Effect of the phosphoryl substituent in the linear nitrone on the spin trapping of superoxide radical and the stability of the superoxide adduct: Combined experimental and theoretical studies

Article abstract of DOI:10.1021/jo061204m

(Chemical Equation Presented) A new phosphorylated linear nitrone N-(4-hydroxybenzyliene)-1-diethoxyphosphoryl-1-methylethylamine N-oxide (4-HOPPN) was synthesized, and its X-ray structure was determined. The spin trapping ability of various kinds of free

Full text of DOI:10.1021/jo061204m

Effect of the Phosphoryl Substituent in the Linear Nitrone on the
Spin Trapping of Superoxide Radical and the Stability of the
Superoxide Adduct: Combined Experimental and
Theoretical Studies
Yang-Ping Liu,†,‡ Lan-Fen Wang,†,‡ Zhou Nie,† Yi-Qiong Ji,† Yang Liu,*,†
Ke-Jian Liu,*,§ and Qiu Tian†
State Key Laboratory for Structural Chemistry of Unstable and Stable Species, Center for Molecular
Science, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100080, China, Graduate School,
The Chinese Academy of Sciences, Beijing 100080, China, and College of Pharmacy, UniVersity of
New Mexico, 2502 Marble NE, Albuquerque, New Mexico 87131
yliu@iccas.ac.cn; jliu@unm.edu
ReceiVed June 12, 2006
A new phosphorylated linear nitrone N-(4-hydroxybenzyliene)-1-diethoxyphosphoryl-1-methylethylamine
N-oxide (4-HOPPN) was synthesized, and its X-ray structure was determined. The spin trapping ability
of various kinds of free radicals by 4-HOPPN was evaluated. Kinetic study of decay of the superoxide
spin adduct (4-HOPPN-OOH) shows the half-life time of 8.8 min. On the basis of the X-ray structural
coordinates, theoretical analyses using density functional theory (DFT) calculations at the B3LYP/6-
31+G(d,p)//B3LYP/6-31G(d) level were performed on spin-trapping reactions of superoxide radical with
4-HOPPN and PBN and three possible decay routes for their corresponding superoxide adducts. The
comparative calculations on the spin-trapping reactions with superoxide radical predicted that both spin
traps share an identical reaction type and have comparable potency when spin trapping superoxide radical.
Analysis of the optimized geometries of 4-HOPPN-OOH and PBN-OOH reveals that an introduction of
the phosphoryl group can efficiently stabilize the spin adduct through the intramolecular H-bonds, the
intramolecular nonbonding attractive interactions, as well as the bulky steric protection. Examination of
the decomposition thermodynamics of 4-HOPPN-OOH and PBN-OOH further supports the stabilizing
role of the phosphoryl group to a linear phosphorylated spin adduct.
Introduction
processes involved in chemical or biochemical environments.
In general, nitrone reacts with transient radical resulting in the
stable nitroxide radical that is detected by electron spin
resonance (ESR). Of all the available nitrones, R-phenyl-N-tert-
butylnitrone (PBN) and 5,5-dimethyl-1-pyrroline N-oxide (DMPO)
as representatives of the linear and cyclic nitrones, respectively,
have received the most attention. However, their uses are limited
due to the low stability of their superoxide spin adducts. The
Spin trapping as a valuable technique for detection of transient
radicals has been extensively utilized to investigate radical
* To whom correspondence should be addressed. (Y.L) Tel: +86 10
62571074. Fax: +86 10 62559373. (K.-J.L.) Tel: (505) 272-9546. Fax: (505)
272-6749.
† Chinese Academy of Sciences.
‡ Both authors contributed equally to this work.
§ University of New Mexico.
10.1021/jo061204m CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/02/2006
J. Org. Chem. 2006, 71, 7753-7762
7753

Liu et al.
CHART 1
SCHEME 1. Synthesis of 4-HOPPN
tion of superoxide radical (the half-life times of the adducts: 6
and 7 min, respectively).5d,8 Therefore, it is of significant impor-
tance to investigate the exact reason the â-phosphorylated spin
traps can enhance the efficiency of spin trapping of superoxide
radical, since this knowledge can open up the possibilities of
developing more efficient spin traps for superoxide radical.
In a previous study,9 we reported the first X-ray structure of
the phosphorylated cyclic nitrone, 5-diethoxyphosphoryl-5-
phenethyl-1-pyrroline N-oxide (DEPPEPO). Thereafter, the
X-ray structures of other cyclic nitrones such as 5-diisopropy-
loxyphosphoryl-5-methyl-1-pyrroline N-oxide (DIPPMPO),10
5-carbamoyl-5-methyl-1-pyrroline N-oxide (AMPO),11 and 5-
butoxycarbonyl-5-methyl-1-pyrroline N-oxide (BMPO)12 were
also experimentally determined. Using these X-ray structural
coordinates as the initial input data to program, a series of theo-
retical calculations11-13 have recently been performed to reason-
ably interpret the spin trapping properties of the cyclic nitrones
containing various substituents. The theoretical calculation has
shown potential for designing better cyclic spin traps,12 but little
theoretical attention has been paid to the â-phosphorylated linear
nitrones.
Herein, in the light of spin-trapping ESR behaviors and X-ray
structural coordination of a new phosphorylated linear nitrone,
[N-(4-hydroxybenzyliene)-1-diethoxyphosphoryl-1-methylethyl-
amine N-oxide (4-HOPPN), and those from PBN, two possible
routes for spin trapping of superoxide radical by 4-HOPPN and
PBN as well as three proper mechanisms for explaining the
decay of their superoxide spin adducts have been comparatively
investigated from the theoretical views. In addition, spin trapping
of various reactive radicals by 4-HOPPN was experimentally
evaluated.
half-life time of DMPO-OOH was under a minute at room
temperature,1 while that of PBN-OOH was even shorter.2
Considering that superoxide radical is a primary upstream
radical of the radical reaction chain inducing oxidative stress3
and is involved in a number of signal transduction pathways,4
many efforts have been devoted to developing new spin traps
for superoxide radical and structurally diverse nitrones have been
synthesized.5 Among them, 5-diethoxyphosphoryl-5-methyl-1-
pyrroline N-oxide (DEPMPO), one â-phosphorylated cyclic
nitrone, appears to be one of the most promising spin traps for
superoxide radical and other reactive oxygen species (ROS)
because of the considerably increased stability of its superoxide
spin adduct (DEPMPO-OOH, half-life time t1/2 ∼ 14 min).5a,6
Because of this, DEPMPO has been widely used in a variety
of biological systems.7 Moreover, for the purpose of obtaining
the lipophilic spin trap for ROS, the phosphoryl group was also
introduced into the â position of PBN to create a linear nitrone,
such as 4-PyOPN or PPN (Chart 1).5d,8 The experimental results
showed that these traps possess better properties of spin trapping
superoxide radical as compared with their nonphosphorylated
analogues. For example, relative to PBN, the phosphorylated
linear nitrones N-benzyliene-1-diethoxyphosphoryl-1-methyl-
ethylamine N-oxide (PPN) and 1-diethoxyphosphoryl-1-methyl-
N-[(1-oxidopyridin-1-ium-4-yl)methylidene] ethylamine N-oxide
(4-PyOPN) exhibited considerable improvements in the detec-
Results and Discussion
Experimental Analysis. Synthesis. As shown in Scheme 1,
condensation of 4-hydroxybenzaldehyde on diethyl (1-hydroxy-
amino-1-methylethyl) phosphonate14 led to 4-HOPPN, and then
the product was recrystallized in a mixture of n-hexane and ethyl
acetate after purification by column chromatography.
