Synthesis, electrochemical, and spectroscopic studies of novel S-nitrosothiols

Article abstract of DOI:10.1016/S0040-4020(01)00694-9

A series of S-nitrosothiol compounds, structurally related to the NO-donor S-nitroso-N-acetyl-D,L-penicillamine (SNAP) that contain amidin groups were synthesised by S-nitrosation of the corresponding thiols and characterised. The kinetics of decomposition were investigated and showed that the two adenine-based thionitrites exhibited an unusual stability in aqueous solution compared to SNAP, suggesting that these compounds may complex the traces of free copper ions present in solution, which is known to catalyze the decomposition of thionitrites. The electrochemical behaviour of these compounds and their nitric oxide-releasing potential were studied by means of cyclic voltammetry techniques on mercury and glassy carbon electrodes.

Full text of DOI:10.1016/S0040-4020(01)00694-9

TETRAHEDRON
Pergamon
Tetrahedron 57 /2001) 7173±7180
Synthesis, electrochemical, and spectroscopic studies of
novel S-nitrosothiols
a
Laurent Soulere, Juan-Carlos Sturm, Luis J. NunÄez-Vergara, Pascal Hoffmann
b
b
a,p
Á
Â
 Âa
and Jacques Perie
a
Á Â Â Ã
Groupe de Chimie Organique Biologique, Laboratoire de Synthese et Physico-Chimie de Molecules d'Interet BiologiqueÐESA-CNRS
Â
5068, Universite de Toulouse III, 118 route de Narbonne, 31062 Toulouse cedex 4, France
bLaboratory of Bioelectrochemistry, Faculty of Chemical and Pharmaceutical Sciences, University of Chile, Casilla 233, Santiago I, Chile
Received 4 April 2001; revised 5 June 2001; accepted 4 July 2001
AbstractÐA series of S-nitrosothiol compounds, structurally related to the NO-donor S-nitroso-N-acetyl-d,l-penicillamine /SNAP) that
contain amidin groups were synthesised by S-nitrosation of the corresponding thiols and characterised. The kinetics of decomposition were
investigated and showed that the two adenine-based thionitrites exhibited an unusual stability in aqueous solution compared to SNAP,
suggesting that these compounds may complex the traces of free copper ions present in solution, which is known to catalyze the decom-
position of thionitrites. The electrochemical behaviour of these compounds and their nitric oxide-releasing potential were studied by means
of cyclic voltammetry techniques on mercury and glassy carbon electrodes. q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
2. Results and discussion
A large body of evidence suggests that nitric oxide
contributes to antiparasitic defence mechanisms during the
infection, and the detrimental effect of NOz from NOz-donors
including S-nitroso-N-acetyl-d,l-penicillamine /SNAP) on
the growth of several parasites such as Trypanosoma cruzi
or Leishmania major has become established.1 This effect is
mainly attributable to the near diffusion-controlled reaction
of nitric oxide and superoxide /O2z2) to form peroxynitrite
/ONOO2), a potent oxidant capable of generating highly
reactive species.2 In a preceding study,3 we detailed the
evaluation of a series of nitric oxide-releasing compounds
that contained amidin groups for their ability to inhibit the
trypanosomal P2 transporter4-mediated uptake of adenosine
on Trypanosoma equiperdum. These compounds were
speci®cally designed to deliver nitric oxide into the parasite
via this purine transporter, so accumulating and displaying
its cytotoxic effect. The S-nitrosothiol /thionitrite) deriva-
tives that bear a melaminyl, adenine or adenosine moiety
were found to have a high af®nity for the P2-transporter
suggesting that these recognition structures could be con-
sidered as an import signal for such drugs. In what follows,
we extend this study to the details of the synthesis, the
decomposition in aqueous solution and the electrochemical
characterisation of these compounds.
2.1. Synthesis
2.1.1. Synthesis of thiols 3a±d ꢀScheme 1). Commercially
Scheme 1. /a) Acetic anhydride, pyridine, 258C, 20 h, 44%. /b) 3a: 2,5-
Diamino-pyridine, CHCl3, 5 h, 258C, 61%; 3b: 2-/4-Aminophenylamino)-
4,6-diamino-1,3,5-triazine /5), DMF, 258C, 18 h, 30%; 3c: 9-/2-
Aminoethyl)-9H-purin-6-ylamine /6), CHCl3/1N NaOH, 258C, 2 h, 61%;
3d: /i) 50-amino-20,30-/O-isopropylidene)-50-deoxyadenosine, CH2Cl2,
258C, 5 h, 85%; /ii) HCOOH/H2O, 258C, 48 h, 100%.
Keywords: thionitrites; nitric oxide; purine; electrochemistry.
Corresponding author. Tel.: 133-05-61-55-64-86; fax: 133-05-61-55-
60-11; e-mail: hoffmann@cict.fr
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020/01)00694-9

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7174
Table 1. NMR Characterization of S-nitrosothiols
Dd /d4i2d3i) /ppm)
1H NMRa 13C NMRb
i
a
0.52±0.58
0.73±0.74
0.59±0.56
0.60±0.60
0.60±0.63
0.67±0.68
12.9
13.7
13.5
13.3
13.2
12.7
b
c
dc
SNAP
a
NMR-shifts of the two inequivalent gem-methyl groups recorded in D6-
DMSO for all compounds except 4b in D5-pyridin.
NMR-shift of /CH3)2-C- recorded in D6-DMSO for all compounds except
4b in D5-pyridin.
b
c
Mixture of the two diastereoisomers.
Scheme 2. /c) /i) 4-Aminoacetanilide, H2O, NaOH 1 equiv., 4.5 h, 1008C,
92%; /ii) aq. HCl, 1.5 h, 1008C, 37%. /d) /i) 2-Bromo-1-[/tert-butyloxycar-
bonyl)amino]ethane, K2CO3, DMF, 18 h, 258C /59%); /ii) TFA, CH2Cl2,
258C, 16 h, 100%.
conditions was preferable to the previous method, giving a
yield of 83%. All thionitrites were stable at the solid stage
and could be stored for long periods. They were charac-
terised by NMR, UV±visible and IR spectroscopies.
