Analysis of the structural and mechanistic factors in antioxidants that preserve mitochondrial function and confer cytoprotection

Article abstract of DOI:10.1016/j.bmc.2012.07.005

Selected pyridinol analogues of the experimental neuroprotective drug idebenone have been synthesized and evaluated as antioxidants capable of preserving mitochondrial function. The compounds, having a different redox core but the same side chain as idebenone, exhibited a range of potencies, reflecting differences in their structures. The results obtained provide guidance in the design of such analogues with improved properties. Analogues were identified that have significantly improved antioxidant activity compared with idebenone in cultured lymphocytes, and which exhibit lesser inhibition of the electron transport chain.

Full text of DOI:10.1016/j.bmc.2012.07.005

Bioorganic & Medicinal Chemistry 20 (2012) 5188–5201
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Analysis of the structural and mechanistic factors in antioxidants
that preserve mitochondrial function and confer cytoprotection
Pablo M. Arce, Ruth Goldschmidt, Omar M. Khdour, Manikandadas M. Madathil, Jennifer Jaruvangsanti,
⇑
Sriloy Dey, David M. Fash, Jeffrey S. Armstrong, Sidney M. Hecht
Center for BioEnergetics, Biodesign Institute and Department of Chemistry and Biochemistry, Arizona State University, Tempe, AZ 85287, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Selected pyridinol analogues of the experimental neuroprotective drug idebenone have been synthesized
and evaluated as antioxidants capable of preserving mitochondrial function. The compounds, having a
different redox core but the same side chain as idebenone, exhibited a range of potencies, reflecting dif-
ferences in their structures. The results obtained provide guidance in the design of such analogues with
improved properties. Analogues were identified that have significantly improved antioxidant activity
compared with idebenone in cultured lymphocytes, and which exhibit lesser inhibition of the electron
transport chain.
Received 30 May 2012
Revised 27 June 2012
Accepted 4 July 2012
Available online 14 July 2012
Keywords:
Idebenone
Mitochondrial dysfunction
Reactive oxygen species
Electron transport chain
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
(CatenaÒ), a synthetic analogue of coenzyme Q10, is currently the
only compound that has reached Phase III clinical trials for the
Mitochondria play a central role in the cellular production of
ATP, and in a range of other metabolic processes, including signal-
ing, cell growth, differentiation and death.1–3 Mitochondria are also
involved in the production of reactive oxygen species through elec-
tron carriers in the respiratory chain;4–6 and the mitochondrion it-
self is quite susceptible to oxidative stress.7,8 A number of human
pathologies result from some mitochondrial dysfunction, including
hereditary mitochondrial cytopathies, diabetes, cardiovascular dis-
ease and neurodegenerative diseases.9–16 One of the neurodegen-
erative diseases, Friedreich’s ataxia (FRDA), is the most common
autosomal recessive ataxia in the Caucasian population,17 and
has been estimated to affect ꢀ1 in 50,000 people in the United
States. There is currently no effective cure or treatment available.
Friedreich’s ataxia disease is believed to arise from reduced syn-
thesis of the iron chaperone frataxin due to impaired gene tran-
scription.17,18 This leads to mitochondrial iron accumulation,
dysfunction of mitochondrial Fe–S containing enzymes, and in-
creased Fenton-mediated reactive oxygen species (ROS) produc-
tion.19,20 An actively pursued strategy for FRDA treatment at the
present time involves the use of antioxidants to slow the progres-
sion of the mitochondrial degradation resulting from oxidative
stress.21–23 While not all such compounds are targeted to the mito-
chondria, they have been shown to preserve the function of mito-
chondria under conditions of oxidative stress. Idebenone
treatment of FRDA.24 The pharmacological and pharmacokinetic
behavior of idebenone has been described.25 However, the clinical
efficacy of idebenone is uncertain,24,26–28 and the degree of
improvement may be limited by the dose currently administered,
suggesting that higher doses may be necessary to have a beneficial
effect on neurological function.29 Unfortunately, biochemical stud-
ies indicate that idebenone is a significant inhibitor of both the
electron transport and proton pumping activities of mitochondrial
respiratory chain complex I, and could thus potentially stimulate
oxygen radical production in mitochondria if used at high
doses.30,31 Accordingly, alternative approaches may have to be con-
sidered. Recently, a few nitrogen heterocycles, including pyrimidi-
nols and pyridinols, have been investigated and have exhibited
strong antioxidant activity in vitro, as judged by their ability to
scavenge peroxyl radical formation in benzene solution. Some of
these have relatively high stability in air, consistent with potential
therapeutic utility.32
To facilitate the delivery of such molecules to the mitochondria
of cells, and enable their permeability in plasma, we have designed
antioxidants bearing lipophilic side chains and having pyrimidinol
or pyridinol redox cores.33,34 An earlier study focused on one pyri-
midinol derivative (compound 6, Fig. 1)34 showed encouraging re-
sults, suggesting that further optimization of the redox core might
afford improved efficacy. Currently, we describe the preparation
and characterization of several novel pyridinol derivatives having
the same side chain as idebenone. This includes an analysis of
structural factors among these pyridinol analogues that afford
⇑
Corresponding author. Tel.: +1 480 965 6625; fax: +1 480 965 0038.
E-mail address: sidney.hecht@asu.edu (S.M. Hecht).
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2012.07.005

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5189
OH
OH
3
OH
2
OH
4
5
N
N 1
6
NH2
N
2
1
OH
OH
OH
OH
N
N
N
N
3
4
OH
OH
OH
OH
N
N
N
N
N
5
6
O
O
CH3O
CH3O
OH
idebenone
Figure 1. Chemical structures of compounds 1–6 and idebenone.
effective cytoprotection, in comparison with the results obtained
2.2. Synthesis of pyridinol analogue 2
with idebenone and compound 6.
Analogue 2 was prepared from 6-amino-3-(benzyloxy)-6-[10-
(methoxymethoxy)decyl]-4-methylpyridine (12) by deprotection
of the aliphatic hydroxyl group under acidic conditions, followed
by hydrogenolysis in 29% yield (Scheme 2).
2. Results and discussion
The approach used for the preparation of this series of antioxi-
dants followed the same general strategy employed for the synthe-
sis of compound 6.34
2.3. Synthesis of pyridinol analogues 3 and 4
Pyridinol analogues 3 and 4 were both synthesized from 3-
(benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-(methoxyme-
thoxy)decyl]-4-methylpyridine (11) (Scheme 2). Treatment under
acidic conditions effected removal of the MOM protecting group.
Hydrogenolysis over 20% palladium hydroxide on carbon in MeOH
at room temperature for 15 min afforded 3 in 17% yield. Hydrogen-
olysis under the same conditions for 16 h afforded 4, albeit only in
low (5%) yield.
2.1. Synthesis of pyridinol analogue 1
6-(Dimethylamino)-2-(10-hydroxydecyl)-4-methylpyridin-3-ol
(1) was synthesized in 8 steps (Scheme 1). First, 6-amino-3-bromo-
2,4-dimethylpyridine (8) was prepared in 84% yield by treating
commercially available 6-amino-2,4-dimethylpyridine with N-
bromosuccinimide. Compound 8 was then treated with 2,5-hexane-
dione in presence of p-toluenesulfonic acid to afford pyrrole-substi-
tuted pyridine 9 in 62% yield. Replacement of the halide moiety by
successive treatments with KOH/Pd2(dba)3 and phosphine ligand
L1, and then with benzyl bromide, afforded 10 in 82% yield. The fully
protected pyridinol core, 3-(benzyloxy)-6-(2,5-dimethyl-1H-pyr-
rol-1-yl)-2,4-dimethylpyridine (10), was then treated with phen-
yllithium in the presence of 1-bromo-9-(methoxymethoxy)nonane
(7) to afford fully protected 11 in 60% yield. None of the regioisomer
resulting from alkylation of the other pyridine methyl group was
observed. Compound 11 was deprotected by treatment with
hydroxylamine hydrochloride in the presence of KOH to obtain 6-
aminopyridinol 12 in 56% yield. This compound was treated with
methyl iodide in presence of sodium hydride to obtain 3-(benzyl-
oxy)-6-(dimethylamino)-2-[10-(methoxymethoxy)decyl]-4-meth-
ylpyridine (13) in 21% yield. Pyridinol analogue 1 was then obtained
in 77% yield by treating 13 successively with HCl and then with
palladium hydroxide under a H2 atmosphere.
2.4. Synthesis of pyridinol analogue 5
6-(Dimethylamino)-2-(10-hydroxydecyl)-4,5-dimethylpyridin-
3-ol (5) was synthesized in seven steps as shown in Scheme 3.
First, 6-amino-2,4,5-trimethylpyridin-3-ol (14) was synthesized
by a literature procedure.35 The exocyclic amine of 14 was then
protected by treatment with 2,5-hexanedione in presence of p-tol-
uenesulfonic acid to afford pyrrole 15 in 75% yield. O-Benzylation
using NaH/BnBr gave 16 in 78% yield. Subsequent treatment of
16 with n-BuLi in the presence of the 1-bromo-9-(methoxyme-
thoxy)nonane (7) gave fully protected analogue 17 in 41% yield.
Compound 17 was deprotected by treatment with hydroxylamine
hydrochloride in presence of KOH, affording 18 in 13% yield. Treat-
ment of 18 with formaldehyde solution and sodium cyanoborohy-
dride in presence of acetic acid gave 19 in 77% yield. Deprotection
of 19 to afford 5 was accomplished in 79% yield by successive

5190
P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
NaH, MOMCl
Br
OH
Br
OMOM
anh THF
70%
7
O
Br
Br
N
1. KOH, Pd2(dba)3, L1
1:1 degassed H2O/dioxane
O
NBS
p-toluenesulfonic acid
N
N
N
toluene, reflux
62%
acetonitrile
84%
2. NaH, BnBr
anh THF
82%
NH2
NH2
8
9
OBn
OBn
OMOM
1. PhLi
HONH2-HCl
KOH
2. 7
N
N
N
N
anh THF
60%
aq ethanol, reflux
56%
11
10
OBn
OBn
OMOM
OMOM
MeI, NaH
anh DMF
21%
N
N
N
NH2
12
13
OH
1. HCl
2. H2, Pd(OH)2/C
OH
N
MeOH
77%
N
1
O
P(t-Bu)2
i-Pr
i-Pr
Pd2(dba)3
=
L1
=
Pd2
3
i-Pr
Scheme 1. Route employed for the synthesis of compound 1.
treatments with HCl and then with palladium hydroxide under a
hydrogen atmosphere.
2. However, the redox core of 1 lacked any activity. Finally, com-
pound 3 had the lowest activity for suppressing lipid peroxidation
among the analogues having a 10-hydroxydecyl side chain.
2.5. Biochemical and biological evaluation of the pyridinol
analogues
2.5.2. Suppression of reactive oxygen species
The ability of the pyridinol analogues to suppress ROS induced
by depletion of glutathione was evaluated in CEM leukemia cells.
ROS was measured in a quantitative FACS experiment, using
dichlorofluorescein diacetate (DCFH-DA) as a substrate for deter-
mining intracellular oxidant production. DCFH-DA is hydrolyzed
by esterases to afford 2,7-dichlorodihydrofluorescein (DCFH), the
latter of which is trapped within the cell. This non-fluorescent mol-
ecule is then oxidized to fluorescent dichlorofluorescein (DCF) by
the action of cellular oxidants. The results are presented in Figure 2,
and show that analogue 4 was the most potent in suppressing ROS,
and did so in a concentration-dependent manner. Compounds 5
and 6 also had quite good activity; interestingly, all of the com-
pounds tested, with the exception of compound 3, were better
2.5.1. Inhibition of lipid peroxidation
The ability of the pyridinol analogues to quench lipid peroxida-
tion has been studied in FRDA lymphocytes depleted of glutathione
by treatment with diethyl maleate (DEM). The cells were pre-incu-
bated with the test compounds for 16 h prior to DEM treatment. Li-
pid peroxidation was measured by a quantitative FACS analysis
using the fluorescent probe, C11-BODIPY581/591, reported to be
highly accurate in measuring lipid peroxidation.36 The results, pre-
sented in Table 1, show the most potent activity for compounds 4
and 5. Both of these compounds were more potent than pyrimidi-
nol 6. Compound 1, lacking a methyl group at position 6, was
slightly more potent than idebenone, and better than compound

