Structural determinants for N1/N7 cyclization of nicotinamide hypoxanthine 5′-dinucleotide (NHD+) derivatives by ADP-ribosyl cyclase from Aplysia californica: Ca2+-mobilizing activity of 8-substituted cyclic inosine 5′-diphosphoribose analogues in T-lymphocytes

Article abstract of DOI:10.1021/jm060275a

A series of nicotinamide hypoxanthine 5′-dinucleotide (NHD ⁺) analogues modified at C-8 (2-5) and 7-deaza-NHD⁺ were synthesized, and cyclization in the presence of Aplysia ADP-ribosyl cyclase was studied. All 8-substituted NHD⁺ analogues were converted into their N1-cyclic forms by the enzyme, while in contrast, 7-deaza-NHD⁺ 17 was hydrolyzed into 7-deazainosine 5′-diphosphoribose (7-deaza-IDPR) 25. Correlations are made showing that the conformation of the NHD⁺ substrate is the key to successful cyclization. The pharmacological activities of these novel cIDPR derivatives were evaluated in both permeabilized and intact Jurkat T-lymphocytes. The results show that in permeabilized cells both 8-iodo 1g and 8-N₃-N1-cIDPR 1d have an activity comparable to that of cADPR, while 8-iodo 1g and 8-phenyl-N1-cIDPR 1c have a small but significant effect in intact cells and can therefore be regarded as membrane-permeant; thus, cIDPR derivatives are emerging as important novel biological tools to study cADPR-mediated Ca²⁺ release in T-cells.

Full text of DOI:10.1021/jm060275a

5162
J. Med. Chem. 2006, 49, 5162-5176
Structural Determinants for N1/N7 Cyclization of Nicotinamide Hypoxanthine 5′-Dinucleotide
(NHD+) Derivatives by ADP-Ribosyl Cyclase from Aplysia californica: Ca2+-Mobilizing Activity
of 8-Substituted Cyclic Inosine 5′-Diphosphoribose Analogues in T-Lymphocytes
Christelle Moreau,† Gerd K. Wagner,†,‡ Karin Weber,§ Andreas H. Guse,§ and Barry V. L. Potter*,†
Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy and Pharmacology, UniVersity of Bath, ClaVerton Down, Bath, BA2
7AY, United Kingdom, and Center of Experimental Medicine, Institute of Biochemistry and Molecular Biology I: Cellular Signal Transduction,
UniVersity Medical Centre Hamburg-Eppendorf, Martinistrasse 52, D-20246 Hamburg, Germany
ReceiVed March 13, 2006
A series of nicotinamide hypoxanthine 5′-dinucleotide (NHD+) analogues modified at C-8 (2-5) and 7-deaza-
NHD+ were synthesized, and cyclization in the presence of Aplysia ADP-ribosyl cyclase was studied. All
8-substituted NHD+ analogues were converted into their N1-cyclic forms by the enzyme, while in contrast,
7-deaza-NHD+ 17 was hydrolyzed into 7-deazainosine 5′-diphosphoribose (7-deaza-IDPR) 25. Correlations
are made showing that the conformation of the NHD+ substrate is the key to successful cyclization. The
pharmacological activities of these novel cIDPR derivatives were evaluated in both permeabilized and intact
Jurkat T-lymphocytes. The results show that in permeabilized cells both 8-iodo 1g and 8-N3-N1-cIDPR 1d
have an activity comparable to that of cADPR, while 8-iodo 1g and 8-phenyl-N1-cIDPR 1c have a small
but significant effect in intact cells and can therefore be regarded as membrane-permeant; thus, cIDPR
derivatives are emerging as important novel biological tools to study cADPR-mediated Ca2+ release in
T-cells.
Introduction
Ca2+ is a ubiquitous intracellular messenger that controls a
diverse range of cellular functions and plays a pivotal role in
processes such as fertilization, gene transcription, muscle
contraction, cell proliferation, and secretion of bioactive com-
pounds. These processes occur despite the fact that in other
circumstances Ca2+ is highly toxic and induces programmed
cell death (apoptosis).1,2 The release of Ca2+ from internal stores
is mediated through three second messengers: D-myo-inositol-
1,4,5-trisphosphate (Ins(1,4,5)P3), nicotinic acid 5′-adenine
dinucleotide 2′-phosphate (NAADP), and cyclic adenosine 5′-
diphosphate ribose (cADPRa). While Ins(1,4,5)P3 acts on the
Ins(1,4,5)P3 receptor, pharmacological studies have revealed that
cADPR elicits Ca2+ release through the ryanodine receptor,
though whether this happens by direct receptor binding or via
a binding protein remains unclear.3-6 Over the years a number
of cADPR analogues have been synthesized and their biological
activities examined in order to characterize the cADPR signaling
pathway and identify the binding protein.7-9
Figure 1. Cyclization behavior of NAD+ and NHD+ by ADPRC.
Hydrolysis of cADPR at the unstable N1 glycosidic bond
linkage to give ADP-ribose occurs both under neutral aqueous
conditions and, rapidly, in the physiological conditions of the
cell.10,11 This chemical and biological instability has limited
studies of the physiological role of cADPR to some extent, and
it was thus recognized that cADPR analogues resistant to both
chemical and enzymatic hydrolysis were required. The enzyme
ADP-ribosyl cyclase (ADPRC) isolated from Aplysia californica
mainly exerts a cyclase activity (though NAD+-glycohydrolase
has been described too)12 and has been used extensively to
convert NAD+ and NAD+ derivatives into cADPR and the
corresponding cADPR analogues, respectively.10,13,14 While
NAD+ and its derivatives are exclusively cyclized at the N1
position by ADPRC, enzymatic cyclization of NHD+ takes place
at N7, leading to biologically inactive N7-cIDPR (Figure 1).15,16
Because of this aberrant cyclization behavior of NHD+, N1-
cIDPR analogues have previously been accessible only by total
chemical synthesis.17-20 Recently, we have shown that place-
ment of a bromo substituent at the 8 position of NHD+ restores
* To whom correspondence should be addressed. Phone: ++44-1225-
386639. Fax: ++44-1225-386114. E-mail: B.V.L.Potter@bath.ac.uk.
† University of Bath.
‡ Current address: School of Chemical Sciences and Pharmacy, Uni-
versity of East Anglia, Norwich, NR4 7TJ, U.K.
§ University Medical Centre Hamburg-Eppendorf.
a Abbreviations: cADPR, cyclic adenosine 5′-diphosphoribose; cIDPR,
cyclic inosine 5′-diphosphoribose; cIDPRE, N1-[(2′′-O-phosphorylethoxy)-
methyl]-5′-O-phosphorylinosine 2′′,5′-cyclicpyrophosphate; NAD+, nico-
tinamide adenosine 5′-dinucleotide; NGD+, nicotinamide guanosine 5′-
dinucleotide; NHD+, nicotinamide hypoxanthine 5′-dinucleotide; ADPRC,
ADP-ribosyl cyclase; TEP, triethyl phosphate; â-NMN, â-nicotinamide 5′-
mononucleotide; CDI, carbonyldiimidazole; TEAB, triethylammonium
bicarbonate; IMP, inosine monophosphate; TDA-1, tris[2-(2-methoxy-
ethoxy)ethyl]amine.
10.1021/jm060275a CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/28/2006

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5163
of 0.05M TEAB against MeCN to yield the desired monophos-
phates 10a-c as their triethylammonium salts. In the following
step, to construct the nucleotide, we sought to apply the
phosphoramidate approach, which involves the coupling of a
monophosphate with a nucleotide phosphoromorpholidate in the
presence of a Lewis acid catalyst.27 This method requires the
activation of the appropriate nucleotide monophosphate to its
corresponding phosphoromorpholidate prior to the coupling
reaction. Therefore, both the 8-chloro- and 8-iodo-IMP (10a
and 10b) were activated as their respective phosphoromorpholi-
dates by treatment with morpholine, dipyridyl disulfide, and
triphenylphosphine.28,29 The reactions proceeded to completion
in about an hour as indicated by 31P NMR spectroscopy (δP 6
ppm); the 8-chloro-IMP/8-iodo-IMP morpholidates (11a and
11b) were then precipitated as their sodium salts and were
subsequently reacted with â-NMN and MnCl2 in formamide
without further purification. The reaction mixtures were stopped
after the starting morpholidates had been consumed and then
purified by reverse-phase chromatography to yield the desired
dinucleotides 8-Cl-NHD+ 2 and 8-I-NHD+ 3 as their triethyl-
ammonium salts.
Figure 2. Outline of the synthesis of 8-substituted N1-cIDPR analogues
directly from 8-bromo-N1-cIDPR 1a.
N1 as the preferred site of cyclization and also confers stability
to chemical hydrolysis at neutral pH and elevated temperature.21
This property was exploited to generate a series of four
analogues from the synthetically versatile parent compound
(Figure 2).22 Pharmacological investigation of N1-cIDPR 1b
itself has revealed, interestingly, that it is equipotent to cADPR
in the release of Ca2+ in permeabilized T-cells illustrating the
potential of this novel class of analogue.
Current synthetic efforts are directed toward the preparation
of further structural analogues of N1-cIDPR as novel pharma-
cological tools for the investigation of cADPR pharmacology.
To use the chemoenzymatic strategy on N1-cIDPR derivatives
to its full potential, it is imperative to understand the structural
requirements for N1 cyclization of NHD+-based ADPRC
substrates. Factors influencing the cyclization behavior of
structurally different NHD+ derivatives include the orientation
of the hypoxanthine nucleobase in the ADPRC binding site,
the respective nucleophilicity of N1/N7, steric bulk at the
8-position, and the conformational equilibrium for rotation about
the glycosidic bond. We report here the synthesis of a series of
NHD+ derivatives and describe and analyze their cyclization
behavior. We demonstrate that our chemoenzymatic approach
provides reliable synthetic access to a structurally diverse family
of novel N1-cIDPR analogues. We have characterized these
The triethylammonium salt of 8-methylinosine 5′-monophos-
phate 10c, however, proved to be insoluble in the reaction
solvent DMSO, and efficient conversion to the free acid was
not practical, with desalination leading to partial degradation
of the monophosphate. An alternative method was thus em-
ployed for the synthesis of 8-methyl-NHD+ 4 with a one-pot
procedure involving the direct condensation of 10c and â-NMN
in the presence of carbonyldiimidazole.30,31 Thus, â-NMN was
first treated with an excess of carbonyldiimidazole to provide
an activated phosphoroimidazolium intermediate, which was
confirmed by 31P NMR spectroscopy with the appearance of a
new singlet at δP -9 ppm. After the remaining CDI was
quenched, the phosphoroimidazolium was treated with 8-methyl-
IMP 10c to form 8-methyl-NHD+ 4 after purification by ion-
exchange chromatography on a Q Sepharose column.
Synthesis of 8-Amino NHD+. We previously reported the
synthesis of 8-amino-N1-cIDPR directly from the 8-bromo-N1-
cIDPR 1 via an 8-azido intermediate.22 To further assess the
influence of an electron-donating group in positon 8 on
cyclization, as well as for comparison of the product obtained
from the cyclization to the 8-amino-N1-cIDPR previously
synthesized from an alternative route, 8-amino-NHD+ was
conceived as a valuable target in its own right. The synthesis
of 8-amino-NHD+ 5 (Scheme 2) commenced with the fully
acetylated inosine 12, which was brominated in a saturated Br2/
Na2HPO4 solution (pH 7)32 to give the 8-bromoinosine deriva-
tive 13 in 84% yield. Removal of the acetate protecting group
followed by nucleophilic displacement of the 8-bromo group
with sodium azide produced 8-azidoinosine 14b, which was then
selectively phosphorylated at the 5′-hydroxyl position under our
standard conditions (POCl3, TEP, and water) to yield 8-N3-IMP
15a. The azido compound was then converted into 8-amino-
IMP 15b by treatment with dithiothreitol in 0.05 M TEAB, pH
8.5, overnight. Evidence of the successful reduction of the azido
group to the amino group was most clearly demonstrated by
analogues pharmacologically and report herein on their Ca2+
mobilizing activity in human T-cells.
-
Results and Discussion
Synthesis of the 8-Chloro-, 8-Iodo-, and 8-Methyl-NHD+
Analogues. The syntheses of 8-chloro-, 8-iodo-, and 8-methyl-
NHD+ (2, 3, and 4, respectively) are summarized in Scheme 1.
Commercially available inosine 6 was first fully protected with
TBDMS groups in 95% yield.23 A lithio anion at C-8 was then
generated with LDA and quenched with toluenesulfonyl chlo-
ride24 or molecular iodine to give the 8-chloro 8a or 8-iodo 8b
products in 73% and 86% yields, respectively. A Stille coupling
of the 8-iodo analogue in the presence of Pd(PPh3)4 and
tetramethyltin afforded the methyl derivative 8c in 96% yield.25
Cleavage of the silyl protecting groups with TBAF in THF
provided the 8-substituted inosines 9a-c, which were then
selectively phosphorylated on the 5′-hydroxyl using POCl3,
triethyl phosphate, and water mixture.26 Interestingly, chroma-
tography of 8-Cl-IMP 10a on a standard Q Sepharose ion-
exchange resin led to cleavage of the glycosidic bond, and under
these conditions the free nucleobase 8-Cl hypoxanthine was
isolated as the sole product. This destabilizing effect of the
electron-withdrawing chloro substituent at position 8 on the
glycosidic bond is in marked contrast to the stability previously
observed for the 8-bromo congener 8-Br IMP, which could be
purified without difficulty on Q Sepharose.22 Purification of the
8-substituted inosine monophosphates was achieved using
reverse-phase Gradifrac chromatography, eluted with a gradient
the shift in the UV spectrum from λmax ) 275.8 nm to λmax
)
264.0 nm. Activation of 8-amino-IMP as its phosphoromor-
pholidate 16, followed by coupling with â-NMN and MnCl2 in
formamide delivered the desired 8-amino NHD+ 5 in good yield.
Synthesis of 7-Deaza-NHD+. A key analogue for our
investigations into the cyclase reaction was 7-deaza-NHD+ 17.
The preferred conformation of 17 was expected to be anti, as
in the case of NHD+, while the formal removal of the ring

