para-Substituted N-Nitroso-N-oxybenzenamine ammonium salts: A new class of redox-sensitive nitric oxide releasing compounds

Article abstract of DOI:10.1016/S0968-0896(99)00300-4

N-Nitroso-N-oxybenzenamine ammonium salts with -OMe, -Me, -H, -F, -Cl, - CF₃, and -SO₂Me substituents at the para position of the phenyl ring constitute a new class of-redox sensitive nitric oxide (NO) releasing compounds. These compounds yield nitric oxide and the corresponding nitrosobenzene derivatives by a spontaneous dissociation mechanism after undergoing a one electron oxidation. Oxidation of these compounds can be achieved through chemical, electrochemical and enzymatic methods. It was observed electrochemically that the amount of NO generated was dependent on the substituent effect and the applied oxidation potential. Electron- withdrawing substituents increase the oxidation potential of the compound. A linear correlation was observed when the peak potentials for the oxidation were graphed versus the Hammett substituent constant. Density functional theory calculations were also performed on this series of compounds. The theoretical oxidation energies of the corresponding anions show a strong linear correlation with the experimental potentials. Furthermore, enzymatic oxidation using horseradish peroxidase showed a similar substituent effect. These results indicate that substitution at the para position of the phenyl ring has a profound effect on the stability, oxidation potential and enzymatic kinetic properties of the compounds. Thus para-substituted N- nitroso-N-oxybenzenamine salts comprise a new class of redox-sensitive nitric oxide releasing agents. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00300-4

Bioorganic & Medicinal Chemistry 8 (2000) 405±412
para-Substituted N-Nitroso-N-oxybenzenamine Ammonium Salts:
A New Class of Redox-sensitive Nitric Oxide Releasing
Compounds
Andrea D. McGill, Wei Zhang, Joanne Wittbrodt, Jianqiang Wang,
H. Bernhard Schlegel and Peng George Wang*
Department of Chemistry, Wayne State University, Detroit, MI 48202-3489, USA
Received 1 September 1999; accepted 7 November 1999
AbstractÐN-Nitroso-N-oxybenzenamine ammonium salts with -OMe, -Me, -H, -F, -Cl, -CF3, and -SO2Me substituents at the para
position of the phenyl ring constitute a new class of-redox sensitive nitric oxide (NO) releasing compounds. These compounds yield
nitric oxide and the corresponding nitrosobenzene derivatives by a spontaneous dissociation mechanism after undergoing a one
electron oxidation. Oxidation of these compounds can be achieved through chemical, electrochemical and enzymatic methods. It
was observed electrochemically that the amount of NO generated was dependent on the substituent e€ect and the applied oxidation
potential. Electron-withdrawing substituents increase the oxidation potential of the compound. A linear correlation was observed
when the peak potentials for the oxidation were graphed versus the Hammett substituent constant. Density functional theory cal-
culations were also performed on this series of compounds. The theoretical oxidation energies of the corresponding anions show a
strong linear correlation with the experimental potentials. Furthermore, enzymatic oxidation using horseradish peroxidase showed
a similar substituent e€ect. These results indicate that substitution at the para position of the phenyl ring has a profound e€ect on
the stability, oxidation potential and enzymatic kinetic properties of the compounds. Thus para-substituted N-nitroso-N-oxyben-
zenamine salts comprise a new class of redox-sensitive nitric oxide releasing agents. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
There are several classes of NO releasing compounds
available today (Fig. 1). Di€erent chemical reactions are
utilized to generate NO from these compounds. In vivo
generation of NO from organic nitrates (e.g. nitrogly-
cerin and isosorbide dinitrate), for example, is believed
to occur through thiol-mediated, reductive metabolic
pathways.13 In vitro generation can occur due to spon-
taneous and non-spontaneous dissociation mechanisms
from NO/nucleophile complexes (NONOates or dia-
zeniumdiolates),14±19 sodium nitroprusside,20 S-nitro-
sothiols,21±24 sydnonimines,25 and furoxanes.26
In the past decade, nitric oxide (NO) has been dis-
covered to be a major signaling agent of profound
importance throughout the animal kingdom.1,2 The
major biological functions of NO include controlling
blood pressure, smoothing muscle tone and platelet
aggregation, assisting the immune system in destroying
tumor cells and intracellular pathogens (parasites, virus
and bacteria) and participating in neuronal synaptic
transmission.3±6 Endogenous NO is generated from
arginine by the catalytic activity of nitric oxide syn-
thase.7,8 Exogenous production from NO releasing
agents serves as a valuable tool in biological research
for studying the functions of NO and o€ers a variety of
pharmaceutical applications.9±12 Thus, the development
of new substances and new approaches to the genera-
tion of NO is a current challenge facing medicinal and
organic chemists.
Here we report a new class of redox-sensitive NO
releasing compounds based on the N-nitroso-N-oxyaryl-
amine structural moiety.27 The N-nitroso-N-oxyalkyl-
amine and N-nitroso-N-oxyarylamine functional groups
occur in a variety of natural and synthetic com-
pounds28±30 (Fig. 2). Alanosine, a potential antiviral
and antitumor drug,31,32 and dopastin, a dopamine b-
hydroxylase inhibitor,33 are natural products containing
the N-nitroso-N-oxyalkylamine moiety. The best known
member of the synthetic N-nitroso-N-oxyarylamine
family is N-nitroso-N-oxybenzenamine ammonium salt
(cupferron, 3, Fig. 3) which is commonly used on a large
Keywords: nitric oxide; NO release; N-nitroso compound; N-nitroso-
N-oxybenzenamine ammonium salt.
