chloro-substituted phenyl, benzyl or phenethyl residues at N9 of the 1,3-dimethyl series led to the discovery of a novel class of potent MAO-B inhibitors, the most potent compound being 9-(3,4-dichlorobenzyl)-1,3-dimethyl- 6,7,8,9-tetrahydropyrimido[1,2-f]purine-2,4(1H,3H)-dione (21g, IC₅₀ human MAO-B: 0.0629 μM), which displayed high selectivity versus the other investigated targets. Potent dually active A₁/A_2A adenosine receptor antagonists were identified, for example, 9-benzyl-1-methyl-3-propargyl-6,7,8,9-tetrahydropyrimido[1,2-f]purine-2,4(1H,3H) dione (19f, K_i, human receptors, A₁: 0.249 μM, A _2A: 0.253 μM). Several compounds showed triple-target inhibition, the best compound being 9-(2-methoxybenzyl)-1-methyl-3-(prop-2-ynyl)-6,7,8,9- tetrahydro pyrimido [1,2-f]purine-2,4(1H,3H)-dione (19g, K_i A ₁: 0.605 μM, K_i A_2A: 0.417 μM, IC ₅₀ MAO-B: 1.80 μM). Compounds inhibiting several different targets involved in neurodegeneration may exhibit additive or even synergistic effects in vivo.

Full text of DOI:10.1016/j.bmc.2013.09.044

Accepted Manuscript
1
,3-Dialkyl-substituted tetrahydropyrimido[1,2-f]purine-2,4-diones as multiple
target drugs for the potential treatment of neurodegenerative diseases
Pierre Koch, Rhalid Akkari, Andreas Brunschweiger, Thomas Borrmann,
Miriam Schlenk, Petra Küppers, Meryem Köse, Hamid Radjainia, Jörg
Hockemeyer, Anna Drabczyńska, Katarzyna Kieć-Kononowicz, Christa E.
Müller
PII:
S0968-0896(13)00824-9
DOI:
http://dx.doi.org/10.1016/j.bmc.2013.09.044
BMC 11128
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
9 August 2013
14 September 2013
17 September 2013
Please cite this article as: Koch, P., Akkari, R., Brunschweiger, A., Borrmann, T., Schlenk, M., Küppers, P., Köse,
M., Radjainia, H., Hockemeyer, J., Drabczyńska, A., Kieć-Kononowicz, K., Müller, C.E., 1,3-Dialkyl-substituted
tetrahydropyrimido[1,2-f]purine-2,4-diones as multiple target drugs for the potential treatment of neurodegenerative
diseases, Bioorganic & Medicinal Chemistry (2013), doi: http://dx.doi.org/10.1016/j.bmc.2013.09.044
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Bioorganic & Medicinal Chemistry
1
,3-Dialkyl-substituted tetrahydropyrimido[1,2-f]purine-2,4-diones as multiple target
drugs for the potential treatment of neurodegenerative diseases
a
a
a
a
Pierre Koch, Rhalid Akkari, Andreas Brunschweiger, Thomas Borrmann, Miriam
a
a
a
a
a
Schlenk, Petra Küppers, Meryem Köse, Hamid Radjainia, Jörg Hockemeyer, Anna
b
b
a,
Drabczyńska , Katarzyna Kieć-Kononowicz, and Christa E. Müller *
a
PharmaCenter Bonn, Pharmaceutical Institute, Pharmaceutical Chemistry I, University of
Bonn, An der Immenburg 4, 53121 Bonn, Germany
b
Jagiellonian University Medical College, Faculty of Pharmacy, Department of Technology
and Biotechnology of Drugs, Medyczna 9, PL-30-688 Kraków, Poland
*Corresponding author. Tel.: +49 228 73 2301; fax: +49 228 73 2567. E-mail:
christa.mueller@uni-bonn.de
1

Abstract
Adenosine receptors and monoamine oxidases are drug targets for neurodegenerative diseases
such as Parkinson’s and Alzheimer’s disease. In the present study we prepared a library of 55
mostly novel tetrahydropyrimido[2,1-f]purinediones with various substituents in the 1- and 3-
position (1,3-dimethyl, 1,3-diethyl, 1,3-dipropyl, 1-methyl-3-propargyl) and broad variation
in the 9-position.
A
synthetic strategy to obtain 3-propargyl-substituted
tetrahydropyrimido[2,1-f]purinedione derivatives was developed. The new compounds were
evaluated for their interaction with all four adenosine receptor subtypes and for their ability to
inhibit monoamine oxidases (MAO). Introduction of mono- or di-chloro-substituted phenyl,
benzyl or phenethyl residues at N9 of the 1,3-dimethyl series led to the discovery of a novel
class of potent MAO-B inhibitors, the most potent compound being 9-(3,4-dichlorobenzyl)-
1
,3-dimethyl-6,7,8,9-tetrahydropyrimido[1,2-f]purine-2,4(1H,3H)-dione (21g, IC50 human
MAO-B: 0.0629 µM), which displayed high selectivity versus the other investigated targets.
Potent dually active A /A2A adenosine receptor antagonists were identified, e.g. 9-benzyl-1-
methyl-3-propargyl-6,7,8,9-tetrahydropyrimido[1,2-f]purine-2,4(1H,3H)dione (19f,
human receptors, A : 0.249 µM, A2A: 0.253 µM). Several compounds showed triple-target
inhibition, the best compound being 9-(2-methoxybenzyl)-1-methyl-3-(prop-2-ynyl)-6,7,8,9-
tetrahydropyrimido[1,2-f]purine-2,4(1H,3H)-dione (19g, K : 0.605 µM, K 2A: 0.417 µM,
1
i
K ,
1
i
A
1
i
A
IC50 MAO-B: 1.80 µM). Compounds inhibiting several different targets involved in
neurodegeneration may exhibit additive or even synergistic effects in vivo.
Keywords: Alzheimer’s disease, anellated xanthines, tetrahydropyrimido[2,1-f]purinediones,
1
adenosine A receptor antagonists, adenosine A2A receptor antagonists, caffeine derivatives,
monoamine oxidase B inhibitors, Parkinson’s disease, tricyclic compounds, synthesis
2

1. Introduction
Parkinson's disease (PD) is a chronic, progressive neurodegenerative disorder characterized
1
by motor dysfunctions that are caused by the loss of dopaminergic neurons. So far only
symptomatic treatment of PD has been available. One approach is the dopamine replacement
therapy using levodopa (L-DOPA), the metabolic precursor of dopamine, which remains the
gold-standard for the treatment of PD. However, after a few years of treatment, levodopa
becomes progressively less effective and causes severe side-effects such as dystonia and
dyskinesia. Levodopa is always combined with a peripherally active DOPA decarboxylase
inhibitor that prevents it from being degraded before reaching the brain. It may additionally be
2
combined with monoamine oxidase type B (MAO-B) inhibitors, and/or catechol-O-
3
methyltransferase (COMT) inhibitors to prevent its metabolism as well as that of dopamine.
Adenosine receptors (ARs), in particular the A2A receptor subtype, have emerged as new
4
targets for the treatment of PD. Several A2A-selective AR antagonists have been evaluated in
4,5
®
preclinical and clinical trials. The 8-styrylxanthine derivative istradefylline (Nouriast , 1)
6
has recently been approved for adjunctive treatment of PD in Japan. The development of a
potent
and
selective
A
2A
antagonist,
preladenant,
an
amino-substituted
triazolopyrazolopyrimidine structurally related to adenine, was discontinued after it had not
shown significant effects in phase III clinical trials, whereas the benzothiazole derivative
7
tozadenant, that was evaluated in a phase II study, showed promising results. There are
indications from preclinical studies that A2AAR antagonists as well as MAO-B inhibitors may
8
-11
exhibit neuroprotective effects.
Therefore, A2A antagonists also show promise for the
12
treatment of Alzheimer’s disease (AD). Drugs that act at multiple targets and provide
symptomatic benefits as well as neuroprotective effects have been proposed to be
13-16
advantageous for the treatment of neurodegenerative diseases like PD and AD.
Dual A
1
1
7, 18
and A2AAR antagonists were found to be highly active in different models of PD.
Mihara
3

et al. reported on a dual A
1
/A2AAR antagonist which improved neuronal degeneration and
motor impairment via A2AAR blockade, and showed positive effects on the cognitive
1
9
1
impairment associated with the disease by blocking A ARs. The latter effect was not
observed with the A2A-selective AR antagonist istradefylline (1, Figure 1), that was
investigated in the same study. Recently, Petzer et al. suggested that dual-targeting of
MAO-B by inhibitors and A2AARs by antagonists may be advantageous for the treatment of
2
0
PD. Since both, MAO-B inhibition and A2AAR blockade, may effect neuroprotection, such a
dual-target-directed approach may act synergistically or at least additively thereby possibly
reducing or stopping the devastating progression of neurodegenerative diseases. Thus,
multiple target drugs that are blocking A
benefit for the treatment of PD and AD.
1
and A2AARs as well as MAO-B may be of clinical
8-m-Chlorostyrylcaffeine (CSC, 2) was the first reported A2AAR antagonist which showed
2
0
ancillary MAO-B inhibitory activity (Figure 1).
Figure 1. Structures and K
i
values of the first marketed A2A-selective adenosine receptor
antagonist istradefylline (1) and of the structurally related dual A2AAR antagonist/MAO-B
2
0, 22
inhibitor 8-m-chlorostyrylcaffeine (CSC, 2) (r = rat, h = human).
The dually active A2A antagonist/MAO-B inhibitor 8-m-chlorostyrylcaffeine (2) had been
2
3
docked into the active site of human MAO-B. In the resulting model the caffeine moiety
was oriented towards the more hydrophilic flavine adenine dinucleotide (FAD) cofactor. The
4