5164 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Scheme 1. Preparation of 8-X-NHD+ (X ) Cl, I, and Me)a
Moreau et al.
a Reagents and conditions: (i) TBDMSCl, imidazole, DMF, room temp, o/n; (ii) LDA, THF, -78 °C, 2 h, then I2 or TsCl, -78 °C, 1 h; (iii) Pd(PPh3)4,
SnMe4, DMF, 100 °C, o/n; (iv) 1 M TBAF in THF, room temp, o/n; (v) POCl3, TEP, 0 °C, 1 h, then H2O, 0 °C to room temp, 15 min; (vi) morpholine,
dipyridyl disulfide, PPh3, room temp, then 0.1 M NaI in acetone, room temp, 1 h; (vii) carbonyldiimidazole, â-NMN, room temp, 4 days; (viii) â-NMN,
MgSO4, 0.2 M MnCl2 in formamide, room temp, o/n.
Scheme 2. Preparation of 8-Amino-NHD+ a
a Reagents and conditions: (i) Ac2O, pyridine, room temp, 1h; (ii) bromine, 10% aqueous Na2HPO4, dioxane, room temp, 5 days; (iii) methanolic ammonia,
room temp, o/n; (iv) NaN3, DMSO, 70 °C, 2 days; (v) POCl3, TEP, H2O, 0 °C, 1 h; (vi) dithiothreitol, 0.05 M TEAB, room temp, o/n; (vii) morpholine,
dipyridyl disulfide, PPh3, DMSO, room temp, 1 h; (viii) â-NMN, MgSO4, 0.2 M MnCl2 in formamide, room temp, o/n.
nitrogen in position 7 of 17 made the N7 cyclization favored
by this conformation impossible. Therefore, 17 represented the
perfect tool to investigate whether in those cases where the anti
conformation was preferred but N7 cyclization was impossible

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5165
Scheme 3. Synthesis of 7-Deaza-NHD+ a
a Reagents and conditions: (i) K2CO3, NaI, reflux, 5 h; (ii) thiourea, NaOEt, EtOH, reflux, 6 h; (iii) 0.2 M HCl, room temp, o/n; (iv) Raney nickel, reflux,
6 h; (v) POCl3, reflux, 2 h; (vi) 18, KOH, TDA-1, HMPT, MeCN, room temp, 4h; (vii) 1 M NaOMe in MeOH, reflux, 6 h; (viii) 75% aqueous TFA, room
temp, 30 min; (ix) TMSCl, NaI, MeCN, reflux, 4 h; (x) POCl3, TEP, H2O, 1 h, 0 °C, then ice/water, 15 min, room temp; (xi) â-NMN, CDI, TEA, DMF,
4 days, room temp.
the enzyme was capable of realigning the substrate in the correct
syn conformation for N1 cyclization. If this was indeed the case,
we expected 7-deaza-NHD+ to show a cyclization behavior
different from that of its parent dinucleotide NHD+ but
potentially similar to that of 7-deaza-NAD+ (i.e., to cyclize at
N1). To synthesize 7-deaza-NHD+, it was first necessary to
synthesize the nucleobase 18. This was achieved in five steps
from ethyl cyanoacetate and bromoacetaldehyde diethylacetyl
by the method of Davoll and Seela (Scheme 3).33,34
The synthesis of the nucleoside then required the regio- and
stereospecific glycosylation of the nucleobase with an appropri-
ate R-halogenose to deliver the protected nucleoside intermedi-
ate. 1-Chloro-2,3-O-isopropylidene-5-O-(tert-butyl)dimethylsilyl-
R-D-ribofuranose 19 has been reported as a reagent for the
preparation of anomerically pure ribonucleosides without the
problem of neighboring group participation.35,36 Thus, treatment
of 1 equiv of nucleobase 18 with KOH and TDA-1 with 1 equiv
of 1-chlororibofuranose 19 gave the desired â-configurated
ribonucleoside 20 in 65% yield (Scheme 3). The cryptand
TDA-1 was used to increase the nucleophilicity of the nucleo-
base, thus allowing the use of equimolar quantities of reagents,
as reported in the literature.37 Further conversion of the chloro
group into a methoxy group and then deprotection of the
TBDMS and isopropylidene groups by treatment with a 75%
aqueous TFA solution and subsequent cleavage of the methyl
ether afforded the desired 7-deazainosine 23 in a 10% overall
yield over 12 steps. Access to the linear dinucleotide 7-deaza-
NHD+ 17 was then achieved as detailed above with selective
phosphorylation at the 5′-hydroxyl followed by a CDI coupling,
delivering after purification on Q Sepharose the desired di-
nucleotide 17 as its triethylammonium salt.
to this site. In practice, however, all four NHD+ derivatives
underwent smooth cyclization exclusively at N1 (Scheme 4).
This finding suggests that the position of cyclization is
independent of the electronic property of the C-8 substituent.
It may well be that the steric demands of the C-8 substituents
direct cyclization away from the neighboring N7 position.
The structures of the cyclic dinucleotides 1e-h were un-
ambiguously confirmed by mass spectrometry and 2D NMR
spectroscopy. The gHMBC experiments revealed crucial cross-
peaks for the three-bond coupling of the H-2 proton of the
nucleobase with the anomeric carbon of the northern ribose and
for the three-bond coupling of the anomeric proton of the
northern ribose with C-2 of the nucleobase. Both cross-peaks
are indicative of the N1/C-1′′ bond characteristic of N1-cyclized
dinucleotides. The 8-amino cIDPR 1e obtained was in every
aspect identical with the product previously synthesized by direct
transformation of 8-bromo-N1-cIDPR 1a.22
In contrast to the 8-substituted NHD+ analogues, when
7-deaza-NHD+ 17 was incubated with ADPRC, the product
obtained from the enzymatic reaction was the linear dinucleotide
7-deaza-IDPR 24 (Scheme 4), as confirmed by the molecular
ion of 558 obtained in the electrospray mass spectrum. This is
in accord with the mass of a linear dinucleotide rather than the
cyclic 7-deaza-cIDPR because this differs by 17 mass units for
1
the additional hydroxyl group. The H NMR spectrum also
showed the presence of two doublets at 5.28 and 5.16 ppm
indicating the â and R configurations of the anomeric protons
of the terminal ribose unit. This result was intriguing because
it has been reported that the corresponding similarly modified
adenine derivative 7-deaza-NAD+ does cyclize readily into
7-deaza-cADPR.13,38,39
Cyclization Behavior. The four linear NHD+ analogues 2-5
were incubated with ADP-ribosyl cyclase from Aplysia cali-
fornica to assess the influence of the 8-substituent on the site
of cyclization (Scheme 4). We reasoned that an electron-
donating substituent at C-8, e.g., CH3 or NH2, would enhance
nucleophilicity at N7 and might therefore redirect cyclization
Additional control experiments were carried out to investigate
the possibility that 7-deaza-NHD+ 17 may not bind to ADPRC
at all and may simply be hydrolyzed chemically under the
conditions of the cyclization reaction. However, when incubated
with HEPES buffer (pH 7.4) alone, dinucleotide 17 was
completely stable. This result demonstrates that hydrolysis of

5166 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Moreau et al.
Scheme 4a
a Reagents and conditions: Aplysia californica, 25 mM HEPES, pH 7.4, room temp.
Table 1. 1H NMR Chemical Shifts (δ) of 8-X-IMP Analogues in D2O
IMP derivativesa
10a
8.19
6.10 (5.84)
5.18 (5.06)
4.65
4.29(
0.92 (0.78)
syn
10b
8.11
6.02 (5.76)
5.32 (5.08)
4.63
4.27
0.70 (0.64)
syn
10c
8.11
5.95 (5.78)
4.97 (4.90)
4.49
4.14
0.98 (0.88)
syn
15b
8-Br-IMPb
24
7.98
6.17 (6.04)
4.58 (4.53)
4.58
4.29
1.59 (1.51)
anti
IMP
H-2
H-1′
H-2′
H-3′
H-4′
∆
7.90
8.13
8.12
5.93 (5.71)
4.67 (4.59)
4.67
4.25
1.26 (1.12)
anti
6.08 (5.96)
5.24 (5.18)
5.24
4.25
0.84 (0.74)
syn
6.05 (6.06)
4.69 (4.70)
4.69
4.34
1.36 (1.36)
anti
c
1′-2′
confd
a The numbers in parentheses represent the resonance values for the corresponding NHD+ analogues. b Data obtained from a previous publication.22
c Difference in chemical shifts between H-1′ and H-2′. d Favored glycosidic bond conformation.
7-deaza-NHD+ 17 requires the presence of ADPRC. Second,
when 7-deaza-NHD+ was co-incubated with ADPRC and a
cyclizable substrate, such as NAD+, cyclization of NAD+ into
cADPR occurred at a much slower rate (data not shown). Indeed,
preliminary kinetic data for the enzymatic inhibition of ADPRC
by 7-deaza-NHD+ showed that at 50 µM in HEPES buffer (pH
7.4) and 25 °C the reaction rate is roughly 30 times slower than
that of NAD+ and Aplysia cyclase alone under the same
conditions. Taken together, these findings support the interpreta-
tion that 7-deaza-NHD+ is indeed recognized by the enzyme
but unusually is hydrolyzed rather than cyclized. Therefore, 17
could potentially be used as a template to design inhibitors of
ADPRC.
The chemical shift values for the protons common to all IMP
analogues prepared during the course of this study, together with
reference data for 8-Br-IMP and 5′-IMP, are listed in Table 1,
and in agreement with earlier reports, we have found that the
NMR resonances of H-2′ can be used very effectively to assign
the favored glycosodic bond conformation. The H-2′ chemical
shift of δ 4.69 for 5′-IMP is ascribed for nucleotide in the anti
conformation, while nucleotides in the syn conformation display
H-2′ values further downfield. Thus, the 8-chloro, 8-iodo,
6-methyl, and 8-bromo analogues exhibit a clear preference for
the syn conformation as evidenced by their respective H-2′ shifts
of >0.3 ppm downfield of the H-2′ value assigned for 5′-IMP.
However, the 8-amino IMP is assigned as anti because of its
H-2′ resonance value of δ 4.67 being nearly identical to that of
5′-IMP. The anti conformation in this case may well arise from
hydrogen bonding between the C-8 NH and the 5′-oxygen and
has been extensively investigated.45,46 The H-2′ NMR resonance
of 7-deaza-IMP (δ 4.58) suggests that this nucleotide has
significant anti glycosyl bond character, a result that is not
surprising in light of its structural resemblance to 5′-IMP.
Moreover, ∆1′,2′, the difference between H-1′ and H-2′ chemical
shifts, appears to be a measure paralleling the trend observed
with the H-2′ resonance. Thus, a lower value suggests a syn
conformation, whereas a larger one (>1 ppm) suggests anti
conformers predominantly.46
Conformational Analysis. It is well-known that when
unsubstituted at C-8, purine nucleosides and nucleotides favor
an anti glycosyl bond orientation40 (e.g., adenosine, inosine,
guanosine, 5′-AMP, 5′-IMP, 5′-GMP are all predominantly anti)
whereas the syn conformation may be induced by the introduc-
tion of bulky substituents at C-8. Early work on conformational
analysis developed by both Ikehara41 and Sundaralingam,42 using
circular dichroism and NMR spectroscopy techniques, has
demonstrated that indeed purine nucleosides with bulky sub-
stituents at C-8 (such as halogen, SMe, hydroxypropyl, etc.)
exist predominantly in the syn conformation in solution. The
predominance in the syn conformation is believed to be due to
unfavorable steric and electrostatic repulsions between the
8-substituent and the ribose ring. Moreover, Stolarski and
Dudycz noted that changes in the base conformation lead to
characteristic changes in chemical shifts of the ribose protons,
and therefore, the difference in chemical shifts of H-2′ in
nucleosides and their corresponding nucleotides can be used as
an indicator of the sugar-base orientation.43,44
The glycosidic bond conformation of NHD+ analogues has
not previously been investigated systematically. Our data show
a similar trend for the H-2′ resonance values in the NHD+ series
as in the IMP series. Therefore, it seems reasonable to propose
that the conformation of the glycosyl bond is unaffected by
pyrophosphate bond formation and that, at least in the hypox-
anthine series, corresponding mononucleotides and dinucleotides