*Corresponding author. Tel.: +1-313-993-6759; fax: +313-577-5831;
e-mail: pwang@chem.wayne.edu
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00300-4

406
A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
zene. Ab initio calculations were carried out to further
study the reaction pathway by providing structural and
reaction energy information for the protonated and
deprotonated N-nitroso-N-oxybenzenamines and their
radicals.
Results and Discussion
Synthesis of ammonium N-nitroso-N-oxybenzenamines
Many methods have been utilized for preparing N-nitroso-
N-oxybenzenamines.38±40 In this study, compounds 1±2
and 4±7 (Fig. 3) were synthesized by nitrosation of the
corresponding N-hydroxyamines, which were readily
obtained by reduction of the corresponding nitro-
benzenes (Scheme 2).40 Products 1±7 are solid and rea-
sonably stable at room temperature. The X-ray crystal
structure of ammonium N-nitroso-N-oxybenzenamine
(3) was determined and discussed later in comparison
with the computational results.
Figure 1. Commonly used NO donor compounds.
Electrochemical generation of NO
Oxidation of N-nitroso-N-oxybenzenamine ammonium
salts to generate NO can be accomplished enzymati-
cally,41 electrochemically,27,42 chemically (e.g. with lead
tetraacetate),43 and thermally and photochemically.44
The common intermediate for such oxidations is an N-
nitrosobenzenamine-N-oxy radical. Balaban and co-
workers have prepared some of these N-oxy radicals
and performed detailed ESR analyses.45 In aqueous
solution, the radical quickly decomposes to generate a
nitric oxide radical and nitrosobenzene. Lawless and co-
workers ®rst demonstrated the electrochemical oxida-
tion of cupferron, 3, while pursuing a method of in situ
nitrosobenzene generation.42
Figure 2. N-Nitroso-N-oxyalkylamines and N-nitroso-N-oxyaryl-
amines.
We have previously reported the preliminary studies on
electrochemical oxidation of para-substituted N-nitroso-
N-oxybenzenamine ammonium salts, 1, 2, 3, 5, 6 and
7.27 Cyclic voltammetry measurements generated three
oxidation peaks (Fig. 4). The ®rst peak corresponds
(0.44±0.60 V) to the one-electron oxidation of the N-ni-
troso-N-oxybenzenamine to the unstable N-nitroso-
benzenamine-N-oxy radical intermediate. We found
that the substituent on the phenyl ring has a profound
e€ect on the value of this ®rst peak potential. Electron-
donating groups lower the oxidation potential, while
electron-withdrawing groups have the opposite e€ect. In
fact, a linear correlation exists between the Hammett
substituent constants and the ®rst oxidation potential
values (Fig. 5). The second and third oxidation peaks in
the electrochemical oxidation of the compounds occur-
red at 0.70‹0.04 and 0.82‹0.05 V, respectively, which
are consistent with the oxidation of NO to nitrite and
the oxidation of nitrite to nitrate, respectively.42
Figure 3. Cupferron, 3, and the synthetic para-substituted ammonium
N-nitroso-N-oxybenzenamines.
scale in industry as a metal chelator34 and as a poly-
merization inhibitor.35,36 The mechanism for the latter
e€ect is related to the thermal dissociation of cupferron
to generate NO.37 At room temperature, both in the
solid state and in solution, cupferron is relatively stable,
generating very little NO. Recently, Balaban and co-
workers found that ortho-substituted analogues of
cupferron have faster decomposition rates due to the
substituent preventing the N-nitroso-N-oxyarylamine
from becoming planar.38 They showed that many ortho-
substituted N-nitroso-N-oxybenzenamines were good NO
donors for both in vitro and in vivo assays. We describe
here the generation of NO from the more stable para-
substituted N-nitroso-N-oxybenzenamines (Scheme 1).
These compounds undergo a one electron oxidation
which results in the generation of NO and nitrosobene-
To investigate NO release rates and conditions from
these compounds, bulk electrolysis46±48 was used to
generate nitric oxide in solution from the ammonium
salts at various controlled potentials while simulta-
neously measuring the concentration of NO generated

A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
407
Scheme 1. electrochemical or enzymatical generation of NO from para-substituted N-nitroso-N-oxybenzenamine ammonium salts.
Scheme 2. Synthesis of para-substituted N-nitroso-N-oxybenzenamines.
with a nitric oxide detection apparatus.27,49 Within the
range of 0.40±0.70 V, the amount of NO generated from
compounds 1±7 was in¯uenced by the substituent on the
phenyl ring (Table 1). For example, at an applied
potential of 0.65 V, the amount of NO generated from
the -OMe substituted compound was approximately 20
times the amount generated from the -SO2Me com-
pound. The concentration of NO generated increased as
the applied electrochemical potential increased from
around 0.40 to 0.70 V. Above 0.70 V, the concentration
of NO decreased as the electrolysis potential increased
due to the oxidation of NO. This result was consistent
with the cyclic voltammetric data which indicated that a
one electron oxidation of NO occurred around 0.70 V
to generate nitrite in aqueous solution. To prove this
under the conditions of our experiment, gaseous NO
was bubbled into an electrolytic solution and the system
was then sealed. Once a stable NO concentration was
obtained and monitored by the nitric oxide detector,
bulk electrolysis was performed while the amount of
NO was continually measured, as previously described.
