Two-step synthesis of novel, bioactive derivatives of the ubiquitous cofactor nicotinamide adenine dinucleotide (NAD)

Article abstract of DOI:10.1021/jm1013852

We report the design and concise synthesis, in two steps from commercially available material, of novel, bioactive derivatives of the enzyme cofactor nicotinamide adenine dinucleotide (NAD). The new synthetic dinucleotides act as sirtuin (SIRT) inhibitors and show isoform selectivity for SIRT2 over SIRT1. An NMR-based conformational analysis suggests that the conformational preferences of individual analogues may contribute to their isoform selectivity.

Full text of DOI:10.1021/jm1013852

ARTICLE
pubs.acs.org/jmc
Two-Step Synthesis of Novel, Bioactive Derivatives of the Ubiquitous
Cofactor Nicotinamide Adenine Dinucleotide (NAD)
Thomas Pesnot,† Julia Kempter,†,‡ J€org Schemies,‡ Giulia Pergolizzi,† Urszula Uciechowska,§ Tobias Rumpf,‡
Wolfgang Sippl,§ Manfred Jung,‡ and Gerd K. Wagner*,†
†School of Pharmacy, University of East Anglia, Norwich, NR4 7TJ, U.K.
‡Institut f€ur Pharmazeutische Wissenschaften, Albert-Ludwigs-Universit€at Freiburg, Albertstrasse 25, 79104 Freiburg, Germany
§Department of Pharmaceutical Chemistry, Martin Luther University, Wolfgang-Langenbeckstrasse 4, 06120 Halle (Saale), Germany
S Supporting Information
b
ABSTRACT: We report the design and concise synthesis, in
two steps from commercially available material, of novel,
bioactive derivatives of the enzyme cofactor nicotinamide
adenine dinucleotide (NAD). The new synthetic dinucleotides
act as sirtuin (SIRT) inhibitors and show isoform selectivity for
SIRT2 over SIRT1. An NMR-based conformational analysis
suggests that the conformational preferences of individual
analogues may contribute to their isoform selectivity.
’ INTRODUCTION
synthetic step which can be plagued by variable yields,14 focusing
instead on the introduction of the 8-substituent on a preformed
NAD scaffold. However, examples for the direct chemical modifica-
tion of NAD are exceedingly rare and, because of the limited
chemical stability of this biomolecule, generally require very mild
conditions.
In principle, the target 8-(hetero)aryl substituted NAD deri-
vatives are accessible by SuzukiꢀMiyaura cross-coupling15 or
direct arylation.16ꢀ18 The direct structural modification of 1 is
complicated by its water solubility, the presence of various unpro-
tected functional groups, and several potential cleavage sites,
including the particularly labile nicotinamide riboside bond.
Previously reported conditions for the direct arylation of adenine
nucleotides are therefore not compatible with the limited che-
mical stability of 1.16ꢀ18 While the SuzukiꢀMiyaura coupling of
unprotected adenine nucleosides19ꢀ22 and nucleotides, includ-
ing adenosine triphosphate (ATP),23 has been reported, no
examples have been described to date for the application of this
chemistry to an intact NAD substrate.
Herein, we report the successful implementation of this novel
and highly efficient synthetic approach. We describe suitable
conditions for the cross-coupling of 8-bromo-NAD 2, including a
one-pot/two-step procedure for the in situ bromination and
cross-coupling of 1. Results from biological assays with two
sirtuin isoforms demonstrate that the target molecules act as
inhibitors of SIRT2 but have generally only weak activity against
SIRT1. A conformational analysis of the new NAD derivatives
suggests that the conformational preferences of individual
The dinucleotide NAD (nicotinamide adenine dinucleotide,
Scheme 1, 1) is a ubiquitous biomolecule with important functions
in cellular energy metabolism and cell signaling.1ꢀ3 Best known
for its role as a redox cofactor, 1 is also an essential cosubstrate for
a range of biologically important nonredox enzymes such as the
poly(ADP ribose) polymerases (PARPs)4,5 and sirtuins (SIRTs).6,7
These enzymes require 1 for covalent modifications during group
transfer reactions. The sirtuins, for example, catalyze the NAD-
dependent deacetylation of acetyllysine residues in various protein
substrates.8 The seven members of the human sirtuin family
(SIRT1ꢀ7) have been implicated in essential cellular processes
including transcriptional control, cell cycle progression, and
aging,9 and individual sirtuins represent promising drug targets
in, for example, cancer and diabetes.10 Structural analogues of 1
are therefore sought as inhibitor candidates and chemical probes for
the investigation of sirtuins and other NAD-dependent enzymes.11
Prompted by our interest in the development of sirtuin in-
hibitors, we have recently designed novel derivatives of 1 with an
additional aromatic or heteroaromatic substituent in position 8
of the adenine base (Scheme 1). Molecular docking studies with
the human sirtuin SIRT212 suggested that the NAD-binding site
of this enzyme might be able to accommodate a sterically demand-
ing substituent in this position (Figure 1). In order to test this
hypothesis, we required efficient synthetic access to the target
8-(hetero)aryl-NAD derivatives. With a view toward analogue
synthesis, we were particularly interested in a synthetic route that
offered maximum structural flexibility. The only previous exam-
ple for an 8-aryl-NAD derivative, 8-phenyl-NAD, has been prepared
via a linear, multistep synthesis centered around the formation of
the pyrophosphate bond.13 Ideally, we wanted to avoid this
Received: October 25, 2010
Published: April 29, 2011
r
2011 American Chemical Society
3492
dx.doi.org/10.1021/jm1013852 J. Med. Chem. 2011, 54, 3492–3499
|

Journal of Medicinal Chemistry
ARTICLE
Scheme 1. Synthesis of 8-substituted NAD derivativesa
a Reagents and conditions i) aq. Br2, aq. NaOAc (pH 4), rt, 1.75 h; ii)
Na2Cl4Pd (2.5 mol %), TXPTS, R-B(OH)2, K2CO3 (2 equiv), H2O,
40 °C, 20ꢀ35 min; iii) neat. Br2, aq. NaOAc (pH 4), rt, 0.5 h; then:
Na2Cl4Pd (2.5 mol %), TXPTS, R-B(OH)2, K2CO3 (5 equiv), H2O,
40 °C, 20ꢀ30 min. For substituents R see Table 1.
analogues may contribute to their isoform selectivity. In princi-
ple, the new cofactor analogues may also be applicable as
chemical probes for other NAD-dependent enzymes. Such applica-
tions will be greatly facilitated by our synthetic protocol, which
provides an extremely short entry from commercially available
material to this novel class of bioactive NAD derivatives.
’ RESULTS AND DISCUSSION
In order to develop a suitable SuzukiꢀMiyaura protocol, we
first tested the hydrolytic stability of the potential cross-coupling
substrate 8-bromo-NAD 2 (Scheme 1) under basic conditions.
