Amphiphilic amide nitrones: A new class of protective agents acting as modifiers of mitochondrial metabolism

Article abstract of DOI:10.1021/jm100212x

Our group has demonstrated that the amphiphilic character of α-phenyl-N-tert-butyl nitrone based agents is a key feature in determining their bioactivity and protection against oxidative toxicity. In this work, we report the synthesis of a new class of am

Full text of DOI:10.1021/jm100212x

J. Med. Chem. 2010, 53, 4849–4861 4849
DOI: 10.1021/jm100212x
Amphiphilic Amide Nitrones: A New Class of Protective Agents Acting as Modifiers
of Mitochondrial Metabolism
†
†
§
Gregory Durand,*,† Burkhard Poeggeler,‡^,§ Stephanie Ortial, Ange Polidori, Frederick A. Villamena, Jutta Boker,
ꢀ
€
ꢀ
€
§
,†
Rudiger Hardeland, Miguel A. Pappolla, and Bernard Pucci*
†Laboratoire de Chimie BioOrganique et des Systeꢁmes Moleꢀculaires Vectoriels, Universiteꢀ d’Avignon et des Pays de Vaucluse,
Faculteꢀdes Sciences, 33 Rue Louis Pasteur, 84000 Avignon, France, ‡Department of Dermatology, University of Luebeck, Ratzeburger Allee 160,
D-23538 Luebeck, Germany, §Abteilung fuer Stoffwechselphysiologie, Institut fuer Zoologie, Anthropologie und Entwicklungsbiologie der Georg
August Universita€t Go€ttingen, Berliner Strasse 28, D-37073 Go€ttingen, Germany, Department of Pharmacology and Center for Biomedical
EPR Spectroscopy and Imaging, The Davis Heart and Lung Research Institute, College of Medicine, The Ohio State University, Columbus,
Ohio 43210, and ^Department of Neurology, Medical University of South Carolina, Charleston, South Carolina 29425
Received February 17, 2010
Our group has demonstrated that the amphiphilic character of R-phenyl-N-tert-butyl nitrone based
agents is a key feature in determining their bioactivity and protection against oxidative toxicity. In this
work, we report the synthesis of a new class of amphiphilic amide nitrones. Their hydroxyl radical
scavenging activity and radical reducing potency were shown using ABTS competition and ABTS•þ
reduction assays, respectively. Cyclic voltammetry was used to investigate their redox behavior, and the
effects of the substitution of the PBN on the charge density of the nitronyl atoms, the electron affinity,
and the ionization potential were computationally rationalized. The protective effects of amphiphilic
amide nitrones in cell cultures exposed to oxidotoxins greatly exceeded those exerted by the parent
compound PBN. They decreased electron and proton leakage as well as hydrogen peroxide formation
in isolated rat brain mitochondria at nanomolar concentration. They also significantly enhanced
mitochondrial membrane potential. Finally, dopamine-induced inhibition of complex I activity was
antagonized by pretreatment with these agents. These findings indicate that amphiphilic amide nitrones
are much more than just radical scavenging antioxidants but may act as a new class of bioenergetic
agents directly on mitochondrial electron and proton transport.
Introduction
species. This unregulated production of ROS has been asso-
ciated with critical pathological events such as myocardial
ischemia-reperfusion or stroke as well as with the etiology of
the aging process and that of many diseases.
Under normal conditions, most of the intracellular reactive
oxygen species (ROSa) are produced within the mitochondria
with the superoxide radical anion (O2-•) being the first
molecule generated during the respiration process. The ability
of O2-• to chemically or enzymatically self-dismutate to H2O2
and subsequently to form HO• via the Fenton reaction as well
as its selective reactivity toward transition metal ions accounts
Therefore, there has been extensive research in the deve-
lopment of effective molecules thatcan scavenge ROS and can
be used either as a probe or therapeutic agents; among them
are nitrone spin-traps. Previous efforts have been mainly
focusedonthe design of intrinsically more potent compounds,
but recently, the selective targeting of antioxidants has been
developed as an alternative approach. Because of the inner
membrane potential of the mitochondria (160-185 mV), the
Skulachev cations consisting of lipophilic triphenylphospho-
nium cations accumulate in the negatively charged compart-
ment. Various antioxidants or probes have been grafted
through their aliphatic chain to this phosphonium carrier.1,2
It has been reported that the PBN conjugate, MitoPBN
(Figure 1), reaches 2.2-4.0 mM within the mitochondria
and blocks the oxygen-induced activation of uncoupling
proteins.3 However, toxicity of triphenylphosphonium deri-
vatives was found at low concentration which may consider-
ably limit their use as therapeutic agents.4 Another approach
is the use of cell penetrating peptides (CPPs) as mole-
cular carriers. Highly cationic CPPs have been engineered
for transport into mitochondria, demonstrating that lipo-
philicity and positive charge in peptidic sequences induce
-•
for its high toxicity. Furthermore, the higher reactivity of O2
at lower pH is due to the formation of the hydroperoxyl
radical (HO2 ) (pKa(HO2 /O2-•) = 4.8), which is more reac-
tive than O2-•. The overproduction of ROS can greatly over-
whelm the endogenous antioxidant system for their detoxi-
fication and hence can lead to an accumulation of deleterious
•
•
*To whom correspondence should be addressed. For G.D.: phone,
þ33 4 9014 4445; fax, 33 4 9014 4491; e-mail, gregory.durand@
univ-avignon.fr. For B.P.: phone, þ33 4 9014 4442; fax, 33 4 9014 4499;
e-mail, bernard.pucci@univ-avignon.fr.
a Abbreviations: ABTS, 2,20-azino-bis(3-ethylbenzthiazoline-6-sulfo-
nic acid); BHT, butylated hydroxytoluene; cmc, critical micellar con-
centration; CPP, cell penetrating peptides, CV, cyclic voltammetry; EA,
electron affinity; ETC, electron transport chain; FCCP, p-trifluoro-
methoxy carbonylcyanide phenylhydrazone; FOX, ferrous oxidation
of xylenol orange; IP, ionization potential; MMP, mitochondrial mem-
brane potential; NARP, neurogenic ataxia retinitis pigmentosa; NBT,
nitro blue tetrazolium; NPA, natural population analysis; PCM, polar-
ized continuum model; PBN, R-phenyl-N-tert-butylnitrone; ROS, re-
active oxygen species; SOD, superoxide dismutase.
r
2010 American Chemical Society
Published on Web 06/09/2010
pubs.acs.org/jmc

4850 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13
Durand et al.
character of the polar head.15 For both series of amphiphilic
PBN, the preliminary biological evaluations showed that
these amphiphilic compounds were in general far more potent
than the parent antioxidant, PBN. This clearly demonstrates
that the amphiphilicity of nitrones is a key feature in deter-
mining their bioactivity and protection against the oxidative
toxicityin vitro andinvivo14,16,17 as exemplified by compound
13 (LPBNAH) exhibiting exceptionally high antioxidant
activity.18,19 In order to extend and explore the importance
of the amphiphilicity of nitrone derivatives in determining
their antioxidant potency, we synthesized hydrophilic, am-
phiphilic, and lipophilic amide analogues of 13. Although
these new analogues exerted good protective effects both in
vitro and in vivo, exceeding the antioxidant activity of many
other compounds, nitrone 13 was found once again to be the
most potent compound.20
We report herein the synthesis of an amphiphilic amide
nitrone 11 (LPBNH15) in which the polar headgroup derived
from lactobionic acid and the C7 alkyl chain are both linked
via an amide bond to the N-tert-butyl group and the phenyl
ring of the PBN moiety, respectively. Nitrones 11 and 13 are
structural isomers because they are differentiated only from
the position of the polar head and the hydrophobic chain. To
explore the importance of amphiphilicity, the fluorinated
analogue 12 (LPBNF13) bearing a 1H,1H,2H,2H-perfluoro-
octyl chain has also been synthesized. Indeed, while amphi-
philic compounds endowed with a hydrocarbon chain can
induce a destabilization of cell membranes when used above
their critical micellar concentration (cmc), fluorinated amphi-
philic analogues are not detergent-like. However, from our
findings with amphiphilic nitrones, the usefulness of fluorinated
tails compared to hydrogenated ones is still unclear, as our lead
compound 13 was found to be much more potent than its
fluorinated analogue 14 (LPBNAF).20 Moreover, because of its
relatively short alkyl chain, no cmc was observed for compound
1314 and it was found to be devoid of any detergent properties or
toxicity even at very high concentrations.18
The hydroxyl radical scavenging activity and the radical
reducing potency of these newly designed amphiphilic amide
nitrones were explored using the ABTS competition and an
ABTS•þ reduction assay, respectively. The effect of the sub-
stitution on the electrochemical properties of the nitronyl
group was investigated by cyclic voltammetry and was corre-
lated with theoretically calculated electron affinities and
ionization potentials using B3LYP/6-31þG(d,p)//B3LYP/
6-31G(d) level of theory. Their protective effects in primary
cortical mixed cell cultures exposed to three oxidotoxins, that
is, hydrogen peroxide, peroxynitrite, and doxorubicin, were
investigated. An in vivo model, using aquatic invertebrate
microorganisms (i.e., rotifers) exposed to lethal concentra-
tions of H2O2 and doxorubicin oxidotoxins, was also em-
ployed. Finally, the ability of amphiphilic amide nitrones to
interact with the mitochondrial energy metabolism at nano-
molar concentration was further investigated. For the sake of
comparison, compound 13 and its fluorinated analogue 14
were also included in the physical-chemical and biological
assays as well as in the computational studies.
Figure 1. Chemical structures of PBN, mito-PBN, and amide nitrones
previously developed.
mitochondrial accumulation.5 For instance, the tetrapeptide
SS-31, consisting of alternating aromatic and basic amino
acids, was able to prevent apoptosis of neurons at nanomolar
concentration because of its 5000-fold accumulation in the
mitochondria.6 Gramicidin S segments also exhibit specific
mitochondrial compartmentalization and exert antiapoptotic
properties when conjugated to nitroxides.7 Another approach
that recently attracted attention is to attach a nitrone group to
the (-)-(R)-carnitine, which is used in the transport of fatty
acids into the matrix of the mitochondria.8
With the expectation that amphiphilic compounds posses-
sing both a hydrophilic polar head and a lipophilic group
would exhibit improved bioavailability and membrane cross-
ing ability, our work over the past10years has been devoted to
the design and the synthesis of amphiphilic antioxidants with
particular attention to nitrone derivatives. Two original series
of amphiphilic compounds, bearing R-phenyl-N-tert-butylni-
trone (PBN) as synthetic antioxidant moiety, have been
developed in our lab. In the first series, the PBN moiety is
grafted onto a fluorinated glycolipidic carrier through an
amide bond. The first amphiphilic nitrone of this series9 was
found to readily diminish superoxide dismutase (SOD) induc-
tion and apoptosis in cultured skin fibroblasts with an isolated
complex V deficiency of the respiratory chain, caused by the
neurogenic ataxia retinitis pigmentosa (NARP) mutation.10
A more general amino acid based amphiphilic carrier consist-
ing of a hydrophilic head derived from lactobionic acid and a
perfluoroalkyl tail was then developed later.11 An amino acid,
that is, lysine or aspartic acid, is used as scaffold upon which
hydrophobic and hydrophilic parts are linked respectively by
carboxyl and amino groups and upon which PBN is grafted to
functional side chain groups.12 This amphiphilicity mediated
delivery strategy was then extended to a broad spectrum of
antioxidants such as lipoic acid, indole-3-propionic acid,
Trolox, and very recently the cyclic nitrone DMPO.13
We also reported the synthesis of a second series of amphi-
philic PBN derivatives in which the nitrone function is fitted
into the core of the molecule14 bridging the polar head grafted
onto the aromatic function and the hydrophobic moiety,
bound to the tert-butyl group. This convergent synthetic
pathway allows modification of the nature, length, and type
ofgrafted hydrophobic moiety, as wellas the ionic ornonionic
Chemistry
Synthesis. The convergent synthetic pathway of the am-
phiphilic nitrones is based on three key steps (Scheme 1):
(i) synthesis of a polar lactobionamide N-tert-butylhydroxyl-
amine from 2-methyl-2-nitropropanol; (ii) synthesis of the