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5167
prefer a similar glycosyl bond orientation (Table 1). Conse-
quently, 8-chloro-, 8-iodo-, and 8-methyl-NHD+ (2, 3, and 4,
respectively) would be predicted to show a clear preference for
the syn conformation, whereas both 8-NH2 and 7-deaza-NHD+
should be predominantly anti.
Mechanistic Study. From the results obtained after incuba-
tion with Aplysia cyclase we can determine that the cyclase
behaves differently in the presence of adenine or hypoxanthine-
based substrates. When the nucleobase is adenine, it appears
that ADPRC is capable of aligning the substrate in the
appropriate conformation for N1 cyclization, regardless of the
preferred conformation for the individual linear dinucleotide.
NAD+ (predominantly anti configuration) gave cADPR, 8-bromo-
NAD+ (predominantly syn conformation) gave 8-bromo-N1-
cADPR, and 7-deaza-NAD+ (predominantly anti conformation)
gave 7-deaza-cADPR. However, when the nucleobase is hy-
poxanthine, it seems that the enzyme utilizes the substrate in
its prearranged conformation and therefore has a low ability to
reorient the substrate in its active site in order to deliver the
preferred N1 cyclized product. 8-Bromo-NHD+ (predominantly
syn conformation) gave 8-bromo-N1-cIDPR, NHD+ (predomi-
nantly anti conformation) gave N7-cIDPR, while 7-deaza-NHD+
(predominantly anti conformation with no nitrogen at N7) gave
the hydrolyzed product.
8-NH2-NHD+ 5 is the one NHD+ analogue that does not
follow this trend. 5 shows a preference for the anti conformation
and would be expected to cyclize at N7 according to the above
interpretation, but cyclization of 5 exclusively gives 8-amino-
N1-cIDPR 1e. This result is comprehensible on the basis of the
nucleophilicity of N1 and N7 of the 8-amino-NHD+ precursor
5. It is well known that the N1 in adenine is more nucleophilic
than the N1 in hypoxanthine bases, whereas the N7 of
hypoxanthine/guanine has greater nucleophilicity than that of
adenine bases.40 It is therefore expected that, as well as the
substrate conformation, these differences in nucleobase nucleo-
philicity could affect the outcome of the cyclization reaction.
Indeed, the incubation of both NHD+ and NGD+ with ADPRC
generated the N7 cyclic dinucleotide,16 whereas all the NAD+
substrates cyclized at N1 despite their unfavorable anti confor-
mation (as in 7-deaza-NAD+). In the case of 8-NH2-NHD+,
however, the amino substituent should be significantly electron-
donating at the 8-position to make the N1 nucleophilic enough
for cyclization at this position and, in addition, most likely the
bulkiness of the 8-substituent would also inhibit N7 cyclization.
Since space within the N1-cyclized cADPR structure is very
crowded, it is very unlikely that an N7-cyclized 8-substituted
analogue could be synthesized enzymatically. Even a total
synthesis approach is expected to be difficult. Indeed, no such
analogue has been synthesized to date by any method. There is
most likely insufficient space in the enzyme active site to tolerate
a substituent at C-8 in the nucleotide anti conformation, leaving
cyclization in the syn conformation as the only possibility.
Rearrangement of 8-NH2-NHD+ 5 from its native anti confor-
mation to the syn conformation required for N1 cyclization
would also explain why it takes 24 h for 8-NH2-NHD+ to cyclize
under our conditions, whereas 8-chloro-, 8-iodo-, and 8-methyl-
NHD+ (nucleotides that already are in the syn conformation)
cyclize in just 5-8 h under comparable conditions.
Figure 3. Schematic representation of the mechanism for N1 cycliza-
tion.
Crystallographic studies of Aplysia cyclase and BST-1, with
ADPRC activity, have been obtained and suggest that three
residues (Glu 179, Trp 140, and residue 98) play a key role in
catalysis. It has been proposed that the mechanism of cyclization
(and hydrolysis) proceeds via a covalent intermediate47 in which
the carboxylate group of the residue Glu 179 displaces nico-
tinamide to form a bond at C-1′′. This covalently bound
intermediate, directed by a π-π interaction between Trp 140
and the purine base, then aligns in a hairpin-like conformation
to allow a nucleophilic attack on C-1′′ by N1 or N7 of the purine
to release the cyclic dinucleotide from the enzyme.48 Residue
98 (glutamate in Aplysia cyclase and serine in BST-1) acts as
a strong hydrogen bond acceptor to align, through a water
molecule, the nucleobase for attack on C-1′′ of the ribose ring.47
ADPRC enzymes catalyze the transformation of numerous
substrates to produce a diverse range of products. The cycliza-
tion of most adenine-based substrates takes place at N1, while
etheno-NAD+, NGD+ (nicotinamide guanosine 5′-dinucleotide),
or NHD+ cyclize at N7. The structures of three adenine-based
ligands complexed with BST-1 showed that the N1 of ATPγS
superimposes with the N7 of both etheno-NAD+ and etheno-
NADP+ (ethenonicotinamide adenine 5′-dinucleotide 2′-phos-
phate), therefore suggesting that residue 98 is crucial for
orienting the nucleobase in the correct conformation for
cyclization. The direction of cyclization (N1 or N7) may be
attributed to the torsion angle of the glycosidic bond, and thus,
steric restraints on the alignment of the base in the cleft may
account for N1 or N7 selective cyclization or hydrolysis.48 More
recently, the crystal structure of the soluble extracellular domain
of the human antigen CD38, which has ectoenzyme ADPRC
activity, has demonstrated that this cyclase activity is regulated
by Glu 146 (residue conserved in Aplysia cyclase but not in
BST-1). Model studies have revealed the existence of a
hydrogen bond between the N7 atom of the adenine ring and
Glu 146, thus showing that Glu 146 helps to fold the purine
base for subsequent cyclization after release of nicotinamide.49
The parallel reaction courses for the conversion of NAD+
into cADPR and 8-X-NHD+ into 8-X-N1-cIDPR implies that
ADPRC is presumably capable of stabilizing the 8-X-NHD+
derivatives in their enol form (Figure 3), although hypoxanthine
bases predominantly exist in their keto form in aqueous
solution.50 Since 7-deaza-NHD+ and indeed NHD+ itself exist
predominantly in the anti conformation and therefore either do
not cyclize or cyclize at N7, it would be reasonable to suggest
that a substitutent at the 8-position of NHD+ could orient the
nucleobase in the appropriate conformation and might also help
stabilize the correct tautomer for N1 cyclization.
The hydrolysis of 7-deaza-NHD+ 17 into 7-deaza-IDPR 23
can also be rationalized considering the nucleophilicity of the
nitrogen in position 1. According to the 1H NMR data, 17 prefers
the anti conformation (condition unfavorable for cyclization)
and the N1 would not be nucleophilic enough for cyclization
at this position, resulting in the hydrolysis of the substrate.

5168 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Moreau et al.
Figure 4. Effect of the 8-substituted N1-cIDPR analogues on Ca2+ signaling in human T-cells. Jurkat T-cells were permeabilized as described in
the Experimental Section. Ca2+ stores were loaded using ATP and an ATP-regenerating system. (A) cADPR, N1-cIDPR, and its 8-substituted
analogues (100 µM) were added as indicated. (B) Combined data are the mean ( SEM (n g 3; error bars are partially not visible because the values
are smaller than the symbol size). (C, D) Intact Jurkat T-cells were loaded with fura2/AM, and [Ca2+]i was determined fluorimetrically in a cell
suspension as described in the Experimental Section. (C) 8-Phenyl-N1-cIDPR 1c or 8-I-N1-cIDPR 1g was added as indicated (final concentration
of 1 mM). Representative results from three experiments are displayed. (D) Combined data are given as the mean ( SEM (n ) 3-5). Addition of
buffer (vehicle) did not induced Ca2+ signaling in intact cells; thus, no bars are visible.
Since the first step of the cyclization is believed to be the
formation of a ribosylated-ADPR intermediate (Figure 3) that
subsequently cyclizes, we can assume that the hydrolysis of
7-deaza-NHD+ implies that limited initial recognition require-
ments are indeed satisfied, but subsequent syn orientation of
the 7-deazahypoxanthine unit is not possible. The higher
nucleophilicity of N7 in NHD+ and NGD+ has been put forward
to explain N7 cyclization for some substrates;16 however, this
is not a factor in 17. Since 7-deaza-NAD+ does indeed cyclize
into 7-deaza-cADPR, it appears that when there is neither N7
nor 6-NH2, the enzyme cannot stabilize and orient the purine
base for cyclization, and therefore, the substrate gets hydrolyzed.
It seems possible that the N6-amino group of NAD+ could be
crucial in stabilizing an interaction with the enzyme that orients
the adenine in the required conformation for cyclization. This
could be via a hydrogen bond donor or acceptor interaction and
may only be present in the transition state for this transformation.
The hypoxanthine or 7-deazahypoxanthine base has only a
potential hydrogen bond acceptor at C-6 in the dominant keto
form, indicating that perhaps the deciding interaction for adenine
is via hydrogen bond donation from the base. In the case of the
8-halo-NHD+ analogues, the halo group forces the base into
the correct orientation without the stabilizing interaction of the
6-amino group but thus allowing cyclization to occur through

