Synthesis and characterization of new curcumin derivatives as potential chemotherapeutic and antioxidant agents

Article abstract of DOI:10.1002/ddr.21158

Preclinical Research The purpose of this work was to synthesize a series of symmetrical analogs (CA2-CA7) of curcumin and determine their efficacy as antioxidant and anticancer agents in vitro. The six analogs were successfully synthesized and characterized, one of which, CA6, had not been previously reported in the literature. With the exception of CA2, the analogs had lower predicted aqueous solubilities and higher partition coefficients than curcumin. Two analogs, CA2 and CA3, had lower potencies as anticancer agents compared with curcumin, while CA6 had a slightly higher IC₅₀ value. Two different trends in the antioxidant capabilities of curcumin and its analogs were determined when assessed in vitro or in cell culture. The in vitro DPPH assay clearly showed curcumin as the strongest antioxidant as compared with the analogs when tested at the same concentration or at their IC₅₀ value. The cell culture-based reactive oxygen species/reactive nitrogen species assay indicated that CA3 and CA6 were equal to curcumin in their free radical scavenging ability at the same concentration, but when curcumin and its analogs were tested at their respective IC₅₀ values, CA4 and CA5 showed excellent antioxidant capacities. These results indicate that in cell culture, the ability of these analogs to produce antioxidant effects may be tied to their downstream effects.

Full text of DOI:10.1002/ddr.21158

DRUG DEVELOPMENT RESEARCH 75 : 88–96 (2014)
Research Article
Synthesis and Characterization of New Curcumin
Derivatives as Potential Chemotherapeutic and
Antioxidant Agents
Roxana Ciochina,1 Chasity Savella,1 Brianna Cote,2 Davis Chang,1 and Deepa Rao2*
1Department of Chemistry, Pacific University, Forest Grove, OR 97116, USA
2School of Pharmacy, Pacific University, Hillsboro, OR 97123, USA
Strategy, Management and Health Policy
Enabling
Preclinical Development Clinical Development
Technology,
Genomics,
Proteomics
Preclinical
Research
Toxicology, Formulation
Drug Delivery,
Pharmacokinetics
Phases I-III
Regulatory, Quality,
Manufacturing
Postmarketing
Phase IV
ABSTRACT
The purpose of this work was to synthesize a series of symmetrical analogs (CA2–CA7) of
curcumin and determine their efficacy as antioxidant and anticancer agents in vitro. The six analogs were
successfully synthesized and characterized, one of which, CA6, had not been previously reported in the
literature. With the exception of CA2, the analogs had lower predicted aqueous solubilities and higher
partition coefficients than curcumin. Two analogs, CA2 and CA3, had lower potencies as anticancer agents
compared with curcumin, while CA6 had a slightly higher IC50 value. Two different trends in the
antioxidant capabilities of curcumin and its analogs were determined when assessed in vitro or in cell
culture. The in vitro DPPH assay clearly showed curcumin as the strongest antioxidant as compared with
the analogs when tested at the same concentration or at their IC50 value. The cell culture-based reactive
oxygen species/reactive nitrogen species assay indicated that CA3 and CA6 were equal to curcumin in
their free radical scavenging ability at the same concentration, but when curcumin and its analogs were
tested at their respective IC50 values, CA4 and CA5 showed excellent antioxidant capacities. These results
indicate that in cell culture, the ability of these analogs to produce antioxidant effects may be tied to their
downstream effects. Drug Dev Res 75 : 88–96, 2014.
© 2013 Wiley Periodicals, Inc.
Key words: curcumin analogs; antioxidant; chemotherapeutics; chemoprevention
INTRODUCTION
curcumin in the treatment of Alzheimer’s disease have
also been reported [Park and Kim, 2002].
Curcumin [Ammon and Wahl, 1991], also known
as diferuloyl methane, is a member of the curcuminoid
family isolated from the spice turmeric of the herb
Curcuma longa. Turmeric has been used for centuries
as a dye, food additive, dietary spice, and therapeutic in
the treatment of wounds, cuts, jaundice, and rheuma-
toid arthritis [Ammon and Wahl, 1991; Shobana and
Naidu, 2000; Naik et al., 2003; Rao, 2003]. Curcumin,
the major bioactive component, possesses numerous
interesting biological activities, such as antioxidant
[Toda et al., 1985; Jayaprakasha et al., 2006], anti-
inflammatory [Nurfina et al., 1997], antimalarial
[Mishra et al., 2008], and anticarcinogenic [Agrawal
Chemically, curcumin contains three important
functional regions: α,β-unsaturated β-diketo group; an
olefinic linker; and an orthomethoxy phenolic hydroxy
group. As shown in Figure 1, curcumin is conjugated
and symmetric with two aromatic ring systems con-
nected by a bis-α,β-unsaturated β-diketone. Therefore,
*Correspondence to: Deepa A. Rao, School of Pharmacy,
Pacific University, 190 SE 8th Avenue, Hillsboro, OR 97123, USA.
