Photochemical reactions of thiols with organic nitrates - Oxygen atom transfer via a thionitrate

Article abstract of DOI:10.1139/V06-056

Nitroglycerin is an organic nitrate that has been used in the clinical treatment of angina for 130 years, yet important details of its mechanism of action remain unanswered. The biological activity of nitrates suggests that they are bioactivated to NO via a three-electron reduction. The involvement of free or bound protein thiols in this reduction has often been proposed. To examine the involvement of thiyl radicals in such a process, the photochemical generation of benzenethiyl radical from thiol and disulfide precursors was studied in the presence of isopropyl nitrate. Analysis of reaction products and kinetics led to the conclusion that photolysis of the nitrate to NO₂ dominated the observed photochemistry. Formation of sulfonothioate and NO as products, and trapping of NO₂ by 4-chlorophenol, indicated a mechanism involving oxygen atom transfer from N to S via a thionitrate intermediate. The results of the study did not indicate a rapid reaction between thiyl radical and organic nitrate. Despite weak nitrate absorption of light >300 nm and a relatively high BDE for homolysis to give NO₂, the photochemistry under thiyl-generating conditions was driven by nitrate photolysis to NO₂. A novel nitrate, containing a phenyl disulfanyl group linked to nitrate groups, did not undergo photolysis to NO₂ or generate sulfonothioate, but did yield NO. These observations suggest that reaction between thiyl radicals and nitrates leading to NO release is a viable pathway, but it is subservient to other competing reactions, such as photolysis, in the case of IPN, and reaction with thiolate, in the case of the novel nitrate.

Full text of DOI:10.1139/V06-056

7
09
Photochemical reactions of thiols with organic
nitrates — Oxygen atom transfer via a thionitrate
1
Jennifer L. Clarke, Irida Kastrati, Linda J. Johnston, and Gregory R.J. Thatcher
Abstract: Nitroglycerin is an organic nitrate that has been used in the clinical treatment of angina for 130 years, yet
important details of its mechanism of action remain unanswered. The biological activity of nitrates suggests that they
are bioactivated to NO via a three-electron reduction. The involvement of free or bound protein thiols in this reduction
has often been proposed. To examine the involvement of thiyl radicals in such a process, the photochemical generation
of benzenethiyl radical from thiol and disulfide precursors was studied in the presence of isopropyl nitrate. Analysis of
reaction products and kinetics led to the conclusion that photolysis of the nitrate to NO dominated the observed
2
photochemistry. Formation of sulfonothioate and NO as products, and trapping of NO by 4-chlorophenol, indicated a
2
mechanism involving oxygen atom transfer from N to S via a thionitrate intermediate. The results of the study did not
indicate a rapid reaction between thiyl radical and organic nitrate. Despite weak nitrate absorption of light >300 nm
and a relatively high BDE for homolysis to give NO , the photochemistry under thiyl-generating conditions was driven
2
by nitrate photolysis to NO . A novel nitrate, containing a phenyl disulfanyl group linked to nitrate groups, did not un-
2
dergo photolysis to NO or generate sulfonothioate, but did yield NO. These observations suggest that reaction between
2
thiyl radicals and nitrates leading to NO release is a viable pathway, but it is subservient to other competing reactions,
such as photolysis, in the case of IPN, and reaction with thiolate, in the case of the novel nitrate.
Key words: nitrate, photolysis, thiyl radical, nitrogen dioxide, nitric oxide.
Résumé : Même si la nitroglycérine est un nitrate organique qui a été utilisé depuis 130 ans dans le traitement de
l’angine, beaucoup de détails importants de son mode d’action restent sans réponse. L’activité biologique des nitrates
suggère qu’ils sont biologiquement activés en NO par le biais d’une réduction à trois électrons. L’implication de grou-
pes thiols de protéines libres ou liées dans cette réduction a été souvent proposée. Afin d’examiner l’implication des
radicaux thiyles dans ce processus, on a étudié la génération photochimique du radical benzènethiyle à partir de précur-
seurs thiols et disulfures, en présence de nitrate d’isopropyle. L’analyse de la cinétique et des produits de la réaction
conduit à la conclusion que la photolyse du nitrate en NO domine la photochimie observée. La formation de sulfono-
2
thioate et de NO comme produits et le piégeage du NO par du 4-chlorophénol suggèrent un mécanisme impliquant le
2
transfert d’un atome d’oxygène du N vers le S par le biais d’un intermédiaire thionitrate. Les résultats de cette étude
indiquent qu’il n’y a pas de réaction rapide entre le radical thiyle et le nitrate organique. Malgré la faible absorption de
la lumière par le nitrate à des valeurs supérieures à 300 nm et une valeur relativement élevée du « BDE » pour
l’homolyse conduisant au NO , la photochimie dans les conditions conduisant à la formation de thiyle est dominée par
2
la photolyse du nitrate en NO . Un nouveau nitrate contenant du groupe phényl disulfanyle lié à des groupes nitrates
2
conduit à du NO, mais ne donne pas lieu à une photolyse conduisant au NO ou à la génération d’un sulfonothionate.
2
Ces observations suggèrent que la réaction entre les radicaux thiyles et les nitrates est une voie viable pour la produc-
tion de NO, mais elle est soumise à d’autres réactions de compétition, dont la photolyse dans le cas du nitrate
d’isopropyle et d’une réaction avec le thiolate dans le cas du nouveau nitrate.
Mots clés : nitrate, photolyse, radical thiyl, bioxyde d’azote, oxyde nitrique.
[
Traduit par la Rédaction] Clarke et al. 719
Received 3 October 2005. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 17 May 2006.
This work is dedicated to Walter A. Szarek in celebration of his 65th birthday and his continuing contributions to medicinal and
carbohydrate chemistry.
J.L. Clarke and L.J. Johnston. Steacie Institute for Molecular Sciences, National Research Council, Ottawa, ON K1A OR6,
Canada.
2
I. Kastrati and G.R.J. Thatcher. Department of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, University of
Illinois at Chicago, 833 S. Wood St. Chicago, IL 60612-7231, USA.
1
This article is part of a Special Issue dedicated to Professor Walter A. Szarek.
Corresponding author (e-mail: thatcher@uic.edu).
2
Can. J. Chem. 84: 709–719 (2006)
doi:10.1139/V06-056
© 2006 NRC Canada

7
10
Can. J. Chem. Vol. 84, 2006
Introduction
The putative bioactivation of organic nitrate to NO is a
three-electron reduction. If bioactivation is mediated by a
thiol, of the three potential reducing agents, thiol, thiolate,
and disulfide radical anion, the latter is clearly the best can-
didate simply based on reduction potentials (27).
The synthesis of nitroglycerin (glyceryl trinitrate, GTN)
was first reported in 1846. In 1878, it was demonstrated that
small doses of GTN taken sublingually provided rapid and
remarkable relief from the pain of angina. This led to clini-
cal use of GTN in the treatment of angina pectoris that has
continued to this day (1, 2). GTN is now applied in con-
trolled hypotension during cardiac surgery and in congestive
heart failure, whereas other organic nitrate vasodilators, in-
cluding isosorbide dinitrate (ISDN), are used in the treat-
ment of angina pectoris. ISDN, in a combination therapy, is
entering the clinic for prevention of heart failure (3). Hybrid
nitrates that conjugate a nitrate group to an established drug
via a labile linker have been the subject of numerous pre-
clinical and clinical studies (4), and novel nitrates are being
explored beyond traditional cardiovascular applications (5).