As expected, the amount of NO in solution began to
decrease only when a potential of 0.70 V or higher was
applied.
other compounds with the hydroxynitrosamine moiety.41
This peroxidase a€ects a one electron oxidation of the
substrate and is regenerated by transferring two elec-
trons to a molecule of H2O2. This enzyme is a model for
in vivo systems which can oxidize these types of com-
pounds to generate NO. Keefer et al. demonstrated the
ability of cupferron to reduce the blood pressure of
rats.50 Although they did not include stability data, they
concluded from Alston's enzymatic oxidation model
that NO must be generated from cupferron due to
enzymatic processes in their in vivo studies.
The enzymatic oxidation rates of these compounds were
also investigated. Km, kcat and kcat/Km were determined
for compounds 2±4 and 6 by following the formation of
the corresponding nitrosobenzene product (Table 2).
These compounds were chosen as representatives of the
set. Although NO generated by this system was con-
®rmed using the NO-detector, more consistent results
were obtained by monitoring nitrosobenzene produc-
tion by UV±vis techniques. As was evident in the elec-
Enzymatic generation of NO
Horseradish peroxidase has been shown by Alston et al.
to be an e€ective oxidizing enzyme for cupferron and
Figure 4. A typical cyclic voltammetry diagram of electrochemical
oxidation of compound 1. Measurements were conducted in 0.1 M
LiClO4 solution with 1.33±1.36 mM of compound at pH 7 (adjusted
with NH4OH and HCl solutions). The scan rate was 20 mV/s starting
at 0 mV with sweeps in three segments, ®rst oxidatively then reduc-
tively then oxidatively.
Figure 5. Hammett plot for compounds 1±7. Equilibrium K and Ko
were calculated as K=e (nFE/RT). E, the ®rst oxidation potentials, were
obtained by cyclic voltammetry experiments (repeated three times)
conducted in 1.3 mM solutions of compounds 1±7 in 0.1 M LiClO4 at
pH 7 (adjusted with NH4OH and HCl solutions) and at a scan rate of
20 mV/s. R2=0.995.

408
A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
Table 1. Maximum NO concentration generated and applied elec-
trochemical potential for compounds 1, 3, 5 and 7a
C6H5[N2O2]H are shown in Figure 7(a) and (b). Both
structures are planar about N1 and have signi®cant N1±
N2 double bond character. This is similar to the results
of Hall and co-workers where, of the various smaller
analogues studied, structure A was planar about N1
only for ON2O2H, indicating that A0 is an important
resonance contributor when an electron donor is avail-
able to stabilize the positive charge on the central
nitrogen.51 Although N1±N2 is longer for tautomer A
(1.314 A) than for B (1.286 A), in each case it is longer
than a normal N±N double bond (ca. 1.22±1.24 A)53 but
shorter than a single bond (1.4±1.45 A).
-OCH3
(1)
-H
(3)
-Cl
(5)
-SO2CH3
(7)
Max. concentration (M) 4.6‹0.5 4.0‹0.4 2.4‹0.2 0.6‹0.1
Potential (V) 0.65 0.70 0.81 0.75
aEach experiment was prepared with 1.3 mM of compound in 0.1 M
LiClO4 solution, 12 mL total volume. Electrolysis was conducted at
each potential for 5 s intervals. The nitric oxide detector was con-
tinuously recording.
Table 2. Rate constant values for the enzymatic oxidation using
horseradish peroxidase
Tautomer B is preferred by 1.5 kcal/mol (Table 3) and,
taking this as the zero of energy, is separated from tau-
tomer A by a transition state at 6.4 kcal/mol. The com-
puted gas phase geometry of B agrees well with the solid
crystal structure,52 the largest error being 1.2% for the
N1±O3 bond length. para-Substitution with X=OMe,
F, CF3 and SO2Me has little e€ect, in terms of both
energetics and geometries. For these substituents, tau-
tomer B is favored by 0.7 to 2.3 kcal/mol, with bond
lengths and angles essentially identical to those of
X=H. Calculations on the HN2O2H system give similar
results, with B favored by 2.2 kcal/mol at the B3LYP/6-
31+G* level of theory. This preference is maintained at
higher levels of theory. Although the MP2 di€erence of
6.7 kcal/mol is somewhat larger, the B3LYP result is in
good agreement with both the QCISD result of 1.2 kcal/
mol and the modi®ed G2 result of 3.9 kcal/mol, thus
indicating the reliability of the B3LYP results for
XC6H4N2O2H.
-CH3
(2)
-H
(3)
-F
(4)
-CF3
(6)
1
kcat (s
Km (mM)
)
177‹9
33.3‹1.8 65.2‹4.3 78.4‹5.1
93.3‹5.6 45.5‹2.5
5.7‹0.4
18.4‹0.7
5.3‹0.4 1.4‹0.2 0.58‹0.07 0.31‹0.04
1
1
kcat/Km (mM
s
)
trochemical assay, the rate of the reaction increased with
the electron donating e€ect of the substituent. In the
case of 6, although the Km indicated it bound to the
enzyme better than the other three, the rate constant
was in line with the substituent e€ect. The kcat/Km
1
values ranged from 0.3 to 5.3 mM
s
1, from the most
electron withdrawing to the most electron donating
substituent. These values indicate that this series of
compounds are reasonably good substrates for horse-
radish peroxidase.
