Substrate specificity of SIRT1-catalyzed lysine Nε-deacetylation reaction probed with the side chain modified Nε-acetyl-lysine analogs

Article abstract of DOI:10.1016/j.bioorg.2009.10.001

Peptides containing l-N^ε-acetyl-lysine (l-AcK) or its side chain modified analogs were prepared and assayed using SIRT1, the prototypical human silent information regulator 2 (Sir2) enzyme. While previous studies showed that the side chain acetyl group of l-AcK can be extended to bulkier acyl groups for Sir2 (including SIRT1)-catalyzed lysine N^ε-deacylation reaction, our current study suggested that SIRT1-catalyzed deacetylation reaction had a very stringent requirement for the distance between the α-carbon and the side chain acetamido group, with that found in l-AcK being optimal. Moreover, our current study showed that SIRT1 catalyzed the stereospecific deacetylation of l-AcK versus its d-isomer. The results from our current study shall constitute another piece of important information to be considered when designing inhibitors for SIRT1 and Sir2 enzymes in general.

Full text of DOI:10.1016/j.bioorg.2009.10.001

Bioorganic Chemistry 38 (2010) 17–25
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
e
Substrate specificity of SIRT1-catalyzed lysine N -deacetylation reaction
e
probed with the side chain modified N -acetyl-lysine analogs
1
*
Nuttara Jamonnak , Brett M. Hirsch, Yi Pang, Weiping Zheng
Department of Chemistry, University of Akron, 190 E. Buchtel Commons, Akron, OH 44325, USA
a r t i c l e i n f o
a b s t r a c t
e
Article history:
Received 23 July 2009
Available online 24 October 2009
Peptides containing L-N -acetyl-lysine (L-AcK) or its side chain modified analogs were prepared and
assayed using SIRT1, the prototypical human silent information regulator 2 (Sir2) enzyme. While previ-
ous studies showed that the side chain acetyl group of -AcK can be extended to bulkier acyl groups for
Sir2 (including SIRT1)-catalyzed lysine N -deacylation reaction, our current study suggested that SIRT1-
catalyzed deacetylation reaction had a very stringent requirement for the distance between the -carbon
and the side chain acetamido group, with that found in -AcK being optimal. Moreover, our current study
showed that SIRT1 catalyzed the stereospecific deacetylation of -AcK versus its -isomer. The results
L
e
Keywords:
a
e
N -acetyl-lysine
L
Analogs
SIRT1
Protein deacetylation
L
D
from our current study shall constitute another piece of important information to be considered when
designing inhibitors for SIRT1 and Sir2 enzymes in general.
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
an avenue for developing novel anti-cancer agents, and SIRT2 inhi-
bition constitutes a novel therapeutic approach to Parkinson’s
Silent information regulator 2 (Sir2) enzymes are a family of
evolutionarily conserved proteins that are present in all kingdoms
of life, with the yeast Sir2 being the founding member [1–3]. In
humans, seven Sir2 orthologues, i.e. SIRT1-7, have been identified.
Due to their well-conserved catalytic domains, Sir2 enzymes are all
disease.
During the past a few years, several classes of human Sir2
enzyme (especially SIRT1 and SIRT2) inhibitors have been identi-
fied [14,19,23,24], however, their inhibitory profiles (potency and
selectivity) among all the human Sir2 enzymes are currently un-
known. Moreover, the mechanism(s) of inhibition for the majority
of currently identified human Sir2 inhibitors still awaits further
elucidation. Therefore, developing novel inhibition strategies and
inhibitors for the human Sir2 enzymes still constitutes a research
area with tremendous current interest.
expected to catalyze the nicotinamide adenine dinucleotide
e
(NAD+)-dependant deacetylation of
L
-N -acetyl-lysine (
L
-AcK) on
e
protein substrates, with the formation of lysine N -deacetylated
protein species and small molecule products, i.e. nicotinamide
and 20-O-acetyl-ADP-ribose which will be non-enzymatically
equilibrated with 30-O-acetyl-ADP-ribose in bulk solution (Fig. 1,
R = CH3) [4–7]. Many acetylated protein substrates for the human
Sir2 enzymes have been identified, such as the core histone pro-
Even though a Sir2 enzyme is a deacetylase enzyme in that it can
catalyze the removal of the side chain acetyl group of
R = CH3), a robust protein depropionylase activity to remove the side
chain propionyl group of -propionyl-lysine (Fig. 1, R = CH2CH3) and
a robust protein debutyrylase activity to remove the side chain bu-
tyryl group of -butyryl-lysine (Fig. 1, R = CH2CH2CH3) have also
L-AcK (Fig. 1,
teins, various transcription factors,
a-tubulin, acetyl-coenzyme A
L
synthetase, and the Tat protein of human immunodeficiency virus
type 1 (HIV-1). This deacetylation reaction has been shown to play
a critical role in regulating multiple important biological processes
such as transcriptional silencing, DNA repair, apoptosis, metabo-
lism, aging, and HIV-1 infection [3,8–13]. As such, chemical modu-
lation of the Sir2 deacetylase activity could offer therapeutic
benefits for treating human diseases [3,14–22]. Indeed, genetic
and pharmacological studies have suggested that human SIRT1
activation may alleviate type 2 diabetes, SIRT1 inhibition may be
L
been recently reported for several Sir2 enzymes, including the hu-
man SIRT1 [25,26], the yeast Hst2 [26,27], and the bacterial enzymes
Sir2Tm [28] and CobB [29]. This steric tolerance of a bulky acyl group
has also been very recently manifested in the study by Asaba et al.
[30] in which a stable potent and selective ‘‘bisubstrate-type” SIRT1
inhibitor was generated in situ from a
group even bulkier than the butyryl group. While it is now known
that the side chain acetyl group of -AcK can be extended to bulkier
L-AcK analog that bore an acyl
L
e
acyl groups for Sir2-catalyzed lysine N -deacylation reaction, in the
current study, we were interested in examining whether the part of
* Corresponding author.
E-mail address: wzheng@uakron.edu (W. Zheng).
Present address: Department of Molecular Biology and Genetics, Johns Hopkins
1
the side chain of
L-AcK between the a-carbon and acetamide can tol-
erate structural perturbation for Sir2-catalyzed deacetylation
University School of Medicine, 725 N. Wolfe Street, Baltimore, MD 21205, USA.