styryl was located in a large hydrophobic pocket. The xanthine core may thus be a promising,
well tolerated scaffold for the design of novel MAO-B inhibitors. Recently, several groups
reported on xanthine-derived inhibitors of MAO-B (Figure 3). Song et al. identified
2
4
8
-substituted benzamido-phenylxanthines (3a and 3b) as weak MAO-B inhibitors. The
styryl group present in CSC was replaced by a substituted phenyl ring. Rivara et al. reported
on 9-deazaxanthine derivatives as dual A2AAR antagonists/MAO-B inhibitors; some of the
compounds were found to be potent and selective MAO-B inhibitors (e.g. compounds 3c and
2
5
3d).
Recently, the first non-xanthine-derived dually active A2A antagonists / MAO-B inhibitors
2
1
were reported.
24, 25
Figure 2. Xanthine-derived MAO-B inhibitors (h = human).
A
1
ARs are widely distributed in the central nervous system (CNS), with high levels being
expressed in many regions of the brain, and in several peripheral organs and tissues, including
2
6
heart, lung, kidney, and fat cells.
A2AARs show a more restricted expression pattern in
midbrain regions of the CNS, such as the striatum. They co-localize and physically associate
27,28
with dopamine D
2
receptors, resulting in reciprocal antagonistic interactions.
5

Consequently, pharmacological blockade of A2AARs has shown beneficial effects in
preclinical animal models of PD and a relief of Parkinsonian symptoms in 1-methyl-4-phenyl-
2
9, 30
1
,2,3,6-tetrahydropyridine (MPTP)-treated non-human primates.
The first AR antagonists described were the alkylxanthine derivatives caffeine (4) and
theophylline (5). Epidemiological studies suggest that caffeine may be protective against AD
3
1, 32
33
and PD.
Both, 4 and 5, are moderately potent, nonselective AR antagonists. The effects
of substituents at positions 1, 3, 7 and 8 on receptor affinity and subtype-selectivity have been
22
extensively explored.
O
O
O
CH3
H
N
H
N
H C
3
H C
3
H C
3
N
N
N
N
1
7
8
3
N
N
N
O
N
O
N
O
N
CH3
CH3
HO
4
Caffeine
5 Theophylline
6 PSB-36
A = 44.9 µM (h)
1
A = 6.77 µM (h)
1
A = 0.0007 µM (h)
1
A = 41.0 µM (r)
1
A = 14.0 µM (r)
1
A = 0.000124 µM (r)
1
A2A = 23.4 µM (h)
A2A = 32.5 µM (r)
A2B = 33.8 µM (h)
A2A = 6.70 µM (h)
A2A = 22.0 µM (r)
A2B = 9.07 µM (h)
A2A = 0.980 µM (h)
A2A = 0.552 µM (r)
A2B = 0.187 µM (h)
A = 13.3 µM (h)
3
A = 22.3 µM (h)
3
A = 2.30 µM (h)
3
O
O
CH3
O
CH3
N
H3C
N
N
N
3
N
N
N
1
9
N
N
N
O
O
N
N
N
O
CH3
OCH3
CH3
F
RO
7
DMPX
8a MSX-2 R = H 8b MSX-3 R = PO3Na2
9
A = 2.50 µM (h) 8c MSX-4
1
A1 = 16.7 µM (h)
A1 > 25.0 µM (r)
A = 11.0 µM (r)
1
O
CH3
CH3
A2A = 5.60 µM (r)
A2B = 4.13 µM (h)
A = 0.900 µM (r)
1
R =
A2A = 0.00538 µM (h)
A2A = 0.00804 µM (r)
A2B >10.0 µM (h)
A2A = 1.880 µM (h)
A2A = 0.147 µM (r)
A2B = 10.0 µM (h)
+
NH3 Cl
-
A > 10.0 µM (r)
3
A3 >10.0 µM (h)
A = 13.3 µM (h)
3
Figure 3. Potent and selective xanthine-derived adenosine receptor antagonists (r = rat, h =
2
2
human).
Introduction of bulky groups at the 8-position of the xanthine core resulted in antagonists
3
4
with high potency at A
1
ARs, e.g. PSB-36 (6). 3,7-Dimethyl-1-propargylxanthine (DMPX, 7,
propargyl = prop-2-ynyl) was among the first described xanthine derivatives with (moderate)
3
5, 36
1
selectivity for A2A over A ARs and was therefore investigated in in vivo experiments.
6

Evaluation of DMPX at all four AR subtypes revealed that DMPX is similarly potent at A2B
2
2
as at A2AARs. Substitution at position 8 of xanthines with a styryl residue led to selectivity
for the A2AAR subtype. The (E)-8-styrylxanthine derivative istradefylline (KW-6002, 1) was
3
7, 38
among the first A2A–selective ARs antagonists reported.
Another example of this
3
9
structural class is MSX-2 (8a), a highly potent and selective A2AARs antagonist. However,
the styryl double bond is susceptible to photoreactions, such as E/Z-isomerization and
4
0, 41
dimerization, leading to compounds with dramatically lower AR affinity.
Another
drawback of styrylxanthines is their high lipophilicity and low water-solubility. To overcome
the latter drawback, phosphoric acid ester (MSX-3; 8b) and L-valine ester prodrugs (8c) have
3
9, 42
been developed, which exhibit increased water-solubility.
Both are widely used as
3
3, 43, 44
pharmacological tools for in vitro and especially for in vivo studies.
We have previously reported on the development of tetrahydropyrimido[2,1-f]purinediones
e.g. compound 9) as AR antagonists, a class of compounds which can be envisaged as
(
4
5-50
tricyclic caffeine derivatives.
They represent analogs of 8-styrylxanthines, that are
sterically constrained by anellation of a tetrahydropyrimidine ring to the 7,8-position of
caffeine mimicking the (E)-configurated styryl substructure of A2A antagonists 1 and 8. So far,
this approach has led to tricyclic compound 9, a relatively potent A2AARs antagonist with a
4
6
i
K -value of 147 nM at the rat A2AAR along with high selectivity. However, its affinity was
lower at human A2AARs.
Crystallographic data of the A2AAR complexed with the xanthine derivative 8-[4-[[[[(2-
aminoethyl)amino]carbonyl]methyl]oxy]phenyl]-1,3-dipropylxanthine (XAC) were recently
51
published. XAC is a competitive antagonist and was shown to bind in the same binding
pocket as the physiological agonist adenosine. It forms a hydrogen bond with Asn253 and a π-
π-stacking interaction with Phe168. The propyl residues at N1 and N3 of XAC showed
hydrophobic contacts to Leu85 and Leu249, and to Ala81, Ile66 and Val 84, respectively
(
Figure S1, Supplementary data). The phenyl residue at position 8 of XAC is coplanar with
7

the xanthine moiety. The polar substituent is located in a channel formed by Tyr9 and Tyr271.
The substituent at position 8 points towards the extracellular space. This opens the
opportunity for the introduction of different (large) moieties at that position, which we probed
in the present study.
In a continuing effort to develop potential drugs for the treatment of neurodegenerative
diseases, including PD and AD, we performed variations at positions 1, 3 and 9 of the
tetrahydropyrimido[2,1-f]purinedione structure (Figure 4). In order to improve the affinity for
the A2AAR we wanted to replace the 1,3-dimethyl substitution of caffeine by a 1-methyl-3-
propargyl substitution pattern (in analogy to the xanthine derivatives DMPX (7) and MSX-2
(8a)), by a 1,3-diethyl substitution (in analogy to istradefylline (1)) and by 1,3-dipropyl
residues. Furthermore, we planned to introduce substituents with one or two methylene
spacers between N9 and the attached aromatic ring (see Figure 4).
Figure 4. Planned structural modifications of lead compound 9.
8

2. Results and discussion
2.1. Chemistry
The new tetrahydropyrimido[2,1-f]purinediones 19-22 were synthesized by a convergent
approach. Diversity at position 9 was introduced in the last step of the synthetic procedure by
ring-closure of 8-bromo-7-(3-chloropropyl)-1,3-disubstituted xanthines 12a-d with a variety
of amines (Scheme 1).
Scheme 1. Synthetic strategy for the preparation of tetrahydropyrimido[2,1-f]purinediones 19-
22.
4
7, 49, 52
Xanthines 12b-d were synthesized in analogy to previously described procedures.
In this present study we developed the first access to 3-propargyl substituted
tetrahydropyrimido[2,1-f]purinediones, analogs of DMPX (7) and MSX-2 (8a). Due to the
reactivity of the propargyl group, bromination at the 8-position of the propargyl-substituted
xanthines could not be performed. Therefore, precursor 12a had to be synthesized in a
5
3
different way than 12b-d. We used 8-bromo-3-methylxanthine (10) as a starting compound
(Scheme 2). Nitrogen N7 of xanthine 10 was selectively alkylated with 1-bromo-3-
chloropropane in the presence of triethylamine yielding 11. The propargyl residue was
introduced by subsequent alkylation at N1 with propargyl bromide in the presence of
potassium carbonate in dimethylformamide (DMF) yielding xanthine 12a in excellent yield.
9