At pH 9, 2 was consumed completely within 15 min at 80 °C and
within 60 min at 60 °C (Supporting Information). However, at
40 °C 2 proved to be remarkably stable for at least 1 h. We
speculated that this stability may open a window of opportunity
for the cross-coupling of 2, provided the reaction could be carried
out swiftly and efficiently. In order to exploit this opportunity, we
set out to identify optimum cross-coupling conditions. Previously,
we have successfully carried out cross-couplings of unprotected
nucleotides and sugar nucleotides in water, using a catalytic system
that combines the water-soluble ligand TPPTS (triphenylpho-
sphine trisulfonate, sodium salt) and the Pd source Na2Cl4Pd.24ꢀ26
However, in preliminary experiments with 8-bromoadenosine and
phenylboronic acid, we found that these conditions necessitated
elevated reaction temperatures not compatible with the limited
stability of 2 (Supporting Information). In contrast, use of the
electron-rich phosphine ligand TXPTS (tris(4,6-dimethyl-3-sul-
fonatophenyl)phosphine trisodium salt)20 allowed the successful
cross-coupling of 8-bromoadenosine at 40 °C within 1 h.
To apply these optimized cross-coupling conditions in the
context of our synthetic strategy, 8-bromo-NAD 2 was required
as the cross-coupling substrate. For the preparation of 2 we adapted
a previously published protocol for the 8-selective bromination
of NAD,27 using saturated bromine water as the brominating
agent instead of neat bromine (Scheme 1). After 1.75 h at room
temperature, RP-HPLC indicated ∼90% conversion but also
suggested the appearance of degradation products. The reaction
was therefore stopped and the material purified by ion-pair chro-
matography, affording 2 in 64% isolated yield. With sufficient
quantities of 2 in hand, we attempted the cross-coupling of 2
with a range of (hetero)arylboronic acids under our optimized
Figure 1. (A) Predicted docking pose for 8-(4-chlorophenyl)-NAD 3b
(cyan) in the cofactor binding site of human SIRT2 (ref 12). Hydrogen
bonds are shown as yellow dashed lines. (B) Schematic representation of the
3bꢀSIRT2 interactions. Hydrogen bonds are shown as green dashed lines.
SuzukiꢀMiyaura conditions (Scheme 1). Pleasingly, the cross-
coupling of 2 was successful with both electron-rich and electron-
poor phenylboronic acids, as well as furan-2-ylboronic acid
(Table 1, entries 1ꢀ5). Because of the mild reaction conditions
and short reaction times, cross-coupling products 3aꢀe were
generally obtained in moderate to good yields. Thus, this highly
efficient approach allowed the preparation of a small library of
8-substituted NAD derivatives from a single synthetic precursor.
In order to further simplify analogue synthesis, we next investi-
gated the combination of the bromination and cross-coupling steps
in a one-pot, two-step procedure (Scheme 1). Using neat bromine
instead of bromineꢀwater, we were able to limit the bromination
time to 30 min. While vigorous mixing was required for an effective
dissolution of the bromine in the sodium acetate buffer, this
protocol allowed us to suppress degradation of 1 or 2 during this
step. Excess bromine was removed by simple extraction with
CHCl3. The aqueous solution was evaporated to dryness, and the
crude residue was used in the subsequent cross-coupling step
3493
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499

Journal of Medicinal Chemistry
ARTICLE
Table 1. SuzukiꢀMiyaura Coupling of 8-Bromo-NAD (2)
Table 2. Sirtuin Inhibition Resultsa
with Various Aryl- and Heteroarylboronic Acids (R-B(OH)2)
compd
scaffoldb
R
SIRT1c
SIRT2d
entry
compd
R
synthesisa
yield (%)b
NAD
2
NAD
H
80 ( 9e
31%
46 ( 12e
28 ( 7
123 ( 4
35 ( 8
43 ( 5
52 ( 12
35%c
1
2
3
4
5
6
7
8
3a
3b
3c
3d
3e
3a
3f
phenyl
i and ii
i and ii
i and ii
i and ii
i and ii
iii
37
54
58
64
45
63
36
41
NAD
bromo
phenyl
4-chlorophenyl
4-methoxyphenyl
4-methylphenyl
furan-2-yl
3a
3b
3c
NAD
17%
NAD
4-chlorophenyl
4-methoxyphenyl
4-methylphenyl
furan-2-yl
bromo
30%
NAD
18%
3d
3e
4a
4b
5a
5b
NAD
20%
phenyl
NAD
16%
adenosine
adenosine
AMP
inactive
inactive
inactive
31 ( 2d
16%c
5-formylthien-2-yl
iii
phenyl
inactivec
inactivec
29 ( 3
3g
5-formylfuran-2-yl
iii
a See Scheme 1. b Isolated yields.
bromo
AMP
phenyl
a Km values for NAD are included for direct comparison. b See Scheme 1
without further purification. As the bromination was carried out
under slightly acidic conditions (pH 4), an additional 3 equiv of
K2CO3 was required under these conditions to reach the optimum
pH of 9 for the subsequent cross-coupling reaction.
and Figure 2. c Inhibition at 50 μM. Inactive: <10% inhibition. d IC50
SEM (μM). e Km, from ref 28.
(
The entire process, including both the bromination and cross-
coupling steps, could be conveniently monitored by RP-HPLC
(Figure S1, Supporting Information). This allowed a quick as-
sessment of reaction progress and helped to avoid potential
degradation of either starting materials or products. As expected,
the cross-coupled products showed a significantly longer reten-
tion time than both 1 and 2. The one-pot cross-coupling reaction
was used successfully to prepare NAD derivatives 3a, 3f, and 3g
in moderate to good yields (Table 1, entries 6ꢀ8). It is noteworthy
that the isolated yield of 3a is comparable to the previously re-
ported yield of 66% from multistep synthesis.13 Advantageously,
our protocol provides access to the target 3a in a single operation
and with short reaction times from commercially available 1. This
new approach therefore considerably facilitates the synthesis of
novel and structurally diverse NAD derivatives on a milligram
scale sufficient for biological experiments.
Figure 2. General structure of 8-substituted derivatives of adenosine
(4a and 4b, R0 = H) and AMP (5a and 5b, R0 = PO32ꢀ) (see Table 2).
indicating the differences in the NAD-binding region, has been
included in the Supporting Information (Figure S2, Supporting
Information).
In order to assess the bioactivity of the novel 8-substituted
NAD derivatives, compounds 2 and 3aꢀe were evaluated as
potential sirtuin inhibitors or cosubstrates. First, we carried out
inhibition experiments with NAD derivatives 2 and 3aꢀe and the
human enzymes SIRT1 and SIRT2 (Table 2). Almost all test
compounds, including the prototype NAD derivative 3a bearing
an 8-phenyl substituent, inhibited SIRT2 at low micromolar
concentrations, while only moderate activity was observed against
SIRT1. The presence of an additional group on the 8-phenyl sub-
stituent (3bꢀd) improved SIRT2 inhibition, while no such im-
provement was observed upon replacement of the phenyl with a
furanyl ring (3e). IC50 values for the new 8-substituted NAD
derivatives against SIRT2 were generally of a similar order of
magnitude as the Km for NAD previously reported for this enzyme28
(Table 2). These results suggest that in general the additional
substituent in position 8 is tolerated at the active site of human
SIRT2, as predicted by molecular docking with the available
X-ray structure12 of this enzyme (Figure 1). At the same time,
this structural modification appears to interfere with cofactor
binding at SIRT1. 8-Substituted NAD derivatives may therefore
provide a template for the development of isoform-selective sirtuin
ligands. Since there is presently no X-ray structure available for
SIRT1, we can only speculate about the structural basis for the
differential effects the 8-substituent has on cofactor binding at the
different sirtuin isoforms. A detailed analysis is complicated further
by the relatively low sequence identity of 44% between SIRT1
and SIRT2. A sequence alignment of human SIRT1, -2, -3, and -5,
In order to explore whether the complete dinucleotide scaffold
is required for biological activity and/or isoform selectivity, we
also tested the corresponding adenosine and adenosine mono-
phosphate (AMP) congeners (Figure 2) of NAD derivatives 2
and 3a against SIRT1 and SIRT2. The 8-substituted derivatives
of adenosine (4a, 4b) and AMP (5a, 5b) were prepared as
previously reported13,22 (Supporting Information). The com-
plete removal of the diphosphate and nicotinamide riboside in
adenosine derivatives 4a and 4b led to a dramatic loss of activity.