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5191
OBn
OH
1. HCl
2. H2, Pd(OH)2/C
OMOM
OH
N
N
MeOH
29%
NH2
NH2
12
2
OH
10
1. HCl
2. H2, Pd(OH)2/C
15 min
OH
N
MeOH
17%
N
OBn
10
OMOM
3
N
N
OH
10
1. HCl
2. H2, Pd(OH)2/C
16 h
OH
N
11
MeOH
5%
N
4
Scheme 2. Routes employed for the syntheses of compounds 2, 3 and 4.
O
1. DMF, SOCl2
2. Zn, AcOH
OH
OH
OH
3. PhNH2, NaNO2, HCl, water
4. Zn, HCOOH, MeOH
O
HO
p-toluenesulfonic acid
N
N
N
toluene, reflux
21%
·HCl
75%
N
NH2
OH
14
15
OBn
OBn
OMOM
1. TMEDA, n-BuLi
NaH, BnBr
2. 7
N
N
N
anh THF
78%
anh THF
41%
N
17
16
formalin
NaBH3CN
AcOH
OBn
HONH2-HCl
OMOM
KOH
N
aq ethanol, reflux
acetonitrile
13%
77%
NH2
18
OBn
OH
OMOM
OH
1. HCl, MeOH
N
N
N
2. H2, Pd(OH)2/C,
MeOH
N
79%
5
19
Scheme 3. Route employed for the synthesis of compound 5.
ROS scavengers than idebenone. Because the use of DCF fluores-
cence as an endpoint has been shown to lack specificity under cer-
tain circumstances,37,38 it was shown in the present study that
the increase in DCF fluorescence induced by DEM was completely

5192
P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
Table 1
ual treatments. Carbonyl cyanide 4-(trifluoromethoxy)phenylhyd-
razone (FCCP), commonly used uncoupler of oxidative
phosphorylation in mitochondria, was employed to dissipate the
chemiosmotic proton gradient (
H+). The lower levels of TMRM
Suppression of lipid peroxidation by pyridinol antioxidants in cultured FRDA
a
lymphocytes treated with diethyl maleate (DEM)a
Compound
Scavenging activity (%)
D
l
fluorescence resulting from FCCP treatment reflect the depolariza-
tion of mitochondrial inner membrane potential. The bar graph, in
the lower panel of Figure 3, summarizes the relative geometric
mean fluorescence intensity (GMFI) of the flow cytometric profiles.
It shows clearly that compounds 4 and 6 had the greatest potency,
and acted in a dose-dependent manner; compound 5 was almost
as good although not strictly dose-dependent in this experiment.
These three compounds were clearly better than idebenone, and
better than compounds 1 and 2, which had activity in the same
range as idebenone. Finally, compound 3 had the lowest activity
and the redox core of compound 1 was completely inactive. The
flow cytometric two-dimensional color density dot plot analyses
for compounds 2, 3, 6 and the redox core of 1 are shown in Figure 1
of the Supplementary data.
2.5
l
M
5 lM
Untreated controlb
100
0
100
0
Treated controlc
1
2
3
4
5
6
71 6.0
62 6.4
20 10
90 2.9
87 5.1
83 4.2
68 6.4
3.5 2.3
82 3.7
74 4.7
39 7.6
97 2.3
97 3.1
91 3.6
77 2.4
3.9 3.0
Idebenone
Redox core 1
a
Values have been calculated as [(100–% mean)/(100–% mean of the untreated
control)] x 100.
b
No DEM treatment.
DEM treatment.
c
2.5.4. Cytoprotection
Cytoprotection was measured initially using cultured CEM leu-
kemia cells (ATCC CCL-119) that were treated with diethyl maleate
to induce oxidative stress through depletion of glutathione. As
shown in Table 2, compounds 4 and 6 were the most effective at
the concentrations tested. Compounds 1 and 5 were also effective
reversed by superoxide dismutase + catalase, or by the antioxidant
N-acetylcysteine (not shown).
2.5.3. Preservation of mitochondrial membrane potential
The ability of the pyridinol analogues to preserve mitochondrial
membrane potential under conditions of oxidative stress was stud-
at 2.5 lM concentration, but not at the two lower concentrations
tested. The ability of the best pyridinols to protect cultured Friedr-
eich’s ataxia lymphocytes from cell death induced by oxidative
stress was also measured (Fig. 2 of the Supplementary data). The
half maximal effective concentrations (EC50) for the best pyridinol
analogues were determined using FRDA lymphocytes; they are
shown in Table 3. Compound 4 was by far the most efficient, hav-
ied. Mitochondrial membrane potential,
Dwm, is an important
parameter of mitochondrial integrity and is essential for maintain-
ing the physiological function of the respiratory chain in ATP syn-
thesis.39
Dwm was estimated using the cationic fluorescent probe
tetramethylrhodamine methyl ester (TMRM), which preferentially
accumulates in the mitochondria due to the negative membrane
potential across the inner mitochondrial membrane, in accordance
ing an EC50 value of 0.15 0.05
more efficient than idebenone. Finally, compound 1 had the lowest
potency, with an EC50 value of 0.84 0.05 M.
lM. Compounds 5 and 6 were both
with the
mitochondria decreases when
D
wm Nernst potential.40 The red fluorescent signal in
l
Dwm is impaired, and can be mea-
sured using flow cytometry. A representative flow cytometric
two-dimensional color density dot plot analyses of the mitochon-
drial membrane potential measurements with and without incuba-
tion in the presence of selected compounds are shown in the upper
panel of Figure 3. The percentage of cells with intact mitochondrial
membrane potential appears in the top right quadrant of individ-
2.5.5. Mitochondrial electron transport chain function
The inhibitory effects of the test compounds on bovine heart
mitochondrial complexes I, III and IV were evaluated using submi-
tochondrial particles (SMP), by measuring NADH oxidase activity.
The results are presented in Table 4, and show that compounds
Figure 2. Flow cytometric analysis of CCRF-CEM leukemia cells pre-treated with compounds 1–6 or idebenone at 2.5, 0.5, and 0.25 lM concentrations for 15 h, and then
treated with diethyl maleate (DEM) for 1 h to induce the production of ROS. The cells were stained with 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA) for 25 min prior
to analysis. Data shown represent the mean SEM of three different experiments run as duplicates. Results are expressed as % ROS scavenging activity.

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5193
untreated
control
treated
control
FCCP
no TMRM
1 (0.5 µM)
1 (1 µM)
4 (0.5 µM)
4 (1 µM)
5 (0.5 µM)
5 (1 µM)
idebenone (0.5 µM)
idebenone (1 µM)
idebenone (2.5 µM)
1 (2.5 µM)
4 (2.5 µM)
5 (2.5 µM)
Figure 3. Effect of pyridinol antioxidants on mitochondrial membrane potential of DEM-treated FRDA lymphocytes. The upper panel shows representative flow cytometric
two-dimensional color density dot plot analyses of mitochondrial membrane potential ( m) in FRDA lymphocyte cells stained with 250 nM TMRM and analyzed using FL2-
H channel as described in the Experimental section. The percentage of cells with intact m is indicated in the top right quadrant of each treatment. Representative example
from at least three independent experiments. In each analysis, 10,000 events were recorded. Bottom panel shows a bar graph of means of the percentage of cells with intact
m recorded by FACS. Data are expressed as means SEM of three independent experiments run in duplicate. The bar graph shows the relative geometric mean fluorescence
intensity (GMFI) of the above flow cytogram profiles calculated using CellQuest software. Data are expressed as means of the SEs (n = 3).
D
w
D
w
D
w
Table 2
Table 3
Cytoprotective effects of pyridinol antioxidants on the viability of cultured CEM
Effect of pyridinol antioxidants on the cellular viability of FRDA lymphocytes treated
with DEM
leukemia cells treated with DEMa
Compound
Viable cells (%)
0.5
Compound
EC50 (lM)
0.1
l
M
lM
2.5 lM
1
4
5
6
0.84 0.05
0.15 0.05
0.56 0.06
0.38 0.1
0.73 0.1
1
2
3
4
5
6
41 9.7
28 13
18 5.6
88 10
42 5.4
76 9.9
34 13
59 14
38 12
23 6.6
90 6.6
47 4.0
93 5.1
49 12
94 3.4
78 14
26 7.5
93 2.4
92 1.6
94 4.1
88 4.3
Idebenone
Idebenone
a
The viability of untreated cells was defined as 100%; cells treated with DEM
ure shows that the redox cores of 1 and 5 have similar redox poten-
tials and that they are both more easily reduced than CoQ0,
consistent with their enhanced ability to suppress ROS (cf Fig. 2).
Figure 4B shows that the redox core of compound 1 was unstable
to repetitive redox cycles while that of compound 5 was com-
pletely stable.
alone had 18 10% viability.
1, 2, 3, and 4 were all less inhibitory to respiratory chain function
with complexes I, III and IV than compound 6 or idebenone. Com-
pound 5 was the most inhibitory of the tested compounds.
2.5.7. Electron spin resonance analysis
2.5.6. Cyclic voltammetric analysis
Cyclic voltammetric analyses of the redox cores of compounds 1
and 5 are presented in Figure 4A in comparison with CoQ0. The fig-
The ability of compound 1 to transfer its phenolic hydrogen
atom and form a stabilized radical has been tested by the use
of ESR. Figure 5A shows the signal obtained from the radical

