Design, synthesis and glutathione peroxidase-like properties of ovothiol-derived diselenides

Article abstract of DOI:10.1016/S0968-0896(03)00504-2

Eleven imidazole diselenides derived from the naturally occurring family of antioxidants, the ovothiols, were synthetised by cyclisation of selenoamides with trimethylsilyltrifluoromethanesulfonate or refluxing of cyanoamines in a selenium/sodium borohydride mixture. These compounds were assayed for their glutathione peroxidase-like (GSH Px-like) activity and their capacity to be reduced by glutathione. The most active molecules of the series were 4 times more potent in the GSH Px-like test than the widely known reference compound, ebselen. This catalytic activity was mediated by the formation of the antioxidant selenol intermediate upon partial but significant exchange reaction with glutathione.

Full text of DOI:10.1016/S0968-0896(03)00504-2

Bioorganic & Medicinal Chemistry 11 (2003) 4623–4630
Design, Synthesis and Glutathione Peroxidase-Like Properties of
Ovothiol-Derived Diselenides
Fabrice Bailly,a,* Nathalie Azaroualb and Jean-Luc Berniera
a
´
Laboratoire de Chimie Organique Macromoleculaire, CNRS UMR 8009, USTL, Batiment C3, UST Lille I,
ˆ
59655 Villeneuve d’Ascq, France
Laboratoire de RMN, Faculte de Pharmacie, CNRS UMR 8009, BP83, 59006 Lille Cedex, France
b
´
Received 7 February 2003; accepted 25 July 2003
Abstract—Eleven imidazole diselenides derived from the naturally occurring family of antioxidants, the ovothiols, were synthetised
by cyclisation of selenoamides with trimethylsilyltrifluoromethanesulfonate or refluxing of cyanoamines in a selenium/sodium bor-
ohydride mixture. These compounds were assayed for their glutathione peroxidase-like (GSH Px-like) activity and their capacity to
be reduced by glutathione. The most active molecules of the series were 4 times more potent in the GSH Px-like test than the widely
known reference compound, ebselen. This catalytic activity was mediated by the formation of the antioxidant selenol intermediate
upon partial but significant exchange reaction with glutathione.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
Ebselen (2-phenyl-1,2-benzisoselenazol-3(2H)-one) was
the first compound suggested for hydroperoxide-inac-
tivating therapy in the presence of glutathione.11,12
Then several attempts have been made to prepare
simple synthetic organoselenium compounds with glu-
tathione peroxidase-like activity like ebselen, which
include ebselen analogues,13 benzoselenazolinones,14
selenamides,15 diaryldiselenides14,16ꢀ21,43 and the semi-
synthetic selenosubtilisin.22 The activities of various
selenides like diarylselenides,23 a-(phenylselenyl)
ketones24 and more recently 3-hydroxypropylselenide25
were also investigated. Two excellent reviews descri-
bed the chemical and pharmacological aspects of
these organoselenium compounds.26,27 All these
researches led to the obtention of compounds with
enzymatic activities much higher than that of the
original ebselen.
For several years, there has been an intensive search for
the use of organoselenium compounds in enzymology
and bioorganic chemistry. The development of these
novel compounds was based on the knowledge of the
various biological functions of selenium,1,2 on the syn-
thetic effort leading to the discovery of new useful reac-
tions and on the pharmacological evaluation of the
synthetised compounds.3ꢀ5 The principal role of sele-
nium in mammals was established as its implication in
the catalytic site of the antioxidant enzyme glutathione
peroxidase (GPx). The enzyme catalytic site includes a
selenocysteine residue in which the selenium undergoes
a redox cycle involving the selenolate anion as the active
form which reduces hydroperoxides.6,7 This led to the
development of organoselenium compounds designed to
mimic this enzymatic activity in vitro. They can be used
for the treatment of a certain number of pathologies in
which an over-production of cytotoxic hydroperoxides
contributes to the functional impairments of cells or
tissues and more particularly for the treatment of
atherosclerosis, inflammatory and/or ischaemic cardio-
vascular, cerebrovascular, digestive, respiratory and
ophtalmic pathologies.8ꢀ10
Recently, we reported the synthesis28 and antioxidant
properties of 4-mercaptoimidazoles derived from the
naturally occurring ovothiols (Fig. 1).28,29 These com-
pounds were shown to be powerful scavengers of
hydrogen peroxide, hydroxyl radicals and hypochlorous
acid and to inhibit copper-induced LDL peroxidation.29
Thiyl radicals were characterised upon radical scaven-
ging.30 Protection against peroxynitrite-induced damage
and glutathione peroxidase-like activity were also evi-
denced.31 According to the in vitro experiments, these
aromatic heterothiols were shown to be largely better
antioxidants than the aliphatic thiol, glutathione.
*Corresponding author. Tel.: +33-3-2033-7231; fax: +33-3-2033-
6309; e-mail: fabrice.bailly@univ-lille1.fr
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00504-2

4624
F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
chromatography led to a major recovery of the nitrile
forms 2 by hydrogen selenide evolving as judged by
thin-layer chromatography (TLC). The selenoamides 3
were identified on TLC plates as strongly absorbing
spots under UV light and as orange spots after iodine
revelation.
Thus, in the first method (method A), hydrogen selenide
generated in situ by hydrolysis of aluminium selenide was
bubbled into a solution of the precursors 2 and triethyla-
mine in dichloromethane until there was not any sig-
nificant evolution of the solution content as judged by
TLC. Then the cyclisation step (step iv) was rapidly per-
formed by treating at ꢀ10 ꢁC the selenoamides 3 with
6 equiv of triethylamine (TEA) and 4 equiv of tri-
methylsilyltrifluoromethanesulfonate (Me3SiOTf).
Figure 1. Structures of ovothiol C, 4-mercaptoimidazoles and imidazole
diselenides.