Compared to the parent thiols, the introduction of the
nitroso group caused a high deshielding effect on the gem-
dimethyl proton resonances, as well as a high down®eld
shift /about 13 ppm) observed on the carbon bearing the
two methyl groups /Table 1).
available d,l-penicillamine 1 was converted to 3-acetami-
do-4,4-dimethylthietan-2-one 2 via cyclisation and con-
comitant acetylation reaction. This reaction provided a
mixture of the desired thietanone 2 /yield: 44%) and of an
apolar product corresponding to 4-isopropylidene-2-
methyloxazol-5/4H)-one which were easily separated.
This activation of penicillamine via its thietanone derivative
had previously been described by Al-Zaidi et al.5 and used
by others for coupling to a number of nucleophiles.6,7
Reaction of 2 with 2,5-diamino pyridine in chloroform at
room temperature gave thiol 3a. Thiol 3b was synthesised
from thietanone 2 and melaminyl derivative 5 /Scheme 2),
which was obtained from 2-chloro-4,6-diamino-1,3,5-tria-
zine and 4-aminoacetanilide and subsequent deacetylation.
Reaction of 2 with adenine derivative 6 /Scheme 2)
gave thiol 3c. Thiol 3d was obtained from 50-amino-20,30-
/O-isopropylidene)-50-deoxyadenosine8 and subsequent
deacetylation.
2.2. Decomposition studies
Decomposition of S-nitrosothiol compounds occurs through
a homolytic cleavage of the S±N bond as primary process to
yield nitric oxide radical and the corresponding thiyl radical,
which dimerizes to give the disul®de. The rate of decom-
position is strongly dependent on the inherent structure of
thionitrites. Most of them are relatively unstable at the solid
stage or in solution to allow a full structural characterisation
or a therapeutic use, but some of them including SNAP,9
S-nitrosoglutathione /GSNO),10 certain thiol proteins /such
as S-nitrosoalbumine) or other structures11 exhibit signi®-
cant stability. The decomposition has been shown to be
catalyzed by a number of parameters including metal ions,
light, pH or temperature; thus as a consequence, stability
studies must be investigated under strictly identical condi-
tions so as to be compared. As thionitrites 4a and 4b were
not water-soluble, we used a mixture of DMSO in a 100 mM
phosphate buffer /pH 7.0) to study their decomposition. The
kinetics were carried out spectrophotometrically following
the disappearance of absorbances at 340 nm for 4b and at
360 nm for 4a, which exhibited a shoulder at around
400 nm. In these conditions, without addition of a metal
2.1.2. Synthesis and spectral properties of S-nitrosothiols
4a±d ꢀScheme 3). S-Nitrosothiols 4a±d were synthesised in
mild conditions by electrophilic nitrosation of their corre-
sponding parent thiols 3a±d. These nitrosation reactions to
obtain the thionitrites required an investigation of various
methods. Nitroso groups are commonly introduced using
either an organic nitrite or sodium nitrite in acetic condi-
tions. Thionitrite 4a was obtained by using tert-butyl nitrite
in chloroform, while 4c and 4d were synthesised with the
same nitrosating reagent, but in acetone as solvent. We used
these two different solvents, which led to precipitation of the
desired products. In contrast, attempts to synthesize 4b by
this method were unsuccessful. In that case, the use of
sodium nitrite as nitrosating agent in acidic methanolic
Scheme 3. /e) 4a: /i) aq. HCl, ethanol, 2 h, 258C; /ii) tert-butyl nitrite,
chloroform, 35 min, 258C /43%). 4b: NaNO2, CH3OH, H2SO4, 1 M HCl,
15 min, 258C /83%). 4c: tert-Butyl nitrite, acetone, 30 min, 258C /62%).
4d: tert-Butyl nitrite, acetone, 30 min, 258C /63%).
Figure 1. Comparative decomposition plots of SNAP, 4c and d /2 mM) at
pH 7.0 /100 mM phosphate buffer) at 258C in the absence of EDTA.

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7175
ion chelator /such as EDTA), reasonable ®rst-order plots
/vs Ag/AgCl) in accordance with the equation:
were obtained with half-life values of around 3 h for both
compounds, i.e. the same order of stability as SNAP in
identical experimental conditions. Decomposition studies
of thionitrites 4c and 4d were followed spectrophotometri-
cally at 258C under stirring by checking the decreasing
absorbance at 340 nm in a 0.1 M phosphate buffer /pH
7.0) in the absence of EDTA. The kinetic experiments
showed that these two adenine-related thionitrites were
more stable than SNAP /Fig. 1), but unlike SNAP, typical
®rst-order behaviours were not observed for these two
compounds. The spontaneous ®rst-order decomposition of
SNAP is probably a reaction, which is catalyzed by traces of
metal ions present in solution. Indeed, a number of reports
demonstrated that the presence of metal ions /such as Cu1,
Cu11) signi®cantly accelerates the decomposition of certain
thionitrites such as SNAP and its derivatives.12 Neverthe-
less, this effect is less marked on S-nitrosoglutathione whose
decomposition proceeds either by a transnitrosation reaction
or by enzymatic cleavage of the peptide bond.10 The Cu1-
catalyzed decomposition of S-nitrosocarboxylic acids
including SNAP is thought to occur via an intermediate in
which copper would be complexed via the carboxylate and
the nitrogen of the S±NvO group.13 In compounds 4c and
4d, as well as in other SNAP derivatives such as glyco-S-
nitrosothiols,7 the carboxylic group is involved in an amidic
bond that should prevent copper coordination, and thus
accounting for the higher stability of such derivatives
compared to the parent molecule SNAP. On the other
hand, based on the observation that copper/II) ions
reversibly denature DNA, it has been shown that adenine
can form cupric complexes,14 suggesting that 4c and 4d
could complex some of the free Cu21 present in solution,
thus increasing the apparent stability of these two adenine-
based thionitrites. To further support this hypothesis, we
used spectrophotometric measurements to evidence the
formation of complexes on the parent molecules
/compounds 3c and 3d) of the two thionitrites 4c and 4d.