The mononucleotide 8-bromo-AMP 5a was also inactive, while
8-phenyl-AMP 5b was of a similar potency as the intact NAD
derivatives. Importantly, however, and in contrast to the correspond-
ing NAD derivative 3a, 5b displayed similar activity against both
sirtuin subtypes. This suggests that in this series of sirtuin inhibitors,
the complete NAD scaffold is crucial for enzyme selectivity.
It is a distinct possibility that, given their close structural
similarity to NAD itself, the new 8-substituted derivatives may
serve as cosubstrates for one or more members of the sirtuin
family. Such a potential cosubstrate activity would have obvious
implications for the interpretation of the observed inhibitory
activity. We therefore also tested the new 8-substituted NAD
derivatives for their reactivity as surrogate cosubstrates in the
sirtuin-catalyzed conversion of an acetyllysine substrate. Under
these conditions, only the 8-phenyl derivative 3a led to a small
conversion of the acetyllysine substrate when used as a substitute
for NAD in this fluorescence-based assay (Figure 3). Impor-
tantly, however, even the conversion of 3a is significantly lower
3494
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499

Journal of Medicinal Chemistry
ARTICLE
depending on the glycosyl torsion.32,33 Information about the
preferred conformation of a given adenine nucleoside or nucleo-
tide can be extracted from 1H NMR data. Thus, the N-type/S-
type ratio has been correlated experimentally with the magnitude
of the H10ꢀH20 coupling constant, and the relative percentage of
the S-type conformer can be calculated as 10J1 2 .
31 The chemical
0
0
shift of the H-20 signal is a good indicator for the position of the
syn/anti equilibrium.32,33 In native NAD, which exists predomi-
nantly in the anti conformation, H-20 resonates at 4.6 ppm. A
downfield shift of this signal is generally indicative of a transition
to the syn conformation.
Figure 3. Substrate conversion with NAD (1) or 8-phenyl-NAD (3a)
as cosubstrate. The 100% conversion refers to the signal obtained with
10.5 μM AMC (7-amino-4-methylcoumarin).
Using this NMR-based approach, we have analyzed important
conformational parameters of the 8-substituted NAD derivatives
2 and 3aꢀe (Table 3). In solution, all analogues show a slight
preference for the S-type conformation. Significantly, this is also
the conformation of native, 8-unsubstituted NAD in the solved
crystal structures with SIRT3,29 SIRT5,34 and bacterial enzymes.30
The same NAD conformation might also be expected for the
structurally related human SIRT2, which so far has been crystal-
lized solely in the apo form.12 This idea is supported by docking
experiments with SIRT2 and the 8-substituted NAD analogues
(Figure S4, Supporting Information). From the conformational
analysis and the docking experiments it therefore appears that the
additional substituent in position 8 does not perturb, and may
indeed stabilize, the bioactive conformation of the adenosine ribose.
For all 8-substituted analogues, we also observed a significant,
albeit variable, downfield shift of the H-20 signal relative to 8-un-
substituted NAD. This suggests that the 8-substituted derivatives
adopt preferentially a syn conformation, which can be attributed
to nonbonded repulsion between the 8-substituent and the ribose
ring.32,33
The results from this conformational analysis raise the intri-
guing possibility that conformational preferences may contribute
to the observed isoform selectivity in this series of NAD deri-
vatives.40 The available structures of SIRTs in complex with
NAD29,30,34 show the cofactor in the anti conformation,12 and
our docking experiments suggest a similar binding pose for the
8-substituted analogues. As all 8-substituted NAD derivatives
adopt preferentially the syn conformation in solution, these cofactor
analogues very likely have to undergo a conformational change
prior to binding at a target sirtuin. The inhibitory activity of 3aꢀd
suggests that in principle both SIRT1 and SIRT2 can induce this
conformational change but that SIRT2 is better equipped to do
so than SIRT1. The differential ability of individual sirtuins to
induce this conformational change may be linked to a noncon-
served three-amino-acid motif in the flexible loop which is close
to the 8-substituent of the ligand (Figures S5 and S6, Supporting
Information). However, in the absence of structural data for
SIRT1 this possibility remains speculative.
than that of NAD itself. These results suggest that none of the
8-substituted NAD derivatives is used productively as a cosub-
strate by either SIRT1 or SIRT2.
To understand the structural basis for this lack of cosubstrate
activity, we analyzed several available SIRT structures in the absence
or presence of various ligands. No crystal structure has been
reported to date for SIRT1, which made it impossible to study the
interaction of the new NAD derivatives with the NAD binding site
of SIRT1 by molecular modeling. However, it is known from the
available crystal structures of human SIRT3 in its apo form and in
complex with adenosine diphosphate ribose (ADPR) that a flexible
loop in the NAD binding site adopts a completely different con-
formation in both structures.29 This flexible loop contains NAD
binding residues, and the relationship between its conformation
and the presence of bound NAD is consistent among all SIRT
homologues for which structures have been solved.30 We there-
fore superimposed the X-ray structure of SIRT3 in complex with
ADPR (PDB code 3GLT) with the structure of SIRT2 contain-
ing the docked inhibitor 3b (Figure S3, Supporting Information).
This superimposition indicates a steric clash between the 8-chlor-
ophenyl substituent and the flexible binding region. Comparison
with the SIRT3 X-ray structure in the apo form (PDB code 3GLS),
on the other hand, shows the flexible loop in a conformation that
can accommodate the 8-chlorophenyl substituent of 3b. The
conformation of the flexible loop is determined by the nature of
the cosubstrate in the NAD binding pocket, and its movement has
been implicated with SIRT catalysis and/or nicotinamide release.30
On the basis of these overlays, it can be hypothesized that the
8-substituent in 2 and 3aꢀe may interfere with this loop move-
ment in SIRT2 and possibly in SIRT1, rendering these NAD
derivatives poor sirtuin cosubstrates, as observed in our experi-
ments.
In order to gain a better understanding of preliminary struc-
tureꢀactivity relationships (SAR) in this series, we carried out a
conformational analysis of compounds 2 and 3aꢀe. It is well-
known that a substituent in position 8 can have a significant effect
on the preferred conformation of NAD derivatives in solution.31ꢀ33
Such conformational preferences may in turn have direct im-
plications for the bioactivity of 8-substituted NAD derivatives.