E-mail: deeparao@pacificu.edu
Received 30 August 2013; Accepted 10 October 2013
Published online in Wiley Online Library (wileyonlinelibrary
and Mishra, 2010]. Studies on the potential role of .com). DOI: 10.1002/ddr.21158
© 2013 Wiley Periodicals, Inc.

CURCUMIN ANALOGS CHEMOTHERAPEUTIC ACTIVITY
89
Fig. 1. Curcumin in its keto–enol equilibrium.
Fig. 2. Structures of curcumin analogs and their abbreviations, CA2–CA7.
it can undergo keto–enol tautomerism, where the enol [Carroll et al., 2011]. Therefore, there is a need to
is the predominant form in solution as a result of con- develop curcumin formulations or analogs to better
jugation and intramolecular hydrogen bonding that address the solubility and the bioavailability issues
increase the stability of the enol form. Considerable related to curcumin dosing. This will also aid in advanc-
structure activity-based research has shown that the ing the use of curcumin in various disease states where
ortho-methoxy phenolic OH group is required for anti- it has shown great promise in vitro.
oxidant activity and the α,β-unsaturated β-diketo group
Therefore, the objective of this work was to syn-
is essential for anticancer activity [Priyadarsini et al., thesize a series of symmetrical analogs (CA2–CA7) of
2003; Chen et al., 2006]. Moreover, the length of the curcumin (Fig. 2) with varying solubilities and partition
olefinic linker is linked to the prevention of protein coefficients, and determine their efficacy as antioxidant
aggregation in Alzheimer’s disease models [Huang and anticancer agents in vitro. As in curcumin, all these
et al., 2005; Anand et al., 2007].
compounds maintain the 7-carbon dienone spacer
Though curcumin has been extensively studied between the rings, but have various substituents on the
and used in various herbal preparations, its efficacy in ring, including nonphenolic groups. These analogs were
vivo is limited by low aqueous solubility and oral low characterized and evaluated for their antioxidant and
bioavailability [Anand et al., 2008]. Curcumin has an antitumor activities in vitro to determine if a correlation
intrinsic aqueous solubility of 0.6 μg/mL [Kurien et al., exists between the structural modifications and the bio-
2007], with the oral bioavailability of curcumin in logical activities.
human being less than 1% [Anand et al., 2007]. The
only clinically beneficial in vivo effects demonstrated
MATERIALS AND METHODS
Cell Titer Blue® was purchased from Promega
with curcumin have been in preventative clinical trials
in colorectal cancer where the high local concentration
of the compound may induce a protective effect (Madison, WI, USA). OxiSelect™ in vitro reactive
Drug Dev. Res.

90
CIOCHINA ET AL.
oxygen species/reactive nitrogen species (ROS/RNS) recorded on an Electrothermal Mel-Temp melting point
assay kit (Green Fluorescence) was purchased from apparatus (Bibby Scientific, Staffordshire, UK) and are
Cell Biolabs, Inc (San Diego, CA, USA). Human presented uncorrected. The IR spectra were recorded
ovarian adenocarcinoma cells, SKOV-3, were obtained on a Thermo Scientific Nicolet iS10 FT-IR Spectrom-
from American Type Culture Collection (Manassas, eter (Waltham, MA), UV Spectra were recorded on a
VA, USA). Cell culture supplies, including RPMI Shimadzu UV-1601 Spectrophotometer (Columbia,
1640 1X, fetal bovine serum, antibiotics penicillin/ MD). H-NMR and C-NMR were recorded on Bruker
streptomycin, and phosphate-buffered saline (PBS) Avance 3000 Ultra Shield 300 MHz NMR Spectrometer
were all purchased through VWR (Radnor, PA, USA). (Billerica, MA) . Chemical shifts are in ppm (δ) relative
All solvents used were of HPLC grade and were to TMS. The partition coefficients and the aqueous
obtained from VWR.
solubilities of curcumin and the analogs were predicted
using ACD Labs Precepta 14.0.0 (Build 1996) (Toronto,
Ontario, Canada) software with Consensus Log P
module and solubility in pure water module.
