Fluoro-curcuminoids and curcuminoid-BF2adducts: Synthesis, X-ray structures, bioassay, and computational/docking study

Article abstract of DOI:10.1016/j.jfluchem.2016.09.009

A series of α-carbonyl fluorinated curcuminoids were synthesized by direct mono- and difluorination with Selectfluor (F-TEDA-BF₄) at r.t. without using a base or additive. Ring fluorinated/trifluoromethylated curcuminoid-BF₂adducts were synthesized by reaction of the corresponding benzaldehydes with acetylacetone-BF₂. Decomplexation of CUR-BF₂adducts under microwave irradiation gave the corresponding curcuminoids. Multinuclear NMR and X-ray analysis confirm that curcuminoids bearing fluorines or trifluoromethyl groups in the aryl rings as well as those that are monofluorinated at the active methylene position all exist as enol tautomers. The α,α-difluorination brings about significant conformational change as these curcuminoids become fixed in the 1,3-diketone configuration. The X-ray structures of CUR-BF₂complexes are consistent with the formation of symmetrical adducts with equal B[sbnd]O bond distances. The anti-proliferative activity of these compounds were tested by in-vitro bioassay against several different cancer cell lines. The corresponding CUR-BF₂adducts exhibited exceptionally high activities at micromolar and in some cases nanomolar concentrations that greatly surpass the activity of parent curcumin. Computational docking calculations were performed to gauge binding energies of these compounds in HER2 protein, and in 20S proteasome, for comparison with the bioassay-derived activity data.

Full text of DOI:10.1016/j.jfluchem.2016.09.009

Journal of Fluorine Chemistry 191 (2016) 29–41
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Fluoro-curcuminoids and curcuminoid-BF2 adducts: Synthesis, X-ray
structures, bioassay, and computational/docking study
a
b
c
d
Kenneth K. Laalia, , Benjamin M. Rathman , Scott D. Bunge , Xin Qi , Gabriela L. Borosky
*
a
Department of Chemistry, University of North Florida, 1 UNF Drive, Jacksonville, FL 32224, USA
Department of Chemistry and Biochemistry, Kent State University, Kent, OH 44242, USA
Department of Medicinal Chemistry, College of Pharmacy, University of Florida, P.O. Box 100484, Gainesville, FL 32610, USA
INFIQC, CONICET and Departamento de Matemática y Física, Facultad de Ciencias Químicas, Universidad Nacional de Córdoba, Ciudad Universitaria,
b
c
d
Córdoba 5000, Argentina
A R T I C L E I N F O
A B S T R A C T
Article history:
A series of
a-carbonyl fluorinated curcuminoids were synthesized by direct mono- and difluorination
Received 30 August 2016
Received in revised form 15 September 2016
Accepted 16 September 2016
Available online 17 September 2016
with Selectfluor (F-TEDA-BF4) at r.t. without using a base or additive. Ring fluorinated/trifluoromethy-
lated curcuminoid-BF2 adducts were synthesized by reaction of the corresponding benzaldehydes with
acetylacetone-BF2. Decomplexation of CUR-BF2 adducts under microwave irradiation gave the
corresponding curcuminoids. Multinuclear NMR and X-ray analysis confirm that curcuminoids bearing
fluorines or trifluoromethyl groups in the aryl rings as well as those that are monofluorinated at the active
Keywords:
Fluorocurcuminoids
Curcuminoid-BF2 adducts
Synthesis
X-ray analysis
In-vitro bioassay
Computational docking
methylene position all exist as enol tautomers. The
a,a-difluorination brings about significant
conformational change as these curcuminoids become fixed in the 1,3-diketone configuration. The X-ray
structures of CUR-BF2 complexes are consistent with the formation of symmetrical adducts with equal
BꢀꢀO bond distances. The anti-proliferative activity of these compounds were tested by in-vitro bioassay
against several different cancer cell lines. The corresponding CUR-BF2 adducts exhibited exceptionally
high activities at micromolar and in some cases nanomolar concentrations that greatly surpass the
activity of parent curcumin. Computational docking calculations were performed to gauge binding
energies of these compounds in HER2 protein, and in 20S proteasome, for comparison with the bioassay-
derived activity data.
ã 2016 Elsevier B.V. All rights reserved.
1. Introduction
the phenolic OH groups (hydrogen bonding interactions), and the
central
b-keto-enolic moiety (participate in H-bonding, chelate
Curcumin (CUR) 1 is a non-toxic natural product extracted from
turmeric and has been used for centuries as dietary supplement
and therapeutic agent in Chinese and Asian medicines. Extensive
studies have shown that curcumin possesses anti-carcinogenic
effects on various types of cancer, as well as anti-inflammatory,
anti-oxidant and other properties that are beneficial to human
health and cancer prevention [1–3]. The central core of CUR
molecule is a conjugated 1,3-diketone that exits almost exclusively
as the enol-tautomer (Fig. 1).
metal cations, and act as Michael acceptor). These favorable
features give curcumin a unique ability to interact with target
proteins.
In spite of these attributes a major obstacle toward the
development of CUR-derived cancer drugs is its rapid metabolism
and poor water solubility. In order to overcome these drawbacks
the synthetic/medicinal chemistry community has devoted
significant effort to modify the CUR structure by synthesizing
analogs that might circumvent these limitations, especially with
regard to metabolic instability.
Whereas the conjugated 1,3-diketone moiety is considered to
be very important in interaction with target proteins [1], it is
believed that this region contributes to its low metabolic stability.
Therefore much research effort has been devoted to structural
modifications with the goal to replace the central portion of the
molecule with metabolically more robust polar moieties.
Despite a relatively simple molecular structure, curcumin has a
combination of features that provide several favorable binding
sites/modes, namely hydrophobic phenyl domains (potential sites
for
p–p interaction with the aromatic side chains in aminoacids),
* Corresponding author.
E-mail address: Kenneth.Laali@UNF.edu (K.K. Laali).
http://dx.doi.org/10.1016/j.jfluchem.2016.09.009
0022-1139/ã 2016 Elsevier B.V. All rights reserved.

30
K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
Fig. 1. Tautomerism in curcumin – exclusive presence of the enol tautomer.
These structural modifications have included introduction of
libraries as potential antitumor agents for lung and prostate cancer
[15–17]. Encapsulation or conjugation with nanoparticles, poly-
meric micelles, or liposomes have been explored as a way to deliver
curcuminoids to cancer cells [3].
enaminone, oxime, and dienone, and replacement of the central
moiety with pyrazole and isoxazole rings [4–9]. It is noteworthy
that many of these synthetic modifications represent significant
departure from CUR’s original skeleton. Other studies have
reported improved bioactivity by transforming the phenolic OH
in CUR to acetates and aminoacid conjugates [10], cinnamic and
succinyl esters [11,12], other types of esters [13], and acetamides
Selective fluorine introduction into pharmaceuticals is
a
powerful strategy for improving metabolic stability and physi-
ochemical properties [18,19], but this approach has remained
greatly under-utilized with respect to curcuminoids.
[14]. Introduction of ester or
active methylene region has been used to prepare curcuminoid
a
,
b
-unsaturated ester linkers into the
Since the standard approach for the assembly of symmetrical
CUR analogs is via a “double aldol” condensation of aldehydes with
Fig. 2. Compound sheet.

