5404
H. Chen et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5401–5405
cell growth (5–9lM).Substitution of the 5-position with
larger lipophilic groups such as t-butyl (2f), phenyl (2g),
and p-t-butylphenyl (2h) resulted in compounds that
were unable to release Ca2+ from intracellular stores,
via the exocyclic methylidene linkage has biological
advantages over the less rigidified system containing a
5-benzyl substitution.
while they still remained potent inducers of eIF2a phos-
In summary, following our original observation that
TRO acts as Ca2+ depleting translation initiation inhib-
itor, we have identified compound 1 with structural fea-
tures and biological profile similar to TRO.We carried
out a limited SAR study, and identified some of the
structural features contributing to increase of potency.
Additionally, several 5-benzylidene-thiazolidine-2,4-
diones (3c, 3d, and 3f), and -thiones (2b, 2e, and 2j) show
intracellular Ca2+ release, strong phosphorylation of
eIF2a, and low lM GI50.These candidates will be used
as leads for further optimization as Ca2+ depleting
translation initiation inhibitors and putative anticancer
agents.
2+
phorylation.The dissociation between Ca
depletion
activity and phosphorylation of eIF2a may suggest an
alternative Ca2+-independent mechanism leading to the
phosphorylation of eIF2a.The different time courses
in which the two assays are conducted (minutes for the
Ca2+ release and P 2h for the eIF2a phosphorylation
assay) could also explain this discrepancy.For example,
compounds with slow Ca2+ release kinetics may still de-
plete the intracellular stores and induce phosphorylation
of eIF2a without causing a detectable change in net
cytosolic Ca2+.Interestingly, while strong electronega-
tive substituents such as nitro (2j) retained the activities
in all assays, the electron-donating substituents (meth-
oxy 2f, piperidine 2k, and morpholine 2l) caused loss
of activity in both Ca2+ depletion and phosphorylation
of eIF2a assays.The overall trend in the 5-benzylid-
ene-thiazolidine-2,4-thione series (2b–e and 2j) pointed
to halogens, nitro and methyl substituents at the 50 posi-
tion to have comparable and favorable biological pro-
files as Ca2+ depleting inhibitors of translation
initiation.
Acknowledgements
These studies were supported in part by NIH National
Cooperative Drug Discovery Group (NCDDG) grant
S U19 CA87427 and NIH CA78411.The authors wish
to thank Drs.M.Tosteson and D.C.Tosteson for con-
tinuous support and Dr.Michael Chorev for comments
and discussion.
There is some overlap in SAR between the thiazolidine-
2,4-thiones 2 described above and the thiazolidine-2,4-
0
diones 3.Substitution at the 5 -position with halogens
References and notes
and methyl (3b–e) results in fully bioactive compounds
with a biological profile similar to that of the corre-
sponding thiazolidine-2,4-thione analogs (2b–e) and
substitution with phenyl, p-t-butyl phenyl and methoxy
(3g–i) results in compromised biological activities simi-
lar to the effect of these substituents in the thiazol-
idine-2,4-thione series (2g–i).
1.(a) Clemens, MJ.;.Bommer, U.
Int. J. Biochem. Cell Biol.
1999, 31, 1–23; (b) Pain, V.M. Eur. J. Biochem. 1996, 236,
747–771; (c) Kozak, M. J. Cell Biol. 1991, 115, 887–903;
(d) Kozak, M. J. Biol. Chem. 1991, 266, 19867–19874.
2. (a) Dua, K.; Williams, T. M.; Beretta, L. Proteomics 2001,
1, 1191–1199; (b) Sonenberg, N. Curr. Opin. Cell Biol.
1993, 955–960.
3.(a) Meric, F;.Hunt, KK. .
Mol. Cancer Ther. 2002, 1, 971–
The impact on biological activities of the 50-NO2 (2j and
3j) and 50-t-butyl (2f and 3f) in the two series is quite
reversed.While 3f, the t-butyl containing analog, dis-
plays good activities in all three assays, with the strong-
est eIF2a phosphorylation activity, 2f does not release
Ca2+ from intracellular stores.On the other hand, while
3j, the NO2 containing analog, is devoid of Ca2+ release
and eIF2a phosphorylation activities the corresponding
2j is fully active in all three assays.
982; (b) Garrett, M.D.; Workman, P. Eur. J. Cancer 1999,
35, 2010–2030.
4. Palakurthi, S. S.; Aktas, H.; Grubissich, L. M.; Morten-
sen, R.M;. Halperin, J.A. Cancer Res. 2001, 61, 6213.
5. Kawasaki, J.; Hirano, K.; Hirano, M.; Nishimura, J.;
Fujishima, M.; Kanaide, H. Eur. J. Pharmacol. 1999, 373,
111–120.
6. (a) Aktas, H.; Fluckiger, R.; Acosta, J. A.; Savage, J. M.;
Proc. Natl. Acad. Sci.
Palakurthi, S.S;. Halperin, J.A.
U.S.A. 1998, 95, 8280; (b) Fan, Y.-H.; Chen, H.; Natara-
jan, A.; Guo, Y.; Harbinski, F.; Iyasere, J.; Christ, W.;
Aktas, H.; Halperin, J. A. Bioorg. Med. Chem. Lett. 2004,
14, 2547; (c) Brostrom, C. O.; Prostko, C. R.; Kaufman,
R.J;. Brostrom, M.A. J. Biol. Chem. 1996, 271, 24995–
25002.
The significance of the exocyclic olefinic bond at the 5-
position was addressed by comparing the most potent
compound (3f) in the 5-benzylidene-thiazolidine-2,4-
dione series and lead compound (1) in the 5-benzylid-
ene-thiazolidine-2,4-thione series with the homologous
5-benzyl analog 4f and 4c in which the exocyclic olefinic
bonds were reduced.In this respect, 4f is a mimic of
TRO.Evidently, while being as potent as 3f in the induc-
tion of Ca2+ release, 4f is inactive in the phosphorylation
of eIF2a and sixfold less potent than 3f in inhibiting cell
growth.On the other hand, 4c has completely lost Ca2+
depletion activity and is an order of magnitude less po-
tent in the growth inhibition assay, GI50 = 2 and 29lM,
respectively.We conclude that unlike TRO, the conju-
gation of the thiazolidine-2,4-dione to the phenyl ring
7. (a) Sudhakar, A.; Ramachandran, A.; Ghosh, S.; Hasnain,
S. E.; Kaufman, R. J.; Ramaiah, K. V. A. Biochemistry
2000, 39, 12929–12938; (b) Prostko, C.R;. Dholakia, J.
N.; Brostrom, M. A.; Brostrom, C. O. J. Biol. Chem. 1995,
270, 6211.
8. Lohray, B. B.; Bhushan, V.; Reddy, A. S.; Rao, P. B.;
Reddy, N. J.; Harikishore, P.; Haritha, N.; Vikramadit-
yan, R. K.; Chakrabarti, R.; Rajagopalan, R.; Katneni, K.
J. Med. Chem. 1999, 42, 2569.
9. (a) Giles, R. G.; Norman, J. L.; Quick, J. K.; Sasse, M. J.;
Urrquhart, M.W.J;. Youssef, L. Tetrahedron 2000, 56,
4531–4537; (b) Bruno, G.; Costantino, L.; Curinga, C.;