Binding of curcumin and its long chain derivatives to the activator binding domain of novel protein kinase C

Article abstract of DOI:10.1016/j.bmc.2009.12.075

Protein kinase C (PKC) is a family of serine/threonine kinases that play a central role in cellular signal transduction. The second messenger diacylglycerol having two long carbon chains acts as the endogenous ligand for the PKCs. Polyphenol curcumin, the active constituent of Curcuma longa is an anti-cancer agent and modulates PKC activity. To develop curcumin derivatives as effective PKC activators, we synthesized several long chain derivatives of curcumin, characterized their absorption and fluorescence properties and studied their interaction with the activator binding second cysteine-rich C1B subdomain of PKCδ, PKCε and PKCθ. Curcumin (1) and its C16 long chain analog (4) quenched the intrinsic fluorescence of PKCδC1B, PKCεC1B and PKCθC1B in a manner similar to that of PKC activator 12-O-tetradecanoylphorbol 13-acetate (TPA). The EC₅₀s of the curcumin derivatives for fluorescence quenching varied in the range of 4-11 μM, whereas, EC₅₀s for TPA varied in the range of 3-6 μM. Fluorescence emission maxima of 1 and 4 were blue shifted and the fluorescence anisotropy values were increased in the presence of the C1B domains in a manner similar to that shown by the fluorescent analog of TPA, sapintoxin-D, confirming that they were bound to the proteins. Molecular docking of 1 and 4 with novel PKC C1B revealed that both the molecules form hydrogen bonds with the protein residues. The present result shows that curcumin and its long chain derivatives bind to the C1B subdomain of novel PKCs and can be further modified structurally to improve its binding and activity.

Full text of DOI:10.1016/j.bmc.2009.12.075

Bioorganic & Medicinal Chemistry 18 (2010) 1591–1598
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Binding of curcumin and its long chain derivatives to the activator
binding domain of novel protein kinase C
*
Anjoy Majhi, Ghazi M. Rahman, Shyam Panchal, Joydip Das
Department of Pharmacological and Pharmaceutical Sciences, College of Pharmacy, University of Houston, Houston, TX 77204, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Protein kinase C (PKC) is a family of serine/threonine kinases that play a central role in cellular signal
transduction. The second messenger diacylglycerol having two long carbon chains acts as the endogenous
ligand for the PKCs. Polyphenol curcumin, the active constituent of Curcuma longa is an anti-cancer agent
and modulates PKC activity. To develop curcumin derivatives as effective PKC activators, we synthesized
several long chain derivatives of curcumin, characterized their absorption and fluorescence properties
and studied their interaction with the activator binding second cysteine-rich C1B subdomain of PKCd,
Received 24 November 2009
Revised 24 December 2009
Accepted 31 December 2009
Available online 6 January 2010
Keywords:
Curcumin
Protein kinase C
Fluorescence
EC50
PKC
PKCdC1B, PKC
13-acetate (TPA). The EC50s of the curcumin derivatives for fluorescence quenching varied in the range of
4–11 M, whereas, EC50s for TPA varied in the range of 3–6 M. Fluorescence emission maxima of 1 and 4
e
and PKCh. Curcumin (1) and its C16 long chain analog (4) quenched the intrinsic fluorescence of
e
C1B and PKChC1B in a manner similar to that of PKC activator 12-O-tetradecanoylphorbol
l
l
were blue shifted and the fluorescence anisotropy values were increased in the presence of the C1B
domains in a manner similar to that shown by the fluorescent analog of TPA, sapintoxin-D, confirming
that they were bound to the proteins. Molecular docking of 1 and 4 with novel PKC C1B revealed that both
the molecules form hydrogen bonds with the protein residues. The present result shows that curcumin
and its long chain derivatives bind to the C1B subdomain of novel PKCs and can be further modified struc-
turally to improve its binding and activity.
Molecular docking
Anisotropy
Activator
Ligand
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
mains. Atypical PKCs contain a single C1 domain that does not bind
DAG. Along with the PKC family, there are six additional families of
Protein kinase C (PKC) is a family of serine/threonine protein ki-
nases involved in the regulation of various aspects of cell functions,
including cell growth, differentiation, metabolism, and apoptosis.1
During the past three decades, it has become clear that PKC iso-
forms play important role in the pathology of several diseases such
as cancer, diabetes, stroke, heart failure, and Alzheimer’s disease.2–8
Therefore, PKC has been a subject of intensive research and drug
development,9 particularly in cancer research.
proteins that contain a DAG-responsive C1 domain.12,13 The C1 do-
mains have become an attractive target in designing the PKC based
drugs. Recently it has been found that alcohol and anesthetics also
bind to the PKC C1 domains.14–16
Several classes of high-affinity ligands that target the DAG bind-
ing C1 domain have been previously described.17 Naturally occur-
ring tumor promoters, phorbol esters, were the first ligands that
were found to bind to the C1 domain of PKC.18 The phorbol ester
binding site in PKCd C1B domain has been characterized by X-ray
crystallography.19 In addition to phorbol esters, naturally occurring
C1 domain ligands include bryostatins, teleocidins, aplysiatoxins,
ingenols, and iridals.18 Most of these C1 domain ligands from nat-
ural sources are highly complex in their chemical structure. Indo-
lactam and benzolactam derivatives act as selective activators of
novel PKC isoenzymes although they are also laborious to synthe-
size and modify.20,21 Therefore it is essential to find simpler tem-
plate for ligand synthesis, whose structure could be easily
modified and fine-tuned in order to achieve selectivity. Using this
concept very recently it was found that isophthalic acid derivatives
bind to the C1 domain and modulate PKC.18
The PKC family has been divided into three main groups: con-
ventional isoforms (
glycerol (DAG) for activation; novel isoforms (d,
that require only DAG and atypical isoforms (f, and k) that require
a, bI, bII and c
) that require Ca2+ and diacyl-
e, g, h and l)
i
neither Ca2+ nor DAG.10 DAG is a second messenger which is gen-
erated by the phospholipase C-catalyzed hydrolysis of membrane
phosphatidylinositol-4,5-bisphosphate (PIP2)11 and selectively
interacts with proteins containing a C1 domain and induces their
translocation to discrete subcellular compartments. In the classical
and novel PKC isoenzymes, the DAG-sensitive C1 domain is dupli-
cated into a tandem C1 domain consisting of C1A and C1B subdo-
Natural polyphenol curcumin (1), the active constituent of Cur-
* Corresponding author. Tel.: +1 713 743 1708; fax: +1 713 743 1884.
