Spin trapping by 5-carbamoyl-5-methyl-1-pyrroline N-oxide (AMPO): Theoretical and experimental studies

Article abstract of DOI:10.1021/jo049244i

The nitrone 5-carbamoyl-5-methyl-1-pyrroline N-oxide (AMPO) was synthesized and characterized. Spin trapping of various radicals by AMPO was demonstrated for the first time by electron paramagnetic resonance (EPR) spectroscopy. The resulting spin adducts for each of these radicals gave unique spectral profiles. The hyperfine splitting constants for the superoxide adduct are as follows: isomer I (80%), α_nitronyl-N = 13.0 G and α _β-H = 10.8 G; isomer II (20%), α_nitronyl-N = 13.1 G, α_β-H = 12.5 G, and α_γ-H = 1.75 G. The half-life of the AMPO-O₂H was about 8 min, similar to that observed for EMPO but significantly shorter than that of the DEPMPO-O ₂H with t_1/2 ～16 min. However, the spectral profile of AMPO-O₂H at high S/N ratio is distinguishable from the spectrum of the ^?OH adduct. Theoretical analyses using density functional theory calculations at the B3LYP/ 6-31+G**//B3LYP/6-31G* level were performed on AMPO and its corresponding superoxide adduct. Calculations predicted the presence of intramolecular H-bonding in both AMPO and its superoxide adduct. The H-bonding interaction was further confirmed by an X-ray structure of AMPO, and of the novel and analogous amido nitrone 2-amino-5-carbamoyl-5-methyl-1-pyrroline N-oxide (NH₂-AMPO). The thermodynamic quantities for superoxide radical trapping by various nitrones have been found to predict favorable formation of certain isomers. The measured partition coefficient in an n-octanol/buffer system of AMPO was similar to those of DMPO and DEPMPO. This study demonstrates the suitability of the AMPO nitrone for use as a spin trap to study radical production in aqueous systems.

Full text of DOI:10.1021/jo049244i

Spin Trapping by 5-Carbamoyl-5-methyl-1-pyrroline N-Oxide
(AMPO): Theoretical and Experimental Studies
Frederick A. Villamena,*,† Antal Rockenbauer,§ Judith Gallucci,‡ Murugesan Velayutham,†
Christopher M. Hadad,*,‡ and Jay L. Zweier*,†
Center for Biomedical EPR Spectroscopy and Imaging, The Davis Heart and Lung Research Institute,
and the Division of Cardiovascular Medicine, Department of Internal Medicine, College of Medicine, and
the Department of Chemistry, The Ohio State University, Columbus, Ohio 43210, and Chemical Research
Center, Institute of Chemistry, H-1025 Budapest, Pusztaszeri 59, Hungary
villamena-1@medctr.osu.edu; hadad.1@osu.edu; zweier-1@medctr.osu.edu
Received May 4, 2004
The nitrone 5-carbamoyl-5-methyl-1-pyrroline N-oxide (AMPO) was synthesized and characterized.
Spin trapping of various radicals by AMPO was demonstrated for the first time by electron
paramagnetic resonance (EPR) spectroscopy. The resulting spin adducts for each of these radicals
gave unique spectral profiles. The hyperfine splitting constants for the superoxide adduct are as
follows: isomer I (80%), anitronyl-N ) 13.0 G and aâ-H ) 10.8 G; isomer II (20%), anitronyl-N ) 13.1 G,
aâ-H ) 12.5 G, and aγ-H ) 1.75 G. The half-life of the AMPO-O2H was about 8 min, similar to that
observed for EMPO but significantly shorter than that of the DEPMPO-O2H with t1/2 ∼16 min.
However, the spectral profile of AMPO-O2H at high S/N ratio is distinguishable from the spectrum
of the •OH adduct. Theoretical analyses using density functional theory calculations at the B3LYP/
6-31+G**//B3LYP/6-31G* level were performed on AMPO and its corresponding superoxide adduct.
Calculations predicted the presence of intramolecular H-bonding in both AMPO and its superoxide
adduct. The H-bonding interaction was further confirmed by an X-ray structure of AMPO, and of
the novel and analogous amido nitrone 2-amino-5-carbamoyl-5-methyl-1-pyrroline N-oxide (NH2-
AMPO). The thermodynamic quantities for superoxide radical trapping by various nitrones have
been found to predict favorable formation of certain isomers. The measured partition coefficient in
an n-octanol/buffer system of AMPO was similar to those of DMPO and DEPMPO. This study
demonstrates the suitability of the AMPO nitrone for use as a spin trap to study radical production
in aqueous systems.
Introduction
Several PBN-type nitrones have been synthesized such
as the R-substituted methoxy, amino, cyano, and mercap-
to nitrones,4,5 as well as the alkoxy-phosphoryl6 and -car-
boxyl7 derivatives, but the DMPO-type analogues are by
far the most commonly used. These DMPO-type spin
traps include the alkoxyphosphorylated nitrones 5-dieth-
oxyphosphoryl-5-methyl-1-pyrroline N-oxide (DEPMPO)8-10
and 5-diisopropyloxyphosphoryl-5-methyl-1-pyrroline N-
oxide (DIPPMPO),11 and the alkoxycarbonyl nitrones 5-
ethoxycarbonyl-5-methyl-1-pyrroline N-oxide (EMPO)12-15
and 5-butoxycarbonyl-5-methyl-1-pyrroline N-oxide (Boc-
MPO).15-18 These substituted DMPO-type nitrones have
Spin trap development has been a major focus in free
radical research and is of great importance for the mea-
surement of free radicals in chemical and biological sys-
tems. Spin trapping by electron paramagnetic resonance
(EPR) spectroscopy has been widely employed to detect
radical adducts with high sensitivity and specificity.1-3
There are generally two families of spin traps, the linear
or PBN-type and the cyclic or DMPO-type nitrones.
(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) Acken, B. J.; Gallis, D. E.; Warshaw, J. A.; Crist, D. R. Can. J.
Chem. 1992, 70, 2076-2080.
(5) Gallis, D. E.; Warshaw, J. A.; Acken, B. J.; Crist, D. R. Collect.
Czech. Chem. Commun. 1993, 59, 125-141.
