Synthesis of tris-hydroxymethyl-based nitrone derivatives with highly reactive nitronyl carbon

Article abstract of DOI:10.1021/jo202098x

A novel series of α-phenyl-N-tert-butyl nitrone derivatives, bearing a hydrophobic chain on the aromatic ring and three hydroxyl functions on the tert-butyl group, was synthesized through a short and convenient synthetic route based on a one-pot reduction/condensation of tris(hydroxymethyl)nitromethane with a benzaldehyde derivative. Because of the presence of hydroxyl functions on the tert-butyl group, an intramolecular Forrester-Hepburn reaction leading to the formation of an oxazolidine-N-oxyl compound was observed by electron paramagnetic resonance (EPR). The mechanism of cyclization was further studied by computational methods showing that intramolecular hydrogen bonding and high positive charge on the nitronyl carbon could facilitate the nucleophilic addition of a hydroxyl group onto the nitronyl carbon. At high nitrone concentrations, a second paramagnetic species, very likely formed by intermolecular nucleophilic addition of two nitrone molecules, was also observed but to a lesser extent. In addition, theoretical data confirmed that the intramolecular reaction is much more favored than the intermolecular one. These nitrones were also found to efficiently trap carbon-centered radicals, but complex spectra were observed due to the presence of oxazolidine-N-oxyl derivatives.

Full text of DOI:10.1021/jo202098x

Article
pubs.acs.org/joc
Synthesis of Tris-hydroxymethyl-Based Nitrone Derivatives with
Highly Reactive Nitronyl Carbon
Fanny Choteau,†,‡ Beatrice Tuccio,§ Frederick A. Villamena,∥,⊥ Laurence Charles,§ Bernard Pucci,†,‡
́
and Gregory Durand*,†,‡
́
†Universite
́
d’Avignon et des Pays de Vaucluse, Equipe Chimie Bioorganique et System
Avignon, France
‡Institut des Biomolec
Montpellier Cedex 05, France
§Aix-Marseille Universite-
CNRS, UMR 7273: Institut de Chimie Radicalaire, Equipe SACS, F-13397 Marseille, France
̀
es Amphiphiles, 33 rue Louis Pasteur, F-84000
́
́
ules Max Mousseron, UMR 5247, CNRS-Universites Montpellier I & II, 15 avenue Charles Flahault, F-34093
́
∥Department of Pharmacology and ⊥The Davis Heart and Lung Research Institute, College of Medicine, The Ohio State University,
Columbus, Ohio 43210, United States
S
* Supporting Information
ABSTRACT: A novel series of α-phenyl-N-tert-butyl nitrone derivatives, bearing a hydrophobic chain on the aromatic ring and
three hydroxyl functions on the tert-butyl group, was synthesized through a short and convenient synthetic route based on a one-
pot reduction/condensation of tris(hydroxymethyl)nitromethane with a benzaldehyde derivative. Because of the presence of
hydroxyl functions on the tert-butyl group, an intramolecular Forrester−Hepburn reaction leading to the formation of an
oxazolidine-N-oxyl compound was observed by electron paramagnetic resonance (EPR). The mechanism of cyclization was
further studied by computational methods showing that intramolecular hydrogen bonding and high positive charge on the
nitronyl carbon could facilitate the nucleophilic addition of a hydroxyl group onto the nitronyl carbon. At high nitrone
concentrations, a second paramagnetic species, very likely formed by intermolecular nucleophilic addition of two nitrone
molecules, was also observed but to a lesser extent. In addition, theoretical data confirmed that the intramolecular reaction is
much more favored than the intermolecular one. These nitrones were also found to efficiently trap carbon-centered radicals, but
complex spectra were observed due to the presence of oxazolidine-N-oxyl derivatives.
EPR spectroscopy. Analyzing the EPR spectrum obtained gives
information on the addend structure.
INTRODUCTION
■
Oxidative stress was defined as a disturbance in the prooxidant−
antioxidant balance in favor of the former, leading to potential
damage, and it has become clear that an overproduction of
prooxidant molecules, also termed reactive oxygen and nitrogen
species (ROS/RNS), is associated with several pathologies such
as neurodegenerative diseases, cancers, and ischemia−reperfu-
sion syndromes to name a few.1 Consequently, any agent that
can trap ROS/RNS to form more stable adducts, and as a result
can alter the course of diseases progression, may be useful as a
novel therapeutic approach.
Of particular interest are nitrone spin-traps, which undergo
addition reaction with free radicals, making them a popular
analytical reagent for the identification of short-lived radicals
using electron paramagnetic resonance (EPR) spectroscopy.2,3
The spin-trapping technique relies on the addition of a
transient radical to a diamagnetic spin-trap (usually a nitrone
or a nitroso compound) to yield a longer lived paramagnetic
spin adduct (a nitroxide) that can be detected by conventional
The linear nitrone, α-phenyl-N-tert-butylnitrone (PBN), and
its derivatives have also been widely used as antioxidant agents
in several biological models including protection against
ischemia-reperfusion injuries, stroke, hearing loss, as well as
increasing life span of mice and rats.4,5 Further studies on
stroke demonstrated the superiority of the disulfonate PBN
derivative (disodium 4-[(tert-butylimino)methyl]benzene-1,3-
disulfonate N-oxide), also referred to as NXY-059, compared to
PBN.6 Moreover, experimental evidence suggest that the
pharmacological activity of linear nitrones involves the
inhibition of signal transduction and gene induction processes
that lead to apoptosis and therefore is not solely due to their
radical trapping capability.4,7
With the expectation that amphiphilic compounds possessing
a hydrophilic polar head and a lipophilic alkyl chain would
Received: October 13, 2011
Published: December 21, 2011
© 2011 American Chemical Society
938
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
exhibit improved bioavailability and membrane permeability,
our work has been devoted to the design of nitrone derivatives
with an amphiphilic character.8−11 Their biological evaluations
have shown that the amphiphilicity is a key feature in
determining bioactivity and protection against the oxidative
toxicity in vitro and in vivo with the LPBNH15 exhibiting the
highest antioxidant activity (Figure 1).12 However, one of its
LiAlH4 was previously used for the reduction of the cyano
group, but we found out in this work that reduction by catalytic
hydrogenation could lead to similar yields in a much simpler
procedure, and therefore, this latter method was preferred. In
parallel, the four carboxylic acids, namely butanoic, hexanoic,
octanoic, and decanoic acid, were activated in the presence of
N-hydroxysuccinimide (HO-Su), N,N-dicyclohexylcarbodii-
mide (DCC), and a catalytic amount of 1-hydroxybenzotriazole
(HOBt) in dry CH2Cl2. The resulting crude mixtures were
directly added to benzylamine 2 in the presence of DIEA in dry
CH2Cl2 to give compounds 3−6 in good yields. Then,
dioxolane group removal was achieved in a 7:3 v/v AcOH/
H2O mixture to give the benzaldehyde derivatives 7−10 in
quantitative yields.
Synthesis of T-PBN. Several general methods have been
described for the preparation of nitrone compounds.15
Migrdichian demonstrated16 the first example of alkylation of
oxime, and despite its lack of regioselectivity, this method has
been used successfully for the synthesis of substituted
nitrones.17 Among the other methods, one can cite the
oxidation of imines or secondary aliphatic amines,18,19 the
condensation between a carbonyl group and a hydroxylamine,
and the in situ reduction of nitro compounds in the presence of
an aldehyde or a ketone.20
Figure 1. Chemical structures of LPBNH15 and the new TCn-PBN
series.
limitations as a potential therapeutic agent may come from its
relatively long synthesis that requires 12 steps. Consequently,
the present report deals with the design of a novel series of
amphiphilic PBN derivatives whose synthesis has been
significantly shortened allowing gram-scale preparation of
pure nitrones (Figure 1). These newly designed nitrones are
comprised of a hydrophobic chain grafted onto the aromatic
ring and three hydroxyl functions grafted onto the tert-butyl
group of the PBN. X-band EPR experiments were performed
and demonstrated a unique reactivity for this series of tris-
hydroxymethyl-based spin-traps due to the presence of
hydroxyl functions on the tert-butyl group. Finally, these
findings were rationalized with the use of computational
methods.
In an attempt at optimizing the formation of the nitronyl
group, we carried out preliminary experiments with the
commercially available benzaldehyde (Scheme 2). The syn-
thesis of T-PBN by reduction of tris(hydroxymethyl)-
nitromethane (also called 2-hydroxymethyl-2-nitro-1,3-pro-
panediol) to the hydroxylamine derivative followed by
condensation with benzaldehyde had previously been reported
by Janzen and Zawalski; however, the overall yield was only
22%.21 Therefore, the oxidation of an imine was the first
method we tested following the procedure by Bernotas et al.18
The imino compound 11 was obtained by reaction of
benzaldehyde with tris(hydroxymethyl)aminomethane in re-
fluxing toluene in a Dean−Stark apparatus. Because of its
instability, the imino compound was not purified, and the crude
compound was directly oxidized by hydrogen peroxide 30% in
the presence of sodium tungstate catalyst according to the
procedure described by Murahashi et al.19 Under these
conditions, the degradation of the imino compound was the
predominant reaction, and no nitrone formation was detected.