E-mail address: jdas@uh.edu (J. Das).
cuma longa, is one of the best studied natural compounds.22,23
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.12.075

1592
A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
Curcumin has been used as a spice to give a specific flavor and yel-
low color to curry, which is consumed daily by millions of people.
Curcumin has been used as traditional medicine for liver disease
(jaundice), indigestion, urinary tract diseases, rheumatoid arthritis
and insect bites.23 This phytochemical has also demonstrated both
anti-cancer and anti-angiogenic properties. Its anti-tumor proper-
ties include growth inhibition and apoptosis induction in a variety
of cancer cell lines in vitro, as well as the ability to inhibit tumero-
genesis in vivo.24–30 Curcumin has also been shown to modulate
the activity of protein kinases,31,32 membrane ATPases33–36 and
transcription factors.37,38 The positive interference of curcumin
with the tumor promoting effects of phorbol esters has presumably
been attributed to its effect of curcumin on the phorbol ester
receptor, PKC.39 There is also well documented modulation of
PKC activity by curcumin in vivo31 and in vitro40 using mem-
brane-free systems. This specificity suggests that curcumin inter-
acts with common domain(s) on target proteins. However, the
molecular mechanism behind the physiological effects of curcumin
is not well understood.
therefore prepared derivatives having different side chains in order
to investigate the length and shape of the alkyl chains required for
binding with the C1B subdomain of PKC. The crystal structure of
the ligand bound C1B domain of PKCd, a novel PKC is known.
Therefore we limited our study only to the C1B domain of novel
PKCs.
2.1. Curcumin and its derivatives
We synthesized the long chain derivatives of curcumin because
of the fact that the physiological PKC activator diacylglycerol con-
tains two oleoyl long chains. The synthesis of the derivatives of
curcumin (Fig. 1) was started from curcumin (1), which was ob-
tained by recrystallization of commercially available (purity
75–80%) curcumin (Aldrich, Inc.).
Curcumin (1), possesses two phenolic OH groups and an
a,b-unsaturated-b-diketone moiety in its chemical structure. The
b-diketone structure can undergo keto–enol tautomerism in solu-
tions.42,43 The relative contributions of the keto and enolic tautom-
ers depend on several factors such as solvent characteristics,
temperature and substitution on curcumin. In general, many of
the diketones exist in solutions predominantly in the enolic form
at room temperature.42,44 The enolic form can exist in different
cis and trans isomeric forms depending on the temperature, polar-
ity or hydrogen bonding nature of the solvents.45,46 The cis-enolic
form should be energetically more stable because of strong intra-
molecular hydrogen bonding. In the present studies, all com-
pounds were found to exist in the enol form because of its
higher thermodynamic stability.
In the present study, we describe design, synthesis and binding
properties of curcumin and its derivatives to the C1B subdomains
of PKCd, PKC
molecular modeling studies. Curcumin (1) and its C16 long chain
derivative (4) bind to the protein with the EC50 of 4–11 M,
whereas, PKC activator 12-O-tetradecanoylphorbol-13-acetate
(TPA) binds with EC50 of 3–6 M under the similar experimental
e and PKCh using fluorescence spectroscopy and
l
l
condition. Fluorescence emission maxima of 1 and 4 were blue
shifted and anisotropy values were increased in the presence of
proteins in a manner similar to that of TPA. In agreement with
the binding results, the molecular modeling indicated that the cur-
cumin derivatives form hydrogen bonds with the residues at the
activator binding site of the proteins.
2.2. Spectral characteristics
Curcumin is poorly soluble in water and shows differential sol-
ubility in several organic solvents. Its solubility is high in polar or-
ganic solvents but it is only sparingly soluble in aliphatic organic
solvents like hexane. Absorption maxima of curcumin (1) was sol-
vent dependent and large red shift was observed in more polar sol-
vents, for example, the maxima are 404 nm and 427 nm in hexane
and methanol, respectively (Table 1). Demethylated compound (2)
also showed red shift in ethanol (432 nm) and water (426 nm)
compared to hexane (406 nm). The same trend was observed for
the long chain derivatives 3–6. Addition of long chain to curcumin
did not show any major changes in the absorption maxima in ace-
tonitrile and hexane. In ethanol and water, however addition of a
single chain shifted the absorption maxima towards blue as com-
2. Results and discussion
We selected curcumin for our studies because it is commer-
cially available, easily derivatizable and known to modulate PKC
activity.36,41 The compound also contains two of the phorbol ester
pharmacophores, namely the hydroxyl and the carbonyl function-
alities within the same molecule. To investigate the hydrophobic-
ity of the compounds that was necessary for binding, we also
prepared one additional hydrophilic derivative and compared its
binding properties with that of the long chain hydrophobic com-
pounds. Because it is unclear how C1 domain interacts with the
membrane, hydrophobic interactions are difficult to model. We
O
OH
OH
OH
HO
HO
2
O
O
O
OH
O
OH
OMe
OH
MeO
HO
OMe
MeO
H33C16
OMe
MeO
HO
OC16H33
O
OC8H17
3
6
1
O
OH
OH
O
OMe
MeO
HO
OMe
OC8H17
MeO
OC16H33
H17C8O
5
4
Figure 1. Curcumin (1) and its analogues (2–6).

A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
1593
Table 1
Absorptiona and fluorescenceb properties of curcumin and its derivatives in different solvent at 25 °C
Compound
Absorbance maximum (kabs), nm
Emission maximum (kem), nm
EtOH
CH3CN
Hexane
Water
EtOH
CH3CN
Hexane
Water
1c
2
3
427(61,864)
432(51,818)
421(46,227)
421(44,855)
419(ND)
417(66,060
416(55,109)
417(51,482)
417(48,055)
419(ND)
404(65,136)
406(35,700)
408(54,745)
409(53,427)
408(ND)
425(23,800)
426(31,082)
405(22,245)
389(19,900)
406(ND)
560
540
542
543
539
544
518
525
519
535
522
536
474, 449
494, 470
473, 445
472, 443
474, 445
476, 446
572
560
581
590
512
514
4
5d
6d
418(ND)
419(ND)
408(ND)
280, 318(ND)
ND, not determined.
a
Concn, 2 ꢂ 10ꢁ6 M. Molar extinction coefficient values in Mꢁ1 cmꢁ1 are in parentheses.
b
c
Concn, 10 ꢂ 10ꢁ6 M in water and 2 ꢂ 10ꢁ6 M in other solvents.
