Novel mitochondria-targeted antioxidants: Plastoquinone conjugated with cationic plant alkaloids berberine and palmatine

Article abstract of DOI:10.1007/s11095-011-0504-8

Purpose To develop effective mitochondria-targeted antioxidants composed entirely of natural constituents. Methods Novel mitochondria-targeted antioxidants were synthesized containing plant electron carrier and antioxidant plastoquinone conjugated by nonyloxycarbonylmethyl residue with berberine or palmatine, penetrating cations of plant origin. These compounds, SkQBerb and SkQPalm, were tested in model planar phospholipid membranes and micelles, liposomes, isolated mitochondria and living cells. Results SkQBerb and SkQPalm penetrated across planar bilayer phospholipid membrane in their cationic forms and accumulated in mitochondria isolated or in living human cells in culture. Reduced forms of SkQBerb and SkQPalm as well as C10Berb and C10Palm (SkQBerb and SkQPalm analogs lacking plastoquinol moiety) revealed radical scavenging activity in lipid micelles and liposomes, while oxidized forms were inactive. In isolated mitochondria and in living cells, berberine and palmatine moieties were not reduced, so antioxidant activity of C10Berb and C10Palm was not detected. SkQBerb and SkQPalm inhibited lipid peroxidation in isolated mitochondria at nanomolar concentrations; their prooxidant effect was observed at 1,000 times higher concentrations. In human cell cuture, nanomolar SkQBerb and SkQPalm prevented fragmentation of mitochondria and apoptosis induced by exogenous hydrogen peroxide. Conclusion This is the first successful attempt to construct mitochondria-targeted antioxidants composed entirely of natural components, namely plastoquinone, nonyl, acetyl and berberine or palmatine residues. Springer Science+Business Media, LLC 2011.

Full text of DOI:10.1007/s11095-011-0504-8

Pharm Res (2011) 28:2883–2895
DOI 10.1007/s11095-011-0504-8
RESEARCH PAPER
Novel Mitochondria-Targeted Antioxidants: Plastoquinone
Conjugated with Cationic Plant Alkaloids Berberine
and Palmatine
Konstantin G. Lyamzaev & Antonina V. Pustovidko & Ruben A. Simonyan & Tatyana I. Rokitskaya & Lidia V. Domnina & Olga Yu.
Ivanova & Inna I. Severina & Natalia V. Sumbatyan & Galina A. Korshunova & Vadim N. Tashlitsky & Vitaly A. Roginsky & Yuriy N.
Antonenko & Maxim V. Skulachev & Boris V. Chernyak & Vladimir P. Skulachev
Received: 9 February 2011 /Accepted: 1 June 2011 /Published online: 14 June 2011
#
Springer Science+Business Media, LLC 2011
ABSTRACT
oxidized forms were inactive. In isolated mitochondria
and in living cells, berberine and palmatine moieties were
not reduced, so antioxidant activity of C10Berb and
C10Palm was not detected. SkQBerb and SkQPalm
inhibited lipid peroxidation in isolated mitochondria at
nanomolar concentrations; their prooxidant effect was
observed at 1,000 times higher concentrations. In human
cell cuture, nanomolar SkQBerb and SkQPalm prevented
fragmentation of mitochondria and apoptosis induced by
exogenous hydrogen peroxide.
Purpose To develop effective mitochondria-targeted antiox-
idants composed entirely of natural constituents.
Methods Novel mitochondria-targeted antioxidants were syn-
thesized containing plant electron carrier and antioxidant
plastoquinone conjugated by nonyloxycarbonylmethyl residue
with berberine or palmatine, penetrating cations of plant origin.
These compounds, SkQBerb and SkQPalm, were tested in
model planar phospholipid membranes and micelles, liposomes,
isolated mitochondria and living cells.
Results SkQBerb and SkQPalm penetrated across planar
bilayer phospholipid membrane in their cationic forms and
accumulated in mitochondria isolated or in living human
cells in culture. Reduced forms of SkQBerb and SkQPalm
as well as C10Berb and C10Palm (SkQBerb and SkQPalm
analogs lacking plastoquinol moiety) revealed radical scav-
enging activity in lipid micelles and liposomes, while
Conclusion This is the first successful attempt to construct
mitochondria-targeted antioxidants composed entirely of natural
components, namely plastoquinone, nonyl, acetyl and berberine
or palmatine residues.
.
.
.
KEY WORDS antioxidant berberine mitochondria
.
.
palmatine plastoquinone SkQ
:
:
:
:
K. G. Lyamzaev A. V. Pustovidko R. A. Simonyan T. I. Rokitskaya
V. A. Roginsky
Institute of Chemical Physics
ul. Kosygina 4
:
:
:
:
L. V. Domnina O. Y. Ivanova I. I. Severina G. A. Korshunova
:
:
Y. N. Antonenko B. V. Chernyak V. P. Skulachev
Belozersky Institute of Physico-Chemical Biology
Lomonosov Moscow State University
Vorobyevy Gory 1/40
Moscow 119977, Russia
Moscow 119899, Russia
M. V. Skulachev
Institute of Mitoengineering, Lomonosov Moscow State University
Vorobyevy Gory 1
Moscow 119992, Russia
:
N. V. Sumbatyan V. N. Tashlitsky
Chemical Faculty, Lomonosov Moscow State University
Vorobyevy Gory 1
Moscow 119991, Russia
B. V. Chernyak (*)
V. P. Skulachev
Department of Bioenergetics, Belozersky Institute
of Physico-Chemical Biology( Lomonosov Moscow State University
Vorobyevy Gory 1
Faculty of Bioengineering and Bioinformatics
Lomonosov Moscow State University
Vorobyevy Gory 1
Moscow 119899, Russia
Moscow 119991, Russia
e-mail: bchernyak@yahoo.com

2884
Lyamzaev et al.
ABBREVIATIONS
ing to the Nernst equation, be 1,000 times higher than in the
cytosol. Synthetic ions penetrating mitochondria were
described in 1969 by E.A. Liberman, one of the present
authors (V.P.S.), and coworkers (3). A phosphonium deriv-
ative was one of the first penetrating cations applied to detect
membrane potential in mitochondria and to verify Mitchell’s
chemiosmotic hypothesis (4). Later we suggested that natural
penetrating cations can be used by mitochondria as
“molecular electric locomotives” to accumulate uncharged
conjugated substances in their matrix (5,6).
Δψ
transmembrane electric potential difference
AAPH
AMVN
BLM
2,2′-azobis(2-amidinopropane) dihydrochloride
2,2′-azodi(2,4′-dimethylvaleronitrile)
bilayer planar phospholipid membrane
bovine serum albumin
BSA
C10Berb
C10Palm
C12TPP
13-(decyloxycarbonylmethyl)berberine
13-(decyloxycarbonylmethyl)palmatine
dodecyltriphenylphosphonium
CM-DCF-DA 5-(−6)-chloromethyl-2′,7′-dichlorodihydro-
fluorescein diacetate
trifluoromethoxycarbonylcyanide
phenylhydrazone
malondialdehyde
10-(6-ubiquinonyl) decyltriphenylphosphonium
reactive oxygen species
cationic derivative of plastoquinone or methyl
plastoquinone
10-(6-plastoquinonyl)
decyltriphenylphosphonium
13-[9- (6-plastoquinonyl) nonyloxycarbonyl-
methyl] berberine
13-[9-(6-plastoquinonyl) nonyloxycarbonyl-
methyl] palmatine
In 2001, Murphy and coworkers designed the
mitochondria-targeted synthetic antioxidant MitoQ where
ubiquinol was conjugated with decyltriphenylphosphonium
(7). In 2007, a new mitochondria-targeted antioxidant,
SkQ1 (10-(6-plastoquinonyl)decyltriphenylphosphonium),
was synthesized and studied by our group. It was shown
that substitution of ubiquinol to plastoquinol significantly
improved the antioxidant properties of the compound in
model membranes as well as in isolated mitochondria,
living cells, and organisms (8). SkQ1was found to be a very
efficient agent at preventing a variety of degenerative
processes and prolonging the lifespan of fungi, inverte-
brates, and vertebrates in nanomolar doses (9–15). We
synthesized and tested several analogs of SkQ1, and in one
of them (SkQR1) the phosphonium cation was substituted
by the fluorescent cation rhodamine 19. Rhodamines were
introduced as selective mitochondrial dyes by L.B. Chen
and coworkers (16) and by our group (for review, see (6))
and were widely used for visualization of mitochondria and
measurements of the membrane potential of mitochondria
and bacteria in vivo. We showed that SkQR1 penetrated
through lipid membranes (9,17) and selectively accumulated
in mitochondria of various cell types in vitro (9) and in
mitochondria of some tissues in vivo (10). The antioxidant
and protective effects of SkQR1 were even stronger than
those of SkQ1 in some cellular models in vitro (9) and
especially in models of kidney and brain infarction in vivo
(10,18).
Neither the alkyltriphenylphosphonium nor rhodamine
cations are found in nature, so certain undesirable effects in
long-term in vivo studies might be expected. This inspired us
to search for natural penetrating cations that could be
conjugated with plastoquinone. Our attention was attracted
by berberine and palmatine, which are isoquinoline
alkaloids derived from plants of the Berberidaceae family.
Extracts containing berberine and palmatine have been
used in traditional Chinese medicine for many centuries.
Pharmacological studies of berberine revealed its antidiar-
rheic (19), antiinflammatory (20), antidiabetic (21), antiar-
rhythmic, and antihypertensive (22) effects. Anticancer
activity of berberine mediated by its antiproliferative and
antiangiogenic properties was also reported (23,24). Data
on possible antioxidant properties of berberine and palma-
FCCP
MDA
MitoQ
ROS
SkQ
SkQ1
SkQBerb
SkQPalm
SkQR1
TBARS
10-(6-plastoquinonyl) decylrhodamine 19
thiobarbituric acid-reactive substances
INTRODUCTION
Mitochondria are one of the major sources of ROS
production in animal cells (1). Mitochondrial ROS and
oxidative damage to mitochondria contributes to a wide
range of pathologies, including diabetes, cardiovascular
disorders and infarction, stroke, and neurodegenerative and
other age-related diseases. This is why the mitochondrion is
an important target for delivery of antioxidants. Several
strategies were suggested to specifically address antioxidants
to mitochondria (for review, see (2)). Historically, the first
and the most popular approach was based on application of
so-called penetrating ions, i.e. charged compounds that can
easily penetrate across the mitochondrial membrane due to
delocalization of the charge. The increased ionic radius of
these compounds effectively lowers the enthalpy associated
with desolvating charged species in a lipid environment (the
Born energy) and penetration into the hydrophobic regions of
the membrane. On entering the cell, penetrating cations will
be selectively accumulated within mitochondria since their
interior is the only cell compartment negatively charged
relative to the cytosol. Since energized mitochondria maintain
very high membrane potential across the inner membrane
(Δψ about 180 mV), the concentration of penetrating
monovalent cation in the mitochondrial matrix will, accord-

