Identification of novel ROS inducers: Quinone derivatives tethered to long hydrocarbon chains

Article abstract of DOI:10.1021/jm501846y

We performed the first synthesis of the 17-carbon chain-tethered quinone moiety 22 (SAN5201) of irisferin A, a natural product exhibiting anticancer activity, and its derivatives. We found that 22 is a potent ROS inducer and cytotoxic agent. Compound 25 (SAN7401), the hydroquinone form of 22, induced a significant release of intracellular ROS and apoptosis (EC₅₀ = 1.3-2.6 μM) in cancer cell lines, including A549 and HCT-116. Compared with the activity of a well-known ROS inducer, piperlongumine, 22 and 25 showed stronger cytotoxicity and higher selectivity over noncancerous cells. Another hydroquinone tethering 12-carbon chain, 26 (SAN4601), generated reduced levels of ROS but showed more potent cytotoxicity (EC₅₀ = 0.8-1.6 μM) in cancer cells, although it lacked selectivity over noncancerous cells, implying that the naturally occurring 17-carbon chain is also crucial for ROS production and a selective anticancer effect. Both 25 and 26 displayed strong, equipotent activities against vemurafenib-resistant SK-Mel2 melanoma cells and p53-deficient H1299 lung cancer cells as well, demonstrating their broad therapeutic potential as anticancer agents.

Full text of DOI:10.1021/jm501846y

Article
pubs.acs.org/jmc
Identification of Novel ROS Inducers: Quinone Derivatives Tethered
to Long Hydrocarbon Chains
Yeonsun Hong,†,§ Sandip Sengupta,‡,§ Wooyoung Hur,*,‡ and Taebo Sim*,†,‡
†KU-KIST Graduate School of Converging Science and Technology, 145 Anam-ro, Seongbuk-gu, Seoul, 136-713, Republic of Korea
‡Chemical Kinomics Research Center, Korea Institute of Science and Technology (KIST), 5 Hwarangro 14-gil, Seongbuk-gu, Seoul,
136-791, Republic of Korea
S
* Supporting Information
ABSTRACT: We performed the first synthesis of the 17-carbon chain-tethered quinone moiety 22 (SAN5201) of irisferin A, a
natural product exhibiting anticancer activity, and its derivatives. We found that 22 is a potent ROS inducer and cytotoxic agent.
Compound 25 (SAN7401), the hydroquinone form of 22, induced a significant release of intracellular ROS and apoptosis (EC50
= 1.3−2.6 μM) in cancer cell lines, including A549 and HCT-116. Compared with the activity of a well-known ROS inducer,
piperlongumine, 22 and 25 showed stronger cytotoxicity and higher selectivity over noncancerous cells. Another hydroquinone
tethering 12-carbon chain, 26 (SAN4601), generated reduced levels of ROS but showed more potent cytotoxicity (EC50 = 0.8−
1.6 μM) in cancer cells, although it lacked selectivity over noncancerous cells, implying that the naturally occurring 17-carbon
chain is also crucial for ROS production and a selective anticancer effect. Both 25 and 26 displayed strong, equipotent activities
against vemurafenib-resistant SK-Mel2 melanoma cells and p53-deficient H1299 lung cancer cells as well, demonstrating their
broad therapeutic potential as anticancer agents.
oxidative stress for several important processes such as
proliferation, angiogenesis, and metastasis.8 However, excessive
ROS levels irreversibly damage DNA and lipids and ultimately
cause apoptosis of cancer cells.9 Recently, pharmacological
elevation of intracellular ROS emerged as an effective strategy
for selectively targeting cancer cells. An exogenous ROS insult
that is within a tolerable level to normal cells could exceed the
threshold that cancer cells can endure, leading to selective
eradication of cancer cells.7,10,11 Moreover, the efficacy of this
redox-modulating method has been demonstrated in models of
drug-resistant cancers as well.12
A number of small molecule ROS inducers have been
discovered with diverse functional groups, including Michael
acceptors, disulfides, isothiocyanates, and phenolic antiox-
idants.9,12 They enhance ROS levels in cells mainly through
their intrinsic redox chemistry, inhibition of intracellular
antioxidant proteins, or perturbation of mitochondrial redox
proteins. Recently, several ROS inducers, including piperlongu-
mine and lanpersone, have been demonstrated that kill cancer
cells with good selectivity over noncancerous cells.13−15 Herein,
INTRODUCTION
■
Natural products have been a rich source of anticancer drugs
and inspired the development of a number of drugs having
novel scaffolds. Natural product-based chemical entities
account for almost 60% of anticancer drugs and show their
efficacy mainly through modulating cancer-related signaling
pathways, such as the p53 and MAPK pathways.1,2
Irisferin A is a resveratrol derivative isolated from the seed
extract of Iris pseudacorus (Iridaceae) that is utilized as a
traditional herbal medicine in Korea. A group in Korea
proposed the structure of irisferin A (Figure 1) and reported
that irisferin A is capable of suppressing proliferation of human
cancer cell lines, including SK-Mel2 (EC50 = 1.38 μM) and
A549 (EC50 = 4.65 μM).3 The antiproliferative activity against
SK-Mel2 melanoma cells is noteworthy because SK-Mel2 cells
are resistant to FDA-approved B-RafV600E inhibitor vemur-
afenib.4 Thus, it has become a crucial issue in targeted
melanoma therapy to identify novel small molecules that
override vemurafenib resistance.5,6
Cancer cells are exposed to a relatively high level of reactive
oxygen species (ROS) compared to that for normal cells,
primarily due to their active metabolism driven by oncogenic
signals.7 In fact, cancer cells take advantage of this moderate
Received: November 29, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Proposed structure of irisferin A.
we report the quinone moiety of a natural product, irisferin A,
and its derivatives as novel ROS inducers and introduce the
analogues that possess potent anticancer activity with decent
selectivity over noncancerous cells.
temperature to afford 30 in 85% yield. The oxidation process
using catalytic salcomine and oxygen bubbling was then
employed to obtain bromo-quinone 31 in 38% yield. The
bromo functionality was then converted to the methoxy group
using Pd(PPh3)4 and Na2CO3 to yield 32 in 65% yield.
Quinone 32 was reduced to hydroquinone 33 using
stoichiometric sodium hydrosulfite in THF/MeOH/H2O
(2:2:1) in 51% yield. Also, quinone 32 was demethylated
regioselectively using perchloric acid (60%) in CH2Cl2 to
obtain hydroxyl quinone 34 in 83% yield.
Biological Evaluations. We measured the antiproliferative
activities of these compounds using an MTT assay against four
cancer cell lines: SK-Mel2 (B-RafWT melanoma), A375 (B-
RafV600E melanoma), A549 (non-small cell lung cancer), and
HCT-116 (colon cancer) cells. HFF-1 fibroblast cells, HaCaT
keratinocytes, and MCF10A breast epithelial cells were also
employed as noncancerous cells. We found that ( )-ε-viniferin
possessed no antiproliferative activity against all four cancer cell
lines, whereas 22 showed potent activities (EC50 = 3.3−5.5
μM). Therefore, we envisioned that the anticancer activity of
irisferin A might be attributed to 22. Although we cannot
completely rule out the active function of (+)-ε-viniferin for
irisferin A, we focused on structure−activity relationship (SAR)
analysis of 22, as we observed antiproliferative activity solely
from 22.
