F. Wu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6430–6433
6431
Scheme 1. General synthetic route for compounds 3–23.
The amines (aliphatic amines, aromatic amines and ami-
no acid esters) were added to the reaction mixture and
heated at 60 ꢁC for 1 h to give the amides 4–23, which
were purified by HPLC and identified by 1H NMR,
13C NMR and HRMS. The compound 3 was formed
as a by-product in the reactions and it was confirmed
used in the experiments. The erythrocytes were washed
three times in PBS (phosphate-buffered saline:
NaCl = 8 g/L, KCl = 0.2 g/L, Na2HPO4 = 1.44 g/L and
KH2PO4 = 0.24 g/L; pH 7.4) and then diluted with
PBS to obtain a 10% suspension. All the tested com-
pounds were dissolved in DMSO at a concentration of
500 lg/mL, and then PBS was added to prepare the test-
ing concentrations ranging from 2.5 to 250 lg/mL. The
final volume of the sample was 1.0 mL. The erythrocyte
suspension (100 lL) was added to the samples to be
tested then the samples were rapidly stirred and incu-
bated at 37 ꢁC with periodic stirring during a 60-min
incubation period. The solutions were then centrifuged
at 3000 rpm for 5 min and the absorbance of the super-
natant was measured at 540 nm using Thermo Multis-
kan MK3 (Labsystems Dragon, Finland, model: 353).
The haemolysis percentage was calculated by compari-
son with the 100% haemolysis caused by distilled water
as maximal haemolytic controls. The haemolytic percent
developed by the PBS control was subtracted from all
groups. The concentration inducing 50% of the maxi-
mum haemolysis is abbreviated as HD50. Each experi-
ment included triplicate at various concentrations and
the results are listed in Table 1.
1
by H NMR, 13C NMR and HRMS to be the product
of EsA coupling with DCU. The by-product formation
appears to be related with the high solubility of DCU
in DMF. This suggested that DMF is not a favourable
solvent for this reaction. When DMF–THF (1:2, v:v)
was used as the solvent, good yields of the desired
amides were achieved and there was no trace of by-prod-
uct 3 found in the reaction mixtures.
The measurement of the inhibitory effects on reactive
oxygen species fluorescence in hCOX-2 expressing sf-9
cells is a rapid method to screen COX-2 inhibitors and
is more effective, less expensive, and does not have iso-
tope contamination in contrast to RIA or ELISA.14
The inhibitory effects of the compounds were deter-
mined by the method described by Zhang.15 Briefly,
recombinant human COX-2 (hCOX-2) was expressed
in insect sf-9 cells and harvested cloned sf-9 cells were
stored in liquid nitrogen until use. Reactive oxygen
species production was stimulated by arachidonic
acid in sf-9 cells and was measured by 20,70-dichlorodi-
hydrofluorescein diacetate (DCDHF-DA) fluorescence.
DCDHF-DA can rapidly permeate into the cells and
is converted into 20,70-dichlorodihydrofluorescein
(DCDHF). DCDHF is not fluorescent, but it can rap-
idly react with a reactive oxygen species to produce fluo-
rescence.16 The sf-9 cells (1 · 105/mL) containing
hCOX-2 protein were preincubated with the tested com-
pounds, including 1, 3–23, and Celecoxib (10 lM), for
30 min in a 96-well black plate, followed by adding
DCDHF-DA (2.5 lM, final concentration) and arachi-
donic acid (5 lM, final concentration). The fluorescence
in the cell suspension was immediately detected using a
PolarStar plate reader with excitation wavelength of
485 nm, emission wavelength of 520 nm. The rate of
fluorescence generation in the first 10 min was recorded.
The biological evaluation of the compounds was investi-
gated at the concentration of 10 lM in hCOX-2 express-
ing sf-9 cells in vitro. Four parallel experiments were
performed for each sample. The inhibitory activities of
the tested compounds are listed in Table 1.
Over the past two decades, we have revealed that EsA
has strong anti-inflammatory activity through effects
on several important inflammatory mediators including
TNF, IL-1 and COX-2. Haemolytic activity is the main
toxicity of EsA, which needs to be overcome. We initi-
ated the optimization of EsA, which aims at increasing
the anti-inflammatory activity and lowering the haemo-
lytic activity.
The compounds 2, 11–16 and 23 bearing aromatic ring
showed higher potency towards COX-2 than EsA and
Celecoxib, but also showed higher haemolytic activity
than EsA. The compounds 4–9 containing aliphatic
groups lost their activity towards COX-2 and also
showed lower haemolytic activity. From this result, we
can make a conclusion that aromatic ring can promote
the biological activity of EsA derivatives. Most notice-
able is compounds 12 and 14 exhibited the highest inhi-
bition rates, 158.2% and 126.1%, respectively, while EsA
and Celecoxib exhibited 21.7% and 40.0%, respectively.
However, the haemolytic activities of the compounds 12
and 14 are also the highest. Interestingly, our research
showed correlativity between COX-2 inhibitory activity
and haemolytic activity of EsA, and there was no report
in the literature, which is worth further research. How-
ever, the results suggested us that the C-28 carboxylic
acid group plays an important role in biological activity.
Haemolysis assay was carried out following the proce-
dure reported by Wang.17 Non-heparinized blood of
healthy New Zealand rabbit (Experimental Animal Cen-
ter of Second Military Medical University, China) was
Wu, Fei
