Development and biological evaluation of C60 fulleropyrrolidine-thalidomide dyad as a new anti-inflammation agent

Article abstract of DOI:10.1016/j.bmc.2008.08.004

Research studies in the field of C₆₀ fullerene derivatives have significantly increased due to the broad range of biological activities that were found for these compounds. We designed and prepared a new C₆₀ fullerene hybrid bearing thalidomide as a potential double-action anti-inflammatory agent, capable of simultaneous inhibition of LPS-induced NO and TNF-α production. The C₆₀ fulleropyrrolidine-thalidomide dyad, CLT, was an effective agent to suppress the release of NO and TNF-α by the LPS-stimulated macrophages RAW 264.7. Ten micromolars of CLT effectively inhibited LPS-induced NO and TNF-α production by 47.3 ± 4.2% and 70.2 ± 4% with respected to the control, respectively. Furthermore, preliminary biochemical investigation revealed that CLT was a potent agent to suppress both LPS-induced intracellular ROS production and iNOS expression, and CLT also inhibited the phosphorylation of ERK which is an important protein kinase involved in the activation of TNF-α synthesis in LPS-activated macrophages. We believed that the studies herein would hold promise for future development of a new generation of potent anti-inflammatory agents.

Full text of DOI:10.1016/j.bmc.2008.08.004

Bioorganic & Medicinal Chemistry 16 (2008) 8619–8626
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Development and biological evaluation of C60 fulleropyrrolidine-thalidomide
dyad as a new anti-inflammation agent
b
c
b
b
Sheng-Tung Huang a, , Chia-Shin Ho , Chun-Mao Lin , Hsu-Wei Fang , Yi-Xiang Peng
*
a Graduate Institute of Biotechnology, National Taipei University of Technology, 1, Sec. 3, Chung-Hsiao E. Road, Taipei 106, Taiwan
b Graduate Institute of Chemical Engineering, National Taipei University of Technology, Taipei, Taiwan
c College of Medicine, Taipei Medical University, Taipei, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
Research studies in the field of C60 fullerene derivatives have significantly increased due to the broad
range of biological activities that were found for these compounds. We designed and prepared a new
C60 fullerene hybrid bearing thalidomide as a potential double-action anti-inflammatory agent, capable
Received 25 June 2008
Revised 1 August 2008
Accepted 2 August 2008
Available online 7 August 2008
of simultaneous inhibition of LPS-induced NO and TNF-
mide dyad, CLT, was an effective agent to suppress the release of NO and TNF-
macrophages RAW 264.7. Ten micromolars of CLT effectively inhibited LPS-induced NO and TNF-
a
production. The C60 fulleropyrrolidine-thalido-
by the LPS-stimulated
pro-
a
a
Keywords:
Anti-inflammation
C60 fulleropyrrolidine
duction by 47.3 4.2% and 70.2 4% with respected to the control, respectively. Furthermore, prelimin-
ary biochemical investigation revealed that CLT was a potent agent to suppress both LPS-induced
intracellular ROS production and iNOS expression, and CLT also inhibited the phosphorylation of ERK
which is an important protein kinase involved in the activation of TNF-a synthesis in LPS-activated mac-
rophages. We believed that the studies herein would hold promise for future development of a new gen-
TNF-a
Nitric oxide
Thalidomide
eration of potent anti-inflammatory agents.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
ties that were found for these compounds.3 Recently, the prepara-
tion of novel water soluble fullerene hybrids bearing a variety of
Inflammation is a complex cellular event, in which parts of the
event involve the secretion of various cytokines by immune cells.
The two pro-inflammatory cytokines, nitric oxide (NO) and tumor
functional moieties such as peptides,4 oligonucleotides,5 porphy-
rins,6 and flavonoids7 have attracted much attention. Such dyad
systems could amplify or alter the biochemical characteristics of
their components or even produce compounds with new biological
properties. In addition, the peculiar structure of C60 fullerene
which is capable of ‘adding’ multiple radicals per molecule can
serve as a ‘radical sponge’.8 The exceptional radical scavenging
property of fullerene makes it a promising pharmacophore to tar-
get diseases caused by the overexpression of radicals.
necrosis factor
a (TNF-a), have been recognized as essential compo-
nents in acute and chronic inflammatory processes.1 NO is associ-
ated with dilating the arteries and increasing the blood flow
during inflammation. The pro-inflammatory enzyme, inducible
forms of nitric oxide synthase (iNOS), is responsible for increasing
the levels of NO, and it is highly expressed in lipopolysaccharide
(LPS)-activated macrophages.2 The polypeptide cytokine, TNF-
a
, is
produced by cells of the monocyte/macrophage lineage. The process
of TNF- production starts with the binding of microbial products to
Current therapeutic approaches to the treatment of inflamma-
tory diseases are centered on the suppression of the NO and TNF-
a
a
production.9 Oxidative stress by oxygen radicals is known to in-
cell surface Toll receptors, which results in initiating a cascade of sig-
duce cellular instability by cascade events leading to inflammation,
and it has been reported that suppression of ROS-mediated NO ele-
vation is beneficial in reducing the development of inflammation.10
nal transduction events which lead to activation of its synthesis.1b
Both NO and TNF-a serve to recruit other inflammatory cells, which
in turn release more cytokines and subsequently amplify the im-
mune response. The cytokine responses to cell injury are regulated
processes and are usually beneficial to the hosts, but overexpression
of these cytokines can cause serious diseases including rheumatoid
arthritis, multiple sclerosis, asthma, and psoriasis.
Inhibiting the TNF-a production through suppressing the expres-
sion of inflammation-activated protein kinases had always been a
hopeful target for the rational development of anti-inflammatory
drugs.1b As part of the program aimed at discovering new anti-
inflammation agents,11 we had designed and prepared several
Research studies in the field of C60 fullerene derivatives have
significantly increased due to the broad range of biological activi-
C60 fullerene-anti-TNF-
a dyads which simultaneously inhibited
the NO and TNF- production during inflammation. Several studies
a
have revealed that C60 fullerene derivatives exhibit potent inhibi-
tory effects on NO-dependent relaxation, and the inhibitory effects
are believed to be associated with C60 fullerene radical scavenging
* Corresponding author. Tel.: +886 02 2771 2171; fax: +886 02 2731 7117.
