Inhibitory effect of β-diketones and their metal complexes on TNF-α induced expression of ICAM-1 on human endothelial cells

Article abstract of DOI:10.1016/j.bmc.2009.07.064

Recent reports show that the natural β-diketone curcumin displays important biological properties regarding the intercellular adhesion molecule-1 (ICAM-1), which plays a critical role in the immune responses and inflammation. In this study the ICAM-1 inhi

Full text of DOI:10.1016/j.bmc.2009.07.064

Bioorganic & Medicinal Chemistry 17 (2009) 6166–6172
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Inhibitory effect of b-diketones and their metal complexes on TNF-
a
induced expression of ICAM-1 on human endothelial cells
b
b
c
d
e
Francesco Caruso a, , Claudio Pettinari , Fabio Marchetti , Miriam Rossi , Cristian Opazo , Sarvesh Kumar ,
*
Sakshi Balwani e, Balaram Ghosh e,
*
a Istituto di Chimica Biomolecolare, Consiglio Nazionale delle Ricerche (CNR), c/o Università di Roma ‘La Sapienza’, Vecchio Istituto Chimico, Piazzale Aldo Moro 5, 00185 Rome, Italy
b Dipartimento Scienze Chimiche, Università di Camerino, Via S. Agostino 1, 62032 Camerino (MC), Italy
c Department of Chemistry, Vassar College, Poughkeepsie, NY 12604-0484, USA
d Academic Computing Services CIS, Vassar College, Poughkeepsie, NY 12604-0748, USA
e Molecular Immunogenetics Laboratory, Institute of Genomics and Integrative Biology, University of Delhi Campus (North), Mall Road, Delhi 110 007, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Recent reports show that the natural b-diketone curcumin displays important biological properties
regarding the intercellular adhesion molecule-1 (ICAM-1), which plays a critical role in the immune
responses and inflammation. In this study the ICAM-1 inhibitory activity of b-diketone compounds,
which are curcumin models lacking aromatic peripheral hydroxyl and methoxy groups, along with some
metal derivatives is investigated. b-Diketones are systematically more active than metal complexes and
the best obtained inhibition is 75% for both groups. The best inhibitors are 4-benzoyl-3-methyl-1-phenyl-
pyrazol-5-one (HQPh) among the ligands, and sodium benzoylacetonato among metal derivatives. These
results appear in line with the reported antitumor activity of related species. Since 4-acyl-5-pyrazolones
posses four tautomeric forms, those corresponding to HQPh were investigated using density functional
theory. Docking of all HQPh tautomers on ICAM-1 protein was performed suggesting one keto-enol form
favored to act in biological environment.
Received 15 April 2009
Revised 22 July 2009
Accepted 26 July 2009
Available online 30 July 2009
Keywords:
Inflammation
Cell adhesion molecules
Human endothelial cells
Anti-inflammatory
Metal-b-diketones
DFT
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
and hence inflammation.4–6 A regulated expression of cell adhesion
molecules is therefore essential for maintaining the body fluidity.
The adhesion of leukocytes to the endothelium is among the
earliest and most important process during any kind of inflamma-
tory response.1 The trans-endothelial migration of the leukocytes
requires the increased expression of cell adhesion molecules on
the surface of endothelial cells that interact with their correspond-
ing receptors on the surface of leukocytes.2 These leukocyte cell
surface proteins are commonly known as cell adhesion molecules.
The major cell adhesion molecules involved in this process are
intercellular cell adhesion molecule-1 (ICAM-1), vascular cell
adhesion molecule-1 (VCAM-1) and E-selectin expressed on endo-
thelium in response to various cytokines or bacterial lipopolysac-
charide.3 The increased expression of cell adhesion molecules
alters the adhesive property of the vasculature leading to indis-
criminate infiltration of the leukocytes across the blood vessels
Inhibition of cell adhesion molecules is a useful therapeutic ap-
proach to regulate inflammatory response.7 Small molecules from
natural and synthetic sources have been used successfully for
down regulating the induced expression of cell adhesion molecules
both in vitro and in vivo.8–12 Several medicinal herbs have been
shown to augment specific cellular and humoral immune re-
sponses.13–16
Curcumin, an important component of curry, is a natural
antioxidant and antitumor polyphenol b-diketone isolated from
Curcuma longa. Recently, it was shown that the expression of
ICAM-1 was attenuated by curcumin at both mRNA and protein
levels.17 Also, pretreatment with curcumin blocked interleukin-1
b induced ICAM-1 expression in A549 cells.18 In addition, the anti-
tumor effect of gemcitabine (the best treatment available for pan-
creatic cancer) is potentiated by curcumin through suppression of
proliferation, angiogenesis and inhibition of nuclear factor-jB-reg-
ulated gene products.19 Moreover, reduction of matrixmetallopro-
teinase-2 activity and inhibition of carcinoma HEP2 cell invasion
by curcumin, strongly indicates the potential of curcumin as an
inhibitor of tumor cell invasion and metastasis.20 Furthermore,
intratracheal injection of bleomycin to rats induces acute lung
Abbreviations: ICAM-1, intercellular adhesion molecule-1; VCAM-1, vascular
cell adhesion molecule-1; TNF-a, tumor necrosis factor-a; HUVEC, human umbilical
vein endothelial cells.
* Corresponding authors. Tel.: +39 06 49913632; fax: +39 06 49913628 (F.C.);
tel.: +91 11 2766 2580; fax: +91 11 2766 7471 (B.G.).