The cupferron anion, which can be formed by proton
loss from either tautomer A or B, is shown in Figure 7(c).
This structure is a hybrid of the protonated geometries,
but is more like A as evidenced by the similar N2±O4
and N1±C5 bond lengths. The computed gas phase
Theoretical calculations
The generation of nitric oxide from para-substituted N-
nitroso-N-oxybenzenamine compounds was examined
via density functional theory calculations on the fol-
lowing structures: XC6H4[N2O2]H (two tautomers); the
corresponding anion, XC6H4[N2O2] ; radical, XC6H4
[N2O2]ꢀ ; and dissociated products, XC6H4NO+ꢀ NO,
for X=OMe, H, F, CF3 and SO2Me. For X=H, the
transition state (TS) separating the two protonated tau-
tomers was also computed.
There has been some concern over the actual structure
of RN2O2H containing compounds, for which there are
two possible tautomers, A and B (Fig. 6). In a theoreti-
cal study, Hall and co-workers found that tautomer B
was preferred for a variety of small molecules contain-
ing the [N2O2]H moiety, including H2N[N2O2]H,
Me[N2O2]H and H[N2O2]H, while for CN[N2O2]H a
slight preference for A was found.51 Hickmann et al.52
have determined that the crystal structure of C6H5-
[N2O2]H corresponds to tautomer B. Optimized
B3LYP/6-31+G* geometries for both tautomers of
Figure 7. B3LYP/6-31+G* optimized structures for X=H. aNumbers
in parentheses for tautomer B correspond to the crystal structure of
Hickmann et al.52 bNumbers in parentheses for anion with counter ion
correspond to the cupferron crystal structure. Bond lengths are in
angstroms, angles are in degrees.
Figure 6. Tautomers of RN2O2H.

A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
Table 3. Relative energetics for XC6H4N2O2H and HN2O2Ha
409
XC6H4N2O2H
Tautomerization energy
A±B
Proton anity
anion±B
Oxidation energy
Radical±anion
Dissociation energy
RNO+NO±radical
X=OCH3
H
F
CF3
2.34
1.51
1.83
2.14
0.70
344.67
342.07
339.23
333.44
329.18
51.44
54.45
57.31
64.44
67.44
6.30
7.27
6.47
7.63
7.79
SO2CH3
HN2O2H
B3LYP
2.20
6.75
1.17
3.89
346.45
346.58
357.56
353.99
51.86
14.14
MP2b
QCISDc
Mod. G2d
40.19
54.31
0.72
3.89
aIn kcal/mol at the B3LYP/6-31+G* level of theory unless otherwise noted. See Figure 7 for selected structures.
bMP2/6-31+G* optimized. Spin contamination problems were encountered for the radicals.
cQCISD/6-311+G** optimized.
dG2 single point corrections on the B3LYP/6-31+G* optimized geometry.
.
anion geometry shows signi®cant errors when compared
to the crystal structure of cupferron, with the N1±N2
bond 4% longer and the N2±O4 bond 2% shorter.
Including Na+ as a counter ion in the calculation (Fig.
7(d)) reduces the N1±N2 error to 1.5% while the other
bonds are in error by less than 1%. The geometries of
the X=OMe and F anions are almost identical to the
X=H geometry, while for X=CF3 and SO2Me, N1±N2
is 0.014±0.019 A longer, N2±O4 is 0.008±0.011 A
shorter, N1±O3 is 0.005±0.007 A shorter, and N1±C5 is
0.012±0.015 A shorter. The proton anities of these
substituted N-nitroso-N-oxybenzenamine anions range
from 345 to 329 kcal/mol, decreasing as the electron
withdrawing ability of the substituent increases (Table
3). These data show a linear correlation (R2=0.974)
when plotted against the Hammett substituent constants.
correlation with the substituent constants. For HN2O2 ,
the dissociation energy is larger, ca. 14 kcal/mol at the
B3LYP/6-31+G* level of theory. (Although we were
unable to ®nd a transition state for this dissociation,
calculations at the same level of theory suggest that if a
TS exists, it is probably less than 2 kcal/mol higher in
energy than the products.) Higher level calculations on
this system indicate that the energy di€erence may dis-
appear, or is much smaller (Table 3). At the QCISD/6-
311+G** level of theory the dissociation energy is
slightly negative. However, this method is susceptible to
spin contamination problems that typically are not
encountered in the G2 method, which gives a dissocia-
tion energy of 4 kcal/mol. Thus it is shown that oxida-
tion of the anion is the key step in the release of nitric
oxide.
For each anion, oxidation to the corresponding radical
(Fig. 7(e)) causes N1±N2 to lengthen by ca. 0.15 A and
N2±O4 to shorten by ca. 0.06 A. The energy required
for oxidation correlates well with both the substituent
constants (Fig. 8) and the experimental oxidation
potentials (Fig. 9). The radicals then easily dissociate
to form RNO+NO with the products 6±8 kcal/mol
higher in energy at the B3LYP/6-31+G* level of theory.
