Novel nicotinamide adenine dinucleotide analogues as selective inhibitors of NAD+-dependent enzymes

Article abstract of DOI:10.1016/j.tet.2004.05.091

Three novel dinucleotide analogues of nicotinamide adenine dinucleotide (NAD⁺) have been synthesised from D-ribonolactone. These compounds incorporate a thiophene moiety in place of nicotinamide and are hydrolytically stable. They have been evaluated as inhibitors of adenosine diphosphate ribosyl cyclase, glutamate dehydrogenase and Sir2 acyltransferase activities. Enzyme specificity and a high level of inhibition was observed for the dehydrogenase.

Full text of DOI:10.1016/j.tet.2004.05.091

Tetrahedron 60 (2004) 6609–6617
Novel nicotinamide adenine dinucleotide analogues as selective
inhibitors of NAD1-dependent enzymes
Nathalie E. Batoux,a Francesca Paradisi,b Paul C. Engelb and Marie E. Migauda
,
*
aSchool of Chemistry, Queen’s University, Belfast BT9 5AG, UK
bDepartment of Biochemistry, University College Dublin, Belfield, Dublin 4, Ireland
Received 8 March 2004; revised 7 May 2004; accepted 27 May 2004
Available online 19 June 2004
Abstract—Three novel dinucleotide analogues of nicotinamide adenine dinucleotide (NADþ) have been synthesised from D-ribonolactone.
These compounds incorporate a thiophene moiety in place of nicotinamide and are hydrolytically stable. They have been evaluated as
inhibitors of adenosine diphosphate ribosyl cyclase, glutamate dehydrogenase and Sir2 acyltransferase activities. Enzyme specificity and a
high level of inhibition was observed for the dehydrogenase.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
quently, whilst most of these compounds act as potent anti-
tumour agents by shutting down the guanosine mono-
phosphate synthesis, they are also highly toxic to healthy
cells.
C-nucleoside analogues of ribosyl nicotinamide (Fig. 1)
have been widely studied and some are currently in clinical
trials.1 They are endowed with several biological effects
ranging from anti-tumour activities to inhibition of
G-protein mediated cellular mechanisms.2 –4 Some of
these effects have been directly related to the capacity
of their adenine dinucleotide derivatives at inhibiting
inosine monophosphate dehydrogenase.2,5,6 Yet, C-nucleo-
sides, such as ribosyl benzamide and tiazofurin (Fig. 1) can
be inhibitors of other nicotinamide adenine dinucleotide
(NADþ) dependent enzymes, such as oxidoreductases,
glycohydrolases and transferases, after their conversion to
dinucleotide cofactor analogues. For instance, BAD, the
NADþ analogue incorporating benzamide is a potent
inhibitor of adenosine diphosphate (ADP) ribosyl cyclase,
an important regulatory enzyme involved in the production
of a modulator of Ca2þ concentration in cells.7 Conse-
Thiophenfurin, the thiophene homologue of ribosyl nicotin-
amide, was shown to possess good selectivity towards
tumour cells in vitro and high-level of conversion to its
dinucleotide form. Yet, in vivo, this derivative was found to
be toxic.8 Unspecific enzyme inhibition and metabolic
modifications of thiophenfurin might be responsible for
such an outcome.
Consequently, in order to optimise enzyme and cell
selectivity and maintain high levels of inhibition, novel
C-nucleosides that possess particular structural features for
specific recognition are required. Many dinucleotide-binding
enzymes bind NADþ via a fold consisting of two mono-
nucleotide-binding motifs.9 When one considers the
Figure 1. Ribosyl nicotinamide and analogues.
Keywords: C-nucleosides; Dinucleotides; NADþ analogues.
*
Corresponding author. Tel.: þ44-2890974339; fax: þ44-2890382117; e-mail address: m.migaud@qub.ac.uk
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.05.091

6610
N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
structural features of all the ribosyl nicotinamide analogues
synthesised thus far, one notices that the amide moiety is a
conserved functionality. Indeed, based on the available
crystallographic data of enzymes co-crystallised with
C-NADþ analogues, the amide moiety is involved in
important hydrogen bonding interactions, acting as a
hydrogen bond donor.10–13 Consequently, a slight modifi-
cation of this important structural feature might result in an
abolition of binding when such interaction accounts for
much of the recognition and stabilisation of the inhibitor–
enzyme complexes. Meanwhile, such modification might
provide the means to improve selectivity. We aim to
establish whether selective enzymatic recognition can be
observed amongst different classes of NADþ-dependent
enzymes by exchanging the amide moiety for an isoelec-
tronic group capable of hydrogen bond interactions. As
such, a nitrile group can be viewed as an isosteric group to
an amide, capable of engaging in hydrogen bond inter-
actions with an enzyme-binding pocket as a hydrogen-bond
acceptor. To the best of our knowledge, there are no nitrile
containing C-nucleosides, analogues of ribosyl nicotina-
mide thus far reported in the literature. Considering the good
selectivity for tumour cells showed by thiophenfurin and
the possibility of improving upon enzyme specificity by
modifying the amide moiety, we have synthesised four
novel C-nucleoside analogues (1aa/b and 1ba/b) that
incorporate a thiophene residue substituted by a nitrile
group (Fig. 1). Three of these C-nucleosides have been
converted to their dinucleotide parents and evaluated
against three NADþ-dependent enzymes that catalyse
different types of reaction.
ribofuranosylthiophene-3-carboxamide and 5-(a/b)-D-ribo-
furanosylthiophene-3-carboxamide using classical Friedel–
Crafts’ conditions starting with the tetraacetate ribofuranose
and the appropriate ethyl thiophene carboxylates.8 When
applied to carbonitrile thiophenes, this method failed to
yield any C-glycosylation product. Attempts to introduce
any other types of thiophenes, including the ethyl thiophene
carboxylate employed by Franchetti and thiophene itself,
either led to no product formation or polymerised
unidentifiable material.
Consequently, the addition of the lithiated derivatives of
thiophenes 2a and 2b to the easily accessible 5-O-tert-
butyldimethylsilyl-2,3-O,O-isopropylidene-D-ribono-1,4-
lactone 3,10 4 followed by removal of the hydroxyl group in
position 1 of lactol 4 was employed.15,16 (Scheme 1) The
thiophene-3 (and -2)-carbonitrile 2a (and 2b), treated with
LDA in THF, were reacted with lactone 3 for 30 min at
278 8C to form the lactol. Quenching of the reaction at
278 8C by addition of a solution of saturated NH4Cl
allowed the isolation of the hemiketals 4a and 4b as single
isomers. Allowing the reaction to warm up or increasing the
reaction time led to the opening of the sugar rings to form
the ketones. The ketones were easily identified by 1H NMR;
for instance, dH-40 shifted from 4.46 ppm of 4b to 3.70 ppm
in the ketone. A shift of the thiophene protons was also
observed; dH-3 and dH-4 were 7.47 and 7.10 ppm for 4b,
respectively and 7.94 and 7.57 ppm for the corresponding
ketone. This observation is in agreement with the strong
deshielding effect a carbonyl has on aromatic protons.