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5169
presumably the minor enol tautomer (Figure 3). Alternatively,
once the required conformation for cyclization is obtained in
8-bromo-NHD+ by the influence of the 8-bromo group, the enol
tautomer at C-6 is stabilized on the enzyme by its mimicry of
the N-H motif at adenine N6.
8-phenyl-N1-cIDPR 1c may be used as novel membrane-
permeant cADPR mimics in intact cells. Such tools are very
important for cell biology, since in many previous studies
cADPR was delivered into cells by either microinjection55 or
infusion via a patch-clamp pipet in the whole-cell configura-
tion.56 The use of broken-cell preparations such as permeabilized
cells11 or cell homogenates57 allows evaluation of the Ca2+
releasing activity of nonmembrane-permeant compounds; how-
ever, only limited conclusions regarding the role of the
respective second messenger system in intact cells can be drawn.
Initial experiments indicate that both 8-chloro-N1-cIDPR 1f
and 8-bromo-N1-cIDPR 1a behave differently from the other
8-X-substituted cIDPR analogues in permeabilized cells. We
have reported 8-bromo-N1-cIDPR 1a to be a membrane-
permeant agonist of cADPR-induced Ca2+ release in intact
cells.21 A more in-depth investigation of these differences is
under way, and the results will be reported in due course.
Regarding the photolabile 8-azido analogue 1d, it is likely
that during the experiment where the cells and the compounds
are exposed to strong UV light, the excitation light for the Ca2+
dye, covalent bonds between the ligand and its receptor are
formed leading to enhanced and prolonged stimulation. A similar
effect was observed with 8-N3-cIDPRE.19
Taken altogether, the chemical syntheses of the 8-substituted
N1-cIDPR analogues reported here result in potentially important
and novel tools to study cADPR-mediated Ca2+ signaling in
permeabilized and intact cells. cIDPR analogues are worthy of
further exploration both chemically and biologically. Indeed,
we recently reported a novel chemical degradation pathway for
8-bromo- and 8-iodo-cIDPR.58 The excellent stability of such
compounds under biological conditions may, in particular, make
them ideal for cocrystallization structure with cADPR binding
protein.
Pharmacology. The pharmacological activity of N1-cIDPR
1b and its 8-substituted analogues was analyzed in permeabilized
and intact human Jurkat T-lymphocytes. In permeabilized cells
rapid Ca2+ release was observed with all analogues reported
here. However, while the Ca2+ mobilizing activity of N1-cIDPR
1b was almost indistinguishable from that of the endogenous
second messenger cADPR,22 8-NH2-N1-cIDPR 1e, 8-CH3-N1-
cIDPR 1h, and 8-phenyl-N1-cIDPR 1c were weaker in their
activity (Figure 4A,B). In contrast, 8-I-N1-cIDPR 1g was
comparable to cADPR, and 8-N3-N1-cIDPR 1d showed even a
somewhat higher, though statistically not significant, efficiency
to release Ca2+ from internal stores (Figure 4A,B). These data
extend our earlier report showing that N1-cIDPR 1b acts via
Ca2+ release from internal stores in human Jurkat T-lympho-
cytes.22
Next, the Ca2+ mobilizing effect of some selected analogues
was analyzed in intact Jurkat T-lymphocytes. Since N1-cIDPR
1b, 8-NH2-N1-cIDPR 1e, and 8-N3-N1-cIDPR 1d were con-
sidered to be too polar to penetrate the plasma membrane, only
the more lipophilic compounds 8-I-N1-cIDPR 1g, 8-phenyl-N1-
cIDPR 1c, and 8-CH3-N1-cIDPR 1h were tested. While 8-CH3-
N1-cIDPR 1h did not much affect [Ca2+]i (data not shown),
there was a small but significant effect of 8-I-N1-cIDPR 1g and
8-phenyl-N1-cIDPR 1c (Figure 4C,D). In contrast to quasi-
physiological stimulation by the anti-CD3 mAb OKT3 (antibody
to the CD3 antigen of human T-cells), there was no biphasic
elevation of [Ca2+]i by 8-I-N1-cIDPR 1g and 8-phenyl-N1-
cIDPR 1c. Instead, both analogues induced a moderate increase
of [Ca2+]i, which remained on a more or less constant level for
at least 10 min (Figure 4D). This discrepancy in shape and
amplitude most likely reflects the complex situation upon
quasiphysiological stimulation versus selective activation of a
single Ca2+ signaling pathway by the cADPR analogues 8-I-
N1-cIDPR 1g and 8-phenyl-N1-cIDPR 1c. Upon quasiphysio-
logical stimulation by the anti-CD3 mAb OKT3, a complex
pattern of formation and metabolism of at least three Ca2+
mobilizing second messengers, NAADP,51 InsP3,52 and
cADPR,53 is observed. This complex pattern in conjunction with
further Ca2+ signaling events, e.g., activation of capacitative
Ca2+ entry, then results in a complex Ca2+ signaling pattern.
On the other hand, the second messenger cADPR, mimicked
here by 8-I-N1-cIDPR 1g and 8-phenyl-N1-cIDPR 1c, is
involved mainly in the sustained phase of Ca2+ signaling upon
T cell receptor/CD3 complex stimulation; thus, the sustained
Ca2+ signal seen upon stimulation by 8-I-N1-cIDPR 1g and
8-phenyl-N1-cIDPR 1c appears indeed to mimic one component
of the physiologically occurring sustained Ca2+ signaling phase.
Experimental Section
General. All reagents and solvents were of commercial quality
and were used without further purification unless described
otherwise. Triethylamine and diisopropylethylamine were dried over
potassium hydroxide, distilled, and then stored over potassium
hydroxide pellets. Morpholine was distilled and dried over KOH,
and formamide was distilled and dried over molecular sieves (4
Å). ADP-ribosyl cyclase was purified from the ovotestis of the
opithobranch mollusk Aplysia californica.59 H2O was of MilliQ
quality. Unless otherwise stated, all reactions were carried out under
an inert atmosphere of nitrogen. All 1H, 13C, and 31P NMR spectra
of final compounds were collected in D2O, either on a JEOL Delta
machine at 270 MHz (1H) or 109 MHz (31P) or on a Varian
Mercury-vx system at 400 MHz (1H) or 100 MHz (13C). Prior to
the recording of 31P NMR spectra a drop of triethylamine was added
to each sample to suppress line broadening and to enhance
resolution. Chemical shifts (δ) are reported in parts per million
(ppm) relative to HDO as internal standard (δ 4.77 ppm, 1H) or to
OP(OPh)3 (δ -18.0 ppm, 31P) as external standard. All 1H and 13
C
NMR assignments are based on gCOSY, gHMBC, gHMQC, and
DEPT experiments. Abbreviations for splitting patterns are as
follows: br, broad; s, singlet; d, doublet; t, triplet; m, multiplet;
etc. UV spectra were collected in aqueous solution on a Perkin-
Elmer Lambda EZ 201 or Lambda 3B spectrophotometer. Infrared
spectra were recorded from KBr disks on a Perkin-Elmer Spectrum
RX1 FT-IR system spectrometer. Peak positions are reported in
cm-1, and the following abbreviations are used for peak intensi-
ties: w, weak; m, medium; s, strong. FAB mass spectra were
recorded on samples in a m-nitrobenzyl alcohol matrix with PEG-
H/PEG-monomethyl ether as reference substances for accurate mass
measurements at the Mass Spectrometry Service Centre, University
of Bath, on a Micromass Autospec instrument. EI mass spectra
were recorded on a Finnigan MAT 95 XP and Micromass Quattro
Ca2+ mobilizing effects of derivatives of N1-cIDPR 1b in
intact cells have been described recently.19,21,22,54 Interestingly,
analogues of N1-cIDPR with substitution of either the northern
ribose by an ether bridge (cIDPRE, 8-N3-cIDPRE, 8-NH2-
cIDPRE) or both riboses by two ether bridges (cIDP-DE)
induced more pronounced initial Ca2+ peaks compared to 8-I-
N1-cIDPR 1g and 8-phenyl-N1-cIDPR 1c when added to intact
Jurkat T-cells.19 It is not yet clear why these different types of
membrane-permeant cADPR analogues resulted in different
initial Ca2+ signals; however, this point will be the subject of
further investigations. Taken together, in addition to the N1-
cIDPR analogues with ether bridges, the 8-I-N1-cIDPR 1g and

5170 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Moreau et al.
II at the EPSRC National Mass Spectrometry Service Centre at
the University of Wales Swansea. HPLC analyses were carried out
on a Waters 2695 Alliance module equipped with a Waters 2996
photodiode array detector (210-350 nm). The chromatographic
system consisted of a Hichrom Guard column for HPLC and a
Phenomenex Synergi 4 µm MAX-RP 80A column (150 mm × 4.60
mm), with elution at 1 mL/min with the following ion-pair buffer:
0.17% (m/v) cetrimide and 45% (v/v) phosphate buffer (pH 6.4)
in MeOH. Preparative chromatography was performed on a
Pharmacia Biotech Gradifrac system equipped with a peristaltic
P-1 pump and a fixed wavelength UV-1 optical unit (280 nm). The
following purification methods were employed: LiChroprep RP-
18 equilibrated with 0.05 M TEAB buffer (pH 6.0-6.4), gradient
0.05 M TEAB buffer against MeCN at 5 mL/min; Q Sepharose
washed with H2O, gradient 1 M TEAB buffer (pH 7.1-7.6) against
H2O at 5 mL/min; AG MP-1 washed with H2O, gradient 150 mM
TFA against H2O at 3 mL/min. Synthetic phosphates were assayed
by an adaptation of the Briggs phosphate test.60
Pharmacology. Cell Culture. Jurkat T-lymphocytes (clone JMP)
were cultured as described earlier.61 In brief, cells were cultured in
RPMI 1640 supplemented with Glutamax I, 25 mM HEPES, 100
units/mL penicillin, 50 µg/mL streptomycin, and 7.5% (v/v)
newborn calf serum.
Loading of Cells with fura2/AM. The cells were loaded with
Fura-2/AM as described in an earlier report.61 Loaded cells were
kept in the dark at room temperature until use.
Hz, H-5′a), 3.78 (dd, 1H, J5′b,5′a ) 11.4 and J5′b,4′ ) 2.5 Hz, H-5′b),
0.94 (s, 9H, tBu), 0.91 (s, 9H, tBu), 0.80 (s, 9H, tBu), 0.12 (s, 6H,
2 × CH3), 0.08 (s, 6H, 2 × CH3), -0.03 (s, 3H, CH3), and -0.19
(s, 3H, CH3). MS (FAB+) m/z 611.3 [(M + H)+, 40%]. HRMS
(FAB+) calcd for C28H55N4O5P2Si3 611.3480 (MH)+; found 611.3480.
Mp 249-250° (lit.23 mp 248-249 °C).
8-Chloro-2′,3′,5′-tris-tert-butyldimethylsilylinosine (8a). To a
solution of diisopropylamine (1.22 mL, 8.71 mmol) in THF (7 mL)
n
at -78 °C was added dropwise BuLi (7.7 mL, 8.88 mmol, 1.15
N). A solution of Tris-TBDMS-inosine 7 (1 g, 1.639 mmol) in THF
(18 mL) was added dropwise at -78 °C, and the mixture was stirred
for further 2 h. Toluenesulfonyl chloride (1.33 g, 6.97 mmol) in
THF (5 mL) was added dropwise, and the reaction mixture was
stirred until completion (3 h) as shown by TLC (CHCl3/EtOAc,
1:1, Rf ) 0.75). The mixture was then warmed to room temperature
and quenched with 1 N NaOAc/AcOH, pH 4. The aqueous phase
was washed with DCM (3 × 50 mL); the combined organic extracts
were washed with 0.5 N NaHCO3, dried (MgSO4), filtered, and
evaporated under reduced pressure giving a yellow oil that was
purified by column chromatography on silica gel (DCM/EtOAc,
9:1) followed by recrystallization with diisopropyl ether to afford
1
the desired product 8a as a white solid (767 mg, 73%). H NMR
(270 MHz, CDCl3) δ 8.20 (s, 1H, H-2), 5.97 (d, 1H, J1′,2′ ) 6.3
Hz, H-1′), 5.27 (dd, 1H, J2′,1′ ) 6.3 and J2′,3′ ) 4.5 Hz, H-2′), 4.45
(dd, 1H, J3′,2′ ) 4.5 and J3′,4′ ) 2.2 Hz, H-3′), 4.11-4.06 (m, 1H,
H-4′), 3.95 (dd, 1H, J5′a,5′b ) 10.9 and J5′a,4′ ) 7.4 Hz, H-5′a), 3.72
(dd, 1H, J5′b,5′a ) 10.9 and J5′b,4′ ) 4.2 Hz, H-5′b), 0.94 (s, 9H,
Determination of [Ca2+]i in Permeabilized Cells. Jurkat T cells
were permeabilized by saponin, and experiments were done as
described in an earlier report.11 Briefly, cells were incubated with
saponin in an intracellular buffer (20 mM HEPES, 110 mM KCl,
2 mM MgCl2, 5 mM KH2PO4, 10 mM NaCl, pH 7.2) in the absence
of extracellular Ca2+. Then saponin was removed by repeated wash
procedures, and the cells were finally resuspended in intracellular
buffer. Then the cells were kept on ice for 2 h to allow for resealing
of intracellular stores. At the start of each individual experiment
the stores of permeabilized cells were reloaded with Ca2+ upon
addition of ATP and an ATP-regenerating system consisting of
creatine phosphate and creatine kinase.61 [Ca2+] was determined
fluorimetrically by fura2/free acid added to the suspension. Changes
in fura2 fluorescence were measured using a Hitachi F-2000
spectrofluorometer (alternating excitation at 340 and 380 nm,
emission 495 nm). When the Ca2+ stores were refilled, test
compounds were added. Usually, the quality of the permeabilized
cell preparation was controlled by its responsiveness to Ins(1,4,5)-
P3 and cADPR on each day. Each experiment was calibrated using
excess CaCl2 and EGTA/Tris to obtain the maximal and minimal
fura2 ratios.
t
t
tBu), 0.85 (s, 9H, Bu), 0.79 (s, 9H, Bu), 0.13 (s, 6H, 2 × CH3),
0.03 (s, 3H, CH3), -0.007 (s, 3H, CH3), -0.05 (s, 3H, CH3), and
-0.33 (s, 3H, CH3). MS (FAB+) m/z 645.2 [(M + H)+, 80%].
HRMS (FAB+) calcd for C28H54N4O5Si335Cl 645.3110 (MH)+,
found 645.3090; calcd for C28H54N4O5Si337Cl 647.3099, found
647.3061 (MH)+. Mp 96-98 °C (lit.24 mp 95-97 °C). Anal.
(C28H53ClN4O5Si3) C, H, N.
8-Chloroinosine (9a). To a solution of 8-chloro-2′,3′,5′-tris-tert-
butyldimethylsilylinosine 8a (720 mg, 1.12 mmol) in THF (4 mL)
was added 1 M TBAF in THF (1 mL, 4.476 mmol). The mixture
was stirred at room temperature until completion of the reaction as
shown by TLC (DCM/MeOH, 8:2, Rf ) 0.66) and was then
quenched with MeOH (2 mL). The solvents were removed under
reduced pressure, and the crude product was purified by column
chromatography on silica gel (DCM/MeOH, 9:1) followed by
recrystallization with aqueous MeOH to afford the desired product
1
9a as a white solid (260 mg, 77%). H NMR (270 MHz, DMSO-
d6) δ 8.16 (s, 1H, H-2), 5.83 (d, 1H, J1′,2′ ) 6.2 Hz, H-1′), 5.52 (br
s, 1H, OH), 5.27 (br s, 1H, OH), 4.99-4.97 (m, 2H, H-2′ and OH),
4.19-4.17 (m, 1H, H-3′), 3.93-3.91 (m, 1H, H-4′), 3.99-3.93 (m,
1H, H-5′a), and 3.72 (dd, 1H, J5′b,5′a ) 10.9 and J5′b,4′ ) 4.2 Hz,
H-5′b). MS (FAB+) m/z 303.0 [(M + H)+, 60%]. HRMS (FAB+)
calcd for C10H12N4O535Cl 303.0500 (MH)+, found 303.0496; calcd
for C10H12N4O537Cl 305.0482, found 305.0466 (MH)+. Anal.
(C10H11ClN4O5‚0.5H2O) C, H, N.
Determination of [Ca2+]i in Intact Cells. [Ca2+]i was measured
in the presence of 1 mM extracellular Ca2+ in Jurkat T-lymphocytes
using fura2/AM as described.61 An F-2000 spectrofluorimeter
(Hitachi) was used in the ratio mode as detailed above for
permeabilized cells. Intracellular [Ca2+] was calculated after
calibration using 0.1% Triton X-100 to obtain the maximal ratio
and EGTA/Tris (8mM/60 mM) to obtain the minimal ratio.
Synthesis of 8-Cl-cIDPR (1f). 2′,3′,5′-Tris-tert-butyldimethyl-
silylinosine (7). To a clear solution of tert-butyldimethylsilyl
chloride (14 g, 93.2 mmol) and imidazole (8.9 g, 130.5 mmol) in
dry DMF (30 mL) was added inosine 6 (5 g, 18.6 mmol) in one
portion. The reaction mixture was stirred at room temperature
overnight, after which TLC analysis (CHCl3/EtOAc, 1:1, Rf ) 0.61)
showed total conversion of starting material in a single product.
The white slurry was poured in a bilayer system of water (100
mL) and DCM (100 mL). The organic layer was separated, and
the aqueous phase was repeatedly extracted with DCM (3 × 50
mL). The combined organic extracts were dried (MgSO4), filtered,
and evaporated under reduced pressure to afford a white solid, which
was recrystallized from isopropyl ether/EtOAc (9:1) giving the
desired product 7 as a white powder (10.7 g, 95%). 1H NMR (270
MHz, CDCl3) δ 8.22 (s, 1H, H-2), 8.06 (s, 1H, H-8), 6.0 (d, 1H,
J1′,2′ ) 4.9 Hz, H-1′), 4.49 (m, 1H, H-2′), 4.29 (m, 1H, H-3′), 4.13-
4.10 (m, 1H, H-4′), 3.98 (dd, 1H, J5′a,5′b ) 11.4 and J5′a,4′ ) 3.7
8-Chloroinosine 5′-Monophosphate (10a). 8-Chloroinosine 9a
(80 mg, 0.264 mmol) was dissolved in triethyl phosphate (1 mL)
by heating with a heat gun. The resulting colorless solution was
cooled to 0 °C, water (2 µL) was added followed by POCl3 (0.1
mL, 1.056 mmol), and the mixture was stirred at 0 °C until HPLC
analysis showed disappearance of starting material and formation
of a single peak (tR(8-Cl-inosine) ) 2.4 min and tR(8-Cl-IMP) ) 7.4
min). After 1 h, the reaction mixture was quenched by addition of
ice/water (15 mL) and stirred for 15 min at 0 °C, after which it
was warmed to room temperature. Triethyl phosphate was extracted
with EtOAc (6 × 6 mL), and the aqueous phase was neutralized
with 2 N NaOH. It was then applied to a reverse-phase Gradifrac
column with elution with a gradient of MeCN in 0.05 M TEAB.
The appropriate fractions were collected and lyophilized overnight
to afford the desired monophosphate 10a as its triethylammonium
1
salt. H NMR (270 MHz, D2O) δ 8.17 (s, 1H, H-2), 6.09 (d, J1′,2′
) 5.6 Hz, 1H, H-1′), 5.20 (app t, 1H, J2′,1′ ) J2′,3′ ) 5.6 Hz, H-2′),
4.61 (dd, 1H, J3′,2′ ) 5.6 and J3′,4′ ) 4.8 Hz, H-3′), 4.27 (app q,
1H, J4′,3′ ) J4′,5′a ) J4′,5′b ) 4.8 Hz, H-4′), and 4.17-4.04 (m, 2H,

