Substrates for efficient fluorometric screening employing the NAD-dependent sirtuin 5 lysine deacylase (KDAC) enzyme

Article abstract of DOI:10.1021/jm300526r

The class III lysine deacylases (KDACs), also known as the sirtuins, have emerged as interesting drug targets for therapeutic intervention in a variety of diseases. To gain a deeper understanding of the processes affected by sirtuins, the development of selective small molecule modulators of individual isozymes has been a longstanding goal. Essential for the discovery of novel modulators, however, are good screening protocols and mechanistic insights with regard to the targets in question. We therefore evaluated the activities of the seven human sirtuin hydrolases against a panel of fluorogenic substrates. Both commonly used, commercially available substrates and novel chemotypes designed to address recent developments in the field of lysine post-translational modification were evaluated. Our investigations led to the discovery of two new fluorogenic ε-N-succinyllysine-containing substrates that enable highly efficient and enzyme-economical screening employing sirtuin 5 (SIRT5). Furthermore, optimized protocols for facile kinetic investigations were developed, which should be valuable for enzyme kinetic investigations. Finally, these protocols were applied to a kinetic analysis of the inhibition of SIRT5 by suramin, a potent sirtuin inhibitor previously shown by X-ray crystallography to bind the substrate pocket of the human SIRT5 KDAC enzyme.

Full text of DOI:10.1021/jm300526r

Article
pubs.acs.org/jmc
Substrates for Efficient Fluorometric Screening Employing the NAD-
Dependent Sirtuin 5 Lysine Deacylase (KDAC) Enzyme
Andreas S. Madsen and Christian A. Olsen*
Department of Chemistry, Technical University of Denmark, Kemitorvet 207, DK-2800, Kongens Lyngby, Denmark
S
* Supporting Information
ABSTRACT: The class III lysine deacylases (KDACs), also
known as the sirtuins, have emerged as interesting drug targets
for therapeutic intervention in a variety of diseases. To gain a
deeper understanding of the processes affected by sirtuins, the
development of selective small molecule modulators of
individual isozymes has been a longstanding goal. Essential
for the discovery of novel modulators, however, are good
screening protocols and mechanistic insights with regard to the
targets in question. We therefore evaluated the activities of the seven human sirtuin hydrolases against a panel of fluorogenic
substrates. Both commonly used, commercially available substrates and novel chemotypes designed to address recent
developments in the field of lysine post-translational modification were evaluated. Our investigations led to the discovery of two
new fluorogenic ε-N-succinyllysine-containing substrates that enable highly efficient and enzyme-economical screening
employing sirtuin 5 (SIRT5). Furthermore, optimized protocols for facile kinetic investigations were developed, which should be
valuable for enzyme kinetic investigations. Finally, these protocols were applied to a kinetic analysis of the inhibition of SIRT5 by
suramin, a potent sirtuin inhibitor previously shown by X-ray crystallography to bind the substrate pocket of the human SIRT5
KDAC enzyme.
In light of the recent discovery of novel post-translational
acylations of lysine ε-amino groups,22 we envisioned that
performance of a systematic screening of the specificities of the
complete panel of recombinant human sirtuins 1−7 against a
series of fluorogenic substrates accommodating variations in
both peptide sequence and identity of the ε-N-acyl group might
yield interesting new tools for some of these enzymes.
INTRODUCTION
■
The sirtuins, silent information regulator 2 (Sir2) enzymes, are
a family of NAD-dependent KDACs comprising seven human
isoforms (SIRT1−7).1−4 These enzymes have been linked to a
variety of phenotypes and are therefore considered targets for
therapeutic intervention in a broad spectrum of diseases,
including various cancers, diabetes, cardiovascular disease, and
neurodegeneration.5−8 The human genome encodes for a total
of 18 different KDACs, 11 of which are Zn2+-dependent
isoforms (classes I, II, and IV), while the sirtuins constitute the
class III KDACs (Figure 1a).1,2 Contrary to the Zn2+-
dependent isozymes, the sirtuins utilize NAD as a cofactor in
a deacylation event that has been studied mechanistically9−12 as
well as supported by several X-ray cocrystal structures with
substrates and/or cofactor bound to the enzyme13−17 (see
Figure 1b for overall reaction). Only sirtuins 1−3 and 5 have
been shown to harbor robust deacetylase activity, while sirtuins
6 and 7 were significantly less active and SIRT4 showed no
activity.18,19 Sirtuin 4 has been shown to facilitate ADP-
ribosylation20 and has therefore been suggested to be an ADP-
ribosyl transferase enzyme rather than a deacetylase. Moreover,
SIRT5 was recently shown to be a potent demalonylase and
desuccinylase of lysine residues in mitochondrial proteins.21
The precise roles of the individual KDAC isoforms in cellular
function are not yet fully understood, however, and new tools
for accurate and efficient profiling of their enzymatic activity are
therefore highly desirable. Furthermore, development of
protocols for high-throughput screening efforts aimed at the
discovery of new inhibitor candidates is an important objective.
RESULTS AND DISCUSSION
■
Substrate Collection. For the profiling, we selected the
collection of substrates shown in Scheme 1a. Fluorogenic
peptide fragments 1−3 (based on p53 fragments), 5a (based on
a histone 4 fragment), and 6a are commercially available
substrates used for assays employing the various KDAC enzyme
isoforms. Substrate 4 was designed based on part of the amino
acid sequence of histone 3 (amino acids 6−9), and it was
included in our series because of work demonstrating activity of
SIRT6 against a H3K9Ac peptide in vitro23 (see Supporting
Information Scheme S1 for synthetic route). Substrates 5b, 5c,
6b, and 6c were designed and included in our panel to address
the recently discovered activities of SIRT5 against ε-N-
succinyllysine (Ksuc) and ε-N-malonyllysine (Kma) containing
substrates. Finally, we included a recently developed substrate
for efficient profiling of class IIa KDACs (5d),24 as well as a
crotonylated peptide (5e) to address the recent discovery of ε-
N-crotonyllysine (Kcr) residues in histone tails.25
Received: April 13, 2012
Published: May 15, 2012
© 2012 American Chemical Society
5582
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
the reported activity was observed with a ∼2700 Da H3
peptide.20 In addition, the reported activity was obtained with
>4 times the amount of enzyme used in the present study,
which would be considered a prohibitively large amount of
enzyme for screening purposes.
The activities of sirtuins 1−3 and 5 were then measured
against substrates 5a−c and 6a−c, which confirmed a selective
demalonylase and desuccinylase activity of SIRT5, while
SIRT1−3 showed deacetylase activity exclusively (Figure 3).
The deacetylation reactions catalyzed by SIRT1 were
performed with two different commercially available prepara-
tions of the enzyme to test whether this would influence the
substrate specificity. Thus, the six acetylated substrates were
treated with full-length SIRT1 enzyme28 or a truncated SIRT1-
GST fusion protein,29 respectively (Supporting Information
Figure S2). Under the applied conditions, the efficacies of these
two different enzymes followed the same general trend but with
higher activities observed for the full-length enzyme except
against substrate 4 where equally good activities were observed
(Supporting Information Figure S2).
Finally, the activities of SIRT1−3, SIRT5, and SIRT6 against
substrates 5d and 5e were recorded under the same conditions
as described vide supra. The very low activities observed against
substrate 5d are in agreement with previous findings.30
Decrotonylase activity was investigated briefly using substrate
5e because of the very recent work showing that crotonylation
of ε-amino groups of lysine residues should now be considered
a possible post-translational modification in histone proteins.25
A weak signal was observed for SIRT1, which may warrant
further investigations with more elaborate substrates in the
future, while a lack of activity was observed for sirtuins 2, 3, 5,
and 6 (Supporting Information Figure S3). As such, our results
do not indicate that sirtuins harbor potent decrotonylase
activity, albeit based on a limited data set.