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4851
Scheme 1a
THF at room temperature without any observable over
reduction. After purification by flash chromatography, the
hydroxylamine 3 was obtained in 31% overall yield from the
starting 2-methyl-2-nitro propanol. It has to be noted that
the hydroxylamine was found to be stable for months at
-20 ꢀC.
4-Cyanobenzaldehyde was used as starting material for
the synthesis of the hydrophobic benzaldehyde derivatives.
Protection of the carbonyl group followed by the reduc-
tion of the cyano group led to compound 4 as previously
reported.21 The resulting amino group was directly con-
densed with octanoic acid or 1H,1H,2H,2H-perfluoronon-
anoic acid in the presence of DCC/HOBT leading to com-
pounds 5 and 6, respectively, in moderate yields (∼50%).
One can note that the yield of the amide bond formation was
significantly improved using BOP as coupling agent instead
of DCC/HOBT. Then transacetalization with acetaldehyde
afforded the benzaldehydic compounds 7 and 8 in ∼45%
overall yield from 4-cyanobenzaldehyde.
The nitrone functional group was finally synthesized by
reaction of the hydroxylamine 3 with the hydrophobic
benzaldehyde derivatives 7 and 8, using a 3:2 (v/v) THF/
AcOH mixture under argon atmosphere and in the absence
of light. It is important to note that such acidic conditions
greatly increase the kinetics of the reaction which is usually
reported to be long as well as giving a low yield of coupling.
O-Acetylated amphiphilic nitrone amides were extensively
purified by flash chromatography and by size exclusion
chromatography. Finally, after purification by aqueous size
ꢀ
exclusion chromatography, the Zemplen de-O-acetylation
procedure led to compounds 11 and 12. The amphiphilic
nitrones 11 and 12, as well as their acetylated derivatives 9
and 10, were fully characterized by 1H, 13C, 19F, DEPT, and
HMQC NMR as well as mass spectrometry (ESIþ). The
purity of 11 and 12 was confirmed by C18 reverse phase
HPLC and was higher than 95% (see Supporting In-
formation).
Physical-Chemical Measurements. All the amphiphilic
amide nitrones were soluble in water, and their solubility
was found to be significantly higher than that of PBN as
shown in Table 1. No significant difference was observed in
a Reagents and conditions: (a) TsCl, Pyr, DCM, 80%; (b) NaN3,
DMF, 89%; (c) PPh3, THF followed by NaOH 2 N; (d) lactobionic acid,
TEA, 2-methoxyethanol; (e) Ac2O/Pyr 1:1, 67% in two steps; (f) Zn,
NH4Cl, THF/H2O 3:1 (v/v), 80%; (g) HOCH2CH2OH, apts, toluene,
90%; (h) LiAlH4, THF, 62%; (i) C7H15COOH, BOP, DIEA, DCM,
85% or C6F13CH2CH2COOH, DCC, HOBt, DIEA, DCM, 52%;
(j) CH3CHO, p-toluenesulfonic acid, 88-96%; (k) THF/AcOH 3:2
(v/v), dark, 76-80%; (i) MeONa, MeOH, 75-90%.
the solubility of 11 and 13, with values of about 500 g L-1 for
3
both compounds. However, because of the high hydropho-
bicity of the perfluorinated chain, compound 12 exhibited
lower solubility than its hydrogenated derivative 11.
Relative Lipophilicities. Relative lipophilicities (log k0W) of
the amphiphilic amide nitrones were measured by a chro-
matographic technique previously used.12,15,20 Despite their
higher water solubility, all the amphiphilic amide nitrones
were also found to be more lipophilic than the parent
compound PBN. Finally, because of the peculiar nature of
their chain, fluorinated derivatives exhibited the highest
values of lipophilicity, which is consistent with our previous
finding.22 Such a behavior demonstrates the interest in
providing amphiphilic character to nitrones, as the water
solubility and the lipophilicity are both increased compared
to that of the parent compound, in agreement with our
finding with the amphiphilic amino acid carriers.12 One
can also note that the grafting of the chain onto the aromatic
ring slightly increased the lipophilicity of the nitrone, as 11
and 12 exhibited higher values of log k0W compared to those
of their substituent-reversed analogues 13 and 14.
hydrogenated and perfluorinated hydrophobic N-4-formyl-
benzylamides from 4-cyanobenzaldehyde; (iii) formation of
the nitronyl function by condensation of the hydroxylamine
to the benzaldehyde derivatives under inert atmosphere.
Lactobionamide-based N-tert-butylhydroxylamine 3 was
synthesized in six steps from the commercially available
2-methyl-2-nitropropanol. First, the hydroxyl group was
converted to the amino group following our previously
reported procedure leading to compound 1.14 For conven-
ience, compound 1 was usually converted to its ammonium
chloride form, allowing its storage for months. After regen-
eration of the amino group, compound 1 was grafted onto
lactobionolactone, followed by acetylation of the hydroxyl
groups leading to 2 after purification by flash chromato-
graphy. Finally, the reduction of the nitro group was mildly
carried out using a mixture of zinc/ammonium chloride in
Self-Aggregation Properties. Self-aggregation properties
were specified using surface tension measurement. Results

4852 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13
Durand et al.
Table 1. Physical-Chemical, Radical Scavenging, and Reduction Properties of Nitrone Agents
self-aggregation
water
solubility (g/L)
quenched ABTS•þ
reduced ABTS•þ
(nmol/30 min)i
7.8 ( 0.3d
nitrones
PBN
log k0W
1.62b
cmcf
γh
(nmol/30 min)i
E1/2 (red)j (V)
-1.93 ( 0.02
26.5
g
g
18.2 ( 0.9d
1.64-1.75c, d, e
2.67-2.76c, e
2.86 b
13
11
14
12
515
530
a
g
g
75.9 ( 0.9e
96.8 ( 0.9b
65.9 ( 1.1e
90.7 ( 1.3b
12.4 ( 0.5e
49.1 ( 1.1b
5.9 ( 0.3e
42.6 ( 0.9b
-1.76 ( 0.03
-1.64 ( 0.04
g
g
4.44-4.65c, e
0.213c
0.180b
20
23
a
270
4.71b
a
a Not determined. b This work. c Data from ref 15. d Data from ref 12. e Data from ref 20. f Critical micellar concentration in mM. g No cmc. h γ, limit surface
tension, in mN/m. i The findings are presented as the mean ( SEM (N = 10). All results are statistically significant different from control (vehicle) at p < 0.01
(ANOVA followed by Bonferroni t test). j 0.15 M NaCl in water, glassy carbon electrode, sweep rate of 0.15 V s-1. Average of three to four measurements.
show that only the fluorinated nitrones were able to form
micelles in pure water, in agreement with the literature
in which these perfluorinated surfactants exhibit very low
critical micellar concentration (cmc) as well as very low
surface tension at the cmc (γcmc).23 No micelle formation
was observed with nitrone 11, suggesting the absence of
detergent properties and related toxicity of this C7 hydro-
genated derivative as it was previously demonstrated with
compound 13.15 According to the slightly higher hydro-
phobic character of 12 compared with its substituent-reversed
analogue 14, the cmc of the former compound was found to
be slightly lower.
ABTS Competition Assay. Radical scavenging was inves-
tigated by means of a scavenger competition assay based on
the formation of the intensely green 2,20-azino-bis(3-ethyl-
benzthiazoline-6-sulfonic acid) ABTS cation radical ini-
tiated by a Fenton reaction.12,24 Radical scavengers suppress
time dependent ABTS cation radical formation by competi-
tion. Because of the micellar aggregation of the fluorinated
derivatives, the measurements were done in the presence of
DMSO to allow a comparison of compounds having a
different degree of amphiphilicity. All amphiphilic nitrones
showed considerable effects on ABTS oxidation. Compound
11 and its fluorinated analogue 12 acted as very potent
radical scavengers and inhibited almost completely the for-
mation of ABTS cation radicals. Their analogues, 13 and 14,
also inhibited the formation of ABTS cation radical but to a
lesser extent. This suggests that the position of the substi-
tuent on the nitrone moiety may affect the reactivity of the
nitronyl group toward radical trapping. High ability in
reducing the burstlike formation of ABTS cation radicals
was already reported for fluorinated amphiphilic amino acid
PBN derivatives, while that of PBN was found to be
limited.12 Moreover, we found that a fluorinated amphiphi-
lic amino acid carrier devoid of PBN derivative was also able
to reduce the ABTS cation radical formation (unpublished
results). Such a result clearly shows that the lactobionamide
polar head may act as a radical trap because of its many
reactive hydroxyl groups.
oxidant capacity of nitrone compounds. The effects of para-
substitution of PBN on the charge density of the nitronyl
group and on the reactivity with superoxide radical anion
and hydroperoxyl radical were theoretically rationalized by
us in a recent effort, demonstrating the importance of the
polar effect on the spin-trapping and antioxidant properties
of linear nitrone.26
Cyclic Voltammetry. Cyclic voltammetry (CV) was used to
investigate the redox behavior of the amphiphilic amide
nitrones (200 μM) in aqueous solution. Nitrone spin traps
are known to be reduced in unbuffered aqueous medium
through an ECE process (a two-electron transfer reac-
tion with an intervening chemical reaction).27 First, a two-
electron reduction of the nitrone leads to a disubstituted
hydroxylamine, which is further dehydrated to an imine
compound. The resulting imine is subsequently two-
electron-reduced to an amine compound. The reduction poten-
tials of PBN and compounds 13 and 11 were measured and
are reported in Table 1. The pronounced shifts in reduction
potentials of nitrones 11 (-1.64 V) and 13 (-1.76 V) relative
to that of PBN (-1.93 V) reflect the effect of substituents at
the aromatic ring and the N-tert-butyl group27,28 and the fact
that 11 and, to a lesser extent, 13 are more easily reduced than
PBN. The slight difference in reduction potential between 11
and 13 may arise from steric hindrance and electronic effects
due to the presence of a lactobionic group or an alkyl chain
on the N-tert-butyl group. As previously reported, electron
withdrawing groups on the N-tert-butyl group induce a
slight decrease in the reduction potentials.27 According to
this, one can expect that the lactobionic group of 11 should
exert an electron-withdrawing effect on the nitronyl group
while the alkyl chain of 13 would exert an electron-donating
effect. The higher degree of hydration of the two amphiphilic
compounds, due to the presence of the very polar lactobion-
amide group, compared to PBN might also play a role in the
reduction of the nitronyl function.
Our preliminary EPR spin trapping experiments with 13 in
the presence of a nonaqueous generating system of O2•- such
as pyridine/H2O2 or DMSO/KO2 gave strong EPR signals,
confirming the spin trapping ability of the amide nitrones
(data not shown). However, because of its amphiphilic
character and formation of secondary products, anisotropic
and complex spectra were observed. The spin trapp-
ing properties of amphiphilic amide nitrones and their
voltammetric behavior in organic solvent will be further
investigated.
Computational Studies. The electron affinities (EAs), io-
nization potentials (IPs), and charge densities of the nitronyl
atoms of compounds 11-13 were calculated at the B3LYP/
6-31þG(d,p)//B3LYP/6-31G(d) level of theory. The bulk
ABTS•þ Reduction Assay. We also investigated the ability
of amide nitrones to act as electron donors in an ABTS•þ
reduction assay.25 Surprisingly, while PBN itself and the first
generation of amide nitrones were found to be very limited
reductants and electron donors, amide nitrones 11 and 12
proved to be good reductants (Table 1). The fact that a
simple reversal of the position of the polar head and the
hydrophobic chain changes the radical reducing potency
dramatically is surprising and suggests that the nature of
the substituents at the aromatic ring system and the N-tert-
butyl group may exert a decisive role in determining anti-