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5171
H-5′a and H-5′b). 31P (109 MHz, D2O) δ 1.6 (s). MS (FAB-) m/z
381.1 [(M - H)-, 70%]. HRMS (FAB-) calcd for C10H11N4O8P35Cl
381.0003 [(M - H)-], found 380.9993; calcd for C10H11N6O8P37Cl
382.9973 [(M - H)-], found 382.9966.
1:1, Rf ) 0.63). The mixture was then warmed to room temperature
and quenched with 1 N NaOAc/AcOH, pH 4. The aqueous phase
was washed with DCM (3 × 30 mL); the combined organic extracts
were washed with 0.5 N NaHCO3, dried (MgSO4), filtered, and
evaporated under reduced pressure to give a yellow oil, which was
purified by column chromatography on silica gel (DCM/EtOAc,
9:1) to afford the desired product 8b as a pale-yellow solid (520
Nicotinamide-8-chlorohypoxanthine 5′-Dinucleotide (8-Chloro-
NHD+ (2). 8-Cl-IMP 10a (90 mg, 0.23 mmol) was dissolved in
dry DMSO (2 mL) and coevaporated with dry DMF (5 × 2.5 mL).
The residue was dissolved in dry DMSO (0.4 mL) to which was
added morpholine (104 µL, 1.2 mmol), dipyridyl disulfide (125
mg, 0.55 mmol), and PPh3 (150 mg, 0.55 mmol), at which point
the solution turned bright-yellow. The reaction was completed
within 1 h as shown by 31P NMR (δP 5.9 ppm). Precipitation
occurred by addition of a solution of 0.1 M NaI in acetone. It was
then filtered, washed with acetone to remove most of the yellow
color, dried under vacuum, and used in the next step without further
purification. A small amount was kept for characterization. 1H NMR
(270 MHz, D2O) δ 8.13 (s, 1H, H-2), 6.04 (d, 1H, J1′,2′ ) 4.5 Hz,
H-1′), 5.24-5.21 (m, 1H, H-2′), 4.69-4.65 (m, 1H, H-3′), 4.20-
4.19 (m, 1H, H-4′), 4.03-3.93 (m, 2H, H-5′a and H-5′b), 3.47 (br
s, 4H, 2 × CH2), and 2.82 (m, 4H, 2 × CH2). 31P (109 MHz, D2O)
δ 8.3. Crude 8-Cl-IMP morpholidate 11a (35 mg, 0.073 mmol),
â-NMN (27 mg, 0.080 mmol), and MgSO4 (17 mg, 0.146 mmol)
were dissolved in a 0.2 M solution of MnCl2 in formamide (0.55
mL) and stirred at room temperature overnight, after which HPLC
analysis showed completion of the reaction (tR(â-NMN) ) 2.1 min
and tR(8-Cl-NHD) ) 4.3 min). Precipitation of the product occurred
by addition of MeCN; it was filtered, dissolved in MilliQ, and
applied to a RP-18 Gradifrac column with elution with a gradient
of MeCN in 0.05M TEAB. The appropriate fractions were collected
and evaporated under reduced pressure to afford the desired product
3 as its triethylammonium salt (23 mg, 46%). 1H NMR (400 MHz,
D2O) δ 9.19 (s, 1H, HN2), 9.01 (d, 1H, J6,5 ) 6.0 Hz, HN6), 8.75
(d, 1H, J4,5 ) 7.8 Hz, HN4), 8.12 (dd, J5,4 ) 7.8 and J5,6 ) 6.0 Hz,
1H, HN5), 7.95 (s, 1H, H-2), 5.89 (d, 1H, J1′,2′ ) 4.7 Hz, H-1′′),
5.83 (d, 1H, J1′,2′ ) 5.4 Hz, H-1′), 5.07-5.04 (m, 1H, H-2′), 4.50
(app t, 1H, J3′,2′ ) J3′,4′ ) 5.0 Hz, H-3′), 4.37-4.25 (m, 5H), and
4.20-4.13 (m, 3H). 31P (109 MHz, D2O) δ -10.5 (d, J ) 21 Hz),
-11.1 (d, J ) 21 Hz). UV (H2O, pH 7.2) λmax 250 nm (ꢀ 12 900).
MS (FAB-) m/z 697.9 [(M - H)-, 50%]. HRMS (FAB-) calcd
for C21H24N6O15P235Cl 697.0463 [(M - H)-], found 697.0484.
Cyclic 8-chloroinosine 5′-Diphosphate Ribose (8-Chloro-
cIDPR, 1f). 8-Chloro-NHD+‚2TEA salt 3 (23 mg, 33 µmol) was
incubated with Aplysia cyclase (0.15 mL) in a 25 mM HEPES
buffer (50 mL, pH 4) at room temperature. After 8 h, HPLC analysis
showed completion of the reaction (tR(nicotinamide) ) 1.7 min and
tR(8-Cl-cIDPR) ) 11.3 min). It was then applied to a Q Sepharose
ion exchange column with elution with 1 M TEAB buffer. The
appropriate fractions were collected and evaporated under vacuum
to afford the desired cyclized product as a glaasy solid in its
triethylammonium form (13 mg, 68%). 1H NMR (400 MHz, D2O)
δ 8.65 (s, 1H, H-2), 5.93 (s, 1H, H-1′′), 5.74 (d, 1H, J1′,2′ ) 5.8
Hz, H-1′), 5.28 (app t, 1H, J2′,1′ ) J2′,3′ ) 5.8 Hz, H-2′), 4.57-
4.56 (m, 1H, H-3′), 4.41-4.38 (m, 1H, H-5′a), 4.34-4.33 (m, 1H,
H-5′′a), 4.28-4.26 (m, 3H, H-2′′, H-3′′, and H-4′′), 4.20-4.19 (m,
1H, H-4′), 4.07 (d, 1H, J5′′a,5′′b ) 10.6 Hz, H-5′′a), and 3.97 (d,
1H, J5′b,5′a ) 10.5 Hz, H-5′b). 31P (109 MHz, D2O) δ -9.3 (br s)
-10.2 (br s). UV (H2O, pH 5.2) λmax 253 nm (ꢀ 10 400). MS
(FAB-) m/z 574.8 [(M - H)-, 100%], 576.8 [(M - H)-, 37%].
HRMS (FAB-) calcd for C15H18N4O14P235Cl 574.9983 [(M - H)-],
found 574.9959; calcd for C15H18N4O14P237Cl 576.9954 [(M - H)-],
found 576.9972.
1
mg, 86%). H NMR (270 MHz, CDCl3) δ 8.15 (s,1H, H-2), 5.89
(d, 1H, J1′,2′ ) 6.5 Hz, H-1′), 5.35 (dd, 1H, J2′,1′ ) 6.5 and J2′,3′
)
4.2 Hz, H-2′), 4.44 (dd, 1H, J3′,2′ ) 4.2 and J3′,4′ ) 2.1 Hz, H-3′),
4.04 (dd, 1H, J4′,3′ ) 2.1 and J4′,5′b ) 4.2 Hz, H-4′), 3.96 (dd, 1H,
J5′a,5′b ) 10.6 and J5′a,4′ ) 7.9 Hz, H-5′a), 3.71 (dd, 1H, J5′b,5′a
)
t
10.6 and J5′b,4′ ) 4.2 Hz, H-5′b), 0.95 (s, 9H, Bu), 0.85 (s, 9H,
t
tBu), 0.78 (s, 9H, Bu), 0.13 (s, 6H, 2 × CH3), 0.029 (s, 6H, 2 ×
CH3), -0.07 (s, 3H, CH3), and -0.35 (s, 3H, CH3). MS (FAB+)
m/z 737.3 [(M + H)+, 80%]. HRMS (FAB+) calcd for C28H54N4O5-
Si3I 737.2467 (MH+), found 737.2446. Mp 105-107 °C. Anal.
(C28H53IN4O5Si3) C, H, N.
8-Iodoinosine (9b). To a solution of 8-iodo-2′,3′,5′-tris-tert-
butyldimethylsilylinosine 8b (1 g, 1.35 mmol) in THF (6 mL) was
added 1 M TBAF in THF (1.3 mL, 5.42 mmol). The mixture was
stirred at room temperature until completion of the reaction as
shown by TLC (DCM/MeOH, 8:2, Rf ) 0.31) and quenched with
MeOH (3 mL). The solvent was removed under reduced pressure,
and the product was isolated by column chromatography on silica
gel (DCM/MeOH, 9:1) followed by recrystallization with MeOH/
H2O (1:1) to afford the title nucleoside 9b as a white solid (390
1
mg, 73%). H NMR (270 MHz, DMSO-d6) δ 7.97 (s, 1H, H-2),
5.73 (d, 1H, J1′,2′ ) 6.6 Hz, H-1′), 4.73-4.68 (m, 1H, H-2′), 4.08
(dd, 1H, J3′,2′ ) 5.5 and J3′,4′ ) 3.0 Hz, H-3′), 3.88 (m, 1H, H-4′),
3.61 (dd, 1H, J5′a,5′b ) 12.0 and J5′a,4′ ) 3.7 Hz, H-5′a), and 3.51-
3.45 (m, 1H, H-5′b). MS (FAB+) m/z 395.0 [(M + H)+, 10%].
HRMS (FAB+) calcd for C10H12N4O5I 395.9886 (MH+), found
395.9886. IR (KBr disk, cm-1) 3357, 1684, 1105. Mp 210-212
°C. Anal. (C10H11IN4O5) C, H, N.
8-Iodoinosine 5′-Monophosphate (10b). Compound 10b was
obtained from 8-iodoinosine 9b following the same procedure as
1
for 10a (84 mg, 70%). H NMR (270 MHz, D2O) δ 8.11 (s, 1H,
H-2), 6.02 (d, 1H, J1′,2′ ) 5.6 Hz, H-1′), 5.34-5.30 (m, 1H, H-2′),
4.65-4.61 (m, 1H, H-3′), 4.27 (dd, 1H, J4′,3′ ) 5.2 and J4′,5′a ) 4.6
Hz, H-4′), 4.14 (dd, 1H, J5′a,5′b ) 11.4 and J5′a,4′ ) 4.6 Hz, H-5′a),
and 4.09 (dd, 1H, J5′b,5′a ) 11.4 and J5′b,4′ ) 5.4 Hz, H-5′b). 31P
(109 MHz, D2O) δ 2.23 (s). MS (FAB-) m/z 473.0 [(M - H)-,
90%]. HRMS (FAB-) calcd for C10H11N4O8PI 472.9354 [(M-H)-],
found 472.9349.
Nicotinamide-8-iodohypoxanthine 5′-Dinucleotide (8-Iodo-
NHD+, 3). 8-Iodo-IMP‚2TEA salt 10b (70 mg, 0.103 mmol) was
dissolved in dry DMSO (2 mL) and coevaporated with dry DMF
(5 × 3 mL). The residue was dissolved in DMSO (400 µL) to which
was added morpholine (47 µL, 0.536 mmol), dipyridyl disulfide
(56 mg, 0.257 mmol), and triphenylphosphine (67 mg, 0.257 mmol),
at which point the solution became bright-yellow. It was stirred
for 1 h at room temperature, after which 31P NMR showed
disappearance of the starting material (δP 1.2 ppm) and appearance
of a new singlet (δP 5.3 ppm) for the morpholidate. Precipitation
of the product occurred by dropwise addition of a solution of NaI
in acetone (0.1M, 8 mL). The resulting precipitate was filtered and
washed with acetone to yield the desired morpholidate 11b as a
pale-yellow solid, which was used in the next step without further
1
purification. H NMR (270 MHz, D2O) δ 7.99 (s, 1H, H-2), 5.88
(d, 1H, J1′,2′ ) 3.7 Hz, H-1′), 5.24-5.20 (m, 1H, H-2′), 4.51 (m,
1H, H-3′), 4.12-4.11 (m, 1H, H-4′), 4.0-3.88 (m, 2H, H-5′a and
H-5′b), 3.38 (m, 4H, 2 × CH2) and 2.73 (m, 4H, 2 × CH2). 31P
(109 MHz, D2O) δ 8.3. MS (FAB-) m/z 542.1 [(M - H)-, 100%].
HRMS (FAB-) calcd for C14H18N5O8IP 541.9937 [(M - H)-],
found 541.9964. Crude 8-iodo-IMP morpholidate 11b (60 mg, 0.106
mmol), â-NMN (47 mg, 0.139 mmol), and MgSO4 (30 mg, 0.256
mmol) were dissolved in a 0.2 M solution of MnCl2 in formamide
(1 mL), and the mixture was stirred at room temperature overnight,
after which HPLC analysis showed completion of the reaction
(tR(â-NMN) ) 2.1 min and tR(8-I-NHD) ) 4.1 min). Precipitation of
Synthesis of 8-Iodo-cIDPR (1g). 8-Iodo-2′,3′,5′-tris-tert-
butyldimethylsilylinosine (8b). To a solution of diisopropylamine
(0.57 mL, 4.09 mmol) in THF (4 mL) at -78 °C was added
n
dropwise BuLi (3.63 mL, 4.44 mmol, 1.15 N in hexane). The
mixture was stirred for 10 min at -78 °C, a solution of Tris-
TBDMS-inosine 7 (0.5 g, 0.819 mmol) in THF (9 mL) was added
dropwise at -78 °C, and stirring continued for further 2 h. Iodine
(0.83 g, 3.276 mmol) in THF (3 mL) was added dropwise until a
deep-yellow color settled. It was stirred for 10 min, after which
TLC analysis showed completion of the reaction (CHCl3/EtOAc,

