Bioassay guided identification of small chaperone proteins α-crystallin and Hsp27 inhibitors from Copaiba oil

Article abstract of DOI:10.1016/j.phytol.2014.08.006

Over-expression of small chaperone proteins in cancer tissue contributes to the resistance of chemotherapy. Targeting these small chaperones including α-crystallin and heat shock protein 27 (Hsp27) is a promising strategy for cancer treatment. Hardwickiic acid (HA), a clerodane diterpenoid from Copaiba oil has been reported to inhibit Hsp27. We expect to identify new small chaperone inhibitors from Copaiba oil that is abundant of diterpenoids. In the current study, cytotoxicity and anti-chaperone assay guided isolation of Copaiba oil led to two major fractions (non-acidic and acidic components) and seven sub-fractions from the acidic components. The non-acidic components and one sub-fraction showed significant cytotoxicity in prostate cancer cells. Four sub-fractions exhibited potent anti-chaperone activity. Three chemical components were identified from these sub-fractions including copalic acid, hardwickiic acid and 3-acetoxycopalic acid. All three compounds inhibited the chaperone activities of α-crystallin and Hsp27. In addition, these compounds enhanced the anti-proliferative activity of the chemotherapeutic agent carboplatin in LNCaP cells. 3-Acetoxycopalic acid slightly decreased the level of Hsp27 client protein signal transducer and activator of transcription 3 (Stat3) in PC3 cells. Overall, several diterpenoids were identified to be small chaperone inhibitors and could be used as lead compounds for the development of more potent derivatives.

Full text of DOI:10.1016/j.phytol.2014.08.006

Phytochemistry Letters 10 (2014) 65–75
Contents lists available at ScienceDirect
Phytochemistry Letters
journal homepage: www.elsevier.com/locate/phytol
Bioassay guided identification of small chaperone proteins
and Hsp27 inhibitors from Copaiba oil
a-crystallin
Rati Lama a, Bo Zhong a, Daniel G. Kulman a, Bin Su a,b,
*
a Department of Chemistry, College of Sciences and Health Professions, Cleveland State University, 2121 Euclid Ave., Cleveland, OH 44115, USA
b Center for Gene Regulation in Health and Disease, College of Sciences & Health Professions, Cleveland State University, 2121 Euclid Ave., Cleveland,
OH 44115, USA
A R T I C L E I N F O
A B S T R A C T
Article history:
Over-expression of small chaperone proteins in cancer tissue contributes to the resistance of
Received 7 February 2014
Received in revised form 22 May 2014
Accepted 5 August 2014
Available online 17 August 2014
chemotherapy. Targeting these small chaperones including
a-crystallin and heat shock protein 27
(Hsp27) is a promising strategy for cancer treatment. Hardwickiic acid (HA), a clerodane diterpenoid
from Copaiba oil has been reported to inhibit Hsp27. We expect to identify new small chaperone
inhibitors from Copaiba oil that is abundant of diterpenoids. In the current study, cytotoxicity and anti-
chaperone assay guided isolation of Copaiba oil led to two major fractions (non-acidic and acidic
components) and seven sub-fractions from the acidic components. The non-acidic components and one
sub-fraction showed significant cytotoxicity in prostate cancer cells. Four sub-fractions exhibited potent
anti-chaperone activity. Three chemical components were identified from these sub-fractions including
copalic acid, hardwickiic acid and 3-acetoxycopalic acid. All three compounds inhibited the chaperone
Keywords:
Copaiba oil
Small chaperone
a-Crystallin
Hsp27
Cancer
activities of
a-crystallin and Hsp27. In addition, these compounds enhanced the anti-proliferative
activity of the chemotherapeutic agent carboplatin in LNCaP cells. 3-Acetoxycopalic acid slightly
decreased the level of Hsp27 client protein signal transducer and activator of transcription 3 (Stat3) in
PC3 cells. Overall, several diterpenoids were identified to be small chaperone inhibitors and could be
used as lead compounds for the development of more potent derivatives.
Published by Elsevier Ireland Ltd on behalf of Phytochemical Society of Europe.
1. Introduction
et al., 2013). In recent years, studies of Copaiba oil and its
components have revealed their potential anti-cancer activity. The
Copaiba oil, which is obtained from the trees of the Copaifera
species, is composed of a mixture of sesquiterpenes and diterpenes
(Leandro et al., 2012). More than 75% of the composition of Copaiba
oil was found to be sesquiterpens. Most sesquiterpens present in
Copaiba oil are hydrocarbons, alcohols and epoxides. Diterpenes,
the minor components in Copaiba oil, were identified to be various
diterpenic carboxylic acids including copalic acid, ent-agathic acid,
kaurenoic acid, hardwickiic acid and others (Leandro et al., 2012;
Soares et al., 2013; Souza Barbosa et al., 2013). Copaiba oil is of
interest because of its wide application in cosmetics and
pharmaceutical industries. Particularly, it has been used as a
traditional medicine for wound healing and general inflammation
(Estevao et al., 2013; Leandro et al., 2012; Pieri et al., 2012; Santos
Copaiba oil from C. multijuga species and its fraction have been
reported to significantly inhibit tumor growth in mice (Lima et al.,
2003). Sesquiterpenes such as
b-caryophyllene and b-elemene
that are the major components of Copaiba oil showed anti-
proliferative activity against several cancer cell lines. In particular,
b
-caryophyllene exhibited synergistic effects by increasing anti-
cancer activity of the chemotherapeutic agent paclitaxel (Leandro
et al., 2012; Legault and Pichette, 2007; Lima et al., 2003). As
another important class of natural products detected in Copaiba
oil, diterpenic acids mainly show antibacterial activities (Leandro
et al., 2012; Pieri et al., 2012; Santos et al., 2012, 2013).
Interestingly, a recent study by Feiella, etc. revealed that hard-
wickiic acid, one of the diterpenic acids existing in Copaiba oil,
inhibits heat shock protein 27 (Hsp27) chaperone activity (Faiella
et al., 2012).