The dissociation energies do not appear to have any
Conclusions
para-Substituted N-nitroso-N-oxybenzenamine ammo-
nium salts constitute a library of redox-sensitive NO
donor compounds, which generate NO electro-
chemically at well-de®ned electrolysis potentials. More
importantly, these potentials are in the window of nitric
Figure 8. B3LYP/6-31+G* oxidation energy of the anions versus the
Hammet substituent constants. R2=0.988.
Figure 9. Experimental oxidation potential versus the B3LYP/6-
31+G* oxidation energy. R2=0.975.

410
A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
oxide generation without oxidation of the nitric oxide.
It has been shown that the rate of NO release can be
modi®ed by varying the para-substituent under appro-
priate oxidation conditions. para-Substitution has a
similar e€ect on the enzymatic oxidation rate when
assayed versus horseradish peroxidase. Theoretical cal-
culations are in good agreement with the experimental
results as shown by the linear correlation between the
computational oxidation energies of the anions and the
experimental oxidation potentials.
s, OCH3), 6.96 (2H, d, ar), 7.81 (2H, d, ar); 13C NMR
(CD3OD) d 120.44, 113.38, 54.58. UV±vis lmax 283 nm.
Anal. calcd for C7H11N3O3: C, 45.41; H, 5.99, N, 22.70.
Found: C, 45.17, H, 5.78, N, 22.46. It is necessary to
store compound 1 at 78 ꢁC since decomposition begins
to occur overnight at room temperature or in a few
weeks at 20 ꢁC.
p-Amino-N-nitroso-N-oxytoluene ammonium salt (2).
Synthesized from 4-nitrotoluene (44%) as a pale-yellow
solid: 1H NMR (CD3OD, 300 MHz) d 2.37 (3H, s,
CH3), 7.26 (2H, d, ar), 7.82 (2H, d, ar); 13C NMR
(CD3OD) d 130.25, 120.37, 21.02; UV±vis lmax 283 nm.
Anal. calcd for C7H11N3O2: C, 49.69; H, 6.55; N, 24.85.
Found: C, 49.66; H, 6.51; N, 24.79.
Such a redox controlled electrochemical NO generating
system has potential for use in a variety of biomedical
applications. The biological activity and toxicity studies
are currently in progress to further evaluate this new
class of NO-donating compounds.
N-Nitroso-N-oxy-p-¯uorobenzenamine ammonium salt
(4). Synthesized from 1-¯uoro-4-nitrobenzene (40%) as
a white solid: 1H NMR (CD3OD, 400MHz) d 7.52 (2H,
t, ar), 8.18 (2H, t, ar); 13C NMR (CD3OD) d 120.61,
114.95, 114.65, 103.47; UV±vis lmax 283 nm. Anal. calcd
for C6H8FN3O2: C, 41.62; H, 4.66. Found: C, 41.71; H,
4.61.
Experimental
General
1H NMR and 13C NMR spectra were recorded on a
Varian Mercury 300 MHz or Varian Mercury 400 MHz
spectrometer. Silica gel plates (Merck F254) were used
for thin-layer chromatography (TLC). Elemental ana-
lyses were performed at Atlantic Microlab. UV±vis
spectra were obtained with a Milton-Roy, Genesys 2
spectrophotometer. Kinetic data were obtained with the
use of an HP 8453 UV±visible spectrophotometer.
Compound 3 was obtained from Wako Pure Chemical
Industries, Ltd., Osaka, Japan.
N-Nitroso-N-oxy-p-chlorobenzenamine ammonium salt
(5). Synthesized from 1-chloro-4-nitrobenzene (54%) as
a white solid: 1H NMR (CD3OD, 400 MHz) d 7.31 (2H,
d, ar), 7.80 (2H, d, ar); 13C NMR (CD3OD, 300 MHz,)
d 128.23, 119.99; UV±vis lmax 300 nm. Anal. calcd for
C6H8ClN3O2: C, 37.98; H, 4.22; N, 22.16; Cl, 18.73.
Found: C, 38.00; H, 4.26; N, 22.07; Cl, 18.61.
p-Amino-N-nitroso-N-oxy-ꢀ,ꢀ,ꢀ-tri¯uorotoluene ammo-
nium salt (6). Synthesized from 4-nitro-a,a,a-tri¯uoro-
toluene (37%) as a light-yellow solid: 1H NMR
(CD3OD, 300 MHz,) d 7.72 (2H, d, ar), 8.11 (2H, d, ar);
13C NMR (CD3OD, 300 MHz) d 125.50, 118.74; UV±
vis lmax 298 nm. Anal. calcd for C7H8F3N3O2: C, 37.67;
H, 3.61. Found: C, 37.39; H, 3.65.
General procedure for the synthesis of para-substituted
N-nitroso-N-oxybenzenamine ammonium salts
Compounds 1±2 and 4±7 were prepared by nitrosating
the corresponding N-hydroxyamine compounds, which
were obtained from the reduction of the corresponding
nitrobenzenes.40 Zinc dust (1.5 equiv) was added to a
stirring solution (water:methanol, 1:5; v:v) of ammoni-
um chloride (1.5 equiv) and the corresponding nitro-
benzene (1 equiv) at 0 ꢁC. The reactions were monitored
by TLC methods and excess ammonium chloride and
zinc dust were added as needed until the starting mat-
erial was depleted, or the reaction began to yield
byproducts. The solution was then ®ltered and the ®l-
trate was again brought to 0 ꢁC. (For compounds 4 and
6, the solvent was evaporated after ®ltration and the
product was dissolved in ether.) Ammonia gas was
bubbled through the solution and after 15 min ethyl
nitrite (1.2 equiv) was added (isoamyl nitrite for com-
pounds 4 and 6). When the reaction was complete, the
product was allowed to crystallize slowly in solution
over several hours at 20 ꢁC. The product was collected
by vacuum ®ltration and rinsed with methylene chlo-
ride. Additional product was collected from the ®ltrate
by rotary evaporation to dryness (without heating) fol-
lowed by recrystallization in ethyl acetate.
