A novel class of adenosine A3 receptor ligands. 2. Structure affinity profile of a series of isoquinoline and quinazoline compounds

Article abstract of DOI:10.1021/jm980037i

1-Substituted 3-(2-pyridinyl)isoquinolines have been shown to form a novel class of adenosine A₃ receptor ligands. In the present study further investigations of this new lead and the structure affinity relationships of this class of compounds

Full text of DOI:10.1021/jm980037i

3994
J . Med. Chem. 1998, 41, 3994-4000
A Novel Cla ss of Ad en osin e A3 Recep tor Liga n d s. 2. Str u ctu r e Affin ity P r ofile
of a Ser ies of Isoqu in olin e a n d Qu in a zolin e Com p ou n d s
J acqueline E. van Muijlwijk-Koezen,* Henk Timmerman, Regina Link,† Henk van der Goot, and
Adriaan P. IJ zerman†
Division of Medicinal Chemistry, Leiden/ Amsterdam Center for Drug Research, Department of Pharmacochemistry,
Vrije Universiteit, De Boelelaan 1083, 1081 HV Amsterdam, The Netherlands, and Division of Medicinal Chemistry,
Leiden/ Amsterdam Center for Drug Research, Gorlaeus Laboratories, Rijks Universiteit Leiden, P.O. Box 9502,
2300 RA Leiden, The Netherlands
Received J anuary 21, 1998
1-Substituted 3-(2-pyridinyl)isoquinolines have been shown to form a novel class of adenosine
A3 receptor ligands. In the present study further investigations of this new lead and the
structure affinity relationships of this class of compounds are described. First, the influence
of an amide group at position 1 of the isoquinoline ring on the adenosine A3 receptor affinity
was determined. A carboxamide proved to be a useful spacer between the isoquinoline and a
phenyl ring. N-[2-(2-pyridinyl)isoquinolin-4-yl]benzamide (VUF8507, compound 6) had an
affinity of 200 nM at the adenosine A3 receptor. Second, we investigated the effects of
substitution of the benzamide ring of 6 with a series of mono- and disubstituted N-[3-(2-
pyridinyl)isoquinoline]benzamides. The ratio of the tautomers of the benzamides was
determined in the solid state and in solution by spectroscopic techniques (IR and NMR).
Affinities were determined in radioligand binding assays at rat brain A1 and A2A receptors
and at cloned human A3 receptor. The benzamides showed higher adenosine A3 receptor affinity
than aliphatic amides. We propose that the adenosine A3 receptor affinity of the different
benzamides is related to their presence in either the iminol or amide form. Ligands present
in the iminol form showed relatively high adenosine A3 receptor affinity. Finally, we explored
the influence of replacement of C4 of the isoquinoline ring by a nitrogen atom. Comparison of
isoquinolines with the corresponding quinazolines revealed that both compounds showed similar
adenosine A3 receptor affinity. These investigations led to potent and selective human
adenosine A3 receptor ligands with affinities in the nanomolar range. The subtype-selective
compound 4-methoxy-N-[2-(2-pyridinyl)quinazolin-4-yl]benzamide (VUF8504, 13) with an
affinity of 17.0 nM at the human adenosine A3 receptor might become a useful tool in the
pharmacological characterization or the investigation of the physiological function of this
receptor.
In tr od u ction
A3 antagonists and found 3-(2-pyridinyl)isoquinoline
derivatives as a novel class of A3 ligands.1
Extracellular adenosine modulates a wide range of
physiological functions by activation of G protein-
coupled adenosine receptors. Four subclasses of adeno-
sine receptors have been identified: A1, A2A, A2B, and
A3. The latter subtype has only been cloned recently2,3
and is the subject of intensive pharmacological char-
acterization.4-7
Activation of the adenosine A3 receptor causes hy-
potension and promotes the release of inflammatory
mediators from mast cells.8,9 Selective A3 antagonists
could act as antiasthmatic,10 cerebroprotective,11,12 and
antiinflammatory agents.7
For the A1 and A2A adenosine receptors, potent and
selective antagonists have been found by optimization
of the xanthine skeleton,13 but these classical adenosine
antagonists usually bind only weakly at adenosine A3
receptors.14 Recently, nonxanthine structures with high
affinity at the adenosine A3 receptor have been
reported.15-17 We have started a search for nonxanthine
The previous study from our group1 has suggested
that an additional phenyl group on 2-pyridinylisoquino-
line increased adenosine A3 receptor affinity, provided
that this aromatic ring is coupled by a spacer allowing
conjugation. We also have reported on the influence of
different spacers between the isoquinoline ring and this
second aromatic ring. In Figure 1 different spacers, e.g.,
diacylamine (1), methyl ketone (2), and amidine (3), are
shown.
In the present study we describe another spacer, the
amide moiety. The influence on adenosine receptor
affinity of substituents on the aromatic ring coupled to
this amide moiety was investigated using a series of
substituted benzamides. Moreover, we explored the
effects of the replacement of C4 in the isoquinoline ring
by a nitrogen atom, yielding quinazoline derivatives.
The structure-activity relationship of these derivatives
is discussed as well.
Ch em istr y
* Author for correspondence. Address: LACDR, Division of Medici-
nal Chemsitry, De Boelelaan 1083, 1081 HV Amsterdam. Tel: (+31)
20 4447604. Fax: (+31) 20 4447610. E-mail: muylwyk@chem.vu.nl.
† Rijks Universiteit Leiden.
Intermediate 1-amino-3-(2-pyridinyl)isoquinoline was
prepared under strong alkaline conditions from 2-meth-
S0022-2623(98)00037-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/19/1998

Structure Affinity Profile of Isoquinolines
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 3995
ligands cannot be compared unambiguously with affini-
ties of other ligands reported in the literature.