Berberine and Palmatine Derivatives of Plastoquinone as Mitochondria-Targeted Antioxidants
2885
tine are controversial. Studies in vivo strongly indicated
antioxidant effects of berberine (25) or its metabolites (26).
Scavenging of radicals by berberine in the micellar model
of lipid peroxidation was reported, and this effect was
inherent only in its reduced form (27).
afforded the ester (6), followed by oxidation with N-
bromosuccinimide (NBS) to yield 50% of the target
compound, which was purified by HPLC. A brief descrip-
tion of the target compounds including NMR spectra are
given below.
Earlier, we analyzed the permeability of planar phos-
pholipid bilayer membrane (BLM) and some natural
membranes to berberine and palmatine and found that
both compounds penetrate through BLM and bacterial
plasma membrane as cations (28). It was suggested that the
membrane potential-driven accumulation of these alkaloids
in bacteria contributed to their antimicrobial activity.
Membrane potential-linked uptake of berberine by isolated
mitochondria and by mitochondria in living mammalian
cells was demonstrated (24,29), but the possible role of
mitochondrial targeting in therapeutic action of berberine
remains to be clarified.
In the present work, we report the synthesis of novel
mitochondria-targeted antioxidants where plastoquinone is
conjugated with berberine or palmatine moieties (SkQBerb
and SkQPalm) and also control compounds lacking
quinone moieties C10Berb and C10Palm (Fig. 1a).
SkQBerb and SkQPalm at nanomolar concentrations were
found to inhibit lipid peroxidation in isolated mitochondria
and to prevent cell death induced by exogenous hydrogen
peroxide. The reduced forms of these compounds revealed
radical scavenging activity in liposomes, while the oxidized
forms were inactive. In isolated mitochondria and in living
cells, berberine and palmatine moieties were not reduced,
so the antioxidant activity is due to their plastoquinone
moieties.
SkQBerb, 13-[9-(4,5-dimethyl-3,6-dioxocyclohexa-1,4-
dien-1-yl)nonyloxycarbonyl-methyl]berberine: TLC: Rf
(chloroform—methanol, 65:10)=0.10; Rf (chloroform—
methanol, 4:1)=0.38, HPLC: τ=7.21 min (40–95% B
during 11 min; A: 0.05% TFA, B: 0.05% TFA in MeCN;
Luna C18(2)′ 0.46×15 cm, 5 μm, 1 ml/min). UV (ethanol):
λmax 262 nm, 350 nm, 430 nm. ESI MS: m/z calculated
1
for C39H44NO8 654.77; found 654.18. H NMR (CDCl3,
AV-600, 293 k): 1.38 ppm {10H, m, (4′-8′)-5CH2};
1.49 ppm {2H, quint., J=7.44 Hz, 3′-CH2}; 1.72 ppm
{2H, quint., J=7.52 Hz, 2′-CH2}; 2.02 ppm {3H, s, 3′-
CH3}; 2.03 ppm {3H, s, 4′-CH3}; 2.40 ppm {2H, t, J=
7.74 Hz, 9′-CH2}; 3.16 ppm {2H, s, 5-CH2}; 4.09 ppm
{3H, s, 19-OCH3}; 4.21 ppm {3H, s, 9-OCH3}; 4.29 ppm
{2H, t, J=6.77 Hz, 1′-CH2}; 4.32 ppm {2H, s, 14-CH2};
4.92 ppm {2H, s, 6-CH2}; 6.12 ppm {2H, s, OCH2O};
6.50 ppm {1H, s, 1′-CH}; 6.92 ppm {1H, s, 4-CH};
7.26 ppm {1H, s, 1-CH}; 7.76 ppm {1H, d, J=9,26 Hz,
12-CH}; 7.89 ppm {1H, d, J=9.24 Hz, 11-CH};
10.02 ppm {1H, s, 8-CH}. 13C NMR (CDCl3, AV-600,
293 k): 12.01 ppm {3″-CH3}; 12.39 ppm {4″-CH3};
27.84 ppm {2′-CH2}; 28.28 ppm {5-CH2}; 28.53 ppm
{2′-CH3}; 29.07–29.73 ppm{4′,5′,6′,7′,8′-5CH2};
37.07 ppm {14-CH2}; 56.88 ppm {10-OCH3};
57.50 ppm {6-CH2}; 62.22 ppm {9-OCH3}; 66.53 ppm
{1′-CH2}; 102.27 ppm {OCH2O}; 108.58 ppm {4-CH};
109.31 ppm (1-CH}; 119.31 ppm {12-CH}; 119.73 ppm
{4-CHCCH2-5}; 121.63 ppm {12-CHCCH-13};
126.28 ppm{11-CH}; 126.39 ppm {13-C}; 132.15 ppm
{1″-CH}; 133.56 ppm {1-CHC}; 133.79 ppm {8-CHCC-
9}; 137.85 ppm {NC}; 140.57 ppm {4″-C}; 141.07 ppm
{3″-C}; 146.00 ppm {9-C}; 146.08 ppm {8-CH};
147.70 ppm {2-C); 150.36 ppm {10-C}; 150.38 ppm {3-
C}; 170.70 ppm {15-C=O}; 187.65 ppm {2′-C=O};
187.78 ppm {5″-C=O}.
MATERIALS AND METHODS
The formulas of compounds used in this study and the
scheme of their synthesis are shown in Fig.1. All berberine
and palmatine analogs were synthesized from berberine or
palmatine by the condensation of the cesium salt of 13-
(carboxymethyl)tetrahydroberberine (or tetrahydropalma-
tine) with the appropriate bromide derivatives. SkQBerb
was prepared from berberine bisulfate in six steps.
Berberine (1) was reduced with sodium borohydride in
pyridine (30 min, r.t.), yielding 91% of dihydroberberine
(2), followed by enamine alkylation with methyl bromoace-
tate (1 h, 100°C) producing the 13-substituted iminium salt
as an unstable product, which was reduced immediately
with sodium borohydride in methanol (30 min, r.t.) to give
the 13-substituted tetrahydroberberine (3) with 80% yield.
Hydrolysis of (3) with 2% aqueous solution of lithium
hydroxide in methanol (1.5 h, reflux) gave 61% of the
tetrahydroberberine acid (4) (30). Coupling of the cesium
salt of (4) with bromide derivative (5) (48 h, 60°C) then
SkQPalm, 13-[9-(4,5-dimethyl-3,6-dioxocyclohexa-1,4-
dien-1-yl)nonyloxycarbonyl-methyl]palmatine: TLC: Rf
(chloroform—methanol, 65:10)=0.10; Rf (chloroform—
methanol, 4:1)=0.40. HPLC: τ=7.11 min (40–95% B
during 11 min; A: 0.05% TFA, B: 0.05% TFA in MeCN;
Luna C18(2)′ 0.46×15 cm, 5 μm, 1 ml/min). UV (ethanol):
λmax 262 nm, 350 nm, 430 nm. ESI MS: m/z calculated
1
for C40H48NO8670.34; found 670.41. H NMR (CDCl3,
AV-600, 293 k): 1.32 ppm {10H, m, (4′-8′)-5CH2};
1.49 ppm {2H, quint., J=7.36 Hz, 3′-CH2}; 1.72 ppm
{2H, quint., J=7.32 Hz, 2′-CH2}; 2.03 ppm {3H, s, 3″-
CH3}; 2.04 ppm {3H, s, 4″-CH3}; 2.41 ppm {2H, t, J=
7.36 Hz, 9′-CH2}; 3.21 ppm {2H, s, 5-CH2}; 3.87 ppm