E-mail address: Ws75624@ntut.edu.tw (S.-T. Huang).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.08.004

8620
S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
and direct NO-quenching abilities.8,12 We expected that the excep-
tional radical scavenging activity and direct NO-quenching abilities
of C60 fullerene would be beneficial to suppress the ROS-mediated
NO elevation during inflammation. However, our previous results
had shown that water soluble C60 derivatives alone lacked the effi-
ERK activation by LPS inhibited the induction of TNF-
further investigation revealed that the inhibition of ERK phosphor-
a release, and
ylation prevented the transport of TNF-a mRNA from the nucleus
to the cytoplasm during inflammation.14 The influence of CLT on
modulating the phosphorylation of ERK had also been studied in
LPS-activated macrophages by Western blot.
cacy to suppress the TNF-
phage.11 By covalently linking a known anti-TNF-
agent to the C60 fullerene moiety, the new C60 fulleropyrrolidine-
anti-TNF- agent dyad expressed potent synergistic anti-inflam-
a
production in the LPS-activated macro-
a
production
2. Results and discussion
a
matory effects that inhibited both the NO and TNF-
a production
in LPS-activated macrophages at low micro molar concentration.11
Based on our previous results, we explored the modularity of
our previous design, probing its applicability to incorporate an-
2.1. CLT synthesis and aqueous solubility
The synthetic scheme of CLT is outlined in Scheme 2. The syn-
thesis began with the preparation of N-alkyated glycine 3. Coupling
the known acid 115 with an amine salt 316 under standard peptide
coupling conditions gave the corresponding amide. Removing two
benzyl protecting groups in the corresponding amide by a palla-
dium-catalyzed hydrogenation reaction furnished the N-alkyated
glycine 3 in 2 steps in 64% yield. The fulleropyrrolidine moiety in
CLT was constructed by 1–3 dipolar cycloaddition reactions. Con-
densing the N-alkyated glycine 3 with known di-methoxyethoxy-
other anti-TNF-
lerene hybrid bearing thalidomide (CLT) as
a
pharmacophore. We synthesized a new C60 ful-
new anti-
a
inflammation agent (Scheme 1). Thalidomide is a hypnotic/seda-
tive drug that was withdrawn from the market because of its se-
vere teratogenicity.13 In spite of this, research into thalidomide
was not halted, and thalidomide has been established as a potent
immunomodulator with biological effects on both cytokine stimu-
lation and cell-mediated immunity.13 The basic chemical architec-
ture of CLT is composed of thalidomide linked to the nitrogen of a
C60 fulleropyrrolidine moiety through ethylene glycol chains, and
an additional polar side chain was strategically positioned on the
fullerene spheroid of the dyad to provide the extra hydration sites
in order to enhance its solubility in aqueous media (Scheme 1). In
this paper, we describe the synthesis of CLT, and we evaluate its
anti-inflammatory effects by determining the amount of LPS-in-
methyl-protected-ketone
417
formed
the
corresponding
azomethine ylide (not shown) which underwent 1–3 dipolar cyclo-
addition reactions with C60 fullerene to yield the desired fullero-
pyrrolidine-thalidomide dyad, CLT, and the yield was 25%. CLT
was soluble in 1% DMSO aqueous solution with a maximum con-
centration of 1.16 Â 10À5 M. The UV–vis spectra of CLT in 1% DMSO
aqueous solutions showed that the main absorption of the fuller-
ene core was still evident, although in analogy to other fullerene
derivatives dissolved in polar solvents, the bands are broader and
less structured, especially in the low-energy region (Supplemen-
tary materials). Amphiphilic fullerene monomers are known to
form a colloidal in aqueous solutions;18 we also examined the size
distribution of CLT in the 1% DMSO aqueous solution. The intensity
size distribution of CLT was measured by a dynamic light scatter-
ing (DLS) particle size analyzer ( Supplementary materials). The
duced NO and TNF-
a released by the macrophages RAW 264.7.
We also investigated the mechanism of CLT in suppressing NO pro-
duction by examining its abilities to modulate the intracellular ROS
accumulation and the expression of iNOS in LPS-activated macro-
phages. Furthermore, mitogen-activated protein kinase (MAPK),
an extracellular signal-regulated kinase (ERK), is an important
inflammation-activated protein kinase which is involved in the
cascade of TNF-a synthesis. Dmuitru et al. revealed that abolishing
average hydrodynamic diameter for 10 lM of CLT was 78.9 nm,
and this value is similar to other known water soluble fullerenes
measured in aqueous media.18 Recently, there have been several
papers describing the preparation of novel fulleropyrrolidine hy-
brids bearing promising biological components, but their best sol-
ubility in a 10% DMSO aqueous solution was reported to be in the
range of ꢀ10À5 M,7 and this presents a challenge the further eval-
uation in a mammalian cell culture model. The concentration of
DMSO exceeding 1% in cell media caused instability of mammalian
cells during culturing and biological evaluation. The synthesis of
the CLT was completed in 3 steps from the known starting materi-
als, and it is soluble in 1% DMSO aqueous solution in which it is
suitable for evaluation with a mammalian culture model.
2.2. CLT cytotoxicity
We first analyzed the cytotoxic effects of CLT and its thalido-
mide-lacking structural analogues CL (Scheme 1) against a RAW
264.7 cell culture lineage via MTT assay; the cell viability results
are shown in the Supplementary materials. Cytotoxicity studies
after 24 h of continuous exposure to the various concentrations
of CLT and CL with or without LPS stimulation were measured with
a colorimetric assay based on the ability of mitochondria in viable
cells to reduce MTT as described previously.19 The RAW 264.7 cells
pretreated with 10 lM of either CLT or CL in the presence of 1 lg/
ml of LPS for 24 h had viability close to 95% with respect to the con-
trol. One interesting observation was that we observed a slight in-
crease in the cells’ viability when the RAW 264.7 culture was
incubated with 10
lM of CLT without LPS stimulation. CLT and
CL were not cytotoxic to the RAW 264.7 cell culture lineage.