Scheme 2. Synthesis of 3-methyl-1-prop-2-ynylxanthine-derivative 12a. Reagents and
conditions: (a) 1-bromo-3-chloropropane, NEt , DMF, 16 h, rt, 50%; (b) propargyl bromide,
CO , DMF, 16 h, rt, 99%.
3
K
2
3
In order to introduce large, basic moieties into the target structures, ω-amino-1-(4-
phenyl/benzyl-1-piperazinyl)ethanones and -propanones 17a-f and 18a,d were prepared as
precursors by a convenient two-step procedure (Scheme 3). N-boc-protected glycine (14a) or
β-alanine (14b) was condensed with N-aryl- or N-benzylpiperazines 13a-f in the presence of
the coupling reagent 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide-HCl (EDC). In the last
step, the boc-protection group was cleaved off under acidic conditions.
O
O
O
H
H
N
x HCl
HN
a
N
b
Boc
N
H N
2
Boc
OH
N
N
n
n
n
N
R
N
R
R
1
5a-f
17a-f
1
3a-f
14a,b
16a,d
18a,d
R = a phenyl
b 4-chlorophenyl e 4-trifluoromethylbenzyl
c benzyl f 4-methoxybenzyl
d 4-chlorobenzyl
n = 1 14a, 15a-f, 17a-f
n = 2 14b, 16a,d, 18a,d
Scheme 3: Synthesis of ω-amino-1-(4-phenyl/benzyl-1-piperazinyl)ethanones and –propan-
ones 17a-f and 18a, d. Reagents and conditions: (a) EDC, MeOH, 1 h, rt; (b) 4 M HCl in
dioxane 1 h, rt; 58-88%, over two steps.
8-Bromo-7-(3-chloropropyl)-1,3-dialkylxanthines 12a-d were reacted with commercially
available amines or with ω-amino-1-(4-phenyl/benzyl-1-piperazinyl)ethanones or –propan-
ones 17a-f or 18a, d in the presence of a base in DMF under microwave irradiation (Scheme
4
).
1
0

Scheme 4: Ring closing reaction yielding tetrahydropyrimido[2,1-f]purinediones 19-22.
Reagents: (a) NEt , 1,2-dimethoxyethane, 1 h, 90 °C, 150 W, 7-99% for compounds 19, 21c,
d, g-i, k and 20; K CO , NEt , DMF, 1 h, 90 °C, 150 W, 18-81%, compounds 21l-q and 22
for details see Experimental Section).
3
2
3
3
(
The structure of all compounds was confirmed by NMR and electrospray ionization mass
spectra. Melting points were determined for all new compounds. The purity of all tested
compounds was at least 95% (except for compound 21n, 93.8%) as confirmed by high-
performance liquid chromatography (HPLC) coupled to electrospray ionization mass
spectrometry (ESI-MS) (for details, see Experimental Section).
2
.2 Biological Evaluation
The new compounds were initially evaluated in radioligand binding assays for their affinity at
and A2AARs in rat brain cortical membrane and rat brain striatal membrane preparations,
respectively. Selected compounds were further investigated at human A and A2AARs
recombinantly expressed in Chinese hamster ovary (CHO) cells. All compounds were
additionally investigated at human A2B and A ARs expressed in CHO cells in order to
A
1
1
3
3
6
3
54
3
determine their selectivity. [ H]2-Chloro-N -cyclopentyladenosine ([ H]CCPA), [ H]3-(3-
3
55
hydroxypropyl)-1-propargyl-7-methyl-8-(m-methoxystyryl)xanthine
([ H]MSX-2),
3
3
56
[
H]8-(4-(4-(4-chlorophenyl)piperazine-1-sulfonyl)phenyl)-1-propylxanthine ([ H]PSB-603)
3
and
[ H]2-phenyl-8-ethyl-4-methyl-(8R)-4,5,7,8-tetrahydro-1H-imidazo[2,1-i]purine-5-one
1
1

3
57
(
1 3
[ H]PSB-11) were used as radioligands in A , A2A, A2B and A ARs binding studies,
respectively. The synthesized compounds were further investigated for inhibition of human
MAO-A and MAO-B. Results are presented in Table 1 and Table 2. Data of standard ligands
are included for comparison.
1
2

Table 1. Adenosine receptor affinities of tetrahydropyrimido[1,2-f]purinediones 19-22 and
related xanthine derivatives
H3C
O
O
O
O
O
H3C
H3C
N
N
N
N
N
N
N
N
N
N
N
N
H3C
N
N
N
N
N
N
N
N
9
R
R
R
O
N
O
N
R
O
N
O
N
O
N
CH3
CH3
CH3
H3C
F
9
19
20
21
22
CH3
K
i
± SEM (µM)
human (h); rat (r)
A
1
vs.
A
2A vs.
A
2B vs.
A
3
vs.
3
a
3
a
3
b
3
b
Compound
R
[ H]CCPA
[ H]MSX-2
[ H]PSB-603
[ H]PSB-11
c
c
c
c
Caffeine (1)
44.9 (h)
23.4 (h)
32.5 (r)
33.8 (h)
30.0 (r)
13.3 (h)
c
c
c
c
4
1.0 (r)
>10.0 (r)
d
c
c
DMPX (7)
MSX-2 (8a)
9
nd (h)
1
nd (h)
5.60 (r)
4.13 (h)
>10.0 (h)
c
c
c
1.0 (r)
c
c
c
2.50 (h)
.900 (r)
0.00538 (h)
0.00804 (r)
>10.0 (h)
>10.0 (h)
c
c
0
c
c
c
c
16.7 (h)
25.0 (r)
1.18 (h)
0.147 (r)
10.0 (h)
13.3 (h)
c
c
>
1
-Methyl-3-propargyltetrahydropyrimido[1,2-f]purinediones (19)
9a
3.61 ± 0.82 (h)
.19 ± 0.29 (r)
e
e
e
1
2.12 ± 0.68 (h)
1.94 ± 0.51 (r)
>1.00 (h) (7%)
>10.0 (h) (22%)
1
e
e
e
1
9b
9.49 ± 2.15 (h)
nd (h)
3.03 ± 0.60 (r)
>1.00 (h) (8%)
>1.00 (h) (20%)
>10.0 (h) (34%)
>1.00 (h) (28%)
0.469 ± 0.027 (r)
e
a
1
9c
>10.0 (h) (21%)
0.284 ± 0.072 (h)
0.932 ± 0.094 (r)
3.86 ± 0.29 (h)
0.313 ± 0.074 (r)
e
19d
9.65 ± 1.68 (h)
nd (h)
>1.00 (h) (9%)
0.228 ± 0.040 (r)
2.04 ± 0.04 (r)
e
a
19e
nd (h)
nd (h)
>1.00 (h) (17%)
1.01 ± 0.19 (h)
5.22 ± 0.69 (r)
1.48 ± 0.05 (r)
a
e
19f
0.249 ± 0.060 (h)
.135 ± 0.012 (r)
0.253 ± 0.015 (h)
0.533 ± 0.078 (r)
3.52 ± 0.73 (h)
>10.0 (h) (34%)
0
e
a
19g
0.605 ± 0.157 (h)
.06 ± 0.19 (r)
0.417 ± 0.068 (h)
0.641 ± 0.100 (r)
>1.00 (h) (16%)
4.41 ± 0.71 (h)
1
e
e
1
9h
1.33 ± 0.37 (h)
0.322 ± 0.034 (h)
0.696 ± 0.076 (r)
>1.00 (h) (8%)
>1.00 (h) (16%)
>1.00 (h) (14%)
0
.227 ± 0.033 (r)
e
e
1
9i
10.2 ± 2.7 (h)
nd (h)
>1.00 (h) (7%)
0.555 ± 0.132 (r)
3.03 ± 0.24 (r)
1
3

