Structure-activity relationships studies of the anti-angiogenic activities of linomide

Article abstract of DOI:10.1016/S0960-894X(03)00047-7

The synthesis and anti-angiogenic activities of linomide and its analogues are reported. Three of the analogues are 3.3-69 times more potent than linomide at inhibiting blood vessel formation in the CAM angiogenesis assay. These compounds possessed consid

Full text of DOI:10.1016/S0960-894X(03)00047-7

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1187–1189
Structure–Activity Relationships Studies of the Anti-Angiogenic
Activities of Linomide
Jiandong Shi,a Zili Xiao,a Michael A. Ihnat,b Chandrashekhar Kamat,b
Bulbul Pandit,a Zhigen Hua and Pui-Kai Lia,*
aDivision of Medicinal Chemistry and Pharmacognosy, College of Pharmacy, The Ohio State University,
Columbus, OH 43210, USA
bThe University of Oklahoma, School of Medicine, Oklahoma City, OK 73104, USA
Received 23 October 2002; accepted 2 December 2002
Abstract—The synthesis and anti-angiogenic activities of linomide and its analogues are reported. Three of the analogues are 3.3–69
times more potent than linomide at inhibiting blood vessel formation in the CAM angiogenesis assay. These compounds possessed
considerable anti-proliferative activity against isolated HUVEC cells with no activity against epithelial-derived prostate tumor cells.
# 2003 Elsevier Science Ltd. All rights reserved.
Linomide (N-phenylmethyl-1,2-dihydro-4-hydroxyl-1-
methyl-2-oxo-quinoline-3-carboxamide) has been shown
to be an immunomodulator. In clinical trials, it has been
shown to have potential in the treatment of auto-
immune disease such as rheumatoid arthritis, systemic
lupus erythematosis and multiple.1À3 Recently, linomide
has also been reported to have antiangiogenic activ-
ity.4,5 Angiogenesis is the formation of new blood ves-
sels from pre-existing ones. It has been shown that
without angiogenesis a tumor can only reach the size of
1–2 mm3, small enough to be easily resected or treated
with conventional cytotoxic chemotherapeutic agents.
In vitro, linomide inhibited endothelial cells prolifera-
tion and migration4,6 as well as invasion through base-
ment membrane at high concentration (>100 mg/mL).4
In vivo, linomide has been demonstrated to cause a dose
dependent, antiangiogenic activity using a Matrigel-
based quantitative in angiogenic assay in rats.4,7 In
addition, linomide has been reported to have in vivo
antitumor effect against rat and human prostatic
cancers.8À10
synthesized several analogues (2–7) of linomide to
investigate its structure–activity relationships (Fig. 1),
particularly on the importance of 1-methyl (R1),
4-hydroxy (R4), N-methyl (R2) and N-phenyl (R3)
group in regard to their antiangiogenic activity.
Linomide 1, compounds 2 and 3 were prepared accord-
ing to the procedure described by Edgar et al., as out-
lined in Figure 2.11 Briefly, stirring N-methylisatoic
anhydride 9 with dimethylmalonate 11 in the presence
of sodium hydride and DMF at 120 ꢀC yielded the
hydroxyquinoline carboxylic ester 12. Linomide 1,
compounds 2 and 3 were obtained by coupling the ester
12 with N-methylaniline, dimethylamine and aniline,
respectively. The syntheses of compounds 4 and 5 were
similar to the syntheses of linomide and compound 3
except isatoic anhydride 10 was used as the starting
material.
The antiangiogenic activity of linomide has been known
for more than 10 years. Surprisingly, no structure–
activity relationship studies of the antiangiogenic activ-
ity of linomide have been reported. In our continued
effort to search for potent antiangiogenic agents, we
*Corresponding author. Fax: +1-614-688-8556; e-mail: li.27@osu.edu
Figure 1. Structures of linomide and its analogues.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00047-7

