In vitro and in vivo studies of 6,8-(diaryl)imidazo[1,2-a]pyrazin-3(7H)-ones as new antioxidants

Article abstract of DOI:10.1016/j.bmc.2009.05.025

A series of 5-aryl and 3,5-diaryl-2-amino-1,4-pyrazines and the derived imidazopyrazinones has been synthesized to study the chemical oxidative degradation of the bicyclic systems in vitro. Imidazopyrazinones mainly degraded following two independent pathways producing their precursors, namely aminopyrazines, and the corresponding amidopyrazines, respectively. Despite the fact that there is no influence of the substituent of the 3-aryl group on the ratio of the products aminopyrazine/amidopyrazine, diarylimidazopyrazinones and diarylaminopyrazines are good antioxidants in vivo. They protected against microvascular damages in ischemia/reperfusion with similar efficiencies.

Full text of DOI:10.1016/j.bmc.2009.05.025

Bioorganic & Medicinal Chemistry 17 (2009) 4336–4344
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
In vitro and in vivo studies of 6,8-(diaryl)imidazo[1,2-a]pyrazin-3(7H)-ones
as new antioxidants
Frederic De Wael a, Paul Jeanjot a,z, Cédric Moens a,b, Tony Verbeuren c, Alex Cordi c, Eliete Bouskela d,
Jean-François Rees b, Jacqueline Marchand-Brynaert a,
*
a Unité de Chimie Organique et Médicinale, Université catholique de Louvain, Bâtiment Lavoisier, Place Louis Pasteur 1, B-1348 Louvain-la-Neuve, Belgium
b Institut des Sciences de la vie, Université catholique de Louvain, Bâtiment Carnoy, Place Croix du Sud 5, B-1348 Louvain-la-Neuve, Belgium
c Institut de Recherches Servier, Rue des Moulineaux 11, F-92150 Suresnes, France
d Laboratorio de Pesquisas em Microcirculaçao, Universidade do estado do Rio de Janeiro, Rua Sao Francisco Xavier 524, 20550-013 Rio de Janeiro, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 January 2009
Revised 6 May 2009
Accepted 11 May 2009
Available online 15 May 2009
A series of 5-aryl and 3,5-diaryl-2-amino-1,4-pyrazines and the derived imidazopyrazinones has been
synthesized to study the chemical oxidative degradation of the bicyclic systems in vitro. Imidazopyraz-
inones mainly degraded following two independent pathways producing their precursors, namely
aminopyrazines, and the corresponding amidopyrazines, respectively. Despite the fact that there is no
influence of the substituent of the 3-aryl group on the ratio of the products aminopyrazine/amidopyr-
azine, diarylimidazopyrazinones and diarylaminopyrazines are good antioxidants in vivo. They protected
against microvascular damages in ischemia/reperfusion with similar efficiencies.
Keywords:
Imidazopyrazinone
Aminopyrazine
Antioxidant
Ó 2009 Elsevier Ltd. All rights reserved.
Ischemia/reperfusion
1. Introduction
Coelenterazine 1a (Scheme 1) and more generally imidazo[1,2-
a]pyrazin-3(7H)-ones (imidazopyrazinones) are bioluminescent
Reactive oxygen species (ROS) and reactive nitrogen species
(RNS) are produced in aerobic organisms as part of the normal
physiological and metabolic processes. They are very important
mediators of cell injury or death due to the damages they can in-
flict if they are produced in excessive concentrations or in wrong
locations. The damages that ROS/RNS cause, essentially on biolog-
ical macromolecules (membrane lipids, proteins, nucleic acids, . . .),
are directly or indirectly implicated in the pathogenesis of various
disorders such as cardiovascular diseases, reperfusion injury, Alz-
heimer’s and other neurodegenerative diseases, cancer develop-
ment and progression, inflammation as well as in the aging
process.1–9
Therefore the interest for the protective role of antioxidants in
medicine has been growing over the last 15 years. Antioxidants
are considered as potential drugs due to their ability to reduce or
inhibit the free radical reactions initiated by ROS/RNS. Currently
available radical scavengers are structurally related to natural anti-
oxidants (vitamin A/b-carotene, vitamin C, vitamin E, green tea
extracts, flavonoids, . . .) and to industrial compounds such as
highly hindered ortho-substituted phenols.10–14
substrates of luciferases,15–17 naturally designed to react with oxy-
gen in light-producing reactions. Numerous studies have been
devoted to the oxidation mechanism of coelenterazine and related
derivatives 1, essentially in the context of colored light production
via an efficient biochemical process and the development of ana-
lytical tools in biochemistry.18–23 A few years ago, our research
group considered coelenterazine and other imidazopyrazinones 1
as potential leads in the discovery of novel antioxidants for thera-
peutic use.24 We showed that imidazopyrazinones have good anti-
oxidant properties as they are highly reactive with radical anion
superoxide, lipid radicals, nitrofurantoin-derived radicals and per-
oxynitrite; LDL protection by compounds 1 has been further dem-
onstrated in vitro. Finally, cellular protection against oxidative
stress and UV-irradiation damages has been observed, in the pres-
ence of compounds 1, on various cell lines (human keratinocytes,
rat hepatocytes, rat neuronal cells, fish erythrocytes).24 During
these previous studies, we found that the chemical precursors of
imidazopyrazinones synthesis, namely the 2-amino-1,4-pyrazine
derivatives 3, are also endowed with remarkable antioxidant prop-
erties providing R3 is a para-phenol (or a catechol) moiety.25,26
The main metabolite of luciferase-catalyzed oxidation of
coelenterazine (1a) is coelenteramide (2a) which is initially pro-
duced in the excited state; this species emits light during decay
(Scheme 1).27,28 Similarly, numerous synthetic analogs 1 display
* Corresponding author. Tel.: +32 10 472740; fax: +32 10 474168.
E-mail address: jacqueline.marchand@uclouvain.be (J. Marchand-Brynaert).
Accidentally deceased on January 19, 2004.
z
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.05.025

F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
4337
Scheme 1. Oxidative degradation of imidazopyrazinones 1.
bioluminescence and chemiluminescence properties under en-
zyme (luciferase) processing or chemical oxidation conditions.15
The so-produced amides 2, formally derived from aminopyrazines
3, were however inactive in antioxidant tests.24,25
In the course of chemical and air oxidation studies of coelenter-
azine (1a), we observed the formation of coelenteramine (3a), in
quite significant amounts, together with the expected coelentera-
mide (2a). This interesting observation was general: the oxidative
degradation of imidazopyrazinones 1 led always to both amides 2
and aminopyrazines 3. Thus besides the luminescence pathway of
coelenterazine (1) decomposition giving an inactive compound (2),
another route can take place which produces in situ a daughter-
antioxidant (3) (Scheme 1).
Aiming to optimize this unique antioxidant system acting in
cascade (3 prolonging the activity of its mother-compound 1), we
studied the outcome of the five-membered ring R2 substituent
and the R1 substituents effect on the 2:3 product ratios formed
when processing imidazopyrazinones 1 under aerobic conditions
in the laboratory. Here we report the HPLC analysis of a model
reaction (R1 = H, R2 = Ph), the synthesis of novel derivatives 1
(R2 = Me) with R1 being various para-substituted phenyl groups,
their in vitro oxidative decomposition pathways and their in vivo
biological activity in the ‘hamster cheek pouch’ assay.
Although the intermediates in the luminescence reactions of
imidazopyrazinones (1) have not yet been firmly established, it is
generally accepted that the emission of light originates from the
thermal decomposition of a dioxetanone species. The anion of 1
reacts (more easily than the neutral substrate) with triplet molec-
ular oxygen to give the anion peroxide intermediates A or B. Both
can conduct to the key-dioxetanone intermediate C which decom-
position, with the loss of carbon dioxide, generates a singlet-ex-
cited state of amidopyrazine anion D. Amide 2 is then produced
via protonation of D and emission of light.15 Hydrolysis of 2 into
aminopyrazine 3 is not likely under the smooth conditions of air
oxidation (protic or aprotic polar solvent, added with a weak
organic base or an aqueous buffer). Therefore, we hypothesized
that aminopyrazine 3 could be formed via a parallel route to the
luminescence pathway, involving the hydrolysis of the protonated
intermediates B or C, namely E and F, by water nucleophilic attack
on C2 followed by C2-N1 bond cleavage, the aminopyrazinyl moi-
ety being a good leaving group. Such a mechanism would simulta-
neously produce the acid 4 bearing the R2 substituent. The possible
occurrence of this alternative route of oxidative decomposition has
been examined on a simple imidazopyrazinone 1 with a view to
easily identify and quantify the three products 2, 3 and 431 by using
HPLC as analytical method.
