Subsequently, other N-thioacetyl-lysine analogs were developed with IC 50
values of 0.24–23 μM for SIRT1 and 1.8–75 μM for SIRT2 with some of those
compounds showing moderate inhibition toward SIRT6 [59, 66, 67]. The most
potent compound of these analogs was not isoform-selective as it inhibited also
SIRT3 (IC 50 ¼ 3.89 μM); it exhibited an antiproliferative effect in lung carcinoma
and breast carcinoma cells [67]. Although N-thioacetyl-lysine analogs lack isoformselectivity, they can be used as a chemical probe to investigate the mechanism of
sirtuin deacetylation reaction or as a starting point in the design of novel smallmolecule inhibitors.
In attempts to improve the isoform-selectivity of N-thioacetyl-lysine analogs,
a set of compounds with various aliphatic acyl groups such as thiobutyryl-,
thioheptanoyl-, and thiomyristoyl-moieties were developed [65]. Interestingly, it
was observed that the length of the aliphatic acyl group exerted an influence on
the isoform-selectivity of sirtuins. The selective inhibitor was thiomyristoyl-lysine
(compound 17 in Fig. 3) with the IC 50 values of 28 nM and 98 μM toward SIRT2 and
SIRT1, respectively, and it did not inhibit SIRT3 at all. Thiomyristoyl analogs,
which mimic a tumor necrosis factor alpha (TNFα) or a histone H3 lysine 9 (H3K9)
peptide sequence, were also potent inhibitors of SIRT6. The most potent inhibitor,
the H3K9 derivative, inhibited both SIRT6 demyristoylation and deacetylation
activity with IC 50 values of 1.7 μM and 8.2 μM, respectively. This compound also
inhibited SIRT6 catalyzed defatty-acylation of TNFα in human embryonic kidney
(HEK) 293 T cells. Thiomyristoyl-lysine had a broad anticancer effect on various
human cancer cells and mouse models of breast cancer [65].
Inhibitors have been developed also for SIRT5; these have been based on the
H3K9 peptide sequence [68]. The 11-residues succinyl-lysine peptide was the
most potent inhibitor for SIRT5 having an IC 50 value of 5 μM. A pentapeptide
with thiosuccinyl-lysine at the middle of sequence seemed to increase the potency
of the inhibition as compared to compounds with the corresponding group at
the C-terminus or N-terminus. Carbamoyl phosphate synthetase peptides (CPS1)
containing succinyl-lysine were demonstrated to inhibit specifically SIRT5; these
displayed the inhibitory constant (K i ) values from 5 to 100 μM [69]. Zang and
coworkers [70] identified a potent thiourea-type linear peptide which contained N
ε -
carboxyethyl-thiocarbamoyl-lysine (called the “SIRT5 inhibitory warhead”). This
compound was selective, but the warhead seemed to result in a metabolically
unstable compound. Later, the cyclic pentapeptide-based counterpart was observed
to be more stable, cell permeable, and selective toward SIRT5 with an IC 50 value of
7.5 μM for SIRT5, but, for other sirtuins, it was a very weak inhibitor [71].
Bivalent sirtuin inhibitors were designed to bind into two binding pockets at the
same time; the N-thioacetyl-lysine part could bind to the peptide substrate binding
site, and the other moiety occupied the pocket of the NAM moiety of NAD
+ . The
most potent bivalent inhibitor exhibited an IC 50 value of 12 μM for SIRT1, but only
weak inhibition toward SIRT2 and SIRT3 (compound 18 in Fig. 3). This bivalent
inhibitor was more isoform-selective than the monovalent counterpart and had
IC 50 values of 39 μM, 22 μM, and 54 μM for SIRT1, SIRT2, and SIRT3,
respectively [72].
Sirtuin Inhibitors and Activators
63
values of 0.24–23 μM for SIRT1 and 1.8–75 μM for SIRT2 with some of those
compounds showing moderate inhibition toward SIRT6 [59, 66, 67]. The most
potent compound of these analogs was not isoform-selective as it inhibited also
SIRT3 (IC 50 ¼ 3.89 μM); it exhibited an antiproliferative effect in lung carcinoma
and breast carcinoma cells [67]. Although N-thioacetyl-lysine analogs lack isoformselectivity, they can be used as a chemical probe to investigate the mechanism of
sirtuin deacetylation reaction or as a starting point in the design of novel smallmolecule inhibitors.
In attempts to improve the isoform-selectivity of N-thioacetyl-lysine analogs,
a set of compounds with various aliphatic acyl groups such as thiobutyryl-,
thioheptanoyl-, and thiomyristoyl-moieties were developed [65]. Interestingly, it
was observed that the length of the aliphatic acyl group exerted an influence on
the isoform-selectivity of sirtuins. The selective inhibitor was thiomyristoyl-lysine
(compound 17 in Fig. 3) with the IC 50 values of 28 nM and 98 μM toward SIRT2 and
SIRT1, respectively, and it did not inhibit SIRT3 at all. Thiomyristoyl analogs,
which mimic a tumor necrosis factor alpha (TNFα) or a histone H3 lysine 9 (H3K9)
peptide sequence, were also potent inhibitors of SIRT6. The most potent inhibitor,
the H3K9 derivative, inhibited both SIRT6 demyristoylation and deacetylation
activity with IC 50 values of 1.7 μM and 8.2 μM, respectively. This compound also
inhibited SIRT6 catalyzed defatty-acylation of TNFα in human embryonic kidney
(HEK) 293 T cells. Thiomyristoyl-lysine had a broad anticancer effect on various
human cancer cells and mouse models of breast cancer [65].
Inhibitors have been developed also for SIRT5; these have been based on the
H3K9 peptide sequence [68]. The 11-residues succinyl-lysine peptide was the
most potent inhibitor for SIRT5 having an IC 50 value of 5 μM. A pentapeptide
with thiosuccinyl-lysine at the middle of sequence seemed to increase the potency
of the inhibition as compared to compounds with the corresponding group at
the C-terminus or N-terminus. Carbamoyl phosphate synthetase peptides (CPS1)
containing succinyl-lysine were demonstrated to inhibit specifically SIRT5; these
displayed the inhibitory constant (K i ) values from 5 to 100 μM [69]. Zang and
coworkers [70] identified a potent thiourea-type linear peptide which contained N
ε -
carboxyethyl-thiocarbamoyl-lysine (called the “SIRT5 inhibitory warhead”). This
compound was selective, but the warhead seemed to result in a metabolically
unstable compound. Later, the cyclic pentapeptide-based counterpart was observed
to be more stable, cell permeable, and selective toward SIRT5 with an IC 50 value of
7.5 μM for SIRT5, but, for other sirtuins, it was a very weak inhibitor [71].
Bivalent sirtuin inhibitors were designed to bind into two binding pockets at the
same time; the N-thioacetyl-lysine part could bind to the peptide substrate binding
site, and the other moiety occupied the pocket of the NAM moiety of NAD
+ . The
most potent bivalent inhibitor exhibited an IC 50 value of 12 μM for SIRT1, but only
weak inhibition toward SIRT2 and SIRT3 (compound 18 in Fig. 3). This bivalent
inhibitor was more isoform-selective than the monovalent counterpart and had
IC 50 values of 39 μM, 22 μM, and 54 μM for SIRT1, SIRT2, and SIRT3,
respectively [72].
Sirtuin Inhibitors and Activators
63