In continuation of this work, we present in this paper
the synthesis of the organoselenium analogues obtained
as the diselenides 4 (Fig. 1) and their ability to catalyse
decomposition of hydrogen peroxide in the presence of
glutathione. The glutathione peroxidase-like activity
was assessed in vitro by means of the NADPH–glu-
tathione reductase system. Exchange reactions with
glutathione of these new derivatives were also investi-
gated by means of two tests: measurement of glu-
tathione disulphide (GSSG) formed via the NADPH–
glutathione reductase system and measurement of sele-
nol formed via scavenging by the stable 2,2-diphenyl-1-
picrylhydrazyl radical (DPPH).
In the second method, the precursors 2 were treated at
room temperature with a 2:3 molar ratio of sodium
borohydride to selenium in ethanol. The crude product
obtained after concentration in vacuo and filtration of
the solid residues was used as well without further
purification for the next cyclisation step.
A serendipituous observation led to an alternative
method (stepv). A by-product formed during the reac-
tion of sodium borohydride with selenium in presence
of precursors 2, which was identified by NMR of the
crude product as the cyclised diselenides 4. It seemed
that the initial heat of the reaction between selenium
and sodium borohydride provoked the cyclisation of the
selenoamides in the mixture. This was corroborated by
the fact that the proportion of cyclised product was
strongly diminished when the reaction was conducted at
0 ꢁC. Thus, slight heating at reflux under the experi-
mental conditions led to the progressive cyclisation of a
major part of the selenoamide present in solution. This
alternative method (method B) was greatly advanta-
geous since the overall yields for conversion of nitriles 2
into diselenides 4 were similar to those obtained by
combining stepiii and stepiv (method A). Yields ranged
from 15 to 45%.
Results
Chemistry
The synthesis of aromatic diselenides 4a–k (Fig. 2) was
accomplished by a method similar to that developed by
Spaltenstein et al. for the synthesis of 4-mercaptoimi-
dazoles28,32 as shown in Scheme 1. Strecker condensa-
tions gave the aminonitriles 1. Transformation of
amines in amides and conversion of nitriles to selenoa-
mides gave the precursors 3. For the third step, two
reactions were performed: to produce hydrogen sele-
nide: hydrolysis of aluminium selenide33,34 and reaction
of selenium with sodium borohydride (molar ratio 3:2)
in ethanol.35 Owing to the well-known instability of
primary alkyl selenoamides when compared to the aro-
matic ones,36,37 the precursors 3 could not be con-
veniently isolated and purified. Attempts to isolate the
selenoamides 3 by extractive workupand silica gel
Scheme 1. Reagents and conditions: (i) H2O–MeOH, HCl or AcOH;
(ii) HCO2H–Ac2O or RCOCl, Et3N; (iii) Al2Se3, H2O, Et3N; (iv)
Me3SiOTf, CH2Cl2, Et3N, ꢀ10 ꢁC; (v) Se, NaBH4, EtOH, Et3N, room
temperature then slight reflux.
Figure 2. Structures of the imidazole diselenides 4a–k.

F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
4625
Antioxidant assays
60% and here again, compounds 4e and 4k were the less
reactive of the series. A thin precipitate appeared in the
cuvette during experiments with 4e, which could explain
the particularly low exchange rate observed for this
compound.
Glutathione peroxidase-like activity evaluation. Figure 3
and Table 1 show the abilities of the diselenides 4 to
catalyse the reduction of hydrogen peroxide in presence
of glutathione. The most active compounds of the series
were compounds 4f with 2397% catalysis and 4h with
2337% catalysis. They were about 4 times more potent
than the reference compound, ebselen (613% catalysis).
A large part of the remaining diselenides presented
roughly the same catalytic activities (1800–1900%),
except for compounds 4e, 4i and 4k, which are only two
times as effective as ebselen (1400% catalysis).
In the second test, the same amounts of diselenides
(50 mM) and glutathione (100 mM) as in the precedent
study were incubated with DPPH (100 mM). If the
exchange reaction is total, all the initial DPPH quantity
should be reduced by the strongly reactive selenol form.
Indeed, the characteristic DPPH absorbance at 517 nm
was extinguished in some extent in presence of the sele-
nol formed upon exchange reactions (Fig. 4; Table 1).
Scavenging of DPPH by glutathione was extremely
weak and did not interfere in the test. Here again, the
most effective compounds were 4f and 4h (65% DPPH
Exchange reactions with glutathione. As encountered for
ovothiols and imidazole disulphides,31,38 diselenides are
readily reduced by glutathione (Table 1). In presence of
a 2-fold glutathione excess, 70–80% of the initial di-
selenide concentration (compounds 4f, 4h) reacted with
glutathione to yield glutathione disulphide. For the
other test compounds, exchange rates ranged from 40 to
Figure 3. GSH Px-like activity of diselenides. Reactions were initiated
by addition of 1.0 mM H2O2 to the assay mixture containing 1.0 mM
GSH, 0.26 mM NADPH, 1U/mL GSSG reductase and 50 mM com-
pound 4f (^), compound 4h (*), compound 4j (~), compound 4i
(&), compound 4k (*), compound 4b (&) and compound 4e (^).
Figure 4. DPPH scavenging (initial concentration 100 mM) by an
ethanolic solution of 100 mM glutathione and 50 mM compound 4f
(^), compound 4h (*), compound 4j (~), compound 4i (&), com-
pound 4a (~), compound 4d (&), compound 4k (*) and diphenyldi-
selenide (^).
Table 1. Antioxidant activities of diselenides 4a–k and reference compounds
Compd
% Catalysisa
% Ebselenb
% Exchangec
DPPH scavenging
% 2 mind
% 4 mine
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
1828
1928
1877
1800
1487
2397
1893
2337
1400
1916
1028
613
298
314
306
293
242
391
309
381
228
312
167
100
261
37.6
48.5
45.6
31.6
16.6
72.8
62.5
80.4
45.8
59.8
27.6
—
34.6
28.3
27.0
28.4
33.6
53.2
35.1
51.2
25.1
39.3
30.9
—
48.7
42.2
39.7
42.6
48.5
67.0
49.6
64.7
37.7
53.0
43.9
—
Ebselen
Diphenyl diselenide
1600
23.0
7.7
13.0
aThe
catalyst’s
percentage
increase
of
the
basal
reaction
rate
between
GSH
and
H2O2
was
calculated
as:
ðrate of NADPH consumption þ catalystÞ=ðrate of NADPH consumption þ vehicleÞ Â 100.
bCalculated as: ðrate of NADPH consumption þ catalystÞ=ðrate of NADPH consumption þ ebselenÞ Â 100.
cFraction of test compound (initial concentration 50 mM) reduced in presence of glutathione (100 mM) after 4-min incubation.
dFraction of DPPH (initial concentration 100 mM) reduced in presence of test compound (50 mM) and glutathione (100 mM) after 2-min incubation.
eAs for d, after 4-min incubation.