Synthesis of Curcumin Analogs
An aliquot of 0.050 mol of appropriate aldehyde
was added to 25.0 mL of ethyl acetate. To this mixture,
0.100 mol of tributyl borate was added. Separately, a
solution of 0.025 mol of acetyl acetone and 0.0180 mol
of boric anhydride was prepared and added to the reac-
tion mixture. While stirring, 0.125 mL of n-butylamine
was added dropwise every 10 min for 40 min. The reac-
tion was stirred for an additional 4 h and allowed to
stand overnight. The following day, 37.5 mL of 0.4 M
HCl was heated to 60°C, added to the reaction mixture,
and stirred for 1 h. The layers were separated and the
water layer washed three times with 12.5 mL of ethyl
acetate. The organic layers were combined, washed
with brine, and dried over MgSO4. Using a rotary
evaporator, the solution was allowed to evaporate to a
volume of about 20 mL. Then 12.5 mL of methanol
was added and the reaction mixture was allowed to sit
for 2 h in an ice bath. Solid was collected via vacuum
filtration and recrystallized from CH2Cl2/methanol
mixture.
DPPH Radical Scavenging Activity
The free radical scavenging activity of curcumin
(compound 1) and its derivatives was measured by the
scavenging of DPPH (2,2-diphenyl-1-picrylhydrazyl)
radical by modifying a previously reported method
[Sreejayan and Rao, 1996]. Two assays were performed.
In the first study, DPPH (3.0 mL, 0.5 mM) was incu-
bated for 30 min with curcumin (1; Curumin has been
identified as compound 1 in the DPPH free radical
scavenging activity and the same nomenclature is con-
tinued in Table 1) or curcumin derivatives (CA2–CA7)
or trolox (9.0 mL, 0.3 mM) at final concentrations of
225 μM. In the second study, DPPH (3.0 mL, 0.5 mM)
was incubated for 30 min with curcumin (1) or
curcumin derivatives (CA2–CA6) or trolox (9.0 mL,
various concentrations) to have final concentrations cor-
responding to the IC50 value for each compound except
for CA7. Trolox could also not be solubilized at its IC50
value and the highest concentration that could be
solubilized was 324 μM. The solvent used in preparing
all solutions was dichloromethane. The absorbance was
read after 30 min at 517 nm. In order to determine the
Characterization of Curcumin Analogs
Analogs were characterized by their melting DPPH scavenging activity of each compound, the fol-
points, IR, UV, and NMR spectra. Melting points were lowing equation was used:
TABLE 1. Yields and Melting Points of Curcumin Analogs
Yield
(%)
Referenced melting
point (oC)
Analog
CA2
Chemical name
Melting point (oC)
(1E,4Z,6E)-5-hydroxy-1,7-bis(3-hydroxy-4-methoxyphenyl)
hepta-1,4,6-trien-3-one
(1E,4Z,6E)-1,7-bis(3,4-dimethoxyphenyl)-5-hydroxyhepta-
1,4,6-trien-3-one
63
38
185–187 (from MeOH/CH2Cl2) 190–192 (20)
130–131 (from MeOH) 128–130 (7)
CA3
CA4
CA5
(1E,4Z,6E)-5-hydroxy-1,7-diphenylhepta-1,4,6-trien-3-one
dimethyl 5,5'-(1E,3Z,6E)-3-hydroxy-5-oxohepta-1,3,6-triene-1,7-
diyl)bis(2-hydroxybenzoate)
10
51
140–142 (from Et2O/pet ether) 140.5 (19)
211–214 (from MeOH/CH2Cl2) 210–212 (21)
CA6
CA7
(1E,4Z,6E)-5-hydroxy-1,7-di(naphthalen-1-yl)hepta-1,4,6-trien-3-one 54
(1E,4Z,6E)-5-hydroxy-1,7-di(naphthalen-2-yl)hepta-1,4,6-trien-3-one 50
167–170 (from MeOH/CH2Cl2) None available
256–257 (from MeOH/CH2Cl2) None available
Drug Dev. Res.

CURCUMIN ANALOGS CHEMOTHERAPEUTIC ACTIVITY
91
Fig. 3. General synthetic scheme for curcumin analogs, CA2–CA7.
centrifuged at 10,000 rpm for 20 min at 4°C. Superna-
tant was collected and assayed per manufacturer
instructions. Briefly, 50 μL of supernatant was added to
a 96-well plate, with each sample run in duplicate. Cata-
lyst (50 μL) was added, mixed, and allowed to incubate
for 5 min at room temperature followed by the addition
of 100 μL of DCFH (2', 7'-dichlorodihydrofluorescein)
solution to each well. The samples were covered to
protect from light, allowed to incubate for 45 min, and
the fluorescence read (excitation wavelength = 480 nm,
emission wavelength = 530 nm). One-way ANOVA with
Dunnett’s posttest was performed using GraphPad
Prism version 5.0 for Windows (GraphPad Software,
http://www.graphpad.com) to compare the antioxi-
dant capacity. All experiments were performed in tripli-
cate and data were presented as the mean ROS/RNS
Activity SD.
