1872
P. R. Tatum et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1871–1874
NH2
O
NH2
O
NH2
O
NH2
I
O
I
NH
NH
NH
NH
b
a
m
O
m
O
p
O
p
m
N
HN
O
H3CO
O
5
7
HO
CH3
p
SIRT2 IC50 > 100 µM
SIRT1 IC50 56 µM
SIRT2 IC50
SIRT1 IC50 > 300 µM
=
1.0 µM
9a
9b
N
8a
8b
: meta
: para
: meta
: para
=
H3CO
CH3
10a
10b
: meta
: para
6
SIRT2 IC50
=
17 µM
NH2
NH2
SIRT1 IC50 > 100 µM
O
O
Figure 2. SIRT2-selective inhibitors discovered by our group.
c
NH
d
NH
m
m
O
p
p
H
NH2
A1
A2
: meta
: para
NH2
11a
11b
: meta
: para
O
NH
O
N
Scheme 1. Reagents and conditions: (a) (i) SOCl2, CH2Cl2, reflux, (ii) N,O-dim-
ethylhydroxylamine hydrochloride, Et3N, CH2Cl2, rt, 91% for 9a, 83% for 9b; (b) 2-
aminobenzamide, Pd2dba3, 2-dicyclohexylphosphino-20,40,60-triisopropyl biphenyl,
K2CO3, tert-BuOH, reflux, 15% for 10a, 81% for 10b; (c) LiAlH4, THF, 0 °C, 45% for 11a,
q. y. for 11b; (d) dimethyl 1-diazoacetonylphosphonate, MeOH, rt, 20% for A1, 22%
for A2.
NH
N
N
R
+
m
N
N
Azides (B)
m
p
p
N
R
Alkynes (A)
SIRT2-Selective
Inhibitor Candidates (AB)
and azides (10
lM) used for this study and a mixture of CuSO4
Figure 3. Design of triazole-containing SIRT2-selective inhibitors.
(2 lM), sodium ascorbate (10
l
M) and TBTA (2 M) were totally
l
inactive. As shown in Figure 5, two hits emerged from the screen.
Crude compounds A1B11 and A2B57 showed SIRT2-inhibitory
activity comparable to AGK2 (1). Furthermore, at the concentration
of 10 lM, crude A1B11 and A2B57 were totally inactive against
SIRT1 and SIRT3 (data not shown), suggesting the SIRT2 selectivity
17
a representative reaction in click chemistry.18 Following these
studies, we performed a further click chemistry approach, seeking
to find novel SIRT2-selective inhibitors. We describe here the rapid
identification of novel SIRT2-selective inhibitors via the use of click
chemistry to generate a library of meta- and para-substituted 2-
anilinobenzamides as SIRT inhibitor candidates.
Based on the previous SAR studies of anilinobenzamide deriva-
tives,14 we designed a library of candidates as SIRT inhibitors bear-
ing a triazole group at the meta or para position (Fig. 3).
Accordingly, we prepared two alkynes A1 and A2 as shown in
Scheme 1. Treatment of iodobenzoic acids 8 with thionyl chloride
afforded the corresponding acid chlorides which were converted to
Weinreb amides 9 by the reaction with N,O-dimethylhydroxyl-
amine. Buchwald–Hartwig reaction between 2-aminobenzamide
of A1B11 and A2B57.
Because A1B11 and A2B57 used in the primary screening were
crude compounds, compounds A1B11 and A2B57 were resynthe-
sized and purified. Scheme 2 illustrates the resynthesis of triazoles
A1B11 and A2B57. Cu(I)-catalyzed coupling of alkyne A1 with B11
and A2 with B57 afforded tirazoles A1B11 and A2B57, respectively.
The resynthesized compounds A1B11 and A2B57 were purified by
column chromatography and recrystallization.22
The pure A1B11 and A2B57 were then examined for inhibitory
effects on SIRT2, SIRT1 and SIRT3.21 The results of the enzyme as-
says are shown in Table 1. In these assays, compounds A1B11 and
A2B57 showed potent SIRT2-inhibitory activity with IC50s of 5.3
and compounds
9
using 2-dicyclohexylphosphino-20,40,60-tri-
isopropylbiphenyl as a ligand gave diphenylamine compounds
10. Compounds 10 were allowed to react with lithium aluminum
hydride at 0 °C to give aldehydes 11 and subsequent Seyferth–Gil-
bert homologation using dimethyl 1-diazoacetonylphosphonate
yielded the desired alkynes A1 and A2.
and 6.3 lM, respectively. The SIRT2-inhibitory activity of A1B11
and A2B57 was comparable to that of AGK2 (1). Furthermore,
while AGK2 (1) inhibited SIRT1 and SIRT3 with IC50s of 30 and
91
lM, respectively, A1B11 and A2B57 did not inhibit either SIRT1
Next, the 114-member 2-anilinobenzamide library was assem-
bled using click chemistry in microtiter plates. Each of the two al-
kynes A (1 equiv) was mixed with each of the 57 azides B (Fig. 4)
(1.2 equiv), which were previously prepared by us,16,17,19 in the
presence of CuSO4 (0.2 equiv), sodium ascorbate (1 equiv), and
or SIRT3 at concentrations up to 100
l
M, showing high selectivity
for SIRT2 over both SIRT1 and SIRT3. Thus, A1B11 and A2B57 are
more selective SIRT2 inhibitors than AGK2 (1) in these enzyme
assays.
In summary, we have rapidly identified novel SIRT2-selective
inhibitors by using click chemistry to generate libraries of candi-
date molecules. The advantage of this work over others is the rap-
idness of identifying novel SIRT2 inhibitors. It should be possible to
obtain even more potent and selective SIRT2 inhibitors by means of
further structural development. SIRT2-selective inhibitors are
thought to have considerable potential both for the development
of novel therapeutic agents and as tools for biological research. De-
tailed studies of the SIRT2-selective inhibitors and their analogues
are under way.
tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl]amine
(TBTA)
(0.2 equiv) in a solvent mixture of DMSO/H2O (1:1).20 The disap-
pearance of the starting materials and the quantitative formation
of the triazole products were confirmed by TLC and LC–MS (HPLC
purity of the resulting triazole and remaining azide: >95%).
These 114 triazole compounds could be screened for SIRT inhib-
itory activity without further purification. To find SIRT2-selective
inhibitors, compounds A1B1ꢀA2B57 were initially tested for activ-
ity against SIRT2 at 10 l lM)
M.21 In this SIRT2 assay, alkynes (10