4626
F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
scavenging after 4-min incubation). DPPH scavenging
rates were very close for all the other diselenides (40–
50% DPPH scavenging after 4-min incubation). This
must be explained by the rapid reaction between DPPH
and the selenol forms, which displaced in some extent
the exchange equilibriums to the right. All the imidazole
diselenides were largely superior to diphenyl selenide
(13% DPPH scavenging). For ebselen no selenol could
be detected as previously reported.39 The major pro-
ducts formed under the experimental conditions were
the selenenylsulfide conjugate and the diselenide.
Experimental
General chemical methods
Chemicals and reagents were of the highest quality
available and were purchased from Sigma Aldrich
Company. Dichloromethane and triethylamine were
distilled from calcium hydride. Kieselgel 60 (70–230
mesh) and (40–63 mesh) of Merck were used for column
chromatography and flash column chromatography,
respectively. Analytical TLC’s were performed on pre-
coated Merck silica gel 60F254 plates and spots were
detected under UV light and after iodine exposition.
The proton and carbon nuclear magnetic resonance (1H
NMR and 13C NMR) spectra were recorded on a Bru-
ker AM 300 WB (300 MHz) spectrometer. Chemical
shifts are reported from tetramethylsilane as an internal
reference and are given in d/ppm units; coupling con-
stants are given in Hz. Abbreviations used for signal
patterns are: br s, broad singlet; s, singlet; d, doublet; t,
triplet; q, quadruplet; m, multiplet. The 77Se-NMR
spectra were recorded at 25 ꢁC and at 52.22 MHz on a
Bruker DPX 300 (300 MHz) spectrometer using
DMSO-d6 as NMR solvent and diphenyldiselenide as
an external standard. Chemical shifts are reported rela-
tive to dimethylselenide (d 0.0 ppm) by assuming that
the resonance of the standard is at d 461.0 ppm. UV
measurements were made on a Uvikon 932 spectro-
photometer. Mass spectral analyses were performed on
an Applied Biosystems Voyager DE STR mass spectro-
meter (MALDI-TOF). High-resolution mass spectra
were recorded on a Kratos MS80 instrument. Elemental
analyses were determined by the CNRS microanalysis
center. Compounds 1a, 1f–1i, 2a, 2f–2i, 2k were pre-
pared as we have described before.28 Compounds 1b–1e
were synthesised according to a literature procedure.44
Conclusions
Our study provides evidence that diselenides derived
from the naturally occurring ovothiols exhibit anti-
oxidant properties expressed either by their glutathione
peroxidase-like activity or their capacity to be reduced
by glutathione. In this series no clear structure–activity
relationshipemerged from the antioxidant studies. The
most effective compounds were the ones substituted on
position 5 of the imidazole ring, close to the selenium
atom. Thus compounds 4f and 4h were 4 times more
potent than the reference compound, ebselen in the
GSH Px-like test. The introduction of the selenium
atom led to a major increase in the GSH Px-like activity
since the ovothiol-derived 4-mercaptoimidazoles were
approximately 8 times less effective at a 5-fold higher
dose.31
The catalytic cycle of ebselen has been clearly detailed.
It has been proved that ebselen reacts with one equiva-
lent of glutathione to produce a selenenyl sulphide and
then with a second equivalent of glutathione to produce
a selenol. This predominant antioxidant species is
responsible for the GSH Px-like activity of ebselen.40ꢀ42
In presence of hydroperoxides it is oxidised to selenenic
acid, which regenerates the selenenyl sulphide by reac-
tion with glutathione. This catalytic cycle characterises
glutathione peroxidase, certain diaryl diselenides,16,19
diphenyldiselenide19 and a-(phenylselenenyl)ketones.24
Here we have also shown the intervention of the anti-
oxidant selenol form as product from the reaction of
imidazole diselenides with glutathione. The former cat-
alytic cycle can only be hypothetically formulated for
these imidazole diselenides. The catalytic intermediates
remain to be further characterised. The main advantage
of imidazole diselenides over diphenyldiselenide lies in
their ability to be reduced by glutathione to a larger
significant extent. The most potent candidates are 3–5
times more reactive towards glutathione than diphenyl
diselenide, as evidenced by two complementary experi-
ments. These selenol/diselenide exchange reactions of
imidazole diselenides with glutathione may be relevant in
vivo. Partial but significant reduction by GSH may be
sufficient to maintain in living cells a concentration of the
selenol form suitable for an effective antioxidant potency.
2-(3-Chlorophenylamino)acetonitrile (1b). Formaldehyde
(37% solution; 2.70 mL, 36.2 mmol) was added drop-
wise to a mixture of 3-chloroaniline (3.0 g, 23.5 mmol) in
acetic acid (15 mL) and potassium cyanide (1.94 g,
29.8 mmol) in water (3.5 mL) at 0 ꢁC. The solution was
stirred for 15 min at 0 ꢁC and for 4 h at 30 ꢁC. The
solution was diluted with water (50 mL) and CH2Cl2
(150 mL). The organic layer was separated, washed with
1 M aq NaOH and water, dried over Na2SO4 and eva-
porated in vacuo. The residue was chromatographied
on silica gel (solvent system: CH2Cl2) to give 1b as a
ꢁ
white solid in 75% yield; mp64 C; 1H NMR (CDCl3) d
4.02 (d, J=6.8 Hz, 2H), 4.30 (t, J=6.8 Hz, 1H), 6.55 (d,
J=8.0 Hz, 1H), 6.67 (t, J=2.2Hz, 1H), 6.84 (d, J=8.0 Hz,
1H), 7.16 (t, J=8.0 Hz, 1H); 13C NMR (CDCl3) d 32.2,
111.5, 113.3, 116.6, 119.7, 130.4, 135.1, 146.1.
