1-Substituted 4-[chloropyrazolyl][1,2,4]triazolo[4,3-a]quinoxalines: Synthesis and structure-activity relationships of a new class of benzodiazepine and adenosine receptor ligands

Article abstract of DOI:10.1002/(SICI)1521-4184(199805)331:5<163::AID-ARDP163>3.0.CO;2-3

Starting from 3-(3-chloro-1H-pyrazol-5-yl)-1H-quinoxalin-2-one (2) a series of substituted [1,2,4]triazolo[4,3-a]quinoxalines (3a-f) was prepared via a multistep reaction sequence. Affinities of the novel derivatives 3a-f for benzodiazepine as well as for adenosine A₁- and A(2A)-receptors of rat brain were determined by radioligand binding assays. 1-Methyl-4-(3-chloro- 1H-pyrazol-5-yl) derivative 3a exhibited submicromolar affinity for the benzodiazepine binding site of GABA(A) receptors (K(i) = 340 nM) and was less potent at A₁- (K(i) = 7.85 μM) and A(2A)-(K(i) = 1.43 μM) adenosine receptors (AR). Derivatives with larger substituents in the 1-position showed reduced binding to benzodiazepine and A(2A)-AR, but increased A₁-AR affinity, the 2-thienylmethyl derivative 3f being the most potent and selective A₁-AR ligand of the present series (K(i) = 200 nM).

Full text of DOI:10.1002/(SICI)1521-4184(199805)331:5<163::AID-ARDP163>3.0.CO;2-3

Substituted [1,2,4]triazolo[4,3-a]quinoxalines
163
1-Substituted 4-[Chloropyrazolyl][1,2,4]triazolo[4,3-a]quinoxalines:
Synthesis and Structure-Activity Relationships of a New Class of
Benzodiazepine and Adenosine Receptor Ligands✩
a)
b)
b)
Barbara Matuszczak *, Elšbieta Pekala , and Christa E. Müller
a) Institute of Pharmaceutical Chemistry, University of Innsbruck, Innrain 52a, A-6020 Innsbruck, Austria
b)
Institute of Pharmacy and Food Chemistry, Pharmaceutical Chemistry, Julius-Maximilians-University Würzburg, Am Hubland, D-97074 Würzburg,
Germany
Key Words: Adenosine receptors; GABA receptor; benzodiazepine; 4-[chloropyrazolyl][1,2,4]triazolo[4,3-a]quinoxalines;
A
structure-activity relationships
toxicity. They are widely used as anxiolytics, antidepressants,
Summary
[5a–c]
hypnotics, anticonvulsants, and muscle relaxants
. Due
to the side effects such as psychological and physical depend-
ence, there is continuing interest in the development of novel
central benzodiazepine receptor ligands with improved prop-
erties. It has been found that many compounds, whose struc-
tures were apparently unrelated to benzodiazepines, are able
Starting from 3-(3-chloro-1H-pyrazol-5-yl)-1H-quinoxalin-2-one
(2) a series of substituted [1,2,4]triazolo[4,3-a]quinoxalines (3a–f)
was prepared via a multistep reaction sequence. Affinities of the
novel derivatives 3a–f for benzodiazepine as well as for adenosine
A1- and A2A-receptors of rat brain were determined by radioligand
binding assays. 1-Methyl-4-(3-chloro-1H-pyrazol-5-yl) derivative
3a exhibited submicromolar affinity for the benzodiazepine bind-
ing site of GABAA receptors (Ki = 340 nM) and was less potent at
A1- (Ki = 7.85 µM) and A2A- (Ki = 1.43 µM) adenosine receptors
(AR). Derivatives with larger substituents in the 1-position showed
reduced binding to benzodiazepine and A2A-AR, but increased
A1-AR affinity, the 2-thienylmethyl derivative 3f being the most
potent and selective A1-AR ligand of the present series (Ki =
200 nM).
to bind competitively to the benzodiazepine binding site of
[5b–c]
the GABA receptor
, e.g. [1,2,4]triazolo[1,5-c]quin-
A
[3a]
[3d]
azolin-5(6H)-ones , [1,2,4]triazolo[4,3-b]pyridazines
and imidazo[1,5-a]quinoxaline derivatives
,
[6]
.
Adenosine receptors (AR) – which belong to the super-
family of G-protein-coupled receptors – can be divided into
two ‘high affinity’ subtypes (A and A ) and two ‘low
1
2A
[7]
affinity’ subtypes (A and A ) . Whereas the first types of
2B
3
receptors are stimulated by adenosine in nanomolar concen-
trations the latter require micromolar concentrations of
[7]
adenosine for activation . Antagonists of the A -adenosine
1
receptors are of interest for the treatment of dementias (e.g.
Alzheimer’s disease), depression, or as diuretics, antihyper-
Introduction
Recently we have observed a novel type of a 1,2-diazine →
1,2-diazole ring transformation . The product of this reac-
tion – a pyrazolyl substituted quinoxalinone (2) – was found
to represent an attractive educt for the development of poten-
tensives, antiarrhythmics, and for the prevention of acute
[1]
[8a–b]
renal failure
. On the other hand, adenosine A antago-
2A
nists may be used as drugs for the treatment of Morbus
[8a–b]
Parkinson
. During recent years a number of potent and
[2a–b]
[8a–c]
tially bio-active compounds
.
selective AR antagonists have been discovered
, and
some of them are currently under clinical development. Most
of the adenosine receptor antagonists are aromatic nitrogen-
containing heterocyclic compounds, e.g. xanthines, pyra-
zolo[1,5-a]pyridines, pyrrolo[2,3-d]pyrimidines, pyr-
azolo[3,4-d]pyrimidines, [1,2,4]triazolo[1,5-a]quinoxalines,
[4b–c, 8a–c]
and [1,2,4]triazolo- [4,3-a]quinoxalines
.