5172 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Moreau et al.
1
the product occurred by addition of MeCN; it was filtered, dissolved
in MilliQ, and applied to a reverse-phase Gradifrac column with
elution with a gradient of 0.05 M TEAB buffer against MeCN.
Further treatment with Chelex-100 to remove any paramagnetic
particles afforded the desired product 3 as its sodium salt (51 mg,
(350 mg, 97%). H NMR (270 MHz, DMSO-d6) δ 8.01 (s, 1H,
H-2), 5.77 (d, 1H, J1′,2′ ) 6.5 Hz, H-1′), 5.40 (d, 1H, JOH,2′ ) 6.6
Hz, 2′-OH), 5.24 (d, 1H, JOH-3′ ) 4.4 Hz, 3′-OH), 5.11 (m, 1H,
5′-OH), 4.74 (dd, 1H, J2′,1′ ) 6.5 and J2′,3′ ) 2.9 Hz, H-2′), 4.12
(m, 1H, H-3′), 3.92 (dd, 1H, J4′,3′ ) 7.2 and J4′,5′a ) 3.9 Hz, H-4′),
3.65 (dd, 1H, J5′a,5′b ) 12.0 and J5′a,4′ ) 3.9 Hz, H-5′a), 3.52 (m,
1H, H-5′b), and 2.50 (s, 3H, CH3 hidden under DMSO peak). MS
(FAB+) m/z 283.0 [(M + H)+, 30%]. HRMS (FAB+) calcd for
1
62% over two steps). H NMR (400 MHz, D2O) δ 9.18 (s, 1H,
HN2), 8.96 (d, 1H, J6,5 ) 6.3 Hz, HN6), 8.77 (d, 1H, J4,5 ) 8.2 Hz,
HN4), 8.07 (dd, 1H, J5,4 ) 8.2 and J5,6 ) 6.3 Hz, HN5), 7.98 (s,
1H, H-2), 5.88 (d, 1H, J1′′,2′′ ) 5.8 Hz, H-1′′), 5.76 (d, 1H, J1′,2′
)
C11H15N4O5 283.1042 (MH)+, found 283.1053. IR (KBr disk, cm-1
)
3495, 1681, 1094. Mp 228-230 °C. Anal. (C11H14N4O5‚1.5H2O)
5.3 Hz, H-1′), 5.08 (app t, 1H, J2′,1′ ) J2′,3′ ) 5.3 Hz, H-2′), 4.48
(dd, 1H, J3′,2′ ) 5.3 and J3′,4′ ) 4.2 Hz, H-3′), 4.26-4.23 (m, 2H,
H-2′′ and H-4′), 4.21-4.19 (m, 1H, H-3′′), and 4.13-4.05 (m, 5H,
H-4′′, H5′a, H5′b, H5′′a, H5′′b). 31P (109 MHz, D2O) δ -10.5 (d,
J ) 19.3 Hz) -11.0 (d, J ) 19.3 Hz). MS (FAB-) m/z 789.9 [(M
- H)-, 10%]. HRMS (FAB-) calcd for C21H25N6O15P2I 789.9897
[(M - H)-], found 789.9895. UV (H2O, pH 5.7) λmax 257 nm (ꢀ
14 700).
C, H, N.
8-Methylinosine 5′-Monophosphate (10c). Compound 10c was
obtained from 8-methylinosine 9c following the same procedure
as for 10a (89 mg, 61%). 1H (400 MHz, D2O) δ 8.11 (s, 1H, H-2),
5.95 (d, 1H, J1′,2′ ) 5.1 Hz, H-1′), 4.97 (app t, 1H, J2′,1′ ) J2′,3′
)
5.1 Hz, H-2′), 4.49 (app t, 1H, J3′,2′ ) J3′,4′ ) 5.1 Hz, H-3′), 4.14-
4.13 (m, 1H, H-4′), 4.03-4.0 (m, 2H, H-5′a and H-5′b), and 2.71
(s, 3H, CH3). 31P (109 MHz, D2O) δ 0.6 (s). MS (FAB-) m/z 360.9
[(M - H)-, 85%]. HRMS (FAB-) calcd for C11H14N4O8P 361.0549
[(M - H)-], found 361.0634.
Cyclic 8-iodoinosine 5′-Diphosphate Ribose (8-Iodo-cIDPR,
1g). 8-Iodo-NHD+ 3 (51 mg, 64 µmol) was incubated with Aplysia
cyclase (160 µL) in a 25 mM HEPES buffer (160 mL, pH 4) at
room temperature, and it was monitored by HPLC. It was stopped
after 5 h (tR(nicotinamide) ) 1.7 min and tR(8-I-cIDPR) ) 12.2 min) and
applied to a Q Sepharose column with elution with a linear gradient
of 1 M TEAB against MilliQ water. The appropriate fractions were
collected, evaporated under vacuum, and coevaporated with MeOH
to afford the desired cyclized product 1g as a glassy solid in the
Nicotinamide 8-methylhypoxanthine 5′-Dinucleotide (8-Meth-
yl NHD+, 4). To a solution of â-NMN (81 mg, 0.242 mmol) in
dry DMF (600 µL) was added carbonyldiimidazole (149 mg, 0.921
mmol) and triethylamine (37 µL, 0.266 mmol). The mixture was
stirred at room temperature for 3 h, after which a small amount of
MeOH was added to quench the excess CDI. The solvents were
removed under vacuum, and the residue was coevaporated three
times with DMF. 8-Me-IMP‚1.5 TEA salt 10c (82 mg, 0.173 mmol)
was added along with DMF (600 µL), and the mixture was stirred
at room temperature for 5 days. The solvent was removed under
reduced pressure, and the residue was applied to Q Sepharose (1
M TEAB) followed by AG MP-1 ion exchange (150 mM TFA) to
produce the desired dinucleotide 4 as a free acid (65 mg, 55%). 1H
NMR (270 MHz, D2O) δ 9.20 (s, 1H, HN2), 9.02 (s, 1H, HN6),
8.73 (d, 1H, J4,5 ) 7.1 Hz, HN4), 8.09 (br s, 1H, HN5), 7.98 (s, 1H,
H-2), 5.95 (d, 1H, J1′′,2′′ ) 4.3 Hz, H-1′′), 5.78 (d, 1H, J1′,2′ ) 5.8
Hz, H-1′), 4.91-4.88 (m, 1H, H-2′), 4.42-4.09 (m, 9H, Hsugar),
and 2.43 (s, 3H, CH3). 31P (109 MHz, D2O) δ -10.9 (br s). MS
(FAB-) m/z 677.1 [(M - H)-, 10%]. HRMS (FAB-) calcd for
C22H27N6O15P2 677.1010 [(M - H)-], found 677.1019. UV (H2O,
pH 5.6) λmax 249 nm (ꢀ 10 600).
Cyclic 8-Methylinosine 5′-Diphosphate Ribose (8-Methyl
cIDPR, 1h). 8-Methyl-NHD+ free acid 4 (90 mg, 96 µmol) was
incubated with Aplysia cyclase (270 µL) in a 25 mM HEPES buffer
(210 mL, pH 4) at room temperature. After 6 h, HPLC analysis
showed completion of the reaction (tR(nicotinamide) ) 1.7 min and
tR(8-Me-cIDPR) ) 7.7 min). It was then applied to a Q Sepharose ion
exchange column with elution with 1 M TEAB buffer. The
appropriate fractions were collected, evaporated under vacuum, and
coevaporated with MeOH to afford the desired cyclized product
1h as a triethylammonium salt (25 mg, 34%). 1H NMR (400 MHz,
D2O) δ 8.73 (s, 1H, H-2), 5.92 (s, 1H, J1′′,2′′ ) 5.7 Hz, H-1′′), 5.83
(br s, 1H, H-1′), 5.20 (app t, 1H, J2′,1′ ) J2′,3′ ) 5.2 Hz, H-2′), 4.52
(dd, 1H, J3′,2′ ) 5.2 and J3′,4′ ) 2.4 Hz, H-3′), 4.47-4.43 (s, 1H,
H-5′a), 4.33-4.30 (m, 1H, H-5′′a), 4.26-4.21 (m, 4H, H-2′′, H-3′′,
H-4′′, and H-4′), 4.07-4.02 (m, 1H, H-5′′b), 3.97-3.93 (m, 1H,
H-5′b), and 2.45 (s, 3H, CH3). 31P (109 MHz, D2O) δ -10.1 (br s)
-10.8 (br s). MS (FAB-) m/z 554.9 [(M - H)-, 60%]. HRMS
(ES+) calcd for C16H23N4O14P2 557.0681 (MH)+ found 557.0683.
UV (H2O, pH 5.7) λmax 251 nm (ꢀ 15 400).
1
triethylammonium salt form (16 mg, 37%). H NMR (400 MHz,
D2O) δ 8.75 (s, 1H, H-2), 5.94 (d, 1H, J1′′,2′′ ) 5.7 Hz, H-1′′), 5.92
(br s, 1H, H-1′), 5.28 (app t, 1H, J2′,1′ ) J2′,3′ ) 5.7 Hz, H-2′),
4.59-4.58 (m, 1H, H-3′), 4.45-4.44 (m, 1H, H-5′a), 4.36-4.33
(m, 1H, H-5′′a), 4.29-4.26 (m, 4H, H-2′′, H-3′′, H-4′′, and H-4′),
4.07 (d, 1H, J5′′a,5′′b ) 11.0 Hz, H-5′′a), and 3.97 (d, 1H, J5′b,5′a
)
11.0 Hz, H-5′b). 31P (109 MHz, D2O) δ -9.3 (br s) -10.2 (br s).
UV (H2O, pH 5.8) λmax 257 nm (ꢀ 10 000). MS (FAB-) m/z 666.8
[(M - H)-, 100%]. HRMS (FAB-) calcd for C15H18N4O14P2I
666.9340 [(M - H)-], found 666.9379.
Synthesis of 8-Methyl-cIDPR (1h). 8-Methyl-2′,3′,5′-tris-tert-
butyldimethylsilylinosine (8c). In a round-bottomed flask equipped
with a condenser was dissolved 8-iodo-2′,3′,5′-tris-tert-butyldi-
methylsilylinosine 8b (50 mg, 0.068 mmol) in dry DMF. Pd(PPh3)4
(16 mg, 0.013 mmol) was added followed by tetramethyltin (0.1
mL, 0.684 mmol). The resulting yellow solution was stirred at 100
°C under argon for 16 h, after which TLC analysis showed
completion of the reaction (CHCl3/EtOAc, 1:1, Rf ) 0.42). The
mixture was then cooled to room temperature and washed with a
saturated aqueous solution of NH4Cl (20 mL) and extracted with
DCM (30 mL). The organic extract was dried (Na2SO4), filtered,
and evaporated under reduced pressure, giving a brown crude oil
which was purified by column chromatography on silica gel (DCM/
EtOAc, 9:1) followed by recrystallization with diisopropyl ether
to yield the desired protected nucleoside 8c as a pale-yellow solid
1
(81 mg, 96%). H NMR (270 MHz, CDCl3) δ 8.0 (s, 1H, H-2),
5.76 (d, 1H, J1′,2′ ) 6.7 Hz, H-1′), 5.16 (dd, 1H, J2′,1′ ) 6.7 and
J2′,3′ ) 4.5 Hz, H-2′), 4.35 (dd, 1H, J3′,2′ ) 4.5 and J3′,4′ ) 2.0 Hz,
H-3′), 4.02-3.99 (m, 1H, H-4′), 3.93 (dd, 1H, J5′a,5′b ) 10.7 and
J5′a,4′ ) 7.2 Hz, H-5′a), 3.69 (dd, 1H, J5′b,5′a ) 10.7 and J5′b,4′ ) 4.0
Hz, H-5′b), 2.56 (s, 1H, CH3), 0.90 (s, 9H, tBu), 0.82 (s, 9H, tBu),
0.71 (s, 9H, tBu), 0.09 (s, 6H, 2 × CH3), -0.005 (s, 6H, 2 × CH3),
-0.03 (s, 3H, CH3), and -0.42 (s, 3H, CH3). MS (FAB+) m/z 625.4
[(M + H)+, 35%]. HRMS (FAB+) calcd for C29H57N4O5Si3
625.3637 (MH)+, found 625.3586. Mp 245-249 °C. Anal.
(C29H56N4O5Si3) C, H, N.
Synthesis of 8-Amino-cIDPR (1e). 2′,3′,5′-Tri-O-acetyl-8-
bromoinosine (13). Compound 13 was synthesized according to a
known procedure,32 giving a shiny white solid (84% yield). 1H NMR
(270 MHz, CDCl3) δ 8.24 (s, 1H, H-2), 6.21 (dd, 1H, J2′,3′ ) 5.9
and J2′,1′ ) 4.9 Hz, H-2′), 6.08 (d, 1H, J1′,2′ ) 4.9 Hz, H-1′), 5.78-
5.74 (m, 1H, H-3′), 4.51-4.46 (m, 1H, H-4′), 4.40-4.27 (m, 2H,
H-5′a and H-5′b), 2.14 (s, 3H, OAc), 2.09 (s, 3H, OAc), and 2.05
(s, 3H, OAc).
8-Methylinosine (9c). To a solution of 8-methyl-2′,3′,5′-tris-
tert-butyldimethylsilylinosine 8c (800 mg, 1.28 mmol) in THF (10
mL) was added 1 M TBAF in THF (4 mL, 5.12 mmol). The mixture
was stirred at room temperature until completion of the reaction as
shown by TLC (DCM/MeOH, 8:2, Rf ) 0.28) and quenched with
MeOH (6 mL). The solvents were removed under reduced pressure,
and the crude product was purified by column chromatography on
silica gel (DCM/MeOH, 9:1) followed by recrystallization with
MeOH/H2O (1:1) to afford the desired product 9c as a white solid
8-Bromoinosine (14a). 2′,3′,5′-Tri-O-acetyl-8-bromoinosine 13
(1.5 g, 3.17 mmol) was dissolved in a methanolic ammonia solution
(saturated at 0 °C, 10 mL), and the mixture was stirred at room