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4853
Table 2. Calculated Electron Affinities, Ionization Potentials, and Charge Densities of the Nitronyl Atoms of 13, 12, 11, and PBN at the B3LYP/
6-31Gþ(d,p)//B3LYP/6-31G(d) Level of Theorya
charge densities of the nitronyl atoms (e)
nitrones
electron affinity (eV)
ionization potential (eV)
C
N
O
13
12
0.79 (1.68)
0.76 (2.15)
0.68 (2.14)
0.12 (2.05)
7.09 (5.78)
7.21 (5.74)
6.90 (5.75)
7.37 (5.75)
-0.024 (0.024)
0.012 (0.027)
0.016 (0.014)
-0.007 (0.009)
-0.030 (0.061)
0.062 (0.059)
0.075 (0.078)
0.077 (0.073)
-0.539 (-0.601)
-0.551 (-0.605)
-0.559 (-0.607)
-0.548 (-0.606)
11
PBN
a Values in parentheses are PCM(water)/B3LYP/6-31Gþ(d,p) B3LYP/6-31G(d).
Table 3. Inhibition of Induced Cell Death in Mixed Cortical Cultures and Percentage of Viable Rotifer after Treatment with Nitrone Agentsa
% inhibition of cell death in mixed cortical culturesa % viable rotifersb
compd (10 μM)
H2O2 (200 μM)
peroxynitrite (200 μM)
doxorubicin (200 μM)
H2O2 (200 μM)
doxorubicin (200 μM)
controlc
PBNc
13d
11.9 ( 0.4
25.6 ( 0.8
83.0 ( 1.1
87.6 ( 1.4
40.6 ( 1.2
30.1 ( 1.5
15 ( 0.6
25.0 ( 0.9
81.0 ( 1.0
93.0 ( 1.1
63.4 ( 1.1
42.6 ( 1.0
20.6 ( 0.7
61.0 ( 1.3
90.1 ( 1.9
40.1 ( 1.0
30.3 ( 1.1
24.2 ( 0.8
82.0 ( 1.3
93.0 ( 0.7
34.9 ( 0.8
32.8 ( 0.8
18.7 ( 0.5f
85.8 ( 1.7
96.5 ( 0.7
30.6 ( 1.2
46.7 ( 1.3
11e
14d
12e
a Cultures were treated for 24 h with oxidotoxins at 200 μM. Nitrones were simultaneously added at 10 μM. Shown is the percentage inhibition of
trypan blue absorbance indicating enhanced survival of the cells. b Rotifers were treated for 24 h with oxidotoxin at 200 μM. Nitrones were
simultaneously added at 10 μM. Shown is the percentage of viable organisms after exposure to the oxidotoxins. The findings are presented as the mean (
SEM (N = 10). Unless otherwise indicated, all results were statistically significantly different from results of the control (vehicle) at p < 0.01 (ANOVA
followed by Bonferroni t test). c Data from ref 12. d Data from ref 20. e This work. f Nonsignificant versus doxorubicin with vehicle.
dielectric effect of water on the energetics was also investi-
gated at the PCM/B3LYP/6-31þG(d,p)//B3LYP/6-31G(d)
level, and the results are shown in Table 2. Results indicate
that PBN exhibits the lowest electron affinity but no sig-
nificant difference was observed in the EAs among the
amphiphilic nitrones in both gaseous and aqueous phases.
This suggests that PBN substitution facilitates reduction of
the nitrone, consistent with the experimentally observed
reduction potentials as determined by cyclic voltammetry.
The IP value was also highest for PBN in the gaseous phase,
indicating that PBN is harder to reduce compared to the
amphiphilic nitrones, but no significant difference in the IPs
was observed in aqueous phase. This is in agreement with the
ABTS•þ reduction assay demonstrating that amphiphilic
amide nitrones are better reductants and electron donors
than the PBN parent compound. However, only a slight
difference was observed in the IP values between 11 and 13
while the ABTS•þ reduction assay showed the superiority of
the reverse-analogue 11. The NPA charges in aqueous phase
show the following order of increasing charge on the nitronyl
C: PBN < 11 < 13 < 12, demonstrating that the substitu-
tion of the nitronyl function alters its charge density which in
turn may affect its reactivity toward free radicals. Finally,
both experimental and theoretical data showed that substi-
tution of the PBN by hydrophilic and lipophilic groups alters
its redox properties, with the amphiphilic amide nitrones
being easier to oxidize and reduce than the parent PBN.
Moreover, the position of the substituent on the nitronyl
group is also of importance for the redox behavior, with the
reverse analogue 11 also exhibiting at the same time better
reducing capacity and electron donating properties than 13.
peroxynitrite, or doxorubicin) at 200 μM. While hydrogen
peroxide induces damage to the whole cell mainly because of
its ability to cross membranes and so is usually referred to
as a global and nonselective toxin, doxorubicin is a mito-
chondria-selective oxidotoxin also referred to as a specific
mitotoxin.29,30 The cytotoxicity of the doxorubicin is likely
due to its accumulation in the mitochondria lipid membrane
and its high affinity toward binding to cardiolipin, an inner
mitochondrial-membrane specific lipid. At physiological
pH, peroxynitrite decomposes through a proton-dependent
mechanism to form the hydroxyl radical (HO•) and the
•
nitrite radical (NO2 ). Peroxynitrite was also found to be
able to induce oxidative damages to the mitochondrial
electron transport chain (ETC). It is produced by reaction
of superoxide with nitric oxide in the mitochondria but can
also diffuse through cell membranes or anion channels to
induce mitochondrial damage.31 Its ability to induce apop-
tosis was also reported32 as well as its oxidant property which
leads to reaction with biomolecules.
No surviving neuronal cells in vehicle oxidotoxin treated
cultures were detected. Under such conditions of lethal
toxicity (200 μM oxidotoxin) only very potent protective
agents are able to rescue the cells and prevent their degenera-
tion and death.12 Nitrones were added simultaneously with
the oxidotoxin at only 10 μM to evaluate their putative
cytoprotective effects. Regardless of the nature of the oxido-
toxin, all the amphiphilic nitrones exhibited significant
neuroprotective activity and were significantly more effec-
tive than the parent compound PBN. Both hydrogenated
derivatives exhibited very high protective activity against the
toxicity induced by the three stressors, with compound 11
being more potent than 13. Worth noting is the very pro-
nounced cytoprotection against the mitochondria-selective
toxin doxorubicin. This may indicate an enhanced affinity of
these amphiphilic amide nitrones for the mitochondria com-
partment. Fluorinated derivatives also exhibited good pro-
tection of cells against hydrogen peroxide but to a lesser
extent than their hydrogenated analogues. One can note that
the protection toward the mitochondria selective oxidotoxin
Biological Results
Primary Cortical Mixed Cell Cultures. We examined the
prevention of cell death in three paradigms of lethal oxida-
tive stress in primary cortical mixed cell cultures, and the
results are shown in Table 3. Cultures were treated for 24 h
with three different oxidotoxins (viz. hydrogen peroxide,

4854 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13
Durand et al.
was significantly lower, suggesting that these fluorinated
compounds are less prone to reduce endogenous oxidative
stress. Surprisingly, fluorinated amphiphilic amino acid
based PBN derivatives have also been shown to reduce
oxidative stress and to exert a higher efficacy than nitrones
12 and 14 toward mitochondria selective oxidotoxins such as
doxorubicin.12 This suggests that the nature of the fluori-
nated chain is not the only parameter that accounts for the
affinity to the mitochondrial compartment but also indicates
that the chemical structure of nitrone molecules may equally
affect their compartmentalization.
Rotifer Cultures. The antioxidant properties of amphiphi-
lic amide nitrones were confirmed in in vivo experiments
using rotifers, a well established animal model in experi-
mental gerontology.18 This animal model allows for a large
scale analysis of the protective effects of antioxidant drugs
tested on individually housed organisms of defined and
uniform age and has been used successfully in aging studies
by other investigators.33-37 Since rotifers of the genus
Philodina are aquatic organisms, single housed animals can
be exposed to exactly the same toxin and drug regime in a
controlled environment, a great advantage over models
requiring dosing of the test compounds by food and water
administration or injection.
Rotifers were treated 24 h with oxidotoxin at 200 μM in the
presence or absence of nitrones at 10 μM to evaluate their
putative protective effect (Table 3). The fluorinated nitrones
12 and 14 exerted substantial protection against the toxicity
of hydrogen peroxide and doxorubicin toward these organ-
isms, whereas only a limited protection by PBN was ob-
served for the rotifers exposed to hydrogen peroxide. Once
again, these two compounds exhibited a lower efficacy than
the fluorinated amphiphilic amino acid based PBN deriva-
tives.12 On the contrary, the two hydrogenated nitrones 11
and 13 were very effective against the lethal toxicity of the
nonselective agent hydrogen peroxide and against the mito-
chondria-selective oxidotoxicity of doxorubicin, these agents
being ∼3 and ∼5 times more active than the parent com-
pounds in the two oxidative stress paradigms, respectively.
The almost complete protection by the new hydrogenated
amphiphilic amide nitrone 11 confirms previous findings
that enhanced bioavailability is a key feature in determining
antioxidant protection in cells and organisms.12,15,18,20 To
study the molecular mechanisms of the antioxidant protec-
tion of the two very efficient hydrogenated amphiphilic
amide nitrones 11 and 13, we next carried out a series of
biochemical experiments on mitochondrial preparations as
described below.
Mitochondrial Preparation. Determination of Soluble
Hydroperoxides Formation. The ability of amide nitrones
to reduce peroxide formation in respiring mitochondria
under physiological conditions was first investigated. We
used the xylenol orange method (also referred to as the FOX
assay) to determine hydroperoxides in mitochondria as
described by Hermes-Lima et al.38 A microassay variant of
this method was adapted to measure soluble hydroperoxides
as described by Lyras et al.39 Because of protein precipitation
during homogenization, the assay as applied does not detect
total peroxides but only the soluble hydroperoxides and thus
almost exclusively the abundant hydrogen peroxide (at least
98% in the mitochondrial preparations, data not shown). As
shown in Figure 2, both amide nitrones were found to reduce
the peroxide formation in respiring mitochondria under
physiological conditions at submicromolar concentrations;
Figure 2. Inhibition of the water-soluble peroxide formation (top)
and inhibition of the NBT reduction (bottom) by 11 (1) and 13 (b)
in isolated rat brain mitochondria. Glutamate and succinate con-
centrations were 6 mM. Mitochondrial preparations were harvested
30 min after incubation with nitrones. The findings are presented
as the mean ( SEM of at least four independent mitochondrial
preparations.
however, compound 11 was much more potent than 13. For
instance, at only 0.4 μM, 13 decreases peroxide formation by
∼10%, while at the same concentration 11 leads to more than
50% of inhibition. As the concentration of nitrones in-
creases, the trend remains the same with 20% and 80% of
inhibition of peroxide formation by 0.8 μM 13 and 11,
respectively. This result demonstrates that 11 and to a lesser
extent 13 can prevent oxidant formation with extremely high
efficacy. Interestingly, both nitrones prevent the time depen-
dent increase in hydroperoxide formation in rotifers cultures
when maintained over a period of 4 weeks (unpublished
results). These findings indicate the role for antioxidant
protection in mediating at least some of the beneficial effects
of amphiphilic amide nitrones in promoting survival, there-
by protecting the mitochondria, cells, and organisms against
oxidant formation.
Determination of Electron Leakage. Amphiphilic amide
nitrones and mitochondrial complex I interaction was stu-
died in rat brain mitochondrial preparations. The deter-
mination of electron leakage induced by rotenone in mito-
chondrial preparations was assayed by the NBT reduction
test.40 Nitro blue tetrazolium is an alternative electron
acceptor, which is reduced by active respiring mitochondria
to a formazan dye. Specificity of the NBT method to
quantify electron leakage from the mitochondrial electron
transport chain was rigorously examined and confirmed as
described by Hensley et al.40 As shown in Figure 2, the
nitrones greatly reduced electron leakage in rat brain