Scavenging Activity % = 1 − A A ×100
( )
(
(
)
s
c
where, Ac is the absorbance of DPPH (control) and AS
is the absorbance of DPPH in the presence of the test
compound. Each compound was run in triplicate and
the data have been presented as mean scavenging effect
(%) SD. A student’s t-test was performed using
GraphPad Prism version 5.00 for Windows (GraphPad
Software, San Diego, CA, USA, http://www.graphpad
.com) to compare the DPPH radical scavenging effect
of each of the treatments to curcumin.
Determination of Cytotoxicity of Curcumin and Its
Analogs in SKOV-3 Cells
SKOV-3 cells were maintained in RPMI 1640, 1×
with 10% fetal bovine serum and 1% penicillin strepto-
mycin at 37°C with 5% CO2 and 95% humidity.
SKOV-3 cells were seeded in 96-well plates at a con-
centration of 5000 cells per well and incubated for 24 h
for attachment. The cells were treated with curcumin
(1) or its analogs (CA2–CA5) in the concentration
range of 0.01–270 μM. Due to the limited solubility of
CA6 in DMSO, treatments were in the concentration
range of 0.01–17 μM. Due to the insolubility of CA7 in
DMSO, no further cell culture-based analysis was pos-
sible with this analog. The final concentration of DMSO
in all the wells was 1%. Posttreatment, the cells were
incubated for 48 h and treated with 20 μL of Cell Titer
Blue® for 2 h to assess cell viability. The cells were
analyzed by fluorescence at excitation wavelength of
560 nm and emission wavelength of 590 nm. All experi-
ments were performed in quadruplicate and data are
presented as Mean IC50 SD.
RESULTS
Synthesis of Curcumin Analogs
Curcumin derivatives were prepared by the con-
densation of the appropriate aldehydes with acetyl
acetone-boric oxide complex in ethyl acetate (Fig. 3) in
the presence of tributyl borate and n-butylamine, as
reported earlier [Pabon, 1964]. Upon recrystallization,
yellow to orange powders were obtained in 10–63%
yields (unoptimized yields). The analog, CA6 thus far,
has not been reported in literature. All curcumin
analogs were characterized by interpretation of spectral
data (IR, UV, 1H-NMR, and 13C-NMR) and compared
with the reported data (Table 1).
Compound CA6 has been synthesized for the first
time, and its structure has been established from a
comparison of the observed IR, UV, 1H-NMR, and 13C-
NMR spectral data. The IR spectrum (Fig. 4a) shows a
broad shallow band around 3000/cm, which corre-
Quantification of ROS and RNS in SKOV-3 Cells
SKOV-3 cells were seeded in 6-well plates at a cell sponds to the enolic O-H bond. The 1614/cm peak
density of 4 × 105 cells/well. The cells were allowed to indicates carbonyl C = O group involved in intramo-
attach overnight at 37°C, 5% CO2, and 95% H2O. At lecular hydrogen bonding. The UV spectrum (Fig. 4b)
24 h, the cells were treated with 10 μM or IC50 concen- in 10−5 M DMSO solution shows a strong λmax at 422 nm
trations of curcumin (1), or its analogs (CA2–CA6), or corresponding to n→π* transition, and a much weaker
trolox and incubated for an additional 24 h. The plates λmax at 268 nm corresponding to π→π* transition.
1
were then centrifuged at 4000 rpm for 5 min, media The H-NMR spectrum (Fig. 4c) shows a singlet at
were aspirated, and cells were washed twice with PBS. δ = 5.97 ppm indicating the presence of the methine
The cells were lysed with Triton X–100 cold lysis buffer, proton on C4 with no keto–enol equilibrium observed.
scraped and collected into a microcentrifuge tube, and The chemical shift of the hydrogen-bonded hydroxyl
Drug Dev. Res.

92
CIOCHINA ET AL.