Herein we report on the synthesis of tricyclic compounds
of type 3 structurally related to recently described adenosine
[4b–c]
receptor antagonists 4
. Moreover we present the results
Scheme 1. Formation of the pyrazolyl-substituted quinoxalinone (2)
from (1).
of biological evaluation as benzodiazepine and adenosine A -
1
and A -receptor ligands.
2A
In view of the interesting biological activities described for
[1,2,4]triazolo-annelated heterocycles (e.g. benzodiazepine
Results and Discussion
[3a–d]
receptor binding
and adenosine receptor antagonistic
[4a–c]
properties
), the pyrazolyl substituted [1,2,4]triazoloqui-
Chemistry
noxalines of type 3 became an object of our interest as
potential benzodiazepine and/or adenosine receptor ligands.
Benzodiazepine derivatives have attracted the attention of
The [1,2,4]triazolo[4,3-a]quinoxaline derivatives of type 3
were prepared as shown in Scheme 2. Starting from the ring
[1]
researchers owing to their interesting activities and their low contraction product 2 , the chloro substituted quinoxaline
Arch. Pharm. Pharm. Med. Chem.
© WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
0365-6233/98/0505/0163 $17.50 +.50/0

164
Matuszczak, Pekala, and Müller
Figure 1. Structures of the target compounds of type 3 and adenosine receptor antagonists 4 [4b–c]
.
5 was prepared by refluxing 2 in phosphorus oxychloride in synthesised derivatives is based on correct elemental analyses
[2a]
1
the presence of pyridine
. The novel 2-hydrazino deriva- as well as on IR, H-NMR, and mass spectral data.
tive 6 became accessible by heating of 5 in hydrazine hydrate
at 150 °C.
Biological Evaluation of Compounds 3a–f as Ligands for
Acylation of the hydrazino function was achieved by treat-
ment of 6 in dry N,N-dimethyl formamide with carboxylic
acid derivatives (acetyl chloride, benzoyl chloride, phenyl-
acetyl chloride, furan-2-carbonyl chloride, thiophene-2-car-
bonyl chloride, and thiophen-2-yl-acetyl chloride,
respectively) in the presence of triethylamine. The hydrazide
derivatives 7a–f were converted into the tricyclic compounds
of type 3 by heating in a mixture of 1,2-dichloroethane and
phosphorus oxychloride. Interestingly, cyclisation of com-
pound 7d was already observed while recrystallising the
compound in a mixture of 1,4-dioxane and N,N-dimethyl
formamide. Under these conditions a 68% yield of 3d was
obtained. However, it turned out that this facilitated cyclisa-
tion was restricted to compound 7d and the approach could
not be generalised.
Benzodiazepine Binding Sites
The compounds were investigated in radioligand binding
assays at rat brain cortical membranes. Initially, all com-
pounds were screened for their potency to displace [ H]di-
azepam from its binding site at a single concentration
(10 µM). For the most potent compound, i.e. 3a, the K value
3
i
was determined additionally.
Structure-Activity Relationships
The results obtained (see Table 1) indicate that the methyl
substituted tricycle is the most potent derivative in this series
(K = 340 nM) which exhibits an approximately 40-fold lower
i
activity than the reference compound diazepam. Formal re-
placement of the methyl substituent by a (hetero)aromatic
(phenyl, 2-thienyl, 2-furyl) or (hetero)arylmethyl (benzyl,
2-thienylmethyl) moiety was found to result in a decrease in
benzodiazepine receptor affinity.
Analytically pure hydrazides as well as the cyclised prod-
ucts could be obtained by recrystallisation from appropriate
solvents (see Experimental). Structural proof for all newly
Scheme 2. Synthesis of 4-[3(5)-chloro-1H-pyrazol-5(3)-yl][1,2,4]triazolo[4,3-a]quinoxalines (3a-f). Reagents used: (i) POCl3, pyridine, reflux; (ii) hydrazine
monohydrate, reflux; (iii) R-COCl, N(C2H5)3 in N,N-dimethyl formamide, room temperature; (iv) POCl3, 1,2-dichloroethane, reflux or heating in a mixture
of N,N-dimethyl formamide and 1,4-dioxane (for R = 2-furyl).
Arch. Pharm. Pharm. Med. Chem. 331, 163–169 (1998)

Substituted [1,2,4]triazolo[4,3-a]quinoxalines
165
Table 1. Inhibition of [3H]diazepam binding to rat brain by test compounds.
——————————————————————————————————————————————
Compd.
R
Percent specific inhibition ± SEM (n = 3)
of [3H]diazepam binding to rat brain cortical
membranes by test compound (10 µM)
Ki ± SEM (n = 3)
[µM]
——————————————————————————————————————————————
diazepam
n.d.a)
0.009 ± 0.003 [9]
0.34 ± 0.08
n.d.a)
n.d.a)
n.d.a)
3a
3b
3c
3d
3e
3f
methyl
85 ± 1
57 ± 5
42 ± 2b)
51 ± 9
57 ± 2
39 ± 19
phenyl
benzyl
2-furyl
2-thienyl
2-thienylmethyl
n.d.a)
n.d.a)
——————————————————————————————————————————————
a) n.d. = not determined
b) Inhibition of [3H]diazepam binding may be underestimated due to low solubility of 3c (estimated < 3 µM).