While combinations of silane and Lewis acids are known to
reduce hemiketals, attempts to carry out direct dehydrox-
ylation of 4a and 4b were unsuccessful. The chemical
instability of similar carbohydrates towards such reaction
conditions had been previously observed.16,17
2. Results and discussion
2.1. Chemistry
Townsend had reported a procedure involving an acetylated
hemiketal intermediate prepared in situ by trapping of the
Franchetti et al. have reported the synthesis of 2-(a/b)-D-
Scheme 1. Synthesis of carbonitrile thiophene C-nucleoside analogues.

N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
6611
alkoxide, addition product of the lithiated aryl onto the
lactone, with acetic anhydride at 270 8C.18 This two-step
reaction yielded the acetylated hemiketals 5a and 5b,
quantitatively. It should be noticed that unlike previously
reported, only one single ‘anomer’ was formed. Attempts to
observed during the nucleophilic addition on lactone 3 to the
literature precedents, one must consider the combination of
effects due to the steric bulk of the nucleophile, the steric
bulk of the 50-substituent on the lactone, the activation of the
lactone through complexation and finally the electronic
nature of the nucleophile.
0
1
identify the stereochemistry at C-1 either by H NMR and
proton-proton nuclear Overhauser effect (NOE) difference
experiments or by crystallographic analyses were unsuc-
cessful. Therefore, the assignments of the anomeric
The subsequent removal of the acetoxy residue under Lewis
acid conditions was optimised when TMSOTf and Et3SiH in
DCM were used in excess and the reaction was carried out
over 30 min. Partial desilylation was detected and, conse-
quently, after addition of triethylamine, TBAF was added
to yield the desilylated C-nucleosides 1aa/b and 1ba/b.
The C-nucleosides 1 were recovered in moderate yields
(47–50%) after purification as the a and b isomers. The a/b
ratio was always found to be in favour of the a isomer at 1.4/
1 when TBDMS was used as protecting group at C-50. The
yields for this reaction were also slightly low when
compared with those reported in the literature for similar
C-glycosides.15,17 The a and b anomers were identified
1
configurations were based on the H NMR-Dd values of
the isopropylidene methyl groups.19 Dd values superior to
0.20 ppm indicate that the acetate group is cis to the
isopropylidene moiety.
The formation and isolation of a single isomer whether the
arylation reaction is quenched with NH4Cl or Ac2O at low
temperature is at odds with the results reported by Town-
send18 and by Dondoni.16 Yet, Benhinda recognised that the
condensation step was very sensitive to steric factors
imposed by the incoming nucleophile,17 which could
explain the results described by Townsend. However, the
nucleophile employed in the study by Dondoni, a thiazolyl
anion, is isosteric to the thiophene-3 (and -2) -carbonitrile
2a (and 2b) used in the present study. When the
tribenzylated ribonolactone was used instead of 3, no
addition product could be detected. Consequently, acti-
vation of the lactone and a-configuration of the product
could be attributed to a complexation of the lactone oxygen
atoms at position 1 and 2 by the lithium cation, thus
puckering the ribonolactone ring such that only an axial
nucleophilic attack can be favoured. In addition, it is
expected that steric effects due to the isopropylidene
substituent would direct the attack of the nucleophile to
the less hindered face of the lactone. When 5-O-TBDPS-
2,3-O,O-isopropylidene ribonolactone was used instead of
3, the two anomers a and b were identified by 1H NMR (2:1
ratio). Yet no ring-open form could be detected when the
reaction was quenched with Ac2O at low temperature,
indicating that the two stereoisomers were addition but not
rearrangement products. Such explanation should tally with
Dondoni’s results when thiazole was used as nucleophile,
yet it did not, as Dondoni observed the formation of opened
compounds. The electronic effect of the nucleophile must
therefore be considered.
1
by H NMR NOE experiments as the anomeric protons
were distinctively deshielded (d¼5.65 ppm for 1aa versus
d¼5.20 ppm for 1ab). When selectively irradiated, the
signal of the a anomer 1aa0 H-10 gives an intensity
enhancement of the H-20 and H-3 signals. The glycosylation
position was determined by 1H NMR. The NMR pattern for
the aromatic protons of 1a, a doublet, confirms the position of
glycosylation of the thiophene at C-2. When the H-10 signal of
1b was irradiated, an NOE effect was observed at H-3.
Various conditions have been tried in order to increase
yields and selectivity. Alternative Lewis acids, solvent and
reducing reagents were considered, yet TMSOTf/Et3SiH
remained the only suitable combination. Competitive
desilylation and removal of the isopropylidene group were
thought to take place when BF3·Et2O was used as Lewis
acid. Unidentifiable material was formed when DIBAH was
used in combination with AlCl3, while no reaction occurred
with Y(OTf)3. Finally, SnCl4 yielded only the a-anomer
products in poor yields. Optimised conditions were found to
be 5 equiv. of Et3SiH, 2.5 equiv. of TMSOTf in DCM
starting the reaction at 0 8C and warming up to room
temperature (Scheme 2).
The phosphorylation reaction of the C-nucleosides 1aa/b
and 1ba/b was carried out following the Yoshikawa
procedure,7 this employing triethyl phosphate as solvent
and activator. Addition of POCl3 was done at room
temperature and the reaction was monitored by HPLC
(SAX column, K2HPO4·50 mM, 5% MeOH, pH 3.5). Once
the reaction complete, the excess of POCl3 is quenched
by addition of water. This resulted in the simultaneous
removal of the isopropylidene group. Purifications of the
C-nucleotides were achieved on AG-MP1 anion exchange
resin using a gradient of TFA with an average recovery of
25%. The purification was not optimised.
Since the thiophene-3 (and -2)-carbonitrile 2a (and 2b),
both yield a single anomeric adduct product when reacted
with 3, it is highly probable that both reagents yield the
kinetic addition products, which are also the most thermo-
dynamically stable compounds under the reaction con-
ditions. Compared with carbonitrile thiophenene, thiazole
is a more electron-deficient aromatic. The formation of
the thermodynamic product enol–ether intermediate in the
a/b-rearrangement of the glycosylation product, obtained
by addition of the lithiothiazole on 3, might result from such
a chemical characteristic. Such an enol is stabilised through
conjugation between the thiazolyl substituent and the enol
double bond.16 Carbonitrile thiophenyl substituents, on the
other hand, are sufficiently stable to not require extra
conjugation that entails ring opening and electron delocali-
sation between the ketone and the thiophene ring. Indeed, no
enol acetate was ever detected during this investigation.