However, despite the historical and growing importance of
nitrates in disease therapy, their mechanism of action is not
well understood. One prevailing theory is that nitrates func-
tion as NO prodrugs, requiring bioactivation (or mechanism-
·
–
–
[
[
[
1a] RS + e → RS
E° = 0.78 V
E° = 1.35 V
E° = –1.42 V
·
–
+
1b] RS + e + H → RSH
–
·–
1c]
RSSR + e → (RSSR)
The reactions of nitrates with disulfide radical anions are
worthy of study in this respect. Disulfide radical anions can
be formed by the reversible combination of thiyl radical with
thiolate (eq. [2]) and by reduction of disulfides (eq. [1c]).
kf
·
–
(RSSR)·–
[
2]
RS + RS
k−
f
It is apparent that thiyl radicals are likely to be present in
any system designed to generate a disulfide radical anion. It
is also possible to draw a mechanism of reaction for nitrate
with thiyl leading to NO.
based biotransformation) of the nitrate moiety (RONO ) to
NO (6, 7).
Nitrate esters (or organic nitrates) contain the nitrooxy
2
·
·
·
[
3]
R′S + RONO → [RON(O)OSR′] → R′SO
2
functional group (-ONO ); conversion of the nitrooxy group
+ RONO
2
to NO is a three-electron reduction that must involve oxygen
atom transfer. To date, no purified protein system has been
demonstrated to mediate the direct reduction of nitrates to
yield relevant quantities of NO, although this is widely held
to be the biologically active product of nitrate bioactivation
+
[
4]
RONO + 1eG + H → ROH + NO
Therefore, it is essential to explore the reactivity of thiyl
radicals themselves with nitrates. However, the clean genera-
tion of thiyl radicals (and disulfide radical anions) presents a
challenge since there is no ready thermal source as has been
described for other biological radicals, such as superoxide
(
8, 9). The biotransformation of GTN yields the dinitrate
metabolites, glyceryl-1,2-dinitrate (1,2-GDN) and glyceryl-
,3-dinitrate (1,3-GDN), as products. Several proteins have
1
(
28). Conversely, a number of photochemical systems for
been identified that are capable of mediating the denitration
thiyl radical generation have been described. In this paper,
thiyl radicals generated from disulfide and thiol photochemi-
cal precursors are studied in the presence of nitrates.
–
of GTN, yielding GDN and NO2 as products, including he-
moglobin, myoglobin, xanthine oxidoreductase, old yellow
enzyme, glutathione S-transferase (GST), cytochrome P450
oxidase, and cytochrome P450 reductase (10–17). More re-
cently, it was confirmed that ALDH2 (aldehyde dehydro-
genase) was capable of mediating denitration of classical
nitrates (18–20).
The vascular effects of nitrates are believed to be medi-
ated primarily via activation of soluble guanylyl cyclase (sGC),
which converts GTP to the secondary messenger, cGMP.
sGC is activated by binding of NO to its ferrous-heme cen-
tre, inducing a conformational change that triggers activation
Experimental section
All reagents were supplied by Sigma-Aldrich (Oakville,
Ontario), with the exception of GTN and GT-130 that were
synthesized by literature methods (29). All solvents used
were HPLC grade. NO was supplied by Praxair Products
Inc. (Mississauga, Ontario) and was used with a NO specific
regulator. Phosphate buffer (PB) (0.1 mol/L, pH 7.4) was ad-
justed to pH and microfiltered before use.
1
(
21, 22). However, in the context of nitrates and GTN, there
All H NMR analyses were done on a Bruker Avance-400
(400 MHz) spectrometer. A Varian Cary 3 UV–vis diode ar-
ray spectrometer was used, with 1 cm × 1 cm quartz cells.
Kinetic measurements were performed on a HP 8452A diode
array spectrophotometer at 40 °C. Chemiluminescence mea-
surements were made using a SIEVERS 270B NO chemi-
luminescence analyzer and a SIEVERS 207B nitric oxide
analyzer (Boulder, Colorado). The UV light sources used
were 300, 350, and mixed 300/350 nm Rayonet reactors
equipped with lamps from the Southern New England Ultra-
violet Company (Brantford, Connecticut).
The lasers used in the laser flash photolysis experiments
were an EX-530 Excimer-500 (Lumonics, Inc.), an UV-24
N2 laser (Laser Photonics), and a HY 750 ytrium–
aluminum–garnet (YAG) laser (Lumonics, Inc.). The detec-
©
2006 NRC Canada

Clarke et al.
711
tor consisted of a Digikrom 240 monochromator (CVI In-
struments) with a six dynode photomultiplier tube. The
oscilloscope used was a Tektronix 7912 AD programmable
digitizer and the delay generator was a DG535 four-channel
digital delay/pulse generator (Stanford Research Systems).
All gas chromatography (GC) work was done on a
Hewlett Packard 5890 series II gas chromatograph with
flame ionization detection. The column used was a HP-1
with dimensions 25 m × 0.2 mm × 0.33 µm, column pres-
sure 20 psi He (1 psi = 6.894 757 kPa), and splitflow
using 300 and 350 nm Rayonets, respectively. All experi-
ments were done in duplicate. Reactions were analyzed by
GC-FID using 3 mmol/L anhydrous decane as standard. Rel-
ative peak areas from the GC traces were calibrated and con-
verted to concentration plots using GC calibration curves.
These calibration curves were obtained using authentic
PhSH, PhSSPh, and PhSO SPh. For PhSO SPh, a photo-
degradation calibration was necessary because it is photo-
labile and breaks down during the course of the experiment.
The use of a photodegradation calibration run under identi-
cal conditions to the experiment provided an estimated cor-
rection. The calibrated concentrations were transformed to
molar “% conversion” relative to the starting concentration
of PhSH.
2
2
3
0 cc/min. One microliter injection volumes were made by
an autoinjector using a 10 µL syringe that was cleaned with
acetonitrile (MeCN). The GC–MS used was a HP 5890 gas
chromatograph with a HP 5970 series mass selective detec-
tor. The MS library used was the Wiley138 database. One
microliter manual injections were made. The column used
was a HP-5 with dimensions 25 m × 0.2 mm × 0.33 µm, col-
umn pressure 10 psi He, and splitflow 30 cc/min. The opti-
mal GC method for analyzing the thiol-nitrate samples used
an injection port temperature of 200 °C, a detector tempera-
ture of 300 °C, and an initial oven temperature of 50 °C with
a gradient of 10 °C/min to a final temperature of 250 °C.
This temperature was maintained for an additional 9 min.
All high-pressure liquid chromatography (HPLC) analyses
were done at room temperature on Hewlett Packard series II
IPN (0.5 mol/L) with and without PhSH (10 mmol/L) in
MeCN–N , GTN (5 mmol/L), and GT-130 (5 mmol/L) in
2
MeCN–PB–N (40:60), were each irradiated using Rayonets
2
with 300 and (or) 350 nm lamps, and NO release was quan-
tified using a NO chemiluminescence (CL) detector. CL was
calibrated by the use of the diazeniumdiolate NO donor,
SPE–NO: initial rates data were collected for NO release
over 10 min and calibrated by comparison to the first-order
rate constant obtained by spectrophotometric analysis of
SPE–NO decay.