2-(4-Chlorophenylamino)acetonitrile (1c). This com-
pound was prepared in a similar manner to that reported
ꢁ
for 1b; white solid (70%); mp66 C; 1H NMR (CDCl3) d
3.99 (d, J=6.9 Hz, 2H), 4.15 (t, J=6.9 Hz, 1H), 6.58 (d,
J=8.9 Hz, 1H), 7.19 (d, J=8.9 Hz, 2H); 13C NMR
(CDCl3) d 32.7, 114.7, 116.5, 124.7, 129.3, 143.4.
We are presently investigating in vitro the protective effects
of these materials against the cytotoxic oxygen-derived
species. Their lipophilicity will also lead us to evaluate their
antioxidant potency in peroxidation systems.
2-(3-Trifluoromethylphenylamino)acetonitrile (1d). 18 h
reaction at 30 ꢁC. Purification by column chromato-
graphy was not necessary. Brown oil (89%); H NMR
1

F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
4627
1
3
(CDCl3) d 4.03 (d, J=7.1 Hz, 2H), 4.51 (t, J=7.1 Hz,
1H), 6.82 (d, J=8.0 Hz, 1H), 6.93 (s, 1H), 7.11 (d,
J=8.0 Hz, 1H), 7.32 (t, J=8.0 Hz, 1H); 13C NMR
123.2 (q, JCꢀF=272.50 Hz), 124.2 (q, JCꢀF=3.90 Hz),
126.8, 130.7, 131.8 (q, 2JCꢀF=32.70Hz), 139.5, 161.5.
3
(CDCl3) d 31.4, 109.2 (q, JCꢀF=4.0 Hz), 115.4 (q,
N-(Cyanomethyl)-N-(4-trifluoromethylphenyl)formamide
(2e). Column chromatography on silica gel (solvent sys-
tem: petroleum ether–ethyl acetate, 70:30). Beige solid
(72%); mp48 C; 1H NMR (CDCl3) d 4.65 (s, 2H), 7.32
3JCꢀF=4.0 Hz), 115.9, 116.7, 123.8 (q, 1JCꢀF=272.5 Hz),
129.6, 131.0 (q, 2JCꢀF=31.8 Hz), 145.2.
ꢁ
2 - (4 - Trifluoromethylphenylamino)acetonitrile (1e). 18 h
reaction at 30 ꢁC. Column chromatography on silica gel
(solvent system: CH2Cl2–MeOH, 95:5). White solid
(89%); mp106 ꢁC; 1H NMR (CDCl3) d 4.14 (d,
J=6.8 Hz, 2H), 4.21 (t, J=6.8 Hz, 1H), 6.73 (d,
J=8.6 Hz, 2H), 7.50 (d, J=8.6 Hz, 2H); 13C NMR
(d, J=8.3 Hz, 2H), 7.62 (d, J=8.3 Hz, 2H), 8.43 (s, 1H);
13C NMR (CDCl3) d 32.3, 114.8, 122.5, 123.3 (q,
3
1JCꢀF=272.1 Hz), 126.7 (q, JCꢀF=3.90 Hz), 127.0 (q,
3JCꢀF=3.90 Hz), 128.7 (q, 2JCꢀF=33.0 Hz), 141.9, 161.2.
N-[1-Cyano-1-(4-trifluoromethylphenyl)methyl]-N-methyl-
formamide (2j). Column chromatography on silica gel
(solvent system: petroleum ether–ethyl acetate, 70:30).
2
(CDCl3) d 32.0, 112.8, 116.3, 121.7 (q, JCꢀF=32.7 Hz),
1
3
124.5 (q, JCꢀF=270.6 Hz), 126.9 (q, JCꢀF=3.7 Hz),
147.6.
1
Yellow oil (83%); H NMR (CDCl3) d 2.59, 2.74 (2s,
ratio 1:4, 3H), 6.09, 6.74 (2s, ratio 1:4, 1H), 7.43 (d,
J=8.3 Hz, 2H), 7.54 (d, J=8.3 Hz, 2H), 8.05, 8.37 (2s,
ratio 4:1, 1H); 13C NMR (CDCl3) d 26.0, 30.0, 45.7,
2-(Methylamino)-2-(4-trifluoromethylphenyl)acetonitrile
(1j). A solution of 4-(trifluoromethyl)benzaldehyde
(3.00 g, 17.2 mmol) in methanol (5 mL) was added
dropwise to a solution of potassium cyanide (1.35 g,
20.7 mmol) and methylamine hydrochloride (1.40 g,
20.7 mmol) in water (5 mL) at 0 ꢁC. The solution was
stirred for 1 h at 0 ꢁC and for 24 h at 30 ꢁC. After
extraction with CH2Cl2, drying (Na2SO4) and concen-
tration in vacuo of the extract, the residue was chroma-
tographied on silica gel (solvent system: CH2Cl2) to give
compound 1j as a yellow oil (77%). 1H NMR (CDCl3) d
1.75 (br s, 1H), 2.50 (s, 3H), 5.20 (s, 1H), 7.43 (d,
J=8.2 Hz, 2H), 7.57 (d, J=8.2 Hz, 2H); 13C NMR
1
52.9, 114.5, 114.8, 123.2 (q, JCꢀF=272.2 Hz), 125.5 (q,
3
3JCꢀF=3.90 Hz), 125.9 (q, JCꢀF=3.90 Hz), 127.1,
127.3, 130.9 (q, 2JCꢀF=32.7 Hz), 131.2 (q,
2JCꢀF=32.7 Hz), 134.7, 134.9, 156.1, 156.3.
Bis-4-[1-[(40 -methoxy)phenyl]-4-seleno-imidazole] (4a).