Extinction coefficient of curcumin (Sigma). Different values reported in Refs. 45,46.
Extinction coefficient not determined due to poor solubility.
d
pared to curcumin. Addition of the second chain blue shifted the
absorption maxima further (Table 1). Figure 2A shows the absorp-
tion spectra of curcumin derivative 4 recorded in different sol-
vents. A strong and intense absorption band was observed in the
350–450 nm wavelength regions. The largest red shifts were ob-
served in polar solvents capable of forming hydrogen bond. In
non-polar solvents, as proposed for simple b- diketones, curcumin
may exist in keto–enol equilibrium. The enol forms are most solu-
ble in polar solvent than non-polar solvent, and its contribution
may result in blue shifted absorption maxima. The presence of long
aliphatic chains can make the environment more non-polar. There-
fore, with increasing the number of the hexadecyl aliphatic chain
in curcumin, absorption maxima shifted more towards blue. The
extinction coefficient of curcumin47,48 and its derivatives in most
of these solvents is in the range ꢀ19,900–66,060 Mꢁ1 cmꢁ1
(Table 1). Addition of two chains decreased the solubility of curcu-
min drastically. Therefore, determination of the extinction coeffi-
cient was not possible for 5 and 6.
The curcumin and its derivatives also showed significant sol-
vent-dependent shifts in their fluorescence emission maxima
(Table 1). The emission spectra of curcumin and its derivatives
(1–6) in hexane were significantly different from other solvents
showing two fluorescence maxima at 449 nm and 474 nm. In hex-
ane, for compounds with single or double chains, there was not
much difference in the absorption spectra. In aprotic solvents like
acetonitrile, curcumin and its derivatives showed emission max-
ima in the region of 518–536 nm. In these solvents, the polarity af-
fected the intramolecular hydrogen bonding. In polar solvents like
alcohols, addition of long chain derivatives showed blue shifted
emission maxima from 560 nm to 539–544 nm. These solvents,
due to intermolecular hydrogen bonding, may induce change in
conformations and thereby shifting the emission maxima towards
blue. Thus the solvent dependent fluorescence spectral shifts con-
firmed that the photo-excitation of curcumin and its derivatives
were influenced both by the intramolecular and intermolecular
proton transfer reactions. In water, curcumin and its derivative
showed much lower fluorescence intensity compared to the organ-
ic solvents indicating lower quantum yields in water. Emission
spectra were also very broad in water. Figure 2B represents the
emission spectra of 4 in different solvents. The fluorescence maxi-
mum red shifted from 472 nm and 443 nm in hexane to 590 nm in
water. For compounds 3 and 4, the emission maxima followed the
order: water > ethanol > acetonitrile > hexane. Addition of a second
chain showed large blue shift in the emission maxima at 512 nm
and 514 nm in compounds 5 and 6, respectively and followed the
order: ethanol > acetonitrile > hexane > water.
2.3. Binding of curcumin and its derivatives with PKC C1B
To determine the binding affinity of curcumin and its deriva-
tives with the PKC C1B subdomains, we used fluorescence quench-
ing technique and measured the EC50 for each of the compounds.
1.0
1.0
A
B
a
c
d
b
c
d
b
a
0.8
0.6
0.4
0.2
0.0
0.8
0.6
0.4
0.2
0.0
300
350
400
Wavelength(nm)
450
500
450
500
550
600
650
Wavelength(nm)
Figure 2. (A) Effect of solvent polarity on the absorption properties of curcumin and its derivatives. Normalized absorption spectra of (1E,4Z,6E)-7-(4-(hexadecyloxy)-3-
methoxyphenyl)-5-hydroxy-1-(4-hydroxy-3-methoxyphenyl)hepta-1,4,6-trien-3-one (4), 2 ꢂ 10ꢁ6 M in (a) water, (b) hexane, (c) acetonitrile and (d) ethanol. (B) Effect of
solvent on the emission properties of curcumin and its derivatives. Normalized fluorescence emission spectra of (1E,4Z,6E)-7-(4-(hexadecyloxy)-3-methoxyphenyl)-5-
hydroxy-1-(4-hydroxy-3-methoxyphenyl)hepta-1,4,6-trien-3-one (4) in (a) water, (b) hexane, (c) acetonitrile and (d) ethanol. The concentration of 4 is 10 ꢂ 10ꢁ6 M in water
and 2 ꢂ 10ꢁ6 M in other solvents.

1594
A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
1.0
0.8
0.6
0.4
0.2
0.0
100
d
b
80
60
40
c
a
5
10
15
20
25
30
20
[Ligand], µM
Figure 3. Binding of curcumin and its derivatives with PKCd C1B. Plot of
fluorescence intensity of PKCd C1B (2 M) in buffer (50 mM Tris, 150 mM NaCl,
2 mM DTT, 50 M ZnSO4, pH 7.2) in the presence of varying concentration of 1
(filled triangle), (filled circle), and TPA (filled square), where and F0 are
fluorescence intensities in the presence and absence of the ligand, respectively.
Solid lines indicate the fit using Hill equation. The corresponding EC50 of 1, 4 and
l
l
4
F
500 520 540 560 580 600 620 640 660
Wavelength (nm)
Figure 4. Blue shift in emission maxima of 1 and 4 (5 ꢂ 10ꢁ6 M) in the presence of
PKCd C1B (50 ꢂ 10ꢁ6 M). Emission maximum of (a) 1 in buffer, (b) 1 in the presence
of PKCd C1B, (c) 4 in buffer, (d) 4 in the presence of PKCd C1B. Buffer used, 50 mM
TPA are 10.6, 8.9 and 5.1 lM and corresponding Hill coefficients are 1.6, 2 and 1.8,
respectively. Excitation wavelength used was 280 nm.
Tris, 150 mM NaCl, 2 mM DTT, 50 lM ZnSO4, pH 7.2.
