Antiproliferative and uncoupling effects of delocalized, lipophilic, cationic gallic acid derivatives on cancer cell lines. Validation in vivo in singenic mice

Article abstract of DOI:10.1021/jm500174v

Tumor cells principally exhibit increased mitochondrial transmembrane potential (Δψ_m) and altered metabolic pathways. The therapeutic targeting and delivery of anticancer drugs to the mitochondria might improve treatment efficacy. Gallic acid e

Full text of DOI:10.1021/jm500174v

Article
pubs.acs.org/jmc
Antiproliferative and Uncoupling Effects of Delocalized, Lipophilic,
Cationic Gallic Acid Derivatives on Cancer Cell Lines. Validation in
Vivo in Singenic Mice
Jose A. Jara,† Vicente Castro-Castillo,§ Jorge Saavedra-Olavarría,§ Liliana Peredo,† Mario Pavanni,†
́
Fabian Jana,† María Eugenia Letelier,‡ Eduardo Parra,∥ María Ines Becker,⊥ Antonio Morello,†
́
́
̃
Ulrike Kemmerling,# Juan Diego Maya,† and Jorge Ferreira*,†
†Clinical and Molecular Pharmacology Program, Institute of Biomedical Sciences (ICBM), Faculty of Medicine, University of Chile,
Independencia 1027, Santiago 8380453, Chile
‡Department of Pharmacological and Toxicological Chemistry, Faculty of Chemical and Pharmaceutical Sciences, University of Chile,
Olivos 1007, Santiago 8380492, Chile
§Department of Chemistry, Faculty of Basic Sciences, Metropolitan Educational Sciences University, Santiago 7760197, Chile
∥Laboratory of Experimental Biomedicine, University of Tarapaca,
́
Campus Esmeralda, Iquique 001, Chile
⊥Development of Science & Technology Foundation (FUCITED), Santiago 7750269, Chile
#Program Anatomy and Developmental Biology, Institute of Biomedical Sciences (ICBM), Faculty of Medicine, University of Chile,
Independencia 1027, Santiago 8380453, Chile
ABSTRACT: Tumor cells principally exhibit increased mito-
chondrial transmembrane potential (ΔΨm) and altered metabolic
pathways. The therapeutic targeting and delivery of anticancer
drugs to the mitochondria might improve treatment efficacy.
Gallic acid exhibits a variety of biological activities, and its ester
derivatives can induce mitochondrial dysfunction. Four alkyl
gallate triphenylphosphonium lipophilic cations were synthesized,
each differing in the size of the linker chain at the cationic moiety.
These derivatives were selectively cytotoxic toward tumor cells.
The better compound (TPP+C10) contained 10 carbon atoms
within the linker chain and exhibited an IC50 value of appro-
ximately 0.4−1.6 μM for tumor cells and a selectivity index of
approximately 17-fold for tumor compared with normal cells. Consequently, its antiproliferative effect was also assessed in vivo. The
oxygen consumption rate and NAD(P)H oxidation levels increased in the tumor cell lines (uncoupling effect), resulting in a ΔΨm
decrease and a consequent decrease in intracellular ATP levels. Moreover, TPP+C10 significantly inhibited the growth of TA3/Ha
tumors in mice. According to these results, the antineoplastic activity and safety of TPP+C10 warrant further comprehensive evaluation.
of ROS: (a) the oxidation of xanthine by xanthine oxidase,
without inhibiting uric acid formation, (b) the peroxidation of
linoleic acid catalyzed by lipoxygenase-1, (c) the synthesis of
prostanoids and hydroxyeicosatetranoic acids catalyzed by COX-
2 and 12-LOX, respectively, and (d) the catalysis of superoxide
dismutation into oxygen and hydrogen peroxide by superoxide
dismutase.9,10 In addition, gallates exhibit myeloperoxidase
inhibitory activity and hypochlorous acid-scavenging proper-
ties.11 Moreover, gallates can chelate transition-metal ions, such
as Fe2+ and Cu2+, which are powerful promoters of free radical
damage.10 Likewise, gallates can inhibit proreductant enzymes,
such as ribonucleotide reductase, which catalyzes the rate-
limiting step of de novo DNA synthesis and is significantly
upregulated in malignant compared with nonmalignant cells.9,10
INTRODUCTION
■
For over 60 years, propyl, octyl, and lauryl gallates have been
permitted for use as antioxidant additives in the food and
pharmaceutical industries because they have demonstrated
significantly low toxicities both in vitro and in vivo.1,2 They act
as antioxidants in a wide variety of ways, including quenching of
reactive oxygen (ROS) and sulfur (RSS) species3 and acting as
potent peroxynitrite (ONOO−) scavengers.4 The reactivity of
alkyl gallates toward radicals depends on the length of their
alkyl chains, and their differential reactions are attributable to
differences in alkyl gallate partitioning between the lipidic and
aqueous pseudophases.5−7 The intramicellar rate constants of
reactions involving gallates with significantly different hydro-
phobicities and radicals incorporated into micelles are
independent of the gallate alkyl chain length.8 The activity of
alkyl gallates also includes the inhibition of several of the fol-
lowing pro-oxidant enzymes that are involved in the production
Received: November 8, 2013
Published: February 25, 2014
© 2014 American Chemical Society
2440
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Gallic acid (GA) and its ester derivatives affect the activity of a
variety of both DNA and RNA animal viruses.12−14 They elicit
antimicrobial effects on Gram-positive bacteria, specifically by
inhibiting the plasma membrane respiratory chain and by
sensitizing antibiotic-resistant mutants to antibiotics.15−17
Furthermore, in numerous fungal and yeast species,3,18,19 gallates
inhibit the mitochondrial oxidative phosphorylation system,
thereby augmenting the effectiveness of fungicides.20 Gallates
not only play an important role in the prevention of
inflammatory processes and malignant transformation10 but
also selectively inhibit proliferation, metastasis, and antiapoptotic
and angiogenic processes. These studies have revealed consider-
able tumor-specific cytotoxic effects in a wide variety of human
and murine tumor cell lines, such as leukemia, lung, stomach,
colon, melanoma, breast, pheochromocytoma, and fibrosarcoma
cell lines.4−6,8,21−23 In contrast, gallates prevent amyloid β (Aβ)-
induced apoptotic neuronal death.11
of any tissue.28,34 Therefore, the differences between normal
and tumor cells suggest that mitochondria might represent a
therapeutic target. This difference could represent a new way to
circumvent the resistance to apoptotic stimuli that is exhibited by
cancer cells. Considering the increased ΔΨm in tumor cells, small
molecules have been proposed to selectively accumulate within
the mitochondria of these cells.35 To improve the cytotoxic
effects of GA esters, we synthesized different lipophilic
delocalized cations gathering a GA ester with the triphenylphos-
phonium (TPP) moiety. The resulting compound might
selectively accumulate within tumor mitochondria to elicit its
cytotoxic effect, exploiting the fact that the IMM potential differs
between cancer and noncancer cells.27,28 These new GA cations
would bind directly to the phospholipid bilayers due to their large
hydrophobic surface area, which reduces the activation energy
required for the uptake of the TPP moiety.36,37 Because of their
positive charge, these cations are driven by the potential of the
cytoplasm and the mitochondrial membrane to accumulate
selectively within the IMM by 500-fold.37 This process would
inhibit vital cellular systems by diminishing ATP synthesis and
inhibiting the glycolytic pathway, which is highly dependent on
the ATP produced by oxidative phosphorylation.38 To evaluate
the above statements in this study, we assessed the cytotoxicity of
the GA derivative compounds and the mechanisms through
which these new molecules diminished ATP levels in tumor
cells. Here, we describe the synthesis of a new cytotoxic
compound that targets mitochondria and exhibits an uncoupling
action on ΔΨm. We further show that decreasing ATP levels
is one of its principal mechanisms of inducing cell death. In
addition, the most cytotoxic compound, triphenyl(10-((3,4,5-
trihydroxybenzoyl)oxy)decyl)phosphonium (TPP+C10), also
exhibited tumor growth inhibitory activity in vivo. Because the
mitochondria appear to represent a selective anticancer target,
we determined the effect of the concentration on (i) the growth
and vitality of tumor and normal cells in culture, (ii) the oxidative
phosphorylation process in mitochondria, (iii) ΔΨm genera-
tion in isolated mitochondria, (iv) the intracellular level of the
NAD(P)H/NAD(P+) ratio, (v) the opening of the mitochon-
drial permeability transition pore in isolated mitochondria, (vi)
intracellular levels of ATP, and (vii) cellular death. Thus, the new
molecules derived from GA may represent a new therapeutic
anticancer modality cancer; their selectivity for mitochondria
would minimize or avoid the adverse effects of current anticancer
drugs.
Our main interest in GA derivatives is related to their selective
antitumoral activities. n-Alkyl gallates induce apoptosis in tumor
cell lines predominantly by activating the mitochondrion-
dependent apoptotic pathway due to their association with
altered mitochondrial oxidative metabolism and increased
intracellular Ca2+ levels. n-Alkyl gallates can elicit the following
effects: (a) decrease mitochondrial membrane potential (ΔΨm);
(b) induce permeability transition pore opening; (c) promote the
release of cytochrome c, apoptosis-inducing factor (AIF), some
procaspases that will be further activated, and endonuclease G
(Endo G); (d) upregulate the expression of Bcl-2-associated X
protein (BAX) and caspase-4, caspase-9, and caspase-3 proteins,
subsequently causing DNA damage and fragmentation.21,24
However, the mechanisms that trigger the apoptosis induced by
GA and its n-alkyl ester derivatives in tumoral cells remain poorly
understood. Increased concentrations of the alkyl esters of GA
were first reported to elicit an uncoupling effect on the oxidative
phosphorylation system, followed by an inhibition of electron
flow through the respiratory chain, predominantly at the
NADH−CoQ step. Both effects prevent ATP synthesis, which
results in cell death. The structure and lipophilicity of the alkyl
side chain of gallates are relevant for their antitumoral activities.22
Moreover, mitochondria play a vital role in regulating energy
metabolism, cytosolic calcium concentration, ROS production,
and apoptosis. They exhibit significant differences in terms of the
oxidative phosphorylation process between tumor and non-
tumor cells; the activities of certain important enzymes involved
in this process are decreased in tumor cells.25,26 Carcinoma cells
exhibit high mitochondrial membrane potential (ΔΨm) and
low respiration rates due to mitochondrial dysfunction.27−29
Tumor cells also exhibit the increased expression of several
proteins involved in glucose metabolism.30,31 In some cancer
cells, aerobic glycolysis may contribute to approximately 50% of
total synthesized ATP, becoming a significant source of power
generation in these cells.32 However, several studies have shown
the importance of ATP production by oxidative phosphoryla-
tion in tumor cells. This process allows the direct delivery of
newly synthesized ATP to the hexokinase II (HK-II) due to its
interaction with the outer mitochondrial membrane proximal
to the voltage-dependent anionic channel (VDAC). HK-II uses
this ATP to phosphorylate and convert glucose to glucose
6-phosphate, supporting increased rates of glycolysis.33 Fur-
thermore, the inner mitochondrial membrane (IMM) of tumor
cells exhibits a ΔΨm of approximately 150−180 mV; it is
negatively charged on its inner face. This potential is greater than
that of any other cellular organelle and that of nontumor cells
RESULTS
■
Synthesis of Delocalized Lipophilic Cations of GA
Derivatives. The synthetic route (see the Experimental Section)
allowed us to derive a series of lipophilic cations that differed in
the number of carbon atoms in the alkyl chains between the GA
and triphenylphosphonium moieties (Figure 1A,B).