The SAR with respect to cytotoxicity was similar for the four
cancer cell lines. (Hydro)quinones bearing the 17-carbon chain
showed a higher level of potency against cancer cells than
against three noncancerous cell lines (Table 1). We also found
that, compared with 22, its hydroquinone form 25 showed
slightly better cellular activities (EC50 = 1.3−2.6 μM). Phenolic
analogue 19 had significantly lower activity and dimethox-
ybenzene analogue 33 showed even lower cytotoxicity than that
of 25. Installation of an extra hydroxy (34) or methoxy group
(32, 33) to the (hydro)quinone ring significantly reduced the
activity. Despite the decreased activity, however, 32, 33, and 34
maintained fairly good cytotoxicity (EC50 = 5.1−7.8 μM),
particularly against SK-Mel2 melanoma cells, and showed about
12−15-fold selectivity over HFF-1 noncancerous fibroblasts. In
addition, 2-methoxyhydroquinone itself showed no cellular
activity, suggesting that the 17-carbon chain is critical for
cellular activity. This implies that the anticancer activity of the
compounds might be mediated in part through direct
interaction with certain intracellular targets.
We also found that when we replaced the 17-carbon chain of
22 and 25 with a 12-carbon chain, the resulting compounds
(23, 26) showed slightly better antiproliferative activity against
the cancer cell lines but with little selectivity over noncancerous
cells. This indicates that the 17-carbon chain of irisferin A
might be crucial for mediating the selective anticancer effect. In
addition, there was little change in cytotoxicity when the cis-
internal olefin of 26 was changed to the trans form (27) or
RESULTS AND DISCUSSION
■
Chemistry. Irisferin A consists of (+)-ε-viniferin and the
hydrocarbon-tethered quinone moiety 22 (SAN5201), and the
two moieties are linked through a C−C bond (Figure 1). We
hypothesized that there might be a local pharmacophore within
irisferin A responsible for its antiproliferative activity. In order
to find the structural element critical for this activity, we
assessed the two moieties separately.
The quinone moiety, 22, and its derivatives were synthesized
as shown in Scheme 1. Our synthesis began with commercially
available 3,5-dimethoxybenzyl bromide 8, which was trans-
formed to the corresponding phosphonium salt (9) at 95%
yield. We also prepared the benzyloxy aldehydes bearing two
different carbon chain lengths. Two diols (1 and 2) were
monoprotected with a benzyl group in moderate yields, and the
remaining free hydroxyl group was oxidized by PCC to afford
the desired aldehydes (3 and 4).
Wittig reaction between the phosphonium salt 9 and the
benzyloxy aldehydes (3 and 4) using n-BuLi at 0 °C furnished
mixtures of the corresponding cis/trans olefins (10 and 11) in
75 and 72% yields, respectively. Double bond reduction and
benzyl group deprotection on the olefins were simultaneously
carried out by Pd/C-catalyzed hydrogenation to yield the
corresponding primary alcohols (12 and 13) in 91 and 94%
yields. Swern oxidation followed by Wittig reaction with
phosphonium salt 6 using t-BuOK resulted in the desired cis
olefins (16 and 17) as sole products in 65−68% yields for two
steps, whereas Swern oxidation followed by modified Julia
olefination with sulfone 7 using KHMDS as a base resulted in
the trans olefin (18) as the sole product in 68% yield over 2
steps.
Crucial methoxyphenol intermediates (19, 20, and 21) were
successfully obtained through selective monodemethylation
using NaH/C2H5SH in DMF in 92−95% yields. Oxidation
reaction using catalytic salcomine and oxygen bubbling for 24 h
afforded the desired quinone compounds (22, 23, and 24) in
41 to 42% yields. The quinones were reduced to the
corresponding hydroquinones (25, 26, and 27) using
stoichiometric sodium hydrosulfite in THF/MeOH/H2O
(2:2:1) in 61 to 62% yields. Hydroquinone 25 (SAN7401)
was methylated using MeI/K2CO3 to afford 28 in 82% yield.
Hydroquinone 26 (SAN4601) was hydrogenated using Pd/C
catalyst to furnish 29 in 98% yield.
In order to synthesize dimethoxy (hydro)quinone deriva-
tives, methoxyphenol intermediate 19 was brominated
regioselectively using NBS in carbon tetrachloride at room
B
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Scheme 1. Synthesis of the Quinone Moiety of Irisferin A and Its Derivatives
saturated (29), indicating that the geometry of the 12-carbon
chain does not affect the cellular activity.
The two hydroquinones, 25 and 26, turned out to be the
most potent analogues against the cancer cells among all
C
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Antiproliferative Activity (EC50’s, μM) of 22 and Its Derivatives against Four Cancer Cell Lines and Three
a
Noncancerous Cell Lines
a
Data is presented as the mean and standard deviation obtained from three independent experiments. n.d., not determined.
derivatives having the 17-carbon chain and the 12-carbon chain,
respectively. It is noteworthy that compound 25 showed
stronger anticancer activities (EC50 = 1.3−2.6 μM) and higher
selectivity over noncancerous cells compared with that of a
well-known ROS inducer, piperlongumine. Compound 26
exhibited stronger antiproliferative activities than 25 against the
cancer cell lines (EC50 = 0.8−1.6 μM). Because 25 and 26 were
most potent against A549 and HCT116 cells, we used these
two cancer cell lines for detailed analysis.
Several studies have demonstrated that resveratrol derivatives
and fatty acids are able to generate ROS in cells.16,17 Since the
proposed structure of irisferin A contains both a resveratrol
derivative and a lipophilic long hydrocarbon chain, we first
investigated whether our synthetic analogues are able to
generate ROS in A549 cells. We monitored intracellular ROS
level using a ROS-sensitive fluorogenic dye (dichlorodihydro-
fluorescein diacetate, DCF-DA) after treating the cells with
compounds (2 and 10 μM) for 3 h. Imaging analysis revealed
that 22 strongly generated ROS in A549 cells, and its
hydroquinone form 25 induced ROS more profoundly. We
observed moderate ROS generation from 26, 27, and 29, and
almost no ROS induction from the remaining compounds
(Figure 2A).
We observed that hydroquinone analogues generated higher
ROS than their corresponding quinones. This is consistent with
the well-known concept that quinones are bioreduced and the
resulting hydroquinones act as the active species for intra-
cellular ROS generation. When we installed an extra electron-
donating group (hydroxyl or methoxy group) on the
(hydro)quinone ring, ROS generation and antiproliferative
activity were greatly decreased. The electron-donating sub-
stituents on hydroquinone might prohibit the tendency to
transfer a single electron to molecular oxygen or might render
the enol form (hydroquinone) disfavored between keto−enol
equilibrium, resulting in reduced ROS generation.18 Moreover,
25 and 26 generated a more intense ROS signal in A549 lung
cancer cells than in noncancerous HFF-1 cells (Figure 2B).
Particularly, 25 showed ROS generation in A549 cells but
barely promoted the release of ROS in HFF-1 cells. This
suggests that selective cytotoxicity of 25 against A549 cells over
D
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. Compound 22 and its cytotoxic analogues showed evident induction of intracellular ROS in A549 lung cancer cells. Cells were treated with
the indicated concentrations of compounds for 3 h before staining with 10 μM DCF-DA for 30 min and then fixing with 4% formaldehyde for 20
min. (A) Compound 22 strongly generated ROS, and its hydroquinone form, 25, does so to a greater extent. A moderate ROS level was observed
from 26, 27, and 29, whereas almost no ROS induction was detected from the rest of the compounds. Fluorescence signal was detected and
quantified using an Operetta high-content imaging system. (B) Compounds 25 and 26 showed more prominent ROS induction in A549 cancer cells
than that in HFF-1 fibroblasts. Selective ROS generation was particularly evident from 25. In addition, only marginal ROS levels were observed from
2-methoxyhydroquinone in A549 cells. Fluorescence was detected using a fluorescence microscope. (C) Similar experiments were done in A549 cells
using three different ROS probes (DCF-DA, DHE, APF) with pretreatment with 10 mM NAC or 0.5 mM AA. Compound 25 increased ROS levels
most significantly in A549 cells, whereas 22, 26, and piperlongumine increased ROS levels to a lesser extent. Pretreatment with NAC or AA greatly
diminished ROS generation.