Am id e Gr ou p a s Sp a cer . We used 1-substituted
3-(2-pyridinyl)isoquinolines, mentioned in our previous
report,1 as lead structure in an effort to find antagonists
for the human adenosine A3 receptor. We described the
optimization of this class and found that addition of a
phenyl ring via a spacer allowing conjugation to the 3-(2-
pyridinyl)isoquinoline skeleton largely increased the
affinity at the A3 receptor.
We investigated several spacers between the phenyl
ring and the isoquinoline moiety. Among these were
diacylamine (in 1), methyl ketone (in 2), and amidine
(in 3) moieties (Figure 1). The rather bulky compound
1, containing an extra phenylacyl group compared to
compounds 2 and 3, showed significant A3/A1 selectivity.
Compounds 2 and 3 showed equal affinities at the
human adenosine A3 receptor, but the A3/A1 selectivity
was at least 4-fold more favorable for the amidine.
Compounds 2 and 3 can be present in two tautomeric
forms. In the enol or iminol form, the hydroxyl group
may form a hydrogen bond with N(2) of the isoquinoline
according to IR data.1,24 In compound 1 this hydrogen
bond could also exist according to its resonance struc-
tures. We decided to explore the influence of an amide
bond as spacer on the adenosine A3 receptor affinity
with a monosubstituted amide. The importance of the
aromatic ring at position 1 of the isoquinoline ring was
checked by comparing the adenosine receptor affinities
of compounds 4 and 5 with 6.
Table 1 showed that replacement of the methyl group
of 4 by a phenyl substituent (6) resulted in a 55-fold
increase of adenosine A3 receptor affinity. The isopropyl
derivative 5 showed an affinity of approximately 10 mM,
not very different from the methyl derivative 4. From
these data we concluded that an aromatic ring has to
be used. The affinity for the human adenosine A3
receptor of 6 increased more than 3-fold compared to
compounds 2 and 3, and compound 6 showed high A3/
A1 selectivity, too. The difference in selectivity between
compounds 3, 6, and 2 was remarkable. The amidine
derivative 3 and the amide derivative 6 showed sub-
stantial A3 selectivity, whereas methyl ketone derivative
2 showed a small preference for the adenosine A1
receptor. Comparison of compounds 6 and 1 showed
about equal affinities as well as A3/A1 selectivities. To
investigate the influence of substituents at the phenyl
ring on the adenosine receptor affinities unambiguously,
we decided to use the monosubstituted amide bond as
spacer.
F igu r e 1. 1-Substituted 3-(2-pyridinyl)isoquinoline deriva-
tives with different spacers between the isoquinoline ring and
the second aromatic ring.
ylbenzenecarbonitrile and picolinonitrile according to a
method described by De Zwart et al.18 with some
modifications. At a temperature of -40 °C potassium
was added to condensed ammonia with ferric nitrate as
catalyst. Subsequently, the nitriles were added to this
freshly prepared KNH2 and stirred overnight at room
temperature to evaporate the ammonia. Hydrolysis
took place by addition of aqueous ammonium chloride
solution to obtain a NH3/NH4+ buffer in situ. After acid/
base extraction the compound was obtained by sublima-
tion.
3,4-Dimethoxy-N-[3-(2-pyridinyl)isoquinolin-1-yl]benz-
amide (19) was prepared according to a modified
method described by De Zwart et al.18 In this procedure
1-amino-3-(2-pyridinyl)isoquinoline was condensed with
the appropriate acyl chlorides, yielding the correspond-
ing amide. These reactions are depicted in Scheme 1.
The acylation took place in two steps: (i) abstraction of
an amine proton and (ii) nucleophilic attack of the acid
chloride. After proton abstraction, an intermediate was
formed in which the negative charge is delocalized. This
could be the reason that the yields of the benzamides
were only moderate (10-60%), although the isomeric
amide was not detected or isolated during the purifica-
tion steps.
Compound 28, 4-methoxy-N-[2-(2-pyridinyl)quinazo-
lin-4-yl]benzamide, was synthesized similarly to the
corresponding isoquinoline benzamide, viz. by acylation
of 4-amino-2-(2-pyridinyl)quinazoline. Preparation of
this precursor was carried out according to Linschoten
et al.19 The synthetic route yielding quinazolines 27 and
28 is depicted in Scheme 2.
Resu lts a n d Discu ssion
Ar om a tic Su bstitu tion of Ben za m id e. Next, the
influence of substituents at the benzamide ring was
investigated. The affinities of substituted benzamides
7-19 are shown in Table 2. A chloro substituent at the
meta position resulted in lower adenosine A3 receptor
affinity (compound 7), whereas a slightly electron-
donating group (methyl compound 8) showed equal
affinity in comparison to the unsubstituted benzamide
6. The electron-donating methoxy substituent on the
meta position (compound 9) raised the adenosine A3
receptor affinity slightly.
Bin d in g Stu d ies. All compounds were tested in
radioligand binding assays to determine their affinities
at adenosine A3, A1, and A2A receptors. The affinities
at adenosine A1 and A2A receptors were determined on
rat brain cortex and rat striatum with [3H]DPCPX or
[3H]CGS 21680 as the radioligands, respectively.20,21
The affinity at adenosine A3 receptors was determined
on membranes from HEK 293 cells, stably expressing
the human A3 receptor with [125I]AB-MECA as radio-
ligand.22,23 These radioligand binding data are sum-
marized in Tables 1, 2, and 4.
As mentioned in our previous report,1 we found Ki
values of some reference compounds different from
literature data. Therefore, the absolute Ki values of the
For the para substituted benzamides a trend similar
to that observed for the substituted N-[3-(2-pyridinyl)-
isoquinolin-1-yl]benzamidines1 was apparent. Para elec-

3996 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
van Muijlwijk-Koezen et al.