Cytotoxicity of Delocalized Lipophilic Cations of GA
Derivatives. The most potent cytotoxic activity on both mouse
mammary adenocarcinoma (both sensitive and resistant to
chemotherapy) and human leukemia cells was exhibited by
TPP+C10 (Figure 1C−E, Table 1). The IC50 values obtained
from the concentration−response curves (as Figure 1D,E)
showed that GA and triphenylphosphine (Ph3P) were the least
cytotoxic. As expected, the GA conversion in ester derivatives
conferred to the resulting compound greater cytotoxicity, which
was dependent on the length of the hydrocarbon side chain. As
well, as we supposed, the new delocalized lipophilic cationic
derivative by conjugation of the respective ester of GA with
2441
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Figure 1. Delocalized lipophilic cations of gallic acid derivatives and their cytotoxic effects. (A, B) Synthesis schedule of lipophilic cations of gallic acid
derivatives. (C) TA3 antiproliferative effect. (D) TA3/Ha sigmoidal dose−response curve (variable slope) showing the cytotoxic effect at 48 h of
TPP+C10 exposure. (E) TA3-MTX-R sigmoidal dose−response curve showing the cytotoxic effect at 48 h of TPP+C10 exposure. The results were
obtained using the neutral red assay. The values represent the mean values of at least three independent experiments SEM. Each assay was performed
in triplicate.
Ph3P showed higher cytotoxic activity. Among the group of alkyl
gallate triphenylphosphonium lipophilic cations, the derivative
with the fewest carbon atoms (TPP+C8) was the least cytotoxic
(Table 1). Strikingly, the derivative with 11 carbon atoms
(TPP+C11) was less cytotoxic than TPP+C10 and TPP+C12,
suggesting that an odd number of carbons can hinder proper
interaction with the IMM. However, further experiments are
necessary to confirm this hypothesis (Table 1).
Comparisons of the IC50 values between tumor and nontumor
cells from the mouse mammary epithelium revealed that
TPP+C10 was the most selective compound of the series, with
a selectivity of approximately 17-fold. Therefore, the TPP+C10
derivative is more cytotoxic to tumor cells than nontumoral cells.
These results suggest that the mechanism of cytotoxicity might
be related either to the perturbation of the negative membrane
charge exerted by the cations or to the degree to which the com-
pounds intercalate into the IMM. Indeed, the latter mechanism
would explain the differences observed between the compounds
in this series. Furthermore, although the TPP+C12 compound
proved to be highly cytotoxic to tumor cells, it was also toxic to
nontumoral cells and is therefore not a suitable anticancer drug
candidate. The TPP+C10 compound, however, exhibited tumor-
cell-selective cytotoxicity.
Uncoupling Effect of GA Derivatives. Figure 2A shows
representative traces of oxygen consumption by isolated
TA3/Ha mitochondria recorded using a Clark oxygen-sensitive
2442
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
least potent (K0.5 ≈ 2 × 10−4 M). This figure also shows
the conventional behavior exhibited by TPP+C12; doubling the
concentration of the cations approximately doubled their
respective uncoupling effects. Thus, a plot of the log change in
respiration rate versus log cation concentration was delineated by
a straight line with a slope of approximately 1. We further plotted
the observed respiration rate at each cation concentration after
the addition of an excess of FCCP (full uncoupling). These
results showed that the cations elicited no confounding effect
on the maximum respiratory capacity. The dynamic range of
the concentrations from the lowest measured uncoupling to the
maximum uncoupling was approximately 10 (n = 1.0). Similar
results were previously reported by Lou et al.39 for the classic
uncouplers CCCP and DNP (2,4-dinitrophenol). Interestingly,
TPP+C10 can be considered a relatively safer uncoupler because
its uncoupling effect increased gradually as its concentration
increased (the slope of the log−log plot was approximately 0.42−
0.54), thereby widening the distance between therapeutic and
toxic doses and providing a therapeutic range of approximately
500-fold. Moreover, uncoupling by TPP+C10 at concentrations
below 10−5 M was only completely inhibited by atractyloside,
and the slopes of the log−log plots were steeper (n ≈ 1.0),
suggesting that uncoupling at these concentrations required
adenine nucleotide translocase activity.39
Intracellular ATP Content. The uncoupling effect induced
by the GA derivatives is expected to result in decreased intra-
cellular ATP content. We therefore evaluated the effect of acute
exposure to these compounds. The results are shown in Figure 3.
For all compounds, the ATP levels decreased significantly in
the TA3/Ha cell line (Figure 3); TPP+C10 (Figure 3B) and
TPP+C12 (Figure 3D) were the more potent compounds, and
TPP+C8 (Figure 3A) and TPP+C11 (Figure 3C) were the weaker
compounds. Moreover, decreased ATP levels occurred in a
concentration-dependent manner. This decrease was not due
to cell death; at the concentration range tested, the percentage
of cell death was relatively small compared with the decrease in
ATP content (Figure 3E).
Metabolic State of TA3/Ha Tumor Cells. To assess the
metabolic state of mitochondria in tumor cells, we studied
the oxidation of reduced nicotinamide adenine nucleotide
(NAD(P)H). The oxidative transition state was evaluated by
measuring the metabolic NAD(P)H/NAD(P)+ ratio. Rotenone
is a known inhibitor of NADH oxidation; it increases the relative
NAD(P)H/NAD(P)+ ratio (reduced state). Accordingly, the
addition of 12 μM rotenone increased NAD(P)H autofluor-
escence (Figure 4A). The transition is consistent with increased
NAD(P)H levels as a result of the inhibitory effect of rotenone
on electron flow. To decrease the NAD(P)H/NAD(P)+ ratio
(i.e., to favor NAD(P)H oxidation), an uncoupler might be
added to the mitochondrial preparation. FCCP is a classical
electron transport chain uncoupler that increases electron flow
and favors the oxidation of NAD(P)H, thereby decreasing the
NAD(P)H/NAD(P)+ ratio. Relative to untreated mitochondria,
the addition of 6 μM FCCP decreased the autofluorescence
of NAD(P)H, thus decreasing the NAD(P)H levels, which is
consistent with its known uncoupling properties (Figure 4A).
The addition of the most cytotoxic cationic derivative (TPP+C10)
to the mitochondrial preparation produced an effect similar to
that observed with FCCP (i.e., decreased NAD(P)H autofluor-
escence) and was dependent on the concentration of TPP+C10.
Therefore, these results suggest that TPP+C10 promotes the
oxidation of NAD(P)H and the reduction in the NAD(P)H/
NAD(P)+ ratio.
Table 1. Cytotoxicity of GA Ester Derivatives in the Tumor
Cell Lines TA3/Ha, TA3-MTX-R, and CCRF-CEM and the
Nontumoral Cell Line MM3MG
a
IC50 (μM)
compd
GA
TA3/Ha
TA3-MTX-R
CCRF-CEM
MM3MG
163.1 2.3
35.9 0.6
7.06 0.1
13.4 0.01
0.40 0.01
6.4 1.0
160.2 1.3
29.8 0.3
3.00 0.03
14.4 0.01
0.38 0.02
3.7 0.8
1.2 0.3
>200
>200
>200
>200
OG
58,5 1,4
15,46 0,1
34.6 0.02
1.65 0.2
19.75 0.01
0.58 0.21
LG
30
7
TPP+C8
TPP+C10
TPP+C11
TPP+C12
Ph3P
29.4 8.7
7.1 3.2
14.3 2.5
1.38 0.4
>200
0.74 0.09
190 2.5
184
5
a
IC50 = the concentration resulting in cytotoxicity in 50% of cultured
cells after 48 h of treatment. Values were calculated from the
respective sigmoidal dose−response curves, representing the mean
SEM values of at least three independent experiments. Each assay was
performed in triplicate.
electrode. During glutamate plus malate oxidation in the absence
of ADP, the mitochondrial respiration rate was low, but its rate
was abruptly increased after the addition of ADP (state 3). This
stimulation was transient, and the respiration rate spontaneously
declined and remained low again (state 4). We found that the
amount of oxygen consumed during this accelerated respiration
phase was stoichiometrically correlated with the amount of ADP
added (ADP/O = 2.3 in this example) and depended on the
nature of the oxidized respiratory substrate, indicating that
mitochondrial respiration is coupled to ADP phosphorylation.
The added ADP was phosphorylated in parallel with stimulated
respiration, and the synthesized ATP accumulated in the mito-
chondrial suspension. These transitions in respiration rate were
repeated several times until the mitochondrial suspension
became anaerobic. Afterward, the ATP synthase inhibitor
oligomycin was added to the same mitochondrial suspension
in state 4, and we did not observe the abruptly increased oxygen
consumption rate after the addition of ADP (state 3 inhibition).
Subsequently, the addition of TPP+C8 induced an abrupt and
permanent increase in respiration until the oxygen concentration
was exhausted (trace a), and this increase was inhibited by
cyanide (trace b). Consequently, TPP+C8 exhibited features of
an uncoupler because it stimulated oxygen consumption with a
consequent decrease in the ΔμH+ caused by energy dissipation
and without a concomitant increase in ATP synthesis.
Figure 2B shows the result of titrating respiration with
TPP+C8. Similar to conventional uncouplers, such as carbonyl
cyanide m-chlorophenylhydrazone (CCCP) or carbonyl cyanide
p-(trifluoromethoxy)phenylhydrazone (FCCP),39 respiration
was markedly stimulated by 10−4 M TPP+C8 and increased
proportionally as the concentration was increased up to 6-fold.
GA esters represent conventional but weak uncouplers.22 They
are lipid-soluble molecules, but the pK values of their phenol
groups are relatively high (pK = 8.1−12.9). Consequently, the
uncoupling activity is limited by low concentrations of the
anionic forms.
We observed increased oxygen consumption rates in mouse
mammary adenocarcinoma cells, which indicate that these
cations essentially behave as uncoupling agents (Figure 2C).