2. Results and discussion
2.2. HPLC study of a model reaction
Imidazopyrazinone 1b (R1= H, R2= Ph, R3= 4-OH-Ph) was se-
lected as model compound for the study of chemical oxidative deg-
radation, indeed: (i) 1b features the minimum pharmacophoric
pattern for antioxidant activity26 and is readily accessible; (ii) the
reference products 2b and 3b are also easily prepared; (iii) the acid
4 that should be produced is benzoic acid, a commercial reference
visible in HPLC with UV detection; (iv) a structurally related mol-
ecule, stable in oxidative conditions (because the phenol moiety
is protected), namely aminopyrazine 3c, can be used as internal
standard (Scheme 1).
2.1. Possible pathways involved in imidazopyrazinone
oxidation (Scheme 1)
Results from the literature suggest that bioluminescence and
chemiluminescence of imidazopyrazinones (1) might involve a
similar mechanism for the production of amides (2) and light.15
However, the formation of aminopyrazines (3) is occasionally
mentioned.29,30 The main questions concern the outcome of the
imidazole moiety (which bears the substituent R2) from precursors
1 and the production of 2 and 3 via parallel routes or via consec-
utive reactions (i.e., formation of 2 and subsequent transformation
into 3).
The synthesis of 1b, 3b and 3c was performed as previously
described.32–35 Amide 2b was independently prepared from 2-ami-

4338
F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
no-5-[p-(t-butyldimethylsilyloxy)-phenyl]-1,4-pyrazine 3d (see
Scheme 2),36 by benzoylation (benzoyl chloride, pyridine, reflux)
followed by deprotection of the phenol moiety (Bu4NF, THF,
20 °C). For HPLC analysis all compounds and benzoic acid (4) were
dissolved in acetonitrile (ACN)–EtOH (96:4, v/v) at 8 ꢀ 10ꢁ5 M final
concentration. We used a C18 column and a gradient of ACN–H2O
(see Section 4) for the elution; benzoic acid is visible at 220 nm,
while the other components are more conveniently detected at
280 nm. The following retention times were recorded: 4,
RT = 2.1 min; 3b, RT = 14 min; 1b, RT = 16 min; 3c, RT = 17.8 min;
2b, RT = 19.5 min.
A 1:1 mixture of imidazopyrazinone 1b and standard 3c
(8 ꢀ 10ꢁ5 M) in 96:4 ACN–EtOH solution was placed in the dark,
at 37 °C and controlled by HPLC, each hour, during 18 h. Air-oxida-
tion was a very slow process37: 80% of 1b is still present and only
20% has been transformed into amide 2b (<5%), aminopyrazine 3b
(9%) and other unidentified derivatives. In the presence of azo-bis-
isobutyronitrile (AIBN, 4 ꢀ 10ꢁ3 M), oxidation of 1b at 37 °C was
complete within 18 h, leading to 2b (26%) and 3b (44%) as the main
products (Fig. 1). Concentration evolution as a function of time,
depicted in Figure 1, shows clearly the simultaneous appearance
of 2b and 3b; amide 2b is not the precursor of aminopyrazine 3b
(this has been also controlled with pure 2b placed in the same
experimental conditions).
Figure 1. Evolution of concentrations during air-oxidation of imidazopyrazinone
1b at 37 °C in the presence of AIBN in large excess.
Air-oxidation of 1b (8 ꢀ 10ꢁ5 M) was further examined in a
micellar solution of linoleic acid (1.6 ꢀ 10ꢁ4 M, phosphate buffer
pH 7.4 ACN–EtOH; 85:12.5:2.5, v/v/v). In this case, the disappear-
ance of 1b was rapid (less than 6 h), and the concomitant forma-
tion of amide 2b and amine 3b was observed as above. The final
ratio of oxidation products 2b/3b was 66%/30%. Benzoic acid was
detected in similar amount as 3b (4: 27%, Fig. 2). Addition of AIBN
(4 ꢀ 10ꢁ3 M) did not change dramatically the reaction profile: 1b is
totally oxidized within 5 h, at 37 °C in the dark as previously, to
furnish 2b, 3b and 4 in, respectively, 57%, 40% and 33%. Differences
were visible after several hours, when monitoring the products sta-
bility during one day: 2b (amide) and 4 (benzoic acid) are stable in
both conditions; 3b (aminopyrazine) is stable in the absence of
AIBN, but is gradually consumed in its presence. This observation
is consistent with the capacity of aminopyrazine derivatives to
inhibit lipid peroxidation.38
Figure 2. Evolution of concentrations during air-oxidation of imidazopyrazinone
1b at 37 °C in the presence of linoleic acid micelles.
degradation (Scheme 1), leading respectively to amidopyrazine
2b and aminopyrazine 3b accompanied by a carboxylic acid 4 bear-
ing the R2 substituent. Depending on the experimental conditions,
the reaction was (very) slow (P18 h) or rapid (65 h), and the cor-
responding product ratios 2b/3b were in favor of aminopyrazine
3b (26/44) or amidopyrazine 2b (66/30). This is in agreement with
previous reports showing that chemiluminescence efficiency of 1 is
This HPLC study of in vitro air-oxidation of imidazopyrazinone
1b has pointed out the occurrence of two independent routes of
Scheme 2. Synthesis of aminopyrazines 6 and imidazopyrazinones 7.

F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
4339
9-C
Table 2
strongly enhanced in micelle solutions of surfactants or in the
presence of cyclodextrins mimicking the hydrophobic active site
Typical 1H and 13C NMR data (d) of 7 (atom numbering of Scheme 2)
of luciferase.39–41
Entry
5-H
2-Me
2-C
3-C
5-C
6-C
8-C
aa
ba
cb
db
eb
fb
8.74
8.67
8.40
8.48
8.55
8.50
8.48
2.09
2.46
2.54
2.56
2.54
2.54
2.42
125.2
123.9
Not visible
122.8
119.8
116.8
nd
125.5
124.7
139.0
139.3
138.8
135.6
nd
109.6
107.9
109.2
109.7
110.1
106.6
nd
133.8
126.4
142.0
141.5
143.5
140.2
nd
144.1
143.9
145.5
144.7
144.5
142.0
nd
136.8
136.9
126.X
126.8
126.5
123.4
nd
2.3. Synthesis of novel imidazopyrazinones (Scheme 2)
We had already established24,25 that the antioxidant properties
of imidazopyrazinones 1 is linked to the hetero-bicyclic core itself
and not so much to the decorating substituents. On the other hand,
the activity of 2-aminopyrazines 3 is mainly due to their para-sub-
stituent (phenol/catechol moiety), the ortho-substituent appearing
less crucial. Accordingly, we decided to prepare a series of novel
imidazopyrazinones 7 (Scheme 2) derived from 2-amino-3,5-dia-
ryl-1,4-pyrazines 6 (introduction of R1 substituents with different
electronic effects) and methyl glyoxal (introduction of the smallest
R2 substituent), in view to possibly promote the route of oxidative
degradation leading to the ‘cascade effect’, namely the production
of aminopyrazines.