Optimization of SIRT5 Desuccinylase Assays. Because
of the high efficacy of SIRT5 observed against succinylated
substrates 5b and 6b (Figure 3), we envisioned that these
substrates might potentially prove useful for efficient screening
of putative inhibitor compounds and decided to address these
substrates in further detail.31 First, we performed titrations to
determine appropriate amounts of enzyme necessary for SIRT5
screening with 5b or 6b, which showed that an enzyme loading
as low as 50 ng of SIRT5 per reaction (30 nM) was sufficient
for end point readings with both substrates (Supporting
Information Figure S4). Then kinetic constants were
determined by measuring initial rate velocities as a function
of substrate concentration and fitting the data to the
Michaelis−Menten equation to give the Km and Vmax values
as shown in Table 1. The kcat values were calculated from the
relationship kcat = Vmax[enzyme]−1, based on enzyme stock
concentration and purity given by the enzyme suppliers.
The turnover numbers (kcat) and catalytic efficiencies
(kcatKm−1) obtained for both of these substrates were somewhat
lower than those reported for longer non-fluorophore-
containing peptide sequences (about 5- to 50-fold).21 The
detection of a released fluorophore, however, enables
significantly higher signal sensitivity compared to HPLC-
based screening methods, thus enabling the low enzyme
loading of 30 nM, as compared to the 1 μM recently reported
for an HPLC-based assay.27 The fluorometric method
furthermore allows for easy performance of continuous rate
experiments as long as an appropriate secondary enzyme is
chosen. The secondary enzyme in the assay, which serves to
Figure 1. (a) Sequence-based classification of the human lysine
deacetylases. (b) Proposed overall mechanism of the NAD-dependent
lysine deacetylation of the active sirtuins 1−3. Complete details can be
found in the referenced literature.9−12,19
The synthesis of substrate 5d has been reported previously;24
however, we were interested in devising an efficient strategy
that would deliver 5a−e in a convergent manner from
inexpensive starting materials. As several attempts to introduce
the AMC fluorophore at a late-stage failed, we applied the
protocol of Jung and co-workers26 followed by mild Fmoc
removal to prepare intermediate 8 in high yield. Subsequent
standard peptide coupling using DIC and HOBt as coupling
reagents and Boc group removal furnished peptide 10, which
could be readily functionalized to give substrates 5a−e
(Scheme 1b). The intermediate 8 was also used for preparation
of substrates 6b and 6c (Supporting Information Scheme S2).
Substrate Profiling with SIRT1−7. For comparison we
first tested for deacylation using 1 μg of enzyme per reaction.
The deacylations were allowed to proceed for 1 h before adding
trypsin “developer” to release the fluorophore of deacylated
substrate molecules and thereby obtain end point assay
readings (see Supporting Information Figure S1 for schematic
representation of the assay). Not surprisingly, on the basis of
the literature,19 SIRT1−3 generally exhibited good activities
toward the acetylated (Kac) substrates, whereas no measurable
activities were observed for sirtuins 4, 6, and 7 (Figure 2).
Sirtuin 5 exhibited poor activity against the Kac substrates in
our collection, with only a poor signal-to-background ratio
observed against substrate 2, which is somewhat surprising
because SIRT5 has been classified as an active deacetylase
enzyme.19 This, however, agrees with more recent inves-
tigations reporting low affinity for Kac substrates with SIRT5.21
On the other hand, SIRT5 exhibited very efficient desuccinylase
activity with substrate 5b, which is also in agreement with
recent contributions in the literature using nonfluorogenic
substrates.21,27 Unfortunately, substrate 4, which was designed
in the hopes of achieving activity of SIRT6, did not furnish a
measurable signal under the applied conditions. This indicates
that more elaborate H3K9Ac-containing fragments are
necessary in order to obtain SIRT6 deacetylase activity, as
5583
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
a
Scheme 1. (a) Structures of the Analyzed Substrates and (b) Synthesis of Substrates 5a−e
a
Reagents and conditions: (a) POCl3 (3 equiv), pyridine (10.6 equiv), 7-amino-4-methylcoumarin (AMC, 1.15 equiv), THF, 0 °C → rt, 2.5 h
(quant).26 (b) Piperidine (1.1 equiv), MeCN, CH2Cl2, 0 °C → rt, 24 h (88%). (c) Ac-Leu-Gly-OH (1.02 equiv), diisopropylcarbodiimide (DIC, 1.1
equiv), hydroxybenzotriazole (1.1 equiv), CH2Cl2, rt, 50 min (80%). (d) Trifluoroacetic acid−CH2Cl2 (1:9, v/v), rt, 30 min. (e) Ac2O (1.5 equiv), i-
Pr2NEt (6 equiv), CH2Cl2, MeCN, 0 °C → rt, 1 h (34%, two steps). (f) Succinic anhydride (1.7 equiv), i-Pr2NEt (6 equiv), CH2Cl2, rt, 55 min
(70%, 2 steps). (g) (1) Potassium methyl malonate (2 equiv), i-Pr2NEt (2.5 equiv), HATU (2 equiv), MeCN, 0 °C, 20 min; (2) LiOH (7.3 equiv),
THF−H2O, rt, 15 min (49%, 3 steps from 9). (h) Trifluoroacetic anhydride (9 equiv), i-Pr2NEt (14 equiv), CH2Cl2, 0 °C → rt, 22 h (60% from 9).
(i) trans-Crotonic anhydride (1.8 equiv), i-Pr2NEt (6 equiv), CH2Cl2, rt, 10 min (59% from 9).
Figure 3. Bar graph for deacylation of substrates 5a−c and 6a−c. The
data were obtained as end point readings after incubation for 1 h with
each substrate at 37 °C and represent at least two individual
experiments performed in duplicate.
Figure 2. Results of sirtuin deacetylation and desuccinylation of
substrate 5b. The data were obtained as end point readings after
incubation for 1 h with each substrate at 37 °C and represent at least
two individual experiments performed in duplicate. The amount of
recombinant enzyme added to each reaction was 1 μg/50 μL as
determined from stock solution concentrations and purities given by
the suppliers (SIRT1-GST fusion protein was used as SIRT1 source).
The RFU values are normalized to control wells without sirtuin
enzyme present. See Supporting Information Figures S2 and S3 for
additional profiling data.
Table 1. Kinetic Parameters for SIRT5 Desuccinylation of
a
Fluorogenic Substrates 5b and 6b
−1
substrate
Km (μM)
Vmax (μM s−1
)
kcat (s−1
)
kcatKm (s−1 M−1
)
5b
6b
33 1.8
84 22
1.9 × 10−3
5.6 × 10−4
3 × 10−2
9 × 10−3
9.2 × 102
1.1 × 102
a
release the fluorophore after the initial deacylation reaction of
interest (Supporting Information Figure S1), must meet two
important demands to be a successful in situ developer enzyme.
First, it must be faster than the initial enzyme in order to
The Km value for NAD measured with substrate 6b at 200 μM was
150 45 μM. Michaelis−Menten plot is shown in Figure S5a.
5584
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
Figure 4. (a) Inhibitor structures and sirtuin 5 inhibition data using substrate 6b. Data are based on two individual assays performed in duplicate.
The values in parentheses were previously reported for inhibition of desuccinylation determined in an HPLC−MS based assay with nonfluorogenic
peptides.27 (b) Geometry of NAD and a succinyllysine-containing peptide cocrystallized with SIRT5 (PDB accession code, 3RIY).21
instantaneously process the intermediate product to give a
readout, and second, it is essential that the sirtuin enzyme will
not be degraded under the assay conditions because of the
inherent peptidase activity of the developer.
sirtinol,34 the SIRT2 inhibitor AGK2,35 and suramin,36 a drug
used to treat sleeping sickness. The IC50 values were all in
agreement with previously published data obtained using an
HPLC-based assay27 (Figure 4a, top).