E
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Both 25 and 26 activated Akt and multiple MAPKs in A549 and HCT-116 cells in a ROS-dependent manner. (A) Treatment with 25 or
26 for 3 h enhanced phosphorylation of Akt, JNK, p38, and ERK1/2 in A549 and HCT-116 cells in a dose-dependent manner. (B) A549 and HCT-
116 cells were pretreated with 10 mM NAC for 2 h prior to treatment with 25 or 26 at 10 μM for 3 h. NAC significantly attenuated the activation of
Akt, JNK, p38, and ERK1/2, indicating that the compounds induced ROS-mediated activation of Akt, JNK, p38, and ERK1/2.
HFF-1 cells might be associated with its selective ROS
production in A549 cells. In addition, 2-methoxyhydroquinone,
which showed no cytotoxicity (Table 1), barely induced ROS in
cells, indicating that not only hydroquinone but also the
naturally occurring 17-carbon chain plays a major role in
escalating intracellular ROS levels.
treatment with the compounds (10 μM), ROS generation was
greatly attenuated.
We also found that compounds with a 12-carbon chain (e.g.,
26) showed better cytotoxicity but released ROS to a lesser
extent than those with a 17-carbon chain (e.g., 25). In fact, the
discrepancy between ROS generation and cytotoxicity has been
reported by others.14,20 For example, the cytotoxicity of
piperlongumine analogues is well-correlated with the ability
to deplete cellular glutathione (GSH) rather than the ability to
induce ROS in cells.14 Thus, we compared the ability of 25, 26,
and piperlongumine to deplete intracellular GSH in A549 cells
(Figure S1, Supporting Information). Compound 25 sparingly
affected the total GSH level and marginally reduced the
reduced GSH level, whereas it slightly increased the level of
oxidized glutathione (GSSG). This suggests that the high level
of ROS induced by 25 caused oxidation of intracellular GSH
into GSSG. On the other hand, consistent with previous
reports,13,14,20 piperlongumine significantly suppressed the level
of total GSH. Compound 26 depleted the total GSH level more
significantly than piperlongumine, which might be associated
with its stronger cytotoxicity. Thus, unlike 25, whose
mechanism of action seems to be mainly attributed to ROS
We then compared 22, 25, and 26 with piperlongumine with
regard to their capability of inducing ROS in A549 cells.
Besides DCF-DA, which detects a broad range of ROS
including hydrogen peroxide, hydroxyl radical, and peroxy
radicals, we employed two other ROS sensors that probe more
specific ROS in cells.19 APF (3′-(p-aminophenyl) fluorescein)
mainly senses hydroxy radical, whereas dihydroethidium
(DHE) selectively detects superoxide anion in cells. Fluo-
rescence imaging with the three different ROS probes showed
consistent results for the ROS inducers. Notably, the level of
ROS generated by 25 outcompeted that induced by
piperlongumine, whereas 22, 26, and piperlongumine showed
similar levels of ROS induction (Figure 2C). When we
pretreated cells with either 10 mM N-acetyl cysteine (NAC)
or 0.5 mM ascorbic acid (AA) as an antioxidant prior to
F
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Both 25 and 26 induced apoptosis in A549 and HCT-116 cells through generation of ROS. (A) Western blot analysis showed that
treatment with 25 or 26 for 48 h caused apoptosis in A549 and HCT-116 cells in a dose-dependent manner. The levels of apoptotic markers (PARP,
AIF, BIM, BAX, BAK, and p53) were increased, whereas antiapoptotic marker Bcl-2 was downregulated. (B) Pretreatment with 10 mM NAC or 0.5
mM AA prior to 25 or 26 at 10 μM blocked the generation of PARP cleavage and p53 in A549 and HCT-116 cells, indicating that 25 and 26
induced apoptosis through ROS generation.
generation, piperlongumine and 26 might have mechanisms for
perturbing GSH homeostasis as well, contributing to
apoptosis.21
NAC itself also downregulated the basal phospho-levels of Akt,
JNK, p38, and ERK1/2. Activation of these kinases is usually
associated with cell proliferation and survival. However, it has
also been reported that increased MAPK and Akt phosphor-
ylation sensitize cells to oxidative stress, thereby inducing
apoptosis.26−29 Thus, the increased ROS levels and the
enhanced kinase activity induced by the compounds might be
synergistically exerted in both of these cancer cell lines,
inducing apoptosis.
Then, we examined apoptosis in A549 and HCT-116 cells
following treatment with the compounds for 48 h. Western blot
analysis showed that both 25 and 26 increased the levels of
apoptotic markers, including cleaved PARP, cleaved caspase-3,
BIM, BAK, and p53, and decreased antiapoptotic marker Bcl-2
in a dose-dependent manner (Figure 4A). Likewise, pretreat-
ment with 10 mM NAC or 0.5 mM AA attenuated the
escalation of p53 and cleaved PARP (Figure 4B), indicating
that 25 and 26 induced apoptosis through generation of ROS.
We also confirmed apoptosis from FACS analysis of A549 and
It has been reported that in response to ROS, MAPK
pathways such as JNK, p38, and ERK1/2 are activated,
ultimately leading to apoptosis.22−25 As 25 and 26 induced
prominent ROS generation, we investigated whether they
upregulated MAPK pathways. Western blot analysis using A549
and HCT-116 cancer cells showed that treatment with 25 or 26
for 3 h increased the phosphorylation of JNK1/2, p38, and
ERK1/2 in dose-dependent manner (Figure 3A). We also
observed that phosphorylation of Akt was increased in a similar
fashion in both A549 and HCT-116 cells. To confirm that the
Akt/MAPK pathways were activated by ROS generation, cells
were pretreated with 10 mM NAC prior to treatment with 10
μM 25 or 26, and then the phospho-levels were examined.
NAC negated the increase of both phospho-Akt and phospho-
MAPKs, confirming that 25 and 26 activated Akt and MAPK
pathways in a ROS-dependent manner (Figure 3B). Because
the antioxidant removed the basal level of ROS, treatment with
G
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 5. Compounds 22, 25, and 26 disrupted mitochondrial membrane integrity in A549 cells through generation of ROS. A549 cells were treated
with 10 μM 22, 25, or 26 for 24 h before JC-1 staining. Compound 26 greatly reduced mitochondrial membrane potential, whereas 22 and 25
reduced it to a lesser extent. Pretreatment with 10 mM NAC or 0.5 mM AA significantly negated the effect of the compounds, indicating that the
three compounds induced a loss of mitochondrial membrane integrity in A549 cells through generation of ROS.
Figure 6. Compounds 25 and 26 potently blocked proliferation of p53-deficient H1299 lung cancer cells.
HCT-116 cells following treatment with 10 μM 26 for 48 h
(Figure S2, Supporting Information).
μM EC50’s) against H1299 cells, comparable to the activity
against p53+/+ A549 lung cancer cells (Figure 6). This result
suggests that our compounds induce cell death in a p53-
independent manner as well and demonstrates their broad
spectrum of anticancer effects.