The over-expression of Hsp27 in cancer tissues is correlated
with anti-cancer drug resistance, which makes it a promising
molecular target for drug development (Andrieu et al., 2010; Arts
et al., 1999; Concannon et al., 2003; Garrido et al., 2006). Hsp27
*
Corresponding author at: Department of Chemistry, College of Sciences and
Health Professions, Cleveland State University, 2121 Euclid Ave., Cleveland, OH
44115, USA. Tel.: +1 216 687 9219; fax: +1 216 687 9298.
E-mail addresses: B.su@csuohio.edu, subin624@hotmail.com (B. Su).
http://dx.doi.org/10.1016/j.phytol.2014.08.006
1874-3900/Published by Elsevier Ireland Ltd on behalf of Phytochemical Society of Europe.

66
R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
15
14
13
O
O
14
11
20
13
O
15
11
20
16
17
H
15
OH
1
16
16
17
13
12
17
H
20
5
11
OH
1
9
1
9
3
5
9
7
5
O
19
3
7
H
3
7
O
H
18
21
19
18
19
CO2H
Hardwickiic acid (HA)
22
3-Acetoxycopalic acid (ACA)
Copalic acid (CA)
Fig. 1. Structures of hardwickiic acid (HA), copalic acid (CA), and 3-acetoxycopalic acid (ACA).
blocks the apoptosis of cancer cells induced by anti-cancer agents
through its molecular chaperone functions. Hsp27 directly
interacts with its client proteins affecting multiple steps in
apoptosis, including cytochrome c release, formation of apopto-
some complex, and caspase-activation (Andrieu et al., 2010; Bruey
et al., 2000; Concannon et al., 2003; Garrido et al., 1999). Preclinical
studies have reported that antisense oligonucleotides (ASO) or
siRNA decrease Hsp27 expression to suppress tumor growth
(Baylot et al., 2011; Cayado-Gutierrez et al., 2013). However, these
strategies are associated with difficulties in drug delivery. Small
molecule Hsp27 inhibitors without administration problems will
be more desirable in cancer treatment. In addition to Hsp27,
chromatography was performed on silica gel TLC plates with
fluorescence indicator 254 nm (Fluka). Flash column chromatog-
˚
raphy was performed using silica gel 60 A (BDH, 40–63
mM). Mass
spectra were obtained on the Bruker ion-trap mass spectrometer at
Cleveland State University MS facility Center. All the 1H NMR
spectra were recorded on a Bruker 400 MHz spectrometer using
CDCl3 as solvent. Chemical shifts (d
) for 1H NMR spectra were
reported in parts per million to residual solvent protons.
2.1.1. General procedure for fractionation of crude Copaiba oil
Crude Copaiba oil (2.3 g) in ethyl ether (10 ml) was extracted
twice with aqueous 5% KOH (10 ml). The organic layer was washed
with water and dried over anhydrous sodium sulfate, then
concentrated under vacuum to give non-acidic fraction (B) with
a yield around 65%. The combined aqueous layers were neutralized
with 3 N HCl and then extracted with ethyl ether. The organic
layers were washed with water and dried over anhydrous sodium
sulfate, then concentrated under vacuum to give acidic fraction (C)
with yield around 28%. 0.66 g Fraction C was subjected to column
chromatograph using gradient elution with pure hexane, hexane/
ethyl acetate (95:5), hexane/ethyl acetate (90:10), hexane/ethyl
acetate (80:20), to give seven subfractions (D–J) with yields of 39%,
21%, 8%, 5%, 5%, 12%, 4% respectively. Fraction H contains one single
another small chaperone protein,
play a cytoprotective role in apoptotic cancer cells as well (Arrigo
et al., 2007; Hamann et al., 2013). -Crystallin targeting agents
may sensitize cancer cells to apoptosis inducing agents. Overall,
a-crystallin has been reported to
a
small chaperone inhibitors are expected to have
application in cancer treatment.
a broad
The discovery of Hsp27 inhibitory activity of hardwickiic acid
raised our interest in Copaiba oil, specifically to its diterpene
components (Faiella et al., 2012). The inhibition of small chaperone
protein by hardwickiic acid may contribute to the anti-cancer
activity of Copaiba oil. We speculate that other diterpenic acids in
Copaiba oil might interfere with small chaperone proteins as well.
Since the diterpene components share similar structures. On the
other hand, small chaperones have been identified as novel anti-
cancer targets (Zoubeidi and Gleave, 2012). However, so far very
few inhibitors of the small chaperones have been discovered.
Copaiba oil may be a good recourse to isolate lead compounds that
inhibit small chaperone proteins. To elucidate the anti-cancer
mechanism of Copaiba oil and search for small molecule chaperone
protein inhibitors, the oil resin was initially separated into two
major fractions including sesquiterpenes and diterpenic acids. The
diterpenic acid components were further fractionated. The two
major fractions from Copaiba oil and the seven sub-fractions from
diterpenes were examined for their potential anti-cancer and anti-
small chaperone activities. Four sub-fractions showed potent anti-
small chaperone activities. Further isolation of the individual
components in the active sub-fractions followed by structural
elucidation and biological activity examination led to the
identification of several small chaperone inhibitors including,
copalic acid, hardwickiic acid and 3-acetoxycopalic acid (Fig. 1).
component, which is 3-acetoxycopalic acid. 1H NMR
d5.689 (1H, s),
4.899 (1H, s), 4.543 (2H, m), 2.433 (1H, m), 2.341 (1H, m), 2.188
(3H, s), 2.074 (3H, s), 2.014 (2H, m), 1.824–1.742 (3H, m), 1.704–
1.505 (4H, m), 1.423 (1H, m), 1.285 (1H, m), 1.200 (1H, m), 0.896
(3H, s), 0.870 (3H, s), 0.739 (3H, s); ESI-MS calculated for C22H33O4
[MÀH]À 361.2, found 361.2.