Consequently, in order to correlate the present results
The diphosphate linkage formation between 6ab and 6ba/b
and adenosine monophosphate employed the commercially
available
adenosine
50-monophosphomorpholidate
(4-morpholine-N,N0-dicyclohexylcarboxamidine salt).20
The reaction was carried out in 4 days in a 0.2 M solution
of MnCl2 in formamide in presence MgSO4 and was

6612
N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
Scheme 2. Dinucleotide analogues synthesis.
monitored by HPLC (SAX column, K2HPO4 50 mM, 5%
MeOH, pH 3.5). Purifications of the three dinucleotides 7ab
and 7ba and 7bb were carried out on DEAE sepharose
columns with a triethylammonium formate gradient. The
isolated yields ranged from 30 to 42% after purification as
calculated by phosphorus titration using Ames’ assay.21
observation indicated that the initial recognition and binding
of NADþ must occur at the northern ribose (NMN-end of
NADþ). Co-crystallisation of the cyclase with nicotinamide
indicated strong hydrogen-bonding interaction between the
amide moiety of nicotinamide and polar residues present in
the binding pocket. As could be anticipated, when the three
carbonitrile thiophene adenine dinucleotides 7ab, 7ba and
7bb were evaluated against Aplysia ADP ribosyl cyclase in
the presence of NADþ (data not shown), none of them was
found to be an inhibitor. When high concentrations of
2.2. Enzyme inhibition
Franchetti has carried out extensive crystallography and
computational studies on furafurin and thiophenfurin.8 In
his study, he observed that the S-atom remained cis to the
furanose-oxygen and ab initio calculations suggested that
this conformation was stabilised by an electrostatic
interaction between a positively charged thiophene sulfur
and a negatively charged furanose oxygen. Therefore, based
on these observations, we could assume similar restricted
rotation around the C-glycosidic bond of 7ab and 7bb.
With such assumption, none of the C-dinucleotides are
expected to bind like NADþ since the carbonitrile
substituent will not be in an appropriate position to share
hydrogen-bonding interactions with the amino acid residues
that are involved in bonding interactions with the amide
moiety of nicotinamide in NADþ. Yet, the following results
describe the effect of each of these C-NADþ analogues on
three enzymes, each of which catalyses a distinct chemical
reaction and each known to bind NADþ via a slightly
different pattern.
C-dinucleotides were used (c.a. 500 mM, KmNAD
,
120 mM), activation of the cyclase was clearly detected
(data not shown). While such activation has never been
reported for the Aplysia Californica cyclase, the human
homologue, CD38, has been shown to be modulated by
nucleotides such as GTP and ATP.25,26 It is possible that at
high concentration of C-dinucleotides, Aplysia cyclase is
itself allosterically modulated.
2.2.2. Glutamate dehydrogenase. In the case of glutamate
dehydrogenase (GDH) from C. symbiosum, NADþ is the co-
factor which allows the oxidative deamination of L-
glutamate to obtain the corresponding 2-oxo-glutarate.
Baker et al.27 showed that ‘the dinucleotide is bound in an
extended conformation with the nicotinamide moiety deep
in the cleft between the two domains’ of GDH. In this case,
the NH2 group of the nicotinamide in the syn conformation
is involved in a hydrogen bond with the oxygen of the
asparagine in position 240, while the hydroxyl group of
threonine 209 interacts with the CO of the same nicotina-
mide moiety. On the other side, the enzyme provides a
pocket for the adenine ring and several amino acids are
involved in the stabilization of the ribose and the phosphate
groups. These characteristics appear to be conserved in all
the hexameric GDH sequences.
2.2.1. Adenosine diphosphate ribosyl cyclase. Based on
crystallographic data, this enzyme was thought to bind
NADþ via the two nucleoside moieties (i.e., adenosine and
ribosyl nicotinamide) with little binding stabilisation
involving the pyrophosphate linkage.22,23 Indeed this
assumption was further supported by the fact that the
cyclase catalyses the cyclisation of an NADþ derivative for
which the pyrophosphate group was partially protected with
o-nitro-phenol moieties.24 Furthermore, nicotinamide
mononucleotide (NMN) is a known potent inhibitor of the
cyclase, while adenosine monophosphate is not.12 This
An initial screening of compounds 7ab, 7ba and 7bb was
performed in presence of a standard concentration of NADþ
(1 mM) and L-glutamate (40 mM)28 and showed the
strongest inhibition effect for 7bb as reported in Figure 2.

N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
6613
Figure 2. Effect of 7ab, 7bb, 7ba on the specific activity (reported as
100% without inhibitor) of GDH.
Figure 4. Kms of 7bb obtained for different concentrations of the substrate.
Despite the fact that the concentration of 1 mM NADþ is
,10 fold the Km, 0.5 mM of 7bb inhibited the activity of
the enzyme almost completely. In marked contrast, 7ba had
very little effect.
2.2.3. Histone deacetylase, Sir2. Sir2 histone deacetylase
has recently come under scrutiny when it was discovered
that unlike most histone deactylase, it was an NADþ
dependent 0 enzyme.29 The nature of the product of the
reaction (1 -O-Ac- vs 20-O-Ac-ADP) and the mechanism by
which the products are formed are still controversial and
under investigation. The assay methods currently available
are also limited and while chromatographic analyses
developed for the kinetic evaluations of Aplysia cyclase
were initially deemed appropriate, the results obtained were
inconsistent and no conclusive results could be drawn with
regards to the effect of these dinucleotides on this
deacetylase. We are currently optimising methods to assay
for deacetylase activity more reliably.
A more detailed experiment was carried out with compound
7bb to define its Ki. In this case, the concentration of NADþ
in solution was varied from 100 to 500 mM and for each
concentration of NADþ a set of five concentrations of the
inhibitor (0.1–5 mM) was tested. The reactions were
performed at 25 8C in phosphate buffer 0.1 M at pH 7.0,
and the activity was measured by following the formation of
NADH at 340 nm (UV Spec, Cary 50). The results are
shown in Figure 3.
From each experiment, a Km value was calculated and
plotted versus the concentration of the inhibitor in Figure 3
to obtain the inhibition constant Ki measured as the negative
abscissa intercept (Fig. 4).
3. Conclusion
Both b-NADþ analogues were selective inhibitors of the
dehydrogenase, yet potency was dependent of the substi-
tution pattern on the thiophene ring. Potent inhibition was
obtained when the thiophene carbonitrile group could act as
a hydrogen bond acceptor. To position the nitrile group and
achieve appropriate binding, the electrostatic interactions
between the positively charged thiophene sulfur and the
negatively charged furanose oxygen must have been
compensated for. One can therefore anticipate that a
carbonitrile-benzene derivative will achieve similar inter-
action with the dehydrogenase-binding pocket and display
better inhibition due to the improved overall stabilisation.
We are currently synthesising such carbonitrile-containing
NADþ analogues to establish whether improved selectivity
compared to that of the benzamide derivatives and potency
compared to that of the thiophene series described here
could be observed.
This elaboration yielded a Ki¼1.09 mM for 7bb, confirming
the strong inhibition effect produced by this NADþ
analogue. From the observation of the Vmax in Figure 2,
we can also conclude that 7bb is a predominantly
competitive inhibitor.
4. Experimental
4.1. Generalities
Chemicals were purchased from Sigma-Aldrich Chemical
Company, Lancaster or ACROS. Solvents for extractions
and chromatography were technical grade. Solvents used in
reactions were freshly distilled from appropriated drying
agents before use. All other reagents were recrystallised or
Figure 3. Lineweaver–Burk plot of the inhibition effect of 7bb on the
GDH reaction. The inhibitor concentrations were 5 mM ( ), 1 mM ( ),
0.5 mM ( ), 0.25 mM ( ), 0.1 mM ( ), and 0 mM ( ).