2886
Lyamzaev et al.
Fig. 1 Formulas of SkQBerb (I), SkQPalm (II), C10Berb (III), C10Palm (IV) (a) and the scheme of synthesis of SkQBerb (b). The same scheme was used
for synthesis of C10Berb, SkQPalm, and C10Palm.
{3H, s, 2-OCH3}; 4.02 ppm {3H, s, 3-OCH3}; 4.09 ppm
{3H, s, 10-OCH3}; 4.24 ppm {3H, s, 9-OCH3}; 4.27 ppm
{2H, t, J=6.87 Hz, 1′-CH2}; 4.36 ppm {2H, s, 14-CH2};
5.32 ppm {2H, s, 6-CH2}; 6.51 ppm {1H, s, 1″-CH};
6.94 ppm {1H, s, 4-CH}; 7.39 ppm {1H, s, 1-CH};
7.75 ppm {1H, d, J=9.49 Hz, 12-CH}; 7.88 ppm {1H, d,
J=9.49 Hz, 11-CH}; 10.10 ppm {1H, s, 8-CH}. 13C
(CDCl3, AV-600, 293 k): 12.08 ppm {3″-CH3}; 12.45 ppm
{4″-CH3}; 25.89 ppm {3′-CH2}; 27.84 ppm {5-CH2};
28.58 ppm {2′-CH3}; 29.06–29.75 ppm{4′,5′,6′,7′,8′-
5CH2}; 37.15 ppm {14-CH2}; 56.24 ppm {2-OCH3};
56.33 ppm {3-OCH3}; 57.02 ppm {10-OCH3};
57.59 ppm {6-CH2}; 62.37 ppm {9-OCH3}; 66.53 ppm
{1′-CH2}; 110.80 ppm {4-CH}; 112.40 ppm (1-CH};
118.66 ppm {12-CH}; 119.13 ppm {4-CHCCH2};
121.70 ppm {12-CHCCH-13}; 125.71 ppm{11-CH};
126.29 ppm {13-C}; 132.06 ppm {1″-CH}; 132.12 ppm
{8-CHCC-9}; 133.67 ppm {1-CHC}; 138.19 ppm {NC};
140.57 ppm {4″-C}; 141.07 ppm {3″-C}; 146.19 ppm {9-
C}; 146.39 ppm {8-CH}; 149.04 ppm {2-C); 150.26 ppm
{10-C}; 151.78 ppm {3-C}; 171.05 ppm {15-C=O};
187.65 ppm {2″-C=O}; 187.77 ppm {5″-C=O}.
C10Berb, 13-(decyloxycarbonyl-methyl)berberine: TLC,
Rf (chloroform—methanol, 9:1)=0.33; Rf (chloroform—
methanol, 4:1)=0.62. HPLC: τ=6.56 min (40–95% B
during 11 min; A: 0.05% TFA, B: 0.05% TFA in MeCN;

Berberine and Palmatine Derivatives of Plastoquinone as Mitochondria-Targeted Antioxidants
2887
Luna C18(2)′ 0.46×15 cm, 5 μm, 1 ml/min). UV (ethanol):
λmax 262 nm, 350 nm, 430 nm. ESI MS: m/z calculated
for C32H40NO6534.66; found 534.40. 1H (CDCl3, AV-
600, 293 k): 0.88 ppm {3H, t, J=7.26 Hz, 10′-CH3};
1.27 ppm {14H, m, (2′-9′)-7CH2}; 1.72 ppm {2H, quint.,
J=7.05 Hz, 2′-CH2}; 3.25 ppm {2H, s, 5-CH2}; 4.07 ppm
{3H, s, 10-OCH3}; 4.28 ppm {4H, m, (1′,14)-2CH2};
4.38 ppm {3H, s, 9-OMe}; 5.31 ppm {2H, s, 6-CH2};
6.10 ppm {2H, s, OCH2O}; 6.91 ppm {1H, s, 4-CH};
7.25 ppm {1H, s, 1-CH}; 7.73 ppm {1H, d, J=9.39 Hz,
12-CH}; 7.86 ppm {1H, d, J=9.39 Hz, 11-CH};
10.62 ppm {1H, s, 8-CH}. 13C (CDCl3, AV-600, 293 k):
14.10 ppm {10′-CH3}; 22.68 ppm {9′-CH2}; 25.91 ppm
{3′-CH2}; 28.61–29.53 ppm{2′,4′,5′,6′,7′,5-6CH2};
31.88 ppm {8′-CH2}; 37.18 ppm {14-CH2}; 57.03 ppm
{10-OCH3}; 57.40 ppm {6-CH2}; 63.17 ppm {9- OCH3};
66.50 ppm {1′-OCH2}; 102.15 ppm {OCH2O};
108.70 ppm {4-CH}; 109.16 ppm (1-CH}; 119.38 ppm
{12-CH}; 119.83 ppm {4-CHCCH2-5}; 121.84 ppm {12-
CHCC-13}; 125.69 ppm{13-C}; 125.89 ppm {11-CH};
133.47 ppm {9-C}; 134.25 ppm {1-CHC}; 137.61 ppm
{8-CHCC-9}; 146.70 ppm {3-C}; 147.43 ppm {8-CH};
147.52 ppm {NC}; 150.22 ppm {2-C}; 150.75 ppm {10-
C); 170.71 ppm {15-C=O}.
The length of aliphatic linker in these structures was
chosen when their octanol:water distribution coefficient was
taken into account. A series of analogs with -(CH2)n-
aliphatic chains, where n varied from 4 to 10, was
synthesized. Their distribution coefficients were estimated
using a method based on correlation of this parameter with
retention time in a reversed-phase high performance liquid
chromatography (HPLC) column (31). It was previously
shown for related compounds (SkQ1 and MitoQ) that this
method estimates the coefficient with satisfactory accuracy
(32). The distribution coefficient for analogs with -(CH2)9-
chain (SkQBerb and SkQPalm) and the control compounds
lacking quinone moiety with -(CH2)10- chain (C10Berb and
C10Palm) were of the same order of magnitude as for
SkQ1 according to HPLC method. They were equivalent
to octanol:water distribution ratio within the range 2,000–
3,000:1. Stability of SkQBerb and SkQPalm (as well as
SkQ1) in aqueous medium was high in contrast to SkQR1.
The similar rates of biotransformation in HeLa cells were
observed for all members of SkQ family. These data
indicated that the ester bond in SkQBerb and SkQPalm
was not attacked in living cell, probably due to high
lipophillicity of the compounds. The similar ester bond in
SkQR1 also was not cleaved in the cells, and no
rhodamine-containing products were detected.
C10Palm, 13-(decyloxycarbonyl-methyl)palmatine:
TLC, Rf (chloroform—methanol, 9:1)=0.30; Rf (chloro-
form—methanol, 4:1)=0.60. HPLC: τ=6.45 min (5–95%
B during 11 min; A: 0.05% TFA, B: 0.05% TFA in MeCN;
Luna C18(2)′ 0.46×15 cm, 5 μm, 1 ml/min). UV (ethanol):
λmax 262 nm, 350 nm, 430 nm. ESI MS: m/z calculated
for C33H44NO6550.71; found 550.36. 1H (CDCl3, AV-
600, 293 k): 0.88 ppm {3H, t, J=7.35 Hz, 10′-CH3};
1.27 ppm {14H, m, (2′-9′)-7CH2}; 1.71 ppm {2H, quint.,
J=7.80 Hz, 2′-CH2}; 3.29 ppm {2H, s, 5-CH2}; 3.85 ppm
{3H, s, 2-OCH3};4.01 ppm {3H, s, 3-OCH3}; 4.07 ppm
{3H, s, 10-OCH3}; 4.27 ppm {2H, t, J=6.92 Hz 1′-CH2};
4.33 ppm {2H, s, 14-CH2} 4.38 ppm {3H, s, 9- OCH3};
5.31 ppm {2H, s, 6-CH2}; 6.94 ppm {1H, s, 4-CH};
7.38 ppm {1H, s, 1-CH}; 7.73 ppm {1H, d, J=9.44 Hz,
12-CH}; 7.86 ppm {1H, d, J=9.42 Hz, 11-CH};
10.62 ppm {1H, s, 8-CH}. 13C (CDCl3, AV-600, 293 k):
14.10 ppm {10′-CH3}; 22.68 ppm {9′-CH2}; 25.89 ppm
{3′-CH2}; 28.07–29.51 ppm{2′,4′,5′,6′,7′,5-6CH2};
31.86 ppm {8′-CH2}; 37.19 ppm {14-CH2}; 56.22 ppm
{2-OCH3}; 56.35 ppm {3-OCH3}; 57.02 ppm {10-
OCH3}; 57.50 ppm {6-CH2}; 63.17 ppm {9- OCH3};
66.53 ppm {1′-OCH2}; 110.84 ppm {4-CH}; 112.39 ppm
(1-CH}; 118.79 ppm {4-CHCCH2-5}; 119.12 ppm {12-
CH}; 121.86 ppm {12-CHCC-13}; 125.15 ppm{13-C};
125.84 ppm {11-CH}; 132.36 ppm {9-C}; 133.55 ppm {1-
CHC}; 138.00 ppm {8-CHCC-9}; 146.71 ppm {3-C};
147.41 ppm {8-CH}; 148.26 ppm {NC}; 150.60 ppm {2-
C}; 151.67 ppm {10-C); 171.06 ppm {15-C=O}.
All reagents for synthesis and biochemical studies (if not
indicated otherwise) were from Sigma-Aldrich. Phospholi-
pids were from Avanti Polar Lipids. MitoTracker Green,
MitoTracker Red, CM-DCF-DA (5-(−6)-chloromethyl-
2′,7′-dichlorodihydrofluorescein diacetate), and BOD-
IPY 581/591 C11 were from Molecular Probes. Reagents
for cell cultures were from Gibco.
Permeability of Planar Bilayer Phospholipid
Membrane (BLM)
A planar bilayer phospholipid membrane was formed on a
0.6-mm aperture in a Teflon septum separating the
experimental chamber into two compartments. Escherichia
coli phospholipids (57% phosphatidylethanolamine, 15%
phosphatidylglycerol, 10% cardiolipin, 18% others) were
used to form the BLM. The phospholipids were dissolved in
decane (final concentration, 2%). The two compartments
initially contained equal concentrations of the studied
cation (10–7 M), and a transmembrane electric potential
difference was measured after addition of increasing
concentration of the same cation into one of the compart-
ments. Electric parameters were measured with AgCl
electrodes and a VA-J-5 electrometer (for details, see (33)).
Oxidation of methyl ester of linoleate in micelles of
Triton X-100 was induced by azo-initiator 2,2′-azobis(2-
amidinopropane) dihydrochloride (AAPH, Polysciences)
and followed by measuring O2 consumption with a