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5173
temperature for 5 h. The resulting precipitate was filtered and
washed with a small portion of methanol, yielding 8-bromoinosine
14a (1 g, 95%). H NMR (270 MHz, DMSO-d6) δ 8.12 (s, 1H,
treatment with Chelex-100 to remove any paramagnetic particles
afforded the desired product 5 as a sodium salt (20.2 mg, 37%
1
1
over two steps). H NMR (400 MHz, D2O) δ 9.19 (s, 1H, HN2),
H-2), 5.81 (d, 1H, J1′,2′ ) 6.4 Hz, H-1′), 5.53 (br s, 1H, OH), 5.32
(br s, 1H, OH), 5.02 (m, 1H, H-2′), 4.20-4.17 (m, 1H, H-3′), 3.92
(dd, 1H, J4′,3′ ) 8.2 and J4′,5′a ) 5.0 Hz, H-4′), 3.65 (dd, 1H, J5′a,5′b
) 11.8 and J5′a,4′ ) 5.0 Hz, H-5′a), 3.50 (dd, 1H, J5′b,5′a ) 11.8 and
J5′b,4′ ) 4.7 Hz, H-5′b), and 3.22 (br s, 1H, OH). MS (ES-) m/z
345 [(M - H)-, 100%]. Mp 196-198 °C (lit.62 mp 196-198 °C).
8-Azidoinosine (14b). To a solution of 8-bromoinosine 14a (100
mg, 0.288 mmol) in dry DMSO (1 mL) was added sodium azide
(60 mg, 0.922 mmol). The mixture was stirred at 70 °C in the dark
for 24 h, after which another portion of sodium azide (60 mg, 0.922
mmol) was added and stirring continued for a further 24 h. The
solvent was removed under reduced pressure, and the residue was
purified by column chromatography on silica gel with elution with
DCM/MeOH (8:2) to yield the desired product 14b as a pale-yellow
9.05 (d, 1H, J ) 6.5 Hz, HN6), 8.71 (d, 1H, J ) 7.9 Hz, HN4),
8.07 (dd, 1H, J ) 7.9 and 6.5 Hz, HN5), 7.81 (s, 1H, H-2), 6.03 (d,
1H, J1′′,2′′ ) 4.5 Hz, H-1′′), 5.71 (d, 1H, J1′,2′ ) 6.6 Hz, H-1′), 4.59
(app t, 1H, J2′,1′ ) J2′,3′ ) 6.6 Hz, H-2′), 4.47 (br s, 1H, H-4′′),
4.39-4.33 (m, 4H, H-2′′, H-3′, H-3′′, and H-5′a), and 4.18-4.15
(m, 4H, H-4′, H-5′b, H-5′′a, and H-5′′b). 31P (109 MHz, D2O) δ
-11.6. MS (ES-) m/z 678.3 [(M - H)-,100%]. HRMS (ES-) calcd
for C21H27N7O15P2 678.0968 [(M - H)-], found 678.0964. UV
(H2O, pH 6.0) λmax 260 nm (ꢀ 12 500).
Cyclic 8-Aminoinosine 5′-Diphosphate Ribose (8-Amino-
cIDPR, 1e). 8-Amino-NHD+ 5 (20 mg, 29 µmol) was incubated
with Aplysia cyclase (60 µL) in a 25 mM HEPES buffer (200 mL,
pH 4) at room temperature. After 24 h, HPLC analysis showed
completion of the reaction (tR(nicotinamide) ) 1.7 min and tR(8-amino-cIDPR)
) 8.8 min). The crude product was then applied to a Q Sepharose
ion exchange column with elution with a linear gradient of 1 M
TEAB buffer against MilliQ water. The appropriate fractions were
collected, evaporated under reduced pressure, coevaporated with
MeOH to remove excess triethylammonium salt, and treated with
Chelex-100 to afford the desired cyclized product 1e as a sodium
1
solid (76 mg, 86%). H NMR (270 MHz, DMSO-d6) δ 8.06 (s,
1H, H-2), 5.64 (d, 1H, J1′,2′ ) 6.2 Hz, H-1′), 5.45 (d, 1H, JOH,2′
)
6.2 Hz, 2′-OH), 5.21 (d, 1H, JOH,3′ ) 5.0 Hz, 3′-OH), 4.90 (t, 1H,
JOH,5′ ) 5.5 Hz, 5′-OH), 4.80 (m, 1H, H-2′), 4.13 (dd, 1H, J3′,4′
)
8.4 and J3′,2′ ) 5.1 Hz, H-3′), 3.89-3.84 (dd, 1H, J4′,3′ ) 8.2 and
J4,5′a ) 4.9 Hz, H-4′), 3.61 (dd, 1H, J5′a,5′b ) 11.9 and J5′a,4′ ) 4.9
Hz, H-5′a), and 3.48 (dd, 1H, J5′b,5′a ) 11.9 and J5′b, 4′ ) 5.2 Hz,
H-5′b). MS (ES-) m/z 308.3 [(M - H)-, 100%]. Mp 210-212 °C
(lit.62 mp >200 °C).
8-Azidoinosine 5′-Monophosphate (15a). Compound 15a was
obtained from 8-azidoinosine 14b following the same procedure
as for 10a (67 mg, 84%). 1H NMR (270 MHz, D2O) δ 8.08 (s, 1H,
H-2), 5.88 (d, 1H, J1′,2′ ) 5.4 Hz, H-1′), 4.97 (dd, 1H, J2′,3′ ) 5.7
and J2′,1′ ) 5.4 Hz, H-2′), 4.48 (dd, 1H, J3′,2′ ) 5.7 and J3′,4′ ) 4.9
Hz, H-3′), 4.15 (app q, 1H, J4′,3′ ) J4′,5′ ) 4.9 Hz, H-4′), and 4.04-
3.88 (m, 2H, H-5′a and H-5′b). 31P (109 MHz, D2O), 1.9 (s). MS
(ES-) m/z 388.1 [(M - H)-, 50%]. HRMS (ES-) calcd for
C10H11N7O8P 388.0412 [(M - H)-], found 388.0411.
1
salt (12.5 µmol, 43%). H NMR (400 MHz, D2O) δ 8.73 (s, 1H,
H-2), 6.02 (br s, 1H, H-1′′), 5.81 (d, J1′,2′ ) 6.1 Hz, 1H, H-1′),
5.36-5.33 (m, 1H, H-2′), 4.63 (dd, J3′,2′ ) 4.8 and J3′,4′ ) 2.5 Hz,
1H, H-3′), 4.52-4.47 (m, 1H, H-5′a), 4.43-4.39 (m, 1H, H-1′′a),
4.30 4.26-4.16 (m, 3H, H-2′′, H-3′′, and H-4′′), 4.29-4.27 (m,
1H, H-4′), 4.15 (dd, J5′′b,5′′a ) 11.7 and J5′′b,4′′ ) 4.3 Hz, 1H, H-5′′b),
and 4.07-4.02 (m, 1H, H-5′b). 31P (109 MHz, D2O) δ -10.3 (br
s) -11.2 (br s). MS (ES-) m/z 556.1 [(M - H)-, 30%]. HRMS
(ES-) calcd for C15H20N4O14P2 556.0487 [(M - H)-], found
556.0492. UV (H2O) λmax 251 nm (ꢀ 15 365).
Synthesis of 7-Deaza-NHD+ (17). 4-Chloropyrrolo[2,3-d]-
pyrimidine (18). Compound 18 was synthesized in five steps by a
procedure developed by Davoll,33 giving a gray solid (28% yield
8-Aminoinosine 5′-Monophosphate (15b). 8-Azido-IMP 15a
(65 mg, 0.158 mmol) was dissolved in 0.05 M TEAB (100 mL,
pH 8.5), and dithiothreitol (243 mg, 1.58 mmol) was added. The
reaction mixture was stirred at room temperature in the dark for
18 h, after which HPLC analysis showed completion of the reaction
1
over five steps). H (270 MHz, DMSO-d6) δ 8.58 (s, 1H, H-2),
7.69 (s, 1H, H-7) and 6.60 (d, 1H, J ) 3.8 Hz, H-8). MS (EI) m/z
153.0 [M+, 15%]. HRMS (ES+) calcd for C6H5N335Cl 154.0167
(MH)+, found 154.0166. Mp 184-186 °C (lit.33 mp 189-190 °C).
5-O-tert-Butyldimethylsilyl-2,3-O-isopropylidene-â-D-ribo-
furanose (19). Compound 19 was synthesized in two steps by
known procedures,63,64 giving a white crystalline solid (71% yield
(tR(8-azido-IMP) ) 9.0 min, λmax ) 275.8 nm and tR(8-amino-IMP)
)
8.4 min, λmax ) 268 nm). The solvent was removed in vacuo, and
the residue was purified on a reverse-phase column with elution
with a gradient of MeCN in 0.05 M TEAB to afford the desired
product 15b as its triethylammonium salt (55 mg, 68%). 1H NMR
(270 MHz, D2O) δ 7.90 (s, 1H, H-2), 5.93 (d, 1H, J1′,2′ ) 7.7 Hz,
H-1′), 4.67 (dd, 1H, J2′,1′ ) 7.7 and J2′,3′ ) 6.0 Hz, H-2′), 4.48 (dd,
1H, J3′,2′ ) 6.0 and J3′,4′ ) 2.5 Hz, H-3′), 4.26-4.24 (m, 1H, H-4′),
and 4.03-4.06 (m, 2H, H-5′a and H-5′b). 31P (109 MHz, D2O) δ
1.5 (s). MS (ES-) m/z 362.0 [(M - H)-, 100%]. HRMS (ES-)
calcd for C10H13N5O8P 362.0507 [(M - H)-], found 362.0507.
Nicotinamide 8-aminohypoxanthine 5′-Dinucleotide (8-Amino
NHD+, 5). 8-Amino-IMP‚1.5TEA salt 15b (42 mg, 0.081 mmol)
was dissolved in dry DMSO (1 mL) and coevaporated with dry
DMF (5 × 3 mL). The residue was dissolved in DMSO (400 µL)
to which was added morpholine (38 µL, 0.440 mmol), dipyridyl
disulfide (46 mg, 0.214 mmol), and triphenylphosphine (56 mg,
0.214 mmol), at which point the solution became bright-yellow.
The mixture was stirred for 1 h at room temperature, after which
HPLC analysis showed completion of the reaction (tR(8-amino-IMP)
) 8.7 min and tR(8-amino-IMP morpholidate) ) 2.8 min). Precipitation of
the product occurred by dropwise addition of a solution of NaI in
acetone (0.1 M, 8 mL). The resulting precipitate 16 was filtered,
washed with acetone, dried, and then reacted with â-NMN (42 mg,
0.125 mmol) and MgSO4 (27 mg, 0.228 mmol) in a 0.2 M solution
of MnCl2 in formamide (0.85 mL) at room temperature overnight,
after which HPLC analysis showed completion of the reaction
(tR(â-NMN) ) 2.1 min and tR(8-amino-NHD) ) 3.3 min). Precipitation
of the product occurred by addition of MeCN; it was filtered,
dissolved in MilliQ water, and applied to a reverse-phase column
with elution with a gradient of MeCN in 0.05 M TEAB. Further
1
over two steps). H (270 MHz, CDCl3) δ 5.36 (d, 1H, J1,2 ) 11.9
Hz, H-1), 4.77 (d, 1H, J2,3 ) 5.9 Hz, H-2), 4.57 (d, 1H, J3,2 ) 5.9
Hz, H-3), 4.44 (br s, 1H, H-4), 3.83 (m, 2H, H-5a and H-5b), 1.56
(s, 3H, CH3), 1.40 (s, 3H, CH3), 1.0 (s, 9H, tBu), and 0.27 (s, 6H,
2 × CH3). MS (FAB+) m/z 287 (M+ + H - OH, 100%). Mp 47-
49 °C (lit.63 mp 47 °C).
4-Chloro-7-(5′-O-tert-butyldimethylsilyl-2′,3′-O-isopropylidene-
â-D-ribofuranosyl)pyrrolo[2,3-d] pyrimidine (20). A solution of
5-O-tert-butyldimethylsilyl-2,3-O-isopropylidene-D-ribofuranose 19
(500 mg, 1.644 mmol) and CCl4 (240 µL, 2.545 mmol) in THF (6
mL) was cooled to -78 °C and treated dropwise with hexa-
methylphosphorus triamide (380 µL, 2.134 mmol) over 15 min.
After being stirred at -78 °C for 2 h, the mixture was slowly
warmed to room temperarure, evaporated to half its volume, and
used as such in the next step (hexane/EtOAc, 8:2, Rf ) 0.8). A
suspension of finely powdered KOH (169 mg, 3.01 mmol), TDA-1
(8 µL, 0.027 mmol), and 4-chloropyrrolo[2,3-d]pyrimidine 18 (209
mg, 1.369 mg) in dry MeCN (8 mL) was stirred for 15 min, after
which the freshly prepared solution of the chloro sugar 19 (530
mg, 1.643 mmol, assuming 100% yield) was added and the mixture
was stirred for 6 h at room temperature under nitrogen. The solvents
were removed under reduced pressure, and the crude mixture was
purified on silica gel with elution with hexane/EtOAc (9:1) to
produce the desired protected nucleoside 20 as a pale-yellow oil
(391 mg, 65%). 1H NMR (270 MHz, CDCl3) δ 8.63 (s, 1H, H-2),
7.54 (d, 1H, J7,8 ) 3.8 Hz, H-7), 6.59 (d, 1H, J8,7 ) 3.8 Hz, H-8),
6.38 (d, 1H, J1′,2′ ) 3.0 Hz, H-1′), 5.04 (dd, 1H, J2′,3′ ) 7.2 and
J2′,1′ ) 3.0 Hz, H-2′), 4.93 (dd, 1H, J3′,2′ ) 7.2 and J3′,4′ ) 3.0 Hz,