We were particularly interested in the conformation of the ribose
ring in the adenosine fragment and in the orientation of the
adenine base relative to the ribose. In solution, the ribose ring in
nucleosides and nucleotides exists in an equilibrium between S-type
(southern) conformations, which include the classical C(20)-
endo and C(30)-exo conformers, and N-type (northern) confor-
mations (Table 3).31 The two main orientations of the adenine
base are either toward the ribose (syn) or away from it (anti),
The general idea that the conformational preferences of the
new 8-substituted sirtuin ligands underpin their isoform selec-
tivity is further supported experimentally by the biological results
for furan-2-yl derivative 3e. 3e shows the largest downfield shift
for H-20 and thus the strongest preference for the syn conforma-
tion. Locked in the syn conformation, 3e is the weakest SIRT
inhibitor in this series and the only analogue not to show
preferential inhibition of SIRT2. It therefore appears that in this
new series of NAD derivatives, the relative ease with which a
given analogue can adopt the anti conformation may be directly
correlated with its SIRT2 inhibitory activity. These results may
therefore provide a template for the rational development of
SIRT2-selective inhibitors.
3495
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499

Journal of Medicinal Chemistry
ARTICLE
Table 3. Conformational Analysis of 8-Substituted NAD Derivatives
compd
R
S-type conformer (%) (10J1,2
)
H-20 (ppm)
ΔH-20 (ppm)a
syn/anti
2
bromo
phenyl
59
5.23
5.19
5.19
5.20
5.17
5.36
0.61
0.57
0.57
0.58
0.55
0.74
syn
syn
syn
syn
syn
syn
3a
3b
3c
3d
3e
59ꢀ60
58ꢀ61
52ꢀ60
56ꢀ60
52ꢀ56
4-chlorophenyl
4-methoxyphenyl
4-methylphenyl
furan-2-yl
a ΔH-20 values have been calculated as the difference between the chemical shift of H-20 of the 8-substituted derivatives and H-20 of NAD (4.62 ppm).
’ CONCLUSION
(25 cm ꢁ 4.6 mm, particle size 5 μm), a diode array detector (detection
wavelengths of 254 and 280 nm), and a column oven (temperature of
25 °C). For details of the chromatographic conditions, see Supporting
Information.
We have developed a synthetic protocol for the Suzukiꢀ
Miyaura coupling of an intact NAD substrate, which has provided
access to a small library of novel, 8-substituted NAD derivatives
with biological activity against the human sirtuin SIRT2. For isoform
selectivity, the complete NAD scaffold is required, illustrating the
usefulness of our efficient synthetic route toward such analogues.
Our protocol will facilitate further analogue synthesis and allow
the systematic exploration of SAR in this exciting series of novel
cofactor analogues. The conformational analysis of the novel
NAD derivatives suggests that conformational preferences may
underpin their isoform selectivity. These findings may therefore
provide a starting point for the rational design of further analogues
with improved activity and isoform selectivity.
General Synthetic Method 1: SuzukiꢀMiyaura Cross-
Coupling of 8-Br-NAD (2). A round-bottom flask was charged with
8-Br-NAD 2 (1 equiv), Na2Cl4Pd (0.1 equiv), TXPTS (0.25 equiv),
K2CO3 (1.5 equiv), and arylboronic acid (1.2 equiv) and purged with N2
for 15 min. Degassed H2O (2 mL) was added to the reaction vessel, and the
reaction mixture was heated at 40 °C for 25 min. Reaction progress was
monitored by RP-HPLC. Upon completion of the reaction, the reaction
mixture was filtered through a syringe filter (pore size of 0.45 μm). The
filtrate was concentrated under reduced pressure, and the residual crude was
purified sequentially by ion-pair and ion-exchange chromatography
(purification methods 2 and 3; see Supporting Information).
General Synthetic Method 2: “One-Pot” Bromination and
SuzukiꢀMiyaura Coupling of NAD. To a solution of 1 (1 equiv) in
aqueous sodium acetate buffer (0.5 M, pH 4, 670 μL), neat bromine
(16 μL, 10 equiv) was added dropwise under vigorous stirring. The
mixture was stirred at room temperature for 30 min. The excess bromine
was extracted with CHCl3 (3ꢁ), and the aqueous solution was evaporated
to dryness in vacuo. To the residual pale orange film were added
Na2Cl4Pd (0.1 equiv), TXPTS (0.25 equiv), K2CO3 (5.0 equiv), and
phenylboronic acid (1.2 equiv). The reaction vessel was purged with N2
for 15 min. Degassed H2O was added under a N2 atmosphere, and the
mixture (pH 9) was heated at 40 °C for the given time. Reaction progress
was monitored by RP-HPLC. Upon completion of the reaction, the
reaction mixture was filtered through a syringe filter (pore size of
0.45 μm). The filtrate was concentrated under reduced pressure, and
the residual crude was purified sequentially by ion-pair and ion-exchange
chromatography (purification methods 2 and 3; see Supporting
Information).
’ EXPERIMENTAL SECTION
General. All chemicals were obtained commercially and used as
received unless stated otherwise. TXPTS was synthesized from tris(2,4-
dimethylphenyl)phosphane35 as described by Gulyas and co-workers.36
8-Bromoadenosine 4a, 8-phenyladenosine 4b, 8-bromo-AMP 5a, and
8-phenyl-AMP 5b were prepared, with minor modifications, as pre-
viously reported.13,22 TLC was performed on precoated aluminum plates
(silica gel 60 F254, Merck). Compounds were visualized by exposure to
UV light (254 and 365 nm). All test compounds were characterized
analytically by NMR (1H, 13C, and where appropriate 31P), high-
resolution mass spectrometry (HRMS), and HPLC and met the required
purity criteria (>95% by HPLC). NMR spectra were recorded at 298 K
on a Varian VXR 400 S spectrometer at 400 MHz (1H) or on a Bruker
Avance DPX-300 spectrometer at 300 MHz (1H). Chemical shifts (δ)
are reported in ppm and referenced to methanol (δH 3.34, δC 49.50) for
solutions in D2O. Coupling constants (J) are reported in Hz. Resonance
allocations were made with the aid of COSY and HSQC experiments.
Accurate electrospray ionization mass spectra (HR ESI-MS) were
obtained on a Finnigan MAT 900 XLT mass spectrometer at the EPSRC
National Mass Spectrometry Service Centre, Swansea, U.K. Preparative
chromatography was performed on a Biologic LP chromatography
system equipped with a peristaltic pump and a 254 nm UV optics
module. Analytical chromatography (HPLC) was carried out on an
Agilent 1200 machine equipped with a Supelcosil LC-18T column
8-Bromo-NAD (2)13. To a solution of NAD (60 mg, 90.4 μmol) in
aqueous sodium acetate (0.5 M, 2 mL, pH 4), saturated aqueous bromine
(1.7 mL) was added dropwise. The reaction mixture was stirred at room
temperature for 1.75 h. The excess bromine was extracted with chloro-
form (3ꢁ). The aqueous solution was reduced in vacuo to give a pale
orange film which was purified by ion-pair chromatography (purification
method 1). The triethylammonium salt of the title compound was
obtained as a glassy solid in 64% yield (42.8 mg, 57.6 μmol). HPLC:
4.52 min (98%). δH (400 MHz, D2O) 4.18ꢀ4.48 (8H, m, H-30, H-300,
3496
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499

Journal of Medicinal Chemistry
ARTICLE
H-40, H-400, H-50, H-500), 4.60 (1H, m, H-200), 5.23 (1H, t, J = 5.9 Hz,
H-20), 5.95ꢀ5.99 (2H, m, H-10, H-100), 7.91 (1H, s, H-2), 8.21 (1H, dd,
J = 5.6 and 8.5 Hz, Hn5), 8.82 (1H, d, J = 8.1 Hz, Hn4), 9.11 (1H, d, J =
6.4 Hz, Hn6), 9.29 (s, 1H, Hn2); δC (75.5 MHz, D2O) 65.7 (C-50), 66.4
(C-500), 70.2 (C-300), 71.5 (C-200), 71.7 (C-30), 78.5 (C-20), 84.2 (C-400),
87.9 (C-40), 90.0 (C-10), 100.9 (C-100), 119.8 (C-5), 128.9 (Cn5), 129.8
(C-8), 134.7 (Cn3), 140.7 (Cn2), 143.5 (Cn6), 146.8 (Cn4), 151.1 (C-4),
153.9 (C-2), 155.0 (C-6), 166.2 (CONH3); δP (121 MHz, D2O) ꢀ11.7
(d, JP,P = 20.6 Hz), ꢀ11.3 (d, JP,P = 20.6 Hz). m/z (ESI) 740.0138 [M ꢀ
H]ꢀ, C21H2579BrN7O14P2 requires 740.0124.