(6) Zeghdaoui, A.; Tuccio, B.; Finet, J.-P.; Cerri, V.; Tordo, P. J.
Chem. Soc., Perkin Trans. 2 1995, 12, 2087-2089.
(7) Allouch, A.; Roubaud, V.; Lauricella, R.; Bouteiller, J.-C.; Tuccio,
B. Org. Biomol. Chem. 2003, 1, 593-598.
(8) Frejaville, C.; Karoui, H.; Tuccio, B.; Le Moigne, F.; Culcasi, M.;
Pietri, S.; Lauricella, R.; Tordo, P. J. Med. Chem. 1995, 38, 258-265.
(9) Liu, K. J.; Miyake, M.; Panz, T.; Swartz, H. Free Radical Biol.
Med. 1999, 26, 714-721.
(10) Stolze, K.; Udilova, N.; Nohl, H. Free Radical Biol. Med. 2000,
29, 1005-1014.
* Address correspondence to these authors. F.A.V.: fax (614)-292-
8454. C.M.H.: fax (614)-292-1685. J.L.Z.: fax: (614)-247-7799.
† Center for Biomedical EPR Spectroscopy and Imaging, The Davis
Heart and Lung Research Institute, College of Medicine, The Ohio
State University.
§ Chemical Research Center, Institute of Chemistry.
‡ Department of Chemistry, The Ohio State University.
(1) Rosen, G. M.; Cohen, M. S.; Britigan, B. E.; Pou, S. Free Radical
Res. Commun. 1990, 9, 187-195.
(2) Rhodes, C. J. Toxicology of the Human Environment; Taylor and
Francis Ltd.: London, UK, 2000.
10.1021/jo049244i CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/19/2004
7994
J. Org. Chem. 2004, 69, 7994-8004

5-Carbamoyl-5-methyl-1-pyrroline N-Oxide
•-
demonstrated a relatively higher rate of superoxide (O2
)
now report the spin trapping and spin adduct character-
istics of AMPO.
trapping and the corresponding O2•- adducts have longer
half-lives as compared to that of DMPO. Although the
spin trapping capabilities of both the PBN and DMPO-
type nitrones have been demonstrated, they have certain
critical limitations. PBN-type nitrones are limited by
their ability to discern different radicals, while the
DMPO-type nitrones are limited by the stability of the
adducts formed and the overall spin trapping efficiency.
Results and Discussion
Theoretical Analysis. Figure S19 shows the naming
system used in this study while Table S15 gives the
pertinent dihedral angles in each •O2H adduct. The N-O
and CdN bond distances for all nitrones are in the range
of 1.25-1.27 and 1.30-1.31 Å, respectively, consistent
with previously reported experimental data of 1.294(1)22
and 1.307(2) Å23 for N-O and CdN bonds, respectively.
The X-ray structure of AMPO shows bond distances of
1.310(2) Å for N-O and 1.286(2) Å for the CdN bond.
The preferred B3LYP/6-31G* geometry for AMPO is the
Newman projection shown here and in Figure 1 (top) in
In our previous studies, we explored how theoretical
analysis can provide insights into the spin trapping
properties and spin adduct stability of nitrones and their
corresponding nitroxyl spin adducts, respectively.16,19 On
the basis of our recent theoretical studies with a density
functional theory approach, we showed that 5-carbamoyl-
5-methyl-1-pyrroline N-oxide (AMPO)20 and its spin
adduct (AMPO-OH) have similar electronic and ther-
modynamic properties compared to the phosphorylated
and carboxylated nitrones and their adducts. This com-
parison, therefore, motivated us to explore the spin
trapping characteristics of AMPO. The compound AMPO
which the carbonyl oxygen is oriented opposite to the
N-O direction. Optimization attempts with different
initial starting conformations only gave the same con-
formation. The calculated dihedral angle for CH3-C-
CNH2-O was θ ) 99.4° and for NH2-AMPO C6-C1-
C5-O2 ) 90.1(2)°. The range of nitroxyl N-O bond
distances is 1.28-1.34 Å, comparable to the experimental
value of 1.267 (5) Å23 for 2,2,5,5-tetramethyl-3-carbami-
dopyrroline-l-oxyl (TMCO).
Natural population analysis revealed that spin densi-
ties on the isolated HOO• radical shows 73% distribution
on the terminal O while 27% on the internal O. The aver-
age spin densities within the N-O moiety in the radical
adducts are 39% and 56% on the N and O atoms,
respectively. The O-O bond distance in the free HOO•
radical is shorter, i.e., 1.33 Å, compared to the O-O bond
distance in the •O2H adducts, which has a mean average
of 1.455 (2) Å, similar to that observed in 1 of 1.4599(17)
Å.24
has been previously synthesized21 as a model system to
study the oxidation of hydroxyamino-amide, but it has
not been evaluated as a spin trap. For the first time, we
(11) Chalier, F.; Tordo, P. J. Chem. Soc., Perkin Trans. 2 2002,
2110-2117.
(12) Olive, G.; Mercier, A.; Le Moigne, F.; Rockenbauer, A.; Tordo,
P. Free Radical Biol. Med. 2000, 28, 403-408.
(13) Zhang, H.; Joseph, J.; Vasquez-Vivar, J.; Karoui, H.; Nsan-
zumuhire, C.; Martasek, P.; Tordo, P.; Kalyanaraman, B. FEBS Lett.
2000, 473, 58-62.
(14) Stolze, K.; Udilova, N.; Nohl, H. Biol. Chem. 2002, 383, 813-
820.
(15) Stolze, K.; Udilova, N.; Rosenau, T.; Hofinger, A.; Nohl, H. Biol.
Chem. 2003, 384, 493-500.
(16) Villamena, F.; Zweier, J. J. Chem. Soc., Perkin Trans. 2 2002,
1340-1344.
(17) Zhao, H.; Joseph, J.; Zhang, H.; Karoui, H.; Kalyanaraman, B.
Free Radical Biol. Med. 2001, 31, 599-606.
(18) Tsai, P.; Ichikawa, K.; Mailer, C.; Pou, S.; Halpern, H. J.;
Robinson, B. H.; Nielsen, R.; Rosen, G. M. J. Org. Chem. 2003, 68,
7811-7817.