e
e
19j
nd (h)
nd (h)
>0.300 (h) (1%)
>1.00 (h) (4%)
1.83 ± 0.86 (r)
6.22 ± 1.53 (r)
a
e
1
9k
0.410 ± 0.093 (h)
0.626 ± 0.045 (h)
0.259 ± 0.069 (r)
1.74 ± 0.11 (h)
>10.0 (h) (30%)
>1.00 (h) (10%)
0.114 ± 0.021 (r)
e
e
e
1
9l
11.6 ± 1.4 (h)
1.20 ± 0.30 (h)
0.827 ± 0.157 (r)
>1.00 (h) (3%)
0.632 ± 0.113 (r)
e
19m
8.26 ± 1.59 (h)
0.346 ± 0.102 (h)
0.276 ± 0.136 (r)
>10.0 (h) (12%)
>10.0 (h) (28%)
>1.00 (h) (13%)
0.501 ± 0.126 (r)
e
e
19n
4.40 ± 0.25 (h)
0.508 ± 0.112 (h)
0.288 ± 0.042 (r)
>1.00 (h) (8%)
3.14 ± 0.39 (r)
e
e
19o
nd (h)
10.0 (r) (18%)
nd (h)
0.723 ± 0.027 (r)
>1.00 (h) (-2%)
>1.00 (h) (5%)
e
>
e
e
19p
1.41 ± 0.19 (r)
.990 ± 0.164 (r)
0.317 ± 0.074 (h)
0.710 ± 0.056 (r)
>1.00 (h) (16%)
>1.00 (h) (20%)
0
a
e
e
1
9q
0.866 ± 0.358 (h)
.07 ± 0.17 (r)
2.36 ± 0.49 (h)
0.231 ± 0.041 (h)
0.380 ± 0.038 (r)
3.97 ± 0.46 (h)
>10.0 (h) (13%)
>10.0 (h) (25%)
1
1
9r
9s
0.311 ± 0.079 (h)
0.361 ± 0.069 (r)
17.1 ± 1.3 (h)
0.515 ± 0.030 (r)
e
e
1
1.15 ± 0.29 (h)
1.11 ± 0.11 (h)
1.43 ± 0.17 (r)
>1.00 (h) (6%)
>10.0 (h) (9%)
0.169 ± 0.066 (r)
e
e
1
9t
1.87 ± 0.53 (h)
0.510 ± 0.101 (h)
0.498 ± 0.104 (r)
>1.00 (h) (-5%)
>1.00 (h) (-2%)
>0.100 (h) (5%)
0.219 ± 0.034 (r)
e
e
19u
nd (h)
0.628 ± 0.074 (h)
2.47 ± 0.57 (r)
>1.00 (h) (3%)
e
>10.0 (r) (32%)
1
,3-Diethyltetrahydropyrimido[1,2-f]purinediones (20)
0a
6.65 ± 1.57 (h)
e
e
e
2
nd (h)
1.88 ± 0.51 (r)
>1.00 (h) (11%)
>10.0 (h) (29%)
>10.0 (h) (29%)
0
.545 ± 0.092 (r)
e
2
0b
nd (h)
10.0 (r) (30%)
nd (h)
10.9 ± 1.3 (r)
>1.00 (h) (22%)
>1.00 (h) (21%)
>1.00 (h) (20%)
e
>
OH
e
e
2
0c
nd (h)
.24 ± 0.02 (r)
nd (h)
5.37 ± 0.47 (r)
6.80 ± 0.86 (h)
1
a
20d
3.92 ± 1.01 (h)
nd (h)
3.03 ± 0.20 (h)
0.506 ± 0.063 (r)
1.83 ± 0.34 (r)
e
a
2
0e
nd (h)
10.0 (r) (31%)
nd (h)
2.65 ± 0.09 (r)
>1.00 (h) (29%)
>1.00 (h) (29%)
3.41 ± 0.36 (h)
e
>
e
a
2
0f
nd (h)
10.0 (r) (20%)
2.53 ± 0.310 (h)
0.624 ± 0.148 (r)
6.32 ± 1.24 (h)
e
e
>
e
e
20g
nd (h)
10.0 (r) (35%)
nd (h)
1.88 ± 0.12 (r)
>1.00 (h) (15%)
>10.0 (h) (32%)
>
1
4

e
e
e
2
0h
nd (h)
10.0 (r) (14%)
nd (h)
3.87 ± 1.42 (r)
>1.00 (h) (6%)
>10.0 (h) (24%)
e
>
e
e
2
0i
1.04 ± 0.34 (h)
.321 ± 0.049 (r)
nd (h)
1.70 ± 0.22 (r)
>1.00 (h) (3%)
>10.0 (h) (37%)
>1.00 (h) (13%)
0
e
2
0j
nd (h)
nd (h)
>0.300 (h) (2%)
1.07 ± 0.06 (r)
>10.0 (r) (44%)
1
,3-Dimethyltetrahydropyrimido[1,2-f]purinediones (21)
e
e
e
2
1a
nd (h)
.42 ± 0.86 (r)
>2.50 (r))
nd (h)
3.76 ± 0.15 (r)
(>2.50 (r))
>1.00 (h) (-4%)
>10.0 (h) (33%)
>10.0 (h) (15%)
>1.00 (h) (12%)
>10.0 (h) (21%)
4
(
46
46
e
21b
>1.00 (h) (10%)
>
10.0 (r) (30%)
25.0 (r) (47%)
1.21 ± 0.18 (r)
1.12 ± 0.43 (r)
4
6
46
≥
e
e
e
2
1c
nd (h)
10.0 (r) (28%)
nd (h)
4.70 ± 1.46 (r)
>1.00 (h) (3%)
>
e
21d
nd (h)
10.0 (r) (21%)
4.61 ± 1.12 (h)
5.54 ± 0.48 (r)
>1.00 (h) (11%)
>1.00 (h) (10%)
>
e
a
2
1e
>10.0 (h) (21%)
.432 ± 0.027 (r)
1.42 ± 0.08 (r))
2.03 ± 0.16 (h)
0.802 ± 0.156 (r)
(1.12 ± 0.14 (r))
15.2 ± 5.5 (h)
0
4
7
47
(
e
e
e
e
e
e
2
1f
>1.00 (h) (1%)
>1.00 (h) (5%)
>1.00 (h) (4%)
>10.0 (h) (36%)
>10.0 (h) (30%)
>10.0 (h) (14%)
>
10.0 (r) (29%)
7.23 ± 2.27 (r)
4.28 ± 0.44 (r)
47
47
>25.0 (r) (27%)
21g
nd (h)
.01 ± 0.46 (r)
nd (h)
3.69 ± 0.61 (r)
4
21h
1.70 ± 0.42 (h)
3.78 ± 0.36 (h)
1.64 ± 0.30 (r)
0.165 ± 0.053 (r)
e
e
2
1i
nd (h)
.76 ± 0.20 (r)
1.35 ± 0.30 (h)
0.166 ± 0.021 (r)
>1.00 (h) (1%)
>1.00 (h) (8%)
>10.0 (h) (43%)
1
e
e
e
21j
nd (h)
.03 ± 0.68 (r)
6.47 ± 1.15 (r))
5.61 ± 1.45 (h)
0.515 ± 0.110 (r)
(0.480 ± 0.111 (r))
>10.0 (h) (38 %)
>10.0 (h) (23 %)
3
48
48
(
e
21k
nd (h)
5.25± 0.67 (h)
>1.00 (h) (10%)
7.24 ± 0.68 (r)
0.780 ± 0.143 (r)
e
e
21l
nd (h)
1.50 (r) (3%)
nd (h)
>10.0 (r) (40%)
>1.00 (h) (-15%)
>10.0 (h) (32%)
e
e
>
1
5

e
e
2
1m
nd (h)
1.50 (r) (0%)
nd (h)
>10.0 (r) (39%)
>1.00 (h) (-7%)
>1.00 (h) (-7%)
>1.00 (h) (24%)
e
e
>
e
e
e
e
e
2
1n
nd (h)
1.50 (r) (0%)
nd (h)
14.0 ± 3.4 (r)
>0.300 (h) (8%)
>10.0 (h) (23%)
>10.0 (h) (36%)
>10.0 (h) (15%)
e
>
e
2
1o
1p
1q
O
nd (h)
>1.50 (r) (-1%)
nd (h)
9.42 ± 1.56 (r)
>1.00 (h) (-12%)
e
Cl
N
N
e
2
nd (h)
1.50 (r) (2%)
nd (h)
>10.0 (r) (27%)
>1.00 (h) (2%)
e
e
e
>
e
2
nd (h)
1.50 (r) (0%)
nd (h)
>10.0 (r) (41%)
>1.00 (h) (-8%)
e
>
1
,3-Dipropyltetrahydropyrimido[1,2-f]purinediones (22)
2a
nd (h)
e
a
2
nd (h)
>10.0 (r) (27%)
>1.00 (h) (6%)
2.45 ± 0.14 (h)
e
e
>1.50 (r) (18%)
e
a
22b
nd (h)
1.50 (r) (19%)
nd (h)
>10.0 (r) (36%)
>1.00 (h) (7%)
0.762 ± 0.177 (h)
e
e
>
e
e
22c
nd (h)
nd (h)
>1.00 (h) (0%)
>1.00 (h) (28%)
e
e
e
>
>
>
1.50 (r) (17%)
8.22 ± 2.54 (r)
e
a
2
2d
nd (h)
1.50 (r) (23%)
nd (h)
7.31 ± 1.37 (r)
>1.00 (h) (15%)
0.541 ± 0.087 (h)
e
a
2
2e
nd (h)
1.50 (r) (18%)
nd (h)
>10.0 (r) (33%)
>1.00 (h) (4%)
1.60 ± 0.24 (h)
e
e
e
e
2
2f
nd (h)
1.50 (r) (5%)
nd (h)
>10.0 (r) (14%)
>1.00 (h) (-5%)
>1.00 (h) (-4%)
>1.00 (h) (21%)
e
>
e
e
2
2g
nd (h)
1.50 (r) (8%)
nd (h)
>10.0 (r) (40%)
>1.00 (h) (23%)
e
e
>
a
b
c
22 d
n=3; n=2, unless otherwise noted; literature data taken from ref ; nd = not determined;
inhibition of radioligand binding at indicated concentration.
e
1
6