(1H, t, J = 5.6 Hz, H-200), 5.17 (1H, t, J = 5.6 Hz, H-20), 5.79 (1H, d,
J = 6.0 Hz, H-10), 5.89 (1H, d, J = 5.2 Hz, H-100), 7.25 (2H, d, J = 8.4 Hz,
Ph), 7.39 (2H, d, J = 8.0 Hz, Ph), 8.16 (1H, t, J = 7.0 Hz, Hn5), 8.16 (1H,
s, H-2), 8.72 (1H, d, J = 8.4 Hz, Hn4), 9.08 (1H, d, J = 6.0 Hz, Hn6), 9.22
(1H, s, Hn2); δC (125 MHz, D2O) 21.1, 65.4, 66.4, 70.0, 71.0, 71.1, 78.2,
83.5, 87.6, 89.7, 100.7, 118.9, 125.1, 129.6, 129.7, 130.3, 134.3, 140.6,
142.7, 143.1, 146.5, 150.7, 152.8, 153.5, 155.4, 166.2; δP (121 MHz,
D2O) ꢀ11.8 (d, JP,P = 20.6 Hz), ꢀ11.3 (d, JP,P = 19.4 Hz). m/z (ESI)
752.1501, [M ꢀ H]ꢀ, C28H32N7O14P2 requires 752.1488.
8-(Furan-2-yl)-NAD (3e). The triethylammonium (TEA) salt of
the title compound was obtained as a glassy solid from 2 (15.8 mg,
21.3 μmol) and furan-2-ylboronic acid in 45% yield (8.0 mg, 0.9 equiv of
TEA) according to general synthetic method 1 (reaction time, 20 min).
HPLC: 8.46 min (98%). δH (400 MHz, D2O) 4.06ꢀ4.40 (8H, m, H-30,
H-300, H-40, H-400, H-50, H-500), 4.61 (1H, t, J = 5.6 Hz, H-200), 5.36 (1H,
t, J = 5.6 Hz, H-20), 5.79 (1H, d, J = 5.2 Hz, H-10), 6.12 (1H, d, J = 6.0 Hz,
H-100), 6.65 (1H, dd, J = 1.6 and 3.2 Hz, fur3), 7.07 (1H, d, J = 3.6 Hz,
fur2), 7.73 (1H, d, J = 1.2 Hz, fur4), 8.12 (1H, t, J = 7.2 Hz, Hn5), 8.19
(1H, s, H-2), 8.66 (1H, d, J = 8.4 Hz, Hn4), 8.96 (1H, d, J = 6.0 Hz, Hn6),
9.17 (1H, s, Hn2); δC (125 MHz, D2O) 65.6, 66.6, 70.0, 71.2, 71.4, 78.4,
83.8, 87.8, 89.8, 100.7, 113.2, 115.9, 119.2, 129.6, 134.2, 140.4, 142.5,
143.0, 143.5, 146.2, 146.7, 150.3, 153.1, 155.3, 165.7; δP (121 MHz,
D2O) ꢀ11.9 (d, JP,P = 21.8 Hz), ꢀ11.4 (d, JP,P = 20.6 Hz). m/z (ESI)
728.1125, [M ꢀ 2H]ꢀ, C25H28N7O15P2 requires 728.1124.
8-Phenyl-NAD (3a)13. From 2. The triethylammonium (TEA)
salt of the title compound was obtained as a glassy solid from 2 (17.0 mg,
22.9 μmol) and phenylboronic acid in 37% yield (7.0 mg, 0.8 equiv of
TEA) according to general synthetic method 1 (reaction time, 25 min).
From 1. The triethylammonium (TEA) salt of the title compound was
obtained as a glassy solid from 1 (20.0 mg, 30.1 μmol) and phenylboronic
acid in 63% yield (16.6 mg, 1.0 equiv of TEA) according to general
synthetic method 2 (cross-coupling time, 30 min). HPLC: 10.90 min
(96%). δH (400 MHz, D2O) 4.12ꢀ4.42 (8H, m, H-30, H-300, H-40, H-400,
H-50, H-500), 4.44 (1H, t, J = 5.6 Hz, H-200), 5.19 (1H, t, J = 6.0 Hz, H-20),
5.83 (1H, d, J = 5.9 Hz, H-10), 5.94 (1H, d, J = 5.2 Hz, H-100), 7.59ꢀ7.68
(5H, m, Ph), 8.17 (1H, t, J = 7.6 Hz, Hn5), 8.22 (1H, s, H-2), 8.72 (1H, d,
J = 8.0 Hz, Hn4), 9.10 (1H, d, J = 6.1 Hz, Hn6), 9.25 (1H, s, Hn2); δC
(125 MHz, D2O) 65.5, 66.4, 70.0, 71.0, 71.1, 78.3, 83.6, 87.6, 89.7, 100.7,
118.9, 128.2, 129.5, 129.8, 130.1, 131.9, 134.2, 140.5, 143.0, 146.3, 150.6,
152.9, 153.6, 155.4, 165.6; δP (121 MHz, D2O) ꢀ11.7 (d, JP,P = 20.6
Hz), ꢀ11.3 (d, JP,P = 20.6 Hz). m/z (ESI) 738.1328, [M ꢀ 2H]ꢀ,
C27H30N7O14P2 requires 738.1331.