All the cations exhibited an uncoupling effect on the TA3/Ha
and TA3-MTX-R cell lines which was comparable to the effect
exerted by the classical uncouplers we used as a control. The
derivative with the fewest carbon atoms (TPP+C8) was the
2443
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Figure 2. (A) Representative traces of the effect of ADP, oligomycin, CN−, and TPP+C8 on the O2 consumption in mitochondria from the TA3/Ha cell
line in the presence of glutamate plus malate as substrates. (B) Representative traces of the effects of TPP+C8 on the time course of O2 consumption by
the TA3/Ha mitochondria (oxidation of glutamate and malate). (C) Uncoupling effect on intact cancer cells by the cationic GA derivatives: double
logarithmic plot of the uncoupling effects of the different GA derivatives on respiration in the TA3 and TA3-MTX-R cell lines. Closed circles represent
the titration of respiration rates induced by each GA derivative. Closed inverted triangles represent the titration of the respiration rates with each GA
derivative in the presence of 5 nmol of ATC/mg of protein (added before succinate). Open circles represent the respiration rate at each concentration of
each GA derivative after the addition of FCCP to attain the maximal cellular respiratory capacity. The values shown are the mean values SEM of at least
three independent experiments. The lines were fitted by linear regression obtained from GraphPad Prism (4.0) plot software. Then the slope of the log−
log plot and K0.5 (M) values were calculated from the linear regression equation. This latter value corresponds to 50% of the maximal cellular respiratory
capacity.
ΔΨm Decrease Produced by TPP+C10. Decreased ΔΨm is
widely used as an early measure of cell death. This parameter was
evaluated using the lipophilic cations safranin and JC-1.
These results showed that, upon the addition of TPP+C10 to
mitochondria isolated from TA3/Ha cells, the ΔΨm decreased
sharply in a manner comparable to that exerted by CCCP
2444
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Figure 3. Lipophilic cation effect on intracellular ATP content. (A−D) A total of 2 × 105 cells/mL were incubated for 2 h with each cation, and the ATP
content was measured as described in the Experimental Section. (A) Effects of TPP+C8 on ATP levels. (B) Effects of TPP+C10 on ATP levels. (C) Effects
of TPP+C11 on ATP levels. (D) Effects of TPP+C12 on ATP levels. (E) Effects of increasing concentrations of lipophilic cations on cellular viability. The
control test of cellular membrane integrity was measured by propidium iodide (PI) incorporation using a flow cytometer. A total of 2 × 105 cells/mL
were seeded and incubated for 24 h at 37 °C and 5% CO2. Next the culture medium was changed, the cationic GA derivatives were added, and the cells
were incubated for 2 and 4 h. PI (1 mg/mL) was then added. The values shown are the mean values SEM of at least three independent experiments.
The error bars correspond to the confidence interval of 95%. An asterisk indicates P < 0.05 compared to the control group (ANOVA) followed by the
Bonferroni post-test. Key: ***, P < 0.001.
(Figure 4C,D). The same effect was observed when testing
TPP+C10 on TA3/Ha cells treated with JC-1 (Figure 4E). Unlike
the expenditure of ΔΨm when ATP is synthesized, the level of
ΔΨm apparently remained constant (Figure 4B) due to com-
pensation by an increased respiration rate (Figure 2A). This effect
occurred early in tumor cells exposed to TPP+C10, suggesting that it
plays a pivotal role in cell death triggered by this GA derivative.
Regulation of the Mitochondrial Permeability Tran-
sition Pore (mPTP). Figure 5A shows an increase in
mitochondrial swelling proportional to the concentration of
TPP+C10 on both TA3/Ha cells and their multi-drug-resistant
variant (TA3-MTX-.R). However, this phenomenon was not
completely reversed by the classical inhibitor of this process,
cyclosporine A (CsA) (Figure 5B). Furthermore, the mitochon-
drial swelling was partially reversed by protective molecules from
the oxidation of sulfhydryl groups present in adenine nucleotide
translocase (ANT), such as NEM (Figure 5C), DTT (Figure 5D),
and NAC (Figure 5E). These data suggest that the GA derivatives
previously oxidized in the respiratory chain might interact with the
sulfhydryl groups present in ANT, thereby regulating the opening
of the mPTP.
GA Derivatives Induce Cell Death. Early and late apoptosis
and necrosis were expressed as the percentage of annexin V+/PI−,
annexin V+/PI+, and annexin V−/PI+ (PI = propidium iodide)
cells, respectively.40 In Figure 6A, we evaluated the activation
of caspase 3 at concentrations near the IC50 for TPP+C10. We
observed that this compound activated caspase 3, whereas the
other derivatives, which are less potent, did not and required
higher concentrations to elicit the same effect. Furthermore,
TPP+C10 induced a cell cycle arrest in the G0/G1 phase, which is
a characteristic effect of uncouplers (Figure 6B). Parts C−F of
Figure 6 show that cells treated with GA derivatives are primarily
in the early stages of apoptosis at 24 and 48 h, and the most potent
compound was TPP+C10. We also found that the percentage of
2445
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
PI-permeable cells was negligible, indicating that necrosis is
irrelevant as the type of cell death induced by these lipophilic
cations. These phenomena are consistent with the low lactate
dehydrogenase activity observed in Figure 6G.
TPP+C10 Induces Antitumor Effects in a Syngenic
Model of TA3/Ha-Derived Tumors. We examined the
inhibitory effect of TPP+C10 on tumor growth in vivo using a
well-established syngenic tumor mouse model.41,42 This model
consists of the inoculation of tumor cells of the same species as
the recipient animal and has the advantage that the tumor grows
in an immunocompetent host.
A dose of 10 mg of TPP+C10/kg or vehicle intraperitoneally
(ip) every 48 h for 30 days resulted in a survival rate of mice
without significant differences compared with the control group.
This dose significantly inhibited tumor growth in CAF1-Jax mice
compared with the control group during the treatment period
(Figures 7), without apparent signs of toxicity (data not shown).
DISCUSSION
■
Currently, extensive effort is focused on obtaining, isolating, and
synthesizing new compounds that inhibit pathways involved in
energy metabolism in tumor cells.43 The emphasis on tumor cell
metabolism is primarily due to specific metabolic changes that
occur that allow tumor cells to evade the immune response and
chemotherapeutic treatments.44 Here, we chemically modified
GA, a well-known herbal antioxidant with various biological
activities. The chemical modifications were designed to produce
mitochondrion-targeted molecules that were selectively toxic
to tumor cells.45,46 The proposed synthetic route allowed us to
obtain molecules that were linked to triphenylphosphonium,
a chemical group that drives mitochondrial uptake by exploiting
the charge differences of the inner and outer faces of the IMM.37
The negative charge of the inner face attracts cations toward
the matrix, and despite its charge, cations can cross the cell
membrane easily because of their lipophilicity.36 We designed
these new molecules to identify new cancer-selective drugs that
can inhibit ATP synthesis and the subsequent phosphorylation of
glucose by HK-II in tumor cells. The GA ester derivatives that we
synthesized have different chain lengths between the polyhy-
droxybenzene and TPP moieties. The design of these molecules
with different chain lengths meets the need of evaluating a
suitable chain length to allow a better intercalation of the cationic
compounds in the inner mitochondrial membrane, since various
authors have assayed a set of GA esters with different chain
lengths having antibacterial, antifungal, or antiviral action. There
are also data showing that in general the esters with less than
8 carbon atoms are less potent and those with a total of close to
15 carbons are highly toxic and nonspecific.7,18 Thus, differences
in compound efficacy depended on the length of the alkyl side
chain. Perhaps more important than compound size is its shape,
because it gives the possibility to have some degree of affinity
with proteins of some inner membrane transport system. TPP+C10
was the most cytotoxic and selective. It may be considered a
relatively safer uncoupler because its uncoupling effect increased
gradually as its concentration increased (Figure 2C), thereby
widening the distance between therapeutic and toxic doses and
providing a therapeutic range of approximately 500-fold. More-
over, only the uncoupling by TPP+C10 at concentrations below
10−5 M was completely inhibited by atractyloside, suggesting that
uncoupling at these concentrations required adenine nucleotide
translocase activity.39 TPP+C8 was the least cytotoxic. Despite
exhibiting toxicity similar to that of TPP+C10, TPP+C12 was
significantly less selective, exhibiting similar IC50 values for tumor
Figure 4. (A) Effects of TPP+C10 on NAD(P)H oxidation on the TA3/
Ha cell line. Time course plots of NAD(P)H oxidation induced by
TPP+C10 (2.5 mg of protein/mL) in the TA3/Ha cell line. We used 12
μM rotenone and 6 μM FCCP. (B) Effect of ADP on the
transmembrane potential (ΔΨm) of the mitochondria of TA3/Ha
cells. (C) Effects of TPP+C10 on the ΔΨm of the TA3-MTX-R cells. (D)
Effects of CCCP on the ΔΨm of the TA3/Ha cells. The experiments
were performed using 1 mg/mL mitochondrial protein, 20 μM safranin,
250 μM ADP, 5 mM malate, and either 5 mM glutamate or 0.2 nmol of
CCCP. The measurements were collected for 16.7 min. The values
represent the average of at least three independent experiments SD.
(E) Effects of TPP+C10 on the ΔΨm of TA3/Ha cells measured using the
JC-1 fluorescent probe in the presence of 3 μM FCCP and 0.5 μM
TPP+C10. The values shown are the mean values SEM of at least three
independent experiments. The error bars correspond to the confidence
interval of 95%. An asterisk indicates P < 0.05 compared to the control
group (ANOVA) and followed by the corresponding post-test
(Bonferroni). Three asterisks indicate P < 0.001.
2446
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Figure 5. Effects of lipophilic cations on mitochondrial swelling. (A) Effects of increasing concentrations of TPP+C10 on the mitochondria of TA3/Ha
(black bars) and TA3-MTX-R (solid bars). (B) Effects of increasing concentrations of cyclosporine on the mitochondria of TA3/Ha cells in the
presence of 0.15 mM TPP+C10. (C) N-Ethylmaleimide (NEM), (D) dithiothreitol (DTT), and (E) N-acetylcysteine (NAC) prevent the effect of 0.15
mM TPP+C10 on the mitochondria of TA3/Ha cells. The values shown are the mean values SEM of at least three independent experiments. The error
bars correspond to the confidence interval of 95%. An asterisk indicates P < 0.05 compared to the control group (ANOVA) and followed by the
corresponding post-test (Bonferroni). Key: *, P < 0.05; **, P < 0.005; ***, P < 0.001.
and nontumor cells. Therefore, the low selectivity of TPP+C12 is
likely to result in adverse effects. Thus, further experiments are
necessary for evaluating the toxicity of TPP+C12 inhuman cells and
in animals. Indeed, tumor cell specificity remains one of the most
critical issues in the search for new anticancer drugs.
All the compounds tested increased oxygen consumption in
the presence of oligomycin, indicating that, at the concentrations
tested, all the derivatives are uncouplers. This effect is different
from that observed for GA esters that lack the triphenylphos-
phonium moiety because these molecules act as inhibitors of
electron flow through the respiratory chain in tumor cells.