E-mail addresses: caruso@vassar.edu (F. Caruso), bghosh@igib.res.in (B. Ghosh).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.07.064

F. Caruso et al. / Bioorg. Med. Chem. 17 (2009) 6166–6172
6167
Table 1
injury accompanied by increased lung levels of intercellular ICAM-
ICAM-1 inhibitory activity of b-diketones and their metal complexes
1 seven days after bleomycin administration. However, curcumin
treatment significantly inhibited bleomycin-induced effects and
restored the levels of antioxidant enzymes to normal values.21
Several years ago we initiated a systematic study using 4-acyl-
5-pyrazolones (Scheme 1) as ligands for several metal cations and
some of their related complexes showed marked antitumor activ-
ity.22,23 4-Acyl-5-pyrazolones are a class of asymmetric b-dike-
tones that differ from curcumin by lacking aromatic hydroxyl
and methoxy substituents. Since their biological activity should
be due only to the diketo moiety, they provide indirect information
about the mechanism of action of curcumin. In this study, we
investigate the inhibition of cell surface expression of ICAM-1 by
several diketones along with their metal complexes, and attempt
to clarify the underlying chemical mechanism behind the inhibi-
tion. The study includes an ICAM-1 docking for the best found
inhibitor, clarifying the related tautomer preference.
Name of
compound
Structure of
compound
Concentration
(lg/ml)
% Inhibition
% Viability
DMSO
C-1
C-2
C-3
C-4
C-5
C-6
L-1
L-2
L-3
L-4
L-5
L-6
Solvent
0.25%
120
35
80
70
80
80
60
60
60
60
60
60
—
97.1
98.3
96.4
95.8
97.1
96.0
98.7
96.2
95.8
96.1
98.6
97.1
97.9
Ca(QnPe)2(EtOH)2
Zn(QnPe)2(H2O)2
NaQPh
40
35
25
75
45
15
55
75
60
35
35
20
Na(bzac)
Na(bzfc)
Zn(QtBu)2(H2O)2
HQFur
HQPh
HQt-But
HQEtCp
HQnPe
HQPh,Ph,Me
HQPh is shown in Figure 2 with IC50 of 36.3
shows only 62% ICAM-1 inhibition at the highest dose of 271.4
ml (IC50 = 98.5
M/ml) whereas curcumin26 shows up to 90%
l
g/ml. However, HQPh
2. Results and discussion
lM/
l
The tested compounds were found to be non-toxic up to their
maximum tolerable concentrations as shown in Table 1. We have
found that more than 95% cells were viable at these concentrations
(Table 1). At the same time, DMSO (used as a solvent) was also not
causing any cytotoxic effect to endothelial cells. These data show
that these compounds are safe for endothelial cells.
ICAM-1 inhibition at the highest dose of 50 lM/ml. These data sug-
gest that curcumin is more active than HQPh in inducing the cyto-
kine-induced expression of I-CAM-1 on HUVEC.
4-Acyl-5-pyrazolones can display several tautomeric forms
shown in Scheme 3. To estimate the potential active form acting
on ICAM-1 we calculate the molecular structure of each tautomer
in water for the most active ligand found in this study, HQPh, using
density functional theory (DFT) and compare with X-ray structures
in the literature,27,28 see Table 2. Structural agreement between X-
ray27 and DFT methods is excellent for the N–H diketo form I; tor-
sion angles of keto-enol II will be discussed later. DFT calculations
show both diketo forms displaying the two carbonyls with OÁ Á ÁO
separations longer than 3 Å due to electronic repulsion, Figures 3
and 6, whereas both keto-enol structures allow a closer approach
of both oxygens due to the linking intramolecular H-bond, Figures
4 and 5. Diffraction studies show 4-acyl-5-pyrazolonato metal
complexes have a tendency for co-planarity of the pyrazole ring
and its attached Ph. The expected torsion angles (closely related
to the dihedral angle between both rings) should be about 0°,
but crystal packing29 can increase this value. In this study the
DFT structures are exempt of packing effects as the molecule is cal-
culated in water solution. We conclude that the X-ray torsion angle
of the keto-enol tautomer II (À30.8° and 24.7°) do not compare
well with the calculated DFT value (À2.0°) because of packing ef-
fects in the crystal27 and so the DFT structure resembles better,
in this case, the situation in the biological environment.
Similar torsion angles, far from co-planarity, found in the X-
ray27 (20.6°) and DFT (23.0°) structures of N–H diketo form are ex-
plained by a different factor, namely, the steric hindrance between
the H bound to N and an ortho-H atom in the Ph ring. On the con-
trary, the C–H tautomer displays the H (of C) away from the Ph and
the DFT structure shows more co-planarity with a torsion angle of
6.2°. There is no preference among both keto-enol forms (II and III)
as their energy difference is only 1.0 kcal/mol. This contrasts with
metal derivative structures that show preference for M–O(pyrazol-
onato) bond shorter than M–O(acyl), consistent with tautomer
form II. A dynamic H atom moving between both oxygens would
explain the DFT result. The energy difference (Delta E = 4.5 kcal/
mol) between both diketo forms (C–H and N–H) favors the latter
tautomer, which is consistent with the fact that the N–H tautomer
is the one found in the crystal.27 The energy difference between
calculated structures of tautomers found in the crystalline state
(keto-enol II and diketo N–H I) favors the former (Delta
E = 3.7 kcal/mol), suggesting keto-enol as potentially more prone
to act in biological environment.
