2
T. Zhou et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
inhibition of b5 subunit could achieve therapeutic effects.25–31
However, co-inhibition of proteasome b1 or b2 subunit have
attracted much attention, since inhibition of b1 or b2 subunit is also
important in some cancer cell lines.32 So far the effect of inhibition
of b1 or b2 subunit has not been fully elucidated mainly due to the
lack of specific inhibitor. Therefore, exploration of selective protea-
some b1 or b2 inhibitors may provide us more information to study
the mechanism and further develop novel drug candidates, as well
as avoiding side effects. Herein we report the copper Schiff base
complexes as selective proteasome b2 inhibitors with valuable
structure–activity relationship.
The Schiff base contains a C@N double bond as a functional
group prepared by condensation of an aldehyde or ketone with a
primary amine. As ligand, Schiff base coordinates metals in various
states by the lone pair of the nitrogen atom of the C@N moiety and
other electron-rich functional groups.33 In the previous Letters,
various copper Schiff base complexes have been designed with dif-
ferent substituted groups. The synthetic method has also been
explored and optimized.34–36 In this work, we hypothesized that
modification of the Schiff base ligand would change the docking
mode of the inhibitor in the active site of proteasome. As shown
in Figure 1, various diamines (a) and salicylaldehydes (s) were
employed to construct tridentate and tetradentate Schiff base
ligands by typical methods.
Table 1
Schiff base ligands as proteasome inhibitorsa
b
Ligand
IC50
NIc
(l
M)
Ligand
IC50 (lM)
a1s1
a4s4
b1 = 1.26 0.02
b2 = 1.19 0.01
b5 = 1.29 0.08
b1 = 1.65 0.06
b2 = 1.71 0.08
b5 = 1.82 0.18
b1 = 9.38 0.27
b2 = 12.47 0.29
b5 = 10.24 0.30
NI
a2(s3)1/2
a2s3
NI
NI
a4s5
a5s3
a6s1
a3s2
b1 = 7.35 0.16
b2 = 8.20 0.18
b5 = 8.02 0.14
NI
b1 = 9.52 0.27
b2 = 12.32 0.28
b5 = 10.05 0.29
NI
a3s3
a3s5
a6s3
a7s1
NI
NI
a4s1
a8s3
a9s3
NI
NI
a4s3
NI
Bortezomibd
4.2 0.34
a
Values represent the mean SD of three independent experiments, each based
on four biological replicates.
b
Percent inhibition at 25
NI: no inhibition.
Positive control.
l .
g mLÀ1
c
d
Before assessment of the copper Schiff base complexes, we pre-
pared various Schiff base ligands and measured their activity
toward proteasome inhibition (Table 1). Different diamines and
salicylaldehydes were chosen to construct a library for further
study of structure–activity relationships (Scheme 1). It is revealed
that the synthesized Schiff base ligands showed no selectivity
toward all three b subunits though some have inhibitory activities.
To increase the water solubility of the ligands, sulfonate groups
were introduced to the salicylaldehyde moiety (s3). However, all
the Schiff base ligands containing s3 showed no inhibitory activity
substituted diamine a4 to build Schiff base ligands with disubsti-
tuted 3,5-dichloro and 3,5-dibromo salicylaldehydes (s4, s5). To
our delight, a4s4, a4s5 showed obvious improvement of protea-
some inhibitory activity around 1 lM despite a4s3 was incompe-
tent. Bulky diamine moiety binaphthyl diamine a9 was used to
replace a4, but the activity disappeared. Furthermore, when the
salicylaldehyde moiety s5 was fixed, a3s5 was prepared to com-
pare with a4s5. Although both of them have activity, a3s5 showed
higher IC50 approximately 6 times to a4s5. Therefore, both the dia-
mine and the salicylaldehyde are crucial, and precise collaboration
of these two moieties is essential for the rational design of metal
Schiff base complexes as proteasome inhibitors.
Then we synthesized series of copper(II) and platinum(II) com-
plexes in different solvents with modest to good yields (Scheme 1).
To confirm the accurate structure of the copper(II) complex, proper
diffusion of ethanol into the aqueous solution of Na2[Cu(a2s3)]
afforded brownish red crystals. Before this, crystal structure of
Na2[Cu(a2s3)] with half-hydrolyzed C@N bond has been reported
by Rogez and co-workers.37 In this work, we obtained the intact
single crystal of Na2[Cu(a2s3)] which crystallizes in the triclinic
P1 space group. The X-ray diffraction reveals a complete Schiff base
ligand in the space lattice consists of NNOO tetradentate coordina-
tion with a central copper ion structure. It is also clear to see the
two deprotonated sulfonate groups on the salicylaldehydes bond-
ing sodium ions (Fig. 2).
except a5s3 which exhibited IC50 10 lM for all b subunits. a2s3,
a3s3, a4s3, a6s3, a8s3, a9s3 and the tridentate ligand a2(s3)1/2
showed negligible activities. Moreover, tetradentate ligands
including a1s1, a4s1, a6s1, a7s1 showed no activities neither.
To identify the diamine moiety, we used the 5-carboxylic acid
Although complexation of copper ion with Schiff base ligand
might change the structure significantly, the primary results of
proteasome inhibition of the Schiff base ligands are still valuable.
We considered that complexation of copper ion with Schiff base
ligand might change the activity by alteration of the molecular
configuration and further the bonding mode in the active site of
proteasome. Before the evaluation of copper(II) complexes, we
examined the fluorescence spectrum of our compounds including
all the Schiff base ligands and copper(II) complexes in the presence
of the reaction buffer, all of them showed negligible fluorescence
intensity under the test condition of proteasome inhibition exper-
iment. Furthermore, we tested the fluorescence of Schiff base
ligand with addition of magnesium ion in HEPEs buffer but no
obvious change of fluorescence spectrum has been observed
which indicated that magnesium ion has no influence to our
Figure 1. Different diamines and salicylaldehydes used in this work.