5194
P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
Table 4
volved a new strategy (Scheme 3). These six analogues all have the
same side chain as idebenone at the 2-position of the aromatic
ring, have an amino group at the 6-position, and a methyl group
at the 4-position (Fig. 1). Compounds 1–5 were of interest as
new examples of multifunctional radical quenchers, a type of coen-
zyme Q analogue capable of conferring cytoprotection to cells un-
der oxidative stress by a few different mechanisms that operate
concurrently.33,34
Inhibitory effects of compounds 1–6, and idebenone on bovine heart mitochondrial
NADH oxidase activity
Compound
NADH oxidase activitya (%)
10
l
M
5
lM
1 lM
Idebenone
15 2.5
25 5.2
48 2.7
43 5.5
27 4.0
53 3.5
42 5.8
52 3.6
49 3.5
59 3.8
65 2.5
85 9.1
74 2.2
71 4.5
78 4.0
27 0.8
49 4.3
1
2
3
4
5
6
The antioxidant properties of the new pyridinol analogues were
tested and analyzed in selected biological and biochemical assays.
Cultured CEM leukemia cells and Friedreich’s ataxia lymphocytes
were challenged with diethyl maleate, which depletes cellular glu-
tathione.41–43 Diethyl maleate induces greater oxidative stress
than is likely to be encountered physiologically, permitting the
identification of compounds anticipated to function robustly under
pathophysiological conditions. Compounds 1–5 showed a range of
potencies in blunting the effects of the resulting oxidative stress. In
addition, an experiment was carried out to determine whether, and
to what extent, compounds 1–5 inhibited the function of the mito-
chondrial electron transport chain (ETC). This was done because
quinone derivatives are known to bind to the ETC, and thereby in-
hibit mitochondrial respiration. Many compounds have been re-
ported to inhibit mitochondrial complexes I44–47 and III48,49 and
these invariably strongly diminish cell viability. Since idebenone
has been found to significantly inhibit complex I,30,31 it seemed
logical to consider that this property might limit its useful antiox-
idant properties. Accordingly, we evaluated compounds 1–6 for
this property using a biochemical assay that measures the activity
of NADH oxidase, which encompasses mitochondrial complexes I,
III, and IV. In fact, compounds 1–4 were found to be superior both
to idebenone and compound 6 in their lesser inhibition of NADH
oxidase activity (Table 4). While these experiments do not defini-
tively establish the biochemical locus of inhibition by compounds
1–5, complex I seems likely to be the relevant locus.
Summarizing the results obtained for compounds 1–6 in the
several assays studied, it appears that compounds 1, 4, 5, and 6
have superior antioxidant activity as compared with idebenone.
Compounds 4, 5 and 6 were also more potent cytoprotective agents
than idebenone (Tables 2 and 3). The greater potency of pyridinols
4 and 5 in suppressing ROS in cultured CEM leukemia cells as com-
pared with idebenone is entirely consistent with the reduction of
their redox core at more oxidizing potentials than the redox core
of idebenone (and coenzyme Q10) (Fig. 4A). Further, introduction
of N-atoms into the quinone moiety has been shown to lower
the bond dissociation energy of the phenolic OH group of the re-
duced form of the cofactor (analogues),32 consistent with the supe-
rior activities of most of the pyridinols in quenching lipid
peroxidation (Table 1).
6
0.2
9
0.3
22 6.4
26 6.7
a
Relative to untreated control.
OH
A
2.0x10-5
N
N
1.0x10-5
0.0
O
O
MeO
MeO
-1.0x10-5
-2.0x10-5
-3.0x10-5
-4.0x10-5
OH
N
N
1.0 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8
Potential (V)
OH
B
5.0x10-7
0.0
N
N
-5.0x10-7
-1.0x10-6
-1.5x10-6
-2.0x10-6
-2.5x10-6
OH
N
N
0
1
2
3
4
5
6
7
8
9
Time (min)
All the new compounds, other than compound 5, exhibited a
lesser inhibitory effect on complexes I, III and IV of the mitochon-
drial electron transport chain than idebenone (Table 4), which
seems likely to be related to their greater potency as cytoprotective
agents (Tables 2 and 3). The diminished potency of 1 is likely re-
lated to its lack of stability under redox cycling conditions (vide
infra).
Figure 4. (A) Cyclic voltammetric analysis of redox cores 1 (scan rate 0.1 V sꢁ1
,
5 mM concentration) and 5 (scan rate 0.04 V sꢁ1, 5 mM concentration) versus CoQ0
(scan rate 0.04 V sꢁ1, 1 mM concentration). (B) Normalized peak current (ip(n)ꢁip(1)
)
plotted against time in continuous cyclic voltammetric analysis of core 1 versus
core 5, where ip(n) is the peak current for the nth scan while ip(1) is the peak current
for the first scan.
From Table 1 and Figure 3, we can easily conclude from the data
for the redox core of compound 1, which lacked any antioxidant
activity, that the lipophilic side chain is crucial for compound effi-
cacy. Interestingly, it appears that compound 6 had better antiox-
idant activity than 1, despite the fact that it was more inhibitory to
NADH oxidase (Table 4). This is undoubtedly due to the low stabil-
ity of compound 1 to the redox cycling (Fig. 4B), which we posit to
be essential to its antioxidant function.34 This is also consistent
with the higher EC50 value obtained with 1 (Table 3). One strategy
to increase the stability of compound 1 was to synthesize com-
pound 5, having a C-methyl group at position 5 of the pyridinol
activation of compound 1 in the presence of illuminated di-tert-bu-
tyl peroxide. Figure 5B shows the resonance contributors of the
stabilized 3-pyridinoxyl radical formed.
3. Discussion
New pyridinol analogue 1 has been synthesized using a strategy
similar to that used to synthesize its aza analogue 6 (Scheme 1).34
Compounds 2, 3, and 4 were prepared from intermediates on the
pathway to 1 (Scheme 2), while the preparation of compound 5 in-

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5195
A
B
simulated
O
N
O
N
O
N
O
N
R
R
R
R
N
N
N
N
H
H
H
H
R
O
N
O
N
O
N
O
N
O
N
R
R
R
R
N
N
N
N
N
H
H
H
H
H
experimental
polar structures
3400
3420
3440
3460
3480
3500
3520
Gauss
Figure 5. (A) ESR signal and simulation of compound 1 (20 mM) in dry de-aerated benzene at 25 °C during continuous illumination, in the presence of di-tert-butyl peroxide.
The hyperfine splitting constants are as follows: aH (3H, o-methyl) = 3.5 G, aH (2H, o-carbon chain) = 3.3 G, aH (1H, o-carbon chain) = 3.4 G, aH (1H, m-ring) = 1.9 G, aH (6H, p-
NMe2) = 4.2 G, aN (1 N, ring) = 1.1 G, aN (1 N, p-NMe2) = 5.9 G, and g = 2.00493. The simulation corresponds to the experimental signal with a relative error of R = 0.983. (B)
Resonance stabilized 3-pyridinoxyl radical resulting from abstraction of the O–H hydrogen atom.
OH
O
H C
3
OH
H C
3
OH
[O]
N
N
H
H
N
N
H C
3
CH
3
+
CH
H C
3
3
3
steric hindrance
Figure 6. Steric interactions putatively arising from the oxidation of compound 3.
ring (Fig. 4B). This resulted in a compound having increased anti-
oxidant activity, equal or superior to 6 in the various assays used
for their evaluation. Unfortunately the increased stability was also
accompanied by an increase in NADH oxidase inhibition, resulting
in an EC50 of 0.56 0.06 lM for 5 compared to 0.38 0.1 lM for
compound 6 in the FRDA lymphocyte protection assay (Tables 3
As suggested in our previous report,34 and predicted for this
family of molecules,32,50 the mechanism for the scavenging of
ROS and lipid radicals by these new pyridinols is through the for-
mation of a pyridinoxyl radical. To test this hypothesis, we studied
the ability of compound 1 to form a radical upon hydrogen abstrac-
tion of the hydroxyl group at position 3. As described above, the
radical derived from compound 1 was formed by excitation of di-
tert-butyl peroxide, and the signal obtained was measured using
an ESR spectrophotometer (Fig. 5). The signal was characterized
by its g factor of 2.0049 and the small hyperfine coupling of the
pyridine ring nitrogen (1.1G). In agreement with the work of Pratt
and co-workers50 this is attributed to the electron donating charac-
ter of N(CH3)2 responsible for an increase in contribution to stabil-
ity arising from the higher energy polar structures (Fig. 5B).
It is interesting that the best compounds described in this
report function more effectively than idebenone as regards the
several parameters studied, and were more potent overall as cyto-
protective agents. Idebenone has also been reported to form a
hydroquinone in situ via the action of NAD(P)H: quinone reduc-
tases, which enables the transfer of electrons to complex III.51
Augmentation of ATP production was not evaluated for compounds
and 4).
Another possible way to improve compound 1 stability involved
changing the electronic character and steric bulk of the substituted
group at the 6-position. Toward that end, compounds 2, 3 and 4 were
synthesized and tested. It appears from the data that compound 4
was the most effective of these compounds. This presumably reflects
the steric effects of the C-methyl groups of the pyrrolidine moiety in
preventing side reactions at the unsubstituted position of the
pyridine ring. Additionally, the more favorable properties of 4 as
compared with 3 is consistent with the possibility that electron-
donating substituents at the 6 position of the pyridine ring may
stabilize the radical putatively formed upon quenching of lipid per-
oxidation by 4. It may be noted that the modest effects of compound
3 as an antioxidant may also be related to the steric interactions that
would be anticipated for the oxidized form (Fig. 6).