Table 2
EC50 values for the binding of curcumin and its derivatives and PKC activators with
long chain derivative (4) showed two possible hydrogen bonds and
the C score of 2 when docked into delta (Fig 5, Table 5). In theta,
however, there was an additional hydrogen bond, although the C
score value was 5 for both. No significant difference was observed
for 3 having eight carbon chains and 4 having sixteen carbon
chains. Demethylated compound 2 showed improved binding with
theta and epsilon as compared to delta. For epsilon and theta, the
long chain derivative 4 showed comparable binding to that of DiC8,
but showed higher binding than that of DiC18. On the other hand, in
delta, it showed comparable binding to that of DiC18, but showed
lowed binding than that of DiC8.
the C1Ba subdomains of PKC
e
, PKCd and PKCh measured by fluorescence quenching
EC50 M)
Epsilon (
Compound
(l
Delta (d)
e
)
Theta (h)
1
2
3
4
10.67 0.35
12.50 0.22
8.69 0.22
8.98 0.25
5.17 0.28
16.15
8.81 0.21
11.14 0.21
9.79 0.20
10.64 0.31
5.56 0.11
8.19 0.30
14.27 0.77
11.08 0.76
6.57 0.71
4.00 0.0
5.95 0.21
3.19 0.56
5.24 0.51
8.67 2.13
TPA
DiC8
DiC18
7.49 0.59
a
Protein concentration is 2 lM in buffer (50 mM Tris, 150 mM NaCl, 2 mM DTT,
To further investigate the binding of curcumin (1) and its long
chain derivative (4) with the PKC C1B subdomains, the effect of
proteins on their emission maxima and fluorescence anisotropies
were studied and compared with the fluorescent PKC activator
sapintoxin-D (SAPD). In the presence of 10-fold excess each of
the PKC C1B subdomains, the emission maxima of 1 and 4 were
blue shifted. The emission maximum of 1 shifted from 572 nm in
buffer to 559 nm in the presence of PKCd C1B, and for 4 it shifted
from 590 nm in buffer to 560 nm. SAPD also showed blue shift
from 442 nm in buffer to 423 nm in the presence of the C1B subdo-
main. Emission maxima of curcumin 1 and its derivative 4 also
were blue shifted in the presence of epsilon and theta (Fig. 4,
Table 3). Among the three compounds studied, 4 showed largest
blue shift of 30 nm indicating strongest interaction amongst the
three. These observations indicated that addition of long chain im-
proved the binding affinity.
50 lM ZnSO4, pH 7.2).
The results were compared with the values obtained for known
PKC activator diacylglycerol (DAG) such as 1,2-dioleoyl-sn-glycerol
(DiC18) and 1,2-dioctanoyl-sn-glycerol (DiC8) and phorbol ester
such as 12-O-tetradecanoylphorbol-13-acetate (TPA).18,19,49 Curcu-
min and its derivatives quench the protein intrinsic fluorescence in
a concentration dependent manner similar to those of DAG and
phorbol esters. The intrinsic fluorescence of the PKC C1B subdo-
mains is due to the presence of a single tryptophan (Trp-252 in del-
ta, Trp-264 in epsilon and Trp-253 in theta) and tyrosine residues
(Tyr-236 and Tyr-238 in delta, Tyr-250 in epsilon and Tyr-249 and
Tyr-251 in theta). Figure 3 shows the plot of fluorescence quench-
ing data for PKCd C1B in the presence of different concentrations of
1, 4, and TPA. The EC50 values determined by fitting the fluores-
cence data using Hill equation are shown in Table 2. Among all
the compounds studied, TPA was found to bind to the C1B subdo-
Further, the observation that the fluorescence anisotropy values
of 1 and 4 increased in the presence of proteins supported that 1
and 4 bind to the proteins. For 1, anisotropy changed from
main with highest affinity having the EC50 in the range of 3–6
Among the three PKC subtypes, the binding affinity was slightly
higher for the PKCh C1B as compared to the PKCd C1B and PKC
l
M.
0.3994 in buffer to 0.4341 in PKC
0.4407 in PKCh C1B, respectively, whereas for 4, the changes were
from 0.2638 in buffer to 0.3355 in PKC C1B, 0.3117 in PKCd C1B
e C1B, 0.4501 in PKCd C1B and
e
e
C1B. For all the proteins, binding affinity for curcumin was less
compared to TPA. This can be explained by inspecting the molecu-
lar model of the ligand bound protein that showed there were five
possible hydrogen bonds between phorbol 13-OAc and PKCd C1B,
on the contrary, only two hydrogen bonds were possible for curcu-
min (Fig. 5). Addition of the long chain to the curcumin moiety im-
proved its affinity for the delta and theta. Both curcumin (1) and its
and 0.3206 in PKCh C1B (Table 4). In the presence of 10-fold excess
of proteins, the anisotropy value of SAPD increased from 0.072 in
buffer to 0.1142 in PKCe C1B, 0.1051in PKCd C1B and 0.1163 in
PKCh C1B. Although the value of anisotropy and the magnitude
of the increment were different for the curcumin and its deriva-
tives and SAPD, this experiment clearly showed that in the pres-
ence of the proteins the curcumin and its derivatives experienced

A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
1595
Table 3
Emmison maxima (kem in nm) of 1, 4 and SAPD in the presence of PKC C1B domains
3. Conclusion
Compound
Buffer
PKCd
PKC
e
PKCh
We described here the synthesis, spectral properties and bind-
ing of curcumin and its derivatives to the C1B subdomains of novel
protein kinase C. Our results showed that curcumin and its long
chain derivatives can bind to the activator binding domain of
PKC by forming hydrogen bonds with the residues at the activator
binding site. Our results also indicated that curcumin and its deriv-
atives can influence PKC activation and its membrane translocation
properties differently depending on the nature of the PKC subtype.
In contrast to phorbol esters, which are tumorogenic, curcumin is
non toxic and has anti-cancer properties. Therefore, development
of suitable curcumin derivative as target for specific PKC isozyme
has high potential to be used as drug. On the basis of the results
presented here, we are currently working towards improving the
affinity and selectivity of the curcumin derivatives for the C1 do-
main containing proteins.
1a
572
590
442
559
560
423
562
560
426
561
561
427
4a
SAPDb
Compounds 1 and 4, 5 ꢂ 10ꢁ6 M and protein, 50 ꢂ 10ꢁ6 M.
a
SAPD, 0.5 ꢂ 10ꢁ6
M
and protein, 5 ꢂ 10ꢁ6 M; Spectra were recorded after
b
incubating the compound for 1 h at 25 °C.
restricted motion in the protein environment by binding to it in a
manner similar to that of SAPD. Similar increase in anisotropy of
curcumin was observed when it bound to serum albumins.47,50
Usually the dissociation constants (Kd) for the PKC and its li-
gands are determined using radioactive phorbol ester binding as-
says in presence of membrane.51 In this case however, we used a
lipid free system that also showed the binding of the ligands.
Although the binding constant values are expected to be different
in the presence of lipid, our fluorescence binding assay clearly indi-
cate that curcumin and its derivatives can bind to the C1 domain of
PKC. These results further indicated that curcumin derivatives
interact differently with the three different PKCs and suitably mod-
ified curcumin derivative could bind to the C1 domain and modu-
late PKC enzymes specifically.