2.3. Structure-activity relationships at adenosine receptors
2.3.1. 1-Methyl-3-propargyltetrahydropyrimido[1,2-f]purinediones
In the 1-methyl-3-propargyltetrahydropyrimido[1,2-f]purinedione series 19, compound 19a
bearing a 4-fluorophenyl ring at position N9 was a moderately potent dual A
antagonist with K values at both AR subtypes in the low micromolar range. Comparison of
1
/A2AARs
i
compound 19a with the lead compound 9 (both are bearing a para-fluorophenyl moiety at
position 9) revealed that replacement of the N3-methyl residue by a propargyl substituent led
1
to an increase of A AR affinity, but decreased A2AAR affinity at the rat receptor by more than
one order of magnitude. However, comparable A2AAR affinities were observed at the human
receptors for both compounds. Compound 19b with a methoxy group in the meta-position of
the phenyl ring showed a K
µM) and in the low micromolar range at the rat A2AAR. However, at the human A
dramatic decrease in affinity by more than one order of magnitude was observed (for
correlation of human and rat A AR data see Figure S2, Supplementary data). Introduction of
another methoxy group in the meta-position resulted in a slight improvement of A and
2AAR affinity at the rat receptors (19b vs. 19d). Compound 19c with a 3-bromophenyl
i
value in the sub-micromolar range at the rat A
1
AR (K
i
= 0.469
1
AR a
1
1
A
1
moiety at position N9 displayed a slight (3-fold) preference for the rat A AR compared to the
rat A2AAR, as well as low affinity at the human A2BAR. However, comparison of the
affinities of 19c at human A and A2AAR revealed a more than 35-fold preference for the
= 0.284 µM) over A (K > 10.0 µM). Replacement of the fluoro substituent by
an isopropyl moiety in the para-position of the phenyl ring resulted in a decrease in affinity at
the rat A AR, but an increase in A AR affinity (19a vs. 19e).
The unsubstituted N9-benzyl derivative 19f was a relatively potent dual human A
antagonist with only low affinity at the A2BAR. Introduction of a methoxy group in the ortho-
position of the benzyl ring resulted in a decrease of A and A2BAR affinity, but in an increase
in affinity for the A AR (19g vs. 19f). Shifting the methoxy group from the ortho- to the
1
A
2AARs (K
i
1
i
1
3
1
/A2AAR
1
3
1
7

meta- (19h) or the para-position (19i) resulted in a decrease in affinity for human A
1
and
A
3
ARs, and an increase in affinity at the rat A AR.
1
Bioisosteric replacement of the benzyl ring at position N9 with a thiophenylmethyl (19k) or a
pyridinylmethyl ring (19j) was investigated. For the latter substituent, a decrease in affinity at
all four ARs was observed. In contrast, bioisosteric replacement of the benzyl ring by a
thiophenylmethyl residue resulted in K
2AARs, combined with low affinity for the A2BAR as observed for the analogous benzyl
derivative 19f.
Introduction of a phenethyl moiety at position 9 (19l) resulted in a moderately potent dual
/A2AAR antagonist at the rat receptors; however, a more than 18-fold loss of affinity was
observed for the human A AR compared to the rat receptor (also see Figure S2). para-
i 1
values in the submicromolar range at both, A and
A
A
1
1
Substitution of the phenethyl ring with a hydroxyl or a methoxy residue (19m, 19n) led to
compounds with a preference for A2AARs (Figure S3, Supplementary data).
Compounds bearing saturated rings (cyclobutyl or cyclohexyl) at the N9-position showed
moderate A2AAR affinity with a slight selectivity over A ARs, and low A2BAR affinity (19q
1
and 19r). The introduction of a 1-benzylpiperidin-4-yl moiety at position 9 resulted in a
moderately potent A2AAR antagonist (19u). The bulky moiety was somewhat better tolerated
by the human than by the rat A2AAR.
In order to evaluate differences in affinity of the investigated compounds at human versus rat
A
1
and A2AARs the pK
Figure S2 and S3). At A
cyclobutyl moiety at position 9, respectively, showed a higher pK
than at the rat receptors (Figure S2). All other tested compounds of this series displayed a
preference for the rat over the human A AR. Especially compounds 19c, 19b, 19d, 19i, 19l
and 19m were considerably more potent at the rat than at the human A AR.
i
values of the 1-methyl-3-propargyl derivatives (19) were correlated
ARs only 19g and 19q bearing an ortho-methoxybenzyl and a
value at the human A AR
(
1
i
1
1
1
1
8

Species differences of compounds 19 at A2AARs were less pronounced. Most of the tested
compounds within this series showed higher affinity for the human A2AAR than at the rat
receptor but for some derivatives the opposite was true (Figure S3).
2.3.2. 1,3-Diethyl-substituted tetrahydropyrimido[1,2-f]purinediones
Replacement of the methyl substituents in position 1 and 3 of lead compound 9 by ethyl
residues resulted in a loss of A2A-selectivity (Table 1, 9 vs. 20a). In the 1,3-diethyl series (20),
N9-phenylethyl-substituted compounds with a meta- and/or para-methoxy substituent (20e-h)
displayed low to moderate A2AAR affinity and selectivity versus A
single methoxy substituent at the phenyl or phenethyl ring (20c-f) showed some affinity for
the A AR. Comparison of 1,3-diethyl-substituted derivatives 20c, 20f, 20h, 20i with the
corresponding 3-methyl-1-propargyl derivatives 19b, 19n, 19o, 19r clearly showed that the
1
ARs. Derivatives with a
3
1-methyl-3-propargyl substitution pattern was superior for interaction with the A2AAR.
2.3.3. 1,3-Dimethyl-substituted tetrahydropyrimido[1,2-f]purinediones
Compounds of the 1,3-dimethyl series bearing a chloro-substituted phenyl, benzyl or
phenethyl moiety at position 9 (21a-k) were investigated. Most of the derivatives showed a
trend towards A2A-selectivity versus A
However affinity for the human A2A receptor was only moderate, while it was higher at the rat
2AAR for some derivatives, the most potent one being the m-chlorophenethyl derivative 21i
with a K value of 0.166 µM at the rat A2AAR combined with selectivity versus the other AR
subtypes.
1 3
, and only negligible inhibition of A2B and A ARs.
A
i
2.3.4. 9-(Piperazinyl-oxoethyl/oxopropyl)-substituted tetrahydropyrimido[1,2-f]purinediones
1
9

1,3-Dimethyl-substituted tetrahydropyrimido[1,2-f]purinediones bearing 4-benzyl- or 4-
phenyl-substituted 2-(piperazin-1-yl)-2-oxoethyl moieties at position N9 (21l-q) generally
lacked or displayed only low AR affinity at all AR subtypes.
The same was true for the 1,3-dipropyl-substituted tetrahydropyrimido[1,2-f]purinediones 22,
with the exception of the A
3
AR to which their affinity was increased. The best A
3
AR
antagonists of this series were 22b (K
i
human A : 0.762 µM) and 22d (K human A : 0.541
3
i
3
µM), which were selective versus the other AR subtypes. Increase in the spacer lengths
between N9 and the piperazine ring resulted in inactive compounds (compare e.g. 22a and
2
2f).
2.4. Structure-activity relationships as MAO inhibitors
All compounds were investigated for their potency to inhibit human MAO-B; selected
compounds were further tested at human MAO-A (Table 2). All investigated
tetrahydropyrimido[1,2-f]purinedione derivatives showed virtually no inhibition of human
MAO-A at a test concentration of 10 µM. At MAO-B, compounds of the 1-methyl-3-
propargyltetrahydropyrimido[1,2-f]purinedione series 19 bearing a para-isopropylphenyl,
benzyl or thiophenylmethyl moiety displayed inhibition of 50% or more at an initial test
concentration of 10 µM, whereas most other derivatives were inactive. 1,3-Diethyl-substituted
tetrahydropyrimido[1,2-f]purinediones 20 were generally more potent MAO-B inhibitors than
the 1-methyl-3-propargyl-substituted derivatives 19. In agreement with structural data and
molecular
modeling
studies
(see
Introduction)
1,3-dimethyl-substituted
tetrahydropyrimido[1,2-f]purinediones bearing a mono- or di-chloro-substituted phenyl,
benzyl or phenethyl ring at position N9 (21a-k) were identified as potent MAO-B inhibitors.
The substitution pattern and the number of chlorine atoms was found to be crucial in the
benzyl (21e-21h) and phenethyl (21i-k) series showing the following rank order of potency:
3,4-dichloro > 3,5-dichloro > 4-chloro > 3-chloro. As a result, 3,4-dichloro-substitution at the
2
0