0045-2068/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2009.10.001

18
N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
NH2
O
NAD+
N
1'
O
ADP
R
HO
R
HO
O
O
NH
H
H
O
N
H
O
1'
O
N
H
ADP
NH
Acylated substrate
O
N
O-alkylamidate
intermediate
O
H
HO
Nicotinamide
NH2
B:
O
H
Deacylated product
H3N
NH
O
ADP
1'
O
O
R
O
NH
H
O
ADP
O
1'
OH
N
H
O
ADP
1' OH
HO
O
Bicyclic intermediate
Non-enzymatic
equilibration
OH
O
O
HO
R
R
3'-O-acyl-ADP-ribose
2'-O-acyl-ADP-ribose
e
Fig. 1. The proposed chemical mechanism of the Sir2-catalyzed NAD+-dependent lysine N -deacylation on protein substrates. The C10 position is indicated. ADP, adenosine
diphosphate. R is CH3, CH2CH3, or CH2CH2CH3.
reaction. For this purpose, we prepared a series of the peptides that
contained different -AcK analogs with subtly different distances be-
tween the -carbon and the side chain acetamido group (Fig. 2). We
showed that, as compared to the -AcK-containing peptide, none of
these peptides were able to be deacetylated under our experimental
conditions by SIRT1, the model human enzyme that we chose for our
study. These results suggested that the SIRT1-catalyzed deacetyla-
tion reaction had a very stringent requirement for the distance
dioxane, silica gel (70–230 mesh, 60 Å), N-methylmorpholine
(NMM), piperidine, phenol, thioanisole, ethanedithiol; Alfa Aeser:
L
a
acetamidomethanol; Acros:
L-homoserine; EMD biosciences:
L
dichloromethane, trifluoroacetic acid (TFA), N,N-dimethylformam-
a
ide (DMF), acetonitrile; Bachem:
Fmoc-protected amino acids except the analogs of
D-AcK; Novabiochem: all N -
a
L
-N -Fmoc-AcK
that were synthesized in the current study, resins for the solid phase
peptidesynthesis(SPPS), 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetram-
ethylaminium hexafluorophosphate (HBTU, the coupling reagent
used for SPPS), N-hydroxybenzotriazole (HOBt, the additive used
between the
a-carbon and the side chain acetamido group, with that
found in -AcK being optimal. The results from our current study
L
shall constituteanotherpiece of importantinformationto be consid-
ered when designing Sir2 enzyme inhibitors.
for
SPPS),
N-(9-fluorenylmethyloxy-carbonyloxy)succinimide
(Fmoc-OSu); Fisher: sodium carbonate, anhydrous diethyl ether.
1H and 13C NMR spectra were obtained on a Varian Mercury 300
spectrometer or a Varian GEMINI 300 spectrometer. Mass spectra
were recorded on a Bruker Reflex-III MALDI-TOF mass spectrome-
ter or a Bruker Esquire-LC ion trap mass spectrometer (with elec-
trospray ionization). High resolution mass spectra (HRMS) were
obtained on a VG ZAB high resolution mass spectrometer at the
high resolution mass spectrometry facility of the University of Cal-
ifornia, Riverside.
2. Experimental
2.1. General
The following are the materials obtained from respective vendors
a
for the synthesis of N -Fmoc-protected amino acids and peptides
The following are the materials obtained for the expression and
purification of GST-SIRT1 and the SIRT1 assays used in the current
used in the current study. All the materials were used as received
without further purification. Sigma–Aldrich: DL-homocysteine, 1,4-
H3C
HN
S
H2N
S
H3C
HN
H3C
CH3
NH
O
O
O
CH3
NH
CH3
NH
O
O
O
O
HN
O
S
H
H
H
H
O
H
OH
OH
OH
OH
OH
OH
OH
H2N
DL-S-AcK
H2N
H2N
H2N
H2N
H2N
DL-S-Car-K
H2N
D-AcK
O
O
O
O
O
O
L-CysAcm
L-AcK
L-AcOrn
L-O-AcK
Peptide 1: H2N-HK-(L-AcK)-LM-COOH
Peptide 2: H2N-HK-(L-AcOrn)-LM-COOH
Peptide 7: H2N-QP-(D-S-AcK)-QI-COOH
Peptide 8: H2N-QP-(L-S-AcK)-QI-COOH
Peptide 3: H2N-HK-(L-CysAcm)-LM-COOH Peptide 9: AcNH-HK-(D-AcK)-LM-CONH2
Peptide 4: H2N-HK-(L-O-AcK)-LM-COOH
Peptide 5: H2N-HK-(D-S-AcK)-LM-COOH
Peptide 6: H2N-HK-(L-S-AcK)-LM-COOH
Peptide 10: H2N-HK-(L-S-Car-K)-LM-COOH
Peptide 11:H2N-HK-(D-S-Car-K)-LM-COOH
Fig. 2.
L
-AcK and its side chain modified analogs, and the corresponding peptides used in the current study. Peptides 5 and 7 were arbitrarily assigned to harbor
D-S-AcK, and
peptides 6 and 8 were arbitrarily assigned to harbor
L
-S-AcK. Peptide 10 was arbitrarily assigned to harbor
L
-S-Car-K, and peptide 11 was arbitrarily assigned to harbor -S-
D
Car-K. Ac, acetyl.

N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
19
study: Plasmid pGEX2TK-PꢀSIRT1 (human full length) was a kind
gift from Prof. Tony Kouzarides. Escherichia coli strain BL21(DE3)
was purchased from Stratagene. Tryptone, yeast extract, isopro-
pyl-1-thio-b-D-galacto-pyranoside (IPTG), ampicillin, NaCl, KCl,
and NaH2PO4 were purchased from Fisher. Bacto Agar was pur-
chased from BD Biosciences. Glycerol was from Acros. DL-dithio-
(5 mL)anda10%(w/v)aqueousNa2CO3 solution(5 mL). To the result-
ing stirred solution was added dropwise a solution of Fmoc-OSu
(675 mg, 2 mmol) in 1,4-dioxane (5 mL) at room temperature. After
the addition was complete, the reaction mixture was stirred at room
temperature for 5 h before ddH2O (50 mL) was added. The excess
Fmoc-OSu was extracted away with diethyl ether (2 ꢁ 100 mL). The
aqueous layer was acidified with a 6.0 M aqueous HCl solution to
pH ꢂ 1 at 0 °C before being extracted with ethyl acetate
(3 ꢁ 100 mL). The organics were combined, dried with anhydrous
Na2SO4, filtered, and concentrated under reduced pressure, affording
a white solid residue from which the desired product was isolated by
reversed-phase high pressure liquid chromatography (HPLC) on a
preparative C18 column (100 Å, 2.14 ꢁ 25 cm). The HPLC column
was eluted with a gradient of ddH2O containing 0.05% (v/v) of TFA
and acetonitrile containing 0.05% (v/v) of TFA. The pooled HPLC frac-
tions were stripped of acetonitrile and lyophilized to afford the de-
sired product as a white solid (260 mg, 61% after two steps): 1H
NMR (300 MHz, CDCl3): d (ppm) 7.78–7.23 (m, 8H, Harom), 4.54–
4.22 (m, 4H, Fluorenyl H9, CH2O, and Halpha), 2.82–2.09 (m, 8H,
CH2CH2-S-CH2CH2); 13C NMR (75 MHz, CD3OD): d (ppm) 174.4
(COOH), 174.1 (C(@O)NH2), 155.6 (NHC(@O)O), 143.9 (Carom), 141.4
(Carom), 127.6 (Carom), 127.0 (Carom), 125.1 (Carom), 119.7 (Carom), 66.8
(CH2O), 53.0 (Calpha), 47.0 (Fluorenyl C9), 35.6 (CH2), 31.5 (CH2), 28.0
(CH2), 27.0 (CH2); HRMS (FAB) calcd. for C22H25N2O5S ([M + H]+)
429.1471; found: 429.1468.
threitol (DTT) was from EMD biosciences. Glutathione–agarose, L-
glutathione reduced, trizma, b-NAD+, nicotinamide, guanidinium
chloride, and a 1.0 M solution of MgCl2 (molecular-biology grade)
were purchased from Sigma. All the materials obtained from com-
mercial sources were used as received without further purification.