In order to assess whether observed apoptosis was mediated
through mitochondria, we monitored mitochondrial outer
membrane potential. JC-1 (5,5′,6,6′-tetrachloro-1,1′,3,3′-tetrae-
thylbenzimidazolocarbocyanine iodide) is a mitochondrial
potential-sensitive fluorometric probe.30 It reversibly changes
from a red aggregate to a green monomer as the membrane
potential decreases due to apoptosis. Imaging analysis using JC-
1 showed that 24 h treatment with 25 or 26 at 10 μM increased
the green monomeric form of JC-1 in A549 cells (Figure 5),
indicating mitochondrial membrane depolarization and apop-
tosis. Moreover, 26 generated a higher JC-1 green fluorescence
signal than 25, which was consistent with the higher
antiproliferative activity of 26 measured with the MTT assay
(Table 1). Cotreatment with either 10 mM NAC or 0.5 mM
AA attenuated the loss of mitochondrial membrane potential,
indicating that disruption of mitochondrial membrane integrity
was dependent on ROS generation.
We then performed a soft agar assay using A549 cells with
treatment of 25, 26, and piperlongumine at 10 μM for 10 days
(Figure S3, Supporting Information). Piperlongumine and 25
significantly decreased the colony size, whereas 26 reduced the
size more significantly. This result showed that 25 and 26 were
effective in blocking anchorage-independent A549 cell growth
in vitro.
Herein, we described the synthesis of the 17-carbon chain-
tethered quinone part (22) of the proposed structure of natural
product irisferin A and its derivatives. There is no report yet
about 22 and its derivatives as ROS inducers. Thus, we report
that they are novel ROS inducers. We found that 22 itself is a
strong ROS inducer and kills cancer cells with good potency
and selectivity. Among our synthetic analogues of 22, 25 and
26 induced ROS generation and exhibited the highest
cytotoxicity against cancer cells. Compound 25, the hydro-
quinone form of 22, was the most potent ROS inducer and
selectively produced high levels of ROS in cancer cells. It
turned out to be superior to piperlongumine in terms of ROS
induction capability and antiproliferative activity and was
selective against cancer cells over noncancerous cells.
Compound 26, bearing a shorter 12-carbon chain, induced
lower levels of ROS compared with those from 25, but it
exhibited higher antiproliferative activities against A549 and
HCT-116 cells, indicating that, aside from ROS generation,
there might be another mechanism for the cytotoxicity elicited
by 26.
It is well-documented that (hydro)quinones produce ROS,
thereby damaging cancer cells. Interestingly, our study revealed
that the naturally occurring 17-carbon chain moiety attached to
the (hydro)quinone backbone also plays a critical role in
generating ROS in cancer cells and confers selectivity over
noncancerous cells. To the best of our knowledge, this is the
first report demonstrating that the long hydrocarbon chain on
Both JC-1 staining and western blot analysis indicated that
25 and 26 are potent inducers of apoptosis. As both
compounds enhanced p53 levels (Figure 4), they induced
p53-mediated apoptosis. We examined the cytotoxicity of 25
and 26 against p53-deficient H1299 lung cancer cells as well.
Interestingly, both compounds showed potent activities (ca. 1
H
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
DMF (20 mL), and refluxed overnight. After being cooled to room
temperature, the solution was acidified with 2 N HCl and extracted
with ether. The organic layer was washed with brine solution, dried
over MgSO4, and concentrated under vacuum. The resulting crude
product was purified by silica gel column chromatography (EtOAc/
hexane 1:4) to afford compound 19 (1.8 g, 94%) as a pale yellow
liquid. Rf = 0.2 (20% EtOAc/hexane); IR (neat, KBr): 3405, 2926,
(hydro)quinone is crucial for ROS generation and cytotoxicity
in cancer cells. It is also worth mentioning that our ROS
inducers showed strong inhibitory activities against vemur-
afenib-resistant SK-Mel2 melanoma cells and p53-null H1299
lung cancer cells as well. Our study demonstrates that further
tuning of our compounds could lead to decent small molecule
agents that effectively and selectively target various cancers.
1
2855, 1615, 1597, 1458, 1147 cm−1; H NMR (400 MHz, CDCl3):
6.34 (d, J = 2.6 Hz, 1H), 6.29 (t, J = 2.6 Hz, 2H), 5.36−5.34 (m, 2H),
3.78 (s, 3H), 2.54 (t, J = 7.6 Hz, 2H), 2.05−1.96 (m, 4H), 1.64−1.56
(m, 2H), 1.33−1.28 (m, 20H), 0.88 (t, J = 6.7 Hz, 3H); 13C NMR
(100 MHz, CDCl3): δ 160.6, 156.4, 145.7, 129.9, 129.8, 108.0, 106.7,
98.7, 55.2, 36.0, 32.5, 31.7, 31.1, 29.7(2), 29.6, 29.5(2), 29.3(2), 28.9,
27.2, 22.6, 14.1; HRMS (ESI): calcd. for C24H40O2Na [M + Na]+,
383.2926; found, 383.2918.
(Z)-2-(Heptadec-10-en-1-yl)-6-methoxycyclohexa-2,5-diene-1,4-
dione (22). Into a stirred solution of 19 (220 mg) and salcomine (25
mg) in DMF (5 mL) was bubbled oxygen at room temperature for 24
h. The mixture was diluted with water and extracted with ether. The
organic layer was washed with brine solution, dried over MgSO4, and
concentrated under vacuum. The resulting crude product was purified
by silica gel column chromatography (EtOAc/hexane 1:20) to afford
22 (95 mg, 42%) as brown liquid. Rf = 0.5 (20% EtOAc/hexane); IR
(neat, KBr): 2950, 2854, 1680, 1652, 1601, 1485 cm−1; 1H NMR (400
MHz, CDCl3): 6.47 (d, J = 2.2 Hz, 1H); 5.86 (d, J = 2.2 Hz, 1H); 5.34
(t, J = 5.5 Hz, 2H); 3.81 (s, 3H); 2.42 (t, J = 8.3 Hz, 2H); 2.00 (q, J =
6.3 Hz, 4H); 1.53−1.45 (m, 2H); 1.32−1.21 (m, 20H), 0.88 (t, J = 7.0
Hz, 3H); 13C NMR (100 MHz, CDCl3): δ 187.5, 181.9, 158.7, 147.4,
132.7, 129.7, 129.6, 106.9, 56.1, 31.6, 29.6, 29.4, 29.3(3), 29.2, 29.1,
28.8, 28.5, 27.6, 27.1, 27.0, 22.5, 13.9; HRMS (ESI): calcd. for
C24H38O3Na [M + Na]+, 397.2719; found, 397.2715.
(Z)-2-(Dodec-5-en-1-yl)-6-methoxycyclohexa-2,5-diene-1,4-dione
(23). Into a stirred solution of 20 (100 mg) and salcomine (15 mg) in
DMF (3 mL) was bubbled oxygen at room temperature for 24 h. The
mixture was diluted with water and extracted with ether. The organic
layer was washed with brine, dried over MgSO4, and concentrated
under vacuum. The resulting crude product was purified by silica gel
column chromatography (EtOAc/hexane 1:20) to afford 23 (41 mg,
41%) as a yellow liquid. Rf = 0.5 (20% EtOAc/hexane); IR (neat,
KBr): 2926, 2853, 1660, 1602, 1458, 1230 cm−1; 1H NMR (400 MHz,
CDCl3): 6.47 (m, 1H), 5.86 (d, J = 2.3 Hz, 1H), 5.42−5.27 (m, 2H),
3.81 (s, 3H), 2.43 (t, J = 8.5 Hz, 2H), 2.08−1.93 (m, 4H), 1.55−1.46
(m, 2H), 1.44−1.36 (m, 2H), 1.34−1.21 (m, 8H), 0.87 (t, J = 7.0 Hz,
3H); 13C NMR (100 MHz, CDCl3): δ 187.5, 182.0, 158.7, 147.3,
132.8, 130.5, 128.9, 107.0, 56.2, 31.7, 29.6, 29.2, 28.9, 28.5, 27.2(2),
26.7, 22.5, 14.0; HRMS (ESI): calcd. for C19H28O3Na [M + Na]+,
327.1937; found, 327.1936.