2888
Lyamzaev et al.
computerized 5300 Oxygen Biological Monitor (Yellow
Springs Instruments), using a Clark electrode as an O2
sensor as described (34,35). Experiments were performed at
37°C in 50 mM phosphate buffer (pH 7.4) containing
20 mM methyl linoleate, 50 mM Triton X-100, and 3 mM
AAPH. The tested compounds were reduced by NaBH4 in
acidic phosphate buffer (pH 5.0).
Lipid peroxidation was evaluated by measuring thio-
barbituric acid (TBA)-reactive substances (TBARS) as
described in (39). Amounts of TBARS (adducts of MDA
and TBA) are expressed as nanomoles of malondialdehyde
per mg protein. Light absorption was read at 535 and
572 nm with an Aminco DW2000 dual wavelength
spectrophotometer. MDA formation was initiated by
addition of 10 mM ascorbate and 100 μM FeSO4 to
mitochondria oxidizing 4 mM glutamate and 1 mM malate
in medium containing 250 mM sucrose, 10 mM Mops-
KOH, 0.1 mM EGTA, pH 7.4, and mitochondria (0.05 mg
protein/ml).
Production of hydrogen peroxide in mitochondria was
measured using Amplex Red (10-acetyl-3,7-dihydroxy-
phenoxazine; Invitrogen). Resorufin, the fluorescent
product of hydrogen peroxide-induced Amplex Red
oxidation, was measured by its fluorescence (excitation/
emission maxima, 550/595 nm) in the same medium
containing 1 U/ml horseradish peroxidase, 4 μM
Amplex Red, and mitochondria (0.25 mg protein/ml).
The fluorescence was recorded with a Hitachi MPF4
spectrofluorimeter.
Respiration was measured using an Oroboros
Oxygraph-2 k in medium containing 250 mM sucrose,
10 mM HEPES, 1 mM EGTA, 2 mM MgCl2, 2 mM
KH2PO4, BSA 0.1%, pH 7.4, and mitochondria (0.1 mg
protein/ml) at 25°C. Additions of 2.5 mM ADP, 4 mM
glutamate, and 1 mM malate or 5 mM succinate were
made where indicated in the figures.
Cell culture HeLa cells and human subcutaneous diploid
fibroblasts (Cell Culture Collection, Institute of Medical
Genetics, Russian Academy of Medical Sciences, Moscow)
were cultured in Dulbecco’s modified Eagle’s medium
(DMEM) supplemented with 10% fetal calf serum, strepto-
mycin (100 U/ml), and penicillin (100 U/ml). Fibroblasts
between 5 and 10 passages were used for experiments.
Accumulation of the antioxidants was analyzed after
incubation of substrate-attached HeLa cells with 5 μM
SkQBerb or SkQPalm. Fifty μM verapamil was added to
the cells 15 min before SkQBerb or SkQPalm. Cells were
detached with trypsin and EDTA, washed, and lysed in the
“RIPA” medium, and fluorescence (excitation/emission
maxima, ~350/530 nm) was measured using the
FluoroMax-3 spectrofluorimeter.
Liposomes were prepared by evaporation under a
stream of nitrogen of a solution of egg yolk phosphatidyl-
choline and heart cardiolipin (80%:20%, w:w) in chloro-
form, followed by hydration with buffer solution containing
100 mM KCl, 10 mM Tris, pH 7.4. The mixture was
vortexed, passed through a cycle of freezing and thawing,
and extruded through a 0.1-μm pore size Nucleopore
polycarbonate membrane, using an Avanti Mini-Extruder.
Lipid oxidation was induced by 2,2′-azodi(2,4′-dimethylva-
leronitrile) (AMVN, Wako Pure Chemical Industries) and
measured via fluorescence of BODIPY 581/591 C11 as
described previously (36,37). The fluorescence at 594 nm
(excitation at 540 nm) was monitored with a Panorama
Fluorat 02 spectrofluorimeter (Lumex, Russia). Experi-
ments were carried out using 0.1 mg/ml liposomes,
0.4 μM BODIPY 581/591C11, and 0.25 mM AMVN at
60°C in 100 mM KCl, 10 mM Tris, pH 7.4. The tested
compounds were reduced by NaBH4 in aqueous buffer
solution at neutral pH.
Mitochondria from rat heart were isolated as in (38) and
suspended in medium containing 250 mM sucrose, 10 mM
Mops-KOH, pH 7.4, 1 mM EGTA, and 0.1% BSA.
Protein concentration of mitochondria was measured with
bicinchoninic acid according to the reagent producer’s
instruction (Pierce), using BSA as a standard. Uptake of
SkQBerb and SkQPalm by mitochondria was measured in
medium containing 250 mM sucrose, 10 mM HEPES,
1 mM EGTA, 2 mM MgCl2, 2 mM KH2PO4, BSA 0,1%,
pH 7,4, (“Medium A”), 4 mM glutamate, 1 mM malate,
and SkQBerb or SkQPalm (5 μM in both cases) and
mitochondria (0.1 mg protein/ml) at 25°C. After incuba-
tion, the mitochondria were sedimented, lysed in medium
containing 150 mM NaCl, 50 mM Tris (pH 7,4), 0.1%
SDS, 0.25% sodium deoxycholate, and 1% Triton X 100
(“RIPA”), and fluorescence (excitation/emission maxima,
350/530 nm) was measured using a FluoroMax-3 spectro-
fluorimeter (Jobin Yvon Horiba).
Reduction of SkQBerb and SkQPalm (5 μM in both
cases) in mitochondria was measured after 30 min incuba-
tion in “Medium A” containing 5 mM succinate and 1 μM
rotenone. In some assays, 5 mM KCN was added. After
incubation, the mitochondria were sedimented and lysed in
medium “RIPA,” and fluorescence spectra (excitation at
350 nm) were measured using the FluoroMax-3 spectroflu-
orimeter. The spectra of the reduced compounds were
measured after addition of NaBH4 in “RIPA” lysates.
For microscopy, HeLa cells were incubated with 0.5 μM
SkQBerb or SkQPalm for 60 min. Two hundred nM
MitoTracker Red was added 15 min before the observa-
tion. In experiments on fragmentation of mitochondria, the
organelles were visualized with MitoTracker Green
(200 nM, 15 min) and analyzed with an Axiovert
microscope (Carl Zeiss) or with an LSM 510 confocal
microscope (Carl Zeiss). A total of 300 cells was counted in
each sample.