We exploited a strategy set up by our group for the preparation
of unsymmetrical 2-amino-3,5-diaryl-1,4-pyrazines, based on two
successive Suzuki coupling reactions.35 TBS protected 2-amino-3-
bromo-5-[4-(t-butyl-dimethyl-silyloxy)phenyl]-1,4-pyrazine (5)
was obtained by monobromination in position 5 of commercially
available 2-amino-1,4-pyrazine, then Suzuki coupling to introduce
the protected p-phenol moiety in position 5 giving 3d and finally
bromination of position 3. Reaction of 3,5-dibromo-2-amino-1,4-
pyrazine with one equivalent of p-(t-butyl-dimethyl-silyl-
oxy)phenyl boronic acid in the presence of Pd-catalyst furnished
5 in low yield (mixture of mono- and di-substituted products
and starting material).42
The target-compounds 7 were readily obtained from 5 after Su-
zuki coupling in position 3 with the required para-substituted
phenylboronic acids, deprotection of t-butyldimethylsilyl ether
and condensation with methylglyoxal40 in a mixture of ethanol
and aqueous HCl at reflux. With this sequence of reactions, we
had in hands seven couples of 3,5-disubstituted-2-aminopyrazines
6 and imidazopyrazinones 7 of which five were not previously de-
scribed (i.e., couples c, d, e, f, g).34,38
Compounds 6 and 7 were characterized as usual (see Section 4).
Tables 1 and 2 summarize typical NMR values, showing only small
differences between the series of compounds 6 and 7, respectively,
that could be attributed to the electron-donating/withdrawing
character of the R substituents, as well as to a solvent effect when
the spectra are recorded in different solvents. Carbon atoms 2 and
3 of compounds 7 could not be attributed precisely due to the lack
of correlation with 1H and the occurrence of enol tautomeric forms
(see Scheme 2): C-3 is highly downfield shifted comparatively to
‘normal’ carbonyl-type carbons.
ga
a,b
See Table 1; nd: not determined.
experimental conditions were set up (aerated solution of dichlo-
romethane (DCM) and wet isopropanol (iPrOH) containing dieth-
ylamine (DEA) as base for accelerating the oxidation), taking into
account both the solubility and the chromatographic resolution of
all the members a–g, for the three series of compounds, namely
the precursors 7, and the oxidation products 8 (amides) and 6
(amines) (Scheme 2), the concomitantly formed acetic acid (4
with R2 = Me) being not detectable by HPLC.43 The reference
acetamides 8 were obtained according to Jeanjot et al.36 The se-
lected internal standard, stable under air-oxidation conditions,
was 3,5-diphenyl-2-amino-1,4-pyrazine.42
For HPLC analysis, all compounds (7, 6 and 8) were dissolved in
CH2Cl2/isopropanol (90:10, v/v) with 0.1% diethylamine (DEA), at
10ꢁ3 M final concentration. We used laboratory packed silica col-
umns and different types of isocratic elution (CH2Cl2/isopropa-
nol/DEA, A 90:10:0.1, B 95:5:0.1 or C 96:4:0.1). All compounds
were detected at 280 nm. The retention times (in min) of the prod-
ucts are given in Table 3.
Calibration curves were established to further quantify the
amounts of products formed by aerobic oxidation of the various
precursors 7. A typical experiment is described below: a mixture
of imidazopyrazinone 7a (10ꢁ3 M) and internal standard (5 ꢀ 10
ꢁ4 M) in 90:10:0.1 CH2Cl2/isopropanol/DEA solution was intro-
duced in a hermetic vial and placed under white light (to accel-
erate the oxidation) at 23 °C. After 19 h, 7a was totally
decomposed (disappearance of 7a on TLC) and transformed into
6a (66%), 8a (15%) and other unidentified derivatives. Both the
internal standard and 8a were stable in those conditions, but
6a slowly degraded with time. The concentration of 6a dimin-
ished of about ꢂ50% after another 24 h. The instability under
oxidative conditions of aminopyrazines was already observed
with 3b (see above).
All the results of imidazopyrazinones (7) oxidative degradation
are collected in Table 3. This HPLC study is illustrated with a typical
chromatogram shown in Figure 3. The series of 6,8-diarylimi-
dazo[1,2-a]pyrazin-3(7H)-ones produced aminopyrazines (6) as
major products and the corresponding amides (8) as minor prod-
ucts. The experimental 6:8 ratios were within 3.0 and 4.8; no cor-
relation with the electronic effect of R substituents could be drawn.
2.4. Air-oxidation of imidazopyrazinones 7
The effect of R substituents on the degradation pathways of
imidazopyrazinones 7a–g has been examined by HPLC. Common
Table 1
Table 3
Typical 1H and 13C NMR data (d) of 6 (atom numbering of Scheme 2)
HPLC results–products ratio at the end of the oxidation
Entry
6-H
2-C
3-C
5-C
6-C
Entry
6
8
6:8
Elution mode
ac
ba
ca
db
ea
fb
8.46
8.35
8.40
8.20
8.49
8.31
8.38
152.3
151.1
152.5
152.4
152.4
152.5
152.5
139.2
138.0
143.9
141.8
141.8
139.9
143.9
149.2
140.3
143.8
143.6
140.6
143.8
143.8
139.4
135.8
138.5
136.1
137.8
137.8
138.5
tR
6.2
7.4
15.5
12.3
11.7
13.5
6.8
%
tR
9.3
10.7
31.1
20.2
19.0
21.7
10.4
%
a
b
c
d
e
f
66
77
71
68
78
76
67
15
16
21
18
16
17
22
4.4
4.8
3.4
3.8
4.8
4.5
3.0
A
A
C
B
B
B
A
ga
a
Spectra recorded in DMSO-d6.
Spectra recorded in methanol-d4.
Spectra recorded in acetone-d6.
g
b
c
tR internal standard = 3.6, 6.1, 7.8 min elution mode A, B, C, respectively.

4340
F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
Table 4
2.5. In vivo activity
Results of ‘hamster cheek pouch’ assay
The protective effect of compounds 6 and 7 against microvascu-
lar damages in ischemia/reperfusion was assayed in vivo. The eval-
uation was realized with the experimental model of ‘hamster
cheek pouch’44 allowing quantitative studies of macromolecular
permeability by direct observation on microscope. Fluorescent-la-
beled dextran was injected intravenously and changes in the num-
ber of microvascular leaky sites were measured after local
ischemia/reperfusion. Animals were treated by gavage with the
tested compounds at 3 mg/kg, 30 min before anesthesia. Results
are given in percentages of inhibition of leaky sites, determined
30 min after the start of reperfusion. In this test, Apocynin (a flavo-
noid derivative used as reference) gave about 40% inhibition at
3 mg/kg.
Results collected in Table 4 showed that all compounds pro-
vided a moderate to good protection against the increase of micro-
vascular permeability due to ischemia/reperfusion. However, the
mother-antioxidants 7 were not systematically more active than
the corresponding daughter-antioxidant 6: in two cases, 7 was
more potent than 6 (b, R = OH; c, R = OMe), in two other cases, 7
was less potent than 6 (d, R = NMe2; f, R = F) and in two last cases,
7 and 6 were almost equally active (a, R = H; e, R = CF3). Disap-
pointingly, the ‘cascade effect’, disclosed by the HPLC studies under
laboratory oxidation conditions, is not a relevant guide principle
for optimizing the biological activity of imidazopyrazinones.
Indeed, the in vivo test takes into account not only the intrinsic
antioxidant activity of the couples 7/6, but also the oral bioavail-
ability, metabolic stability and distribution of the tested com-
pounds in the specific tissue where the efficacy is measured.
6
% Inhibition
3 mg/kg
7
% Inhibition
3 mg/kg
a
b
c
d
e
f
18
29
10
50
14
42
a
b
c
d
e
f
12
40
44
21
17
34
3. Conclusion
Amongst the various strategies of drug discovery, we became
interested in the ‘non-natural natural products’ approach which,
after about 25 years of ‘combinatorial chemistry technologies’, is
nowadays reconsidered as very promising by distinguished scien-
tists in the field.45
Coelenterazine (a marine luciferine, 1a) was our natural ‘lead’
compound for the discovery of potential antioxidants, due to its
intrinsic capacity of quenching oxygen and ROS/RNS. Two pharma-
cophores were identified from our previous studies, namely the
bicyclic core of 1a (imidazo[1,2-a]pyrazin-3(7H)-one) and 2-ami-
no-5-(p-hydroxyphenyl)-1,4-pyrazine (3b). In this paper, we have
experimentally proved that an aminopyrazine is one of the oxida-
tion products of the imidazopyrazinone, formed via a parallel route
to the luminescence process. However, the so-called ‘cascade
effect’ (i.e., a mother-antioxidant decomposes into a daughter-anti-
oxidant), could not be optimized by varying the electronic proper-
Figure 3. HPLC analysis of oxidative degradation of imidazopyrazinone 7e after 4 h with elution mode A.