(1) Buettner, G. R.; Oberley, L. W. Biochim. Biophys. Res. Commun.
1978, 83, 69.
(2) Thomas, C. E.; Ohlweiler, D. F.; Carr, A. A.; Nieduzak, T. R.; Hay,
D. A.; Adams, G.; Vaz, R.; Bernotas, R. C. J. Biol. Chem. 1996, 271, 3097.
(3) (a) Halliwell, B. Am. J. Med. 1991, 91 (3C), 14S. (b) Halliwell, B.;
Gutteridge, J. M. C. Am. J. Med. 1984, 219, 1. (c) Halliwell, B.; Chirico,
S.; Am. J. Clin. Nutr. 1993, 57 (S), 715S.
(4) (a) Finkel, T. J. Leukocyte Biol. 1999, 65, 337. (b) Wolin, M. S.
Arterioscler. Thromb. Vasc. Biol. 2000, 20, 1430. (c) Droge, W. Physiol.
ReV. 2002, 82, 47.
(5) (a) Fre´javille, C.; Karoui, H.; Tuccio, B.; Le Moigne, F.; Culcasi,
M.; Pietri, S.; Lauricella, R.; Tordo, P. J. Chem. Soc., Chem. Commun.
1994, 1793. (b) Olive, G.; Mercier, A.; Le Moigne, F.; Rockenbauer, A.;
Tordo, P. Free Radic. Biol. Med. 2000, 8, 403. (c) Zhao, H.; Joseph, J.;
Karoui, H.; Kalyanaraman, B. Free Radic. Biol. Med. 2001, 31, 599. (d)
Zeghdaoui, A.; Tuccio, B.; Finet, J. P.; Cerri, V.; Tordo, P. J. Chem. Soc.,
Perkin Trans. 2 1995, 2087.
X-ray Structure. It is valuable to explore the X-ray structure
of 4-HOPPN since this experimental data can provide substantial
(9) Xu, Y. K.; Chen, Z. W.; Sun, J.; Liu, K.; Chen, W.; Shi, W.; Wang,
H. M.; Liu, Y. J. Org. Chem. 2002, 67, 7624.
(10) Chalier, F.; Tordo, P. J. Chem. Soc., Perkin Trans. 2 2002, 2110.
(11) Villamena, F. A.; Rockenbauer, A.; Gallucci, J.; Velayutham, M.;
Hadad, C. M.; Zweier, J. L. J. Org. Chem. 2004, 69, 7994.
(12) Villamena, F. A.; Hadad, C. M.; Zweier, J. L. J. Phys. Chem. A
2003, 107, 4407.
(6) Fre´javille, C.; Karoui, H.; Tuccio, B.; Le Moigne, F.; Culcasi, M.;
Pietri, S.; Lauricella, R.; Tordo, P. J. Med. Chem. 1995, 38, 258.
(7) (a) Shi, H.; Timmins, G.; Monske, M.; Burdick, A.; Kalyanaraman,
B.; Liu, Y.; Cle´ment, J. L.; Burchiel, S.; Liu, K. J. Arch. Biochem. Biophys.
2005, 437, 59. (b) Liu, K.; Sun, J.; Song, Y. G.; Liu, B.; Xu, Y. K.; Zhang,
S. X.; Tian, Q.; Liu, Y. Photosyn. Res. 2004, 81, 41. (c) Vasquez-Vivar, J.;
Kalyanaraman, B.; Martasek, P.; Hogg, N.; Masters, B. S. S.; Karoui, H.;
Tordo, P.; Pritchard, K. A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 9220.
(8) (a) Tuccio, B.; Zeghdaoui, A.; Finet, J. P.; Cerri, V.; Tordo, P. Res.
Chem. Intermed. 1996, 22, 393. (b) Roubaud, V.; Lauricella, R.; Tuccio,
B.; Bouteiller, J. C.; Tordo, P. Res. Chem. Intermed. 1996, 22, 405.
(13) (a) Villamena, F. A.; Hadad, C. M.; Zweier, J. L. J. Am. Chem.
Soc. 2004, 126, 1816. (b) Rosen, G. M.; Beselman, A.; Tsai, P.; Pou, S.;
Mailer, C.; Ichikama, K.; Robinson, B. H.; Nielsen, R.; Halpern, H. J.;
Mackerell, A. D. J. Org. Chem. 2004, 69, 1321. (c) Cle´ment, J. L.; Ferre´,
N.; Siri, D.; Karoui, H.; Rockenbauer, A.; Tordo, P. J. Org. Chem. 2004,
69, 1198. (d) Villamena, F. A.; Hadad, C. M.; Zweier, J. L. J. Phys. Chem.
A 2005, 109, 1662. (e) Villamena, F. A.; Merle, J. K.; Hadad, C. M.; Zweier,
J. L. J. Phys. Chem. A 2005, 109, 6083. (f) Villamena, F. A.; Merle, J. K.;
Hadad, C. M.; Zweier, J. L. J. Phys. Chem. A 2005, 109, 6089.
(14) Petrov, K. A.; Chauzov, V. A.; Pastukhova, L. V.; Bogdanov, N.
N. Zh. Obshch. Khim. 1976, 46, 1246.
7754 J. Org. Chem., Vol. 71, No. 20, 2006

Spin Trapping of Superoxide Radical
FIGURE 1. ORTEP view of X-ray structure of 4-HOPPN. Broken
lines indicate sites of intramolecular hydrogen bonds, and their distances
and angles are also given.
FIGURE 2. Representative ESR spectra of the spin adducts of
4-HOPPN: (A) the superoxide spin adduct (1) generated by UV irradi-
ation of the aqueous solution containing 4-HOPPN (45 mM), riboflavin
(0.1 mM), and DTPA (1 mM), (b), the signal derived from the electron
donor (DTPA), (*) the three-line signal resulting from a decomposition
product of 4-HOPPN; (B) the superoxide spin adduct produced by
addition of H2O2 (1%) to 4-HOPPN (50 mM) in pyridine; (C) the CH3O
spin adduct generated by adding a small amount of Pb(OAc)4 to a solu-
tion of 4-HOPPN (50 mM) and methanol (5 M) in DMSO; (D) the
p-ClPh• spin adduct obtained by UV photolysis of a small amount of
para-chlorophenyl diazonium tetrafluoraborate in the presence of
4-HOPPN (50 mM) in DMSO. Spectrometer settings: microwave
power ) 12.9 mW; modulation frequency ) 100 kHz; modulation
amplitude ) 0.03 mT; time constant ) 0.33 s; sweep time ) 163 s;
sweep width ) 12 mT.
structural information for the phosphorylated analogues of PBN.
The X-ray crystallographic structure is illustrated in Figures 1,
and the corresponding structural details are listed in the Sup-
porting Information as Table S1-S7. It can be seen from Fig-
ure 1 that 4-HOPPN contains two intramolecular H-bonds of
C(3)-H(3)‚‚‚O(2) (2.167 Å; 119.9°) and C(11)-H(11a)‚‚‚O(2)
(2.286 Å; 131.2°). Given that there exists only one intra-
molecular H-bond (2.180 Å; 119.9°) in a crystal of PBN15 and
no H-bond in a crystal of phosphorylated cyclic nitrone,
9
DEPPEPO, the presence of two intramolecular H-bonds is
probably a special characteristic of the â-phosphorylated ana-
logues of PBN. Additionally, an intermolecular H-bond of
O(1)-H(1)‚‚‚O(3) (1.860 Å; 174.4°) between two adjacent
4-HOPPN is also observed in Table S6 (Supporting Informa-
tion). Since PPN exists as a viscous oil at room temperature,8b
the intermolecular H-bond might be one of the important factors
to solidify 4-HOPPN. Analysis of the bond lengths on the
nitronyl group indicates that, as shown in Table S3 (Supporting
Information), the phosphorylated 4-HOPPN has the shorter NdC
bond distance of 1.302(3) Å relative to PBN (1.330(12) Å).15
Further examination of the X-ray structure suggests that when
the spin-trapping reaction occurs, due to the steric blocking
effect of the bulky phosphoryl group on C8, attacking radicals
should prefer to approach the opposite position relative to the
phosphoryl group.
be obtained by nuleophilic addition of H2O2 to 4-HOPPN in
pyridine, followed by oxidation of the subsequent hydroxyl-
amine (Figure 2B). As previously observed for the superoxide
spin adducts obtained with other nitrones,17 the lifetime of
4-HOPPN-OOH is significantly longer in pyridine, lasting more
than 20 h at room temperature.