Scheme 1. Chemical structure of thalidomide, CLT, and CL.

S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
8621
Scheme 2. Reagents and conditions: (a) EDCI, HOBT, Et3N, DMF, rt, 16 h, 70%; (b) H2 (60 psi), Pd/C, 18 h, 91%; (c) toluene, reflux, 25%.
2.3. CLT anti-inflammatory evaluation
We evaluated the anti-inflammatory effects of thalidomide, CL,
an equimolar concentration of thalidomide–CL cocktail, and CLT on
the LPS-activated macrophages RAW 264.7 by determining the
amount of NO and TNF-
a released into the cell supernatants. The
RAW 264.7 cells were pretreated with thalidomide, CL, an equimo-
lar concentration of thalidomide–CL cocktail, or CLT at different
concentrations (10, 1, and 0.1
lM) for 3 h, followed by stimulating
the cells with LPS (1 g/mL) for 24 h. RAW 264.7 stimulated with
l
LPS alone was the control. The amount of NO was determined by
measuring the nitrite concentrations in the cell supernatants after
24 h treatment.20 The concentration of TNF-
a was determined by
the ELISA method as described according to the known prior
art.21 The results are shown in Figures 1 and 2. All tested herein
agents showed an inhibitory effect on LPS-induced NO production
in a concentration-dependent manner (Fig. 1). The RAW 264.1 pre-
Figure 2. TNF-
a
production in RAW 264.7 cells pretreated with the indicated
concentration of CL, Thal (thalidomide), CL + Thal (an equimolar thalidomide–CL
cocktail), and CLT for 3 h. The cells were activated with LPS (1 g/ml). Control cells
were incubated with 1% DMSO alone. Culture supernatants were collected after 24-
h activation. Error bars represent the standard error of mean (n = 3). Significant
difference (P < 0.01) compared with the control treated with LPS.
l
treated with 10
exhibited a moderate reduction of NO synthesis, and the concen-
trations of NO produced were 32.3 1.6 and 35.7 3.1 M, respec-
lM of either the CL or the thalidomide–CL cocktail
l
tively, and those values were 30.9 3.5% and 23.7 6.7% reduction
with respect to the control. In contrast, thalidomide and CLT were
potent agents to reduce the LPS-induced NO secreted by the mac-
rophages. The concentrations of NO detected in the cell superna-
phages pretreated with either thalidomide, thalidomide–CL
cocktail or CLT exhibited a significant reduction in the TNF-
thesis. RAW 264.7 pretreated with 10 M of either thalidomide or
thalidomide–CL cocktail exhibited a reduction in TNF- , and the
concentrations of TNF- detected in the cell medium were
a syn-
l
tants were 26.9 2.5 and 25.2 1.9
lM for cells pretreated with
a
10 M of thalidomide and CLT, respectively, and those values were
l
a
44.1 5.8% and 47.3 4.2% reduction with respect to the control.
All agents tested herein had no quenching effect on the Griess re-
agent at the concentrations used. On the other hand, the macro-
9.2 0.8 and 11.9 1.7 ng/ml, respectively, and those values were
55.5 3.6% and 41.9 8% reduction with respect to the control
(Fig. 2). CLT exhibited an impressive effect on suppressing the
LPS-induced TNF-
a
synthesis at either 1 or 10
lM in the pretreat-
ments; the concentrations of TNF-
a
detected in the cells medium
were 11.5 1.5 and 5.3 1.0 ng/ml, respectively, and those values
were 44.4 7.2% and 70.2 4% reduction with respect to the con-
trol. The non-covalently-linked thalidomide CL cocktail is less
effective in reducing the LPS-induced NO and TNF-
a production
compared to the fulleropyrrolidine-thalidomide dyad, CLT. Fur-
thermore, CLT is nearly tenfold more potent than thalidomide
alone in suppressing the LPS-induced TNF-a production by the
macrophages. CLT is the most potent agent herein tested in sup-
pressing NO and TNF-
macrophages.
a
synthesis in the LPS-activated
2.4. The scavenging effect of thalidomide, CL, and CLT to ROS
Intracellular ROS is an important signal mediator in the inflam-
mation process, and its concentration elevates during inflamma-
tion. We evaluated the influence of thalidomide, CL, an equimolar
concentration of thalidomide–CL cocktail, and CLT on modulating
intracellular ROS concentration in the LPS-activated macrophages
RAW 264.7. The formation of intracellular ROS was visualized by
Figure 1. Nitrite production in RAW 264.7 cells pretreated with the indicated
concentration of CL, Thal (thalidomide), CL + Thal (an equimolar thalidomide–CL
cocktail), and CLT for 3 h. The cells were activated with LPS (1 lg/ml). Control cells
were incubated with DMSO alone. Culture supernatants were collected after a 24-h
activation. Error bars represent the standard error of mean (n = 3). Significant
difference (P < 0.01) compared with the control treated with LPS.