5196
P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
1–5 in the present study, but optimization of this property could
plausibly further enhance the favorable properties of the pyridinols
studied here. Likewise, cellular distribution of the pyridinols, espe-
cially as regards mitochondrial localization, could also materially
affect their biological function, and is worthy of characterization.
MelTemp apparatus and are uncorrected. High resolution mass
spectra were obtained in the Arizona State University CLAS High
Resolution Mass Spectrometry Laboratory.
5.1.1. 1-Bromo-9-(methoxymethoxy)nonane (7)34
To a stirred solution containing 5.00 g (22.4 mmol) of 9-bromo-
1-nonanol in 60 mL of anh THF was added 5.10 mL (5.41 g;
67.2 mmol) of MOMCl followed by 1.79 g (44.8 mmol) of a 60% sus-
pension of NaH in mineral oil. The reaction mixture was stirred at
23 °C overnight. The reaction mixture was carefully quenched with
satd aq sodium bicarbonate, poured into 200 mL of water and ex-
tracted with two 150-mL portions of ether. The combined organic
solution was washed with 200 mL of brine, dried (MgSO4) and then
concentrated under diminished pressure. The residue was purified
by chromatography on a silica gel column (15 ꢂ 5 cm). Elution
with 9:1 hexanes/ethyl acetate afforded the expected product as
a colorless oil: yield 4.21 g (70%); silica gel TLC Rf 0.45 (9:1 hex-
anes/ethyl acetate); 1H NMR (CDCl3) d 1.28–1.39 (br, 10H), 1.56
(quint, 2H, J = 7.2 Hz), 1.82 (quint, 2H, J = 7.2 Hz), 3.33 (s, 3H),
3.37 (t, 2H, J = 6.8 Hz), 3.48 (t, 2H, J = 6.8 Hz), and 4.59 (s, 2H);
13C NMR (CDCl3) d 26.1, 28.1, 28.7, 29.28, 29.33, 29.7, 32.8, 34.0,
55.0, 67.7, and 96.4; mass spectrum (APCI), m/z 267.0953 (M+H)+
(C11H24O2Br requires 267.0960).
4. Conclusions
In the present study, the effects of several structural changes
have been shown to influence both the antioxidant activities of
the compounds as well as inhibition of the mitochondrial ETC.
The conclusions include the requirement that a lipophilic side
chain is essential for good antioxidant activity, and a redox core
that undergoes electron transfer at more oxidizing potentials than
CoQ0 is preferable. Also, good stability at the pyridinoxyl-radical
level is important and can be achieved by full substitution of the
3-pyridinol redox core. Finally, resonance delocalization increasing
the resonance contribution due to polar structures of the radical is
observed; these are likely stabilized by increasing the electron
donating character of the substituted 6-amino group. The data ob-
tained for inhibition of NADH oxidase activity do not provide any
simple strategy for achieving low inhibition. However, they do
underscore the importance of minimizing such inhibitory effects.
Finally, it can be noted that compound 4 has the best antioxidant
activity among the tested compounds, including idebenone and
compound 6. Compound 4 has the lowest EC50 value (0.15
5.1.2. 6-Amino-3-bromo-2,4-dimethylpyridine (8)55
To a stirred solution containing 2.00 g (16.3 mmol) of 2-amino-
4,6-dimethylpyridine in 25 mL of acetonitrile was added 2.90 g
(16.3 mmol) of N-bromosuccinimide. The reaction mixture was
stirred at room temperature for 5 h. The formed precipitate was fil-
tered and dried to afford 6-amino-3-bromo-2,4-dimethylpyridine
(8) as a colorless solid: yield 2.76 g (84%); mp 143–145 °C; silica
gel TLC Rf 0.15 (2:1 hexanes/ethyl acetate); 1H NMR (CDCl3) d
2.25 (s, 3H), 2.48 (s, 3H), 4.39 (br s, 2H), and 6.22 (s, 1H); 13C
NMR (CDCl3) d 23.3, 25.1, 108.1, 112.3, 148.6, 155.2, and 156.3;
mass spectrum (APCI), m/z 201.0032 (M+H)+ (C7H10N2Br requires
201.0027).
0.05 lM) in FRDA lymphocytes, and was the most potent in confer-
ring cytoprotection to cultured CEM leukemia cells (Tables 2 and 3).
It also has the best ROS and lipid peroxidation scavengingproperties,
as well as the best ability to maintain mitochondrial membrane po-
tential. Importantly, compound 4 is also less inhibitory to NADH oxi-
dase than compound 6 and idebenone, making this compound a
good candidate for further evaluation.
While the focus of the current study is Friedreich’s ataxia, there
are several mitochondrial diseases involving a significant compo-
nent of energy dysfunction (e.g. Alzheimer’s disease, Parkinson’s
disease, LHON, and Leigh syndrome).10,11,15,52–54 The underlying
biochemistry at the level of mitochondrial dysfunction is likely
similar for these diseases. Therefore, patients with these disorders
may well benefit from therapeutic agents designed by employing
the concepts that arise from the present results.
5.1.3. 3-Bromo-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4-
dimethylpyridine (9)
To a stirred solution containing 2.76 g (13.8 mmol) of 6-amino-
3-bromo-2,4-dimethylpyridine (8) in 25 mL of toluene were added
2.02 mL (1.98 g; 17.2 mmol) of 2,5-hexanedione followed by
130 mg (0.68 mmol) of p-toluenesulfonic acid. The cooled reaction
mixture was stirred at reflux for 14 h. The cooled reaction mixture
was poured into 150 mL of water and then extracted with 200 mL
of ethyl acetate. The organic phase was washed with 150 mL of
brine, dried (MgSO4) and then concentrated under diminished
pressure. The residue was purified by chromatography on a silica
gel column (15 ꢂ 6 cm). Elution with 6:1 hexanes/ethyl acetate
afforded 3-bromo-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4-dimethyl-
pyridine (9) as an orange oil: yield 2.37 g (62%); silica gel TLC Rf
0.70 (6:1 hexanes/ethyl acetate); 1H NMR (CDCl3) d 2.12 (s, 6H),
2.47 (s, 3H), 2.70 (s, 3H), 5.88 (s, 2H), and 6.93 (s, 1H); 13C NMR
(CDCl3) d 13.1, 23.4, 25.6, 106.9, 121.16, 121.24, 122.5, 128.4,
149.6, and 157.4; mass spectrum (APCI), m/z 279.0502 (M+H)+
(C13H16N2Br requires 279.0497).
5. Experimental section
5.1. Chemistry
1H NMR spectra were recorded on a Varian Inova 400 MHz,
using chloroform-d, methanol-d4 or acetonitrile-d3. 1H NMR chem-
ical shifts were reported relative to residual chloroform at
7.26 ppm, or to residual methanol at 3.31 ppm, or to residual ace-
tonitrile at 2.10 ppm. 13C NMR chemical shifts were reported rela-
tive to residual chloroform at 77.1 ppm, or to residual methanol at
49.0 ppm, or to residual acetonitrile at 1.89 ppm and 116.43 ppm.
All solvents were analytical grade and were used without further
purification. All chemicals were purchased from Aldrich Chemical
Company and were used without further purification. The reac-
tions were carried out under an argon atmosphere unless specified
otherwise. Column chromatography was carried out using silica gel
(Silicycle R10030B, 60 particle size, 230–240 mesh). Analytical thin
layer chromatography separations were carried out on glass plates
coated with silica gel (60, particle size F254, E. Merck 5608/7). The
TLC chromatograms were developed using UV irradiation or by
immersing the plates in 2.5% potassium permanganate in ethanol
or 2% anisaldehyde + 5% sulfuric acid + 1.5% glacial acetic acid in
ethanol, followed by heating. Melting points were recorded on a
5.1.4. 3-(Benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4-
dimethylpyridine (10)
To a stirred solution containing 3.13 g (11.2 mmol) of 3-bromo-
6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4-dimethylpyridine
(9)
in
50 mL of 1:1 degassed dioxane/water was added 612 mg
(0.67 mmol) of Pd2(dba)3 followed by 284 mg (0.67 mmol) of 2-
di-tert-butylphosphino-20,40,60-triisopropylbiphenyl
(L1)
and
1.88 g (33.6 mmol) of KOH. The reaction mixture was stirred at
100 °C for 3 h. The reaction mixture was poured into 200 mL of

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5197
water and extracted with two 100-mL portions of ethyl acetate.
The aqueous layer was acidified with HCl (to pH 2–3) and then ex-
tracted with two 100-mL portions of ethyl acetate. The combined
organic layer was washed with a 100-mL portion of brine, dried
(MgSO4) and concentrated under diminished pressure. The residue
was dissolved in 50 mL of anh THF and 1.99 mL (2.86 g;
16.8 mmol) of benzyl bromide followed by 807 mg (22.4 mmol)
of 60% sodium hydride suspension in mineral oil were added.
The reaction mixture was stirred at 23 °C for 16 h. The reaction
mixture was poured into 150 mL of water and extracted with
two 100-mL portions of diethyl ether. The combined organic layer
was washed with 100 mL of brine, dried (MgSO4) and concentrated
under diminished pressure. The residue was purified by chroma-
tography on a silica gel column (15 ꢂ 6 cm). Elution with hexanes
(removal of unreacted benzyl bromide) and then with 4:1 hexanes/
diethyl ether afforded 3-(benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-
1-yl)-2,4-dimethylpyridine (10) as a yellowish oil: yield 2.82 g
(82%); silica gel TLC Rf 0.65 (5:1 hexanes/ethyl acetate); 1H NMR
(CDCl3) d 2.17 (s, 6H), 2.37 (s, 3H), 2.58 (s, 3H), 4.94 (s, 2H), 5.91
(s, 2H), 6.98 (s, 1H), and 7.40–7.50 (m, 5H); 13C NMR (CDCl3) d
13.1, 16.2, 19.5, 74.6, 106.4, 121.9, 127.9, 128.31, 128.34, 128.6,
136.7, 142.0, 146.6, 151.0, and 152.0; mass spectrum (APCI), m/z
307.1801 (M+H)+ (C20H23N2O requires 307.1810).
dried (MgSO4) and concentrated under diminished pressure. The
residue was purified by chromatography on a silica gel column
(10 ꢂ 3 cm). Elution with 9:1 dichloromethane/methanol afforded
6-amino-3-(benzyloxy)-2-[10-(methoxymethoxy)decyl]-4-meth-
ylpyridine (12) as a yellowish oil: yield 113 mg (56%); silica gel TLC
Rf 0.45 (9:1 dichloromethane/methanol); 1H NMR (CDCl3) d 1.24–
1.31 (br m, 12H), 1.53–1.63 (m, 4H), 2.16 (s, 3H), 2.62 (dd, 2H,
J = 8.0 and 8.0 Hz), 3.33 (s, 3H), 3.48 (t, 2H, J = 6.8 Hz), 4.24 (br s,
2H), 4.59 (s, 2H), 4.71 (s, 2H), 6.16 (s, 1H), and 7.31ꢁ7.43 (m,
5H); 13C NMR (CDCl3) d 16.7, 29.2, 29.40, 29.46, 29.48, 29.6, 29.7,
29.8, 32.2, 53.4, 55.1, 67.9, 75.4, 96.3, 108.1, 127.7, 128.0, 128.5,
137.4, 142.5, 144.9, 153.5, and 154.1; mass spectrum (APCI), m/z
415.2957 (M+H)+ (C25H39N2O3 requires 415.2961).
5.1.7. 3-(Benzyloxy)-6-(dimethylamino)-2-[10-
(methoxymethoxy)decyl]-4-methylpyridine (13)
To a stirred solution containing 218 mg (0.53 mmol) of 6-ami-
no-3-(benzyloxy)-2-[10-(methoxymethoxy)decyl]-4-methylpyri-
dine (12) in 3 mL of anh DMF was added 100 lL (228 mg;
1.59 mmol) of methyl iodide followed by 64.0 mg (1.59 mmol) of
a 60% suspension of NaH in mineral oil. The reaction mixture
was stirred at 23 °C for 3 h. The reaction was quenched with
0.5 mL of water and concentrated under diminished pressure.
The residue was dissolved in 50 mL of dichloromethane and
washed with 50 mL of brine. The organic solution was dried
(MgSO4) and then concentrated under diminished pressure. The
residue was purified by chromatography on a silica gel column
(8 ꢂ 3 cm). Elution with 3:1 hexanes/ethyl acetate afforded 3-
(benzyloxy)-6-(dimethylamino)-2-[10-(methoxymethoxy)decyl]-
4-methylpyridine (13) as a yellowish oil: yield 27 mg (21%); silica
gel TLC Rf 0.30 (3:1 hexanes/ethyl acetate); 1H NMR (CDCl3) d 1.22–
1.34 (br m, 12H), 1.56 (quint, 2H, J = 6.8 Hz), 1.71 (quint, 2H,
J = 7.6 Hz), 2.21 (s, 3H), 2.68 (dd, 2H, J = 7.6 and 7.6 Hz), 3.01 (s,
6H), 3.43 (s, 3H), 3.49 (t, 2H, J = 6.8 Hz), 4.60 (s, 2H), 4.71 (s, 2H),
6.16 (s, 1H), and 7.32–7.45 (m, 5H); 13C NMR (CDCl3) d 16.7,
26.2, 28.7, 29.4, 29.5, 29.6, 29.73, 29.76, 31.6, 32.0, 38.3, 55.0,
67.9, 75.3, 96.4, 104.9, 127.8, 127.9, 128.5, 137.7, 141.2, 143.4,
152.9, and 155.7; mass spectrum (APCI), m/z 443.3290 (M+H)+
(C27H43N2O3 requires 443.3274).
5.1.5. 3-(Benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-
(methoxymethoxy)decyl]-4-methylpyridine (11)
To a stirred solution at ꢁ78 °C containing 906 mg (2.95 mmol)
of 3-(benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4-dimethyl-
pyridine (10) in 20 mL of anh THF was added 789 mg (2.95 mmol)
of 1-bromo-9-(methoxymethoxy)nonane (7) followed by 1.99 mL
(3.54 mmol) of a 1.80 M solution of PhLi in hexane. The reaction
mixture was stirred at ꢁ78 °C for 30 min then the reaction mixture
was allowed to warm to 23 °C slowly; the reaction mixture was
then stirred at 23 °C for 30 min. The reaction was quenched with
satd aq ammonium chloride and then poured into 80 mL of water.
The mixture was then extracted with two 80-mL portions of ethyl
acetate. The combined organic layer was then washed with 100 mL
of brine, dried (MgSO4) and concentrated under diminished pres-
sure. The residue was purified by chromatography on a silica gel
column (15 ꢂ 5 cm). Elution with 4:1 hexanes/ethyl acetate affor-
ded
3-(benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-
5.1.8. 6-(Dimethylamino)-2-(10-hydroxydecyl)-4-
(methoxymethoxy)decyl]-4-methylpyridine (11) as a light yellow
oil: yield 877 mg (60%); silica gel TLC Rf 0.30 (9:1 hexanes/ethyl
acetate); 1H NMR (CDCl3) d 1.18–1.22 (br m, 12H), 1.50 (quint,
2H, J = 6.8 Hz), 1.65 (m, 2H), 2.06 (s, 6H), 2.26 (s, 3H), 2.75 (dd,
2H, J = 7.6 and 7.6 Hz), 3.28 (s, 3H), 3.43 (t, 2H, J = 6.8 Hz), 4.53
(s, 2H), 4.82 (s, 2H), 5.78 (s, 2H), 6.79, (s, 1H) and 7.29–7.40 (m,
5H);13C NMR (CDCl3) d 13.3, 16.4, 26.2, 28.9, 29.40, 29.48, 29.50,
29.55, 29.6, 29.7, 32.1, 55.1, 67.9, 75.2, 96.4, 106.4, 121.7, 127.7,
128.3, 128.6, 128.9, 138.8, 142.0, 146.8, 150.8, and 155.9; mass
spectrum (APCI), m/z 493.3426 (M+H)+ (C31H45N2O3 requires
493.3430).
methylpyridin-3-ol (1)
To a stirred solution containing 27.0 mg (0.09 mmol) of 3-
(benzyloxy)-6-(dimethylamino)-2-[10-(methoxymethoxy)decyl]-
4-methylpyridine (13) in 3 mL of methanol was added one drop of
concentrated aq HCl. The reaction mixture was stirred at reflux for
16 h. Then 3 mg of 20% palladium hydroxide on carbon (Degussa
type E101 NE/N) was added to the cooled reaction mixture, which
was stirred at 23 °C under a H2 atmosphere for 15 min. The reac-
tion mixture was filtered through Celite and the filtrate was con-
centrated under diminished pressure. The residue was purified
by chromatography on a silica gel column (10 ꢂ 1 cm). Elution
with 9:1 dichloromethane/methanol afforded 6-(dimethylamino)-
2-(10-hydroxydecyl)-4-methylpyridin-3-ol (1) as a colorless oil:
yield 10 mg (77%); silica gel TLC Rf 0.50 (9:1 dichloromethane/
methanol); 1H NMR (methanol-d4) d 1.28–1.34 (br m, 12H), 1.49
(m, 2H), 1.63 (m, 2H), 2.18 (s, 3H), 2.66 (dd, 2H, J = 7.6 and
7.6 Hz), 2.93 (s, 6H), 3.51 (t, 2H, J = 6.4 Hz) and 6.31 (s, 1H); 13C
NMR (methanol-d4) d 15.5, 25.5, 28.2, 29.16, 29.20, 29.22, 29.24,
29.3, 31.2, 32.2, 38.1, 67.9, 106.0, 138.1, 141.0, 147.1 and 154.2;
mass spectrum (APCI), m/z 309.2553 (M+H)+ (C18H33N2O2 requires
309.2542).
5.1.6. 6-Amino-3-(benzyloxy)-2-[10-(methoxymethoxy)decyl]-
4-methylpyridine (12)
To a stirred solution containing 240 mg (0.49 mmol) of 3-(ben-
zyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-(methoxymethoxy)
decyl]-4-methylpyridine (11) in 10 mL of 9:1 ethanol/water was
added 338 mg (4.87 mmol) of hydroxylamine hydrochloride fol-
lowed by 273 mg (4.87 mmol) of KOH. The reaction mixture was
stirred at reflux for 6 h, then
a second portion of 338 mg
(4.87 mmol) of hydroxylamine hydrochloride was added, followed
by 273 mg (4.87 mmol) of KOH. The reaction mixture was stirred at
reflux for 16 h. The cooled reaction mixture was poured into 50 mL
of water and extracted with two 50-mL portions of dichlorometh-
ane. The combined organic layer was washed with 60 mL of brine,
5.1.9. 6-Amino-2-(10-hydroxydecyl)-4-methylpyridin-3-ol (2)
To a stirred solution containing 113 mg (0.27 mmol) of 6-amino-
3-(benzyloxy)-2-[10-(methoxymethoxy)decyl]-4-methylpyridine