The above results allow us to conclude that 4-HOPPN can
be used to detect superoxide radical. As a spin trap for super-
oxide radical, like PPN, one advantage of 4-HOPPN over the
commonly used nitrones, such as DMPO, EMPO, and DEP-
MPO, is that it is a solid such that its ease of purification makes
4-HOPPN less susceptible to contamination from paramagnetic
impurities.
(ii) Spin Trapping of Other Oxygen-, Carbon-, and Sulfur-
Centered Radicals with 4-HOPPN. To further evaluate the
potential of 4-HOPPN in the detection of short-lived radicals,
a series of oxygen-, carbon-, and sulfur-centered radicals were
generated and trapped in DMSO or aqueous solutions. Table 1
lists a summary of all the observed ESR spectral parameters.
Whatever the spin adduct considered, the ESR spectrum
recorded always consists of 12 lines, due to hyperfine splitting
constants (hfsc) with a nitrogen (IN-14 ) 1), a â-hydrogen (IH-1
) 1/2), and a large phosphorus coupling (IP-31 ) 1/2). Owing
to the presence of the additional phosphorus coupling, these
spin adducts give more information as compared with that of
PBN.
As reported previously, a nucleophilic addition of the alcohol
to the nitrone, assisted by Pb(OAc)4, could occur followed by
a rapid oxidation of the hydroxylamine to the corresponding
alkoxyl spin adduct.9,18 In the present study, when a large excess
of methanol (7.4 M) was added to a solution of 4-HOPPN in
DMSO in the presence of a small amount of Pb(OAc)4, a pure
Spin Trapping
(i) Spin Trapping of Superoxide Radical with 4-HOPPN.
When superoxide radical was produced using the light-
riboflavin-DTPA superoxide-generating system in the presence
of 4-HOPPN, an overlapped ESR spectrum, shown in Figure
2A, was observed. The main signal (1) is composed of 12-line
peaks attributable to a nitroxide that could be suppressed upon
the addition of SOD, and therefore, it is assigned to a superoxide
spin adduct of 4-HOPPN (4-HOPPN-OOH). Like PPN-OOH,5d,8
4-HOPPN-OOH affords the relatively small RH (Table 1).
Besides the superoxide spin adduct, there are also two additional
weak signals that are attributed to one signal (b) derived from
the electron donor (DTPA)16 and one three-line signal (*) result-
ing from a decomposition product of 4-HOPPN, respectively.5d,8
However, these paramagnetic species do not limit the use of
4-HOPPN in spin trapping of the superoxide radical, since they
do not interfere with the ESR signal of the superoxide spin
adduct. The pure signal corresponding to 4-HOPPN-OOH could
(17) (a) Reszka, K.; Bilski, P.; Chignell, C. F. Free Radical Res.
Commun. 1992, 17, 377. (b) Tuccio, B.; Lauricella, R.; Fre´javille, C.;
Bouteiller, J. C.; Tordo, P. J. Chem. Soc., Perkin Trans. 2 1995, 295.
(18) Karoui, H.; Nsanzumuhire, C.; Le Moigne, F.; Tordo, P. J. Org.
Chem. 1999, 64, 1471.
(15) Liu, Y.; Wang, X.; Chen, B.; Xu, G. J. Struct. Chem. 1987, 6, 17.
(16) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J. J. Am. Chem. Soc.
1980, 102, 4994.
J. Org. Chem, Vol. 71, No. 20, 2006 7755

Liu et al.
TABLE 1. ESR Hyperfine Splitting Constants of 4-HOPPN Spin Adductsa
adduct source
solvent
RP (mT)
RN (mT)
RH (mT)
HOO•
H2O2
riboflavin/hν
pyridine
water
3.86
3.88
3.89
3.89
3.91
3.88
3.89
3.90
3.91
3.93
3.92
3.93
3.93
3.92
3.89
3.91
3.90
3.87
4.20
1.31
1.32
1.34
1.34
1.33
1.31
1.33
1.33
1.34
1.34
1.34
1.34
1.34
1.34
1.31
1.34
1.32
1.33
1.40
0.16
0.18
0.28
0.29
0.27
0.28
0.21
0.26
0.28
0.28
0.28
0.27
0.28
0.27
0.26
0.28
0.27
0.27
0.29
CH3O•
CH3OH/Pb(OAC)4
C2H5OH/Pb(OAC)4
n-C4H9OH/Pb(OAC)4
(t-BuO)2/hν
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
C2H5O•
n-C4H9O•
t-BuO•
•
C2H5
C2H5I/hν
i-Bu•
i-Bu(O)COSnCy3/hν
Ph(O)COSnCy3/hν
m-BrPh(O)COSnCy3/hν
m-ClPh(O)COSnCy3/hν
m-FPh(O)COSnCy3/hν
(2,4,6-tri-ClPhOCH2)(O)COSnCy3/hν
R(O)COSnCy3/hν
Ph•
m-BrPh•
m-ClPh•
m-FPh•
•
2,4,6-tri-ClPhOCH2
R•
p-IPh•
p-IPh(O)COSnCy3/hν
p-MePh(O)COSnCy3/hν
p-NO2PhOSnCy3/hν
(PhCH2S)2/hν
p-MePh•
p-NO2PhO•
PhCH2S•
p-ClPh•
p-ClPhN2+BF4-/hν
a
ESR signal assigned to methoxyl spin adduct was observable
(Figure 2C). However, with a smaller amount of methanol (1.2
M), another weak signal was also detected besides the methoxyl
spin adduct signal (data not shown). The same results were also
observed in the case of the ethoxyl and n-butoxyl spin adducts.
A most likely explanation is that when only a relatively small
amount of the alcohol is present, the nitronyl group in one
molecule of 4-HOPPN would react with the phenolic group from
another one to yield the hydroxylamine that is subsequently
oxidized to the corresponding aminoxyl. Therefore, we suggest
that one should be careful to use the method of the nucleophilic
addition to produce spin adducts of the nitrones containing the
phenolic or hydroxyl group because the side product(s) might
occur.
Using the UV photolytic method,9 we obtained a variety of
the aromatic radical adducts (Table 1). When a solution contain-
ing 4-HOPPN and a small amount of p-chlorophenyl diazonium
tetrafluoraborate was irradiated, the strong ESR signal attributed
to the p-ClPh• spin adduct was observed (Figure 2D).