Tested compounds in this study are shown in Scheme 2 while
Table 1 indicates that although the maximum level of inhibition
is the same for ligands and metal complexes (75%), the concentra-
tion needed for such an endeavor is always higher for the latter. It
appears that the ligands are more effective in the used model be-
cause of a straightforward interaction with the target. Therefore,
the effect of the metal in their complexes seems to be to decrease
ligand availability. On the contrary, antitumor studies of titanium
b-diketonates show the Ti-b-diketone binding is strong and neces-
sary for biological activity. In fact, titanium complexes were the
most active metal derivatives in studies performed by Keppler.24,25
We conclude that there is no potential benefit of ligand protection
though metal coordination in the present model. The antitumor
activity in Keppler’s studies show benzoylacetonato to be the best
ligand for Ti complexes. Interestingly, benzoylacetonato, used in
compound C-4, gives the best inhibition in this study among metal
derivatives (C-1 to C-6). Consistently, HQPh is the the most active
among all tested ligands (L-1 to L-6), and forms a Ti complex with
antitumor activity22 of the same order (T/C around 300%) than the
best one in Keppler studies.25 Therefore, we selected HQPh and pro-
ceeded to test its activity at lower concentration. First we analyzed
viability, Figure 1, and found that HQPh is less cytotoxic than curcu-
min, as the maximum tolerable doses are 100
lg/ml = 271.4 lM/
ml and 50
l
M/ml,26 respectively. The specific inhibitor activity of
Scheme 1. 4-Acyl-5-pyrazolones, only compounds containing R1 = Ph were used in
the present study.

6168
F. Caruso et al. / Bioorg. Med. Chem. 17 (2009) 6166–6172
Scheme 2. Tested compounds.
Figure 1. Cytotoxicity of HQPh on human umbilical vein endothelial cells (HUVEC).
Figure 2. ICAM-1 inhibition of HQPh on HUVEC.
Therefore, we performed a docking study on ICAM-1 to see its
preference for HQPh tautomers. Although there are no crystal struc-
tures including guests for this protein, a number of ICAM-1-bind-
ing isolates have been identified and analyzed in detail;30 they
all involve the L43 loop of ICAM-1 (aminoacids 42–49). Such loop
has been recently successfully employed to design an inhibitor of
Plasmodium falciparium, (+)-epigalloyl-catechin-gallate [(+)EGCG],
after exploring in silico over 3 million commercial compounds.31
We proceeded extracting coordinates of ICAM-1 from the Pro-
tein Data Bank, PDB code 1IAM.32 and performed the docking for
the four tautomers plus (+)EGCG, operating with Discovery Studio
2.1. After CHARMm force-field H generation H coordinates in the
active site were made consistent with the crystal structure. In a
first simulation, the active site was kept fixed and the four tautom-
ers and (+)EGCG in turn were allowed to fit. Ten poses were calcu-
lated for each guest and binding energy, which describes guest
affinity for the active site pocket, was calculated for each mole-
cule’s pose. Later, ligand minimizations for the active site includ-
ing, in turn, all docked molecules were performed. Energy
affinity, Table 3, of the best fit molecules show that keto-enol (II)
fits better than keto-enol (III), and, as mentioned earlier, this is
consistent with all HQPh metal derivatives showing structures hav-
ing M–O(pyrazolonato) bond shorter than M–O(acyl). Interest-
ingly, the affinity of the CH diketo (IV) tautomer is highest for all

F. Caruso et al. / Bioorg. Med. Chem. 17 (2009) 6166–6172
6169
Scheme 3. Tautomer forms of 4-acyl-5-pyrazolones. I: N–H diketo, II: keto-enol(pyrazolone), III: keto-enol(acyl), IV: C–H diketo.
Table 2
Tautomer structures of HQPh
Tautomer
Method
Torsion
d(C@O)
d(C–OH)
d(OÁ Á ÁO)
3.08
3.034
2.66
2.62
2.495
2.523
3.257
Energy (–)
Diketo N–H (I)
Diketo N–H (I)
Keto-enol (II) (a)
Keto-enol (II) (b)
Keto-enol (II)
X-ray DEBFARR1
DFT
20.6
23.0
À30.8
24.7
À2.0
À8.4
6.2
1,23, 1.24
1.239, 1.245
1.25
1.24
1.277
916.439002
X-ray YUYDOL
1.32
1.32
1.323
1.431
X-ray YUYDOLR2
DFT
DFT
DFT
916. 444969
916.446720
916.431832
Keto-enol (III)
Diketo C–H (IV)
1.266
1.230, 1.232
Notes: R1 (DEBFAR) and R2 (YUYDOL) are refcodes in the crystal structure CSD database. The keto-enol form (YUYDOL) crystallizes with 2 independent molecules in the unit
cell (a and b); their geometrical values are displayed in rows 4th and 5th. Energy unit: hartree; distances: Å.
Figure 5. DFT calculated keto-enol form III, showing intramolecular hydrogen
bond.
Figure 3. DFT calculated diketo N–H form I.
Figure 6. DFT calculated diketo, C–H form IV.
Figure 4. DFT calculated keto-enol form II, showing intramolecular hydrogen bond.
3. Experimental section
tautomers, but this chemical form has not been verified experi-
mentally and since its internal energy is 8.3 kcal/mol higher than
keto-enol (II) and 4.5 kcal/mol higher than NH diketo (I), the prob-
ability of its existence is very low at room temperature; the good
docking performance of C–H diketo (IV) may stimulate further
studies to isolate such tautomer form. We conclude that keto-enol
(II), depicted in Figure 7, is the most probable tautomer acting at
ICAM-1 active site.