This is in agreement with the findings of Bernotas et al., who
found that direct oxidation of benzazepines to nitrones under
similar reaction conditions was unsuccessful.22 We overcame
this problem by using a two-step procedure consisting of first
reducing the imine 11 before oxidizing the resulting amino
compound into its nitronyl form. Reduction using NaBH4 in
methanol or in ethanol led to compound 12 after purification
by chromatography on silica gel but in low yield (32% and 26%,
respectively, in two steps from benzaldehyde). On the contrary,
reduction by catalytic hydrogenation in the presence of Pd/C
was found to be more efficient, yielding compound 12 in good
yield. The last step involved oxidation of the amino group and
this was achieved using the procedure described by Murahashi
et al.19 To optimize the yield of nitrone formation, several
conditions were used and data are presented in Scheme 2. The
yields of oxidation remained below 40%, but the best
conditions appeared to be 2 equiv of H2O2 at 0 °C, leading
to nitrone 13 (also called T-PBN) in 36% yield. Therefore, the
overall yield of T-PBN formation using that synthetic strategy
was 24%, similar to that of Janzen and Zawalski (22%).21
RESULTS AND DISCUSSION
■
Synthesis. Synthesis of Hydrophobic Benzaldehyde
Derivatives. 4-Cyanobenzaldehyde was used as starting ma-
terial for the synthesis of the hydrophobic benzaldehyde
derivatives following our previous procedure (Scheme 1).13,14
Scheme 1. Synthesis of the Hydrophobic Benzaldehyde
Derivatives 7−10
The carbonyl group was first protected as a dioxolane, and then
reduction of the cyano group by catalytic hydrogenation (10%
Pd/C) in EtOH was carried out affording the benzylamine 2 in
77% yield in two steps (Scheme 1). It has to be noted that
939
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
Scheme 2. Synthesis of T-PBN 13
Scheme 3. Synthesis of TCn-PBN Compounds 14−17
The mildest and the most selective route to prepare nitrones
is the condensation of N-alkylhydroxylamines with carbonyl
compounds.15 Nevertheless, the drawback of this method is
that only a few hydroxylamines are commercially available, and
their preparation is often not straightforward. Moreover, during
the coupling reaction between the aldehyde and the hydroxyl-
amine, the oxidation of this latter can also occur, producing
nitroso side products. To improve the formation of the nitronyl
function, we next focused our attention on a one-pot method
which involves in situ reduction of tris(hydroxymethyl)-
nitromethane in the presence of a benzaldehyde derivative.
As a wide range of nitro compounds are commercially available;
this method has the advantage of being broad and general and
usually gives high yields.20 For instance, Gautheron-Chapou-
laud et al. demonstrated that the in situ zinc mediated
reduction of nitro compounds in the presence of sugar
constitutes a direct and efficient method to synthesize a wide
range of glycosylated nitrones.23 Following this strategy, 2
equiv of tris(hydroxymethyl)nitromethane and benzaldehyde
were solubilized in ethanol in the presence of 6 equiv of acetic
acid. Zinc dust (4 equiv) was added at 0 °C, and then the
reaction was heated at 60 °C for 20 h. After purification by
chromatography and two successive crystallizations from ethyl
acetate/n-hexane, pure T-PBN was obtained in 72% yield,
which is much higher than the procedure used by Janzen and
Zawalsky.21
followed by two successive crystallizations from ethanol, TCn-
PBN compounds 14−17 were obtained in good yields ranging
from 64% to 72%.
Physical−Chemical Properties. Partition Coefficients.
The relative lipophilicities (log k′W) of nitrones were measured
by a HPLC technique, and as shown in Figure 2, log k′W values
○
●
Figure 2. Water solubility ( ) and lipophilicity ( ) of TCn-PBN
compounds (n = 3, 5, 7, and 9) and T-PBN (n = 0).
are increased with the number of carbon of the alkyl chain,
demonstrating that for a given polar head this parameter is
linearly correlated to the length of the tail. Moreover, TC7-PBN
and LPBNH1512 which have the same alkyl chain but a
different polar head exhibited comparable hydrophobic proper-
ties with log k′W values of 2.54 and 2.86, respectively. Such a
result strongly confirms that the volume and the nature of the
polar group have a low impact on the log k′w value, in perfect
agreement with our previous observations.8,24
Synthesis of Tris-hydroxymethyl-Based Nitrones. Having
demonstrated the superiority of the second synthetic strategy,
we next used this latter for the preparation of the TCn-PBN
series. Condensation of tris(hydroxymethyl)nitromethane with
the four aldehyde derivatives 7−10 was done in the presence of
zinc and acetic acid in ethanol (Scheme 3). Owing to the
possible oxidation of hydroxylamine by free radical processes,
the reaction was carried out in the dark and under argon.
Addition of molecular sieves was also very useful in increasing
the kinetics of the reaction. After chromatography on silica gel
940
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
Solubility. The solubility of the nitrones in water was also
determined, and the data are in agreement with the partition
coefficient values, the longer the chain, the lower the solubility.
However, when comparing these values with that of PBN (26.5
g/L),12 despite the presence of a tris(hydroxymethyl)-
aminomethane-based polar head, none of the TCn-PBN nor
the T-PBN exhibited a higher solubility than PBN. This
suggests that the formation of hydrogen bonds between
hydroxyl groups is likely responsible for the low water solubility
of this series. In agreement with this, we have recently
demonstrated that tris(hydroxymethyl)-based acrylamide
monomers with a related structure to that of the TCn-PBN
polar head were able to form hydrogen bondings.25
Spin-Trapping. Prior to any spin-trapping assay, the
supernatant of a saturated aqueous solution of T-PBN obtained
after heating a heterogeneous mixture of 34 mg of T-PBN in 1
mL of water (0.15 mol·L−1, that is ∼2× the solubility limit in
water) was analyzed by EPR leading to the signal shown in
Figure 3a. Simulation of this signal revealed the presence of two
type nitrone (aN = 1.44 mT and aH = 0.18 mT, ∼25%).26
Though the identification of this second species remains
hypothetical, it could likely correspond to adduct 23, obtained
after bimolecular nucleophilic addition of two nitrone
molecules, a mechanism largely favored by the high nitrone
concentrations used (see Scheme 4b). An EPR analysis of a less
concentrated T-PBN solution (0.008 mol·L−1, ∼10× below the
solubility limit in water) showed the presence of the sole
nitroxide 18, confirming that the formation of 23 occurs only at
high nitrone concentrations.
Similar results were obtained from EPR analysis of T-PBN
solutions prepared in dimethylformamide (DMF), a solvent
that favors nucleophilic addition processes.28 Using 0.15
mol·L−1 solution of T-PBN in DMF, a mixture of nitroxide
18 (aN = 1.48 mT and aH = 1.98 mT, ∼80%) and of nitroxide
23 (aN = 1.40 mT and aH = 0.17 mT, ∼20%) was observed,
while only nitroxide 18 was detected when 0.008 mol·L−1
solution was used. It is also worth noting that 18 was the only
species observed when T-PBN was dissolved in methanol (aN =
1.51 mT and aH = 1.98 mT), whatever the nitrone
concentration. Very similar observations were made with
nitrones 14−17 solubilized in methanol and DMF and with
nitrone 14 solubilized in water. The EPR parameters of such
nitroxides 18−27 thus formed are reported in Table 1.
An estimation of the proportion of compound 18 in T-PBN
solutions (0.01−0.05 mol·L−1) was determined by using 3-
carboxyproxyl (3-CP) as an internal reference and by
comparing the area of the signal of 3-CP with that of
compound 18. Whatever the solvent, the concentration of the
oxazolidine-N-oxyl compound 18 was found lower than 0.2%,
i.e., 0.16% in H2O, 0.15% in DMF and 0.10% in MeOH. The
same observations were made with TC5-PBN solutions, and the
concentration of nitroxide 20 was found to be 0.25% in H2O,
0.27% in DMF, and 0.22% in MeOH. Worth noting is that no
increase of the oxazolidine-N-oxyl concentration was observed
over a period of 2 h. This suggests that a rapid equilibrium is
reached during the very first minutes of mixing, and then the
concentration of the cyclic nitroxide remains constant. The
stability of T-PBN (0.075 mol·L−1) in organic solvents was also
Figure 3. EPR spectra of various nitroxides derived from T-PBN: (a)
signal recorded from the supernatant of a saturated aqueous T-PBN
solution; (b) signal recorded after generating •OH using a standard
Fenton system in the presence of 0.008 mol·L−1 T-PBN in phosphate
1
surveyed by TLC and H NMR, and after 24 h of stirring at
•
buffer; (c) signal recorded after generating CH3 using a standard
room temperature, no significant degradation of T-PBN was
observed either in methanol or DMF (data not shown).