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4855
mitochondrial preparations respiring on physiological sub-
strates. Compound 11 was much more effective than 13 in
this regard, but both compounds acted at nanomolar con-
centration to reduce formazan formation in respiring mito-
chondria at physiological glutamate and malate concen-
trations. After treatment with only 0.8 μM nitrones, the
inhibition of the NBT reduction reached 30% and 75% for
13 and 11, respectively. This indicates that both nitrones may
safeguard electron flow by preventing leakage, probably
acting as alternative electron acceptors being reduced to
active nitroxides or even to the fully reduced hydroxyl-
amines. Hensley et al. reported that PBN interacts with
mitochondrial flavin dehydrogenases to alter the dynamics
of electron transit and to prevent leakage from these
enzymes.40 More recently, Genova et al. suggested Fe-S
cluster N2 as the site of electron leakage within complex I
to direct reaction with oxygen and with water-soluble
quinone.41,42 The iron sulfur cluster N2 is localized in an
amphiphatic ramp extruding from the inner membrane to
the mitochondrial matrix, and because of their amphiphilic
nature, amide nitrones 11 and 13 may target this specific site
and act as alternative electron acceptors. This confirms
previous findings where 13 efficiently prevented the inhibi-
tion of the activity of Fe-S cluster N2 in submitochondrial
particles after incubation with oxidotoxins (viz. H2O2, per-
oxynitrite, and doxorubicin), while PBN only exerted limited
protection.20 By preventing electron leakage, amphiphilic
amide nitrones may reduce superoxide radical formation,
which in turn may reduce hydrogen peroxide and hydroper-
oxide formation. The data observed with the FOX assay
further support this hypothesis.
Increase of Mitochondrial Membrane Potential. Oxidative
stress can induce a depolarization of the mitochondrial
membrane potential (mmp) by inhibiting the activities of
Krebs cycle enzymes. Therefore, maintaining intact mito-
chondria is of importance for the survival of cells exposed to
oxidative stress, as the collapse of mmp has been identified as
a critical event in cell injury during ischemia.43 Mitochon-
drial membrane potential restoration after protonophore
FCCP-induced partial collapse was measured directly by
recording rhodamine 123 fluorescence,44,45 and results are
shown in Figure 3. FCCP at 0.2 μM induced a 50% decrease
of the mmp, resulting in a dequenching of rhodamine 123
and thereby increasing the detectable fluorescence in the
mitochondrial preparations. Compound 11 almost comple-
tely prevented this loss of mmp at 0.8 μM. Compound 13 was
also very potent in preserving mmp under these conditions.
This demonstrates that pretreatment with extremely low
concentrations of amide nitrones prevents FCCP-induced
decrease of proton potential and results in reenergized
mitochondria that are able to sustain ATP production. These
findings suggest a direct effect of the amide nitrones on
mitochondrial energy metabolism by preventing uncoupling.
Lee et al. reported that the inhibitory effect of ciclopirox
(a synthetic antifungal agent) on H2O2-induced depolariza-
tion of mmp is not caused by its antioxidant properties
because it does not scavenge H2O2, hydroxyl, and superoxide
radicals or peroxynitrite.45 These findings indicate that
amphiphilic amide nitrones are much more than just radical
scavenging antioxidants and may act directly on mitochon-
drial electron and proton transport.
Figure 3. Increase of the membrane potential in isolated rat brain
mitochondria after protonophore-induced collapse by treatment
with 11 (1) and 13 (b). Mitochondrial preparations were harvested
30 min after incubation with nitrones. The protonophore FCCP was
added 15 min after incubation with nitrones. The findings are
presented as the mean ( SEM of at least four independent mito-
chondrial preparations.
Table 4. Complex I Activity in Rat Cortical Mitochondriaa
complex 1 activity
((nmol/mg protein)/min)
compd (10 μM)
control
PBN
99 ( 1.2
107 ( 1.1
116 ( 1.3
124 ( 1.9
24 ( 0.4
60 ( 1.3
79 ( 1.2
110 ( 1.6
11 ( 0.3
13 ( 0.4
19 ( 0.5
31 ( 0.7
13
11
dopamine
dopamine þ PBN
dopamine þ 13
dopamine þ 11
15
15 þ PBN
15 þ 13
15 þ 11
a Complex I activity in rat cortical mitochondrial preparations was
determined by measuring the reduction of ferric cyanide in the presence
of antimycin A and azide as previously published.20,50 Submitochondrial
particles were incubated for 30 min with nitrone agents and dopamine.
The findings are presented as the mean ( SEM (N = 10).
mitochondrial respiration in a dual manner, via the for-
mation of ROS or by direct interaction with complex I.
Ben-Shachar et al. have shown the ability of dopamine to
inhibit mitochondrial function through its interaction with
complex I in disrupted brain mitochondria46 as well as in
intact viable human neuroblastoma SH-SY5Y cells.47-49
Indeed, in intact cells while dopamine inhibits respiration
through complex I, no inhibition was observed through
complexes II and III. Moreover, the ability of dopamine to
specifically inhibit complex I activity in disrupted mitochon-
dria, possibly interacting at a site located between the bind-
ing site for NADH and the iron-sulfur clusters but not that
of complexes II, IV, and V, further supports this hypothesis.
The relief of dopamine-induced inhibition of complex I
activity by nitrone amides 11 and 13 was assayed by the ferric
cyanide reduction method20,50 and compared to that of PBN
(Table 4).20 The two amphiphilic nitrones and the parent
compound PBN induce a slight increase of the complex
activity compared to the control, demonstrating the non-
toxicity of these agents on the complex I. Dopamine inhibits
complex I activity immediately, leading to a strong toxicity,
and this inhibition is relieved by pretreatement with the
Determination of Complex I Activity. The mechanisms by
which dopamine inhibits complex I are not fully identified
and characterized. Yet dopamine is believed to interfere with

4856 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13
Durand et al.
nitrone compounds but fully only by the amide nitrone 11.
This shows that nitrone agents are able to prevent dopamine-
induced complex I toxicity, with the amphiphilic amide
nitrones being more efficient. Once again, we observed that
the positions of the hydrophilic and lipophilic substituents
on the nitronyl group play an important role in the bio-
activity of this series of amide nitrones, with the reverse analogue
11 being ∼40% more potent than its analogue 13 in prevent-
ing dopamine toxicity. Interestingly, dopamine did not
induce an increase of hydroperoxide levels (data not shown).
These findings indicate that the effects of such mitochondrial
metabolism modifiers not only are mediated by their anti-
oxidant properties but also may be related to many other
effects like direct interactions with a key component of the
mitochondrial respiratory chain, the complex I. Indeed,
Ben-Shachar et al. found that neither the antioxidant BHT
nor the monoamine oxidase A and B inhibitors prevented
dopamine-induced inhibition of complex I activity.47 More-
over, the dopamine-induced dissipation of mitochondrial
membrane potential in intact cells, as demonstrated by
confocal microscopy and flow cytometry, was not associated
with the production of ROS and was without any effect on
cell survival.48,49 This suggests that dopamine oxidation is
not involved in the inhibition of complex I activity and
further confirms the nonantioxidant mediated protection
by amphiphilic amide nitrones.
Previous work by Singer et al. demonstrated that dopa-
mine inhibitory effect is detected in the presence of ferry-
cyanide, which accepts electrons from Fe-S clusters.51,52 This
suggests that dopamine may interact at a site that lies
between the binding site for NADH and the Fe-S cluster
N1, before the more hydrophobic part of the complex.47 This
hypothesis is further supported by the work by Harmon and
Crane who observed that NADH dehydrogenase activity is
inhibited by iron chelators, likely by displacing the labile
sulfur of the Fe-S cluster and chelating the iron.53 As shown
by Lenaz et al., capsaicin, a class B inhibitor, completely
prevents ROS production from complex I.54 Like capsaicin,
dopamine may bind to Fe-S clusters in complex I blocking
the reducing center or inducing a conformational change
making it less accessible, which in turn could prevent electron
escape from complex I. Moreover, preliminary findings
suggest that nitrones might displace [3H]dopamine from its
mitochondrial complex I binding site (unpublished results).
This would result in a reversal of complex I inhibition by
nitrones as observed in Table 4, acting as ligands of the Fe-S
centers at the same sides also targeted by dopamine and
capsaicin. By increasing and preserving complex I activity,
amide nitrones and to a lesser extend PBN may increase
energy metabolism efficacy. In this regard, the effects of the
nitrones on mitochondrial respiration and oxidative phos-
phorylation have to be investigated further in future studies
involving a broad spectrum of additional methods such as
oxygen consumption and ATP formation. As specific mito-
chondrial metabolism modifiers, amphiphilic nitrones may
constitute a new unique class of bioenergetic agents that
target mitochondria and mediate their effects by directly
modulating the activity of the respiratory chain and its key
components such as complex I.
agreement with a previous work,20 that lipophilic nitrone
agents may induce acute toxicity. Moreover, none of the
PBN and nitrones 11 and 13 were able to significantly relieve
the 15-induced complex I inhibition, as it was observed with
the dopamine. This suggests that the inhibition of complex I
activity induced by treatment with the lipophilic 15 may
proceed through a different mechanism than that of dopa-
mine. The understanding of such toxicity will be further
investigated.
Conclusion
We have synthesized a novel series of amphiphilic amide
nitrones in which a lactobionic acid based polar headgroup
and an alkyl chain are both linked via an amide bond to the
N-tert-butyl group and the phenyl ring of the PBN moiety,
respectively. To explore the importance of amphiphilicity, the
hydrogenated derivative, called 11, and its fluorinated ana-
logue, called 12, weresynthesized. Forthe sake of comparison,
physical-chemical and biological properties of 11 and 12
were compared to those previously synthesized structural
isomers, 13 and 14, respectively, in which the positions of
the polar head and the alkyl chain on the PBN moiety are
reversed. Experimental and theoretical data showed that
substitution of the PBN by hydrophilic and lipophilic groups
alters its redox properties, with the amphiphilic amide ni-
trones being easier to oxidize and reduce than the parent PBN.
Moreover, we found that the position of the substituents
on the nitronyl group is also of importance for their redox
behavior, with the reverse analogue 11 exhibiting better
reducing properties than 13. Very high protective effects were
observed in in vitro and in vivo experiments for the hydro-
genated derivatives 11 and 13, while the fluorinated com-
pounds were found to be less potent. Specific interactions of
hydrogenated amide nitrones with the mitochondria were also
demonstrated. Compounds 11 and 13 decreased peroxide
formation, reduced electron leakage, and maintained mito-
chondrial membrane potential at nanomolar concentration.
These findings indicate that amphiphilic amide nitrones are
more than just radical scavenging antioxidants but may act as
a new class of bioenergetic agents targeted to the mitochon-
drial electron and proton transport chain. Thus, by increasing
and preserving complex I activity, they may increase energy
and oxygen metabolism efficacy and hence protect the
cell against oxidative insults maintaining its function and
viability.
Experimental Section
Synthesis. All starting materials were purchased from either
Sigma-Aldrich or Acros except 1H,1H,2H,2H perfluorooctyl
iodide from Elf Atochem. All solvents were distilled and dried
according to standard procedures. TLC analyses were per-
formed on aluminum sheets precoated with silica gel 60F254
(Merck). Compound detection was achieved by exposure to UV
light (254 nm), by spraying a 5% sulfuric acid solution in
methanol or 5% ninhydrin solution in ethanol, and then by
heating at 150 ꢀC. Flash chromatography was carried out on
Merck silica gel Geduran Si 60 (0.063-0.200 mm), and size
exclusion chromatography was carried out on Sephadex LH
20 or Sephadex G-15 (Amersham Biosciences). Optical rota-
tions were determined with a Perkin-Elmer MC241 polarimeter.
Mass spectra were recorded on a triple quadripolar spectro-
meter API III Plus Sciex for ESIþ. Melting points were mea-
sured on an Electrothermal IA9100 apparatus and have not
been corrected. The purity of all final amphiphilic amide
Very surprisingly the lipophilic amide nitrone 15
(EPBNAH)20 (Figure 1) was found to strongly inhibit com-
plex I activity, thereby disrupting the rate-limiting step of
mitochondrial respiration (Table 3), while the other nitrones
were found to be devoid of any toxicity. This indicates, in