CONCLUSIONS
■
Here, we synthesized the 17-carbon chain-tethered quinone
moiety (22) within the proposed structure of natural product
irisferin A and its derivatives. For the first time, we found them
to be novel ROS inducers and report that the naturally
occurring 17-carbon chain is critical for ROS generation and
selective cytotoxicity toward cancer cells over noncancerous
cells. We found that 25, the reduced form of 22, generated an
exceptionally high level of intracellular ROS and potently killed
cancer cells (EC50 = ca. 1.3−2.6 μM). Interestingly, another
analogue, 26, bearing a shorter 12-carbon chain showed a
higher cytotoxicity (EC50 = ca. 0.8−1.6 μM) but generated a
lower level of ROS than that with 25, which suggest that
another mode of action might be involved in its potent
cytotoxicity. Both 25 and 26 activated Akt and multiple MAPK
pathways in a ROS-dependent fashion, thereby inducing
apoptosis. Both compounds showed stronger antiproliferative
activity than piperlongumine against four cancer cell lines
(A549, A375, SK-Mel2, and HCT116 cells). In particular,
compound 25 turned out to be superior to piperlongumine not
only in terms of antiproliferative activity against cancer cells but
also selectivity over noncancerous cells (HFF-1, MCF10A, and
HaCaT cells). Their potent antiproliferative activities against
various cancer cells, including vemurafenib-resistant SK-Mel2
melanoma cells and p53-null H1299 lung cancer cells, also
demonstrate the broad therapeutic potential of 25 and 26 as
anticancer agents.
EXPERIMENTAL SECTION
■
Chemical Synthesis. General. Reactions were monitored by TLC
using 0.25 mm Merck precoated silica gel plates (60F254). Reaction
progress was monitored by TLC analysis using a UV lamp or
ninhydrin/p-anisaldehyde stain for detection purposes. Commercially
available reagents were used without further purification. All solvents
were purified by standard techniques. Purification of reaction products
was carried out by silica gel column chromatography using Kieselgel 60
Art. 9385 (230−400 mesh). The purity of all compounds was over
95% and was analyzed using a Waters LC/MS system (Waters 2998
photodiode array detector, Waters 3100 mass detector, Waters SFO
system fluidics organizer, Waters 2545 binary gradient module, Waters
reagent manager, and Waters 2767 sample manager) using a
SunFireTM C18 column (4.6 × 50 mm, 5 μm particle size). The
solvent gradient was 60% (or 95%) A at 0 min and 1% A at 5 min.
Solvent A was 0.035% TFA in H2O; solvent B was 0.035% TFA in
(Z)-2-(Heptadec-10-en-1-yl)-6-methoxybenzene-1,4-diol (25). To
a stirred solution of 22 (90 mg, 0.24 mmol) in THF/MeOH/H2O
(4:3:3, 10 mL) was added Na2S2O4 (170 mg, 0.96 mmol) at room
temperature. The reaction mixture was stirred for 24 h, condensed
under vacuum, and extracted with EtOAc. The organic layer was
washed with water and brine solution, dried over MgSO4, and
concentrated under vacuum. The residue was purified by silica gel
column chromatography (EtOAc/hexane 1:15) to afford 25 (56 mg,
62%) as a yellow solid. Rf = 0.4 (20% EtOAc/hexane); IR (neat, KBr):
1
3551, 3325, 1649, 1601, 1458 cm−1; H NMR (400 MHz, CDCl3):
MeOH. The flow rate was 3.0 (or 2.5) mL/min. H and 13C NMR
1
6.31 (d, J = 2.7 Hz, 1H), 6.20 (d, J = 2.7 Hz, 1H), 5.35 (t, J = 4.6 Hz,
2H), 5.22 (brs, 1H), 3.84 (s, 3H), 2.56 (t, J = 7.6 Hz, 2H), 2.05−1.99
(m, 4H), 1.61−1.53 (m, 2H), 1.31−1.24 (m, 20H), 0.88 (t, J = 7.0 Hz,
3H); 13C NMR (100 MHz, CDCl3): δ 148.3, 146.7, 137.3, 129.9,
129.8, 129.0, 107.9, 97.0, 55.5, 31.7, 29.7(4), 29.6, 29.5(4), 29.3, 28.9,
27.2, 22.6, 14.1; HRMS (ESI): calcd. for C24H40O3Na [M + Na]+,
399.2875; found, 399.2862.
(Z)-2-(Dodec-5-en-1-yl)-6-methoxybenzene-1,4-diol (26). To a
stirred solution of 23 (50 mg, 0.16 mmol) in THF/MeOH/H2O
(4:3:3, 4 mL) was added Na2S2O4 (25 mg, 0.48 mmol) at room
temperature. The reaction mixture was stirred for 24 h, condensed
under vacuum, and extracted with EtOAc. The organic layer was
washed with water and brine solution, dried over MgSO4, and
spectra were obtained using a Bruker 400 MHz FT-NMR (400 MHz
for 1H and 100 MHz for 13C) spectrometer. Chemical shifts are
1
reported relative to CHCl3 (δ = 7.26) for H NMR and CHCl3 (δ =
77.0) for 13C NMR. Standard abbreviations are used to denote signal
multiplicities. Infrared spectra were measured on an FT-IR Nicolet
iS10 spectrometer. Samples were recorded neat or as KBr optics. High-
resolution mass spectra (HRMS) were recorded on a Q-TOF mass
spectrometer.
(Z)-3-(Heptadec-10-en-1-yl)-5-methoxyphenol (19). To a suspen-
sion of 60% NaH (963 mg, 24 mmol) in DMF (20 mL) was added a
solution of ethanethiol (1.86 mL, 24.34 mmol). The mixture was
stirred for 5 min, treated with a solution of 16 (2 g, 5.35 mmol) in
I
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
concentrated under vacuum. The residue was purified by silica gel
column chromatography (EtOAc/hexane 1:15) to give 26 (30 mg,
61%) as a white solid. Rf = 0.4 (20% EtOAc/hexane); IR (neat, KBr):
3558, 3393, 2924, 2858, 1678, 1602, 1379, 1218 cm−1; 1H NMR (400
MHz, CDCl3): 6.30 (d, J = 2.7 Hz, 1H), 6.22 (d, J = 2.7 Hz, 1H), 5.35
(t, J = 4.9 Hz, 2H), 5.26 (s, 1H), 3.82 (s, 3H), 2.58 (t, J = 7.8 Hz, 2H),
2.08−1.94 (m, 4H), 1.64−1.56 (m, 2H), 1.44−1.36 (m, 2H), 1.36−
1.26 (m, 8H), 0.88 (t, J = 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3):
δ 148.3, 146.7, 137.3, 130.1, 129.6, 128.8, 107.9, 97.1, 55.9, 31.7, 29.7,
29.5, 29.4, 29.3, 28.9, 27.2, 27.0, 22.6, 14.06; HRMS (ESI): calcd. for
C19H30O3Na [M + Na]+, 329.2092; found, 329.2085.