Sch em e 1
Sch em e 2
The methoxy substituent at the para position con-
tributed most to adenosine A3 receptor affinity. VUF8504
(13) showed a Ki value as low as 17 nM at the A3
receptor. This potent A3 antagonist was also a highly
selective ligand. Only very low affinity was found for
the A1 and A2A receptor with 14 and 0% displacement
of the radioligand at a concentration of 10-5 M, respec-
tively.
We also tested disubstituted benzamides to investi-
gate the influence of the position of various substituents
(Table 2). The 3,4-dichlorobenzamide 10 showed an
affinity for the adenosine A3 receptor equal to that of
benzamide 6. Larger differences were seen in the
dimethyl derivatives 14-18. The 3,4-dimethylbenz-
amide derivative 17 showed the highest adenosine A3
receptor affinity from this series of disubstituted com-
pounds. The adenosine A3 receptor affinities for the
2,5-, 2,4-, and 3,5-dimethylbenzamides (compounds 15,
16, and 18) were almost equal. The 2,6-dimethylbenz-
amide 14, however, showed a 10-fold lower affinity. An
explanation for this large difference in adenosine A3
receptor affinity could be steric hindrance of the ortho
substituents which could influence the tautomeric equi-
librium.
In both pairs of compounds 8-9 and 12-13, the
methoxy derivatives showed higher adenosine A3 recep-
tor affinities than did the methyl derivatives. Because
of the relatively high adenosine A3 receptor affinity of
17, higher than the unsubstituted or the 3-methyl
derivative and comparable to the 4-methyl derivative,
we decided to synthesize and measure adenosine A3
receptor binding of the 3,4-dimethoxy derivative 19.
However, the adenosine A3 receptor affinity of 19
decreased compared to 17.
tron-donating substituents present in 12 and 13 ren-
dered these analogues more potent than the unsubsti-
tuted phenyl analogue 6, and an electron-withdrawing
group as in 11 resulted in an equally potent compound.
In the case of benzamides, para substitution contributed
more convincingly to adenosine A3 affinity. An electron-
withdrawing chloro substituent, which in the series of
benzamidines resulted in lower A3 affinity,1 showed no
effect here (adenosine A3 receptor affinity of 200 nM for
both 11 and 6). The influence of an electron-donating
group, however, seemed to be larger since a 2-fold
increase in adenosine A3 receptor affinity was observed
for 4-methylbenzamide 12.
From compounds 8, 12, 17, and 18 we concluded that
a para electron-donating group is necessary for high
adenosine A3 receptor affinity and that methyl groups
on the meta position were allowed (17) but did not
contribute to positive ligand-receptor interaction (18).

Structure Affinity Profile of Isoquinolines
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 3997
Ta ble 1. Affinities at Adenosine Receptors of 1-Substituted 3-(2-Pyridinyl)isoquinoline Derivatives
a
b
d
e
compound
(1)
R
A3
A1
A2A
A1/A3
19
N(COPh)2
CH2COPh
NdC(NH2)Ph
NH(CO)CH3
NH(CO)CH(CH3)2
NH(CO)Ph
0.23 ( 0.091
0.66 ( 0.23
0.74 ( 0.15
13 ( 1.9
4.3 ( 2.8c
27
22
4
18
48
0
(2)
(3)
(4)
(5)
0.24 ( 0.1c
0.36
34
24
38
48% (10-5 M)
0.20 ( 0.04
VUF8507 (6)
3.2 ( 0.3c
16
a
Displacement of specific [125I]AB-MECA binding at human adenosine A3 receptors expressed in HEK 293 cells, expressed as Ki (
b
SEM in mM (n ) 3-5) or percentage displacement of specific binding at a concentration of 10 mM (n ) 3). Displacement of specific
[3H]DPCPX binding in rat cortical membranes, expressed as percentage displacement of specific binding at a concentration of 10 mM (n
) 2-3). Displacement of specific [3H]DPCPX binding in rat cortical membranes, expressed as Ki ( SEM in mM (n)3). Displacement
of specific [3H]CGS 21680 binding in rat striatal membranes, expressed as percentage displacement of specific binding at a concentration
of 10 mM (n ) 2-3). e A1/A3 ) KiAi/KiA3.
c
d
Ta ble 2. Affinities at the Adenosine Receptors of Aromatic N-[3-(2-Pyridinyl)isoquinolin-1-yl]amides
a
b
d
compound
R2
R3
R4
R5
R6
A3
A1
3.2 ( 0.3c
42
35
22
0
11
37
14
2
29
14
16
8
A2A
VUF8507 (6)
H
H
H
H
H
H
H
H
CH3
CH3
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
H
H
CH3
H
H
H
H
H
H
H
H
H
CH3
H
H
H
H
0.20 ( 0.042
0.77 ( 0.50
0.24 ( 0.084
0.15 ( 0.023
0.21 ( 0.049
0.20 ( 0.049
0.096 ( 0.026
0.017 ( 0.0017
2.6 ( 0.74
0.43 ( 0.10
0.36 ( 0.13
0.069 ( 0.021
0.32 ( 0.13
0.31 ( 0.12
0
0
0
0
0
0
0
0
0
7
0
0
0
(7)
(8)
(9)
(10)
(11)
Cl
CH3
OCH3
Cl
Cl
Cl
CH3
OCH3
H
H
H
H
H
H
H
(12)
VUF8504 (13)
(14)
(15)
(16)
(17)
(18)
(19)
H
CH3
CH3
H
CH3
CH3
OCH3
OCH3
H
41
20
a
Displacement of specific [125I]AB-MECA binding at human adenosine A3 receptors expressed in HEK 293 cells, expressed as Ki (
SEM in nM (n ) 3-5). Displacement of specific [3H]DPCPX binding in rat cortical membranes, expressed as percentage displacement
b
of specific binding at a concentration of 10 mM (n ) 2-3). c Displacement of specific [3H]DPCPX binding in rat cortical membranes,
expressed as Ki ( SEM in mM (n ) 3). Displacement of specific [3H]CGS 21680 binding in rat striatal membranes, expressed as percentage
d
displacement of specific binding at a concentration of 10 mM (n ) 2-3).