FIGURE 3. Decay plot of the superoxide spin adduct of 4-HOPPN
after 1 min of UV irradiation of the aqueous solution containing
riboflavin (0.1 mM), 4-HOPPN (45 mM), and DTPA (1 mM).
time (t1/2 ∼ 8.8 min) that was slightly longer than that of the
superoxide adduct of PPN.8b
Attempts to trap HO• in aqueous media, unfortunately, were
never successful regardless of the HO• generator. For example,
when the Fenton reaction (H2O2/Fe2+) was used to generate HO•
in the presence of 4-HOPPN, the ESR spectrum observed was
too complex to assign the presence of the hydroxyl spin adduct
(not shown). This may be closely associated with the com-
petition of the phenolic group with the nitronyl group for
HO•, since PPN that does not contain the phenolic group
(Chart 1) had been shown to be able to trap HO• in aqueous
Theoretical Analysis
Optimized Geometries. Selected bond distances for the
relevant moieties for the nitrone spin traps and the respective
spin adducts are listed in Table 2 based on the optimized
geometries at the B3LYP/6-31G(d) level of theory. As expected,
most of the bond distances, including the C-N and N-O bond
distances for 4-HOPPN or PBN, are consistent with their X-ray
structural data.15 Meanwhile, the optimized bond distances of
three superoxide spin adducts, including C-N, N-O, C-O2,
O-O, and O-H bonds, appear to be comparable to the crystal
data from 2,2,5,5-tetramethyl-3-carbamidopyrroline-l-oxyl (TM-
CO)19 and 3,4,4a,5,6,7,8,8a-octahydro-8a-hydroperoxy-2-quino-
lone,20 respectively.
5d,8a
media.
(iii) Kinetics of Decay of Superoxide Spin Adduct. The
light-riboflavin-DTPA system was utilized to investigate
kinetics of decay of the superoxide spin adduct of 4-HOPPN.
The ESR signal of the spin adduct persisted for as long as 35
min, as shown in Figure 3. Peak intensities were normalized
and plotted. As mentioned above, the presence of the two minor
signals does not interfere with the low field peak. Analysis of
data shows that the decay plot of the superoxide spin adduct is
well fitted with a pure first-order process. Based on the values
from the computer simulation, we could calculate the half-life
Thermodynamics of Spin Trapping of Superoxide Radical.
The pKa values for various alkyl hydroperoxides (R-OOH) that
(19) Turley, J. W.; Boer, F. P. Acta Crystallogr. 1972, B28, 1641.
(20) Alini, S.; Citterio, A.; Farina, A.; Fochi, M. C.; Malpezzi, L. Acta
Crystallogr. 1998, C54, 1000.
7756 J. Org. Chem., Vol. 71, No. 20, 2006

Spin Trapping of Superoxide Radical
TABLE 2. Comparison of Selected X-ray Crystallographic Bond
Lengths with Calculated Bond Lengths at the B3LYP/6-31G(d)
Level of Theory
overall endoergic with 6.61 and 14.48 kcal/mol, respectively,
indicating that formation of cis-isomer is more thermodynam-
ically favored. In the mechanism A, a large endoergic step (step
1) is first involved in the formation of the O2•- adducts (21.65
kcal/mol for the cis-isomer and 29.72 kcal/mol for the trans-
isomer), and then a subsequent protonation of the anionic adduct
(step 2) is high exoergic with ∆G values of -15.04 kcal/mol
for the cis-isomer and -15.24 kcal/mol for the trans-isomer.
Comparatively, ∆G value of the formation of HOO• (step 3) in
mechanism B is only endoergic with 6.57 kcal/mol, and the
subsequent trapping of HOO• (step 4) is slightly endothermic
with 0.04 kcal/mol for the formation of cis-isomer and 7.91
kcal/mol for the formation of trans-isomer. The comparison
evidently indicates that mechanism B is more thermodynam-
ically favorable for both isomers, cis-4-HOPPN-OOH and trans-
4-HOPPN-OOH, than mechanism A. Our result on 4-HOPPN
appears to be distinct from the mechanisms of spin trapping of
superoxide radical by the cyclic nitrones such as DMPO13e and
AMPO11 in which a direct addition of O2•- to the nitrones (like
mechanism A) is more thermodynamically favorable than the
addition of HOO• (like mechanism B). That two kinds of
nitrones (linear and cyclic) bear two distinct reaction mecha-
nisms is associated with the difference of their reactivity for
superoxide radical, resulting from their structural discrepancy.27
Once the spin-trapping reaction occurs, the π-system conjugation
formed between a benzene ring and a nitronyl group in a linear
PBN analogue, 4-HOPPN, would be diminished. Thus, it
appears that more energy is required for the linear nitrones in
their spin-trapping reactions. For example, the dihedral angle
for C4-C7-N-O is 179.5° in 4-HOPPN, almost coplanar,
but for two isomers of its superoxide spin adduct, cis-4-HOPPN-
OOH and trans-4-HOPPN-OOH, the angles change to 103.5°
and 112.1°, respectively. On the other hand, in the cyclic
nitrones, spin-trapping reaction leads to the steric decompression
in the five-membered ring associated with the change from a
trigonal to a tetrahedral carbon. The above results imply that
the cyclic nitrones have a stronger reactivity for superoxide
radical than the linear nitrones.
bond distance (Å)
bonds
calcd values
exptl values
CdN
1.31a
1.302(3)d
1.330(12)15
1.293(3)d
1.303(8)15
1.4714(19)d
1.5647(19)d
1.5608(19)d
1.832(2)d
1.5022
1.32b
N-O
1.29a
1.27b
PdO
P-OR
1.49a
1.62a
1.61a
P-C
1.88a
C-N
N-O
C-O2
O-O
O-H
1.48-1.53c
1.29-1.32c
1.39-1.43c
1.43-1.48c
0.99c
1.2722
1.44123
1.459923
1.0223
a Calculation from 4-HOPPN. b Calculation from PBN. c Calculations
from their superoxide spin adducts. d The crystal data from this work (4-
HOPPN).
were experimentally determined to be in the range of 11.65-
12.8 are normally smaller than that for water (15.75).21,22
Similarly, a recent theoretical study demonstrated that the pKa
value for the superoxide spin adduct of DMPO (DMPO-OOH)
is about 14.9 ( 0.5.13e Thus, it is the hydroperoxyl form of the
spin adduct (4-HOPPN-OOH) rather than its anionic form (4-
HOPPN-O2•-) that mainly exists in the neutral water.
As theoretically proposed, the main species trapped by cyclic
DMPO13e or AMPO11 at neutral media was superoxide anionic
radical (O2•-) rather than hydroperoxyl radical (HOO•) due to
the small pKa value of HOO• (approximately 4.823 or 4.424).
However, it was experimentally verified that HOO• as the acidic
•-
form of O2•- appears to be more reactive than O2
,
16,25 which
could be further confirmed by the higher reduction potential of
HOO• (1.06 V) relative to O2 (0.94 V). In neutral media,
to examine whether the low ratio of HOO• is possible to take
part in spin trapping of superoxide radicals with a linear
4-HOPPN, we comparatively calculated the thermodynamics
of two possible routes in which formation of the hydroperoxyl
spin adducts was performed by using HOO• or O2•- as an attack
species, respectively, as illustrated in the mechanisms A and B
of Scheme 2.
•-
26
To distinguish the influence of a p-hydroxyl group in the
4-HOPPN, we further conducted a calculation on PPN, a pure
phosphorylated analogue of PBN. Analysis of the optimized
geometry of PPN evidently finds that its main structural feature
is almost as same as that in 4-HOPPN, as listed in Table S8
(Supporting Information). For example, the CdN distance for
4-HOPPN (1.314 Å) is nearly identical with that for PPN (1.314
Å) and as for the N-O bond distance they have similar values
(1.290 Å for 4-HOPPN and 1.286 Å for PPN). Consequently,
it can be predicted that both 4-HOPPN and PPN have
comparable reactivity to trap superoxide radical, and this allows
us to evaluate an effect of phosphoryl group in the linear nitrones
on the spin trapping reactivity by a comparison between
4-HOPPN and PBN.