(19) Villamena, F.; Hadad, C. M.; Zweier, J. J. Phys. Chem. A 2003,
107, 4407-4414.
Since the pKa of HOO• is approximately 4.8,25 it could
(20) Villamena, F.; Hadad, C. M.; Zweier, J. J. Am. Chem. Soc. 2004,
126, 1816-1829.
•-
certainly be assumed that the attacking species is O2
(21) Alderson, G. W.; Black, D. S.; Clark, V. M.; Todd, L. J. Chem.
Soc., Perkin Trans. 1 1976, 1955-1960.
at neutral pH.26,27 Therefore, we explored the most
J. Org. Chem, Vol. 69, No. 23, 2004 7995

Villamena et al.
SCHEME 1. Possible Mechanisms for Formation of the AMPO-O2H with Free Energies of Reaction
∆GRxn,298K (kcal/mol) in the Gas Phase at the B3LYP/6-31+G**//B3LYP/6-31G* Levela
a Values in parentheses are in the aqueous phase at the B3LYP/6-31+G**(PCM)//B3LYP/6-31G* level of theory.
favorable route for superoxide adduct formation using
computational methods as shown in Scheme 1 with the
corresponding reaction free energies (∆G298K,rxn in
kcal/mol) for each step both in gaseous and aqueous
phases (using the PCM model for water). Several con-
formational as well as configurational isomers have been
considered in the approximation of the energetics of each
of the reaction steps, and the ∆G298K,rxn values only reflect
the energy difference for the most thermodynamically
favored isomer of reactants and products (see Supporting
Information Tables S19, S20, and S22 for the structures
and thermodynamic parameters of these isomers).
•-
Once O2 is formed in solution, it could add directly
to the nitrone (step 1), undergo protonation (step 3), or
abstract a proton from an amide group (step 5). Steps 3
and 5 gave endoergic gas-phase ∆Grxn values of 39.07 and
5.93 kcal/mol compared to an exoergic value of -16.75
kcal/mol for step 1. Mechanism B should only occur at
acidic pH given that the pKa of water and RCONH2 are
(22) Villamena, F. A.; Dickman, M. H.; Crist, D. R. Inorg. Chem.
1998, 37, 1446-1453.
(25) Bielski, B. H. J. Photochem. Photobiol. 1978, 28, 645-649.
(26) Halliwell, B.; Gutteridge, J. M. C. In Free Radicals in Biology
and Medicine; Oxford Univeristy Press: Oxford, UK, 1999; pp 60-61.
(27) Finkelstein, E.; Rosen, G. M.; Rauckman, E. J. J. Am. Chem.
Soc. 1980, 102, 4995.
(23) Turley, J. W.; Boer, F. P. Acta Crystallogr. 1972, B28, 1641-
1644.
(24) Alini, S.; Citterio, A.; Farina, A.; Fochi, M. C.; Malpezzi, L. Acta
Crystallogr. 1998, C54, 1000-1003.
7996 J. Org. Chem., Vol. 69, No. 23, 2004

5-Carbamoyl-5-methyl-1-pyrroline N-Oxide
about 15.75 and 25,28 respectively, and that HOO• is
relatively acidic at 25 °C. Mechanism A, therefore, seems
to be the most plausible route for superoxide adduct
formation, in which O2•- is predominant in solution and
its addition to AMPO yields the AMPO-O2-, which is
subsequently protonated to yield the hydroperoxyl ad-
duct, AMPO-O2H. Although the proton abstraction
processes in steps 2 and 3 are both endoergic in the gas
phase (43.81 and 39.07 kcal/mol, respectively), formation
TABLE 1. Reaction Enthalpies ∆Hrxn (in kcal/mol) for
the Formation of the Hydroperoxyl Adductsa at the
B3LYP/6-31+G**//B3LYP/6-31G* Level at 298 K
spin adducts
HMPO
HMPO-O2H
DMPO
∆Hrxn
spin adducts
∆Hrxn
EMPO-O2H trans-1
-24.16
-23.24
-26.23
-23.67 EMPO-O2H trans-2
EMPO-O2H trans-3
-24.95 DEPMPO-1
DEPMPO-O2H cis-1
DMPO-O2H
AMPO
-26.70
-26.56
-23.09
AMPO-O2H cis-2a -24.18b DEPMPO-O2H cis-2
AMPO-O2H cis-2b -20.40 DEPMPO-O2H cis-3
-
of AMPO-O2H from the AMPO-O2 and H2O (step 2)
AMPO-O2H cis-3
-17.81 DEPMPO-O2H cis-3H -24.36
may be possible in H2O due to more favorable thermo-
dynamic parameters when compared to the direct forma-
tion of HOO• from O2•- and H2O (step 3). This is further
supported by the experimental pKa reported for various
alkyl hydroperoxides in water at 25 °C, which is in the
range of 11.65-12.8.29 Moreover, experimental evidence
AMPO-O2H trans-2 -23.55 DEPMPO-O2H trans-1 -26.08
EMPO-1
DEPMPO-O2H trans-2 -27.05b
-26.07b DEPMPO-O2H trans-3 -24.87
-21.68 DEPMPO-2
EMPO-O2H cis-1
EMPO-O2H cis-2
EMPO-O2H cis-3
-25.16 DEPMPO-O2H cis-1
-28.57
-28.42
-24.96
EMPO-O2H trans-1 -23.91 DEPMPO-O2H cis-2
EMPO-O2H trans-2 -22.99 DEPMPO-O2H cis-3
•-
shows that the addition of O2 to 5-tert-butoxycarboxy-
EMPO-O2H trans-3 -25.97 DEPMPO-O2H cis-3H -26.22
5-methyl-1-pyrroline N-oxide at various pH values has
been reported to be relatively slower at pH 7.0 with kapp
) 75.0 M-1 s-1 compared to kapp ) 239.2 M-1 s-1 at pH
5.0.18 This difference in rate constants at various pH
EMPO-2
DEPMPO-O2H trans-1 -27.95
-25.88b DEPMPO-O2H trans-2 -28.91b
-21.48 DEPMPO-O2H trans-3 -26.74
-24.97 DEPMPO-3
EMPO-O2H cis-1
EMPO-O2H cis-2
EMPO-O2H cis-3
•-
EMPO-O2H trans-1 -23.71 DEPMPO-O2H cis-1
EMPO-O2H trans-2 -22.80 DEPMPO-O2H cis-2
EMPO-O2H trans-3 -25.78 DEPMPO-O2H cis-3
-24.57
-24.43
-20.96
values further supports that O2 dominates in neutral
pH while HOO• is formed at acidic pH, and correlates
well with the experimental reduction potentials of E° )
1.06 and 0.94 V for HOO• and O2•-, respectively, dem-
onstrating their relative reactivity.30 Mechanism D shows
the possibility of intramolecular proton abstraction of the
amide proton by the superoxide moiety. Although step 7
is exoergic, ∆Grxn ) -12.19 kcal/mol in the aqueous
phase, typical pKa values for RCONH2 of 2528 and ROOH
of 11.65-12.829 would indicate that the equilibrium
should favor reactants over the products; however, in this
case, intramolecular hydrogen bonding biases the system
to strongly favor the products.