K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
31
acetylacetone [20,21], ring fluorinated derivatives can be synthe-
sized via this route, however very few examples have been
reported, and with limited NMR data [22].
To the best of our knowledge only two examples of
curcuminoids bearing a single fluorine at the active methylene
position (along with an ester linker or a methyl group) have been
reported [16,17]. The reported methods required additional steps,
namely deprotonation (NaH/DMF) followed by fluorination, as
well as additional steps for protection and deprotection of the
phenolic OH in the case of parent CUR.
Considering the importance of the 1,3-diketo moiety in the CUR
skeleton for interaction with target proteins, development of
methods for direct fluorine introduction into the active methylene
position is highly desirable, but no such methods have hitherto
been reported.
It is also noteworthy that despite extensive bioassay studies on
curcuminoids [1–3], the potential of CUR-BF2 complexes as anti-
proliferation agents has remained largely unexplored. Review of
the published literature shows that these compounds are usually
treated as “intermediates” en-route to curcuminoids, and only few
examples exist in which they have been characterized, albeit with
limited NMR data [20], except in two independent recent studies
[23], where a series of CUR-BF2 complexes bearing various donor
end-groups were synthesized and their photophysical properties
were explored.
binding energies in HER2 protein relative to a well-known ligand
(SYR), versus trends from bioactivity data for breast cancer (IC50
MDA231), and in proteasome in comparison with bioassay data in
leukemia (IC50 MOLT-4).
2. Results and discussion
2.1. Direct fluorination of parent curcumin 1
Preliminary feasibility studies were performed to identify
suitable solvents for the reaction of 1 with Selectfluor. Experiments
were performed by using 0.1 mmol of CUR and 1.1 and 2.1
equivalents of Selectfluor, and the progress of the reactions were
monitored by 19F NMR (and by TLC). 19F NMR provided a
convenient and direct method to gauge chemoselectivity in
different solvents. Apart from F-CUR and F2-CUR, formation of
CUR-BF2 adduct was observed in some solvents, notably in water
and in acetone under microwave irradiation (more in Section 2.2).
Presence of a distinct doublet at
d
ꢀ195 ppm (JHF = 50 Hz) signified
the 1,3-diketo-tautomer of F-CUR present in different proportions
in some solvents. Application of microwave (MW) increased the
proportion of CUR-BF2 adduct in acetone as solvent (Table 1).
Based on these studies, MeOH was selected as solvent of choice
for direct difluorination of curcuminoids, whereas MeCN was
chosen for monofluorination.
Inspired by our earlier fluorination study of carbonyl compound
[24], we report here the synthesis of novel curcuminoids that are
mono- and difluorinated at the active methylene position by direct
one-pot fluorination with Selectfluor (F-TEDA-BF4) without using a
base or additive (see Fig. 2). In addition to parent CUR (1) and its
symmetrical O-dimethylated analog (DMC; 6), symmetrical
curcuminoids bearing fluorines (11 and 15) and trifluoromethyl
groups (20) in the phenyl rings were also synthesized, and
subsequently mono- and difluorinated at the active methylene
position (see Fig. 2). The corresponding curcuminoid-BF2 adducts
which serve as synthetic intermediates en-route to these
curcuminoids were isolated and characterized. The X-ray crystal
structures were obtained for CUR-BF2 adducts 5, 10, and 14, and for
the F-CUR 8. The cell growth inhibitory and apoptosis inducing
effects of these compounds were examined by in-vitro assays
against leukemia (MOLT-4), prostate cancer (PC3 and LNCap), as
well as lung (A549) and breast cancer (MDA231) cell lines. A
selected group of compounds were screened by the NCI 60 cell
panel. Computational docking studies were also performed on
CUR-BF2 adducts and fluorinated curcuminoids to compare their
To minimize the formation of other species observed in 19F
monitoring study in MeCN solvent, the monofluorinations were
performed at 0 ꢁC and continued at r.t. A noteworthy feature in
these studies was that competing ring fluorination products were
not detected, and further fluorination beyond the formation of F2-
CUR was not observed in a control experiment using excess
Selectfluor (4 equiv.). Based on these studies compounds 3 and 4
were synthesized and isolated according to Fig. 3.
2.2. Development of a general and improved synthetic method for the
assembly of the curcuminoid skeleton
The base-catalyzed “double aldol” condensation of aldehydes
with acetyl acetone serves as a standard method to assemble the
curcumin skeleton. Earlier methods employing boric oxide or boric
acid as additive had drawbacks with respect to reproducibility, and
a more reliable method involving the reaction of acetylacetone-BF2
complex with aldehydes followed by hydrolysis of the complex
was developed [20]. Microwave-assisted one-pot methods that
Table 1
19F NMR monitoring of solvent effect on fluorination selectivitya.
Solventb,c
19F NMR signal at
ꢀ115 ppm (s)
F2-CUR
19F NMR signal at
ꢀ140 ppm (s)
CUR-BF2
19F NMR signal at
19F NMR signal at ꢀ195 ppm (d) Diketo-tautomer of F-CUR
ꢀ176 ppm (s)
F-CUR
Acetonitrile/r.t.d
Methanol/r.t.e
Ethanol/refluxd
DMF/refluxd
THF/r.t.d
–
11%
–
55%
–
34%
–
100%
61%
21%
24%
15%
66%
24%
–
33%
–
6%
–
43%
32%
24%
23%
6%
36%
17%
25%
8%
6%
–
27%
36%
3%
64%
100%
THF/MWd
Acetone/r.t.e
Acetone/MWd
Water/refluxe
–
a
Reported percentages are based on relative integrals of these species in 19F NMR (not taking into account signals due to unreacted Selectfluor and its byproducts).
b
c
R.T. reactions were allowed to run overnight and reflux temperature was limited to 70 ꢁC.
MW conditions: 200 W until temperature reached 140 ꢁC.
Selectfluor (1.1 equiv.).
d
e
Selectfluor (2.1 equiv.).

32
K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
Fig. 3. Synthesis and isolation of F2-CUR (3) and F-CUR (4).
Scheme 1. Synthetic sequence for the preparation of CUR-BF2 compounds and curcuminoids.
employed calcium oxide, and boric acid/sodium sulfate as
additives in toluene solvent had also been reported [21,25].
In the present study, we found the reported hydrolytic/
decomplexation process involving multiple steps [20] to be
inefficient. Instead, the curcuminoid-BF2 adducts and their
corresponding curcuminoids were synthesized as outlined in
Scheme 1. A microwave-assisted method reported for decom-
plexation of curcumin-quinolone hybrids [26] was adopted but
modified by addition of sodium oxalate. Sodium oxalate serves as a
bidentate ligand and chelating agent to preferentially coordinate
Fig. 4. X-ray structure of CUR-BF2 adducts 5, 14, and 10.

K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
33
with BF2, possibly forming sodium-difluoro(oxalato)borate which
is a known compound [27].
In initial studies of solvent effect on fluorination selectivity
using Selectfluor (Table 1), formation of curcumin-BF2 adduct was
unexpectedly observed under microwave irradiation in acetone or
in water. Whereas the method represents an interesting alternative
for the synthesis of CUR-BF2 adducts, in practice the isolated yields
of this MW-assisted method were lower (see experimental) as
compared to the method outlined in Scheme 1, and it was therefore
not used in subsequent reactions.
By using the method outlined in Scheme 1, symmetrical
curcuminoids 6,11,15, and 20 were synthesized (see Fig. 2). As with
parent 1, these curcuminoids are also exclusively present in the
enolic form (NMR). In addition, the corresponding CUR-BF2
adducts 5, 10, 14, and 19 (Fig. 2) were isolated in high yields
and fully characterized (see exp. section).
Fig. 6. Optimized structure of (CF3)2-CUR-BF2 19 by B3LYP/6-311 + G(d,p).
The X-ray structures for CUR-BF2 adducts 5, 10, and 14 (Fig. 4,
and Supplementary data), confirm the symmetrical coordination
of BF2 to the two oxygens in the enolic configuration. The
asymmetric unit in CUR-BF2 5 corresponds to one molecule with a
total of four molecules in each cell. Each molecule interacts with an
adjacent molecule via one H–F contact. A packing diagram is
shown in Fig. 5 (see Supplementary data for additional informa-
tion).
Whereas attempts to obtain X-ray quality crystals for the bis-
trifluoromethyl-CUR-BF2 adduct 19 were unsuccessful, its DFT-
optimized structure (Fig. 6) fully agrees with a symmetrically BF2-
coordinated adduct.
Fig. 7. Optimized structure of 12 by B3LYP/6-311 + G(d,p).
mixtures in the monofluorination reactions, which could be
minimized by running the reactions initially at 0 ꢁC, then warming
to r.t. and continuing to stir at room temperature.
2.3. Synthesis of
features
a-carbonyl fluorinated curcuminoids and structural
The
a-carbonyl monofluorinated compounds exhibit a distinc-
The corresponding
a-carbonyl difluorinated analogs (7, 12, 16,
tive two-bond C/F coupling with the carbonyls, giving rise to a 20–
23 Hz doublet in 13C NMR. The 1,3-diketone tautomer can be
readily recognized by the presence of a ꢂ50 Hz doublet in 19F and
1H NMR due to geminal H/F coupling (see experimental).
Whereas the monofluorinated curcuminoids 4,13,17, and 22 are
exclusively present in the enolic form, compound 8 was present as
a 75:25 (enol to ketone) tautomeric mixture (by 19F NMR).
Interestingly, the X-ray structure of 8 shows only the enolic
tautomer and exhibits intramolecular hydrogen bonding and one
H-F contact with a distance of 2.44 Å (Fig. 8).
and 21) were synthesized in good to moderate isolated yields by
using 2.1 equivalents of Selectfluor, following the same procedure
applied to fluorination of parent 1 (Fig. 1). These compounds
exhibit distinctive two-bond C/F coupling with the carbonyls,
giving rise to a ꢂ28 Hz triplet in 13C NMR (see Experimental). The
DFToptimized structure of tetrafluorinated curcuminoid 12 (Fig. 7)
shows notable conformational changes resulting from tautome-
rization.
The
a-carbonyl monofluorinated analogs (8, 13, 17, and 22)
were synthesized by reaction with 1.1 equivalent of Selectfluor in
MeCN (analogous to synthesis of 4; Fig. 3). Concomitant formation
of the difluorinated analogs was observed in the crude reaction
The asymmetric unit for 8 corresponds to three crystallograph-
ically nonequivalent molecules with a total of six molecules in each
unit cell (see Supplementary data for more detail).
Fig. 5. Crystal packing diagram of 5.