Because all of the spin trapping experiments were performed
in the aqueous media, we therefore evaluated the varieties of
free energy (∆G), both in spin-trapping reactions and in the
•-
decomposition of the O2 adducts, using the conductor-like
polarizable continuum medium (CPCM) model. The solvation
contributions to the ∆G values are demonstrated in Tables S9
and S10.
It is worth noting that although mechanisms A and B have
different sequences of steps, the overall ∆G are the same.
Formation of the cis- and trans-isomers of 4-HOPPN-OOH are
In contrast to 4-HOPPN-OOH, the superoxide spin adduct
of PBN (PBN-OOH) only provides one isomer due to the
absence of the phosphoryl group. On the basis of the most pre-
ferred conformation of PBN and its corresponding reaction
products, we have likewise calculated two mechanisms (mech-
anisms A and B) of the formation of PBN-OOH (Table 3). It is
clear that mechanism B is thermodynamically preferred for the
formation of PBN-OOH, which is the same as that of 4-HOPPN-
(21) Richardson, W. H.; Hodge, V. F. J. Org. Chem. 1970, 35, 4012.
(22) Gordon, A. J.; Ford, R. A., Eds. The Chemist’s Companion: A
Handbook of Practical Data, Techniques, and References; Wiley-Inter-
science: New York, 1972.
(23) Behar, D.; Czapski, G.; Rabani, J.; Dorfman, L. M.; Schwarz, H.
A. J. Phys. Chem. 1970, 74, 3209.
(24) Czapski, G.; Bielski, B. H. J. Phys. Chem. 1967, 67, 2180.
(25) (a) Tsai, P.; Elas, M.; Parasca, A. D.; Barth, E. D.; Mailer, C.;
Halpern, H. J.; Rosen, G. M. J. Chem. Soc., Perkin Trans. 2 2001, 875. (b)
Tsai, P.; Ichikama, K.; Mailer, C.; Pou, S.; Halpern, H. J.; Robinson, B.
H.; Nielsen, R.; Rosen, G. M. J. Org. Chem. 2003, 68, 7811.
(26) Buettner, G. R. Arch. Biochem. Biophys. 1993, 300, 535.
(27) Lauricella, R.; Allouch, A.; Roubaud, V.; Bouteiller, J. C.; Tuccio,
B. Org. Biomol. Chem. 2004, 2, 1304.
J. Org. Chem, Vol. 71, No. 20, 2006 7757

Liu et al.
SCHEME 2. Possible Mechanisms for the Formation of 4-HOPPN-OOH with Free Energies of Reaction (kcal/mol) at the
CPCM-B3LYP/6-31+G(d,p)//B3LYP/6-31G(d) Level
TABLE 3. Reaction Free Energies (kcal/mol) of Various Routes
for the Formation of 4-HOPPN-OOH and PBN-OOH at the
CPCM-B3LYP/6-31+G(d,p)//B3LYP/6-31G(d) Level
oxide spin adduct of cyclic DEPPEPO. With a goal of further
clarifying the stabilizing role of the superoxide spin adduct pro-
vided by the phosphoryl group of linear nitrone, we compara-
tively analyzed the optimized geometries of 4-HOPPN-OOH
and PBN-OOH.
reaction scheme
4-HOPPN-OOH
PBN-OOH
mechanism A
step 1
21.65 (cis)
29.72 (trans)
-15.24 (cis)
-16.52 (trans)
25.36
(i) Intramolecular H-Bond. Two optimized configurations,
i.e., cis- and trans-isomers, are assigned for 4-HOPPN-OOH
indicating the position of the diethoxyphosphoryl group relative
to the HOO group (Figure 4). Analysis of the optimized cis
and trans isomeric structures reveals that while one strong
intramolecular H-bond involving -P-O‚‚‚HOO- (1.785 Å) is
formed in the cis-isomer, the trans-isomer affords two intra-
molecular H-bonds involving -N-O‚‚‚HOO- (2.038 Å) and
-P-O‚‚‚H -C(7) (2.328 Å). According to the optimized
geometry of PBN-OOH, we can find a weak H-bond (2.147 Å)
on a -N-O‚‚‚HOO- bond motif as a consequence of the
absence of the phosphoryl group. This indicates that the presence
of the relatively stronger H-bonds in two isomers of 4-HOPPN-
OOH, to some extent, accounts for its higher stability as
compared with PBN-OOH. Therefore, we could predict that the
phosphoryl group may play a significant role in stabilization of
the superoxide spin adduct partly through participation in the
formation of the strong intramolecular H-bond. This is in accord
with a previous proposal in which the intramolecular H-bond
is formed in the superoxide spin adduct of the phosphorylated
cyclic nitrone DEPMPO and significantly stabilizes its molecular
structure.13d
(ii) Intramolecular Nonbonding Interaction. Besides the
intramolecular H-bond, interestingly, there exist many intra-
molecular nonbonding attractive interactions between the O atom
and the neighboring H atom by observations of the optimized
geometries of cis- and trans-4-HOPPN-OOH (Figure 4). All of
the distances between the O atom and the neighboring H that
vary from 2.464 to 2.946 Å are listed in Table 4 and are totally
within the range of nonbonding interaction as previously
demonstrated by Tsuzuki (from 2.591 to 2.963 Å).29 Some non-
bonding attractive interactions are also present in the optimized
geometry of PBN-OOH (Table 4), but when compared with two
isomers of 4-HOPPN-OOH, the interactions are evidently
weaker due to the absence of the phosphoryl group. As a result,
the intramolecular nonbonding attractive interaction involving
the O atoms and the H atoms of the phosphoryl group can be
expected to be another contributing factor toward the increased
stability of 4-HOPPN-OOH.
step 2
-20.86
mechanism B
step 3
6.57
6.57
step 4
0.04 (cis)
7.91 (trans)
-2.07
OOH. An overall variation of free energy in the spin trapping
of superoxide radical by PBN is 4.50 kcal/mol, which indicates
that formation of cis-4-HOPPN-OOH (∆G ) 6.61 kcal/mol) is
almost as feasible as that of PBN-OOH, but trans-4-HOPPN-
OOH does not (∆G ) 14.48 kcal/mol). Therefore, PBN nearly
has an identical thermodynamic tendency of spin trapping of
superoxide radical as compared with 4-HOPPN. In other words,
an introduction of the phosphoryl group seems not to influence
its thermodynamic tendency.
Structural Stability of the Superoxide Spin Adducts. It
can be experimentally determined that 4-HOPPN-OOH (t1/2
∼ 8.8 min) has a similar half-life time with PPN-OOH (t1/2
∼ 6 min)5a,d in aqueous media, which indicates that an intro-
duction of the hydroxyl group has no obvious effect on the
stability of these phosphorylated spin adducts. On the other hand,
the half-life time of PBN-OOH could not be determined because
of the very fast decay of the spin adduct.2 These observations
imply that the phosphoryl group may play an important role
in stabilizing superoxide spin adducts. Since the introduction
of the first cyclic phosphorylated spin trap DEPMPO, many
efforts have been made toward uncovering the role of the phos-
phoryl group. Tordo et al.28 suggested that the strong electron-
withdrawing effect and the large steric hindrance of the
phosphoryl group were responsible for the stabilities of the linear
and cyclic nitrone spin adducts. More recent theoretical data13a,d
also revealed that the intramolecular H-bond played an important
role in the stability of the superoxide and hydroxyl spin adducts
of DEPMPO. However, our previous X-ray structural data9
predicted that besides the steric exclusion and the electron-
withdrawing effect, the microenviroment around the OOH group
also was a significant contributing factor to stabilize the super-
(29) Tsuzuki, S.; Uchimaru, T.; Tanabe, K.; Hirano, T. J. Phys. Chem.