Ta u t om er ism a n d Con for m a t ion . In general,
benzamides can exist in the amide or in the iminol form.
It appeared that the unsubstituted compound 6 was
present in solution in the iminol form as detected by
NMR and IR spectroscopy. Because of the solubility
properties of these benzamides, chloroform instead of
water was used as solvent. In DMSO the NMR spectra
were comparable to those in chloroform. The iminol
group in compound 6 formed a hydrogen bond with the
nitrogen atom of the isoquinoline ring which was shown
by a signal at 16.3 ppm for the intramolecular hydrogen
bond in the NMR spectrum. However, in the solid state,
benzamide 6 existed in the amide form, according to the
absorption of the CdO vibration at 1660 cm-1 in the IR
spectrum of 6 in a KBr pellet. A solution IR spectrum
of 6 in chloroform, however, showed that the carbonyl
absorption had disappeared and an absorption at 3400
cm-1 according to the OH vibration had arisen. From
these measurements we concluded that, in chloroform
solution, compound 6 existed predominantly in the
highly conjugated iminol form (Table 3). Similar find-
ings have been reported by other authors.25-28 They
have investigated the equilibria of the two tautomers
of different amides of 1-aminoisoquinolines by photo-
chemistry and mass spectrometric studies.
The structures of the benzamides 6 and 12-18 were
established in the solid phase (KBr platelets) with IR
spectroscopy and in chloroform solution with IR and
NMR spectrometry (Table 3).18 Increased affinities of
the disubstituted benzamides paralleled their presence

3998 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
van Muijlwijk-Koezen et al.
Ta ble 3. Tautomeric Forms of N-[3-(2-Pyridinyl)isoquinolin-1-yl]benzamides
solution
solid phase
IR KBr platelets (cm-1
compound
1H NMR CDCl3 δ (ppm)
IR chloroform solution A (cm-1
)
)
affinity A3 Ki (nM)
VUF8507 (6)
(12)
VUF8504 (13)
16.30 (N-HO)18
16.33 (N-HO)18
16.26 (N-HO)18
10.92 (NH)18
16.20 (N-HO)
16.18 (N-HO)
16.24 (N-HO)18
16.22 (N-HO)18
15.15 (0.2 N-HO)
10.96 (0.8 NH)a
3400 (OH)
3400 (OH)
3400 (OH)
1660 (CdO)
3400 (OH)/1660 (CdO)
3400 (OH)/1660 (CdO)
3400 (OH)
3400 (OH)
3400 (OH)
3240 (NH)/1660 (CdO)18
3240 (NH)/1655 (CdO)18
3400 (OH)18
204 ( 42
96.3 ( 26
17.0 ( 1.7
2560 ( 740
426 ( 101
358 ( 129
68.8 ( 21
322 ( 133
306 ( 115
(14)
(15)
(16)
(17)
(18)
(19)
3170 (NH)/1640 (CdO)18
1660 (CdO)
1660 (CdO)
3400 (OH)18
3400 (OH)18
1650 (CdO)
a
Measured in DMSO solution because of its low solubility in chloroform.
in the iminol form. Compound 14, existing in the amide
form both in the solid state and in chloroform solution,
had very low affinity at the adenosine A3 receptor (2.56
mM). Compounds 15 and 16 were in the amide form
in the solid state and for the larger part in the iminol
form in solution. Indeed, CdO absorption in the IR
spectrum was markedly reduced in the chloroform
solution, whereas in the KBr pellets, high CdO absorp-
tion was observed. The two compounds had moderate
affinity at the adenosine A3 receptor (0.43 and 0.36 mM,
respectively). Compound 17 existed in the iminol form
in the solid as well as in the liquid phase and showed
high adenosine A3 receptor affinity (69 nM). From these
observations we suggest that the influence of the
positions of the substituents on affinity at the human
adenosine A3 receptor could be two-fold; first, direct
steric and electronic effects of the substituents govern
the conformations of the molecule, and second, the
substituents determine the tautomeric form in which
the ligand is present. For example, 2,6-dimethyl-N-[3-
(2-pyridinyl)isoquinolin-1-yl]benzamide (14) contained
two methyl groups on the ortho positions. This caused
steric hindrance with the amide moiety and forced the
phenyl ring out of the plane of the isoquinoline ring and
amide moiety, resulting in the decoupling of the amide-
aromatic system interaction. Because of this decrease
in conjugation, the amide form would be preferred over
the iminol form. The amide tautomer could fit the
adenosine A3 receptor pocket worse for yet unknown
reasons. Second, if the adenosine A3 receptor prefers a
nearly coplanar conformation of the isoquinolinylbenz-
amides, as in the iminol form, the substituents have
large influence. In that case, the ortho substituents of
14 prevented a good fit on the receptor.