8622
S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
confocal microscopy a with latent fluorescent probe, DHR 123,
whose its intense fluorescent rhodamine could be selectively re-
vealed by the ROS oxidation, and the results are shown in Figure
of iNOS in RAW 264.7 cells pretreated with the agents tested here-
in after LPS stimulation, and the Western blot results are shown in
Figure 5. The macrophages stimulated with LPS alone after 24 h
exhibited a large expression of iNOS (Fig. 5, lane 2). The iNOS
expression in the LPS-activated macrophages pretreated with
3.22 The RAW 264.7 cells were pretreated with 10
l
M of either
CLT, CL, thalidomide, or an equimolar concentration of thalido-
mide–CL cocktail followed by stimulating them with LPS (1 g/
l
either 1 or 10 lM of thalidomide or CLT had a minimum or essen-
mL) for 3 h. The RAW 264.7 stimulated with LPS alone was the con-
trol. The RAW 264.7 stimulated with LPS alone exhibited an in-
tense fluorescent signal (Fig. 3b). In contrast, the cells pretreated
with all of the agents herein tested resulted in the diminished
intensity of the fluorescent signal within cells via visualizing this
through a confocal microscope (Fig. 3c–f). Flow cytometric detec-
tion of intracellular ROS was also preformed by examining the oxi-
dation of green DHR 123 fluorescence by ROS in the LPS-stimulated
RAW 264.7 cells (Fig. 4). Once again, the cells pretreated with all of
the agents herein tested showed inhibition of the ROS-associated
green DHR 123 fluorescence, and these results were consistent
with the results of confocal microscopy (Fig. 4c–f). The cells pre-
treated with thalidomide or CLT exhibited the weakest fluorescent
signal compared to other agents herein tested (Fig. 4d and f). Be-
cause radical scavenging by C60 fullerene has been widely demon-
strated in vitro, interestingly it was noted that CL alone also
exhibited a potent effect on reduction of the intensity of the DHR
123 fluorescent signal (Figs. 3c and 4c). The results of the above
experiments indicated that all of the reagents tested herein were
effective in decreasing the accumulation of intracellular ROS con-
centration in the LPS-activated RAW 264.7
tially no expression of iNOS protein (Fig. 5, lanes 5, 6, 9, and 10). On
the other hand, the cells pretreated with an equimolar concentra-
tion of thalidomide–CL cocktail (10
lM) exhibited a moderate
iNOS expression. Interestingly, cells pretreated with CL (10
lM)
also exhibited a minor reduction of the expression of iNOS after
LPS stimulation. CLT and thalidomide were the most potent agents
tested herein on reducing the expression of iNOS.
Suppression of the ROS-mediated NO released by macrophages
through scavenging intracellular ROS has been widely demon-
strated in vitro. The chemical structure of CL contains a ‘radical
sponge’ C60 fullerene moiety which exhibited a potent effect on
reducing the intracellular ROS concentration; thus, it suppressed
the LPS-induced NO production by the macrophages. However,
CL lacked efficacy in suppressing iNOS expression. From this
study, we observed that the C60 fullerene hybrid with an iNOS
expression suppressing agent, thalidomide, dyad (CLT), exhibited
a synergistic effect on suppressing the NO production; macro-
phages pretreated with CLT were more potent in reducing the
NO production compared to CL pretreatment alone. Unfortu-
nately, we did not observe much improvement in the efficacy of
inhibiting NO production compared between cells pretreated with
either CLT or thalidomide alone. The chemical structure of CLT
contains both a ‘radical sponge’ and an iNOS suppressing agent
exhibiting potent synergistic anti-inflammatory effects that inhib-
ited the LPS-induced NO synthesis by the macrophages. CLT inhi-
bition of NO synthesis by the LPS-stimulated macrophages is
attributed to its abilities to scavenge intracellular ROS and sup-
press the iNOS expression.
2.5. The influence of thalidomide, CL, and CLT on iNOS
expression
The pro-inflammatory enzyme, iNOS, is responsible for increas-
ing the levels of NO, and it is highly expressed in lipopolysaccha-
ride (LPS)-activated macrophages.2 We examined the expression
Figure 3. The intracellular reactive oxygen species (ROS) production in RAW 264.7 cells visualized with the oxidized DHR123 green fluorescence through confocal a
microscope. Confocal images of green oxidized DHR123 fluorescence in the RAW 264.7 cells with (a) alone, (b) 1-h LPS (100 ng/ml) stimulation alone, (c) CL (10 M) 3-h
pretreatment, (d) thalidomide (10 M) 3-h pretreatment, (e) an equimolar thalidomide–CL cocktail (10 M) 3-h pretreatment, or (f) CLT (10 M) 3-h pretreatment, followed
by 1-h LPS (100 ng/ml) stimulation and stained with 10 M of DHR123 for 30 min.
l
l
l
l
l

S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
8623
Figure 4. The intracellular reactive oxygen species (ROS) production in RAW 264.7 cells measured by the green fluorescence of oxidized DHR123 through a flow cytometer.
The FL1-H fluorescence intensity, represented by mean values, on a FACScan flow cytometer was evaluated for LPS stimulation (100 ng/ml) for 30 min. Flow cytometer
spectra of green oxidized DHR123 fluorescence in the RAW 264.7 cells with (a) alone, (b) 1-h LPS stimulation alone, (c) CL (10
lM) 3-h pretreatment, (d) thalidomide (10 lM)
3-h pretreatment, (e) an equimolar thalidomide–CL cocktail (10 M) 3-h pretreatment, or (f) CLT (10 M) 3-h pretreatment.
l
l
Figure 5. The expression of iNOS and pERK was assessed by immunoblot. The RAW 264.7 cells pretreated with various concentrations (1 and 10
lM) of CL, Thal
(thalidomide), CL + Thal (an equimolar thalidomide–CL cocktail), and CLT for 3 h followed by LPS (1 g/ml) stimulation, and cell lysate prepared for analysis. Lane 1, shows is
l
the untreated cell control, and lane 2, shows is the cells with LPS stimulation alone. The immunoblots shown have been equilibrated for protein loading. The data shown are
representative of three separate experiments.
2.6. The influence of thalidomide, CL, and CLT on ERK
phosphorylation
Many studies had demonstrated that thalidomide inhibited TNF-
production through suppressing p-38 expression thus destabiliz-
ing the TNF-
mRNA for translation.13 However, thalidomide sup-
a
a
Finally, ERK is an important inflammation-activated protein ki-
pression of pERK expression in the LPS-activated macrophages
RAW264.7 has not been documented. Our preliminary biochemical
mechanism study of both thalidomide and CLT on inhibiting LPS-
nase which is involved in the cascade of TNF-
phosphorylated ERK (pERK) was known to be involved in the trans-
location of the TNF- mRNA from the nucleus to the cytoplasm. We
a synthesis. The
a
induced TNF-a production revealed that they both inhibited the
also looked into the expression of pERK 1 and 2 in RAW 264.7 cell
pretreated with the agents here in tested after LPS stimulation for
30 min, and the Western blot results are also shown in Figure 5.