2.1.2. General procedure for isolation of copalic acid and hardwickiic
acid
To a solution of fraction E (0.436 g), 4-methoxyphenol (0.178 g,
1.434 mmol), dicyclohexyl carbodiimide (0.325 g, 1.577 mmol) in
dichloromethane was added 4-dimethylaminopyridine (0.192 g,
1.577 mmol) and stirred at room temperature for 4 h. The
insoluble solid was filtered. The filtrate was concentrated under
vacuum and the residue was purified by column chromatograph to
yield 0.36 g 4-methoxyphenol ester of copalic acid. 0.2 g 4-
methoxyphenol ester of copalic acid in the mixture of 3 N aqueous
NaOH and ethanol (1:1) was refluxed for 1 h. The reaction mixture
was neutralized by addition of 3 N HCl and subsequently extracted
with ethyl acetate. The organic layer was concentrated under
vacuum followed by purification by column chromatograph to
2. Materials and methods
yield 0.12 g copalic acid. 1H NMR
d 5.697 (1H, s), 4.873 (1H, s),
4.516 (1H, s), 2.414 (1H, m), 2.344 (1H, m), 2.195 (3H, s), 2.054–
1.955 (2H, m), 1.782–1.700 (2H, m), 1.636–1.410 (6H, m), 1.344
(1H, m), 1.207 (1H, m), 1.110 (1H, s), 1.022 (1H, m), 0.896 (3H, s),
0.826 (3H, s), 0.708 (3H, s); ESI-MS calculated for C20H31O2 [MÀH]À
303.2, found 303.4.
2.1. Chemistry
Crude Copaiba oil was manufactured by Natural Joint Solutions
Inc. (FL). Human recombinant insulin and a-crystallin from bovine
eye lens were from Sigma–Aldrich. Human recombinant Hsp27
were from Biosciences. Other commercial chemicals and solvents
were used as received without further purification. Thin layer
Isolation of hardwickiic acid from fraction F followed the
similar procedure with a yield close to 15%. 1H NMR
d 7.373 (1H,

R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
67
m), 7.227 (1H, s), 6.884 (1H, m), 6.281 (1H, s), 2.458 (1H, m), 2.398–
2.316 (2H, m), 2.270–2.163 (2H, m), 1.746–1.397 (9H, m), 1.285
(3H, s), 1.198 (1H, m), 0.859 (3H, d, J = 6.8 Hz), 0.789 (3H, s). ESI-MS
calculated for C20H27O3 [MÀH]À 315.2, found 315.2.
for 24 h, and were exposed to fractions or compounds dissolved in
DMSO (final concentration ꢀ0.1%) in media for 72 h. Controls
received DMSO vehicle at a concentration equal to that in fractions
or compounds-treated cells. The medium was removed, replaced
by 200
mL of 0.5 mg/mL of 3-(4,5-dimethylthiazol-2-yl)-2,5-
2.2. Cell culture
diphenyl-2H-tetrazolium bromide in fresh media, and cells were
incubated in the CO2 incubator at 37 8C for 2 h. Supernatants were
removed from the wells, and the reduced 3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyl-2H-tetrazolium bromide dye was solubilized in
PC3 and LNCaP human prostate cancer cell lines were obtained
from ATCC (Rockville, MD). The cells were maintained in RPMI1640
medium supplemented with 10% fetal bovine serum (FBS),
200
mL/well DMSO. Absorbance at 570 nm was determined on a
2 mmol/L
L
-glutamine, and 100 U/mL penicillin-streptomycin.
plate reader. Statistical and graphical information was determined
using GraphPad Prism software (GraphPad Software Incorporated)
and Microsoft Excel (Microsoft Corporation). IC50 values were
determined using nonlinear regression analysis.
FBS was heat inactivated for 30 min in a 56 8C water bath before
use. Cell cultures were grown at 37 8C, in a humidified atmosphere
of 5% CO2 in a VWR CO2 incubator (Bridgeport, NJ).
2.3. Anti-chaperone assay
2.5. Western blot
2.3.1. Endpoint measurement
PC3 and LNCaP cells were treated with three diterpenic acids
respectively for 24 h. The cells were lysed, briefly sonicated, and
A mixture of 0.24 mg/mL human recombinant insulin, 150
mg/
mL
a
-crystallin, and crude Copaiba oil (A) or non-acidic
centrifuged at 16,000 Â g for 30 min. 30
mg of protein for each
components (B) or acidic components (C) or subfractions (D–J)
or copalic acid or hardwickiic acid or 3-acetoxycopalic acid in
sample was boiled with 2Â loading buffer for 5 min, electrophor-
esed on a 10% SDS-polyacrylamide gel, and transferred onto
polyvinylidene difluoride (PVDF) membrane. The membrane was
blocked for 1 h with 5% nonfat milk in PBST and then incubated
with specific primary antibody (Cell Signaling). After washing, the
membrane was incubated with horseradish-conjugated secondary
antibody (Cell Signaling). The bands were visualized by chemilu-
minescence with ECL reagent (Thermo Scientific).
98
m
L sodium phosphate buffer, pH 7.4, was incubated at 37 8C for
L of 1 M DTT in the water was added to
5 min, whereupon 2
m
initiate the insulin aggregation. The absorbance (A) at 400 nm was
recorded after 45 min incubation at 37 8C in a plate reader
(molecule device). A mixture of 0.24 mg/mL insulin in the absence
or presence of 150
mg/mL a-crystallin with DMSO was used as
control. For evaluation of effects of crude Copaiba oil (A) and non-
acidic and acidic components (B–C) and sub-fractions (D–J) on
DTT-induced insulin aggregation, a mixture of 0.24 mg/mL insulin
and 20 mM DTT in the presence of DMSO or crude Copaiba oil (A) or
non-acidic components (B) or acidic components (C) or individual
subfractions (D–J) was incubated for 45 min at 37 8C followed by
the measurement of the absorbance (A) at 400 nm.
3. Results and discussion
3.1. Fractionation of diterpenic acid components in Copaiba oil via
flash column chromatography
We focused on diterpenic acid components in Copaiba oil to
search for potential small chaperone inhibitors. Flash column
chromatograph was used to separate diterpenic acid components
into several main portions. This was followed by further
purification of the major components in these fractions. As showed
in Scheme 1, partitioning of crude Copaiba oil in Et2O and 5% aq.