Berberine and Palmatine Derivatives of Plastoquinone as Mitochondria-Targeted Antioxidants
2889
Cell viability was determined with CellTiter-Blue
reagent (Promega). Intensity of fluorescence was measured
with a Fluoscan Ascent fluorimeter (Thermo).
followed by measuring O2 consumption. Antioxidant chain-
breaking activity of the quinol conjugates was characterized
by the rate constant k1 for the reaction between the peroxyl
•
radical of oxygenated linoleate residue (LO2 ) and the
quinol (QH2):
RESULTS AND DISCUSSION
Model Lipid Membranes
LOꢀ2 þ Q H2 ! LOOH þ Q Hꢀ;
which competes with the reaction of chain propagation:
LOꢀ2 þ LH ! LOOH þ Lꢀ:
In the first series of experiments, the novel compounds were
tested for their ability to penetrate through planar
phospholipid bilayer membrane (BLM). If a cation pene-
trates through a membrane separating two compartments
differing in concentrations of this cation, a 60 mV
transmembrane electric potential difference (Δψ) should
be generated for ten-fold cation gradient according to the
Nernst equation. SkQBerb and SkQPalm were shown to
generate Δψ of proper direction (compartment with lower
cation positive). Within the concentration range from 10–7
to 10–6 M, SkQBerb and SkQPalm generated almost the
same Δψ as SkQ1 (Fig. 2), indicating that permeabilities of
BLM for these three compounds were similar. Theoretical
(Nernstian) potential was not reached at such low concen-
trations of the cations. Further increase in the concen-
trations was limited by poor solubility of SkQBerb and
SkQPalm in aqueous solutions.
SkQ1, SkQBerb, and SkQPalm were reduced by
NaBH4 before the experiment. The addition of the reduced
compounds resulted in a strong decrease in the rate of
oxygen consumption, which then increased progressively
with consumption of the antioxidant. Importantly, the
oxidized forms of the compounds were inactive in this
system. The absolute values of k1 calculated as described in
(34,35) were 1.3·105 M−1 −1
1.9·105 M−1 −1
for SkQBerb, while the value for SkQ1
measured earlier was 2.2·105 M−1 −1
. Although the rate
s
for SkQPalm and
s
s
constants for the chain-breaking reaction were similar for
all three compounds, the periods of inhibition were
significantly longer with SkQBerb and SkQPalm, indicating
some additional antioxidant activity of these compounds
compared with SkQ1. We analyzed reduced C10Berb and
C10Palm (lacking the quinol residue) in the same assay and
found some chain-breaking activity. Again the oxidized
forms of the compounds were almost inactive. These data
are in line with the reports on antioxidant activity of
berberine and palmatine in various models (25–27,41).
In experiments on isolated mitochondria, it was shown
that cardiolipin was critical for initiation of phospholipid
peroxidation and that SkQ1 effectively protected mito-
chondrial cardiolipin against oxidation (42). To compare
the antioxidant activity of the novel compounds in
cardiolipin-containing membranes, we used liposomes
prepared from egg phosphatidylcholine and heart cardio-
lipin (80%:20%, w:w). Peroxidation was initiated by azo-
initiator AMVN and measured using the fluorescent
lipophilic probe BODIPY 581/591C11 as described (40).
The addition of reduced SkQBerb and SkQPalm prior to
AMVN led to concentration-dependent inhibition of
BODIPY 581/591C11 oxidation, while the oxidized forms
of the compounds were substantially less active (Fig. 3a). It
was found that reduced forms of SkQBerb and SkQPalm
exhibited stronger antioxidant activity compared to SkQ1
at the same concentration. This suggests that the palmatine
and berberine moieties in the molecules possess additional
antioxidant activity in this model. In line with this
suggestion, it was found that C10Palm reduced by NaBH4
also protected BODIPY 581/591C11 from oxidation
initiated by AMVN (Fig. 3b). Similar data were obtained
Antioxidant activity of SkQBerb and SkQPalm was
analyzed using lipid micelles as described earlier (40).
Peroxidation of the methyl ester of linoleate in micelles of
Triton X-100 was induced by the azo-initiator AAPH and
Fig. 2 Generation of transmembrane electric potential difference (Δψ)
across a BLM. Two compartments of the experimental chamber separated
by a BLM (formed on an aperture in a Teflon septum) initially contained
equal concentrations (10–7 M) of the studied cation: SkQBerb (−●-),
SkQPalm (−▲-) and SkQ1 (−■-).Δψ was measured after addition of
increasing cation concentration into one compartment. The theoretical
Nernstian dependence of Δψ on the gradient of a monovalent cation is
shown for comparison.

2890
Lyamzaev et al.
Fig. 3 Antioxidant activity of SkQBerb and SkQPalm (a) and C10Palm (b) in phosphatidylcholine/cardiolipin liposomes. All the tested compounds were
added at 2 μM concentration . Lipid oxidation was induced by azo-initiator AMVN added at zero time and monitored as a decrease in fluorescence of
BODIPY 581/591 C11. In certain cases, the tested compounds were pretreated by NaBH4 which reduced the plastoquinone as well as the berberine and
palmatine moieties of the studied compounds (red curves).
with C10Berb (data not shown). Oxidized forms of
C10Berb and C10Palm were ineffective. These results
together with the data of experiments with lipid peroxida-
tion in micelles demonstrated high antioxidant activity of
SkQBerb and SkQPalm and indicated a contribution of the
cationic residues (in reduced form) to this activity.
formation initiated by Fe2+ and ascorbate. It was shown
earlier that under these conditions cardiolipin was rapidly
oxidized, and this effect was strongly lowered by 100 nM
SkQ1 (9). The data shown in Fig. 5 demonstrate that
SkQBerb and SkQPalm are almost as effective as SkQ1 in
prevention of MDA accumulation. Uncoupler significantly
decreased the efficiency of SkQBerb, consistent with the
data on Δψ-dependent accumulation of SkQBerb in
mitochondria (Fig. 5). C10Berb and C10Palm, the analogs
lacking the quinol residue, were inactive, confirming the
inability of the mitochondrial electron transport chain to
reduce the palmatine and berberine residues.
We also estimated the prooxidant action of SkQBerb and
SkQPalm by measuring H2O2 production in mitochondria
oxidizing glutamate and malate in State 4. Figure 6 shows
that the prooxidant activity became pronounced at micro-
molar concentration of these compounds (cf. antioxidant
effects developing at nanomolar concentrations, see above
Fig. 5). Prooxidant activity of SkQBerb and SkQPalm
proved to be lower than in the case of SkQ1. To analyze
this difference, we addressed the effects of SkQBerb and
SkQPalm on respiration of mitochondria. We found that
these compounds at 10–6 – 10–5 M severely inhibited
respiration in State 3 when glutamate and malate were used
as substrates (Fig. 7a). Oxidation of succinate was much less
sensitive. Addition of succinate after almost complete
inhibition of glutamate+malate oxidation strongly stimulated
oxygen consumption (Fig. 7b), indicating that the inhibitory
effect was not related to inhibition of ATP synthase or ATP/
ADP transporter. These data are consistent with earlier
observations on inhibition of Complex I by berberine
(29,43). Taking into account the slow development of the
inhibition, we preincubated mitochondria for 20 min with
Isolated Mitochondria
Uptake of SkQBerb and SkQPalm by isolated mitochon-
dria was measured using their fluorescence. It was shown
that accumulation of both compounds was completed in
15–20 min. Addition of uncoupler FCCP discharging Δψ
inhibited accumulation only partially, consistent with high
distribution coefficient of SkQBerb and SkQPalm in
membrane:water systems (data not shown). The similar
time course of accumulation and the effect of uncoupler
were observed earlier for SkQ1 using a delocalized cation-
sensitive electrode (9).
It was shown earlier that the plastoquinone residue of
SkQ1 is reduced by the mitochondrial respiratory chain (9).
The same parameters of plastoquinone reduction were
observed for SkQBerb and SkQPalm based on measure-
ments of absorbance at 274 nm (not shown). However, no
reduction of the berberine or palmatine residues was
observed either with NAD-linked substrates or with
succinate. Inhibition of respiration with rotenone (Fig.4)
or with KCN (not shown), which shifted redox balance of
mitochondria to reduced state, was not able to induce
reduction of the berberine or palmatine residues.
To estimate antioxidant activity of SkQBerb and
SkQPalm in isolated mitochondria, we measured preven-
tion by these compounds of malondialdehyde (MDA)

