New NSAIDs-NO hybrid molecules with antiproliferative properties on human prostatic cancer cell lines

Article abstract of DOI:10.1016/j.bmcl.2008.07.018

The design of profen hybrids containing a NO donor moiety connected to an aliphatic spacer led to compounds with a similar cyclooxygenase inhibition compared to their parent profen and with significant antiproliferative activities on PC3 cells. However, inhibition of COX-2 pathway alone did not seem sufficient to inhibit cancer cell proliferation, and NO-release in a time-dependent manner strongly contributes to this activity.

Full text of DOI:10.1016/j.bmcl.2008.07.018

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4655–4657
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
New NSAIDs-NO hybrid molecules with antiproliferative properties
on human prostatic cancer cell lines
Nicolas Bézière a, Laurence Goossens a, Jean Pommery a, Hervé Vezin b, Nadia Touati b,
Jean-Pierre Hénichart a, Nicole Pommery a,
*
a Université de Lille 2, 3 rue du Professeur Laguesse, BP 83, 59006 Lille Cedex, France
b Laboratoire de Chimie Organique et Macromoléculaire, UMR-CNRS 8009, USTL Bât C4, 59655, Villeneuve d’Ascq Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 February 2008
Revised 2 July 2008
Accepted 4 July 2008
Available online 10 July 2008
The design of profen hybrids containing a NO donor moiety connected to an aliphatic spacer led to com-
pounds with a similar cyclooxygenase inhibition compared to their parent profen and with significant
antiproliferative activities on PC3 cells. However, inhibition of COX-2 pathway alone did not seem suffi-
cient to inhibit cancer cell proliferation, and NO-release in a time-dependent manner strongly contributes
to this activity.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Cancer
Cyclooxygenase
EPR
Nitric oxide
Profens
In vivo, in vitro and epidemiological studies have all demon-
strated that non-steroidal anti-inflammatory drugs (NSAIDs)
appear to be effective in the chemoprevention and possible
treatment of many cancers.1 On the other hand, recently and
initially in order to counteract the side effects of these drugs on
the gastric mucosa, nitric oxide (NO) donating NSAIDs have been
synthesized to take advantage of the multiple roles of this ubiquitous
messenger. In addition to its similar properties to prostaglandins
within the gastric mucosa, evidence is accumulating for the role
of NO as a new oncopreventive agent and as a novel therapeutic
to overcome cell resistance.2 Numerous new chemical entities with
enhanced anticancer properties, safety and efficacy versus NO-free
control compounds were then developed, and the archetypical
hybrid drug (in preclinical development) is nitric oxide-donating
aspirin (NO-ASA).3 However, even if NO-ASA seems, in cultured
cancer cells, 1000-fold more potent than aspirin, the role of NO
and NSAID moiety is more complex than that previously thought.
Several mechanisms of action have been proposed including (or
not) COX-2 activity,4 and several signalling factors such as
tion of these new therapeutic derivatives and of each part of the
molecule, we studied the effect of several new NO-NSAIDs on the
growth of human prostate cancer cells, and the influence of NO
release kinetics.
In an attempt to rationalize the link between cyclooxygenase
inhibition, NO liberation and cell proliferation and to study the
Cl
O
O
O
OH
OH
O
S
O
N
H
OH
A
C
B
Ketoprofen (1)
Suprofen (2)
Carprofen (3)
Figure 1. Structure of profens 1-3.
R 1
R 1
R 1
NF-jB, IFN-b, Hsp90, and p75 and latest hypothesis even evoke
a
b
Br
Br
O
OH
ONO2
R2
ONO2
the active influence of the spacer used to connect NO-donating
O
group and NSAID.5
4a,b
5a,b
R2 = A 6a, 7b
R2 = B 8a, 9b
Based on these considerations, which emphasize the need of
additional essential data on mechanism of metabolisation and ac-
a
b
R1 = H
R2 = C 10a, 11b
R1 = CH3
* Corresponding author. Tel.: +33 (0)3 20 96 49 03; fax: +33 (0)3 20.95 90 09.
E-mail address: nicole.pommery@univ-lille2.fr (N. Pommery).
Scheme 1. Reagents and conditions: (a) 70% HNO3, 95% H2SO4, 2 h, 0 °C; (b) 1—3,
Cs2CO3, DMF, 12 h, 20 °C.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.018