34
K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
Fig. 8. Ball and stick plot of 8.
A notable feature in the X-ray crystallographic data is absence of
any significant
stacking. Intermolecular interactions appear to be
curcuminoids were also more active than their precursors, in
particular against leukemia. The -carbonyl-difluorinated analogs
3 and 12 exhibited increased potency (against MOLT-4) as
a
p
–
p
predominantly through short H-F contacts.
compared to their curcuminoid precursors 1 and 11. Overall,
except for compound 11 which exhibited lower activity relative to
curcumin, all others were more active than parent curcumin, but
those shown in Fig. 9 were clearly superior.
2.4. Cell growth inhibitory effects against human cancer cells –
bioassay
The CUR-BF2 adducts 5, 10, and 19, and the bis-trifluoromethyl-
CUR 20 and its difluoro-derivative 21 were screened at the National
Cancer Institute using NCI-60 cell one-dose protocol (10ꢀ5 M).
Compounds 10, 19, 20, and 21 exhibited relatively modest growth
inhibition activities (10–20%) that were below the NCI’s threshold
criteria for further testing. The CUR-BF2 adduct 5 on the other hand
proved highly active, with growth inhibition 70%-80% and cell
death up to 33%, and has been selected by NCI for further testing.
To evaluate the cell growth inhibitory and apoptosis inducing
effects of the fluorinated curcuminoids and their BF2 adducts, in-
vitro bioassay tests were performed on four different cancer cell
lines namely: MOLT-4 (human leukemia suspension cancer cell
line), PC3 and LNCap (human androgen sensitive and insensitive
prostate cancer cell lines respectively), A549 (lung cancer), and
MDA231 (breast cancer), and the results are sketched in Table 2.
Based on the magnitude of IC50 values (Table 2), compounds 5,
6, 8, 10 and 14 (Fig. 9) exhibited high potencies toward multiple
cancer cell lines, with measured activities at low micro-molar
concentrations, and in the case of compound 5 in nano-molar
concentration for leukemia.
2.5. Computational docking study
In an effort to shed some light on the factors determining the
bioactivity of these curcumin analogues, molecular docking
calculations were performed in the active site of HER2. Human
epidermal growth factor receptor 2 (HER2) is one of the tyrosine
kinase receptors in EGFR family, which is known to play a central
role in the pathogenesis of several human cancers [28]. Amplifica-
tion or overexpression of HER2 occurs in breast, prostate, gastric/
gastroesophageal, ovarian, endometrium, bladder, lung, colon, and
head and neck cancers. Therefore, it is a drug target for cancer
therapy focusing on inhibiting HER2 to reduce tumor growth [28].
Taking this into account, binding energies in HER2 were compared
These anti-proliferative activity trends far exceed those of
parent curcumin. It is noteworthy that three of these compounds
are CUR-BF2 adducts.
Comparing the IC50 data for compound 6 with the
a-carbonyl-
monofluorinated analog 8 indicates increased activity as a result
fluorine introduction, notably against breast cancer. Compound 6
(DMC) was previously tested as an anti-prostate cancer agent.17
The reported IC50 values against PC3 and LNCap are near identical
to those in the present study. Some of the other fluorinated
Table 2
In vitro cell viability bioassay
¼.
Leukemia
IC50
M)a,b
Prostate Cancer
Lung Cancer
IC50
M)a,c
Breast Cancer
IC50
M)a,c
(
m
IC50
(
m
M)a,c
(
m
(m
Curcuminoid
Parent 1
4
3
10
11
13
6
MOLT-4
19
3.6*
0.66**
1.5*
28.0
PC3
9.0
7.4
15.0
1.9*
14.0
38.0
1.0*
3.2*
nd
LNCap
9.7
13.0
28.0
4.3
30.0
7.0
1.2*
0.37**
29.2
0.19**
21.3
2.98*
13.2
A549
18.0
20.0
26.0
7.3
29.0
60.0
7.1
1.5*
nd
nd
nd
nd
MDA231
11.5
6.2
2.2*
1.0*
20.0
12.0
0.28**
2.0*
6.7
0.023***
2.1*
2.8*
0.043***
0.56 (nM)****
1.32*
0.076***
1.07*
5
12
8d
7
14
15
nd
nd
nd
nd
0.10**
1.58*
1.7*
20
nd
¼
Curcuminoids with: IC50 < 4 micro-molar are considered active*; IC50 <1 micro-molar considered highly active**; IC50 < 0.1 micro-molar considered potent***; IC50 values
in the nano-molar range are considered highly potent****; compounds were chemically stable at r.t. in DMSO (solvent used for bioassay). nd = not determined.
a
IC50 is drug concentration that can inhibit 50% of cell growth.
b
Cell viability was analyzed by the CellTiter-Glo1 Luminescent Cell Viability Assay.
c
Cell viability was analyzed by the MTT assay.
Contained circa 25% of the diketo-tautomer 9.
d

K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
35
Fig. 9. Most potent curcuminoids identified based on bioassay.
with the bioactivity data (IC50 MDA231; breast cancer) in the
search for correlations between the HER2 inhibitory activity and
inhibition of cancer cells growth.
insight into the role played by electrostatic forces in the
interactions between biomolecules and their ligands.
For this family of compounds the most notable observation is
the negative electrostatic potential developed by the BF2 and the
keto-enol moiety.
The curcuminoid derivatives fitted nicely in the tunnel-like
binding pocket of HER2, where they mainly established hydropho-
bic contacts. The more bioactive analogs (as in 14) interacted with
the residues Leu726, Thr798, Lys753, Leu796, Thr862, Asp863,
Val734, and Leu852 (Figs. 10 and 11), similar to the potent HER2
tyrosine kinase domain (HER2-TK) inhibitor SYR127063 (2-{2-[4-
({5-chloro-6-[3-(trifluoromethyl))phenoxy]pyridin-3-yl}amino)-
5H-pyrrolo[3,2-d]pyrimidin-5-yl]ethoxy}ethanol). In addition to
its numerous hydrophobic interactions, SYR also formed hydrogen
bonds with Met801, Asp863 and Asn850. A hydrogen bond with
Met801 was also observed for the difluorinated curcuminoid 3.
Table S2 (Supplementary data) provides a comparison between
docking energies in HER2 and the measured IC50 values for
MDA231, showing that curcuminoids-BF2 adducts 10, 14, and 5
exhibit both favorable docking energy and cytotoxicity against
breast cancer. Inhibition of tumor cellular proteasome has been
suggested as the mechanism by which curcumin arrests the
proliferation of acute promyelocytic leukemia (APL) cells [5].
Considering the high degree of potency against leukemia observed
in the present study, in particular by 5, 6, 8, and 14, docking
calculations were also performed in proteasome (Table S3,
3. Comparative discussion and summary
A series of fluorinated curcumins and curcumin-BF2 adducts
have been synthesized and characterized. The
a-carbonyl mono-
and difluorinations were achieved by direct fluorination with
Selectfluor, and ring fluorinated/trifluoromethylated CUR-BF2
adducts were assembled in one-pot from the corresponding
aldehydes. Structural features of the new curcuminoids were
examined by multinuclear NMR, and by X-ray analysis (for three
CUR-BF2 adducts and a mono-fluorinated analog). A notable