Berberine and Palmatine Derivatives of Plastoquinone as Mitochondria-Targeted Antioxidants
2891
Fig. 4 Alkaloid moieties of SkQBerb (a) and SkQPalm (b) were not reduced in isolated mitochondria. Mitochondria were incubated for 30 min with
5 μM SkQBerb or SkQPalm in medium containing 5 mM succinate and 1 μM rotenone. Fluorescence spectra (excitation at 350 nm) were measured after
solubilization of mitochondria with detergent mixture (curves SkQBerb and SkQPalm). The spectrum of mitochondria without SkQBerb or SkQPalm
(“blank”) and the spectra of the reduced SkQBerb and SkQPalm measured after addition of NaBH4 to solubilized mitochondria are shown.
the tested compounds and measured constant oxygen
consumption rates after addition of glutamate + malate
and ADP. Half-maximal inhibitory concentrations (C1/2) are
presented in Table I.
We found that SkQBerb and C10Berb inhibited respira-
tion with similar efficiency, while the palmatine conjugates
were slightly less efficient (Table I). The effects of SkQBerb
and C10Berb were observed at 5–10 times lower concen-
trations than that of berberine, probably due to greater
hydrophobicity of the former compounds. This conclusion is
consistent with the analysis of Complex I inhibition by
aliphatic conjugates of berberine (43). It should be men-
tioned that the inhibitory (as well as the prooxidant) action of
SkQBerb and SkQPalm was observed at 1,000 times higher
concentrations than their antioxidant effects. The window
between anti- and prooxidant concentrations of SkQBerb
and SkQPalm was practically the same as for the other
members of SkQ family (9) confirming that plastoquinone is
the only redox active component of these compounds.
Cell Culture
We studied the interaction of SkQBerb and SkQPalm with
HeLa carcinoma cells and human subcutaneous fibroblasts
Fig. 5 Antioxidant activity of SkQBerb and SkQPalm in isolated
mitochondria. Mitochondria were incubated with antioxidants for
20 min. Then lipid peroxidation was initiated by addition of 10 mM
ascorbate and 100 μM FeSO4 . Accumulation of TBARS was measured
and expressed as nanomoles of malondialdehyde per mg protein. 1 μM
FCCP was added where indicated.
Fig. 6 Prooxidant action of SkQBerb (−●-) and SkQPalm (−○-) in
isolated mitochondria. Production of hydrogen peroxide was measured
using Amplex Red and horseradish peroxidase in medium containing
glutamate (4 mM) and malate (1 mM) as respiratory substrates.

2892
Lyamzaev et al.
Fig. 7 Effects of SkQBerb on respiration of isolated mitochondria. (a) Time course of oxygen consumption in medium containing glutamate (4 mM) and
malate (1 mM) as respiratory substrates and ADP (2.5 mM). SkQBerb was added when indicated. (b) Rate of respiration recorded by an Oxygraph-2 k
(Oroboros) in the same medium as in (a). SkQBerb (5 μM) and succinate (5 mM) were added when indicated. For other additions, see (a).
in culture. Both compounds were accumulated in HeLa
cells during 1–1.5 h as seen by increase in cell fluorescence
(for SkQBerb, see Fig. 8a). The fluorescence did not change
after accumulation, indicating that the living cell, just as
isolated mitochondria, cannot reduce berberine and pal-
matine moieties of SkQBerb and SkQPalm (data not
shown). Removal of SkQBerb and SkQPalm from the
medium was followed by slow release of the compounds
from the cells. The uncoupler FCCP prevented SkQBerb
and SkQPalm accumulation and stimulated their efflux,
confirming the membrane potential-driven nature of the
accumulation of the cations in intracellular mitochondria.
Verapamil, an inhibitor of P-glycoprotein, stimulated
accumulation of SkQBerb and SkQPalm in HeLa cells,
indicating that these compounds are substrates of the
multidrug resistance (MDR) pump. Similar kinetics of
accumulation and release of SkQBerb and SkQPalm were
observed in human fibroblasts (data not shown). The only
difference between HeLa cells and fibroblasts was that in
the latter case verapamil was ineffective, which is in line
with a low content of P-glycoprotein in fibroblasts in
comparison with HeLa cells. Subcellular localization of
SkQBerb and SkQPalm coincided with mitochondria
stained with MitoTracker Red, indicating high selectivity
of the novel mitochondria-targeted antioxidants (for
SkQBerb, see Fig. 8b). All these observations closely
resemble the behavior of SkQR1 (9,44,45). An important
difference was found in parameters of release. The release
of SkQR1 from both cell types was incomplete even in the
presence of uncoupler, while SkQBerb and SkQPalm were
released almost completely. These data might be related to
a lower distribution coefficient of SkQBerb and SkQPalm
in the octanol:water system as compared to SkQR1.
Moreover, it is also possible that some irreversible
modifications of cellular components inherent to SkQR1
is absent in the case of the novel antioxidants.
Our earlier studies using mitochondria-targeted antiox-
idants demonstrated (9,44,46) that oxidative stress induced
by exogenous H2O2 results in an excessive production of
endogenous ROS in mitochondria. That is why we have
exploited this model in the following experiments. One of
the earliest consequences of oxidative stress is fragmenta-
tion of extended tubular mitochondria into small spherical
ones (the “thread–grain transition” (47)). It was shown
earlier that incubation with SkQ1 at nanomolar concen-
trations for 2 h prevents the H2O2-induced fragmentation
of mitochondria even while the level of intracellular ROS
(measured with (5-(−6)-chloromethyl-2′,7′-dichlorodihydro-
fluorescein diacetate, CM-DCF-DA) remains unaffected
(44). This indicates that mitochondrial fragmentation is a
result of oxidation of some components of mitochondrial
membrane initiated by ROS produced inside mitochon-
Tab le I Inhibition of Mitochondrial Respiration by Berberine- and
Palmatine-Containing Compounds. Mitochondria were Preincubated for
20 min with the Tested Compounds, and Half-Maximal Inhibitory
Concentrations (C1/2) were Measured in the Presence of Glutamate,
Malate and ADP as in Fig. 7a
Compound
C1/2, M
SkQBerb
SkQPalm
C10Berb
C10Palm
2.5×10−7
1.5×10−6
3.5×10−7
6.0×10−6