The changes in the absorption spectra were very similar for
both compounds: lmax and emax values signi®cantly changed
in acetonitrile when the thiols were in presence of 10 equiv.
NOz $ NO1 1 e2
ꢀi†
However, in aqueous solution the following equilibrium
exists:
NO1 1 H2O $ HNO2 1 H1
nitrous acid being the prevalent species in acidic diluted
ꢀii†
solutions.16
2
Hence, the overall electrode reaction leading to the NO2
formation in the phosphate buffer may be described by the
equation:
NOz 1 H2O $ HNO2 1 H1 1 e2
ꢀiii†
Nitrite, which is also a degradation product of NOz in solu-
tions containing oxygen, is electroactive at a very similar
potential to NOz and is converted in a further step to nitrate.
To improve sensitivity and selectivity, modi®ed electrodes
with electropolymerised ®lms of metalloporphyrins17 or
phthalocyanine18 and surfaces covered with Na®one and/
or cellulose acetate were employed. Nevertheless, the use of
Na®on to suppress the oxidation of nitrite could lead to
misleading results since, with relatively thick Na®on layers,
the transport of NOz to the electrode surface is restricted and
lower currents are obtained. On the other hand, a direct
reduction of NOz on bare glassy carbon is not observed,
except in the presence of electrocatalysts such as iron substi-
tuted polyoxotungstates19 or iron±alizarin complexone.20
Depending on the experimental conditions, the electro-
reduction of nitric oxide leads to the formation of ammonia
or other intermediate compounds.
So as to check the redox response of thionitrites 4a±d and
obtain information about their NOz generating properties,
cyclic voltammograms on a wide potential window
/21500 to 1300 mV) were recorded using glassy carbon
electrodes. Firstly, the glassy carbon electrode response
was checked on pure NOz solutions obtained by direct
bubbling gaseous nitric oxide in phosphate buffer electro-
lyte /0.1 M, pH 7.0). Only one irreversible oxidation wave
was observed at 1000 mV /Fig. 2). A linear relationship of
peak current and concentration was obtained. Addition of
nitrite ions to the solution resulted in the appearance of a
peak at a more cathodic potential /830 mV). In contrast,
of copper ion /3c: 260 nm /11,750 M21 cm21), 3c/Cu21
:
265 nm /5280); 4c: 259 nm /12,300), 4c/Cu21: 266 nm
/7292)). Such variations were a clear indication of the
involvement of the purine group in the coordination of
copper, that might account for the higher stability of these
two thionitrites.
2.3. Redox behaviour of SNAP and S-nitrosothiols 4a±d
The direct measurement of nitric oxide in solution from
decomposition of NO-donors is dif®cult owing to its low
concentration due to both a slow decomposition reaction
and a high reactivity of the radical NOz with a number of
species. As a consequence of its redox properties, the
electrochemical-based detection methods are preferred
since they are suitable for real-time measurements of NOz
with adequate sensitivity.15 The electrochemical behavior of
NOz was studied on glassy carbon and mercury surfaces by
cyclic voltammetry. Nitric oxide is oxidised at glassy
carbon /GC) or graphite electrodes leading to the formation
of the nitrosonium cation /NO1) at around 10.9 to 11.0 V
Figure 2. Oxidation of pure nitric oxide solutions at different concentra-
tions on glassy carbon electrode in 0.1 M phosphate buffer pH 7.0.

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7176
2R-Sz ! R-SS±R
ꢀ2†
ꢀ3†
ꢀ4†
ꢀ5†
R-SNO 1 e2 ! R-S2 1 NOz
R-S2 1 H1 $ R-SH
R-SH 1 Hg $ ꢀHgSR†ads 1 H1 1 e2
Reaction /Eq. /1)) corresponds to the spontaneous homo-
lytic cleavage of the S±N bond, which is in fact catalyzed by
a number of parameters such as metal ions /Cu1, Cu21),
light, pH or temperature. The thiyl radical then dimerizes
to give the corresponding dissul®de /Eq. /2)). The electro-
chemical reaction /Eq. /3)) results in the formation of nitric
oxide and R-S2, which protonates to form a free thiol /Eq.
/4)). In presence of Hg, the thiol reacts to form the mercury
mercaptide, which is reversibly oxidised or reduced electro-
chemically /Eq. /5)).
Figure 3. Time course of SNAP /1 mM) in 0.1 M phosphate buffer /pH
7.0), 10% DMSO on glassy carbon electrode at a scan rate of 100 mV s21
/Ð 5 min; - - - 35 min; WWW 90 min).
addition of sodium nitrate, another oxidative product of NOz
did not induce any detectable signal.
For comparison purposes, we used SNAP as a model
compound. On a GC electrode, SNAP exhibited only a
single reduction peak at 2876 mV /vs Ag/AgCl, 3 M
KCl) in phosphate buffer at a scan rate of 100 mV s21
/Fig. 3). This peak was broad and has an irreversible nature,
and the potential was found to be independent of solution
pH. When the scan rate increased, the reduction potential
shifted to more negative potentials indicating a slow hetero-
geneous electron transfer rate. To further elucidate the
electron transfer mechanism of thionitrite decomposition,
Wang et al.21 carried out controlled potential electrolysis
experiments and concluded that NOz was formed during
the bulk electrolysis, and the amount of NOz generated
correlated linearly with the applied potential between
2800 and 21150 mV. They also determined that one
electron per equivalent was transferred during the bulk
electrolysis. Thus, on the basis of these previous studies, it
can be concluded that this wave at 2876 mV is due to the
electrochemical cleavage of the S±NO bond. As time
progressed, the wave diminished, revealing the spontaneous
decomposition of SNAP /Fig. 3). This decrease correlated
with the spectrophotometrically spectral change at 340 nm,
associated with the ®rst order decomposition reaction. On
the oxidation side, two waves were observed which were not
affected by pH changes: the wave at Epˆ810 mV which
increased with time corresponded to nitrite, and the second
at Epˆ1120 mV was due to NOz oxidation /Fig. 3).