Method A. The all-glass reaction apparatus consisted of
a 100 mL two-neck reaction flask (A) in which hydrogen
selenide was generated by the action of H2O upon
Al2Se3. This was connected to a 200 mL three-neck
reaction flask (B) by means of a glass tube. A granular
anhydrous calcium chloride cartridge was placed
between the two flasks, which served to dry hydrogen
selenide prior to its contact with the cyano amide com-
pound 2a in flask B. Flask A was charged with alumi-
nium selenide (0.78 g, 2.7 mmol) and fitted with a
dropping funnel containing 3 mL of water. A solution of
compound 2a (0.36 g, 1.9 mmol) and triethylamine
(0.79 mL, 5.7 mmol) in CH2Cl2 (50 mL) was prepared in
flask B. The apparatus was then evacuated and flushed
with dry argon.
1
(CDCl3) 27.0, 48.3, 114.1, 122.4 (q, JCꢀF=272.2 Hz),
125.0
(q,
3JCꢀF=3.9 Hz),
126.8,
130.6
(q,
2JCꢀF=32.7 Hz), 134.7.
N - (Cyanomethyl) - N - (3 - chlorophenyl)formamide (2b).
Acetic anhydride (36.1 mL, 383 mmol) was added drop-
wise to a 96% formic acid solution (14.4 mL, 383 mmol)
at 0 ꢁC. The mixture was stirred for 15 min at 60 ꢁC. The
mixed anhydride was then treated dropwise with the
cyano amine 1b (2.90 g, 17.4 mmol) in CH2Cl2 (15 mL)
at 0 ꢁC. Stirring was continued for 1 h at 0 ꢁC and 2 h at
room temperature. After several washings with 1 M aq
NaOH and water, drying of the organic layer (Na2SO4)
and removal of the volatiles in vacuo, the residue was
chromatographied on silica gel (solvent system:
CH2Cl2–MeOH, 98:2) to give compound 2b as a col-
Water was added dropwise to the aluminium selenide so
that the rate of evolution of H2Se gas was constant. The
mixture was stirred until there was not any evolution of
the content of flask B (TLC analysis) and cooled at
ꢀ10 ꢁC. Then trimethylsilyltrifluoromethanesulfonate
(0.69 mL, 3.8 mmol) was added dropwise and advance-
ment of the reaction was followed by TLC. If the reac-
tivity of the silane was too low after 1 h stirring, 3 equiv
(0.79 mL) of TEA and 2 equiv (0.69 mL) of Me3SiOTf
could be further added and reaction was run for 15 h at
0 ꢁC. The mixture was washed with water. The organic
layer was dried (Na2SO4) and evaporated in vacuo. The
residue was chromatographied on silica gel (solvent
system: CH2Cl2–MeOH, 98:2) to give 4a, which crystal-
lised as an orange solid in ethyl acetate in 30% yield.
1
ourless oil (74% yield). H NMR (CDCl3) d 4.58 (s,
2H), 7.11 (d, J=8.0 Hz, 1H), 7.20 (t, J=2.2 Hz, 1H),
7.24 (d, J=8.0 Hz, 1H), 7.32 (t, J=8.0 Hz, 1H), 8.30 (s,
1H); 13C NMR (CDCl3) d 32.3, 114.6, 121.1, 123.0,
127.3, 130.6, 134.6, 139.6, 161.0.
N - (Cyanomethyl) - N - (4 - chlorophenyl)formamide (2c).
Same method as for the preparation of 2b. 15 h reac-
tion at room temperature. Beige solid (80%); mp 53 ꢁC;
1H NMR (CDCl3) d 4.60 (s, 2H), 7.19 (d, J=8.9 Hz,
2H), 7.40 (d, J=8.9 Hz, 2H), 8.31 (s, 1H); 13C NMR
(CDCl3) d 32.4, 114.5, 124.6, 129.5, 132.9, 136.9, 160.9.
Method B. To a solution of compound 2a (0.60 g,
3.1 mmol) and triethylamine (0.48 mL, 3.4 mmol) in dry
ethanol (25 mL) were added finely ground gray selenium
powder (1.5 g, 19.0 mmol) and sodium borohydride
(0.48 g, 12.7 mmol). The resulting mixture was stirred at
N-(Cyanomethyl)-N-(3-trifluoromethylphenyl)formamide
(2d). Yellow oil (76%); 1H NMR (CDCl3) d 4.67 (s, 2H),
7.44–7.54 (m, 2H), 7.58–7.60 (m, 2H) 8.39 (s, 1H); 13C
NMR (CDCl3) d 32.7, 114.9, 120.2 (q, 3JCꢀF=3.90 Hz),

4628
F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
room temperature until formation of selenoamide 3a
was maximal and was then slightly refluxed overnight.
Ethanol was removed in vacuo and CH2Cl2 (100 mL)
was added. The solid residues were filtered. The
organic phase was extracted with 0.2 M aq HCl and
water, dried over Na2SO4 and the solvent was
removed in vacuo. Column chromatography of the
residue on silica gel using 98/2 CH2Cl2/MeOH affor-
ded the title compound 4a, which crystallised in ethyl
acetate in 36% yield. Orange powder; mp 127 C; H
NMR (CDCl3) d 3.84 (s, 3H), 6.96 (d, J=9.0 Hz, 2H),
7.30 (d, J=9.0 Hz, 2H), 7.46 (s, 1H), 7.77 (s, 1H); 13C
NMR (CDCl3) d 55.4, 114.8, 122.9, 124.2, 128.5,
129.8, 136.6, 159.0; 77Se NMR (DMSO-d6) d 422.5; m/
z (MALDI-TOF) 507.0 (MH+, 100); HRMS calcd
(M+) 505.9760; Found: 505.9767. Anal. calcd for
C20H18N4O2Se2: C, 47.63; H, 3.60; N, 11.11; Found:
C, 47.71; H, 3.58; N, 11.16.
(MH+, 100); HRMS calcd (M+) 581.9296; Found:
581.9301. Anal. calcd for C20H12F6N4Se2: C, 41.40; H,
2.08; N, 9.66; Found: C, 41.34; H, 2.07; N, 9.71.
Bis-4-[5-[(40-methoxy)phenyl]-4-seleno-1-methylimidazole]
(4f). Method A: 41% yield; Method B: 53% yield.