benzyl or the phenethyl residue led to the most potent MAO-B inhibitors of the present series
with IC50 values in the nanomolar range. The most potent compounds were 21g (IC50 human
MAO-B = 62.9 nM) and 21k (IC50 human MAO-B = 86.9 nM). Both compounds were found
to be selective versus MAO-A and versus ARs. They represent a new class of potent and
selective MAO-B inhibitors and may serve as lead structures for the development of potent
MAO-B inhibitors as therapeutic agents.
Table 2. MAO-A and MAO-B inhibitory potencies of tetrahydropyrimido[1,2-f]purinediones
19-22 and standard compounds.
IC50 ± SEM (µM)
IC50 ± SEM (µM)
a
a
a
a
Compd
MAO-A
MAO-B
Compd
DMPX (7)
Safinamide
MAO-A
MAO-B
>10.0
b
b
CSC (2)
>10.0 (23%)
0.0178 ± 0.0034
>10.0 (3%)
nd
c
Selegiline
nd
0.00613 ± 0.00085
0.00767 ± 0.00181
1-Methyl-3-propargyltetrahydropyrimido[1,2-f]purinediones (19)
b
b
1
9a
nd
>10.0 (17%)
ca. 10.0 (46%)
ca. 10.0 (64%)
1.80 ± 0.16
19b
19d
19f
19h
19j
19l
nd
>10.0 (36%)
b
b
b
1
1
9c
nd
nd
>10.0 (30%)
b
b
b
b
9e
>10.0 (5%)
>10.0 (4%)
>10.0 (5%)
3.76 ± 0.33
b
1
9g
9i
9k
9m
>10.0 (5%)
ca. 10.0 (51%)
b
b
b
1
>10.0 (11%)
ca. 10.0 (50%)
ca. 10.0 (51%)
nd
nd
nd
nd
nd
nd
>10.0 (21%)
b
b
b
1
>10.0 (0%)
>10.0 (2%)
b
b
1
nd
nd
nd
nd
nd
>10.0 (37%)
19n
19p
19r
19t
>10.0 (25%)
b
b
19o
>10.0 (29%)
>10.0 (17%)
b
b
19q
>10.0 (19%)
>10.0 (26%)
b
b
1
9s
>10.0 (30%)
>10.0 (33%)
b
19u
>10.0 (22%)
1
1
,3-Diethyltetrahydropyrimido[1,2-f]purinediones (20)
b
b
b
2
0a
nd
>10.0 (4%)
nd
ca. 10.0 (48%)
6.39 ± 0.71
20b
20d
20f
>10.0 (11%)
>10.0 (28%)
ca. 10.0 (56%)
b
b
2
0c
0e
3.26 ± 0.60
4.76 ± 1.48
b
b
2
>10.0 (33%)
>10.0 (2%)
b
b
2
0g
>10.0 (4%)
5.87 ± 1.90
3.72 ± 0.70
20h
20j
nd
nd
>10.0 (25%)
b
b
20i
>10.0 (-3%)
>10.0 (34%)
,3-Dimethyltetrahydropyrimido[1,2-f]purinediones (21)
2
1

b
b
b
b
2
1a
>10.0 (15%)
>10.0 (12%)
4.96 ± 0.98
ca. 10.0 (58%)
0.344 ± 0.021
21b
21d
21f
21h
21j
21l
>10.0 (6%)
nd
ca. 10.0 (50%)
b
b
2
2
1c
>10.0 (40%)
b
b
1e
>10.0 (2%)
>10.0 (14%)
0.218 ± 0.024
0.193 ± 0.008
0.624 ± 0.057
ca. 10.0 (60%)
ca. 10.0 (55%)
3.48 ± 0.44
b
b
2
1g
1i
1k
1m
>10.0 (10%)
0.0629 ± 0.0070
1.35 ± 0.16
>10.0 (1%)
b
b
b
b
2
>10.0 (-4%)
>10.0 (-2%)
>10.0 (4%)
b
b
b
2
0.0869 ± 0.0042
2.20 ± 0.35
>10.0 (1%)
b
2
>10.0 (13%)
21n
21p
>10.0 (4%)
b
b
21o
nd
nd
>10.0 (30%)
>10.0 (6%)
b
21q
>10.0 (30%)
1
,3-Dipropyltetrahydropyrimido[1,2-f]purinediones (22)
b
b
b
2
2a
nd
nd
ca. 10.0 (46%)
22b
22d
22f
>10.0 (-2%)
ca. 10.0 (52%)
2.98 ± 0.98
b
b
2
2c
2e
>10.0 (15%)
>10.0 (5%)
nd
b
b
2
nd
>10.0 (30%)
>10.0 (30%)
b
22g
>10.0 (1%)
3.60 ± 0.68
a
b
c
n = 3; enzyme inhibition at indicated concentration; nd not determined.
Tetrahydropyrimido[1,2-f]purinediones 21l-q and 22, compounds bearing a chloro- or
trifluoromethyl-substituted (benzyl/phenyl-piperazinyl)oxoethyl moiety at position 9 showed
low MAO-B inhibition. This finding confirms the assumption that aromatic chloro-substituted
moieties at position 9 are favourable for potent MAO-B inhibition.
3
. Conclusions
In search for novel drugs suitable for the treatment of neurodegenerative diseases we
synthesized and investigated series of 55 mostly novel
tetrahydropyrimido[2,1-f]purinediones. A synthetic scheme was developed to gain access to
-methyl-3-propargyl-substituted tetrahydropyrimido[2,1-f]purinedione derivatives, which
a
1
had previously not been reported. The compounds were investigated at targets that are thought
to be relevant for neuroprotective activity. We were successful in discovering 1,3-dimethyl-
substituted tetrahydropyrimido[2,1-f]purinediones as a new class of potent and selective
MAO-B inhibitors. Furthermore, several 1-methyl-3-propargyl-substituted derivatives
exhibited dual A
1
/A2A (e.g. 19f) or triple (A
1
/A2A/MAO-B) activity, which is expected to
result in additive or even synergistic activity. While A2A antagonism and MAO-B inhibition
2
2

may confer anti-Parkinsonian effects combined with neuroprotective properties, A
antagonism may ameliorate cognitive deficits. The identified dually active A /A2A adenosine
receptor antagonists identified in the present study, such as 9-benzyl-1-methyl-3-propargyl-
1
1
6,7,8,9-tetrahydropyrimido[1,2-f]purine-2,4(1H,3H)dione (19f, K
i 1
, human receptors, A :
0.249 µM, A2A: 0.253 µM), were somewhat less potent than recently described dual
antagonists. However, they represent a new scaffold and provide a starting point for further
improvement. The most potent, selective MAO-B inhibitor of the present series, 9-(3,4-
dichlorobenzyl)-1,3-dimethyl-6,7,8,9-tetrahydropyrimido[1,2-f]purine-2,4(1H,3H)-dione (21g,
IC50 human MAO-B: 0.0629 µM), was only 8-fold less potent than the MAO-B inhibitor
safinamide which is currently evaluated in advanced clinical trials.
2
3

4
4
4
. Experimental section
.1. Chemistry
.1.1 Material and Methods
All commercially available reagents and solvents were used without further purification. The
reactions were monitored by thin layer chromatography (TLC) using aluminum sheets coated
with silica gel 60 F254 (Merck). Melting points were determined on a Büchi 530 melting point
apparatus and are uncorrected. Preparative HPLC was performed on a C18 column (250 × 20
mm, particle size 10 μm, Eurospher 100) using a mixture of methanol and H
2
O as eluent at a
flow rate of 20 mL/min. Microwave reactions were performed in a CEM Discover microwave
1
13
reactor. H NMR and C NMR data were recorded on a Bruker Avance spectrometer at 500
MHz for proton and 125 MHz for carbon. Shifts are given in ppm relative to the remaining
protons of the deuterated solvents. Mass spectra were recorded on an API 2000 mass
spectrometer (electron spray ion source, Applied Biosystems, Darmstadt, Germany) coupled
with an Agilent 1100 HPLC system using a Phenomenex Luna HPLC C18 column (50 × 2.00
mm, particle size 3 µm). The purity of the tested compounds was determined by HPLC-UV
obtained on an LC-MS instrument (Applied Biosystems API 2000 LC-MS/MS, HPLC
Agilent 1100) using the procedure as follows: dissolving of the compounds at a concentration
of 1.0 mg/mL in methanol and if necessary sonication to complete dissolution. Then, 10 µL of
the solution was injected into a Phenomenex Luna C18 HPLC column (50 × 2.00 mm,
particle size 3 µm) and elution was performed for 30 min at a flow rate of 250 µL/min with a
gradient of water:methanol either containing 2 mM ammonium acetate from 90:10 up to
0
:100, starting the gradient after 10 min (system A), or containing 2 mM ammonium acetate
and 0.1 % formic acid from 90:10 up to 0:100, starting the gradient after 10 min (system B),
or containing 2 mM ammonium acetate from 60:40 up to 0:100 for 30 min, starting the
2
4