Because of its sensitivity and the structural information it can
provide, electrospray mass spectrometry (ESI-MS) was
employed to study the reactivity of T-PBN in water, methanol,
and DMF. Unfortunately, tandem mass spectrometry experi-
ments did not allow us to discriminate the nitrone 13 from the
oxazolidine-N-oxyl compound 18. Indeed, ESI-MS generally
allows the observation of the hydroxylamine form obtained
after reduction of the nitroxide compound rather than the
paramagnetic nitroxide itself.29,30 Therefore, since the cyclic
hydroxylamine derived from nitroxide 18 and the parent T-
PBN are isomers, they would be MS-detected at the same m/z
value. An aqueous solution of T-PBN (0.035 mol·L−1) was
prepared and EPR analyzed before being submitted to ESI-MS
showing the presence of nitroxides 18 and 23. MS analysis of
this solution led to the detection of a main signal at m/z 226.1
(Figure S28, Supporting Information), which could be assigned
either to the protonated cyclic hydroxylamine or to the
protonated nitrone (or both). Moreover, the main m/z 122.1
product ion generated upon dissociation of the m/z 226.1 ion
(Figure S28, Supporting Information) could be accounted for
by considering either the reduced form of 18 or the nitrone 13
Fenton system in phosphate buffer/DMSO (80/20) in the presence of
0.008 mol·L−1 T-PBN; (d) signal recorded after generating •CH3 using
a concentrated Fenton system in phosphate buffer/DMSO (80/20) in
the presence of 0.15 mol·L−1 T-PBN.
nitroxides, both showing a six-line spectrum due to hyperfine
couplings of the unpaired electron with nitrogen and β-
hydrogen nuclei. In the major signal (aN = 1.52 mT and aH =
1.97 mT, ∼75%), the hyperfine coupling constant (hfcc) with
the β-hydrogen was too high for being consistent with a PBN-
type spin adduct,26 but surprisingly, in agreement more with a
five-membered cyclic nitroxide. This signal could be assigned to
the oxazolidine-N-oxyl compound 18, which very likely
originates from intramolecular nucleophilic addition of a
hydroxyl group onto the nitronyl carbon followed by fast
autoxidation of the resulting hydroxylamine, the so-called
Forrester-Hepburn mechanism (see Scheme 4a).27 The
possible formation of such cyclic nitroxides from hydroxy-
PBN compounds was first proposed by Janzen and Zawalski in
the late 1970s,21 but to the best of our knowledge this has
remained unproven. On the basis of its EPR parameters, the
minor species could be assigned to an alkoxy-adduct of a PBN-
941
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
Scheme 4. (a) Mechanisms Proposed for the Cyclization Reaction Leading to the T-PBN and TCn-PBN-Derived Nitroxides 18−
22; (b) Mechanism Proposed for the Bimolecular Reaction Leading to the Formation of Nitroxides 23−27; (c) Spin-Trapping
a
of a Free Radical •Y by the Nitrone 13 Yielding the Spin Adduct 13/•Y
a
Also shown are the free energies of reaction (ΔG298K, in kcal/mol) in water and in DMSO (in parentheses) for compound 13 (R = H) at the PCM/
B3LYP/6-31G+(d,p)// B3LYP/6-31G(d) level of theory.
as the precursor ion. Indeed, this unique dissociation pathway
consists of the release of 2-(hydroxymethyl)prop-1-ene-1,3-diol
(104 Da), whose formation can be envisaged upon activation of
both isomers. The study of the reactivity of nitrone 15 led to
similar conclusions (Figure S29, Supporting Information), with
the detection of a signal at m/z 353.2 in the MS mode and two
successive dissociation reactions (i.e., loss of 104 Da to yield m/
z 249.2, followed by elimination of N-pentylformamide to
produce m/z 134.1). As for the second species that could arise
from bimolecular nucleophilic addition of two nitrone
molecules, its concentration was much too low for MS
detection regardless of the nitrone studied and the nature of
the solvent. It has to be noted that the same results were
obtained after EPR and MS analysis of solutions of nitrones 13
and 15 either in DMF or methanol.
simulation showed another spin adduct present other than
nitroxide 18 whose EPR signal consists of six lines (aN = 1.46
mT and aH = 0.21 mT) which could possibly be assigned to the
13/•OH adduct; however, it could also correspond to a carbon-
centered radical adduct 13/•C, since •OH radicals are known to
react with alcohols by hydrogen abstraction on the α-
carbon.31,32 Therefore, in the presence of 13, OH could
•
generate a carbon-centered radical that could subsequently be
•
•
trapped by the nitrone. In the next step, CH2OH and CH3
were produced by generating •OH in the presence of methanol
and DMSO, respectively. In each case, a minor six line EPR
signal corresponding to either 13/•CH2OH or 13/•CH3 was
detected other than that of nitroxide 18. To illustrate this, the
EPR signal obtained when •CH3 was produced in the presence
of T-PBN is shown in Figure 3c. More complex signals were
observed when the same experiments were performed by
raising the nitrone concentration to 0.15 mol·L−1 and by
generating twice more •OH (Figure 3d). It appears that many
radical reactions could occur between T-PBN and •OH,
yielding a complex mixture of several radical species. With
nitrones 14-17, the spin-trapping experiments were limited to
Despite the tendency of nitrones 13−17 to form oxazolidine-
N-oxyl compounds, their spin-trapping properties were never-
theless surveyed with a series of free radicals. In order to
simplify the notation, the spin adduct obtained after addition of
a radical •Y on a nitrone X will be noted X/•Y, as indicated in
Scheme 4c. We first studied the spin-trapping properties of T-
PBN at 0.008 mol·L−1. EPR analysis of the medium always
revealed the presence of the cyclic nitroxide 18, regardless of
the radical generating system. Moreover, we found that the
concentration of nitroxide 18 remained constant during the
spin-trapping experiments, which suggests that its rapid
formation is not reversed by the spin-trapping reaction. This
species was the only radical detected when O2•− was produced
by the xanthine/xanthine oxidase system in pH 7.2 phosphate
•
•CH3 and CH2OH radicals owing to solubility issues. As
previously observed, the signals of the various spin adducts
were detected in addition to those of cyclic nitroxides 19−22,
and very complex spectra were systematically observed when
higher amounts of •OH and nitrone were used. All of the EPR
data collected for the various nitroxides detected are listed in
Table 1. It should be noted that very large amounts of DMSO
were used as •OH scavenger, which could result in the
formation of an oxygen-centered radical derived from DMSO
beside •CH3. Consequently, there is uncertainty in the
•
buffer. When OH was generated using a standard Fenton
system, the EPR signal given in Figure 3b was obtained. Its
942
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
oxidation to nitroxide were calculated for T-PBN at the PCM/
B3LYP/6-31G+(d,p)//B3LYP/6-31G(d) level of theory in
water and in DMSO (Scheme 4). In water, results indicate that
intramolecular nucleophilic addition of one alcohol group to
the nitronyl-C to form the oxazolidine−hydroxylamine is
endoergic by 10.4 kcal/mol but its subsequent oxidation to
nitroxide by oxygen to form hydroperoxyl radical is less
endoergic at 5.1 kcal/mol (Scheme 4a). This indicates that the
formation of the oxazolidine-N-oxyl compound 18 is probable
considering that nucleophilic addition of superoxide to DMPO
was previously calculated to be endoergic as well by 11.9 kcal/
mol at the same level of theory in water.33 Intermolecular
nucleophilic addition is more endoergic with ΔG298K of 33.1
kcal/mol indicating that in aqueous solution, the predominant
hydroxylamine would result from intramolecular OH addition
onto the nitrone (Scheme 4b). However, the oxidation of the
resulting alkoxyhydroxylamine is highly favorable similar to the
oxazolidine-N-oxyl formation with ΔG298K = 2.8 kcal/mol. In
DMSO, intra- and intermolecular nucleophilic addition as well
as the oxidation process are only slightly less endoergic by <1
kcal/mol than in water. Therefore, the thermodynamics of
nitroxide formation is in good agreement with the experimental
observations which suggest that the intramolecular mechanism
is more favored than the intermolecular one, the latter being
observed only at high nitrone concentrations either in DMF or
water with ∼20 and ∼25% molar ratio, respectively. The
favorability of intramolecular cyclization could also be driven by
the intramolecular H-bonding that can facilitate the nucleo-
philic addition of the alcohol onto the nitronyl carbon.
As shown in Figure 4, the various H-bond motifs favor two
conformations in which one of the hydroxylic hydrogen H-
bonds to the nitronyl oxygen (Figure 4a and 4b). On the
contrary, the conformation that shows no H-bonding with the
nitronyl-O but three H-bonding between the three hydroxyl
groups is the least favorable (Figure 4c). Also worth noting is
that the two favored conformers gave much higher positive
charge density on the nitronyl-C, the site of nucleophilic
addition, with values in water of 0.044 and 0.035 e, while the
least favored one gave charge density of 0.025 e (Table 2). In
comparison, that of PBN is significantly lower (0.009 e).34
These findings strongly suggest that H-bonding with the
nitronyl-O results in a higher charge density on the nitronyl-C,
Table 1. EPR Hyperfine Coupling Constants for Various
Nitroxides Derived from Nitrones 13−17
Solvent
Water
MeOH
DMF
aN
(mT)
aH
(mT)
aN
(mT)
aH
(mT)
aN
(mT)
aH
(mT)
nitroxides
a
18
1.52
1.50
1.97
1.84
1.51
1.49
1.49
1.49
1.49
1.98
1.84
1.83
1.84
1.84
1.48
1.46
1.46
1.46
1.46
1.98
1.83
1.83
1.83
1.83
a
19
a
20
a
21
a
22
a
23
1.44
0.18
23−27
1.40
0.17
spin adducts
solvent
aN (mT)
aH (mT)
b
•
•
13/ OH or 13/ C
water
water
1.46
0.21
c
•
PBN/ OH
1.53−1.57
1.48
0.27−0.29
0.26
b
•
13/ CH3
water/DMSO (80/20)
water
c
•
PBN/ CH3
1.65−1.66
1.44
0.35−0.37
0.25
•
13/ CH2OH
water/MeOH (80/20)
water
c
•
PBN/ CH2OH
1.59−1.69
1.49
0.38−0.40
0.28
b
•
14/ CH3
water/DMSO (40/60)
water/MeOH (40/60)
water/DMSO (30/70)
water/MeOH (30/70)
water/DMSO (30/70)
water/MeOH (30/70)
water/DMSO (25/75)
b
b
b
b
•
14/ CH2OH
1.48
0.24
b
•
15/ CH3
1.48
0.29
•
15/ CH2OH
1.47
0.25
b
•
16/ CH3
1.47
0.27
•
16/ CH2OH
1.46
0.25
b
•
17/ CH3
1.47
0.26
•
17/ CH2OH
water/MeOH (25/75)
1.45
0.25
a
b
c
g value: 2.0062. g value: 2.0061. Parameters obtained from the
following Web site: http://www.chm.bris.ac.uk/stdb/.
identification of specific spin adducts observed by EPR in the
presence of 14−17, but the values aN of 0.26−0.29 mT
obtained correspond to carbon-centered radical adducts in
general. Regardless, all these results pointed out that the
presence of several hydroxymethyl groups yields a particular
reactivity of 13−17 that notably led to the formation of cyclic
nitroxides or to complex mixtures of paramagnetic species in
the presence of •OH.