1188
J. Shi et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1187–1189
Figure 2. Syntheses of linomide 1 and compounds 2–5. Reagents and conditions: (a) NaH, DMF, 120 ꢀC, 2.5 h, 44.2–72.7%; (b) R2R3NH, toluene,
reflux 4 h, 55.4–80.5%.
Figure 3. Syntheses of compounds 6 and 7. Reagents and conditions: (a) TsOH, toluene, 36 h, 70.5%; (b) Fe/AcOH, 100 ꢀC, 6 h, 66.6%;
(c) (i) NaOH/MeOH/H2O, 60 ꢀC, 12 h, 93%; (ii) (1) SOCl2; (2) aniline/DMF, rt, overnight, 56.2%; (d) (i) NaOH/MeOH/H2O, 60 ꢀC, 12 h, 93%;
(ii) (1) SOCl2; (2) N-methylaniline/DMF, rt, overnight, 53.1%; (e) CH3I/NaH/DMF, rt, overnight, 85.6%.
The syntheses of compounds 6 and 7 (Fig. 3) have never
been reported. Quinolinecarboxylate 16 is the common
intermediate which was prepared from compound 14
according to the reported procedure.12 Compound 6
was obtained by reacting 16 with thionyl chloride fol-
lowed by condensation with aniline. Substituting aniline
with N-methylaniline yielded 17 which was methylated
at the N-1 position to yield compound 7. The com-
pounds were purified by column chromatography and
recrystallization and their purities were confirmed by
elemental analyses.
replaced by H, are devoid of anti-angiogenic activity
(>100 mg/embryo). Interestingly, replacing the 1-methyl
(R1) or N-methyl (R2) with H resulted in significantly
increased anti-angiogenic activity. The ED50 of com-
pounds 3 and 4 are 18 and 7.5 mg/embryo, respectively
3.3 and 8 times more potent than linomide. Eliminating
both of the methyl groups in linomide yielded com-
pound 5 which is 69 times more potent than linomide
with an ED50 of 0.87 mg/embryo. In addition, the
N-phenyl group is not optimal for anti-angiogenic
activity. Replacing the phenyl group in linomide with a
methyl moiety to form compound 2 increases the anti-
angiogenic activity (ED50=8.5 mg/embryo).
The antiangiogenic activities of linomide and its ana-
logues were evaluated using two established angiogenic
assays. The first assay we used was an in vivo chicken
chorioallantoic membrane (CAM) assay.13,14 In the
CAM assay, human VEGF-165 and bFGF (200 ng
each) were added to saturation to a microbial testing
disk and placed onto the CAM of 10-day-old chicken
embryos.15 Eight h after the growth factor treatment,
antiangiogenic compounds were added to the same
microbial testing disk and the growth factors and
embryos allowed to incubate an additional 40 h. CAMs
were removed, paraformaldehyde fixed, placed onto
Petri dishes, and digital images taken at 7.5Â using a
dissecting microscope. A 1 Â 1-cm grid was then added
to the digital CAM images and the number of vessels
within each grid counted as a measure of vascularity.
Table 1 represents the CAM assay data of linomide and
its analogues and dose in mg/embryo of compound
necessary to reduce the blood vessel number to 50%
that of the VEGF/bFGF treated group. Linomide is
weakly active with ED50 of 60.1 mg/embryo. It appears
that the 4-OH group (R4) is critical for anti-angiogenic
activity since both compounds 6 and 7, with 4-OH
Because we found that linomide and its analogues were
active in a functional angiogenesis assay, we next sought
to define the cellular target for the compounds. First, we
Table 1. Effect of linomide and its analogues on CAM angiogenesis
and human umbilical vein endothelial cell (HUVEC) proliferation
Compd
R1
R2
R3
R4
ED50
IC50 HUVEC
(mg/embryo) proliferation
(mM)
Linomide 1 CH3 CH3 Ph OH
60.1
8.5
18.0
7.5
13.95
16.47
7.20
6.55
4.0
2
3
4
5
6
7
CH3 CH3 CH3 OH
CH3
H
H
H
Ph OH
CH3 Ph OH
Ph OH
H
0.87
>100
>100
CH3 CH3 Ph
Ph
H
H
20.58
22.26
H
H