2.2. Molecular modeling
The HyperChemÒ 8.0 molecular modeling software was used to
build and optimize the molecular geometry of
L-AcK and each of its
L-analogs with the Austin Model 1 (AM1) setting. The atom-to-
atom distance was directly measured from the optimized struc-
ture. The HOMOs were also generated from the optimized molecu-
lar structures.
a
2.3. Synthesis of DL-N -Fmoc-S-AcK
(a) To a stirred solution of DL-homocysteine (135 mg, 1 mmol) in
TFA (4.5 mL) was added in three portions (at 0, 15, and 30 min) of
acetamidomethanol (3 ꢁ 30 mg, 1 mmol) at room temperature.
After the third portion was added, the reaction mixture was stirred
at room temperature for another 15 min before TFA was removed
under reduced pressure. (b) To the resulting oily residue was added
double deionized water (ddH2O) (5 mL), and the acidic solution was
neutralized at 0 °C while stirring with a 10% (w/v) aqueous Na2CO3
solution. Another portion of the 10% (w/v) Na2CO3 solution (5 mL)
was further added, and to the resulting solution was added dropwise
a solution of Fmoc-OSu (675 mg, 2 mmol) in 1,4-dioxane (5 mL) at
room temperature. After the addition was complete, the reaction
mixture was stirred at room temperature for 5 h before ddH2O
(50 mL) was added. The excess Fmoc-OSu was extracted away with
diethyl ether (2 ꢁ 100 mL). The aqueous layer was acidified with a
6.0 M aqueous HCl solution to pH ꢂ 1 at 0 °C before being extracted
with ethyl acetate (3 ꢁ 100 mL). The organics were combined, dried
with anhydrous Na2SO4, filtered, and concentrated under reduced
pressure, affording a white solid residue from which the desired
product was isolated via silica gel column chromatography as a
white solid (270 mg, 63% after two steps): 1H NMR (300 MHz,
CDCl3): d (ppm) 7.78–7.19 (m, 8H, Harom), 4.38–4.02 (m, 6H, Fluore-
nyl H9, CH2O, CH2N and Halpha), 2.54–1.99 (m, 4H, CH2CH2), 1.91 (s,
3H, CH3); 13C NMR (75 MHz, CDCl3): d (ppm) 175.1 (COOH), 173.2
(C(@O)NH), 156.9 (NHC(@O)O), 143.6 (Carom), 141.5 (Carom), 128.0
(Carom), 127.3 (Carom), 125.2 (Carom), 120.3 (Carom), 67.5 (CH2O), 60.1
(Calpha), 47.1 (Fluorenyl C9), 41.4 (CH2NH), 32.5 (CH2), 27.4 (CH2),
22.6 (CH3); HRMS (FAB) calcd. for C22H25N2O5S ([M + H]+)
429.1484; found: 429.1481.
a
2.5. Synthesis of
L
-N -Fmoc-O-AcK
Essentially the same synthetic procedure as that described
a
above for the synthesis of DL-N -Fmoc-S-AcK was followed except
that L-homoserine (119 mg, 1 mmol) was used instead of DL-homo-
cysteine. After two steps of reactions (Scheme 2), the desired prod-
uct was isolated via silica gel column chromatography as a white
solid (180 mg, 44% after two steps): 1H NMR (300 MHz, CDCl3): d
(ppm) 7.44–6.98 (m, 8H, Harom), 4.31–3.88 (m, 6H, Fluorenyl H9,
CH2O, OCH2N and Halpha), 3.27–1.74 (m, 4H, CH2CH2), 1.69 (s, 3H,
CH3); 13C NMR (75 MHz, CDCl3): d (ppm) 175.2 (COOH), 173.0
(C(@O)NH), 156.7 (NHC(@O)O), 143.8 (Carom), 141.4 (Carom), 128.0
(Carom), 127.3 (Carom), 125.3 (Carom), 120.2 (Carom), 70.4 (OCH2NH),
67.3 (CH2O), 64.7 (side chain CH2O), 58.7 (Calpha), 47.2 (Fluorenyl
C9), 31.7 (CH2), 25.5 (CH3); HRMS (FAB) calcd. for C22H25N2O6
([M + H]+) 413.1713; found: 413.1710.
a
2.6. Synthesis of
D
-N -Fmoc-AcK
To a stirred solution of D-AcK (188 mg, 1 mmol) in aqueous 10%
(w/v) Na2CO3 (10 mL) was added dropwise a solution of Fmoc-OSu
(675 mg, 2 mmol) in 1,4-dioxane (5 mL) at room temperature. After
the addition was complete, the reaction mixture was stirred at room
temperature for 5 h before ddH2O (50 mL) was added. The excess
Fmoc-OSu was extracted away with diethyl ether (2 ꢁ 100 mL).
The aqueous layer was acidified with a 6.0 M aqueous HCl solution
to pH ꢂ 1 at 0 °C before being extracted with ethyl acetate
(3 ꢁ 100 mL). The organics were combined, dried with anhydrous
Na2SO4, filtered, and concentrated under reduced pressure, affording
an oily residue from which the desired product was isolated via silica
gel column chromatography as a white solid (231 mg, 56%): 1H NMR
(300 MHz, CD3OD): d (ppm) 7.79–7.26 (m, 8H, Harom), 4.34–4.07 (m,
4H, Fluorenyl H9, CH2O, and Halpha), 3.14 (t, J = 6.6 Hz, 2H, CH2N), 1.89
(s, 3H, CH3), 1.89–1.32 (m, 6H, CH2CH2CH2); 13C NMR (75 MHz,
CDCl3): d (ppm) 172.5 (COOH), 164.2 (C(@O)NH), 158.0 (NHC(@O)O),
144.5(Carom), 141.9 (Carom), 128.1 (Carom), 127.5 (Carom), 125.6 (Carom),
120.2 (Carom), 67.2 (CH2O), 54.9 (Calpha), 47.4 (Fluorenyl C9), 39.6
(CH2N), 29.2 (CH2), 25.6 (CH2), 23.6 (CH2), 21.8 (CH3); MS (ESI): m/z
411 [M + H]+, 433 [M + Na]+. HRMS (FAB) calcd. for C23H25N2O5
([MꢃH]ꢃ) 409.1758; found: 409.1759.
a
2.4. Synthesis of DL-N -Fmoc-S-Car-K
(a) To a stirred solution of 3-bromopropionamide (152 mg,
1 mmol) in freshly degassed methanol (33 mL) was added dropwise
atroomtemperatureasolutionofDL-homocysteine(135 mg, 1 mmol)
in a 1.0 M aqueous buffer solution of triethylammonium bicarbonate
(pH 8.1–8.5) (33 mL). After the addition was complete, the reaction
mixture was stirred at room temperature overnight under nitrogen
before methanol was removed under reduced pressure. To the result-
ing residue was addedethyl acetate (60 mL) to extractawaythe unre-
acted 3-bromopropionamide. The aqueous phase was then
lyophilized overnight, affording a white solid residue. (b) To the
above-obtained white solid residue was added while stirring ddH2O

20
N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
2.7. Peptide synthesis and purification
incubated at 37 °C until quenched at different time points (0, 30,
60, and 120 min) with the following stop solution: 3.2 M guanidi-
All peptides were synthesized using the Fmoc chemistry-based
SPPS [31] on a PS3 peptide synthesizer (Protein Technologies Inc.,
nium chloride in 0.1 mM of sodium phosphate buffer (pH 6.8). One
portion (40
lL) of the SIRT1 assay mixture was quenched with
Tucson, AZ, USA). One to two equivalents of the analogs of
80 L of the above stop solution and was analyzed by reversed-
l
a
L
-N -Fmoc-AcK that were synthesized in the current study (i.e.
phase HPLC with a C18 analytical column (100 Å, 0.46 ꢁ 25 cm).