feature observed in the X-ray crystal structures is detection of
intermolecular interactions via short H-F contacts. The IC50 data
from in-vitro cell growth inhibitory bioassay indicated that the
majority of curcuminoids were more active relative to parent
curcumin, but those shown in Fig. 10 were clearly superior. The
bioassay data along with the NCI-60 screening suggest that CUR-
BF2 complexes bearing activating substituents in the phenyl rings
(as in 5) may be promising drug candidates, and that introduction
of F or CF3 groups into the phenyl rings is not particularly
beneficial. The finding that 8 is more potent than 6 implies that
Supplementary data). The structure of the 20S proteasome (
b5
and 6 subunits) was obtained from the Protein Data Bank.
b
Treatment with proteasome inhibitors results in decrease prolif-
eration, induction of apoptosis, and sensitization of a variety of
tumor cells to chemotherapeutic agents and irradiation. Docking
calculations for curcuminoids resulted in binding energies that
were similar to parent curcumin itself, but judging from their IC50
values these compounds are more active.
Considering the high bioactivity and favorable docking energies
of compounds 5, 6, 8, 10, and 14, their electrostatic potential maps
were computed and are shown in Fig. 12 for a better visualization
of the electronic properties of these molecules. Molecular
electrostatic potential is a useful tool for interpreting and getting
monofluorinaton at the a-carbonyl position can be beneficial. The
fact that activity of F2-CUR analogs (3, 12, 7, and 16) were not
particularly impressive reinforces the earlier conclusions that the
enolic tautomer is an important feature in binding of curcumin to
proteins [1,5]. Model molecular docking calculations in the active
site of HER2 indicated that the curcuminoid derivatives can fit
nicely in the tunnel-like binding pocket of HER2 by establishing
hydrophobic contacts, leading to favorable docking energies. In
summary, the present study has provided notable structural clues
that when brought to bear could lead to the synthesis of effective
drug targets based on curcumin. Synthesis and characterization of

36
K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
reactions were performed in a miniature 400 W lab microwave
in 5 mL vials with magnetic stirring. Melting points were measured
in open capillaries and are not corrected.
4.2. Typical procedure for difluorination of curcuminoids
Selectfluor (2.2 equiv.) was added in one portion to a solution of
the curcuminoid (0.75 mmol) in methanol (25 mL) at r.t. (at 55 ꢁC in
case of compounds 6 and 11) with efficient stirring under a
nitrogen atmosphere for the requisite time (either 4 h at 55 ꢁC or
overnight for r.t.) until completion (TLC monitoring). The MeOH
solvent was then removed in vacuo and the reaction mass was
dissolved in DCM (3 ꢄ10 mL), washed with deionized water
(3 ꢄ10 mL), dried (sodium sulfate) and filtered through a coarse
sintered glass funnel. The DCM was removed in vacuo and the
crude mixture was purified by flash chromatography eluting with
hexane and ethyl acetate (refer to analytical data), ramping of the
elution solvent was employed in all cases.
4.3. Specific procedure – difluorination of 11
Selectfluor (371 mg, 1.04 mmol, 2.5 equiv.) was added in one
portion to a solution of 11 (0.42 mmol,131 mg) in methanol (20 mL)
under a nitrogen atmosphere, and the mixture was stirred under
mild reflux at 55 ꢁC overnight. Completion of the reaction was
confirmed by TLC and the solvent was removed in vacuo. The
reaction mass was dissolved in DCM (3 ꢄ10 mL) and washed with
deionized water (3 ꢄ10 mL). The organic layer was dried (sodium
sulfate), filtered through a coarse sintered glass funnel, and the
solvent was removed in vacuo. The difluoro-derivative (compound
12) precipitated out of the crude reaction mixture as a white
crystalline solid by addition of hexane/EtOAc (20%); 62 mg,
0.18 mmol, 42% yield.
4.4. Typical procedure for monofluorination of curcuminoids
Selectfluor (0.8 mmol,1.05 equiv.) was added in one portion to a
solution of the curcuminoid (0.75 mmol) in acetonitrile (25 mL) at
0 ꢁC under a nitrogen atmosphere, and the mixture was stirred for
6 h at this temperature, followed by overnight stirring at r.t. Upon
completion (verified by TLC) the solvent was removed in vacuo and
the reaction mass was dissolved in DCM (3 ꢄ10 mL). The reaction
mixture was washed with deionized water (3 ꢄ10 mL), dried over
sodium sulfate and filtered through a coarse sintered glass funnel.
The solvent was removed in vacuo and the crude reaction mixture
was purified by flash chromatography eluting with hexane/ethyl
acetate (refer to analytical data), ramping of the elution solvent
was employed in all cases).
Fig. 10. Binding mode in the active site of HER2 of compound 14 as a model
curcuminoid analog (2D-plot).
other fluorocurcuminoids, including those bearing heterocyclic
aryl rings, are ongoing in this laboratory and collaborative studies
aimed at understanding how fluorination contributes to inhibition
of cell viability are planned.
4. Experimental section
4.5. Specific procedure – monofluorination of 11
4.1. General
Selectfluor (286 mg, 0.81 mmol, 1.05 equiv.) was added in one
portion to a solution of 11 (119 mg, 0.38 mmol) in acetonitrile
(20 mL) at 0 ꢁC under a nitrogen atmosphere, and the mixture was
stirred for 6 h at this temperature, followed by overnight stirring at
r.t. Upon completion (monitored by TLC), the acetonitrile was
removed in vacuo and the reaction mass was dissolved in DCM
(3 ꢄ10 mL). Following the steps outlined in the typical procedure
and flash chromatography using hexane/EtOAc (5%), the mono-
fluorinated product 13 was obtained as a yellow solid (54 mg,
0.16 mmol, 43% yield).
Synthetic curcumin, fluorinated aldehydes (p-fluorobenzalde-
hye, 6-fluorovetraldehyde, vetraldehyde, and p-trifluoromethyl-
benzaldehyde), Selectfluor, and acetylacetone were all high purity
commercially available samples and were used without further
purification. Regular solvents used for synthesis (MeCN, acetone,
DCM, hexane, and EtOAc) were all of sufficient purity and were
used as received. Column chromatography was performed on silica
gel (63–200 mesh). NMR spectra were recorded on a 500 MHz
instrument using CDCl3, DMSO-d6, or MeCN-d3 as solvent. 19F NMR
and 11B NMR spectra were referenced relative to external CFCl3 and
4.6. Synthesis of acetylacetone-difluoroboron adduct
BF3 Et2O respectively. HRMS analyses were performed on an LC–
ꢃ
MS instrument in electrospray mode using DMSO as solvent. FT-IR
spectra were recorded in ATR mode in solvent. Microwave
Following a similar procedure described by Fraser et al. [29], to a
mixture of acetylacetone (10 mmol, 1.00 g) in dry DCM (50 mL,