3.1. Isolation and culture of endothelial cells
Primary endothelial cells were isolated from human umbilical
cord as described before.33 Endothelial cells were cultured in M-
199 medium supplemented with 15% fetal calf serum, 2 mM
mine, 100 units/ml penicillin, 100 units/ml streptomycin, 0.25
ml amphotericin and 50 g/ml-endothelial-cell growth factor
L
-gluta-
l
g/
l

6170
F. Caruso et al. / Bioorg. Med. Chem. 17 (2009) 6166–6172
Table 3
incubated in 3% defatted milk for 3 h at 37 °C, then cells were incu-
bated with anti ICAM-1 monoclonal antibody overnight at 4 °C.
Next day cells were washed with PBS and incubated with peroxi-
dase conjugated secondary antibody for 3 h at 37 °C. After washing
with PBS, the cells were exposed to the peroxidase substrate (O-
phenylenediamine dihydrochloride 40 mg/100 ml in citrate buffer
pH 4.5). The color development reaction was stopped by 2 N sulfu-
ric acid and absorbance was read at 490 nm by an automated
microplate reader (Spectromax 190, Molecular Devices, USA).
Binding energy (kcal/mol) for guests in the pocket of ICAM-1 active site (line 2), and
Delta energy for isolated ligands in water (line 3)
(+)EGCG
NH diketo (I)
Keto-enol (II)
Keto-enol (III)
CH diketo (IV)
À77.5
À4.6
À33.2
0.4
0.0
À54.7
4.8
1.0
9.3
supplemented with heparin (5 U/ml) in gelatin coated tissue culture
flasks. For subculture, the cells were dislodged using 0.125% trypsin
containing 0.01 M EDTA solution in puck saline and HEPES buffer.
Each donor yields about 9–10 million cells after three passages.
The cells thus obtained were used as primary cells in our
experiments.
3.4. Chemistry
All solvents were dried, degassed, and distilled prior to use. Ele-
mental analyses (C, H, N) were performed with a Fisons Instru-
ments 1108 CHNS-O Elemental analyzer. IR spectra were
recorded from 4000 to 100 cmÀ1 with a Perkin–Elmer System
2000 FTIR instrument. 1H NMR spectra, referenced to Si(CH3)4,
were recorded on a VXR-300 Varian spectrometer (300 MHz for
1H). Relative intensity of signals is given in square brackets and J
in hertz.
3.2. Cell cytotoxicity assay
The cytotoxic effect of tested compounds was determined by
using trypan blue exclusion test as described8 and was further con-
firmed by colorimetric methyl thiazol tetrazolium (MTT) assay as
described.11 Briefly, endothelial cells were treated with vehicle,
DMSO alone (0.25%) or with different concentrations of tested for
24 h. Four hours before the end of incubation, medium was re-
3.4.1. 4-Furancarbonyl-3-methyl-1-phenylpyrazol-5-one, HQFur
(L-1)
moved and 100
After 4 h incubation MTT was removed, cells were washed out with
PBS, and 50 l DMSO was added to each well to dissolve water
insoluble MTT-formazan crystals. Absorbance was recorded at
570 nm in an ELISA reader (BIO RAD, Model 680, USA).
ll of MTT (5 mg/ml in PBS) was added to each well.
To a hot dioxane solution of 3-methyl-1-phenylpyrazole-5-one
(15 g, 0.088 mol) Ca(OH)2 (12 g, 0.162 mol) was added and the
resulting mixture refluxed for 30 min. Then 2-furoylchloride
(12.61 g, 0.086 mol) was added dropwise to the suspension and
the reaction mixture refluxed for 24 h. The yellow precipitate
formed was treated with 350 ml of 2 N HCl and then filtered off
and re-crystallized from MeOH. Mp 103–105 °C. Anal. Calcd for
C15H12N2O3: C, 67.16; H, 4.51; N, 10.44. Found: C, 67.19; H, 4.57;
l
3.3. Cell-ELISA for measurement of ICAM-1
N, 10.40. IR (Nujol, cmÀ1): 2700br
. . . . . .
m
(OÁ Á ÁH), 1625m, 1585s; 1531s
The cell surface expression of ICAM-1 on endothelial monolayer
was quantified using cell-ELISA.33 Briefly, human umbilical vein
endothelial cells were grown to confluency in 96 well flat bottom
ELISA plate that was already gelatinized. The cells were incubated
with or without b-diketones (L-1 to L-6) or metal diketonates (C-1
to C-6), solubilized in DMSO for desired time, followed by induc-
m
(C O, C C), 690s; 604s; 585s, 509s; 391m; 355s; 282s. 1H
•
•
NMR (CDCl3, 300 MHz): d 2.59 (s, 3H, 3-CH3QFur), 6.62 (t, 1H,
C4H3O), 7.25 (t, 1H, C4H3O), 7.42 (m, 3H, C6H5), 7.69 (d, 1H,
C4H3O), 7.85 (d, 2H, C6H5).
3.4.2. 4-Benzoyl-3-methyl-1-phenylpyrazol-5-one, HQPh (L-2)
It was synthesized as described for HQFur (L-1). Anal. Calcd for
C17H14N2O2: C, 73.22; H, 5.27; N, 9.98. Found: C, 73.16; H, 5.16;
tion with TNF-a. The incubation was continued for 16 h for
ICAM-1. The cells were washed with phosphate buffer saline
(PBS, pH 7.4) and incubated with 1% gluteraldehyde for 30 min at
4 °C. To inhibit the non-specific binding of antibody, cells were
N, 10.12. Mp 94–96 °C. IR (Nujol, cmÀ1): 2700br
m(OÁ Á ÁH), 1599s,
Figure 7. Best fit structures of (+)EGCG (left) and keto-enol (II) tautomer (right) in the active site (Leu42-Arg49) of ICAM-1.