Computational Studies. The thermodynamics of nitrone
cyclization to form the hydroxylamine and its subsequent
Figure 4. Various conformations of T-PBN (13) (in gas phase) showing the H-bond distances as well as the relative free energies and charge
densities on the nitronyl-C at the PCM(water)/B3LYP/6-31G+(d,p)//B3LYP/6-31G(d) level of theory.
943
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
Finally, the nitronyl-C charge density of conformer (A)
(0.044 e) is unprecedented for linear nitrones since this value is
in the same range as that of the highly reactive cyclic nitrones
EMPO (0.040 e) and DEPMPO (0.043 e) and only slightly
lower than that of the lead AMPO (0.060).33 This suggests that
the novel T-PBN series may exhibit a high reactivity for the
trapping of free radicals, but because of that intrinsic reactivity
and the presence of three hydroxyl groups, the Forrester−
Hepburn mechanism is unfortunately more favored. Although
the Forrester-Hepburn mechanism has been well documented
in the literature over the past four decades, we report here a
unique and specific intramolecular Forrester-Hepburn mecha-
nism of N-tert-butyl-substituted PBN compounds, which was
only suggested once in the literature.21 To understand and
overcome this uncontrolled reactivity, a thorough investigation
of the reactivity of this tris-hydroxymethyl-PBN series as well as
the development of mono- and dihydroxylated PBN derivatives
are underway. Moreover, since the synthetic strategy of this
novel series of amphiphilic nitrones is simple and straightfor-
ward allowing gram-scale synthesis of pure compounds, their
potential pharmacological interest will be investigated in the
near future.
Table 2. Nitronyl-C and Nitronyl-H Charge Densities of the
Nitronyl Group of PBN and T-PBN
a
compd
PBN
nitronyl-C
nitronyl-H
0.009
(0.010)
0.044
0.253
(0.248)
0.266
T-PBN (A)
T-PBN (B)
T-PBN (C)
(0.044)
0.035
(0.264)
0.265
(0.035)
0.026
(0.253)
0.255
(0.025)
(0.250)
a
NBO charges were determined from a natural population analysis at
the PCM/B3LYP/6-31G+(d,p)//B3LYP/6-31G(d) level of theory in
water and in DMSO (in parentheses).
therefore increasing the nitronyl reactivity toward nucleophilic
additions, which in turn could facilitate cyclization reactions.
The effect of H-bonding on the reactivity of nitronyl group had
been previously demonstrated with amide-substituted nitro-
nes33,35 and was also observed upon protonation of the
nitronyl-O.36
The effect of H-bonding on the nitronyl atoms charge
1
densities was also investigated using H NMR spectroscopy.
CONCLUSION
■
We previously demonstrated a good correlation between the
charge densities of the nitronyl group of para-substitued-PBN
compounds with the β-hydrogen shifts, the chemical shift
increasing linearly with the nitronyl-H charge density.34 As
shown in Figures S30 and S31, Supporting Information, the
A novel series of amphiphilic PBN derivatives, bearing both a
hydrophobic chain on the aromatic ring and three hydroxyl
functions on the tert-butyl group, was synthesized through a
short and convenient synthetic route based on a one-pot
method which involves in situ reduction of tris-
(hydroxymethyl)nitromethane in the presence of a benzalde-
hyde compound. The water solubility of this series was found
to be significantly affected by the length of the alkyl chain, the
longer the chain, the lower the solubility, and an opposite trend
was obviously observed as for the lipophilicity of the
compounds. The relatively low solubility of these derivatives
when compared to that of PBN may come from the formation
of intramolecular H-bonds between the hydroxyl groups, which
was supported by theoretical data obtained at the PCM/
B3LYP/6-31G+(d,p)//B3LYP/6-31G(d) level of theory. Sur-
prisingly, an intramolecular Forrester−Hepburn reaction
leading to the formation of a cyclic nitroxide was observed by
EPR when these compounds were dissolved in water or in
organic solvents. Theoretical data showed that H-bonding
between one hydroxyl function and the nitronyl oxygen gave
much higher positive charge density on the nitronyl carbon,
therefore increasing the nitronyl group reactivity toward
nucleophilic additions, which in turn could facilitate cyclization
reactions. At high nitrone concentration, a second paramagnetic
species, very likely formed by intermolecular nucleophilic
addition of two nitrone molecules, was also observed but to a
lesser extent. The free energies of reaction of both mechanisms
were computationally determined and were in full agreement
with the experimental observations for the favorability of
formation of adducts via Forrester−Hepburn mechanism.
Despite the tendency of these tris-hydroxymethyl-based
nitrones to form nitroxides, they were also found to trap
carbon-centered radicals. Finally, the very high nitronyl-C
charge density of T-PBN when compared to PBN suggests that
this novel series of compounds may exhibit intrinsic high
reactivity for the trapping of free radicals. Therefore, in the
development of novel linear nitrone compounds with improved
trapping capability, the importance of H-bonding with the
nitronyl group will have to be taken into account. Finally, since
1
nitronyl H and 13C chemical shifts of T-PBN were measured
in several solvents but no significant dependence to the solvent
polarity parameter was observed. In DMSO-d6, the chemical
shift of the β-hydrogen of T-PBN was upfield shifted (7.60
ppm), which is characteristic of a shielding of the methine
proton, while in CDCl3, the peak of resonance was downfield
shifted (7.89 ppm). Surprisingly, the opposite trend was
observed as for PBN with β-hydrogen shifts of 7.85 and 7.53
ppm, respectively in DMSO-d6 and CDCl3, in agreement with
the literature. Indeed, Janzen et al. had previously demonstrated
1
the nitronyl H chemical shift sensitivity to solvent of PBN
based on the equation: δ = 0.022 × ET(30) + 6.7045, and the
conclusion was that polar solvents could stabilize the resonance
form of PBN having the greatest phase separation (i.e., form
(1)).37
As shown in Table 2, higher nitronyl-H charge densities were
observed for the two conformers involving H-bonding with the
nitronyl-O, with NBO charges densities in water of 0.266 e for
(A) and 0.265 e for (B) while, conformer (C) exhibited a very
close value (0.255 e) to that of PBN (0.253 e); the same trend
was observed in DMSO. Therefore the upfield shift of the T-
PBN methine proton observed in DMSO-d6 suggests that T-
PBN may adopt a particular conformation in which
intermolecular H-bonds with solvent molecules may be
occurring leading to a lower positive nitronyl-H charge density.
On the contrary, the highest β-hydrogen chemical shift in
CDCl3 indicates that T-PBN may adopt a conformation for
which the nitronyl-H charge density is the highest, in
agreement with the fact that no H-bond with the solvent
molecules is expected in such a nonpolar solvent.
944
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
nm; HR-MS (ESI+, m/z) calcd for C16H24NO3 [(M + H)+] 278.1762,
found 278.1756.
gram-scale amounts of tris-hydroxymethyl-based nitrones are
now available, their protective activities will be carefully
investigated in the near future in in vitro and in vivo biological
models of oxidative-mediated diseases.
N-[4-(1,3-Dioxacyclopent-2-yl)benzyl]octanamide (5). The
synthetic procedure was essentially the same as for compound 3.
Octanoic acid (2.18 g, 15.1 × 10−3 mol), HO-Su (2.09 g, 18.2 × 10−3
mol), and DCC (3.74 g, 18.2 × 10−3 mol) were used as starting
materials. The resulting crude ester compound (2.92 g, 12.1 × 10−3
mol) and compound 2 (1.80 g, 10.1 × 10−3 mol) were used for the
reaction, and after purification by flash chromatography (EtOAc/
cyclohexane 2:8 v/v) compound 5 (2.19 g, 7.2 × 10−3 mol, 70%) was
obtained as a white powder: Rf 0.48 (EtOAc/cyclohexane 6:4 v/v);
mp 77.2−78.4 °C; 1H NMR (CDCl3, 250 MHz) δ 7.47 (2H, d, J = 8.1
Hz), 7.31 (2H, d, J = 8.1 Hz), 5.82 (2H, m), 4.46 (2H, d, J = 5.7 Hz),
4.09 (4H, m), 2.22 (2H, t, J = 7.5 Hz), 1.66 (2H, m), 1.30 (8H, m),
0.90 (3H, m); 13C NMR (CDCl3, 62.86 MHz) δ 173.0 (CO), 139.8,
137.1 (C), 127.8, 126.6 (CH), 103.5 (CH-O), 65.3, 43.3, 36.8, 31.8,
29.4, 29.0, 25.8, 22.6 (CH2), 13.8 (CH3). The spectral data of
compound 5 were in agreement with those reported by Durand et al.12
N-[4-(1,3-Dioxacyclopent-2-yl)benzyl]decanamide (6). The
synthetic procedure was essentially the same as for compound 3.