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4857
nitrones used in biological testing was assessed by reverse phase
HPLC with UV detection, confirming >95% purity. HPLC
analysis were performed on a Varian apparatus equipped with a
Microsorb C18 column (5 μm, 4.6 mm ꢀ250 mm i.d.) at 0.8 mL/
min. UV/vis spectra were recorded on a Cary Win Varian
spectrophotometer with a double-compartment quartz cell of
4.36 ꢀ 10-3 mol, 1.1 equiv), compound 421 (0.70 g, 3.95 ꢀ 10-3
mol, 1 equiv), DCC (0.97 g, 4.70 ꢀ 10-3 mol, 1.2 equiv), and a
catalytic amount of HOBt were dissolved in DCM with DIEA
(pH ∼9). After 48 h of being stirred at room temperature, the
mixture was filtered over a pad of Celite and the solvent was
concentrated under vacuum. Purification by flash chromato-
graphy and elution with EtOAc/cyclohexane (2:8 v/v) gave
compound 6 (1.13 g, 2.04 ꢀ 10-3 mol, 52% yield) as a white
1
10 mm length (Suprasil). The H, 13C, and 19F NMR spectra
€
were recorded on a Bruker AC-250 spectrometer at 250, 66.86,
1
powder. Rf = 0.57 in EtOAc/cyclohexane (6:4 v/v). H NMR
and 235 MHz, respectively. Chemical shifts are given in parts per
million (ppm) relative to the solvent residual peak as a hetero-
nuclear reference for 1H and 13C. Abbreviations used for signal
patterns are as follows: s, singlet; d, doublet; t, triplet; q, quartet;
m, multiplet; dd, doublet of doublet.
(CDCl3) δ 7.47 (2H, d, J = 8.0 Hz), 7.30 (2H, d, J = 8.0 Hz),
5.97 (1H, m), 5.81 (1H, s), 4.45 (2H, d, J = 5.7 Hz), 4.25-3.90
(4H, m), 2.70-2.30 (4H, m). 13C NMR (CDCl3) δ 169.7 (CO),
138.9, 137.5 (C), 128.1, 127.0, 103.4 (CH), 65.3, 43.6, 33.9, 26.4
(CH2). 19F NMR (CDCl3) δ -80.7 (3F, CF3), -114.1 (2F, CF2),
-121.8 (2F, CF2), -122.8 (2F, CF2), -123.5 (2F, CF2), -126.1
(2F, CF2).
N-(4-Formylbenzyl)octanamide (7). At 10 ꢀC, compound 5
(0.92 g, 3.01 ꢀ 10-3 mol) and a catalytic amount of p-toluene-
sulfonic acid were dissolved in 15 mL of ethanol. After 14 h of
being stirred at 10 ꢀC the solvent was concentrated under
vacuum and the crude mixture was dissolved in EtOAc. The
organic phase was washed with saturated NaHCO3 solution
(2ꢀ), brine (2ꢀ), dried over Na2SO4, then concentrated under
vacuum to give compound 7 (0.75 g, 2.87 ꢀ 10-3 mol, 95% yield)
as a beige powder. Rf = 0.47 in EtOAc/cyclohexane (6:4 v/v). 1H
NMR (CDCl3) δ 9.98 (1H, s), 7.84 (2H, d, J = 8.1 Hz), 7.42 (2H,
d, J = 8.0 Hz), 6.18 (1H, m), 4.50 (2H, d, J = 5.9 Hz), 2.26 (2H,
t, J = 7.6 Hz), 1.63 (2H, m), 1.29 (8H, m), 0.88 (3H, t, J = 6.8
Hz). 13C NMR (CDCl3) δ 191.9, 173.0 (CO), 145.7, 135.5 (C),
130.1, 128.1 (CH), 43.1, 36.8, 31.7, 29.3, 29.0, 25.8, 22.6 (CH2),
14.1 (CH3).
N-(Octa-O-acetyllactobionyl)-2-methyl-2-nitropropanamide (2).
A solution of compound 114 (0.97 g, 6.27 ꢀ 10-3 mol, 1 equiv)
and lactobionolactone (2.72 g, 7.99 ꢀ 10-3 mol, 1.2 equiv) in
methoxyethanol with TEA (pH ∼9) was stirred at 55 ꢀC for 20 h.
The solvent was evaporated under vacuum, and a solution of
Ac2O/pyridine (1:1 v/v) was added dropwise to the residue at
0 ꢀC. After 12 h of being stirred at ambient temperature, the
mixture was poured into cold 1 N HCl and extracted with
CH2Cl2 (3ꢀ). The organic layer was washed with brine (3ꢀ),
dried over Na2SO4, and concentrated in vacuum. Purification
by flash chromatography and elution with EtOAc/cyclohexane
(4:6 v/v), gave compound 2 (3.23 g, 4.06 ꢀ 10-3 mol, 65% yield)
as a white foam. Rf = 0.18 in EtOAc/cyclohexane (6:4 v/v). 1H
NMR (CDCl3) δ 6.64 (1H, t, J = 6.2 Hz), 5.55 (2H, m), 5.40
(1H, m), 5.25-4.90 (3H, m), 4,64 (1H, d, J = 7.7 Hz), 4.49 (1H,
m), 4.35-3.85 (5H, m), 3.73 (2H, m), 2.30-1.90 (24H, m), 1.58
(3H, s), 1.57 (3H, s). 13C NMR (CDCl3) δ 170.5, 170.5, 170.2,
170.1, 169.9, 169.7, 169.6, 169.2, 168.0 (CO), 101.8 (CH), 88,5
(C), 77.6, 71.8, 71.1 71.0, 69.8, 69.1, 68.9, 66.8 (CH), 61.6, 61.0,
46,0 (CH2), 24.0, 20.8, 20.8, 20.7, 20.7, 20.6, 20.6 (CH3).
N-(Octa-O-acetyllactobionyl)-2-methyl-2-hydroxylaminopro-
panamide (3). At 0 ꢀC, compound 2 (3.32 g, 4.18 ꢀ 10-3 mol,
1 equiv) and NH4Cl (0.32 g, 5.98 ꢀ 10-3 mol, 1.4 equiv) were
dissolved in a THF/H2O (3:1 v/v) mixture while stirring. Zinc
powder (1.08 g, 1.65 ꢀ 10-2 mol, 4 equiv) was added as a small
fraction in order to keep the temperature below 30 ꢀC. After 1 h
of being stirred at room temperature, the mixture was filtered
over a pad of Celite, the zinc salt was washed with hot THF, and
the solvents were concentrated under vacuum. Purification by
flash chromatography and elution with EtOAc/cyclohexane (8:2
v/v) gave compound 3 (2.60 g, 3.33 ꢀ 10-3 mol, 80% yield). Rf =
0.52 in EtOAc/CH3OH (9:1 v/v). [R]2D0 þ27.0 (c, l, CH2Cl2). 1H
NMR (CDCl3) δ 6.62 (1H, t, J = 6.0 Hz), 5.65-5.50 (2H, m),
5.39 (1H, d, J = 3.2 Hz), 5.25-4.95 (3H, m), 4.68 (1H, d, J = 8.0
Hz), 4.60-4.45 (2H, m), 4.35 (1H, dd, J = 3.3 Hz and J = 6.8
Hz), 4.30-3.85 (5H, m) 3.27 (2H, m), 2.25-1.90 (24, m), 1.06
(6H, s). 13C NMR (DMSO-d6) δ 170.7, 170.3, 170.1, 170.1,
170.0, 169.6, 169.6, 166.7 (CO), 101.1, 78.4, 72.1, 70.8, 70.2,
70.0, 69.5, 69.3, 67.6 (CH), 61.7, 61.4, 57.4 (CH2), 45.4 (C), 22.8,
22.7, 21.1, 21.0, 20.9, 20.8 (CH3).
N-(4-Formylbenzyl)-2H,2H,3H,3H-perfluorononamide (8).
The synthetic procedure was essentially the same as for com-
pound 7. Compound 6 (1.13 g, 2.04 ꢀ 10-3 mol) was reacted to
give compound 8 (0.97 g, 1.90 ꢀ 10-3 mol, 93%) as a white
1
powder. Rf = 0.57 in EtOAc/cyclohexane (6:4 v/v). H NMR
(CDCl3) δ 10.02 (1H, s), 7.86 (2H, d, J = 8.2 Hz), 7.45 (2H, d,
J = 8.2 Hz), 6.05 (1H, m), 4.56 (2H, d, J = 6.0 Hz), 2.58 (4H, m).
13C NMR (CDCl3) δ 191.8, 169.9 (CO), 144.8, 135.8 (C), 130.2,
128.1 (CH), 43.5, 33.9, 26.7 (CH2). 19F NMR (CDCl3) δ -80.7
(3F, CF3), -114.1 (2F, CF2), -121.8 (2F, CF2), -122.8 (2F,
CF2), -123.5 (2F, CF2), -126.1 (2F, CF2).
N-[1,1-Dimethyl-1-(octa-O-acetyl-lactobionamidomethyl)]-r-
[4-octanamidomethyl)phenyl]nitrone (9). Compound 3 (0.66 g,
8.45 ꢀ 10-4 mol, 0.6 equiv) and compound 7 (0.368 g, 1.41 ꢀ
10-3 mol, 1 equiv) were dissolved in THF/AcOH mixture (3:2
v/v) under argon. After the mixture was stirred for 2 h at 60 ꢀC
under argon, compound 3 (0.443 g, 5.68 ꢀ 10-4 mol, 0.4 equiv)
was added and the mixture was heated overnight. Then more of
the compound 3 (0.110 g, 1.41 ꢀ 10-4 mol, 0.1 equiv) was
added and the heating was continued for 2 h. The solvent was
concentrated under vacuum. Purification by flash chromato-
graphy and elution with EtOAc/cyclohexane (7:3 v/v) followed
by size exclusion chromatography using CH2Cl2/CH3OH (1:1
v/v) gave compound 9 (0.98 g, 9.57 ꢀ 10-4 mol, 67% yield) as a
white powder. Rf = 0.25 in EtOAc. [R]2D0 þ44.6 (c, l, CH2Cl2).
1H NMR (CDCl3) δ 8.25 (2H, d, J = 8.3 Hz), 7.49 (1H, s), 7.35
(2H, d, J = 8.3 Hz), 7.17 (1H, m), 5.96 (1H, m), 5.51 (2H, m),
5.29 (1H, m), 5.15 (1H, m), 4.98 (2H, m), 4.68 (1H, d, J = 7.9
Hz), 4.55-4.40 (3H, m), 4.33 (1H, m), 4.25-3.85 (4H, m),
3.85-3.50 (2H, m), 2.26 (2H, t, J = 7.6 Hz), 2.20-1.85 (24H,
m), 1.67 (2H, m), 1.59 (3H, s), 1.57 (3H, s), 1.29 (8H, m), 0.87
(3H, t, J = 6.8 Hz). 13C NMR (CDCl3) δ 173.2, 170.4, 170.2,
169.8, 168.7, 169.6, 169.2, 167.7 (CO), 141.5 (C), 131.9 (CH),
129.4 (C), 129.4 127.7, 101.8, 77.7 (CH), 73.4 (C), 72.1, 71.1,
70.9, 69.8, 68.9, 66.8 (CH), 61.6, 60.9, 47.1, 43.2, 36.8, 31.7,
29.3, 29.0, 25.8 (CH2), 25.1, 24.3 (CH3), 22.6 (CH2), 20.9, 20.8,
20.7, 20.6, 20.5, 20.4, 14.1 (CH3). MS (ESIþ, m/z): 1024.8
[M þ H]þ, 1041.8 [M þ NH4]þ, 1046.7 [M þ Na]þ, 1062.7
[M þ K]þ.
N-[4-(1,3-Dioxacyclopent-2-yl)benzyl]octanamide (5). Octa-
noic acid (1.74 g, 12.08 ꢀ 10-3 mol, 1.2 equiv), compound 4
(1.80 g, 10.17 ꢀ 10-3 mol, 1 equiv), and BOP (5.75 g, 13.01 ꢀ
10-3 mol, 1.3 equiv) were dissolved in DCM with DIEA (pH
∼9). After 48 h of being stirred at room temperature, the solvent
was concentrated under vacuum. Purification by flash chroma-
tography and elution with EtOAc/cyclohexane (2:8 v/v) gave
compound 5 (2.60 g, 8.51 ꢀ 10-3 mol, 84% yield) as a beige
1
powder. Rf = 0.47 in EtOAc/cyclohexane (6:4 v/v). H NMR
(CDCl3) δ 7.47 (2H, d, J = 8.0 Hz), 7.30 (2H, d, J = 8.0 Hz),
5.81 (2H, m), 4.45 (2H, d, J = 5.7 Hz), 4.2-3.9 (4H, m), 2.21
(2H, t, J = 7.6 Hz), 1.63 (2H, m), 1.29 (8H, m), 0.88 (3H, t, J =
6.8 Hz). 13C NMR (CDCl3) δ 173.0 (CO), 139.7, 137.0 (C),
127.9, 126.8, 103.5 (CH), 65.3, 43.3, 36.8, 31.7, 29.3, 29.0, 25.8,
22.6 (CH2), 14,1 (CH3).
N-[4-(1,3-Dioxacyclopent-2-yl)benzyl]-2H,2H,3H,3H-perfluoro-
nonamide (6). 1H,1H,2H,2H-Perfluorononanoic acid (1.71 g,