1
Fig. 4. CA6 spectra—IR (a), UV (b), H-NMR (c), and 13C-NMR (d).
proton (H8) is δ 16.04 ppm, exceptionally downfield as (1E,4Z,6E)-5-hydroxy-1,7-di(naphthalen-1-yl)
with most enol tautomers of 1,3-diketones. The alkenyl hepta-1,4,6-trien-3-one (CA6)
protons correspond to the chemical shift at 8.56 ppm
Yield 54%, mp 167–170°C (from MeOH/CH2Cl2).
(H1 and H7) and 6.78 ppm (H2 and H6), respectively. Found: 85.82; H, 5.42. Calc. for C27H20O2: C, 86.14; H,
1
The trans geometry was identified from the observed J 5.36. H-NMR: δ (CDCl3) 16.51 (s, 1H), 8.56 (d, 2H,
value (Jd = 15.8 Hz).
J = 15.6 Hz), 8.28 (d, 2H, J = 8.3 Hz), 7.90 (m, 4H),
Drug Dev. Res.

CURCUMIN ANALOGS CHEMOTHERAPEUTIC ACTIVITY
93
TABLE 2. Predicted Solubilities and Log Distribution Coefficients
for Curcumin and its Analogs, CA2–CA6
TABLE 3. The IC50 Values of Curcumin and its Analogues, CA2–
CA7, and Trolox in SKOV-3 Cells Post 48 hr Treatment
Solubility in pure
water (mg/mL)
Log partition
coefficient
Compound
Mean (μM)
SD (μM)
Compound
Curcumin
CA2
CA3
CA4
CA5
CA6
CA7
Trolox
6.70
5.58
3.51
75.15
72.70
14.47
ND
2.44
2.00
0.74
1.19
3.84
3.87
ND
Curcumin (1)
CA2
CA3
CA4
CA5
0.12
0.12
0.069
0.0041
0.016
4.8 × 10−5
6.5 × 10−6
2.64
2.56
3.22
3.91
4.81
6.05
6.04
CA6
CA7
1607.89
66.37
SD, standard deviation; ND, not determined.
other analogs matched the ability of curcumin to scav-
enge DPPH radicals at 225 μM or at their respective
IC50 values. Statistical analysis indicates that the analogs
tested had significantly lower free radical scavenging
capability compared with curcumin. The trends in the
antioxidant capacity as compared to curcumin were
similar at both 225 μM and at the respective IC50 values
of all the compounds.
Determination of Cytotoxicity of Curcumin and Its
Analogs in SKOV-3 Cells
Data for the cytotoxicity of curcumin and its
analogs as Mean IC50 SD are presented in Table 3.
This shows that CA3 is the most potent analog in com-
parison with curcumin. Both, CA2 and CA3 had
cytotoxicities similar to curcumin.
Fig. 5. DPPH scavenging by curcumin, its analogs (CA2–CA7) and
trolox at 225 μM or at their respective IC50 values (except CA7). Data
are presented as mean scavenging effect (%) SD. n = 3; * indicates
statistically significant difference in effect at the different concentra-
tions (P < 0.05); ϕ indicates statistically significant difference as com-
é
Quantification of ROS and RNS in SKOV-3 Cells
pared with curcumin at 225 μM (P < 0.05);
indicates statistically
significant difference of the analogs at their IC50 values as compared
The ROS/RNS activity of curcumin, its analogs
and trolox in SKOV-3 cells at 10 μM, and their respec-
tive IC50 values, were compared with control ROS/RNS
(Fig. 6). Based on the data, curcumin, CA3, and CA6
were significantly more potent at scavenging these
ROS/RNS species in SKOV-3 at 10 μM. Further analy-
sis comparing curcumin to CA3 or CA6 at 10 μM was
not significant, indicating that these analogs are as
potent at scavenging ROS/RNS species as curcumin at
this concentration. At the IC50 values, curcumin still
demonstrated significant ROS/RNS scavenging ability.
CA4 and CA5 were the only curcumin analogs that
showed significant antioxidant effects at their IC50
values. While CA4 is not significantly different as com-
pared with curcumin in its antioxidant effect, CA5
showed a significantly higher antioxidant effect as com-
pared with curcumin at its IC50 value.
with curcumin at its IC50 values (P < 0.05).
7.83 (d, 2H, J = 7.2 Hz), 7.59 (m, 4H), 7.53 (d, 2H,
J = 8.0 Hz), 6.78 (d, 2H, J = 15.5 Hz), 5.97 (s, 1H); 13C-
NMR: δ (CDCl3) 183.3, 137.5, 133.7, 132.4, 131.6,
130.4, 128.7, 126.9, 126.6, 126.2, 125.5, 124.9, 123.5,
102.2.