On these grounds, we then examined the electrochemical
properties of thionitrites 4a±4d by recording a series of
cyclic voltammograms. As expected, subsequent scans of
potential revealed that 4c and 4d exhibit a slower chemical
decomposition process compared to SNAP, in accordance
with the spectrophotochemical studies. Surprisingly, in
these electrochemical experimental conditions /i.e. 10%
DMSO in 0.1 M phosphate buffer, pH 7.0), 4a and 4b also
exhibited a greater stability than SNAP, suggesting that
their stability is highly affected by the solvent. Indeed,
these two compounds were found to be much more stable
in organic solvents such as methanol or DMSO. On the
oxidation, a single process was observed for the four
2
compounds corresponding to NO2 oxidation, in relation
with the release of nitric oxide, and its subsequent conver-
sion to NO22. The NOz signal itself could not be observed,
probably due to the small concentration reached. On the
other hand, on the reduction side, small waves were
observed on glassy carbon corresponding to the S±NO
cleavage. This electrochemical reaction occurred at more
negative potentials than SNAP in the same experimental
conditions /around 2900 mV, Table 2 and Fig. 4), and
smaller current intensity were observed. These data are
consistent with a greater stability of these compounds in
the experimental conditions compared to SNAP. In addi-
tion, a wave at more positive potentials was observed for
each compound 4a-4d which might be attributed to their
inherent structure or to oxygen traces present in solution
/Epˆ2610 mV).
On a mercury electrode, in addition to the cleavage of the
S±NO bond /at Epˆ21140 mV), a reversible wave was
observed at E1/2ˆ2336 mV at pH 7.0, which can be
attributed to the formation of a mercury mercaptide /not
shown). This was con®rmed with N-acetyl-penicillamine
which have a half wave potential of 2338 mV in the
same experimental conditions. This reversible wave shifted
to negative potentials with increasing pH with a slope of
256 mV/pH, suggesting that one proton is involved in the
reaction mechanism. The presence of mercury mercaptide
accounts for a chemical rupture of the S±NO bond. Based
on our experiments and literature results,21 the following
reaction mechanism is suggested:
Table 2. Electrochemical potentials of S-nitrosothiols, nitric oxide and
nitrile ion
Oxidation Ep /mV)
Reduction Ep /mV)
SNAP
4a
4b
810/1120
840
696
897
865
2876
2644/2900
2638
2698/2962
2596
4c
4d
NOz
NO2
950
835
±
±
2
R-SNO ! R-Sz 1 NOz
ꢀ1†

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7177
Figure 4. Cyclic voltammograms of compounds 4b±d on glassy carbon electrode in 0.1 M phosphate buffer /pH 7.0) and 10% DMSO.
3. Conclusions
/60F254, Merck) were used for thin layer chromatography.
All chemicals were obtained from Aldrich, and used without
further puri®cation. 2-Bromo-1-[/tert-butyloxycarbonyl)-
amino]ethane was synthesised from 2-bromoethylamine
hydrobromide and di-BOC dicarbonate. Elemental analyses
S-Nitrosothiols display a number of pharmacological
effects, and their biological activity is mainly attributed to
their capacity to release nitric oxide.22 Moreover, there is an
interest in their reactivity and they have been used as
reagents in some chemical reactions such as disul®de forma-
tion23 or decarboxylative nitrosation reactions.24 Conse-
quently, there is an increasing interest in the development
of new nitric oxide-releasing compounds capable of gener-
ating NOz quantitatively at a predictable rate. The present
work deals with the preparation and the decomposition and
electrochemical studies of new S-nitrosothiols which
contain amidin groups that were designed for better target-
ting. All these compounds were prepared by electrophilic
nitrosation of their corresponding parent thiols using either
tert-butyl nitrite or sodium nitrite in acidic conditions. Two
of these compounds /4c and 4d), bearing purine moieties,
exhibited a remarkable stability in aqueous solution
compared to SNAP, probably due to their capacity to retain
trace quantities of Cu21 ions present in aqueous solution, as
shown by spectroscopic studies. The electrochemical
properties of the different compounds con®rm this
difference in stability of 4c and 4d compared to SNAP;
results were similar on mercury and glassy carbon
electrodes, demonstrating irreversible processes attributed
Â
were performed by the Ecole Nationale Superieure de
Chimie de Toulouse, France.
Electrochemical measurements were carried out on a BAS
CV-50 Electrochemical Analyzer /Bioanalytical Systems
Inc.) and a PAR model 273 Potentiostat/Galvanostat
/EG&G Princeton Applied Research). A conventional
three electrode-cell with a glassy carbon electrode /BAS,
2.5 mm f) as working electrode, Ag/AgCl reference and
platinum counter electrode was used. A phosphate buffer
/0.1 M, pH 7.0) was used as supporting electrolyte.
Compounds were dissolved in DMSO /0.5 mL) and this
solution was added to 5 mL of supporting electrolyte.
Solutions were deoxygenated by argon bubbling.
4.1.1. 3-Acetamido-4,4-dimethylthietan-2-one ꢀ2). To an
ice-cooled suspension of /1/2) penicillamine /6.5 g,
43.5 mmol) in dry pyridine /25 mL), acetic anhydride
/12.52 g, 123 mmol) was added for 30 min. The reaction
mixture was stirred for 20 h at room temperature. The solu-
tion was dissolved in chloroforme /250 mL), washed with
0.1 M HCl, water and dried over magnesium sulphate. The
solvent was evaporated to dryness and the residue was tri-
turated with light petroleum ether and ®ltered to yield 2 as a
pale yellow crystalline solid /3.31 g, 19.1 mmol, 44%). IR
/KBr) 3053, 1758, 1658 cm21. 1H NMR /CDCl3, 250 MHz):
d 1.59 and 1.82 /s, 6H), 2.02 /s, 3H), 5.66 /d, Jˆ8 Hz, 1H),
6.98 /Br, 1H). 13C NMR /CDCl3, 60 MHz): d 22.4, 26.1,
30.2, 51.2, 76.4, 169.9, 193.3.