Decanoic acid (5.00 g, 29.0 × 10−3 mol), HO-Su (4.00 g, 38.8 × 10−3
mol), and DCC (7.17 g, 34.8 × 10−3 mol) were used as starting
materials. The resulting crude ester compound (5.44 g, 20.2 × 10−3
mol) and compound 2 (3.00 g, 16.8 × 10−3 mol) were used for the
reaction, and after purification by flash chromatography (EtOAc/
cyclohexane 3:7 v/v) compound 6 (3.40 g, 10.2 × 10−3 mol, 62%) was
obtained as a white powder: Rf 0.42 (EtOAc/cyclohexane 5:5 v/v); mp
EXPERIMENTAL SECTION
■
Synthesis. All reagents were from commercial sources and were
used as received. All solvents were distilled and dried according to
standard procedures. TLC analysis was performed on aluminum sheets
coated with silica gel (40−63 μm). Compound detection was achieved
either by exposure to UV light (254 nm) and by spraying a 5% sulfuric
acid solution in ethanol or a 2% ninhydrin solution in ethanol and then
by heating at ∼150 °C. Flash chromatography was carried out on silica
gel (40−63 μm). Size-exclusion chromatography was carried out on
hydroxypropylated cross-linked dextran. UV spectra were recorded on
UV/vis spectrometer equipped with a double-compartment quartz cell
of 10-mm length. Melting points have not been corrected. The 1H and
13C NMR spectra were recorded at 250 and 62.86 MHz, respectively.
Chemical shifts are given in ppm relative to the solvent residual peak
1
as a heteronuclear reference for H and 13C. Abbreviations used for
signal patterns are: s, singlet; d, doublet; t, triplet; q, quartet; sext,
sextuplet; m, multiplet; dd, doublet of doublet. Exact mass was
obtained by electrospray ionization in positive mode (M + H)+.
4-(1,3-Dioxacyclopent-2-yl)benzylamine (2).14 Under stirring
and at 0 °C, 4-cyano-1,3-dioxacyclopent-2-ylbenzyl (1.00 g, 5.8 × 10−3
mol) was dissolved in a 99:1 (v/v) ethanol/acetic acid mixture, and
0.46 g of 10% Pd/C was portionwise added under stirring. The
reaction mixture was submitted to a hydrogen atmosphere for 12 h (8
bar), then the crude mixture was filtered off through a pad of Celite
and the solvent was evaporated under vacuum to give compound 2
(0.98 g, 5.5 × 10−3 mol, 96%) as a yellow oil. Compound 2 was
directly used in the next step without further purification. The spectral
data of compound 2 were in agreement with those reported by Ouari
et al.14
N-[4-(1,3-Dioxacyclopent-2-yl)benzyl]butanamide (3). Under
stirring, butanoic acid (1.33 g, 15.1 × 10−3 mol), HO-Su (2.09 g, 18.1
× 10−3 mol), and DCC (3.74 g, 18.2 × 10−3 mol) were dissolved in dry
CH2Cl2 at room temperature. After 6 h of being stirred, the reaction
mixture was filtered off, and the solvent was removed under vacuum.
The resulting crude ester compound (2.25 g, 12.1 × 10−3 mol) and
compound 2 (1.80 g, 10.1 × 10−3 mol) were dissolved in dry CH2Cl2
with DIEA (pH = 8−9). The mixture was stirred for 20 h at room
temperature, and the solvent was removed under vacuum. The crude
mixture was purified by flash chromatography (EtOAc/cyclohexane
2:8 v/v) to give compound 3 (1.40 g, 3.8 × 10−3 mol, 58%) as a white
powder: Rf 0.32 (EtOAc/cyclohexane 6:4 v/v); mp 86.4−87.6 °C; 1H
NMR (CDCl3, 250 MHz) δ 7.47 (2H, d, J = 8.1 Hz), 7.31 (2H, d, J =
8.1 Hz), 5.81 (2H, m), 4.46 (2H, d, J = 5.7 Hz), 4.10 (4H, m), 2.20
(2H, t, J = 7.4 Hz), 1.70 (2H, sext, J = 7.4 Hz), 0.97 (3H, t, J = 7.4
Hz); 13C NMR (CDCl3, 62.86 MHz) δ 172.8 (CO), 139.5, 137.2 (C),
127.9, 126.8 (CH), 103.5 (CH-O), 65.3, 43.3, 38.7, 19.2 (CH2), 13.8
(CH3). UV (CH2Cl2) λmax 230 nm; HR-MS (ESI+, m/z) calcd for
C14H20NO3 [(M + H)+] 250.1443, found 250.1443.
1
80.1−80.9 °C; H NMR (CDCl3, 250 MHz) δ 7.47 (2H, d, J = 8.1
Hz), 7.31 (2H, d, J = 8.1 Hz), 5.82 (2H, m), 4.46 (2H, d, J = 5.7 Hz),
4.10 (4H, m), 2.22 (2H, t, J = 7.5 Hz), 1.66 (2H, m), 1.28 (12H, m),
0.90 (3H, m); 13C NMR (CDCl3, 62.86 MHz) δ 173.0 (CO), 139.5,
137.2 (C), 127.9, 126.9 (CH), 103.5 (CH-O), 65.3, 43.3, 36.8, 31.9,
29.5, 29.4, 29.3, 25.8, 25.6, 22.7 (CH2), 14.1 (CH3); UV (CH2Cl2)
λmax 230 nm; HR-MS (ESI+, m/z) calcd for C20H32NO3 [(M + H)+]
334.2376, found 334.2382.
N-(4-Formylbenzyl)butanamide (7). Under stirring, compound
3 (0.90 g, 3.6 × 10−3 mol) was dissolved in a 7:3 (v/v) AcOH/H2O
mixture. After one night of stirring, 30 mL of EtOAc was added, and
the organic layer was successively washed with NaHCO3 saturated
solution and brine, dried over Na2SO4, and evaporated under vacuum
to give compound 7 (0.72 g, 3.5 × 10−3 mol, 96%) as a white powder:
1
Rf 0.29 (EtOAc/cyclohexane 6:4 v/v); mp 72.2−73.0 °C; H NMR
(CDCl3, 250 MHz) δ 9.97 (1H, s), 7.84 (2H, d, J = 8.1 Hz), 7.43 (2H,
d, J = 8.1 Hz), 6.31 (1H, m), 4.51 (2H, d, J = 6.0 Hz), 2.25 (2H, t, J =
7.4 Hz), 1.70 (2H, sext, J = 7.4 Hz), 0.96 (3H, t, J = 7.4 Hz); 13C
NMR (CDCl3, 62.86 MHz) δ 191.9 (CHO), 173.4 (CO), 145.6, 135.6
(C), 130.1, 128.1 (CH), 43.2, 38.5, 19.2 (CH2), 13.8 (CH3); UV
(CH2Cl2) λmax 256 nm; HR-MS (ESI+, m/z) calcd for C12H16NO2
[(M + H)+] 206.1173, found 206.1181.
N-(4-Formylbenzyl)hexanamide (8). The synthetic procedure
was essentially the same as for compound 7. From compound 4 (1.12
g, 4.1 × 10−3 mol), compound 8 (0.90 g, 3.9 × 10−3 mol, 94%) was
obtained as a white powder: Rf 0.34 (EtOAc/cyclohexane 6:4 v/v); mp
59.2−59.8 °C; 1H NMR (CDCl3, 250 MHz) δ 9.99 (1H, s), 7.84 (2H,
d, J = 8.2 Hz), 7.43 (2H, d, J = 8.2 Hz), 6.18 (1H, m), 4.52 (2H, d, J =
6.0 Hz), 2.26 (2H, t, J = 7.6 Hz), 1.68 (2H, m), 1.34 (4H,m), 0.91
(3H, t, J = 6.7 Hz); 13C NMR (CDCl3, 62.86 MHz) δ 192.1 (CHO),
174.0 (CO), 145.7, 135.4 (C), 130.1, 128.0 (CH), 43.1 (CH2), 36.5,
31.4, 25.4, 22.3 (CH2), 13.9 (CH3); UV (CH2Cl2) λmax 256 nm; HR-
MS (ESI+, m/z) calcd for C14H20NO2 [(M + H)+] 234.1488, found
234.1494.
N-(4-Formylbenzyl)octanamide (9). The synthetic procedure
was essentially the same as for compound 7. From compound 5 (2.18
g, 7.1 × 10−3 mol), compound 9 (1.82 g, 7.0 × 10−3 mol, 98%) was
obtained as a white powder: Rf 0.40 (EtOAc/cyclohexane 6:4 v/v); mp
62.5−63.5 °C; 1H NMR (CDCl3, 250 MHz) δ 10.0 (1H, s), 7.86 (2H,
d, J = 8.1 Hz), 7.45 (2H, d, J = 8.1 Hz), 6.06 (1H, m), 4.54 (2H, d, J =
6.0 Hz), 2.27 (2H, t, J = 7.6 Hz), 1.68 (2H, m), 1.30 (8H, m), 0.89
(3H, t, J = 6.7 Hz); 13C NMR (CDCl3, 62.86 MHz) δ 192.9 (CHO),
173.0 (CO), 139.7, 137.0 (C), 127.9, 126.8 (CH), 43.3 (CH2), 36.8,
N-[4-(1,3-Dioxacyclopent-2-yl)benzyl]hexanamide (4). The
synthetic procedure was essentially the same as for compound 3.