K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
37
to come to room temperature and quenched with DI water (15 mL).
It was transferred to a large separatory funnel, the DCM layer was
separated, and the aqueous layer was discarded (the aqueous layer
was quite acidic with a pH of about 1 or less). The reaction mixture
was subsequently washed several times with DI water (3 ꢄ15 mL)
until the aqueous layer had a pH of about 7. The organic layer was
dried (sodium sulfate), filtered through a coarse sintered glass
funnel, and the solvent was removed in vacuo to afford a brown
crystalline solid (1.42 g, 9.61 mmol, 96.0% yield), which was shown
to be pure by NMR.
4.7. Typical procedure for the synthesis of curcuminoid-difluoroboron
adducts
These compounds were synthesized by a slight modification of
the procedure described by Rao and Sudheer [20]. To a mixture of
acetylacetone-BF2 complex (887 mg, 6 mmol) in ethyl acetate
(60 mL) under stirring and nitrogen atmosphere, the respective
aldehyde (2.2 equivalence, 13.2 mmol) was added in one portion,
followed by N-butylamine (0.22 eq., 1.32 mmol, 96.5 mg, 130 mL)
over a period of 20 min, with continuous stirring at room
temperature overnight. The completion of the reaction was
confirmed by TLC. The desired product precipitates from ethyl
acetate. The reaction mixture was cooled to 0 ꢁC in an ice bath and
the product was filtered, washed with cold (0 ꢁC) ethyl acetate and
dried for 30 min. The purity of this cut was exceptional (NMR) and
no further purification was required (ꢂ60% isolated yield). The
filtrate was transferred to a round bottom flask and concentrated
under vacuum and re-filtered to obtain a second cut which was
slightly less pure by NMR (combined yield: typically > 80%, except
for compound 2 which was 64%).
Fig. 11. Binding mode in the active site of HER2 of compound 14 as a model
curcuminoid analog (3D-plot).
distilled from P2O5) under a nitrogen atmosphere BF3
ꢃEt2O (ꢂ48%;
2.13 g, 1.89 mL, 15 mmol, 1.5 equiv.) was slowly added over a period
of 5 min, and the reaction mixture was refluxed (41 ꢁC) for 12 h.
Upon completion of the reaction, the reaction mixture was allowed
Fig. 12. Electrostatic potential maps for the most active curcuminoid derivatives.

38
K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
4.8. Specific procedure – synthesis of curcuminoid-BF2 adduct 10
5. Characterization data
To a mixture of acetylacetone-BF2 (887 mg, 6.00 mmol) in ethyl
acetate (60 mL) under stirring and nitrogen atmosphere, was
added p-fluorobenzaldehyde (2.2 equiv., 13.2 mmol, 1.638 g) in one
portion, followed by slow addition (over 20 min) of N-butylamine
5.1. Curcumin-BF2 adduct (2)
Yield: 64% (using BF3
solid, mp > 260 ꢁC. Rf 0.16 (40% EtOAc in hexane). 1H NMR (CD3CN,
500 MHz): 7.96 (d, J = 16.0 Hz, 2H), 7.42 (br s, 2OH), 7.37 (d,
ꢃEt2O) and 41% (by using Selectfluor), red
(0.22 equiv.,1.32 mmol, 96.5 mg,130
m
L). The reaction mixture was
d
stirred continuously at room temperature overnight whereupon
compound 10 precipitated from ethyl acetate. The reaction
mixture was cooled to 0 ꢁC in an ice bath, filtered, and the product
was washed with cold (0 ꢁC) ethyl acetate and dried for 30 min to
afford compound 11 as a yellow solid (1.72 g, 4.78 mmol, 80% yield)
which was confirmed by NMR to be highly pure.
J = 2.1 Hz, 2H), 7.30 (dd, J = 8.2 and 2.1 Hz, 2H), 6.94 (d, J = 8.6 Hz, 2H),
6.89 (d, J = 16.0 Hz, 2H), 6.28 (s, 1H), 3.95 (s, 6H, 2OMe). 13C NMR
(CD3CN, 125 MHz):
d 180.5, 151.4, 148.7, 147.6, 127.8, 128.0, 119.2,
116.2, 112.3, 102.4, 56.7.19F NMR (CD3CN, 470 MHz):
d
ꢀ140.97 (s,
11B F), ꢀ140.91 (s, 10B F). 11B NMR (CD3CN, 160.3 MHz):
d
0.95 (s).
5.2. (1E,6E)-4,4-Difluoro-1,7-bis(4-hydroxy-3-methoxyphenyl)
hepta-1,6-diene-3,5-dione (3)
4.9. General procedure for decomplexation of curcuminoid-BF2
Yield: 33%, red solid, mp 117–118 ꢁC. Rf 0.53 (40% EtOAc in
Using
a modified microwave assisted method [26], the
hexane). 1H NMR (CD3CN, 500 MHz):
d
7.85 (d, J = 16.0 Hz, 2H), 7.35
(d, J = 1.5 Hz, 2H), 7.29 (br s, 2OH), 7.25 (dd, J = 8.0 and 1.5 Hz, 2H),
7.12 (d, J = 16 Hz, 2H), 6.88 (d, J = 8.0 Hz, 2H), 3.90 (s, 6H, 2OMe). 13
NMR (CD3CN, 125 MHz):
186.5 (t, 2JCF 26.7 Hz), 150.5, 148.8, 147.8,
126.2, 125.5, 115.2, 116.5, 111.1, 111.7 (t, JCF = 263.2 Hz), 58.9. 19F
curcuminoid-BF2 complex (0.3 mmol) and sodium oxalate (2
equiv) were added to a clean/dry microwave vial equipped with
magnetic stirrer. Aqueous methanol (5 mL, 8:2 MeOH/H2O) was
added and the vial was sealed with a crimp-able cap with septa
using a crimping tool and the sealed vial was irradiated for 6 min at
140 ꢁC. The vial was cooled to room temperature and the cap
removed. The reaction mixture was transferred to a round bottom
flask and the methanol removed under vacuum. Upon addition of
deionized-water (20 mL) a precipitate was formed which was
collected by filtration, washed with 40 mL of deionized-water and
dried for 30 min. The resulting curcuminoid product was >98% pure
(by NMR).
C
d
1
NMR (CD3CN, 470 MHz):
d
- 115.3. HRMS (ESI): m/z [M + H]+ calcd
for C21H19O6F2: 405.1149; found: 405.065. IR (cmꢀ1, CH2Cl2/
MeCN): 3415 (br, OH), 3182-2847 (C-H package), 1693 (CO),
1681, 1568, 1506, 1431, 1271, 1207, 1122, 1064.
5.3. (1E,4E,6E)-4-Fluoro-5-hydroxy-1,7-bis(4-hydroxy-3-
methoxyphenyl)hepta-1,4,6-trien-3-one (4)
Yield: 52%, purple solid, mp 154–155 ꢁC. Rf 0.60 (40% EtOAc in
hexane). 1H NMR (CD3CN, 500 MHz):
d 14.1 (br, enolic OH), 7.63 (d,
4.10. Typical procedure – decomplexation of curcuminoid-BF2 adduct
10
J = 16 Hz, 2H), 7.31 (d, J = 16 Hz, 2H), 7.20 (dd, J = 8.5 and 1.5 Hz, 2H),
7.10 (dd, J = 16 and 3.5 Hz, 2H), 7.05 (s, 2H, 2OH), 6.87 (d, J = 8.0 Hz,
2H), 3.91 (s, 6H, 2OMe). 13C NMR (CD3CN, 125 MHz): 172.1 (d,
The curcuminoid-BF2 adduct 10 (101 mg, 0.28 mmol) and
sodium oxalate (75 mg, 0.56 mmol, 2 equiv.) were added to a
clean/dry microwave vial equipped with magnetic stirrer. Upon
addition of aqueous methanol (5 mL, 8:2 MeOH/H2O) a suspension
was formed. The vial was sealed and irradiated for 6 min at 140 ꢁC.
The vial was cooled to r.t. and the sealed cap was removed. Removal
of solvent and addition of deionized-water gave a precipitate
which was washed and dried under vacuum to give compound 11
as a yellow solid (79 mg, 0.25 mmol, 90% yield) which was pure by
NMR.
1
3
2JCF = 23 Hz), 149.5, 147.7, 143.0 (d, JCF = 238.0 Hz), 141.9 (d, JCF
=
2.9 Hz), 127.4, 123.8, 115.1, 114.3, 110.7, 55.8. 19F NMR (CD3CN,
470 MHz):
d
- 176.5 (t, JHF = 2.8 Hz). HRMS (ESI): m/z [M + H]+ calcd
for C21H20O6F: 387.1243; found: 387.080. IR (cmꢀ1, CH2Cl2/MeCN):
3417 (br, OH), 3062-2937 (C-H package), 1620 (CO), 1568, 1504,
1429. 1267, 1122, 1029.
5.4. Tetramethoxy-curcuminoid-BF2 complex (5)
Yield 83%, purple solid, mp 224–226 ꢁC, Rf 0.16 (40% EtOAc in
hexane). 1H NMR (DMSO-d6, 500 MHz):
d
7.97 (d, J = 16.0 Hz, 2H),
7.49 (d, J = 2.0 Hz, 2H), 7.46 (dd, J = 8.0 and 2.0 Hz, 2H), 7.11 (d,
J = 16 Hz, 2H), 7.07 (d, J = 8.0 Hz, 2H), 6.50 (s, 2H), 3.84 (s, 6H). 13
NMR (DMSO-d6, 125 MHz): 179.5, 153.0, 149.6, 147.3, 127.6, 125.6,
4.11. Microwave assisted formation of curcumin-BF2 adduct using
selectfluor
C
To a clean dry microwave vial (5 mL) curcumin 1 (276 mg,
0.75 mmol) was added followed by acetone (5 mL) to dissolve the
curcumin. Selectfluor (292 mg, 0.82 mmol, 1.1 equiv) was then
added directly to the solution, and the vial was sealed with a
crimp-able cap with septa using a crimping tool. The absorbance
was set to very high and the mixture was irradiated at 200 W for
97 s until it reached a temperature of 138 ꢁC and a pressure of
10 bar. The reaction was monitored by TLC at intervals by removing
ꢂ0.1 mL through the septa with a syringe and diluting in DCM.
Upon completion, the acetone was removed in vacuo and reaction
mass was dissolved in DCM, washed with deionized water
(3 ꢄ10 mL), dried (sodium sulfate) and filtered through a coarse
sinter glass funnel. The solvent was removed in vacuo, and the
crude mixture was purified by flash chromatography eluting with
ethyl acetate/hexane (40:60). The resulting curcumin-BF2 complex
2 (128 mg, 0.307 mmol, 41% yield) was pure as confirmed by NMR.
For comparison, a 64% isolated yield was obtained for this
d
119.3, 112.3, 111.7, 101.8, 56.3, 56.1.19F NMR (DMSO-d6, 470 MHz):
d
- 137.9 (s, 11B F), ꢀ137.8 (s, 10B F). 11B NMR (DMSO-d6,
160.3 MHz):
d
0.90 (s). IR (cmꢀ1, CH2Cl2): 3003-2841 (CH package),
1610 (CO), 1529, 1508, 1263, 1136.
5.5. (1E,4E,6E)-5-Hydroxy-1,7-bis(3,4-dimethoxyphenyl)hepta-1,4,6-
trien-3-one (6)
Yield 94%, red solid, mp 122–125 ꢁC. Rf 0.37 (40% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d
ꢂ14.8 (br, enolic OH),7.62 (d,
J = 16.0 Hz, 2H), 7.17 (dd, J = 8.0 and 1.6 Hz, 2H), 7.09 (d, J = 1.5 Hz,
2H), 6.89 (d, J = 8 Hz, 2H), 6.51 (d, J = 16 Hz, 2H), 5.85 (s, 1H), 3.94 (s,
6H), 3.93 (s, 6H). 13C NMR (DMSO-d6, 125 MHz):
d 183.6, 151.4,
149.5, 140.9, 128.0, 123.5, 122.5, 112.1, 110.8, 101.5, 56.0, 56.1. HRMS
(ESI): m/z [M+H]+ calcd for C23H25O6: 397.1651; found: 397.140. IR
(cmꢀ1, DCM/CH2Cl2): 3005-2837 (CH package), 1624 (CO), 1581,
1510, 1462, 1259, 1136, 1022.
decomplexation by using BF3 Et2O instead of Selectfluor.
ꢃ

