1192
X. Ouyang et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1191–1196
H
oxadiazole (Scheme 3). Therefore, oxadiazole deriva-
tives could be obtained in two easy steps without purifi-
cation by column chromatography.
O
S
O
N
O
N
O
NH
N
N
H
OMe
N
Compounds of the general formula given in Table 1
were evaluated in two assays. Compounds that showed
inhibition of tubulin polymerization in vitro were evalu-
ated for inducing cell cycle arrest in A431 human cancer
cells (Table 1). To identify an alternative group to 3,5-
dimethoxyaniline, we initially kept R2 constant as either
2,3-dihydro-benzo[1,4]dioxin-6-yl or benzo[1,3]dioxol-5-
yl groups (preferred substitutions at R2 position ob-
tained from the triazole series8) and used various
alkylamines (R1CH2NH2) to determine the effect of R1
on activity (1–21, Table 1). This strategy proved to be
successful as compounds with a variety of groups at
R1 were active. Aryl groups were the most favorable
as R1 substituents: compounds displayed double digit
nanomolar activity in cellular assays when R1 was alk-
oxylphenyl (1–3), fluorophenyl (4–6), methanesulfon-
amidophenyl (7–8), pyridyl (9–14), pyrazinyl (15) or
furanyl (17) moiety. R1 alkyl groups generally demon-
strated a weaker activity than aromatic analogs. For
example, tetrahydrofuran analogs (19) and (20) had
EC50s >150 nM. Compounds with polar groups at R1
(i.e., 21) lost functional and cellular activities. Notewor-
thy, compounds with a 2,3-dihydro-benzo[1,4]dioxin-6-
yl group at R2 are often more potent than closely related
analogs with a benzo[1,3]dioxol-5-yl group. Benzodioxin
analogs (5), (7), (11), (13), (15), and (17) are at least
twice as potent as their corresponding benzodioxole
analogs (6), (8), (12), (14), (16), and (18), indicating that
a benzodioxin group is the preferred substitution at R2.
Cl
OH
D-24851
ABT-751
N
N
N
N
NH
NH
N
N
H
O
N
NH
O
N
NH
O
O
O
10
MeO
OMe
Triazole
Oxadiazole
Figure 1. Structures of D-24851, ABT-751, triazole, and oxadiazole.
and have shown that the presence of the electron-rich
3,5-dimethoxyphenyl moiety is necessary for cellular
potency (Fig. 1).8 In an effort to improve the physico-
chemical properties and pharmacokinetic profile of
these compounds, we designed a series of oxadiazoles
where the 3,5-dimethoxyaniline group is not required
for potent antimitotic activity in cells (i.e., 10 in
Fig. 1). The present report describes the synthesis and
evaluation of oxadiazoles as a novel class of tubulin
inhibitors that are potent and effective against tumor
cells including a cell line with MDR. Structure–activity
relationships (SARs) and pharmacokinetic studies of
this novel tubulin inhibitor class are reported herein.
In contrast, our attempts to find a replacement for the
benzodioxin or benzodioxole groups at the R2 position
were unsuccessful. While keeping R1 constant, various
substituted phenyl groups at R2 were evaluated. Only
a 4-methoxyphenyl analog (22) demonstrated apprecia-
ble activity, whereas 3-methoxyphenyl, 4-methylphenyl,
3-methyl, and 4-hydroxyphenyl analogs were inactive,
indicating potential hydrogen bond that involved the
oxygen atom at the para position. Compounds contain-
ing substitutions at the R2 phenyl ring, including elec-
tron-donating groups such as 3,4-dimethylamino,
amino, and electron-withdrawing groups such as cyano,
difluoro, and nitro, had diminished activity in tubulin
polymerization assays (see supporting information sec-
tion). Difluoro-benzodioxole analog (23) displayed
much weaker activity than the corresponding benzodi-
oxole analog (14).
Oxadiazole compounds in this report were synthesized
according to Method I described by Sunder (Scheme
1).9 Ester A was treated with hydrazine monohydrate
to yield hydrazide B, which was reacted further with
an isothiocyanate to form thiourea intermediate C.
Finally, C was cyclized to produce oxadiazole D by
heating with DCC. Oxadiazole analogs similar to (10)
(Fig. 1) were synthesized by the displacement of chlorine
atom from 2-chloronicotinic acid ethyl ester by a variety
of amines under thermal conditions. As a result, the 2-
alkylaminonicotinic acid ethyl ester E was generated
and then converted via Method I to the oxadiazole ana-
logs listed in Table 1 (Scheme 2).
Compounds listed in Table 1 can also be prepared via a
very efficient protocol utilizing cyclization of commer-
cially available 2-fluoronicotinic acid with a thiosemi-
carbazide provided oxadiazole intermediate F, which is
substituted with an amine of choice to give a final
Biaryl ether analogs such as compound (24, X = O),
synthesized by a modified procedure at step (i) in
O
R2
NH
O
O
N
H
H
N
(iii)
(i)
(ii)
N
N
Ar
N
R2
Ar
OEt
Ar
NHNH2
O
H
Ar
S
A
C
B
D
Scheme 1. Method I. Reagents and conditions: (i) hydrazine monohydrate, isopropanol, 120 °C, 2–4 h, 60–90%; (ii) R2NCS, CHCl3, 30–60 °C,
2–6 h, 80–95% (iii) DCC, toluene, 110 °C, 8–16 h, 60–80%.