Consistent with the mechanism of action of these compounds
on mitochondrial metabolism, rotenone and FCCP induced
changes in the autofluorescence of NAD(P)H. In mitochondria
of TA3/Ha cells, the most potent compound (TPP+C10) induced
a concentration-dependent decrease in the NAD(P)H/
NAD(P)+ ratio (an effect similar to that produced by FCCP),
indicating that the electron transport chain was actively using
NADH as an electron donor.47 Furthermore, the availability of
NADPH is relevant in the process of glutathione (GSH)
recovery via glutathione reductase and thioredoxin regeneration
through the action of thioredoxin reductase, among other critical
processes for the tumor cell.48
The uncoupling effect elicited by these new molecules also
decreased the ATP synthesis that is coupled to the oxidative
phosphorylation system. This decrease might occur, for example,
via a decrease in glucose phosphorylation catalyzed by HK-II,
leading to decreased glycolytic flow. HK-II is a large isoenzyme
that is overexpressed in many tumor cells and is crucial for the
glycolytic pathway.43,49 The uncoupling effect can be achieved
through a variety of mechanisms. For example, uncoupling
can occur as a result of the neutralization of a negative charge
2447
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Figure 6. Type of cell death. (A) Western blot measurement of caspase 3 activation on the TA3/Ha cells in the presence of each compound at 0.5 μM for
24 h. The assay was conducted under the conditions listed in the Experimental Section. (B) Cell cycle analysis of TA3/Ha cells after 24 h of treatment
with increasing concentrations of TPP+C10. (C, D) Representative dot plots of PI and annexin V double staining on the TA3/Ha cells in the presence of
a 0.5 μM concentration of each cation or 10 μM staurosporine (SP) as a positive control for (C) 24 h or (D) 48 h of treatment. (E) Quantifications of
(C) and (D), respectively. The values shown are the mean values SEM of at least three independent experiments. The error bars correspond to the
confidence interval of 95%. An asterisk indicates P < 0.05 compared to the control group (ANOVA) followed by the Bonferroni post-test. Key: *, P <
0.05; **, P < 0.005; ***, P < 0.001.
located on the inner face of the IMM by positively charged
cations (i.e., a potential-sensitive uncoupling effect). Another
possible mechanism is the delivery of protons into the
mitochondrial matrix and the subsequent recruitment of anionic
2448
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Figure 7. Effect of TPP+C10 on tumor growth in CAF1-Jax mice. (A) Syngenic model: Mice were administered 10 mg of TPP+C10/kg or vehicle ip every
48 h for 30 days. The tumor sizes were measured externally three times per week. The graph shows the average tumor volumes (mm3) from two
independent experiments (n = 6 mice per group). (B) Representative images of the tumor volumes. The error bars correspond to the confidence interval
of 95%. An asterisk indicates P < 0.05 compared to the control group (nonparametric Mann−Whitney test followed by the Wilcoxon post-test).
species by ANT. Next, compounds might interact near ANT,
capturing a proton from the intermembrane space and trans-
ferring it back to the mitochondrial matrix, acting as a proton
shuttle from the intermembrane space into the matrix. This
mechanism is facilitated by the interaction of the molecules with
ANT. Many studies have reported that conventional uncoupling,
such as that observed in DNP-induced apoptosis of Calu-6 lung
cancer cells, can occur as a result of the production of ROS
production, cell cycle arrest in the G1 phase, and a decrease of
reduced glutathione levels.50 The uncoupler FCCP also inhibits
the growth of Calu-6 cells through apoptosis and cell cycle arrest
by decreasing cyclin-dependent kinase (CDK) and activating
caspases, respectively. A close relationship among the uncoupling
effect, decreased reduced glutathione levels, and the ability of
these compounds to trigger cell death has also been suggested.50
Furthermore, protonated weak lipophilic acids of aromatic
compounds can augment the H+ permeability of the IMM by
facilitating H+ transport across the hydrophobic barrier. Many
other mechanisms that increase mitochondrial H+ conductance
have been described, including the following: (a) H+ leakage via
different uncoupling proteins (UCPs), which is inhibited by
GDP, (b) allosteric stimulation by AMP on H+ leakage through
ANT, which is inhibited by atractyloside and derivatives, (c) the
transmembrane cycling of protonated/deprotonated free fatty
acids, and (d) a H+ subconductance state of the mPTP. These
mechanisms might be activated by various events that induce
mitochondrial dysfunction, including ROS generation, PTP
opening, or increases in AMP concentration. However, whether
increased H+ conductance plays a deleterious or protective role
in terms of cellular viability remains poorly understood.51
In contrast, the mitochondrial swelling assay revealed that
the compounds can induce the opening of the mPTP. This effect
was not reversed by the classical inhibitor CsA. However, redox-
active molecules, which avoid thiolic group oxidation, partially
reversed the opening of the mPTP (Figure 5). As a consequence,
partially oxidized GA appeared to be able to interact with
the protein sulfhydryl groups present in ANT. This interaction
could result in a permanent stimulation of mPTP openinga
hypothesis that remains to be studied.52,53
death, leading to the release of classical mitochondrial proteins,
such as cytochrome c, apoptosis-inducing factor, endonuclease
G, Smac/DIABLO, and Omi/HtrA2, which lead to apoptotic or
necrotic cell death.54−57
Furthermore, these decreases exert a strong signal to inhibit
cellular proliferation and to activate the death machinery, as
reflected in a concentration-dependent accumulation of cells
in the G1 phase and in the activation of caspase 3 observed
approximately at the IC50 values of TPP+C10. The death induced
by these derivatives was principally apoptosis. This effect is
reflected by the exposition of phosphatidylserine (PS), the low
rate of PI incorporation, and lactate dehydrogenase (LDH)
activity. The effects listed above suggest that targeting of GA to
the mitochondria of tumor cells through the TPP moiety does
not modify the pharmacologic mode of action of GA.
Nonetheless, preliminary experiments in animal models have
yielded promising results. At a concentration of 10 mg/kg of
body weight, the treated mouse group exhibited a statistically
similar survival curve compared with the control group.
Furthermore, the final volumes of the TA3/Ha tumors were
significantly lower than those of the control group at 25 days after
the ip administration of TPP+C10 (Figure 7B). Interestingly, a
preclinical study performed with tetradecyl gallate reported a
decrease in tumor size and metastasis of melanoma in an animal
model. According to the authors, this effect was attributed to the
inhibition of GSH synthesis and to the depletion of ATP, with
no apparent signs of toxicity.58 Therefore, further experiments
are under way to evaluate the toxicity and effectiveness of these
compounds as potential antineoplastic cancer agents.59
CONCLUSION
■
Extensive studies have reported the diverse range of effects of
both GA and its esters. Most of these reports describe the increase
of oxidized glutathione levels as a primary cause of cytotoxicity
against various cell types, including tumor cells. Through this
mechanism, most of these reports have attributed the cytotoxicity
to the ability of GA to produce free oxygen radicals. The
production of ROS can exert a variety of deleterious effects at the
intracellular level. For this reason, we chemically modified the GA
derivatives to improve their activity and selectivity. Moreover, we
propose a primary mechanism of action that targets the highly
conserved ΔΨm and that consequently reduces ATP synthesis
by the oxidative phosphorylation system, which is essential for
sustaining the metabolic reprogramming exhibited by tumor cells.
The mechanism of action described for these lipophilic cations is
Furthermore, we assume that the GA ester derivatives bound
to the TPP group would exhibit improved cytotoxic activity with
low IC50 values. They exhibit an important selectivity toward
tumor cells, causing mitochondrial dysfunction, which results
in severely decreased ΔΨm and intracellular ATP levels. Both of
these effects are potent triggers of mitochondrion-mediated cell
2449
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
(DMF) (50 mL) was treated with a solution of N,N-dicyclohexylcarbo-
diimide (DCC; 272 mg, 1.3 mmol) in dry DMF (30 mL). The mixture
was cooled to 0 °C, and a solution of 1 (520 mg, 1.1 mmol) in DMF
(10 mL) and N,N-dimethylpyridin-4-amine (DMAP, in catalytic
amounts) in DMF (5.0 mL) was added. The reaction was stopped the
next day, and any resulting precipitate was removed by filtering. The
solvent was then removed, producing a brown residue that was subjected
to chromatography on silica gel (DCM, MeOH) to yield compound 5 as
consistent with most of the effects that characterize GA and esters.
We consider that the effects described by previous studies occur
downstream of the decreased synthesis of ATP. Furthermore,
there are many examples of the use of lipophilic cations to drive
molecules, mainly with antioxidant activity, to mitochondria.
So far we can only find molecules mainly with antioxidant
activity.59−62 However, in this work we have shown that this
strategy is used to selectively deliver cytotoxic molecules to
mitochondria to disrupt the mitochondrial functioning in tumor
cells both in vivo and in vitro.
1
a light yellow oil (282 mg, 41%). H NMR (400 MHz, DMSO-d6): δ
1.17−1.55 (m, 14H, CH2), 3.55 (t, J = 6.5 Hz, 2H, CH2), 6.92 (s, 2H,
ArH), 7.75−7.88 (m, 15H, ArH). HRMS: m/z 622.0301 (calcd for
C33H36BrO5P 622.1487).
Synthesis of Triphenyl(10-((3,4,5-trihydroxybenzoyl)oxy)-
decyl)phosphonium Bromide (TPP+C10Br−). Following the same
procedure as described above for obtaining TPP+C8Br−, we synthesized
TPP+C10Br− as a light yellow oil (701 mg, 36%). 1H NMR (400 MHz,
DMSO-d6): δ 1.18−1.52 (m, 18H, CH2), 3.53 (t, J = 6.3 Hz, 2H, CH2),
6.92 (s, 2H, ArH), 7.72−7.88 (m, 15H, ArH). 13C NMR (DMSO-d6): δ
20.24, 20.74, 22.08, 25.82, 28.45, 29.00, 29.22, 30.19, 32.82, 61.04,
109.14, 118.50, 119.34, 125.41, 130.52−130.65 (m), 133.86−133.96
(m), 135.22 (m), 141.15, 145.76, 169.44 (CO). HRMS: m/z 650.1109
(calcd for C35H40BrO5P 650.1797).
Synthesis of Triphenyl(11-((3,4,5-trihydroxybenzoyl)oxy)-
undecyl)phosphonium Bromide (TPP+C11Br−). Following the
same procedure as described for obtaining TPP+C8Br−, we synthesized
TPP+C11Br− as a light yellow oil (581 mg, 44%). 1H NMR (400 MHz,
DMSO-d6): δ 1.18−1.56 (m, 20H, CH2), 3.63 (t, J = 6.1 Hz, 2H, CH2),
6.99 (s, 2H, ArH), 7.70−7.89 (m, 15H, ArH). 13C NMR (DMSO-d6): δ
20.21, 20.72, 22.09, 24.74, 25.83, 28.45, 29.03, 29.27, 32.83, 33.65, 61.05,
109.21, 118.48, 119.33, 130.09, 130.52−130.65 (m), 133.85−133.95
(m), 136.80 (m), 137.09, 145.73, 170.03 (CO). HRMS: m/z 664.1632
(calcd for C36H42BrO5P 664.1953).