The macrophages stimulated with LPS alone for 30 min exhibited
a large amount of pERK 1 and 2 expressions (Fig. 5, lane 2). In con-
trast, cells pretreated with either thalidomide or CLT after LPS
stimulation exhibited minimum amount of pERK 1 and 2 expres-
sions compared to LPS-stimulated cells alone (Fig. 5, lanes 5, 6, 9,
TNF- synthesis through suppressing the pERK expression. Fur-
a
thermore, the C60 fullerene moiety alone was unable to modulate
the activation of ERK, and CLT containing the bulky C60 fullerene
moiety did not hamper the effect of thalidomide on suppressing
pERK expression. Further details of molecular pharmacological
studies involving CLT on modulating the activity of other kinases
in the cascade of the TNF-a synthesis are underway.
From this study, we observed different efficacies between CLT
and the thalidomide–CL cocktail treatment on LPS-induced macro-
phages. Although the thalidomide–CL cocktail contains both phar-
macophores as does CLT, the thalidomide–CL cocktail treatment is
and 10). The cells pretreated with 10 lM of CLT exhibited the low-
est expressions of pERK 1 and 2 compared to the other agents here-
in tested (Fig. 5, lane 9). However, cells pretreated with an
equimolar concentration of thalidomide–CL cocktail (10
l
M) and
far less effective in suppressing LPS-induced NO and TNF-
sis, iNOS and pERK expression compared to CLT treatment alone.
LPS-stimulatedmacrophagespretreated withCLT(10 M) exhibited
47.3 4.2% and 70.2 4% reduction of NO and TNF- production
with respect to the control. However LPS-stimulated macrophages
pretreated with thalidomide–CL cocktail (10 M) only exhibited
23.7 6.7% and 41.9 8% reduction of NO and TNF- production
a synthe-
CL alone (10 M) exhibited only moderate or no inhibitory effects
l
on pERK expression after LPS stimulation. CLT is the most potent of
the agents tested herein in suppressing the pERK expression in the
LPS-activated macrophages.
The biochemical mechanism of thalidomide-inhibited TNF-
production in the LPS-activated macrophages had been studied.
l
a
a
l
a

8624
S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
with respect to the control. We postulated that the potent effect of
CLT on reversing the inflammatory effects induced by LPS might be
associated with the C60 fullerene moiety enhancing the delivery of
the thalidomide into the cells. Yang et al. had demonstrated that
the C60 fullerene moiety enhanced the delivery of amino acids into
the cells.23 In this study, we also observed similar effects in which
the C60 fullerene moiety in CLT enhanced the delivery of the thalid-
omide moiety to cross the cell barriers; therefore, CLT was more
effective in reversing the inflammatory effects induced by LPS. In
contrast, the cocktail treatment does not assure equal delivery of
both agents, and the thalidomide also lacks the enhancement of
the delivery acted on by the C60 fullerene moiety. CLT is a C60 fuller-
ene thalidomide dyad that amplifies the biochemical characteristics
of thalidomide and produces a new biological property that exhibits
potent synergistic anti-inflammatory effects.
(0.75 g, 5.3 mmol) in 10 mL of DMF was stirred for 1 h and then
added dropwise to the solution of 4 at room temperature. After
16 h, thesolutionwasdilutedwithethylacetate(40 mL)andwashed
with acidic water (3 Â 40 mL) and dried with MgSO4, and the solvent
was evaporated under reduced pressure. The crude material was
purified by flash chromatography (eluent DCM/methanol, 95:5,
Rf = 0.40) to afford the desired compound as an oil (1.39 g, 2.0 mmol,
yield 77%). 1H NMR (500 MHz, CDCl3): d = 2.12 (m, 1H), 2.82 (m, 2H),
2.95 (m, 1H) 3.40–3.61(m, 12H), 4.18 (d, 2H, J = 20Hz), 4.42–4.52(m,
2H), 5.08 (d, 2H, J = 14Hz), 5.12 (m, 1H), 5.16 (d, 2H, J = 5.1Hz), 6.52
(1H, NH), 7.24–7.34 (m, 10H), 7.73–7.74 (m, 2H), 7.85 (m, 2H). 13
C
NMR (125 MHz, CDCl3): 22.0, 31.5, 39.4, 42.9, 47.9, 48.5, 49.8, 50.3,
66.8, 67.4, 67.6, 69.7, 70.1, 123.7, 127.0, 127.7, 127.8, 127.9, 128.0,
128.1, 128.2, 128.3, 128.4, 128.5, 128.6, 131.7, 134.4, 135.4, 135.5,
136.3, 136.4, 156.0, 156.3, 166.5, 167.2, 168.4, 170.0, 170.6. FT-IR:
m
~ = 3363, 3064, 3034, 2950, 1681, 1614, 1499, 1457, 1240, 1172,
1135 cmÀ1. MS (FAB+): m/z = 729 [M+1]. Calcd for C38H40N4O11, 728.
3. Conclusion
4.2.2. {2-[2-(2-{2-[3-(1,3-Dioxo-1,3-dihydro-isoindol-2-yl)-2,6-
dioxo-piperidin-1-yl]-acetylamino}-ethoxy)-ethoxy]-ethylamino}-
acetic acid (3)
We successfully prepared a new C60 fulleropyrrolidine-thalido-
mide dyad, CLT. Our biological evaluation using a cell lineage to
simulate the type of macrophages present during inflammation
had demonstrated that CLT was a potent agent in reducing both
A
solution of above protected N-alkyated glycine (0.41 g,
0.6 mmol) in 10 mL of methanol was stirred for 18 h under hydro-
gen pressure (60 psi), with 10% Pd/C (0.04 g) as catalyst. The crude
material was purified by filtration through Celite; the solvent was
then evaporated under reduced pressure, yielding a solid (0.26 g,
0.6 mmol, yield 91%), 1H NMR (500 MHz, D2O): d = 1.97 (s, 1H),
2.29 (m, 1H), 2.80 (m, 1H), 3.08 (m, 2H), 3.30 (t, 2H, J = 4.9 Hz),
3.47 (t, 2H, J = 4.9 Hz), 3.67–3.75 (m, 10H), 3.79–3.81 (t, 2H,
J = 4.9 Hz), 4.53 (d, 1H, J = 16 Hz), 4.62 (d, 1H, J = 16 Hz), 5.39 (dd,
1H, J = 5.44 Hz, 13 Hz), 7.91–7.93 (m, 2H), 7.96–7.98 (m, 2H). 13C
NMR (125 MHz, CDCl3): 21.4, 31.4, 39.4, 43.2, 47.2, 49.5, 50.2,
65.8, 69.2, 69.9, 70.0, 124.2, 131.4, 135.6, 169.5, 170, 171.5,
NO and TNF-
inhibition of both NO and TNF-
a
released into the cell medium. Furthermore, CLT
synthesis in LPS-activated macro-
a
phages is attributed to its abilities to scavenge intracellular ROS,
suppress the iNOS expression and inhibit ERK phosphorylation.