Ta ble 4. Affinities at Adenosine Receptors of Quinazoline and
Isoquinoline Compounds
a
b
c
compound
X
R
A3
A1
A2A
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
N
N
N
NH(CO)CH3
NH(CO)CF3
N(COPh)2
11 ( 5.4
86 ( 5.2
0.47 ( 0.06
27 ( 10
25
8
3
21
16
47
17
0
3
35
10
42
17
2
CH OH
OH
CH NdCHNH2
N
22 ( 9.2
20 ( 6.6
N
N
N
NdCHNH2
NH(CO)Ph
NH(CO)(4-OCH3Ph) 0.15 ( 0.12 41
4 (10-5 M)d 38
0.23 ( 0.02
0
20
a
Displacement of specific [125I]AB-MECA binding at human
adenosine A3 receptors expressed in HEK 293 cells, expressed as
b
Ki ( SEM in mM (n ) 3-5). Displacement of specific [3H]DPCPX
binding in rat cortical membranes, expressed as percentage
displacement of specific binding at a concentration of 10 mM (n )
2-3). c Displacement of specific [3H]CGS 21680 binding in rat
striatal membranes, expressed as percentage displacement of
d
specific binding at a concentration of 10 mM (n ) 2-3). Dis-
placement of specific [125I]AB-MECA binding at human adenosine
A3 receptors expressed in HEK 293 cells, expressed as percentage
displacement of specific binding at a concentration of 10 mM (n )
3).
thines29 and the deazaadenines.30,31 In analogy, we
studied the adenosine receptor affinities of the closely
related quinazolines (Table 4).
Aliphatic amides 4 and 20 showed equal affinities at
the different adenosine receptors as did the quinolone
23 and quinazolone 24. The adenosine A3 receptor
affinities of the tertiary amides 1 and 22 differed by a
factor 2 in favor of the isoquinoline. Quinazoline 21
having a bulky, strong electron-withdrawing group
showed a decrease in adenosine A3 receptor affinity
compared to quinazoline 20. Because of the fact that
the relatively potent compound 22, having two phenyl
groups, was bulky too, the decrease in affinity could be
subscribed to the electronic effect.
Whereas the tertiary amides showed a small differ-
ence in adenosine A3 receptor affinities, the secondary
benzamides 6 and 27 showed equal affinities. The
largest difference between quinazoline and the “1-deaza
compound” isoquinoline was seen in the amidine series.
The quinazoline 26 showed no adenosine A3 receptor
affinity (only 4% displacement of the radioligand at a
concentration of 10 mM), whereas the corresponding
isoquinoline 25 showed a Ki value of 20 mM at the
The preference of the human adenosine A3 receptor
for a flat conformation of the ligand might also cause
the low adenosine A3 receptor affinity of the aliphatic
amides 4 and 5, besides their lower lipophilicity and
bulkiness. The iminol form is favored by conjugation
of an aromatic phenyl group with the isoquinoline
moiety.25-28 The adenosine A3 receptor affinity de-
creased without this conjugation, as explained above.
Also, in case of the substituted benzamidines, conjuga-
tion is highly conducive to adenosine A3 receptor affin-
ity.1 All in all the tautomeric preference observed in
KBr platelets may resemble the receptor-bound confor-
mation of the ligands most.
Aza a n d Dea za An a logu es. It has been shown that
the influence of each nitrogen atom on the affinity at
adenosine receptors can be established very well with
the aid of deaza compounds such as the deazaxan-

Structure Affinity Profile of Isoquinolines
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21 3999
Instruments 6030 Galaxy Series spectrophotometer. Melting
points were measured on a Electrothermal IA9200 apparatus
and are uncorrected.
adenosine A3 receptor. From these data it can be
concluded that in these cases the nitrogen atom at the
1-position of the quinazoline ring or the carbon atom at
the 4-position of the isoquinoline ring does not play an
important role in the ligand-adenosine A3 receptor
interaction.
In two cases, an interesting difference between iso-
quinoline and quinazoline in their adenosine A1 receptor
affinities was seen. Compound 1 showed an adenosine
A1 receptor affinity of 4.3 mM, whereas the correspond-
ing quinazoline derivative 22 showed no adenosine A1
receptor affinity at all. The aromatic benzamides 6 and
27 differed likewise. Isoquinoline derivative 6 showed
an adenosine A1 receptor affinity of 3.2 mM, whereas
the corresponding quinazoline derivative 27 showed no
adenosine A1 receptor affinity at all.
This latter finding prompted us to prepare compound
28 which is the quinazoline derivative of the potent
adenosine A3 antagonist 13 in order to increase the A3/
A1 selectivity even more. However, the adenosine A3
receptor affinity decreased almost 10-fold by this sub-
stitution, whereas the adenosine A1 receptor affinity
slightly increased.
1-Am in o-3-(2-p yr id in yl)isoqu in olin e. In a flame-dried
three-necked flask, equipped with a mechanical stirrer, con-
taining 50 mL condensed ammonia under nitrogen atmo-
sphere, 30 mmol of potassium was added in portions at -40
°C together with some crystals of ferric nitrate as catalyst.
The gray suspension was then cooled to -78 °C, and 30 mmol
of 2-methylbenzonitrile in 20 mL of anhydrous THF was added
slowly with appearance of a dark red color, directly followed
by the addition of 30 mmol of 2-cyanopyridine in 20 mL of
anhydrous THF. The reaction mixture was stirred overnight
at room temperature to evaporate the ammonia and hydro-
lyzed with 40 mL of 2 M aqueous ammonium chloride solution,
creating a buffer of pH ≈ 9. THF was removed under reduced
pressure, and 30 mL of diethyl ether was added. The organic
layer was separated and dried with Na2SO4, and after filtra-
tion, ether/HCl was added dropwise under stirring until
precipitation ceased. The dark precipitate was isolated by
filtration, washed with ether, and solved in ether/0.5 M NaOH.
The organic layer was separated, dried with Na2SO4, filtrated,
and evaporated until dryness under reduced pressure. Sub-
limation yielded 58% free base as white crystals.