Methyl p-nitrophenylsulphone. Methyl 4-nitrophenyl
sul®de (2.54, 0.015 mol) was dissolved in 60 mL of
acetone. H2O2 (10 mL) was added and the reaction was
heated to 60 ꢁC. After stirring for 30 min, two more
equivalents of H2O2 (10 mL) were added and the reac-
tion was stirred for 2 h. Most of the solvent was
removed through rotary evaporation. The residue was
extracted with CHCl3. The organic phase was dried over
MgSO4, ®ltered and then concentrated. The product
was isolated by column chromatography (silica gel,
hexane:ethyl acetate; 1:1 then 1:2) as a pale-yellow solid
(23%): mp: 102.0±104.5 ꢁC; 1H NMR (CDCl3, 400
MHz) d 3.14 (3H, s, CH3), 8.18 (2H, d, ar), 8.43 (2H, d,
ar); MS (m/e): 201 (M+, 53), 186 (44), 139 (100), 122
(36), 109 (18), 92 (14), 76 (16), 63 (17).
N-Nitroso-N-oxy-p-methylsulfonylbenzenamine ammoni-
um salt (7). Synthesized from methyl p-nitrophenyl-
1
sulphone (42%) as a white, crystalline solid: H NMR
(400 MHz, CD3OD) d 2.05 (s, 3H, CH3), 3.05 (s),
7.91 (d, 2H, ar), 8.09 (d, 2H, ar); UV±vis lmax 321 nm.
Anal. calcd for C7H11N3O4S: C, 36.05; H, 4.72; N, 18.03;
S 13.73. Found: C, 36.09; H, 4.70; N, 17.87; S, 13.64.
N-Nitroso-N-oxy-p-methoxybenzenamine ammonium salt
(1). Synthesized from 4-nitroanisole (3%) as a light-
brown solid: 1H NMR (CD3OD, 300 MHz) d 3.82 (3H,

A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
411
Stability measurement of N-nitroso-N-oxybenzenamines.
A solution of each compound 1±7 was prepared in
sodium phosphate bu€er (50 mM, pH 7.4). The con-
centration of each solution was adjusted in order to
achieve an absorption of 1±2 at the corresponding
maximum wavelength. The stability of the solution was
observed by following the absorption over time.
Theoretical calculations. Geometries were fully opti-
mized in redundant cartesian coordinates and energies
computed at the B3LYP/6-31+G* level of theory54
using the Gaussian9455 series of programs. The B3LYP
method was chosen because it includes electron correla-
tion, yet is less expensive than MP2 and exhibits less
spin contamination for radicals as well. To assess the
accuracy of these results, structures in the HN2O2H
system were optimized at several levels of theory;
B3LYP/6-31+G*, MP2/6-31+G* and QCISD/6-
311+G**.56 Furthermore, approximate QCISD(T)/6-
311+G(3df,2p) energetics were computed for the
B3LYP structures using the single point corrections of
the high accuracy G2 method.57 (The other portions of
this method, zero point and empirical corrections, were
not included.) To verify the nature of the ®nal geomet-
ries, frequency calculations were performed for all of
the B3LYP and MP2 optimized HN2O2H structures.
Due to the size of the larger XC6H4N2O2H systems,
frequencies were computed only for the two protonated
structures and the associated TS of X=H.
Electrochemical measurements. A platinum working
electrode, 1.6 mm in diameter (from Bioanalytical Sys-
tems (BAS), West Lafayette, IN), a platinum wire
counter electrode (BAS), and a Ag/AgCl reference elec-
trode (BAS) were employed in all cyclic voltammetry
measurements. Working and counter platinum gauze
electrodes, 13Â13 mm (160 mm2 surface area) ¯ag-like
in shape, were used in all electrolysis experiments.
(These electrodes were made by annealing platinum
mesh onto platinum wire. The metals were obtained
from Fisher Scienti®c, Pittsburgh, PA.) The supporting
electrolyte, LiClO4, was recrystallized in ethanol, and
dried under vacuum. Doubly distilled water was used as
the solvent. In each case, the solution was argonated
before each experiment and measurements were per-
formed under an argon atmosphere. Cyclic voltammo-
grams were recorded at 20 mV/s and all measurements
were performed at room temperature. An airtight, dual
chamber electrochemical cell was glass blown in
house.46±48 An inlet was made for each of the electrodes
and NO detector probe as well as for the argon tubes.
An airtight seal was assured by using o-rings and inlet
caps. Clamps were used to enhance the ®t of the cell
caps. Electrolysis was conducted using the BAS 100W
program to operate the BAS 100B electrochemical ana-
lyzer.
Enzymatic assay. A stock solution of 1.13 mM horse-
radish peroxidase was prepared in 50 mM sodium
phosphate bu€er (pH 7.4) containing 1 mM EDTA.