gradient after 0 min and ending after 20 min (system C). UV absorption was detected from
220 to 400 nm using a diode array detector.
4
.1.1. 8-Bromo-7-(3-chloropropyl)-3-methyl-1H-purine-2,6(3H,7H)-dione (11).
5
3
To a stirred suspension of 8-bromo-3-methyl-1H-purine-2,6(3H,7H)-dione (10) (2.35 g, 9.6
mmol) and triethylamine (2.95 mL, 21.1 mmol) in DMF (72 mL) 1-bromo-3-chloropropane
(
0.95 mL, 9.6 mmol) was added. After stirring overnight at rt, the reaction mixture was
concentrated under reduced pressure and the residue was purified by column chromatography
(
silica gel, ethyl acetate) to provide compound 11 as a colorless solid. Yield: 50 %; mp:
1
3
3
2
25 °C; H-NMR (CDCl
6.3 Hz, 2H, N7-CH -CH
) δ 153.4 (C6), 150.5 (C2), 150.4 (C4), 128.3 (C8), 109.1 (C5), 45.4 (N7-
), 41.1 (N7-CH ), 32.8 (N7-CH -CH ), 29.2 (N3-CH ). ESI-MS: positive mode
3
) δ 8.46 (s, 1H, N1-H), 4.45 (t, J = 7.1 Hz, 2H, N7-CH
2
), 3.57 (t, J
=
2
2
-CH ), 3.51 (s, 3H, N3-CH ), 2.33-2.29 (m, 2H, N7-CH
2
3
2
-CH ).
2
1
3
C-NMR (CDCl
CH -CH -CH
23.3 [M+H] .
.1.2. 8-Bromo-7-(3-chloropropyl)-3-methyl-1-(prop-2-ynyl)-1H-purine-2,6(3H,7H)-
3
2
2
2
2
2
2
3
+
3
4
dione (12a).
To a stirred suspension of 8-bromo-7-(3-chloropropyl)-3-methyl-1H-purine-2,6(3H,7H)-dione
(11) (1.17 g, 3.6 mmol) and potassium carbonate (754 mg, 5.5 mmol) in DMF (18 mL)
propargyl bromide (0.60 mL, 5.4 mmol) was added. After stirring overnight at rt, the reaction
mixture was concentrated under reduced pressure and the residue was purified by column
chromatography (silica gel, ethyl acetate) to provide the desired compound 12a as a colorless
1
4
3 2
solid. Yield: 99 %; mp: 99 °C; H-NMR (CDCl ) δ 4.74 (d, J = 2.5 Hz, 2H, N1-CH ), 4.45 (t,
3
3
J = 7.1 Hz, 2H, N7-CH
2
), 3.58 (t, J = 6.3 Hz, 2H, N7-CH
2
-CH
2
-CH
2
3
), 3.54 (s, 3H, N3-CH ),
4
13
2
.32-2.28 (m, 2H, N7-CH
2
-CH
2
), 2.15 (d, J = 2.5 Hz, 1H, N1-CH
2
-C≡CH). C-NMR
(
CDCl
3
) δ 153.0 (C6), 150.4 (C2), 148.8 (C4), 128.1 (C8), 108.7 (C5), 78.3 (N1-CH
0.7 (N1-CH -C≡CH), 45.3 (N7-CH -CH -CH ), 41.2 (N7-CH ), 32.8 (N7-CH -CH
), 29.9 (N3-CH ). ESI-MS: positive mode 361.3 [M+H] .
2
-C≡CH),
7
2
2
2
2
2
2
3
), 30.6
+
(
N1-CH
2
3
2
5

4.1.3 General procedure for the preparation of ω-amino-1-(4-phenyl/benzyl-1-
piperazinyl)ethanone and -propanone hydrochlorides (17a-f, 18a,d)
N-boc-glycine (14a) or N-boc-β-alanine (14b) (10.0 mmol) was dissolved in methanol (10
mL), and EDC (1.92 g, 10.0 mmol) followed by N-benzyl- or -phenylpiperazine 13a-f (11.0
mmol) dissolved in methanol (10 mL) was added. The solution was stirred for 1 h at rt. Then,
the solvent was removed by rotary evaporation and water was added to the residue to
crystallize
the
products.
The
products,
N-boc-ω-amino-1-(4-phenyl/benzyl-1-
piperazinyl)ethanones or -propanones (15a-f, 16a,d), were filtered off and dried. To remove
the boc-protective group, compounds 15a-f, 16a,d were dissolved in 4N HCl in dioxane (8
mL) and stirred for 1 h at rt. Then, diethyl ether (50 mL) was added. The product was filtered
off, washed with diethyl ether (3 × 20 mL) and dried.
4
.1.3.1. 2-Amino-1-(4-phenylpiperazin-1-yl)ethanone hydrochloride (17a)
1
+
Yield: 85 % over 2 steps; mp: 253 °C; H-NMR (DMSO-d
6
3
) δ 8.37 (br s, 3H, NH ), 7.42-
3
3
3
7
.38 (m, 4H, phe), 7.15 (dd, J = 6.5 Hz, J = 6.6 Hz, 1H, C4-H, phe), 3.93 (q, J = 5.6 Hz, 2H,
CH
piperazine), 3.37-3.35 (m, 2H, piperazine). C-NMR (DMSO-d
phe), 129.5 (C3/C5, phe), 124.2 (C4, phe), 118.6 (C2/C6, phe), 50.9 and 50.7 (C3 and C5,
2
-C=O), 3.85 (br s, 2H, piperazine), 3.77-3.75 (m, 2H, piperazine), 3.45-3.42 (m, 2H,
1
3
6
) δ 164.8 (C=O), 146.7 (C1,
2
piperazine), 42.8 (C2/C6, piperazine), 40.1 (CH -C=O). ESI-MS: positive mode 220.1
+
[
M+H] .
4.1.3.2. 2-Amino-1-(4-(4-chlorophenyl)piperazin-1-yl)ethanone hydrochloride (17b)
1
+
Yield: 88 % over 2 steps; mp: 242 °C; H-NMR (DMSO-d
6
3
) δ 8.30 (br s, 3H, NH ), 7.32-
3
7
.29 (m, 2H, C3-/C5-H, phe), 7.11-7.09 (m, 2H, C2-/C6-H, phe), 3.90 (q, J = 5.6 Hz, 2H,
CH
piperazine), 3.21-3.19 (m, 2H, piperazine). C-NMR (DMSO-d
phe), 128.8 (C3/C5, phe), 124.3 (C4, phe), 118.1 (C2/C6, phe), 48.8 and 48.5 (C3 and C5,
2
-C=O), 3.69-3.67 (m, 2H, piperazine), 3.59-3.57 (m, 2H, piperazine), 3.27-3.25 (m, 2H,
1
3
6
) δ 164.6 (C=O), 148.3 (C1,
2
6

2
piperazine), 43.4 (C2/C6, piperazine), 40.8 (CH -C=O). ESI-MS: positive mode 254.1
+
[
M+H] .
4.1.3.3. 2-Amino-1-(4-benzylpiperazin-1-yl)ethanone dihydrochloride (17c)
1
+
Yield: 74 % over 2 steps; mp: 296 °C; H-NMR (DMSO-d
6
) δ 12.18 (s, 1H, NH ), 8.34 (br s,
+
3
4
1
H, NH
3
), 7.66-7.65 (m, 2H, C3-/C5-H, phe), 7.45-7.44 (m, 3H, C2-/C4-/C6-H, phe), 4.40-
.34 (m, 3H), 3.94-3.86 (m, 3H), 3.69-3.65 (m, 1H), 3.39 (br s, 3H), 3.10 (br s, 1H), 2.91 (br s,
1
3
6
H). C-NMR (DMSO-d ) δ 164.9 (C=O), 131.4 (C3/C5, phe) 129.4 (C4, phe), 129.4 (C1,
phe), 128.7 (C2/C6, phe), 58.4 (CH
piperazine), 38.0 (CH
2
), 50.1 and 49.6 (C3 and C5, piperazine), 40.8 (C2/C6,
+
2
-C=O). ESI-MS: positive mode 234.3 [M+H] .
4.1.3.4. 2-Amino-1-(4-(4-chlorobenzyl)piperazin-1-yl)ethanone dihydrochloride (17d)
1
+
Yield: 78 % over 2 steps; mp: 290 °C; H-NMR (DMSO-d
6
) δ 12.20 (s, 1H, NH ), 8.33 (br s,
+
3
3
3
H, NH
phe), 4.34 (s, 2H, CH
s, 2H), 2.91 (br s, 2H). C-NMR (DMSO-d
phe), 133.6 (C3/C5, phe), 128.9 (C2/C6, phe), 57.6 (CH2), 50.2 and 49.8 (C3 and C5,
3
), 7.70 (d, J = 8.5 Hz, 2H, C3-/C5-H, phe), 7.51 (d, J = 8.5 Hz, 2H, C2-/C6-H,
2
-C=O), 3.62 (s, 2H, phe-CH
2
), 3.38 (br, s, 2H), 3.28 (br, s, 2H), 3.09 (br,
1
3
6
) δ 165.2 (C=O), 134.7 (C1, phe), 133.7 (C4,
2
piperazine), 42.0 (C2/C6, piperazine), 38.2 (CH -C=O). ESI-MS: positive mode 268.1
+
[
M+H] .
4
.1.3.5.
chloride (17e)
Yield: 69 % over 2 steps; mp: 298 °C; H-NMR (DMSO-d
2-Amino-1-(4-(4-(trifluoromethyl)benzyl)piperazin-1-yl)ethanone
dihydro-
1
+
6
) δ 12.42 (br s, 1H, NH ), 8.30 (br
+
3
3
s, 3H, NH
3
), 7.92 (d, J = 8.0 Hz, 2H, C3-/C5-H, phe), 7.43 (d, J = 8.2 Hz, 2H, C2-/C6-H,
phe), 4.46-4.40 (m, 3H), 3.94–3.88 (m, 3H), 3.69 (br s, 1H), 3.41 (br s, 3H), 3.14 (br s, 1H),
1
3
2
1
2
6
.97 (br s, 1H). C-NMR (DMSO-d ) δ 165.1 (C=O), 134.3 (C1, phe), 132.6 (C2/C6, phe),
2
3
1
29.9 (q, JC,F = 31.7 Hz, C4, phe), 125.7 (q, JC,F = 3.4 Hz, C3/C5, phe), 124.1 (q, JC,F
=
70.6 Hz, CF
3
), 57.7 (phe-CH
2
), 50.5 and 50.0 (C3 and C5, piperazine), 41.0 (C2/C6,
+
piperazine), 38.2 (CH
2
-C=O). ESI-MS: positive mode 302.0 [M+H] .
2
7