K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
39
5.6. (1E,6E)-4,4-Difluoro-1,7-bis(3,4-dimethoxyphenyl)hepta-1,6-
diene-3,5-dione (7)
7.67–7.63 (m, 4H), 7.15–7.12 (m, 4H), 7.06 (d, J = 16.5 Hz, 2H). 13C
2
NMR (CDCl3, 125 MHz):
d
186.7 (t, JCF 27.9 Hz), 164.9 (d,
1JCF = 254.7 Hz), 147.3, 131.4 (d, JCF = 8.5 Hz), 129.9 (d, JCF = 3.9 Hz),
Yield: 40%, orange solid, mp 148–149 ꢁC. Rf 0.42 (40% EtOAc in
117.6 (d, JCF = 2.9 Hz), 116.4 (d, JCF = 22 Hz), 111.4 (t, 1JCF = 265.1 Hz).
hexane). 1H NMR (CDCl3, 500 MHz):
d
7.87 (d, J = 15.5 Hz, 2H), 7.24
19F NMR (CDCl3, 470 MHz):
d
- 106.2 (m, 2F), ꢀ115.3 (s, 2F). HRMS
(dd, J = 8.0 and 2.0 Hz, 2H), 7.13 (d, J = 2.0 Hz, 2H), 6.98 (d, J = 15.5 Hz,
2H), 6.90 (d, J = 8.0 Hz, 2H), 3.95 (s, 6H, 2OMe), 3.96 (s, 6H, 2OMe).
(ESI): m/z [M+H]+ calcd for C19H13O2F4: 349.0851; found: 349.040.
IR (cmꢀ1, DCM/CDCl3): 3078, 2929, 1697, 1608, 1585, 1508, 1417,
1234, 1159, 1112, 1099, 1058.
2
13C NMR (CDCl3, 125 MHz):
d
186.6 (t, JCF 26.8 Hz), 152.7, 149.4,
148.8, 126.7, 125.1, 115.6, 112.0 (t, 1JCF = 264.1 Hz), 111.0, 110.0, 59.0.
19F NMR (CDCl3, 470 MHz):
d
- 115.3 (s, 11B isotope) and ꢀ115.4 (s,
5.11. (1E,4E,6E)-4-Fluoro-1,7-bis(4-fluorophenyl)-5-hydroxy-hepta-
1,4,6-trien-3-one (13)
10B isotope). HRMS (ESI): m/z [M+H]+ calcd for C23H23O6F2:
433.1462; found: 433.0886. IR (cmꢀ1, CH2Cl2/DCM): 3003-2937
(CH package), 1681(CO), 1587, 1573, 1510, 1463, 1421, 1265, 1139.
Yield: 43%, yellow solid, mp 152–153 ꢁC. Rf 0.45 (5% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d
ꢂ13.8 (br, enolic OH), 7.68 (d,
5.7. (1E,4E,6E)-4-Fluoro-5-hydroxy-1,7-bis(3,4-dimethoxyphenyl)
hepta-1,4,6-trien-3-one (8) and (1E,6E)-4-Fluoro-1,7-bis(3,4-
dimethoxyphenyl)hepta-1,6-diene-3,5-dione (9)
J = 15.9 Hz, 2H), 7.62 (dd, J = 8.5 and 5.5. Hz, 4H), 7.12 (t appearance,
J = 8.5, 4H), 7.06 (dd, J = 15.7 and 3.7 Hz, 2H). 13C NMR (CDCl3,
2
125 MHz): 172.0 (d, JCF = 21.8 Hz), 164.0 (d, 1JCF = 251.6 Hz), 145.5,
143.1 (d, 1JCF = 238.4 Hz), 140.6 (d, JCF = 1.9 Hz), 131.2 (d, JCF = 2.9 Hz,
Tautomeric mixture (75:25 ratio by 19F NMR). Yield 33%, red
130.3 (d, JCF = 8.7 Hz), 116.8, 116.2 (d, JCF = 22.0 Hz). 19F NMR (CDCl3,
solid, mp 145–146 ꢁC. Rf 0.52 (40% EtOAc in hexane). Analytical
470 MHz):
d
- 108.9 (m, 2F), ꢀ175.2 (t, J = 3.3 Hz,1F). HRMS (ESI): m/
data for 8. 1H NMR (CDCl3, 500 MHz):
d
ꢂ14.0 (br, enolic OH), 7.67
z [M+H]+ calcd for C19H14O2F3: 331.0945; found: 331.110. IR (cmꢀ1
,
(d, J = 16 Hz, 2H), 7.21 (dd, J = 8.0 and 2.0 Hz, 2H), 7.14 (d, J = 2 Hz,
2H), 6.99 (dd, J = 16.0 and 3.5 Hz, 2H), 6.90 (d, J = 8.5 Hz, 2H), 3.96 (s,
DCM/CDCl3): 3072, 2920, 1633, 1597, 1508, 1417, 1319, 1232, 1157.
OMe), 3.94 (s, OMe).19F NMR (CDCl3, 470 MHz):
d
- 175.7 (s).
Analytical data for 9: 1H NMR (CDCl3, 500 MHz):
d 7.78 (d, J = 16 Hz,
5.12. Tetramethoxydifluoro-curcuminoid-BF2 adduct (14)
2H), 7.22 (dd, not fully resolved), 7.12 (d, J = 2 Hz), 6.88 (d, J = 8.5 Hz),
Yield 90%, red solid, mp 253–255. Rf 0.21 (40% EtOAc in hexane).
5.70 (d, JHF = 50 Hz), 3.93 (s, OMe), 3.94 (OMe, not fully resolved).
1H NMR (DMSO-d6, 500 MHz):
d
8.05 (d, J = 16.0 Hz, 2H), 7.46 (d,
J = 7.0 Hz, 2H), 7.16 (d, J = 15.5 Hz, 2H), 7.08 (d, J = 12 Hz, 2H), 6.65 (s,
1H). 13C NMR (DMSO-d6, 125 MHz): 179.3, 157.2 (d, 1JCF = 250 Hz),
19F NMR (CDCl3, 470 MHz):
d
ꢀ 194.9 (d, JHF = 50 Hz). Data for 8 (a)
and 9 (b): 13C NMR (CDCl3, 125 MHz): 189.9 (d, 2JCF = 20 Hz)b, 172.0
(d, 2JCF = 22 Hz)a, 152.3, 151.4a, 149.3, 149.2a, 146.9 (d, 3JCF = 2.7 Hz)b,
143.0 (d, 1JCF = 236 Hz)a, 141.7 (d, 3JCF = 3 Hz)a, 128.0a, 126.9b, 123.2a,
117.1b, 115.0a, 111.1a, 111.0, 109.9a, 97.5 (d, 1JCF = 199 Hz)b, 56.0, 55.0.
HRMS (ESI): m/z [M+H]+ calcd for C23H24O6F: 415.1556; found:
415.120. IR (tautomeric mixture; cmꢀ1, CH2Cl2/CDCl3): 3400 (br,
OH), 2933, 2839, 1685 (CO), 1589, 1510, 1463, 1265, 1139, 1022.
d
153.9 (d, JCF = 11.4 Hz), 145.9, 138.3 (d, JCF = 2.9 Hz), 120.7 (d,
JCF = 5.7 Hz), 113.1 (d, JCF = 11.4 Hz), 110.4 (d, JCF = 3.8 Hz), 101.7,
100.7 (d, JCF = 28.6 Hz), 56.5, 56.2. 19F NMR (DMSO-d6, 470 MHz):
d -
119.0 (unresolved dd, 2F), ꢀ137.7 (s, 11B F), ꢀ137.6 (s, 10B F).11B NMR
(DMSO-d6, 160.3 MHz):
d 0.89 (br s).
IR (cmꢀ1, CH2Cl2): 2954, 2922, 2852, 1714, 1597, 1514, 1462,
1278, 1193, 1001.
5.8. Difluorocurcuminoid-BF2 adduct (10)
Yield 80%, orange solid, mp 258–260 ꢁC. Rf 0.05 (5% EtOAc in
5.13. (1E,4E,6E)-5-Hydroxy-1,7-bis(3,4-dimethoxy-6-fluorophenyl)
hepta-1,4,6-trien-3-one (15)
hexane). 1H NMR (acetone-d6, 500 MHz):
d 8.07 (d, J = 15.7 Hz, 2H),
7.96 (dd, J = 8.5 and 7.0 Hz, 4H), 7.30 (t appearance, J = 9.5 Hz, 4H),
7.13 (d, J = 15.7 Hz, 2H), 6.65 (s,1H).13C NMR (acetone-d6,125 MHz):
Yield 94%, orange solid, mp 154–156 ꢁC. Rf 0.45 (40% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d 7.74 (d, J = 16.0 Hz, 2H), 6.99
d
181.7,165.7 (d,1JCF = 251.7 Hz),146.3,132.7 (d, JCF = 8.6 Hz),131.9 (d,
JCF = 3.9 Hz), 122.2, 117.2 (d, JCF = 22 Hz), 103.1.19F NMR (acetone-d6,
(d, JHF = 7.5 Hz, 2H), 6.67 (d, JHF = 12.5 Hz, 2H), 6.60 (d, J = 16.5 Hz,
2H), 5.88 (s, 1H), 3.92 (s, 12H). 13C NMR (CDCl3, 125 MHz):
d 183.2,
470 MHz):
(acetone-d6, 160.3 MHz):
d
-108.7(m,2F),ꢀ140.2(s,11BF),ꢀ140.1(s,10BF).11BNMR
1
1ꢃ01 (s). IR (cmꢀ1, CH2Cl2): 3107, 3041,
156.6 (d, JCF = 249.0 Hz), 151.8 (d, JCF = 10.4 Hz), 145.6 (d, JCF = 2.0
Hz), 133.0, 124.0 (d, JCF = 6.6 Hz), 114.1 (d, JCF = 13.3 Hz), 109.6 (d,
d
2922, 2850, 1620 (CO), 1589, 1548, 1508, 1404, 1232, 1155.
J
CF = 4.8 Hz), 101.4, 100.2 (d, JCF = 28.5 Hz), 56.4, 56.3. 19F NMR
(CDCl3, 470 MHz):
d - 120.1 (dd, JHF = 11.7 and 5.6 Hz). HRMS (ESI):
5.9. (1E,4E,6E)-1,7-Bis(fluorophenyl)-5-hydroxy-hepta-1,4,6-trien-3-
one (11)
m/z [M+H]+ calcd for C23H23O6F2: 433.1462; found: 433.142. IR
(cmꢀ1, DCM/CH2Cl2): 3005-2835 (CH package), 1614 (CO), 1510,
1440, 1363, 1292, 1273, 1211, 1192, 1139, 1109.
Yield 90%, yellow solid, mp 158–160 ꢁC, Rf 0.32 (5% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d
ꢂ15.9 (br, enolic OH), 7.64 (d,
5.14. (1E,6E)-4,4-Difluoro-1,7-bis(3,4-dimethoxy-6-fluorophenyl)
hepta-1,6-diene-3,5-dione (16)
J = 16.0 Hz, 2H), 7.56 (dd, J = 8.5 and 15.0 Hz, 4H), 7.10 (t appearance,
J = 8.5 Hz, 4H), 6.51 (d, J = 15.9 Hz, 2H), 5.82 (s, 1H). 13C NMR (CDCl3,
125 MHz):
Hz),129.9 (d, JCF = 8.6 Hz),123.7,116.1 (d, JCF = 22 Hz),101.8.19F NMR
(CDCl3, 470 MHz):
- 109.7 (m). HRMS (ESI): m/z [M+H]+ calcd for
19H15O2F2: 313.1040; found: 3313.100. IR (cmꢀ1, DCM): 3066-2850
d
183.1, 163.8 (d, 1JCF = 250.8 Hz), 139.4, 131.2 (d, JCF = 2.9
Yield: 26%, brown solid, mp 162–163 ꢁC. Rf 0.53 (40% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d 8.05 (d, J = 16.0 Hz, 2H), 7.05
d
(d, J = 16.0 Hz, 2H), 7.02 (d, JHF = 6.7 Hz, 2H), 6.67 (d, JHF = 11.6 Hz,
C
2H), 3.93 (s, 6H, OMe), 3.92 (s, 6H, OMe). 13C NMR (CDCl3,
(CH package), 1631 (CO), 1593, 1508, 1414, 1234, 1150.
125 MHz):
d
186.7 (t, 2JCF = 27.6 Hz), 158.0 (d, 1JCF = 252.6 Hz), 153.8
(d, JCF = 10.5 Hz), 145.8 (d, JCF = 2.0 Hz), 140.9, 117.0 (d, JCF = 6.6 Hz),
5.10. (1E,6E)-4,4-Difluoro-1,7-bis(4-fluorophenyl)hepta-1,6-diene-
3,5-dione (12)
1
113.1 (d, JCF = 12.4 Hz), 111.7 (t, JCF = 264.2 Hz, CF2), 109.3 (d,
J
CF = 3.8 Hz), 100.1 (d, JCF = 28.6 Hz), 56.5, 56.4. 19F NMR (CDCl3,
Yield 42%, white solid, mp 72–73 ꢁC, Rf 0.35 (5% EtOAc in
470 MHz):
d
- 115.4 (s, 2F), ꢀ117.6 (dd, JHF = 11.7 and 5.8 Hz, 2F).
HRMS (ESI): m/z [M+H]+ calcd for C23H21O6F4: 469.1274; found:
hexane). 1H NMR (CDCl3, 500 MHz):
d 7.88 (d, J = 16.5 Hz, 2H),