Stock solutions for each compound were prepared by
making a saturated sodium phosphate solution of the
compound. A stock solution of H2O2 was prepared,
adjusting the concentration with water to match the
concentration of the respective compound. In kinetic
measurements, the formation of the para-substituted
nitrosobenzene from 3.6 to 68.3 mM of the para-
substituted N-nitroso-N-oxybenzenamines and equal
amounts of H2O2 with 0.1±4.6 mM horseradish perox-
idase was monitored by UV±vis at 365 nm for com-
pounds 2 and 3, 351 nm for compound 4 and 254 nm
for compound 6. Km and kcat values were calculated
from a Lineweaver±Burk plot for each compound.
NO measurements. NO concentration measurements
were conducted using an ISO-NOP 2 mm electrode
probe, ISO-NOP Mark II meter and data collected
using the DUO.18 software from World Precision
Instruments, Inc. (Sarasota, FL).
Acknowledgments
Electrolysis and NO concentration measurements. Each
experiment was run with 1.35 mM of compound in 0.1
M LiClO4 solution. Electrolysis was conducted at each
potential ranging from 0.25 to 1.00 V at 0.05 V incre-
ments. The nitric oxide detector was continuously
recording. Once the baseline current of the NO detector
was constant, a potential was applied to the system for 5
s with constant stirring at 1100 rpm. Once the con-
centration peak of the NO detector stabilized again to a
constant value, the baseline current was subtracted from
the NO peak (or plateau) value in order to determine
the amount of NO generated. Key values are listed in
Table 2. Five seconds was enough time to minimize
experimental error and achieve reproducible results. A
longer time period would only aid in exhausting the
starting material in solution, thus adding equilibrium
shift errors into the measurement. Also, the computer
program driven meter only records up to 4.8 mM of
nitric oxide before it gets saturated. The total amount of
NO generated in each experiment was well below the
solubility limit of NO in solution.
This work was supported by the NIH (GM54074) and
American Heart Association, Florida Aliation.
Andrea McGill thanks the National Science Founda-
tion for support through her predoctoral fellowship.
Thanks are also due to Dr. Petersen for the use of
his electrochemical equipment, to David Daenzer for
constructing the electrochemical cell, and to Mary
Jane Heeg for elucidation of the crystal structure of
cupferron.
References and Notes
1. Kerwin, Jr., J. F.; Lancaster, Jr., J. R.; Feldman, P. L. J.
Med. Chem. 1995, 38, 4343.
2. Williams, R. J. P. Chem. Soc. Rev., 1996, 77.
3. Moncada, S.; Palmer, R. M. J.; Higgs, E. A. Pharmacol.
Rev. 1991, 43, 109.
4. Bredt, D. S.; Snyder, S. H. Annu. Rev. Biochem. 1994, 63,
175.

412
A. D. McGill et al. / Bioorg. Med. Chem. 8 (2000) 405±412
5. Schmidt, H. H. W.; Walter, U. Cell 1994, 78, 919.
6. Feldman, P. L.; Grith, O. W.; Stuehr, D. J. Chem. & Eng.
News 1993, 71 (20 December), 26.
7. Nathan, C.; Xie, Q.-W. Cell 1994, 78, 915.
8. Marletta, M. A. Cell 1994, 78, 927.
9. Hanson, S. R.; Hutsell, T. C.; Keefer, L. K.; Mooradian, D.
L.; Smith, D. J. In Nitric Oxide: Biochemistry, Molecular
Biology and Therapeutic Implications; Ignarro, L.; Murad, F.,
Eds.; Academic Press: San Diego, 1995; p 383.
10. Guo, Z.-M.; McGill, A.; Yu, L.-B.; Li, J.; Ramirez, J.;
Wang, P. G. Bioorg. Med. Chem. Lett. 1996, 6, 573.
11. Yu, L.-B.; McGill, A.; Ramirez, J.; Wang, P. G.; Zhang, Z.
Y. Bioorg. Med. Chem. Lett. 1995, 5, 1003.
34. Paneli, M.; Ouguenoune, H.; David, F.; Bolyos, A. Anal.
Chim. Acta. 1995, 304, 177.
35. Scates, M. O.; Dougherty, E. F. European Patent, 0 301
879 A2, 1989.
36. Ulbricht, J.; Hoering, S. Plaste Kaut. 1968, 15, 396.
37. Kende, I.; Sumegi, L.; Tudos, F. Polymer Bulletin. 1980, 3,
325.
38. Gar®eld, R. E.; Balaban, A. T.; Seitz, W. A.; Klein, D. J.;
Lesko, M. International Patent, WO 96/36326, 1996.
39. George, M. V.; Kierstead, R. W.; Wright, G. F. Can. J.
Chem. 1959, 37, 679.
40. Balaban, A. T.; Gar®eld, R. E.; Lesko, M. J.; Seitz, W. A.
Org. Prep. Proced. Int. 1998, 30, 439.
12. Ramirez, J.; Yu, L.-B.; Li, J.; Braunschweiger, P. G.;
Wang, P. G. Bioorg. Med. Chem. Lett. 1996, 6, 2575.
13. Bauer, J. A.; Booth, B. P.; Fung, H.-L. In Nitric Oxide:
Biochemistry, Molecular Biology and Therapeutic Implications;
Ignarro, L.; Murad, F., Eds.; Academic Press: San Diego,
1995; p 361.
41. Alston, T. A.; Porter, D. J. T.; Bright, H. J. J. Biol. Chem.
1985, 260, 4069.