Synthesis of Triphenyl(12-((3,4,5-trihydroxybenzoyl)oxy)-
dodecyl)phosphonium Bromide (TPP+C12Br−). Following the
same procedure as described for obtaining TPP+C8Br−, we synthesized
TPP+C12Br− as a light yellow oil (818 mg, 39%). 1H NMR (400 MHz,
DMSO-d6): δ 1.16−1.56 (m, 22H, CH2), 3.64 (t, J = 6.6 Hz, 2H, CH2),
6.95 (s, 2H, ArH), 7.76−7.88 (m, 15H, ArH). 13C NMR (DMSO-d6): δ
19.87, 20.37, 21.67, 25.19, 25.43, 28.06, 28.62, 29.61, 29.77, 32.44, 33.89,
61.61, 108.80, 118.11, 118.96, 120.70, 130.09−130.21 (m), 133.47−
133.57 (m), 134.79 (m), 137.83, 145.42, 167.62 (CO). HRMS: m/z
678.1967 (calcd for C37H44BrO5P 678.2110).
Cell Lines and Cell Culture. The mouse mammary adenocarcinoma
TA3/Ha cell line was kindly provided by Dr. Gasic, University of
Pennsylvania, and has been used by our laboratory since 1989.63 Its multi-
resistant variant, TA3-MTX-R, was generated under the same conditions
as previously described.64 This cell line exhibits MTX resistance and
cross-resistance to CPT, DOX, 5-fluorouracil, and vinblastine.64 Both cell
lines were propagated until the day of the assay by weekly ip inoculation
of ascitic fluid into young adult male CAF1/J mice and harvested after
5−7 days as described previously.64 The mice were purchased from
the animal facility of the Faculty of Medicine of the University of Chile,
where they were housed and fed under the same conditions as previously
described.64 The University of Chile Committee on Animal Welfare and
CONICYT approved all the animal protocols used in this study, and all
precautions were taken to ensure that the animals did not suffer unduly.
The murine tumor cells and epithelial mammary gland cell line MM3MG
(ATCC, Manassas, VA; catalog no. CRL-6376) were cultured in DMEM
supplemented with 10% inactivated fetal calf serum (FCS) and 1%
antibiotics (penicillin/streptomycin) in a humidified atmosphere (37 °C
and 5% CO2). The human acute lymphoblastic leukemia cell line CCRF-
CEM (ATCC, catalog no. CCL-119) was maintained in RPMI-1640
culture medium with 10% inactivated FCS and 1% antibiotics under the
same conditions as described above.
EXPERIMENTAL SECTION
■
Materials. Atractyloside (ATC), bovine serum albumin (BSA),
CCCP, FCCP, cyclosporine A, dichlorofluorescein diacetate, digitonin,
Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum
(FBS), glutamate, glutamine, malate, neutral red, oligomycin, PI, RPMI
1640 culture medium, Triton X-100, trypan blue 0.4% solution, primary
antibodies for β-tubulin, and succinate were purchased from Sigma
Chemical Co. (St. Louis, MO). The primary antibodies for caspases 3
were purchased from Cell Signaling Technology (Boston, MA). All
other organic compounds and inorganic salts, acids, and solvents
were purchased from Merck (Darmstadt, Germany) unless otherwise
specified. All reagents and solvents used for the synthesis of the
derivatives were purchased from Sigma-Aldrich (St. Louis, MO) or
Merck and were used without further purification.
General Experimental Procedures. We recorded the 1H and 13
C
nuclear magnetic resonance (NMR) spectra using a Bruker Avance 400
spectrometer equipped with a Bruker inverse 5 mm Z gradient probe
operating at 400.13 and 100.62 MHz, respectively. All experiments were
conducted at a probe temperature of 300 K using solutions in DMSO-d6
with tetramethylsilane (TMS) as an internal standard. The chemical
1
shifts are reported as δ (ppm) downfield from TMS for H NMR.
Coupling constants (J) are presented in hertz. Electron image mass
spectroscopy was run on a Thermo Finnigan MAT 95XP instrument
with electron impact ionization at 70 eV and perfluorokerosene as a
reference. Elemental analyses were performed using a Fisons Carlo Erba
EA1108 elemental analyzer; values were within 0.5% of the calculated
values. Consequently, these compounds meet the criteria of >95%
purity, allowing them to show good in vitro and in vivo activity.
Synthesis of (8-hydroxyoctyl)triphenylphosphonium Bromide
(1). A solution of 8-bromooctan-1-ol (244 mg, 1.16 mmol) in dry
acetonitrile (100 mL) was treated with triphenylphosphine (310 mg,
1.18 mmol). The solution was refluxed with stirring for 48 h. The solvent
was removed in a vacuum, and the crude product was subjected to
chromatography on silica gel (EtOAc, MeOH) to yield compound 1 as a
colorless oil (384 mg, 70%). 1H NMR (400 MHz, DMSO-d6): δ 1.12−
1.51 (m, 14H, CH2), 3.57 (t, J = 6.8 Hz, 2H, CH2), 7.73−7.91 (m, 15H,
ArH). HRMS: m/z 470.1041 (calcd for C26H32BrOP 470.1374).
Synthesis of (10-Hydroxydecyl)triphenylphosphonium Bro-
mide (2). Following the same procedure as described for obtaining
compound 1, we synthesized compound 2 as a colorless oil (1.2 g, 63%).
1H NMR (400 MHz, DMSO-d6): δ 1.01−1.51 (m, 18H, CH2), 3.57 (t,
J = 7.1 Hz, 2H, CH2), 7.73−7.91 (m, 15H, ArH). HRMS: m/z 485.4312
(calcd for C27H33BrOP 485.4357).
Synthesis of (11-Hydroxyundecyl)triphenylphosphonium
Bromide (3). Following the same procedure as described for obtaining
compound 1, we synthesized compound 3 as a colorless oil (230 mg,
31%). 1H NMR (400 MHz, DMSO-d6): δ 1.13−1.64 (m, 20H, CH2),
3.32 (t, J = 6.6 Hz, 2H, CH2), 7.72−7.88 (m, 15H, ArH). HRMS: m/z
512.0890 (calcd for C29H38BrOP 512.1843).
Synthesis of (12-hydroxydodecyl)triphenylphosphonium
Bromide (4). Following the same procedure as described for obtaining
compound 1, we synthesized compound 4 as a colorless oil (242 mg,
53%). 1H NMR (400 MHz, DMSO-d6): δ 1.18−1.59 (m, 20H, CH2),
3.34 (t, J = 6.4 Hz, 2H, CH2), 7.50−7.90 (m, 15H, ArH). HRMS: m/z
527.5061 (calcd for C30H40BrOP 527.5152).
Mitochondrial Preparation. Mitochondrial suspensions of
approximately 40−50 mg of protein/mL were prepared from the
tumor cells as described previously64,65 with the following modifications:
the mitochondrial fractions were washed twice, centrifuged at 12000g for
10 min, and resuspended in a minimal volume of their respective medium
in the absence of BSA to eliminate hydrophobic compound adsorption.
Synthesis of Triphenyl(8-((3,4,5-trihydroxybenzoyl)oxy)-
octyl)phosphonium Bromide (TPP+C8Br−). In an atmosphere of
N2, a solution of GA (225 mg, 1.3 mmol) in dry dimethylformamide
2450
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
Protein concentrations were determined by the Lowry reaction, and the
results were standardized with serum albumin.66
absorbance prior to and after the addition of the stimulus. Each point
corresponds to the mean values
experiments.
SEM of three independent
Cellular Viability Assay Using Neutral Red. The viability of the
cell lines treated with different concentrations of each triphenylphos-
phonium lipophilic cation was analyzed using the neutral red test as
previously described.67 The test conditions were 1 × 105 cells/mL for
TA3, TA3-MTX-R, and CCRF-CEM cells and 4 × 104 cells/mL for
MM3MG cells. The tumor cells were grown in 96-well tissue plates in
a humidified atmosphere at 37 °C and 5% CO2 for 48 h in the absence
(controls) or presence of different concentrations of the lipophilic
cations. After incubation, a neutral red solution (100 μg/mL) was
dissolved in serum-free culture medium for 2 h at 37 °C and 5% CO2 and
the resulting solution washed twice with phosphate-buffered saline
(PBS) by centrifugation at 1400g for 15 min. Next, 0.1 mL of extraction
solution (ethanol/acetic acid, 50%/1%) was added to each well. The
absorbance of the incorporated neutral red was measured at 540 nm
using a microplate ELISA reader (Bio-Rad, Hercules, CA).
Oxygen Consumption Assay. The rates of cell respiration were
measured polarographically as described previously.64,68 The 0.6 mL
reaction mixture contained the following: PBS, pH 7.4, 0.013 μg of
digitonin/mg of cellular protein (for permeabilizing the plasma
membrane), 10 mM succinate as the substrate, and 2.5 mg of protein/
mL of tumor cells supplemented with 1 μg/mL oligomycin. Respiration
was initiated by adding the lipophilic cation derivatives of GA, which were
added into 3 μL of DMSO using DMSO stocks prepared at different
concentrations. At the end of the run, 0.15 μM FCCP (TA3 cell line) and
0.25 μM FCCP (TA3-MTX-R cell line) were added to fully uncouple
respiration, as previously described.39 Mitochondrial oxygen consump-
tion was also measured polarographically. The 2.0 mL reaction medium
consisted of the following components: 200 mM sucrose, 50 mM KCl,
3.0 mM potassium phosphate, 3.0 mM Hepes, pH 7.0, 2.0 mM MgCl2,
and 1.0 mg/mL mitochondrial protein.64,68 Substrate concentrations
were 2.0 mM glutamate plus 2.0 mM malate and 0.5 mM ADP. Other
additions included 2.0 μg of oligomycin and KCN (0.1 mM). The system
was equilibrated with mitochondria at 25 °C for 2 min. Next, the rates of
O2 consumption were determined. All measurements were made as
quickly as possible after the isolation of mitochondria.
Intracellular NAD(P)H Levels. NAD(P)H fluorescence (excitation
and emission, 340 and 420 nm, respectively) was measured using a
JASCO FP-6200 spectrofluorometer (JASCO Corp., Tokyo, Japan)
equipped with magnetic stirring and thermostatic controls (25 °C). The
final reaction mixture (2.0 mL) contained the following components:
PBS, pH 7.4, 2.5 mM glutamine as the substrate and 2.5 mg of protein/
mL of the TA3 tumor cells.71 Rotenone, FCCP, or TPP+C10 was added
as indicated by arrows in the figures.