Orange powder; mp 160 ꢁC; 1H NMR (CDCl3) d 3.47 (s,
3H), 3.55 (s, 3H), 6.84 (d, J=8.8 Hz, 2H), 7.10 (d,
J=8.8 Hz, 2H), 7.46 (s, 1H); 13C NMR (CDCl3) d 32.8,
55.3, 113.6, 121.3, 125.7, 131.7, 137.5, 138.4, 159.6;
77Se NMR (DMSO-d6) d 430.2; m/z (MALDI-TOF)
534.9 (MH+, 100); HRMS calcd (M+) 534.0073;
Found: 534.0067. Anal. calcd for C22H22N4O2Se2: C,
49.64; H, 4.17; N, 10.52; Found: C, 49.73; H, 4.21; N,
10.57.
ꢁ
1
Bis-4-[5-[(20 -chloro)phenyl]-4-seleno-1-methylimidazole]
(4g). Method A: 43% yield; Method B: 37% yield.
Orange powder; mp 210 ꢁC; 1H NMR (CDCl3) d 3.46 (s,
3H), 7.14–7.48 (m, 4H), 7.51 (s, 1H); 13C NMR (CDCl3)
d 32.7, 125.9, 126.6, 126.8, 128.1, 129.6, 130.8, 133.8,
135.3, 139.0; 77Se NMR (DMSO-d6) d 420.3; m/z
(MALDI-TOF) 542.9 (MH+, 100); HRMS calcd (M+)
541.9082; Found: 541.9078. Anal. calcd for
C20H16Cl2N4Se2: C, 44.39; H, 2.98; N, 10.35; Found: C,
44.48; H, 2.93; N, 10.34.
Bis - 4 - [1 - (30 - chlorophenyl) - 4 - selenoimidazole] (4b).
Method A: 27% yield; Method B: 23% yield.
Orange powder; mp 208 ꢁC; 1H NMR (DMSO-d6) d
7.43 (d, J=8.0 Hz, 1H), 7.51 (d, J=8.0 Hz, 1H), 7.65
(d, J=8.0 Hz, 1H), 7.85 (t, J=2.0 Hz, 1H), 8.05 (s,
1H), 8.44 (s, 1H); 13C NMR (DMSO-d6) d 112.4,
113.8, 117.3, 120.7, 121.2, 125.0, 127.8, 130.8, 130.9;
77Se NMR (DMSO-d6) d 420.7; m/z (MALDI-TOF)
514.9 (MH+, 100); HRMS calcd (M+) 513.8769;
Found: 513.8761. Anal. calcd for C18H12Cl2N4Se2: C,
42.13; H, 2.36; N, 10.92; Found: C, 42.07; H, 2.30;
N, 11.02.
Bis-4-[5-[(40-chloro)phenyl]-4-seleno-1-methyl-imidazole]
(4h). Method A: 31% yield; Method B: 42% yield.
Orange powder; mp 238 ꢁC; 1H NMR (CDCl3) d 3.52 (s,
3H), 7.15 (d, J=8.4 Hz, 2H), 7.32 (d, J 8.4 Hz, 2H), 7.62
(s, 1H); 13C NMR (CDCl3) d 33.0, 125.9, 127.3, 128.5,
131.7, 134.6, 136.3, 138.9; 77Se NMR (DMSO-d6) d
425.9; m/z (MALDI-TOF) 542.9 (MH+, 100); HRMS
calcd (M+) 541.9082; Found: 541.9089. Anal. calcd for
C20H16Cl2N4Se2: C, 44.39; H, 2.98; N, 10.35; Found: C,
44.36; H, 3.02; N, 10.32.
Bis - 4 - [1 - (40 - chlorophenyl) - 4 - selenoimidazole] (4c).
Method A: 38% yield; Method B: 29% yield. Orange
brown powder; mp 215 ꢁC; 1H NMR (CDCl3) d 7.29 (d,
J=8.7 Hz, 2H), 7.38 (s, 1H), 7.44 (d, J=8.7 Hz, 2H),
7.79 (s, 1H); 13C NMR (CDCl3) d 115.9, 117.6, 121.4,
123.5, 125.3, 128.8, 130.9; 77Se NMR (DMSO-d6) d
421.2; m/z (MALDI-TOF) 514.9 (MH+, 100); HRMS
calcd (M+) 513.8769; Found: 513.8766. Anal. calcd for
C18H12Cl2N4Se2: C, 42.13; H, 2.36; N, 10.92; Found: C,
42.16; H, 2.44; N, 10.97.
Bis-4-[5-[(20 -trifluoromethyl)phenyl]-4-seleno-1-methyl
imidazole] (4i). Method A: 30% yield; Method B: 26%
yield. Lemon powder; mp 224 ꢁC; H NMR (CDCl3) d
1
3.29 (s, 3H), 6.87 (s, 1H), 7.25–7.71 (m, 4H); 13C NMR
1
(CDCl3) d 32.0, 123.4 (q, JCꢀF=274.0 Hz), 126.2 (q,
Bis-4-[1-[(30 -trifluoromethyl)phenyl]-4-selenoimidazole]
(4d). Method A: 37% yield; Method B: 26% yield.
3JCꢀF=5.0 Hz), 127.7, 129.7, 130.8 (q, JCꢀF=33.0 Hz),
2
131.2, 131.7, 134.0 (q, 3JCꢀF=5.0 Hz), 138.2,
138.5;77SeNMR (DMSO-d6) d 419.4; m/z (MALDI-
TOF) 611.0 (MH+, 100); HRMS calcd (M+) 609.9609;
Found: 609.9615. Anal. calcd for C22H16F6N4Se2: C,
43.44; H, 2.65; N, 9.21; Found: C, 43.52; H, 2.60; N,
9.27.
ꢁ
1
Lemon powder; mp 181 C; H NMR (CDCl3) d 7.60–
7.64 (m, 5H), 7.90 (s, 1H); 13C NMR (CDCl3) d 118.3
3
1
(q, JCꢀF=3.9 Hz), 122.1, 123.2 (q, JCꢀF=272.0 Hz),
3
123.6, 124.6 (q, JCꢀF=3.9 Hz), 129.9, 130.8, 132.6 (q,
2JCꢀF=33.0 Hz), 136.4, 137.1; 77Se NMR (DMSO-d6) d
420.2; m/z (MALDI-TOF) 582.9 (MH+, 100); HRMS
calcd (M+) 581.9296; Found: 581.9288. Anal. calcd for
C20H12F6N4Se2: C, 41.40; H, 2.08; N, 9.66; Found: C,
41.50; H, 1.99; N, 9.63.