The solubilities and log partition coefficients for
curcumin and its analogs, CA2–CA7 structures, were
predicted using ACD Labs Prospecta software (Toronto,
Ontario, Canada) with the Consensus Log P Module and
the Solubility in Pure Water Module (Table 2).
DPPH Radical Scavenging Activity
The DPPH radical scavenging effect (Fig. 5) indi-
cates that curcumin and trolox have very similar scav-
enging effects at 225 μM, but at the IC50 concentration
of curcumin compared with trolox at 324 μM, trolox is a
DISCUSSION
Based on the solubility predictions, the structural
far more potent antioxidant. However, none of the modifications on the rings in all but CA2 have lower
Drug Dev. Res.

94
CIOCHINA ET AL.
cytotoxic agent. The unsubstituted naphthalene deriva-
tive did decrease the potency of the analog in compari-
son with curcumin, possibly due to steric hindrance.
CA7 could not be tested due to its insolubility in
DMSO; however, given its structural similarity to CA6,
it is possible that the same steric hindrance would be an
issue. The analogs showing the least potency were CA4
and CA5. For CA4, the lack of any substituents might
be responsible for the decreased efficacy, while in the
case of CA5, shifting of the methyl group away from the
ring may decrease cytotoxicity. So, while the methoxy
group itself might not be the most important factor in
cytotoxicity, substitution on the ring at the 5’, 4’, and 3’
position appears to have significant effects on the
cytotoxicity in ovarian cancer cells. Curcumin exerts its
anticancer activities by altering deregulated cell cycle
via cyclin- and p53-dependent and independent path-
ways [Sa and Das, 2008]. Therefore, interaction of
curcumin or its analogs with these pathways is critical to
their functionality. The downstream effects of these
analogs need to be further investigated to determine
the precise role of the substituents in eliciting cell
Fig. 6. The ROS/RNS activities of trolox, curcumin (1), and curcumin
analogs CA2–CA6 in SKOV-3 cells at 10 μM or at their respective IC50
values (except CA7). Data are presented as Mean ROS/RNS Activity
SD. n = 3; * indicates statistically significant difference in effect at
the different concentrations (P < 0.05); ϕ indicates statistically signifi-
cant difference as compared with curcumin at 10 μM (P < 0.05);
é
indicates statistically significant difference of the analogs at their
IC50 values as compared with curcumin at its IC50 values (P < 0.05).
predicted aqueous solubilities than curcumin itself. In death.
addition, CA6 and CA7, derivatives of naphthalene,
Interestingly, neither CA3 nor CA6 contain the
had far lower solubilities in the μg/mL range, com- ortho-methoxy phenolic OH group previously cited as
pared with the other ring substituents, consistent with necessary for the antioxidant effect [Priyadarsini et al.,
the ring structure. The predicted log partition coeffi- 2003; Chen et al., 2006]. This discrepancy may be due
cients for all analogs except CA2 were higher than to the fact that most antioxidant studies employed in
curcumin, which correlates with their lower solubility. structure activity relationship determination are not
The naphthalene derivatives have the highest partition based in cell culture or in in vivo models, where the
coefficients and resulted in far lower aqueous solu- interaction of the molecules with a dynamic redundant
bilities. Therefore, CA2 is the only structural deriva- system may produce different effects. The trends in the
tive of curcumin that has a similar solubility with a DPPH assay for antioxidant effects remain consistent
lower partition coefficient.
when tested at the same concentration or at the respec-
Prior research has demonstrated that curcumin is tive IC50 values of curcumin, its analogs (CA2–CA6),
a more potent free radical scavenger than vitamin E and trolox. However, in the case of trolox, solubility was
[Zhao et al., 1989]. It has been suggested that the limited at 324 μM, which is a quarter of its IC50 con-
methoxy group on the phenol ring [Sreejayan and Rao, centration. In the DPPH assay, there is a concentration-
1996; Priyadarsini et al., 2003] as well as the β- dependent correlation with the ability of the molecule
diketone moiety [Jovanovic et al., 2001] in curcumin are in scavenging DPPH radicals, which may be antici-
the key functional groups responsible for the antioxi- pated, based on the assay principle. DPPH is a free
dant and free radical scavenging properties. The stable radical accepting a hydrogen radical or an elec-
present results indicate that the position of the hydroxyl tron. DPPH in the free radical form has an absorbance
and methoxy groups is important, as noted from the at λmax = 517 nm, which disappears after accepting the
scavenging activity of CA2, where the OH group is on hydrogen radical from an antioxidant. Curcumin and its
C3’ and OMe group is on C4’. All other compounds analogs provide these hydrogen radicals, so by decreas-
showed very weak scavenging effects.