1993, 97, 1346.
(28) Tordo, P. Electron Paramagn. Reson. 1998, 16, 116.
7758 J. Org. Chem., Vol. 71, No. 20, 2006

Spin Trapping of Superoxide Radical
nitroxide radicals.30 Recently, in the attempts to develop better
spin traps for the superoxide radicals, Nohl et al. found that the
steric shielding of the spin-bearing nitroxyl group and the
peroxyl group was a crucial factor that governs stability of the
EMPO-derived superoxide spin adducts.31 Therefore, it is con-
ceivable that a similar steric effect may exist in 4-HOPPN-OOH.
As can be seen from Figure 4, the phosphoryl substituent in
the cis-HOPPN-OOH is held in close proximity to the hydro-
peroxyl and nitroxyl groups through an intramolecular H-bond
and some nonbonding attractive interactions. The steric proxim-
ity can be evidently demonstrated by comparison of some spatial
angles around C(8); i.e., the angle of N(1)-C(8)-P(1) (106.5°)
is smaller than the usual bond angle of sp3 hybrid orbital
(109.5°). It can be thus predicted that the large phosphoryl group
would contribute to a steric protection of two vulnerable groups,
-OOH and nitroxyl, thereby preventing or inhibiting the attack
from other molecules (i.e., water) in the solution and, conse-
quently, stabilizing the superoxide adduct. Similarly, in the
optimized geometry of trans-4-HOPPN-OOH, the H-bond
(2.328 Å) involving -PdO‚‚‚H(7) and some nonbonding attrac-
tive interactions (Table 4) also provide a steric protection of
the phosphoryl group on the hydroperoxyl and the nitroxide
moiety (the angle of N(1)-C(8)-P(1) is 106.5°). Compara-
tively, the steric protection provided by a methyl group in PBN-
OOH is much weaker in contrast to the phosphoryl group since
the steric proximity of the methyl group (the corresponding angle
for N(1)-C(8)-C(9) is 107.7°) to the nitroxyl group and the
-OOH is relatively weaker due to the absence of H-bond and
weaker nonbonding attractive interactions, as well as its rela-
tively less bulkiness than the phosphoryl group. As a result,
the steric protection via a bulky phosphoryl group seems to be
another important factor for increasing the stability of 4-HOPPN-
OOH.
Thermodynamics of the Decomposition of the Superoxide
Adducts. Although the exhaustive elucidation of the decay
mechanisms of the nitrone radical adducts, especially their super-
oxide spin adducts, is valuable in the development of better
spin traps, its theoretical study has not received much attention.
Until recently, only one study using DFT theory has shed light
on the thermodynamics of decay of DMPO-OOH and DEP-
MPO-OOH.13f To give further insight into the stability of
4-HOPPN-OOH and to provide new clues for stabilizing some
other linear or cyclic superoxide spin adducts, we herein con-
ducted theoretical calculation on the thermodynamics of the
4-HOPPN-OOH decay.
FIGURE 4. 4-HOPPN-OOH (cis), 4-HOPPN-OOH (trans), and PBN-
OOH in optimized structures at the B3LYP/6-31G(d) level of theory.
Broken lines indicate sites of intramolecular hydrogen bond, and their
distances are also given.
TABLE 4. Selected Nonbonding Distances (Å) and Bond Angles
(deg) in the Optimized Geometries of cis-4-HOPPN-OOH,
trans-4-HOPPN-OOH, and PBN-OOH at the B3LYP/6-31G(d) Level
of Theory
As experimentally observed,17b decay of the nitrone super-
oxide spin adducts usually included first- or second-order kinetic
process. However, the experimental ESR observation discussed
above has indicated that only first-order kinetics occurs during
the 4-HOPPN-OOH decay (Figure 2), which is consistent with
the decays of PPN-OOH and 4-PyOPN-OOH.8b Therefore, the
following section focuses mainly on the unimolecular decay of
4-HOPPN-OOH.
Although the actual first-order decay mechanism of the
superoxide spin adducts produced from the phosphorylated linear
nitrones have not received much attention, there is some indirect
evidence that could aid in elucidating the relevant reaction type.
O (2)-H(5)
2.701 O(3)-H(10a) 2.946
2.464 O(4)-H(9a) 2.787
2.605 O(5)-H(9b) 2.621
2.858 O(5)-H(10b) 2.697
O (2)-H(9a)
O (2)-H(11a)
O (2)-H(12a)
cis-4-HOPPN-OOH
N(1)-C(8)-P(1) 106.5
O (2)-H(2)
O (2)-H(9a)
O (2)-H(9b)
2.492 O(3)-H(11a) 2.568
2.643 O(3)-H(13a) 2.800
2.531 O(4)-H(9a) 2.706
2.945 O(5)-H(9c) 2.644
2.849 O(5)-H(10a) 2.720
trans-4-HOPPN-OOH O (2)-H(12a)
O (3)-H(5)
N(1)-C(8)-P(1) 106.5
O (1)-H(2)
2.697 O(1)-H(11a) 2.728
2.452
PBN-OOH
O (1)-H(10a)
N(1)-C(8)-C(9) 107.7
(30) Kocherginsky, N.; Swartz, H. M. Nitroxide Spin Labels. Reaction
in Biology and Chemistry; CRC Press: Boca Raton, New York, London,
Tokyo, 1995; p 27.
(31) (a) Stolze, K.; Udilova, N.; Rosenau, T.; Hofinger, A.; Nohl, H.
Biol. Chem. 2003, 384, 493. (b) Stolze, K.; Udilova, N.; Rosenau, T.;
Hofinger, A.; Nohl, H. Biochem. Pharmacol. 2003, 66, 1717.
(iii) Steric Protection via the Phosphoryl Group. Steric
protection via bulky groups on the adjacent ring carbons (usually
methyl groups but sometimes other alkyl groups) has been long
considered as one of the most important factors for stabilizing
J. Org. Chem, Vol. 71, No. 20, 2006 7759

Liu et al.
SCHEME 3. Possible Mechanism of Various Decomposition Pathways of 4-HOPPN-OOH and PBN-OOH with Free Energies
of Reaction (kcal/mol) at the CPCM-B3LYP/6-31+G(d,p)//B3LYP/6-31G(d) Level
For instance, Kotake and Janzen32 previously proposed that a
unimolecular reaction involving the C-N bond cleavage might
occur during the decay of the hydroxyl adduct of PBN. Other
evidence of this cleavage could be provided by analysis of the
decomposition products of the hydroxyl spin adduct of the linear
phosphorylated33 and esterated34 nitrones. More recent theoreti-
cal studies further demonstrated that the C-N bond-breaking
process played a crucial role in the stability of hydroxyl13a and
superoxide13f spin adducts for cyclic nitrones. Similarly, the
C-N bond-breaking process is probably involved in the
unimolecular decay of 4-HOPPN-OOH, a superoxide spin
adduct obtained from linear 4-HOPPN. On the other hand, a
previous study35 found that the peroxyl spin adducts of PBN
underwent decomposition via the C-Hâ bond cleavage, generat-
ing bezoyl tert-butyl nitroxide. The nitroxyl ketone was also
observable during the decomposition of PBN-OOH in the
presence of cyclodextrins.36 Further evidence revealed that the
cleavage might be labile during the decomposition of DMPO-
OOH.37 Hence, the C-Hâ bond cleavage should be considered
during the decay of the superoxide spin adduct; In addition,
the reduction of nitroxide to hydroxylamine took place when
both an electron and a H+ donor were present in the solution.30
The reduction route was much more labile in the biological
systems containing sulfhydryl reagents,38 ascorbic acid,39 mito-
chondrial electron transport chain,40 cytochrome P-450,41 and
bacterial electron transport systems.42
On the basis of the above analysis, we separately analyzed
the three possible routes for the unimolecular decomposition
of 4-HOPPN-OOH, as illustrated in Scheme 3. In the decay
process involving C-N bond cleavage (mechanism A), 4-
HOPPN-OOH decomposes through two steps. The hemolytic
cleavage of the hydroperoxyl O-O bond43 (step 1) produces a
diradical intermediate 2 (in triplet or singlet) and a hydroxyl
radical, and then the diradical 2 undergoes a C-N bond cleavage
to yield a nitroso 3 and a 4-hydroxybenzaldehyde 4 (step 2).