2
to the electrochemical oxidation reaction of NO2 and to
the electrochemical cleavage of S±NO bond. Beyond the
speci®c interest of the physical properties of present nitro-
gen oxide derivatives, these results will help to de®ne the
best compromise between the relative stability of these NO
donors in solution, and their ability to deliver this species
within the cell in the presence of metal catalysts and/or
under the action of enzymes. This aspect is at present
being investigated.
4.1.2. N1-ꢀ6-Amino-3-pyridyl)-2-ꢀacetylamino)-3-methyl-
3-sulfanylbutanamide ꢀ3a). A solution of 2,5-diamino-
pyridine /0.155 g, 1.42 mmol) and 3-acetamido-4,4-
dimethylthietan-2-one /0.295 g, 1.70 mmol) in chloroform
/17 mL) was stirred for 5 h at room temperature. The
solvent was evaporated to dryness and the residue was
treated with a mixture of ethyl acetate and methanol
/85:15). The solution was ®ltered and the ®ltrate was evapo-
rated to dryness. The residue was puri®ed by chromato-
graphy /ethyl acetate/methanol/32% aq. ammonium
hydroxide, 85:15:0.25) to yield 3a as a pink powder
4. Experimental
4.1. General
1H NMR and 13C NMR spectra were recorded on Brucker
AC200, AC250 or AC300 spectrometers. Ultraviolet spec-
tra were recorded on a HP8453 spectrophotometer /Hewlett
Packard), and IR spectra on a Perkin±Elmer 1600 FTIR
spectrophotometer. Silica gel 60 /70±230 mesh, Merck)
was used for column chromatography and silica plates
1
/0.245 g, 61%). H NMR /D6-DMSO, 250 MHz): d 1.37

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7178
1
/s, 3H), 1.40 /s, 3H), 1.94 /s, 3H), 2.75 /s, 1H), 4.65 /d,
Jˆ9 Hz, 1H), 5.75 /s, 2H), 6.41 /d, Jˆ9 Hz, 1H), 7.55 /dd,
Jˆ9, 1 Hz, 1H), 8.08±8.12 /m, 2H), 9.91 /s, 1H). 13C NMR
/D6-DMSO, 50 MHz): 22.4, 28.7, 30.0, 46.2, 60.9, 107.4,
124.8, 130.4, 139.7, 156.5, 167.6, 169.4. Mass spectrometry
FAB /glycerol/thioglycerol) m/z 283 [M1H]1. Anal. Calcd
for C12H18N4O2S: C, 51.04; H, 6.43; N, 19.84. Found C,
50.99; H, 6.41; N 19.81.
/KBr) 1651, 649 cm21. H NMR /CDCl3, 250 MHz): d
1.27, 1.34 /s, 3H), 1.30 /s, 3H), 1.42, 1.47 /s, 3H), 1.56 /s,
3H), 2.00 /s, 3H), 2.50, 2.58 /s, 1H), 3.32±3.48, 3.95±4.06
/m, 2H), 4.41±4.46 /m, 1H), 4.58, 4.68 /d, Jˆ9 and 9.5 Hz,
1H), 4.78±4.84 /m, 1H), 5.26±5.34, 5.40±5.44 /m, 1H),
5.82, 5.88 /d, Jˆ4 Hz, 1H), 6.79, 6.86 /s, 2H), 6.89±6.93
/m, 1H), 7.91, 7.88 /s, 1H), 8.38, 8.65 /s, 1H), 8.76, 8.90±
8.93 /m, 1H). 13C NMR /CDCl3, 50 MHz): d 23.2, 23.3,
25.1, 25.2, 27.4, 28.1, 29.0, 31.3, 31.6, 41.0, 41.5, 45.7,
46.0, 60.8, 81.6, 82.1, 82.5, 82.9, 83.1, 83.8, 92.1, 92.5,
114.6, 114.7, 120.7, 120.8, 140.3, 148.8, 153.1, 153.6,
156.0, 156.1, 170.1, 170.3, 170.7, 170.8. Anal. Calcd for
C20H29N7O5S, 1.5 H2CO3: C, 46.57; H, 5.77; N, 18.10.
Found C, 46.06; H, 5.84; N, 18.08.
4.1.3. N1-4-[ꢀ4,6-Diamino-1,3,5-triazin-2-yl)amino]phenyl-
2-ꢀacetylamino)-3-methyl-3-sulfanylbutanamide ꢀ3b). A
solution of 5 /0.3 g, 1.38 mmol) and 3-acetamido-4,4-
dimethylthietan-2-one /0.285 g, 1.65 mmol) in dimethyl-
formamide /10 mL) was stirred overnight. The solvent
was evaporated to dryness. The residue was triturated with
hot ethyl acetate. The solution was ®ltered and the ®ltrate
evaporated to dryness. The residue was puri®ed by chroma-
tography /dichloromethane/methanol/aq. 32% ammonium
hydroxide, 90:10:0.25) to yield 3b as a white powder
/ii) A solution of the 20,30-O-isopropylidene derivative of 3d
/0.36 g, 0.75 mmol) in water /1 mL) and formic acid
/1.5 mL) was stirred at room temperature for 48 h. The
solvent was evaporated to dryness. Several co-evaporatings
of the residue with ethanol afforded 3d as a white powder
1
/0.16 g, 30%). H NMR /D6-DMSO, 200 MHz): d 1.39 /s,
1
3H), 1.43 /s, 3H), 1.96 /s, 3H), 2.73 /s, 1H), 4.70 /d,
Jˆ9 Hz, 1H), 6.19 /s, 4H), 7.45 /d, Jˆ8.5 Hz, 2H), 7.67
/d, Jˆ9 Hz, 2H), 8.03 /d, Jˆ9 Hz, 2H), 8.7 /s, 1H), 10.01
/s, 1H). 13C NMR /D6-DMSO, 50 MHz): d 22.4, 28.6, 30.1,
46.3, 61.1, 119.5, 119.7, 132.2, 136.5, 164.6, 167.0, 167.6,
169.4. Mass spectrometry /FAB, glycerol/thioglycerol) m/z
391 [M1H]1. Anal. Calcd for C16H22N8O2S, 1.5 H2CO3:
C, 43.47; H, 5.21; N, 23.18. Found C, 42.71; H, 5.67; N
22.98.