4656
N. Bézière et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4655–4657
I
ONO2
ONO2
2
R
2
R
a
b
2
R
2
R
OH
O
O
O
O
I
O
O
O
R2 = A
R2 = B
1
2
14
15
12
13
Scheme 2. Reagents and conditions: (a) Allyl bromide, Cs2CO3, DMF, 20 °C, 12 h; (b) AgNO3, I2, CH3CN, 0 °C, then AgNO3, reflux.
dissociation kinetics of such compounds, hybrid molecules were
designed from representative compounds of these kinds of activity.
Profens, that is, Ketoprofen 1, Suprofen 2 (Fig. 1), were chosen be-
cause of their interesting inhibitory activity against cyclooxygen-
ase and of their different selectivity versus the two isoforms
COX-1/COX-2. Carprofen 3, even if globally less effective, can be
representative of a more selective COX-2 inhibition.
the literature. As expected, the modified profens have a very sim-
ilar IC50 value (Table 1), and this can be explained by the esterase
activity in human whole blood assays, enzymes able to cleave the
ester bond by which the nitric oxide-donating moiety is linked to
the parent cyclooxygenase inhibitor. Only the Carprofen deriva-
tives present a significant difference: with a spacer and a NO donor,
cyclooxygenase inhibition activity is cut in half. The release of this
parent compound was verified by evaluating the amount of profen
released via HPLC (Table 1). Results seem to indicate that the
release was effective and almost complete after 22 h of incubation
in human whole blood. In addition to the cleavage of the ester
bond, biodegradation of the profen itself can and probably has
occurred during the incubation time, hence the sub-100% release
values are obtained.
The NO-donor moiety was carried by a nitrooxy alkyl function
brought by the
x
Àbromoalkanols 4a,b: the nitric ester group
was chosen for its kinetic profile of metabolization.6 The aliphatic
spacer was modulated in an attempt to modify the profen-release
kinetics.
The NO-donating profens 6–11, 14, 15 were synthesized as de-
scribed on Schemes 1 and 2.
The set of compounds 6–11 containing one nitrooxy function
was obtained (Scheme 1) via nitration of the bromoalkanols 4a,b
using a 70% HNO3 95% H2SO4 mixture, giving the corresponding
nitrooxyalkyl bromides 5a,b in quantitative yield. The nitro-
oxyalkyl esters 6–11 were prepared by condensation of the bro-
mides 5a,b with profens 1–3 in the presence of Cs2CO3.
The dinitrooxy-containing compounds 14,15 were prepared
(Scheme 2) from profens 1,2. Allyl esters 12,13 were prepared by
nucleophilic substitution of allyl bromide by the carboxylic acid
function of Ketoprofen 1 or Suprofen 2 (DMF, Cs2CO3). A subse-
quent vicinal iodination of the allyl function (products not isolated)
followed by a two-step nitration of the primary and secondary ha-
lide carbons (AgNO3, acetonitrile), according to a previously de-
scribed procedure7, led to dinitrooxy esters 14 and 15. This route
failed with the allyl ester of Carprofen 3 because of the instability
of the intermediate iodonium ion. Every structure was confirmed
by IR and NMR spectrums as well as LC/MS chromatograms, and
purity of both profens and final products exceeded 98%.
Assays for the inhibition of COX pathways were performed in
human whole blood (HWB),8 and Ketoprofen, Suprofen and
Carprofen displayed inhibition values similar to those found in
The first series of assays on PC3 cell proliferation was done
using the standard incubation time of 72 h with the compounds.10
We first had to witness almost no activity at all from the profens on
these very resistant cells. However, when linked to a NO-donating
moiety, the derivatives showed some potency that can apparently
be accounted only to the NO release, even though the difference
between molecules having one and two nitric esters is not signifi-
cant (Table 1).
Three molecules were then incubated for 24, 48 (data not
shown) and 72 h in a proliferation kinetic assay. Moreover, mole-
cules were chosen to evaluate if the inactive profen moiety had
any influence on proliferation when linked to the exact same nitric
oxide donor. Results are shown in Fig. 2. As expected, we observed
that the effect of NO was much more visible after 24 h incubation;
this is probably due to the metabolic activity.
EPR spectroscopy11 was simultaneously performed to see if a
link could be found between the amount of NO released in a given
amount of time and the proliferation kinetics profile. The metabo-
lization process of our molecules includes the cleaving of the ester
bond by esterase, the release of ONO2À and the subsequent 3 elec-
trons reduction to obtain NO. And since this radical half life is very
Table 1
COX pathways inhibitiona, profen releaseb in HWB assays and PC3 cells proliferation assays (percentage of inhibition at 100
lM or IC50)
R 1
R 3
R2
O
O
Compound
R1
R2
R3
Inhibition of COX-1 pathway IC50
Ma
,
Inhibition of COX-2 pathway IC50
Ma
,
Profen release ratio %,
10
PC3
l
l
lM
Ketoprofen
6
7
14
—
H
CH3
–ONO2
—
A
A
A
—
0,33
0,45
0,60
0,58
0,69
0,77
0,73
0,74
100
73
70
0
–ONO2
–ONO2
—CH2ONO2
26
26
21
79
Suprofen
8
9
15
—
H
CH3
–ONO2
—
B
B
B
—
0,56
0,58
0,57
0,48
2,75
5,80
4,56
2,70
100
46
86
0
7
13
33
–ONO2
–ONO2
–CH2ONO2
73
Carprofen
10
11
—
H
CH3
—
C
C
—
–ONO2
–ONO2
59
110
123
38
65
45
100
88
100
5
IC50 = 48
IC50 = 48
l
l
M
M
a
All IC50 values represent the mean of at least four experiments.
Profen release ratio measured after an incubation time of 22 h, using the same blood pool as the COX inhibition experiments.
b