ing the concentration of these compounds, we expected
CA2 and CA3, where the position of the methoxy an increased in the absorbance of DPPH at 517 nm,
is reversed (CA2) or the methoxy is removed (CA3), thus a weaker scavenging affect. The IC50 concentra-
appear to indicate that this has little effect on tions are all much lower than 225 μM, and according to
cytotoxicity, supporting prior findings that the α,β- our predictions, all scavenging effects are lower.
unsaturated β-diketo group is essential for anticancer
In comparing the antioxidant activity of curcumin
activity [Priyadarsini et al., 2003; Chen et al., 2006] and and its analogs in the DPPH assay and in cell culture as
not the methoxy group. CA6 also showed efficacy as a assessed by the ROS/RNS assay, it is evident that the
Drug Dev. Res.

CURCUMIN ANALOGS CHEMOTHERAPEUTIC ACTIVITY
95
data show different trends in activity. This can be used as chemotherapeutics, while CA4 and CA5 have
expected as the DPPH assay directly assesses the ability excellent potential to be chemopreventative as antioxi-
of a compound to scavenge free radicals, while a cell dants due to their low cytotoxic potential and strong
culture-based system like SKOV-3 involves multiple free radical scavenging ability in cell culture. We have
cellular processes that can be affected by antioxidants also shown that two different in vitro methods of assess-
like curcumin. Curcumin can induce antioxidant and ing antioxidant activity can have demonstrably different
detoxifying enzymes via Nrf2/EpRE signaling pathways trends in antioxidant effect. However, we believe that
[Erlank et al., 2011]. Therefore, in the cell culture this may be due to the activation of downstream effec-
treatment, we may be witnessing the combination of tors in the in vitro cell culture model and further study
direct free radical scavenging ability of the compound is needed to understand these processes.
and its downstream effects in triggering inherent
detoxifying enzymes within the cell. In this context, in
cell culture, CA3 and CA6 have an equivalent ability to
curcumin in decreasing the total free radicals present at
10 μM. However, at their respective IC50 values, CA3
and CA6 are unable to scavenge any ROS/RNS species.
Whether CA3 and CA6 work through the same mecha-
nism remains to be elucidated. Interestingly, the posi-
tive control, trolox, had ROS/RNS scavenging effects at
10 μM, but at its IC50 value no antioxidant effects were
observed. Interestingly, at their respective IC50 concen-
ACKNOWLEDGMENTS
This work was supported by M. J. Murdock
Charitable Trust and Pacific University. We thank
Richard V. Whiteley and David Cordes for reviewing
this manuscript.
REFERENCES
trations, only curcumin, CA4, and CA5 show antioxi- Agrawal DK, Mishra PK. 2010. Curcumin and its analogues: poten-
tial anticancer agents. Med Res Rev 30:818–860.
dant effects. CA4 and CA5 have much higher IC50
(∼70 μM) values compared with curcumin (6.5 μM).
Therefore, the antioxidant effect for CA4 and CA5 may
be concentration dependent, and by modulating the
concentration between 10 μM and 75 μM we may be
able to use these analogs, especially CA5, which has a
pronounced antioxidant effect at its IC50 value. Based
on our results, we can classify these analogs as potential
antioxidants or chemotherapeutics. CA2, CA3, and
CA6 may prove to have chemotherapeutic potential,
especially if combined with other agents, while CA4
and CA5, with additional studies, might be more useful
as chemopreventative based on their low cytotoxicity
and strong antioxidants effects.
Ammon HPT, Wahl MA. 1991. Pharmacology of Curcuma longa.
Planta Med 57:1–7.
Anand P, Kunnumakkara AB, Newman RA, Aggarwal BB. 2007.
Bioavailability of curcumin: problems and promises. Mol Pharm
4:807–818.
Anand P, Thomas SG, Kunnumakkara AB, Sundaram C, Harikumar
KB, Sung B, Tharakan ST, Misra K, Priyadarsini IK, Rajasekharan
KN, et al. 2008. Biological activities of curcumin and its analogues
(Congeners) made by man and Mother Nature. Biochem
Pharmacol 76:1590–1611.
Carroll RE, Benya RV, Turgeon DK, Vareed S, Neuman M,
Rodriguez L, Kakarala M, Carpenter PM, McLaren C, Meyskens
FL, et al. 2011. Phase IIa clinical trial of curcumin for the preven-
tion of colorectal neoplasia. Cancer Prev Res (Phila) 4:354–364.