For the present protocol, we considered the Lys-C protease,
which has been used as a mild nondegrading option in HDAC
experiments,32,33 and trypsin, which is known from end point
assays to be very potent. Initial rate experiments for Lys-C at 4
milliunits of enzyme per reaction revealed a Vmax that was
considerably higher than those of SIRT5 against 5b and 6b,
indicating that Lys-C could be a useful option (Supporting
Information Figure S6). With trypsin our concern was the
possibility of premature inactivation of SIRT5 due to
degradation. At a final concentration of 10 μg/mL of trypsin
in the assay, however, the rate increase in fluorescence signal
was stable for recordings up to one hour (Supporting
Information Figure S7). Protocols were thus established for
end point assays, where interference of the developer enzyme is
absent, as well as for continuous assays employing two different
in situ developer enzymes.
Evaluation of Inhibitors. Though substrate 6b furnished
slightly lower catalytic efficacy than 5b, its simple chemical
structure and hence easy accessibility was appealing, and it was
therefore decided to perform inhibition experiments using
substrate 6b. Initially, IC50 values were determined by nonlinear
regression of dose−response data obtained by end point assays.
The selected sirtuin inhibitors were the SIRT1 inhibitor
In an attempt to achieve binding affinity by interaction of the
active site arginine (Arg-105, Figure 4b) with a carboxylic or
hydroxamic acid functionality, three known HDAC inhibitors
that are all in clinical use were tested as well (Figure 4a, bottom
panel). This proved fruitless with the simple structures tested
herein, although evidence that a carboxylic acid moiety may
contribute to the binding affinity of SIRT5 inhibitors was
recently described for ε-N-thiosuccinyllysine-containing inhib-
itors,27 which were designed by analogy to ε-N-thioacetyllysine-
containing SIRT1 inhibitors.30,37,38
Because of previously reported assay artifacts related to
fluorophore-containing substrates, which has come under
scrutiny with respect to activation of SIRT1,39−42 we performed
kinetic inhibition experiments using the continuous assay
methods discussed vide supra to further validate the developed
substrates and protocols. We chose the commercially available
inhibitor suramin, which conveniently has been cocrystallized
with SIRT5,14 for our kinetic studies.
First, we measured steady-state velocities from a titration of
substrate 6b at saturating NAD concentration and varying
concentrations of suramin to give a secondary double reciprocal
plot (Lineweaver−Burk plot)43 as shown in Supporting
Information Figure S8. Intersection of the lines with the y-
5585
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
Figure 5. (a) Progress curves for enzyme inhibition of suramin against SIRT5 using substrate 6b and Lys-C as the secondary enzyme (linear
regression lines for the 30−120 min regions are shown in black). (b) Progress curves for suramin inhibition of SIRT5 using substrate 6b and trypsin
as developer (linear regression lines for the 10−50 min regions are shown in black). (c) Nonlinear fitting of the data to eq 133,44 using the GraphPad
Prism software.
−1
axis at constant Vmax shows that the inhibitor is competitive
with respect to the fluorogenic substrate (6b), which
gratifyingly agrees with the X-ray structure showing that
suramin binds to the acyllysine substrate pocket of the
enzyme.14 Next, we measured progression curves for SIRT5
desuccinylase activity in the presence of varying concentrations
of suramin using the two different developer enzymes
discussed. When using Lys-C as the coupled secondary enzyme,
we observed a lag time of almost 30 min before the system
reached a steady state. On the other hand, the resulting
progress curves remained linear up to 2 h (Figure 5a). Trypsin
development allowed for the system to reach steady state after
just 5−10 min, giving rise to linear curves up to at least 50 min
(Figure 5b). Linear graphs in these plots represent a constant
increase in the fluorescence signal, which in turn reflects a
constant turnover rate of the SIRT5 enzyme in agreement with
a fast-on−fast-off mode of inhibition. Finally, fitting the data to
eq 1, which is in accord with the fast-on−fast-off mechanism,
furnished similar Ki values regardless of the developer enzyme
used (Figure 5c). To our knowledge, these findings conclude
the first disclosure of a kinetic investigation of SIRT5 inhibition
using the progression curve method.
significantly lower deacetylase activities of SIRT6 and SIRT7
compared to SIRT1−3.18 These findings, however, do not rule
out that these enzymes may deacylate substrates not yet
discovered. Sirtuin 5, on the other hand, exhibited somewhat
surprisingly low deacetylase activities but very potent
demalonylase and desuccinylase activities as recently described
for nonfluorogenic peptide substrates.
Thus, two new substrates were discovered, which enable
highly efficient SIRT5 desuccinylation activity at significantly
lower enzyme loading than previously described. As such, these
readily prepared substrates should be suitable for screening
efforts toward the discovery of novel SIRT5 inhibitor
compounds. Furthermore, optimized conditions for facile
kinetic investigations of SIRT5 interaction with substrates
and/or inhibitors were developed, which will be valuable for
mechanistic investigations. The well-known inhibitor suramin,
which has been cocrystallized with SIRT5, was used to validate
our two different continuous assay protocols by performing
progression curve experiments.
Additionally, the results of the first decrotonylation experi-
ments performed with sirtuin enzymes indicated that sirtuins
are not likely to be potent decrotonylases.
−1
⎛
⎜
⎜
⎞
⎟
⎟
EXPERIMENTAL METHODS
■
vi
v0
[I]
Materials. Recombinant SIRT1-GST fusion protein, SIRT2, and
Lys-C (Lysobacter enzymogenes) were purchased from Merck-Millipore
(Darmstadt, Germany), and SIRT1 (full length His-tagged), SIRT3,
SIRT5, and SIRT6 were from Enzo Life Sciences (Postfach,
Switzerland). Recombinant SIRT4 and SIRT7 were from Abnova
(Taipei, Taiwan). Sirtuin assay buffer was prepared as described in the
Biomol International product sheets BML-AK-555 [Tris-Cl (50 mM),
NaCl (137 μM), KCl (2.7 μM), MgCl2 (1 μM), pH 8.0, and bovine
serum albumin (1 mg/mL)]. The SIRT2 substrate (3), suramin,
sirtinol, and AGK2 inhibitors were from Enzo Life Sciences (Postfach,
Switzerland). 4-Phenylbutyric acid, valproic acid, NAD, and trypsin
(10 000 units/mg, TPCK treated from bovine pancreas) were from
Sigma-Aldrich (Steinheim, Germany).
Sirtuin Deacylation Assays. In brief, all reactions were performed
in black low binding NUNC 96-well microtiter plates. End point
assays were performed by incubation of the appropriate substrate,
NAD, sirtuin, and inhibitor (if applicable) in assay buffer (50 μL final
volume). Control wells without enzyme were included in each plate.
After 1 h at 37 °C, a developer mixture (50 μL) containing
nicotinamide (2 mM) and trypsin (0.4 mg mL−1) was added, and the
assay was incubated for 15−30 min at room temperature before the
=
+ 1
[S]
Km
⎜
⎝
⎟
⎠
Ki 1 +
(
)
(1)
CONCLUSIONS
■
In summary, we have evaluated the activities of the seven
human sirtuin hydrolases against a series of fluorogenic
substrates in vitro. The tested substrate collection was designed
to address enzyme specificities toward the most commonly
used commercial substrates as well as the recently discovered
lysine post-translational modifications (i.e., Kma, Ksuc, and
Kcr).22 Convergent synthetic routes for preparation of these
new chemotypes were devised to gain easy access to the
substrate collection.