6614
N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
distilled as necessary. All reactions requiring anhydrous or
inert conditions were carried out in oven-dried glassware
under a positive atmosphere of argon. Solutions or liquids
were introduced using oven dried syringes or cannula
through rubber septa. All reactions were stirred magneti-
cally using Teflon-coated stirs bars. In the cases requiring
278 8C cooling, the reactions were chilled with a dry
ice/acetone bath. Removal of solvents was accomplished
using a rotary evaporator at water aspirator pressure or
under high vacuum (0.5 mm Hg). Analytical TLC was
performed with Merck Silica gel 60 F254 plates. Visualisa-
tion was accomplished by UV-light (l¼254 nm) and/or
staining with an anisaldehyde solution, followed by heating.
Column chromatography were carried out using Fluoro-
(s, 9H, (CH3)3CSi), 1.35 and 1.42 (s£2, 6H, C(CH3)2), 3.74
(dd, 1H, J¼11.3, 1.4 Hz, H-5a), 3.83 (dd, 1H, J¼11.3,
2.1 Hz, H-5b), 4.55 (m, 1H, H-4), 4.65 and 4.67 (2£d, 2H,
J¼5.6, 5.6 Hz, H-2, H-3). d 13C ppm (CDCl3, 75 MHz):
25.4 and 25.2 ((CH3)2Si), 18.6 (SiC(CH3)3), 26.0, 26.1,
27.8 (C(CH3)2, SiC(CH3)3), 63.3 (C-5), 76.2, 78.8 (C-2,
C-3), 82.6 (C-4), 113.4 (C(CH3)2), 174.5 (CvO-1). IR
cm21(KBr): 2989, 2954, 2859 (CH), 1775 (CvO).
Method B. D-Ribose (5.0 g, 33.30 mmol) was suspended in
acetone and 2,2-dimethoxypropane (8.2 mL, 66.60 mmol,
2 equiv.) followed by HCR-W2 Hþ-Dowex resin, was
added to the mixture. When the solution had become clear,
the resin was filtered off and rinsed with acetone. After
concentration, the crude yellow oil was dissolved in dry
dichloromethane (15 mL) and DMAP (0.41 g, 3.33 mmol,
0.1 equiv.) and triethylamine (4.75 mL, 36.60 mmol,
1.1 equiv.) were added to the solution. After cooling the
solution to 0 8C, tertbutyldimethylsilyl chloride was added
(5.52 g, 36.60 mmol, 1.1 equiv.) and the solution was stirred
under argon atmosphere for 6 h. The mixture was then
filtered over celite and washed with saturated aqueous
NH4Cl. The aqueous layer was extracted twice with
dichloromethane. The combined organic layers were dried
over MgSO4, concentrated and used without purification.
The crude mixture was dissolved in 300 mL of acetone and
10.53 g (66.60 mmol, 2 equiv.) of potassium permanganate
are added. The solution was stirred at 50 8C for 6 h. The
mixture was filtered over celite and concentrated. The same
purification as described previously yielded 3 (4.00 g, 40%
over the three steps).
˚
chem Silica gel 40263 mm, 60 A. HPLC monitoring was
˚
accomplished using a Supelcosil SAX1 (120 A, 5 mm,
25 cm£4.6 mm) using a buffer KH2PO4 50 mM/MeOH
95/5, pH¼3.5 at a flow rate of 1 mL/min. Purification of
phosphate compounds were carried out as specified on
BioRad AGwMP-1M resin (100–200 mesh, chloride form)
or Amersham Biosciences DEAE Sepharosee Fast Flow
1
resin. H, 13C and 2D (H-COSY, HMQC) NMR spectra
were all recorded on Bru¨ker avance DPX 300 and Bru¨ker
avance DPX 500. Infrared spectra were recorded on a Perkin
Elmer Spectrum RX 1 FT-IR System, using KBr discs. UV
spectra were recorded using a Perkin Elmer Lambda 800
UV/Vis spectrometer. For known compounds, NMR data
are compared to the one given in the literature. The names
given between parenthesis use casual nomenclature.
4.2. Synthetic Procedures
5-O-tert-butyldimethylsilyl-2,3-O,O-isopropylidene-g-
ribonolactone 3.14 Method A. g-Ribonolactone (500 mg,
3.37 mmol) was suspended in 10 mL of acetone and 2,2-
dimethoxypropane (0.83 mL, 13.49 mmol, 2 equiv.) and
HCR-W2 Hþ-Dowex resin (100 mg) were added. After
stirring at room temperature for 4 h, the mixture was filtered
and the resin rinsed with acetone. The filtrate was
concentrated to give in a quantitative yield the intermediate
as a white powder, which was used without further
2,2-Dimethyl-4-(3-cyano-thiophen-2-yl)-6-hydroxy-
methyl-tetrahydro-furo[3,4-a][1,3]dioxole 1a. (2-(20,30-
O,O-isopropylidene-10-deoxyribofuranosyl)thiophene-3-
carbonitrile).
Intermediate 4a. 6-(tert-Butyl-dimethyl-silan-
oxymethyl)-2,2-dimethyl-4-(3-cyano-thiophen-2-yl)-
tetrahydro-furo[3,4-a][1,3]dioxol-4-ol. (5-O-tert-butyldi-
methylsilyl-2,3-O,O-isopropylidene-1-(2-thiophene-3-carbo-
nitrile)-a-D-ribofuranose).
1
purification. d H ppm (CDCl3, 500 MHz): 1.39 and 1.48
(s£2, 6H, C(CH3)2), 2.81 (br, 1H, 5-OH), 3.81 (bd, 1H,
J¼12.3 Hz, H-5a), 4.00 (bd, 1H, J¼12.3 Hz, H-5b), 4.64
(t, 1H, J¼1.9 Hz, H-4), 4.77 (d, 1H, J¼5.6 Hz, H-3), 4.84
(d, 1H, J¼5.7 Hz, H-2). d 13C ppm (CDCl3, 75 MHz): 25.8
and 27.1 (C(CH3)2), 62.3 (C-5), 76.1, 78.7 (C-2,C-3), 83.2
(C-4), 113.5 (C(CH3)2), 175.5 (CvO-1). IR cm21(KBr):
3463 (OH), 2991, 2934 (CH), 1752 (CvO).
n-Butyl lithium (2 M solution in pentane) (1.25 mL,
2.42 mmol, 1.5 equiv.) was added to a freshly distilled
THF (20 mL) solution of diisopropylamine (0.35 mL,
2.42 mmol, 1.5 equiv.) at 278 8C under argon atmosphere.
The solution was stirred at 278 8C for 5 min. Then
thiophene-3-carbonitrile 2a (0.226 mL, 2.42 mmol,
1.5 equiv.) was added and the solution was stirred at
278 8C for 20 min. To this mixture was added by cannula a
solution of 3 (500 mg, 1.66 mmol, 1 equiv.) in dry THF
(20 mL). The solution was stirred at 278 8C for 30 min then
the reaction was quenched at 260 8C by addition of ether
(30 mL) and a saturated solution of NH4Cl (30 mL). The
organic layer was washed with a saturated solution of
NaHCO3 then water. The resulting organic layers were
gathered, dried on MgSO4 and concentrated. For analysis
purposes, it was purified by silica gel chromatography
column (hexane/EtOAc 95/5 to 1/1) to obtain 4a (468 mg,
69%) as a light yellow oil. However, this compound is
To a dry DCM solution (5 mL) of the partially protected
lactone was added DMAP (41 mg, 0.34 mmol, 0.1 equiv.)
and triethylamine (0.48 mL, 3.71 mmol, 1.1 equiv.). After
cooling the solution to 0 8C, tert-butyldimethylsilyl chloride
(560 mg, 3.71 mmol, 1.1 equiv.) was added. The mixture
was stirred overnight under argon atmosphere. DCM
(20 mL) was added to the solution and washed with a
saturated solution of NH4Cl then water. The resulting
organic layer was dried over MgSO4 and concentrated.