KOH enabled the separation of diterpene components from the
sesquiterpene components. Then the non-acidic components (B),
sesquiterpenes were dissolved in organic phase, while diterpenic
carboxylic acids (C) were remained in basic aqueous phase. The
separation of the acidic mixture by column chromatograph
produced seven fractions (D–J). Each fraction was subjected to
ESI–MS and 1H NMR analysis to determine the major components.
Chemical compositions of fractions D, G, H, and I were complex.
Only fraction H consisted of one single component that gave
molecular ions at m/z 361.2 [MÀH]À and m/z 723.4 [2MÀH]À in the
mass spectrum (MS). The MS data of fraction H matches the
molecular weight of 3-acetoxycopalic acid that was reported to be
an abundant constituent of Copaiba oil. Analysis of its 1H NMR data
confirmed fraction H to be 3-acetoxycopalic acid (Cavin et al.,
2006). The predominant molecular ions at m/z 303.4 [MÀH]À and
m/z 608.0 [2MÀH]À for fraction E indicated that this fraction
mainly contains copalic acid (Cavin et al., 2006). Additionally, the
crude 1H NMR data of fraction E has resonance for three protons,
one at 5.70 ppm and the other two at 4.87 and 4.52 ppm, which are
consistent with the signals reported for vinylic protons of copalic
acid (one proton at C14 and two protons at C17). Fraction F gave
major molecular ions at m/z 315.2 [MÀH]À and m/z 631.1
[2MÀH]À, suggesting the existence of hardwickiic acid (Braun
2.3.2. Kinetic measurement
A mixture of 0.24 mg/mL human recombinant insulin, 150
-crystallin, and copalic acid or hardwickiic acid or 3-
acetoxycopalic acid in 98 L sodium phosphate buffer, pH 7.4,
was incubated at 37 8C for 5 min, whereupon 2 L of 1 M DTT in
mg/
mL
a
m
m
the water was added to initiate the insulin aggregation. The
absorbance (A) at 400 nm was monitored over 45 min using a
plate reader. A mixture of 0.24 mg/mL insulin in the absence or
presence of 150
control. The -crystallin inhibition potency (%) of compounds at
45 min were determined by A(a-crystallin+compound+DTT) À A(a-crystal-
lin+DTT)/ADTT À A(a-crystallin+DTT) where A(a-crystallin+compound+DTT)
A(a-crystallin+DTT) represents an increase of insulin aggregation level
in the presence of -crystallin with compounds compared to
that in the presence of -crystallin without compounds and
ADTT À A(a-crystallin+DTT) represented a decrease of insulin aggrega-
tion level caused by -crystallin.
mg/mL a-crystallin with DMSO was used as
a
,
À
a
a
a
Evaluation of the Hsp27 inhibition potency of copalic acid and
hardwickiic acid and 3-acetoxycopalic acid followed the same
procedure with the exception of using Hsp27 instead of
crystallin. The final concentration of Hsp27 in the assay buffer was
15 g/mL.
a-
m
2.4. Cell viability analysis
The effects of crude Copaiba oil and its fractions and three
diterpenic acids on human prostate cancer cells PC3 and LNCaP
viability were assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl-2H-tetrazolium bromide assay in six replicates. Cells
were grown in RPMI1640 medium in 96-well, flat-bottomed plates
and Breitenbach, 1977; Costa et al., 1999). The 1H NMR signals at
7.37 and 7.23 and 6.28 are attributed to the protons at the furan
ring of hardwickiic acid (Braun and Breitenbach, 1977; Costa et al.,
d

68
R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
Scheme 1. Separation and purification of three diterpenic carboxylic acids from Copaiba oil.
1999). Another proton signal at
d
6.88 is the conjugated vinylic
chaperone proteins.
functional architecture
oligomeric assembly and chaperone functions of both proteins.
Moreover, the three-dimensional structures of -crystallin-
domains in both the proteins are highly overlapped (Fig. 2). This
evidence indicates that -crystallin and Hsp27 may be inhibited by
the same inhibitors. The two small chaperone proteins show strong
similarity and both contribute to cancer drug resistance (Arrigo
et al., 2007). We expect to identify new lead compounds that target
these small chaperones. As shown in Fig. 3, dithiothreitol (DTT)-
a
-Crystallin and Hsp 27 share a common
proton. The crude 1H NMR of fraction F also indicated the existence
of copalic acid. The ratio of hardwickiic acid and copalic acid in
fraction F was about 2:1.
a
-crystallin-domain that mediates the
a
3.2. Screening of Copaiba oil (A), non-acidic and acidic components (B
and C), and seven acidic sub-fractions (D-J) with chaperone assay and
cell proliferation assay
a
The crude Copaiba oil (A), non-acidic and acidic components (B
and C), and seven acidic sub-fractions (D–J), were evaluated for
their potency to inhibit in vitro chaperone functions of small
induced insulin aggregation in the presence of
about 50% less than that in the absence of -crystallin, which is
attributed to the chaperone function of -crystallin in the insulin
a-crystallin was
a
a
aggregation. Hsp27 showed more potent chaperone activity in the
same assay. We initiated the chaperone inhibition screening with
a
-crystallin. The crude Copaiba oil (A) and acidic sub-fractions (E–
H) showed inhibitory activity against -crystallin chaperone
function. More specifically, at 3.5 g/mL the crude Copaiba oil
(A) inhibited -crystallin function by 20% while the acidic sub-
fractions E, F, G and H exhibited more potent inhibition of
a
m
a
a
-
crystallin function by 41%, 34%, 46%, and 47% respectively. The
non-acidic components (B), acidic components (C), and acidic sub-
fractions (D and J) did not show significant inhibition of the
chaperone function of
insulin aggregation in the presence of
fraction (I) was significantly less than that in the presence of
crystallin without the sub-fraction (I), suggesting that sub-fraction
(I) can enhance the chaperone function of -crystallin. Moreover,
all the fractions and the whole oil were evaluated at the
concentration of 3.5 g/mL in the absence of -crystallin in this
a
-crystallin. Unexpectedly, DTT-induced
a
-crystallin and the sub-
a
-
a
m
a
assay to rule out the non-specific effects. The results revealed that
they exhibited no direct effects on DTT-induced insulin aggrega-
tion (Fig. 3 (2)).