8-(4-Chlorophenyl)-NAD (3b). The triethylammonium (TEA)
salt of the title compound was obtained as a glassy solid from 2 (19.4 mg,
26.1 μmol) and 4-chlorophenylboronic acid in 54% yield (13.7 mg, 1.2
equiv of TEA) according to general synthetic method 1 (reaction time,
35 min). HPLC: 14.72 min (98%). δH (400 MHz, D2O) 4.12ꢀ4.40
(8H, m, H-30, H-300, H-40, H-400, H-50, H-500), 4.48 (1H, t, J = 5.5 Hz,
H-200), 5.19 (1H, t, J = 6.1 Hz, H-20), 5.78 (1H, d, J = 5.8 Hz, H-10), 5.94
(1H, d, J = 5.2 Hz, H-100), 7.50 (2H, d, J = 8.0 Hz, Ph), 7.56 (2H, d, J = 8.4
Hz, Ph), 8.18 (1H, t, J = 7.2 Hz, Hn5), 8.19 (1H, s, H-2), 8.74 (1H, d, J =
8.0 Hz, Hn4), 9.10 (1H, d, J = 6.3 Hz, Hn6), 9.25 (1H, s, Hn2); δC (125
MHz, D2O) 65.4, 66.3, 70.0, 70.9, 71.1, 78.2, 83.5, 87.6, 89.6, 100.7,
118.7, 126.4, 129.4, 129.8, 131.0, 134.2, 137.5, 140.5, 143.0, 146.3, 150.5,
151.9, 153.1, 155.3, 165.6; δP (121 MHz, D2O) ꢀ11.8 (d, JP,P = 20.6
Hz), ꢀ11.3 (d, JP,P = 20.6 Hz). m/z (ESI) 772.0952, [M ꢀ H]ꢀ,
C27H29Cl35N7O14P2 requires 772.0942.
8-(4-Methoxyphenyl)-NAD (3c). The triethylammonium (TEA)
salt of the title compound was obtained as a glassy solid from 2 (17.0 mg,
22.9 μmol) and 4-methoxyphenylboronic acid in 58% yield (11.7 mg, 1.0
equiv of TEA) according to general synthetic method 1 (reaction time,
20 min). HPLC: 12.93 min (98%). δH (400 MHz, D2O) 3.86 (3H, s,
MeO), 4.12ꢀ4.40 (8H, m, H-30, H-300, H-40, H-400, H-50, H-500), 4.49
(1H, t, J = 5.8 Hz, H-200), 5.20 (1H, t, J = 6 Hz, H-20), 5.78 (1H, d, J = 5.2
Hz, H-10), 5.90 (1H, d, J = 5.2 Hz, H-100), 6.99 (2H, d, J = 8.8 Hz, mPh),
7.49 (2H, d, J = 9.2 Hz, oPh), 8.158 (1H, s, H-2), 8.162 (1H, t, J = 7.2 Hz,
Hn5), 8.72 (1H, d, J = 8.1 Hz, Hn4), 9.08 (1H, d, J = 6.2 Hz, Hn6), 9.22
(1H, s, Hn2); δC (125 MHz, D2O) 56.1, 65.4, 66.3, 70.0, 70.9, 71.1, 78.2
(C-20), 83.4, 87.6, 89.7, 100.7, 115.0, 118.8, 120.4, 129.4, 131.4, 134.1,
140.4, 143.0, 146.3, 150.6, 152.4, 153.1, 155.0, 161.6, 165.6; δP (121
MHz, D2O) ꢀ11.8 (d, JP,P = 20.6 Hz), ꢀ11.3 (d, JP,P = 20.6 Hz). m/z
(ESI) 768.1426, [M ꢀ 2H]ꢀ, C28H32N7O15P2 requires 768.1437.
8-(4-Methylphenyl)-NAD (3d). The triethylammonium (TEA)
salt of the title compound was obtained as a glassy solid from 2 (15.4 mg,
20.7 μmol) and 4-methylphenylboronic acid in 64% yield (12.1 mg, 1.0
equiv of TEA) according to general synthetic method 1 (reaction time,
20 min). HPLC: 13.91 min (99%). δH (400 MHz, D2O) 2.34 (3H,
s, Me), 4.12ꢀ4.42 (8H, m, H-30, H-300, H-40, H-400, H-50, H-500), 4.48
8-(5-Formylthien-2-yl)-NAD (3f). The triethylammonium (TEA)
salt of the title compound was obtained as a glassy solid from 1 (20.0 mg,
30.1 μmol) and 5-formylthien-2-ylboronic acid in 36% yield (10.2 mg, 1.7
equiv of TEA) according to general synthetic method 2 (cross-coupling
time, 30 min). δH (400 MHz, D2O) 4.06ꢀ4.40 (8H, m, H-30, H-300, H-40,
H-400, H-50, H-500), 4.61 (1H, t, J = 5.6 Hz, H-200), 5.50 (1H, t, J = 6.0 Hz,
H-20), 5.85 (1H, m, H-10), 5.97 (1H, d, J = 5.9 Hz, H-100), 7.64 (1H, d,
J = 4.1 Hz, Th), 7.95 (1H, d, J = 4.1 Hz, Th), 8.16 (1H, dd, J = 6.4 and 7.8
Hz, Hn5), 8.22 (1H, s, H-2), 8.69 (1H, d, J = 8.1 Hz, Hn4), 9.02 (1H, d,
J = 6.4 Hz, Hn6), 9.21 (1H, s, Hn2), 9.83 (1H, s, CHO); δP (121 MHz,
D2O) ꢀ11.9 (d, JP,P = 21.8 Hz), ꢀ11.4 (d, JP,P = 20.6 Hz). m/z (ESI)
772.0838 [M ꢀ H]ꢀ, C26H28N7O15P2S requires 772.0845.
8-(5-Formylfuran-2-yl)-NAD (3g). The triethylammonium (TEA)
salt of the title compound was obtained as a glassy solid from 1 (20.0 mg,
30.1 μmol) and 5-formylfuran-2-ylboronic acid in 41% yield (12.2 mg,
2.1 equiv of TEA) according to general synthetic method 2 (cross-
coupling time, 20 min). δH (400 MHz, D2O) 4.06ꢀ4.40 (8H, m, H-30,
H-300, H-40, H-400, H-50, H-500), 4.70 (1H, t, J = 5.6 Hz, H-200), 5.34 (1H,
t, J = 5.9 Hz, H-20), 5.83 (1H, d, J = 4.3 Hz, H-10), 6.17 (1H, d, J = 5.3 Hz,
H-100), 7.26 (1H, d, J = 3.6 Hz, fur), 7.57 (1H, d, J = 3.8 Hz, fur), 8.12
(1H, t, J = 7.2 Hz, Hn5), 8.19 (1H, s, H-2), 8.67 (1H, d, J = 8.4 Hz, Hn4),
8.99 (1H, d, J = 6.0 Hz, Hn6), 9.17 (1H, s, Hn2), 9.57 (1H, s, CHO); δP
(121 MHz, D2O) ꢀ11.9 (d, JP,P = 21.8 Hz), ꢀ11.4 (d, JP,P = 20.6 Hz).
m/z (ESI) 756.1059 [M ꢀ H]ꢀ, C26H28N7O16P2 requires 756.0173.
Sirtuin Bioassay. Recombinant Expression of SIRT1 and SIRT2.
Human SIRT1 was expressed as a GST-tagged fusion protein, and human
SIRT2 was expressed as a N-terminally tagged His6 fusion protein. Both
enzymes were purified as described with minor modifications.37 Identity
and purity of the produced enzymes were verified using SDS electro-
phoresis. The deacetylase activity of the sirtuins was dependent on the
cofactor NAD and could be inhibited with the endogeneous sirtuin in-
hibitor nicotinamide.