4. Experimental
4.1. General
Sapintoxin-D (SAPD) was purchased from LC Laboratories,
Woburn, MA. Curcumin and TPA were from Sigma. 1,2-dioleoyl-
sn-glycerol (DiC18), 1,2-dioctanoyl-sn-glycerol (DiC8) were pur-
chased from Avanti Polar Lipids. Solvents were purchased from
VWR and Fisher. All other reagents were purchased from Sigma–
Aldrich and used without further purification. Progress of chemical
reaction was monitored through thin layer chromatography (TLC)
on pre-coated glass plates (Silica Gel 60 F254, 0.25 mm thickness)
purchased from EMD chemicals. 1H NMR and 13C NMR spectra
were recorded on a GE QE-300 spectrometer. Unless otherwise
specified, all NMR spectra were obtained in deuterated chloroform
(CDCl3) and referenced to the residual solvent peak; chemical
shifts were reported in parts per million (ppm), and coupling con-
stants in hertz (Hz). Multiplicities were reported as follows: s (sin-
glet), d (doublet), t (triplet), m (multiplet) and br (broadened).
Mass spectra were obtained on either a VG 70-S Nier Johnson or
JEOL Mass Spectrometer. Absorption spectra were recorded on Hit-
achi U-2910 spectrophotometer and fluorescence spectra were re-
corded on PTI LPS 220B.
In the crystal structure19 of phorbol ester (phorbol-13-O-ace-
tate) bound PKCd C1B, the hydroxyl groups attached to C20 and
C4 and the carbonyl group on C3 formed hydrogen bonds with
the protein residues. Phorbol ester hydroxyl group (C-20) was
hydrogen bonded to the backbone amide proton of Thr-242 and
the carbonyls of Thr-242 and Leu-251. The C3 carbonyl group
formed hydrogen bond with the backbone amide proton of
Gly-253. Another hydrogen bond was observed with C-4 hydroxyl
group with the backbone carbonyl of Gly-253 (Fig. 5A). We con-
ducted molecular docking experiments to verify if the phorbol es-
ter-like hydrogen bonding with protein residues was feasible for
curcumin and its derivatives. Our docked model with dC1B domain
showed that hydroxyl group of 1 is hydrogen bonded to the car-
bonyl of backbone Leu-251. Another hydrogen bond was observed
between –OMe of 1 and amide proton of Gln-257 (Fig. 5B). Docked
model of 4 with dC1B domain showed hydrogen bond between hy-
droxyl group of 4 and backbone hydroxyl of Ser-240. Another
hydrogen bond was observed between the enol of 4 and backbone
carbonyl of Met-239 (Fig. 5C). Molecular docking of curcumin and
its derivatives into the PKC C1B subdomains was carried out using
Surflex dock module of Sybyl 7.3. The C score values are presented
in Table 5, where higher C score value indicates higher binding.
When phorbol 13-OAc was docked into the unliganded PKCdC1B,
four hydrogen bonds were observed with a C score value of 3, in
contrast to five hydrogen bonds in the crystal structure (Fig. 5).
The observation that the C score values obtained from the models
do not always corroborate with the experimental binding data
indicate the fact that both the proteins and ligands can undergo
conformational changes in solutions.
Table 5
C score values obtained from the docking of curcumin (1) and its derivatives (2–4),
TPA, DiC8 and DiC18 into the PKC C1B domains using Surflex dock module of Sybyl 7.3
Compound
d C1B
e
C1B
h C1B
1
2
3
4
2
3
4
2
3
2
3
4
5
4
4
2
4
2
5
5
3
5
3
5
3
Phorbol-13-OAc
DiC8
DiC18
Table 4
Anisotropy valuesa of 1, 4 and SAPD in the presence and absence of the C1B domain of PKC
e
, PKCd and PKCh at 25 °C
Compound
Bufferc
d C1B
e
C1B
h C1B
1b
0.3994(0.0066)
0.2638(0.0060)
0.0720(0.0018)
0.4501(0.0068)
0.3117(0.0056)
0.1051(0.0014)
0.4341(0.0044)
0.3355(0.0077)
0.1142(0.0024)
0.4407(0.0072)
0.3206(0.0072)
0.1163(0.0033)
4b
SAPDc
a
b
c
Values in the parentheses indicate standard deviations.
Compounds 1 and 4, 5 ꢂ 10ꢁ6 M; protein, 50 ꢂ 10ꢁ6 M in buffer (50 mM Tris, 150 mM NaCl, 2 mM DTT, 50
lM ZnSO4, pH 7.2). Incubation was done for 30 min at 25 °C.
SAPD, 0.5 ꢂ 10ꢁ6 M; protein, 5 ꢂ 10ꢁ6 M in the same buffer.

1596
A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
Figure 5. Structures of ligand bound PKC dC1B. (A) Crystal structure of phorbol 13-O-acetate bound PKCd C1B; (B) modeled structure of curcumin (1) docked into PKCd C1B;
(C) modeled structure of long chain curcumin derivative (4) docked into PKCd C1B. The modeled structures are generated using the autodock module of Sybyl 7.3. The oxygen
atoms in the ligand structures are shown in red. The dotted line indicates possible hydrogen bonds, small spheres indicate Zn atoms.
4.2. General procedure for preparation of 2
J = 2.0 Hz), 7.00 (1H, d, J = 8.2 Hz), 6.94 (1H, d, J = 8.2 Hz), 6.59
(1H, d, J = 16.2 Hz), 6.54 (1H, d, J = 16.2 Hz), 5.80 (1H, s), 5.88
(1H, br), 4.04 (2H, t, J = 7.0 Hz), 3.93 (3H, s), 3.91 (3H, s), 1.85
(2H, m), 1.23–1.45 (10H, m), 0.88 (3H, t, J = 6.0 Hz); 13C NMR
(CDCl3): d 183.3, 183.2, 150.7, 149.4, 146.7, 140.5, 140.4, 127.7,
122.8, 122.6, 121.7, 114.8, 112.3, 110.1, 109.5, 101.2, 69.0, 56.0,
55.9, 31.7, 29.3, 29.1, 29.0, 25.9, 22.64, 14.0; ESI-MS m/z 481
[M+H]+.