Hexanoic acid (2.00 g, 17.2 × 10−3 mol), HO-Su (2.37 g, 20.6 × 10−3
mol), and DCC (4.25 g, 20.6 × 10−3 mol) were used as starting
materials. The resulting crude ester compound (2.10 g, 9.9 × 10−3
mol) and compound 2 (1.60 g, 9.0 × 10−3 mol) were used for the
reaction, and after purification by flash chromatography (EtOAc/
cyclohexane 2:8 v/v) compound 4 (1.60 g, 4.3 × 10−3 mol, 64%) was
obtained as a white powder: Rf 0.44 (EtOAc/cyclohexane 6:4 v/v); mp
1
73.5−74.3 °C; H NMR (CDCl3, 250 MHz) δ 7.48 (2H, d, J = 8.2
Hz), 7.31 (2H, d, J = 8.2 Hz), 5.82 (2H, m), 4.46 (2H, d, J = 5.7 Hz),
4.11 (4H, m), 2.22 (2H, t, J = 7.6 Hz), 1.69 (2H, m), 1.32 (4H, m),
0.91 (3H, t, J = 6.7 Hz); 13C NMR (CDCl3, 62.86 MHz) δ 172.9
(CO), 139.5, 137.2 (C), 127.9, 126.8 (CH), 103.5 (CH−O), 65.3,
43.3, 36.8, 31.8, 25.4, 22.4 (CH2), 14.0 (CH3); UV (CH2Cl2) λmax 230
945
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
31.7, 29.3, 29.0, 25.8, 22.6 (CH2), 14.1 (CH3). The spectral data of
compound 5 were in agreement with those reported by Durand et al.12
N-(4-Formylbenzyl)decanamide (10). The synthetic procedure
was essentially the same as for compound 7. From compound 6 (1.78
g, 5.3 × 10−3 mol), compound 10 (1.49 g, 5.1 × 10−3 mol, 97%) was
obtained as a white powder: Rf 0.46 (EtOAc/cyclohexane 5:5 v/v); mp
69.0−69.8 °C; 1H NMR (CDCl3, 250 MHz) δ 10.0 (1H, s), 7.85 (2H,
d, J = 8.2 Hz), 7.45 (2H, d, J = 8.2 Hz), 6.08 (1H, m), 4.54 (2H, d, J =
6.0 Hz), 2.27 (2H, t, J = 7.6 Hz), 1.68 (2H, m), 1.28 (12H, m), 0.90
(3H, m); 13C NMR (CDCl3, 62.86 MHz) δ 191.9 (CHO), 173.3
(CO), 145.6, 135.6 (C), 130.1, 128.1 (CH), 43.2 (CH2), 36.7, 31.9,
29.5, 29.4, 29.3, 25.8, 25.6, 22.7 (CH2), 14.1 (CH3); UV (CH2Cl2)
λmax 256 nm; HR-MS (ESI+, m/z) calcd for C18H28NO2 [(M + H)+]
290.2118, found 290.2120.
α-Phenyl-N-(2-hydroxymethyl-1,3-dihydroxy-2-propyl)-
imine (11). Under stirring, benzaldehyde (6.00 g, 56.5 × 10−3 mol)
and tris(hydroxymethyl)aminomethane (8.22 g, 67.8 × 10−3 mol) were
dissolved in toluene, and the mixture was heated at reflux using a
Dean−Stark apparatus. After 6 h of being refluxed, the reaction
mixture was cooled and filtered, and the solvent was removed under
vacuum to give compound 11 (11.0 g, 52.5 × 10−3 mol, 93%) as a
white oil, which was used without further purification.
α-Phenyl-N-(2-hydroxymethyl-1,3-dihydroxy-2-propyl)-
amine (12). Under stirring and at 0 °C, compound 11 (2.00 g, 9.6 ×
10−3 mol) was dissolved in a 99:1 (v/v) ethanol/acetic acid mixture,
and 0.57 g of 10% Pd/C was portionwise added. The reaction mixture
was submitted to a hydrogen atmosphere for 4 h (8 bar), and then the
crude mixture was filtered off through a pad of Celite and the solvent
was removed under vacuum. The crude mixture was purified by flash
chromatography (EtOAc/methanol 9:1 v/v) to give compound 12
(1.46 g, 6.9 × 10−3 mol, 72%) as a white powder: Rf 0.39 (EtOAc/
methanol 8:2 v/v); mp 86.4−87.6 °C; 1H NMR (CDCl3, 250 MHz) δ
7.36−7.21 (5H, m), 4.34 (4H, m), 3.73 (2H, s), 3.41 (6H, s); 13C
NMR (CDCl3, 62.86 MHz) δ 142.3 (C), 128.6, 128.5, 126.9 (CH),
61.6 (CH2), 60.5 (C), 45.9 (CH2); UV (MeOH) λmax 212 nm; HR-
MS (ESI+, m/z) calcd for C11H18NO3 [(M + H)+] 212.1286, found
212.1287.
phere, compound 7 (0.80 g, 3.9 × 10−3 mol), tris(hydroxymethyl)-
nitromethane (1.17 g, 7.8 × 10−3 mol), and AcOH (1.33 mL, 23.34 ×
10−3 mol) were dissolved in EtOH. The mixture was cooled to 0 °C,
and then zinc powder (1.00 g, 15.6 × 10−3 mol) was slowly added in
order to keep the temperature at 15 °C. The mixture was stirred at
room temperature for a couple of minutes and then heated at 60 °C in
the dark for 10 h in the presence of molecular sieves (4 Å). The
reaction mixture was filtered off through a pad of Celite, and the
solvent was removed under vacuum. The crude mixture was purified
by flash chromatography (EtOAc/methanol 9:1 v/v) followed by two
successive crystallizations from ethanol to give compound 14 (0.90 g,
2.8 × 10−3 mol, 72%) as a white powder: Rf 0.40 (EtOAc/methanol
1
8:2 v/v); mp 160.2−160.6 °C; H NMR (DMSO, 250 MHz) δ 8.35
(1H, t, J = 6.0 Hz), 8.30 (2H, d, J = 8.0 Hz), 7.57 (1H, s), 7.29 (2H, d,
J = 8.0 Hz), 4.94 (3H, t, J = 5.5 Hz), 4.30 (2H, d, J = 6.0 Hz), 3.81
(6H, d, J = 5.5 Hz), 2.14 (2H, t, J = 7.2 Hz), 1.56 (2H, sext, J = 7.2
Hz), 0.87 (3H, t, J = 7.2 Hz); 13C NMR (DMSO, 62.86 MHz) δ 172.5
(CO), 142.1 (C), 133.6 (CH), 130.2 (C), 129.3, 127.2 (CH), 80.3
(C), 60.4, 43.2, 37.8, 19.2 (CH2), 14.1 (CH3); UV (MeOH) λmax 299
nm; HR-MS (ESI+, m/z) calcd for C16H25N2O5 [(M + H)+]
325.1758, found 325.1760.
α-(4-Hexanamidomethyl)phenyl-N-(2-hydroxymethyl-1,3-
dihydroxy-2-propyl)nitrone (15). The synthetic procedure was
essentially the same as for compound 14. Compound 8 (0.60 g, 2.6 ×
10−3 mol), tris(hydroxymethyl)nitromethane (0.78 g, 5.1 × 10−3 mol),
AcOH (1.18 mL, 20.6 × 10−3 mol), and zinc powder (0.89 g, 13.7 ×
10−3 mol) were used as starting materials. The crude mixture was
purified by flash chromatography (EtOAc/methanol 9:1 v/v) followed
by two successive crystallizations from ethanol to give compound 15
(0.58 g, 1.6 × 10−3 mol, 64%) as a white powder: Rf 0.44 (EtOAc/
methanol 8:2 v/v); mp 137.6−138.2 °C; 1H NMR (DMSO, 250
MHz) δ 8.35 (1H, t, J = 5.9 Hz), 8.30 (2H, d, J = 8.0 Hz), 7.57 (1H,
s), 7.29 (2H, d, J = 8.0 Hz), 4.94 (3H, t, J = 5.4 Hz), 4.30 (2H, d, J =
5.9 Hz), 3.81 (6H, d, J = 5.4 Hz), 2.14 (2H, t, J = 7.2 Hz), 1.56 (2H,
m), 1.27 (4H, m), 0.87 (3H, t, J = 7.2 Hz); 13C NMR (DMSO, 62.86
MHz) δ 172.5 (CO), 142.1 (C), 133.6 (CH), 130.2 (C), 129.3, 127.2
(CH), 80.3 (C), 60.4, 43.2, 35.8, 31.3, 25.5, 22.3 (CH2), 14.4 (CH3);
UV (MeOH) λmax 299 nm; HR-MS (ESI+, m/z) calcd for
C18H29N2O5 [(M + H)+] 353.2057, found 353.2057.
α-(4-Octanamidomethyl)phenyl-N-(2-hydroxymethyl-1,3-di-
hydroxy-2-propyl)nitrone (16). The synthetic procedure was
essentially the same as for compound 14. Compound 9 (0.80 g, 3.1
× 10−3 mol), tris(hydroxymethyl)nitromethane (0.92 g, 6.1 × 10−3
mol), AcOH (1.05 mL, 18.4 × 10−3 mol), and zinc powder (0.80 g,
12.2 × 10−3 mol) were used as starting materials. The crude mixture
was purified by flash chromatography (EtOAc/methanol 9:1 v/v)
followed by two successive crystallizations from ethanol to give
compound 16 (0.78 g, 2.1 × 10−3 mol, 68%) as a white powder: Rf
0.36 (EtOAc/methanol 9:1 v/v); mp 144.6−145.4 °C; 1H NMR
(DMSO, 250 MHz) δ 8.38 (1H, t, J = 6.0 Hz), 8.29 (2H, d, J = 8.4
Hz), 7.58 (1H, s), 7.28 (2H, d, J = 8.4 Hz), 4.96 (3H, t, J = 5.5 Hz),
4.29 (2H, d, J = 6.0 Hz), 3.81 (6H, d, J = 5.5 Hz), 2.15 (2H, t, J = 7.3
Hz), 1.52 (2H, m), 1.25 (8H, m), 0.87 (3H, t, J = 7.1 Hz); 13C NMR
(DMSO, 62.86 MHz) δ 172.8 (CO), 142.1 (C), 133.6 (CH), 130.2
(C), 129.3, 127.2 (CH), 80.2 (C), 60.4, 42.3, 35.8, 31.7, 29.1, 28.9,
25.8, 22.5 (CH2), 14.4 (CH3); UV (MeOH) λmax 299 nm; HR-MS
(ESI+, m/z) calcd for C20H33N2O5 [(M + H)+] 381.2384, found
381.2384.