We are currently exploring the utility of CLT on modulating other
cytokines in the process of inflammation. We believed that the
agents synthesized herein would hold promise for future develop-
ment of a new generation of potent anti-inflammatory agents.
4. Experimental
4.1. General
171.6, 174.4. FT-IR:
m~ = 3452, 3426, 3416, 3072, 2961, 1769,
1721, 1687, 1559, 1494, 1468, 1457, 1173, 1130, 1102 cmÀ1. MS(E-
SI-): m/z = 504 [M]. Calcd for C23H28N4O9, 504.
All reactions were carried out under argon by using standard
techniques. Solvent were dried under nitrogen by standard proce-
dures, distilled before use, and store under argon. NMR spectra were
recorded on a Bruker AMX-500 spectrometer. Chemical shifts were
reported in ppm relative to tetramethylsilane (d units). Fast atom
bombardment (FAB) mass spectra and Electrospray ionization (ESI)
at the Analytical Facility of The National Taiwan University. IR spec-
tra were obtained on Perkin-Elmer Spectrum RXI FT-IR system.
ChemicalswerepurchasedfromAcros, Aldrich, orTCIandusedwith-
out future purification. CL was prepared according to our previous
published procedure.11 The RAW 264.7 cells were mouse macro-
phage cell line, and were obtained from American Type Culture Col-
lection (ATCC, TIB-67). Fetal bovine serum (FBS) was purchased from
Hyclone (Logan, UT, USA). Dulbecco’s modified Eagle’s medium
(DMEM), penicillin, streptomycin, Lipopolysaccharide (LPS from
Escherichia coli 026:B6) were purchased from Sigma (St. Louis, MO,
USA). The stock solution of thalidomide, CL, and CLT prepared in
DMSO and added to FBS before use, and the DMSO concentrations
never exceeded 1% (v/v) in all the experiments. The enzyme-linked
immunosorbent assay (ELISA) kit used for the determination of
4.2.3. CLT
A mixture of C60 (0.05 g, 0.07 mmol), amino acid 3 (0.07 g,
0.14 mmol), and ketone 6 (0.35 mmol) in toluene (30 mL) was
heated to reflux for 5 h. After evaporation of the solvent, the crude
material was purified by flash chromatography. (DCM/methanol,
98:2, Rf = 0.39) Final product: 25 mg (25%) 1H NMR (500 MHz,
CDCl3/d-acetone): d = 2.81–2.84 (m, 2H), 2.98 (m,1H), 3.38 (s, 7H),
3.48–3.60 (m, 10H), 3.71–3.81 (m, 8H), 3.92 (m, 2H), 4.49–4.53 (m,
3H), 4.60 (d, 2H, J = 10 Hz), 4.82 (s, 4H), 4.88 (s, 2H), 5.11 (dd, 1H,
J = 5.44 Hz, 13.02 Hz), 6.35 (br, 1H, NH), 7.74–7.77 (m, 2H), 7.86–
7.88 (m, 2H). 13C NMR (125 MHz, CDCl3): 22.1, 29.7, 31.6, 39.5,
43.0, 53.3, 59.0, 67.1, 67.4, 67.5, 67.6, 67.8, 68.5, 69.7, 70.3, 70.7,
71.6, 71.7, 88.9, 92.6, 131.8, 134.4, 136.2, 141.8, 141.9, 142.1,
142.2, 142.6, 143.1, 145.3, 145.4, 145.7, 146.1, 146.2, 154.1, 155.0,
166.4, 167.2, 168.4, 170.5. MS (ESI+): m/z = 1430 [M+1]. Calcd. for
C93H48N4O13, 1429.3952 Anal. Calcd for C93H48N4O13: C, 78.14; H,
3.38; N, 3.92. Found: C, 77.69; H, 3.35; N, 3.81.
4.3. Cell culture
TNF-a was purchased from R&D systems (Minneapolis, MN, USA).
The macrophage cell line RAW 264.7 was cultured in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 10% heat-
inactivated fetal bovine serum (FBS), 100 U/ml penicillin, and
4.2. Synthetic procedures
4.2.1. (Benzyloxycarbonyl-{2-[2-(2-{2-[3-(1,3-dioxo-1,3-dihydro-
isoindol-2-yl)-2,6-dioxo-piperidin-1-yl]-acetylamino}-ethoxy)-
ethoxy]-ethyl}-amino)-acetic acid benzyl ester
100 lg/ml streptomycin at 37 °C under 5% CO2 humidified air.
4.4. Cell viability
Triethylamine (390 lL, 2.8 mmol) was added to a solution of acid
1 (1.13 g, 2.6 mmol) in 2 mL of DMF. In a separate flask, a solution of
amine salt 3 (0.81 g, 2.9 mmol), EDCI (1.02 g, 5.3 mmol), and HOBT
The 3-(4,5-dimethyldiazol-2-yl)-2,5 diphenyl tetrazolium bro-
mide (MTT) which could be reduced to yield formazone in cells’

S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
8625
mitochrondrial-dependent reduction was used as an indicator of
cell viability. One day before drug application, the RAW 264.7 cells
were seeded in 24-well flat-bottomed microtiter plates
(2 Â 105 cells/well). The RAW 264.7 cells were stimulated or not
detection of intracellular ROS was performed as described previ-
ously.19 RAW 264.7 cells were suspended in phenol red-free media
at a concentration of 2 Â 105 cells/ml. Cells were stained with
10 lM DHR123 in the dark for 30 min, and then analyzed using
with LPS(1
l
g/ml) alone or in combination with thalidomide, CL,
FACScan (Becton–Dickinson, Franklin Lakes, NJ). Oxidation of the
green DHR123 fluorescence in living cells was detected using the
FL1-H wavelength band. The fluorescence signals of 10,000 cells
were processed using a logarithmic amplifier.