Under nitrogen atmosphere, a suspension of curcumin (368 mg,
1 mmol) in 15 mL of anhydrous dichloromethane was cooled to
ꢁ78 °C using a dry ice/acetone bath. To this stirred suspension
was added 0.57 mL (6 equiv) boron tribromide. After 20 min the
cooling bath was removed and solution was left to stir for 5 h.
The reaction mixture was then carefully poured into saturated so-
dium bicarbonate solution with stirring. Water and dichlorometh-
ane layers were separated and the water layer was extracted twice
with diethyl ether. The dichloromethane layer was removed by ro-
tary evaporator and the resulting solid redissolved in diethyl ether.
The combined ether layer was washed with water until the ex-
tracts were neutral and dried over anhydrous sodium sulfate. After
filtration, the diethyl ether was removed under reduced pressure
using a rotary evaporator. The compound was purified by column
chromatography (hexane/EtOAc/MeOH; 60:38:2 %).
4.3.2. (1E,4Z,6E)-7-(4-(Hexadecyloxy)-3-methoxyphenyl)-5-
hydroxy-1-(4-hydroxy-3-methoxyphenyl)hepta-1,4,6-trien-
3-one
Yield 49%; 1H NMR (CDCl3): d 7.59 (2H, d, J = 16.0 Hz), 7.12 (2H,
dd, J = 8.0, 2.0 Hz), 7.09 (1H, dd, J = 8.0, 2.0 Hz), 7.04 (1H, d,
J = 2.0 Hz), 6.94 (1H, d, J = 8.0 Hz), 6.84 (1H, d, J = 8.0 Hz), 6.50 (1H,
d, J = 16.2 Hz), 6.45 (1H, d, J = 16.2 Hz), 5.84 (1H, br s), 5.79 (1H, s),
4.03 (2H, t, J = 7.1 Hz), 3.93 (3H, s), 3.90 (3H, s), 1.84 (2H, m),
1.24–1.44 (26H, m), 0.86 (3H, t, J = 6.2 Hz); 13C NMR (CDCl3): d
183.5, 183.3, 150.7, 149.4, 141.3, 140.5, 129.9, 129.2, 122.8, 122.6,
122.4, 121.7, 115.9, 114.7, 112.3, 110.3, 100.5, 69.0, 56.0, 53.1,
31.9, 29.6, 29.5, 29.3, 29.0, 25.9, 23.0, 22.6, 14.0; ESI-MS m/z 593
[M+H]+.
4.2.1. (1E,4Z,6E)-1,7-bis(3,4-dihydroxyphenyl)-5-
hydroxyhepta-1,4,6-trien-3-one
Yield 80%; 1H NMR (CD3OD): d 7.58 (2H, d, J = 16.5 Hz), 7.23
(2H, d, J = 2.0 Hz), 7.12 (2H, dd, J = 8.0, 2.0 Hz), 6.92 (2H, d,
J = 8.0 Hz), 6.65 (2H, d, J = 16.5 Hz), 6.02 (1H, s); ESI-MS m/z 341
[M+H]+.
4.3.3. (1E,4Z,6E)-5-Hydroxy-1,7-bis(3-methoxy-4-(octyloxy)phenyl)-
hepta-1,4,6-trien-3-one
4.3. General procedure for preparation of compounds 3, 4, 5
and 6
Yield 52%; 1H NMR (CDCl3): d 7.60 (2H, d, J = 15.6 Hz), 7.10 (2H,
dd, J = 8.1, 1.8 Hz), 7.07 (2H, d, J = 1.8 Hz), 6.87 (2H, d, J = 8.1 Hz),
6.48 (2H, d, J = 15.6 Hz), 5.81 (1H, s), 4.05 (4H, t, J = 6.9 Hz), 3.91
(3H, s), 1.85 (4H, m), 1.25–1.49 (20H, m), 0.88 (6H, t, J = 6.3 Hz);
13C NMR (CDCl3): d 183.2 (2), 150.69, 149.4, 140.45, 127.8, 122.6,
121.8, 112.3, 110.1, 101.2, 69.0, 56.0 (2), 34.2, 31.7, 29.3, 29.1,
29.0, 25.9, 22.6, 14.0; ESI-MS m/z 593 [M+H]+.
Bromooctane (1 equiv for 3 and 2 equiv for 5) or bromohexa-
decane (1 equiv for 4 and 2 equiv for 6) were added to a stirred
solution of curcumin (368 mg, 1 mmol) and K2CO3 (1 equiv for 3
and 4 and 2 equiv for 5 and 6) in dry acetone (10 ml) and the mix-
ture was refluxed for 24 h. After cooling the mixture to room tem-
perature and filtering, solvent was removed in vacuo. The resulting
residue was subjected to column chromatography (n-hexane–
EtOAc) to purify the corresponding products.
4.3.4. (1E,4Z,6E)-1,7-Bis(4-(hexadecyloxy)-3-methoxyphenyl)-
5-hydroxyhepta-1,4,6-trien-3-one
Yield 59%; 1H NMR (CDCl3): d 7.60 (2H, d, J = 15.6 Hz), 7.12 (2H,
dd, J = 8.1, 1.8 Hz), 7.07 (2H, d, J = 1.8 Hz), 6.87 (2H, d, J = 8.1 Hz),
6.48 (2H, d, J = 15.6 Hz), 5.81 (1H, s), 4.04 (4H, t, J = 6.9 Hz), 3.91
(6H, s), 1.85 (4H, m), 1.23–1.41 (52H, m), 0.87 (3H, t, J = 6.3 Hz);
13C NMR (CDCl3): d 183.2 (2), 150.6, 149.4, 140.4, 127.8, 122.6,
4.3.1. (1E,4Z,6E)-5-Hydroxy-1-(4-hydroxy-3-methoxyphenyl)-7-
(3-methoxy-4-(octyloxy)phenyl)hepta-1,4,6-trien-3-one
Yield 56%; 1H NMR (CDCl3): d 7.65 (2H, d, J = 16.2 Hz), 7.20 (2H,
dd, J = 8.2, 2.0 Hz), 7.15 (1H, dd, J = 8.2, 2.0 Hz), 7.12(1H, d,

A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
1597
121.8, 112.3, 110.1, 101.3, 69.0, 55.9 (2), 31.9, 29.6 (2), 29.5 (2),
docking studies. The average structure from the combined 20
structures for the PKChC1B was selected using INSIGHT II.
29.4, 29.3, 29.0, 25.9, 22.6, 21.0, 14.1; ESI-MS m/z 816 [M]+.