Radioligand could be absorbed to the precipitate formed and thus result in high counts feigning low affinity.
3
[ H]3-(3-hydroxypropyl)-8-(3-methoxystyryl)-7-methyl-1-
[11]
Table 2. A1-Adenosine receptor affinities:
——————————————————————————————
propargylxanthine (MSX-2)
as A ligand, caffeine was
2A
chosen as a standard reference compound. The results thus
obtained are shown in Tables 2 and 3.
Structure-Activity Relationships
The results of these binding assays indicate that all new
compounds prepared show affinity to adenosine receptors.
Compd.
R
A1-affinity
Ki ± SEM [µM] (n = 2)
Rat brain cortical membranes
[3H]CCPA
A -Affinity of the novel derivatives is higher than that of
1
caffeine, which was tested as a reference compound. Further-
more, the methyl substituted tricyclic compound was about
——————————————————————————————
20-fold more potent at A -receptors than caffeine. Since a
2A
caffeine
3a
23.5 ± 3.0
virtually intact ribose moiety is required for agonistic activity
[8]
at AR
it can be assumed that triazoloquinoxalines 3a–f,
methyl
7.85 ± 0.95
1.62 ± 0.68
0.31 ± 0.09a)
0.41 ± 0.16
0.71 ± 0.06
0.20 ± 0.01
which are lacking a ribose residue, are antagonists at AR, like
caffeine.
A substituent in position 1 of the tricyclic system has a
3b
phenyl
3c
benzyl
profound effect on affinity as well as selectivity.
3d
2-furyl
For A -adenosine receptor affinity the following structure-
1
3e
2-thienyl
2-thienylmethyl
activity relationships could be established: the methyl substi-
tuted derivative 3a exhibits only low affinity, (formal)
exchange of methyl by a (hetero)aryl or (hetero)arylmethyl
moiety leads to increased potency. Comparing these com-
3f
——————————————————————————————
a) Estimated value through extrapolation of the curve. A complete curve
could not be recorded due to low solubility of the compound.
pounds, the rank order of A -AR affinity according to the
1
nature of the substituent is as follows: phenyl (3b) < 2-thienyl
(3e) < 2-furyl (3d) < benzyl (3c) < 2-thienylmethyl (3f). In
this series, the heteroaryl substituted derivatives (3d/3e)
turned out to be about 2- to 4-fold more potent than the
corresponding phenyl substituted compound 3b. Formal re-
placement of the (hetero)aryl moiety by a (hetero)arylmethyl
Biological Evaluation of Compounds 3a–f as Adenosine Re-
ceptor Antagonists
The pyrazolyl-substituted [1,2,4]triazolo[4,3-a]quinoxal-
ines 3 were further tested in radioligand binding assays for
affinity at A - and A -adenosine receptors in rat cortical
1
2A
substituent, however, leads to a further increase in A -AR
1
membrane and rat striatal membrane preparations, respec-
3
6
affinity. The most active compound in this series (3f) was
was used as A ligand and approximately 120-fold more potent than caffeine.
tively. The A -selective agonist [ H]2-chloro-N -cyclopen-
1
[10]
tyladenosine (CCPA)
1
Arch. Pharm. Pharm. Med. Chem. 331, 163–169 (1998)

166
Matuszczak, Pekala, and Müller
Table 3. A2A-Adenosine receptor affinities
——————————————————————————————————————————————
Compd.
R
A2A-affinity
Ki
Percent specific inhibition ± SEM (n = 2)
of [3H]MSX-2-binding to rat striatal membranes
by test compound (25 µM)
[µM]
——————————————————————————————————————————————
caffeine
3a
36 (at 30 µM)
96 ± 1
32.5
1.43 a)
methyl
3b
phenyl
0
n.d.b)
n.d.b)
n.d.b)
n.d.b)
n.d.b)
3c
benzyl
6.1 ± 1.5c)
3d
2-furyl
31 ± 3
3e
2-thienyl
2-thienylmethyl
22 ± 13
21 ± 10
3f
——————————————————————————————————————————————
a) result from single experiment
b) n.d. = not determined
c) Inhibition of [3H]MSX-2 binding may be underestimated due to low solubility of 3c (estimated < 3 µM). Radioligand
could be absorbed to the precipitate formed and thus result in high counts feigning low affinity.
These structure-activity relationships are in accordance Compound 3a is 5-fold selective for the A -AR versus A ,
2A
1
with pharmacophore models described for A -adenosine re- while compounds with larger substituents are A -selective.
1
1
ceptor antagonists, as shown for the potent A -selective AR
In conclusion, the pyrazolyl-substituted [1,2,4]triazolo[4,3-
1
[8b,12]
antagonist ADPEP
tive (see Figure 2
, a pyrrolo[2,3-d]pyrimidine deriva- a]quinoxaline derivatives show affinity to benzodiazepine
). A -AR ligands typically contain a and/or adenosine receptors. Whereas the 1-(hetero)aryl and
[8b]
1
hydrogen bond donor at a certain distance from a hydrogen 1-(hetero)arylmethyl substituted derivatives represent selec-
bond acceptor (see Figure 2). A lipophilic domain increases tive A -adenosine receptor ligands, compound 3a (bearing a
1
A selectivity of AR antagonists versus A -AR since the methyl group in position 1 of the tricycle) shows affinity to
1
2A
A
-AR shows restricted bulk tolerance in that receptor benzodiazepine and A -adenosine receptors (A :A = 5),
2A
2A 2A 1
[8b, 12d]
domain
. A major drawback of the derivatives bearing with four-fold higher affinity to benzodiazepine receptors as
a lipophilic moiety (especially benzyl), however, is the low compared to A adenosine receptors. Based on these struc-
2A
solubility of these compounds.
ture-activity relationships further structural modifications are
in progress in order to increase affinity and selectivity of this
new class of AR antagonists and ligands for the ben-
zodiazepine binding site of GABA receptors.