F. Caruso et al. / Bioorg. Med. Chem. 17 (2009) 6166–6172
6171
. . .
. . .
1570s, 1560s, 1554s, 1536sh, s
m
(C O, C C), 602m, 533m, 507m,
(10 ml) and dried under reduced pressure at 50 °C. Re-crystalliza-
tion from hot ethanol gave colorless crystals on cooling. Mp 329–
331 °C, yield 84%. Anal. Calcd for C36H50CaN4O6: C, 64.07; H,
7.47; N, 8.30. Found: C, 63.82; H, 7.55; N, 8.46. IR (Nujol, cmÀ1):
•
•
412w, 400w, 393w, 361w, 328w, 296w, 281w. 1H NMR (CDCl3,
300 MHz): d 2.11 (s, 3H, 3-CH3QPh), 7.3, 7.5, 7.9 (m, 14H, C6H5),
10.5 (s br, 1H, OH). 13C NMR (CDCl3): d 15.8 (s, CH3), 120.8,
126.7, 128.3, 129.1, 131.9, 137.3, 137.6 (s, Carom of C6H5), 103.6
(s, C3), 148.0 (s, C4), 162.5 (s, C5), 192.0 (s, CO).B.
3500–2400br,
m
(OHÁ Á ÁN) 1644s, d(OH); 1591vs, n(C]] O); 440vs,
409m, msym (Ca–O); 377m, 253vs (br), 233m,
m
asym(Ca–O). 1H
NMR (CDCl3): d 2.39 (s, 6H, CH3), 2.43 (s, 4H, CH2), 0.90 (s, 18H,
CH3); 6.67 (d) 7.05 (t), 7.24 (t), (10H, aromatics); 1.71 (t, 6 H, CH3E-
tOH), 3.64 (q, 4 H, CH2EtOH).
3.4.3. 4-t-Butancarbonyl-3-methyl-1-phenylpyrazol-5-one,
HQtBu (L-3)
It was synthesized as described for compound L-1. Mp 88–
89 °C. Anal. Calcd for C15H18N2O2: C, 69.80; H, 7.02; N, 10.81.
Found: C, 69.54; H, 7.43; N, 10.65. IR (Nujol, cmÀ1): 3300br–
3.4.8. bis(4-tert-Butylacetyl-3-methyl-1-phenylpyrazol-5-
onato)bis(aquo)zinc(II), [Zn(QnPe)2(H2O)2], (C-2)
. . .
. . .
2800br
m
(OÁ Á ÁH), 1644s, 1604s, 1555s
m
(C O, C C), 610sw,
It was synthesized reacting Zn(O2CCH3)2Á2H2O (0.219 g,
1.0 mmol) with HQnPe (0.545 g, 2.0 mmol) in 30 ml of MeOH at rt.
From the solution a pale yellow solid slowly precipitated. The sus-
pension was stirred 4 h and filtered. The precipitate was washed
with Et2O and dried to constant weight under reduced pressure.
Recrystallized from CHCl3/n-hexane. Mp 185–189 °C, yield 79%.
Anal. Calcd for C32H40N4O5Zn: C, 61.39; H, 6.44; N, 8.95. Found: C,
•
•
596m, 504m, 444w, 395w, 333m, 311w. 1H NMR (DMSO-d6,
300 MHz): d 1.40 (s, 9H, CH3tBu) 2.50 (s, 3H, 3-CH3), 7.3, 7.5, 7.9
(m, 14H, C6H5). 13.0 (s br, 1H, OH).
3.4.4. 4-(3-Cyclopentylpropanoyl)-3-methyl-1-phenylpyrazol-
5-one, HQETCP (L-4)
It was synthesized as described for compound L-1. Mp 77–78 °C.
Anal. Calcd for C18H22O2N2: C, 72.48; H, 7.38; N, 9.40. Found: C,
61.50; H, 6.60; N, 9.21. Km (in acetone): 5.0 X
À1 cm2 molÀ1. IR (Nu-
jol, cmÀ1): 3000–3200br (O–HÁ Á ÁN), 1608s (C@O), 466sbr, 350m
(Zn–O). 1H NMR (CDCl3): d, 1.10 (s) (18 H, (CH3)3CCH2C(@O)), 2.15
(br) (2H, H2O), 2.50 (s) (6H, 3-CH3), 2.65 (s) (18H, (CH3)3CCH2C(@O)),
7.30 (m), 7.92 (m) (10H, C6H5).
71.97; H, 7.23; N, 9.69. IR (Nujol, cmÀ1): 3000–2800
m
(OÁ Á ÁH),
1
. . .
m(C O, C C), 753, 691, 632, 503. H NMR
•
. . .
1631, 1594, 1553, 1498
•
(CDCl3, 295 K, 300 MHz): d 1.2–2.0 (m, 9H, C5H9), 2.51 (s, 3H, 3-
CH3pz), 2.75 (m, 2H, CH2), 7.27–7.85(m, 5H, C6H5), 12.2 (s br, 1H, OH).
3.4.9. (4-Benzoyl-3-methyl-1-phenylpyrazol-5-onato)sodium,
[NaQPh], (C-3)
3.4.5. 4-tert-Butylacetyl-3-methyl-1-phenylpyrazol-5-one,
HQnPe (L-5)
It was synthesized reacting Na(OMe) (0.108 g, 2.0 mmol) with
HQPh (0.545 g, 2.0 mmol) in 30 ml of MeOH at rt. The clear solution
was stirred overnight, then evaporated and the residue washed
with Et2O and dried to constant weight under reduced pressure.