EMPO-3
DEPMPO-O2H cis-3H -22.22
-26.32b DEPMPO-O2H trans-1 -23.95
-21.93 DEPMPO-O2H trans-2 -24.92b
-25.42 DEPMPO-O2H trans-3 -22.74
EMPO-O2H cis-1
EMPO-O2H cis-2
EMPO-O2H cis-3
a See Figures S19 and Table S15 for the corresponding structure
of the naming system used. b Most stable conformer.
ence in the formation of cis adducts compared to forma-
tion of trans adducts. It is predicted that there are a
lesser number of conformational and configurational
isomers of AMPO-O2H (Table 1) compared to EMPO-
O2H and DEPMPO-O2H. However, assuming that the
g values of individual isomers are the same, this predic-
tion on the number of isomers was not observed as the
experimental line widths among the O2•- adducts did not
show any significant differences, i.e., EMPO-O2H (2.3
G) and DEPMPO-O2H (1.8 G) and AMPO-O2H (2.3 G).
Table 1 shows the enthalpy changes for the formation
of various •O2H adducts from the corresponding nitrones
and HOO•. Examination of the thermodynamic quantities
for trapping HOO• by various nitrones reveals that
enthalpies of reaction are less exothermic, i.e., ∆Hrxn
∼
18-29 kcal/mol exothermic (see Table 1), compared to
•-
•
Consistent with the EPR spectra for O2 adducts of
the trapping of OH, which was predicted to be on the
EMPO12 and DEPMPO8,31 (Figure 2), the lowest field
peak is broadened, possibly due to the presence of
order of ∆Hrxn ∼ 51-58 kcal/mol exothermic.19,20 The most
energetically preferred isomers are AMPO-O2H cis-2a,
EMPO-O2H cis-1, and DEPMPO-O2H trans-2. Analysis
of all these optimized structures predicted the presence
of intramolecular H-bonding, i.e., -OOH- - -O-N in all
adducts and an additional -N-O- - -H-N bonding mode
for AMPO-O2H cis-2a (see Figure 1). Interestingly, the
formation of DEPMPO-O2H cis-3H with intramolecular
H-bonding involving -PdO- - -H-OO- was predicted to
be 2.7 kcal/mol less exothermic than DEPMPO-O2H
trans-2, contrary to the DEPMPO-OHs reported previ-
ously in which the presence of -PdO- - -H-O- in
DEPMPO-OH cis stabilizes the adduct by 2.4 kcal/mol
compared to DEPMPO-OH trans-2 adduct.20
•-
diastereomeric mixtures of O2 adducts.
Four isomers were predicted for AMPO-O2H as shown
in Figure 1 with different modes of intramolecular
H-bonding, i.e., -N-O- - -H-N, -OOH- - -OdC, and
-N-O- - -HOO-. The AMPO-O2H cis-3 adduct has the
shortest intramolecular H-bond distance of 1.90 Å on an
OOH- - -OdC- bond motif, but it is not the most ther-
modynamically preferred isomer. A closer look at the spin
density distribution on various AMPO-O2H using natu-
ral population analysis indicate a 0.1-0.3% delocalization
of spin on the amido-N compared to about 1.9-2.5%
delocalization on the phosphoryl-P in the DEPMPO-O2H
(Table S16). In general, at least 1 γ-H in all adducts
exhibit 0.1% spin density distribution while 0.8-2.0%
was predicted for the â-H’s. Values of isotropic hyperfine
splitting constant aX for the various adducts can be
calculated from the nuclear spin density Frx (Fermi
On the basis of Table 1, it is apparent that the enthalpy
changes for the formation of AMPO-O2H, EMPO-O2H,
and DEPMPO-O2H show less than 0.1 kcal/mol differ-
(28) Gordon, A. J.; Ford, R. A., Eds. The Chemist’s Companion: A
Handbook of Practical Data, Techniques, and References; Wiley-
Interscience: New York, 1972.
(29) Richardson, W. H.; Hodge, V. F. J. Org. Chem. 1970, 35, 4012-
4016.
(30) Buettner, G. R. Arch. Biochem. Biophys. 1993, 300, 535-543.
(31) Barbati, S.; Clement, J. L.; Olive, G.; Roubaud, V.; Tuccio, B.;
Tordo, P. In Free Radicals in Biology, Environment; Minisci, F., Ed.;
Kluwer Academic Publishers: Dordrecht, The Netherlands, 1997; pp
39-47.
J. Org. Chem, Vol. 69, No. 23, 2004 7997

Villamena et al.
FIGURE 1. A view of the most stable B3LYP/6-31G* optimized structures of AMPO (top) and its superoxide adducts.
contact terms) based on eq 1, where g0 is the isotropic
aX ) 8π/3(ge/g0)γXâXFrx (1)
their spectral profiles as will be discussed later in the
Spin Trapping section.