The crude mixture is purified by silica gel column
chromatography (PE/acetone 99/1) to give 3 as white
1
readily decomposed upon storage. 4a: d H ppm (CDCl3,
300 MHz): 0.17, 0.18 (s£2, 3H£2, (CH3)2Si), 0.96 (s, 9H,
1
crystals (688 mg, 68% over the two steps). 3: d H ppm
(CDCl3, 300 MHz): 0.01 and 0.02 (s£2, 6H, (CH3)2Si), 0.84
(CH3)3CSi), 1.25, 1.48 (s£2, 3H£2, C(CH3)2), 3.84 (dd, 1H,

N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
6615
J¼2.1, 11.2 Hz, H-5a), 3.90 (dd, 1H, J¼2.2, 11.2 Hz, H-5b),
4.56 (m, 1H, H-4), 4.71 (d, 1H, J¼5.7 Hz, H-2), 4.90 (dd,
1H, J¼1.2, 5.7 Hz, H-3), 7.22 (d, 1H, J¼5.2 Hz, HTh-4),
7.36 (d, 1H, J¼5.2 Hz, HTh-5). d 13C ppm (CDCl3,
75 MHz): 25.2 ((CH3)2Si), 18.7 ((CH3)3CSi), 25.0, 26.2,
26.4 (C(CH3)2, (CH3)3CSi), 64.9 (C-5), 82.0 (C-3), 87.1
(C-4), 89.0 (C-2), 106.1 (C-1), 109.5, 113.7, 115.6 (CTh-3,
CN, C(CH3)2), 126.3 (CTh-4), 130.3 (CTh-5), 152.2 (CTh-2).
quenched by a saturated solution of NH4Cl and extracted
with dichloromethane. The organic layer was washed with
water, dried on MgSO4 and concentrated. After purification
on silica gel chromatography column, 1aa (74 mg, 29%)
and 1ab (54 mg, 21%) were isolated. 1ab: d 1H ppm
(CDCl3, 300 MHz): 1.37, 1.64 (s£2, 3H£2, C(CH3)2), 2.23
(m, 1H, OH-50), 3.85 (ddd, 1H, J¼12.4, 3.6, 8.5 Hz, H-50a),
4.00 (ddd, 1H, J¼12.4, 20.8, 4.45 Hz, H-50b), 4.25 (m, 1H,
H-400), 4.63 (m, 1H, H-2 ), 4.87 0(dd, 1H, J¼3.6, 6.6 Hz,
H-3 ), 5.20 (d, 1H, J¼5.7 Hz, H-1 ), 7.23 (d, 1H, J¼5.3 Hz,
HTh-4), 7.33 (d, 1H, J¼5.3 Hz, HTh-5). d 13C ppm (CDCl3,
75 MHz): 25.7, 27.9 (C(CH3)2), 62.9 (C-50), 81.8, 81.8
(C-10, C-30), 85.5 (C-40), 87.4 (C-20), 108.1, 115.4, 115.8
(CTh-3, CN, C(CH3)2), 126.2 (CTh-5), 129.8 (CTh-4), 153.1
(CTh-2). IR (KBr) cm21: 3447 (OH), 2926 (CH), 2229 (CN).
MS (LSIMS) MþH calculated: 282.0800; measured:
(s£2, 3H£2, C(CH3)2), 1.91 (m, 1H, OH-50), 3.85 (m, 2H,
H-50a and H-50b), 4.34 (m, 1H, H-40), 4.88 (dd, 1H, J¼0.9,
5.9 Hz, H-30), 4.97 0(dd, 1H, J¼4.0, 5.9 Hz, H-20), 5.65 (d,
1H, J¼4.0 Hz, H-1 ), 7.19 (d, 1H, J¼5.3 Hz, HTh-4), 7.38
(d, 1H, J¼5.3 Hz, HTh-5). d 13C ppm (CDCl3, 75 MHz):
25.1, 26.5 (C(CH3)2), 63.2 (C-50), 79.3 (C-10), 82.1 (C-20),
83.3 (C-30), 85.1 (C-40), 108.1, 113.7, 115.1 (CTh-3, CN,
C(CH3)2), 127.6 (CTh-5), 128.0 (CTh-4), 151.5 (CTh-2). IR
(KBr) cm21: 3436 (OH), 2924 (CH), 2230 (CN). MS
(LSIMS) MþH calculated: 282.0800; measured: 282.0806.
Intermediate 5a. Acetic acid 6-(tert-butyl-dimethyl-
silanoxymethyl)-2,2-dimethyl-4-(3-cyano-thiophen-2-
yl)-tetrahydro-furo[3,4-a][1,3]dioxol-4-yl ester. (1-O-
acetyl-5-O-tert-butyldimethylsilyl-2,3-O,O-isopropylidene-
1-(2-thiophene-3-carbonitrile)-a-D-ribofuranose).
n-Butyl lithium (2 M solution in pentane) (1.25 mL,
2.42 mmol, 1.5 equiv.) was added to a freshly distilled
THF (20 mL) solution of diisopropylamine (0.35 mL,
2.42 mmol, 1.5 equiv.) at 278 8C under argon atmosphere.
The solution was stirred at 278 8C for 5 min. Then
thiophene-3-carbonitrile 2a (0.226 mL, 2.42 mmol,
1.5 equiv.) was added and the solution was stirred at
278 8C for 20 min. To this mixture was added by cannula a
solution of 3 (500 mg, 1.66 mmol, 1 equiv.) in dry THF
(20 mL). The solution was stirred at 278 8C for 30 min then
warmed to 270 8C/260 8C and acetic anhydride (0.78 mL,
8.28 mmol, 5 equiv.) was added. After stirring another
30 min at 260 8C, the reaction was quenched at 260 8C by
addition of ether (30 mL) and a saturated solution of NH4Cl
(30 mL). The organic layer was washed with a saturated
solution of NaHCO3 then water. The combined aqueous
layers were washed with ether and the resulting organic
layers were gathered, dried on MgSO4 and concentrated.
This crude mixture was nearly pure and used without
purification for the next step. For analysis purposes, it was
purified by silica gel chromatography column (PE/Acetone
9/1) to obtain 5a (713 mg, 95%) as a light yellow oil
However this compound is readily decomposed upon
1
282.0795. 1aa: d H ppm (CDCl3, 300 MHz): 1.34, 1.49
2,2-Dimethyl-4-(2-cyano-thiophen-5-yl)-6-hydroxy-
(5-(20,
methyl-tetradro-furo[3,4-a][1,3]dioxole
1b.
30-O,O-isopropylidene-10-deoxyribofuranosyl)thiophene-2-
carbonitrile).