(E)-2-(Dodec-5-en-1-yl)-6-methoxybenzene-1,4-diol (27). To a
stirred solution of 24 (40 mg, 0.13 mmol) in THF/MeOH/H2O
(4:3:3, 4 mL) was added Na2S2O4 (21 mg, 0.40 mmol) at room
temperature. The reaction mixture was stirred for 24 h, condensed
under vacuum, and extracted with EtOAc. The organic layer was
washed with water and brine solution, dried over MgSO4, and
concentrated under vacuum. The residue was purified by silica gel
column chromatography (EtOAc/hexane 1:15) to give 27 (24 mg,
61%) as an off-white solid. Rf = 0.4 (20% EtOAc/hexane); IR (neat,
KBr): 3557, 3390, 2925, 2858, 1678, 1600, 1379, 1218 cm−1; 1H NMR
(400 MHz, CDCl3): 6.30 (d, J = 2.7 Hz, 1H), 6.20 (d, J = 2.7 Hz, 1H),
5.40−5.34 (m, 2H), 5.22 (s, 1H), 3.84 (s, 3H), 2.60−2.55 (m, 2H),
2.08−1.94 (m, 4H), 1.65−1.55 (m, 2H), 1.36−1.20 (m, 10H), 0.88 (t,
J = 6.6 Hz, 3H); 13C NMR (100 MHz, CDCl3): δ 148.2, 146.7, 137.4,
130.5, 130.1, 128.8, 107.9, 97.1, 55.9, 32.4, 31.7, 29.7, 29.6, 29.5, 29.3,
28.9, 27.2, 22.6, 14.0; LCMS (ESI): 307 (M + H)+.
(Z)-1-(Heptadec-10-en-1-yl)-2,3,5-trimethoxybenzene (28). A
mixture of 25 (20 mg, 0.05 mmol), methyl iodide (0.1 mL), and
K2CO3 (25 mg) in acetone (10 mL) was refluxed overnight. After
being cooled to room temperature, the reaction mixture was diluted
with water and extracted with EtOAc. The organic layer was washed
with water and brine solution, dried over MgSO4, and concentrated
under vacuum. The residue was purified by silica gel column
chromatography (EtOAc/hexane 1:30) to afford 28 (17 mg, 82%)
as a yellow liquid. Rf = 0.2 (5% EtOAc/hexane); IR (neat, KBr): 2925,
2853, 1600, 1492, 1466, 1222, 1058 cm−1; 1H NMR (400 MHz,
CDCl3): 6.34 (d, J = 2.8 Hz, 1H), 6.28 (d, J = 2.8 Hz, 1H), 5.35 (t, J =
4.6 Hz, 2H), 3.83 (s, 3H), 3.77 (s, 3H), 3.75 (s, 3H), 2.58 (t, J = 7.9
Hz, 2H), 2.06−1.97 (m, 4H), 1.61−1.54 (m, 2H), 1.32−1.27 (m,
20H), 0.88 (t, J = 6.7 Hz, 3H); 13C NMR (100 MHz, CDCl3): δ 155.8,
153.2, 141.1, 136.9, 130.3, 129.9, 104.9, 97.6, 60.8, 55.6, 55.4, 32.6,
31.7, 30.8, 30.1, 29.7(2), 29.6, 29.5(3), 29.3, 28.9, 27.2, 22.6, 14.1;
HRMS (ESI): calcd. for C26H44O3Na [M + Na]+, 427.3188; found,
427.3196.
Hz, 2H), 2.03−1.96 (m, 4H), 1.41−1.26 (m, 24H), 0.87 (t, J = 6.7 Hz,
3H); 13C NMR (100 MHz, CDCl3): δ 183.6, 182.4, 158.7, 155.8,
130.7, 129.9, 129.8, 105.3, 61.3, 56.3, 31.7, 29.7(2), 29.6(2), 29.5, 29.4,
29.3(2), 28.9, 28.6, 27.2, 23.0, 22.6, 14.1; HRMS (ESI): calcd. for
C25H40O4Na [M + Na]+, 427.2824; found, 427.2815.
(Z)-3-(Heptadec-10-en-1-yl)-2,5-dimethoxybenzene-1,4-diol (33).
To a stirred solution of 32 (30 mg, 0.07 mmol) in THF/MeOH/H2O
(4:3:3, 2 mL) was added Na2S2O4 (50 mg, 0.96 mmol) at room
temperature. The reaction mixture was stirred for 24 h, condensed
under vacuum, and extracted with EtOAc. The organic layer was
washed with water and brine solution, dried over MgSO4, and
concentrated under vacuum. The residue was purified by silica gel
column chromatography (EtOAc/hexane 1:15) to give 33 (12 mg,
51%) as a yellow liquid. Rf = 0.5 (20% EtOAc/hexane); IR (neat,
KBr): 3553, 3370, 1608, 1482, 1452 cm−1; 1H NMR (400 MHz,
CDCl3): 6.43 (s, 1H), 5.34 (t, J = 4.7 Hz, 2H), 5.26 (s, 1H), 5.23 (s,
1H), 3.82 (s, 3H), 3.74 (s, 3H), 2.64 (t, J = 7.9 Hz, 2H), 2.02−1.97
(m, 4H), 1.63−1.55 (m, 2H), 1.43−1.28 (m, 20H), 0.88 (t, J = 7.0 Hz,
3H); 13C NMR (100 MHz, CDCl3): 158.8, 155.9, 138.9, 137.1, 130.7,
129.9, 105.4, 96.8, 61.7, 56.1, 31.8, 29.9, 29.7, 29.6, 29.5(3), 29.4(4),
28.7, 27.2, 22.6, 14.1.
(Z)-3-(Heptadec-10-en-1-yl)-2-hydroxy-5-methoxycyclohexa-2,5-
diene-1,4-dione (34). To a stirred solution of 32 (50 mg) in CH2Cl2
(5 mL) was added a catalytic amount of HClO4 (60%) at room
temperature. The reaction mixture was stirred for 12 h, washed with
NaHCO3 solution and brine solution, dried over MgSO4, and
concentrated under vacuum. The residue was purified by silica gel
column chromatography (EtOAc/hexane 1:5) to afford 34 (40 mg,
83%) as yellow solid. Rf = 0.2 (20% EtOAc/hexane); IR (neat, KBr):
3344, 2921, 2851, 1660, 1596, 1384, 1202 cm−1; 1H NMR (400 MHz,
CDCl3): 7.21 (brs, 1H), 5.83 (s, 1H), 5.34 (t, J = 4.9 Hz, 2H), 3.85 (s,
3H), 2.43 (t, J = 7.4 Hz, 2H), 2.04−1.98 (m, 4H), 1.51−1.40 (m, 2H),
1.27−1.24 (m, 20H), 0.88 (t, J = 6.7 Hz, 3H); 13C NMR (100 MHz,
CDCl3): δ 182.7, 181.6, 161.1, 151.5, 130.3, 129.8, 119.2, 102.1, 56.7,
32.6, 31.7, 31.5, 29.7(2), 29.5(2), 29.4, 29.3, 28.9, 28.0, 27.2, 22.6(2),
14.1; HRMS (ESI): calcd. for C24H38O4Na [M + Na]+, 413.2668;
found, 413.2656.
Biological Experiments. Cell Proliferation Assay (MTT Assay).