Mp 185–189 °C, yield 90%. Anal. Calcd for C17H13N2NaO2: C,
68.00; H, 4.36; N, 9.33. Found: C, 67.45; H, 4.60; N, 9.12. IR (Nujol,
3-Methyl-1-phenylpyrazol-5-one (15.0 g, 0.088 mol) and dry
1,4-dioxane (80 ml) were placed in a flask equipped with a stir-
rer, separating funnel and a reflux condenser. To this warmed
clear solution calcium hydroxide (12.0 g, 0.162 mol) was added.
Afterwards, t-butylacetyl chloride (11.9 g, 0.086 mol) was drop-
wise added, in about 10 min. The mixture was refluxed for 4 h
and poured into 2 mol dmÀ3 HCl (300 ml) to decompose the cal-
cium complex. The light brown immediately formed precipitate
was filtrated and dried under reduced pressure at 50 °C. Re-crys-
tallization was performed by treating the solid with hot metha-
nol and slow cooling of the solution afforded a yellow crystalline
powder. Mp 85–87 °C, yield 84%, Anal. Calcd for C8H10NO: C,
70.60; H, 7.40; N, 10.32. Found: C, 70.50; H, 7.50; N, 10.30. IR
cmÀ1): 1620sh, 1605s, 1592s
m(C O, C C), 602m, 540m, 504m,
• •
. . .
. . .
404w, 329w. 1H NMR (CDCl3, 300 MHz): d 1.85 (s, 3H, 3-CH3QPh),
7.2, 7.4, 7.7 (m, 14H, C6H5).
3.4.10. Sodium(benzoylacetonato), [Na(bzac)]ÁH2O, (C-4)
It was synthesized reacting Na(OMe) (0.108 g, 2.0 mmol) with
benzoylacetone, bzacH (0.160 g, 1.0 mmol) in 30 ml of MeOH at
rt. The clear solution was stirred overnight, then evaporated and
the residue washed with Et2O and dried to constant weight under
reduced pressure. Mp 100–104 °C, yield 90%. Anal. Calcd for
C10H10O3Na: C, 59.41; H, 5.48. Found: C, 59.23; H, 5.24. IR (Nujol,
(Nujol, cmÀ1): 1642vs,
m(CO); 579s, 509vs, 448w, 419w, 396w,
370m, 356m, 333m, 305w, 270m, 243w and 224w. 1H NMR
(CDCl3): d 2.48 (s, 3 H, CH3QnPe); 2.63s (s, 2H, CH2), 1.12 (s, 9H,
CH3); 7.25 (t), 7.46 (t), 7.90 (d), (5H, aromatics), 11.5 (br, 1H,
cmÀ1): 1605s, 1554s, 1503s,
m(C O, C C), 588m, 547m, 497w,
• •
. . .
. . .
. . .
OH O). UV/VIS (CDCl3):246 (sh) (11740) and 268 nm
428w, 398w, 301w. 1H NMR (CD3OD, 300 MHz): d 2.28 (s, 3H,
•
(15,380 dm3 molÀ1 cmÀ1).
CH3), 7.4m (4H, C6H5 + CHbzac) 7.8 (2H, C6H5bzac).
3.4.6. 4-Benzoyl-3-methyl-1-phenylpyrazol-5-one, HQPh,Ph,Me
(L-6)
It was synthesized as described for compound L-1. Mp 145–
147 °C. yield: 72%. MS; m/z: 278 [M+]. Anal. Calcd for
C17H14N2O2: C, 73.37; H, 5.07; N, 10.07. Found: C, 73.16; H, 5.16;
3.4.11. Sodium(benzoyltrifluoroacetonato), [Na(bzfc)], (C-5)
It was synthesized reacting Na(OMe) (0.108 g, 2.0 mmol) with
benzoyltrifluoroacetone, bzfcH (0.216 g, 1.0 mmol) in 30 ml of
MeOH at rt. The clear solution was stirred overnight, then evapo-
rated and the residue washed with Et2O and dried to constant
weight under reduced pressure. Mp 256–261 °C, yield 80%. Anal.
Calcd for C10H6F3O2Na: C, 50.44; H, 2.54. Found: C, 50.23; H,
N, 10.12. IR (Nujol, cmÀ1):
m
(OHÁ Á ÁO) 2800br; 1622vs (br). 1H
NMR (CDCl3): d 2.18 (s, 3H, CH3), 3.20 (s br, 1H, NHÁ Á ÁO), 7.25m,
7.33d, 7.60m, 7.92d (10H, C6H5). 13C NMR (CDCl3): d 26.8 (s,
CH3(C@O)), 120.9, 126.8, 128.5, 128.7, 129.1, 129.3, 129.4, 132.9,
137.3 (s, C of C6H5), 104.0 (s, C3), 151.3 (s, C4), 160.5 (s, C5),
195.3 (s, CO).
2.76. IR (Nujol, cmÀ1): 1605s, 1554s, 1503s,
m(C O, C C), 633s,
• •
. . .
. . .