Intermediate 4b. 6-(tert-Butyl-dimethyl-silan-
oxymethyl)-2,2-dimethyl-4-(2-cyanothiophen-5-yl)-
tetrahydro-furo[3,4-a][1,3]dioxol-4-ol. (5-O-tert-butyldi-
methylsilyl-2,3-O,O-isopropylidene-1-(5-thiophene-2-carbo-
nitrile)-a-D-ribofuranose).
1
storage. 5a: d H ppm (CDCl3, 300 MHz): 0.06 and 0.07
(s£2, 6H, (CH3)2Si)), 0.86 (s, 9H, (CH3)3CSi), 1.39, 1.73
(s£2, 3H£2, C(CH3)2), 2.25 (s, 3H, OAc), 3.83 (dd, 1H,
J¼11.3, 2.6 Hz, H-5a), 3.91 (dd, 1H, J¼11.3, 3.2 Hz, H-5b),
4.49 (m, 1H, H-4), 4.61 (d, 1H, J¼6.4 Hz, H-2), 4.83 (dd,
1H, J¼6.4, 2.1 Hz, H-3), 7.14 (d, 1H, J¼5.3 Hz, HTh-4),
7.33 (d, 1H, J¼5.3 Hz, HTh-5). d 13C ppm (CDCl3,
75 MHz): 25.2 and 24.9 ((CH3)2Si), 18.6 ((CH3)3CSi),
21.8 (CH3 Ac), 26.1 (), 26.2 ((CH3)3CSi), 63.0 (C-5), 81.1
(C-2), 84.9 (C-4), 88.7 (C-3), 103.7 (C-1), 106.8, 114.5,
116.3 (CTh-3, CN, C(CH3)2), 125.8 (CTh-5), 130.4 (CTh-4),
153.8 (CTh-2), 168.8 (OAc). IR (KBr) cm21: 2932, 2858
(CH), 2232 (CN), 1763 (CvO). MS (LSIMS) M-OAc: 394.
The same procedure as the one used to prepare 4a was used
to prepare 4b for analysis purposes. However, the
compound is readily decomposed upon storage. 4b d H
1
ppm (CDCl3, 300 MHz): 0.18, 0.19 (s£2, 3H£2, (CH3)2Si),
0.96 (s, 9H, (CH3)3CSi), 1.30, 1.56 (s£2, 3H£2, C(CH3)2),
3.83 (dd, 1H, J¼2.1, 11.3 Hz, H-5a), 3.87 (dd, 1H, J¼2.0,
11.3 Hz, H-5b), 4.46 (m, 1H, H-4), 4.59 (d, 1H, J¼5.7 Hz,
H-2), 4.90 (dd, 1H, J¼4.6, 5.7 Hz, H-3), 7.24 (d, 1H,
J¼4.1 Hz, HTh-4), 7.67 (d, 1H, J¼4.1 Hz, HTh-3). d 13C
ppm (CDCl3, 75 MHz): 25.3((CH3)2Si), 18.7 ((CH3)3CSi),
25.2, 26.2, 26.8 (C(CH3)2, (CH3)3CSi), 64.9 (C-5), 82.2
(C-2), 86.8 (C-4), 88.9 (C-3), 105.8, 110.2, 113.6, 115.0
(C-1, CTh-2, CN, C(CH3)2), 126,9 (CTh-4), 137.3 (CTh-3),
150.5 (CTh-5). IR (KBr) cm21: 3328 (OH), 2933, 2859
(CH), 2221.1 (CN).
The crude mixture of 5a (1.19 mmol) was dissolved in dry
dichloromethane (10 mL) with molecular sieves, under
argon atmosphere. Triethylsilane (0.95 mL, 5.97 mmol,
5 equiv.) and freshly distilled trimethylsilyl trifluoro-
methanesulfonate (0.54 mL, 2.98 mmol, 2.5 equiv.) were
added to the solution at 0 8C. The reaction was then warmed
up to room temperature and stirred for 30 min. No
remaining starting material was detected on TLC, triethyl-
amine (2 mL) were added to the solution and then
tetrabutylammonium fluoride (1 M solution in THF)
(2.4 mL, 2.38 mmol, 2 equiv.). After 4 h, the mixture was
Intermediate 5b. Acetic acid 6-(tert-butyl-dimethyl-
silanoxymethyl)-2,2-dimethyl-4-(3-cyano-thiophen-5-
yl)-tetrahydro-furo[3,4-a][1,3]dioxol-4-yl ester. (1-O-
acetyl-5-O-tert-butyldimethylsilyl-2,3-O,O-isopropylidene-
1-(5-thiophene-2-carbonitrile)-a-D-ribofuranose).
The same procedure as the one used to prepare 5a was used
to prepare 5b. Purification for analysis purposes on silica gel

6616
N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
column (PE/Acetone 95/5) afforded pure 5b as a powder
(95%). However the compound is readily decomposed upon
storage. 5b: d H ppm (CDCl3, 300 MHz): 0.05 and 0.06
(C-10), 84.4 (d, JC–C–O–P¼8.5 Hz, C-40), 108.1, 115.7 (C-3,
CN), 128.1 (C-5), 129.5 (C-4), 154.6 (C-2). d 31P ppm (D2O,
121 MHz): 1.3 (s). MS (ESI-) M2H: 320.
1
(s£2, 3H£2, (CH3)2Si), 0.85 (s, 9H, (CH3)3CSi),1.39, 1.69
(s£2, 3H£2, C(CH3)2), 2.18 (s, 3H, OAc), 3.81 (dd, 1H,
J¼2.5, 11.3 Hz, H-5a), 3.90 (dd, 1H, J¼2.8, 11.3 Hz, H-5b),
4.48 (m, 1H, H-4), 4.61 (d, 1H, J¼6.4 Hz, H-2), 4.83
(dd, 1H, H-3, J¼2.1, 6.4 Hz, H-2), 7.10 (d, 1H, J¼3.9 Hz,
HTh-4), 7.47 (d, 1H, J¼3.9 Hz, HTh-3). d 13C ppm (CDCl3,
75 MHz): 25.2, 25.0 ((CH3)2Si), 18.6 ((CH3)3CSi),
23.0 (OAc), 25.9, 26.2 (C(CH3)2, (CH3)3CSi), 63.1 (C-5),
81.0 (C-2), 85.0 (C-4), 88.3 (C-3), 124.9 (CTh-4), 137.6
(CTh-3).
Compound 6ab (11 mg, 34.3 mmol) was dissolved in a
0.2 M solution of manganese chloride in formamide left on
molecular sieves during several days (0.5 mL). MgSO4
(8 mg, 68.5 mmol, 2 equiv.) and adenosine 50-monopho-
sphomorpholidate·4-morpholine-N,N0-dicyclohexylcarbox-
amidine salt (48.6 mg, 68.5 mmol, 2 equiv.) were added to
this solution. The mixture was sonicated and left to react for
4 days after which no evolution was observed by HPLC
SAX monitoring. 20 mL of water were added to quench the
reaction. The mixture was then purified on a DEAE
sepharose resin column eluted with a gradient of triethy-
lammonium formate 20 to 250 mM to give 7ab with 39%
yield. 7ab: d 1H ppm (D2O, 500 MHz): 3.60–4.45 (m,
protons of the sugars), 4.73 (d, 1H, J¼6.04 Hz, HTh-10), 5.67
(d, 1H, J¼4.0 Hz, HAd-10), 6.68 (d, 1H, J¼4.9 Hz, HTh-4),
6.90 (d, 1H, J¼4.7 Hz, HTh-5), 7.70 (Ad). d 31P ppm (D2O,
121 MHz): 29.53 (m). lmax (H2O)¼253 nm. MS (ES-)
M2H calculated: 649.0525, measured: 649.0528.