Homology model for PKCeC1B has been generated using UNI-
4.4. Bacterial expression and purification of the PKC
and hC1B subdomains
e
C1B, dC1B
PROT (http://www.uniprot.org/) and Expasy SWISS Model work-
space (http://swissmodel.expasy.org/workspace/), the web based
tools for the automatic homology model generation. PKCh C1B
(PDB code: 2ENZ) has nearly 66% sequence homology with
The PKC C1B subdomains fused with glutathione S-transferase
(GST) were expressed in BL21(DE3) gold E. coli and purified as de-
scribed earlier.16 Briefly, cell pellets were treated with 1% Triton X-
100 and lysozyme (1 mg/ml), followed by sonication and centrifu-
gation. The clarified supernatant was then applied to a glutathione-
Sepharose column. The bound protein was thoroughly washed
with phosphate buffer saline, released by thrombin cleavage,
eluted with phosphate-buffered saline, and concentrated by
ammonium sulfate precipitation (80%). Proteins were further puri-
fied by fast performance liquid chromatography (Akta Purifier)
using a Superdex™ 75 column (GE Healthcare Biosciences), a mo-
bile phase of 50 mm Tris, 100 mm NaCl, pH 7.2, and a flow rate of
0.5 ml/min.
PKCeC1B and thus used as a template. Energy minimization
(ꢁ1741.670 kJ/mol) was done using the same program. The model
was validated using VERIFY3D and all the amino acids residues had
an acceptable score above zero. The model space analysis for
PKCeC1B was done using the Ramachandran Plot (http://dic-
soft1.physics.iisc.ernet.in/rp/index.html). The plot indicated that
100% residues were within the allowed region (52.08% in the Fully
Allowed Region (FAR), 35.42% in the Additional Allowed Region
(AAR) and 12.50% were in the Generously Allowed Region (GAR))
thereby, validating the model.
Molecular docking was performed on Surflex module of Sybyl 7.3
using Threshold-0.5, Bloat-2.0 and Radius-3 Å for the protomol gen-
eration. Residues Tyr-239, Lys-240, Ser-241, Pro-242, Thr-243, Phe-
244, Leu-251, Leu-252, Trp-253, Gly-254, Leu-255 and Glu-258 of
PKC theta; Tyr-238, Met-239, Ser-240, Pro-241, Thr-242, Phe-243,
Leu-250, Leu-251, Trp-252, Gly-253, Leu-254 and Gln-257 for PKC
delta and Tyr-250, Lys-251, Val-252, pro-253, Thr-254, Phe-255,
Leu-262, Leu-263, Trp-264, Gly-265, Leu-266 and Gln-269 for PKC
epsilon were used. These residues were selected by comparing the
PKCactivator phorbolester bindingsitein PKC deltaC1B.53 Ringflex-
ibility and Post Dock energy minimization were applied on each
structure. Higher C-score values represent better fitting.
4.5. Fluorescence studies
Fluorescence spectra were recorded on PTI LPS 220B equipped
with temperature and stirring control systems. A 1.5-ml cuvette
(Hellma) with a Teflon stopper was used for fluorescence measure-
ments. For fluorescence quenching experiments, protein (2
and varying concentration of ligands (2–34 M) were incubated
in a buffer solution (50 mM Tris, 150 mM NaCl, 2 mM DTT,
50 M ZnSO4, pH 7.2) at 25 °C. Protein was excited at 280 nm
lM)
l
l
and emission spectra were recorded from 300 nm to 650 nm. Fluo-
rescence intensity data, (F0 ꢁ F)/F were plotted against the ligand
concentration to generate the binding curves, where F and F0 rep-
resented the fluorescence intensity at 350 nm in the presence and
in the absence of ligand, respectively. For EC50 measurement, all
curves were fitted with the Hill equation using Igor Pro 4. Effect
of proteins on the emission maxima of the compounds was mea-
Acknowledgements
This research has been supported by the National Institutes of
Health Grant G096452. Molecular docking studies were performed
at the Center for Experimental Therapeutics and Pharmaco Infor-
matics at the College of Pharmacy, University of Houston.
sured by using 5
phorbol ester, 0.5
Tris, 150 mM NaCl, 2 mM DTT, 50
l
l
M each of 1 and 4 with 50
M SAPD and 5 M protein in buffer (50 mM
M ZnSO4, pH 7.2). 1, 4 and SAPD
lM protein and for
l
l
References and notes
were incubated with the proteins for 1 h and excited at 425 nm,
389 nm and 355 nm, respectively. The wavelength maxima of the
emission spectra were determined by fitting the symmetrical top
of the spectra to a Gaussian function with Igor Pro 4 (WaveMetrics,
Inc., Lake Oswego, OR).
Fluorescence anisotropy was measured in the same fluorimeter
using parallel and perpendicular polarizers. The steady-state
anisotropy, (r), is defined as
1. Battaini, F.; Mochly-Rosen, D. Pharmacol. Res. 2007, 55, 461.
2. Koivunen, J.; Aaltonen, V.; Peltonen, J. Cancer Lett. 2006, 235, 1.
3. Griner, E. M.; Kazanietz, M. G. Nat. Rev. Cancer 2007, 7, 281.
4. Evcimen, N.; King, G. L. Pharmacol. Res. 2007, 55, 498.
5. Bright, R.; Mochly-Rosen, D. Stroke 2005, 36, 2781.
6. Chou, W. H.; Messing, R. O. Trends Cardiovasc. Med. 2005, 15, 47.
7. Sabri, A.; Steinberg, S. F. Mol. Cell. Biochem. 2003, 251, 97.
8. Alkon, D. L.; Sun, M. K.; Nelson, T. J. Trends Pharmacol. Sci. 2007, 28, 51.
9. Hofmann, J. Curr. Cancer Drug Targets 2004, 4, 125.
10. Newton, A. C. Chem. Rev. 2001, 101, 2353.
ðrÞ ¼ ðIII ꢁ IAÞ=ðIII þ 2IAÞ
where III and IA are the polarized fluorescence intensities in the
directions parallel and perpendicular, respectively, to the excitation
polarization.
11. Nishizuka, Y. Science 1992, 258, 607.
12. Colon-Gonzalez, F.; Kazanietz, M. G. Biochim. Biophys. Acta 2006, 1761, 827.
13. Yang, C.; Kazanietz, M. G. Trends Pharmacol. Sci. 2003, 24, 602.
14. Das, J.; Addona, G. H.; Sandberg, W. S.; Husain, S. S.; Stehle, T.; Miller, K. W. J.
Biol. Chem. 2004, 279, 37964.