5198
P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
(12) in 10 mL of methanol was added two drops of concentrated
HCl. The reaction mixture was stirred at reflux for 16 h. To the reac-
tion mixture was added 5 mg of 20% palladium hydroxide on car-
bon (Degussa type E101 NE/N). The reaction mixture was stirred
at 23 °C under a H2 atmosphere for 15 min. The reaction mixture
was filtered through Celite and the filtrate was concentrated under
diminished pressure. The residue was purified by chromatography
on a silica gel column (10 ꢂ 1 cm). Elution with 17:3 dichlorometh-
ane/methanol afforded 6-amino-2-(10-hydroxydecyl)-4-methyl-
pyridin-3-ol (2) as a colorless solid: yield 22 mg (29%); mp
138–139 °C; silica gel TLC Rf 0.20 (17:3 dichloromethane/metha-
nol); 1H NMR (methanol-d4) d 1.28–1.35 (m, 12H), 1.48–1.58
(m, 4H), 2.13 (s, 3H), 2.60 (dd, 2H, J = 7.6 and 7.6 Hz), 3.51 (t, 2H,
J = 6.8 Hz) and 6.25 (s, 1H); 13C NMR (methanol-d4) d 15.2, 25.5,
28.8, 29.15, 29.20, 29.22, 29.3, 31.5, 32.2, 61.6, 108.4, 139.2,
141.7, 146.9 and 152.5; mass spectrum (APCI), m/z 281.2231
(M+H)+ (C16H29N2O2 requires 281.2229).
5.1.12. 6-Amino-2,4,5-trimethylpyridin-3-ol (14)35
To a stirred solution containing 10.0 g (48.6 mmol) of 4,5-
bis(hydroxymethyl)-2-methylpyridin-3-ol hydrochloride in 37 mL
of thionyl chloride was added 370 lL (372 mg; 0.51 mmol) of
DMF. The reaction mixture was stirred at reflux for 2 h. The cooled
reaction mixture was treated with 20 mL of diethyl ether. The
resulting suspension was stirred for 1 h and the formed precipitate
was filtered and washed with 20 mL of ether. The solid obtained
was dissolved in 42 mL of glacial acetic acid and 9.80 g (150 mmol)
of zinc dust was added in three portions. The resulting suspension
was stirred and heated at reflux for 2 h. The cooled reaction mix-
ture was filtered and washed with glacial acetic acid. The filtrate
was neutralized with 6 M NaOH and the formed precipitate was fil-
tered and washed with a small amount of brine. The orange precip-
itate was dissolved in 10 M HCl solution and the HCl adduct was
salted out with NaCl. The solid so obtained was dissolved in
200 mL of saturated aq. sodium bicarbonate. Then a diazonium
salt, freshly prepared by slowly mixing 4.00 mL (4.09 g;
43.9 mmol) of aniline in 40 mL of 6 M HCl at 0 °C with a solution
containing 3.00 g (44.0 mmol) of NaNO2 in 15 mL of water, was
added to the reaction mixture dropwise. After 1 h, a red precipitate
that had formed was filtered. The diazo intermediate was dissolved
in 80 mL of 1:1 methanol/formic acid and 14.1 g (220 mmol) of
zinc dust was added in three portions. The reaction mixture was
then stirred at reflux for 2 h. The cooled reaction mixture was then
filtered and the filtrate was washed with hot MeOH. The MeOH
was concentrated and the resulting white precipitate was filtered,
washed with ether, dried, and then dissolved in hot water and ad-
justed to pH 8.0 with 6 M NaOH. Upon cooling the solution, a white
precipitate formed. The precipitate was filtered, dissolved in EtOH
and filtered through a silica gel pad. The filtrate was concentrated
under diminished pressure to afford 6-amino-2,4,5-trimethylpyri-
din-3-ol (14) as a pale orange solid; yield 1.54 g (21%); mp 169–
171 °C, lit35 mp 170–172 °C; silica gel TLC Rf 0.20 (9:1 dichloro-
methane/methanol); 1H NMR (DMSO-d6) d 1.88 (s, 3H), 2.00 (s,
3H), 2.12 (s, 3H), 4.86 (br s, 2 H), and 7.40 (br s, 1H); 13C NMR
(DMSO-d6) d 12.9, 13.4, 19.4, 112.7, 135.0, 140.2, 141.4, and 151.3.
5.1.10. 6-(2,5-Dimethyl-1H-pyrrol-1-yl)-2-(10-hydroxydecyl)-4-
methylpyridin-3-ol (3)
To a stirred solution containing 321 mg (0.65 mmol) of 3-(ben-
zyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-(methoxymethoxy)
decyl]-4-methylpyridine (11) in 5 mL of methanol was added two
drops of concentrated HCl. The reaction mixture was stirred at re-
flux for 16 h. To the cooled reaction mixture was added 10 mg of
20% palladium hydroxide on carbon (Degussa type E101 NE/E)
and the reaction mixture was then stirred at 23 °C under a H2
atmosphere for 15 min. The reaction mixture was filtered through
Celite and the filtrate was concentrated under diminished pres-
sure. The residue was purified by chromatography on a silica
gel column (13 ꢂ 3 cm). Elution with 1:1 hexanes/ethyl acetate
afforded 6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-(10-hydroxydecyl)-4-
methylpyridin-3-ol (3) as a colorless solid: yield 40 mg (17%);
mp 98–99 °C; silica gel TLC Rf 0.42 (1:1 hexanes/ethyl acetate);
1H NMR (CDCl3) d 1.29–1.35 (m, 12H), 1.53 (quint, 2H, J = 6.8 Hz),
1.68 (quint, 2H, J = 6.8 Hz), 2.03 (s, 6H), 2.24 (s, 3H), 2.77 (dd, 2H,
J = 7.6 and 7.6 Hz), 3.60 (t, 2H, J = 6.8 Hz), 5.80 (s, 2H), and 6.80
(s, 1H), 13C NMR (CDCl3) d 12.9, 15.9, 25.6, 28.2, 29.24, 29.25,
29.30, 29.36, 29.38, 32.0, 32.6, 63.0, 106.0, 121.8, 128.5, 134.4,
143.5, 147.9, and 148.3; mass spectrum (APCI), m/z 359.2697
(M+H)+ (C22H35N2O2 requires 359.2699).
5.1.13. 6-(2,5-Dimethyl-1H-pyrrol-1-yl)-2,4,5-trimethylpyridin-
3-ol (15)
To a stirred solution containing 1.36 g (8.94 mmol) of 6-amino-
2,4,5-trimethylpyridin-3-ol (14) in 50 mL of toluene was added
1.32 mL (1.28 g; 11.2 mmol) of 2,5-hexanedione followed by
86.0 mg (0.45 mmol) of p-toluenesulfonic acid. The reaction mix-
ture was stirred and heated at reflux for 14 h using a Dean-Stark
apparatus. The reaction mixture was poured into 80 mL of water
and then extracted with 100 mL of ethyl acetate. The organic solu-
tion was washed with 80 mL of brine, dried (MgSO4) and concen-
trated under diminished pressure. The residue was purified by
chromatography on a silica gel column (15 ꢂ 6 cm). Elution with
2:1 hexanes/ethyl acetate afforded 6-(2,5-dimethyl-1H-pyrrol-1-
yl)-2,4,5-trimethylpyridin-3-ol (15) as colorless needles: yield
1.54 g (75%); mp 218–219 °C; silica gel TLC Rf 0.25 (2:1 hexanes/
ethyl acetate); 1H NMR (methanol-d4) d 1.75 (s, 3H), 1.86 (s, 6H),
2.24 (s, 3H), 2.38 (s, 3H), and 5.77 (s, 2H); 13C NMR (methanol-
d4) d 12.4, 12.8, 13.6, 18.6, 106.7, 129.1, 131.1, 136.7, 142.7,
144.0, and 151.2; mass spectrum (APCI), m/z (M+H)+ 231.1492
(C14H19N2O requires 231.1497).
5.1.11. 6-(2,5-Dimethyl-1H-pyrrolidin-1-yl)-2-(10-
hydroxydecyl)-4-methylpyridin-3-ol (4)
To a stirred solution containing 479 mg (0.65 mmol) of 3-(ben-
zyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-(methoxymethoxy)
decyl]-4-methylpyridine (11) in 5 mL of methanol was added two
drops of concentrated HCl. The reaction mixture was stirred at re-
flux for 16 h. To the cooled mixture was added 10 mg of 20% palla-
dium hydroxide on carbon (Degussa type E101 NE/E) and the
reaction mixture was then stirred at 23 °C under a H2 atmosphere
for 16 h. The reaction mixture was filtered through Celite and the
filtrate was concentrated under diminished pressure. The residue
was purified by chromatography on
a silica gel column
(13 ꢂ 3 cm). Elution with 1:1 toluene/ethyl acetate afforded 6-
(2,5-dimethyl-1H-pyrrolidin-1-yl)-2-(10-hydroxydecyl)-4-methyl-
pyridin-3-ol (4) as a colorless oil: yield 17 mg (5%); silica gel TLC Rf
0.42 (1:1 hexanes/ethyl acetate); 1H NMR (CD3CN) d 1.24–1.34 (m,
18H), 1.46 (m, 2H), 1.68 (m, 4H), 2.01 (m, 2H), 2.25 (s, 3H), 2.59 (t,
2H, J = 7.2 Hz), 3.47 (t, 2H, J = 6.4 Hz), 3.87 (m, 2H), and 6.11 (s,
1H); 13C NMR (CD3CN) d 16.1, 16.8, 19.6, 22.5, 22.6, 26.6, 29.0,
30.2, 30.24, 30.27, 30.33, 30.9, 32.6, 32.8, 33.6, 56.1, 62.6, 106.5,
136.9, 141.3, 143.6 and 152.9; mass spectrum (APCI), m/z
363.3018 (M+H)+ (C22H39N2O2 requires 363.3012).
5.1.14. 3-Benzyloxy-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4,5-
trimethylpyridine (16)
To a stirred solution containing 1.54 g (6.69 mmol) of 6-(2,5-di-
methyl-1H-pyrrol-1-yl)-2,4,5-trimethylpyridin-3-ol (15) in 50 mL
of anh THF was added 1.19 mL (1.71 g; 10.0 mmol) of benzyl bro-
mide followed by 803 mg (20.1 mmol) of 60% sodium hydride sus-