Bis-4-[5-[(40 -trifluoromethyl)phenyl]-4-seleno-1-methyl
imidazole] (4j). Method A: 28% yield; Method B: 33%
yield. Lemon powder; mp 180 ꢁC; H NMR (CDCl3) d
1
3.53 (s, 3H), 7.35 (d, J=8.5 Hz, 2H), 7.49 (s, 1H),
7.62 (d, J=8.5 Hz, 2H); 13C NMR (CDCl3) d 33.0,
Bis-4-[1-[(40 -trifluoromethyl)phenyl]-4-selenoimidazole]
(4e). Method A: 32% yield; Method B: 29% yield.
Lemon powder; mp 189 ꢁC; 1H NMR (DMSO-d6) d
7.82 (d, J=8.9 Hz, 2H), 7.87 (d, J=8.9 Hz, 2H), 8.07 (s,
1H), 8.51 (s, 1H); 13C NMR (DMSO-d6) d 120.9, 124.0
1
3
123.9 (q, JCꢀF=272.2 Hz), 125.1 (q, JCꢀF=3.9 Hz),
2
126.7, 130.3 (q, JCꢀF=33.0 Hz), 130.7, 132.7, 136.0,
139.2; 77Se NMR (DMSO-d6) d 423.7; m/z (MALDI-
TOF) 611.0 (MH+, 100); HRMS calcd (M+)
609.9609; Found: 609.9605. Anal. calcd for
C22H16F6N4Se2: C, 43.44; H, 2.65; N, 9.21; Found: C,
43.37; H, 2.72; N, 9.15.
1
3
(q, JCꢀF=272.2 Hz), 124.1, 127.2 (q, JCꢀF=3.9 Hz),
2
127.6 (q, JCꢀF=32.3 Hz), 128.1, 137.5, 139.3; 77Se
NMR (DMSO-d6) d 420.2; m/z (MALDI-TOF) 582.9

F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
4629
Bis-4-[2-[(30-chloro)phenyl]-4-seleno-1-methyl-imidazole]
References and Notes
(4k). Method A: 23% ꢁyield; Method B: 30% yield.
Lemon powder; mp 154 C; 1H NMR (CDCl3) d 3.75 (s,
3H), 7.32 (s, 1H), 7.36–7.51 (m, 3H), 7.67 (s, 1H); 13C
NMR (CDCl3) d 34.7, 126.6, 127.6, 128.8, 128.9, 129.0,
129.7, 131.5, 134.5, 147.4; 77Se NMR (DMSO-d6) d
420.3; m/z (MALDI-TOF) 542.9 (MH+, 100); HRMS
calcd (M+) 541.9082; Found: 541.9089. Anal. calcd for
C20H16Cl2N4Se2: C, 44.39; H, 2.98; N, 10.35; Found: C,
44.36; H, 3.05; N, 10.43.
1. Neve, J. Biological Functions of Selenium. In Selenium in
Medicine and Biology; Neve, J., Favier, A., Eds.; De Gruyter
W.: Berlin, 1988; p97.
2. Burk, R. F., Ed. Selenium in Biology and Human Health;
Springer: New York, 1994.
3. Schamberger, R. J. Biochemistry of Selenium; Plenum: New
York, 1983.
4. Klayman, D. L., Gunther, W. H. H., Eds. Organic Sele-
nium Compounds: Their Chemistry and Biology; Wiley: New
York, 1973.
5. Parnham, M. J.; Graf, E. Prog. Drug. Res. 1991, 36, 9.
6. Gunzler, W. A.; Steffens, G. J.; Grossman, A. Hoppe-Sey-
ler’s Z. Physiol. Chem. 1982, 365, 194.
7. Reich, H. J.; Jasperse, C. P. J. Am. Chem. Soc. 1987, 109,
5549.
8. Vitoux, D.; Chappuis, P.; Arnaud, J.; Bost, M.; Accom-
inotti, M.; Roussel, A. M. Ann. Biol. Clin. (Paris) 1996, 54,
181.
Antioxidant assays
Drug preparation. Compounds were solubilised in N,N-
dimethylformamide (DMF) to yield 10 or 5 mM stock
solutions. Ebselen and diphenyldiselenide were used as
references.
Measurement of the glutathione peroxidase-like (GSH
Px-like) activity. GSH Px-like activity was determined
by the reduction of GSSG formed via the NADPH–
glutathione reductase system as a continuous indicator
system.11 Experiments were conducted at room tem-
perature (23 ꢁC). The decrease in NADPH monitored
spectrophotometrically at 366 nm (e=3300/M/cm) is a
measure of GSH Px activity. The assay mixture (1.0 mL)
contained 50 mM phosphate buffer (prepared with
Milli-Q water, conductivity <10ꢀ18/ꢀ/cm), pH 7.6,
0.1 mM EDTA and 1.0 mM NaN3. Test compound
(50 mM final concentration), glutathione (1.0 mM),
NADPH (280 mM) and GSSG reductase (1 unit) were
added and the absorbance was recorded for 4 min to
estimate the background and stability of the prepara-
tion. Reaction was initiated by the subsequent addition
of hydrogen peroxide at concentration 1.0 mM. Appro-
priate blanks were also run in the absence of test com-
pound and in the presence of H2O2. For compounds 4b–
e and diphenyldiselenide, DMF (200 mL) was included
in the incubations to improve the solubility of test
compounds in buffer. Catalyst activity (% catalysis) was
determined as the increase of the basal reaction between
GSH and H2O2 in presence of test compound (as
compared with the uncatalysed process in presence of
vehicle).
9. Sies, H. In Oxidative Stress; Sies, H., Ed.; Academic: Lon-
don and New York, 1991.