For assessing the possible SIRT1-catalyzed deacetylation, the C18
analytical column was eluted with the following gradient of ddH2O
containing 0.05% (v/v) TFA (mobile phase A) and acetonitrile con-
taining 0.05% (v/v) TFA (mobile phase B): linear increase from 0%
B to 50% B from 0 to 40 min (1 mL/min), and ultraviolet (UV) mon-
itoring at 214 nm. For assessing the possible SIRT1-catalyzed nico-
tinamide production, the C18 analytical column was eluted with
the following gradient of mobile phase A and mobile phase B: lin-
ear increase from 0% B to 15% B from 0 to 40 min (1 mL/min), and
monitored with UV at 261 nm.
a
a
a
DL-N -Fmoc-S-AcK, DL-N -Fmoc-S-Car-K,
L
-N -Fmoc-O-AcK, and
-N -Fmoc-AcK) were used in 9-h coupling reactions to incorporate
the respective -AcK analogs into peptides. However, for all other
a
D
L
a
N -Fmoc-protected amino acids, four equivalents were used in 1-h
coupling reactions. All the peptides were cleaved from the resin by
reagent K (83.6% (v/v) TFA, 5.9% (v/v) phenol, 4.2% (v/v) ddH2O,
4.2% (v/v) thioanisole, 2.1% (v/v) ethanedithiol), precipitated in cold
diethyl ether, and purified by reversed-phase HPLC on a preparative
C18 column (100 Å, 2.14 ꢁ 25 cm). The HPLC column was eluted
with a gradient of ddH2O containing 0.05% (v/v) of TFA and acetoni-
trile containing 0.05% (v/v) of TFA. The pooled HPLC fractions were
stripped of acetonitrile and lyophilized to give all peptides as puffy
white solids. Peptide purity (>95%) was verified by reversed-phase
HPLC on an analytical column (100 Å, 0.46 ꢁ 25 cm), and their
molecular weights were confirmed by either matrix assisted laser
desorption ionization-time of flight (MALDI-TOF) or electrospray
ionization (ESI) mass spectrometric analysis.
2.10. HPLC-based SIRT1 inhibition assay
A SIRT1 inhibition assay solution (100 lL) contained the fol-
lowing components: 25 mM TrisꢀHCl (pH 8.0), 137 mM NaCl,
2.7 mM KCl, 1 mM MgCl2, 0.5 mM b-NAD+, 0.1 mM peptide 1,
350 nM GST-SIRT1, and a pentapeptide (peptides 5, 6, 10, or 11)
varied from 0 to 1.6 mM. An enzymatic reaction was initiated
by the addition of GST-SIRT1 at 37 °C and was incubated at
2.8. Expression and purification of GST-SIRT1
37 °C for 10 min before quenched with 25 lL of a stop solution
The following procedure was derived with modifications from
those described previously [32–34]. Following the transformation
of pGEX2TK-P ꢀ SIRT1 (human full length) into the Escherichia coli
strain BL21(DE3), one of the resulting colonies was used to inoc-
ulate a 10-mL Luria Broth containing 100 lg/mL of ampicillin, and
the culture was grown at 37 °C overnight (18 h). This culture was
composed of 0.1 M HCl and 0.16 M acetic acid. The enzymatic
deacetylation product (i.e. peptide 1a in Fig. 5) was detected
and quantified by reversed-phase HPLC with a C18 analytical col-
umn (100 Å, 0.46 ꢁ 25 cm), eluting with the following gradient of
ddH2O containing 0.05% (v/v) TFA (mobile phase A) and acetoni-
trile containing 0.05% (v/v) TFA (mobile phase B): linear increase
from 0% B to 50% B from 0 to 40 min (1 mL/min), and UV moni-
toring at 214 nm.
subsequently used to inoculate a 1-L Luria Broth also containing
100 lg/mL of ampicillin, and the culture was grown at 37 °C until
the optical density (OD) at 600 nm reached 0.6. Fresh 1.0 M IPTG
was then added to the culture to a final concentration of 0.3 mM,
and the culture was grown at 16 °C for an additional 20 h. The
cells were harvested by centrifugation at 3500 rpm and 4 °C for
20 min, and the pellet was resuspended in 25 mL of the ice-cold
lysis buffer (25 mM sodium phosphate buffer (pH 7.4), 130 mM
NaCl, 5 mM DTT, 10% (v/v) glycerol, and 1 mM phenylmethylsul-
phonyl fluoride (PMSF)). Resuspended cells were lysed by double
passage through a French pressure cell at 700 psi and centrifuged
at 15,000 rpm and 4 °C for 30 min to remove cell debris. The
supernatant was loaded onto a plastic column packed with
350 mg of the glutathione–agarose that had been pre-swollen in
30 mL of ddH2O and equilibrated with 50 mL of the ice-cold lysis
buffer without PMSF. The glutathione–agarose beads were then
washed with 40 mL of a high salt buffer (25 mM sodium phos-
phate buffer (pH 7.4), 387 mM NaCl, 5 mM DTT, 10% (v/v) glyc-
erol). The GST-SIRT1 protein was eluted off of the column with
the elution buffer (25 mM sodium phosphate buffer (pH 7.4),
3. Results and discussion
3.1. Synthesis of the side chain modified
corresponding peptides
L-AcK analogs and the
All the peptides used in the current study were synthesized
using the Fmoc chemistry-based SPPS [31] (see Section 2). In or-
der to incorporate L-AcK and its analogs into peptides, their cor-
a
responding N -Fmoc-protected derivatives were either obtained
from commercial sources if available or synthesized as described
a
a
a
below.
that were used to incorporate respectively
-CysAcm into peptides 1, 2, and 3 were commercially available.