CLT or an equimolar of thalidomide-CL cocktail (dissolved in
DMSO). The final concentration of DMSO was 1% (v/v). The DMSO
percentage allows the optimal solubilization of C60 fulleropyrroli-
dines in aqueous solutions. After 24 h of incubation at 37 °C, cells
were washed with PBS once, and 0.2% MTT (100
lL) was added
4.8. iNOS and ERK1/2 protein expression assay
to the well for 3 h. Then, the supernatant was removed, and the
formazone crystals were dissolved using DMSO. The optical den-
sity (OD570) for the thalidomide, CL, CLT, or equal molar of CL
and thalidomide cocktail mixture were compared to the OD of con-
The expressions of iNOS, ERK1/2 were detected by Western blot.
After treatment with 1
thalidomide (10 M), an equimolar thalidomide-CL cocktail
(10 M), or CLT (10 M) for 3 h, total cellular protein (for iNOS
lg/ml LPS or pretreatment with CL (10 lM),
l
trol or LPS(1
lg/ml)-stimulated wells to assess the cytotoxicity.
l
l
The absorbance was read at 570 nm with an ELISA plate reader (TE-
CAN SunriseTM, Switzerland). Data are report as mean SEM values
of three independent determinations.
and ERK1/2) was prepared using lysis buffer containing 10% Glyc-
erol, 1% Triton X-100, 1 mM sodium orthovanadate, 1 mM EGTA,
5 mM EDTA, 10 mM NaF, 1 mM sodium pyrophosphate, 20 mM
Tris–HCl, pH 7.9, 100 mM B-glycerophosphate, 137 mM NaCl,
1 mM phenylmethylsulfonyl fluoride (PMSF), and 10 mg/ml aproti-
nin. Thirty to fifty milligrams of proteins were separated on so-
dium dodecyl sulfate–polyacrylamide gel (SDS–PAGE) (10% for
iNOS, phospho-ERK1/2, ERK1/2, and GAPDH) and electro trans-
ferred to polyvinylidene difluoride (PVDF) membrane (Immobi-
lonP, Millipore, Bedford, MA, USA). The membrane was
preincubated in Tris-buffered saline (TBS) containing 0.01% Tween
20, 1% bovine serum albumin (BSA) for 1 h, and then incubated
with anti-GAPDH monoclonal antibody (St. Louis, MO, USA), anti-
phospho (Thr202/Tyr204)-specific p44/42 MAP kinase antibody
(St. Louis, MO, USA), anti-p44/42 MAP kinase antibody (St. Louis,
MO, USA), or anti-iNOS polyclonal antibodies overnight at 4 °C (di-
luted 1:2000). After incubation with horseradish peroxidase-con-
jugated or alkaline phosphatase-conjugated anti-rabbit/mouse
IgG antibody, the immunoreactive bands were visualized with
the enhanced chemiluminescence reagents (ECL, Amersham).
4.5. Nitrite assay
Nitric oxide (NO) production in cell culture supernatant was
evaluated by measuring the nitrite concentration. The nitrite con-
centration was detected with the Griess reaction. The RAW 264.7
were plated at a density of 2 Â 105 cells/ml in 24-well plates for
24 h, followed by treatment different concentrations of the indi-
cated compounds, with thalidomide, CL, CLT, or an equimolar of
thalidomide-CL cocktail for 3 h, then treatment with LPS (1 lg/
ml) for 24 h. One hundred microliters of each supernatant was
mixed with the same volume of Griess reagent (1% sulfanilamide
in 5% phosphoric acid and 0.1% naphthylethylendiamine dihydro-
chloride in water) and incubated at room temperature for
10 min. The absorbance was read at 550 nm with an ELISA plate
reader (TECAN SunriseTM, Switzerland). The amount of nitrite in
the samples (in micromolar) was calculated from a sodium nitrite
standard curve freshly prepared in culture medium. Data are re-
port as mean SEM values of three independent determinations.
Acknowledgments
4.6. TNF-
a
assay
This work was supported by the Nation Science Council (NSC-
96-2113-M-027-002-MY3). We are also grateful to the National
Center for High-performance computing for the computer time
and facilities.
Soluble cytokine was tested in the supernatants of cultured
RAW 264.7 macrophage by ELISA. The RAW 264.7 were plated at
a density of 2 Â 105 cells/ml in 24-well plates for 24 h, followed
by treatment different concentrations of the indicated compounds
such as with thalidomide, CL, CLT, or an equimolar of thalidomide-
Supplementary data
CL cocktail mixture for 3 h, then treatment with LPS (1
lg/ml) for
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.08.004.
24 h. The TNF- -protein in cell supernatant was detected using
a
the ELISA kit (R&D, Catalogue No. DY410) and conducted according
to the manufacturer instructions. The absorbance was read at
540 nm with an enzyme-linked immunosorbent assay (ELISA)
plate reader (TECAN SunriseTM, Switzerland). Data are report as
mean SEM values of three independent determinations.
References and notes
1. (a) Jialal, I.; Devaraj, S.; Venugopal, S. K. Free Radical Res. 2002, 36, 1331–1336;
(b) Suryaprasad, A. G.; Prindiville, T. Autoimmun. Rev. 2003, 2, 346–357.
2. (a) Palmer, R. M.; Ashton, D. S.; Moncada, S. Nature 1988, 333, 664–666;
(b) Duval, D. L.; Miller, D. R.; Collier, J.; Billings, R. E. Mol. Pharmacol. 1996,
50, 277–284.