10. Parnham, M. J.; Graf, E. Biochem. Pharmacol. 1987, 36,
3095.
11. Muller, A.; Cadenas, E.; Graf, P.; Sies, H. Biochem.
Pharmacol. 1984, 33, 3235.
12. Wendel, A.; Fausel, M.; Safayhi, H.; Tiegs, G.; Otter, R.
Biochem. Pharmacol. 1984, 33, 3241.
13. Jacquemin, P. V.; Christiaens, L. E.; Renson, M. J.; Evers,
M. J.; Dereu, N. Tetrahedron Lett. 1992, 33, 3863.
14. Galet, V.; Bernier, J. L.; Henichart, J. P.; Lesieur, D.;
Abadie, C.; Rochette, L.; Lindenbaum, A.; Chalas, J.; Renaud
de la Faverie, J. F.; Pfeiffer, B.; Renard, P. J. Med. Chem.
1994, 37, 2903.
15. Back, T. G.; Dyck, B. P. J. Am. Chem. Soc. 1997, 119,
2079.
16. Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1994, 116,
2557.
17. Tailhan-Lomont, C.; Erdelmeier, I.; Moutet, M.; Chau-
diere, J.; Yadan, J.C. European Patent Application EP 850,
924, A1, 1998.
18. Wirth, T. Molecules 1998, 3, 164.
19. Mugesh, G.; Panda, A.; Singh, H. B.; Punckar, N. S.;
Butcher, R. J. J. Am. Chem. Soc. 2001, 123, 839.
20. Mugesh, G.; Panda, A.; Singh, H. B.; Punckar, N. S.;
Butcher, R. J. Chem. Commun. 1998, 2227.
21. Wilson, S. R.; Zucker, P. A.; Huang, R.-R. C.; Spector, A.
J. Am. Chem. Soc. 1989, 111, 5936.
22. Wu, Z.-P.; Hilvert, D. J. Am. Chem. Soc. 1990, 112, 5647.
23. Engman, L.; Stern, D.; Frisell, H.; Vessman, K.; Ber-
glund, M.; Ek, B.; Andersson, C. M. Bioorg. Med. Chem.
1995, 3, 1255.
24. Engman, L.; Andersson, C.; Morgenstern, R.; Cotgreave,
I. A.; Andersson, C. M.; Halberg, A. Tetrahedron 1994, 50,
2929.
25. Back, T. G.; Moussa, Z. J. Am. Chem. Soc. 2002, 124,
12104.
26. Mugesh, G.; du Mont, W.-W.; Sies, H. Chem. Rev. 2001,
101, 2125.
27. Mugesh, G.; Singh, H. B. Chem. Soc. Rev. 2000, 29, 347.
28. Zoete, V.; Bailly, F.; Catteau, J. P.; Bernier, J. L. J. Chem.
Soc., Perkin Trans. 1 1997, 2983.
29. Zoete, V.; Bailly, F.; Vezin, H.; Teissier, E.; Duriez, P.;
Fruchart, J. C.; Catteau, J. P.; Bernier, J. L. Free Radic. Res.
2000, 32, 515.
30. Zoete, V.; Vezin, H.; Bailly, F.; Vergoten, G.; Catteau,
J. P.; Bernier, J. L. Free Radic. Res. 2000, 32, 525.
31. Bailly, F.; Zoete, V.; Vamecq, J.; Catteau, J. P.; Bernier,
J. L. FEBS Lett. 2000, 486, 19.
Exchange reaction with glutathione. Two methods were
used to investigate the exchange reaction of diselenides
with glutathione. The first one used the NADPH-glu-
tathione reductase system, which reduced GSSG formed
upon exchange reaction. For this purpose, diselenide
(50 mM) and glutathione (100 mM) were incubated for 4
min at room temperature. NADPH (100 mM) and
GSSG reductase (1 U/mL) were subsequently added.
The absorbance decrease at 366 nm was measured and
compared to that obtained in presence of glutathione
disulphide alone (50 mM). The ratio of both NADPH
losses gave the amount of glutathione/diselenide
exchange. The second method used the stable radical
DPPH. Diselenide (50 mM) and glutathione (100 mM)
were added to 1.0 mL of a solution of DPPH (100 mM)
in ethanol. The absorbance decrease at 517 nm was
measured for 10 min. Control was also run in absence of
test compound.

4630
F. Bailly et al. / Bioorg. Med. Chem. 11 (2003) 4623–4630
32. Spaltenstein, A.; Holler, T. P.; Hopkins, P. B. J. Org.
Chem. 1987, 52, 2977.
33. Andrew, D.; Hastings, D. J.; Weedon, A. C. J. Am. Chem.
Soc. 1994, 116, 10870.
39. Haenen, G. R. M. M.; D Rooij, B. M.; Vermeulen,
N. P. E.; Bast, A. Mol. Pharmacol. 1990, 37, 412.
40. Maiorino, M.; Roveri, A.; Coassin, M.; Ursini, F. Bio-
chem. Pharmacol. 1988, 37, 2267.
34. Wright, S. W. Tetrahedron Lett. 1994, 35, 1331.
35. Klayman, D. L.; Griffin, T. S. J. Am. Chem. Soc. 1973, 95,
197.
36. Lai, L.-L.; Reid, D. H. Synthesis 1993, 870.
37. Ruan, H.-D.; Zhang, P-F.; Tao, Y.; Fan, W-Q. Synth.
Commun. 1996, 26, 2617.
41. Morgenstern, R.; Cotgreave, I. A.; Engman, L. Chem-
Biol. Ineract. 1992, 84, 77.
42. Cotgreave, I. A.; Morgenstern, R.; Engman, L.; Ahokas,
J. Chem-Biol. Interact. 1992, 84, 69.
43. Chaudiere, J.; Yadan, J.C.; Erdelmeier, I.; Tailhan-
Lomont, C.; Moutet, M. In Oxidative Processes and Anti-
oxidants; Paoletti, R., Ed.; Raven: New York, 1994; p165.
44. Marxer, A. Helv. Chim. Acta 1954, 37, 166.
38. Weaver, K. H.; Rabenstein, D. L. J. Org. Chem. 1995, 60,
1904.