Gen er a l P r oced u r e for P r ep a r a tion of Der iva tives 19
a n d 28. A solution of 20 mmol of 1-amino-3-(2-pyridinyl)-
isoquinoline or 4-amino-2-(2-pyridinyl)quinazoline in 40 mL
of anhydrous THF was cooled to -10 °C in
a nitrogen
Con clu sion s
atmosphere, and 12.5 mL of 1.6 M n-butyllithium in hexane
was added dropwise while stirring. After the mixture was
stirred for 10 min at this temperature, a solution of 20 mmol
of freshly distilled acyl chloride in 5 mL of anhydrous THF
was added, and stirring was continued for 1 h at -10 °C. After
warming to room temperature, the mixture was hydrolyzed
with water and extracted with chloroform (3 × 50 mL), and
the combined organic layers were washed twice with a 5%
NaHCO3 solution, dried over sodium sulfate, and evaporated
to dryness under reduced pressure. The residue was purified
by crystallization from methanol/ethyl acetate.
In this study we optimized the affinity of isoquinoline
compounds selectively for the adenosine A3 receptor
versus A1 and A2A receptors by use of an amide bond
on position 1 of the isoquinoline ring and by aromatic
substitution of the benzamide ring.
These investigations resulted in the potent and selec-
tive antagonist VUF 8504 (13) for the human adenosine
A3 receptor with an affinity of 17 nM. This compound
showed negligible affinity at the rat A1 or A2A receptor.
This potent and selective compound may become a
useful tool in a further pharmacological characterization
of the human adenosine A3 receptor and in the inves-
tigation of its physiological role. We are currently
exploring analogues of VUF8504 as potential adenosine
A3 receptor antagonists.
3,4-Dim eth oxy-N-[3-(2-pyr idin yl)isoqu in olin -1-yl]ben z-
1
a m id e (19): yield 46%; mp 157 °C; H NMR ([D6]DMSO, ref
3
DMSO[D5H] ) 2.49 ppm) δ 3.87 (s, 6H, OCH3), 7.15 (d, J 5′′6′′
3
3
) 8.4 Hz, 1H, H5′′), 7.47 (ddd, J 4′3′ ) 4.9 Hz, J 4′5′ ) 7.5 Hz,
4J 4′6′ ) 1.2 Hz, 1H, pyr-H4′), 7.64-7.77 (m, 4H, H5, H6, H7,
and H2′′), 7.82-7.98 (m, 2H, pyr-H3′ and H6′′), 8.20 (ddd, 3J 5′6′
3
4
) 8.0 Hz, J 5′4′ unsolved, J 5′3′ ) 1.7 Hz, 1H, pyr-H5′), 8.44 (d,
3J 87 ) 8.1 Hz, 1H, H8), 8.74 (ddd, J 6′5′ ) 8.0 Hz, J 6′4′ ) 1.2
Hz, J 6′3′ unsolved, 1H, pyr-H6′), 8.83 (s, 1H, H4), 10.96 (s,
3
4
5
Exp er im en ta l Section
0.8H, NH) and 15.15 (bs, 0.2H, OH). Anal. (C23H19N3O3) C,
H, N.
Ma ter ia ls. n-Butyllithium was purchased from ACROS
(Belgium) as a 2.5 M solution in hexane. p-Anisic acid,
picolinonitrile, 3,4-dimethoxybenzoic acid, ferric nitrate, and
2-methylbenzonitrile were purchased from ACROS (Belgium).
THF was predried over CaH2 and distilled from LiAlH4. All
other solvents used were of analytical grade. 1-Amino-3-(2-
pyridinyl)isoquinoline was prepared by a modified method of
de Zwart et al.,18 and 4-amino-2-(2-pyridinyl)quinazoline was
prepared as described by Linschoten et al.19 The synthesis
and identification of compounds 1-3, 23, and 25 were de-
scribed in the previous paper.1 Compounds 4-18,18 20-22,19
24,19 and 26-2719 are from our laboratory stock (VU Amster-
dam).
4-Me t h oxy-N -[2-(2-p yr id in yl)q u in a zolin -4-yl]b e n z-
a m id e (28): yield 52%; mp 147-148 °C; HRMS (EI) m/z
1
356.1270 (M+), 356.1273 (C21H16N4O2); H NMR ([D6]DMSO,
3
ref DMSO[D5H] ) 2.49 ppm) δ 3.87 (s, 3H, CH3), 7.07 (d, J AB
) 8.9 Hz, 2H, AA′BB′ ArH), 7.70-7.77 (m, 3J 4′5′ ) 7.5 Hz, 2H,
H8, and pyr-H4′), 7.90-8.18 (m, 3H, H6, H7, and pyr-H5′),
3
3
8.39 (d, J BA ) 8.9 Hz, 2H, AA′BB′ ArH), 8.58 (dd, J 56 ) 8.1
4
3
4
Hz, J 57 ) unsolved, 1H, H5), 8.78 (dd, J 6′5′ ) 8.2 Hz, J 6′4′
)
1.4 Hz, 1H, pyr-H6′), 8.89 (ddd, 3J 3′4′ ) 4.2 Hz, 4J 3′5′ unresolved,
5J 3′6′ unresolved, 1H, pyr-H3′), and 15.53 (bs, 1H, OH). Anal.
(C21H16N4O2) C, H, N.
P h a r m a cology.
R a d ioliga n d
Bin d in g
St u d ies.