Our screening confirmed that SIRT1−3 all exhibit relatively
broad-spectrum deacetylase activities. Sirtuins 4, 6, and 7 were
practically inactive under our conditions, which is in agreement
with previous literature reporting lack of SIRT4 activity and
5586
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
plates were analyzed using a Perkin-Elmer Enspire plate reader with
excitation at 360 nm and emission at 460 nm. Rate experiments for
determination of kinetic parameters were performed by incubation of
substrate, NAD, and sirtuin in assay buffer (100 μL final volume) with
in situ fluorophore cleavage by Lys-C or trypsin. Fluorescence readings
were recorded every 30 s for at least 30 min at 25 °C to obtain initial
linear rates ν0 (RFU min−1) for each concentration. The data were
analyzed using GraphPad Prism to afford Km (μM) and Vmax (RFU
min−1) values. The kcat values were calculated based on the [7-amino-
4-methylcoumarin]−RFU standard curve (Supporting Information
Figure S5) and the enzyme purities and concentrations given by the
suppliers. The dilution series for dose−response inhibitor experiments
were prepared in SIRT assay buffer from DMSO stock solutions.
Substrate 6b was used for inhibition experiments, and the SIRT5
enzyme concentration was 50 ng/well (∼30 nM in 50 μL final
volume) for end point assays or 200 ng/well (∼60 nM in 100 μL final
volume) for continuous assays.
1H NMR (DMSO-d6) δ = 10.39 (s, 1H, NHAMC), 8.34 (t, J = 5.8, 1H,
NHGly), 8.11 (d, J = 7.2, 1H, NHLeu), 8.02 (d, J = 7.5, 1H, NHα,Lys),
7.81 (t, J = 5.5, NHε,Lys), 7.80 (d, J = 2.0, 1H, H8AMC), 7.73 (d, J = 8.7,
1H, H5AMC), 7.52 (dd, J = 8.7 and 2.0, 1H, H6AMC), 6.27 (d, J = 1.4,
1H, H3AMC), 4.32−4.41 (m, 1H, Hα,Lys), 4.21 (q, J = 7.4, 1H, Hα,Leu),
3.74 (mABX,A, J = 16.8 and 5.7, 1H, Hα,Gly,A), 3.71 (mABX,B, J = 16.8 and
5.9, 1H, Hα,Gly,B), 3.00 (q, J = 6.1, 2H, CH2,ε,Lys), 2.40 (d, J = 1.3, 3H,
CH3,AMC), 1.84 (s, 3H, CH3CONHLeu), 1.77 (s, 3H, CH3CONHε,Lys),
1.51−1.74 (m, 3H, CHγ,Leu and CH2,β,Lys), 1.16−1.50 (m, 6H,
CH2,β,Leu, CH2,γ,Lys,CH2,δ,Lys), 0.88 (d, J = 6.5, 3H, CH3,Leu,A), 0.83
(d, J = 6.5, 3H, CH3,Leu,B). 13C NMR (DMSO-d6) δ = 172.9 (COLeu),
171.4 (COLys), 169.7 (CONHLeu), 169.02 (COGly/CONHε,Lys), 168.96
(COGly/CONHε,Lys), 160.0 (C2AMC), 153.6 (C8aAMC), 153.1 (C4AMC),
142.1 (C7AMC), 126.0 (C5AMC), 115.3 (C6AMC), 115.2 (C4aAMC),
112.4 (C3AMC), 105.8 (C8AMC), 53.6 (Cα,Lys), 51.5 (Cα,Leu), 42.0
(Cα,Gly), 40.5 (Cβ,Leu), 38.3 (Cε,Lys), 31.4 (Cβ,Lys), 28.9 (Cδ,Lys), 24.2
(Cγ,Leu), 22.9 (CH3,Leu,A), 22.6 (Cγ,Lys), 22.5 (CH3CO), 21.7
(CH3,Leu,B), 18.0 (CH3,AMC). UPLC−MS tR = 1.49 min, m/z 558.3
Chemistry. General Methods and Materials. 2-Chlorotrityl
chloride resin was from Sigma-Aldrich (Milwaukee, WI). All α-amino
acids, chemicals, and solvents were analytical grade and used without
further purification as obtained from commercial suppliers. Anhydrous
THF was stored over sodium and freshly distilled prior to use.
Anhydrous MeCN was dried by storage over 3 Å molecular sieves.
Anhydrous CH2Cl2, i-Pr2NEt, and pyridine were dried by storage over
activated 4 Å molecular sieves. Dry column vacuum chromatography
was performed on silica gel 60 (particle size 0.015−0.040 mm).
Preparative reversed-phase HPLC was performed on a [250 mm × 20
mm, C18(2) Phenomenex Luna column (5 μm, 100 Å)] using an
Agilent 1260 LC system equipped with a diode array UV detector and
an evaporative light scattering (ELS) detector. A gradient with eluent
A (95:5:0.1, water−MeCN−TFA) and eluent B (0.1% TFA in
acetonitrile) rising linearly from 0% to 95% of B during t = 5−45 min
was applied at a flow rate of 20 mL/min. The isolated substrates were
lyophilized from the HPLC solvents and stored at −20 °C. UPLC−
MS analyses were performed on a Waters Acquity ultrahigh
performance liquid chromatography system to confirm purities of
>95% of all compounds, and high resolution mass spectroscopic
measurements were performed using ultrahigh performance liquid
chromatography−high resolution mass spectrometry (UHPLC-
HRMS) on a maXis G3 quadrupole time-of-flight (TOF) mass
spectrometer (Bruker Daltonics, Bremen, Germany) equipped with an
electrospray (ESI) source. NMR spectra were recorded on a Varian
+
([M + H]+, C28H40N5O7 calcd 558.3). HRMS m/z 558.2916 ([M +
+
H]+, C28H40N5O7 calcd 558.2922).
Ac-Leu-Gly-Lys(succinyl)-(7-amino-4-methylcoumarin) (5b).
Crude Ac-Leu-Gly-Lys-(7-amino-4-methylcoumarin) trifluoroacetate
salt (10, 50 mg) was suspended in anhydrous CH2Cl2 (2 mL) and i-
Pr2NEt (0.1 mL) under N2 at room temperature. Succinic anhydride
(13 mg, 133 μmol) was added, immediately resulting in a clear
solution. After being stirred for 55 min, the reaction mixture was
evaporated to dryness. Preparative reversed-phase HPLC purification
of the crude residue afforded Ac-Leu-Gly-Lys(succinyl)-(7-amino-4-
methylcoumarin) (5b) as a white fluffy material after lyophilization
(29 mg, 70% from 9). 1H NMR (DMSO-d6) δ 10.38 (s, 1H, NHAMC),
8.35 (t, J = 5.8, 1H, NHGly), 8.11 (d, J = 7.3, 1H, NHLeu), 8.02 (d, J =
7.5, 1H, NHα,Lys), 7.83 (t, J = 5.5, 1H, NHε,Lys), 7.80 (d, J = 2.0, 1H,
H8AMC), 7.72 (d, J = 8.7, 1H, H5AMC), 7.52 (dd, J = 8.7, 2.0, 1H,
H6AMC), 6.27 (d, J = 1.2, 1H, H3AMC), 4.32−4.42 (m, 1H, Hα,Lys), 4.22
(q, J = 7.4, 1H, Hα,Leu), 3.74 (mABX,A, J = 16.7 and 5.5, 1H, Hα,Gly,A),
3.71 (mABX,B, J = 16.7, 5.8, 1H, Hα,Gly,B), 3.01 (q, J = 6.3, 2H, CH2,ε,Lys),
2.34−2.42 (m, 5H, CH3,AMC, HO2CCH2), 2.23−2.32 (m, 2H,
CH2CONHε,Lys), 1.84 (s, 3H, CH3CONHLeu), 1.51−1.77 (m, 3H,
CHγ,Leu, CH2,β,Lys), 1.18−1.51 (m, 6H, CH2,β,Leu,CH2,γ,Lys,CH2,δ,Lys),
0.88 (d, J = 6.5, 3H, CH3,Leu,A), 0.83 (d, J = 6.4, 3H, CH3,Leu,B). 13C
NMR (DMSO-d6) δ 173.9 (CO2H), 173.0 (COLeu), 171.5 (COLys),
170.8 (NHε,LysCO), 169.8 (CH3CO), 169.0 (COGly), 160.1 (C2AMC),
153.7 (C8aAMC), 153.1 (C4AMC), 142.2 (C7AMC), 126.0 (C5AMC),
115.3 (C6AMC), 115.2 (C4aAMC), 112.4 (C3AMC), 105.8 (C8AMC), 53.6
(Cα,Lys), 51.5 (Cα,Leu), 42.1 (Cα,Gly), 40.5 (Cβ,Leu), 38.4 (Cε,Lys), 31.4
(Cβ,Lys), 30.0 (CH2CONHε,Lys), 29.2 (HO2CCH2), 28.8 (Cδ,Lys), 24.2
(Cγ,Leu), 23.0 (CH3,Leu,A), 22.9 (Cγ,Lys), 22.5 (CH3CO), 21.7
(CH3,Leu,B), 18.0 (CH3,AMC). UPLC−MS tR = 1.47 min, m/z 616.5
1
Mercury 300 instrument. H NMR, 13C NMR, and 19F NMR were
recorded at 300, 75, and 282 MHz, respectively. Chemical shifts are
reported in ppm relative to deuterated solvent peaks as internal
standards (δH, DMSO-d6 2.50 ppm; δC, DMSO-d6 39.52 ppm) or
external calibration (δF). Coupling constants (J) are given in hertz
1
(Hz). Multiplicities of H NMR signals are reported as follows: s,
+
([M + H]+, C30H42N5O9 calcd 616.3). HRMS m/z 616.2976 ([M +
singlet; d, doublet; dd, doublet of doublets; t, triplet; q, quartet; p,
pentet; m, multiplet; br, broad signal. Protons that were exchangeable
upon addition of D2O are denoted “ex”. Assignments of NMR spectra
are based on correlation spectroscopy (COSY, HSQC, HMQC, and/
or HMBC spectra).