The crude Copaiba oil and its fractions were also tested for
their anti-proliferative activity against LNCaP and PC3 prostate
cancer cell lines (Fig. 4). Fractions B and D showed the most potent
Fig. 2. Overlap of the crystal structures of
and -crystallin (green). (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of the article.)
a-crystallin domains in Hsp27 (magenta)
a

R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
69
(1)
esterification with dimethyl sulfate or diazomethane, which is
followed by isolation as methyl ester derivatives and then ester
hydrolysis (Costa et al., 1999). However, hydrolysis of hardwickiic
acid methyl ester proved to be difficult. In this study, we isolated
copalic acid and hardwickiic acid as 4-methoxyphenol ester
derivatives. Fractions E or F reacted with 4-methoxylphenol and
generated 4-methoxylphenol esters. After separation of the esters
by flash column chromatograph, the cleavage of the 4-methox-
yphenoxy group in 3 N NaOH/MeOH generated pure copalic acid
and hardwickiic acid (Scheme 1). The purification of sub-fraction G
by this method did not work, which could be due to the complex
components in the mixture.
0.30
*
0.25
0.20
0.15
0.10
*
*
*
3.4. Dose-dependent inhibition of small chaperones and cytotoxicity
of the three diterpenic acids
In order to confirm the inhibition of
hardwickiic acid, and 3-acetoxycopalic acid, the kinetics of DTT-
induced insulin aggregation was monitored in the presence of
crystallin with or without these three diterpenic acids. As shown
in Fig. 5 (1), at a concentration of 3.5 g/mL, copalic acid,
hardwickiic acid and 3-acetoxycopalic acid exhibited potent
inhibition of -crystallin by 85% and 38% and 45% respectively
(corresponding micromole concentration of the three compounds
are listed as well). The results are consistent with the -crystallin
a-crystallin by copalic acid,
Insulin
DTT
a-crystallin
Compounds
+
+
-
-
+
+
+
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
E
+
+
+
F
+
+
+
+
+
+
+
+
+
I
+
+
+
J
a
-
A
B
C
D
G
H
m
(2)
0.35
a
0.30
0.25
0.20
0.15
0.10
0.05
a
inhibitory activity showed by sub-fractions E, F and H. These three
diterpenic acids at the same concentration did not interfere with
DTT-induced insulin aggregation (data not shown). Inhibition of
crystallin by three diterpenic acids at three different concentra-
tions was further evaluated. All three diterpenic acids inhibit
a-
a
-
crystallin in a dose-dependent manner in Fig. 5 (2)–(4). To the best
of our knowledge, this is the first time that the three diterpenic
acids were identified to be small chaperone a-crystallin inhibitors.
Copalic acid, hardwickiic acid and 3-acetoxycopalic acid were
further evaluated for the inhibition of Hsp27. These three
diterpenic acids at 3.5
mg/mL inhibited Hsp27 by 38%, 29%, and
0.00
Insulin
DTT
20% respectively (Fig. 6). Inhibition of chaperone function of Hsp27
by hardwickiic acid has been reported (Faiella et al., 2012).
However, our results firstly revealed copalic acid and 3-
acetoxycopalic acid as inhibitors of Hsp27. All the three diterpenic
acids are less potent for inhibition of Hsp27 compared with the
+
+
+
+
A
+
+
B
+
+
C
+
+
D
+
+
E
+
+
F
+
+
G
+
+
H
+
+
I
+
+
J
Compounds
-
Fig. 3.
crude Copaiba oil and its fractions. (1) DTT-induced insulin aggregation in the
absence of -crystallin, or in the presence of -crystallin, or in the presence of
crystallin with crude Copaiba oil (A) or fractions (B and C) or sub-fractions (D–J).
*p < 0.01 fraction treatment vs vehicle in the presence of -crystallin; (2) DTT-
a-Crystallin prevents DTT induced insulin aggregation in the presence of
inhibition of
chaperone potency of Hsp27 and
a
-crystallin. The results also demonstrated different
-crystallin for inhibition of DTT-
a
a
a-
a
induced insulin aggregation. In both the assays, the concentrations
of insulin and DTT were identical. The final concentration of Hsp27
a
induced insulin aggregation in the presence of crude Copaiba oil (A) or fractions (B
and C) or sub-fractions (D–J). The mixture of insulin and DTT with or without other
components in the assay buffer was incubated for 45 min at 37 8C and the
absorbance at 400 nm was measured. Each bar represents insulin aggregation level
after 45 min incubation at different conditions.
in the assay buffer was 15
m
g/mL, while
a
-crystallin was 150
mg/
mL. As shown in Figs. 5 and 6, 15
m
g/mL Hsp27 and 150
m
g/mL a-
crystallin inhibited DTT-induced insulin aggregation at similar
level, suggesting Hsp27 as a small chaperone is much more potent
than
Since copalic acid, hardwickiic acid and 3-acetoxycopalic acid
inhibited small chaperones Hsp27 and -crystallin, the three
diterpenic acids may block cancer cell growth. Therefore, the anti-
proliferative activities of these diterpenic acids were evaluated
with two prostate cancer cell lines LNCaP and PC3. As shown in
Fig. 7, the diterpenic acids showed weak growth inhibition of both
cell lines. Hardwickiic acid as an anti-proliferative agent is more
potent than the other two diterpenic acids. It inhibited cell growth
a-crystallin.
anti-proliferative activity at 175
and D at three different concentrations (175, 35, and 7
showed dose-dependent growth inhibition of both cell lines.
However, both fractions only weakly inhibited the cell growth at
m
g/mL. Further evaluation of B
g/mL)
a
m
35
mg/mL, suggesting that the Copaiba oil fractions are not potent
cytotoxic agents.