N. Bézière et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4655–4657
4657
and chemical degradation of the NO-scavenger complex begin to
be faster than the actual release.
24h incubation time
120
100
80
60
40
20
0
These results first confirm that it is excluded to directly link
inhibition of COX activity and antiproliferative properties. The rad-
ical NO was showed much more effective, but our data suggest that
no general conclusion will ever be valid and that each part of hy-
brid molecules (NSAIDs, linkers, NO-donating group) and above
all targeted tissues contribute to the effectiveness of such drugs.
7
9
11
References
1
10
100
-20
Concentrations (µM)
1. Zha, S.; Yegnasubramanian, V.; Nelson, W. G.; Isaacs, W. B.; De Marzo, A. M.
Cancer Lett. 2004, 215, 1.
2. Stefano, F.; Distrutti, E. Curr. Top. Med. Chem. 2007, 7, 277.
3. Chiroli, V.; Benedini, F.; Ongini, E.; Del Soldato, P. Eur. J. Med. Chem. 2003, 38,
441.
72h incubation time
120
100
80
60
40
20
0
4. Kashfi, K.; Rigas, B. Biochem. Pharmacol. 2005, 70, 969.
5. Kashfi, K.; Rigas, B. Biochem. Biophys. Res. Commun. 2007, 358, 1096.
6. Yap, S. K.; Rhodes, C. T.; Fung, H.-L. Anal. Chem. 1975, 47, 1183.
7. Lazzarato, L.; Rolando, B.; Lolli, M. L.; Tron, G. C.; Fruttero, R.; Gasco, A.; Deleide,
G.; Guenther, H. L. J. Med. Chem. 2005, 48, 1322.
7
9
11
8. HPLC analysis of arachidonic acid metabolites in whole blood.9 Fresh blood was
collected in heparinized tubes from normal volunteers. Then, 1 mL aliquots
were transferred to heparinized tubes preloaded with either 4
(DMSO) or 4 L of test compounds and incubated for 15 min at 37 °C (final
compound concentrations used: 10À7; 5.10À7; 10À6; 5.10À6; 10À5; 5.10À5; 10À4
and 5.10À4 M as solubility allows). Then, 40
M of calcium ionophore A 23187
alone was added, and incubation was carried out at 37 °C for 20 h to stimulate
the COX-1 pathway. For the COX-2 pathway, samples stimulated with
23187+LPS (500 g) were incubated for 20 h. Eicosanoids were extracted from
lL of vehicle
l
1
10
100
-20
l
Concentrations (µM)
A
Figure 2. Effect of 7, 9 and 11 on PC3 cell proliferation after 24 and 72 h incubation.
l
the samples into ethyl acetate. Samples were then evaporated to dryness under
nitrogen, re-suspended in the mobile phase to precipitate proteins and then
centrifuged. The supernatant was directly analyzed by HPLC using Hypersil
6000
5000
4000
3000
2000
1000
0
ODS
3
l
M
columns (12.5 mm  0.46 mm),
a methanol/water/acetic acid
(80:20:0.08) mobile phase at a flow rate of 1 mL/min and UV detection firstly
at 270 nm and 12 min later at 245 nm.
9. Pommery, J.; Pommery, N.; Hénichart, J.-P. Prostaglandins, Leukot. Essent. Fatty
Acids 2005, 73, 411.
10. Cell culture and cell proliferation assays. Human prostate PC3 cancer cells were
grown at 37 °C in RPMI 1640 medium supplemented with 10% fetal calf serum,
in a humidified incubator containing 5% CO2. In the cell proliferation assay,
cells were plated (1.8 Â 104 cells/well) on 24-well plates. After 3 days, the cell
medium was changed to serum free medium, and the cells were starved for
24 h for culture synchronization. Cells were then incubated in culture medium
that contained various concentrations of test compounds (1; 10; 50 and
100
lM as solubility allows), each dissolved in less than 0.1% DMSO. After
0
5
10
15
incubating for
a
given amount of time (usually 72 h), cell growth was
estimated by the colorimetric MTT test.
Time (h)
11. EPR spectra were recorded using a Bruker ELEXYS E580 spectrometer operating
at 9.5 GHz with a 100 kHz high frequency modulation and a modulation
amplitude of 1 G. The sample solutions were examined in quartz cell inserted
in standard thermoregulated cavity for the EPR spectra at liquid nitrogen
Figure 3. NO released from 7 evaluated by EPR spectroscopy.
short (a few seconds in HWB), it requires the use of iron (II) N,N-
diethyldithiocarbamate, a radical scavenger.12
According to our measurement, NO release from the nitric ester
in human whole blood is slow, and it shows a peak in concentra-
tion after approximately 6 h (Fig. 3). Afterwards, metabolic activity
temperature. The EPR signal of NO° trapped by iron (II)
diethyldithiocarbamate consists in three lines with nitrogen hyperfine
splitting constant of 13 G. The concentration of NO° released was quantified by
double integration of EPR signal using 2,2,6,6-tetramethyl-1-piperidinyloxy
TEMPO as standard.
N,N-
a
12. Samouilov, A.; Zweier, J. L. Methods Enzymol. 2002, 352, 506.