/yield: 100%). H NMR /D6-DMSO, 250 MHz): d 1.24,
1.30 /s, 3H), 1.32, 1.33 /s, 3H), 1.87, 1.92 /s, 3H), 2.68,
2.72 /s, 1H), 3.37±3.65 /m, 2H), 3.94±4.07 /m, 2H), 4.51±
4.59 /m, 1H), 4.66±4.70 /m, 1H), 5.24 /s, 1H), 5.44 /s, 1H),
5.82, 5.86 /m, 1H), 7.31 /s, 2H), 7.95±8.07 /m, 1H), 8.24,
8.29±8.36 /m, 2H), 8.35±8.46, 8.54±8.65 /m, 1H). 13C
NMR /CD3OD, 50 MHz): d 22.6, 29.2, 29.7, 30.7, 31.1,
42.2, 42.6, 46.3, 46.4, 63.3, 63.5, 72.9, 73.1, 74.5, 74.6,
85.3, 85.4, 90.2, 90.9, 121.2, 142.4, 150.4, 153.8, 154.2,
157.3, 172.1, 173.2, 173.3. Mass spectrometry /FAB,
glycerol/thioglycerol) m/z 430 [M1H]1, 462 [M1Na]1.
4.1.4. N1-[2-ꢀ6-Amino-9H-9-purinyl)ethyl]-2-ꢀacetylamino)-
3-methyl-3-sulfanylbutanamide ꢀ3c). A solution of 3-acet-
amido-4,4-dimethylthietan-2-one /0.273 g, 1.58 mmol) in
chloroform /1.8 mL) and a solution of 6 /1.26 mmol) in
aq. NaOH 1N /2.7 mL) were stirred for 2 h. The aqueous
layer was separated and evaporated to dryness. The residue
was treated with a mixture of ethyl acetate/methanol/32%
aq. ammonium hydroxide, 75:25:0.25. The solution was
®ltered, and the ®ltrate evaporated to dryness. The residue
was puri®ed by chromatography /ethyl acetate/methanol/
32% aq. ammonium hydroxide, 75:25:0.25) to yield 3c as
a white powder /0.27 g, 61%). IR /KBr) 3293, 3210, 1654,
1602, 648 cm21. 1H NMR /D6-DMSO, 200 MHz): d 1.23 /s,
3H), 1.26 /s, 3H), 1.91 /s, 3H), 2.69 /s, 1H), 3.51±3.53 /m,
2H), 4.21 /t, Jˆ6 Hz, 2H), 4.38 /d, Jˆ9 Hz, 1H), 7.21 /s,
2H), 7.95 /d, Jˆ9 Hz, 1H), 8.02 /s, 1H), 8.13 /s, 1H), 8.31±
8.34 /m, 1H). 13C NMR /D6-DMSO, 50 MHz): d 22.4, 28.9,
29.7, 38.1, 42.4, 45.6, 60.8, 118.7, 140.9, 149.5, 152.2,
155.8, 169.3, 169.5. Mass spectrometry /FAB, glycerol/
thioglycerol) m/z 352 [M1H]1, 374 [M1Na]1. Anal.
Calcd for C14H21N7O2S, 1.5 H2CO3: C, 41.89; H, 5.44; N,
22.06. Found C, 41.01; H, 5.40; N, 22.06.
4.1.6. S-Nitrosothiol 4a. To a solution of 3a /0.15 g,
0.53 mmol) in ethanol /3 mL) was added 12 M hydrochloric
acid /0.044 mL). After 2 h at room temperature, the solution
was ®ltered to yield the hydrochloride salt of 3a as a white
powder /0.1 g, 0.32 mmol). To a solution of 3a hydro-
chloride in chloroform /4.5 mL) was added tert-butyl nitrite
/2.08 g, 20.18 mmol). After 35 min at room temperature,
the solution was ®ltered to yield 4a /0.08 g, 43%). IR
/KBr) 1657, 1531, 667 cm21. UV /50 mM phosphate buffer
pH 7.4) l nm /e, M21 cm21): 206 /7500), 234 /6930), 400
/sh). H NMR /D6-DMSO, 250 MHz): d 1.89 /s, 3H), 1.98
1
/s, 3H), 2.03 /s, 3H), 5.40 /d, Jˆ9.5 Hz, 1H), 7.06 /d,
Jˆ9.5 Hz, 1H), 7.96 /d, Jˆ9 Hz, 1H), 8.41 /s, 1H), 8.69
/d, Jˆ9 Hz, 1H), 10.94 /s, 1H). 13C NMR /D6-DMSO,
60 MHz): 22.2, 24.7, 26.4, 59.1, 59.5, 114.1, 124.9, 125.0,
137.6, 151.3, 167.6, 169.5.
4.1.7. S-Nitrosothiol 4b. To a solution of 3b /0.1 g,
0.26 mmol) in methanol /1 mL), concentrated sulfuric acid
/1 mL) and 1 M HCl /1 mL), was slowly added a 1 M
aqueous solution of sodium nitrite /1 mL). After 15 min
under vigorous stirring, the precipitate was ®ltered and
washed with water to yield 4b /0.090 g, 83%). IR /KBr)
1654, 1510, 668 cm21. UV /50 mM phosphate buffer pH
7.4/5% methanol) l nm /e, M21 cm21): 212 /4380), 276
4.1.5. N1-[ꢀ3S,4R)-5-ꢀ6-Amino-9H-9-purinyl)-3,4-dihydroxy-
tetrahydro-2-furanyl]methyl-2-ꢀacetylamino)-3-methyl-3-
sulfanylbutanamide ꢀ3d). /i) To a solution of 50-amino-
20,30-/O-isopropylidene)-50-deoxyadenosine
/0.305 g,
1
1 mmol) in CH2Cl2 /12 mL) was added 3-acetamido-4,4-
dimethylthietan-2-one /0.208 g, 1.2 mmol). After 5 h at
room temperature, the solvent was evaporated to dryness.