A
Acknowledgement
The authors are very grateful to Dr. Dietmar Rakowitz (Institute of
Pharmaceutical Chemistry, University of Innsbruck) for recording the mass
spectra. E.P. was on leave from the Department of Chemical Technology of
Drugs, laboratory of Dr. habil. Kiec-Kononowicz, Collegium Medicum of
the Jagiellonian University, Kraków, Poland, with support from ‘Deutscher
Akademischer Austauschdienst’ (DAAD), which is gratefully acknow-
ledged.
Experimental
Melting points were determined on a Kofler hot-stage microscope
(Reichert) and are uncorrected. IR spectra were taken on a Mattson Galaxy
Series FT-IR 3000 spectrophotometer (KBr pellets). Mass spectra were
obtained on a Finnigan MAT SSQ 7000. 1H and 13C NMR spectra were
recorded on a Varian Gemini 200 spectrometer (1H: 199.98 MHz). The centre
of the solvent multiplet ([D6]DMSO) was used as internal standard (chemical
shifts in δ ppm), which was related to TMS with δ 2.49 ppm for 1H. Reactions
were monitored by TLC using Polygram® SIL G/UV254 (Macherey-Nagel)
plastic-backed plates (0.25 mm layer thickness). Microanalyses were per-
formed at the Institute of Physical Chemistry (Mag. J. Theiner), University
Figure 2. Pharmacophore model for A1-adenosine receptor antagonists [8b,
12d]
.
Structure-activity relationships for compounds 3 at A
-
2A
AR can be described as follows: the 1-methyl substituted
derivative 3a represents the most active A -AR ligand in
2A
this series, (formal) replacement of methyl by (hetero)aryl or
(hetero)arylmethyl leads to a reduction of receptor affinity.
Arch. Pharm. Pharm. Med. Chem. 331, 163–169 (1998)

Substituted [1,2,4]triazolo[4,3-a]quinoxalines
167
Anal. C,H,N.– IR (KBr): 3461, 3230, 1637 cm–1.– 1H-NMR ([D6]DMSO +
1 drop of D2O): δ = 7.92–7.91 (m, 2H), 7.62 (‘s’ br, 1H), 7.51 (‘s’ br, 1H),
7.20 (‘s’ br, 1H) (quinoxaline-H-5, -H-6, -H-7, -H-8, pyrazole-H-4, furan-H-
5), 7.29 (d, J = 3.5 Hz, 1Hz, furan-H-3), 6.70 (dd, J = 3.5 Hz, J = 1.8 Hz, 1H,
furan-H-4).– EI MS (70 eV): m/z = 354 [M+].
of Vienna, Austria. All analyses were within ±0.4% of the calculated values.
Light petroleum refers to the fraction of bp 40–60 °C. The yields are not
optimised.
Syntheses
Starting materials: 2-Chloro-3-(3-chloro-1H-pyrazol-5-yl)-quinoxaline
(5) was prepared by treatment of (2) [1] with phosphorus oxychloride and
pyridine [2]. 3-(3-Chloro-1H-pyrazol-5-yl)-1H-quinoxalin-2-one (2) became
accessible by reaction of N-(2-aminophenyl)-3,6-dichloropyridazine-4-car-
3-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-2-(2-thenoyl)-hydrazinoquinoxaline
(7e)
boxamide (1) with sodium hydride in dry N,N-dimethyl formamide [2]
.
Recrystallisation from a mixture of 1,4-dioxane and N,N-dimethyl for-
mamide (ca. 2:1) yielded 65% of yellow needles, mp 195–200 °C.–
C16H11ClN6OS (370.82) Anal. C,H,N.– IR (KBr): 3282, 2923, 1643 cm–1.–
1H-NMR ([D6]DMSO + 1 drop of D2O): δ = 8.00–7.98 (m, 1H), 7.92–7.83
(m, 2H), 7.62–7.61 (m, 2H), 7.52–7.44 (m, 1H), 7.25–7.20 (m, 2H) (quinox-
aline-H-5, -H-6, -H-7, -H-8, pyrazole-H-4, thiophene-H-3, -H-4, -H-5).– EI
MS (70 eV): m/z = 370 [M+].
3-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-2-hydrazinoquinoxaline (6)
[2]
A mixture of 2-chloro-3-(3-chloro-1H-pyrazol-5-yl)-quinoxaline (5)
(0.512 g, 2 mmol) in hydrazine monohydrate (15 mL) was heated at 150 °C
until TLC indicated completion of the reaction (ca. 2 h). After cooling, the
mixture was poured into ice-water (150 mL). The crystals thus obtained were
collected by filtration, washed with water and light petroleum and were
recrystallised from tetrahydrofuran to yield 460 mg (88%) of a yellow
powder, mp 243–250 °C.– C11H9ClN6 (260.69) Anal. C,H,N.– IR (KBr):
3308, 3147 cm–1.– 1H-NMR ([D6]DMSO + 1 drop of D2O): δ = 7.71 (d, J =
7.8 Hz, 1H), 7.50–7.48 (m, 2H), 7.31–7.22 (m, 1H) (quinoxaline-H-5, H-6,
H-7, H-8), 7.08 (s, 1H, pyrazole-H-4).– EI MS (70 eV): m/z = 260 [M+].