3.3. Isolation of three diterpenic acids from the fractions that inhibited
by 95% for LNCaP and 75% for PC3 at the concentration of 175 mg/
a
-crystalline
mL. It is difficult to conclude whether the anti-proliferative
activities of the three diterpenic acids are due to their chaperone
inhibition, since the drug concentration in the cell proliferation
assay is much higher than it in the chaperone activity assay. More
direct evidence is needed to confirm that the inhibition of small
chaperones by the three diterpenic acids is correlated to their
cytotoxicity.
The promising
a-crystallin inhibition of sub-fractions (E–H)
prompted us to isolate the individual active components and
confirm their biological activity. Fraction H contains a single
component and fraction G has a complex composition. The general
procedure for separation of acidic components in Copaiba oil is

70
R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
Fig. 4. Cytotoxicity evaluation of Copaiba oil (A), non-acidic and acidic components (B and C), and seven acidic sub-fractions (D–J) against LNCaP and PC3 cells. (1) LNCaP cells
were treated with DMSO or 175 g/mL of A–J for 72 h, and cell viability was determined with MTT assay. The cytotoxicity of B and D was further evaluated at three different
concentrations. (2) The similar cytotoxicity evaluation against PC3 cells.
m

R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
71
Fig. 5.
a-Crystallin prevent DTT induced insulin aggregation in the presence of diterpenic acids. (1) The kinetics of the DTT-induced insulin aggregation was monitored in the
absence of
a
-Crystallin, or in the presence of -crystallin without or with diterpenic acids; (2–4) dose-dependent inhibition of -crystallin by diterpenic acids. Each bar
a
a
represents insulin aggregation level after 45 min incubation at different conditions.
Insulin + DTT
Insulin + HSP27 + DTT
3.5. The small chaperone inhibitors enhanced the cytotoxicity of
chemotherapeutic agents in prostate cancer cells
Insulin + HSP27 + 3.5 µg/ml (12 µM) copalic acid + DTT
Insulin + HSP27 + 3.5 µg/ml (11 µM) hardwickiic acid + DTT
0.35
It has been demonstrated that Hsp27 inhibitors can potentiate
the cell growth inhibitory activity of other chemotherapeutic
agents (Zhong et al., 2013). In this study, we investigated the
potential synergistic effects of these three diterpenic acids in
combination with known chemotherapeutic agents. Three repre-
sentative chemotherapeutic agents including Carboplatin, Vin-
blastine, and Camptothecin were used. Carboplatin is a direct DNA
interfering agent. Vinblastine is a tubulin inhibitor that inhibits the
assembly of microtubules. Camptothecin is a topoisomerase I
inhibitor. The treatment of LNCaP cells with Carboplatin of
Insulin + HSP27 + 3.5 µg/ml (10 µM) acetoxycopalic acid + DTT
0.30
0.25
0.20
0.15
0.10
0.05
0.00
different doses in the absence or presence of 35
mg/mL of
individual diterpenic acid showed that all three diterpenic acids
can promote the activity of carboplatin (Fig. 8A). As shown in
Fig. 8B, 3-acetoxycarpalic acid can also potentiate the activity of
Vinblastine against LNCaP cells. However, the combination of
vinblastine with copalic acid and hardwickiic acid did not show
significant synergistic effects. These three diterpenic acids did not
potentiate either camptothecin against LNCaP cells or the other
chemotherapeutic agents against PC3 cells (Fig. 8C–F). The results
suggest that the three diterpenic acids selectively potentiate anti-
cancer activities of some chemotherapeutic agents. The synergistic
0
10
20
30
40
Time (min)
Fig. 6. Hsp27 prevents DTT induced insulin aggregation in the presence of
diterpenic acids. The kinetics of the DTT-induced insulin aggregation was
monitored in the absence of Hsp27, or in the presence of Hsp27 without or with
diterpenic acids.

72
R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
Fig. 7. Cytotoxicity evaluation of the three identified diterpenic acids against LNCaP and PC3 cells. (1) LNCaP cells were treated with DMSO or various concentrations of copalic
acid (CA), hardwickiic acid (HA) or 3-acetoxycopalic acid (ACA) for 72 h, and cell viability was determined with MTT assay. (2) The similar cytotoxicity evaluation against PC3
cells.

R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
73
Fig. 8. Diterpenic acids and chemotherapeutic agents combination treatment on prostate cancer cell proliferation. (A–C) LNCaP cells were treated with three different
chemotherapeutic agents of different concentrations in the absence or presence of 35 g/mL individual diterpenic acid copalic acid (CA), hardwickiic acid (HA) or 3-
m
acetoxycopalic acid (ACA). After 72 h, cell viability was evaluated. (D–F) The similar combination treatment on PC3 cells.
effects could be affected by various factors such as mechanism of
the chemotherapeutic agents, cancer cell lines, and the cell
membrane permeability of the diterpenic acids.
3.6. Diterpenic acids slightly affected the Hsp27 client protein
Mammalian Hsp27 can form oligomeric complexes ranging
from 100 to 800 kDa (Garrido, 2002). This highly dynamic process
is regulated by the phosphorylation status of the protein Hsp27
(Garrido, 2002; Lelj-Garolla and Mauk, 2006). In order to
investigate whether these three diterpenic acids change Hsp27
phosphorylation, we checked the levels of phosphorylated Hsp27
(p-Hsp27) and total Hsp27 in LNCaP and PC3 cells after 24 h
treatment with the three diterpenic acids (Fig. 9). The concen-
tration of diterpenic acids used to treat LNCaP cells was lower
than that for PC3 treatment, which is due to their cytotoxicity in
LNCaP cells. p-Hsp27 bands were not detected in both cell lines,
suggesting that three diterpenic acids did not significantly induce
Fig. 9. Effects of copalic acid (CA) and hardwickiic acid (HA) and 3-acetoxycopalic
the phosphorylation of Hsp27 to regulate the oligomerization of
this protein. Additionally, no clear changes in Hsp27 total protein
were observed. Hsp27 is an important chaperone protein and can
stabilize many client proteins (Andrieu et al., 2010; Gibert et al.,
acid (ACA) on Hsp27 and several Hsp27 interactive proteins. LNCaP and PC3 cells
were treated with DMSO, three diterpenic acids for 24 h. Levels of p-Hsp27, Hsp27,
eIF4E, Stat3, Hsp70, and
b-actin were analyzed by Western blotting of cell extracts
with specific antibodies as described in Section 2.