Berberine and Palmatine Derivatives of Plastoquinone as Mitochondria-Targeted Antioxidants
2893
Fig. 9 SkQBerb and SkQPalm prevented fragmentation of mitochondria
induced by H2O2. Fibroblasts were incubated with antioxidants for 2 h and
treated with 400 μM H2O2 for 3 h. Mitochondria were visualized with
MitoTracker Green, and cells with elongated mitochondria were counted.
protective effects of mitochondria-targeted antioxidants
suggests slow redistribution of this compound in the
population of mitochondria in the cell. It was shown in
various cell types, including fibroblasts, that the cells
contain a fraction of mitochondria with low membrane
potential that would accumulate cationic antioxidants only
slowly (46). These same “impaired” mitochondria probably
produced the major part of ROS under oxidative stress (as
observed in other cell models (48,49). Apparently, just
these ROS determine the fate of the cell. An alternate
explanation is that slow development of the effects of
mitochondria-targeted antioxidants could be related to
induction of endogenous antioxidant systems in the cell.
Fig. 8 Accumulation of SkQBerb in cells. (a) HeLa cells were incubated with
5 μM SkQBerb . 50 μM verapamil and 10 μM FCCP were added to the cells
15 min before SkQBerb. After 2 h, the medium was changed to the similar
one but without SkQBerb (indicated with vertical line). Fluorescence
(excitation/emission maxima, ~350/530 nm) was measured after detachment,
washing, and lysis of the cells with detergent mixture. (b) HeLa cells were
incubated with 0.5 μM SkQBerb for 60 min and 200 nM MitoTracker Red
was added, incubated for 15 min, and analyzed with an Axiovert microscope
(Carl Zeiss) equipped with a Neofluar 100×NA 1.3 objective.
dria. These observations were confirmed with SkQBerb
and SkQPalm (Fig. 9). The control compounds C10Berb
and C10Palm were ineffective in protection of mitochon-
drial structure, indicating that the nanomolar concentra-
tions oxidized forms of berberine and palmatine moieties of
SkQBerb and SkQPalm do not have detectable antioxidant
properties in living cells as well as in isolated mitochondria
and model membranes.
In contrast to protection of mitochondrial structure,
prevention of fibroblast cell death induced by exogenous
H2O2 required prolonged (24 h) preincubation with SkQ1,
and the protective effect was correlated with the decrease in
intracellular ROS accumulation (9,44). Prevention of the
cell death by SkQBerb and SkQPalm also required 24 h
preincubation, and these antioxidants (as well as SkQ1)
were effective at nanomolar concentrations (Fig. 10). Again,
C10Berb and C10Palm were ineffective in protection of the
cells. The possible explanation of delayed antioxidant and
Fig. 10 SkQBerb and SkQPalm protected cells against H2O2-induced
damage. Fibroblasts were incubated with the antioxidants for 24 h and
treated with 400 μM H2O2 for the next 24 h. Cell viability was
determined with CellTiter-Blue reagent (Promega).

2894
Lyamzaev et al.
7. Kelso GF, Porteous CM, Coulter CV, Hughes G, Porteous WK,
Ledgerwood EC, et al. Selective targeting of a redox-active
ubiquinone to mitochondria within cells: Antioxidant and anti-
apoptotic properties. J Biol Chem. 2001;276:4588–96.
8. Skulachev VP. A biochemical approach to the problem of aging:
“Megaproject” on membrane-penetrating ions. The first results
and prospects. Biochemistry (Mosc). 2007;72:1385–96.
9. Antonenko YN, Avetisyan AV, Bakeeva LE, Chernyak BV,
Chertkov VA, Domnina LV, et al. Skulachev VP. Mitochondria-
targeted plastoquinone derivatives as tools to interrupt execution
of the aging program. 1. Cationic plastoquinone derivatives:
synthesis and in vitro studies. Biochemistry (Mosc). 2008;73:1273–
87.
However, at nanomolar concentrations of SkQBerb and
SkQPalm, we did not detect any sign of prooxidant
action of these compounds, which might be a stimulus
for induction. No induction of expression of the major
antioxidant enzymes at the level of mRNA was detected
(data not shown). Prooxidant activity of SkQBerb and
SkQPalm was pronounced at micromolar concentrations,
resulting in reversal of the protective action on mito-
chondrial structure and cell viability under oxidative
stress (Figs. 9, and 10).
In conclusion, novel mitochondria-targeted antioxidants
where plastoquinone is conjugated with berberine or
palmatine moieties (SkQBerb and SkQPalm) were synthe-
sized and tested in various models. Their antioxidant
activity in living cells depends on the plastoquinone
moieties and is not significantly different from that of
conjugates of plastoquinone with synthetic cations triphe-
nylphosphonium (SkQ1) or rhodamine 19 (SkQR1). The
isoquinoline plant alkaloids derived from the Berberidaceae
family have been used in traditional Chinese medicine for a
long time and reveal numerous therapeutic effects. The
new mitochondria-targeted antioxidants constructed exclu-
sively from natural components could be promising
candidates for therapeutic applications.
10. Bakeeva LE, Barskov IV, Egorov MV, Isaev NK, Kapelko VI,
Kazachenko AV, et al. Mitochondria-targeted plastoquinone
derivatives as tools to interrupt execution of the aging program.
2. Treatment of some ROS- and age-related diseases (heart
arrhythmia, heart infarctions, kidney ischemia, and stroke).
Biochemistry (Mosc). 2008;73:1288–99.
11. Agapova LS, Chernyak BV, Domnina LV, Dugina VB, Efimenko
AY, Fetisova EK, et al. Mitochondria-targeted plastoquinone
derivatives as tools to interrupt execution of the aging program.
3. Inhibitory effect of SkQ1 on tumor development from p53-
deficient cells. Biochemistry (Mosc). 2008;73:1300–16.
12. Neroev VV, Archipova MM, Bakeeva LE, Fursova A, Grigorian
EN, Grishanova AY, et al. Mitochondria-targeted plastoquinone
derivatives as tools to interrupt execution of the aging program. 4.
Age-related eye disease. SkQ1 returns vision to blind animals.
Biochemistry (Mosc). 2008;73:1317–28.
13. Anisimov VN, Bakeeva LE, Egormin PA, Filenko OF, Isakova
EF, Manskikh VN, et al. Mitochondria-targeted plastoquinone
derivatives as tools to interrupt execution of the aging program. 5.
SkQ1 prolongs lifespan and prevents development of traits of
senescence. Biochemistry (Mosc). 2008;73:1329–42.
ACKNOWLEDGMENTS & DISCLOSURES
We thank M. L. Domninskaya and A. A. Pashkovskaya for
help in some experiments. This study was supported by
research contract with Mitotech LLC, the Russian Ministry
of Education and Science (grants “Leading Scientific
Schools” No. 5762.2008.4 and “Centers for Science and
Education” No. 14.740.11.0010) and the Russian Founda-
tion for Basic Research (grant Nos. 09-04-01454, 09-04-
00890, 10-04-91331).
14. Skulachev VP, Anisimov VN, Antonenko YN, Bakeeva LE,
Chernyak BV, Erichev VP, et al. An attempt to prevent
senescence: a mitochondrial approach. Biochim Biophys Acta.
2009;1787:437–61.
15. Skulachev MV, Antonenko YN, Anisimov VN, Chernyak BV,
Cherepanov DA, Chistyakov VA, et al. Mitochondrial-targeted
plastoquinone derivatives. Effect on senescence and acute age-
related pathologies. Current Drug Targets. 2011; (accepted).
16. Johnson LV, Walsh ML, Chen LB. Localization of mitochondria
in living cells with rhodamine 123. Proc Natl Acad Sci USA.
1980;77:990–4.
17. Rokitskaya TI, Klishin SS, Severina II, Skulachev VP, Antonenko
YN. Kinetic analysis of permeation of mitochondria-targeted
antioxidants across bilayer lipid membranes. J Membr Biol.
2008;224:9–19.
18. Plotnikov EY, Silachev DN, Chupyrkina AA, Danshina MI,
Jankauskas SS, Morosanova MA, et al. New-generation Skulachev
ions exhibiting nephroprotective and neuroprotective properties.
Biochemistry (Mosc). 2010;75:145–50.
19. Taylor CT, Baird AW. Berberine inhibition of electrogenic ion
transport in rat colon. Br J Pharmacol. 1995;116:2667–72.
20. Ckless K, Schlottfeldt JL, Pasqual M, Moyna P, Henriques JA,
Wajner M. Inhibition of invitro lymphocyte transformation by the
REFERENCES
1. Skulachev VP. Mitochondrial physiology and pathology; concepts
of programmed death of organelles, cells and organisms. Mol
Aspects Med. 1999;20:139–84.
2. Yousif LF, Stewart KM, Kelley SO. Targeting mitochondria with
organelle-specific compounds: strategies and applications. Chem-
biochem. 2009;10:1939–50.
3. Liberman EA, Topaly VP, Tsofina LM, Jasaitis AA, Skulachev
VP. Mechanism of coupling of oxidative phosphorylation and the
membrane potential of mitochondria. Nature. 1969;222:1076–8.
4. Mitchell P. Coupling of phosphorylation to electron and hydrogen
transfer by a chemi-osmotic type of mechanism. Nature.
1961;191:144–8.
5. Severin SE, Skulachev VP, Iaguzhinskii LS. Possible role of
carnitine in the transport of fatty acids through the mitochondrial
membrane. Biokhimiia (rus). 1970;35:1250–3.
isoquinoline alkaloid berberine.
J Pharm Pharmacol.
1995;47:1029–31.
21. Zhou J, Zhou S, Tang J, Zhang K, Guang L, Huang Y, et al.
Protective effect of berberine on beta cells in streptozotocin- and
high-carbohydrate/high-fat diet-induced diabetic rats. Eur J
Pharmacol. 2009;606:262–8.
22. Hong Y, Hui SC, Chan TY, Hou JY. Effect of berberine on
regression of pressure-overload induced cardiac hypertrophy in
rats. Am J Chin Med. 2002;30:589–99.
6. Skulachev VP. Membrane bioenergetics. Berlin: Springer; 1988.
p. 442.