4858 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13
Durand et al.
N-[1,1-Dimethyl-1-(octa-O-acetyllactobionamidomethyl)]-r-
[4-(2H,2H,3H,3H-perfluorononamidomethyl)phenyl]nitrone (10).
The synthetic procedure was essentially the same as for com-
pound 9. Compound 3 (1.03 g, 1.32 ꢀ 10-3 mol, 1.1 equiv) and
compound 8 (0.615 g, 1.21 ꢀ 10-3 mol, 1 equiv) were reacted to
give compound 10 (1.162 g, 9.13 ꢀ 10-4 mol, 75% yield) as a
white powder. Rf = 0.27 in EtOAc. [R]2D0 þ36.2 (c, l, CH2Cl2). 1H
NMR (CDCl3) δ 8.24 (2H, d, J = 8.3 Hz), 7.50 (1H, s), 7.34 (2H,
d, J = 8.3 Hz), 7.22 (1H, m), 6.35 (1H, m), 5.60-5.40 (2H, m),
5.29 (1H, d, J = 2.9 Hz), 5.25-5.10 (1H, m), 5.05-4.85 (2H, m),
4.68 (1H, d, J = 8.0 Hz), 4.50 (2H, d, J = 5.7 Hz) 4.45-4.25 (2H,
m), 4.22-3.85 (6H, m), 3.85-3.45 (2H, m), 2.56 (2H, m),
2.20-1.90 (24H, m), 1.59 (3H, s), 1.57 (3H, s). 13C NMR
(CDCl3) δ 170.5, 170.4, 170.2, 170.1, 169.9, 169.8, 169.7, 169.3,
167.6 (CO), 140.8 (CH), 131.7 (CH), 129.6 (C), 129.4, 127.7,
101.8, 77.9 (CH), 73.4 (C), 72.6, 71.9, 71.2, 70.8, 69.8, 68.9, 66.8
(CH), 61.6, 61.0 (CH2), 47.2, 43.4, 26.9, 26.7 (CH2), 25.1, 24.3,
20.8, 20.7, 20.6, 20.5, 20.4 (CH3). 19F NMR (CDCl3) δ -80.7 (3F,
CF3), -114.1 (2F, CF2), -121.8 (2F, CF2), -122.8 (2F, CF2),
-123.5 (2F, CF2), -126.2 (2F, CF2). MS (ESIþ, m/z): 1272.6 [M
þ H]þ, 1289.6 [M þ NH4]þ, 1294.6 [M þ Na]þ, 1310.6 [M þ K]þ.
N-[1,1-Dimethyl-1-(lactobionamidomethyl)]-r-[4-(octanamido-
methyl)phenyl]nitrone (11). Compound 9 (0.500 g, 4.88 ꢀ 10-4
mol) and a catalytic amount of sodium methoxide (pH ∼8) were
dissolved in MeOH. After the mixture was stirred for 4 h, 1 N
HCl solution was added dropwise to neutralize the mixture and
the solvents were concentrated under vacuum. Purification by
size exclusion chromatography using water followed by lyophi-
lization gave compound 11 (0.25 g, 3.63 ꢀ 10-4 mol, 75% yield)
as a white and hygroscopic foam. Rf = 0.38 in EtOAc/MeOH/
H2O (7:2:1 v/v/v). [R]D20 þ10.6 (c, l, DMSO). UV (CH3OH):
extrapolated to 100% aqueous to give the log k0W values listed in
Table 1.
Surface Tension Measurements. Nitrone solutions were pre-
pared by using Milli-Q water (resistivity of 18.2 MΩ cm; surface
3
tension of 72.8 mN/m). All measurements were carried out at
25 ꢀC. The surface tensions of nitrone solutions were measured
€
with a Kruss K12ST tensiometer controlled by the Kruss
Labdesk software (Kruss, Germany) by the Wilhelmy plate
€
€
technique. The critical micelle concentration (cmc) and the
surface tension (γcmc) were determined from the breakpoint of
the surface tension and logarithm of concentration curve.
Radical Scavenging. Radical scavenging was investigated
using a scavenger competition assay based on the burstlike
formation of the intensely green colored 2,20-azino-bis(3-ethyl-
benzthiazoline-6-sulfonic acid) ABTS cation radical initiated by
a Fenton reaction.55 Measurements were made at 420 nm and
monitored for a period of 30 min in the absence or presence of
scavenging competitor at 1 mM dissolved in DMSO. The
reaction system consisted of 0.9 mL of distilled water, 0.15 mL
of ABTS at 1 mM, 0.15 mL of FeSO4 at 0.5 mM, 0.15 mL of the
test compound at 1 mM dissolved in DMSO, and 0.15 mL of
H2O2 at 10 mM. Controls received the DMSO vehicle only, with
the solvent reaching a final concentration of 10% in the assay.
The reaction was started with the addition of H2O2 and the
rapidly increasing concentration of the ABTS cation radicals
was determined every minute. Radical scavengers suppressed
this time dependent ABTS cation radical formation by competi-
tion.
Radical Reduction. ABTS cation radical reduction was as-
sayed according to Re et al.25 with the following modifica-
tions:24 assay concentration of the ABTS cation radical was
105 μM, final concentration of the test compounds was 0.5 mM.
Controls received DMSO vehicle only, reaching a final concen-
tration of 2.5% in the assay. The reactions were started with the
addition of the nitrone agent, and reduction of the ABTS cation
radical was monitored every minute at 420 nm for 30 min.
Cyclic Voltammetry. Cyclic voltammetry was performed on a
potentiostat and computer-controlled electroanalytical system.
Electrochemical measurements were carried out in a 10 mL cell
equipped with a glassy carbon working electrode, a platinium-
wire auxiliary electrode, and a Ag/AgCl reference electrode.
Solutions were degassed by bubbling with nitrogen gas before
λ
max = 298 nm. 1H NMR (DMSO-d6) δ 8.35 (3H, m), 7.82 (1H,
s), 7.57 (1H, t, J = 6.2 Hz), 7.29 (2H, d, J = 8.3 Hz), 5.45-3.10
(25H, m), 2.15 (2H, t, J = 7.3 Hz), 1.49 (8H, m), 1.25 (8H, s),
0.87 (3H, t, J = 6.6 Hz). 13C NMR (CD3OD) δ 174.4, 174.2
(CO), 142.4 (C), 135.1, 129.9 (CH), 129.2 (C), 127.1, 104.3, 81.7,
75.8 (CH), 73.8 (C), 73.3, 72.5, 71.7, 71.3, 71.1, 68.9 (CH), 62.3,
61.3, 45.7, 42.4, 35.7, 31.5, 28.9, 28.7, 25.7 (CH2), 23.5, 23.4
(CH3), 22.3 (CH2), 13.0 (CH3). MS (ESIþ, m/z): 688.6 [M þ
H]þ, 705.5 [M þ NH4]þ, 710.6 [M þ Na]þ, 726.5 [M þ K]þ.
HR-MS (ESIþ, m/z) for [C32H53N3O13 þ H]þ: calcd 688.3651,
found 688.3630. RP-HPLC tR =10.05 min (CH3OH/H2O
60:40 v/v).
N-[1,1-Dimethyl-1-(lactobionamidomethyl)]-r-[4-(2H,2H,3H,3H-
perfluorononamidomethyl)phenyl]nitrone (12). The synthetic pro-
cedure was essentially the same as for compound 11. Compound
10 (0.5 g, 3.93 ꢀ 10-4 mol) was reacted to give compound 12 (0.33
g, 3.53 ꢀ 10-4 mol, 90% yield) as a white and hygroscopic foam.
[R]2D0 þ6.1 (c, l, DMSO). UV (CH3OH): λmax = 298 nm. 1H NMR
(DMSO-d6) δ 8.60 (1H, m) 8.32 (2H, d, J = 8.3 Hz), 7.83 (1H, s),
7.56 (1H, m), 7.31 (2H, d, J = 8.3 Hz), 5.40-3.20 (25H, m),
2.80-2.40 (4H, m), 1.48 (6H, s). 13C NMR (DMSO) δ 173.1,
169.9 (CO), 141.6 (C), 131.0 (CH), 130.4 (C), 129.2, 127.4, 105.1,
83.5, 76.1 (CH), 73.8 (C), 73.7, 72.4, 71.8, 71.5, 71.0, 68.7 (CH),
62.8, 61.1, 46.2, 42.6, 26.3 (CH2), 24.8, 24.7 (CH3). 19F NMR
(DMSO) δ -80.2(3F, CF3), -113.5(2F, CF2), -121.7 (2F, CF2),
-122.6 (2F, CF2), -123.1 (2F, CF2), -125.7 (2F, CF2). MS
(ESIþ, m/z): 936.5 [M þ H]þ, 958.5 [M þ Na]þ, 974.5 [M þ K]þ.
RP-HPLC tR =11.99 min (CH3OH/H2O 70:30 v/v).
Determination of log k0W Values. Methanol solutions of the
nitrones (1.0 mg/mL) were injected into an HPLC equipped with
a Microsorb C18 column (250 mm ꢀ 4.6 mm, 5 μm). The
nitrones were eluted using a MeOH/water mixture (9:1 to 7:3
v/v) using a flow rate of 0.8 mL/min. The column temperature
was 23 ꢀC, and the UV detector wavelength was λ = 298 nm.
The log k0 values were calculated by using the equation log k0 =
log((t - t0)/t0), where t is the retention time of the nitrone and t0
is the elution time of MeOH, which is not retained on the
column. Linear regression analysis (r2>0.997) was performed
on the three data points for each nitrone and the resulting line
recording the voltammograms at a scan rate (ν) of 0.15 V s-1
.
Nitrone concentration was 1 mM in doubly distilled water, and
3
the supporting electrolyte was 0.15 M NaCl.
Computational Methods. All calculations were performed at
the Ohio Supercomputer Center. The minimization of initial
structures using MMFF9456 was performed with MacroModel
9.6.57 Conformational search was then carried out using
MMFF9456 via the Monte Carlo multiple minimum method
coupled with generalized Born/surface area (GB/SA) conti-
nuum solvation model using water as the solvent58 as im-
plemented in the MacroModel package. Density functional
theory59,60 was further applied in this study to determine the
optimized geometry of each species61-64 at the B3LYP/6-31G(d)
level using Gaussian 03.65 Electron spin densities (populations)
were obtained from a natural population analysis (NPA) ap-
proach using single-point energies evaluated at the B3LYP/
6-31Gþ(d,p) level with 6d functions.66 The effect of solvation
on the gas-phase calculations was also investigated using the
polarized continuum model (PCM).67-70 Only the bottom-of-
the-well energies were considered in thisstudy, since the size of the
molecules limits the calculation of the vibrational frequencies.
Primary Cortical Mixed Cell Cultures. Mixed cortical cell
cultures from E15 Sprague-Dawley rat embryos were prepared,
cultured, and treated as described by Law et al.71 The cultures
were exposed to toxins (200 μM) and antioxidant agents (10 μM)
for 24 h. Cell death was quantified by the trypan blue absor-
bance assay.72 The absorbance of the dye was measured at
590 nm.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4859
Rotifers. Survival of rotifers was investigated in standardized
brane potential was measured directly in the preparations by
recording the quenching of rhodamine 123 fluorescence due to
the accumulation of the dye in respiring mitochondria and the
dequenching after addition of the protonophore FCCP due to
a partial collapse of the membrane potential, release of rhoda-
mine 123, and a resulting increase in detectable rhodamine
123 fluorescence according to Lee et al.45 with the modifica-
tions as described above in the paragraph on incubation
procedures. Mitochondrial membrane potential were calcu-
lated according to the methods as described by Scaduto and
Grotyohann.44
Complex I Activity. Ferric cyanide reductase activity in sub-
mitochondrial particles was measured by monitoring the reduc-
tion of ferric cyanide from the wavelength difference at 440 and
490 nm in the presence of antimycin and azide20,50 at an
incubation temperature of 37.5 ꢀC. This recently developed
method allows for a specific and sensitive measurement of
mitochondrial iron sulfur cluster N2 activity in complex I, as
it is completely abolished in the presence of the specific ligand
and inhibitor capsaicin when added at 50 μM. Dopamine added
at equimolar concentration was almost as effective as capsaicin
in acting as a specific antagonist at this site of the respiratory
chain. Preparations were preincubated for 30 min with the
nitrone agent or the dopamine or the addition of both com-
pounds prior to the measurement of ferric cyanide reduction.
Basal activity was 1200 (nmol/min)/mg protein.
viability assays as described previously.18 Experiments on the
controls receiving vehicle only resulted in zero percentage
survival and were carried out in parallel on animals of the same
stock culture to ensure absolutely identical exposure conditions
for all animals.
Rat Brain Mitochondrial Preparations. The mitochondrial
preparations from the forebrain of 3-month-old Sprague-
Dawley rats were obtained from Charles River Laboratories
(Wilmington, MA) fed at libitum and housed under a 12-12 h
light-dark cycle. The mitochondrial preparations were isolated
and prepared as previously described.40 The rat brain mitochon-
drial preparations containing at least 80% of structurally fully
conserved and intact mitochondria with very little debris present
were resuspended to allow for respiration with the physiological
substrates glutamate and malate in phosphate buffered Tris-
KCl medium: 10 mM Tris-HCl, pH 7.4, and 5 mM potassium
phosphate buffer, pH 7.4, with 100 mM KCl and 1.5 mM
MgCl2.
Incubation Procedures Using Rat Brain Mitochondrial Prep-
arations. Incubation with the nitrones and respiration in the
mitochondrial preparations were started with following sub-
strates: 6 mM glutamate and 6 mM malate at 37.5 ꢀC. In the
experiment on the mitochondrial membrane potential (mmp),
the protonophore FCCP was added at 0.2 μM after 15 min to
induce a 50% reduction (partial collapse of mitochondrial
membrane potential). Electron leakage, hydrogen peroxide
formation, and mitochondrial membrane potential were as-
sayed after 30 min of incubation by adding extraction solvent
(NBT), acid (hydroperoxides) or doing direct measurements on
quenching of rhodamine 123 fluoresence by mitochondria in the
preparations themselves to determine the recovery of the mmp.
Determination of Soluble Hydroperoxides Formation in Mito-
chondrial Preparations from Rat Forebrain by the FOX Assay.
Xylenol orange method (FOX assay) was used to determine
soluble hydroperoxides in mitochondria as described by Hermes-
Lima et al.38 The sensitivity and specifity of the peroxide
measurements was increased by adding 100 mM modifier
sorbitol as described by Butterfield et al.73 Mitochondrial
preparations were harvested after 30 min of incubation and
collected on ice. Then 800 μL of the mitochondrial preparations
was added to 200 μL of 0.35 M H2SO4 and homogenized in a
Potter-Elvehjem homogenizer with a tightly fitting glass pestle.
The acidic homogenate was centrifuged for 3 min at 10000g.
Soluble hydroperoxides in supernatants were determined by a
microassay variant39 to increase yield, sensitivity, and specificity
by 1 order of magnitude; 0.07 M final assay concentration of
H2SO4 was used to ensure stability and reproducible measure-
ments by obtaining an optimal pH with 250 μL of homogenate
or reagent blank containing 0.07 M H2SO4 and 750 μL of
reagent solution containing 2.5 mM FeSO4, 1 mM xylenol
orange, 100 mM sorbitol, and 0.07 M H2SO4. Additional
homogenate blanks contained a reagent solution devoid of xyl-
enol orange. After being vortexed, mixtures were incubated for
30 min at 25 ꢀC in the dark. After another centrifugation for
3 min at 10000g the Fe(III)-xylenol orange complex was
determined at 570 nm. Because of protein precipitation by the
acid used already during homogenization, the assay as applied
does not detect total peroxides but only the soluble hydroper-
oxides, which are mainly hydrogen peroxide (at least 98% in
mitochondrial preparations).
Acknowledgment. This work was supported by Associa-
tion Franc-aise contre les Myopathies (AFM) Grant No.
12674 (2005-2007). S.O. was the recipient of a fellowship
ꢀ
^
from the “Region Provence Alpes Cote d’Azur” and from
Targeting System Pharma Company. B.P. was supported by
the grant PO 662/1-1 of the German Research Foundation
(“Deutsche Forschungsgemeinschaft—DFG”). F.A.V. was
supportedby the OhioSupercomputer Centerand NIHGrant
HL81248.
Supporting Information Available: NMR and MS spectra of
compounds 9-12; HPLC chromatograms of 11 and 12; cyclic
voltammograms of 11 and 13. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Coulter, C. V.; Kelso, G. F.; Lin, T.-K.; Smith, R. A. J.; Murphy,
M. P. Mitochondrially targeted antioxidants and thiol reagents.
Free Radical Biol. Med. 2000, 28, 1547–1554.
(2) Muratovska, A.; Lightowlers, R. N.; Taylor, R. W.; Wilce, J. A.;
Murphy, M. P. Targeting large molecules to mitochondria. Adv.
Drug Delivery Rev. 2001, 49, 189–198.
(3) Murphy, M. P.; Echtay, K. S.; Blaikie, F. H.; Asin-Cayuela, J.;
Cocheme, H. M.; Green, K.; Buckingham, J. A.; Taylor, E. R.;
Hurrell, F.; Hughes, G.; Miwa, S.; Cooper, C. E.; Svistunenko,
D. A.; Smith, R. A. J.; Brand, M. D. Superoxide activates uncou-
pling proteins by generatingcarbon-centered radicals and initiating
lipid peroxidation: studies using a mitochondria-targeted spin trap
derived from R-phenyl-tert-butylnitrone. J. Biol. Chem. 2003, 278,
48534–48545.
(4) Covey, M. V.; Murphy, M. P.; Hobbs, C. E.; Smith, R. A. J.;
Oorschot, D. E. Effect of the mitochondrial antioxidant, mito
vitamin E, on hypoxic-ischemic striatal injury in neonatal rats:
a dose-response and stereological study. Exp. Neurol. 2006, 199,
513–519.
(5) Horton, K. L.; Stewart, K. M.; Fonseca, S. B.; Guo, Q.; Kelley,
S. O. Mitochondria-penetrating peptides. Chem. Biol. 2008, 15,
375–382.
(6) Zhao, K.; Luo, G.; Giannelli, S.; Szeto, H. H. Mitochondria-
targeted peptide prevents mitochondrial depolarization and apop-
tosis induced by tert-butyl hydroperoxide in neuronal cell lines.
Biochem. Pharmacol. 2005, 70, 1796–1806.
(7) Wipf, P.; Xiao, J.; Jiang, J.; Belikova, N. A.; Tyurin, V. A.; Fink,
M. P.; Kagan, V. E. Mitochondrial targeting of selective electron
scavengers: synthesis and biological analysis of hemigramicidin-
TEMPO conjugates. J. Am. Chem. Soc. 2005, 127, 12460–12461.
Determination of Electron Leakage in Mitochondrial Prep-
arations from Rat Forebrain by the Nbt Reduction Assay. Nitro
blue tetrazolium (NBT) reduction was measured as described by
Hensley et al.40 in the presence of 3 μM rotenone and 3 mM azide
with the following modifications: incubation temperature was
37.5 ꢀC, and formazan was extracted according to the method
described by Vrablic et al.74
Increase of Mitochondrial Membrane Potential in Mitochon-
drial Preparations from Rat Forebrain. Mitochondrial mem-