575s, 513w, 492m, 441w, 367s br, 305w. 1H NMR (CD3OD,
300 MHz): 5.99 (s, 1H, CHbzfc) 7.36m, 7.79m (5H, C6H5bzfc).
3.4.12. bis(4-Pivaloyl-3-methyl-1-phenylpyrazol-5-onato)bis
(aqua)zinc(II), [Zn(QtBu)2(H2O)2], (C-6)
3.4.7. bis(4-t-Butylacetyl-3-methyl-1-phenylpyrazol-
5onato)bis(ethanol)calcium(II), [Ca(QnPe)2(EtOH)2] (C-1)
An ethanolic solution (30 ml) of HQnPe (2 mmol) and KOH
(2 mmol) were added to an aqueous solution (10 ml) of calcium
dichloride (1 mmol). In a few minutes a white precipitate formed.
After 1 h stirring the precipitate was filtered, washed with water
It was synthesized reacting Zn(O2CCH3)2Á2H2O (0.219 g,
1.0 mmol) with HQtBu (0.530 g, 2.0 mmol) in 30 ml of MeOH at rt.
From the solution a pale yellow solid slowly precipitated. After
stirring 4 h the suspension was filtered. The precipitate was
washed with Et2O and dried to constant weight under reduced

6172
F. Caruso et al. / Bioorg. Med. Chem. 17 (2009) 6166–6172
pressure. Re-crystallised from CHCl3/n-hexane. Mp 176–180 °C,
yield 79%. Anal. Calcd for C30H38N4O6Zn: C, 58.49; H, 6.22; N,
9.09. Found: C, 58.55; H, 6.30; N, 9.13. IR (Nujol, cmÀ1): 3640s
(O–HÁ Á ÁN), 1608s, 1593s, 1583s, 1502s (C@O), 519w, 503m,
489m, 420s, 411w, 386w, 363w, 324w, 299w (Zn–O). 1H NMR
(CDCl3): d, 1.25 (s) (18 H, (CH3)3CC(@O)), 2.60 (br, 6 H, 3-CH3),
7.20, 7.70 (m, 10H, C6H5).
References and notes
1. Osborn, L. Cell 1990, 62, 3.
2. Springer, T. A. Cell 1994, 76, 301.
3. Collins, T.; Read, M. A.; Neish, A. S.; Whitley, M. Z.; Thanos, D.; Maniatis, T.
FASEB J. 1995, 9, 899.
4. Gorski, A. Immunol. Today 1994, 15, 251.
5. Mantovani, A.; Bussolino, F.; Introna, M. Immunol. Today 1997, 18, 231.
6. Butcher, C. E. Cell 1991, 67, 1033.
7. Weiser, M. R.; Gibbs, S. A. L.; Hechtman, H. B. In Adhesion Molecules in Health
and Disease; Paul, L. C., Issekutz, T. B., Eds.; Marcel Dekker: New York, 1997; p
55.
3.5. Theoretical calculations
8. Madan, B.; Batra, S.; Ghosh, B. Mol. Pharmacol. 2000, 58, 526.
9. Kumar, S.; Singhal, V.; Roshan, R.; Rambotkar, G. W.; Ghosh, B. E. J. Pharmacol.
2007, 73, 1602.
10. Kumar, S.; Singh, B. K.; Kalra, N.; Kumar, V.; Kumar, A.; Raj, H. G.; Prasad, A. K.;
Parmar, V. S.; Ghosh, B. Bioorg. Med. Chem. 2005, 13, 1605.
11. Kumar, S.; Arya, P.; Mukherjee, C.; Singh, B. K.; Singh, N.; Parmar, V. S.; Prasad,
A. K.; Ghosh, B. Biochemistry 2005, 44, 15944.
12. Madan, B.; Ghosh, B. Shock 2003, 19, 91.
13. Singh, N.; Kumar, S.; Singh, P.; Raj, H. G.; Prasad, A. K.; Parmar, V. S.; Ghosh, B.
Phytomedicine 2008, 4, 284.
14. Shibata, S. Yakugaku Zasshi 2000, 120, 849.
15. Ram, A.; Mahabalirajan, U.; Das, M.; Bhattacharya, I.; Dinda, A. K.; Gangal, S. V.;
Ghosh, B. Int. Immunopharmacol. 2006, 9, 1468.
16. Kumar, S.; Sharma, A.; Madan, B.; Singhal, V.; Ghosh, B. Biochem. Pharmacol.
2007, 73, 1602.
17. Kim, Y. S.; Ahn, Y.; Hong, M. H.; Joo, S. Y.; Kim, K. H.; Sohn, I. S.; Park, H. W.;
Hong, Y. J.; Kim, J. H.; Kim, W.; Jeong, M. H.; Cho, J. G.; Park, J. C.; Kang, J. C. J.
Cardiovasc. Pharmacol. 2007, 50, 41.
The structural features of four HQPh tautomers (Scheme 3) were
analyzed with density functional theory using the Accelrys34 pack-
35–37
age Materials Studio 4.1, code DMOL3
.
The effect of water sol-
vent was included using a COSMO algorithm.38 We used the
general gradient approximation (GGA) and the Becke exchange
(BP) functional.39 The highest available DMOL3 level of theory was
used including a double numeric basis set35 with polarization func-
tions (DNP)39,40 for an all electron calculation. Initial coordinates
were those obtained from crystal structures, N–H diketo27 and
keto-enol28 for tautomers I and II, respectively. Forms III and IV
were calculated after modification of calculated structures I and
II using DMOL3 subroutines. Converged energies are shown in Table
2. Docking studies for all 4 tautomers and the reference compound
(+)EGCG were performed with the CDOCKER package in Discovery
Studio 2.1 from Accelrys,41 the active site radius was of 12 Å in the
conformer simulation for each ligand.