+
H]+, C30H42N5O9 calcd 616.2977).
Ac-Leu-Gly-Lys(malonyl)-(7-amino-4-methylcoumarin) (5c).
Potassium monomethyl malonate (50 mg, 0.319 mmol) was
suspended in MeCN (2 mL) and i-Pr2NEt (0.07 mL) under Ar at 0
°C. HATU (122 mg, 0.321 mmol) was added affording a yellow
solution. After preincubation for 5 min, the crude Ac-Leu-Gly-Lys-(7-
amino-4-methylcoumarin) trifluoroacetate salt (10, 99 mg) was added,
and the resulting suspension was stirred at 0 °C. After the mixture was
stirred for 20 min, MeOH (1 mL) was added, and the reaction mixture
was evaporated to dryness. The resulting residue was purified by
column chromatography (0−8% MeOH in CH2Cl2), to afford a white
foam [Ac-Leu-Gly-Lys((methyl)malonyl)-(7-amino-4-methylcoumar-
in) (84 mg). UPLC−MS tR = 1.56 min, m/z 616.3 ([M + H]+,
C30H42N5O9+ calcd 616.3)]. This crude material was dissolved in THF
(1.0 mL) and aqueous LiOH (1.0 M, 1.0 mL). After being stirred for
15 min, the reaction mixture was concentrated to approximately 1 mL,
then neutralized by addition of aqueous HCl (2 M, 0.3 mL). The
resulting solution was purified by preparative reversed-phase HPLC to
afford Ac-Leu-Gly-Lys(malonyl)-(7-amino-4-methylcoumarin) (5c) as
Ac-Leu-Gly-Lys(Ac)-(7-amino-4-methylcoumarin) (5a).45
Crude Ac-Leu-Gly-Lys-(7-amino-4-methylcoumarin) trifluoroacetate
salt (10, 49 mg) was suspended in anhydrous CH2Cl2 (4 mL) and
anhydrous MeCN (3 mL) at 0 °C under N . Hunig's base (63 mg, 488
̈
2
μmol) and Ac2O (12 mg, 115 μmol) were added. After being stirred
for 1 h at room temperature, the reaction mixture was taken up in
brine. The aqueous phase was extracted with CH2Cl2 (3 × 20 mL),
and the combined organic phase was washed with aqueous HCl (1 M,
2 × 10 mL). The acidic aqueous phase was back-extracted with
CH2Cl2 (3 × 20 mL), then 10% MeOH in CH2Cl2 (5 × 20 mL). The
combined organic phase was then washed with saturated aqueous
NaHCO3 (10 mL) and the aqueous phase back-extracted with CH2Cl2
(3 × 20 mL). The combined organic phase was dried over MgSO4,
filtered, and evaporated to dryness. Preparative reversed-phase HPLC
purification of the crude residue afforded 5a as a white fluffy material
after lyophilization (13 mg, 34% from 9). Full characterization details
were not given in the original report and are therefore included here.45
1
a white fluffy material after lyophilization (40 mg, 49% from 9). H
NMR (DMSO-d6) δ = 10.39 (s, 1H, NHAMC), 8.35 (t, J = 5.8, 1H,
5587
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
NHGly), 8.11 (d, J = 7.2, 1H, NHLeu), 7.99−8.08 (m, 2H, NHα,Lys and
NHε,Lys), 7.80 (d, J = 1.9, 1H, H8AMC), 7.72 (d, J = 8.7, 1H, H5AMC),
7.52 (dd, J = 8.7 and 2.0, 1H, H6AMC), 6.27 (d, J = 1.2, 1H, H3AMC),
4.32−4.43 (m, 1H, Hα,Lys), 4.22 (q, J = 7.4, 1H, Hα,Leu), 3.74 (mABX,A, J
= 16.6 and 5.7, 1H, Hα,Gly,A), 3.71 (mABX,B, J = 16.6 and 5.6, 1H,
μmol) was added, resulting in a clear solution after stirring for 30 s.