3-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-2-(2-thienylacetyl)-hydrazino-
quinoxaline (7f)
Recrystallisation from 1,4-dioxane yielded 61% of a yellow powder, mp
223–229 °C, crystallised at 230 °C, then mp 250–252 °C.– C17H13ClN6OS
(384.85) Anal. C,H,N.– IR (KBr): 3210, 3041, 1640 cm–1.– 1H-NMR
([D6]DMSO + 1 drop of D2O): δ = 8.23 (dd, J = 6.0 Hz, J = 3.5 Hz, 1H),
8.12 (dd, J = 6.0 Hz, J = 3.5 Hz, 1H), 7.73–7.63 (m, 2H) (quinoxaline-H-5,
-H-6, -H-7, -H-8), 7.62 (s, 1H, pyrazole-H-4), 7.41 (dd, J = 4.9 Hz, J = 1.5
Hz, 1H, thiophene-H-5), 6.96–6.92 (m, 2H, thiophene-H-3, -H-4), 5.18 (s,
2H, CH2).– EI MS (70 eV): m/z = 384 [M+].
General Procedure for the Synthesis of Compounds of Type 7:
To a stirred mixture of (6) (0.261–0.521 g, 1.0–2.0 mmol) and triethyl-
amine (0.121–0.242 g, 1.2–2.4 mmol, 1.2 equiv.) in dry N,N-dimethyl
formamide (5–10 mL) was added slowly a solution of the corresponding
carboxylic acid chloride (1.1 equiv.) (acetyl chloride, benzoyl chloride,
phenylacetyl chloride, furan-2-carbonyl chloride, thiophene-2-carbonyl
chloride, and thiophen-2-yl-acetyl chloride, respectively) in 1–2 mL of dry
N,N-dimethyl formamide at 0 °C under a nitrogen atmosphere. The mixture
was allowed to warm to room temperature and stirred over night at this
temperature. Then the mixture was poured into 150–300 mL of water. The
crystals thus obtained were collected by filtration, washed with water and
light petroleum and were recrystallised (see below).
General Procedure for the Synthesis of Compounds of Type 3a–c and 3e–f:
A mixture of the corresponding 2-acylhydrazino-3-[3(5)-chloro-1H-pyra-
zol-5(3)-yl]-quinoxaline (7a–c, 7e–f) (0.5–1.0 mmol) and phosphorus oxy-
chloride (4–8 mL) in 1,2-dichloroethane (10–20 mL) was heated under reflux
until TLC indicated completion of the reaction (ca. 3 h). After cooling, the
mixture was poured into ice-water (150 mL) and was extracted exhaustively
with ethyl acetate. The organic layer was washed with 2N NaOH, water and
brine, dried over sodium sulphate and evaporated in vacuo. The products thus
obtained were purified by recrystallisation (see below).
2-Acetylhydrazino-3-[3(5)-chloro-1H-pyrazol-5(3)-yl]-quinoxaline (7a)
Recrystallisation from ethyl acetate yielded 83% of a dark yellow powder,
mp 293–297 °C.– C13H11ClN6O × 0.2 CH3COOC2H5 (320.35) Anal.
C,H,N.– IR (KBr): 3217, 1640 cm–1.– 1H-NMR ([D6]DMSO + 1 drop of
D2O): δ = 7.89 (d, J = 8.0 Hz, 1H), 7.66–7.60 (m, 2H), 7.55–7.42 (m, 1H)
(quinoxaline-H-5, H-6, H-7, H-8), 7.16 (s, 1H, pyrazole-H-4), 2.00 (s, 3H,
CH3).– EI MS (70 eV): m/z = 302 [M+].
4-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-1-methyl-[1,2,4]triazolo[4,3-a]-
quinoxaline (3a)
Recrystallisation from 1,4-dioxane/H2O (ca. 2:1) yielded 71% of light
beige needles, mp 283–289 °C (sublimation above 230 °C).– C13H9ClN6
(284.71) Anal. C,H,N.– IR (KBr): 3421, 3100 cm–1.– 1H-NMR
([D6]DMSO): δ = 14.22 (s, 1H, NH), 8.42–8.37 (m, 1H), 8.14–8.09 (m, 1H),
7.85–7.70 (m, 2H) (H-6, H-7, H-8, H-9), 7.60 (s, 1H, pyrazole-H-4), 3.14 (s,
3H, CH3).– EI MS (70 eV): m/z = 284 [M+].
2-Benzoylhydrazino-3-[3(5)-chloro-1H-pyrazol-5(3)-yl]-quinoxaline (7b)
Recrystallisation from tetrahydrofuran/ethyl acetate (ca. 1:1) yielded 80%
of a yellow powder, mp 222–225 °C.– C18H13ClN6O (364.80) Anal. C,H,N.–
IR (KBr): 3447, 3217, 1636 cm–1.– 1H-NMR ([D6]DMSO): δ = 14.35 (s, br,
1H, D2O-exchangeable, NH), 10.88 (s, 1H, D2O-exchangeable, NH), 10.22
(br, 1H, D2O-exchangeable, NH), 8.00–7.89 (m, 3H), 7.62–7.50 (m, 6H)
(quinoxaline-H-5, -H-6, -H-7, -H-8, phenyl-H), 7.22 (s, 1H, pyrazole-H-4).–
EI MS (70 eV): m/z = 364 [M+].