Solutions of 10 mmol/L PhSH, 1.0 mol/L IPN, and vary-
ing concentrations of 0–100 mmol/L 4-chlorophenol in
1
090 liquid chromatographs using diode array spectropho-
tometers (HPIB 15). An autosampler was used with 25 µL
injections. Both C8 and C18 reverse-phase columns were
used. The C8 column was a Waters Nova-Pak HPLC col-
umn, with a 4 µm diameter and dimensions of 3.9 mm ×
MeCN–N were irradiated for 0 and 10 min in a 300 nm
2
Rayonet reactor. To determine the extent of photolysis at
longer wavelength, solutions of
5 mmol/L PhSSPh,
1.0 mol/L IPN, and 0–100 mmol/L p-chlorophenol were
deoxygenated with nitrogen and irradiated for 0 and 45 min
in a 350 nm Rayonet reactor. Anhydrous decane was used as
a standard in both experiments and product analysis was
done by GC-FID.
1
50 mm. The C18 column was a Waters Spherisorb ODS2
column with a diameter of 5 µm and dimensions of 4.6 mm ×
50 mm. For all experiments, the eluant was a mixture of
2
methanol (MeOH) and Millipore-filtered water, with a flow
rate of 1 mL/min. Three HPLC methods were used, where A
Nitric oxide gas was purified according to the method of
Bostrup et al. (30). Using this purification setup, three
25 mL portions of MeCN were each saturated with argon.
One was flushed with purified NO for 25 min (until satu-
rated) and another with unpurified NO. A stock solution was
made with 20 mmol/L PhSH and 20 mmol/L decane stan-
dard. One milliliter of this solution was saturated with nitro-
gen and diluted with 1 mL of either (i) argon-saturated,
(ii) purified NO-saturated, or (iii) unpurified NO-saturated
MeCN, giving final concentrations of 10 mmol/L PhSH and
10 mmol/L decane. Another stock solution was made with
60 mmol/L p-chlorophenol and 20 mmol/L decane. This so-
lution was deoxygenated with nitrogen and diluted with the
satd. MeCN solutions similar to the above method. The
resulting concentrations were 30 mmol/L and 10 mmol/L p-
chlorophenol and decane, respectively. All samples were an-
alyzed by GC-FID with decane as a standard.
is H O and B is MeOH. Method 1 used a gradient from 35%
2
to 95% B over 16 min, followed by a plateau at 95% B for
6
min and gradient to 35% B over 2 min. Method 2 used a
gradient from 35% to 95% B over 30 min, followed by a
min plateau. Products from the PhSH/IPN photoreaction
5
were detected using HPLC Method 2 and the C18 column.
HPLC Method 2 and the C8 column were used for the GT-
1
30 photochemical and thermal product analysis. Detection
of GDNs from GTN electron capture was done using
Method 1 with the C18 column.
Benzenethiyl radical was generated by irradiation of
benzenethiol (1) (PhSH, 10 mmol/L), diphenyl disulfide
(
PhSSPh, 4 or 1 mmol/L), or GT-130 (1 mmol/L) in the
presence of 100 mmol/L 2,3-dimethyl-1,3-butadiene in oxygen-
or nitrogen-saturated acetonitrile (MeCN/N ) using either a
2
3
00 nm Rayonet reactor (arrangement of UV lamps) or a la-
ser flash photolysis setup, respectively. Decane (3 mmol/L,
anhydr.) was used as a standard for GC-FID analysis of the
Rayonet reactions. GC–MS was first used to identify the
Theoretical methods
Geometry optimizations were performed with density
functional theory using the (U)B3LYP function (31) with the
6-311++G(d,p) basis set implemented in the Gaussian 03
suite of programs (32). Calculation of structure and energy
of the excited state of IPN was carried out in Spartan 04
(Wavefunction, Inc.) using TDDFT at the B3LYP/6-31G*
level. For calculation of the thermodynamics of homolysis,
minima were fully optimized and characterized by harmonic
1
,4- and 1,2-addition products, the identity of which was
1
further confirmed by H NMR analysis and quantified by
GC-FID.
Benzenethiyl radical was also generated in the presence of
isopropyl nitrate (IPN, 100 mmol/L or 1.0 mol/L) by irradi-
ating PhSH (10 mmol/L or 100 mmol/L) or PhSSPh
(
5 mmol/L) in nitrogen-saturated acetonitrile (MeCN–N2)
©
2006 NRC Canada

7
12
Can. J. Chem. Vol. 84, 2006
vibrational frequency analysis, followed by further optimiza-
tion and energy evaluation using the complete basis set
DMB, the only product observed by GC-FID analysis was
disulfide. However, with excess quantities of DMB present,
the yield of disulfide was reduced and both the 1,4- and 1,2-
(
CBS-QB3) methodology, known to provide thermo-
1
chemical estimates approaching experimental accuracy. En-
ergy minimum structures were confirmed to have adequate
convergence and zero imaginary vibrational frequencies. For
the radical species, spin determinant calculations were eval-
uated to check for spin contamination.
adducts were observed by GC-FID and H NMR analysis,
with the major product being that of conjugate addition.
Thermal generation of benzenethiyl at 95 °C and subsequent
addition to DMB has previously been investigated, where
the yield of the 1,4-adduct was found to be 98% (35). DMB
is therefore an effective thiyl trap as predicted by the rate
constants determined by LFP analysis. Addition of IPN to
this reaction system led to a decrease in the yield of addition
product in a concentration dependent manner (data not
shown), suggesting that IPN inhibited reaction of thiyl radi-
cal with the conjugated diene.
Results
The photochemical sources of thiyl radicals used were
photolysis of the RS—H bond (eq. [5]) and photolysis of the
S—S bond (eq. [6]) of the corresponding disulfide (33). Hy-
drogen abstraction from RSH by photochemically generated
To examine whether a reaction between benzenethiyl and
IPN was responsible for the observed reduction in formation
of the 1,4-DMB adduct, PhSH (10 mmol/L) was irradiated
in the presence of IPN (1.0 mol/L) alone. A new product
was observed in the GC-FID chromatogram, which was
identified as a sulfonothioate (S-phenyl benzenesulfonothio-
ate; RS(O)2SR, R = Ph) by GC–MS and by comparison to
an authentic sample. Formation of this product increased
with irradiation time (Fig. 1); however, the sulfonothioate is
a photolabile compound and its quantity is therefore under-
estimated by this assessment, especially at longer irradiation
times. As a result, the sulfonothioate product stability was
measured by GC-FID analysis as a function of irradiation
time under the reaction conditions. This enabled construc-
tion of a corrected calibration curve that was used to more
accurately estimate sulfonothioate formation (Fig. 1).
It is evident from Fig. 1 that in addition to a decrease in
PhSH with irradiation time, a decrease in the disulfide prod-
uct is also observed. The disulfide absorbs light at 300 nm
and is cleaved into thiyl radicals, though requiring longer ir-
radiation times than the corresponding thiol. This is compat-
ible with the increase in sulfonothioate product observed at
irradiation times where the PhSH source has been depleted.·
The sulfonate product therefore results from reaction of PhS
formed from photolysis of both the thiol and the disulfide.
·
tert-butoxyl radicals (t-BuO ) is an alternative route (eq. [7])
that is complicated by the potential side reactions of the
alkoxyl radical in solutions containing nitrates.
hv
·
·
[
[
5]
6]
RSH ⎯⎯→ RS + H
hv
·
RSSR ⎯⎯→ 2RS
hv
·
RSH
·
[
7]
t-BuO–O-t-Bu ⎯⎯→ t-BuO ⎯⎯⎯→ RS
t-BuOH
A restriction to these methods is the chromophore of the
+
source, which is often weak. Alkyl thiyl radicals typically
absorb at 330 nm, which is similar to the optical spectra of
their sources, and have low extinction coefficients (ε ~
–
1
–1
5
00 (mol/L) cm ) (27). They are therefore difficult to de-
tect and monitor using photochemical techniques (27, 34).