4.7. ROS assay
3. Nakamura, E.; Isobe, H. Acc. Chem. Res. 2003, 36, 807.
4. Yang, J.; Alemany, L. B.; Driver, J.; Hartgerink, J. D.; Barron, A. R. Chem. Eur. J.
2007, 13, 2530–2545.
5. Da Ros, T.; Bergamin, M.; Vazquez, E.; Spalluto, G.; Baiti, B.; Moro, S.; Boutorine,
A. S.; Prato, M. Eur. J. Org. Chem. 2002, 3, 405–413.
Levels of cellular oxidative stress were measured using the fluo-
rescent probe, dihydrorhodamine 123 (DHR123), as described pre-
viously. The DHR 123 dye enters mitochondria and fluoresces
when oxidized by ROS. After treatment with 100 ng/ml LPS or pre-
6. (a) Imahori, H.; Fukuzumi, S. Adv. Funct. Mater. 2004, 14, 525; (b) Guldi, D. M.
Pure Appl. Chem. 2003, 75, 1069.
treatment with CL (10
thalidomide-CL cocktail (10
stained with 10 M DHR123 for 30 min, and washed with phos-
lM), thalidomide (10
lM), an equimolar
7. (a) de la Torre, M. D. L.; Rodrigues, A. G. P.; Tome, A. C.; Silva, A. M. S.; Cavaleiro,
J. A. S. Tetrahedron 2004, 60, 3581; (b) del la Torre, M. D. L.; Marcorin, G. L.; Pirri,
G.; Tome, A. C.; Silva, A. M. S.; Cavaleiro, J. A. S. Tetrahedron Lett. 2002, 43, 1689;
(c) de la Torre, M. D. L.; Tome, A. C.; Silva, A. M.; Cavaleiro, J. A. S. Tetrahedron
Lett. 2002, 43, 4617.
8. Krusic, P. J.; Wassermann, E.; Keizer, P. N.; Morton, J. R.; Preston, K. F. Science
1991, 254, 1183.
9. Kulkarni, R. G.; Achaiah, G.; Sastry, G. N. Curr. Pharm. Des. 2006, 12, 2437.
l
M), or CLT (10 M) for 3 h, cells were
l
l
phate-buffered saline. Cells were imaged using a confocal micro-
scope (Leica TCS SP5, Bensheim, Germany). Cells were located
under bright-field optics and then scanned once with a laser (with
excitation at 488 nm and emission at 510 nm). Flow cytometric

8626
S.-T. Huang et al. / Bioorg. Med. Chem. 16 (2008) 8619–8626
10. (a) Komatsu, W.; Ishihara, K.; Murata, M.; Saito, H.; Shinohara, K. Free Radical
Biol. Med. 2003, 34, 1006; (b) Yamakoshi, Y.; Umexawa, N.; Ryu, A.; Arakane, K.;
Miyata, N.; Goda, Y. J. Am. Chem. Soc. 2003, 125, 12803; (c) Gharbi, N.; Pressac,
M.; Hadchouel, M.; Szwarc, H.; Wilson, S. R.; Moussa, F. Nano Lett. 2005, 5,
2578.
11. Huang, S. T.; Liao, J. S.; Fang, H. W.; Lin, C. M. Bioorg. Med. Chem. Lett. 2008, 18,
99–103.
12. (a) Satoh, M.; Matsuo, K.; Takanashi, Y.; Takayanagi, I. Gen. Pharmacol. 1995, 26,
1533; (b) Satoh, M.; Matsuo, K.; Kiriya, H.; Mashino, T.; Nagano, T.; Hirobe, M.
Eur. J. Pharmacol. 1997, 327, 175; (c) Satoh, M.; Mashino, T.; Nagano, T.; Hirobe,
M.; Takayanagi, I. Fullerene Sci. Technol. 2001, 9, 141.
16. Kordatos, K.; Da Ros, T.; Bosi, S.; Vázquez, E.; Bergamin, M.; Cusan, C.; Pellarini,
F.; Véronique, T.; Benedetta, B.; Davide, P.; Georgakilas, V.; Giampiero, S.; Prato,
M. J. Org. Chem. 2001, 66, 4915.
17. Wolleb, H.; Pfander, H. Helv. Chim. Acta 1986, 69, 1505.
18. (a) Hu, Z.; Guam, W.; Wang, W.; Huang, L.; Tang, X.; Xu, H.; Zhu, Z.; Xie, X.;
Xing, H. Carbon 2008, 46, 99–109; (b) Partha, R.; Lackey, M.; Hirsch, A.;
Casscells, W. S.; Conyers, J. L. J. Nanobiotechnology 2007, 5, 6.
19. Huang, S. T.; Tsai, H. D.; Kuo, H. S.; Yang, Y. P.; Peng, Y. C.; Lin, Y. L.
ChemBioChem 2004, 5, 797.
20. Green, L. C.; Wagner, D. A.; Glogowski, J.; Skipper, P. L.; Wishnok, J. S. Anal.
Biochem. 1982, 126, 131.
13. Melchert, M.; List, A. Int. J. Biochem. Cell Biol. 2007, 39, 1489–1499.
14. Dumitru, C. D.; Ceci, J. D.; Tsatsanis, C.; Kontoyiannis, D.; Stamatakis, K.; Lin, J.-
H.; Patriotis, C.; Jenkis, N. A.; Copeland, N. G.; Kollias, G. Cell 2000, 103, 1071–
1083.
21. Cho, J. Y.; Park, J.; Yoo, E. S.; Yoshikawa, K.; Baik, K. U.; Lee, J. Arch. Pharm. Res.
1998, 21, 12.
22. Chang, Y. C.; Lee, F. W.; Chen, C. S.; Huang, S. T.; Tsai, S. H.; Huang, S. H.; Lin, C.
M. Free Radic. Biol. Med. 2007, 43, 1541–1551.
15. Hess, S.; Akermann, M. A.; Wnendt, S.; Zwingenberger, K.; Eger, K. Bioorg. Med.
Chem. 2001, 9, 1279–1291.
23. Yang, J.; Wang, K.; Driver, J.; Yang, J.; Barron, A. R. Org. Biomol. Chem. 2007, 5,
260–266.