The reaction mixture was stirred for 10 min and was then evaporated
to dryness. The resulting crude was purified by preparative reversed-
phase HPLC to afford the title compound as a white fluffy material
1
after lyophilization (24 mg, 59% from 9). H NMR (DMSO-d6) δ =
H
α,Gly,B), 3.10 (s, 2H, CH2(CO)2), 3.04 (q, J = 6.5, 2H, CH2,ε,Lys), 2.39
10.38 (s, 1H, NHAMC), 8.34 (t, J = 5.8, 1H, NHGly), 8.11 (d, J = 7.3,
1H, NHLeu), 8.03 (d, J = 7.5, 1H, NHLys), 7.85 (t, J = 5.6, 1H, NHε,Lys),
7.79 (d, J = 2.0, 1H, H8AMC), 7.72 (d, J = 8.7, 1H, H5AMC), 7.52 (dd, J
= 8.7 and 2.0, 1H, H6AMC), 6.56 (dq, J = 15.3 and 6.9, 1H, CH
CHCH3), 6.27 (d, J = 1.4, 1H, H3AMC), 5.86 (dq, J = 15.3 and 1.6, 1H,
CHCHCH3), 4.30−4.42 (m, 1H, Hα,Lys), 4.22 (q, J = 7.4, 1H,
(d, J = 1.2, 3H, CH3,AMC), 1.84 (s, 3H, CH3CO), 1.51−1.79 (m, 3H,
CHγ,Leu and CH2,β,Lys), 1.19−1.51 (m, 6H, CH2,β,Leu, CH2,γ,Lys, and
CH2,δ,Lys), 0.88 (d, J = 6.5, 3H, CH3,Leu,A), 0.83 (d, J = 6.4, 3H,
CH3,Leu,B). 13C NMR (DMSO-d6) δ = 173.0 (COLeu), 171.5 (COLys),
169.8 (CH3CO), 169.6 (CO2H), 169.1 (COGly), 165.6 (CONHε,Lys),
160.1 (C2AMC), 153.7 (C8aAMC), 153.1 (C4AMC), 142.2 (C7AMC),
126.0 (C5AMC), 115.3 (C6AMC), 115.2 (C4aAMC), 112.4 (C3AMC),
105.8 (C8AMC), 53.6 (Cα,Lys), 51.5 (Cα,Leu), 42.6 (CH2(CO)2), 42.1
(Cα,Gly), 31.4 (Cβ,Lys), 28.7 (Cδ,Lys), 24.2 (Cγ,Leu), 23.0 (CH3,Leu,A), 22.9
(Cγ,Lys), 22.5 (CH3CO), 21.7 (CH3,Leu,B), 18.0 (CH3,AMC). UPLC−MS
H
α,Leu), 3.74 (mABX,A, J = 16.9 and 5.4, 1H, Hα,Gly,A), 3.71 (mABX,B, J =
16.9 and 6.1, 1H, Hα,Gly,B), 3.08 (q, J = 6.4, 2H, CH2,ε,Lys), 2.39 (d, J =
1.3, 3H, CH3,AMC), 1.84 (s, 3H, CH3CONHLeu), 1.75 (dd, J = 6.9 and
1.6, 3H, CHCHCH3), 1.51−1.72 (m, 3H, CHγ,Leu and CH2,β,Lys),
1.17−1.51 (m, 6H, CH2,β,Leu, CH2,γ,Lys, and CH2,δ,Lys), 0.88 (d, J = 6.5,
3H, CH3,Leu,A), 0.83 (d, J = 6.4, 3H, CH3,Leu,B). 13C NMR (DMSO-d6)
δ = 172.9 (COLeu), 171.4 (COLys), 169.7 (CH3CO), 169.0 (COGly),
164.8 (NHCOε,Lys), 160.0 (C2AMC), 153.6 (C8aAMC), 153.1 (C4AMC),
142.1 (C7AMC), 137.3 (CHCHCH3), 126.0 (C5AMC,CHCHCH3),
115.3 (C6AMC), 115.2 (C4aAMC), 112.4 (C3AMC), 105.8 (C8AMC), 53.6
(Cα,Lys), 51.5 (Cα,Leu), 42.0 (Cα,Gly), 40.5 (Cβ,Leu), 39.0 (Cε,Lys), 31.4
(Cβ,Lys), 28.9 (Cδ,Lys), 24.2 (Cγ,Leu), 23.0 (CH3,Leu,A), 22.9 (Cγ,Lys), 22.5
(CH3CO), 21.7 (CH3,Leu,B), 18.0 (CH3,AMC), 17.3 (CHCHCH3).
UPLC−MS tR = 1.62 min, m/z 584.4 ([M + H]+, C30H42N5O7+ calcd
+
tR = 1.49 min, m/z 602.3 ([M + H]+, C29H4+0N5O9 calcd 602.3).
HRMS m/z 602.2823 ([M + H]+, C29H40N5O9 calcd 602.2821).
Ac-Leu-Gly-Lys(Tfa)-(7-amino-4-methylcoumarin) (5d).24
Crude Ac-Leu-Gly-Lys-(7-amino-4-methylcoumarin) trifluoroacetate
salt (10, 52 mg) was suspended in anhydrous CH2Cl2 (4 mL) and
MeCN (3 mL) under N2 at 0 °C. i-Pr2NEt (62 mg, 477 μmol) and
(CF3CO)2O (33 mg, 159 μmol) were added, and the mixture was
stirred for 100 min. Additional portions of i-Pr2NEt (100 μL, 0.58
mmol) and (CF3CO)2O (127 mg, 0.61 mmol) were added over a time
course of 20 h to drive the reaction to completion. Then the reaction
mixture was taken up in aqueous HCl (1 M, 10 mL) and CH2Cl2 (20
mL). The aqueous phase was extracted with CH2Cl2 (3 × 20 mL), and
the combined organic phase was washed with aqueous HCl (1 M, 10
mL), after which the aqueous phase was back-extracted with CH2Cl2
(3 × 20 mL). The organic phase was then washed with saturated
aqueous NaHCO3 (10 mL), dried over MgSO4, filtered, and
evaporated to dryness. The resulting residue was purified by column
chromatography (0−5% MeOH in CH2Cl2), affording the desired Ac-
Leu-Gly-Lys(Tfa)-(7-amino-4-methylcoumarin) (5d) as a white solid
(26 mg, 60% from 9). The 13C NMR data did not correspond fully
with previously reported data, since we did not observe additional
peaks attributed to a mixture of conformers.24 1H NMR (DMSO-d6) δ
= 10.39 (s, 1H, NHAMC), 9.42 (t, J = 5.7, 1H, NHCOCF3), 8.34 (t, J =
5.8, 1H, NHGly), 8.10 (d, J = 7.3, 1H, NHLeu), 8.05 (d, J = 7.5, 1H,
NHLys), 7.79 (d, J = 2.0, 1H, H8AMC), 7.72 (d, J = 8.7, 1H, H5AMC),
7.51 (dd, J = 8.7 and 2.0, 1H, H6AMC), 6.27 (d, J = 1.4, 1H, H3AMC),
4.32−4.44 (m, 1H, Hα,Lys), 4.22 (q, J = 7.5, 1H, Hα,Leu), 3.74 (mABX,A, J
= 16.8 and 5.4, 1H, Hα,Gly,A), 3.71 (mABX,B, J = 16.8 and 5.7, 1H,
+
584.3). HRMS m/z 584.3079 ([M + H]+, C30H42N5O7 calcd
584.3079).
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional figures, additional experimental details, and
1
compound characterization data, as well as H, 13C, and 19F
NMR spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +45 45252105. E-mail: cao@kemi.dtu.dk.
Notes
The authors declare no competing financial interest.
Hα,Gly,B), 3.16 (q, J = 6.7, 2H, CH2,ε,Lys), 2.40 (d, J = 1.4, 3H,
CH3,AMC), 1.84 (s, 3H, CH3CONHLeu), 1.19−1.80 (m, 9H, CH2,β,Leu
,
ACKNOWLEDGMENTS
CHγ,Leu, CH2,β,Lys, CH2,γ,Lys, and CH2,δ,Lys), 0.88 (d, J = 6.5, 3H,
CH3,Leu,A), 0.83 (d, J = 6.4, 3H, CH3,Leu,B). 13C NMR (DMSO-d6) δ =
172.9 (COLeu), 171.4 (COLys), 169.7 (CH3CO), 169.0 (COGly), 160.0
(C2AMC), 156.1 (q, J = 35.9, COCF3), 153.6 (C8aAMC), 153.1
(C4AMC), 142.1 (C7AMC), 126.0 (C5AMC), 116.0 (q, J = 288.0, CF3),
115.3 (C6AMC), 115.2 (C4aAMC), 112.4 (C3AMC), 105.8 (C8AMC), 53.5
(Cα,Lys), 51.5 (Cα,Leu), 42.0 (Cα,Gly), 40.5 (Cβ,Leu), 39.0 (Cε,Lys), 31.3
(Cβ,Lys), 28.0 (Cδ,Lys), 24.2 (Cγ,Leu), 22.9 (CH3,Leu,A), 22.7 (Cγ,Lys), 22.5
(CH3CO), 21.6 (CH3,Leu,B), 18.0 (CH3,AMC). 19F NMR (DMSO-d6) δ
= −75.7 (CF3). UPLC−MS tR = 1.77 min, m/z 612.4 ([M + H]+,
■
Financial support from the Lundbeck Foundation (Young
Group Leader Fellowship), the Danish Independent Research
CouncilNatural Sciences (Steno Grant No. 10-080907), and
the Carlsberg Foundation is gratefully acknowledged.