L
-N -Fmoc-AcK,
L
-N -Fmoc-AcOrn, and
L
-N -Fmoc-CysAcm
L
-AcK, -AcOrn, and
L
L
a
a
Scheme 1 depicts the synthesis of DL-N -Fmoc-S-AcK and DL-N -
Fmoc-S-Car-K. Of note, DL-N -Fmoc-S-AcK was synthesized using
a modified literature procedure [35]. Even though L-homocysteine
a
241 mM NaCl, 5 mM DTT, 50 mM
L
-glutathione reduced, 10% (v/
is also commercially available, we expected that racemization of
this starting material and the product might occur under the gi-
ven reaction condition (room temperature, overnight, pH 8.1–
v) glycerol). The pooled fractions (ꢂ10 mL) were dialyzed against
the storage buffer (25 mM TrisꢀHCl (pH 7.5), 100 mM NaCl, 5 mM
DTT, 10% (v/v) glycerol), aliquoted, snap frozen with liquid N2,
and stored at ꢃ80 °C. The protein concentration (1.5 mg/mL)
was determined by the Bradford assay.
a
8.5) during the first step of the synthesis of DL-N -Fmoc-S-Car-
K, therefore, we used the commercially available racemic starting
material (i.e. DL-homocysteine) for the synthesis. DL-homocysteine
a
was also used to synthesize DL-N -Fmoc-S-AcK. However, even
a
a
2.9. HPLC-based SIRT1 time-course assay
though DL-N -Fmoc-S-AcK and DL-N -Fmoc-S-Car-K were obtained
in our synthesis, following their use to incorporate DL-S-AcK and
DL-S-Car-K into peptides, we were able to, in each case, separate
A SIRT1 time-course assay solution contained the following
components: 25 mM TrisꢀHCl (pH 8.0), 137 mM NaCl, 2.7 mM
KCl, 1 mM MgCl2, 0.5 mM b-NAD+, 350 nM GST-SIRT1, and
0.3 mM of a pentapeptide (i.e. peptides 1–11). An enzymatic reac-
tion was initiated by the addition of GST-SIRT1 at 37 °C and was
the two diastereomeric peptides harboring either the
D- or the
L
-enantiomer of an analog with HPLC, affording peptides 5–8,
a
10, and 11. Scheme 2 depicts the synthesis of
and
L
-N -Fmoc-O-AcK
a
D
-N -Fmoc-AcK that were used to incorporate respectively

N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
21
H3C
HN
S
O
H3C
O
HN
HS
S
O
i
ii
OH
N
H
O
OH
OH
O
H2N
H2N
O
O
DL-Nα-Fmoc-S-AcK
DL-Homocysteine
H2N
O
H2N
O
S
S
HS
O
iii
ii
OH
N
H
O
OH
OH
O
H2N
H2N
O
O
DL-Nα-Fmoc-S-Car-K
DL-Homocysteine
Scheme 1. Reagents: (i) acetamidomethanol, TFA. (ii) Fmoc-OSu, 1,4-dioxane, 10% (w/v) aq. Na2CO3. (iii) 3-Bromopropionamide, degassed MeOH, 1.0 M Et3NHþ ꢀ HCO3ꢃ (pH
8.1–8.5).
H3C
O
H3C
HN
O
HN
O
O
HO
H
O
ii
i
OH
H
H
N
H
O
OH
OH
O
H2N
H2N
O
O
L-Nα-Fmoc-O-AcK
L-Homoserine
CH3
NH
O
CH3
O
NH
H
O
ii
OH
H
N
O
OH
H
O
H2N
O
D-Nα-Fmoc-AcK
D-AcK
Scheme 2. Reagents: (i) acetamidomethanol, TFA. (ii) Fmoc-OSu, 1,4-dioxane, 10% (w/v) aq. Na2CO3.
L
-O-AcK and
the S-acetamidomethyl (S-Acm)-containing side chains of
sAcm and -(or )-S-AcK as well as the O-acetamidomethyl (O-
Acm)-containing side chain of -O-AcK were observed to be
D
-AcK into peptides 4 and 9. It should be noted that
3.2. Design of the side chain modified
corresponding peptides
L-AcK analogs and the
L
-Cy-
D
L
L
3.2.1. Peptide sequence selection
chemically stable enough to survive our experimental conditions
for the peptide synthesis and purification, as well as the SIRT1 as-
says. This chemical stability is consistent with the fact that Acm
can be used as an acid/base stable thiol protecting group (e.g.
for cysteine) during SPPS [36] and with the literature report of
the peptides containing the O-Acm group on serine and threonine
side chains, as well as the acid/base stable serine analog whose
side chain contained O-Acm [37].
Fig. 2 shows the structures of L-AcK and its side chain modified
analogs, as well as the corresponding peptides used in the current
study. Peptides 1–6 and 9–11 were constructed based on the se-
quence (amino acid residue 380–384) derived from the C-terminal
region of the human p53 protein [38]. Since the 382 position was
shown to be the in vivo SIRT1 deacetylation site of the human
p53 protein [39],
L-AcK or its analogs were thus each incorporated
into this position in a peptide. Furthermore, the 5-residue peptide

22
N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
(i.e. peptide 1) was shown by Garske et al. [40] and our current
study to serve as a good SIRT1 substrate, therefore, we also used
this 5-residue sequence to construct peptides 2–6 and 9–11. Since
Among all the side chain structures shown in Fig. 3, only one is
not terminated with the acetamido group, that is the side chain of
L
-S-Car-K that is terminated with a primary carboxamide. How-
ever, the distance between the -carbon and the side chain car-
bonyl oxygen atom of this analog (i.e. 7.78 Å) is the closest
distance to that in -AcK (7.64 Å) among all the distances shown
in Fig. 3, and because of this, the peptide containing this analog
(i.e. peptide 10, Fig. 2) and the peptide containing its -isomer
(i.e. peptide 11, Fig. 2) were also prepared (see Section 3.1) and as-
sayed with SIRT1 (see Section 3.3). Even though the side chain of
Garske et al. also reported previously that H2N-QP-(
L
-AcK)-QI-
a
COOH was a ꢂ9-fold better SIRT1 substrate than H2N-HK-(
L-
AcK)-LM-COOH, as judged by kcat/Km ratios [40], peptides 7 and 8
were also prepared in the current study to examine whether the
use of a better SIRT1 peptide substrate sequence could have an ef-
L
D
fect on SIRT1 processing of a given L-AcK analog.
L
-
S-Car-K terminates with a primary carboxamide instead of an acet-
amide, the oxygen atom of the carboxamide is presumably able to
function similarly to that of acetamide for the nucleophilic attack
at the C10 atom of NAD+ (Fig. 1) as long as the oxygen atom is posi-
tioned correctly relative to the C10 atom of NAD+.
3.2.2. Design of
3.2.2.1. -Analogs with varied side chain carbonyl oxygen-
distances. As described in Section 3.3, deleting one methylene
group from the side chain of -AcK seemed to be a too drastic side
chain length perturbation for the SIRT1-catalyzed deacetylation
reaction to take place, since the SIRT1-catalytzed deacetylation
L-AcK analogs
L
a carbon
L
All together, two L-analogs with longer distances and three
L-analogs with shorter distances as shown in Fig. 3 were designed.
was not observed for the peptide containing
-AcOrn) (i.e. peptide 2, Fig. 2) under our experimental conditions,
which is in stark contrast to a robust SIRT1-catalyzed deacetylation
of the -AcK-containing peptide (i.e. peptide 1, Fig. 2). Therefore,
we were interested in coming up with further -AcK analogs whose
distances between the -carbon and the side chain acetamide
will be between that in -AcOrn and that in -AcK. In addition,
we were also interested in evaluating those side chains that are
L-acetyl-ornithine
It should be noted that, in order to subtly change the length be-
(L
tween the
S and O atoms as the isosteric replacements for CH2. In the side
chain of -S-Car-K, we also made the isosteric swapping of the
a-carbon and the side chain acetamide of L-AcK, we used
L
L
L
NH and CH3 groups of the acetamide.
a
L
L
3.2.2.2. Predicted similar side chain carbonyl oxygen nucleophilicity in
longer than that of
lene group.