Intracellular Levels of ATP. Intracellular ATP levels were
measured using the CellTiter-Glo Luminescent Assay (Promega,
Madison, WI) according to the manufacturer’s protocols. Briefly,
2 × 105 cells/mL were incubated in a 24-well plate with the different
cation GA derivatives for the indicated periods of time. Next, an aliquot
of the cell suspension was transferred to an opaque 96-well plate,
maintained at room temperature for 10 min in the dark, and then
measured for luminescence using a Thermo Scientific Varioskan Flash
spectral scanning multimode reader. Cell membrane integrity was
assessed by PI incorporation as previously described.72 This method was
used as a control assay for cell viability for the measurement of
intracellular ATP content. Briefly, 105 cells/well were placed in a 24-well
plate and incubated with different concentrations of the compounds
for the indicated times. Next, PI (10 μg/mL) was added to each well,
and the cells were analyzed by flow cytometry (FACS Canto, BD
Biosciences, San Jose, CA).
Western Blotting. After the treatments, the cells were washed in
PBS and then lysed with RIPA (radioimmunoprecipitation assay) buffer
(Tris−Cl [50 mM], NaCl [150 mM], sodium dodecyl sulfate [SDS;
0.1%]) containing proteinase inhibitor cocktail and phosphatate
inhibitor cocktail (Roche, Indianapolis, IN). The lysate was centrifuged
at 3000g at 4 °C for 10 min. The supernatants were collected, and the
protein concentrations were determined using a BCA Protein Assay Kit
(Pierce, Rockford, IL). A 50 μg sample of proteins was separated by
12% SDS−polyacrylamide gel electrophoresis (SDS−PAGE), and the
resolved proteins were transferred to a methylcellulose membrane
(Millipore, Billerica, MA). The membrane was blocked with 5% nonfat
milk in Tris-buffered saline containing 0.1% Tween-20 (TBST) at room
temperature for 1 h and then incubated with the primary antibodies
at 4 °C overnight. After being washed with TBST, the membrane was
incubated with antimouse or antirabbit horseradish peroxidase-
conjugated secondary antibodies. Antibodies against caspases 3
(dilution 1:1000) or tubulin (dilution 1:10000) were incubated
overnight at 4 °C. The secondary antibodies (dilution 1:5000) were
incubated for 2 h at room temperature. After washing, the membranes
were exposed to enhanced chemoluminescent (ECL) reagent. BioMax
film was used for Western blotting. The blots were quantified by laser
scanning densitometry.
Measurement of Mitochondrial Transmembrane Potential
Changes. Changes in transmembrane potential (ΔΨm) were deter-
mined using the safranin method as previously described.69 Changes of
ΔA511−533nm were assessed by spectrophotometry of the mitochondria.
The reaction medium (2.5 mL) contained the following final con-
centrations: 200 mM sucrose, 50 mM KCl, 3 mM Hepes (pH 7.4), 0.5
mM EGTA, 3 mM potassium phosphate, 2 mM MgCl2, and 1.0 mg/mL
mitochondrial protein. Mitochondria from the tumor cells were
incubated with 20 μM safranin, and readings were collected for 5 min.
Next, 2 mM glutamate and 2 mM malate were added, and new readings
were collected for 5 min. Finally, the cations were added, and the final
readings were obtained.
Apoptosis by Flow Cytometry. The apoptotic events induced by
the cationic GA derivatives were determined by annexin V staining and
PI according to the manufacturer’s protocol for the Annexin V-FITC
Apoptosis Detection Kit (Abcam). For these analyses, we used 1 × 105
cells/mL, which were incubated at 37 °C and 5% CO2 with the various
lipophilic cations for 24 and 48 h. The cells were resuspended in 500 μL
of 1× annexin V-binding buffer, and 5 μL each of annexin V and PI were
added to each sample (per the manufacturer’s protocol). Next, we
incubated the cells for 5 min at room temperature and then subjected
them to flow cytometric analysis (FACS Canto, BD Biosciences) at
the following wavelengths: annexin V−FITC, ex/em = 488/530 nm; PI,
ex/em = 488/575 nm. For each sample, 5000 events were recorded. The
results were analyzed using the Cyflogic program (noncommercial
version, CyFlo Ltd.).73
LDH Activity. Lactate dehydrogenase release into the extracellular
medium was evaluated using an LDH-Cytotoxicity Kit II (Abcam)
according to the manufacturer’s protocol. We used 2 × 105 cells/mL per
assay. The cells were incubated in the presence or absence of the GA
derivatives in an environment with 90% humidity, 5% CO2, and 37 °C
for the indicated times. Next, the plates were gently shaken to
homogenize the LDH content. The cells were centrifuged at 600g for
Determination of Mitochondrial Transmembrane Potential
in Whole Cells Using JC-1. A total of 1 × 106 TA3/Ha cells/mL were
seeded into six-well plates. They were exposed to 10 μM TPP+C10 or
10 μM CCCP (as a positive control). The cells were incubated for 1 h
and then centrifuged at 1500g for 5 min. The resulting precipitates were
resuspended in 1 mL of PBS and then centrifuged at 1500g for 5 min.
The resulting precipitates were resuspended in 1 mL of PBS at 37 °C,
and we added JC-1 (1 mg/mL in DMSO) and incubated the mixtures
for 15 min at 37 °C in the dark. Next the samples were centrifuged at
1500g for 5 min, and the supernatants, which contained unincorporated
JC-1, were discarded. The precipitates were suspended in PBS (3 mL)
and measured for fluorescence using a JASCO FP-6200 spectrofluor-
ometer. For the JC-1 monomers, the wavelengths used for excitation
and emission were 485 and 540 nm, respectively, whereas those for the
J-aggregates were 485 and 590 nm.
Mitochondrial Swelling Assay. To assess mitochondrial swelling,
changes in side scatter of 540 1 nm of a 0.5 mg/mL suspension of
mitochondria in experimental buffer (125 mM KCl, 10 mM Tris−
HCl [pH 7.4], 1 mM Pi, 5 mM glutamate, 2.5 mM malate, and 10 μM
EGTA-Tris [pH 7.4]) were monitored using a JASCO V-560 UV/vis
spectrophotometer.70 The data are expressed as the difference in the
2451
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
3 min, and then 10 μL aliquots were transferred into each well of cell-free
medium in a 96-well plate. Next, we added 100 μL of LDH reaction mix
into each well, and the plate was incubated for 30 min at room
temperature. The absorbance measurements were performed at 450 nm
using a plate reader (BioRad, Hercules, CA).
apoptosis in human breast cancer cells by lauryl gallate. Carcinogenesis
2006, 27, 1699−1712.
(3) Fujita, K.; Kubo, I. Plasma membrane injury induced by nonyl
gallate in Saccharomyces cerevisiae. J. Appl. Microbiol. 2002, 92, 1035−
1042.
Evaluation of the Antitumor Activity of the Alkyltriphenyl-
phosphonium Cation in Vivo. Adult male CAF1-Jax mice were used.
To determine tumor size, an external model was used. Briefly, 1.0 × 106
TA3/Ha cells/0.1 mL of 0.9% NaCl were injected subcutaneously into
the backs of the mice. A 10 mg sample of TPP+C10/kg of body weight
was administered ip to groups of six mice every 48 h for 30 days. During
this period, we measured the minimum and maximum diameters (mm)
of the tumors three times per week. The tumor volumes were calculated
using the following formula: V = [(maximum diameter)(minimum
diameter)2]/2.74 These volumes were compared with those from the
control mouse group, which received a vehicle treatment. This protocol
was duly approved by the bioethics committee of the Faculty of
Medicine, University of Chile.
Statistical and Regression Analysis. The data were analyzed by
two-way ANOVA with a Bonferroni post-test. The IC50 was calculated
using a dose−response curve with a variable slope. GraphPad Prism
was also used to find the best-fit linear regression (log−log plots),
deduce the slopes and differences between slopes, and determine the
significance of the differences between the means by ANOVA. The
differences were considered significant at P < 0.05.
(4) Heijnen, C. G.; Haenen, G. R.; Vekemans, J. A.; Bast, A.
Peroxynitrite scavenging of flavonoids: structure activity relationship.
Environ. Toxicol. Pharmacol. 2001, 10, 199−206.
(5) Campos, A. M.; Ponce, E.; Lissi, E. A. Effects of alkyl chain lengths
of gallates upon their distribution and reactivity towards diphenylpicryl
hydracil radicals in Triton X-100 micellar solutions. J. Phys. Org. Chem.
2009, 22, 1208−1211.
(6) Bridi, R.; Aliaga, C.; Aspee, A.; Abuin, E.; Lissi, E. Distribution and
reactivity of gallates toward galvinoxyl radicals in SDS micellar
solutionseffect of the alkyl chain length. Can. J. Chem. 2011, 89,
181−185.
(7) Takai, E.; Hirano, A.; Shiraki, K. Effects of alkyl chain length of
gallate on self-association and membrane binding. J. Biochem. 2011, 150,
165−171.
(8) Campos, A.; Villalobos, C.; Lissi, E. Reaction of alkyl gallates with
diphenyl picrylhydracyl radicals in SDS micelles. J. Chil. Chem. Soc. 2010,
55, 458−460.
(9) Madlener, S.; Saiko, P.; Vonach, C.; Viola, K.; Huttary, N.; Stark,
N.; Popescu, R.; Gridling, M.; Vo, N. T.; Herbacek, I.; Davidovits, A.;
Giessrigl, B.; Venkateswarlu, S.; Geleff, S.; Jager, W.; Grusch, M.;
Kerjaschki, D.; Mikulits, W.; Golakoti, T.; Fritzer-Szekeres, M.;
Szekeres, T.; Krupitza, G. Multifactorial anticancer effects of digalloyl-
resveratrol encompass apoptosis, cell-cycle arrest, and inhibition of
lymphendothelial gap formation in vitro. Br. J. Cancer 2010, 102, 1361−
1370.
AUTHOR INFORMATION
Corresponding Author
*Phone: +56 2 9786069. Fax: +56 2 7355580. E-mail: jferreir@
med.uchile.cl.
■
(10) Bernhaus, A.; Fritzer-Szekeres, M.; Grusch, M.; Saiko, P.;
Krupitza, G.; Venkateswarlu, S.; Trimurtulu, G.; Jaeger, W.; Szekeres, T.
Digalloylresveratrol, a new phenolic acid derivative induces apoptosis
and cell cycle arrest in human HT-29 colon cancer cells. Cancer Lett.
2009, 274, 299−304.
(11) Rosso, R.; Vieira, T. O.; Leal, P. C.; Nunes, R. J.; Yunes, R. A.;
Creczynski-Pasa, T. B. Relationship between the lipophilicity of gallic
acid n-alquil esters’ derivatives and both myeloperoxidase activity and
HOCl scavenging. Bioorg. Med. Chem. 2006, 14, 6409−6413.
(12) Uozaki, M.; Yamasaki, H.; Katsuyama, Y.; Higuchi, M.; Higuti, T.;
Koyama, A. H. Antiviral effect of octyl gallate against DNA and RNA
viruses. Antiviral Res. 2007, 73, 85−91.