ABBREVIATIONS USED
■
ADPR, adenosine diphosphate ribose; AMC, 7-amino-4-
methylcoumarin; DIC, N,N′-diisopropylcarbodiimide; H3,
histone 3 protein; H4, histone 4 protein; HATU, O-(7-
azabenzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium hexafluor-
ophosphate; HDAC, histone deacetylase; HOBt, hydroxyben-
zotriazole; HPLC, high performance liquid chromatography; h,
hour; KDAC, lysine deacylase; Kac, ε-N-acetyllysine; Ksuc, ε-N-
succinyllysine; Kma, ε-N-malonyllysine; Ktfa, ε-N-trifluoroace-
tyllysine; Kcr, ε-N-crotonyllysine; Ma, malonyl; MS, mass
spectrometry; NAD, nicotinamide adenine dinucleotide; NMR,
nuclear magnetic resonance; RFU, relative fluorescence unit; rt,
room temperature; SIRT, sirtuin; Suc, succinyl; TFA, trifluoro-
acetic acid; tR, retention time; UPLC, ultrahigh performance
liquid chromatography
+
C28H37F3N5O7+ calcd 612.3). HRMS m/z 612.2646 ([M + H]+,
C28H37F3N5O7 calcd 612.2640).
Ac-Leu-Gly-Lys(crotonyl)-(7-amino-4-methylcoumarin) (5e).
Crotonoic acid (5.03 g, 58.4 mmol) was suspended in CH2Cl2 (50
mL) under N2, and the mixture was stirred at 0 °C. DIC (6.0 mL, 38.7
mmol) was added portionwise. The reaction mixture was stirred for 3
days and then evaporated to dryness. The resulting residue was taken
up in heptane (50 mL), and the white precipitate was filtered off and
washed with heptane (3 × 10 mL). The combined organic phases were
evaporated to dryness to afford crude trans-crotonic anhydride, which
was used directly in the acylation step. Crude Ac-Leu-Gly-Lys-(7-
amino-4-methylcoumarin) trifluoroacetate salt (10, 50 mg) was
suspended in anhydrous CH2Cl2 (2.5 mL) and i-Pr2NEt (0.1 mL)
under N2 at room temperature. Crotonyl anhydride (22 mg, 143
5588
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
Barrett, J. C.; Chang, H. Y.; Bohr, V. A.; Ried, T.; Gozani, O.; Chua, K.
F. SIRT6 is a histone H3 lysine 9 deacetylase that modulates telomeric
chromatin. Nature 2008, 452, 492−496.
REFERENCES
■
(1) Biel, M.; Wascholowski, V.; Giannis, A. Epigeneticsan
epicenter of gene regulation: histones and histone-modifying enzymes.
Angew. Chem., Int. Ed. 2005, 44, 3186−3216.
(24) Bradner, J. E.; West, N.; Grachan, M. L.; Greenberg, E. F.;
Haggarty, S. J.; Warnow, T.; Mazitschek, R. Chemical phylogenetics of
histone deacetylases. Nat. Chem. Biol. 2010, 6, 238−243.
(25) Tan, M.; Luo, H.; Lee, S.; Jin, F.; Yang, J. S.; Montellier, E.;
Buchou, T.; Cheng, Z.; Rousseaux, S.; Rajagopal, N.; Lu, Z.; Ye, Z.;
Zhu, Q.; Wysocka, J.; Ye, Y.; Khochbin, S.; Ren, B.; Zhao, Y.
Identification of 67 histone marks and histone lysine crotonylation as a
new type of histone modification. Cell 2011, 146, 1016−1028.
(26) Heltweg, B.; Dequiedt, F.; Marshall, B. L.; Brauch, C.; Yoshida,
M.; Nishino, N.; Verdin, E.; Jung, M. Subtype selective substrates for
histone deacetylases. J. Med. Chem. 2004, 47, 5235−5243.
(27) He, B.; Du, J.; Lin, H. Thiosuccinyl peptides as SIRT5-specific
inhibitors. J. Am. Chem. Soc. 2012, 134, 1922−1925.
(2) Schemies, J.; Uciechowska, U.; Sippl, W.; Jung, M. NAD+-
dependent histone deacetylases (sirtuins) as novel therapeutic targets.
Med. Res. Rev. 2010, 30, 861−889.
(3) Denu, J. M. The Sir2 family of protein deacetylases. Curr. Opin.
Chem. Biol. 2005, 9, 431−440.
(4) Cole, P. A. Chemical probes for histone-modifying enzymes. Nat.
Chem. Biol. 2008, 4, 590−597.
(5) Haigis, M. C.; Sinclair, D. A. Mammalian sirtuins: biological
insights and disease relevance. Annu. Rev. Pathol.: Mech. Dis. 2010, 5,
253−295.
(6) Herranz, D.; Serrano, M. SIRT1: recent lessons from mouse
models. Nat. Rev. Cancer 2010, 10, 819−823.
(7) Blum, C. A.; Ellis, J. L.; Loh, C.; Ng, P.-Y.; Perni, R. B.; Stein, R.
L. SIRT1 modulation as a novel approach to the treatment of diseases
of aging. J. Med. Chem. 2011, 54, 417−432.
(28) Vaziri, H.; Dessain, S. K.; Ng Eaton, E.; Imai, S. I.; Frye, R. A.;
Pandita, T. K.; Guarente, L.; Weinberg, R. A. hSIR2(SIRT1) functions
as an NAD-dependent p53 deacetylase. Cell 2001, 107, 149−159.
(29) Solomon, J. M.; Pasupuleti, R.; Xu, L.; McDonagh, T.; Curtis,
R.; DiStefano, P. S.; Huber, L. J. Inhibition of SIRT1 catalytic activity
increases p53 acetylation but does not alter cell survival following
DNA damage. Mol. Cell. Biol. 2006, 26, 28−38.
(30) Smith, B. C.; Denu, J. M. Mechanism-based inhibition of Sir2
deacetylases by thioacetyl-lysine peptide. Biochemistry 2007, 46,
14478−14486.
(31) While the present work was in progress, a patent application
[WO2012006391 (2A), 2012] was published by Lin, H. Although,
similar fluorogenic succinyllysine-containing substrates were men-
tioned in that patent application, optimized conditions with respect to
enzyme loading and the secondary developer enzyme for continuous
mechanistic assays were not dealt with. Furthermore, the minimal
requirements of the peptide composition of the substrate were not
investigated; thus, the substrates reported in this work are considerably
more readily available by chemical synthesis.
(32) Xu, C.; Soragni, E.; Chou, C. J.; Herman, D.; Plasterer, H. L.;
Rusche, J. R.; Gottesfeld, J. M. Chemical probes identify a role for
histone deacetylase 3 in Friederich’s Ataxia gene silencing. Chem. Biol.
2009, 16, 980−989.
(33) (a) Chou, C. J.; Herman, D.; Gottesfeld, J. M. Pimelic
diphenylamide 106 is a slow, tight-binding inhibitor of class I histone
deacetylases. J. Biol. Chem. 2008, 283, 35402−35409. (b) Vickers, C.
J.; Olsen, C. A.; Leman, L. J.; Ghadiri, M. R. Discovery of HDAC
inhibitors that lack an active site Zn2+-binding functional group, ACS
Med. Chem. Lett. [Online early access]. DOI: 10.1021/ml300081u.
Published Online: Apr 26, 2012.
(34) Grozinger, C. M.; Chao, E. D.; Blackwell, H. E.; Moazed, D.;
Schreiber, S. L. Identification of a class of small molecule inhibitors of
the sirtuin family of NAD-dependent deacetylases by phenotypic
screening. J. Biol. Chem. 2001, 276, 38837−38843.