Generation of aryl thiyl radicals is more facile owing to
higher wavelength chromophores with higher extinction co-
–
1
–1
efficients (ε ~ 2500 at 460 nm and 10 000 (mol/L) cm at
2
95 nm) (27). Consequently, aryl thiols and aryl disulfides
are preferred as thiyl radical sources. A relatively clean
method of generating thiyl radicals was found by irradiation
in MeCN using a YAG laser (λ = 355 nm). Benzenethiyl
·
radical (PhS ) was generated from the disulfide or thiol pre-
·
PhS was also generated from 5 mmol/L PhSSPh using
cursor and characterized by laser flash photolysis (LFP). The
·
LFP. A second-order fit to the decay trace (monitored at
PhS spectrum obtained displayed the well-characterized
4
2
60 nm) gave a rate constant for radical recombination of
× 10 (mol/L) s . Addition of IPN had no effect on the
absorbance maxima at 490 and 450 nm.
7
–1 –1
The thiyl radical was further characterized by measuring
the kinetics of conjugate addition to 2,3-dimethyl-1,3-
butadiene (DMB). Under anaerobic conditions, thiyl is
known to add to olefinic bonds in anti-Markovnikov fashion
decay rate of the radical, even when 80% IPN – 20% MeCN
was used as the solvent, indicating that any reaction with
IPN cannot compete with radical recombination. This effect
was consistent under aerobic and anaerobic conditions, and
questions the viability of a direct and rapid reaction between
thiyl radical and nitrate.
(
27), where the 1,4-adduct is expected to be the major prod-
uct in the case of DMB (35). LFP was used to measure the
addition rates of PhS to DMB in the presence and absence
·
of oxygen. The rate constants thus derived for the addition
·
of PhS to DMB under aerobic and anaerobic conditions
Sulfonothioate as a product of NO reaction
The sulfonothioate product could potentially result from
direct reaction between benzenethiyl radical and NO , since
2
8
8
–1 –1
were found to be 1.3 × 10 and 0.83 × 10 (mol/L) s , re-
spectively. These experiments validate and quantify the for-
mation of thiyl radical, the rate of addition to the diene, and
the utility of the diene trap in measuring thiyl formation.
2
NO is the expected product of IPN photolysis. Smaller
2
amounts of sulfonothioate were observed in photochemical
experiments where longer wavelength irradiation was used:
·
·
Generation of PhS in the presence of IPN — An
reaction of PhS with IPN was performed using a 350 nm
intermolecular model
lamp and PhSSPh as the thiyl precursor. As shown in Fig. 2,
much smaller amounts of sulfonothioate were detected, in
comparison to the irradiation of PhSH and IPN at 300 nm. It
should be noted that at λ = 300 nm, the relative molar
To facilitate product monitoring, photolysis of PhSH in
the presence of DMB was carried out using a Rayonet
photoreactor fitted with 300 nm lamps. In the absence of
©
2006 NRC Canada

Clarke et al.
713
Fig. 1. Left: Decay of reactant (᭹) and growth of products (thiosulfonate (᭿) and disulfide (᭡)) from irradiation of PhSH (10 mmol/L)
in the presence of IPN (1.0 mol/L) in MeCN–N using a 300 nm lamp, as assayed by GC-FID calibrated using authentic samples (av-
2
erage of two trials). Right: Identical reactant and product data incorporating a correction for the photochemical decomposition of the
products under the reaction conditions.
7
6
5
4
3
2
1
0
70
40
3
3
2
2
1
1
5
0
5
0
5
0
5
0
60
5
4
3
2
1
0
0
0
0
0
0
5
10
15
0
Irradiation time (min)
0
5
10
15
Irradiation time (min)
experiment was designed using 4-chlorophenol. Photonitra-
tion of arenes has been known for many years and recent
work has been published reporting the formation of 2- and
4-nitrophenol resulting from irradiation of nitrate or nitrite
ion in the presence of phenol in aqueous solution (36, 37).
The ortho-nitration of tyrosine is a well-known reaction (38),
and the use of 4-chlorophenol has the advantage of limiting
nitration to one isomeric product. Generation of 4-chloro-2-
nitrophenol was shown to result from irradiation of
1.0 mol/L IPN in the presence of 10 mmol/L p-chlorophenol,
where the product identity was confirmed by GC–MS analysis.
Fig. 2. Comparison of PhS(O) SPh yield from IPN (1.0 mol/L)
2
in MeCN–N . Irradiation of PhSH (10 mmol/L) at 300 nm (᭿);
2
irradiation of PhSSPh (5 mmol/L) at 350 nm (᭹). Average of
two trials.
1
.4
.2
1
1
Increasing concentrations of 4-chlorophenol were added
0.8
0.6
0.4
0.2
0
to irradiated solutions of IPN with PhSH or PhSSPh to com-
·
pete with PhS for any NO generated by nitrate photolysis
2
(
Fig. 3). The decrease in relative peak area of 4-chloro-
phenol was the same as the increase observed in the relative
peak area of 4-chloro-2-nitrophenol (data not shown), indi-
cating that nitration was efficient and was the only reaction
occurring with 4-chlorophenol. The observation of 4-chloro-
2
-nitrophenol formation confirmed the generation of NO on
2
irradiation, while nonirradiated solutions gave no nitration
products. Clearly, NO2 is generated by irradiation of
.0 mol/L IPN at 300 nm and to a lesser extent at 350 nm. It
0
5
10 15 20 25 30 35 40 45
1
Irradiation time (min)
also appears from Fig. 3 that the amount of 4-chloro-2-
nitrophenol formed reaches a plateau. A simple rationale
·
would be the lower yield of PhS and of NO owing to com-
absorbance of PhSH to IPN is approximately 10:1, whereas
at 350 nm, the absorbance of IPN is lower by a factor of 40.
2
petitive absorbance by 4-chlorophenol and 2-nitro-4-
chlorophenol in the reaction solution. At 100 mmol/L 4-
chlorophenol, A(300 nm) = 2.4 and A(350 nm) = 0.079;
whereas for PhSH (10 mmol/L), A(300 nm) = 0.85; for
PhSSPh (5 mmol/L), A(350 nm) = 0.75; and for IPN
1.0 mmol/L), A(300 nm) = 3.4 and A(350 nm) = 0.087.
Both the nitration of 4-chlorophenol and the formation of
Thus, it is expected that less NO , and consequently less
2
sulfonate, would be generated at 350 nm as a result of de-
creased light absorption and photolysis of the nitrate. Evi-
dence for thermal decomposition of IPN to NO was seen in
2
(
careful inspection of Fig. 1, which shows sulfonothioate
product present at time zero. Sulfonothioate was formed ow-
ing to thermolysis in the high-temperature GC injection port
sulfonothioate appeared to result from the photolysis of IPN
(
200 °C). To examine the photolysis of IPN, a NO trapping
to NO . Commercial gaseous NO provides a further means
2
2
©
2006 NRC Canada

7
14
Can. J. Chem. Vol. 84, 2006
Fig. 3. Formation of 4-chloro-2-nitrophenol from irradiation of
IPN (1.0 mol/L) in the presence of (top) 10 mmol/L PhSH and
p-chlorophenol using a 300 nm lamp and (bottom) 5 mmol/L
PhSSPh and p-chlorophenol using a 350 nm lamp. Averages of
two trials.
Fig. 4. Quantitation of products from reaction of NO gas with
PhSH (10 mmol/L) in MeCN–N by GC-FID. Argon, unpurified
2
NO–NO gas, or purified NO gas passed through alkaline scrub-
2
bers was bubbled through the reaction solution at comparable
rates. Top: reactant PhSH (open bar), PhS(O) SPh (solid bar),
2
and PhSSPh (hashed bar) were measured. Bottom: the same re-
actions were carried out in the presence of 4-chlorophenol (open
bar) and product 4-chloro-2-nitrophenol (solid bar) was mea-
sured. Triplicate reactions shown.