F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
4341
ties of the C8 substituent of 1 (=C3 substituent in 3); our idea was
4.2. General procedure A for the synthesis of 3,5-diaryl-2-
aminopyrazines
to improve the leaving group capacity of the aminopyrazinyl moi-
ety by increasing the electron-withdrawing character of R1 (see
Scheme 1). Nevertheless, in protic media, the series of non-natural
imidazopyrazinones 7a–g furnished always the corresponding
aminopyrazines 6a–g as the major products of aerobic degradation.
We have further confirmed the interest of both families 7 and 6
in drug discovery by a representative in vivo animal assay, namely
the protection against ischemia/reperfusion injury. Here again, no
particular benefit due to the ‘antioxidant cascade’, nor substituent
effect, could be demonstrated; the most active compounds were
imidazopyrazinones 7b, 7c and aminopyrazines 6d, 6f, respec-
tively, bearing different R1 substituents.
The oxidative release of benzoic acid (see Scheme 1, 4 with
R2 = Ph) we demonstrated by HPLC studies, suggests that imidazo-
pyrazinone derivatives could be developed not only for their
intrinsic antioxidant properties, but also as carriers to deliver
acidic drugs on their site of action. This novel type of potential
dual-action drug or pro-drug is currently under investigation.46
1,4-Bis(diphenylphosphino)butane (6 mol %) was added to a
suspension of bis(benzonitrile)dichloro palladium (5 mol %) in
dry toluene (3 mL/mmol) and was stirred for 30 min under argon
atmosphere at room temperature. 2-Amino-3-bromo-5-(p-tbutyl-
dimethylsilyloxyphenyl)-1,4-pyrazine (1 equiv), arylboronic acid
(1.2 equiv), ethanol (0.4 mL/mmol), 1 M Na2CO3 (1 equiv) and tol-
uene (3 mL/mmol) were added successively and the mixture was
heated to reflux overnight. The mixture was cooled to room tem-
perature, diluted with water and extracted with ethyl acetate
(3ꢀ). The organic layers were combined, washed with brine and
dried over MgSO4, filtered and concentrated under vacuum.
The residue was dissolved in THF (10 mL/mmol) and tetrabutyl-
ammonium fluoride (1 M in THF, 2.5 equiv) was added dropwise at
0 °C. The mixture is stirred 10 min at 0 °C and another 3 h at room
temperature. Water (1 mL/mmol) was added and the mixture was
extracted with ethyl acetate (1 mL/mmol, 4ꢀ). The organic layers
were combined, washed with brine and dried over MgSO4, filtered
and concentrated under vacuum. The residue was purified by flash
chromatography.
4. Experimental
4.1. Chemistry: general methods
4.2.1. 2-Amino-3-(4-methoxyphenyl)-5-(4-hydroxyphenyl)-1,4-
pyrazine (6c)
1H (300 MHz) and 13C (75 MHz) NMR spectra were recorded on
Compound 6c was prepared according to the general procedure
A using 1,4-Bis(diphenylphosphino)butane (33 mg, 0.077 mmol,
6 mol %), bis(benzonitrile)dichloro palladium (23 mg, 0.059 mmol,
5 mol %), dry toluene (2 ꢀ 20 mL), 5 (425 mg, 1.11 mmol, 1 equiv),
corresponding arylboronic acid (204 mg, 1.34 mmol, 1.2 equiv),
a
Varian Gemini-300 spectrometer. 1H (500 MHz) and 13C
(125 MHz) NMR spectra were recorded on a BRUCKER AM-500
spectrometer. 19F (282 MHz) NMR spectra were recorded on a
BRUCKER AVANCE-300 spectrometer. The attributions were estab-
lished by selective decoupling experiments. Chemical shifts are
reported as d values (in ppm) downfield from TMS for 1H and 13C
or from CFCl3 for 19F. The mass spectra (FAB or APCI modes) were
obtained using a Finnigan-MAT TSQ-700 instrument. Elemental
analyses were obtained at the University College of London (UK).
HRMS analysis by the ESI method was performed at the University
of Mons-Hainaut (Belgium). UV spectra were recorded on a UV–
vis–NIR Varian-Cary spectrophotometer (k given in nm). IR spectra
were determined using a FT Biorad FTS 135 apparatus. Thin-layer
chromatography was carried out on Silica Gel 60 plates F254
(Merck, 0.2 mm thick); product visualization was effected with
UV light (k = 254 nm). For flash chromatography we used Merck
Silica Gel 60 of 230–400 mesh ASTM. Melting points (uncorrected)
were determined on an Electrothermal apparatus.
Anhydrous solvents used were of P.A. quality grade and freshly
distilled [(CH2Cl2 and NEt3 from CaH2) and (THF, ether, benzene
and toluene from Na and benzophenone)]. All reagents were
obtained from Aldrich-Fluka, Acros or ROCC.
HPLC analysis of 1b and degradation products was carried out
on Thermoseparation system (Thermoquest, Belgium). The system
consisted of a P2000 binary pump plus a temperature-controlled
AS1000 autosampler and a UV3000 UV–vis detector. All analyses
ethanol (350 lL), 1 M Na2CO3 (1.1 mL, 1 equiv), THF (6 mL) and tet-
rabutylammonium fluoride (2.8 mL, 2.5 equiv); Column chroma-
tography using ether/cyclohexane: 3/1; 249 mg of 6c as pale
yellow solid; 85% yield; Rf 0.6 (ether/ethyl acetate: 6/4); 1H NMR
(300 MHz, DMSO-d6) d 3.84 (s, 3H, 15-H), 6.06 (br s, 2H, NH2),
3
3
6.82 (d, J9,8 = 8.7 Hz, 2H, 9-H), 7.07 (d, J13,12 = 8.7 Hz, 2H, 13-H),
3
3
7.75 (d, J8,9 = 8.7 Hz, 2H, 8-H), 7.80 (d, J12,13 = 8.7 Hz, 2H, 12-H),
8.40 (s, 1H, 6-H), 9.6 (br s, OH). 13C NMR (75 MHz, DMSO-d6) d
26.8 (16-C), 116.6 (9-C), 128.0 (8-C), 129.8 (12-C), 129.9 (13-C),
130.1 (7-C), 138.3 (14-C), 138.5 (6-C), 139.2 (11-C), 143.8 (5-C),
143.9 (3-C), 152.5 (2-C), 159.1 (10-C); MS (FAB+) C17H15N3O2
(MW = 293.33) m/z 294 ((M+H)+), 278 ((MꢁCH3)+); Anal. Calcd for
C17H15N3O2ꢃ0.3H2O (298.52): C, 68.77; H, 5.22; N, 14.16. Found:
C, 68.85; H, 5.41; N, 13.52. UV–vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
365 (0.93), 327 (0.50), 285 (2.5), 251 (1.8); mp 222–224 °C.