(35) Outeiro, T. F.; Kontopoulos, E.; Altmann, S. M.; Kufareva, I.;
Strathearn, K. E.; Amore, A. M.; Volk, C. B.; Maxwell, M. M.; Rochet,
J. C.; McLean, P. J.; Young, A. B.; Abagyan, R.; Feany, M. B.; Hyman,
B. T.; Kazantsev, A. G. Sirtuin 2 inhibitors rescue alpha-synuclein-
mediated toxicity in models of Parkinson’s disease. Science 2007, 317,
516−519.
(36) Trapp, J.; Meier, R.; Hongwiset, D.; Kassack, M. U.; Sippl, W.;
Jung, M. Structure−activity studies on suramin analogues as inhibitors
of NAD+-dependent histone deacetylases (sirtuins). ChemMedChem
2007, 2, 1419−1431.
(8) Milne, J. C.; Denu, J. M. The Sirtuin family: therapeutic targets to
treat diseases of aging. Curr. Opin. Chem. Biol. 2008, 12, 11−17.
(9) Jackson, M. D.; Schmidt, M. T.; Oppenheimer, N. J.; Denu, J. M.
Mechanism of nicotinamide inhibition and transglycosidation by Sir2
histone/protein deacetylases. J. Biol. Chem. 2003, 278, 50985−50998.
(10) Smith, B. C.; Denu, J. M. Acetyl-lysine analog peptides as
mechanistic probes of protein deacetylases. J. Biol. Chem. 2007, 282,
37256−37265.
(11) Smith, B. C.; Hallows, W. C.; Denu, J. M. Mechanisms and
molecular probes of sirtuins. Chem. Biol. 2008, 15, 1002−1013.
(12) Cen, Y.; Sauve, A. A. Transition state of ADP-ribosylation of
acetyllysine catalyzed by Archaeoglobus fulgidus Sir2 determined by
kinetic isotope effects and computational approaches. J. Am. Chem. Soc.
2010, 132, 12286−12298.
(13) Finnin, M. S.; Donigian, J. R.; Pavletich, N. P. Structure of the
histone deacetylase SIRT2. Nat. Struct. Biol. 2001, 8, 621−625.
(14) Schuetz, A.; Min, J.; Antoshenko, T.; Wang, C. L.; Allali-
Hassani, A.; Dong, A.; Loppnau, P.; Vedadi, M.; Bochkarev, A.;
Sternglanz, R.; Plotnikov, A. N. Structural basis of inhibition of the
human NAD+-dependent deacetylase SIRT5 by suramin. Structure
2007, 15, 377−389.
(15) Jin, L.; Wei, W.; Jiang, Y.; Peng, H.; Cai, J.; Mao, C.; Dai, H.;
Choy, W.; Bemis, J. E.; Jirousek, M. R.; Milne, J. C.; Westphal, C. H.;
Perni, R. B. Crystal structures of human SIRT3 displaying substrate-
induced conformational changes. J. Biol. Chem. 2009, 284, 24394−
24405.
(16) Min, J.; Landry, J.; Sternglanz, R.; Xu, R. M. Crystal structure of
a SIR2 homolog-NAD complex. Cell 2001, 105, 269−279.
(17) Avalos, J. L.; Celic, I.; Muhammad, S.; Cosgrove, M. S.; Boeke, J.
D.; Wolberger, C. Structure of a Sir2 enzyme bound to an acetylated
p53 peptide. Mol. Cell 2002, 10, 523−535.
(18) North, B. J.; Marshall, B. L.; Borra, M. T.; Denu, J. M.; Verdin,
E. The human Sir2 ortholog, SIRT2, is an NAD+-dependent tubulin
deacetylase. Mol. Cell 2003, 11, 437−444.
(19) Blander, G.; Guarente, L. The Sir2 family of protein
deacetylases. Annu. Rev. Biochem. 2004, 73, 417−435.
(20) Haigis, M. C.; Mostoslavsky, R.; Haigis, K. M.; Fahie, K.;
Christodoulou, D. C.; Murphy, A. J.; Valenzuela, D. M.; Yancopoulos,
G. D.; Karow, M.; Blander, G.; Wolberger, C.; Prolla, T. A.;
Weindruch, R.; Alt, F. W.; Guarente, L. SIRT4 inhibits glutamate
dehydrogenase and opposes the effects of calorie restriction in
pancreatic beta cells. Cell 2006, 126, 941−954.
(37) Fatkins David, G.; Monnot Andrew, D.; Zheng, W. N-epsilon-
Thioacetyl-lysine: a multi-facet functional probe for enzymatic protein
lysine N-epsilon-deacetylation. Bioorg. Med. Chem. Lett. 2006, 16,
3651−3656.
(38) Huhtiniemi, T.; Salo, H. S.; Suuronen, T.; Poso, A.; Salminen,
A.; Leppanen, J.; Jarho, E.; Lahtela-Kakkonen, M. Structure-based
design of pseudopeptidic inhibitors for SIRT1 and SIRT2. J. Med.
Chem. 2011, 54, 6456−6468.
(21) Du, J.; Zhou, Y.; Su, X.; Yu, J. J.; Khan, S.; Jiang, H.; Kim, J.;
Woo, J.; Kim, J. H.; Choi, B. H.; He, B.; Chen, W.; Zhang, S.; Cerione,
R. A.; Auwerx, J.; Hao, Q.; Lin, H. Sirt5 is a NAD-dependent protein
lysine demalonylase and desuccinylase. Science 2011, 334, 806−809.
(22) Olsen, C. A. Expansion of the lysine acylation landscape. Angew.
Chem., Int. Ed. 2012, 51, 3755−3756.
(23) Michishita, E.; McCord, R. A.; Berber, E.; Kioi, M.; Padilla-
Nash, H.; Damian, M.; Cheung, P.; Kusumoto, R.; Kawahara, T. L. A.;
5589
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590

Journal of Medicinal Chemistry
Article
(39) Borra, M. T.; Smith, B. C.; Denu, J. M. Mechanism of human
SIRT1 activation by resveratrol. J. Biol. Chem. 2005, 280, 17187−
17195.
(40) Kaeberlein, M.; McDonagh, T.; Heltweg, B.; Hixon, J.;
Westman, E. A.; Caldwell, S. D.; Napper, A.; Curtis, R.; DiStefano,
P. S.; Fields, S.; Bedalov, A.; Kennedy, B. K. Substrate-specific
activation of sirtuins by resveratrol. J. Biol. Chem. 2005, 280, 17038−
17045.
(41) Beher, D.; Wu, J.; Cumine, S.; Kim, K. W.; Lu, S. C.; Atangan,
L.; Wang, M. Resveratrol is not a direct activator of SIRT1 enzyme
activity. Chem. Biol. Drug Des. 2009, 74, 619−624.
(42) Pacholec, M.; Bleasdale, J. E.; Chrunyk, B.; Cunningham, D.;
Flynn, D.; Garofalo, R. S.; Griffith, D.; Griffor, M.; Loulakis, P.; Pabst,
B.; Qiu, X.; Stockman, B.; Thanabal, V.; Varghese, A.; Ward, J.;
Withka, J.; Ahn, K. SRT1720, SRT2183, SRT1460, and resveratrol are
not direct activators of SIRT1. J. Biol, Chem. 2010, 285, 8340−8351.
(43) Lineweaver, H.; Burk, D. The determination of enzyme
dissociation constants. J. Am. Chem. Soc. 1934, 56, 658−666.
(44) Tornheim, K. Kinetic applications using high substrate and
competitive inhibitor concentrations to determine Ki or Km. Anal.
Biochem. 1994, 221, 53−56.
(45) Wegener, D.; Wirsching, F.; Riester, D.; Schweinhorst, A. A
fluorogenic histone deacetylase assay well suited for high-throughput
activity screening. Chem. Biol. 2003, 10, 61−68.
5590
dx.doi.org/10.1021/jm300526r | J. Med. Chem. 2012, 55, 5582−5590