The same procedure as the described for 1a was used for the
synthesis of 1b and afford 27% of a isomer and 20% of b
isomer. 1bb: d 1H ppm (CDCl3, 300 MHz): 1.37, 1.60 (s£2,
3H£2, C(CH3)2), 3.77 (dd, 1H, J¼4.1, 12.1 Hz, H-50a), 3.89
(dd, 1H, J¼3.3, 12.1 Hz, H0-50b), 4.22 (m, 1H, H-40), 4.59
(dd,0 1H, J¼5.3, 6.6 Hz, H-2 ), 4.709 (dd, 1H, J¼3.7, 6.6 Hz,
H-3 ), 5.10 (d, 1H, J¼5.2 Hz, H-1 ), 7.07 (d, 1H, J¼3.8 Hz,
HTh-4), 7.54 (d, 1H, J¼3.8 Hz, HTh-3). d 13C ppm (CDCl3,
75 MHz): 25.8, 27.8 (C(CH3)2), 63.0 (C-50), 82.1, 82.9
(C-10, C-30), 85.5 (C-40), 87.2 (C-20), 109.3, 114.5, 115.7
(CTh-2, CN, C(CH3)2), 124.6 (CTh-4), 138.1 (CTh-3), 151.7
(CTh-5). IR (KBr) cm21: 3445 (OH), 2925 (CH), 2228 (CN).
MS (LSIMS) MþH calculated: 282.0800; measured:
b-5(2-cyanothiophene) adenine dinucleotide 7bb
Intermediate 6bb: phosphoric acid mono-[5(R)-(2-
cyano-thiophen-5-yl)-3,4-dihydroxy-tetrahydro-furan-2-
yl methyl] ester. (5-b-D-[10-(50-phosphateribofuranosyl)]
thiophene-2-carbonitrile).
1
282.0807. 1ba: d H ppm (CDCl3, 300 MHz): 1.34, 1.53
(s£2, 3H£2, C(CH3)2), 1.86 (bs, 1H, OH-50), 3.82 (m, 2H,
H-50a and H-50b), 4.30 (m, 1H, H-40), 4.83 (dd, 1H, J¼3.9,
5.9 Hz, H-20), 4.88 0(dd, 1H, J¼1.0, 5.9 Hz, H-30), 5.43 (d,
1H, J¼3.9 Hz, H-1 ), 7.07 (d, 1H, J¼3.8 Hz, HTh-4), 7.52
(d, 1H, J¼3.8 Hz, HTh-3). d 13C ppm (CDCl3, 75 MHz):
25.2, 26.5 (C(CH3)2), 63.5 (C-50), 80.1 (C-10), 82.5 (C-20),
83.3 (C-30), 84.9 (C-40), 110.7, 113.8 (CTh-2, CN, C(CH3)2),
126.2 (CTh-4), 136.8 (CTh-3), 147.8 (CTh-5). IR (KBr)
cm21: 3467 (OH), 2925 (CH), 2226 (CN). MS (LSIMS)
MþH calculated: 282.0800; measured: 282.0785.
The same procedure as described for 6ab was used for the
1
synthesis of 6bb (25%). 6bb: d H ppm (D2O, 500 MHz):
4.20 (m, 2H, H-50a, H-50b), 4.29–4.44 (m, 3H, H-20, H-30,
H-40), 5.25 (d, 1H, J¼7.1 Hz, H-10), 7.36 (d, 1H, J¼3.7 Hz,
HTh-4), 7.85 (d, 1H, J¼3.8 Hz, HTh-3). d 13C ppm (D2O,
75 MHz): 67.3 (d, JC–O–P¼5.0 Hz, C-50), 73.5, 79.5 (C-30,
C-20) 81.3 (C-10), 85.6 (d, JC–C–O–P¼8.7 Hz, C-40), 109.8,
117.0 (CTh-2, CN), 127.8 (CTh-4), 141.2 (CTh-3), 153.4
(CTh-5). d 31P ppm (D2O, 121 MHz): 1.3 (s). MS (ESI-)
M2H: 320. The same procedure as described for 7ab was
used for the synthesis of 7bb (42%). 7bb: d 1H ppm (D2O,
500 MHz): 3.7–3.9 (m, H-50 £4), 4.0–4.4 (m, protons of
the sugars), 5.75 (d, J¼3.6 Hz, HAd-10), 6.74 (HTh-4), 7.22
(HTh-3). d 13C ppm (D2O, 75 MHz): 67.9, 68.5, 69.4, 72.9,
74.1, 77.1, 80.3, 82.4, 86.0, 86.5, 90.3, 109.9, 117.7, 128.1,
141.9, 152.4, 155.0. d 31P ppm (D2O, 121 MHz): 210.0
(m). lmax (H2O)¼260 nm. MS (ES-) M2H calculated:
649.0525, measured: 649.0530.
b-2(3-cyanothiophene) adenine dinucleotide 7ab
Intermediate 6ab. Phosphoric acid mono-[5(R)-(3-
cyano-thiophen-2-yl)-3,4-dihydroxy-tetrahydro-furan-
2-yl methyl] ester. (2-b-D-[10-(50-phosphateribofuranosyl)]
thiophene-3-carbonitrile).
Compound 1ab (32 mg, 0.114 mmol) was solubilised in
triethyl phosphate (1.2 mL) and heated at 50 8C. Then after
cooling down at 0 8C, phosphorus oxychloride (32 mL,
0.342 mmol, 3 equiv.) was added. The mixture was stirred
for 2 days at room temperature and monitored by anion
exchange HPLC (Supelcosil SAX1, buffer KH2PO4 50 mM/
MeOH 95/5). Another 32 mL of POCl3 were added if
needed to complete the reaction. The reaction was quenched
at 0 8C by addition of ice-cold water and stirred for 1 h. The
mixture was extracted with diethyl ether (20 mL£3) and the
aqueous layer was freeze-dried. The crude mixture was
purified on AG MP1 resin with a gradient of 0–150 mM
a-5(2-cyanothiophene) adenine dinucleotide 7ba
Intermediate 6ba: phosphoric acid mono-[5(S)-(2-
cyano-thiophen-5-yl)-3,4-dihydroxy-tetrahydro-furan-2-
yl methyl] ester. (5-a-D-[10-(50-phosphateribofuranosyl)]
thiophene-2-carbonitrile).