The residue was puri®ed by chromatography /ethyl acetate/
methanol, 95:5) to yield the 20,30-/O-isopropylidene)
derivative of 3d as a white powder /0.41 g, 85 %). IR
/3200), 340 /320). H NMR /C5D5N, 300 MHz): d 2.12 /s,
3H), 2.17 /s, 3H), 2.24 /s, 3H), 6.16 /d, Jˆ9.5 Hz, 1H), 6.57
/Br, 6H), 7.89 /d, Jˆ9 Hz, 2H), 7.97 /d, Jˆ9 Hz, 2H), 9.53
/d, Jˆ9.5 Hz, 1H), 10.74 /s, 1H); 11.73 /s, 1H). 13C NMR
/C5D5N, 75 MHz): d 22.9, 25.6, 27.0, 60.0, 61.1, 121.2,
122.2, 134.6, 136.7, 164.4, 168.5, 170.4.

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7179
4.1.8. S-Nitrosothiol 4c. To a solution of 3c /0.27 g,
0.77 mmol) in acetone /11 mL) was slowly added tert-
butyl nitrite /4.91 g, 48 mmol). After 35 min at room
temperature, the solution was ®ltered to yield 4c /0.18 g,
62%). IR /KBr) 1700, 1655, 1512, 668 cm21. UV /50 mM
phosphate buffer pH 7.4) l nm /e, M21 cm21): 212
bonylamino)ethyl]-adenine as a white powder /59%). IR
/KBr) 3364, 3148, 1686, 1598 cm21 1H NMR /D6-
.
DMSO, 250 MHz): d 1.31 /s, 9H), 3.30±3.37 /m, 2H),
4.17 /t, Jˆ6 Hz, 2H), 6.97 /t, Jˆ6 Hz, 1H), 7.17 /s, 2H),
8.00 /s, 1H), 8.13 /s, 1H). 13C NMR /D6-DMSO, 50 MHz):
d 28.0, 38.1, 42.7, 77.7, 118.6, 140.9, 149.6, 151.9, 155.5,
155.6. Mass spectrometry /DCI/NH3) m/z 279 [M1H]1.
/ii) A solution of 9-[2-/tert-butoxycarbonylamino)ethyl]-
adenine /0.35 g, 1.2 mmol) in dichloromethane /4 mL)
and tri¯uoroacetic acid /4 mL) was stirred at room tempera-
ture for 16 h. The solvent was removed by evaporation in
vacuo and the remaining residue produced, after several
co-evaporations with ethanol, 9-/2-aminoethyl)-9H-purin-
1
/26,460), 260 /18,920), 340 /480). H NMR /D6-DMSO,
200 MHz): d 1.82 /s, 6H), 1.85 /s, 3H), 3.59±3.62 /m,
2H), 4.33 /m, 2H), 5.05 /d, Jˆ9.5 Hz, 1H), 8.35 /d,
Jˆ9.5 Hz, 1H), 8.43 /s, 1H), 8.45 /s, 1H), 8.59 /m, 1H),
8.77 /Br, 1H), 9.55 /Br, 1H). 13C NMR /D6-DMSO,
50 MHz): d 22.2, 24.6, 26.7, 38.2, 43.3, 58.8, 59.1, 117.9,
144.1, 144.6, 148.7, 148.8, 168.6, 169.2.
1
6-ylamine 5 as a white powder /100%). H NMR /D6-
4.1.9. S-Nitrosothiol 4d. To a solution of 3d /0.27 g,
0.61 mmol) in acetone /8.7 mL) was slowly added tert-
butyl nitrite /3.93 g, 38 mmol). After 35 min at room
temperature, the solution was ®ltered to give 0.18 g of 4d
/yield: 63%). IR /KBr) 1686, 1654, 1510, 669 cm21. UV
/50 mM phosphate buffer pH 7.4) l nm /e, M21 cm21):
210 /61,400), 258 /23,600), 338 /430). 1H NMR /D6-
DMSO, 250 MHz): d 1.84, 1.90 /s, 3H), 1.92, 1.96 /s,
3H), 1.95 /s, 3H), 3.37±3.50 /m, 2H), 3.99±4.10 /m, 2H),
4.60±4.65 /m, 1H), 5.22, 5.24 /d, Jˆ9.5 Hz, 1H), 5.92±5.94
/m, 1H), 8.40±8.53 /m,2H), 8.67±8.73 /m, 2H). 13C NMR
/D6-DMSO, 60 MHz): d 22.4, 24.5, 24.7, 26.6, 40.0, 40.4,
58.8, 58.9, 59.6, 71.2, 73.2, 83.2, 83.5, 87.5, 87.6, 118.8,
119.0, 142.3, 146.6, 148.4, 148.5,; 151.0, 151.2, 168.5,
169.3.
DMSO1D2O, 200 MHz): d 3.38 /t, Jˆ6 Hz, 2H), 4.50 /t,
Jˆ6 Hz, 2H), 8.34/s, 1H), 8.43 /s, 1H).
Acknowledgements
This work was supported by a CNRS-CONICYT inter-
national scienti®c co-operation program /8726). The
authors would like to thank Dr Fethi Bedioui for fruitful
discussions.
References
1. Bourguignon, S. C.; Alves, C. R.; Giovanni De Simone, S.
Acta Trop. 1997, 66, 109±118.
2. Groves, J. T. Curr. Opin. Chem. Biol. 1999, 3, 226±235.
Á Â Â
3. Soulere, L.; Hoffmann, P.; Bringaud, F.; Perie, J. Bioorg. Med.
Chem. Lett. 2000, 10, 1347±1350.
4.1.10. 2-ꢀ4-Aminophenylamino)-4,6-diamino-1,3,5-tri-
azine ꢀ5). /i) A solution of 4-aminoacetanilide /10.3 g,
68.7 mmol) and 2-chloro-4,6-diamino-1,3,5-triazine /10 g,
68.7 mmol) in water /250 mL) was heated to re¯ux. To
this mixture, 2£17.5 mL of 2 M NaOH were added succes-
sively after 1 and 2 h, and heating was maintained for a total
of 4.5 h. After cooling, the precipitate was ®ltered and
washed with water to give 16.41 g of 2-/4-acetylaminophe-
nyl)-4,6-diamino-1,3,5-triazine /92%). 1H NMR /D6-
DMSO, 250 MHz): d 2.00 /s, 3H), 6.28 /s, 4H), 7.40 /d,
Jˆ7 Hz, 2H), 7.65 /d, Jˆ7 Hz, 2H), 8.75 /s, 1H), 9.77 /s,
1H). 13C NMR /D6-DMSO, 50 MHz): d 23.8, 119.2, 119.8,
133.0, 136.0, 164.6, 167.0, 167.7.
