H. Shih et al. / Bioorg. Med. Chem. Lett. 10 (2000) 487±490
489
Consistent to the cytotoxicity pattern on substituted 1-
tetralone analogues, the 1-indanones, 22±36 showed
slight to remarkable cytotoxicity, especially 31 (IC50 80
nM), 35 (IC50 55 nM) and 36 (Indanorine, IC50 6 nM).
36 is more potent than 35 because the 40-hydroxy group
provides a much stronger binding anity than the 40-
amino to the re®ned region of binding protein. The far
less active derivatives of 36, 37±40, suggest that the
binding protein has a small and restricted region corre-
sponding to the binding of 30,50-dilipophilic-40-hydro-
philic substituents on (E)-2-benzylidene moiety. To
further examine the structure±activity relationships
around the benzene ring of 1-indanone, we found that
the large lipophilic 7-nitro group of 41 might cause an
incorrect geometry to disfavor the binding to the
pocket and resulted in no cytotoxicity. So did the
nontoxic 43. However, 42 expressed remarkable cyto-
toxicity (IC50 1 nM), more potent than 36 (IC50 6 nM)
because the 7-hydrophilic amino group provides an
extra docking to the binding protein. Replacing the 40-
hydroxy of 42 with an amino group oered a less active
44 (IC50 23.5 nM), as expected. Replacing the 7-amino
of 42 with a 7-methoxy gave a less active 45 (IC50 125
nM) because the methoxy is a nonpolar but weak
hydrogen-bond acceptor. This result further provides
the evidence that there is a hydrogen binding site in the
binding protein around the 7 position and 1-carbonyl of
indanone. Demethylation of 36 at the 5 position yielded
a less active 46 (IC50 530 nM) because the 5-hydrophilic
hydroxy substituent not only forms hydrogen bonding
with its neighboring methoxy group but also interferes
with the binding anity of the molecule to the binding
site. Introduction of a nitro group at the 4 position of 46
gave a far less cytotoxic 47 (IC50 9.3 mM). Reduction of
the nitro group of 47 resulted in a marginally cytotoxic
48 (IC50 30 mM). The weak cytotoxic eects of 47 and
48 suggest that there is no binding site in the binding
protein for the 4-substituent of 1-indanone. Instead, the
4-substituent might squeeze or form hydrogen binding
with the neighboring 5-hydroxy to intervene in the
binding anity with the protein. The lack of cytotoxi-
city for monomethoxy compounds, 49 and 50, suggests
that the 5- or 6-substituent of indanone cannot position
correct geometry and loses the binding anity to the
hydrophobic region of the pocket. Compound 51 with
lipophilic 5,6-dioxymethylene substituents expressed no
cytotoxicity because the lipophilic and planar dioxy-
methylene ring cannot dock to the binding site. Com-
pared to the potent 30 with 5,6-dimethoxy substituents,
the inactive 52 with 5,6-dihydroxy substituents provides
further evidence that the hydrogen-bond interaction
between 5,6-dihydroxy substituted 1-indanone causes the
molecule to lose the binding anity to the protein. In
these active compounds, their 5,6-dimethoxy substitu-
ents are lipophilic, repulsive and out of the indanone
plane. This information indicates that there are steric
features in the binding protein around the 5±7 positions
of indanone for binding anity to render cytotoxicity.
Furthermore, these results suggest that in addition to
hydrophobic and hydrogen binding sites for 30,50-dili-
pophilic and 40-hydrophilic substituents on (E)-2-benzyl-
idene moiety, the binding pocket crucially contains a
hydrophobic region around the 5 and 6 positions and a
Figure 1. Proposed pharmacophore of indanocine interacting with an
unidenti®ed binding pocket.
hydrogen binding site around the 7 position and 1-car-
bonyl of (E)-2-benzylidene-1-indanone for binding
anity. Since these active compounds share similar
conformations and properties to render cytotoxicity,
they contain the complementary patterns of ligand
atoms, or pharmacophore to form ligand±receptor
complex. Thus, indanocine 42 possibly ®ts the pocket
very well and demonstrates remarkable cytotoxicity
(Fig. 1). We will take advantage of this model to explore
more antitumor agents.
References and Notes
1. Inoue, A.; Kawai, T.; Wakita, M.; Iimura, Y.; Sugimoto,
H.; Kawakami, Y. J. Med. Chem. 1996, 39, 4460.
2. Cardozo, M. G.; Imura, Y.; Sugimoto, H.; Yamanishi, Y.;
Hop®nger, A. J. J. Med. Chem. 1992, 35, 584.
3. Geldmacher, D. S. Comprehensive Therapy 1997, 23, 492.
4. Abe, A.; Radin, N. S.; Shayman, J. A.; Wotring, L. L.;
Zipkin, R. E.; Sivakumar, B.; Ruggieri, J. M.; Carson, K. G.;
Ganem, B. J. Lip. Res. 1995, 36, 611.
5. Bielawska, A.; Linardic, C. M.; Hannun, Y. A. FEBS 1992,
307, 211.
6. Leoni, L.; Shih, H.; Deng, L.; Carson, D.; Cottam, H. Bio-
chem. Pharmacol. 1998, 55, 1105.
7. Dimmock, J. R.; Kandepu, N. M.; Nazarali, A. J.; Kowal-
chuk, T. R.; Motaganahalli, N.; Quail, J. W.; Mykytiuk, P. A.;
Audette, G. F.; Prasad, L.; Perjesi, P.; Allen, T. M.; Santos, C.
L.; Szydlowski, J.; Clercq, E. D.; Balzarini, J. J. Med. Chem.
1999, 42, 1358.
8. Mp 136±137 ꢀC; ms 347 (MNa+); 1H NMR (DMSO-d6);
2.21 (s, 6H, CH3), 3.81 (s, 3H, OCH3), 3.88 (s, 3H, OCH3),
3.94 (s 2H, CH2), 7.19 (s, 1H, C4H), 7.24 (s, 1H, CH), 7.28 (s,
1H, C7H), 7.34 (s, 2H, C02H and C06H). Anal. for C20H20O4
C, 74.06; H, 6.21, found C, 73.70; H, 6.40.
9. Mp 238±240 ꢀC; ms 340 (MH+); 1H NMR (DMSO-d6);
2.20 (s, 6H, CH3), 3.63 (s, 3H, OCH3), 3.83 (s, 2H, CH2), 3.85
(s, 3H, OCH3), 6.30 (s, 2H, NH2), 6.50 (s, 1H, C4H), 7.15
(s, 1H, CH), 7.29 (s, 2H, C02H and C06H). Anal. for
.
C20H21NO4 0.2CH3OH C, 70.24; H, 6.35; N, 4.05, found C,
70.24; H, 5.97; N, 3.99.
10. Bayer, H.; Batzl, C.; Hartmann, R. W.; Mannschreck, A.
J. Med. Chem. 1991, 34, 2685.
11. Mp 147±148 ꢀC; ms, 238 (MH+); H NMR (DMSO-d6);
1
2.66 (t, 2H, C3H2), 3.08 (t, 2H, C2H2), 3.80 (s, 3H, OCH3),
3.98 (s, 3H, OCH3), 7.46 (s, 1H, C4H). Anal. for C11H11NO4
C, 55.70; H, 4.67; N, 5.90, found C, 55.62; H, 4.52; N, 5.93.