Syn th esis. 1H NMR and 13C NMR spectra were recorded
on a Bruker AC 200 (1H NMR, 200 MHz; 13C NMR, 50.29 MHz)
spectrometer with tetramethylsilane or sodium 3-(trimethyl-
silyl)propionate as an internal standard. 2D-NMR (H-H and
C-H) COSY techniques were frequently used to support
interpretation of 1D spectra. The multiplicity of the carbon
signals was determined by DEPT or APT spectra or by a
combination of a normal decoupled carbon spectrum combined
with a CH correlation. The symbols used are (p) for primary,
(s) for secondary, (t) for tertiary, and (q) for quaternary carbon
signals. HRMS measurements were performed on a Finnigan
MAT 90 mass spectrometer (direct inlet) at an ionization
potential of 70 eV. FT-IR spectra were recorded on a Mattson
Adenosine A1 receptor affinities were determined on rat
cerebral cortex membranes with [3H]DPCPX as radioligand
as described previously.20 Adenosine A2A receptor affinities
were determined on rat striatal membranes with [3H]CGS
21680 as radioligand according to a protocol published previ-
ously.21
Binding of [125I]AB-MECA to human A3 receptors in stably
transfected HEK 293 cell membranes was performed as
described.22,23
Ack n ow led gm en t. The authors thank J acobien
Frijtag von Drabbe Ku¨nzel, Miriam de Groote, and

4000 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 21
Mathieu Mous for measurement of binding at A1, A2A
and A3 receptors and Bas Limmen for his synthetic
work. We thank Dr. Roel Vollinga for helpful discussions
and critically reading the manuscript and Karl-Norbert
Klotz (University of Wu¨rzburg, Germany) for providing
HEK 293 cells expressing the human adenosine A3
receptor. This research was supported by Byk Gulden
(Zwanenburg, The Netherlands).
van Muijlwijk-Koezen et al.
(11) Lubitz, D. K. J . E. v.; Carter, M. F.; Deutsch, S. I.; Lin, R. C. S.;
Mastropaolo, J .; Meshulam, Y.; J acobson, K. A. The effects of
adenosine A3 receptor stimulation on seizures in mice. Eur. J .
Pharmacol. 1995, 275, 23-29.
(12) Lubitz, D. K. J . E. v.; Lin, R. C. S.; Popik, P.; Carter, M. F.;
J acobson, K. A. Adenosine A3 receptor stimulation and cerebral
ischemia. Eur. J . Pharmacol. 1994, 263, 59-67.
(13) J acobson, K. A., Galen, P. J . M., van, and Williams, M. Adenosine
receptors: pharmacology, structure-activity relationships, and
therapeutic potential. J . Med. Chem. 1992, 35, 407-422.
(14) Kim, H. O.; J i, X. D.; Melman, N.; Olah, M. E.; Stiles, G. L.;
J acobson, K. A. Structure-activity relationships of 1,3-dialkyl-
xanthine derivatives at rat A3 adenosine receptors. J . Med.
Chem. 1994, 37, 3373-3382.
,
Abbr evia tion s
(15) Rhee, A. M. v.; J iang, J . L.; Melman, N.; Olah, M. E.; Stiles, G.
L.; J acobson, K. A. Interaction of 1,4-dihydropyridine and
pyridine derivatives with adenosine receptors: selectivity for A3
receptors. J . Med. Chem. 1996, 39, 2980-2989.
(16) Kim, Y.-C.; J i, X.; J acobson, K. A. Derivatives of the triazolo-
quinazoline adenosine antagonist (CGS15943) are selective for
the human A3 receptor subtype. J . Med. Chem. 1996, 39, 4142-
4148.
AcOH
Ac2O
APT
n-BuLi
CI
acetic acid
acetic anhydride
attached proton test
n-butyllithium
chemical ionization
correlated spectroscopy
COSY
DEPT
(17) J acobson, M. A.; Chakravarty, P. K.; J ohnson, R. G.; Norton, R.
Novel selective nonxanthine A3 adenosine receptor antagonists.
Drug Dev. Res. 1996, 37, 131.
distortionless enhancement by polariza-
tion transfer
(18) Zwart, M. A. H. d.; Goot, H. v. d.; Timmerman, H. Synthesis
and copper dependent antimycoplasmal activity of 1-amino-3-
(2-pyridyl)isoquinoline derivatives. 1. Amides. J . Med. Chem.
1988, 31, 716-722.
(19) Linschoten, M. R.; Gaisser, H.-D.; Goot, H. v. d.; Timmerman,
H. Synthesis and copper dependent antimycoplasmal activity
of quinazolinylamidines and amides: a case of concentration
quenching. Eur. J . Med. Chem.-Chim. Ther. 1984, 19, 137-
142.
(20) Pirovano, I. M.; IJ zerman, A. P.; Galen, P. J . M. v.; Soudijn, W.
The influence of molecular structure of N6-(w-aminoalkyl)-
adenosines on adenosine receptor affinity and intrinsic activity.
Eur. J . Pharmacol. 1989, 172, 185-191.
(21) J arvis, M. F., Schulz, R., Hutchison, A. J ., Do, U. H., Sills, M.
A., and Williams, M. [3H]CGS 21680, A selective A2 adenosine
receptor agonist directly labels A2 receptors in rat brain. J .
Pharmacol. Exp. Ther. 1989, 251, 888-893.
(22) Olah, M. E.; Gallorodriguez, C.; J acobson, K. A.; Stiles, G. L.
125I-4-aminobenzyl-5′-N-methylcarboxamidoadenosine, a high
affinity radioligand for the rat A3 adenosine receptor. Mol.
Pharmacol. 1994, 45, 978-982.
(23) Galen, P. J . M. v.; Bergen, A. H. v.; Gallorodriquez, C.; Melman,
N.; Olah, M. E.; IJ zerman, A. P.; Stiles, G. L.; J acobson, K. A. A
binding site model and structure-activity relationships for the
rat A3 adenosine receptor. Mol. Pharmacol. 1994, 45, 1101-
1111.
(24) Serc`ık, J .; Grambal, F. Chemistry Amidines Imidates 1975, 567.