(e)):
4.2.2. 2-Amino-3-(4-(N,N)-dimethylaminophenyl)-5-(4-
hydroxyphenyl)-1,4-pyrazine (6d)
Compound 6d was prepared according to the general procedure
A using 1,4-bis(diphenylphosphino)butane (27 mg, 0.062 mmol,
6 mol %), bis(benzonitrile)dichloro palladium (20 mg, 0.051 mmol,
5 mol %), dry toluene (2 ꢀ 10 mL), 5 (290 mg, 1.04 mmol, 1 equiv),
corresponding arylboronic acid (198 mg, 1.2 mmol, 1.16 equiv),
were performed on a 250 ꢀ 4.6 mm 5
equipped with a precolumn cartridge, maintained at a temperature
of 35 °C. The volume of full-loop injections is 10 L. Solutions are
lm C18 column (Alltech)
ethanol (400 lL), 1 M Na2CO3 (1 mL, 1 equiv), THF (6 mL) and tet-
l
rabutylammonium fluoride (2.8 mL, 2.5 equiv); column chroma-
tography using cyclohexane/ethyl acetate 5/3; 261 mg of 6d as
yellow green solid; 82% yield; Rf 0.31 (cyclohexane/ethyl acetate:
5/4); 1H NMR (500 MHz, CD3OD) d 3.02 (s, 6H, 15-H), 6.85 (d,
made with solvents of HPLC quality grade for organic solutions
and millipore water for aqueous ones.
HPLC system used for the analysis of 7a–g and degradation
products consisted of a Beckman 126P pump connected to Beck-
man 168 PDA detector (chromatograms were extracted at
k = 280 nm). The HPLC column length was 250 mm ꢀ 4.6 mm
3
3J9,8 = 8.8 Hz, 2H, 9-H), 6.89 (d, J13,12 = 8.8 Hz, 2H, 13-H), 7.68 (d,
3
3J12,13 = 8.8 Hz, 2H, 12-H), 7.76 (d, J8,9 = 8.8 Hz, 2H, 8-H), 8.20 (s,
1H, 6-H). 13C NMR (125 MHz, CD3OD) d 40.6 (15-C), 113.4 (9-C),
116.5 (13-C), 126.3 (11-C), 128.1 (8-C), 130.1 (7-C), 130.3 (12-C),
136.1 (6-C), 141.8 (3-C), 143.6 (5-C), 152.4 (2-C), 152.6 (14-C),
158.9 (10-C); MS (FAB+) C18H18N4O (MW = 306.37) m/z 307.3
((M+H)+). Anal. Calcd for C18H18N4O: C, 70.57; H, 5.92; N, 18.29.
packed in the laboratory with Silica 60, 15–40 lm, irregular.
Organic solutions were made with solvents of HPLC quality grade
without drying.
Compounds 6a, 6b, 7a and 7b were obtained according to Refs.
34 and 36.

4342
F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
Found: C, 70.60; H, 6.08; N, 17.95. UV–vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
CD3OD) d 26.8 (16-C), 116.6 (9-C), 128.0 (8-C), 129.8 (12-C),
129.9 (13-C), 130.1 (7-C), 138.3 (14-C), 138.5 (6-C), 139.2 (11-C),
143.8 (5-C), 143.9 (3-C), 152.5 (2-C), 159.1 (10-C); MS (FAB+)
C18H15N3O2 (MW = 305.34) m/z 306.4 ((M+H)+), 186 ((MꢁC8H7O)+),
119 ((C8H7O)+); Anal. Calcd for C18H15N3O2ꢃ0.3H2O (312.57): C,
69.57; H, 5.07; N, 13.52. Found: C, 69.64; H, 4.94; N, 13.20. UV–
(e)): 378 (1.3), 287 (2.06), 267 (2.15); mp 186–188 °C.
4.2.3. 2-Amino-3-(4-trifluoromethylphenyl)-5-(4-
hydroxyphenyl)-1,4-pyrazine (6e)
Compound 6e was prepared according to the general procedure
A
using 1,4-bis(diphenylphosphino)butane (55 mg, 0.13 mmol,
vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
(e)): 377 (0.83), 290 (1.75), 272
(1.9), 259 (1,9); mp 223–225 °C.
6 mol %), bis(benzonitrile)dichloro palladium (39 mg, 0.10 mmol,
5 mol %), dry toluene (2 ꢀ 10 mL), 5 (645 mg, 1.7 mmol, 1 equiv),
corresponding arylboronic acid (410 mg, 2.16 mmol, 1.27 equiv),
4.3. General procedure B for the condensation of 3,5-diaryl-2-
aminopyrazines (6) with methylglyoxal
ethanol (700 lL), 1 M Na2CO3 (1.7 mL, 1 equiv), THF (30 mL) and
tetrabutylammonium fluoride (4.3 mL, 2.5 equiv); column chroma-
tography using cyclohexane/ethyl acetate 5/4; 512 mg of 6e as yel-
low solid; 91% yield; Rf 0.30 (cyclohexane/ethyl acetate: 5/4); 1H
Methylglyoxal (40% w/v aqueous solution, 2 equiv) and HCl 37%
(3.6 equiv) were added to a solution of 6 (1 equiv) in ethanol
(3 mL/mmol) under argon atmosphere. The mixture was heated
at 80 °C for 4 h. After cooling, the mixture was evaporated to dry-
ness, the residue was washed with ether and ethyl acetate. Com-
pounds 7 were precipitated from cold methanol.
NMR (500 MHz, DMSO-d6)
d 6.29 (br s, 2H, NH2), 6.81 (d,
3
3J9,8 = 8.7 Hz, 2H, 9-H), 7.80 (d, J8,9 = 8.7 Hz, 2H, 8-H), 7.84 (d,
3J13,12 = 8.1 Hz, 2H, 13-H), 8.01 (d, J12,13 = 8.1 Hz, 2H, 12-H), 8.49
3
(s, 1H, 6-H), 9.56 (s, 1H, OH). 13C NMR (125 MHz, DMSO-d6) d
1
115.5 (9-C), 124.27 (q, J15,F = 272 Hz, 15-C), 125.42 (q,
3J13,F = 3.1 Hz, 13-C) 126.4 (8-C), 127.7 (7-C), 128.5 (d,
2J14,F = 32 Hz, 14-C), 129.0 (12-C), 135.6 (11-C), 137.8 (6-C), 140.6
(5-C), 141.8 (3-C), 151.4 (2-C), 157.4 (10-C); MS (FAB+)
C17H12F3N3O (MW = 331.29) m/z 332.0 ((M+H)+); Anal. Calcd for
C17H12F3N3Oꢃ0.4H2O (338.50): C, 60.32; H, 3.81; N, 12.41. Found:
4.3.1. 2-Methyl-6-(4-hydroxyphenyl)-8-(4-methoxyphenyl)-
3,7-dihydroimidazo[1,2-a]pyrazin-3-one (7c)
Compound 7c was prepared according to the general procedure
B
using methylglyoxal (115 lL, 2 equiv), HCl 37% (110 lL,
3.6 equiv), 6c (100 mg, 0.34 mmol, 1 equiv), ethanol (10 mL);
96 mg of 7c as orange yellow solid; 81% yield; 1H NMR
(300 MHz, CD3OD) d 2.54 (s, 3H, 10-H), 3.94 (s, 3H, 19-H), 6.89
C, 60.54; H, 3.74; N, 12.01. UV–vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
371 (0.91), 292 (2.18); mp > 250 °C decomposition.
(e)):
3
3
(d, J18,17 = 8.7 Hz, 2H, 13-H), 7.19 (d, J13,12 = 9.0 Hz, 2H, 17-H),
3
3
4.2.4. 2-Amino-3-(4-fluorophenyl)-5-(4-hydroxyphenyl)-1,4-
pyrazine (6f)
7.86 (d, J17,18 = 8.7 Hz, 2H, 12-H), 8.04 (d, J12,13 = 9.0 Hz, 2H, 16-
H), 8.40 (s, 1H, 5-H). 13C NMR (75 MHz, CD3OD) d 9.6 (10-C), 56.1
(19-C), 109.2 (5-C), 115.7 (17-C), 116.8 (13-C), not observed (2-C
or 3-C), 126.2, 126.4, 126.6 (15-C, 11-C, 9-C), 129.4 (12-C), 132.0
(16-C), 139.0 (2-C or 3-C), 142.0 (6-C), 145.5 (8-C), 160.7 (14-C),
164.2 (18-C); MS (FAB+) C20H17N3O3 (MW = 347.38) m/z 348.2
((M+H)+); HRMS for C20H18N3O3: Calcd 348.1348, found
Compound 6f was prepared according to the general procedure
A using 1,4-bis(diphenylphosphino)butane (49 mg, 0.116 mmol,
6 mol %), bis(benzonitrile)dichloro palladium (36 mg, 0.093 mmol,
5 mol %), dry toluene (2 ꢀ 10 mL), 5 (719 mg, 1.89 mmol, 1 equiv),
corresponding arylboronic acid (310 mg, 2.21 mmol, 1.2 equiv),
ethanol (800
lL), 1 M Na2CO3 (1.9 mL, 1 equiv), THF (15 mL) and
348.1348; UV–vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
(e)): 441 (0.40), 351
(0.65), 307 (1.25), 274 (1.58); mp > 300 °C, decomposition.