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5199
pension in mineral oil. The reaction mixture was stirred at 23 °C for
16 h. The reaction mixture was poured into 150 mL of water and
extracted with two 100-mL portions of ether. The combined organ-
ic layer was washed with 100 mL of brine, dried (MgSO4) and con-
centrated under diminished pressure. The residue was purified by
chromatography on a silica gel column (15 ꢂ 6 cm). Elution with
hexanes (separation of unreacted benzyl bromide) and then with
4:1 hexanes/diethyl ether afforded 3-benzyloxy-6-(2,5-dimethyl-
1H-pyrrol-1-yl)-2,4,5-trimethylpyridine (16) as yellowish crystals:
yield 1.67 g (78%); mp 64–65 °C; silica gel TLC Rf 0.60 (2:1 hexanes/
ethyl acetate); 1H NMR (CDCl3) d 1.83 (s, 3H), 1.95 (s, 6H), 2.27 (s,
3H), 2.52 (s, 3H), 4.89 (s, 2H), 5.87 (s, 2H), and 7.40–7.50 (m, 5H);
13C NMR (CDCl3) d 12.5, 13.3, 13.8, 19.5, 74.9, 105.8, 128.0, 128.1,
128.5, 128.8, 129.7, 136.7, 141.2, 145.8 149.5, and 151.6; mass
spectrum (APCI), m/z (M+H)+ 321.1969 (C21H25N2O requires
321.1967).
2H), 4.60 (s, 2H), 4.69 (s, 2H), and 7.32–7.45 (m, 5H); 13C NMR
(CDCl3) d 12.9, 13.1, 21.9, 26.3, 29.46, 29.53, 29.54, 29.63, 29.66,
29.79, 29.87, 32.2, 55.1, 67.9, 96.4, 113.6, 127.8, 128.0, 128.6,
137.5, 140.1, 144.9, 150.1, and 152.8; mass spectrum (APCI), m/z
(M+H)+ 429.3114 (C26H41N2O3 requires 429.3117).
5.1.17. 3-Benzyloxy-6-dimethylamino-2-[10-
(methoxymethoxy)decyl]-4,5-dimethylpyridine (19)
To a stirred solution containing 96.0 mg (0.22 mmol) of 6-ami-
no-3-benzyloxy-2-[10-(methoxymethoxy)decyl]-4,5-dimethylpyr-
idine (18) in 3 mL of 1:1 formalin/acetonitrile was added 83.0 mg
(1.32 mmol) of sodium cyanoborohydride followed by 46.0 lL
(48.2 mg; 0.80 mmol) of glacial acetic acid. The reaction mixture
was stirred at 23 °C for 16 h and then poured into 20 mL of satds
aq. sodium bicarbonate. The mixture was then extracted with
20 mL of ether. The organic phase was washed with 20 mL of brine,
dried (MgSO4) and concentrated under diminished pressure. The
residue was purified by chromatography on a silica gel column
(10 ꢂ 2 cm). Elution with 4:1 hexanes/ether afforded 3-benzyloxy-
6-dimethylamino-2-[10-(methoxymethoxy)decyl]-4,5-dimethyl-
pyridine (19) as a colorless oil: yield 78 mg (77%); silica gel TLC Rf
0.30 (4:1 hexanes/diethyl ether); 1H NMR (CDCl3) d 1.21–1.32 (br
m, 12H), 1.55 (m, 2H), 1.73 (m, 2H), 2.18 (s, 3H), 2.19 (s, 3H),
2.64 (m, 8H), 3.36 (s, 3H), 3.50 (t, 2H, J = 6.8 Hz), 4.62 (s, 2H),
4.73 (s, 2H), and 7.32–7.45 (m, 5H); 13C NMR (CDCl3) d 13.2,
15.0, 26.4, 28.8, 29.6, 29.70, 29.74, 29.8, 29.9, 32.0, 42.7, 55.2,
68.0, 75.4, 96.5, 121.5, 127.9, 128.1, 128.7, 137.7, 140.3, 147.4,
149.6, and 158.2; mass spectrum (APCI), m/z (M+H)+ 457.3437
(C28H45N2O3 requires 457.3430).
5.1.15. 3-Benzyloxy-6-(2,5-dimethyl-1H-pyrrol-1-yl)- 2-[10-
(methoxymethoxy)decyl]-4,5-dimethylpyridine (17)
To a stirred solution at ꢁ78 °C containing 670 mg (2.09 mmol)
of 3-(benzyloxy)-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2,4,5-trimethyl-
pyridine (16) in 30 mL of anh THF was added 314 lL (243 mg;
2.09 mmol) of N,N,N0,N0-tetramethylethylenediamine (TMEDA) fol-
lowed by 1.00 mL (2.51 mmol) of a 2.50 M solution of n-BuLi in
hexane. The reaction mixture was stirred at ꢁ78 °C for 5 min and
then 615 mg (2.30 mmol) of 1-bromo-9-(methoxymethoxy)non-
ane (7) was added. The reaction mixture was stirred at ꢁ78 °C
for 30 min, then the reaction mixture was warmed to 23 °C slowly
and stirred for 30 min. The reaction was quenched with satd aq
ammonium chloride and then poured into 50 mL of water. The
mixture was then extracted with 80 mL of ether. The combined or-
ganic layer was then washed with 100 mL of brine, dried (MgSO4)
and concentrated under diminished pressure. The residue was
purified by chromatography on a silica gel column (15 ꢂ 3 cm).
Elution with 4:1 hexanes/ether afforded 3-benzyloxy-6-(2,5-
dimethyl-1H-pyrrol-1-yl)-2-[10-(methoxymethoxy)decyl]-4,5-
dimethylpyridine (17) as a colorless oil: yield 432 mg (41%); silica
gel TLC Rf 0.25 (4:1 hexanes/ether); 1H NMR (CDCl3) d 1.21–1.32
(br m, 12H), 1.55 (m, 2H), 1.68 (m, 2H), 1.81 (s, 3H), 1.92 (s, 6H),
2.25 (s, 3H), 2.82 (dd, 2H, J = 8.0 and 7.6 Hz), 3.33 (s, 3H), 3.48 (t,
2H, J = 6.7 Hz), 4.59 (s, 2H), 4.85 (s, 2H), 5.85 (s, 2H), and 7.38–
7.50 (m, 5H); 13C NMR (CDCl3) d 12.6, 13.4, 13.8, 26.3, 29.45,
29.54, 29.63, 29.65, 29.68, 29.7, 29.9, 32.4, 55.2, 68.0, 75.6, 96.5,
105.8, 127.8, 128.1, 128.4, 128.8, 129.4, 137.0, 141.2, 146.0,
151.3, and 153.4; mass spectrum (APCI), m/z (M+H)+ 507.3590
(C32H47N2O3 requires 507.3587).
5.1.18. 6-Dimethylamino-2-(10-hydroxydecyl)-4,5-
dimethylpyridin-3-ol (5)
To a stirred solution containing 76.0 mg (0.17 mmol) of 3-ben-
zyloxy-6-dimethylamino-2-[10-(methoxymethoxy)decyl]-4,5-
dimethylpyridine (19) in 3 mL of methanol was added 1 drop of
concentrated HCl. The reaction mixture was stirred at reflux for
2 h. The cooled reaction mixture was concentrated under dimin-
ished pressure and the residue was partitioned between 20 mL of
satd aq NaHCO3 and 20 mL of ethyl acetate. The organic layer
was washed with 20 mL of brine, dried over MgSO4 and concen-
trated under diminished pressure. The residue was dissolved in
2 mL of methanol, then 2 mg of 20% palladium hydroxide on car-
bon (Degussa type E101 NE/N) was added and the reaction mixture
was stirred at 23 °C under a H2 atmosphere for 15 min. The
reaction mixture was filtered through Celite and the filtrate was
concentrated under diminished pressure to afford 6-dimethyl-
amino-2-(10-hydroxydecyl)-4,5-dimethylpyridin-3-ol (5) as a yel-
lowish oil: yield 41 mg (79%); silica gel TLC Rf 0.50 (9:1
dichloromethane/methanol); 1H NMR (methanol-d4) d 1.28–1.34
(br m, 12H), 1.51 (m, 2H), 1.65 (m, 2H), 2.15 (s, 3H), 2.17 (s, 3H),
2.66 (s, 6H), 2.72 (dd, 2H, J = 8.0 and 7.2 Hz) and 3.51 (t, 2H,
J = 6.8 Hz); 13C NMR (methanol-d4) d 12.9, 14.6, 26.9, 29.5, 30.46,
30.56, 30.59, 30.66, 30.69, 30.74, 32.7, 33.7, 43.4, 63.0, 123.7,
136.9, 145.8, 146.4 and 156.3; mass spectrum (APCI), m/z
323.2694 (M+H)+ (C19H35N2O2 requires 323.2699).
5.1.16. 6-Amino-3-benzyloxy-2-[10-(methoxymethoxy)decyl]-
4,5-dimethylpyridine (18)
To a stirred solution containing 849 mg (1.68 mmol) of 3-ben-
zyloxy-6-(2,5-dimethyl-1H-pyrrol-1-yl)-2-[10-(methoxymethoxy)
decyl]-4,5-dimethylpyridine (17) in 10 mL of 9:1 ethanol/water
was added 2.34 g (33.6 mmol) of hydroxylamine hydrochloride fol-
lowed by 2.07 g (37.0 mmol) of potassium hydroxide. The reaction
mixture was stirred and heated at reflux for 16 h. The reaction mix-
ture was then poured into 50 mL of water and extracted with two
50-mL portions of ethyl acetate. The combined organic layer was
washed with 60 mL of brine, dried (MgSO4) and concentrated un-
der diminished pressure. The residue was purified by chromatogra-
5.1.19. Cyclic voltammetry
Cyclic voltammetry (CV) studies were carried out using a Model
1030 multi-potentiostat from CH Instruments. The platinum work-
ing electrode 2 mm diameter (model CHI102), platinum wire coun-
ter electrode (model CHI115), and Ag/AgCl reference electrode
(model CHI111) were obtained from CH Instruments. Cyclic vol-
tammetry measurements were performed in a precut 20 mL vial
at room temperature in 2 mL of a 0.1 M tetrabutylammonium per-
chlorate solution in acetonitrile containing the pyridinol analyte at
a 1 mM concentration. The analytes were dissolved in acetonitrile.
phy on
a
silica gel column (10 ꢂ 3 cm). Elution with 9:1
dichloromethane/methanol afforded 6-amino-3-benzyloxy-2-[10-
(methoxymethoxy)decyl]-4,5-dimethylpyridine (18) as a colorless
oil: yield 96 mg (13%); silica gel TLC Rf 0.45 (9:1 dichlorometh-
ane/methanol); 1H NMR (CDCl3) d 1.21–1.32 (br m, 12H), 1.55
(m, 2H), 1.67 (m, 2H), 2.01 (s, 3H), 2.18 (s, 3H), 2.64 (dd, 2H,
J = 8.0 and 8.0 Hz), 3.34 (s, 3H), 3.50 (t, 2H, J = 6.7 Hz), 4.45 (br s,