42. Lawless, J. G.; Hawley, M. D. Electrochemical Oxidation
of Cupferron Anal. Chem. 1968, 40, 948.
43. Balaban, A. T.; Negoita, N.; Pascaru, I. Rev. Roum. Chim.
1971, 16, 721.
14. Morley, D.; Keefer, L. K. J. Cardiovasc. Pharmacol. 1993,
22 (Suppl. 7), S3.
44. Hwu, J. R.; Yau, C. S.; Tsay, S.-C.; Ho, T.-I. Tetrahedron
Lett. 1997, 38, 9001.
15. Saavedra, J. E.; Dunams, T. M.; Flippen-Anderson, J. L.;
Keefer, L. K. J. Org. Chem. 1992, 57, 6134.
45. Balaban, A. T.; Negoita, N.; Baican, R. J. Mag. Res. 1973,
9, 1.
16. Hrabie, J. A.; Klose, J. R.; Wink, D. A.; Keefer, L. K. J.
Org. Chem. 1993, 58, 1472.
46. Fry, A. Synthetic Organic Electrochemistry; 2nd ed.; Wiley:
New York, 1989.
17. Smith, D. J.; Chakravarthy, D.; Pulfer, S.; Simmons, M.
L.; Hrabie, J. A.; Citro, M. L.; Saavedra, J. E.; Davies, K. M.;
Hutsell, T. C.; Mooradian, D. L.; Hanson, S. R.; Keefer, L. K.
J. Med. Chem. 1996, 39, 1148.
18. Saavedra, J. E.; Southan, G. L.; Davies, K. M.; Lundell,
A.; Markou, C.; Hanson, S. R.; Adrie, C.; Hurford, W. E.;
Zapol, W. M.; Keefer, L. K. J. Med. Chem. 1996, 39, 4361.
19. Keefer, L. K. Chemtech. 1998, 8, 30.
47. Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications. Wiley: New York, 1980.
48. Kaifer, A. E. In Comprehensive Supramolecular Chemistry;
Davies, J. E. D., Ed.; Elsevier Science: London, 1996; Vol. 8,
p 499.
49. The electrochemical cell was designed with an airtight seal
modeled from: Rasmussen, S. C.; Richter, M. M.; Yi, E.;
Place, H.; Brewer, K. J. Inorg. Chem. 1990, 29, 3926.
50. Keefer, L.; Wink, D. A.; Dunams, T. M.; Hrabie, J. A. US
Patent, 5,212,204, 1993.
20. Richter-Addo, G. B.; Legzdins, P. Metal Nitrosyls; Oxford
University Press: New York, 1992.
21. Stamler, J. S.; Singel, D. J.; Loscalzo, J. Science 1992, 258,
1898.
22. Stamler, J. S. Cell 1994, 78, 931±936.
23. Stamler, J. S. Current Topics in Microbiology and Immu-
nology; Koprowski, H.; Maeda, H., Eds.; Springer-Verlag:
New York, 1995; p 19.
24. Williams, D. L. H. Chem. Soc. Rev. 1985, 14, 171.
25. Feelisch, M.; Ostrowski, J.; Noack, E. J. Cardiovasc.
Pharmacol. 1989, 14, S13.
26. Medana, C.; Ermondi, G.; Fruttero, R.; Di Stilo, A.; Fer-
retti, C.; Gasco, A. J. Med. Chem. 1994, 37, 4412.
27. McGill, A. D.; Yang, Y.; Wang, J.; Echegoyen, L.; Wang,
P. G. In Methods in Enzymology; Packer, L., Ed.; Academic
Press: CA, 1998; Vol. 301, Chapter 25, p. 235.
51. Taylor, D. K.; Bytheway, I.; Barton, D. J. R.; Bayse, C.
A.; Hall, M. B. J. Org. Chem. 1995, 60, 435.
52. Hickmann, E., Hadicke, E. and Reuther, W. Tetrahedron
Let. 2457, 1979.
53. Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.;
Orpen, A. G.; Taylor, R. J. Chem. Soc. Perkin Trans. 2 1987,
S1.
54. Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
55. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M.
W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.
A.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.;
Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman,
J. B.; Peng, C. Y.; Ayala, P. Y.; Wong, M. W.; Andres, J. L.;
Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.;
Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94 (Revision
C.3); Gaussian: Pittsburgh, PA, 1995.
28. Kano, K.; Anselme, J.-P. Tetrahedron. 1993, 49, 9453.
29. Kano, K.; Anselme, J.-P. J. Org. Chem. 1993, 58, 1564.
30. Kano, K.; Anselme, J.-P. Tetrahedron. 1992, 48, 10075.
31. Murthy, Y. K. S.; Thiemann, J. E.; Coronelli, C.; Sensi, P.
Nature 1966, 211, 1198.
56. Pople, J. A.; Head-Gordon, M.; Raghavachari, K. J.
Chem. Phys. 1987, 87, 5968.
32. Fumarola, D. Pharmacology 1970, 3, 215.
33. Iinuma, H.; Takeuchi, T.; Kondo, S.; Matsuzaki, H.;
Umezawa, H.; Ohno, M. J. Antibiot. 1972, 25, 497.
57. Curtiss, L. A.; Raghavachari, K.; Trucks, G. W.; Pople, J.
A. J. Chem. Phys. 1991, 94, 7221.