4860 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13
Durand et al.
(8) El Fangour, S.; Marini, M.; Good, J.; McQuaker, S. J.; Shiels,
P. G.; Hartley, R. C. Nitrones for understanding and ameliorating
the oxidative stress associated with aging. Age 2009, 31, 269–276.
(9) Ouari, O.; Polidori, A.; Pucci, B.; Tordo, P.; Chalier, F. Synthesis
of a glycolipidic amphiphilic nitrone as a new spin trap. J. Org.
Chem. 1999, 64, 3554–3556.
nonaqueous electrochemical characterizations of spin traps.
J. Phys. Chem. 1980, 84, 916–921.
(28) Tuccio, B.; Bianco, P.; Bouteiller, J. C.; Tordo, P. Electrochemical
characterisation of [beta]-phosphorylated nitrone spin traps. Elec-
trochim. Acta 1999, 44, 4631–4634.
(29) Gille, L.; Nohl, H. Analyses of the molecular mechanism of
adriamycin-induced cardiotoxicity. Free Radical Biol. Med. 1997,
23, 775–782.
(10) Geromel, V.; Kadhom, N.; Cebalos-Picot, I.; Ouari, O.; Polidori,
€
A.; Munnich, A.; Rotig, A.; Rustin, P. Superoxide-induced massive
(30) Kotamraju, S.; Konorev, E. A.; Joseph, J.; Kalyanaraman, B.
Doxorubicin-induced apoptosis in endothelial cells and cardio-
myocytes is ameliorated by nitrone spin traps and ebselen. Role of
reactive oxygen and nitrogen species. J. Biol. Chem. 2000, 275,
33585–33592.
apoptosis in cultured skin fibroblasts harboring the neurogenic
ataxia retinitis pigmentosa (NARP) mutation in the ATPase-6 gene
of the mitochondrial DNA. Hum. Mol. Genet. 2001, 10, 1221–
1228.
ꢀ
ꢀ
(11) Perino, S.; Contino-Pepin, C.; Jasseron, S.; Rapp, M.; Maurizis,
J.-C.; Pucci, B. Design, synthesis and preliminary biological eva-
luations of novel amphiphilic drug carriers. Bioorg. Med. Chem.
Lett. 2006, 16, 1111–1114.
(31) Radi, R.; Rodriguez, M.; Castro, L.; Telleri, R. Inhibition of
mitochondrial electron transport by peroxynitrite. Arch. Biochem.
Biophys. 1994, 308, 89–95.
ꢀ
(32) Estevez, A. G.; Radi, R.; Barbeito, L.; Shin, J.; Thompson, J. A.;
(12) Ortial, S.; Durand, G.; Poeggeler, B.; Polidori, A.; Pappolla, M. A.;
Boeker, J.; Hardeland, R.; Pucci, B. Fluorinated amphiphilic
amino acid derivatives as antioxidant carriers: a new class of
protective agents. J. Med. Chem. 2006, 49, 2812–2820.
(13) Durand, G.; Prosak, R. A.; Han, Y.; Ortial, S.; Rockenbauer, A.;
Pucci, B.; Villamena, F. A. Spin trapping and cytoprotective
properties of fluorinated amphiphilic carrier conjugates of cyclic
versus linear nitrones. Chem. Res. Toxicol. 2009, 22, 1570–1581.
(14) Durand, G.; Polidori, A.; Salles, J.-P.; Pucci, B. Synthesis of a new
family of glycolipidic nitrones as potential antioxidant drugs for
neurodegenerative disorders. Bioorg. Med. Chem. Lett. 2003, 13,
859–862.
(15) Durand, G.; Polidori, A.; Ouari, O.; Tordo, P.; Geromel, V.;
Rustin, P.; Pucci, B. Synthesis and preliminary biological evalua-
tions of ionic and nonionic amphiphilic alpha-phenyl-N-tert-
butylnitrone derivatives. J. Med. Chem. 2003, 46, 5230–5237.
(16) Durand, G.; Polidori, A.; Salles, J.-P.; Prost, M.; Durand, P.;
Pucci, B. Synthesis and antioxidant efficiency of a new amphiphilic
spin-trap derived from PBN and lipoic acid. Bioorg. Med. Chem.
Lett. 2003, 13, 2673–2676.
(17) Asanuma, T.; Yasui, H.; Inanami, O.; Waki, K.; Takahashi, M.;
Iizuka, D.; Uemura, T.; Durand, G.; Polidori, A.; Kon, Y.; Pucci,
B.; Kuwabara, M. A new amphiphilic derivative, N-{[4-(lactobion-
amido)methyl]benzylidene}-1,1-dimethyl-2-(octylsulfanyl)ethyla-
mine N-oxide, has a protective effect against copper-induced
fulminant hepatitis in Long-Evans cinnamon rats at an extremely
low concentration compared with its original form alpha-phenyl-
N-(tert-butyl) nitrone. Chem. Biodiversity 2007, 4, 2253–2267.
(18) Poeggeler, B.; Durand, G.; Polidori, A.; Pappolla, M. A.;
Vega-Naredo, I.; Coto-Montes, A.; Boeker, J.; Hardeland, R.;
Pucci, B. Mitochondrial medicine: neuroprotection and life exten-
sion by the new amphiphilic nitrone LPBNAH acting as a highly
potent antioxidant agent. J. Neurochem. 2005, 95, 962–973.
(19) Tanguy, S.; Durand, G.; Reboul, C.; Polidori, A.; Pucci, B.;
Dauzat, M.; Obert, P. Protection against reactive oxygen species
injuries in rat isolated perfused hearts: effect of LPBNAH, a new
amphiphilic spin-trap derived from PBN. Cardiovasc. Drugs Ther.
2006, 20, 147–149.
Beckman, J. S. Peroxynitrite-induced cytotoxicity in PC12 cells:
evidence for an apoptotic mechanism differentially modulated by
neurotrophic factors. J. Neurochem. 1995, 65, 1543–1550.
(33) Meadow, N. D.; Barrows, C. H. J. Studies on aging in a bdelloid
rotifer. I. The effect of various culture systems on longevity and
fecundity. J. Exp. Zool. 1971, 176, 303–313.
(34) Meadow, N. D.; Barrows, C. H. J. Studies on aging in a bdelloid
rotifer. II. The effects of various environmental conditions and
maternal age on longevity and fecundity. J. Gerontol. 1971, 26,
302–309.
(35) Enesco, H. E.; Verdone-Smith, C. R-Tocopherol increases lifespan
in the rotifer philodina. Exp. Gerontol. 1980, 15, 335–338.
(36) Sawada, M.; Carlson, J. C. Association between lipid peroxidation
and life-modifying factors in rotifers. J. Gerontol. 1987, 42,
451–456.
(37) Sawada, M.; Carlson, J. C. Biochemical changes associated with
the mechanism controlling superoxide radical formation in the
aging rotifer. J. Cell. Biochem. 1990, 44, 153–165.
(38) Hermes-Lima, M.; Willmore, W. G.; Storey, K. B. Quantification
of lipid peroxidation in tissue extracts based on Fe(III) xylenol
orange complex formation. Free Radical Biol. Med. 1995, 10,
271–280.
(39) Lyras, L.; Cairns, N. J.; Jenner, A.; Jenner, B.; Halliwell, B. An
assessment of oxidative damage to proteins, lipids, and DNA in
brain from patients with Alzheimer’s disease. J. Neurochem. 1997,
68, 2061–2069.
(40) Hensley, K.; Pye, Q. N.; Maidt, M. L.; Stewart, C. A.; Robinson,
K. A.; Jaffrey, F.; Floyd, R. A. Interaction of R-phenyl-N-tert-
butyl nitrone and alternative electron acceptors with complex 1
indicates a substrate reduction site upstream from the rotenone
binding site. J. Neurochem. 1998, 71, 2549–2557.
(41) Genova, M. L.; Pich Merlo, M.; Bernacchia, A.; Bianchi, C.;
Biondi, A.; Bovina, C.; Falasca, A. I.; Formiggini, G.; Parenti
Castelli, G.; Lenaz, G. The mitochondrial production of reactive
oxygen species in relation to aging and pathology. Ann. N.Y. Acad.
Sci. 2004, 1011, 86–100.
(42) Genova, M. L.; Ventura, B.; Giuliano, G.; Bovina, C.; Formiggini,
G.; Parenti Castelli, G.; Lenaz, G. The site of production of
superoxide radical in mitochondrial complex I is not a bound
ubisemiquinone but presumably iron-sulfur cluster N2. FEBS
Lett. 2001, 505, 364–368.
(43) Murakami, K.; Kondo, T.; Kawase, M.; Li, Y.; Sato, S.; Chen,
S. F.; Chan, P. H. Mitochondrial susceptibility to oxidative stress
exacerbates cerebral infarction that follows permanent focal cere-
bral ischemia in mutant mice with manganese superoxide dismu-
tase deficiency. J. Neurosci. 1998, 18, 205–213.
(20) Durand, G.; Poeggeler, B.; Boeker, J.; Raynal, S.; Polidori, A.;
Pappolla, M. A.; Hardeland, R.; Pucci, B. Fine-tuning the amphi-
philicity: a crucial parameter in the design of potent alpha-phenyl-
N-tert-butylnitrone analogues. J. Med. Chem. 2007, 50, 3976–3979.
(21) Ouari, O.; Chalier, F.; Bonaly, R.; Pucci, B.; Tordo, P. Synthesis
and spin-trapping behavior of glycosylated nitrones. J. Chem. Soc.,
Perkin Trans. 2 1998, 2299–2308.
(22) Lebaupain, F.; Salvay, A. G.; Olivier, B.; Durand, G.; Fabiano,
A. S.; Michel, N.; Popot, J. L.; Ebel, C.; Breyton, C.; Pucci, B.
Lactobionamide surfactants with hydrogenated, perfluorinated or
hemifluorinated tails: physical-chemical and biochemical charac-
terization. Langmuir 2006, 22, 8881–8890.
(44) Scadoto, R. C.; Grotyohann, L. W. Measurement of mitochondrial
membrane potential using fluorescent rhodamine derivatives. Bio-
phys. J. 1999, 76, 469–477.
(23) Kissa, E. Fluorinated Surfactants: Synthesis, Properties, Applica-
(45) Lee, S. J.; Jin, Y.; Yoon, H. Y.; Choi, B.-O.; Kim, H. C.; Oh,
Y.-K.; Kim, H.-S.; Kim, W.-C. Ciclopirox protects mitochondria
from hydrogen peroxide toxicity. Br. J. Pharmacol. 2005, 145,
469–476.
(46) Ben-Shachar, D.; Zuk, R.; Glinka, Y. Dopamine neurotoxicity:
inhibition of mitochondrial respiration. J. Neurochem. 1995, 64,
718–723.
(47) Ben-Shachar, D.; Zuk, R.; Gazawi, H.; Ljubuncic, P. Dopamine
toxicity involves mitochondrial complex I inhibition: implications
to dopamine-related neuropsychiatric disorders. Biochem. Phar-
macol. 2004, 67, 1965–1974.
(48) Brenner-Lavie, H.; Klein, E.; Zuk, R.; Gazawi, H.; Ljubuncic, P.;
Ben-Shachar, D. Dopamine modulates mitochondrial function in
viable SH-SY5Y cells possibly via its interaction with complex
I: relevance to dopamine pathology in schizophrenia. Biochim.
Biophys. Acta 2008, 1777, 173–185.
tions; Dekker: New York, 1994; Vol. 7, pp 264-282.
(24) Poeggeler, B.; Thuermann, S.; Dose, A.; Schoenke, M.; Burkhardt,
S.; Hardeland, R. Melatonin’s unique radical scavenging proper-
ties—roles of its functional substituents as revealed by a compar-
ison with its structural analogs. J. Pineal Res. 2002, 33, 20–30.
(25) Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.;
Rice-Evans, C. Antioxidant activity applying an improved ABTS
radical cation decolorization assay. Free Radical Biol. Med. 1999,
26, 1231–1237.
(26) Durand, G.; Choteau, F.; Pucci, B.; Villamena, F. A. Reactivity of
superoxide radical anion and hydroperoxyl radical with R-phenyl-
N-tert-butylnitrone (PBN) derivatives. J. Phys. Chem. A 2008, 112,
12498–12509.
(27) McIntire, G. L.; Blount, H. N.; Stronks, H. J.; Shetty, R. V.;
Janzen, E. G. Spin trapping in electrochemistry. 2. Aqueous and