4.1.3.6. 2-Amino-1-(4-(4-methoxybenzyl)piperazin-1-yl)ethanone dihydrochloride (17f)
1
+
6
Yield: 76 % over 2 steps; mp: 154 °C; H-NMR (DMSO-d ) δ 11.89 (s, 1H, NH ), 8.28 (br s,
+
3
H, NH
3
), 7.57-7.55 (m, 2H, C2-/C6-H, phe), 7.00-6.98 (m, 2H, C3-/C5-H, phe), 4.41-4.38
), 3.66-3.61 (m, 1H), 3.29
) δ 164.9 (C=O), 160.0 (C4,
), 55.2 (OCH ), 49.8
-C=O). ESI-MS:
(
(
m, 1H), 4.30-4.26 (m, 2H), 3.94-3.86 (m, 3H), 3.78 (s, 3H, OCH
3
1
3
br s, 3H), 3.03 (br, s, 1H), 2.87 (br s, 1H). C-NMR (DMSO-d
6
phe), 133.0 (C2/C6, phe), 121.0 (C1, phe), 114.1 (C3/C5, phe), 58.0 (CH
2
3
and 49.4 (C3 and C5, piperazine), 40.9 (C2/C6, piperazine), 38.1 (CH
2
+
positive mode 264.1 [M+H] .
4.1.3.7. 3-Amino-1-(4-phenylpiperazin-1-yl)propan-1-one hydrochloride (18a)
1
+
Yield: 65 % over 2 steps; mp: 271 °C; H-NMR (DMSO-d
6
3
) δ 7.96 (br s, 3H, NH ), 7.33-
7.30 (m, 2H, C3-/C5-H, phe), 7.18-7.17 (m, 2H, C2-/C6-H, phe), 6.98-6.95 (m, 1H, C4-H,
phe), 3.72 (br s, 2H, piperazine), 3.68-3.66 (m, 2H, piperazine), 3.30-3.28 (m, 2H, piperazine),
3
3
.24-3.22 (m, 2H, piperazine), 3.09-3.01 (m, 2H, CH
2
-CH
2
-C=O), 2.77 (t, J = 6.5 Hz, 2H,
13
2 6
CH -C=O). C-NMR (DMSO-d ) δ 168.2 (C=O), 129.2 (C3/C5, phe), 129.0 (C1, phe), 117.0
(
C4, phe), 116.0 (C2/C6, phe), 49.7 and 49.4 (C3 and C5, piperazine), 43.9 (C2/C6,
+
piperazine), 35.1 (CH
.1.3.8. 3-Amino-1-(4-(4-chlorobenzyl)piperazin-1-yl)propan-1-one dihydrochloride
18d)
Yield: 68 % over 2 steps; mp: 151 °C; H-NMR (DMSO-d
2 2 2
-CH -C=O), 29.7 (CH -C=O). ESI-MS: positive mode 234.4 [M+H] .
4
(
1
+
6
) δ 12.21 (s, 1H, NH ), 8.00 (br s,
+
3
3
4
3
3
H, NH ), 7.68 (d, J = 8.5 Hz, 2H, C3-/C5-H, phe), 7.52 (dd, J = 8.5 Hz, J = 1.9 Hz, 2H,
2
C2-/C6-H, phe), 3.62 (s, 2H, CH ), 3.38, 3.28, 3.09 and 2.91 (4x br s, 8H, piperazine). 3.01 (t,
3
3
13
2 2 2 6
J = 6.0 Hz, 2H, CH -CH -C=O), 2.76 (t, J = 6.1 Hz, 2H, CH -C=O). C-NMR (DMSO-d )
δ 168.5 (C=O), 134.3 (C1, phe), 133.3 (C3/C5, phe), 128.7 (C2/C6, phe), 128.3 (C4, phe),
5
7.5 (CH
2
2 2
), 50.1 and 49.8 (C3 and C5, piperazine), 41.4 (C2/C6, piperazine), 34.9 (CH -CH -
+
C=O), 29.6 (CH
2
-C=O). ESI-MS: positive mode 282.4 [M+H] .
2
8

4
.1.4. General procedure for the synthesis of tetrahydropyrimido[1,2-f]purinediones 19,
0 and 21c,d,g-i,k.
2
Compound 12a, 12b or 12c (0.55 mmol), the appropriate amine (1.1 mmol) and 1,2-
dimethoxyethane (1 mL) were heated in a microwave reactor (150 Watt, Power Max On,
9
0 °C, 10 bar) for 1 h. The solvent was removed under reduced pressure and the residue was
purified by column chromatography (silica gel, ethyl acetate).
.1.4.1. 9-(4-Fluorophenyl)-1-methyl-3-(prop-2-ynyl)-6,7,8,9-tetrahydropyrimido[1,2-f]-
4
purine-2,4(1H,3H)-dione (19a)
1
Yield: 70 %; mp: 250 °C; H-NMR (CDCl
3
) δ 7.42-7.39 (m, 2H, C2-/C6-H, phe), 7.10-7.07
4
3
(
m, 2H, C3-/C5-H, phe), 4.77 (d, J = 2.4 Hz, 2H, N3-CH
2
), 4.37 (t, J = 6.1 Hz, 2H, C6-H
2
),
4
3
.83-3.81 (m, 2H, C8-H
2
), 3.45 (s, 3H, N1-CH
3
), 2.34-2.29 (m, 2H, C7-H
2
), 2.16 (t, J = 2.4
1
3
1
Hz, 1H, N3-CH -C≡CH). C-NMR (CDCl
2
3
) δ 159.8 (d, JC,F = 243.8 Hz, C4, phe), 152.8
4
3
(
C4), 151.0 (C2), 149.8 (C9a), 148.6 (C10a), 138.8 (d, JC,F = 2.8 Hz, C1, phe), 125.1 (d, JC,F
2
=
2
8.3 Hz, C3/C5, phe), 115.7 (d, JC,F = 22.5 Hz, C2/C6, phe), 103.0 (C4a), 78.6 (N3-CH -
C≡CH), 70.1 (N3-CH
2
2 3
-C≡CH), 47.7 (C8), 42.1 (C6), 30.1 (N3-CH ), 29.8 (N1-CH ), 21.7
+
(
C7). ESI-MS: positive mode 354.3 [M+H] . HPLC: 98.7 % (A), 99.0 % (B).
4
.1.4.2. 9-(3-Methoxyphenyl)-1-methyl-3-(prop-2-ynyl)-6,7,8,9-tetrahydropyrimido[1,2-
f]purine-2,4(1H,3H)-dione (19b)
1
3
3
Yield: 75 %; mp: 210 °C; H-NMR (CDCl
3
) δ 7.21 (dd, J = 8.2 Hz, J = 8.2 Hz, 1H, C5-H,
4
4
phe), 6.69-6.66 (m, 2H, C4-/C6-H), 6.61 (dd, J = 2.2 Hz, J = 2.2 Hz, 1H, C2-H, phe), 4.77
4
3
(
d, J = 2.4 Hz, 2H, N3-CH
.77 (s, 3H, OCH ), 3.48 (s, 3H, N1-CH
N3-CH
.1.4.3. 9-(3-Bromophenyl)-1-methyl-3-(prop-2-ynyl)-6,7,8,9-tetrahydropyrimido[1,2-f]-
purine-2,4(1H,3H)-dione (19c)
2
), 4.36 (t, J = 6.1 Hz, 2H, C6-H
2
), 3.86-3.83 (m, 2H, C8-H
2
),
4
3
3
3
), 2.33-2.28 (m, 2H, C7-H
2
), 2.16 (t, J = 2.4 Hz, 1H,
+
2
-C≡CH). ESI-MS: positive mode 365.6 [M+H] . HPLC: 97.7 % (A), 96.9 % (B).
4
2
9