Berberine and Palmatine Derivatives of Plastoquinone as Mitochondria-Targeted Antioxidants
2895
23. Lin CC, Ng LT, Hsu FF, Shieh DE, Chiang LC. Cytotoxic effects
of Coptis chinensis and Epimedium sagittatum extracts and their major
constituents (berberine, coptisine and icariin) on hepatoma and
leukaemia cell growth. Clin Exp Pharmacol Physiol. 2004;31:65–
9.
24. Serafim TL, Oliveira PJ, Sardao VA, Perkins E, Parke D, Holy J.
Different concentrations of berberine result in distinct cellular
localization patterns and cell cycle effects in a melanoma cell line.
Cancer Chemother Pharmacol. 2008;61:1007–18.
36. Naguib YM. A fluorometric method for measurement of peroxyl
radical scavenging activities of lipophilic antioxidants. Anal
Biochem. 1998;265:290–8.
37. Zhu M, Qin ZJ, Hu D, Munishkina LA, Fink AL. Alpha-synuclein
can function as an antioxidant preventing oxidation of unsaturated
lipid in vesicles. Biochemistry. 2006;45:8135–42.
38. Palmer JW, Tandler B, Hoppel CL. Biochemical properties of
subsarcolemmal and interfibrillar mitochondria isolated from rat
cardiac muscle. J Biol Chem. 1977;252:8731–9.
25. Hwang JM, Wang CJ, Chou FP, Tseng TH, Hsieh YS, Lin WL,
et al. Inhibitory effect of berberine on tert-butyl hydroperoxide-
induced oxidative damage in rat liver. Arch Toxicol. 2002;76:664–
70.
26. Shia CS, Hou YC, Juang SH, Tsai SY, Hsieh PH, Ho LC, et
al. Metabolism and pharmacokinetics of san-huang-xie-xin-tang,
a polyphenol-rich chinese medicine formula, in rats and ex-vivo
antioxidant activity. Evid Based Complement Alternat Med.
2009
27. Shirwaikar A, Rajendran K, Punitha IS. In vitro antioxidant
studies on the benzyl tetra isoquinoline alkaloid berberine. Biol
Pharm Bull. 2006;29:1906–10.
28. Severina II, Muntyan MS, Lewis K, Skulachev VP. Transfer of
cationic antibacterial agents berberine, palmatine, and benzal-
konium through bimolecular planar phospholipid film and
staphylococcus aureus membrane. IUBMB Life. 2001;52:321–
4.
29. Pereira GC, Branco AF, Matos JA, Pereira SL, Parke D, Perkins
EL, et al. Mitochondrially targeted effects of berberine [natural
yellow 18, 5,6-dihydro-9,10-dimethoxybenzo(g)-1,3-benzodioxolo
(5,6-a) quinolizinium] on k1735-m2 mouse melanoma cells:
comparison with direct effects on isolated mitochondrial fractions.
J Pharmacol Exp Ther. 2007;323:636–49.
30. Samosorn S. Development of berberine-based derivatives as novel
antimicrobal agents. Australia: Department of chemistry: Univer-
sity of Wollongong; 2005. p. 213.
39. Jentzsch AM, Bachmann H, Furst P, Biesalski HK. Improved
analysis of malondialdehyde in human body fluids. Free Radic
Biol Med. 1996;20:251–6.
40. Antonenko YN, Roginsky VA, Pashkovskaya AA, Rokitskaya TI,
Kotova EA, Zaspa AA, et al. Protective effects of mitochondria-
targeted antioxidant SkQ in aqueous and lipid membrane
environments. J Membr Biol. 2008;222:141–9.
41. Martinez LA, Rios JL, Paya M, Alcaraz MJ. Inhibition of
nonenzymic lipid peroxidation by benzylisoquinoline alkaloids.
Free Radic Biol Med. 1992;12:287–92.
42. Skulachev VP, Antonenko YN, Cherepanov DA, Chernyak BV,
Izyumov DS, Khailova LS, et al. Prevention of cardiolipin
oxidation and fatty acid cycling as two antioxidant mechanisms
of cationic derivatives of plastoquinone (SkQs). Biochim Biophys
Acta. 2010;1797:878–89.
43. Mikes V, Yaguzhinskij LS. Interaction of fluorescent berberine
alkyl derivatives with respiratory chain of rat liver mitochondria. J
Bioenerg Biomembr. 1985;17:23–32.
44. Chernyak BV, Izyumov DS, Lyamzaev KG, Pashkovskaya AA,
Pletjushkina OY, Antonenko YN, et al. Production of reactive
oxygen species in mitochondria of HeLa cells under oxidative
stress. Biochim Biophys Acta. 2006;1757:525–34.
45. Fetisova EK, Avetisyan AV, Izyumov DS, Korotetskaya MV,
Chernyak BV, Skulachev VP. Mitochondria-targeted antioxidant
SkQR1 selectively protects MDR (Pgp 170)-negative cells against
oxidative stress. FEBS Lett. 2010;584:562–6.
31. Lombardo F, Shalaeva MY, Tupper KA, Gao F. Elogd: A tool for
lipophilicity determination in drug discovery. 2. Basic and neutral
compounds. J Med Chem. 2001;44:2490–7.
32. Martyushin AA, Tcarev DA, Grigorenko MA, Fedorov II,
Ramenskaya GV, Tashlitsky VN, et al. Estimation of lipophilicity
of new generation of antioxidants. Pharmacia. 2008;5:23–9
(Russ.).
33. Severina II. Nystatin-induced increase in photocurrent in the
system ‘bacteriorhodopsin proteoliposome/bilayer planar mem-
brane’. Biochim Biophys Acta. 1982;681:311–7.
34. Roginsky V. Chain-breaking antioxidant activity of natural
polyphenols as determined during the chain oxidation of methyl
linoleate in Triton x-100 micelles. Arch Biochem Biophys.
2003;414:261–70.
46. Izyumov DS, Domnina LV, Nepryakhina OK, Avetisyan AV,
Golyshev SA, Ivanova OY, et al. Mitochondria as source of
reactive oxygen species under oxidative stress. Study with novel
mitochondria-targeted antioxidants-the “Skulachev-ion” deriva-
tives. Biochemistry (Mosc). 2010;75:123–9.
47. Skulachev VP, Bakeeva LE, Chernyak BV, Domnina LV, Minin
AA, Pletjushkina OY, et al. Thread-grain transition of mitochon-
drial reticulum as a step of mitoptosis and apoptosis. Mol Cell
Biochem. 2004;256–257:341–58.
48. Belousov VV, Fradkov AF, Lukyanov KA, Staroverov DB,
Shakhbazov KS, Terskikh AV, et al. Genetically encoded
fluorescent indicator for intracellular hydrogen peroxide. Nat
Methods. 2006;3:281–6.
49. Zorov DB, Filburn CR, Klotz LO, Zweier JL, Sollott SJ. Reactive
oxygen species (ROS)-induced ROS release: a new phenomenon
accompanying induction of the mitochondrial permeability
transition in cardiac myocytes. J Exp Med. 2000;192:1001–14.
35. Roginsky V, Barsukova T, Loshadkin D, Pliss E. Substituted p-
hydroquinones as inhibitors of lipid peroxidation. Chem Phys
Lipids. 2003;125:49–58.