0
0
0
0
.15
.10
.05
.00
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
trl
trl
trl
d
)
)
d
c
c
c
fied)
fie
fie
ified)
ied)
ied)
if
if
0
.0
20.0
40.0
60.0
80.0
100.0
Ar
Ar
Ar
N
uri
ur
ur
ur
p
p
p
npuri
np
npuri
(
(
(
u
u
u
(
(
(
O
O
N
O
[p-Chlorophenol] (mmol/L)
O
O
N
N
N
NO
0.015
0.010
0.005
0.000
2
1
0
.5
2
.5
1
.5
0
0.0
20.0
40.0
60.0
80.0
100.0
[p-Chlorophenol] (mmol/L)
)
)
)
)
d
trl
trl
trl
d)
d
d
d
fie
fied)
c
c
c
ie
ie
ie
if
ifie
if
if
r
r
r
r
u
Ar
Ar
Ar
uri
u
uri
u
u
p
p
(
p
p
(
np
n
np
u
(
(
u
u
(
(
O
O
O
O
O
O
N
N
N
N
N
N
to test this hypothesis because the unscrubbed gas contains
didate for studying the likelihood of thiyl radical reactivity
with nitrates owing to the proximity of the two reactive
groups. In the presence of DMB and under identical condi-
tions to those used for the study of PhSH–IPN, irradiation of
GT-130 resulted in formation of the same 1,4-thiyl-DMB
adduct, verifying that PhS was produced and therefore con-
firming photolysis of the S—S bond.
NO as the major impurity. Nitration of 4-chlorophenol was
2
measured in solutions of PhSH in MeCN–N treated with
2
gas streams of either argon, the unpurified NO–NO , or NO
2
purified by passage through alkaline scrubbers (Fig. 4, bot-
tom). Only in the samples treated with NO–NO was signifi-
2
·
cant nitration observed, indicating that this system is useful
for studying reactions of NO in the presence of other nitro-
2
gen oxide species and that 4-chlorophenol is an effective
NO trap. In parallel experiments, unreacted thiol, disulfide,
and sulfonothioate were measured by GC-FID analysis in
2
S
S
S
hν
solutions of PhSH in MeCN–N treated with gas streams of
ONO
ONO2
2
2
ONO2
ONO2
S
+
either argon, unpurified NO–NO , or purified NO. Treat-
λ=308nm
2
ment with purified NO resulted in significant oxidation of
thiol to disulfide, but no sulfonothioate formation. In con-
trast, treatment with NO–NO gas led to complete depletion
Using an Excimer laser (308 nm) to excite GT-130, PhS·
was again identified by LFP analysis of its characteristic
spectrum and shown to be a product of GT-130 photolysis.
2
of thiol, yielding both disulfide and sulfonothioate (Fig. 4,
top). The nitration of 4-chlorophenol in PhSH solution
·
flushed with NO–NO was also monitored in a UV–vis spec-
The decay curves of PhS generated by photolysis of GT-130
2
trometer at 350 nm in which 4-chloro-2-nitrophenol was ob-
served to form immediately upon flushing with the gas (data
not shown).
and Ph S at equal concentrations are shown in Fig. 5. As
2
2
expected, the change in optical density from Ph2S2
photolysis was approximately twice that from GT-130
·
photolysis, since twice as many PhS radicals would result
An intramolecular model for thiyl-nitrate interactions —
GT-130
The phenyldisulfanyl dinitrate (GT-130) contains both
disulfide and nitrate moieties, and is therefore a suitable can-
from photolysis of the S—S bond. The decay rate constants
measured for each curve were second order and were found
to be similar, indicating that the recombination pathway for
the GT-130 radicals has similar kinetics to the radical–radi-
©
2006 NRC Canada

Clarke et al.
715
Fig. 6. Comparison of cumulative NO response. Irradiation of
·
Fig. 5. Decay curves of PhS monitored at 450 nm by LFP.
Formed by irradiation of 1 mmol/L GT-130 (ٗ) or 1 mmol/L
5 mmol/L GTN (᭺) and 5 mmol/L GT-130 (᭿) in MeCN–PB–
PhSSPh (᭿) in MeCN–N using an Excimer laser (308 nm).
N (40:60) using mixed 300 and 350 nm lamps (average of three
2
2
trials).
–
2
2
2
2
2
2
2
3
4
.0×10
–
3
.5×10
20
–
–
–
–
–
–
3
2
2
1
1
5
.0×10
15
.5×10
.0×10
.5×10
.0×10
.0×10
0
1
0
5
0
.0
0
2
4
6
8
0
5
10
15
20
25
Time (min)
Irradiation time (min)
·
cal recombination of PhS forming PhSSPh. This suggests
that the PhS radical formed via photolysis undergoes recom-
bination to GT-130 with a rate similar to that of self-
combination to PhSSPh.
To explore possible nitrate photolysis of GT-130, irradia-
tion was performed at 300 nm in the presence of the 4-
chlorophenol trap. Nitration was not observed (data not
·
Fig. 7. Comparison of NO response. Irradiation of 10 mmol/L
PhSH + 0.5 mol/L IPN (᭹) vs. irradiation of 0.5 mol/L IPN (᭿)
in MeCN–N (average of three trials).
2
180
160
shown), indicating that NO either was not formed or was
2
140
120
trapped more rapidly by another species. A reasonable ex-
planation is the low absorbance associated with the nitrate
groups in GT-130 (5 mmol/L), which is substantially less
than in experiments with IPN (1.0 mol/L). Disulfide cleav-
age is therefore the dominant route of photolysis upon irra-
diation of GT-130 at 5 mmol/L.
100
80
60
40
20
0
NO release from reaction of GT-130
The thermal reaction of GT-130 with thiols has been re-
ported to result in production of NO (29). The photolysis of
GT-130 was compared with that of GTN by
chemiluminescence detection of NO in headspace gas of the
reaction mixtures (Fig. 6). Small, submicromolar quantities
of NO were observed from GTN photolysis, but GT-130
photolysis generated more than 20 µmol/L NO after 20 min
irradiation. However, comparison of NO release from ther-
mal reaction of GT-130 with cysteine vs. photochemical re-
action of GT-130 showed that the thermal reaction is more
efficient in generating NO (29). Irradiation of GT-130 under
aerobic or anaerobic conditions exhibited no change in the
NO response (data not shown).
0
10
20
30
40
Irradiation time (min)
have been reported previously (29). The major products were
identified by comparison of HPLC chromatograms with au-
thentic samples, and were found to be PhSSPh and 2-
denitro-GT-130, with an equal amount of unreacted GT-130
present (Scheme 1). Of numerous minor products, the cyclic
thiirane was the most abundant and the symmetrical disul-
fide 1,2-di(3-nitrooxy-2-hydroxypropyl)disulfane was also
identified (Scheme 1). The products identified in the photo-
lysis of GT-130 (excluding the starting material) were, in or-
der of abundance, diphenyldisulfide > 2-denitro-GT-130 >
1,2-di(2,3-dinitrooxypropyl)disulfane > 3-nitrooxypro-
pylthiirane > 1,2-di(3-nitrooxy-2-hydroxypropyl)disulfane
(Scheme 1). In the context of this work, it is important to
Interestingly, irradiated solutions of PhSH–IPN in MeCN–
N also generated NO, with a linear rate of NO release for at
2
least the first 20 min of reaction (Fig. 7). As shown,
photolysis of IPN alone did not produce NO within the de-
tection limits.