74
R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
2012). Could these diterpenic acids block the interaction of
Hsp27 with its client proteins, thereby affecting the stability of
the client proteins? In order to answer this question, we checked
three representative Hsp27 client proteins including eukaryotic
translation initiation factor 4E (eIF4E) (Andrieu et al., 2010),
signal transducer and activator of transcription 3 (Stat3) (Gibert
et al., 2012), and 70 kDa heat shock protein (Hsp70) (Kindas-
Mugge et al., 2002). The results revealed that all these protein
levels in LNCaP and PC3 remained nearly constant after the
treatment with the diterpernic acids except that the protein level
of Stat 3 in PC3 cells was slightly decreased by 3-acetoxycopalic
acid treatment. This result suggested that 3-acetoxycopalic acid
may affect Hsp27 chaperone function for stabilization of Stat3 in
PC3 cells. Overall, the cellular Hsp27 targeting effects of the three
diterpenic acids are not strong enough to affect the stability of
most client proteins of Hsp27. The carboxylic group (pKa ꢁ 5) of
these diterpenic acids is ionized in the cell culture medium,
which makes the molecules negatively charged. As a result, the
cell membrane permeability of these diterpenic acids could
be poor, which may decrease their potency in the cell study.
Future studies will focus on the designing of diterpenic acid
derivatives to improve their cell membrane permeability and
Hsp27 targeting effects.
References
Andrieu, C., Taieb, D., Baylot, V., Ettinger, S., Soubeyran, P., De-Thonel, A., Nelson, C.,
Garrido, C., So, A., Fazli, L., Bladou, F., Gleave, M., Iovanna, J.L., Rocchi, P., 2010.
Heat shock protein 27 confers resistance to androgen ablation and chemother-
apy in prostate cancer cells through eIF4E. Oncogene 29, 1883–1896.
Arrigo, A.P., Simon, S., Gibert, B., Kretz-Remy, C., Nivon, M., Czekalla, A., Guillet, D.,
Moulin, M., Diaz-Latoud, C., Vicart, P., 2007. Hsp27 (HspB1) and alphaB-crystal-
lin (HspB5) as therapeutic targets. FEBS Lett. 581, 3665–3674.
Arts, H.J., Hollema, H., Lemstra, W., Willemse, P.H., De Vries, E.G., Kampinga, H.H.,
Van der Zee, A.G., 1999. Heat-shock-protein-27 (hsp27) expression in ovarian
carcinoma: relation in response to chemotherapy and prognosis. Int. J. Cancer
84, 234–238.
Baylot, V., Andrieu, C., Katsogiannou, M., Taieb, D., Garcia, S., Giusiano, S., Acunzo, J.,
Iovanna, J., Gleave, M., Garrido, C., Rocchi, P., 2011. OGX-427 inhibits tumor
progression and enhances gemcitabine chemotherapy in pancreatic cancer. Cell
Death Dis. 2, e221.
Braun, S., Breitenbach, H., 1977. Strukturaufklarung einer neuen diterpensaure aus
metasequoia glyptostroboides mit hilfe der 13C-NMR-spektroskopie. Tetrahe-
dron 33, 145.
Bruey, J.M., Ducasse, C., Bonniaud, P., Ravagnan, L., Susin, S.A., Diaz-Latoud, C.,
Gurbuxani, S., Arrigo, A.P., Kroemer, G., Solary, E., Garrido, C., 2000. Hsp27
negatively regulates cell death by interacting with cytochrome c. Nat. Cell Biol.
2, 645–652.
Cavin, A.L., Hay, A.E., Marston, A., Stoeckli-Evans, H., Scopelliti, R., Diallo, D.,
Hostettmann, K., 2006. Bioactive diterpenes from the fruits of Detarium micro-
carpum. J. Nat. Prod. 69, 768–773.
Cayado-Gutierrez, N., Moncalero, V.L., Rosales, E.M., Beron, W., Salvatierra, E.E.,
Alvarez-Olmedo, D., Radrizzani, M., Ciocca, D.R., 2013. Downregulation of
Hsp27 (HSPB1) in MCF-7 human breast cancer cells induces upregulation of
PTEN. Cell Stress Chaperones 18, 243–249.
4. Conclusion
Concannon, C.G., Gorman, A.M., Samali, A., 2003. On the role of Hsp27 in regulating
apoptosis. Apoptosis: Int. J. Progr. Cell Death 8, 61–70.
Costa, M., Tanaka, C.M.A., Imamura, P.M., Marsaioli, A.J., 1999. Isolation and syn-
thesis of a new clerodane from Echinodorus grandiflorus. Phytochemistry 50,
117–133.
Estevao, L.R., Medeiros, J.P., Baratella-Evencio, L., Simoes, R.S., Mendonca Fde, S.,
Evencio-Neto, J., 2013. Effects of the topical administration of copaiba oil
ointment (Copaifera langsdorffii) in skin flaps viability of rats. Acta Cir. Bras./
Soc. Bras. Desenvolvimento Pesquisa em Cirurgia 28, 863–869.
Faiella, L., Piaz, F.D., Bisio, A., Tosco, A., De Tommasi, N., 2012. A chemical proteomics
approach reveals Hsp27 as a target for proapoptotic clerodane diterpenes. Mol.
bioSyst. 8, 2637–2644.
Garrido, C., 2002. Size matters: of the small HSP27 and its large oligomers. Cell
Death Diff. 9, 483–485.
Garrido, C., Bruey, J.M., Fromentin, A., Hammann, A., Arrigo, A.P., Solary, E., 1999.
HSP27 inhibits cytochrome c-dependent activation of procaspase-9. FASEB J. 13,
2061–2070.