Trolox is a water-soluble analog of vitamin E and
therefore may not have had sufficient permeability
across the cell membrane in the time allotted to
produce a significant effect in cell lines, limiting its
ability to scavenge free radicals in this type of an experi-
mental setup. Therefore, additional studies must be
conducted to properly utilize trolox as an antioxidant
benchmark for natural compounds in cell culture
models or other compounds might have to be used as
positive controls in cell culture models.
In conclusion, six analogs of curcumin, one never
reported in literature, with differing solubilities and
partition coefficients, have been synthesized and
assessed for their activity as free radical scavengers
and potential chemotherapeutic effects. Based on the
cytotoxicity analysis in ovarian cancer cells, we have
demonstrated that CA2, CA3, and CA6 have lower or
similar potencies to curcumin and potentially can be
Chen WF, Deng SL, Zhou B, Yang L, Liu ZL. 2006. Curcumin and
its analogues as potent inhibitors of low density lipoprotein oxida-
tion: H-atom abstraction from the phenolic groups and possible
involvement of the 4-hydroxy-3-methoxyphenyl groups. Free
Radic Biol Med 40:526–535.
Erlank H, Elmann A, Kohen R, Kanner J. 2011. Polyphenols activate
Nrf2 in astrocytes via H2O2, semiquinones, and quinones. Free
Radic Biol Med 51:2319–2327.
Huang DJ, Ou BX, Prior RL. 2005. The chemistry behind antioxi-
dant capacity assays. J Agric Food Chem 53:1841–1856.
Jayaprakasha GK, Rao LJ, Sakariah KK. 2006. Antioxidant activities
of curcumin, demethoxycurcumin and bisdemethoxycurcumin.
Food Chem 98:720–724.
Jovanovic SV, Boone CW, Steenken S, Trinoga M, Kaskey RB. 2001.
How curcumin works preferentially with water soluble antioxi-
dants. J Am Chem Soc 123:3064–3068.
Kurien B, Singh A, Matsumoto H, Scofield R. 2007. Improving the
solubility and pharmacological efficacy of curcumin by heat treat-
ment. Assay Drug Dev Technol 5:567–576.
Drug Dev. Res.

96
CIOCHINA ET AL.
Mishra S, Karmodiya K, Surolia N, Surolia A. 2008. Synthesis and
exploration of novel curcumin analogues as anti-malarial agents.
Bioorg Med Chem 16:2894–2902.
hydrogen on the free radical reactions and antioxidant activity of
curcumin. Free Radic Biol Med 35:475–484.
Rao BN. 2003. Bioactive phytochemicals in Indian foods and their
potential in health promotion and disease prevention. Asia Pac J
Clin Nutr 12:9–22.
Naik GH, Priyadarsini KI, Satav JG, Banavalikar MM, Sohoni DP,
Biyani MK, Mohan H. 2003. Comparative antioxidant activity of
individual herbal components used in Ayurvedic medicine.
Phytochemistry 63:97–104.
Sa G, Das T. 2008. Anti cancer effects of curcumin: cycle of life and
death. Cell Div 3:14.
Nurfina AN, Reksohadiprodjo MS, Timmerman H, Jenie UA,
Sugiyanto D, van der Goot H. 1997. Synthesis of some symmetri-
cal curcumin derivatives and their antiinflammatory activity. Eur J
Med Chem 32:321–328.
Shobana S, Naidu KA. 2000. Antioxidant activity of selected Indian
spices. Prostaglandins Leukot Essent Fatty Acids 62:107–110.
Sreejayan N, Rao M. 1996. Free radical scavenging activity of
curcuminoids. Arzneimittelforschung 46:169–171.
Pabon HJJ. 1964. A synthesis of curcumin and related compounds.
Rec Trav Chim 83:379–386.
Toda S, Miyase T, Arichi H, Tanizawa H, Takino Y. 1985. Natural
Antioxidants .13. Antioxidative components isolated from
rhizome of curcuma-longa L. Chem Pharm Bull (Tokyo) 33:1725–
1728.
Park SY, Kim D. 2002. Discovery of natural products from Curcuma
longa that protect cells from beta-amyloid insult: a drug discovery
effort against Alzheimer’s disease. J Nat Prod 65:1227–1231.
Zhao BL, Li XJ, He RG, Cheng SJ, Xin WJ. 1989. Scavenging effect
of extracts of green tea and natural antioxidants on active oxygen
radicals. Cell Biophys 14:175–185.
Priyadarsini KI, Maity DK, Naik GH, Kumar MS, Unnikrishnan MK,
Satav JG, Mohan H. 2003. Role of phenolic O-H and methylene
Drug Dev. Res.