Fluorescent Deacetylase Assay. All NAD derivatives were evaluated
for their ability to inhibit recombinant SIRT1 and SIRT2 using a homo-
geneous fluorescent deacetylase assay.38 DMSO stock solutions of the
inhibitors were prepared, and defined amounts were added to an
incubation mixture. The assay was carried out in 96-well plates with a
reaction volume per well of 60 μL, containing enzyme (SIRT1 or SIRT2),
the fluorescent histone deacetylase substrate ZMAL (Z-(Ac)Lys-AMC,
10.5 μM), cofactor NAD (500 μM), and inhibitor. Initially, inhibitors
3497
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499

Journal of Medicinal Chemistry
ARTICLE
were tested at 50 μM only. Inhibitors that showed inhibition greater
than 50% at 50 μM were tested at a wider range of concentrations
(0.1ꢀ500 μM) to allow the determination of IC50. The amount of
enzyme solution used was dependent on the activity of the individual
enzyme preparation, which varied slightly from batch to batch. To maintain
initial state conditions, enzyme solutions were used in amounts that gave
10ꢀ30% conversion of the substrate in the absence of inhibitor. After 4 h
of incubation at 37 °C, the reaction was stopped with a solution of
þ44 (0)20 7848 4747. Fax: þ44 (0)20 7848 4045. E-mail: gerd.
wagner@kcl.ac.uk.
’ ACKNOWLEDGMENT
We thank the EPSRC (First Grant EP/D059186/1 to G.K.
W.) and the Deutsche Forschungsgemeinschaft (Grant Ju 295/
8-1 to M.J., Grant Si 868/7-1 to W.S.) for financial support and
the EPSRC National Mass Spectrometry Service Centre, Swansea,
U.K., for the recording of mass spectra.
trypsin buffer (60 μL) containing trypsin (1 mg mLꢀ1) from bovine
3
pancreas (1000 (BAEE units) mgꢀ1) and the sirtuin inhibitor nicotin-
3
amide (8 mM), and the microplate was incubated with this solution for
20 min at 37 °C. Finally, fluorescence intensity was measured in a plate
reader (BMG Polarstar) with a coumarin filter (λex = 390 nm, λem
=
’ ABBREVIATIONS USED
460 nm). The amount of remaining substrate in the positive control with
inhibitor versus negative control without inhibitor was employed to
calculate inhibition. All inhibition determinations were carried out at least in
duplicate. IC50 values were calculated using GraphPad Prism software.
The same assay was used to assess if the 8-substituted NAD analogues
were accepted as cosubstrates by SIRT1 and SIRT2. The NAD analogues
and, for direct comparison, NAD itself were tested at 500 μM, with a
DMSO concentration of 8.3% (v/v). Under these conditions, only 3a
showed a detectable conversion against the blank without cofactor.
Molecular Modeling. All calculations were performed on a Pen-
tium IV 2.2 GHz based Linux cluster (20 CPUs). The molecular struc-
tures of the NAD derivatives were generated using the crystal structure
of NAD bound to archeal sirtuin Sir2-Af2 (PDB code 1yc2) and the
MOE modeling package (Chemical Computing Group). The modified
NAD and AMP derivatives were energy minimized using the MMFF94s
force field and the conjugate gradient method until the default derivative
ADPR, adenosine diphosphate ribose; AMC, 7-amino-4-methyl-
coumarin; AMP, adenosine monophosphate; ATP, adenosine
triphosphate; HRMS, high-resolution mass spectrometry; NAD,
nicotinamide adenine dinucleotide; SAR, structureꢀactivity re-
lationships; SIRT, sirtuin; TPPTS, triphenylphosphine trisulfo-
nate; TXPTS, tris(4,6-dimethyl-3-sulfonatophenyl)phosphine;
ZMAL, Z-(Ac)Lys-AMC
’ REFERENCES
(1) Ziegler, M. New functions of a long-known molecule—emerging
roles of NAD in cellular signaling. Eur. J. Biochem. 2000, 267,
1550–1564.
(2) Berger, F.; Ramírez-Hernꢀandez, M. H.; Ziegler, M. The new life
of a centenarian: signalling functions of NAD(P). Trends Biochem. Sci.
2004, 29, 111–118.
(3) Lin, H. Nicotinamide adenine dinucleotide: beyond a redox
coenzyme. Org. Biomol. Chem. 2007, 5, 2541–2554.
(4) Rouleau, M.; Patel, A.; Hendzel, M. J.; Kaufmann, S. H.; Poirier,
G. G. PARP inhibition: PARP1 and beyond. Nat. Rev. Cancer 2010,
10, 293–301.
(5) Ferraris, D. V. Evolution of poly(ADP-ribose) polymerase-1
(PARP-1) inhibitors. From concept to clinic. J. Med. Chem. 2010,
53, 4561–4584.
(6) Finkel, T.; Deng, C. X.; Mostoslavsky, R. Recent progress in the
biology and physiology of sirtuins. Nature 2009, 460, 587–591.
(7) Schemies, J.; Uciechowska, U.; Sippl, W.; Jung, M. NAD-
dependent histone deacetylases (sirtuins) as novel therapeutic targets.
Med. Res. Rev. 2010, 30, 861–889.
(8) 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.
(9) Imai, S.; Armstrong, C. M.; Kaeberlein, M.; Guarente, L.
Transcriptional silencing and longevity protein Sir2 is an NAD-depen-
dent histone deacetylase. Nature 2000, 403, 795–800.
(10) Khan, J. A.; Forouhar, F.; Tao, X.; Tong, L. Nicotinamide
adenine dinucleotide metabolism as an attractive target for drug
discovery. Expert Opin. Ther. Targets 2007, 11, 695–705.
(11) Chen, L.; Petrelli, R.; Felczak, K.; Gao, G.; Bonnac, L.; Yu, J. S.;
Bennett, E. M.; Pankiewicz, K. W. Nicotinamide adenine dinucleotide
based therapeutics. Curr. Med. Chem. 2008, 15, 650–670.
(12) Finnin, M. S.; Donigian, J. R.; Pavletich, N. P. Structure of the
histone deacetylase SIRT2. Nat. Struct. Biol. 2001, 8, 621–625.
(13) Zhang, B.; Wagner, G. K.; Weber, K.; Garnham, C.; Morgan,
A. J.; Galione, A.; Guse, A. H.; Potter, B. V. L. 20-Deoxy cyclic adenosine
50-diphosphate ribose derivatives: importance of the 20-hydroxyl motif
for the antagonistic activity of 8-substituted cADPR derivatives. J. Med.
Chem. 2008, 51, 1623–1636.
convergence criterion of 0.01 kcal/(mol Å) was met. The structure of
3
the human SIRT2 protein was taken from the Protein Databank (PDB
code 1j8f, monomer B). Docking of the inhibitors was carried out using
the program GOLD, version 4.1,39 and default settings (Cambridge
Crystallographic Data Centre). All torsion angles in each inhibitor were
allowed to rotate freely. The binding site was defined on Phe620 with a
radius of 15 Å. Goldscore was chosen as fitness function. For each
molecule 30 docking runs were performed. The resulting solutions were
clustered on the basis of the heavy atom rmsd values (1 Å). The top-
ranked poses for each ligand were retained and analyzed graphically
within MOE 2006.08 (Chemical Computing Group). In addition to the
docking complexes derived for human SIRT2, the archaeal Sir2-Af2
crystal structure was used for the current investigation to inspect the
NADꢀenzyme interaction. These docking results suggest that in general
the additional substituent in position 8 is tolerated at the active site of
SIRT2. A comparison of the SIRT2 docking results for the NAD derivatives
modified in position 8, and the X-ray structure of the archeal homologue
Sir2-Af2 in complex with NAD, shows that the ligands are involved in the
same hydrogen bonds to the residues of the binding pocket (Figure S4,
Supporting Information).