Attempts to locate the singlet structure of 2 suffered from failure,
but 3 and 4 were ultimately obtained through the triplet.
Although formation of the diradical 2 (in triplet) is endoergic
with the ∆G value of 23.47 kcal/mol (cis-isomer) and 15.60
kcal/mol (trans-isomer), subsequently, conversion of 2 to the
nitroso and the aldehyde provides a free energy of -52.00 kcal/
mol. As usually observed during the decay of some superoxide
spin adducts, the HO• thus generated may be trapped by the
nitrone to form a hydroxyl spin adduct.10,34a,44 Nevertheless, the
hydroxyl spin adduct of 4-HOPPN has not been experimentally
observed, possibly due to a H-abstraction from 4-HOPPN by
(32) Kotake, Y.; Janzen, E. G. J. Am. Chem. Soc. 1991, 113, 9503.
(33) (a) Rizzi, C.; Lauricella, R.; Tuccio, B.; Bouteiller, J. C.; Cerri, V.;
Tordo, P. J. Chem. Soc., Perkin Trans. 2 1997, 2507. (b) Allouch, A.;
Roubaud, V.; Lauricella, R.; Bouteiller, J. C.; Tuccio, B. Org. Biomol. Chem.
2003, 1, 593. (c) Janzen, E. G.; Kotake, Y.; Hinton, R. Free Radic. Biol.
Med. 1992, 12, 169. (d) Janzen, E. G.; Hinton, R.; Kotake, Y. Tetrahedron
Lett. 1992, 33, 1257.
(38) Gaffney, B. J. Spin Labeling Theory and Application; Academic
Press: New York, 1976; 183.
(39) Smith, I. C. P. Biological Applications of Electron Spin Resonance;
Weily-Interscience: New York, 1972; 483.
(34) Roubaud, V.; Lauricella, R.; Bouteiller, J. C.; Tuccio, B. Arch.
Biochem. Biophys. 2002, 397, 51.
(35) Niki, E.; Yokoi, S.; Tsuchiya, J.; Kamiya, Y. J. Am. Chem. Soc.
1983, 105, 1498.
(36) Karoui, H.; Tordo, P. Tetrahedron Lett. 2004, 45, 1043.
(37) (a) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J.; Paxton, J. Mol.
Pharm. 1979, 16, 676. (b) Briere, R.; Rassat, A. Tatrahedron 1976, 32,
2891. (c) Villamena, F. A.; Dickman, M. H.; Crist, D. R. Inorg. Chem.
1998, 37, 1446. (d) Villamena, F. A.; Dickman, M. H.; Crist, D. R. Inorg.
Chem. 1998, 37, 1454.
(40) Likhtenshtein, G. T. Spin Labeling Methods in Molecular Biology;
Weily-Interscience: New York, 1976; p 190.
(41) (a) Stier, A.; Sackman, E. Biochim. Biophys. Acta 1973, 311, 400.
(b) Rosen, G. M.; Rauckman, E. J. Biochem. Pharmacol. 1977, 26, 675.
(42) Goldberg, J. S.; Rauckman, E. J.; Rosen, G. M. Biochim. Biophys.
Res. Commun. 1977, 79, 198.
(43) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J. Mol. Pharm. 1982,
21, 262.
(44) Karoui, H.; Cle´ment, J. L.; Rockenbauer, A.; Siri, D.; Tordo, P.
Tetrahedron Lett. 2004, 45, 149.
7760 J. Org. Chem., Vol. 71, No. 20, 2006

Spin Trapping of Superoxide Radical
TABLE 5. Reaction Free Energies (kcal/mol) of Various
Decomposition Pathways for 4-HOPPN-OOH and PBN-OOH at the
CPCM-B3LYP/6-31+G(d,p)//B3LYP/6-31G(d) Level
of PBN-OOH and 4-HOPPN-OOH. In fact, hydroxylamines
derived from PBN spin adducts were detected in the in vivo
spin-trapping experiment.42
reaction scheme
4-HOPPN-OOH
PBN-OOH
mechanism A
step 1
Conclusions
23.47 (cis)
15.60 (trans)
-52.00
23.68
A new phosphorylated analogue of PBN, 4-HOPPN, has been
prepared, and its crystal structure has been determined by the
X-ray diffraction. Spin-trapping experiments demonstrate that
4-HOPPN can be used to trap various carbon-, oxygen-, and
sulfur-centered radicals. Like other phosphorylated linear ni-
trones, 4-HOPPN significantly stabilizes its superoxide spin
adduct (t1/2 ∼ 8.8 min). To elucidate the role of the phosphoryl
group, theoretical calculation using DFT were utilized to
investigate the mechanisms of spin trapping of superoxide
radical by 4-HOPPN and PBN, as well as to analyze the struc-
tural stability and the decay mechanisms of 4-HOPPN-OOH
and PBN-OOH. Our calculations show that both 4-HOPPN and
PBN share a common mechanism and have, thermodynamically,
a comparable potency for spin trapping of superoxide radical.
Examination of the optimized geometries of 4-HOPPN-OOH
and PBN-OOH reveals that there exist some stronger intra-
molecular interactions, including the H-bonds and the nonbond-
ing attractive interactions, as well as larger steric protection in
4-HOPPN-OOH, when compared to those in PBN-OOH. All
of the intramolecular interactions and the protections are prob-
ably responsible for stabilizing the superoxide spin adducts.
Finally, the higher stability of 4-HOPPN-OOH has reasonably
been verified by comparing its decomposition thermodynamics
using three various possible routes.
step 2
mechanism B
step 3
-50.53
-79.93
-79.30 (cis)
-86.90 (trans)
mechanism C
step 4
18.09 (cis)
7.03 (trans)
-125.97 (cis)
-124.29 (trans)
8.86
step 5
-121.98
HO•, as mentioned within the “spin trapping” in the Experi-
mental Section.
In the decomposition process via C-Hâ bond cleavage
(mechanism B), possibly induced by HO- or H2O, the C-Hâ
bond cleavage yields a nitroxyl-ketone 5 through the elimination
of a H2O molecule (step 3). The process is more favorable with
∆G values of -79.03 kcal/mol (cis-isomer) and -86.90 kcal/
mol (trans-isomer) relative to the step 1 of mechanism A.
The above theoretical analysis reveals that 4-HOPPN and
PBN possess comparable spin-trapping potency for superoxide
radical, but spin-trapping ESR experiments prove that only
4-HOPPN superoxide adduct is detectable in aqueous media.