4-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-1-phenyl-[1,2,4]triazolo[4,3-a]-
quinoxaline (3b)
Recrystallisation from ethyl acetate yielded 69% of light pink needles, mp
267–270 °C.– C18H11ClN6 (346.78) Anal. C,H,N.– IR (KBr): 3447, 3144,
3074 cm–1.– 1H-NMR ([D6]DMSO): δ = 14.30 (s, 1H, NH), 8.13 (dd, J = 8.0
Hz, J = 1.6 Hz, 1H), 7.83–7.64 (m, 7H), 7.57–7.49 (m, 1H), 7.41–7.36 (m,
1H) (H-6, H-7, H-8, H-9, phenyl-H, pyrazole-H-4).– EI MS (70 eV): m/z =
346 [M+].
3-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-2-phenylacetylhydrazinoquinoxaline
(7c)
Recrystallisation from tetrahydrofuran yielded 49% of a yellow powder,
mp 225–231 °C.– C19H15ClN6O (378.82) Anal. C,H,N.– IR (KBr): 3213,
3040, 1637 cm–1.– 1H-NMR ([D6]DMSO + 1 drop of D2O): δ = 7.91 (d, J =
8.2 Hz, 1H), 7.70–7.63 (m, 2H), 7.55–7.19 (m, 7H), (quinoxaline-H-5, -H-6,
H-7, -H-8, phenyl-H, pyrazole-H-4), 3.64 (s, 2H, CH2).– EI MS (70 eV): m/z
= 378 [M+].
1-Benzyl-4-[3(5)-chloro-1H-pyrazol-5(3)-yl]-[1,2,4]triazolo[4,3-a]-
quinoxaline (3c)
Recrystallisation from tetrahydrofuran/ethyl acetate (ca. 2:1) yielded 62%
of light yellow needles, mp 268–270 °C.– C19H13ClN6 × 0.8 H2O (375.22)
Anal. C,H,N.– IR (KBr): 3108 cm–1.– 1H-NMR ([D6]DMSO): = 14.25 (s,
1H, NH), 8.19–8.08 (m, 2H), 7.73–7.65 (m, 3H), 7.36–7.23 (m, 5H) (H-6,
H-7, H-8, H-9, pyrazole-H-4, phenyl-H), 5.01 (s, 2H, CH2).– EI MS (70 eV):
m/z = 360 [M+].
3-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-2-(2-furoyl)-hydrazinoquinoxaline
(7d)
Recrystallisation from ethyl acetate/tetrahydrofuran (ca. 1:1) yielded 63%
of a yellow powder, mp 212–215 °C.– C16H11ClN6O2 × 0.7 H2O (367.37)
Arch. Pharm. Pharm. Med. Chem. 331, 163–169 (1998)

168
Matuszczak, Pekala, and Müller
4-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-1-(2-thienyl)-[1,2,4]triazolo[4,3-a]-
quinoxaline (3e)
diazepam (5 µM) was used to define non-specific binding. All compounds
were initially tested in a single concentration (10 µM). For compound 3a a
full inhibition curve was determined in triplicate. Ki values were calculated
from IC50 values, determined by the non-linear regression program Prism
version 1.0 (Graphpad, San Diego, California, USA), using the Cheng-
Recrystallisation from tetrahydrofuran yielded 63% of light beige needles,
mp 307–310 °C.– C16H9ClN6S (352.81) Anal. C,H,N.– IR (KBr): 3448,
1
3149 cm–1.– H-NMR ([D6]DMSO): δ = 14.29 (s, 1H, NH), 8.09 (dd, J =
Prusoff equation [14] and a KD value of 4 nM for diazepam [15]
.
5.1 Hz, J = 1.2 Hz, 1H, thiophene-H-5), 8.16–8.12 (m, 1H), 7.76–7.65 (m,
2H), 7.61–7.58 (m, 2H) (H-6, H-7, H-8, H-9, pyrazole-H-4), 7.72 (dd, J =
3.6 Hz, J = 1.2 Hz, 1H, thiophene-H-3), 7.43 (dd, J = 5.1 Hz, J = 3.6 Hz, 1H,
thiophene-H-4).– EI MS (70 eV): m/z = 352 [M+].
Adenosine Binding Assays
Inhibition of binding of [3H]2-chloro-N6-cyclopentyladenosine (CCPA)
[10] to A1-adenosine receptors of rat brain cortical membranes and inhibition
of [3H]3-(3-hydroxypropyl)-8-(3-methoxystyryl)-7-methyl-1-propargyl-
xanthine (MSX-2) [11] to adenosine receptors in rat brain striatal membranes
were assayed in analogy to published procedures [16]. As buffer Tris-HCl,
50 mM, pH 7.4 (at room temp.) was used for all experiments. The incubation
tubes for theA1-assay contained 25 µL of test compound dissolvedin DMSO,
or DMSO alone as a control, respectively, 0.775 mL of buffer, 100 µL of
radioligand solution in buffer to obtain a final concentration of 0.5 nM and
100 µL of membrane suspension (50 µg protein per tube) treated with
adenosine deaminase, to give a final volume of 1 mL.
4-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-1-(2-thienylmethyl)-[1,2,4]triazolo-
[4,3-a]quinoxaline (3f)
Recrystallisation from ethyl acetate/tetrahydrofuran (ca. 1:1) yielded 68%
of light yellow needles, mp 253–258 °C.– C17H11ClN6S (366.83) Anal.
C,H,N.– IR (KBr): 3448, 3124 cm–1.– 1H-NMR ([D6]DMSO): δ = 14.25 (s,
1H, NH), 8.29–8.22 (m, 1H), 8.16–8.08 (m, 1H), 7.46–7.68 (m, 2H) (H-6,
H-7, H-8, H-9), 7.65 (s, 1H, pyrazole-H-4), 7.43 (dd, J = 4.8 Hz, J = 1.5 Hz,
1H, thiophene-H-5), 6.99–6.92 (m, 2H, thiophene-H-3, -H-4), 5.21 (s, 2H,
CH2).– EI MS (70 eV): m/z = 366 [M+].