J. Shi et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1187–1189
1189
examine the anti-proliferative activity of the compounds
on HUVEC cells as previously described.15 We chose
HUVEC cells because they are a standard line for anti-
angiogenic screening, as set by the National Cancer
Institute in the Angiogenesis Resource Center (http://
www.dtp.nci.nih.gov/). Linomide and its analogues
exhibited substabtial anti-proliferative activities with
IC50 values range from 4.0 to 22.26 mM (Table 1).
Except compound 2, there seems to be a direct corre-
lation between the inhibition of CAM blood vessel for-
mation and inhibition of HUVEC cells proliferation
(Table 1).
ture–activity relationships studies of linomide are cur-
rently in progress.
References and Notes
1. Andersen, O.; Lycke, J.; Tollesson, P. O.; Svenningsson,
A.; Runmarker, B.; Linde, A. S.; Astrom, M.; Gjorstrup, P.;
Ekholm, S. Mult. Scler. 1996, 1, 348.
2. Andersen, O.; Lycke, J.; Tollesson, P. O.; Svenningsson,
A.; Runmarker, B.; Linde, A. S.; Astrom, M.; Gjorstrup, P.;
Ekholm, S. Neurology 1996, 47, 895.
3. Karussis, D. M.; Meiner, Z.; Lehmann, D.; Gomori, J. M.;
Schwarz, A.; Linde, A.; Abramsky, O. Neurology 1996, 47, 341.
4. Vukanovic, J.; Passaniti, A.; Hirata, T.; Traystman, R. J.;
Hartley-Asp, B.; Isaacs, J. T. Cancer Res. 1993, 53, 1833.
5. Joseph, I. B.; Vukanovic, J.; Isaacs, J. T. Cancer Res. 1996,
56, 3404.
6. Parenti, A.; Donnini, S.; Morbidelli, L.; Granger, H. J.;
Ziche, M. Br. J. Pharmacol. 1996, 4, 619.
7. Khan, S. R.; Mhaka, A.; Pili, R.; Isaacs, J. T. Bioorg. Med.
Chem. Lett. 2001, 11, 451.
8. Ichikawa, T.; Lamb, J. C.; Christensson, P. I.; Hartley-Asp,
B.; Isaacs, J. T. Cancer Res. 1992, 52, 3022.
In order to ascertain whether linomide and its meta-
bolites are selective for endothelial cells, the effects of
the compared on prostate cancer cells (PC-3) were also
examined. The PC-3 cell line was chosen because lino-
mide has been reported to have in vivo antitumor effect
against human prostatic cancers.8À10 Linomide and its
analogues had no antiproliferative activities against PC-3
prostate cancer cells with IC50 of all the compounds
>100 mM (data not shown).
9. Vukanovic, J.; Isaacs, J. T. Cancer Res. 1995, 55, 1499.
10. Vukanovic, J.; Hartley-Asp, B.; Isaacs, J. T. Prostate
1995, 26, 235.
11. Edgar, E.; Sandberg, E. B.; Stahhandske, J. T. US Patent
4547511, 1982.
12. Suzuki, M.; Kaneko, T.; Kamiyama, H.; Ohuchi, Y.;
Yokomori, S. Heterocycles 2000, 53, 2471.
13. Sheu, J. R.; Fu, C. C.; Tsai, M. L.; Chung, W. J. Anti-
cancer Res. 1998, 18, 4435.
14. Brooks, P. C.; Montgomery, A. M.; Cheresh, D. A.
Methods Mol. Biol. 1999, 129, 257.
15. Marks, M. G.; Shi, J.; Fry, M. O.; Xiao, Z.; Trzyna, M.;
Pokala, V.; Ihnat, M. A.; Li, P. K. Biol. Pharm. Bull. 2002, 25,
597.
In conclusion, this is the first report on the structure–
activity relationship studies of the antiangiogenic activ-
ity of linomide. Several linomide analogues exhibited
higher antiangiogenic activity than linomide. This study
also points to the fact that linomide may not be the true
antiangiogenic moiety in vivo. Compound 4, a known
metabolite of linomide in vivo,16 is 8 times more potent
than linomide at inhibiting blood vessel formation in
the CAM angiogenesis assay. Furthermore, the cellular
target for the linomide analogues appears to be the
endothelial cells lining the blood vessel since these com-
pounds possessed considerable anti-proliferative activity
against isolated HUVEC cells with no activity against
epithelial-derived prostate tumor cells. Detailed struc-
16. Strandgarden, K.; Hoglund, P.; Gronquist, L.; Svensson, L.;
Gunnarsson, P. O. Biopharmaceut. Drug Disposit. 2000, 21, 53.