Structural Analysis. AMPO was synthesized from
EMPO in the absence of NaCN catalyst and was fully
characterized with spectroscopic techniques. One advan-
tage of AMPO over commonly used nitrones such as
DMPO, EMPO, and DEPMPO is that as a solid, its ease
of purification makes AMPO less susceptible to contami-
g-value for the radical, ge the g-value for the free electron,
γX the gyromagnetic nuclear ratio, and âX the nuclear
magneton of the nucleus X.32 Although these small
percentage values for the spin density distribution may
seem insignificant, their corresponding aX in Gauss
(Table S17) could indeed give a significant influence on
1
nation from paramagnetic impurities. H NMR spectra
of AMPO in D2O show fast exchange of the amido H’s
while two nonequivalent amido H’s in CDCl3, relatively
broad peaks at 5.43 and 8.25 ppm, indicate restricted
NH2 group rotation around the C-N bond. This could
(32) Cirujeda, J.; Vidal-Gancedo, J.; Ju¨rgens, O.; Mota, F.; Novoa,
J. J.; Rovira, C.; Veciana, J. J. Am. Chem. Soc. 2000, 122, 11393-
11405.
7998 J. Org. Chem., Vol. 69, No. 23, 2004

5-Carbamoyl-5-methyl-1-pyrroline N-Oxide
FIGURE 2. EPR spectral profile of EMPO-O2H (a) and DEPMPO-O2H (b) adducts after 5 min of irradiation, using the light-
•-
riboflavin system to generate O2 and (c and d) spectra of the same adducts, respectively, taken 1 h after the light was turned
off; (e and f) spectra of EMPO-OH and DEPMPO-OH, using the Fenton reaction, respectively. See experimental methods for
spectrometer settings.
SCHEME 2. Synthesis of AMPO and NH2-AMPO
be partly due to intramolecular H-bonding of the NH2
with the nitronyl O.
It was worth investigating the X-ray structures of
AMPO and NH2-AMPO since these experimental data
can provide important structural information for nitrones
with an amido substituent. The compound NH2-AMPO
was produced from aminolysis of the carbonyl C and a
series of reactions involving an oxidation of the nitrone
functionality as facilitated by the initial nucleophilic
Spin Trapping. The addition of various radicals to
substitution by CN- (Scheme 2). X-ray crystallographic
AMPO was performed in a phosphate buffer and char-
structures for AMPO and NH2-AMPO are shown in
Figure 3.
acteristic spectral profiles were observed for each of the
adducts (Figure 5). Table 2 shows a summary of all of
Theoretical analysis of AMPO predicted intramolecular
H-bonding of amido H with nitronyl O giving a bond
distance of 1.98 Å (Figure 1, top). The X-ray packing
arrangement for AMPO shown in Figure 4 indicates
intermolecular H-bonding of the amido H with the
nitronyl O with bond distances of 1.95 and 2.15 Å,
respectively, while the novel nitrone NH2-AMPO exhib-
ited both inter- and intramolecular H-bonding of the
amide moiety to the nitronyl O with distances of 1.96 and
2.10 Å (see Supporting Information). Crystallographic
details of the compounds AMPO and NH2-AMPO are
shown in the Supporting Information as Tables S1-S6
and Tables S7-S14, respectively.
the observed EPR spectral parameters. Although the
overall spectral profile for the •OH, t-BuO•, and glutathio-
nyl (GS•) adducts are similar, their hyperfine splitting
constant (hfsc) assignments vary. The superoxide adduct
of AMPO gave a spectral profile (Figure 6a,b) similar to
those observed for DMPO and alkoxycarboxylated O2H
adducts. The AMPO-O2H generated from xanthine-
xanthine oxidase, PMA activated neutrophils, or riboflavin-
light systems (see Figures 6 and 7) gave hyperfine
splittings resulting from a mixture of two diastereomers
and simulation of these spectra gave hfsc assignments
for isomer I (80%) of anitronyl-N ) 13.0 G and aâ-H ) 10.8
G and for isomer II (20%) anitronyl-N ) 13.1 G, aâ-H ) 12.5
J. Org. Chem, Vol. 69, No. 23, 2004 7999

Villamena et al.
FIGURE 3. A view of the X-ray structure of AMPO (left) and NH2-AMPO (right). The non-hydrogen atoms are drawn with 50%
probability displacement ellipsoids. The hydrogen atoms are drawn with an arbitrary radius.
characteristic spectra (Figure 2a,b); however, their spec-
tral profiles (Figure 2c,d) become obscure after 1 h due
to their partial decomposition to the •OH adducts as well.
Figure 2, spectra e and f, show the spectra of the
•
independently generated OH adducts and exhibit close
•-
resemblance to the partially decomposed O2 adduct’s
spectra (Figure 2c,d). One probable advantage of AMPO-
O2H is that although it decomposes relatively slow to the
OH adduct, its spectrum is uniquely different from its
•OH adduct spectrum so that even in low concentrations,
the AMPO-O2H spectrum could not be misinterpreted
•
as an OH adduct. As demonstrated in Figure 7, the
AMPO stability and its spin trapping characteristic were
assessed in biological milieu by using PMA activated
neutrophils and showed a distinguishable AMPO-O2H
FIGURE 4. A view of the X-ray structure of AMPO showing
the intermolecular H-bonding interaction between the amido-H
and the nitronyl-O.
spectrum over a period of 1 h. The presence of a
characteristic signature for a “C-centered adduct” is
evident during spin trapping by AMPO, using enzymati-
•-
cally generated O2 (Figures 6a and 7), and is more
G, and aγ-H ) 1.75 G. The theoretically predicted hfsc
with use of eq 1 indicates that additional hfsc could result
from the amide N, i.e., the calculated hfsc’s for the most
stable AMPO-O2H cis-2a are anitronyl-N ) 12.53 G, aamido-N
) 1.86 G, and aâ-H ) 14.53 G (Table S17). However, we
could not make these assignments consistent with the
experimentally obtained spectrum of the 15N-labeled
amide nitrogen of AMPO-OOH, or the relatively large
hfsc due to amide N cannot be observed in the unlabeled
AMPO-OOH. It should be noted, however, that these
calculated aX values do not account for the solvation effect
and could still vary significantly after the environmental
effects are considered.33 Gas-phase calculations of the
isotropic hyperfine splitting constants were also predicted
evident with use of the light-riboflavin superoxide gen-
erating system (Figure 6c). Despite all attempts to obtain
a more purified AMPO (as indicated by our spectroscopic
data) as well as subjecting the finely ground sample to
heating at 40 °C and 1 mmHg for 16 h, the spectra did
not reflect the formation of the “C-centered adduct”.