18. Lin, C.-C.; Lee, C.-W.; Chu, T.-H.; Cheng, C.-Y.; Luo, S.-F.; Hsiao, L.-D.; Yang, C.-
M. J. Cell. Physiol. 2007, 211, 771.
19. Kunnumakkara, A. B.; Guha, S.; Krishnan, S.; Diagaradjane, P.; Gelovani, J.;
Aggarwal, B. B. Cancer Res. 2007, 67, 3853.
20. Mitra, A.; Chakrabarti, J.; Banerji, A.; Chatterjee, A.; Das, B. R. J. Environ. Pathol.
Toxicol. Oncol. 2006, 25, 679.
21. Punithavathi, D.; Babu, M.; Arumugam, V.; Venkatesan, N. Lett. Drug Des.
Discovery 2006, 3, 714.
3.6. Statistical analysis
Results are given as mean SD. Independent two-tailed Stu-
dent’s t-test was performed. Differences were considered statisti-
cally significant for P <0.05. All statistical analyses were performed
using software Microcal Origin (ver 3.0; Microcal Software Inc.,
Northampton, MA).
22. Caruso, F.; Rossi, M.; Tanski, J.; Sartori, R.; Sariego, R.; Moya, S.; Diez, S.;
Navarrete, E.; Cingolani, A.; Marchetti, F.; Pettinari, C. J. Med. Chem. 2000, 43,
3665.
23. Pettinari, C.; Caruso, F.; Zaffaroni, N.; Villa, R.; Marchetti, F.; Pettinari, R.;
Phillips, C.; Tanski, J.; Rossi, M. J. Inorg. Biochem. 2006, 100, 58.
24. Bischoff, H.; Berger, M. R.; Keppler, B. K.; Schmaehl, D. J. Cancer Res. Clin. Oncol.
1987, 113, 446.
25. Keppler, B. K.; Friesen, C.; Vongerichten, H.; Vogel, E. In Metal Complexes in Cancer
Chemotherapy; Keppler, B. K; Ed.; VCH: Weinheim, Germany, 1993. pp 297–323.
26. Kumar, A.; Dhawan, S.; Hardegen, N. J.; Aggarwal, B. B. Biochem. Pharmacol.
1998, 55, 775.
27. O’Connell, M. J.; Ramsay, C. G.; Steel, P. J. Aust. J. Chem. 1985, 38, 401.
28. Akama, Y.; Shiro, M.; Ueda, T.; Kajitani, M. Acta Crystallogr., Sect. C 1995, 51,
1310.
29. Pettinari, C.; Marchetti, F.; Cingolani, A.; Lorenzotti, A.; Mundorff, E.; Rossi, M.;
Caruso, F. Inorg. Chim. Acta 1997, 262, 33.
30. Tse, M. T.; Chakrabarti, K.; Gray, C.; Chitnis, C. E.; Craig, A. Mol. Microbiol. 2004,
51, 1039.
31. Dormeyer, M.; Adams, Y.; Kramer, B.; Chakravorty, S.; Tse, M. T.; Pegoraro, S.;
Whittaker, L.; Lanzer, M.; Craig, A. Antimicrob. Agents Chemother. 2006, 50, 724.
32. Bella, J.; Kolatkar, P. R.; Marlor, C. W.; Greve, J. M.; Rossmann, M. G. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 4140.
33. Kumar, S.; Singh, B. K.; Pandey, A. K.; Kumar, A.; Sharma, S. K.; Raj, H. G.;
Prasad, A. K.; Van der Eychen, E.; Parmar, V. S.; Ghosh, B. Bioorg. Med. Chem.
2007, 15, 1952.
4. Conclusion
In this study the ICAM-1 inhibitory activity of curcumin models
lacking aromatic peripheral hydroxyl and methoxy groups (b-dike-
tones), and their metal derivatives are investigated experimentally
and theoretically. b-Diketones are systematically more active at
lower concentrations than their metal complexes; the highest inhi-
bition for both groups is 75%. Among the ligands, the best inhibitor
is 4-benzoyl-3-methyl-1-phenyl-pyrazol-5-one (HQPh). For this
reason, the four tautomeric forms of HQPh were investigated using
density functional theory. In addition, docking of all HQPh tautom-
ers on ICAM-1 protein was performed and suggested that keto-enol
(II), depicted in Figures 4 and 7, was favored to act in biological
environment.
34. Accelrys Inc. Materials Studio Release 4.1. 10188 Telesis Court, Suite 100, San
Diego, CA 92121, USA.
35. Delley, B. J. Chem. Phys. 1990, 92, 508.
36. Delley, B. J. Chem. Phys. 2000, 113, 7756.
Acknowledgements
37. Delley, B. J. Phys. Chem. 1996, 100, 6107.
38. Klamt, A.; Schrmann, G. J. Chem. Soc., Perkin Trans. 1993, 2, 799.
39. Becke, A. D. Phys. Rev. 1988, A38, 3098.
40. Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh,
D. J.; Fiolhais, C. Phys. Rev. B: Condens. Matter Mater. Phys. 1992, 46, 6671.
41. Wu, G.; Robertson, D. H.; Brooks, C. L., III; Vieth, M. J. Comput. Chem. 2003, 24,
1549.
This work is supported by Council of Scientific and Industrial
Research (CSIR, India), Task Force Project NWP0033, and by CSIR-
CNR (Indio-Italian) bilateral agreement. S.K. and S.B. thank CSIR
for their Fellowships. F.C. thanks Vassar College for facilities pro-
vided in using DMOL3 program.