40
K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
469.0499. IR (cmꢀ1, CDCl3/DCM): 308, 2941, 1707, 1693, 1512, 1442,
1365, 1280, 1193.
(30%)], orange solid, mp 144–146 ꢁC. Rf 0.64 (10% EtOAc in hexane).
1H NMR (CDCl3, 500 MHz):
ꢂ13.5 (br, enolic OH), 7.22 (dd, J = 15.5
and 3.7 Hz, 2H), 7.76-7-67 (unresolved-m, 10H). 19F NMR (CDCl3,
d
5.15. (1E,4E,6E)-4-Fluoro-5-hydroxy-1,7-bis(3,4-dimethoxy-6-
fluorophenyl)hepta-1,4,6-trien-3-one (17)
470 MHz):
d
ꢀ 62.8 (6F, CF3), ꢀ173.9 (t, JHF = 3.2 Hz,1F). HRMS (ESI):
m/z [M+H]+ calcd for C21H14O2F7: 431.0882; found: 431.0804.
Yield 10%, red solid, mp 148–150 ꢁC. Rf 0.58 (40% EtOAc in
6. Bioassay methods
hexane). 1H NMR (CDCl3, 500 MHz):
d
7.82 (d, J = 16 Hz, 2H), 7.08–
7.03 (m, 4H), 6.67 (d, JHF = 11.5 Hz), 3.93 (s, OMe), 3.942(s, OMe).13
C
The cell viability/anti-proliferative activity of the curcuminoids
against PC3 (human androgen-insensitive prostate cancer cell
line), LNCap (human-androgen sensitive prostate cancer cell line),
A549 (lung cancer), and MDA231 (breast cancer) were determined
by means of 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetra-
zolium bromide (MTT) assay (the tetrazolium salt was a commer-
cial sample) [30]. The ability of the curcuminoids to affect
proliferations of suspension cell lines (MOLT-4) was tested by
the CellTiter-Glo1 Luminescent Cell Viability Assay (purchased
from Promega Madison, WI, USA) to determine the number of
viable cells in culture based on quantitation of the ATP present. The
IC50 values were obtained from fitting data with GraphPad
software to determine the growth inhibition in the presence of
test compounds [30].
2
1
NMR (CDCl3, 125 MHz): 172.0 (d, JCF = 21 Hz), 156.9 (d, JCF = 250
Hz), 152.2 (d, JCF = 10.4 Hz), 145.7, 143.1 (d, JCF = 238.5 Hz), 134.2,
116.8 (d, JCF = 5.8 Hz), 114.2 (d, JCF = 12.4 Hz), 109.6 (d, JCF = 3.7 Hz),
100.3, 56.4. 56.3.19F NMR (CDCl3, 470 MHz):
d
ꢀ 119.5 (dd, JHF = 11.7
and 6.6. Hz, 2F), ꢀ175.3 (distorted t, J = 3.3 Hz, 1F). HRMS (ESI): m/z
[M+H]+ calcd for C23H22O6F3 450.1368; found: 451.0632. IR (cmꢀ1
,
CH2Cl2/CDCl3): 2922, 1608, 1548, 1504, 1367, 1276, 1213, 1157, 1066.
5.16. (1E,6E)-4-Fluoro-1,7-bis(3,4-dimethoxy-6-fluorophenyl)hepta-
1,6-diene-3,5-dione (18)
Yield 12% (by NMR). 119F NMR (CDCl3, 470 MHz):
HF = 50 Hz, 1F), ꢀ118.59 (m, 2F)
d
ꢀ195.2 (d,
J
5.17. Bis-trifluoromethylcurcuminoid-BF2 adduct (19)
7. X-ray crystallography
Yield 88%, yellow solid, mp > 260 ꢁC. Rf 0.14 (10% EtOAc in
Suitable single crystals for X-ray diffraction studies were
obtained for compounds 10, 5 and 8 from ethyl acetate and for
14 from dichloromethane. Crystal data of the compounds were
collected by exactly the same method by mounting a crystal onto a
thin glass fiber from a pool of FluorolubeTM and immediately
placing it under a liquid N2 cooled stream, on a Bruker AXS
diffractometer upgraded with an APEX II CCD detector. The
hexane). 1H NMR (DMSO-d6, 500 MHz):
8.14 (d, J = 15.8 Hz, 2H),
8.10 (d, J = 8.5 Hz, 4H), 7.87 (d, J = 8.0 Hz, 4H), 7.44 (d, J = 15.8 Hz, 2H),
6.77 (s, 1H). 13C NMR (DMSO-d6, 125 MHz):
180.6, 145.2, 145.1,
137.7, 131.0 (q, JCF = 31.4 Hz),130.1,125.9 (q, JCF = 3.8 Hz),124.1, 123.9
d
d
1
(q, JCF = 272.7 Hz, CF3), 103.3. 19F NMR (DMSO-d6, 470 MHz):
d
ꢀ
61.4 (s, CF3), ꢀ136.5 (11B-F), ꢀ136.4 (10B-F). 11B NMR (DMSO-d6,
160.3 MHz):
d
0ꢃ96 (br,s). IR (cmꢀ1, DCM): 1620, 1529, 1514, 1404,
radiation used is graphite monochromatized Mo K
a radiation
1321, 1166, 1111, 1064.
(l= 0.7107 Å). The lattice parameters are optimized from a least-
squares calculation on carefully centered reflections. Lattice
determination, data collection, structure refinement, scaling, and
data reduction were carried out using APEX2 Version 2014.11
software package [31,32]. The data were corrected for absorption
using the SCALE program within the APEX2 software package
[31,32]. The structure were solved using SHELXT [33]. This
procedure yielded a number of the C, B, F and O atoms. Subsequent
Fourier synthesis yielded the remaining atom positions. The
hydrogen atoms are fixed in positions of ideal geometry (riding
model) and refined within the XSHELL software package [34].
These idealized hydrogen atoms had their isotropic temperature
factors fixed at 1.2 or 1.5 times the equivalent isotropic U of the C
atoms to which they were bonded. A few hydrogen atoms could not
be adequately predicted via the riding model within the XSHELL
software [34], these hydrogen atoms were located via difference-
Fourier mapping and subsequently refined. The final refinement of
each compound included anisotropic thermal parameters on all
non-hydrogen atoms. The crystal data for the compounds are given
in Table 1. Packing diagrams [35] and thermal ellipsoid plots along
with selected interatomic distances and bond angles are included
in Supplementary data.
5.18. (1E,4E,6E)-1,7-Bis(4-trifluoromethyl-phenyl)-5-hydroxy-hepta-
1,4,6-trien-3-one (20)
Yield 88%, yellow solid, mp 153–154 ꢁC, Rf 0.56 (10% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d
ꢂ15.9 (br, enolic OH), 7.95 (d,
J = 8.0 Hz, 2H), 7.79 (d, J = 8.0 Hz, 2H), 7.72 (d, J = 15.9 Hz, 2H), 7.13 (d,
J = 15.8 Hz, 2H), 6.28 (s, 1H). 13C NMR (CDCl3, 125 MHz):
182.9,
138.3, 131.5 (q, JCF = 31.5 Hz), 128.9, 126.9, 125.7 (q, JCF = 4 Hz), 124.0
d
(q, 1JCF = 271.8 Hz, CF3), 102.6.19F NMR (CDCl3, 470 MHz):
d - 61.2 (s,
CF3). HRMS (ESI): m/z [M+H]+ calcd for C21H15O2F6: 413.0976;
found: 413.087. IR (cmꢀ1, DCM): 2928, 1635 (CO), 1579, 1413, 1330,
1265, 11681128, 1066.
5.19. (1E,6E)-4,4-Difluoro-1,7-bis(4-trifluoromethyl-phenyl)hepta-
1,6-diene-3,5-dione (21)
Yield 36%, white solid, mp 75–76 ꢁC, Rf 0.72 (10% EtOAc in
hexane). 1H NMR (CDCl3, 500 MHz):
d
7.93 (d, J = 15.9 Hz, 2H), 7.76
(d, J = 8.5 Hz, 4H), 7.71 (d, d, J = 8.5 Hz, 4H), 7.19 (d, J = 16 Hz, 2H). 13
NMR (CDCl3, 125 MHz):
186.6 (t, 2JCF = 27.7 Hz), 146.6, 136.8, 133.2
(q, JCF = 3.8 Hz), 129.3, 126.1 (q, JCF = 3.8 Hz), 123.6 (q, 1JCF = 272.7 Hz,
C
d
CF3), 120.0, 112.1 (t, 1J = 265.1 Hz, CF2).19F NMR (CDCl3, 470 MHz):
d
8. Computational methods
- 63.1 (s, 6F, CF3), ꢀ115.1 (s, 2F, CF2). HRMS (ESI): m/z [M+H]+ calcd
for C21H13O2F8: 449.0787; found: 449.0145. IR (cmꢀ1, CDCl3): 3080,
2933, 1699 (CO), 1608, 1577, 1417, 1319, 1168, 1124, 1066.
Geometry optimizations of the curcumin derivatives were
performed at the B3LYP /6-311 + G(d,p) level [36] with the
Gaussian 09 package [37]. Distribution of the electrostatic
potential derived from the electron density was estimated by
energy calculations at the optimized structures. The programs
AutoDock 4.2 [38] and AutoDock Vina [39] were employed to carry
out automated molecular docking for estimating the interaction
energy and modeling the binding modes between the curcuminoid
5.20. (1E,4E,6E)-4-Fluoro-1,7-bis(4-trifluoromethyl-phenyl)-5-
hydroxy-hepta-1,4,6-trien-3-one (22)
Yield 60% [crude yield, contained 20 (22%) and 19 (18%); 16%
yield after recrystallization (purity by NMR was 70%, contained 19