L
-AcK but less than lengthening by one methy-
L
-AcK and its
the more electronegative heteroatoms S and O (than C) in the de-
signed -analogs may alter the nucleophilicity of the nearby car-
L
-analogs. It should be noted that the introduction of
To help guide our design of such analogs, we used the Hyper-
ChemÒ software to optimize the molecular geometry of
-AcK
and its analogs and then to compare the distance between the
-carbon and the side chain carbonyl oxygen atom from each geo-
L
L
bonyl oxygen atom which, if positioned correctly, will serve as a
nucleophile to attack the electrophilic C10 atom of NAD+, as shown
in Fig. 1. However, when we calculated the highest occupied
molecular orbitals (HOMOs) involving the side chain carbonyl oxy-
a
metrically optimized structure. Of note, we used this distance as
the parameter of comparison because the side chain carbonyl oxy-
gen atoms for
acetamide analogs (i.e.
lobes were observed on the carbonyl oxygen atoms among these
three structures (Fig. 4), suggesting that very similar nucleophilic-
ity is expected when the carbonyl oxygen attacks the C10 atom of
NAD+. In our opinion, the three HOMOs shown in Fig. 4 should
L
-AcK and its two heteroatom (O or S)-containing
-O-AcK and -S-AcK), very similar orbital
gen atom of
NAD+ during the first chemical step of Sir2 enzymatic catalysis
(Fig. 1). Fig. 3 shows the optimized geometry of -AcK and its -ana-
logs, as well as the distance between the -carbon and the side
chain carbonyl oxygen atom from each geometrically optimized
structure. From shortest to longest, these distances are: -AcOrn,
6.62 Å; -CysAcm, 6.80 Å; -O-AcK, 7.47 Å; -AcK, 7.64 Å; -S-Car-
K, 7.78 Å; -S-AcK, 8.00 Å. Since one methylene group is worth of
1.02 Å which is the difference between the distances in -AcK
and -AcOrn (i.e. 7.64–6.62 Å), the above distances form a fine-
tuned spectrum within one methylene group deviation from the
side chain of -AcK. It is apparent from Fig. 3 that all of the geomet-
rically optimized side chains assume the fully extended conforma-
tion, thus mimicking that of the side chain of -AcK bound to its
L
-AcK serves as a nucleophile to attack the C10 atom of
L
L
L
L
a
L
be able to serve as the representatives for all the
in Fig. 3. Specifically, the HOMO of -AcK will also represent those
of -S-Car-K and -AcOrn, and the HOMO of -S-AcK will also repre-
sent that of -CysAcm. It is also noted from Fig. 4 that the lone pair
L-analogs shown
L
L
L
L
L
L
L
L
L
L
L
L
electrons on the heteroatom near the center of the molecule tend
to contribute to the HOMO through hyperconjugation with adja-
cent C–H bonds. Although this heteroatom impact is evident in
L
the HOMO of
appears to be minimal (in terms of the orbital lobe size on the car-
bonyl oxygen). In the case of -S-AcK, contribution to HOMO from
L-O-AcK, its effect on the side chain carbonyl oxygen
L
hydrophobic tunnel at a Sir2 enzyme active site [41,42]. In our
opinion, this conformational mimicry makes the above distance
comparison a reasonable reflection of the corresponding distance
differences among the side chains if they are able to bind to the
SIRT1 active site.
L
the sulfur atom becomes negligible, partly attributing to the longer
C–S bond which weakens the hyperconjugation with C–H bonds.
The above modeling results thus suggested that the side chain
carbonyl oxygen atoms in L-AcK and its L-analogs (Fig. 3) are antic-
H
H
H2N
N
CH3
H2N
S
N
CH3
CH3
L-S-AcK
H
L-AcK
O
H
O
O
7.64 Å
6.62 Å
O
O
8.00 Å
OH
OH
O
H
N
H2N
H2N
O
L-O-AcK
N
H
CH3
H
L-AcOrn
H
O
O
7.47 Å
6.80 Å
OH
OH
O
H2N
H2N
S
NH2
L-S-Car-K
L-CysAcm
S
N
H
CH3
H
H
O
O
O
7.78 Å
OH
OH
Fig. 3. Optimized molecular geometry of
L
-AcK and its
L
-analogs. Atomic color code: H, white; C, light blue; O, red; N, blue; S, yellow. The distances between the side chain
carbonyl oxygen atom and the a-carbon were indicated for L-AcK and each of the analogs. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)

N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
23
Fig. 4. Representative HOMOs involving the side chain carbonyl oxygen atoms (arrow indicated) for L-AcK (left), L-O-AcK (middle), and L-S-AcK (right).
ipated to exhibit very similar nucleophilicity while their relative
positions to the -carbon atoms are varied.
resentative HPLC chromatograms for the assay with peptide 6, one
of the test peptides. Since the reliable detection limit of our HPLC-
a
based assay is ꢂ1
lM and around 40% of substrate deacetylation
3.2.2.3.
-analogs of
peptide containing
examine whether SIRT1 can catalyze the stereospecific deacetyla-
tion of -AcK versus its -isomer. The remaining -analog-contain-
ing peptides include peptides 5 and 7 both containing -S-AcK and
peptide 11 containing -S-Car-K (Fig. 2). From the analog synthesis
standpoint, these latter three -analog-containing peptides were
D
-Analogs. Besides the above-described
-AcK were also involved in the current study. The
-AcK (i.e. peptide 9, Fig. 2) was used to directly
L
-analogs, three
was observed for peptide 1 after a 2-h incubation with GST-SIRT1
(HPLC assay chromatogram not shown), peptides 2–9 were thus
each at least ꢂ120-fold worse than peptide 1 toward deacetylation
by SIRT1 under our experimental conditions.
D
L
D
L
D
D
The fact that same phenomena were observed for peptides 5
and 6 versus peptides 7 and 8 that contain the same pair of
D
L-AcK
D
analogs (i.e. -S-AcK and -S-AcK) argue that our observations in
D
L
D
the current study resulted from the intrinsic properties of the side
chains found in the analogs, because they do not depend on the
peptide sequence context. Since a detectable enzymatic deacetyla-
also produced (see Section 3.1), however, the additional evaluation
of these peptides in SIRT1 assays nevertheless made a further con-
tribution to our understanding of the molecular behavior of the
tion was not observed from the
D-AcK-containing peptide 9 under
L-AcK analogs used in the current study (see Section 3.3).
the same SIRT1 time-course assay condition under which the L-
AcK-containing peptide 1 was shown to give rise to a robust
3.3. SIRT1 assays of peptides containing -AcK or its analogs
L
deacetylation signal, SIRT1 can thus catalyze the stereospecific
deacetylation of L-AcK versus its D-isomer.