5200
P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
The vial was covered with a Teflon cap (CHI223) and sealed using
parafilm. All samples were purged with nitrogen for 5 min to re-
move any oxygen. The substrate was then added with a syringe
and stirred for 1 min. Stirring was stopped and the nitrogen was
moved above the sample so that only the headspace was streamed
with nitrogen. The cell was allowed to sit for 1 min to allow for the
diffusion layer to set up and come to equilibrium. The parameters
used for the starting and vertex potential were ꢁ1.5 V and 1.5 V,
respectively, and the scan rate was 100 mV/s. The cyclic voltamm-
agram of CoQ0 was recorded using an initial reductive sweep;
those of the pyridinols involved initial oxidative sweeps. Between
each CV experiment, the electrodes were rinsed with dichloro-
methane. The platinum working electrode was then rinsed with
water and polished with 0.05 micron aluminum powder and dis-
tilled water on a polishing pad (CHI120). Immediately before each
CV was performed, the electrodes were rinsed with acetonitrile.
Results obtained were verified by running duplicates and repeating
experiments in three independent experiments.
5.2.3. NADH oxidase activity assay
Mitochondria were prepared as described.57,58 One beef heart
was ground and blended in sucrose buffer (0.25 M sucrose,
10 mM Tris-HCl, pH 7.8, containing 0.2 mM EDTA) at 4 °C. Cell deb-
ris was removed by centrifugation at 1200ꢂg for 20 min. The
supernatant was filtered through two layers of cheesecloth. Mito-
chondria were harvested by centrifugation at 26,000ꢂg for
15 min and then homogenized in the same buffer with a Dounce
homogenizer. Mitochondria were harvested by centrifugation at
12,000ꢂg for 30 min, and stored at ꢁ80 °C in sucrose buffer. Sub-
mitochondrial particles (SMPs) were prepared as described by
Linnane and Titchener.59 Mitochondria were sonicated with a sonic
dismembrator (Fisher Scientific) in 0.25 M sucrose, 5 mM MgCl2,
1 mM ATP, 10 mM MnCl2, 1 mM sodium succinate, 10 mM Tris–
HCl, pH 7.8, at 4 °C. Cell debris was pelleted by centrifugation at
20,000ꢂg for 7 min at 4 °C. SMPs were harvested at 152,000ꢂg
for 30 min at 4 °C and stored at ꢁ80 °C in 10 mM Tris–HCl, pH
7.5, containing 0.25 M sucrose, 5 mM MgCl2, 2 mM ATP, 2 mM glu-
tathione, and 1 mM sodium succinate. The protein concentration
was determined by BCA titration (Pierce) and the sample was di-
luted as described below.
5.1.20. Electron spin resonance (ESR) study
The compounds were dissolved in deoxygenated benzene solu-
tion for analysis to a final concentration of 10–20 mM. Then
di-tert-butylperoxide (98%, Aldrich) was added to a final concen-
tration of 0.1 M. The sample was inserted in the ESR cavity and
the mercury lamp was turned on. Illumination was done directly
in the cavity by inserting an optic fiber that was connected to
the lamp. The ESR spectra were recorded at regular intervals on a
X-band Bruker ELEXSYS E580 spectrometer using the following
settings: microwave frequency 9.70–9.77 GHz, power 64 mW,
modulation amplitude 0.1G, center field 3462 G, sweep time 168
s, and time constant 40 ms. The simulations were done using the
Winsim software. Control experiments (dark sample, and sample
without test compounds) were measured and gave no signal.
The inhibitory effects of the test compounds on bovine heart
mitochondrial complexes I, III and IV were evaluated. The com-
pounds were dissolved in dimethylsulfoxide (DMSO), and then
used to make serial dilutions. Maximal DMSO concentrations never
exceeded 2% and had no influence on the control enzymatic activ-
ity. Bovine heart SMPs were diluted to 0.5 mg/mL. The enzymatic
activities were assayed at 30 °C and monitored spectrophotometri-
cally with a Beckman Coulter DU-530 (340 nm, e
6.22 mMꢁ1 cmꢁ1).
5.2. Biochemical and biological evaluation
NADH oxidase activity was determined in a reaction medium
(2.5 mL total volume) containing 50 mM Hepes, pH 7.5, and
5 mM MgCl2. The final mitochondrial protein concentration was
5.2.1. Cell culture experiments
Friedreich’s ataxia lymphocytes (Coriell, GM158150) were cul-
tured in RPMI (Gibco, Grand Island, NY, USA) with 15% fetal bovine
serum (Fisher Scientific, TX, USA), 2 mM glutamine (HyClone,
South Logan, Utah, USA) and 1% penicillin–streptomycin mix (Cell-
gro). Cells were maintained in log phase at a concentration be-
tween 1 ꢂ 105 and 1 ꢂ 106 cells/mL.
30 lg. After the pre-equilibration of SMP with inhibitor for
5 min, the initial rates were calculated from the linear portion of
the traces.
5.2.4. Cell viability (trypan blue exclusion assay)
A hemocytometer-based assay was used to determine the num-
ber of viable cells present in the cell suspensions. Briefly, CEM leu-
kemia or FRDA cells were seeded at a density of 5 ꢂ 105 cells/mL
and treated with different concentrations of the test compounds.
Cells were incubated at 37 °C in a humidified atmosphere of 5%
CO2 in air for 17 h. Oxidative stress was then induced by incubation
with 5 mM DEM for 6 h followed by evaluation of cytoprotection.
Cell viability was determined by the use of 0.4% trypan blue. Cyto-
protection by the test compounds was assessed with respect to the
untreated controls. Cells not treated with DEM had >90% cell via-
bility whereas DEM treatment reduced cell viability to <20%. The
ability of the test compounds to protect the cells against the effects
of DEM was determined. Cell viability was expressed as the per-
centage of untreated control. Data are expressed as means S.E.M.
(n = 3).
CCRF-CEM leukemia lymphocytes (ATCC, CRL-119) were cul-
tured in RPMI (Gibco) with 10% fetal bovine serum (Fisher Scien-
tific), 2 mM glutamine (HyClone) and 1% penicillin–streptomycin
mix (Cellgro). Cells were maintained in log phase at a concentra-
tion between 1 ꢂ 105 and 1 ꢂ 106 cells/mL. For the experiments
involving cell viability, the media did not contain glutamine.
5.2.2. Lipid peroxidation assay
A quantitative FACS analysis of lipid peroxidation of FRDA lym-
phocytes, which had been treated with diethyl maleate following
incubation in the presence and absence of the test compounds,
was measured as described previously.56 Briefly, FRDA lymphocytes
(5 ꢂ 105 cells/mL) were treated with the test compounds at final
concentrations of 2.5 and 5 lM and incubated at 37 °C for 16 h in a
humidified atmosphere containing 5% CO2 in air. Cells were treated
with 1
l
M C11 BODIPY581/591 in phenol red-free RPMI-1640 media
5.2.5. Scavenging of reactive oxygen species (ROS)
and incubated at 37 °C in the dark for 30 min. Oxidative stress was
induced with 5 mM DEM in phenol red-free RPMI-1640 media for
2 h. Treated cells were collected by centrifugation at 300ꢂg for
3 min and then washed with phosphate buffered saline. Cells were
Intracellular ROS production was measured using the oxidant-
sensitive fluorescent probe 2,7-dichlorodihydrofluorescein diace-
tate (DCFH-DA) (Molecular Probes, Eugene, OR, USA). One mL of
CEM lymphocytes (2.5 ꢂ 105 cell/mL) were plated in a 24-well
plate, treated with the test compounds and incubated for 16 h at
37 °C, in a humidified atmosphere containing 5% CO2 in air. Cells
were treated with 5 mM diethyl maleate (DEM) for 1 h, collected
by centrifugation at 300ꢂg for 3 min and then washed twice with
phosphate buffered saline (PBS) (Invitrogen, NY, USA). Cells were
resuspended in 250 lL of phosphate buffered saline and analyzed
by FACS (FACS Calibur flow cytometer, Becton Dickinson) to monitor
the change in intensity of the C11 BODIPY581/591-green (oxidized)
fluorescence signal. In each analysis, 10,000 events were recorded.