4. Carter, N. S.; Fairlamb, A. H. Nature 1993, 361, 173±176.
Carter, N. S.; Berger, B. J.; Fairlamb, A. H. J. Biol. Chem.
1995, 270, 28153±28157. Tye, C.-K.; Kasinathan, G.; Barrett,
M. P.; Brun, R.; Doyle, V. E.; Fairlamb, A. H.; Weaver, R.;
Gilbert, I. H. Bioorg. Med. Chem. Lett. 1998, 8, 811±816.
È
Maaser, P.; Suatterlin, C.; Kralli, A.; Kaminsky, R. Science
1999, 285, 242±244.
È
5. Al-Zaidi, S. M. R.; Crilley, M. M. L.; Stoodley, R. J. J. Chem.
Soc. Perkin Trans. 1 1983, 2259±2265.
6. Moynihan, H. A.; Roberts, S. M. J. Chem. Soc., Perkin Trans.
1 1994, 797±806.
/ii) A solution of 2-/4-acetylaminophenyl)-4,6-diamino-
1,3,5-triazine /16 g, 62 mmol) in 1.2 M HCl /500 mL) was
re¯uxed for 1.5 h. After cooling, 250 mL 2.5 M NaOH were
added, and the precipitate was ®ltered and washed with hot
methanol /150 mL). The ®ltrate was then evaporated to
dryness under reduced pressure to yield 5 g of 6 /37%).
1H NMR /D6-DMSO, 250 MHz): d 4.67 /s, 2H), 6.13 /s,
4H), 6.47 /d, Jˆ7 Hz, 2H), 7.29 /d, Jˆ8.5 Hz, 2H); 8.3 /s,
1H). 13C NMR /D6-DMSO, 50 MHz): d 113.7, 122.0, 129.6,
143.5, 164.8, 167.0.
7. Ramirez, J.; Yu, L.; Li, J.; Braunschweiger, P. G.; Wang, P. G.
Bioorg. Med. Chem. Lett. 1996, 6, 2575±2580.
8. Kolb, M.; Danzin, C.; Barth, J.; Claverie, N. J. Med. Chem.
1982, 25, 550±556.
9. Field, L.; Dilts, R. V.; Ravichandran, R.; Lenhert, P. G.;
Carnahan, G. E. J. Chem. Soc., Chem. Commun. 1978, 249±
250.
10. Singh, S. P.; Wishnok, J. S.; Keshive, M.; Deen, W. M.;
Tannenbaum, S. R. Proc. Natl. Acad. Sci. USA 1996, 93,
14428±14433.
4.1.11. 9-ꢀ2-Aminoethyl)-9H-purin-6-ylamine ꢀ6). /i) A
solution of adenine /2 g, 14.8 mmol), 2-bromo-1-[/tert-
butyloxycarbonyl)amino]ethane /3.32 g, 14.8 mmol) and
potassium carbonate /4.09 g, 29.6 mmol) in dry N,N0-
dimethylformamide /60 mL) was stirred at room tempera-
ture for 18 h. After ®ltering, the solvent was evaporated to
dryness under reduced pressure. The residue was triturated
with water and ®ltered to give 2.4 g of 9-[2-/tert-butoxycar-
11. Goto, K.; Hino, Y.; Kawashima, T.; Kaminaga, M.; Yano, E.;
Yamamoto, G.; Takagi, N.; Nagase, S. Tetrahedron Lett.
2000, 41, 8479±8483. Arulsamy, N.; Bohle, D. S.; Butt,
J. A.; Irvine, G. J.; Jordan, P. A.; Sagan, E. J. Am. Chem.
Soc. 1999, 121, 7115±7123.
12. Williams, D. L. H. Methods Enzymol. 1996, 268, 399±308.
Singh, R. J.; Hogg, N.; Joseph, J.; Kalyanaraman, B. J. Biol.
Chem. 1996, 271, 18596±18603.

Á
L. Soulere et al. / Tetrahedron 57 >2001) 7173±7180
7180
13. Askew, S. C.; Barnett, D. J.; McAninly, J.; Williams, D. L. H.
J. Chem. Soc., Perkin Trans. 2 1995, 741±745.
14. De Meester, P.; Goodgame, D. M.; Price, K. A.; Skapski, A. C.
Nature 1971, 229, 191±192.
20. Zhang, J.; Lever, A. B. P.; Pietro, W. J. Inorg. Chem. 1994, 33,
1392±1392.
21. Hou, Y.; Wang, J.; Arias, F.; Echegoyen, L.; Wang, P. G.
Bioorg. Med. Chem. Lett. 1998, 8, 3065±3070.
22. For a recent review on biological aspects of S-nitrosothiols,
see: Hogg, N. Free Radic. Biol. Med. 2000, 28, 1478±1486.
23. Demir, A. S.; Igdir, A. C.; Mahasneh, A. S. Tetrahedron 1999,
55, 12399±12404.
15. Turney, T. A.; Wright, G. A. Chem. Rev. 1959, 59, 497±513.
16. Pires, M.; Rossi, M. J.; Ross, D. S. Int. J. Chem. Kinet. 1994,
26, 1207±1228.
17. Malinski, T.; Taha, Z. Nature 1992, 358, 676±678.
18. Pontie, M.; Lecture, H.; Bedioui, F. Sensors Actuators B 1999,
56, 1±5.
24. Girard, P.; Guillot, N.; Motherwell, W. B.; Hay-Motherwell,
R. S.; Potier, P. Tetrahedron 1999, 55, 3573±3584.
19. Toth, J. E.; Anson, F. C. J. Am. Chem. Soc. 1989, 111, 2444±
2451.