A2A-assay tubes contained 25 µL of compound dissolved in DMSO, or
DMSO as a control, respectively, 0.775 mL of buffer, 100 µL of radioligand
solution in buffer to obtain a final concentration of 1 nM and 100 µL of
membrane suspension (70 µg protein per tube) treated with adenosine de-
aminase, to give a final volume of 1 mL. 2-Chloroadenosine (50 µM) was
used to define nonspecific binding. DMSO concentration was 2.5% (V/V) in
all experiments. Incubation was performed at 23 °C for 1.5 h (A1-assay), or
at 23 °C for 30 mins. (A2A-assay). Incubation was terminated by rapid
filtration through glass fiber (Schleicher & Schüll GF51) filters using a cell
harvester. In order to minimise nonspecific binding of the radioligand
[3H]MSX-2 on the filters, those were soaked in aqueous polyethylenimine
solution (0.3%) at least 1 h before use. Filters were washed twice with 5 mL
of ice-cold buffer. The wet filter papers were incubated with scintillation
cocktail for at least 6 h before radioactivity was counted. Inhibition of the
receptor-radioligand binding was determined by a range of 5 to 6 concentra-
tions of the compounds. The Cheng-Prusoff equation [14] and KD of 0.2 nM
for [3H]CCPA [10] and 8 nM for [3H]MSX-2 [11] were used to calculate the
Ki values from the IC50 values, determined by the non-linear curve fitting
program PRISMTM (GraphPad, San Diego, California, USA).
4-[3(5)-Chloro-1H-pyrazol-5(3)-yl]-1-(2-furyl)-[1,2,4]triazolo[4,3-a]-
quinoxaline (3d)
Compound 7d (crude product) (0.250 g, 0.70 mmol) was suspended in a
mixture of dry N,N-dimethyl formamide and dry 1,4-dioxane (ca. 10 mL,
ratio ca. 2:1) and the mixture was heated until a solution was obtained. Then
the solution was cooled to room temperature and poured into ice water
(100 mL). The crystals thus obtained were filtered, washed with water and
light petroleum and dried in vacuo. Recrystallisation from ethyl acetate/tetra-
hydrofuran (ca. 3:1) yielded 0.160 g (68%) of light beige needles, mp
320–327 °C.– C16H9ClN6O (336.74) Anal. C,H,N.– IR (KBr): 3120 cm–1.–
1H-NMR ([D6]DMSO): δ = 14.30 (s, 1H, NH), 8.21–8.14 (m, 2H), 8.14–8.09
(m, 1H), 7.80–7.66 (m, 3H), 7.38–7.33 (m, 1H), (H-6, H-7, H-8, H-9,
pyrazole-H-4, furan-H-5), 7.29 (dd, J = 3.4 Hz, J = 0.8 Hz, 1H, furan-H-3),
6.94 (dd, J = 3.4 Hz, J = 1.8 Hz, 1H, furan-H-4).– EI MS (70 eV): m/z = 336
[M+].
Benzodiazepine and Adenosine Receptor Binding Assays
[3H]Diazepam (NET564, 3071 GBq/mmol, 83 Ci/mmol) and [3H]CCPA
(NET1026, 1110 GBq/mmol, 30 Ci/mmol) were obtained from New England
Nuclear (NEN) Life Science Products (Köln, Germany). The 7-desmethyl
precursor of [3H]MSX-2 was synthesised in our laboratory [11] and radiola-
belled by American Radiolabeled Chemicals Inc. (St. Louis, Minnesota,
USA) through Biotrend (Köln, Germany) (ART691, 3145 GBq/mmol,
85 Ci/mmol).
References and Notes
✩
Dedicated to Professor Dr. Gottfried Heinisch, Innsbruck, on the occa-
sion of his 60th birthday.
[1] G. Heinisch, B. Matuszczak, K. Mereiter, Heterocycles 1994, 38,
2081–2089.
Drug Solutions
[2] a) B. Matuszczak, K. Mereiter, Heterocycles 1997, 45, 2449–2462; b)
B. Matuszczak, T. Langer, K. Mereiter, J. Heterocycl. Chem. 1998, 35,
113–115.
The compounds were dissolved in DMSO and further diluted with tris(hy-
droxymethyl)aminomethane- (Tris-) HCl buffer (50 mM, pH 7.4) (final
DMSO concentrations: 2.5% for A1- and A2A-binding assays and 1% for
benzodiazepine-binding assays, respectively).
[3] a) J.E. Francis, W.D. Cash, B.S. Barbaz, P.S. Bernard, R.A. Lovell,
G.C. Mazzenga, R.C. Friedmann, J.L. Hyun, A.F. Braunwalder, P.S.
Loo, D.A. Bennett, J. Med. Chem. 1991, 34, 281–290; b) D. Catarzi, L.
Cecchi, V. Colotta, F. Melani, G. Filacchioni, C. Martini, L. Giusti, A.
Lucacchini, J. Med. Chem. 1994, 37, 2846–2850; c) W. Zhang, R. Liu,
Q. Huang, P. Zhang, K.F. Koehler, B. Harris, P. Skolnick, J.M. Cook,
Eur. J. Med. Chem. 1995, 30, 483–496; d) J.D. Albright, D.B. Moran,
W.B. Wright, Jr., J.B. Collins, B. Beer, A.S. Lippa, E.N. Greenblatt, J.