α-Phenyl-N-(2-hydroxymethyl-1,3-dihydroxy-2-propyl)-
nitrone (13). (a) First Synthetic Strategy. Under stirring and argon
atmosphere, compound 12 (0.80 g, 3.8 × 10−3 mol) was dissolved in
EtOH, and then Na2WO4 (0.038 g, 1.15.10−3 mol) dissolved in 1 mL
of water was added to the mixture. The mixture was cooled to 0 °C,
and H2O2 30% in water (0.78 g, 6.9.10−3 mol) was added dropwise,
and then the mixture was stirred at room temperature for 18 h. The
solvent was removed under vacuum, and the crude mixture was
purified by flash chromatography (EtOAc) followed by two successive
crystallizations from EtOAc/n-hexane to give compound 13 (0.160 g,
7.10−4 mol, 36%) as a white powder. (b) Second Synthetic Strategy.
Under stirring and argon atmosphere, benzaldehyde (1.00 g, 9.4 ×
10−3 mol), tris(hydroxymethyl)nitromethane (2.84 g, 18.9 × 10−3
mol), and AcOH (3.23 mL, 56.5 × 10−3 mol) were dissolved in EtOH.
The mixture was cooled to 0 °C, and then zinc powder (2.45 g, 37.7 ×
10−3 mol) was slowly added in order to keep the temperature at 15 °C.
The mixture was stirred at room temperature for a couple of minutes
and then heated at 60 °C in the dark for 10 h in the presence of
molecular sieves (4 Å). The reaction mixture was filtered off through a
pad of Celite, and the solvent was removed under vacuum. The crude
mixture was purified by flash chromatography (EtOAc) followed by
two successive crystallizations from EtOAc/n-hexane to give
compound 13 (1.52 g, 6.7 × 10−3 mol, 72%) as a white powder: Rf
0.28 (EtOAc/methanol 9.5:0.5 v/v); mp 83.4−84.8 °C; 1H NMR
(DMSO, 250 MHz) δ 8.36 (2H, m), 7.60 (1H, s), 7.43 (3H, m), 4.94
(3H, t, J = 5.1 Hz), 3.82 (6H, d, J = 5.1 Hz); 13C NMR (DMSO, 62.86
MHz) δ 133.7 (C), 131.7 (CH), 130.2, 129.3, 128.6 (CH), 80.4 (C),
60.4 (CH2); UV (MeOH) λmax 295 nm; MS (ESI+, m/z) 248.1 [M +
Na]+, 226.1 [M + H]+. The spectral data of compound 13 were in
agreement with those reported by Janzen and Zawalski except for the
melting point that was found to be 89−91 °C.21
α-(4-Decanamidomethyl)phenyl-N-(2-hydroxymethyl-1,3-
dihydroxy-2-propyl)nitrone (17). The synthetic procedure was
essentially the same as for compound 14. Compound 10 (1.20 g, 4.2 ×
10−3 mol), tris(hydroxymethyl)nitromethane (1.26 g, 8.3 × 10−3 mol),
AcOH (1.43 mL, 25.0 × 10−3 mol), and zinc powder (1.08 g, 16.6 ×
10−3 mol) were used as starting materials. The crude mixture was
purified by flash chromatography (EtOAc/methanol 9:1 v/v) followed
by two successive crystallizations from ethanol to give compound 17
(1.16 g, 2.8 × 10−3 mol, 68%) as a white powder: Rf 0.40 (EtOAc/
methanol 9:1 v/v); mp 150.8−151.6 °C; 1H NMR (DMSO, 250
MHz) δ 8.38 (1H, t, J = 6.0 Hz), 8.30 (2H, d, J = 8.3 Hz), 7.57 (1H,
s), 7.28 (2H, d, J = 8.3 Hz), 4.97 (3H, t, J = 5.5 Hz), 4.29 (2H, d, J =
α-(4-Butanamidomethyl)phenyl-N-(2-hydroxymethyl-1,3-di-
hydroxy-2-propyl)nitrone (14). Under stirring and argon atmos-
946
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
6.0 Hz), 3.80 (6H, d, J = 5.5 Hz), 2.15 (2H, t, J = 7.3 Hz), 1.54 (2H,
m), 1.25 (12H, m), 0.87 (3H, t, J = 6.8 Hz); 13C NMR (DMSO, 62.86
MHz) δ 172.7 (CO), 142.0 (C), 133.5 (CH), 130.2 (C), 129.3, 127.2
(CH), 80.3 (C), 60.4, 43.2, 35.8, 31.8, 29.4, 29.2, 29.2, 29.1, 25.8, 22.6
(CH2), 14.5 (CH3); UV (MeOH) λmax 299 nm; HR-MS (ESI+, m/z)
calcd for C22H37N2O5 [(M + H)+] 409.2697, found 409.2690.
Determination of log k′W Values. Compounds were dissolved in
MeOH at 1.0 mg/mL and were injected onto a Microsorb C18
reversed-phase column (250 mm × 4.6 mm, 5 μm). The compounds
were eluted at various MeOH and water ratios (9:1 to 3:7 v/v) using a
flow rate of 0.8 mL/min. The column temperature was 25 °C, and the
UV detector wavelength was λ = 298 nm. Linear regression analysis
were performed on three data points for TC5-PBN (from 6:4 to 4:6; r2
= 0.9998) and TC7-PBN (from 8:2 to 6:4; r2 = 1) and four points for
T-PBN (from 7:3 to 3:7; r2 = 0.9988), TC3-PBN (from 6:4 to 3:7; r2 =
0.9981) and for TC9-PBN (from 9:1 to 6:4; r2 = 0.9986). The log k′
values were calculated by using the equation: log k′ = log((t − t0)/t0),
where t is the retention time of the nitrone and t0 is the elution time of
MeOH, which is not retained on the column.
acetate (3 mM) using a dilution factor of 1/103. Diluted sample
solutions were immediately introduced in the ionization source at a 5
μL·min−1 flow rate using a syringe pump.
Computational Methods. All calculations were performed at the
Ohio Supercomputer Center. Density functional theory39,40 was used
in this study to determine the optimized geometry, vibrational
frequencies, and single-point energy of all stationary points41−43 using
Gaussian 03.44 The effect of aqueous solvation was also investigated
using the polarizable continuum model (PCM).45−47 Single-point
energies were obtained at the B3LYP/6-31+G** level based on the
optimized B3LYP/6-31G* geometries. Charge densities were obtained
from a natural population analysis (NPA)48 at the single point PCM/
B3LYP/6-31+G** level. These calculations used six Cartesian d
functions. Stationary points for nitrones and their respective adducts
have zero imaginary vibrational frequency as derived from a vibrational
frequency analysis (B3LYP/6-31G*). A scaling factor of 0.981 was
used for the zero-point vibrational energy (ZPE) corrections for the
B3LYP/6-31G* level.49 Spin contamination for all of the stationary
point of the radical structures was negligible, i.e., ⟨S2⟩ = 0.75.
Electron Paramagnetic Resonance and Spin-Trapping
Experiments. A saturated solution of T-PBN was first prepared at
0.15 mol·L−1 (∼2× the solubility limit in water). The supernatant of
that solution was used for the analysis and was referred to as the
concentrated solution. In other experiments, a 0.008 mol·L−1 nitrone
aqueous solution was prepared, and it was referred to as the low
concentrated solution. The concentrations of 18 and 20 were
evaluated in water, methanol, and DMF by simulation of the EPR
signal recorded from a solution containing the nitrone 13 or 15
(0.01−0.05 mol·L−1, depending on the nitrone and the solvent) in the
presence of 3-carboxyproxyl (3-CP, 0.4 mmol·L−1) used here as an
internal standard. In the various spin-trapping experiments, the free
radicals were produced in 10 mmol·L−1 phosphate buffer at pH 7.2, in
the presence of various amounts of methanol or DMSO, and mostly
using a 0.008 mol·L−1 nitrone concentration, though this parameter
could be varied in between 0.004 mol·L−1 and 0.15 mol·L−1. High
nitrone concentrations were obtained by dissolving the spin trap in
DMF or DMSO before adding phosphate buffer and the radical
generator. The superoxide generating system employed used 0.4
mmol·L−1 xanthine and 0.1 unit·mL−1 xanthine oxidase. Hydroxyl
radical was produced by a standard Fenton system consisting of 0.1%
H2O2, 1 mmol·L−1 ethylenediaminetetraacetic acid (EDTA), and 1
mmol·L−1 FeSO4. For the concentrated Fenton system, the
concentrations of FeSO4 and H2O2 were doubled. Carbon-centered
radicals •CH3 and •CH2OH were generated by adding DMSO (20−
75%) or methanol (20−75%), respectively, to the standard Fenton
system described above. EPR assays were carried out in capillary tubes
and the spectra were recorded at room temperature (20−22 °C) on a
continuous wave X-band Bruker EMX spectrometer, equipped with an
NMR gaussmeter for magnetic field calibration in order to evaluate g
values. The following conditions were used: modulation frequency,
100 kHz; non saturating microwave power, 1−20 mW; modulation
amplitude, 0.1−0.15 mT; receiver gain, 105−106; time constant, 1.28−
655 ms; scan time, 60−180 s. Simulations were carried out using the
EPR Software WinSim2002.38
Mass Spectrometry. High-resolution MS and MS/MS experi-
ments were performed using a QStar Elite (AB Sciex) mass
spectrometer equipped with an electrospray ionization source operated
in the positive mode. The capillary voltage was set at +5500 V and the
cone voltage at +70 V. In this hybrid instrument, ions were measured
using an orthogonal acceleration time-of-flight (oa-TOF) mass
analyzer. A quadrupole was used for selection of precursor ions to
be further submitted to collision-induced dissociation (CID) in MS/
MS experiments. Air was used as the nebulizing gas (10 psi) while
nitrogen was used as the curtain gas (20 psi) as well as the collision
gas. Collision energy was set according to the experiments. Samples
were prepared in water, methanol or DMF and kept at room
temperature. Nitrone solutions were prepared as described in the EPR
and spin trapping experiments section. After 1−2 h, an aliquot was
submitted to an EPR analysis while the rest of the sample was diluted
in methanol (1/10) and then in a methanol solution of ammonium
ASSOCIATED CONTENT
■
S
* Supporting Information
1
Complete ref 41; H NMR spectra of compounds 3, 4, 6−8,
10, 12, and 14−17; 13C NMR spectra of compounds 3−6, 12,
and 14−17; HRMS spectra of compounds 14−17; HPLC
1
chromatograms of compounds 14−17; plot of the nitronyl H
and 13C NMR chemical shift of T-PBN vs the solvent polarity
parameter; ESI MS/MS spectrum of T-PBN and TC5−PBN;
thermodynamic parameters of the nitrones and nitroxides in
aqueous and DMSO phases at the PCM/B3LYP/6-31+g**//
B3LYP/6-31g* level of theory; Cartesian coordinates. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Tel: +33 (0)4 9014 4445. Fax: +33 (0)4 9014 4441. E-mail:
gregory.durand@univ-avignon.fr.