(13) Hurtado, C.; Bustos, M. J.; Sabina, P.; Nogal, M. L.; Granja, A. G.;
Gonzalez, M. E.; Gonzalez-Porque, P.; Revilla, Y.; Carrascosa, A. L.
Antiviral activity of lauryl gallate against animal viruses. Antiviral Ther.
2008, 13, 909−917.
(14) Kratz, J. M.; Andrighetti-Frohner, C. R.; Kolling, D. J.; Leal, P. C.;
Cirne-Santos, C. C.; Yunes, R. A.; Nunes, R. J.; Trybala, E.; Bergstrom,
T.; Frugulhetti, I. C.; Barardi, C. R.; Simoes, C. M. Anti-HSV-1 and anti-
HIV-1 activity of gallic acid and pentyl gallate. Mem. Inst. Oswaldo Cruz
2008, 103, 437−442.
(15) Shibata, H.; Nakano, T.; Parvez, M. A.; Furukawa, Y.; Tomoishi,
A.; Niimi, S.; Arakaki, N.; Higuti, T. Triple combinations of lower and
longer alkyl gallates and oxacillin improve antibiotic synergy against
methicillin-resistant Staphylococcus aureus. Antimicrob. Agents Chemo-
ther. 2009, 53, 2218−2220.
(16) Shibata, H.; Kondo, K.; Katsuyama, R.; Kawazoe, K.; Sato, Y.;
Murakami, K.; Takaishi, Y.; Arakaki, N.; Higuti, T. Alkyl gallates,
intensifiers of β-lactam susceptibility in methicillin-resistant Staph-
ylococcus aureus. Antimicrob. Agents Chemother. 2005, 49, 549−555.
(17) Stapleton, P. D.; Shah, S.; Anderson, J. C.; Hara, Y.; Hamilton-
Miller, J. M.; Taylor, P. W. Modulation of β-lactam resistance in
Staphylococcus aureus by catechins and gallates. Int. J. Antimicrob. Agents
2004, 23, 462−467.
Author Contributions
J.A.J., A.M., and J.F. designed the study, analyzed the data, and
wrote the manuscript. J.A.J., M.P., L.P., and F.J. performed most
of the experimental work. V.C. and J.S.-O. performed the organic
synthesis. U.K., M.E.L., M.I.B., E.P., and J.D.M. contributed to
project development and data interpretation.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by FONDECYT Grants 1090075 and
1130772, Anillo ACT-112, and a Ph.D. fellowship to J.J.
(CONICYT Grant 21080199).
ABBREVIATIONS USED
■
ANT, adenine nucleotide translocase; ATC, atractyloside; CsA,
cyclosporine A; CPT, cisplatin; CCCP, carbonyl cyanide
m-chlorophenylhydrazine; FCCP, carbonyl cyanide
p-(trifluoromethoxy)phenylhydrazone; DNP, 2,4-dinitrophenol;
DOX, doxorubicin; DTT, dithiothreitol; GA, gallic acid; HK-II,
hexokinase II; IMM, inner mitochondrial membrane; LG, lauryl
gallate; ΔΨm, mitochondrial membrane potential; MTX,
methotrexate; NAC, N-acetylcysteine; NEM, N-ethylmaleimide;
OG, octyl gallate; Ph3P, triphenylphosphine; TPP, triphenyl-
phosphonium moiety
REFERENCES
■
(1) Fiuza, S. M.; Gomes, C.; Teixeira, L. J.; Girao da Cruz, M. T.;
Cordeiro, M. N.; Milhazes, N.; Borges, F.; Marques, M. P. Phenolic acid
derivatives with potential anticancer propertiesa structure-activity
relationship study. Part 1: methyl, propyl and octyl esters of caffeic and
gallic acids. Bioorg. Med. Chem. 2004, 12, 3581−3589.
(2) Calcabrini, A.; Garcia-Martinez, J. M.; Gonzalez, L.; Tendero, M. J.;
Ortuno, M. T.; Crateri, P.; Lopez-Rivas, A.; Arancia, G.; Gonzalez-
Porque, P.; Martin-Perez, J. Inhibition of proliferation and induction of
(18) Leal, P. C.; Mascarello, A.; Derita, M.; Zuljan, F.; Nunes, R. J.;
Zacchino, S.; Yunes, R. A. Relation between lipophilicity of alkyl gallates
and antifungal activity against yeasts and filamentous fungi. Bioorg. Med.
Chem. Lett. 2009, 19, 1793−1796.
2452
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
(19) Hsu, F. L.; Chen, P. S.; Chang, H. T.; Chang, S. T. Effects of alkyl
chain length of gallates on their antifungal property and potency as an
environmentally benign preservative against wood-decay fungi. Int.
Biodeterior. Biodegrad. 2009, 36, 543−547.
(20) Kim, J. H.; Campbell, B. C.; Mahoney, N.; Chan, K. L.; Molyneux,
R. J.; Xiao, C. L. Use of chemosensitization to overcome fludioxonil
resistance in Penicillium expansum. Lett. Appl. Microbiol. 2010, 51, 177−
183.
(21) Yeh, R. D.; Chen, J. C.; Lai, T. Y.; Yang, J. S.; Yu, C. S.; Chiang, J.
H.; Lu, C. C.; Yang, S. T.; Yu, C. C.; Chang, S. J.; Lin, H. Y.; Chung, J. G.
Gallic acid induces G(0)/G(1) phase arrest and apoptosis in human
leukemia HL-60 cells through inhibiting cyclin D and E, and activating
mitochondria-dependent pathway. Anticancer Res. 2011, 31, 2821−
2832.
(22) Frey, C.; Pavani, M.; Cordano, G.; Munoz, S.; Rivera, E.; Medina,
J.; Morello, A.; Maya, J. D.; Ferreira, J. Comparative cytotoxicity of alkyl
gallates on mouse tumor cell lines and isolated rat hepatocytes. Comp.
Biochem. Physiol., A:Mol. Integr. Physiol. 2007, 146, 520−527.
(23) de Cordova, C. A. S.; Locatelli, C.; Assuncao, L. S.; Mattei, B.;
Mascarello, A.; Winter, E.; Nunes, R. J.; Yunes, R. A.; Creczynski-Pasa,
T. B. Octyl and dodecyl gallates induce oxidative stress and apoptosis in
a melanoma cell line. Toxicol. Vitro 2011, 25, 2025−2034.
(24) Locatelli, C.; Leal, P. C.; Yunes, R. A.; Nunes, R. J.; Creczynski-
Pasa, T. B. Gallic acid ester derivatives induce apoptosis and cell
adhesion inhibition in melanoma cells: The relationship between free
radical generation, glutathione depletion and cell death. Chem.-Biol.
Interact. 2009, 181, 175−184.
(25) Lopez-Rios, F.; Sanchez-Arago, M.; Garcia-Garcia, E.; Ortega, A.
D.; Berrendero, J. R.; Pozo-Rodriguez, F.; Lopez-Encuentra, A.;
Ballestin, C.; Cuezva, J. M. Loss of the mitochondrial bioenergetic
capacity underlies the glucose avidity of carcinomas. Cancer Res. 2007,
67, 9013−9017.
(26) Putignani, L.; Raffa, S.; Pescosolido, R.; Aimati, L.; Signore, F.;
Torrisi, M. R.; Grammatico, P. Alteration of expression levels of the
oxidative phosphorylation system (OXPHOS) in breast cancer cell
mitochondria. Breast Cancer Res. Treat. 2008, 110, 439−452.
(27) Modica-Napolitano, J. S.; Singh, K. K. Mitochondrial dysfunction
in cancer. Mitochondrion 2004, 4, 755−762.
(28) Gourley, P. L.; Hendricks, J. K.; McDonald, A. E.; Copeland, R.
G.; Barrett, K. E.; Gourley, C. R.; Singh, K. K.; Naviaux, R. K.
Mitochondrial correlation microscopy and nanolaser spectroscopy
new tools for biophotonic detection of cancer in single cells. Technol.
Cancer Res. Treat. 2005, 4, 585−592.
(29) Rossignol, R.; Gilkerson, R.; Aggeler, R.; Yamagata, K.;
Remington, S. J.; Capaldi, R. A. Energy substrate modulates
mitochondrial structure and oxidative capacity in cancer cells. Cancer
Res. 2004, 64, 985−993.
bioenergetics and free radical biology. Biochemistry (Moscow) 2005, 70,
222−230.
(37) Murphy, M. P.; Smith, R. A. Targeting antioxidants to
mitochondria by conjugation to lipophilic cations. Annu. Rev. Pharmacol.
Toxicol. 2007, 47, 629−656.
(38) Moreno-Sanchez, R.; Rodriguez-Enriquez, S.; Marin-Hernandez,
A.; Saavedra, E. Energy metabolism in tumor cells. FEBS J. 2007, 274,
1393−1418.
(39) Lou, P. H.; Hansen, B. S.; Olsen, P. H.; Tullin, S.; Murphy, M. P.;
Brand, M. D. Mitochondrial uncouplers with an extraordinary dynamic
range. Biochem. J. 2007, 407, 129−140.
(40) Jana, F.; Faini, F.; Lapier, M.; Pavani, M.; Kemmerling, U.;
Morello, A.; Maya, J. D.; Jara, J.; Parra, E.; Ferreira, J. Tumor cell death
induced by the inhibition of mitochondrial electron transport: the effect
of 3-hydroxybakuchiol. Toxicol. Appl. Pharmacol. 2013, 272, 356−364.
(41) Zhang, L.; Yang, N.; Garcia, J. R.; Mohamed, A.; Benencia, F.;
Rubin, S. C.; Allman, D.; Coukos, G. Generation of a syngeneic mouse
model to study the effects of vascular endothelial growth factor in
ovarian carcinoma. Am. J. Pathol. 2002, 161, 2295−2309.
(42) Ebos, J. M.; Lee, C. R.; Cruz-Munoz, W.; Bjarnason, G. A.;
Christensen, J. G.; Kerbel, R. S. Accelerated metastasis after short-term
treatment with a potent inhibitor of tumor angiogenesis. Cancer Cell
2009, 15, 232−239.
(43) Rodriguez-Enriquez, S.; Marin-Hernandez, A.; Gallardo-Perez, J.
C.; Carreno-Fuentes, L.; Moreno-Sanchez, R. Targeting of cancer
energy metabolism. Mol. Nutr. Food Res. 2009, 53, 29−48.
(44) Pathania, D.; Millard, M.; Neamati, N. Opportunities in discovery
and delivery of anticancer drugs targeting mitochondria and cancer cell
metabolism. Adv. Drug Delivery Rev. 2009, 61, 1250−1275.
(45) You, B. R.; Park, W. H. Gallic acid-induced lung cancer cell death
is related to glutathione depletion as well as reactive oxygen species
increase. Toxicol. Vitro 2010, 24, 1356−1362.