Med. Chem. 1981, 24, 592–600.
Benzodiazepine Binding Assays
Frozen rat brains were obtained from Pel-Freez, Rogers, Arkansas, USA.
The cortex was dissected and inhibition of binding of [3H]diazepam to rat
brain cortical membranes was determined as previously described [9, 13]. In
a final volume of 1 mL, each test tube contained 790 µL of Tris-HCl buffer
(50 mM, pH 7.4), 10 µL of drug solution (see above), 100 µL of rat cerebral
cortical membrane preparation with a protein concentration of ca. 100 µg per
tube, and 100 µL of [3H]diazepam solution, to give a final concentration of
1 nM. DMSO (final concentration: 1%) was necessary since the compounds
showed a rather low water-solubility. Higher concentrations of DMSO,
however, were not tolerated by the receptors. DMSO without test compound
served as a control.
[4] a) J.E. Francis, W.D. Cash, S. Psychoyos, G. Ghai, P. Wenk, R.C.
Friedmann, C. Atkins, V. Warren, P. Furness, J.L. Hyun, G.A. Stone,
M. Desai, M. Williams, J. Med. Chem. 1988, 31, 1014–1020; b) B.K.
Trivedi, R.F. Bruns, J. Med. Chem. 1988, 31, 1011–1014; c) R. Sarges,
H.R. Howard, R.G. Browne, L.A. Lebel, P.A. Seymour, B.K. Koe, J.
Med. Chem. 1990, 33, 2240–2254; d) V. Colotta, L. Cecchi, D. Catarzi,
G. Filacchioni, C. Martini, P. Tacchi, A. Lucacchini, Eur. J. Med.
Chem. 1995, 30, 133–139.
Incubations were performed at 2 °C for 1 h and were terminated by rapid
filtration through glass fiber filters (Schleicher & Schüll GF51) using a
Brandel cell harvester M-24 (Brandel, Gaithersburg, Maryland, USA). Three
5 mL washes with ice-cold Tris-HCl buffer were performed. Unlabelled
[5] a) T.A. Hamor, I.L. Martin in Progress in Medicinal Chemistry (Eds.:
G.P. Ellis, G.B. West), Elsevier, 1983, 20, 157–223; b) R.I. Fryer in
Arch. Pharm. Pharm. Med. Chem. 331, 163–169 (1998)

Substituted [1,2,4]triazolo[4,3-a]quinoxalines
169
Comprehensive Medicinal Chemistry (Eds.: C. Hansch, P.G. Sammes,
J.B. Taylor, Vol.-Ed.: J.C. Emmett), Pergamon Press,1990, 3, 539–566;
c) S.P. Gupta in Progress in Drug Research (Ed.: E. Jucker), Birk-
häuser, 1995, 45, 67–106.
[10] K.-N. Klotz, M.J. Lohse, U. Schwabe, G. Cristalli, S. Vittori, M.
Grifantini, Naunyn-Schmiedeberg’s Arch. Pharmacol. 1989, 340, 697–
683.
[11] C.E. Müller et al., unpublished results.
[6] E.J. Jacobsen, R.E. TenBrink, L.S. Stelzer, K.L. Belonga, D.B. Carter,
H.K. Im, W.B. Im, V.H. Sethy, A.H. Tang, P.F. von Voigtlander, J.D.
Petke, J. Med. Chem. 1996, 39, 158–175.
[12] a) M.J. Dooley, R.J. Quinn, J. Med. Chem. 1992, 35, 211–216; b) E.
van der Wenden, A.P. IJzerman, W. Soudijn, J. Med. Chem. 1992, 35,
629–635; c) M.J. Dooley, K. Motomichi, F. Suzuki, Bioorg. Med.
Chem. 1996, 4, 923–934; d) C.E Müller, U. Geis, B. Grahner, W.
Lanzner, K. Eger, J. Med. Chem. 1996, 39, 2482–2491.
[7] a) P.J.M. van Galen, G.L. Stiles, G. Michaels, K.A. Jacobsen, Med. Res.
Rev. 1992, 12, 423–471 and literature cited therein; b) C.E: Müller, T.
Scior, Pharm. Acta Helv. 1993, 68, 77–111;
[13] U. Geis, B. Grahner, M. Pawlowski, A. Drabczynska, M. Gorczyca,
C.E. Müller, Pharmazie 1995, 50, 333–336.
[8] a) C.E. Müller, B. Stein, Curr. Pharm. Design 1996, 2, 501–530 and
literature cited therein; b) C.E. Müller, Exp. Opin. Ther. Patents 1997,
7, 419–440 and literature cited therein; c) P.G. Baraldi, B. Cacciari, G.
Spalluto, A. Borioni, M. Viziano, S. Dionisotti, E. Ongini, Current
Med. Chem. 1995, 2, 707–722.
[14] Y. Cheng, W. Prusoff, Biochem. Pharmacol. 1973, 22, 3099–3108.
[15] a) H. Möhler, T. Okada, Life Sci. 1977, 20, 2101–2110; b) R.F. Squires,
C. Braestrup, Nature 1977, 266, 732–734.
[16] C.E. Müller, R. Sauer, U. Geis, F. Frobenius, P. Talik, M. Pawlowski,
Arch. Pharm. Pharm. Med. Chem. 1997, 330, 181–189.
[9] U. Geis, K. Kiec-Kononowicz, C.E. Müller, Sci. Pharm. 1996, 64,
383–390.
Received: December 15, 1997 [FP268]
Arch. Pharm. Pharm. Med. Chem. 331, 163–169 (1998)