K.K. Laali et al. / Journal of Fluorine Chemistry 191 (2016) 29–41
41
ligands and the enzymes HER2 and proteasome. The three-
dimensional coordinates of the proteins were obtained from the
Protein Data Bank (PDB codes 3PP0 [40] (HER2) and 3SDK [41] (20S
[7] M.W. Amolins, L.B. Peterson, B.S.J. Blagg, Bioorg. Med. Chem. 17 (2009) 360–
367.
[8] M. Labbozzetta, R. Baruchello, P. Marchetti, M. Gueli, P. Poma, M. Notarbartolo,
D. Simoni, N. D’Alessandro, Chem. Biol. Interact. 181 (2009) 29–36.
[9] R. Narlawar, M. Pickhardt, S. Leuchtenberger, K. Bauman, S. Krause, T. Dyrks, S.
Weggen, E. Mandelkow, B. Schmidt, ChemMedChem 3 (2008) 165–172.
[10] S.B. Wan, H. Yang, Z. Zhou, Q.-C. Cui, D. Chen, J. Kanwar, I. Mohammad, Q.P. Dou,
T.H. Chan, Int. J. Mol. Med. 26 (2010) 447–455.
[11] L. Feng, Z.F. Song, H.J. Li, Q.-Y. Huai, Chem. Pharm. Bull. 63 (2015) 873–881.
[12] W. Wichitnithd, U. Nimmannit, S. Wacharasindhu, P. Rojsitthisak, Molecules 16
(2011) 1888–1900.
[13] Z. Cheikh-Ali, J. Caron, S. Cojean, C. Bories, P. Couveur, P.M. Loiseau, D.
Desmaele, E. Poupon, P. Champy, ChemMedChem 10 (2015) 411–418.
[14] R. Sribalan, M. Kirubavathi, G. Banuppriya, V. Padmini, Bioorg. Med. Chem. Lett.
25 (2015) 4282–4286.
[15] K. Wada, J.Y. Lee, H.Y. Hung, Q. Shi, L. Lin, Y. Zhao, M. Goto, P.C. Yang, S.-C. Kuo,
H.W. Chen, K.-H. Lee, Bioorg. Med. Chem. 23 (2015) 1507–1514.
[16] L. Lin, Q. Shi, C.-Y. Su, C.C.-Y. Shih, K.-H. Lee, Bioorg. Med. Chem. 14 (2006)
2527–2534.
proteasome)). Chain A of HER2, and chains K (b5 subunit) and L
(b
6 subunit) of 20S proteasome were selected as target templates
for the docking calculations. Co-crystalized ligands and crystallo-
graphic water molecules were removed. Addition of hydrogens,
merger of non-polar hydrogens to the atom to which they were
linked, and assignment of partial charges were achieved with
AutoDockTools. Merz-Kollman partial atomic charges were
employed for proteins, and Gasteiger charges were assigned to
ligands. The docking area was defined using the AutoDock module
AutoGrid. The docking area, defined using the AutoDock module
AutoGrid, was constrained to a 30 ꢄ 26.2 ꢄ 30 Å box centered at the
active site, providing proper space for rotational and translational
movement of the ligands. With AutoDock 4.2, the Lamarckian
genetic algorithm (LGA) was used, default parameters were
applied, and the maximum number of energy evaluations was
set to 1.0 ꢄ107. For each of the 100 independent runs performed for
each ligand a maximum number of 2.7 ꢄ104 genetic algorithm
[17] L. Lin, Q. Shi, A.K. Nyarko, K.F. Bastow, C.-C. Wu, C.-Y. Su, C.C.-Y. Shih, K.-H. Lee, J.
Med. Chem. 49 (2006) 3964–3972.
[18] J. Wang, M. Sanchez-Rosello, J.L. Acena, C. del Pozo, A.E. Sorochinsky, S. Fustero,
V.A. Soloshonok, H. Liu, Chem. Rev. 114 (2014) 2432–2506.
[19] H.-J. Bohm, D. Banner, S. Bendels, M. Kansy, B. Kuhn, K. Muller, U. Obst-Sander,
M. Stahl, ChemBioChem 5 (2004) 637–643.
[20] E.V. Rao, P. Sudheer, Indian J. Pharm. Sci. 73 (2011) 262–270.
[21] S. Elavarasan, D. Bhakiaraj, T. Elavarasan, M. Gopalakrishnan, J. Chem. (2013) 1–
8, doi:http://dx.doi.org/10.1155/2013/640936.
[22] (a) S. Padhye, H. Yang, A. Jamadar, Q.C. Cui, D. Chavan, K. Dominiak, J.
McKinney, S. Banerjee, Q.P. Dou, F.H. Sarkar, Pharm. Res. 26 (2009) 1874–1880;
(b) S. Padhye, S. Banerjee, D. Chavan, S. Pandye, K.V. Swamy, S. Ali, J. Li, Q.P. Dou,
F.H. Sarkar, Pharm. Res. 26 (2009) 2438–2445.
operations were generated on
a single population of 150
individuals. Operator weights for crossover, mutation, and elitism
were default parameters, 0.80, 0.02, and 1, respectively. The default
parameters were used for Vina.
[23] (a) K. Kamada, S. Senatore, C. Metthews, P.-F. Lenne, O. Maury, C. Andraud, M.
Ponce-Vargas, B. Le Guennic, D. Jacquemin, P. Agbo, D.D. An, S.S. Gauny, X. Liu,
R.J. Abergel, F. Fages, A. D’Aleo, Chem. Eur. J. 22 (2016) 5219–5232;
(b) G. Bai, C. Yu, C. Cheng, E. Hao, Y. Wei, X. Mu, L. Jiao, Org. Biomol. Chem. 12
(2014) 1618–1626.
[24] A.S. Reddy, K.K. Laali, Tetrahedron Lett. (2015) 5495–5499.
[25] M.G. Shioorkar, M.B. Ubale, S.A. Jadhav, R.K. Pardeshi, Der Chemie Sinica 6
(2015) 110–113.
Acknowledgements
B.R. and K.L. thank The Camille and Henry Dreyfus Foundation
for summer research support (to B. R.) in the context of a Jean
Dreyfus Boissevain Lectureship award to UNF. This research was
supported in part by a grant from the University of North Florida’s
John A. Delaney Presidential Professorship (to K.L.) and by grants
from Consejo Nacional de Investigaciones Científicas y Técnicas
(CONICET) and the Secretaría de Ciencia y Tecnología de la
Universidad Nacional de Córdoba (Secyt-UNC) (to G.B.). We thank
Vivian S. Hogan (Kent State) for assistance with crystallization,
University of Florida for access to computational facilities at UF
High-Performance Computing Center, and NCI of NIH develop-
mental therapeutics program (DTP) for NCI 60 screening.
[26] S. Raghavan, P. Manogaran, K.K.G. Narasimha, B.K. Kuppusami, P. Mariyappan,
A. Gopalakrishnan, G. Venkatraman, Bioorg. Med. Chem. Lett. 25 (2015) 3601–
3605.
[27] J. Chen, Z. Huang, C. Wang, S. Porter, B. Wang, W. Lie, H.K. Liu, Chem. Commun.
51 (2015) 9809–9812.
[28] N. Iqbal, N. Iqbal, Mol. Biol. Int. 2014 (2014) 852748.
[29] S. Xu, R.E. Evans, T. Liu, G. Zhang, J.N. Demas, C.O. Trindle, C.L. Fraser, Inorg.
Chem. 52 (2013) 3597–3610.
[30] R.N. Khaybullin, M. Zhang, J. Fu, X. Liang, T. Li, A.R. Katritzky, P. Okunieff, X. Qi,
Molecules 19 (2014) 18676–18689.
[31] APEX II, Analytical X-ray Systems, Bruker AXS, Inc., 5465 East Cheryl Parkway,
Madison WI, 2005, pp. 53711–55373.
[32] SAINT Version 6.45A, Analytical X-ray Systems, Bruker AXS, Inc., 5465 East
Appendix A. Supplementary data
Cheryl Parkway, Madison WI, 2003, pp. 53711–55373.
[33] G.M. Sheldrick, SHELXT Acta Cryst. A71 (2015) 3–8.
[34 SHELXTL V6.12, Analytical X-ray Systems, Bruker AXS, Inc., 5465 East Cheryl
Parkway, Madison WI, 2002, pp. 53711–55373.
[35] K. Brandenburg, DIAMOND, Crystal Impact GbR, Bonn, Germany, 1999.
[36] (a) A.D. Becke, J. Chem. Phys. 98 (1993) 5648–5652;
(b) C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785–789;
(c) B. Miehlich, B.A. Savin, H. Stoll, H. Preuss, Chem. Phys. Lett. 57 (1989) 200–
206.
Supplementary data associated with this article can be found, in
the
online
version,
at
http://dx.doi.org/10.1016/j.
jfluchem.2016.09.009.
References
[37] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, et al., Gaussian 09,
Gaussian, Inc, Wallingford, CT, 2009.
[1] S.C. Gupta, S. Prasad, J.H. Kim, S. Patchva, L.J. Webb, I.K. Priyadarsini, B.B.
Aggarwal, Nat. Prod. Rep. 28 (2011) 1937–1955 (Review).
[2] A. Vyas, P. Dandawate, P.S. Padhye, A. Ahmad, A.F. Sarkar, Curr. Pharm. Des. 19
(2013) 2047–2069 (Review).
[3] M. Mimeault, S.K. Batra, Chin. Med. 6 (2011) 1–19 (Review).
[4] D. Simoni, M. Rizzi, R. Rondanin, R. Baruchello, P. Marchetti, F.P. Invidiata, M.
Labbozzetta, P. Poma, V. Carina, M. Notarbartolo, A. Alaimo, N. D’Alessandro,
Biorg. Med. Chem. Lett. 18 (2008) 845–849.
[38] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J.
Olson, J. Comput. Chem. 30 (2009) 2785–2791.
[39] O. Trott, O.A.J. Olson, J. Comput. Chem. 31 (2010) 455–461.
[40] K. Aertgeerts, R. Skene, J. Yano, B.-C. Sang, H. Zou, G. Snell, A. Jennings, K.
Iwamoto, N. Habuka, A. Hirokawa, T. Ishikawa, T. Tanaka, H. Miki, Y. Ohta, S.J.
Sogabe, J. Biol. Chem. 286 (2011) 18756–18765.
[41] C. Blackburn, C. Barrett, J.L. Blank, F.J. Bruzzese, N. Bump, L.R. Dick, P. Fleming,
K. Garcia, P. Hales, Z. Hu, M. Jones, J.X. Liu, D.S. Sappal, M.D. Sintchak, C. Tsu, K.
M. Gigstad, Bioorg. Med. Chem. Lett. 20 (2010) 6581–6586.
[5] K.-L. Tan, S.-B. Koh, R.P.-L. Ee, M. Khan, M.-L. Go, ChemMedChem 7 (2012)
1567–1579.
[6] C.L. Nieto, M.P. Cabildo, M.P. Cornago, D. Sanz, R.M. Claramunt, M.C. Torralba,
M.R. Torres, J. Elguero, J.A. Garcia, A. Lopez, D. Acuna-Castroviejo, Molecules 20
(2015) 15643–15665.