Moreover, using the same experimental conditions for
DMPO and DEPMPO did not yield such “C-centered
adducts”. The presence of a C-centered spectrum may be
a result of other spin-trapping processes simultaneously
•-
occurring in solution or from an unidentified form of O2
adduct that is in equilibrium with AMPO-O2H, or its
paramagnetic decomposition product.
•-
for the most stable O2 adducts of DMPO, EMPO, and
Spin Adduct Stability. The lowest field peak inten-
sity of the nitrone-O2H spectrum was monitored, and no
signal was observed before irradiation. However, peak
formation does occur when the light source was turned
on. Peak intensity begins to rise as a function of time
and reaches a plateau over a certain period of time, then
decreases when the light source is turned off. Decay plots
(Figure 8) of the O2 adducts were fitted by using eq 2,
since fitting with a pure first or second order shows
deviation from linearity. Therefore parallel first- and-
DEPMPO, and gave good agreement with their experi-
mental values in the aqueous phase (see Table S17).
The AMPO-O2H showed slow decomposition to the
corresponding OH (Figure 5a for the OH adduct) unlike
the DMPO-O2H.34 Figure 2 shows a comparison of O2
•-
adducts of EMPO and DEPMPO and 1 h after their
formation. Both EMPO-O2H and DEPMPO-O2H gave
•-
(33) Improta, R.; Barone, V. Chem. Rev. 2004, 104, 1231-1253.
(34) Buettner, G. R.; Oberley, L. W. Biochem. Biophys. Res. Com-
mun. 1978, 83, 69-74.
8000 J. Org. Chem., Vol. 69, No. 23, 2004

5-Carbamoyl-5-methyl-1-pyrroline N-Oxide
FIGURE 5. Experimental (left) and simulated (right) EPR spectra of AMPO radical adducts with (a) •OH, (b) CO2•-, (c) GS•, (d)
SO3•-, (e) t-BuO•, and (f) CH3 CHOH. See experimental methods for radical generation and spectrometer settings.
•
second-order reactions with a differential rate equation
d[RA]
order rate constants, respectively. A coefficient of 2 was
included in the second-order expression to avoid ambigu-
ity and to denote that two nitrone-O2Hs undergo adis-
-
) k1[RA] + 2k2[RA]2
(2)
•-
proportionation reaction. Half-lives of the O2 adducts
dt
were approximated from the first-order rate constant and
are shown in Table 3. Results show that the half-lives of
AMPO-O2H (t1/2 ∼ 8 min) and EMPO-O2H (t1/2 ∼ 10
was employed, where [RA] is the concentration of ni-
trone-O2H, while k1 and k2 are the first- and second-
J. Org. Chem, Vol. 69, No. 23, 2004 8001

Villamena et al.
TABLE 2. EPR Parameters of Simulated Radical
Adducts of AMPOa
hyperfine coupling
constants (G)
diastereo-
mers (%)
â
γ
radicals
generating system
HX/XOg
aN
aH
aH
•-
O2
80
20
69
31
47
53
46
54
100
56
44
90
10
13.0
13.1
14.0
14.0
10.8
12.5
13.5
12.5
1.75
•OH
CO2
Fe2+-H2O2
Fe2+-H2O2-NaHCO2
Fe2+-H2O2-Na2SO3
14.25 18.15
14.53 16.48
13.47 15.93
13.47 14.67
•-
b
•- c
SO3
CH3 CHOHd Fe2+-H2O2-EtOH
(CH3)3CO• e (CH3)3COOC(CH3)3
UV
14.8
21.4
•
-
14.19 13.64
13.85 12.79
14.26 14.96
14.39 12.06
GS• f
GSSG-UV
a Based on the simulation program by Rockenbauer et al.35
b Simulated spectrum contains 19% C-centered adduct and 12%
OH adduct. c Simulated spectrum contains 23% C-centered adduct.
d Simulated spectrum contains 13% OH adduct. e Simulated spec-
trum contains 10% C-centered adduct and 12% OOH-like adduct.
f Simulated spectrum contains 12% C-centered adduct. g Hypo-
xanthine/xanthine oxidase.
min) are similar to the literature value of t1/2 ) 8 min36
or 8.6 min14 for EMPO-O2H. The AMPO-O2H half-life
of 8 min is relatively shorter compared to that found for
DEPMPO-O2H of t1/2 ) 16 min but much longer than
that of DMPO-O2H of t1/2 ∼ 1 min.
Partition Coefficient. Values obtained in this study
(Table 3) are within an order of magnitude of those
previously reported of 0.06,31 0.15,14 and 0.137 for DEP-
MPO, EMPO, and DMPO, respectively. The partition
coefficient value obtained for AMPO is similar to Kp
values observed for most of the DMPO-type nitrones. Due
to this similarity in Kp values, it is expected that
permeability of AMPO through lipid bilayer membranes
would be similar to that of DMPO and DEPMPO reported
by Anzai et al.38
FIGURE 6. EPR spectral profile of AMPO-O2H (a) generated
by xanthine-xanthine oxidase; (b) simulated spectrum based
on the parameters described in Table 2; (c) spectrum generated
by the light-riboflavin system (note the significant contribution
from a C-centered adduct). See experimental methods for
spectrometer settings. The arrow indicates the peak being
monitored during kinetic studies.
Experimental Section
2-Amino-5-carbamoyl-5-methyl-1-pyrroline N-Oxide
(NH2-AMPO). NH2-AMPO was prepared by using the
procedure described previously39 with cyanide as catalyst for
the aminolysis of esters. A solution of 100 mg (0.584 mmol) of
EMPO in 25 mL of ca. 12 N NH3 in MeOH and 28 mg (0.058
mmol) of NaCN was heated to 60 °C in a sealed tube for 40 h.
The solvent was evaporated and the residue was redissolved
in CH2Cl2. The organic phase was extracted with a minimal
amount of water and dried over MgSO4. Evaporation of the
solvent gave a mixture of AMPO and NH2-AMPO. The crude
product was purifed by column chromatography, using silica
gel and methanol-ethyl acetate (30:70) as solvent. The product
was further purified twice by column chromatography with
EtOH as solvent, which afforded NH2-AMPO (5 mg), mp >200
H, m, C(3)H). IR (neat film) 3346 and 1661 (N-H), 1682 (Cd
O), 1627 (CdN), 1201 (N-O). ESI-MS calcd for C6H11N3O2-
Na+ m/z 180.0749, found 180.0743 amu.