Copaiba oil showed numerous biological and pharmacological
activities such as antimicrobial, anti-inflammatory, anti-mutagenic,
wound healing and cytotoxic activities. In this study, we focused on
small chaperone inhibition of Copaiba oil and its chemical
components. The oil resin was initially fractionated to non-acidic
components (B) and acidic components (C). The acidic components
were subjected to column chromatography on silica gel, leading to
seven sub-fractions (D–J). Three diterpenic acids including copalic
acid, hardwickiic acid, and 3-acetoxycopalic acid were obtained
from the sub-fractions that showed chaperone inhibition. Their
anti-chaperone activity was further confirmed by kinetic anti-a-
crystallin and anti-Hsp27 assays. The three diterpenic acids also
increased the anti-proliferation activity of carboplatin against
LNCaP cells. The additive effects may be associated with the small
chaperone inhibition of the three diterpenic acids. In addition, 3-
acetoxycopalic acid slightly decreased the level of Hsp27 client
protein Stat 3 in PC3 cells.
In brief, we found that copalic acid, hardwickiic acid, and
3-acetoxycopalic acid from Copaiba oil inhibited mammalian small
chaperone proteins. The activities of the chemical components
in Copaiba oil may account for the anti-cancer efficacy of this
natural oil resin. From drug discovery perspective, very few
small molecules have been reported to exhibit such novel
biological activity. These novel diterpenic acids could be used as
new lead compounds to develop more potent small chaperone
inhibitors.
Garrido, C., Brunet, M., Didelot, C., Zermati, Y., Schmitt, E., Kroemer, G., 2006. Heat
shock proteins 27 and 70: anti-apoptotic proteins with tumorigenic properties.
Cell Cycle (Georgetown, Tex.). 5, 2592–2601.
Gibert, B., Eckel, B., Fasquelle, L., Moulin, M., Bouhallier, F., Gonin, V., Mellier, G.,
Simon, S., Kretz-Remy, C., Arrigo, A.P., Diaz-Latoud, C., 2012. Knock down of heat
shock protein 27 (HspB1) induces degradation of several putative client pro-
teins. PloS One 7, e29719.
Hamann, S., Metrailler, S., Schorderet, D.F., Cottet, S., 2013. Analysis of the cyto-
protective role of alpha-crystallins in cell survival and implication of the
alphaA-crystallin C-terminal extension domain in preventing Bax-induced
apoptosis. PloS One 8, e55372.
Kindas-Mugge, I., Rieder, C., Frohlich, I., Micksche, M., Trautinger, F., 2002. Charac-
terization of proteins associated with heat shock protein hsp27 in the squamous
cell carcinoma cell line A431. Cell Biol. Int. 26, 109–116.
Leandro, L.M., Vargas Fde, S., Barbosa, P.C., Neves, J.K., da Silva, J.A., da Veiga-Junior,
V.F., 2012. Chemistry and biological activities of terpenoids from copaiba
(Copaifera spp.) oleoresins. Molecules (Basel, Switzerland) 17, 3866–3889.
Legault, J., Pichette, A., 2007. Potentiating effect of beta-caryophyllene on anticancer
activity of alpha-humulene, isocaryophyllene and paclitaxel. J. Pharm. Pharma-
col. 59, 1643–1647.
Lelj-Garolla, B., Mauk, A.G., 2006. Self-association and chaperone activity of Hsp27
are thermally activated. J. Biol. Chem. 281, 8169–8174.
Acknowledgements
Lima, S.R., Junior, V.F., Christo, H.B., Pinto, A.C., Fernandes, P.D., 2003. In vivo and in
vitro studies on the anticancer activity of Copaifera multijuga hayne and its
fractions. Phytother. Res. 17, 1048–1053.
Pieri, F.A., Mussi, M.C., Fiorini, J.E., Moreira, M.A., Schneedorf, J.M., 2012. Bacterio-
static effect of copaiba oil (Copaifera officinalis) against Streptococcus mutans.
Braz. Dental J. 23, 36–38.
Santos, A.O., Izumi, E., Ueda-Nakamura, T., Dias-Filho, B.P., Veiga-Junior, V.F.,
Nakamura, C.V., 2013. Antileishmanial activity of diterpene acids in copaiba
oil. Memorias Inst. Oswaldo Cruz. 108, 59–64.
This research was supported by Center for Gene Regulation in
Health and Disease (GRHD) of Cleveland State University and Ohio
Department of Development (ODOD), and also supported by CSU
Undergraduate Research and Creative Achievement award from
CSU.
Santos, R.C., dos Santos Alves, C.F., Schneider, T., Lopes, L.Q., Aurich, C., Giongo,
J.L., Brandelli, A., de Almeida Vaucher, R., 2012. Antimicrobial activity
of Amazonian oils against Paenibacillus species. J. Invertebr. Pathol. 109,
265–268.
Soares, D.C., Portella, N.A., Ramos, M.F., Siani, A.C., Saraiva, E.M., 2013. Trans-beta-
Caryophyllene: an effective antileishmanial compound found in commercial
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.phytol.2014.08.006.

R. Lama et al. / Phytochemistry Letters 10 (2014) 65–75
75
copaiba oil (Copaifera spp.). Evidence-based complementary and alternative
medicine. eCAM 2013, 761323.
Souza Barbosa, P.C., Moreira Wiedemann, L.S., da Silva Medeiros, R., de Tarso
Barbosa Sampaio, P., Vieira, G., Florencio da Veiga-Junior, V., 2013. Phytochem-
ical fingerprints of copaiba oils (Copaifera multijuga Hayne) determined by
multivariate analysis. Chem. Biodiver. 10, 1350–1360.
Zhong, B., Chennamaneni, S., Lama, R., Yi, X., Geldenhuys, W.J., Pink, J.J., Dowlati,
A., Xu, Y., Zhou, A., Su, B., 2013. Synthesis and anticancer mechanism
investigation of dual Hsp27 and tubulin inhibitors. J. Med. Chem. 56,
5306–5530.
Zoubeidi, A., Gleave, M., 2012. Small heat shock proteins in cancer therapy and
prognosis. Int. J. Biochem. Cell Biol. 44, 1646–1656.