The same procedure as described for 6ab was used for the
1
synthesis of 6ba (15%). 6ba: d H ppm (D2O, 300 MHz):
TFA to give 6ab (10 mg, 25%). 6ab: d H ppm (D2O,
3.70–4.39 (m, 5H, H-20, H-30, H-40, H-50a, H-50b), 5.36
(broad s, 1H, H-10), 7.02 (d, 1H, J¼3.6 Hz, HTh-4), 7.59 (d,
1H, J¼3.6 Hz, HTh-3). d 13C ppm (D2O, 75 MHz): 65.4 (d,
JC–O–P¼4.83 Hz, C-50), 72.4, 73.7(C-20, C-30), 79.5 (C-10),
80.9 (d, JC–C–O–P¼7.9 Hz, C-40), 108.7, 115.7 (CTh-2,
1
5000MHz): 3.95 (m, 2H, H-50a, H-50b), 4.100–4.17 (m, 3H,
H-2 , H-30, H-40), 5.17 (d, 1H, J¼7.1 Hz, H-1 ), 7.18 (d, 1H,
J¼5.0 Hz, H-4), 7.43 (d, 1H, J¼5.0 Hz, H-5). d 13C ppm
(D2O, 125 MHz): 65.9 (C-50), 72.1, 78.4 (C-20, C-30), 79.4

N. E. Batoux et al. / Tetrahedron 60 (2004) 6609–6617
6617
CN), 126.8 (CTh-4), 139.1 (CTh-3), 149.0 (CTh-5). d 31P ppm
(D2O, 121.45 MHz): 1.7 (s). MS (ESI-) M2H: 320.
10. Goldstein, B. M.; Colby, T. D. Adv. Enzyme Regul. 2000, 40,
405–426.
11. Baker, P. J.; Waugh, M. L.; Wang, X. G.; Stillman, T. J.;
Turnbull, A. P.; Engel, P. C.; Rice, D. W. Biochemistry 1997,
36, 16109–16115.
The same procedure as described for 7ab was used for the
1
synthesis of 7ba (30%). 7ba: d H ppm (D2O, 500 MHz):
3.7–4.5 (m, protons of the sugars), 5.05 (s, 1H, HTh-10), 5.90
12. Sauve, A. A.; Schramm, V. L. Biochemistry 2002, 41,
8455–8463.
(d, 1H, J¼5.05 Hz, HAd-10), 6.71 (d, 1H, HTh-4), 7.32 (d,
1H, J¼3.0 Hz, HTh-3), 8.05 (s, HAd-2), 8.34 (s, HAd-8). d 31
P
13. Munshi, C.; Aarhus, R.; Graeff, R.; Levitt, T. F.; Lee, H. C.
J. Biol. Chem. 2000, 275, 21566–21571.
ppm (D2O, 121 MHz): 29.79 (m). lmax (H2O)¼261 nm.
MS (ES-) M2H calculated: 649.0525, measured: 649.0527.
14. Kaskar, B.; Heise, G.; Michalak, R. S.; Vishnuvajjala, B. R.
Synthesis 1990, 1031–1032.
15. Dondoni, A.; Sherrmann, M. C. Tetrahedron 1993, 34,
7319–7322.
Acknowledgements
16. Dondoni, A.; Sherrmann, M. C. J. Org. Chem. 1994, 59,
6404–6412.
We thank the MRC for funding (Grant number G0001004)
and the EPSRC National Mass Spectrometry Service
Centre, Swansea for mass spectra. We also thank Dr S.D.
Bell from the Hutchinson/MRC Research Centre, Cancer
Cell Unit in Cambridge for supplying Sir2 samples.
17. Guianvarc’h, D.; Fourrey, J. L.; Dau, M. E. T. H.; Guerineau,
V.; Benhida, R. J. Org. Chem. 2002, 67, 3724–3732.
18. Gudmundsson, K. S.; Drach, J. C.; Townsend, L. B. J. Org.
Chem. 1997, 62, 3453–3459.
19. (a) Rayner, B.; Tapiero, C.; Imbach, J. L. Carbohydr. Res.
1976, 47(2), 195–202. (b) MacCoss, M.; Robins, M. J.;
Rayner, B.; Imbach, J. L. Carbohydr. Res. 1977, 59(2),
575–579.
References and notes
20. Lee, J.; Churchil, H.; Choi, W. B.; Lynch, J. E.; Roberts, F. E.;
Volante, R. P.; Reider, P. J. Chem. Commun. 1999, 729–730.
21. Ames, B. N. Methods Enzymol. 1966, 8, 115–118.
22. Prasad, G. S.; McRee, D. E.; Stura, E. A.; Stout, C. D.; Levitt,
T. F.; Lee, H. C. Nature Struct. Biol. 1996, 3, 957–964.
23. Munshi, C.; Thiel, D. J.; Mathews, I. I.; Aarhus, R.; Walseth,
T. F.; Lee, H. C. J. Biol. Chem. 1999, 274, 30770–30777.
24. Walseth, T. F.; Aarhus, R.; Gurnack, M. E.; Lee, H. C.; Wong,
L.; Breitinger, H. G. A.; Gee, K. R. Methods Enzymol. 1997,
280, 294–305.
1. Pankiewicz, K. W.; Watanabe, K. A.; Lesiak-Watanabe, K.;
Goldstein, B. M.; Jayaram, H. N. Curr. Med. Chem. 2002, 9,
733–741.
2. Pankiewicz, K. W. Pharmacol. Ther. 1997, 76, 89–100.
3. Franchetti, P.; Cappellacci, L.; Perlini, P.; Jayaram, H. N.;
Butler, A.; Schneider, B. P.; Collart, F. R.; Huberman, E.;
Grifantini, M. J. Med. Chem. 1998, 41, 1702–1707.
4. Franchetti, P.; Cappellacci, L.; Marchetti, S.; Martini, C.;
Costa, B.; Varani, K.; Borea, P. A.; Grifantini, M. Bioorg.
Med. Chem. 2000, 8, 2367–2373.
25. Hotta, T.; Asai, K.; Fujita, K.; Kato, T.; Higashida, H.
J. Neurochem. 2000, 74, 669–675.
5. Franchetti, P.; Grifantini, M. Curr. Med. Chem. 1999, 6,
599–614.
26. Okamoto, H. Mol. Cell. Biochem. 1999, 193, 115–118.
27. Baker, P. J.; Britton, K. L.; Engel, P. C.; Farrants, G. W.;
Lilley, K. S.; Rice, D. W.; Stillman, T. J. Proteins 1992, 12,
75–86.
6. Franchetti, P.; Cappellacci, L.; Grifantini, M. Farmaco 1996,
51, 457–469.
7. Migaud, M. E.; Pederick, R. L.; Bailey, V. C.; Potter, B. V. L.
Biochemistry 1999, 38, 9105–9114.
28. Syed, S. E.-H.; Engel, P. C.; Parker, D. M. Biochim. Biophys.
Acta 1991, 1115, 123–130.
8. Franchetti, P.; Cappellacci, L.; Grifantini, M.; Barzi, A.;
Nocentini, G.; Yang, H. Y.; O’Connor, A.; Jayaram, H. N.;
Carrell, C.; Goldstein, B. M. J. Med. Chem. 1995, 38,
3829–3837.
29. (a) Wardleworth, B. N.; Russell, R. J. M.; Bell, S. D.; Taylor,
G. L.; White, M. F. EMBO J. 2002, 21(17), 4654–4662.
(b) Bell, S. D.; Botting, C. H.; Wardleworth, B. N.; Jackson,
S. P.; White, M. F. Science 2002, 296, 148–151.
9. Bottoms, C. A.; Smith, P. E.; Tanner, J. J. Protein Sci. 2002,
11, 2125–2137.