tetrabutylammonium fluoride (2.6 mL, 2.5 equiv); Column chro-
matography using cyclohexane/ethyl acetate 4/5; 367 mg of 6f as
yellow solid; 69% yield; Rf 0.32 (cyclohexane/ethyl acetate: 5/4);
1H NMR (500 MHz, CD3OD) d 4.62 (br s, 2H, NH2), 6.85 (d,
4.3.2. 2-Methyl-6-(4-hydroxyphenyl)-8-(4-(N,N)-
dimethylaminophenyl)-3,7-dihydroimidazo[1,2-a]pyrazin-3-
one (7d)
Compound 7d was prepared according to the general procedure
B using methylglyoxal (90 lL, 2 equiv), HCl 37% (90 lL, 4.2 equiv),
3
3J9,8 = 9 Hz, 2H, 9-H), 7.24 (t, J13,12 = 3J13,F = 8.8 Hz, 2H, 13-H),
3
3
7.76 (d, J8,9 = 9 Hz, 2H, 8-H), 7.82 (dd, J12,13 = 8.8 Hz and
3J12,F = 5.2 Hz, 2H, 12-H), 8.31 (s, 1H, 6-H). 13C NMR (125 MHz,
CD3OD) d 116.6 (9-C), 116.7 (d, J13,F = 23.6 Hz, 13-C), 128.4 (8-C),
129.8 (7-C), 131.7 (d, J12,F = 8.4 Hz, 12-C), 135.2 (11-C), 137.8 (6-
2
6d (80 mg, 0.26 mmol, 1 equiv), ethanol (10 mL); 66 mg of 7d as
3
yellow solid; 70% yield; 1H NMR (300 MHz, CD3OD) d 2.56 (s, 3H,
3
C), 139.9 (3-C), 143.8 (5-C), 152.5 (2-C), 159.0 (10-C), 164.46 (d,
1J14-F = 247 Hz, 14-C); MS (FAB+) C16H12FN3O (MW = 281.28) m/z
282.3 ((M+H)+); Anal. Calcd for C16H12FN3Oꢃ0.1H2O (283.08): C,
67.89; H, 4.34; N, 14.84. Found: C, 68.01; H, 4.63; N, 14.39. UV–
10-H), 3.35 (s, 6H, 19-H), 6.93 (d, J18,17 = 8.4 Hz, 2H, 13-H), 7.58
3
3
(d, J13,12 = 9.0 Hz, 2H, 17-H), 7.89 (d, J17–18 = 8.4 Hz, 2H, 12-H),
8.21 (d, J12,13 = 9.0 Hz, 2H, 16-H), 8.48 (s, 1H, 5-H). 13C NMR
3
(125 MHz, CD3OD) d 9.9 (10-C), 44.2 (19-C), 109.7 (5-C), 116.9
(13-C), 118.6 (17-C), 122.8 (2-C or 3-C), 126.1 (11-C), 126.8 (9-C),
128.7 (15-C), 129.6 (12-C), 132.4 (16-C), 139.3 (2-C or 3-C), 141.5
(6-C), 144.7 (8-C), 146.8 (18-C), 160.8 (14-C); MS (APCI)
C21H19N4O2 (MW = 360.42) m/z 361.4 ((M+H)+); HRMS for
C21H21N4O2: calculated 361.1665, found 361.1660; UV–vis (MeOH,
vis (MeOH, 5 ꢀ 10–5 M, kmax
(2.3); mp 232–234 °C.
(e)): 365 (0.79), 329 (0.51), 287
4.2.5. 2-Amino-3-(4-acetylphenyl)-5-(4-hydroxyphenyl)-1,4-
pyrazine (6g)
Compound 6g was prepared according to the general procedure
A using 1,4-bis(diphenylphosphino)butane (21 mg, 0.048 mmol,
6 mol %), bis(benzonitrile)dichloro palladium (15 mg, 0.039 mmol,
5 mol %), dry toluene (2 ꢀ 15 mL), 5 (275 mg, 0.72 mmol, 1 equiv),
corresponding arylboronic acid (126 mg, 0.77 mmol, 1.05 equiv),
5 ꢀ 10ꢁ5 M, kmax
(e)): 453 (1.08), 394 (1.62), 260 (2.27);
mp > 300 °C, decomposition.
4.3.3. 2-Methyl-6-(4-hydroxyphenyl)-8-(4-
trifluoromethylphenyl)-3,7-dihydroimidazo[1,2-a]pyrazin-3-
one (7e)
ethanol (300 lL), 1 M Na2CO3 (0.75 mL, 1 equiv), THF (13 mL)
and tetrabutylammonium fluoride (1.8 mL, 2.5 equiv); Column
chromatography using cyclohexane/ethyl acetate: 2/3; 165 mg of
6g as yellow solid; 78% yield; Rf 0.3 (cyclohexane/ethyl acetate:
2/3); 1H NMR (500 MHz, CD3OD) d 2.67 (s, 3H, 16-H), 4.6 (br s,
2H, NH2), 6.86 (d, J9,8 = 8.7 Hz, 2H, 9-H), 7.79 (d, J8–9 = 8.7 Hz,
2H, 8-H), 7.97 (d, 3J12,13 = 8.7 Hz, 2H, 12-H), 8.15 (d,
3J13,12 = 8.7 Hz, 2H, 13-H), 8.38 (s, 1H, 6-H). 13C NMR (125 MHz,
Compound 7e was prepared according to the general procedure
using methylglyoxal (280 lL, 2.2 equiv), HCl 37% (250 lL,
B
3.6 equiv), 6e (282 mg, 0.84 mmol, 1 equiv), ethanol (15 mL);
325 mg of 7e as pale yellow solid; 92% yield; 1H NMR (300 MHz,
3
3
3
CD3OD) d 2.54 (s, 3H, 10-H), 6.85 (d, J18,17 = 8.4 Hz, 2H, 13-H),
3
3
7.88 (d, J17,18 = 8.4 Hz, 2H, 12-H), 7.96 (d, J13,12 = 8.1 Hz, 2H, 17-
H), 8.20 (d, J12,13 = 7.9 Hz, 2H, 16-H), 8.55 (s, 1H, 5-H). 13C NMR
3

F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
4343
(125 MHz, CD3OD) d 9.2 (10-C), 110.1 (5-C), 116.8 (13-C), 119.8 (2-
azine-N0-2-ethanesulfonic acid (HEPES)-supported HCO3ꢁ-buffered
saline solution (composition in mM: NaCl 110.0, KCl 4.7, CaCl2 2.0,
MgSO4 1.2, NaHCO3 18.0, Hepes 15.39 and Hepes Na+-salt 14.61)
whose temperature was maintained at 36.5 °C and the superfusion
rate was 6 ml/min. The pH was set to 7.40 by bubbling the solu-
tions continuously with 5% CO2 in 95% N2. Thirty min after the
completion of the preparative procedure, fluorescein isothiocya-
nate (FITC)-dextran, molecular weight 150,000 Da (Sigma), with a
degree of substitution of 2 FITC molecules per 1000 glucose mole-
cules in the polysaccharide chain was given in a dose of 25 mg/
100 g body weight as an intravenous injection of 5% solution in
0.9% saline. Observations were made with a Leitz Ortholux micro-
scope with X3.5 objective and X10 oculars. The light source was a
100 W mercury DC lamp (Irem model EL-XH5 P/L). The specific
light filters used for observations in fluorescent light (Leitz BG12,
BG38, GG455 and KP490) were positioned between the light
source and the condenser to give a light for optimal excitation at
490 nm of FITC-dextran. A barrier filter (K530) was placed between
the objective and the eyepieces. Observations of the number of
leakage sites (=leaks) were made by scanning manually the total
observation area twice at X35 magnification. The fluorescent spots
formed at leakage sites could be traced when they reached a cer-
tain minimal size and fluorescent intensity. Each site was classified
3
C or 3-C), 126.5 (9-C), 126.7 (11-C), 127.1 (d, JC–F = 3.6 Hz, 17-C),
2
129.1 (12-C), 130.8 (16-C), 133.9 (d, JC–F = 32.8 Hz, 18-C), 138.4
(15-C), 138.8 (2-C or 3-C), 143.5 (6-C), 144.5 (8-C), 160.7 (14-C).