EPR Measurements. EPR measurements were carried out
on an X band spectrometer with HS resonator at room
temperature. General instrument settings are as follows unless
otherwise noted: microwave power, 10 mW; modulation am-
plitutude, 1.0 G; receiver gain 3.17-3.56 × 105; for field sweep
the scan time was 40 s and the time constant was 328 ms; for
kinetic measurements at fixed magnetic field, the acquisition
time was 335-2680 s and the time constant was 1310 ms. All
spin-trapping studies were carried out in phosphate buffered
saline containing 0.1 mM diethylenetriaminepentaacetic acid
(DTPA). The total volume of all solutions used for EPR
measurement was 50 µL, which was loaded into 50-µL mi-
cropipets. Samples were irradiated with a 150-W light source
positioned 5 cm away from the sample cavity.
Decay Kinetics. In a typical decay kinetic study, 50 µL of
solution containing 25 mM of the nitrone and 100 µM
riboflavin was irradiated for 3 min in the cavity. The lowest
field peak decay was monitored as a function of time over a
period of 2680 s after the light source was turned off. All data
were the average of three or more measurements.
1
dec. H NMR (400 MHz, D2O, 4.58 ppm) δ 1.40 (3 H, s, C(5)-
Me), 1.90-2.06 and 2.22-2.35 (2 H, m, C(4)H), 2.59-2.65 (2
(35) Rockenbauer, A.; Korecz, L. Appl. Magn. Reson. 1996, 10, 29-
43.
(36) Zhang, H.; Joseph, J.; Vasquez-Vivar, J.; Karoui, H.; Nsan-
zumuhire, C.; Martasek, P.; Tordo, P.; Kalyanaraman, B. FEBS Lett.
2000, 473, 58-62.
(37) Janzen, E. G.; West, M. S.; Kotake, Y.; DuBose, C. M. J.
Biochem. Biophys. Methods 1996, 32, 183-190.
(38) Anzai, K.; Aikawa, T.; Furukawa, Y.; Matsushima, Y.; Urano,
S.; Ozawa, T. Arch. Biochem. Biophys. 2003, 415, 251-256.
(39) Hogberg, T.; Strom, P.; Ebner, M.; Ramsby, S. J. Org. Chem.
1987, 52, 2033-2036.
Miscellaneous Spin-Trapping Studies. (a) Fenton Re-
action System. A 50-µL 0.1 M phosphate buffer solution
8002 J. Org. Chem., Vol. 69, No. 23, 2004

5-Carbamoyl-5-methyl-1-pyrroline N-Oxide
TABLE 3. First-Order Approximation Half-Lives of
Nitrone-Superoxide Adducts and Partition Coefficient
of the Nitrone Spin Traps at pH 7.2 and 23 °C
Kp(n-octanol/
spin trap
k1/10-4 s-1
t1/2/mina
ref
water)b
AMPO
EMPO
14.0 ( 1.2
11.6 ( 0.5
8.3 ( 0.7 this work
9.9 ( 0.4 this work
0.03
0.33
8.6
8.0
14
36
24.6 ( 2.7 18
DEPMPO
DMPO
7.5 ( 0.7 15.5 ( 1.4 this work
8.13
129.0
0.16
0.06
14.2
0.9
16
16
a Based on the first-order rate constant; values are the mean
average of 3-6 measurements. b Shaken for 2 h at 37 °C.
NaHCO2, Na2SO3, or ethanol was prepared with 65 mM freshly
prepared FeSO4. The mixture was then transferred to a 50-
µL capillary tube and the EPR spectrum of the adduct was
recorded over a 5 min time period.
(c) Trapping of GS• and t-BuO• Radicals. A 50-µL 0.1
M phosphate buffer solution containing 30 mM AMPO and 100
mM GSSG or (CH3)3CO-OC(CH3)3 was prepared. The mixture
was then transferred to a 50-µL capillary tube and the radicals
were generated by UV photolysis. The EPR spectrum of the
adduct was recorded over a 5 min time period.
•-
(d) Trapping of O2•-. Typical O2 trapping experiments
utilized the riboflavin-light system as previously described.16
•-
An alternative O2 generating system used a solution of 0.4
mM hypoxanthine and 0.5 unit/mL xanthine oxidase, or 10
nM PMA and 8 × 105 neutrophil cells in 25 mM AMPO.
Spectra were acquired over a period of 15 min.
Partition Coefficient. To a 100-µL aqueous solution of 400
mM nitrone was added 100 µL of n-octanol. The mixture was
vigorously mixed for 2 h at 37 °C and centrifuged. The
concentration of the nitrones was analyzed by using a UV-
vis spectrophotometer. Standard curves were constructed for
each of the nitrones at wavelengths of 226 nm for DMPO and
EMPO, 233 nm for DEPMPO, and 230 nm for AMPO. The
concentration of the nitrone in the organic phase was deter-
mined by subtracting the original concentration of the nitrone
from the concentration in the aqueous layer after vigorous
FIGURE 7. Superoxide adduct formation by 25 mM AMPO,
using PMA-activated neutrophils: top to bottom 2, 10, 30, 60
min after the addition of PMA.
mixing. Partition coefficient Kp is expressed as [nitrone]n-octanol
/
[nitrone]buffer
.
Computational Methods. Density functional theory40,41
was applied in this study to determine the optimized geometry,
vibrational frequencies, and single-point energy of all station-
ary points.42-45 The effect of solvation on the gaseous phase
calculations was also investigated by using the polarized
continuum model (PCM).46-49 All calculations were performed
with Gaussian 9850 at the Ohio Supercomputer Center. Single-
point energies were obtained at the B3LYP/6-31+G** level
based on the optimized B3LYP/6-31G* geometries. Stationary
points for both the nitrone spin traps and spin adducts have
no imaginary vibrational frequencies as derived from a vibra-
tional frequency analysis (B3LYP/6-31G*). A scaling factor of
(40) Labanowski, J. W.; Andzelm, J. Density Functional Methods
in Chemistry; Springer: New York, 1991.