’ ASSOCIATED CONTENT
S
Supporting Information. Chromatography conditions;
b
protocols for the synthesis of 4a, 4b, 5a, and 5b; stability testing
of 2; 1H, 13C, and 31P NMR spectra; Figures S1ꢀS6; and docking
scores. This material is available free of charge via the Internet at
http://pubs.acs.org.
’ AUTHOR INFORMATION
(14) Lee, J.; Churchil, H.; Choi, W. B.; Lynch, J. E.; Roberts, F. E.;
Volante, R. P.; Reider, P. J. A chemical synthesis of nicotinamide adenine
dinucleotide (NAD). Chem. Commun. 1999, 729–730.
(15) Suzuki, A. Carbonꢀcarbon bonding made easy. Chem. Commun.
2005, 4759–4763.
Corresponding Author
*Current address: Institute of Pharmaceutical Science, School of
Biomedical Sciences, King’s College London, Franklin-Wilkins
Building, 150 Stamford Street, London SE1 9NH, U.K. Phone:
3498
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499

Journal of Medicinal Chemistry
ARTICLE
(16) Cerna, I.; Pohl, R.; Hocek, M. The first direct CꢀH arylation of
purine nucleosides. Chem. Commun. 2007, 4729–4730.
(17) Storr, T. E.; Firth, A. G.; Wilson, K.; Darley, K.; Baumann,
C. G.; Fairlamb, I. J. S. Site-selective direct arylation of unprotected
adenine nucleosides mediated by palladium and copper: insights into the
reaction mechanism. Tetrahedron 2008, 64, 6125–6137.
(36) Gulyꢀas, H.; Sz€oll€osy, A.; Hanson, B. E; Bakos, J. A direct
approach to selective sulfonation of triarylphosphines. Tetrahedron Lett.
2002, 43, 2543–2546.
(37) North, B. J.; Schwer, B.; Ahuja, N.; Marshall, B.; Verdin, E.
Preparation of enzymatically active recombinant class III protein
deacetylases. Methods 2005, 36, 338–345.
(18) Storr, T. E.; Baumann, C. G.; Robert, R. J.; De Ornellas, S.;
Whitwood, A. C.; Fairlamb, I. J. S. Pd(0)/Cu(I)-mediated direct arylation of
20-deoxyadenosines: mechanistic role of Cu(I) and reactivity comparisons
with related purine nucleosides. J. Org. Chem. 2009, 74, 5810–5821.
(19) Amann, N.; Wagenknecht, H. A. Preparation of pyrenyl-
modified nucleosides via SuzukiꢀMiyaura cross-coupling reactions.
Synlett 2002, 687–691.
(20) Western, E. C.; Daft, J. R.; Johnson, E. M.; Gannett, P. M.;
Shaughnessy, K. H. Efficient one-step Suzuki arylation of unprotected
halonucleosides, using water-soluble palladium catalysts. J. Org. Chem.
2003, 68, 6767–6774.
(38) Heltweg, B.; Trapp, J.; Jung, M. In vitro assays for the determina-
tion of histone deacetylase activity. Methods 2005, 36, 332–337.
(39) Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R.
Development and validation of a genetic algorithm for flexible docking.
J. Mol. Biol. 1997, 267, 727–748.
(40) When ligand selectivity is discussed, it is important to note that
the SIRT1 and SIRT2 enzymes used in this study have been labeled with
different tags (SIRT1 with GST; SIRT2 with His). In principle, the
different tags could influence the selectivity of sirtuin inhibitors,
although there are no previous reports in the literature for such an effect.
(21) Kohyama, N.; Katashima, T.; Yamamoto, Y. Synthesis of novel
2-aryl AICAR derivatives. Synthesis 2004, 2799–2804.
(22) Collier, A.; Wagner, G. K. SuzukiꢀMiyaura cross-coupling of
unprotected halopurine nucleosides in water-influence of catalyst and
cosolvent. Synth. Commun. 2006, 36, 3713–3721.
(23) Capek, P; Pohl, R.; Hocek, M. Cross-coupling reactions of un-
protected halopurine bases, nucleosides, nucleotides and nucleoside tripho-
sphates with 4-boronophenylalanine in water. Synthesis of (purin-8-yl)- and
(purin-6-yl)phenylalanines. Org. Biomol. Chem. 2006, 4, 2278–2284.
(24) Collier, A.; Wagner, G. K. A facile two-step synthesis of 8-arylated
guanosine mono- and triphosphates (8-aryl GXPs). Org. Biomol. Chem.
2006, 4, 4526–4532.
(25) Collier, A.; Wagner, G. K. A fast synthetic route to GDP-sugars
modified at the nucleobase. Chem. Commun. 2008, 20, 178–180.
(26) Pesnot, T.; Wagner, G. K. Novel derivatives of UDP-glucose:
concise synthesis and fluorescent properties. Org. Biomol. Chem. 2008,
6, 2884–2891.
(27) Panza, J. L.; Russell, A. J.; Beckman, E. J. Synthesis of
fluorinated NAD as a soluble coenzyme for enzymatic chemistry in
fluorous solvents and carbon dioxide. Tetrahedron 2002, 58, 4091–4104.
(28) Du, J.; Jiang, H.; Lin, H. Investigating the ADP-ribosyltransfer-
ase activity of sirtuins with NAD analogues and 32P-NAD. Biochemistry
2009, 48, 2878–2890.
(29) 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 conforma-
tional changes. J. Biol. Chem. 2009, 284, 24394–24405.
(30) Sanders, B. D.; Jackson, B.; Marmorstein, R. Structural basis for
sirtuin function: what we know and what we don’t. Biochim. Biophys. Acta
2010, 1804, 1604–1616.
(31) Altona, C.; Sundaralingam, M. Conformational analysis of the
sugar ring in nucleosides and nucleotides. Improved method for the
interpretation of proton magnetic resonance coupling constants. J. Am.
Chem. Soc. 1973, 95, 2333–2344.
(32) Ikehara, M.; Uesugi, S.; Yoshida, K. Nucleosides and nucleo-
tides. XLVII. Conformation of purine nucleosides and their 50-phos-
phates. Biochemistry 1972, 11, 830–836.
(33) Sarma, R. H.; Lee, H. C.; Evans, F. E.; Yathindra, N.; Sundar-
alingam, M. Probing the interrelation between the glycosyl torsion, sugar
pucker, and the backbone conformation in C(8) substituted adenine
nucleotides by proton and protonꢀphosphorus-31 fast Fourier transfer
nuclear magnetic resonance methods and conformational energy calcu-
lations. J. Am. Chem. Soc. 1974, 96, 7337–7348.
(34) 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.
(35) Culcasi, M.; Berchadsky, Y.; Gronchi, G.; Tordo, P. Anodic
behavior of crowded triarylphosphines. ESR study of triarylphospho-
niumyl radicals. J. Org. Chem. 1991, 56, 3537–3542.
3499
dx.doi.org/10.1021/jm1013852 |J. Med. Chem. 2011, 54, 3492–3499