19F NMR (282 MHz, CD3OD)
d
ꢁ62.7 (19-F); MS (APCI)
C20H14F3N3O2 (MW = 385.35) m/z 386.3 ((M+H)+); Anal. Calcd for
C20H14F3N3O2ꢃHClꢃ1.5H2O (448.84): C, 53.52; H, 3.71; N, 9.36.
Found: C, 53.40; H, 3.78; N, 9.24. UV–vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
(e)): 450 (0.38), 353 (0.47), 307 (1.21), 270 (2.26); mp > 300 °C,
decomposition.
4.3.4. 2-Methyl-6-(4-hydroxyphenyl)-8-(4-fluorophenyl)-3,7-
dihydroimidazo[1,2-a]pyrazin-3-one (7f)
Compound 7f was prepared according to the general procedure
B
using methylglyoxal (190 lL, 2.5 equiv), HCl 37% (190 lL,
4.7 equiv), 6f (137 mg, 0.49 mmol, 1 equiv), ethanol (10 mL);
133 mg of 7f as yellow solid; 73% yield; 1H NMR (300 MHz, CD3OD)
3
d 2.54 (s, 3H, 10-H), 6.86 (d, J17,16 = 8.8 Hz, 2H, 13-H), 7.41 (t,
3
3J13,12 = 3J13–F = 8.7 Hz, 2H, 17-H), 7.88 (d, J16,17 = 8.7 Hz, 2H, 12-
3
4
H), 8.09 (dd, J12,13 = 8.9 Hz and J12–F = 5.3, 2H, 16-H), 8.50 (s,
1H, 5-H). 13C NMR (75 MHz, CD3OD) d 6.21 (10-C), 106.6 (5-C),
113.7 (13-C), 114.3 (d, JC–F = 22.2 Hz, 17-C), 116.8 (2-C or 3-C),
2
123.4 (9-C), 123.7 (11-C), 126.1 (12-C), 127.9 (15-C), 129.6 (d,
3JC–F = 9.0 Hz, 16-C), 135.6 (2-C or 3-C), 140.2 (6-C), 142.0 (8-C),
as a leakage site when its diameter was larger than 100 lm. The
1
157.6 (14-C), 163.2 (d, JC–F = 249 Hz, 18-C). 19F NMR (282 MHz,
number of leakage sites is reported per square centimeter. All ham-
sters with the prepared area showing spontaneous nonfading leaks
or more than 10 fading leaks during the first 30 min control period,
after FITC-dextran was given, were discarded. A group of three ani-
mals were treated by gavage with 0.2 ml of a solution containing
the compound at 3 mg/kg while another group of three animals re-
ceived 0.2 ml of the solvent. Each animal was treated with the
compound or the solvent 30 min before anesthesia. The prepara-
tions were used to investigate the effect of local ischemia, which
was obtained by a cuff, made of thin latex tubing, mounted around
the neck of the everted pouch where it leaves the mouth of the
hamster. The cuff can be placed without any visible interference
with local blood flow. The intratubular pressure of the cuff can
be rapidly increased by air compression using a syringe, and also
be rapidly decreased at evacuation; an intratubular of 200–
220 mmHg resulted in a complete arrest of the microvascular flow
within a few seconds. Throughout the 30-min occlusion period,
minor adjustments of blood movement could be observed in the
larger blood vessels. At evacuation, an immediate increase in blood
flow occurred and the blood flow eventually returned to the pre-
occlusion values. The maximal number of leaky sites per cm2
was determined 30 min after the start of reperfusion. The mean
of the leaky sites was calculated for the solvent and the treated
groups of animals and the percentage of inhibition was calculated
as follow: [1 ꢁ (mean treated)/(mean control)] ꢀ 100.
CD3OD) d ꢁ108.4 (18-F); MS (APCI) C19H14FN3O2 (M.W. = 335.34)
m/z 336.3 ((M+H)+); Anal. Calcd for C19H14FN3O2ꢃHClꢃH2O
(389.82): C, 58.54; H, 4.40; N, 10.78. Found: C, 58.70; H, 4.51; N,
(e)): 444 (0.45), 352
(0.44), 306 (1.21), 268 (2.13); mp > 300 °C, decomposition.
10.57. UV–vis (MeOH, 5 ꢀ 10ꢁ5 M, kmax
4.3.5. 2-Methyl-6-(4-hydroxyphenyl)-8-(4-acetylphenyl)-3,7-
dihydroimidazo[1,2-a]pyrazin-3-one (7g)
Compound 7g was prepared according to the general procedure
B
using methylglyoxal (61 lL, 2.2 equiv), HCl 37% (80 lL,
3.6 equiv), 6g (81 mg, 0.27 mmol, 1 equiv), ethanol (5 mL); 91 mg
of 7g as red clay solid; 85% yield; 1H NMR (300 MHz, DMSO-d6) d
3
2.42 (s, 3H, 10-H), 2.67 (s, 3H, 20-H), 6.89 (d, J19,18 = 8.7 Hz, 2H,
3
3
13-H), 7.99 (d, J18,19 = 8.7 Hz, 2H, 12-H), 8.15 (d, J12,13 = 8.4 Hz,
3
2H, 16-H), 8.48 (s, 1H, 5-H), 8.75 (d, J13,12 = 8.4 Hz, 2H, 17-H);
MS (FAB+) C21H17N3O3 (MW = 359.39) m/z 360.1 ((M+H)+); Anal.
Calcd for C21H17N3O3ꢃHClꢃ0.25H2O (400.36): C, 63.0; H, 4.62; N,
10.49. Found: C, 63.03; H, 4.65; N, 10.31. UV–vis (MeOH,
5 ꢀ 10ꢁ5 M, kmax
(e)): 458 (0.28), 362 (0.65), 312 (1.5), 273 (2.7);
mp > 300 °C, decomposition.
4.4. Biological assays: permeability in the hamster cheek pouch
submitted to ischemia/reperfusion (I/R)
Male Syrian hamsters (Mesocricetus auratus, Engle Labs Farm-
ersburg, Indianapolis, USA), 7–10 weeks old, weighing 90–130 g,
were used for these studies. Anesthesia was induced by an intra-
peritoneal injection of 0.1–0.2 ml of sodium pentobarbital (Sanofi,
Acknowledgements
The authors thank the F.R.S.-FNRS (Belgium), FRIA (Belgium)
and IdRS (France) for the financial support. J.M-B. is senior research
associate of F.R.S.-FNRS (Fonds de la Recherche Scientifique).
F.D.W. is holding a PhD bursary of FRIA (Fonds pour la Formation
à la Recherche dans l’Industrie et l’Agriculture).
France, 60 mg/ml) and maintained with
a-chloralose (100 mg/kg)
administered through the femoral vein. The femoral artery was
cannulated for measurement of arterial blood pressure. Through-
out surgery and the subsequent experiment, the temperature of
the animals was kept at 37.5 °C with a heating pad controlled by
a rectal thermistor. A tracheal tube was inserted to facilitate spon-
taneous breathing. The hamster was placed on a microscope stage.