Once it was established that irradiation of GT-130 resulted
in efficient S—S bond cleavage, the nature of the resulting
organic products was investigated. The products resulting
from thermal reaction of GT-130 with PhSH in neutral aque-
ous solution were analyzed by HPLC with UV detection and
note that no PhS(O) SPh was detected in either the
2
thermochemical or photochemical reactions of GT-130. In
©
2006 NRC Canada

7
16
Can. J. Chem. Vol. 84, 2006
Scheme 1. HPLC retention times are annotated under the conditions described in the text.
Not observed
O
S
S
PhSH
or hv
S
S
S S
S
S
S
S
S
OH
ONO
ONO
ONO
ONO
OH
OH
S
S
ONO
2
2
2
2
2
ONO
2
O
ONO2
ONO2
ONO
2
ONO
2
GT-130
9.3 min
1
8.5 min
21.7 min
14.4 min
16.5 min
12.0 min
4.4 min
13.6 min
contrast, HPLC analysis of reaction mixtures of PhSH and
IPN did reveal sulfonothioate after irradiation.
to result from radical combination based on computational
studies (eqs. [14] and [15]). The lack of disulfide product
formed from reaction of pure samples of thionitrate indi-
cated that rxns. [16]–[18] (and thiyl radical recombination)
were not competitive under the reaction conditions.
Irradiation of GT-130 for 20 min at 300 nm was sufficient
to identify the photoproducts. Further experiments revealed
no change in photochemical products from nitrogen-
deoxygenated to air-equilibrated solutions. It can therefore
be confirmed that oxygen has no effect on the photochemi-
cal reaction of GT-130 leading to the products shown.
[
[
[
[
[
[
[
10] RSH + R′ONO → RSNO + R′OH
2
2
·
·
11] RSNO → [RS NO ]
2
2
·
·
12] [RS NO ] → RSONO or RS(O)NO
2
Discussion
·
13] RSONO → RSO + NO
The release of NO from organic nitrates in complex bio-
logical systems can often be observed using a variety of dif-
ferent methods. In human plasma, NO release was observed
from solutions of nitrates and thiosalicylic acid using
amperometric detection (39). In plasma, the series of 1,2-
dinitrooxy propane derivatives including GTN were all ob-
served to release NO in the presence of TSA with an initial
rate of ~4–80 nmol/L s–1 at 0.25–0.5 mmol/L nitrate concen-
tration. At physiological pH, the reactivity of free thiols to-
·
14] 2RSO → RS(O) SR
2
·
·
15] RSO + RS → RS(O)SR
·
·
·
·
16] [RS NO ] → RS + NO
2
2
–
·
·
–
[17] RS + NO → RS + NO
2
2
–
–
[
18] RS + RSNO → RSSR + NO
2
2
wards nitrates is dependent on thiol pK and is too slow to
account for the rapid bioactivation of GTN; for example,
only 10% reaction of GTN (1 mmol/L) with cysteine
a
GT-150 (2,3-dinitrooxypropane-1-thiol) contains a thiol
group vicinal to a nitrate group. This compound was shown
to spontaneously generate NO in neutral aqueous solution,
as was the related disulfanyl nitrate GT-130 on addition of
thiol (Scheme 1) (29), with no evidence being obtained for
sulfonyl or sulfinyl products (S=O containing). Reaction of
(
3
2 mmol/L) is observed in aqueous solution at pH 7.4,
7 °C, after 1 h (26). Using earlier electrochemical devices,
this reaction was not found to produce NO within the detec-
tion limits, while chemiluminescence was able to detect the
low levels of NO produced in these mixtures (26, 40). Nev-
ertheless, catalysis of the reaction between cysteine thiol and
nitrate at an enzyme active site or at other protein sites con-
taining reactive thiols, such as thioredoxin folds, may lead to
acceleration of reaction and an increased flux of NO (6).
The most popular, contemporary theory for GTN bioacti-
vation invokes ALDH2, which contains three cysteines at the
active site (41, 42). ALDH2 and several other proteins are
capable of accelerating reductive denitration of nitrates to in-
organic nitrite (eq. [8]), however, conversion to NO requires
further reduction.
the disulfanylnitrates was unaffected by O , and other
2
reductants studied did not facilitate NO release. The simplest
mechanism, compatible with these observations was proposed
to be a thiol-disulfide exchange.
–
[19] RSSCH CHRONO + R′S → RSSR′
2
2
+ –
SCH CHRONO2
2
–
[
20]
SCH CHRONO → → NO
2 2
A sequence of reactions postulated to account for NO pro-
duction in the absence of S=O containing products proceeds
–
via a sulfenate (RSO ) and a nitrosothiol (RSNO) intermedi-
–
+
–
[
[
8]
9]
RONO + 2e + 2H → ROH + NO
2 2
ate (eqs. [21]–[24]). The spontaneous release of NO from
nitrosothiols in the presence of thiol is a well-studied phe-
nomenon (46, 47).
NO2– + e + H → HO + NO
–
+
–
A
chemical mechanism that identified an organic
–
–
[
21] RS + R′ONO → RSO + R′ONO
2
thionitrate (RSNO ) as the common intermediate in a mech-
anism leading to either NO2 or NO from the reaction of
thiol with nitrate was first described by Yeates (43). Later
computational and experimental approaches to this reaction
yielded a cohesive thionitrate rearrangement pathway for a
thiol-nitrate reaction leading to NO release (eqs. [10]–[13])
2
–
–
+
[22] RSO + RSH + H → RSSR + H O
2
[
[
23] R′ONO + RSH → RSNO + R′OH
·
24] RSNO → RS + NO
(
44, 45). In an investigation of the aqueous decomposition of
Alternative mechanisms were considered, including an
inner-sphere electron-transfer process via an intermediate in
thiol-disulfide exchange or a disulfide radical anion. Since
tert-butyl thionitrate, the only organic products observed
were sulfonothioate and sulfinothioate, which were proposed
©
2006 NRC Canada

Clarke et al.
717
–
the overall reaction is a 3e reduction, radicals in addition to
NO must be involved or released at some stage of the reac-
tion. In many such reaction mechanisms a thiyl radical is a
by-product.
owing to the negligible absorbance of organic nitrates at
350 nm compared with 300 nm.
Nitration of 4-chlorophenol was used to confirm the ni-
trate photolysis pathway. The nitration of phenols by NO2,
involving oxidation and radical addition, is a well-known
reaction that leads to tyrosine nitration in vivo. 4-Chloro-
To initiate studies into the reactivity of S-centred radicals
with nitrates, thiyl radical, the simplest of such radicals, was
studied using photochemical systems to generate the radical
in the presence of a simple aliphatic nitrate (IPN). The PhS·
radical was generated by photolysis of benzene thiol or the
disulfide and characterized by LFP analysis of its character-
istic spectrum and decay rates in the presence of DMB. The
conjugate addition of the thiyl radical to the DMB diene was
also used as a measure of the ability of the nitrate to react
with thiyl radical and inhibit the addition reaction. Signifi-
cant inhibition of the formation of addition product from
phenol was used as a NO trap, where formation of 4-
2
chloro-2-nitrophenol revealed release of NO upon irradia-
2
tion of PhSH–IPN at 300 nm. In contrast, significantly less
NO2 was observed from irradiation of PhSSPh–IPN at
350 nm, compatible with the very weak absorbance of ni-
trates at higher wavelength.