P. M. Arce et al. / Bioorg. Med. Chem. 20 (2012) 5188–5201
5201
16. Rocha, M.; Apostolova, N.; Hernandez-Mijares, A.; Herance, R.; Victor, V. M.
Curr. Med. Chem. 2010, 17, 3827.
17. Koenig, M. Encycl. Life Sci. 2005, 1.
resuspended in PBS containing 10 mM glucose and incubated at
37 °C in the dark for 25 min with 10 M DCFH-DA. Cells were col-
l
lected by centrifugation at 300ꢂg for 3 min and then washed twice
with PBS. The samples were analyzed immediately by flow cytom-
etry using a 488 nm excitation laser and FL1-H channel 538 nm
emission filter. In each analysis, 10,000 events were recorded after
cell debris was electronically gated. The generation of ROS, mainly
peroxides, was detected as a result of the oxidation of DCFH (kex
488 nm; kem 515–540 nm). Results obtained were verified by run-
ning duplicates and repeating experiments in three independent
experiments. Hydrogen peroxide was used to produce the positive
control.
18. Campuzano, V.; Montermini, L.; Molto, M. D.; Pianese, L.; Cossee, M.;
Cavalcanti, F.; Monros, E.; Rodius, F.; Duclos, F.; Monticelli, A.; Zara, F.;
Canizares, J.; Koutnikova, J.; Bidichandani, S. I.; Gellera, C.; Brice, A.; Trouillas,
P.; De Michele, G.; Filla, A.; De Frutos, R.; Palau, F.; Patel, P. I.; Di Donato, S.;
Mandel, J. L.; Cocozza, S.; Koenig, M.; Pandolfo, M. Science 1996, 271, 1423.
19. Rotig, A.; de Lonlay, P.; Chretien, D.; Foury, F.; Koenig, M.; Sidi, D.; Munnich, A.;
Rustin, P. Nat. Genet. 1997, 17, 215.
20. Koutnikova, H.; Campuzano, V.; Foury, F.; Dolle, P.; Cazzalini, O.; Koenig, M.
Nat. Genet. 1997, 16, 345.
21. Calabrese, V.; Lodi, R.; Tonon, C.; D’Agata, V.; Sapienza, M.; Scapagnini, G.;
Mangiameli, A.; Pennisi, G.; Stella, A. M.; Butterfield, D. A. J. Neurol. Sci. 2005,
233, 145.
22. Sparaco, M.; Gaeta, L. M.; Santorelli, F. M.; Passarelli, C.; Tozzi, G.; Bertini, E.;
Simonati, A.; Scaravilli, F.; Taroni, F.; Duyckaerts, C.; Feleppa, M.; Piemonte, F. J.
Neurol. Sci. 2009, 287, 111.
23. Armstrong, J. S.; Khdour, O. M.; Hecht, S. M. FASEB J. 2010, 24, 2152.
24. Lynch, D. R.; Perlman, S. L.; Meier, T. Arch. Neurol. 2010, 67, 941.
25. Zs.-Nagy, I. Arch. Gerontol. Geriatr. 1990, 11, 175.
26. Rustin, P.; Von Kleist-Retzow, J. C.; Chantrel-Groussard, K.; Sidi, D.; Munnich,
A.; Rotig, A. Lancet 1999, 354, 477.
27. Lagedrost, S. J.; Sutton, M. S.; Cohen, M. S.; Satou, G. M.; Kaufman, B. D.;
Perlman, S. L.; Rummey, C.; Meier, T.; Lynch, D. R. Am. Heart. J. 2011, 161, 639.
28. Meier, T.; Perlman, S. L.; Rummey, C.; Coppard, N. J.; Lynch, D. R. J. Neurol. 2012,
259, 284.
29. Di Prospero, N. A.; Baker, A.; Jeffries, N.; Fischbeck, K. H. Lancet Neurol. 2007, 6,
878.
5.2.6. Maintenance of mitochondrial membrane potential
40,60,61
(Dwm)
Briefly, cells were pre-treated with or without the test com-
pounds for 16 h. The cells were treated with 5 mM DEM for
140 min, collected by centrifugation at 300ꢂg for 3 min and then
washed twice with phosphate buffered saline. The cells were
resuspended in PBS containing 20% glucose and incubated at
37 °C in the dark for 15 min with 250 nM TMRM. Cells were col-
lected by centrifugation at 300ꢂg for 3 min and were then washed
with phosphate buffered saline. The samples were analyzed imme-
diately by flow cytometry using a 488 nM excitation laser and the
FL2-H channel. The results obtained were verified in three inde-
pendent experiments. FCCP, a mitochondrial uncoupler, was used
to produce a negative control. In each analysis, 10,000 events were
recorded.
30. Esposti, M. D.; Ngo, A.; Ghelli, A.; Benelli, B.; Carelli, V.; McLennan, H.; Linnane,
A. W. Arch. Biochem. Biophys. 1996, 330, 395.
31. Briere, J. -J.; Schlemmer, D.; Chretien, D.; Rustin, P. Biochem. Biophys. Res.
Commun. 2004, 316, 1138.
32. Wijtmans, M.; Pratt, D. A.; Valgimigli, L.; DiLabio, G. A.; Pedulli, G. F.; Porter, N.
A. Angew. Chem., Int. Ed. 2003, 42, 4370.
33. Lu, J.; Khdour, O. M.; Armstrong, J. S.; Hecht, S. M. Bioorg. Med. Chem. 2010, 18,
7628.
34. Arce, P. M.; Khdour, O. M.; Goldschmidt, R.; Armstrong, J. S.; Hecht, S. M. ACS
Med. Chem. Lett. 2011, 2, 608.
35. Serwa, R.; Nam, T. G.; Valgimigli, L.; Culbertson, S.; Recto, C. L.; Jeong, B. S.;
Pratt, D. A.; Porter, N. A. Chem. Eur. J. 2010, 16, 14106.
36. Drummen, G. P.; Van Liebergen, L. C.; Op den Kamp, J. A.; Post, J. A. Free Radical
Biol. Med. 2002, 33, 473.
37. Bonini, M. G.; Rota, C.; Tomasi, A.; Mason, R. P. Free Radical Biol. Med. 2006, 40,
968.
38. Karlsson, M.; Kurz, T.; Brunk, U. T.; Nilsson, S. E.; Frennesson, C. I. Biochem. J.
2010, 428, 183.
Acknowledgments
We thank Dr. Marco Flores (Arizona State University, Chemistry
Department ESR facility) for assistance with the ESR experiment.
This work was supported by a research grant from the Friedreich’s
Ataxia Research Alliance.
39. Bernardi, P.; Krauskopf, A.; Basso, E.; Petronilli, V.; Blachly-Dyson, E.; Di Lisa, F.;
Forte, M. A. FEBS J. 2006, 273, 2077.
Supplementary data
40. Ehrenberg, B.; Montana, V.; Wei, M. D.; Wuskell, J. P.; Loew, L. M. Biophys. J.
1988, 53, 785.
41. Miccadei, S.; Kyle, M. E.; Gilfor, D.; Farber, J. L. Arch. Biochem. Biophys. 1988, 265,
311.
42. Tirmenstein, M. A.; Nicholls-Grzemski, F. A.; Zhang, J.-G.; Fariss, M. W. Chem.
Biol. Interact. 2000, 127, 201.
43. Armstrong, J. S.; Whiteman, M.; Rose, P.; Jones, D. P. J. Biol. Chem. 2003, 278,
49079.
Supplementary data (flow cytometric analysis of the ability of
selected analogues to maintain mitochondrial membrane poten-
tial, and of their ability to confer cytoprotection, as well as 1H
and 13C NMR spectral data) associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.bmc.2012.07.005.
44. Esposti, M. D.; Ghelli, A.; Ratta, M.; Cortes, D.; Estornell, E. Biochem. J. 1994, 301,
161.
45. Zharova, T. V.; Vinogradov, A. D. Biochim. Biophys. Acta 1997, 1320, 256.
46. Watabe, M.; Nakaki, T. Mol. Pharmacol. 2008, 74, 933.
47. Fato, R.; Bergamini, C.; Leoni, S.; Lenaz, G. Biofactors 2008, 32, 31.
48. Yagi, M.; Hatefi, Y. J. Biol. Chem. 2001, 276, 19006.
49. Monzote, L.; Stamberg, W.; Patel, A.; Rosenau, T.; Maes, L.; Cos, P.; Gille, L.
Chem. Res. Toxicol. 2011, 24, 1678.
50. Valgimigli, L.; Brigati, G.; Pedulli, G. F.; DiLabio, G. A.; Mastragostino, M.;
Arbizzani, C.; Pratt, D. A. Chem. Eur. J. 2003, 9, 4997.
51. Haefeli, R. H.; Erb, M.; Gemperli, A. C.; Robay, D.; Fruh, I. C.; Anklin, C.; Dallman,
R.; Gueven, N. PLoS One 2011, 6, e17963.
52. Taylor, R. W.; Morris, A. A. M.; Hutchinson, M.; Turnbull, D. M. Eur. J. Hum.
Genet. 2002, 10, 141.
53. Ohta, S.; Ohsawa, I. J. Alzheimers Dis. 2006, 9, 155.
54. Henchcliffe, C.; Beal, M. F. Nat. Clin. Pract. Neurol. 2008, 4, 600.
55. Wijtmans, M.; Pratt, D. A.; Brinkhorst, J.; Serwa, R.; Valgimigli, L.; Pelludi, G. F.;
Porter, N. A. J. Org. Chem. 2004, 69, 9215.
56. Khdour, O. M.; Lu, J.; Hecht, S. M. Pharmaceut. Res. 2011, 28, 2896.
57. Smith, A. L. Methods Enzymol. 1967, 10, 81.
58. Palotti, F.; Lenaz, G. Methods Cell Biol. 2007, 80, 3.
59. Linnane, A. W.; Titchener, E. B. Biochim. Biophys. Acta 1960, 39, 469.
60. Armstrong, J. S.; Yang, H.; Duan, W.; Whiteman, M. J. Biol. Chem. 2004, 279,
50420.
References and notes
1. Duchen, M. R. Diabetes 2004, 53, S96.
2. McBride, H. M.; Neuspiel, M.; Wasiak, S. Curr. Biol. 2006, 16, R551.
3. Mandal, S.; Lindgren, A. G.; Srivastava, A. S.; Clark, A. T.; Banerjee, U. Stem Cells
2011, 29, 486.
4. McLennan, H. R.; Esposti, M. D. J. Bioenerg. Biomembr. 2000, 32, 153.
5. Liu, Y.; Fiskum, G.; Schubert, D. J. Neurochem. 2002, 80, 780.
6. Rhoads, D. M.; Umbach, A. L.; Subbaiah, C. C.; Siedow, J. N. Plant Physiol. 2006,
141, 357.
7. Murakami, K.; Kondo, T.; Kawase, M.; Li, Y.; Sato, S.; Chen, S. F.; Chan, P. H. J.
Neurosci. 1998, 18, 205.
8. Lenaz, G.; Bovina, C.; Formiggini, G.; Parenti Castelli, G. Acta Biochim. Pol. 1999,
46, 1.
9. Santos, D. L.; Palmeira, C. M.; Seica, R.; Dias, J.; Mesquita, J.; Moreno, A. J.;
Santos, M. S. Mol. Cell. Biochem. 2003, 246, 163.
10. Battisti, C.; Formichi, P.; Cardaioli, E.; Bianchi, S.; Mangiavacchi, P.; Tripodi, S.
A.; Tosi, P.; Federico, A. J. Neurol. Neurosurg. Psychiatry 2004, 75, 1731.
11. Emerit, J.; Edeas, M.; Bricaire, F. Biomed. Pharmacother. 2004, 58, 39.
12. Lin, M. T.; Beal, M. F. Nature 2006, 443, 787.
13. Pieczenik, S. R.; Neustadt, J. Exp. Mol. Pathol. 2007, 83, 84.
14. Friederich, M.; Hansell, P.; Palm, F. Curr. Diabetes Rev. 2009, 5, 120.
15. Fernandez-Checa, J. C.; Fernandez, A.; Morales, A.; Mari, M.; Garcia-Ruiz, C.;
Colell, A. CNS Neurol. Disord. Drug Targets 2010, 9, 439.
61. Lu, C.; Armstrong, J. S. Biochem. Biophys. Res. Commun. 2007, 363, 572.