We initially subjected the peptides shown in Fig. 2 that are ter-
minated with the acetamido group (i.e. peptides 1–9) to a HPLC-
based SIRT1 deacetylation time-course assay [43–45]. When the
positive control peptide (i.e. peptide 1) was incubated with GST-
SIRT1 at 37 °C for 1 h, around 20% of substrate deacetylation was
already observed (Fig. 5). However, no enzymatic deacetylation
peptide product was able to be detected for peptides 2–9 even after
a 2-h incubation with GST-SIRT1 at 37 °C. Fig. 5 also shows the rep-
Besides the above-described SIRT1 deacetylation assay to assess
the possible enzymatic formation of a deacetylated peptide prod-
uct, we also performed the HPLC-based time-dependent nicotin-
amide formation assay to assess the effect of the side chain
structural deviation from
L-AcK on SIRT1-catalyzed nicotinamide
production from the -AcK-containing peptide (i.e. the positive
L
control peptide 1), which is the first chemical step along the reac-
tion coordinate of Sir2-catalyzed deacetylation reaction (Fig. 1). In
Fig. 5. Representative reversed-phase HPLC chromatograms from a SIRT1 time-course assay to assess the possible enzymatic formation of a deacetylated peptide product.
Peptide 1a (H2NꢃHKKLMꢃCOOH; MS (MALDI-TOF): m/z 656 [M + H]+, 678 [M + Na]+, 694 [M + K]+) is the SIRT1-catalyzed deacetylation product from peptide 1 (also compare
chromatograms (a) and (b)). All assays were performed in duplicate and essentially the same assay HPLC chromatograms were obtained for duplicates. The small peak with
tR ꢂ 32.5 min in chromatogram (d) was from a minor impurity in the purified peptide 6 sample (as shown in chromatogram (c)), rather than the deacetylated product.

24
N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
specific, we assayed peptide 1 and peptides 2–6 and 9–11 that each
contains a unique analog of -AcK (Fig. 2). In consistent with the
incorrect positioning of its side chain carbonyl oxygen relative to
the C10 atom of NAD+ within the
-AcK binding tunnel.
L
L
observed robust SIRT1-catalyzed deacetylation of peptide 1, a ro-
bust SIRT1-catalyzed nicotinamide production from peptide 1
was also observed in that around 8.9% of NAD+ cleavage (to form
nicotinamide) was already observed following the incubation of
peptide 1 with GST-SIRT1 at 37 °C for 30 min. However, no enzy-
matic nicotinamide formation was detected from peptides 2–6
and 9–11 even after a 2-h incubation with GST-SIRT1 at 37 °C.
While further studies are needed to unambiguously account for
the above-described lack of observable enzymatic deacetylation
The above-described two pairwise SIRT1 inhibition experiments
not only furnished evidence in favor of our hypothesis, but also
suggested that peptide 6 has a weaker binding affinity than pep-
tide 10, even though both seemed to bind weakly as compared to
peptide 1, considering that we used 100
inhibition assay. This could be due to the suboptimal positioning
of the side chains of both -S-Car-K and -S-AcK within the -AcK
binding tunnel as compared to the side chain of -AcK, however,
that of -S-AcK is the least optimal side chain among the three. This
latter notion is consistent with the greater deviation of the dis-
tance between the -carbon and the side chain carbonyl oxygen
from that of -AcK for -S-AcK than -S-Car-K (Fig. 3). The SIRT1
inhibition results with peptides 10 and 6 further suggested that
the -AcK binding tunnel could also accommodate an atom/group
(e.g. S) bulkier than CH2 within the part of the side chain of
lM of peptide 1 in the
L
L
L
L
L
and/or nicotinamide formation from peptides containing
analogs, our current hypothesis is that the side chain of a -analog
is unable to bind to the -AcK binding tunnel at the SIRT1 active
site, whereas the side chain of a -analog is able to bind to the
L-AcK
D
a
L
L
L
L
L
L
-AcK binding tunnel, but with the side chain carbonyl oxygen
L
atom being positioned incorrectly relative to the C10 atom of
L-
NAD+. Of note, as described in Section 3.2, the nucleophilicity of
AcK between the a-carbon and acetamide.
the side chain carbonyl oxygen atoms in
was predicted to be similar based on the calculated HOMOs. As a
support to our hypothesis, the side chains of several -analogs of
-AcK, including -thioacetyl-lysine and those with acyl groups
bulkier than the acetyl group, were previously shown to be able
to be accommodated at the active sites of SIRT1 and several other
Sir2 enzymes when placed within different peptide contexts
[25–29,43,44,46].
L-AcK and its analogs
4. Conclusion
L
L
L
In the current study, peptides containing L-AcK or its side chain
modified analogs were prepared and assayed using human SIRT1.
Our results suggested that SIRT1-catalyzed deacetylation reaction
had a very stringent requirement for the distance between the
a
-carbon and the side chain acetamido group, with that found in
-AcK being optimal. Moreover, our current study showed that
SIRT1 catalyzed the stereospecific deacetylation of -AcK versus
its -isomer. Taken together with the previous demonstration that
the side chain acetyl group of -AcK can be extended to bulkier acyl
To preliminarily address our hypothesis, we first turned our
attention to the pair of the peptides (i.e. peptides 10 and 11,
L
L
Fig. 2) that respectively contain
the distance between the -carbon and the carbonyl oxygen atom
in this side chain (7.78 Å) is the closest one to that in -AcK (7.64 Å)
L-S-Car-K and D-S-Car-K, since
D
a
L
L
e
groups for Sir2-catalyzed lysine N -deacylation reaction, and the
highly conserved nature of the Sir2 enzyme catalytic domain
[1–3], the results from our current study shall constitute another
piece of important information to be considered when designing
inhibitors for SIRT1 and Sir2 enzymes in general.
among all the distances shown in Fig. 3. When these two peptides
were examined for their respective inhibitory potency against
SIRT1-catalyzed deacetylation of peptide 1, we found that, while
peptide 11 did not exhibit any inhibition at 1.6 mM, 81% inhibition
was observed for peptide 10 at 1.6 mM. Since the only difference
between peptides 10/11 and peptide 1 is their containing different
side chains at the central position, the different inhibitory potency
that we observed for peptides 10 and 11 shall also reflect the dif-
ferent potency of these two peptides to compete with the binding
of peptide 1 at SIRT1 active site. Therefore, this SIRT1 inhibition
Acknowledgments
The financial support from the James L. and Martha J. Foght
Endowment, the University of Akron Research Foundation, and
the University of Akron Faculty Research Fellowship is greatly
appreciated. We thank Prof. Tony Kouzarides (University of Cam-
bridge, UK) for the GST-SIRT1 plasmid. We also thank Prof. Chrys
Wesdemiotis and his research group at the University of Akron
for the assistance with mass spectrometric analysis of the com-
pounds used in the current study. We wish to thank The Goodyear
Corporation for donation of the NMR instrument (Varian Mercury
300) used in this work. We also wish to thank The National Science
Foundation (CHE-8808587) for funds used to purchase the NMR
instrument (Varian GEMINI 300) used in this work.
experimental result strongly suggested that the side chain of
Car-K was unable to bind to the -AcK binding tunnel at the SIRT1
active site, whereas the side chain of -S-Car-K was able to, but
D-S-
L
L
with the side chain carbonyl oxygen atom being positioned incor-
rectly relative to the C10 atom of NAD+, which could account for our
inability to detect the enzymatic nicotinamide formation from
both peptides 10 and 11. Within this context, the incorrect side
chain carbonyl oxygen positioning relative to the C10 atom of
NAD+ within the
L
-AcK binding tunnel could also account for our
inability to observe the enzymatic deacetylation and nicotinamide
formation from peptide 4 which contains -O-AcK, another -AcK
analog that also has a fairly close distance between the -carbon
-AcK (i.e. 7.47 Å
L
L
References
a
and the side chain carbonyl oxygen to that in
L
[1] S. Thiagalingam, K.H. Cheng, H.J. Lee, N. Mineva, A. Thiagalingam, J.F. Ponte,
Ann. NY. Acad. Sci. 983 (2003) 84–100.
versus 7.64 Å, Fig. 3).