To clarify the contradiction, we have to comparatively analyze
the decays of both spin adducts, 4-HOPPN-OOH and PBN-
OOH. All the free energies in each step of the PBN-OOH decay
were thermodynamically calculated using similar three reaction
models. As shown in Table 5, it is clear that, both in mechanisms
A and B, there is no significant difference in the overall or
stepwise energies of decay between cis-4-HOPPN-OOH and
PBN-OOH. However, a comparison of decay of the trans-4-
HOPPN-OOH with that of PBN-OOH reveals that the trans-
isomer appears to be thermodynamically unstable in two possible
processes, mechanisms A and B, in which the trans-4-HOPPN-
OOH decay is 9.55 and 7.17 kcal/mol more exoergic than the
PBN-OOH decay, respectively. That is to say, according to the
calculations based on two mechanisms, that neither trans-4-
HOPPN-OOH, nor cis-4-HOPPN-OOH, can decay slower than
PBN-OOH does. These thermodynamic calculations are obvi-
ously contradicted with their ESR observations when comparing
the distinct difference between the lifespan of PBN-OOH and
4-HOPPN-OOH, and thus, two mechanisms do not account for
the decays. Examination of the formation of the corresponding
hydroxylamine (mechanism C) shows that the H+-accepting
reaction (step 4) for cis-4-HOPPN-OOH is more endothermic
with ∆G of 18.09 kcal/mol as compared to the value of 8.83
kcal/mol for PBN-OOH. The subsequent decay via electron
transfer (step 5) is considerably exothermic, and their ∆G values
are approximately same for cis-4-HOPPN-OOH and PBN-OOH.
Accordingly, the rate-limiting step is probably the H+-accepting
reaction (step 4), and in mechanism C, PBN-OOH decay is more
thermodynamically favored than 4-HOPPN-OOH decay. More-
over, in the kinetic view, both the protonation process and the
subsequent electron transfer for 4-HOPPN-OOH are less favor-
able due to the large steric protection of the oxygen of the
nitroxyl group provided by the bulky phosphoryl group as
compared with PBN-OOH. Therefore, mechanism C is a
possible pathway that is responsible for the different lifespan
Experimental Section
Synthesis of N-(4-Hydroxybenzyliene)-1-diethoxyphosphoryl-
1-methylethylamine N-Oxide (4-HOPPN). To a solution of
4-hydroxybenzaldehyde (0.50 g, 4.5 mmol) in anhydrous THF
(30 mL) under nitrogen were added diethyl (1-hydroxyamino-1-
methylethyl)phosphonate (0.85 g, 4 mmol) and a small amount of
dry MgSO4, and the mixture was stirred for 3 h under reflux. The
reaction mixture was filtered and then concentrated in vacuo to
give a pale yellow solid which was purified by silica gel column
chromatography by using a solvent system of dichloromethane and
ethanol (30:1) and crystallized from the mixed solvent of n-hexane
and ethyl acetate. 4-HOPPN was obtained as a colorless crystal
(1.01 g, 80.2%). Mp: 138.5-139.5 °C. UV (C2H5OH; λmax/nm):
312.5. 1H NMR (CDCl3; ppm): δ 1.442 (6H, t), 1.890 (6H, d, J )
15.3 Hz), 4.278 (4H, m), 6.787 (2H, d, J ) 8.7 Hz), 7.408 (H, s),
7.910 (2H, d, J ) 8.6 Hz). 13C NMR (CDCl3; ppm): δ 15.41, 22.30,
62.92, 71.75, 114.71, 120.85, 130.38, 132.32, 158.70. 31P NMR
(CDCl3, H3PO4, external standard; ppm): δ 22.87. MS (ESI, m/z):
316.2 ([M + H]+, 100), 338.2 ([M + Na]+, 38.5), 354.1 ([M +
K]+, 8.2).
Spin Trapping Studies. (a) ESR Measurements. X-band ESR
spectra were recorded at room temperature. The spectrometer
settings were normally at the modulation amplitude of 0.1 mT,
modulation frequency of 100 kHz, microwave power of 12.9 mW,
and microwave frequency of 9.5 GHz. UV photolysis was per-
formed by a 200 W high-pressure mercury lamp.
(b) Spin Trapping of Superoxide Radical. The superoxide
adduct (4-HOPPN-OOH) was generated using two systems: (a)
riboflavin-DTPA-light system, an aqueous solution containing
riboflavin (0.1 mM), 4-HOPPN (45 mM) and DTPA (1 mM),
illuminated with UV light; (b) a pyridine solution of H2O2 (1%)
and 4-HOPPN (45 mM).
(c) Spin Trapping of Alkoxyl Radicals. The spin adducts of
methoxyl, ethoxyl, and n-butoxyl radicals were produced by adding
J. Org. Chem, Vol. 71, No. 20, 2006 7761

Liu et al.
a small amount of Pb(OAc)4 to a solution of 4-HOPPN (50 mM)
and the desired alcohol (7.4 M) in DMSO. tert-Butoxyl radical was
generated by UV photolysis of a solution of di-tert-butyl peroxide
(1 M) in the presence of 4-HOPPN (50 mM) in DMSO.
(d) Spin Trapping of Thiyl Radical. PhCH2S was generated
by UV photolysis of a solution of dibenzyl disulfide (1.5 M) and
4-HOPPN (50 mM) in DMSO.
(e) Spin Trapping of Other Radicals. Other radicals were
generated by UV photolysis of a solution of corresponding sources
(1 M) in the presence of 4-HOPPN (50 mM) in DMSO.
(f) Kinetics of Decay of Superoxide Spin Adduct. In a typical
decay kinetic study, an aqueous solution containing riboflavin (0.1
mM), 4-HOPPN (45 mM), and DTPA (1 mM) was irradiated for
1 min. The spin adduct decay was followed by monitoring the
decrease of the lowest field peak. Computer simulations were
performed by use of Origin 7.0. In these calculations, the monitored
ESR peak intensity is related to the actual radical concentration by
a scale factor.
(CPCM) model.45 All quantum chemical calculations were per-
formed with Gaussian 98.46
Structural Determination. Single crystals of 4-HOPPN were
produced by recrystallization followed by a crystal-growing process.
The recrystallization involved heating the nitrone and proper amount
of the hexane. Ethyl acetate was then slowly added to the mixture
until dissolved. The resulting solution was put into a crystal-growing
bottle while still hot. Hexane vapor was used to slowly diffuse into
the solution until a perfect crystal was produced. Data for 4-HOPPN
was collected on a Rigaku R-axis Rapid IP diffractometer with a
graphite monochromator. The structure was solved by direct
methods and refined by full-matrix least-squares methods on F2.
All calculations were performed by SHELX97.47 Full details of
the crystallographic data for 4-HOPPN are described in the
Supporting Information.
Acknowledgment. The investigation was supported by
NSFC (No. 20473098) and the Outstanding Overseas Chinese
Scholars Fund of Chinese Academy of Sciences (2005-1-12).
Computational Methods. Density functional theory (DFT) was
applied in this study to determine the optimized geometry, vibra-
tional frequencies, and single-point energy of all stationary points.
The elaborate calculation procedures are the following. The
molecular geometries were optimized first by molecular mechanic
method MMX, and then, by B3LYP functional on basis set of
6-31G(d). All the stable structures without virtual vibrational
frequencies were gained. The zero point vibrational energy (ZPVE)
and the vibrational contribution to the enthalpy and entropy were
scaled by a factor of 0.9806.12 Finally, the single-point energies
were gained using B3LYP/6-31+G(d,p). The solvent effect was
also considered by employing the self-consistent reaction field
(SCRF) method with conductor-like polarizable continuum medium
Supporting Information Available: X-ray crystallographic data
(CIF). Energies, enthalpies, and free energies for the spin traps,
their superoxide adducts, and the decompostion products of the
superoxide adducts. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO061204M
(45) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995.
(46) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; et al. Gaussian 98,
revision A.11; Gaussian, Inc.: Pittsburgh, PA, 2001.
(47) (a) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467. (b) Sheldrick,
G. M. Program for the Refinement of Crystal Structures, Universtity of
Go¨ttingen.
7762 J. Org. Chem., Vol. 71, No. 20, 2006