The use of the 4-chlorophenol trap was validated by ob-
servation of nitration by unpurified NO gas containing NO2
impurity, whereas purified NO gave almost no nitration
·
reaction of PhS with DMB was only observed at high con-
product. Treatment of a PhSH solution with NO–NO gave
2
centrations of the nitrate ([IPN] = 1.0 mol/L). The reaction
of the benzene thiyl radical with IPN at this high concentra-
tion was studied by identifying the reaction products and
quantifying formation as a function of irradiation time. In
addition to the expected disulfide product of thiyl radical
substantial formation of PhS(O) SPh, which was not ob-
served with purified NO gas, supporting photolysis of IPN to
2
NO as the source of sulfonothioate. The thermal experi-
2
ments with the gaseous NO–NO source confirm that reac-
2
tion of NO with thiyl radical (eq. [26]) competes both with
2
combination, the sulfonothioate (PhS(O) SPh), was ob-
thiol oxidation and with radical combination with NO
(eqs. [29]–[31]), which is in accord with literature rate data.
2
served. The observation of this product under the anaerobic
reaction conditions demonstrated that the photochemical re-
action of thiyl with nitrate involved O atom transfer from ni-
trogen to sulfur.
Direct reaction of the electrophilic thiyl radical with the
HOMO of the nitrate (localized on the nonbonding orbitals
of the terminal oxygens) would yield a radical intermediate
leading to O atom transfer (eq. [25]). Homolysis of this in-
termediate yields sulfinyl radical and organic nitrite, with
subsequent sulfinyl radical combination leading to the ob-
served sulfonothioate product. An important feature of this
mechanism is that the organic nitrite product is a known
source of NO (48).
·
[
[
[
29] RSH + NO → RS + HNO
2
2
·
30] RS + NO → RSNO
2
2
31] NO + NO → N O
3
2
2
In addition to photolysis to NO , the irradiation of PhSH–
2
IPN solutions generated a reasonable flux of NO, with the
presence of thiol being a requirement. This novel observa-
tion is entirely compatible with the rearrangement of the
thionitrate intermediate to yield NO (eq. [27]), which we
have previously reported in the context of the thermal de-
composition of a thionitrate (44).
A large part of the early work on organic nitrate reactivity,
including photochemistry, is part of the impressive ouevre of
Canadian carbohydrate chemists, and published in this jour-
nal (49, 50). From these and other studies, it is seen that var-
ious benzyl nitrates are photolabile, whereas simple aliphatic
nitrates require extended reaction times to observe photo-
lysis products (51, 52). Work by Csizmadia and Hayward
investigating the photolysis of hydrobenzoin and acenaph-
thenediol dinitrates in benzene found the predominant pro-
[
25]
Ph
0.5 PhS(O) SPh
2
S
PhS
iPrONO2
O
PhSO
•
+
O
N
+
+ iPrONO
O
NO
However, a sulfonothioate is also the major product of the
thermal decomposition of tert-butyl thionitrate (t-BuSNO2)
in neutral aqueous solution at room temperature (44), hinting
at the formation of PhSNO as an intermediate in the photo-
2
cess to be photolysis to NO (49, 50). This was validated by
2
·
chemical reaction of PhS with IPN. The simplest route to
the detection of NO as a reaction product using ESR spec-
2
thionitrate is by the rapid radical combination of NO with
2
troscopy (49). There is no evidence for direct photolysis
leading to NO.
thiyl radical.
·
·
Calculation of the thermodynamics for homolysis of IPN
reveals the BDE and ∆G of reaction to be 43.7 and
31.6 kcal/mol, respectively, (1 cal = 4.184 J) employing
CBS-QB3//B3LYP methodology. The structure of the first
excited state was calculated and the energy shown to be
54 kcal/mol above the ground state (Table 1). The calculated
BDE for a nitrate is substantially larger than that for
homolysis of the S—N bond of a nitrosothiol as shown by
comparison of the thermochemical data calculated for
[
[
[
26] PhS + NO2 → PhSNO2
27] PhSNO → PhSONO → PhSO + NO
2
28] 2PhSO → 0.5PhS(O) SPh
2
Evidence against the direct reaction of thiyl with nitrate
·
was seen in the photochemical generation of PhS from irra-
diation of PhSSPh at 350 nm in which sulfonothioate forma-
tion was greatly reduced from the 35% yield observed on
irradiation of PhSH at 300 nm. This observation is compati-
MeONO and MeSNO, which reveals a BDE for the nitrate
2
ble with thionitrate formation from NO formed by the ini-
tial photolysis of IPN, which would be greatly reduced
greater by 11 kcal/mol. Nitrosothiols are well-known to be
labile to photolysis to NO. Nitrosothiol photolysis is facili-
2
©
2006 NRC Canada

7
18
Can. J. Chem. Vol. 84, 2006
Table 1. Calculated homolysis thermodynamics for IPN com-
8. B.M. Bennett and G.S. Marks. Trends Pharmacol. Sci. 5, 329
pared with methyl nitrate and thionitrite.
(1984).
9
. G.R.J. Thatcher and H. Weldon. Chem. Soc. Rev. 27, 331
1998).
Energy (kcal/mol)
(
Isodesmic reaction
E°a
BDE
∆G
10. J.J. Doel, B.L. Godber, R. Eisenthal, and R. Harrison.
Biochim. Biophys. Acta, 1527, 81 (2001).
i-PrONO → i-PrO + NO2
44.0
42.0
31.0
43.7
43.2
31.9
31.6
30.6
20.9
2
1
1. B.M. Bennett, S.M. Kobus, J.F. Brien, K. Nakatsu, and G.S.
MeONO → MeO + NO2
2
Marks. J. Pharmacol. Exp. Ther. 237, 629 (1986).
MeSNO → MeS + NO
12. J.D. Ratz, J.J. McGuire, D.J. Anderson, and B.M. Bennett. J.
Pharmacol. Exp. Ther. 293, 569 (2000).
13. Y. Meah, B.J. Brown, S. Chakraborty, and V. Massey. Proc.
Natl. Acad. Sci. U.S.A. 98, 8560 (2001).
Note: Calculated using CBS-QB3//B3LYP/6-311++G**.
Total energy including scaled zero-point energy.
a
1
4. B.J. McDonald and B.M. Bennett. Can. J. Physiol. Pharmacol.
tated by significant absorbance and a λ
above 300 nm, in
max
68, 1552 (1990).
contrast to nitrates.
15. J.J. McGuire, D.J. Anderson, B.J. McDonald, R.
Narayanasami, and B.M. Bennett. Biochem. Pharmacol. 56,
881 (1998).
The disulfanyl dinitrate (GT-130) might be anticipated to
undergo competitive photolysis to yield either PhS or NO .
·
2
Good evidence for efficient thiyl radical formation was ob-
16. R. Nigam, T. Whiting, and B.M. Bennett. Can. J. Physiol.
Pharmacol. 71, 179 (1993).
17. R. Nigam, D.J. Anderson, S.F. Lee, and B.M. Bennett. J.
tained, but no evidence for formation of NO was found.
2
This novel nitrate, containing a phenyl disulfanyl group
linked to an organic nitrate, introduces the possibility of an
intramolecular reaction between thiyl radical and the nitrate
functionality. Photolysis of GT-130 (5 mmol/L) was ob-
served to give a reasonable rate of NO flux, but the rate was
lower than from the thermal reaction of GT-130 with thiol.
Taken together, the results of this study provide no sup-
port for a rapid reaction between thiyl radical and organic
nitrate. Despite weak absorption of light >300 nm and a rel-
atively high BDE for homolysis, the dominant photo-
chemistry under conditions required to generate thiyl radical
Pharmacol. Exp. Ther. 279, 1527 (1996).