To extend the above-described pairwise analysis with peptides
10 and 11, we next examined the SIRT1 inhibitory potency of pep-
tides 5 and 6, one of the two pairs of the peptides that respectively
[2] I.V. Gregoretti, Y.M. Lee, H.V. Goodson, J. Mol. Biol. 338 (2004) 17–31.
[3] L.R. Saunders, E. Verdin, Oncogene 26 (2007) 5489–5504.
[4] G. Blander, L. Guarente, Annu. Rev. Biochem. 73 (2004) 417–435.
[5] J.M. Denu, Curr. Opin. Chem. Biol. 9 (2005) 431–440.
[6] A.A. Sauve, C. Wolberger, V.L. Schramm, J.D. Boeke, Annu. Rev. Biochem. 75
(2006) 435–465.
[7] S.C. Hodawadekar, R. Marmorstein, Oncogene 26 (2007) 5528–5540.
[8] X.J. Yang, E. Seto, Oncogene 26 (2007) 5310–5318.
[9] K. Batta, C. Das, S. Gadad, J. Shandilya, T.K. Kundu, Subcell. Biochem. 41 (2007)
193–212.
contain D-S-AcK and L-S-AcK (Fig. 2). It turned out that peptide 5
did not exhibit any inhibition at 1.6 mM, whereas 14% inhibition
was observed for peptide 6 at 1.6 mM. The differential extents of
SIRT1 inhibition exhibited by peptides 5 and 6 also strongly sug-
gested that the side chain of
D-S-AcK was unable to bind to the
[10] H. Yamamoto, K. Schoonjans, J. Auwerx, Mol. Endocrinol. 21 (2007) 1745–
1755.
[11] S. Michan, D. Sinclair, Biochem. J. 404 (2007) 1–13.
[12] B. Schwer, E. Verdin, Cell Metab. 7 (2008) 104–112.
L-AcK binding tunnel, whereas the side chain of
L-S-AcK was able
to. Our inability to observe the enzymatic deacetylation and nico-
tinamide formation from peptide 6 could also be explained by the

N. Jamonnak et al. / Bioorganic Chemistry 38 (2010) 17–25
25
[13] T. Finkel, C.X. Deng, R. Mostoslavsky, Nature 460 (2009) 587–591.
[14] M. Porcu, A. Chiarugi, Trends Pharmacol. Sci. 26 (2005) 94–103.
[15] L. Gan, Drug News Perspect. 20 (2007) 233–239.
[16] P.J. Elliott, M. Jirousek, Curr. Opin. Investig. Drugs 9 (2008) 371–378.
[17] T.F. Outeiro, O. Marques, A. Kazantsev, Biochim. Biophys. Acta 1782 (2008)
363–369.
[18] W. Deppert, Cell Cycle 7 (2008) 2947–2948.
[19] J.C. Milne, J.M. Denu, Curr. Opin. Chem. Biol. 12 (2008) 11–17.
[20] A.A. Sauve, Curr. Pharm. Des. 15 (2009) 45–56.
[32] E. Langley, M. Pearson, M. Faretta, U.M. Bauer, R.A. Frye, S. Minucci, P.G. Pelicci,
T. Kouzarides, EMBO J. 21 (2002) 2383–2396.
[33] A.J. Bannister, A. Cook, T. Kouzarides, Oncogene 6 (1991) 1243–1250.
[34] B.J. North, B. Schwer, N. Ahuja, B. Marshall, E. Verdin, Methods 36 (2005) 338–
345.
[35] R. Lutgring, K. Sujatha, J. Chmielewski, Bioorg. Med. Chem. Lett. 3 (1993) 739–
742.
[36] Merck Biosciences, NovabiochemÒ Guide to the selection of building blocks for
peptide synthesis, 2009.
[21] S. Imai, W. Kiess, Front. Biosci. 14 (2009) 2983–2995.
[22] T. Liu, P.Y. Liu, G.M. Marshall, Cancer Res. 69 (2009) 1702–1705.
[23] R.C. Neugebauer, W. Sippl, M. Jung, Curr. Pharm. Des. 14 (2008) 562–573.
[24] B.C. Smith, W.C. Hallows, J.M. Denu, Chem. Biol. 15 (2008) 1002–1013.
[25] Z. Cheng, Y. Tang, Y. Chen, S. Kim, H. Liu, S.S. Li, W. Gu, Y. Zhao, Mol. Cell.
Proteom. 8 (2009) 45–52.
[26] B.C. Smith, J.M. Denu, J. Biol. Chem. 282 (2007) 37256–37265.
[27] B.C. Smith, J.M. Denu, J. Am. Chem. Soc. 129 (2007) 5802–5803.
[28] W.F. Hawse, K.G. Hoff, D.G. Fatkins, A. Daines, O.V. Zubkova, V.L. Schramm, W.
Zheng, C. Wolberger, Structure 16 (2008) 1368–1377.
[37] H. Lamthanh, C. Roumestand, C. Deprun, A. Ménez, Int. J. Pept. Protein Res. 41
(1993) 85–95.
[38] T. Soussi, P. May, J. Mol. Biol. 260 (1996) 623–637.
[39] H. Vaziri, S.K. Dessain, E. Ng Eaton, S.I. Imai, R.A. Frye, T.K. Pandita, L. Guarente,
R.A. Weinberg, Cell 107 (2001) 149–159.
[40] A.L. Garske, J.M. Denu, Biochemistry 45 (2006) 94–101.
[41] K.G. Hoff, J.L. Avalos, K. Sens, C. Wolberger, Structure 14 (2006) 1231–1240.
[42] J.L. Avalos, I. Celic, S. Muhammad, M.S. Cosgrove, J.D. Boeke, C. Wolberger, Mol.
Cell 10 (2002) 523–535.
[43] D.G. Fatkins, A.D. Monnot, W. Zheng, Bioorg. Med. Chem. Lett. 16 (2006) 3651–
3656.
[29] J. Garrity, J.G. Gardner, W. Hawse, C. Wolberger, J.C. Escalante-Semerena, J.
Biol. Chem. 282 (2007) 30239–30245.
[30] T. Asaba, T. Suzuki, R. Ueda, H. Tsumoto, H. Nakagawa, N. Miyata, J. Am. Chem.
Soc. 131 (2009) 6989–6996.
[44] D.G. Fatkins, W. Zheng, Int. J. Mol. Sci. 9 (2008) 1–11.
[45] N. Jamonnak, D.G. Fatkins, L. Wei, W. Zheng, Org. Biomol. Chem. 5 (2007) 892–
896.
[31] D.A. Wellings, E. Atherton, Methods Enzymol. 289 (1997) 44–67.
[46] B.C. Smith, J.M. Denu, Biochemistry 46 (2007) 14478–14486.