■
ACKNOWLEDGMENTS
■
Particular thanks are due to Marie Rosselin (Universite
d’Avignon et des Pays de Vaucluse) for her assistance in the
study of the stability of T-PBN in organic solvents. Fanny
́
Choteau was the recipient of a fellowship from the “Reg
́
ion
Provence Alpes Cote d’Azur” and Targeting System Pharma
̂
Company. L.C. acknowledges support from Spectropole, the
Analytical facility of Aix-Marseille University, by allowing
special access to the instruments purchased with European
Funding (FEDER OBJ2142-3341).
REFERENCES
■
(1) Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and
Medicine; 4th ed.; Oxford University Press: Oxford, 2007.
(2) Villamena, F. A.; Zweier, J. L. Antiox. Red. Signal 2004, 6, 619−
629.
(3) Rosen, G. M.; Britigan, B., E.; Halpern, H., J.; Pou, S. Free
Radicals: Biology and Detection By Spin Trapping; Oxford University
Press: New York, 1999.
(4) Floyd, R. A.; Hensley, K.; Forster, M. J.; Kelleher-Andersson, J.
A.; Wood, P. L. Mech. Ageing Dev. 2002, 123, 1021−1031.
(5) Floyd, R. A.; Kopke, R. D.; Choi, C.-H.; Foster, S. B.; Doblas, S.;
Towner, R., A. Free Radical Biol. Med. 2008, 45, 1361−1374.
(6) Green, A. R.; Ashwood, T.; Odergren, T.; Jackson, D. M.
Pharmacol. Ther. 2003, 100, 195−214.
(7) Floyd, R. A.; Hensley, K.; Jaffrey, F.; Maidt, L.; Robinson, K.; Pye,
Q.; Stewart, C. Life Sci. 1999, 65, 1893−1899.
947
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948

The Journal of Organic Chemistry
Article
(8) Durand, G.; Poeggeler, B.; Boker, J.; Raynal, S.; Polidori, A.;
Pappolla, M. A.; Hardeland, R.; Pucci, B. J. Med. Chem. 2007, 50,
3976−3979.
(43) Hehre, W. J.; Radom, L.; Schleyer, P. V.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley Interscience: New York, 1986.
(44) Frisch, M. J. et al. Gaussian, Revision B.01; Gaussian, Inc.:
Wallingford, CT, 2004; Pittsburgh PA, 2003.
(45) Barone, V.; Cossi, M.; Tomasi, J. J. Chem. Phys. 1997, 107,
3210−3221.
(46) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999−3093.
̈
(9) Durand, G.; Polidori, A.; Ouari, O.; Tordo, P.; Geromel, V.;
Rustin, P.; Pucci, B. J. Med. Chem. 2003, 46, 5230−5237.
(10) Durand, G.; Prosak, R. A.; Han, Y.; Ortial, S.; Rockenbauer, A.;
Pucci, B.; Villamena, F. A. Chem. Res. Toxicol. 2009, 22, 1570−1581.
(11) Choteau, F.; Durand, G.; Ranchon-Cole, I.; Cercy, C.; Pucci, B.
Bioorg. Med. Chem. Lett. 2010, 20, 7405−7409.
(47) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027−2094.
(48) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
899−926.
(12) Durand, G.; Poeggeler, B.; Ortial, S.; Polidori, A.; Villamena, F.
A.; Boker, J.; Hardeland, R.; Pappolla, M. A.; Pucci, B. J. Med. Chem.
̈
(49) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502−16513.
2010, 53, 4849−4861.
(13) Ortial, S.; Durand, G.; Poeggeler, B.; Polidori, A.; Pappolla, M.
A.; Boeker, J.; Hardeland, R.; Pucci, B. J. Med. Chem. 2006, 49, 2812−
2820.
(14) Ouari, O.; Chalier, F.; Bonaly, R.; Pucci, B.; Tordo, P. J. Chem.
Soc., Perkin Trans. 2 1998, 2299−2307.
(15) Hinton, R. D.; Janzen, E. G. J. Org. Chem. 1992, 57, 2646−2651.
(16) Migrdichian, V. Organic Synthesis; Wiley: New York, 1957; Vol.
2.
(17) Allouch, A.; Roubaud, V.; Lauricella, R.; Bouteiller, J.-C.; Tuccio,
B. Org. Biomol. Chem. 2003, 1, 593−598.
(18) Bernotas, R. C.; Thomas, C. E.; Carr, A. A.; Nieduzak, T. R.;
Adams, G.; Ohlweiler, D. F.; Hay, D. A. Bioorg. Med. Chem. Lett. 1996,
6, 1105−1110.
(19) Murahashi, S.-I.; Mitsui, H.; Shiota, T.; Tsuda, T.; Watanabe, S.
J. Org. Chem. 1990, 55, 1736−1744.
(20) Huie, R.; Cherry, W. R. J. Org. Chem. 1985, 50, 1531−1532.
(21) Janzen, E. G.; Zawalski, R. C. J. Org. Chem. 1978, 43, 1900−
1903.
(22) Bernotas, R. C.; Adams, G.; Carr, A. A. Tetrahedron 1996, 52,
6519−6526.
(23) Gautheron-Chapoulaud, V.; Pandya, S. U.; Cividino, P.; Masson,
́
G.; Py, S.; Vallee, Y. Synlett. 2001, 8, 1281−1283.
(24) Abla, M.; Durand, G.; Pucci, B. J. Org. Chem. 2008, 73, 8142−
8153.
(25) Sharma, K. S.; Durand, G.; Pucci, B. Des. Monomers Polym. 2011,
14, 499−513.
(26) Buettner, G. Free Radic. Biol. Med. 1987, 3, 259−303.
(27) Forrester, A. R.; Alexander, R.; Hepburn, S. P. J. Chem. Soc. C:
Org. 1971, 4, 701−703.
(28) Tuccio, B.; Zeghdaoui, A.; Finet, J.-P.; Cerri, V.; Tordo, P. Res.
Chem. Intermed. 1996, 22, 393−404.
(29) El Hassan, I.; Charles, L.; Lauricella, R.; Tuccio, B. New J. Chem.
2008, 32, 680−688.
(30) Triquigneaux, M.; Tuccio, B.; Lauricella, R.; Charles, L. J. Am.
Soc. Mass Spectrom. 2009, 20, 2013−2020.
(31) Gunther, M. R.; Hanna, P. M.; Mason, R. P.; Cohen, M. S. Arch.
Biochem. Biophys. 1995, 316, 515−522.
(32) Yamazaki, I.; Piette, L. H. J. Am. Chem. Soc. 1991, 113, 7588−
7593.
(33) Villamena, F. A.; Xia, S.; Merle, J. K.; Lauricella, R.; Tuccio, B.;
Hadad, C. M.; Zweier, J. L. J. Am. Chem. Soc. 2007, 129, 8177−8191.
(34) Durand, G.; Choteau, F.; Pucci, B.; Villamena, F. A. J. Phys.
Chem. A 2008, 112, 12498−12509.
(35) Villamena, F. A.; Rockenbauer, A.; Gallucci, J.; Velayutham, M.;
Hadad, C. M.; Zweier, J. L. J. Org. Chem. 2004, 69, 7994−8004.
(36) Burgett, R. A.; Bao, X.; Villamena, F. A. J. Phys. Chem. A 2008,
112, 2447−2455.
(37) Janzen, E. G.; Haire, D. L.; Coulter, G. A.; Stronks, H. J.;
Krygsman, P. H.; Towner, R. A.; Hilborn, J. W. J. Org. Chem. 1989, 54,
2915−2920.
(38) Duling, D. R. J. Magn. Reson., Ser. B 1994, 104, 105−10.
(39) Labanowski, J. W.; Andzelm, J. Density Functional Methods in
Chemistry; Springer: New York, 1991.
(40) Parr, R. G.; Yang, W. Density Functional Theory in Atoms and
Molecules; Oxford University Press: New York, 1989.
(41) Becke, A. D. Phys. Rev. 1988, 38, 3098−3100.
(42) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
948
dx.doi.org/10.1021/jo202098x | J. Org. Chem. 2012, 77, 938−948