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4861
(49) Brenner-Lavie, H.; Klein, E.; Ben-Shachar, D. Mitochondrial
complex I as a novel target for intraneuronal DA: modulation of
respiration in intacts cells. Biochem. Pharmacol. 2009, 78, 85–95.
(50) Moreno-Sanchez, R.; Hogue, B. A.; Bravo, C.; Newman, A. H.;
Basile, A. S.; Chiang, P. K. Inhibition of substrate oxidation in
mitochondria by the peripheral-type benzodiazepine receptor
ligand AHN 086. Biochem. Pharmacol. 1991, 41, 1479–1484.
(51) Gutman, M.; Singer, T. P. EPR studies on the iron sulfur centers of
DPNH dehydrogenase during redox cycle of the enzyme. Biochem.
Biophys. Res. Commun. 1973, 44, 1572–1578.
(52) Lusty, C. J.; Machinist, J. M.; Singer, T. P. Studies on the
respiratory chainlinked nicotinamide adenine dinucleotide dehy-
drogenase. VII. Labile sulfide group in the dehydrogenase and
related proteins. J. Biol. Chem. 1965, 240, 1804–1810.
(53) Harmon, H. J.; Crane, F. L. Inhibition of mitochondrial electron
transport by hydrophilic metal chelators. Determination of dehy-
drogenase topography. Biochim. Biophys. Acta 1976, 440, 45–48.
(54) Lenaz, G.; Fato, R.; Genova, M. L.; Bergamini, C.; Bianchi, C.;
Biondi, A. Mitochondrial complex I: structural and functional
aspects. Biochim. Biophys. Acta 2006, 1757, 1406–1420.
(55) Poeggeler, B.; Reiter, R. J.; Hardeland, R.; Tan, D.-X.; Barlow-
Walden, L. R. Melatonin and structurally-related endogenous
indoles act as potent electron donors and radical scavengers in
vitro. Redox Rep. 1996, 2, 179–184.
J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.;
Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli,
C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.;
Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck,
A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.;
Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox,
D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong,
M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision B.04;
Gaussian, Inc.: Pittsburgh PA, 2003.
(66) Freed, J. H. Theory of Slow Tumbling ESR Spectra for Nitroxides;
Academic: New York, 1976.
(67) Cossi, M.; Barone, V. Analytical second derivatives of the free
energy in solution by polarizable continuum models. J. Chem.
Phys. 1998, 109, 6246–6254.
(68) Cossi, M.; Barone, V.; Cammi, R.; Tomasi, J. Ab initio study of
solvated molecules: a new implementation of the polarizable con-
tinuum model. Chem. Phys. Lett. 1996, 255, 327–335.
(56) Halgren, T. A. Merck molecular force field. I. Basis, form, scope,
parametrization and performance of MMF94. J. Comput. Chem.
1996, 17, 490–519.
(69) Tedeschi, A. M.; D’Errico, G.; Busi, E.; Basosi, R.; Barone, V.
Micellar aggregation of sulfonate surfactants studied by electron
paramagnetic resonance of a cationic nitroxide: an experimental
and computational approach. Phys. Chem. Chem. Phys. 2002,
2180–2188.
(70) Tomasi, J.; Persico, M. Molecular interactions in solution: an
overview of methods based on continuous distributions of the
solvent. Chem. Rev. 1994, 94, 2027–2094.
(71) Law, A.; Gauthier, S.; Quirion, R. Neuroprotective and neuror-
escuing effects of isoform-specific nitric oxide synthase inhibitors,
nitric oxide scavenger, and antioxidant against beta-amyloid toxi-
city. Br. J. Pharmacol. 2001, 133, 1114–1124.
(72) Uliasz, T. F.; Hewett, S. J. A microtiter trypan blue absorbance
assay for the quantitative determination of excitoxic neuronal
injury in cell culture. J. Neurosci. Methods 2000, 100, 157–163.
(73) Butterfield, D. A.; Martin, L.; Carney, J. M.; Hensley, K.
Aβ(25-35) peptide displays H2O2-like reactivity towards aqueous
Fe2þ, nitroxide spin probes, and synaptosomal membrane pro-
teins. Life Sci. 1996, 58, 217–228.
€
(57) MacroModel, version 9.6; Schrodinger, LLC: New York, 2005.
(58) Still, W. C.; Tempczyk, A.; Hawlely, R. C.; Hendrickson, T.
Semianalytical treatment of solvation for molecular mechanics
and dynamics. J. Am. Chem. Soc. 1990, 112, 6127–6129.
(59) Labanowski, J. W.; Andzelm, J. Density Functional Methods in
Chemistry; Springer: New York, 1991.
(60) Parr, R. G.; Yang, W. Density Functional Theory in Atoms and
Molecules; Oxford University Press: New York, 1989.
(61) Becke, A. D. Density-functional exchange-energy approximation
with correct asymptotic behavior. Phys. Rev. A 1988, 38, 3098–
3100.
(62) Becke, A. D. Density-functional thermochemistry. III. The role of
exact exchange. J. Chem. Phys. 1993, 98, 5648–5652.
(63) Hehre, W. J.; Radom, L.; Schleyer, P. V.; Pople, J. A. Ab Initio
Molecular Orbital Theory; John Wiley & Sons: New York, 1986.
(64) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti
correlation-energy formula into a functional of the electron den-
sity. Phys. Rev. B 1988, 37, 785–789.
(65) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi,
(74) Vrablic, A. S.; Albright, C. D.; Craciunescu, C. N.; Salganik, R. I.;
Zeisel, S. H. Altered mitochondrial function and overgeneration
of reactive oxygen species precede the induction of apoptosis by
1-O-octadecyl-2-methyl-rac-glycero-3-phosphocholine in p53-de-
fective hepatocytes. FASEB J. 2001, 15, 1739–1744.