Y. Matsuya et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
O
go ahead with further SAR studies focusing on the meta-amide-
substituted structure.
NO2
Ar
NH
N
S
The chloro-compound 21, precursor of the compound 17, was
subjected to nucleophilic substitution reaction with three second-
ary amines (piperidine, morpholine, and pyrrolidine), and the
products were converted to meta-quinoxaloyl-amide derivatives
22–24, respectively, through meta-nitro group manipulations
(Scheme 4). Modifications of aryl groups on the amide-substituents
were also performed, starting from the piperazine derivative 17
with two steps, to furnish five derivatives 25–29 in satisfactory
yields (Scheme 5). Thus, eight new SRT1720 derivatives possessing
meta-amide-substituents were obtained.
Survey of the meta-amide-substituted derivatives 22–29 on
SIRT1 activation was performed using the same in vitro assay
method, and revealed that 22–24 did not exhibit any activities
(data not shown), suggesting the piperazinyl moiety would be
important for the activity. Concerning variation of the hetero-aro-
matic ring, 2-quinolinyl and 3-quinolinyl derivatives (25 and 26,
a, b
N
S
N
N
Boc
N
N
17
HN
N
25 (Ar = 2-quinolinyl), 26 (Ar = 3-quinolinyl)
27 (Ar = 4-quinolinyl), 28 (Ar = 2-pyrazinyl)
29 (Ar = 2-indolyl)
Scheme 5. Reagents and conditions: (a) NaSH, MeOH/H2O, reflux, 18 h; (b) ArCO2H,
EDC, DMAP, CH2Cl2, rt, 1 h; then TFA, CH2Cl2, rt, 1 h (25: 80%, 26: 82%, 27: 61%, 28:
81%, 29: 99%, in two-steps, respectively).
respectively) exerted ca. 140–160% SIRT1 activation at 10 lM con-
centration, which was comparable to SRT1720 (3) and 2-quinoxali-
nyl derivative (19), and the other derivatives 27–29 were found to
be inactive (Fig. 3).
Concentration dependency was examined for the active deriva-
tives 19, 25, and 26, as well as the reference compound SRT1720
(3). As summarized in Figure 4, all of the three new derivatives
showed SIRT1 activation activity more potent than 3 especially at
100 lM. Thus, modified SRT1720 compounds with meta-amide-
substituents were considered as promising candidates for a novel
SIRT1 activator.
Figure 3. SIRT1 activation assay for the m-substituted derivatives 25–29, using
SIRT1 fluorometric drug discovery kit (AK-555, BIOMOL Research Laboratories). C
represents control experiment. For all experiments, compound concentrations were
To explore the origin of the higher activity of the meta deriva-
tives than that of SRT1720 (3), we studied the binding mode of
25 and SRT1720 with SIRT1 by using Molegro Virtual Docker 5 soft-
10 lM. The SD bars for small value (below 2%) are omitted for clarity.
ware.9 Wu et al. have reported
p/p interaction between an aro-
matic ring of activators and a coumarin ring of labeled substrate
plays an important role for SIRT1 activation.10 The two binding
models in this Letter also indicate that the quinoxaline ring of
SRT1720 and the quinoline ring of 25 can interact with the couma-
rin ring via p/p interaction (Fig. 5). In the simulated SRT1720/pep-
tide substrate/SIRT1 complex, only one hydrogen bond is observed
between the amide hydrogen atom of SRT1720 and the carbonyl
oxygen atom of coumarin ring (Fig. 5A). On the other hand, two
hydrogen bonds are observed in the simulated 25/peptide sub-
strate/SIRT1 complex, where there can be a hydrogen bond be-
tween the piperazine of 25 and the side chain of Gln 294 in
SIRT1, in addition to the possibility of a hydrogen bond between
the amide oxygen and the NH of Lys in the peptide substrate
(Fig. 5B). These calculation results suggest that the two hydrogen
bonds in the 25/peptide substrate/SIRT1 complex may be attrib-
uted to its potent activity as compared with SRT1720.
In this Letter, we described syntheses and in vitro evaluation of
new SIRT1 activator candidates based on SRT1720 possessing the
imidazo[1,2-b]thiazole core, as a lead compound. Easily modifiable
moieties of the SRT1720 structure, namely, piperazine moiety,
quinoxaline ring on the amide group, and position of the amide
function, were changed, and the assay results indicated that piper-
azine ring and hetero-bicyclic ring on the amide would be impor-
tant for SIRT1 activation ability. Novel finding in this Letter is that
meta-substitution pattern of the amide function increases the
activity compared with ortho-substituted SRT1720, which can be
rationalized by docking simulation of drug/peptide substrate/SIRT1
complex. In the docking study, compound 25 was docked into the
catalytic site of SIRT1, however, it has recently been reported that
SIRT1 activators may bind to an allosteric site of the enzyme.11 Fur-
ther studies including SIRT1 activation mechanism analysis of our
activators will be continued in our laboratory.
Figure 4. SIRT1 activation of compounds 3 (Â), 19 (d), 25 (N), and 26 (j) at 1, 10,
and 100 M.
l
With these new SRT1720 derivatives in hand, we screened their
SIRT1 activation activities using a fluorometric in vitro assay meth-
od, and the results are summarized in Figure 2. At 10 lM concen-
tration, the reference compound SRT1720 (3) exhibited ca. 160%
activation. On the other hand, the other derivatives (6–8 and 12–
14), all of which have ortho-amide-substituents, showed almost
no activity at the same concentration. Gratifyingly, we could find
that the meta-amide-substituted derivative 19 activated SIRT1
with comparable potency to SRT1720, whereas para-amide-substi-
tuted derivative 20 had no activity. These results prompted us to
Matsuya, Yuji