Cells (5000−30000 cells/well) were seeded in a 96-well plate. After
18−24 h, cells were treated with the serially diluted (3-fold, 12 point)
compounds and were incubated at 37 °C in a 5% CO2 atmosphere for
72 h. Cells were then subjected to MTT solutions (Promega)
following the manufacturer’s procedure. The optical density at 570 nm
of each well was detected using a plate reader (Envision, PerkinElmer).
The data was analyzed using Prism 5 software (GraphPad).
Western Blot Analysis. Cells were seed at a density of 6 × 105 cells/
well in a 60 mm dish and incubated overnight. Cells were treated with
the compounds at 37 °C in a 5% CO2 atmosphere, washed twice with
cold PBS, and then scraped with RIPA buffer. The resulting cell lysates
were centrifuged at 15 000 rpm for 20 min. Total protein
concentration was determined by BCA assay (Pierce). A total of
25−50 μg of the lysate was used for western blot analysis. Proteins
separated by SDS-PAGE were electrophoretically transferred to
methanol-activated PVDF membranes. After blocking in 1× TBS/T
containing 5% nonfat dry milk for 1 h, membranes were incubated
with primary antibodies overnight at 4 °C. After washing, the blots
were incubated with 1:2000 dilutions of horseradish peroxidase-
conjugated secondary antibodies (Santa Cruz Biotechnology) for 1 h
at room temperature. After washing, the immune-reactive proteins
were visualized using ECL reagents.
Measurement of Intracellular ROS Levels. Cells were seeded at 1 ×
106 cells/well in a 12-well plate onto cover glass and incubated for 18−
24 h. Cells were then treated with the compounds for the indicated
periods of time and dosages, treated with 10 μM DCF-DA (Sigma),
DHE (Sigma), or APF (Invitrogen) for 30 min, and fixed with 4%
formaldehyde for 20 min. The cells were washed with PBS twice, and
fluorescence was detected via fluorescence microscopy (Nikon). For a
similar assay using a high-content cellular imaging system, cells were
seeded at 1 × 104 cells/well in a black 96-well plate and incubated for
18−24 h. Cells were then treated with the compounds for 3 h, treated
2-Dodecyl-6-methoxybenzene-1,4-diol (29). To a stirred solution
of 26 (20 mg, 0.06 mmol) in methanol was added Pd/C (2 mg) in one
portion. The reaction mixture was stirred at room temperature for 2 h
under hydrogen pressure and then filtered through a Celite pad. The
filtrate was concentrated under vacuum to obtain 29 (18 mg, 98%) as a
white solid. Rf = 0.35 (20% EtOAc/hexane); IR (neat, KBr): 3565,
3349, 2954, 2914, 2850, 1653, 1598, 1471, 1233 cm−1; 1H NMR (400
MHz, CDCl3): 6.31 (d, J = 1.7 Hz, 1H), 6.22 (d, J = 1.7 Hz, 1H), 4.63
(brs, 1H), 3.84 (s, 3H), 2.57 (t, J = 7.9 Hz, 2H), 1.43−1.15 (m, 20H),
0.89 (t, J = 7.0 Hz, 3H); 13C NMR (100 MHz, CDCl3): δ 148.2, 146.7,
137.4, 129.0, 107.9, 97.0, 56.0, 31.9, 29.7, 29.6(3), 29.5, 29.3(4), 22.7,
14.1; HRMS (ESI): calcd. for C19H32O3Na [M + Na]+, 331.2249;
found, 331.2226.
(Z)-3-(Heptadec-10-en-1-yl)-2,5-dimethoxycyclohexa-2,5-diene-
1,4-dione (32). To a stirred solution of 31 (100 mg, 0.22 mmol) and
tetrakis(triphenylphosphine)palladium (5 mg) in CH3OH/THF (1:1,
10 mL) under argon was added a 2 N Na2CO3 solution (0.5 mL). The
mixture was stirred for 10 min at 90 °C, concentrated under vacuum,
and partitioned between EtOAc and water. The EtOAc layer was
washed with brine solution, dried over MgSO4, and concentrated
under vacuum. The residue was purified by silica gel column
chromatography (EtOAc/hexane 1:20) to afford 32 (58 mg, 65%)
as a liquid. Rf = 0.5 (20% EtOAc/hexane); IR (neat, KBr): 2924, 2853,
1657, 1598, 1384, 1048 cm−1; 1H NMR (400 MHz, CDCl3): δ 5.72 (s,
1H), 5.38−5.29 (m, 2H), 4.04 (s, 3H), 3.79 (s, 3H), 2.42 (t, J = 7.3
J
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
with 10 μM DCF-DA for 30 min, and fixed with 4% formaldehyde for
20 min. The cells were washed with PBS twice, and fluorescence was
detected via Operetta (PerkinElmer).
Detection of Mitochondrial Membrane Potential. Cells were
seeded at 5 × 105 cells/well in a 24-well plate onto cover glass. After
18−24 h, the cells were treated with 10 mM compounds for 24 h.
Cells were then treated with 10 μM JC-1 (Sigma) for 30 min and
washed with PBS twice. Fluorescence was detected via fluorescence
microscopy (Nikon).
(4) Kim, Y. K.; Ahn, S. K.; Lee, M. Differential sensitivity of
melanoma cell lines with differing B-Raf mutational status to the new
oncogenic B-Raf kinase inhibitor UI-152. Cancer Lett. 2012, 320, 215−
224.
(5) Keating, G. M. Vemurafenib: in unresectable or metastatic
melanoma. BioDrugs 2012, 26, 325−334.
(6) Romano, E.; Pradervand, S.; Paillusson, A.; Weber, J.; Harshman,
K.; Muehlethaler, K.; Speiser, D.; Peters, S.; Rimoldi, D.; Michielin, O.
Identification of multiple mechanisms of resistance to vemurafenib in a
patient with BRAFV600E-mutated cutaneous melanoma successfully
rechallenged after progression. Clin. Cancer Res. 2013, 19, 5749−5757.
(7) Waris, G.; Ahsan, H. Reactive oxygen species: role in the
development of cancer and various chronic conditions. J. Carcinog.
2006, 5, 14.
(8) Wu, W. S. The signaling mechanism of ROS in tumor
progression. Cancer Metastasis Rev. 2006, 25, 695−705.
(9) Gorrini, C.; Harris, I. S.; Mak, T. W. Modulation of oxidative
stress as an anticancer strategy. Nat. Rev. Drug Discovery 2013, 12,
931−947.
(10) Pelicano, H.; Carney, D.; Huang, P. ROS stress in cancer cells
and therapeutic implications. Drug Resist. Updates 2004, 7, 97−110.
(11) Deavall, D. G.; Martin, E. A.; Horner, J. M.; Roberts, R. Drug-
induced oxidative stress and toxicity. J. Toxicol. 2012, 2012, 645460.
(12) Trachootham, D.; Alexandre, J.; Huang, P. Targeting cancer
cells by ROS-mediated mechanisms: a radical therapeutic approach?
Nat. Rev. Drug Discovery 2009, 8, 579−591.
(13) Raj, L.; Ide, T.; Gurkar, A. U.; Foley, M.; Schenone, M.; Li, X.;
Tolliday, N. J.; Golub, T. R.; Carr, S. A.; Shamji, A. F.; Stern, A. M.;
Mandinova, A.; Schreiber, S. L.; Lee, S. W. Selective killing of cancer
cells by a small molecule targeting the stress response to ROS. Nature
2011, 475, 231−234.
(14) Adams, D. J.; Dai, M.; Pellegrino, G.; Wagner, B. K.; Stern, A.
M.; Shamji, A. F.; Schreiber, S. L. Synthesis, cellular evaluation, and
mechanism of action of piperlongumine analogs. Proc. Natl. Acad. Sci.