(41) Parr, R. G.; Yang, W. Density Functional Theory in Atoms and
Molecules; Oxford University Press: New York, 1989.
(42) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
(43) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(44) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(45) Hehre, W. J.; Radom, L.; Schleyer, P. V.; Pople, J. A. Ab Initio
Molecular Orbital Theory; John Wiley & Sons: New York, 1986.
(46) Barone, V.; Cossi, M.; Tomasi, J. J. Chem. Phys. 1997, 107,
3210.
FIGURE 8. Decay plots of O2•- adducts of DEPMPO, EMPO,
and AMPO after 3 min of visible light irradiation of solutions
containing 25 mM of the corresponding nitrone and 100 µM
riboflavin in pH 7.2 phosphate buffer.
containing 30 mM AMPO, 1% H2O2, and 65 mM FeSO4 was
transferred to a 50-µL capillary tube and the EPR spectrum
of the hydroxyl adduct was recorded over a 5 min time period.
(47) Barone, V.; Cossi, M.; Tomasi, J. J. Comput. Chem. 1998, 19,
404.
•
(b) Trapping of SO3•-, CO2•-, and CH3 CHOH Radicals.
(48) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Chem. Phys. Lett.
1996, 255, 327.
(49) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027.
A 50-µL 0.1 M phosphate buffer solution containing 30 mM
AMPO, 1% H2O2, and 100 mM of the respective radical source
J. Org. Chem, Vol. 69, No. 23, 2004 8003

Villamena et al.
0.9806 was used51 for the zero-point vibrational energy (ZPE)
corrections. Spin contamination for all of the stationary points
for the spin adduct radical structures was negligible, i.e., 0.75
< S2 < 0.76. All spin and charge densities were obtained from
natural population analyses (NPA) at the B3LYP/6-31G*
level.52
Conclusion
The spin-trapping properties of AMPO have been
demonstrated for the first time. AMPO is a solid spin trap
that is easy to purify without giving paramagnetic
impurities based on its EPR spectrum. The nitrone
AMPO exhibited characteristic EPR spectra with various
radicals, notably, O2 vs OH adduct. Optimized geom-
etry calculations for AMPO and its O2 adduct show
evidence of intramolecular H-bonding. The most plausible
mechanism for AMPO-O2H formation based on the
calculated thermodynamic data is the addition of O2•- to
AMPO to form the anionic adduct AMPO-O2-, and
subsequent protonation giving the hydroperoxyl adduct,
AMPO-O2H. An X-ray structure determination of the
analogous NH2-AMPO further supports the calculated
structural features of AMPO. The half-life of the
AMPO-O2H is comparable to that of EMPO-O2H but
shorter compared to that of DEPMPO-O2H. Thermody-
namic properties of superoxide radical addition to ni-
trones can predict favorable formation of certain adducts.
The partition coefficient Kp of AMPO is comparable for
all nitrones investigated in this study. This study,
therefore, demonstrates the suitability of AMPO as a spin
trap to study radical production in aqueous systems.
Crystallographic Structure. Single crystals of AMPO
were obtained from slow evaporation at room temperature of
its solution in methanol-benzene. The data collection crystal
for AMPO was a thin colorless plate. Examination of the
diffraction pattern on a CCD diffractometer indicated a
monoclinic crystal system. All work was done at 200 K. The
data collection strategy was set up to measure a quadrant of
reciprocal space with a redundancy factor of 3.1, which means
that 90% of the reflections were measured at least 3.1 times.
A combination of phi and omega scans with a frame width of
2.0° was used. Data integration was done with Denzo,53 and
scaling and merging of the data was done with Scalepack.53
Merging the data and averaging the symmetry equivalent
reflections resulted in an Rint value of 0.044. The structure
was solved by direct methods in SHELXS-86.54 Full-matrix
least-squares refinements based on F2 were performed in
SHELXL-93.55
A pale yellow approximately rectangular plate single crystal
of NH2-AMPO was obtained similar to the procedure de-
scribed for AMPO but with methylene chloride-ethyl acetate
as solvents. The diffraction pattern on a CCD diffractometer
indicated an orthorhombic crystal system. The data collection
strategy was set up to measure an octant of reciprocal space
with a redundancy factor of 4.4. The teXsan56 package indi-
cated the space group to be P 2(1)2(1)2(1).
•-
•
•-
Acknowledgment. The authors wish to thank Prof.
David Hart of the OSU Chemistry Department and
Prof. DeLanson R. Crist of the Georgetown University
Chemistry Department for helpful discussions. The Ohio
Supercomputer Center (OSC) is acknowledged for sup-
port of this research. This work was supported by NIH
grants HL38324, HL63744, and HL65608. C.M.H. ac-
knowledges support from the NSF-funded Environmen-
tal Molecular Science Institute (CHE-0089147). A.R.
acknowledges support from the Hungarian Scientific
Research Fund OTKA T-046953.
Full details of the crystallographic data for AMPO and
NH2-AMPO are described in the Supporting Information
(S17-S34).
(50) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.,
Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J. A. Gaussian 98, revision A.11.3; Gaussian, Inc.: Pittsburgh,
PA, 1998.
Supporting Information Available: Energies, enthalp-
ies, and free energies for all spin traps and their corresponding
adducts as well as 1H NMR, IR, and MS spectra. This material
is available free of charge via the Internet at http://pubs.acs.org.
(51) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502-16513.
(52) Reed, A. E.; Weinhold, F. A.; Curtiss, L. A. Chem. Rev. 1998,
98, 899.
JO049244I
(53) Otwinowski, Z.; Minor, W. In Macromolecular Crystallography,
Part A in Methods in Enzymology; Carter, C. W., Jr., Sweet, R. M.,
Eds.; Academic Press: New York, 1997; Vol. 276, pp 307-326.
(54) SHELXS-86: Sheldrick, G. M. Acta Crystallogr. 1990, A46,
467-473.
(55) SHELXL-93: Sheldrick, G. M. Universitat Gottingen, Germany
1993.
(56) teXsan, version 1.7-2: Crystal Structure Analysis Package;
Molecular Structure Corporation: Woodlands, TX, 1995.
8004 J. Org. Chem., Vol. 69, No. 23, 2004