(46) Punithavathi, V. R.; Prince, P. S. M.; Kumar, R.; Selvakumari, J.
Antihyperglycaemic, antilipid peroxidative and antioxidant effects of
gallic acid on streptozotocin induced diabetic Wistar rats. Eur. J.
Pharmacol. 2011, 650, 465−471.
(47) Mayevsky, A.; Rogatsky, G. G. Mitochondrial function in vivo
evaluated by NADH fluorescence: from animal models to human
studies. Am. J. Physiol.: Cell Physiol. 2007, 292, C615−C640.
(48) Murphy, M. P. Mitochondrial thiols in antioxidant protection and
redox signaling: distinct roles for glutathionylation and other thiol
modifications. Antioxid. Redox Signaling 2012, 16, 476−495.
(49) Rodriguez-Enriquez, S.; Gallardo-Perez, J. C.; Marin-Hernandez,
A.; Moreno-Sanchez, R. The Warburg hypothesis and the ATP supply
in cancer cells. Is oxidative phosphorylation impaired in malig-
nant neoplasias? Curr. Pharm. Biotechnol. [Online early access].
Published Online: June 1, 2012; http://europepmc.org/abstract/
MED/22204369/reload=0;jsessionid=snm6QHBl9C0iY12GCOS3.12
(accessed September 2012).
(30) Christofk, H. R.; Vander Heiden, M. G.; Harris, M. H.;
Ramanathan, A.; Gerszten, R. E.; Wei, R.; Fleming, M. D.; Schreiber,
S. L.; Cantley, L. C. The M2 splice isoform of pyruvate kinase is
important for cancer metabolism and tumour growth. Nature 2008, 452,
230−233.
(50) Han, Y. H.; Moon, H. J.; You, B. R.; Kim, S. Z.; Kim, S. H.; Park,
W. H. Effects of carbonyl cyanide p-(trifluoromethoxy) phenyl-
hydrazone on the growth inhibition in human pulmonary adenocarci-
noma Calu-6 cells. Toxicology 2009, 265, 101−107.
(31) Rodriguez-Enriquez, S.; Marin-Hernandez, A.; Gallardo-Perez, J.
C.; Carreno-Fuentes, L.; Moreno-Sanchez, R. Targeting of cancer
energy metabolism. Mol. Nutr. Food Res. 2009, 53, 29−48.
(32) Pedersen, P. L. The cancer cell’s “power plants” as promising
therapeutic targets: an overview. J. Bioenerg. Biomembr. 2007, 39, 1−12.
(33) Rosano, C. Molecular model of hexokinase binding to the outer
mitochondrial membrane porin (VDAC1): implication for the design of
new cancer therapies. Mitochondrion 2011, 11, 513−519.
(34) Modica-Napolitano, J. S.; Singh, K. K. Mitochondrial dysfunction
in cancer. Mitochondrion 2004, 4, 755−762.
(51) Nadtochiy, S. M.; Tompkins, A. J.; Brookes, P. S. Different
mechanisms of mitochondrial proton leak in ischaemia/reperfusion
injury and preconditioning: implications for pathology and cardiopro-
tection. Biochem. J. 2006, 395, 611−618.
(52) Battaglia, V.; Salvi, M.; Toninello, A. Oxidative stress is
responsible for mitochondrial permeability transition induction by
salicylate in liver mitochondria. J. Biol. Chem. 2005, 280, 33864−33872.
(53) Brustovetsky, T.; Li, T.; Yang, Y.; Zhang, J. T.; Antonsson, B.;
Brustovetsky, N. BAX insertion, oligomerization, and outer membrane
permeabilization in brain mitochondria: role of permeability transition
and SH-redox regulation. Biochim. Biophys. Acta, Bioenerg. 2010, 1797,
1795−806.
(35) Coulter, C. V.; Kelso, G. F.; Lin, T. K.; Smith, R. A.; Murphy, M. P.
Mitochondrially targeted antioxidants and thiol reagents. Free Radical
Biol. Med. 2000, 28, 1547−1554.
(54) Kilbride, S. M.; Prehn, J. H. Central roles of apoptotic proteins in
mitochondrial function. Oncogene 2012, 32, 2703−2711.
(55) Basanez, G.; Soane, L.; Hardwick, J. M. A new view of the lethal
apoptotic pore. PLoS Biol. 2012, 10, e1001399.
(36) Ross, M. F.; Kelso, G. F.; Blaikie, F. H.; James, A. M.; Cocheme, H.
M.; Filipovska, A.; Da Ros, T.; Hurd, T. R.; Smith, R. A. J.; Murphy, M. P.
Lipophilic triphenylphosphonium cations as tools in mitochondrial
2453
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454

Journal of Medicinal Chemistry
Article
(56) Ng, H.; Smith, D. J.; Nagley, P. Application of flow cytometry to
determine differential redistribution of cytochrome c and Smac/
DIABLO from mitochondria during cell death signaling. PLoS One
2012, 7, e42298.
(57) Takeda, M.; Kobayashi, M.; Shirato, I.; Osaki, T.; Endou, H.
Cisplatin-induced apoptosis of immortalized mouse proximal tubule
cells is mediated by interleukin-1 beta converting enzyme (ICE) family
of proteases but inhibited by overexpression of Bcl-2. Arch. Toxicol.
1997, 71, 612−621.
(58) Locatelli, C.; Carvalho, D. R.; Mascarello, A.; de Cordova, C. A.;
Yunes, R. A.; Nunes, R. J.; Pilati, C.; Creczynski-Pasa, T. B.
Antimetastatic activity and low systemic toxicity of tetradecyl gallate
in a preclinical melanoma mouse model. Invest. New Drugs 2012, 30,
870−879.
photodamage and membrane permeabilization. PLoS One 2010, 5,
e12717.
(73) Puig, T.; Vazquez-Martin, A.; Relat, J.; Petriz, J.; Menendez, J. A.;
Porta, R.; Casals, G.; Marrero, P. F.; Haro, D.; Brunet, J.; Colomer, R.
Fatty acid metabolism in breast cancer cells: differential inhibitory
effects of epigallocatechin gallate (EGCG) and C75. Breast Cancer Res.
Treat. 2008, 109, 471−479.
(74) Jensen, M. M.; Jorgensen, J. T.; Binderup, T.; Kjaer, A. Tumor
volume in subcutaneous mouse xenografts measured by microCT is
more accurate and reproducible than determined by 18F-FDG-
microPET or external caliper. BMC Med. Imaging 2008, 8, 16.
(59) Prime, T. A.; Blaikie, F. H.; Evans, C.; Nadtochiy, S. M.; James, A.
M.; Dahm, C. C.; Vitturi, D. A.; Patel, R. P.; Hiley, C. R.; Abakumova, I.;
Requejo, R.; Chouchani, E. T.; Hurd, T. R.; Garvey, J. F.; Taylor, C. T.;
Brookes, P. S.; Smith, R. A.; Murphy, M. P. A mitochondria-targeted S-
nitrosothiol modulates respiration, nitrosates thiols, and protects against
ischemia-reperfusion injury. Proc. Natl. Acad. Sci. U.S.A. 2009, 106,
10764−10769.
(60) Kelso, G. F.; Porteous, C. M.; Coulter, C. V.; Hughes, G.;
Porteous, W. K.; Ledgerwood, E. C.; Smith, R. A.; Murphy, M. P.
Selective targeting of a redox-active ubiquinone to mitochondria within
cells: antioxidant and antiapoptotic properties. J. Biol. Chem. 2001, 276,
4588−4596.
(61) Covey, M. V.; Murphy, M. P.; Hobbs, C. E.; Smith, R. A.;
Oorschot, D. E. Effect of the mitochondrial antioxidant, Mito Vitamin E,
on hypoxic−ischemic striatal injury in neonatal rats: a dose−response
and stereological study. Exp. Neurol. 2006, 199, 513−519.
(62) Brown, S. E.; Ross, M. F.; Sanjuan-Pla, A.; Manas, A. R.; Smith, R.
A.; Murphy, M. P. Targeting lipoic acid to mitochondria: synthesis and
characterization of a triphenylphosphonium-conjugated α-lipoyl de-
rivative. Free Radical Biol. Med. 2007, 42, 1766−1780.
(63) Fones, E.; Amigo, H.; Gallegos, K.; Guerrero, A.; Ferreira, J. t-
Butyl-4-hydroxyanisole as an inhibitor of tumor cell respiration.
Biochem. Pharmacol. 1989, 38, 3443−3451.
(64) Plaza, C.; Pavani, M.; Faundez, M.; Maya, J. D.; Morello, A.;
Becker, M. I.; De Ioannes, A.; Cumsille, M. A.; Ferreira, J. Inhibitory
effect of nordihydroguaiaretic acid and its tetra-acetylated derivative on
respiration and growth of adenocarcinoma TA3 and its multiresistant
variant TA3MTX-R. In Vivo 2008, 22, 353−361.
(65) Moreadith, R. W.; Fiskum, G. Isolation of mitochondria from
ascites tumor cells permeabilized with digitonin. Anal. Biochem. 1984,
137, 360−367.
(66) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. Protein
measurement with the Folin phenol reagent. J. Biol. Chem. 1951, 3,
1125−1131.
(67) Repetto, G.; del Peso, A.; Zurita, J. L. Neutral red uptake assay for
the estimation of cell viability/cytotoxicity. Nat. Protoc. 2008, 3, 1125−
1131.
(68) Proceedings of an international conference on anticancer drugs:
therapeutic index improvement by toxicity reduction. London, 6−7
April 1987. Cancer Treat. Rev. 1987, 14, 157−420.
(69) Pulselli, R.; Arcuri, E.; Paggi, M. G.; Floridi, A. Changes in
membrane potential induced by local anesthetic bupivacaine on
mitochondria within Ehrlich ascites tumor cells. Oncol. Res. 1996, 8,
267−271.
(70) Fantin, V. R.; Berardi, M. J.; Scorrano, L.; Korsmeyer, S. J.; Leder,
P. A novel mitochondriotoxic small molecule that selectively inhibits
tumor cell growth. Cancer Cell 2002, 2, 29−42.
(71) Eng, J.; Lynch, R. M.; Balaban, R. S. Nicotinamide adenine
dinucleotide fluorescence spectroscopy and imaging of isolated cardiac
myocytes. Biophys. J. 1989, 55, 621−630.
(72) Moisenovich, M. M.; Ol’shevskaya, V. A.; Rokitskaya, T. I.;
Ramonova, A. A.; Nikitina, R. G.; Savchenko, A. N.; Tatarskiy, V. V., Jr.;
Kaplan, M. A.; Kalinin, V. N.; Kotova, E. A.; Uvarov, O. V.; Agapov, I. I.;
Antonenko, Y. N.; Shtil, A. A. Novel photosensitizers trigger rapid death
of malignant human cells and rodent tumor transplants via lipid
2454
dx.doi.org/10.1021/jm500174v | J. Med. Chem. 2014, 57, 2440−2454