U.S.A. 2012, 109, 15115−15120.
(15) Shaw, A. T.; Winslow, M. M.; Magendantz, M.; Ouyang, C.;
Dowdle, J.; Subramanian, A.; Lewis, T. A.; Maglathin, R. L.; Tolliday,
N.; Jacks, T. Selective killing of K-ras mutant cancer cells by small
molecule inducers of oxidative stress. Proc. Natl. Acad. Sci. U.S.A. 2011,
108, 8773−8778.
(16) Yuan, H.; Zhang, X.; Huang, X.; Lu, Y.; Tang, W.; Man, Y.;
Wang, S.; Xi, J.; Li, J. NADPH oxidase 2-derived reactive oxygen
species mediate FFAs-induced dysfunction and apoptosis of beta-cells
via JNK, p38 MAPK and p53 pathways. PLoS One 2010, 5, e15726.
(17) Miki, H.; Uehara, N.; Kimura, A.; Sasaki, T.; Yuri, T.; Yoshizawa,
K.; Tsubura, A. Resveratrol induces apoptosis via ROS-triggered
autophagy in human colon cancer cells. Int. J. Oncol. 2012, 40, 1020−
1028.
(18) Dharmaraja, A. T.; Jain, C.; Chakrapani, H. Substituent effects
on reactive oxygen species (ROS) generation by hydroquinones. J.
Org. Chem. 2014, 79, 9413−9417.
(19) Wardman, P. Fluorescent and luminescent probes for
measurement of oxidative and nitrosative species in cells and tissues:
progress, pitfalls, and prospects. Free Radical Biol. Med. 2007, 43, 995−
1022.
(20) Zhu, C.; Hu, W.; Wu, H.; Hu, X. No evident dose−response
relationship between cellular ROS level and its cytotoxicitya
paradoxical issue in ROS-based cancer therapy. Sci. Rep. 2014, 4, 5029.
(21) Anathy, V.; Roberson, E. C.; Guala, A. S.; Godburn, K. E.; Budd,
R. C.; Janssen-Heininger, Y. M. Redox-based regulation of apoptosis:
S-glutathionylation as a regulatory mechanism to control cell death.
Antioxid. Redox Signaling 2012, 16, 496−505.
(22) Deng, Y. T.; Huang, H. C.; Lin, J. K. Rotenone induces
apoptosis in MCF-7 human breast cancer cell-mediated ROS through
JNK and p38 signaling. Mol. Carcinog. 2010, 49, 141−151.
(23) Bragado, P.; Armesilla, A.; Silva, A.; Porras, A. Apoptosis by
cisplatin requires p53 mediated p38alpha MAPK activation through
ROS generation. Apoptosis 2007, 12, 1733−1742.
ASSOCIATED CONTENT
* Supporting Information
■
S
Protocols for biochemical and cellular assays. Figure S1:
Measurement of cellular GSH and GSSG levels in A549 cells
following compound treatment for 4 h. Figure S2: FACS
analysis of apoptosis in A549 and HCT-116 cells after
treatment of 26 for 48 h. Figure S3: Soft agar assay using
A549 cells. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
■
*(W.H.) E-mail: whur@kist.re.kr. Phone: +82-2-958-5193.
*(T.S.) E-mail: tbsim@kist.re.kr, tbsim@korea.ac.kr. Phone:
+82-2-958-6437.
Author Contributions
§Y.H. and S.S. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Korea Institute of Science and
Technology (KIST), the Creative/Challenging Research
Program (2011-0028676) and Global Research Network
Research Program (NRF-220-2011-1-C00042) of the National
Research Foundation of Korea (NRF), and a grant (D33400)
from the Korea Basic Science Institute. We also thank Dr. Yun
Kyung Kim for assistance in Operetta operation and data
analysis.
ABBREVIATIONS USED
■
ROS, reactive oxygen species; NAC, N-acetyl-cysteine; AA,
ascorbic acid; SAR, structure−activity relationship; FACS,
fluorescence-activated cell sorting; DCF-DA, 2′,7′-dichloro-
fluorescein diacetate; APF, 3′-(p-aminophenyl) fluorescein;
DHE, dihydroethidium; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl tetrazolium bromide; JC-1, 5,5′,6,6′-tetrachloro-
1,1′,3,3′-tetraethylbenzimidazolocarbocyanine iodide; EC50,
effective concentration 50%; GSH, glutathione; NBS, N-
bromosuccinimide; TLC, thin-layer chromatography; PCC,
pyridinium chlorochromate; DMF, N,N-dimethylformamide;
MBT, mercaptobenzothiazole
REFERENCES
■
(1) Mann, J. Natural products in cancer chemotherapy: past, present
and future. Nat. Rev. Cancer 2002, 2, 143−148.
(2) Newman, D. J.; Cragg, G. M. Natural products as sources of new
drugs over the 30 years from 1981 to 2010. J. Nat. Prod 2012, 75,
311−335.
(3) Choi, S. W.; Choi, C. H.; Ryu, S. Y.; Kim, Y. S.; Cha, M. R.; Kim,
M. J.; Jeon, J. H.; Kim, Y. K. Anti-cancer substances containing
resveratrol derivatives isolated from the seeds of Iris pseudacorus. Korea
Patent 10-2012-0119693, October 31, 2012.
K
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(24) Wang, X.; Liu, J. Z.; Hu, J. X.; Wu, H.; Li, Y. L.; Chen, H. L.;
Bai, H.; Hai, C. X. ROS-activated p38 MAPK/ERK−Akt cascade plays
a central role in palmitic acid-stimulated hepatocyte proliferation. Free
Radical Biol. Med. 2011, 51, 539−551.
(25) Mao, X.; Yu, C. R.; Li, W. H.; Li, W. X. Induction of apoptosis
by shikonin through a ROS/JNK-mediated process in Bcr/Abl-positive
chronic myelogenous leukemia (CML) cells. Cell Res. 2008, 18, 879−
888.
(26) Dolado, I.; Nebreda, A. R. AKT and oxidative stress team up to
kill cancer cells. Cancer Cell 2008, 14, 427−429.
(27) Esposito, F.; Chirico, G.; Montesano Gesualdi, N.; Posadas, I.;
Ammendola, R.; Russo, T.; Cirino, G.; Cimino, F. Protein kinase B
activation by reactive oxygen species is independent of tyrosine kinase
receptor phosphorylation and requires SRC activity. J. Biol. Chem.
2003, 278, 20828−20834.
(28) Zhuang, S.; Schnellmann, R. G. H2O2-induced transactivation of
EGF receptor requires Src and mediates ERK1/2, but not Akt,
activation in renal cells. Am. J. Physiol.: Renal Physiol. 2004, 286,
F858−F865.
(29) Ponnusamy, M.; Pang, M.; Annamaraju, P. K.; Zhang, Z.; Gong,
R.; Chin, Y. E.; Zhuang, S. Transglutaminase-1 protects renal epithelial
cells from hydrogen peroxide-induced apoptosis through activation of
STAT3 and AKT signaling pathways. Am. J. Physiol.: Renal Physiol.
2009, 297, F1361−F1370.
(30) Perelman, A.; Wachtel, C.; Cohen, M.; Haupt, S.; Shapiro, H.;
Tzur, A. JC-1: alternative excitation wavelengths facilitate mitochon-
drial membrane potential cytometry. Cell Death Dis. 2012, 3, e430.
L
DOI: 10.1021/jm501846y
J. Med. Chem. XXXX, XXX, XXX−XXX