1
1
2
2
2
2
2
2
8. Z. Chen, J. Zhang, and J.S. Stamler. Proc. Natl. Acad. Sci.
U.S.A. 99, 8306 (2002).
9. J. Towell, T. Garthwaite, and R. Wang. Alcohol Clin. Exp.
Res. 9, 438 (1985).
0. N. Mukerjee and R. Pietruszko. J. Biol. Chem. 269, 21664
(
1994).
1. K. Schmidt, A. Schrammel, D. Koesling, and B. Mayer. Mol.
Pharmacol. 59, 220 (2001).
2. Y. Zhao, P.E. Brandish, D.P. Ballou, and M.A. Marletta. Proc.
Natl. Acad. Sci. U.S.A. 96, 14753 (1999).
3. S. Chong and H.L. Fung. Biochem. Pharmacol. 42, 1433
was driven by nitrate photolysis to NO . The observation of
2
both sulfonothioate and NO as products of reaction is readily
explained by the formation of a thionitrate intermediate and
provides further evidence that thionitrates are a source of
NO. A novel nitrate containing a phenyl disulfanyl group did
(
1991).
4. L.J. Ignarro and C.A. Gruetter. Biochim. Biophys. Acta, 631,
21 (1980).
2
5. R.A. Yeates, H. Laufen, and M. Leitold. Mol. Pharmacol. 28,
not undergo photolysis to NO , nor generate sulfonothioate,
555 (1985).
2
but did yield NO. The observations with the novel nitrate
suggest that reaction between thiyl radicals and nitrates lead-
ing to NO release is a viable pathway, but is subservient to
other competing reactions, such as photolysis in the case of
IPN and reaction with thiolate in the case of the novel ni-
trate.
26. J.D. Artz, V. Toader, S.I. Zavorin, B.M. Bennett, and G.R.J.
Thatcher. Biochemistry, 40, 9256 (2001).
27. Z.B. Alfassi. S-Centered radicals. John Wiley and Sons,
Chichester, UK. 1999.
28. K.U. Ingold, T. Paul, M.J. Young, and L. Doiron. J. Am.
Chem. Soc. 119, 12364 (1997).
2
3
9. S.I. Zavorin, J.D. Artz, A. Dumitrascu, A. Nicolescu, D.
Scutaru, S.V. Smith, and G.R.J. Thatcher. Org. Lett. 3, 1113
Acknowledgements
(
2001).
0. O. Bostrup, R.S. Tobias, S.S. Hutcheson, and W.H.
Engelmann. In Inorganic syntheses. Edited by H.F.J.
Holtzclaw. McGraw-Hill, New York. p. 191. 1966.
This work was supported in part by a Natural Sciences
and Engineering Research Council of Canada (NSERC)
grant (245617-01) and a National Institutes of Health (NIH)
grant (CA 102590).
3
3
1. A.D. Becke. J. Chem. Phys. 98, 5648 (1993).
2. M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, J.A. Montgomery, Jr., T. Vreven, K.N.
Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V.
Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian,
J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterxki, P.Y. Ayala, K. Morokuma, G.A. Voth,
P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui,
A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu,
References
1
2
3
. T.L. Brunton. Lancet, 2, 97 (1867).
. W. Murrell. Lancet, 1, 80 (1879).
. J.A. Franciosa, A.L. Taylor, J.N. Cohn, C.W. Yancy, S. Ziesche,
A. Olukotun, E. Ofili, K. Ferdinand, J. Loscalzo, and M. Worcel.
J. Card. Failure, 8, 128 (2002).
4
5
. G.R.J. Thatcher. Curr. Top. Med. Chem. 5, 597 (2005).
. G.R.J. Thatcher, B.M. Bennett, and J.N. Reynolds. Curr. Alz-
heimer Res. 2, 171 (2005).
6
7
. G.R.J. Thatcher, A.C. Nicolescu, B.M. Bennett, and V. Toader.
Free Radicals Chem. Biol. Med. 37, 1122 (2004).
. H.L. Fung. Annu. Rev. Pharmacol. Toxicol. 44, 67 (2004).
©
2006 NRC Canada

Clarke et al.
A. Liashenkko, P. Piskorz, I. Komaromi, R.L. Martin, D.J.
719
42. J. DiFabio, Y. Ji, V. Vasiliou, G.R. Thatcher, and B.M.
Bennett. Mol. Pharmacol. 64, 1109 (2003).
Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara,
M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W.
Wong, C. Gonzalez, and J.A. Pople. Gaussian 03. Revision
C.02 [computer program]. Gaussian, Inc., Wallingford, Con-
necticut. 2004.
43. R.A. Yeates. Arzneim.-Forsch./Drug Res. 42, 1314 (1992).
44. J.D. Artz, K. Yang, J. Lock, C. Sanchez, B.M. Bennett, and
G.R.J. Thatcher. J. Chem. Soc. Chem. Commun. 927 (1996).
45. D.R. Cameron, A.M.P. Borrajo, B.M. Bennett, and G.R.J.
Thatcher. Can. J. Chem. 73, 1627 (1995).
46. D.L.H. Williams. J. Chem. Soc. Chem. Commun. 1085 (1996).
47. M.D. Bartberger, J.D. Mannion, S.C. Powell, J.S. Stamler,
K.N. Houk, and E.J. Toone. J. Am. Chem. Soc. 123, 8868
(2001).
3
3
3. C. Chatgillialoglu and K.-D. Asmus. Life Sci. 197, 451 (1990).
4. D.J. McPhee, M. Campredon, M. Lesage, and D. Griller. J.
Am. Chem. Soc. 111, 7563 (1989).
3
5. A.A. Oswald, K. Griesbaum, W.A. Thaler, and B.E. Hudson,
Jr. J. Am. Chem. Soc. 84, 3897 (1964).
3
3
6. D.L. Bunbury. Can. J. Chem. 43, 1714 (1965).
7. D. Vione, V. Maurino, C. Minero, and E. Pelizzetti.
Chemosphere, 45, 893 (2001).
48. A.C. Nicolescu, J.N. Reynolds, L.R. Barclay, and G.R.J.
Thatcher. Chem. Res. Toxicol. 17, 185 (2004).
49. I.G. Csizmadia and L.D. Hayward. Photochem. Photobiol. 4,
657 (1965).
50. L.D. Hayward, R.A. Kitchen, and D.J. Livingstone. Can. J.
Chem. 40, 434 (1962).
51. J.G. Batelaan, H.J. Hageman, and J. Verbeek. Tetrahedron
3
3
4
4
8. T.B. Johnson and E.F. Kohmann. J. Am. Chem. Soc. 37, 1863
(
1915).
9. J.D. Artz and G.R.J. Thatcher. Chem. Res. Toxicol. 11, 1393
1998).
(
0. D. Jourd’heuil, F.S. Laroux, A.M. Miles, D.A. Wink, and M.B.
Grisham. Arch. Biochem. Biophys. 361, 323 (1999).
1. Z. Chen, M.W. Foster, J. Zhang, L. Mao, H.A. Rockman, T.
Kawamoto, K. Kitagawa, K.I. Nakayama, D.T. Hess, and J.S.
Stamler. Proc. Natl. Acad. Sci. U.S.A. 102, 12159 (2005).
Lett. 28, 2163 (1987).
52. M.P. Turberg, D.M. Giolando, C. Tilt, T. Soper, S. Mason, M.
Davies, P. Klingensmith, and G.A. Takacs. J. Photochem.
Photobiol. A, 51, 281 (1990).
©
2006 NRC Canada