15. Das, J.; Zhou, X.; Miller, K. W. Protein Sci. 2006, 15, 2107.
16. Das, J.; Pany, S.; Rahman, G. M.; Slater, S. J. Biochem. J. 2009, 421, 405.
17. Blumberg, P. M.; Kedei, N.; Lewin, N. E.; Yang, D.; Czifra, G.; Pu, Y.; Peach, M. L.;
Marquez, V. E. Curr. Drug Targets 2008, 9, 641.
18. Gennas, G. B.; Talman, V.; Aitio, O.; Ekokoski, E.; Finel, M.; Tuominen, R. K.; Yli-
Kauhaluoma, J. J. Med. Chem. 2009, 52, 3969. and references cited therein.
19. Zhang, G.; Kazanietz, M. G.; Blumberg, P. M.; Hurley, J. H. Cell 1995, 81, 917.
20. Irie, K.; Nakagawa, Y.; Ohigashi, H. Chem. Rec. 2005, 5, 185.
21. Yanagita, R. C.; Nakagawa, Y.; Yamanaka, N.; Kashiwagi, K.; Saito, N.; Irie, K. J.
Med. Chem. 2008, 51, 46.
4.6. Generation of 3D models of PKC C1B subdomains and
molecular docking
Three-dimensional structures of curcumin and its derivatives
were generated using ChemDraw Ultra 7.0 and Cactus, a web based
program (http://cactus.nci.nih.gov/services/translate/). The struc-
tures were subjected to pre-dock energy minimization using an-
other free web application (http://bioserv.rpbs.jussieu.fr/cgi-bin/
Frog).
22. Aggarwal, B. B.; Sung, B. Trends Pharmacol. Sci. 2009, 30, 85.
23. Singh, S. Cell 2007, 130, 765.
24. Mehta, K.; Pantazis, P.; McQueen, T.; Aggarwal, B. B. Anti-Cancer Drugs 1997, 8,
470.
25. Kuo, M. L.; Huang, T. S.; Lin, J. K. Biochim. Biophys. Acta 1996, 1317, 95.
26. Jee, S. H.; Shen, S. C.; Tseng, C. R.; Chiu, H. C.; Kuo, M. L. J. Invest. Dermatol. 1998,
111, 656.
27. Kawamori, T.; Lubet, R.; Steele, V. E.; Kelloff, G. J.; Kaskey, R. B.; Rao, C. V.;
Reddy, B. S. Cancer Res. 1999, 597.
The crystal structures of PKCd (PDB code: 1PTQ), and the phor-
bol-13-OAc bound PKCd (PDB code: 1PTR).19 NMR structure of the
PKC theta C1B (PDB code: 2ENZ)52 and a homology modeled struc-
ture of PKCe C1B have been used as the receptors for molecular

1598
A. Majhi et al. / Bioorg. Med. Chem. 18 (2010) 1591–1598
28. Singletary, K.; MacDonald, C.; Iovinelli, M.; Fisher, C.; Wallig, M. Carcinogenesis
1998, 19, 1039.
29. Ruby, A. J.; Kuttan, G.; Babu, K. D.; Rajasekeharan, K. N.; Kuttan, R. Cancer Lett.
1995, 94, 79.
30. Kuttan, R.; Bhanumathy, P.; Nirmala, K.; George, M. C. Cancer Lett. 1985, 29, 197.
31. Liu, J. Y.; Lin, S. J.; Lin, J. K. Carcinogenesis 1993, 14, 857.
32. Hasmeda, M.; Polya, G. M. Phytochemistry 1996, 42, 599.
33. Zheng, J.; Ramirez, V. D. Br. J. Pharmacol. 2000, 130, 1115.
34. Logan-Smith, M. L.; Lockyer, P. J.; East, J. M.; Lee, A. G. J. Biol. Chem. 2001, 276,
46905.
41. Garg, R.; Ramchandani, A. G.; Maru, G. B. Carcinogenesis 2008, 29, 1249.
42. Markov, P. In The Chemistry of Enols; Rapport, Z., Ed.; Wiley: Chichester, 1992;
pp 69.
43. Claramunt, R. M.; Lopez, C.; Maria, M. D. S.; Sanz, D.; Elguero, J. Prog. Nucl.
Magn. Reson. Spectrosc. 2006, 49, 169.
44. St. Nikolov, G.; Markov, P. J. Photochem. Photobiol. A: Chem. 1981, 16, 93.
45. Arnaut, L. G.; Formosinno, S. J. J. Photochem. Photobiol. A: Chem. 1993, 75, 1.
46. Arnaut, L. G.; Formosinno, S. J. J. Photochem. Photobiol. A: Chem. 1993, 75, 21.
47. Barik, A.; Priyadarsini, K. I.; Mohan, H. Photochem. Photobiol. 2003, 77, 597.
48. Priyadarsini, K. I. J. Photochem. Photobiol. C: Photochem. Rev. 2009, 10, 81.
49. Marquez, V. E.; Blumberg, P. M. Acc. Chem. Res. 2003, 36, 434. and references
cited therein.
35. Sumbilla, C.; Lewis, D.; Hammerschmidt, T.; Inesi, G. J. Biol. Chem. 2002, 277,
13900.
36. Mahmmoud, Y. A. Br. J. Pharmacol. 2005, 145, 236.
37. Singh, S.; Aggarwal, B. B. J. Biol. Chem. 1995, 270, 24995.
38. Choi, H.; Chun, Y.-S.; Kim, S.-W.; Kim, M.-S.; Park, J.-W. Mol. Pharmacol. 2006,
70, 1664.
39. Lin, J. K.; Chen, Y. C.; Huang, Y. T.; Lin-Shiau, S. Y. J. Cell. Biochem. 1997, 28–29,
39.
50. Moreno, F.; Cortijo, M.; Gonzalez-Jimenez, J. Photochem. Photobiolol. 1999, 69, 8.
51. Nancy, E. L.; Blumberg, P. M. In Methods in Molecular Biology; Alexandra, C. N.,
Ed.; 2003; Vol. 233, pp 129.
52. Solution structure of the second C1 domain from human protein kinase C theta,
Nagashima, T.; Hayashi, F.; Yokoyama, S. (To be published, PDB code: 2ENZ).
53. Quest, A. F.; Bardes, E. S.; Xie, W. Q.; Willott, E.; Borchardt, R. A.; Bell, R. M.
Methods Enzymol. 1995, 252, 153.
40. Reddy, S.; Aggarwal, B. B. FEBS Lett. 1994, 341, 19.