(25) Becher, J .; Lohse, C. The photochemical rearrangement of
isoquinolinium N-acyl ylides to 1-acylaminoisoquinolines via the
excited singlet state. Acta Chem. Scand. 1972, 26, 4041-4048.
(26) Aysola, D. P.; Gibson, M. S. Unusual formation of N-(isoquino-
linyl)benzamide from 1-o- and 1-m-chlorobenzoylisoquinolines.
Can. J . Chem. 1977, 55, 435-437.
(27) Møller, J .; Becher, J .; Lohse, C.; Lempert, K.; Agai, B. Mass
spectrometric studies of 1-acylaminoisoquinolines and of some
N-(2-isoquinolinio)amidates. Acta Chem. Scand. 1973, 27, 3647-
3656.
DMSO
Et2O
FT-IR
[3H]DPCPX
dimethylsulfoxide
diethyl ether
Fourier transformation infrared
[3H]-1,3-dipropyl-8-cyclopentyl-xanthine
[3H]CGS 21680 [3H]-2-[[4-(2-carboxyethyl)phenyl]ethyl-
amino]-5′-N-(ethylcarbamoyl)adenosine
human embryonic kidney cells
hexamethylphosphoric triamide
HEK cells
HMPT
[
125I]AB-MECA 125I]-N6-(4-amino-3-iodobenzyl)-5′-(N-meth-
[
ylcarbamoyl)adenosine
equilibrium inhibition constant
tetrahydrofuran
Ki
THF
TLC
thin layer chromatography
Refer en ces
(1) Muijlwijk-Koezen, J . E. v.; Timmerman, H.; Link, R.; Goot, H.
v. d.; IJ zerman, A. P. A Novel Class of Adenosine A3 Receptor
Ligands. 1. 3-(2-Pyridinyl)Isoquinoline Derivatives. J . Med.
Chem. 1998, 41, 3987-3993.
(2) Zhou, Q.-Y.; Li, C.; Olah, M. E.; J ohnson, R. A.; Stiles, G. L.;
Civelli, O. Molecular cloning and characterization of an adenos-
ine receptor: the A3 adenosine receptor. Proc. Natl. Acad. Sci.
U.S.A. 1992, 89, 7432-7436.
(3) Salvatore, C. A.; J acobson, M. A.; Taylor, H. E.; Linden, J .;
J ohnson, R. G. Molecular cloning and characterization of the
human A3 adenosine receptor. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 10365-10369.
(4) Linden, J . Cloned adenosine A3 receptors: Pharmacological
properties, species differences and receptor functions. Trends
Pharmacol. Sci. 1994, 15, 298-306.
(5) Palmer, T. M.; Gettys, T. W.; Stiles, G. L. Differential interaction
with and regulation of multiple G-proteins by the rat A3
adenosine receptor. J . Biol. Chem. 1995, 270, 16895-16902.
(6) Linden, J ., Taylor, H. E., Robeva, A. S., Tucker, A. L., Stehle, J .
H., Rivkees, S. A., Fink, J . S., and Reppert, S. M. Molecular
cloning and functional expression of a sheep A3 adenosine
receptor with widespread tissue distribution. Mol. Pharm. 1993,
44, 524-532.
(28) Tamura, Y.; Matsugashita, S.; Ishibashi, H.; Ikeda, M. Photo-
chemistry of N-acylimino-isoquinolium and -quinolinium be-
taines. Tetrahedron 1973, 29, 2359-2364.
(29) Grahner, B.; Winiwarter, S.; Lanzner, W.; Muller, C. E. Syn-
thesis and structure-activity relationships of deazaxanthines:
Analogues of potent A1 and A2-adenosine receptor antagonists.
J . Med. Chem. 1994, 37, 1526-1534.
(7) Ramkumar, V.; Stiles, G. L.; Beaven, M. A.; Ali, H. The A3AR is
the unique adenosine receptor which facilitates release of allergic
mediators in mast cells. J . Biol. Chem. 1993, 268, 16887-16890.
(8) Schaick, E. A. v.; J acobson, K. A.; Kim, H. O.; IJ zerman, A. P.;
Danhof, M. Hemodynamic effects and histamine release elicited
by the selective adenosine A3 receptor agonist 2-Cl-IB-MECA
in conscious rats. Eur. J . Pharmacol. 1996, 308, 311-314.
(9) Hannon, J . P.; Pfannkuche, H. J .; Fozard, J . R. A role for mast
cells in adenosine A3 receptor-mediated hypotension in the rat.
Br. J . Pharmacol. 1995, 115, 945-952.
(10) Beaven, M. A.; Ramkumar, V.; Ali, H. Adenosine A3 receptors
in mast cells. Adenosine A3 receptor antagonists may be useful
in the treatment of allergic disorders. Trends Pharmacol. Sci.
1994, 15, 13-14.
(30) Siddiqi, S. M.; J acobson, K. A.; Esker, J . L.; Olah, M. E.; J i, X.
D.; Melman, N.; Tiwari, K. N.; Secrist, J . A.; Schneller, S. W.;
Cristalli, G.; Stiles, G. L.; J ohnson, C. R.; IJ zerman, A. P. Search
for new purine- and ribose-modified adenosine analogues as
selective agonists and antagonists at adenosine receptors. J .
Med. Chem. 1995, 38, 1174-1188.
(31) Mu¨ller, C. E.; Geis, U.; Grahner, B.; Lanzner, W.; Eger, K. Chiral
pyrrolo [2,3-d]pyrimidine and pyrimido[4,5-b]indole deriva-
tives: Structure-activity relationships of potent, highly stereo-
selective A1 adenosine receptor antagonists. J . Med. Chem. 1996,
39, 2482-2491.
J M980037I