The cheek pouch was gently averted and pinned with 4–5 needles
into a circular well filled with silicone rubber to give a flat bottom
layer thus avoiding stretching of the tissue but preventing shrink-
age. In this position, the pouch was submerged in a superfusion
solution that continuously flushed the pool of the microscope
stage. The superfusion solution was a N-2-hydroxyethylpiper-
References and notes
1. Oxidative Stress: Oxidants and Antioxidants; Sies, H., Ed.; Academic Press:
London, 1991.
2. Free Radicals Damage and its Control (New Comprehensive Biochemistry); Rice-
Evans, C. A., Burdon, C. A., Eds.; Elsevier: Amsterdam, 1994; Vol. 28.
3. Free Radicals in Biology and Medicine; Halliwell, B., Gutteridge, J. M. C., Eds.;
Oxford University Press: Oxford, 1999.
4. Beckman, K. B.; Ames, B. N. Physiol. Rev. 1998, 78, 547–581.

4344
F. De Wael et al. / Bioorg. Med. Chem. 17 (2009) 4336–4344
5. Morrissey, P. A.; O’Brien, N. M. Int. Dairy J. 1998, 8, 463–472.
6. Krishna, M. C.; De Graff, W.; Hankovzsky, O. H.; Sar, C. P.; Kalai, T.; Jeko, T.;
Russo, A.; Mitchell, J. B.; Hideg, K. J. Med. Chem. 1998, 41, 3477.
7. Schackelfold, R. E.; Kaufmann, W. K.; Paules, R. S. Free Radical Biol. Med. 2000,
28, 1387.
formation of a a-keto-acid together with etioluciferin, consisted in the
formation of a tertiary alcohol from hydroperoxide intermediate AH (see
Scheme 1) by base catalyzed oxygen elimination or reaction of AH with
unreacted luciferin, and subsequent hydrolysis.30 In our experiments,
compound
4
with RT = 2.1 min is benzoic acid (R2CO2H) and not
8. Kehrer, J. P. Toxicology 2000, 149, 43.
phenylglyoxalic acid (R2COCO2H).
9. Antioxidants in Science, Technology, Medicine and Nutrition; Scoot, G., Ed.; Albion
Publishing Ltd: Chichester, 1997.
10. Natural Antioxidants in Human Health and Disease; Frei, B., Ed.; Academic Press:
San Diego, 1994.
32. Devillers, I.; Dive, G.; De Tollenaere, C.; Falmagne, B.; de Wergifosse, B.; Rees, J.-
F.; Marchand-Brynaert, J. Bioorg. Med. Chem. Lett. 2001, 11, 2305.
33. Hirano, T.; Nishibuchi, S.; Yoneda, M.; Tsujimoto, K.; Ohashi, M. Tetrahedron
1993, 49, 9267.
11. Larson, R. A. Naturally Occurring Antioxidants; Lewis CRC Press: Boca Raton,
1997.
34. Cavalier, J.-F.; Burton, M.; De Tollenaere, C.; Dussart, F.; Marchand, C.; Rees, J.-
F.; Marchand-Brynaert, J. Synthesis 2001, 5, 768.
12. Handbook of Synthetic Antioxidants; Packer, L., Cadenas, E., Eds.; Marcel Dekker:
New York, 1997.
35. Cavalier, J.-F.; Burton, M.; Dussart, F.; Marchand, C.; Rees, J.-F.; Marchand-
Brynaert, J. Bioorg. Med. Chem. 2001, 9, 1037.
13. Andersson, C-M.; Hallberg, A.; Hogberg, A. Adv. Drug Res. 1996, 28, 65.
14. Gordon, M. H. Natural products reports 1996, 265.
15. Teranishi, K. Bioorg. Chem. 2007, 35, 82.
36. Jeanjot, P.; Bruyneel, F.; Arrault, A.; Gharbi, S.; Cavalier, J.-F.; Abels, A.;
Marchand, C.; Touillaux, R.; Rees, J.-F.; Marchand-Brynaert, J. Synthesis 2003, 4,
513.
16. Wu, C.; Kawasaki, K.; Ohgiya, S.; Ohmiya, Y. Tetrahedron Lett. 2006, 47, 753.
17. Auld, D. S.; Southall, N. T.; Jadhav, A.; Jhonson, R. L.; Diller, D. J.; Simeonov, A.;
Austin, C. P.; Inglese, J. J. Med. Chem. 2008, 51, 2372.
18. Kuse, M.; Isobe, M. Tetrahedron 2000, 56, 2629.
19. Vysotski, E. S.; Lee, J. Acc. Chem. Res. 2004, 37, 405.
20. Kotarsky, K.; Antonsson, L.; Owman, C.; Olde, B. Anal. Biochem. 2003, 316, 208.
21. Alvarez, J.; Montero, M. Cell Calcium 2002, 32, 251.
22. Teranishi, K.; Shimomura, O. Anal. Biochem. 1997, 249, 37.
23. Daiber, A.; August, M.; Baldus, S.; Wendt, M.; Oelze, M.; Sydow, K.; Kleschyov,
A. L.; Munzel, T. Free Radical Biol. Med. 2004, 36, 101.
24. Dubuisson, M. L. N.; Rees, J.-F.; Marchand-Brynaert, J. Drug Dev. Ind. Pharm.
2005, 31, 827.
25. Dubuisson, M. L. N.; Rees, J.-F.; Marchand-Brynaert, J. Mini-Rev. Med. Chem.
2004, 4, 421.
26. Janssens, B.; Le-Gall, R.; Rees, J.-F. J. Fish Biol. 2002, 61, 71.
27. Takahashi, Y.; Kondo, H.; Maki, S.; Niwa, H.; Ikeda, H.; Hirano, T. Tetrahedron
Lett. 2006, 47, 6057.
37. The oxygen concentration in aerated acetonitrile at 298 K is 1.7 ꢀ 10ꢁ3 mol Lꢁ1
,
according to: Achord, J. M.; Hussey, C. L. Anal. Chem. 1980, 52, 601. Non-dried
solvents were used, thus containing enough moisture for performing the
hydrolysis of highly reactive intermediates.
38. Burton, M.; De Tollenaere, C.; Cavalier, J.-F.; Marchand, C.; Dussart, F.; Rees, J.-
F.; Marchand-Brynaert, J. Free Radical Res. 2003, 37, 145.
39. Goto, T.; Fukatsu, H. Tetrahedron Lett. 1969, 10, 4299.
40. Teranishi, K.; Tanabe, S.; Hisamatsu, M.; Yamada, T. Luminescence 1999, 14,
303.
41. Teranishi, K.; Nishiguchi, T.; Ueda, H. Carbohydr. Res. 2003, 228, 987.
42. Adamczyk, M.; Akireddy, S. R.; Johnson, D. D.; Mattingly, P. G.; Pan, Y.; Reddy,
R. E. Tetrahedron 2003, 59, 8129.
43. Acetic acid (4, R2 = Me) has not been formally identified, but proposed as by-
product in analogy with the degradation of imidazopyrazinones bearing a
phenyl substituent (R2 = Ph). We cannot exclude the possible formation of
methylglyoxalic acid.30,31
44. Arrault, A.; Dubuisson, M. L. N.; Gharbi, S.; Marchand, C.; Verbeuren, T.; Rupin,
A.; Cordi, A.; Bouskela, E.; Rees, J.-F.; Marchand-Brynaert, J. Bioorg. Med. Chem.
Lett. 2003, 13, 653.
28. Kondo, H.; Igarashi, T.; Maki, S.; Niwa, H.; Ikeda, H.; Hirano, T. Tetrahedron Lett.
2005, 46, 7701.
29. Usami, K.; Isobe, M. Tetrahedron 1996, 52, 12061.
30. Shimomura, O.; Johnson, F. H. Biochem. Biophys. Res. Commun. 1971, 44, 340.
45. Newan, D. J. J. Med. Chem. 2008, 51, 2589.
46. For a recent review on dual action approaches, see: Bremner, J. B.; Ambrus, J. I.;
Samosorn, S. Curr. Med. Chem. 2007, 14, 1459.
31.
a-Keto-b-methyl-n-valeric acid has been identified as by-product of aerobic
oxidation of cypridina luciferin. The proposed mechanism to explain the