5174 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Moreau et al.
H-3′), 4.35-4.31 (m, 1H, H-4′), 3.87 (dd, 1H, J5′a,5′b ) 11.4 and
J5′a′,4′ ) 3.5 Hz, H-5′a), 3.77 (dd, J5′b,5′a ) 11.4 and J5′b,4′ ) 3.8
Hz, 1H, H-5′b), 1.62 (s, 3H, CH3), 1.36 (s, 3H, CH3), 0.85 (s, 9H,
tBu), and 0.03 (s, 6H, 2 × CH3). Rf ) 0.66 (hexane/EtOAc, 8:2).
MS (FAB+) m/z 440.1 [(M + H)+, 65%]. HRMS (FAB+) calcd
for C20H31N3O4Si35Cl 440.1772 (MH)+, found 440.1747; calcd for
C20H31N3O4Si37Cl 442.1742 [(M + H)+], found 442.1732.
6.17 (d, 1H, J1′,2′ ) 6.5 Hz, H-1′), 4.60-4.56 (m, 1H, H-2′), 4.40
(dd, 1H, J3′,2′ ) 5.2 and J3′,4′ ) 3.0 Hz, H-3′), 4.29-4.28 (m, 1H,
H-4′), and 4.09-4.07 (m, 2H, H-5′a and H-5′b). 31P (109 MHz,
D2O) δ 0.73 (s). MS (FAB-) m/z 346.3 [(M - H)-, 10%]. HRMS
(FAB-) calcd for C11H13N3O8P 346.0440 [(M - H)-], found
346.0438.
Nicotinamide 7-deazahypoxanthine 5′-Dinucletide (7-Deaza-
NHD+, 17). To a solution of â-NMN (60 mg, 0.179 mmol) in dry
DMF (500 µL) was added carbonyldiimidazole (110 mg, 0.680
mmol) and triethylamine (27 µL, 0.197 mmol). The mixture was
stirred at room temperature for 3 h, after which a small amount of
MeOH was added to quench the excess CDI. The solvents were
removed under vacuum, and the residue was coevaporated three
times with DMF. 7-Deaza-IMP (50 mg, 0.143 mmol) was added
along with DMF (400 µL), and the mixture was stirred at room
temperature for 4 days. The solvent was removed under reduced
pressure, and the residue was applied to a Q Sepharose (1 M TEAB)
ion-exchange column with elution with a linear gradient of 1 M
TEAB against MilliQ water. The appropriate fractions were
combined and evaporated under vacuum. The residue was coevapo-
rated with MeOH to remove excess salt and lyophilized overnight
to afford the desired dinucleotide 17 as a triethylammonium salt
4-Methoxy-7-(5′-O-tert-butyldimethylsilyl-2′,3′-O-isopropylidene-
â-D-ribofuranosyl)pyrrolo[2,3-d]pyrimidine (21). 4-Chloro-
7-(5′-O-tert-butyldimethylsilyl-2′,3′-O-isopropylidene-â-D-ribo-
furanosyl)pyrrolo[2,3-d]pyrimidine 20 (560 mg, 1.27 mmol) was
added to a freshly prepared solution of 1 N NaOMe (58 mg, 2.55
mmol). The resulting solution was then refluxed for 30 min, after
which TLC analysis showed completion of the reaction (hexane/
EtOAc, 8:2, Rf ) 0.56). The mixture was then neutralized with
Dowex 50WX4-50 ion-exchange and filtered. The filtrate was
concentrated under reduced pressure to leave an oil, which was
purified by flash chromatography with elution with hexane/EtOAc
(9:1) to produce the title compound 21 as a colorless oil (368 mg,
1
67%). H NMR (270 MHz, CDCl3) δ 8.43 (s, 1H, H-2), 7.23 (d,
1H, J8,7 ) 3.6 Hz, H-8), 6.49 (d, 1H, J7,8 ) 3.6 Hz, H-7), 6.33 (d,
1H, J1′,2′ ) 2.9 Hz, H-1′), 5.08 (dd, 1H, J2′,3′ ) 6.4 and J2′,1′ ) 2.9
Hz, H-2′), 4.92 (dd, 1H, J3′,2′ ) 6.4 and J3′,4′ ) 3.2 Hz, H-3′), 4.26
(dd, 1H, J4′,5′a ) 4.1 and J4′,3′ ) 3.2 Hz, H-4′), 4.07 (s, 3H, OMe),
3.82 (dd, 1H, J5′a,5′b ) 11.2 and J5′a,4′ ) 4.1 Hz, H-5′a), 3.73 (dd,
1H, J5′b,5′a ) 11.2 and J5′b,4′ ) 4.1 Hz, H-5′b), 1.59 (3H, s, CH3),
1
(35 mg, 0.055 mmol, 39%). H NMR (400 MHz, D2O) δ 9.24 (s,
1H, HN2), 9.09 (d, 1H, J6,5 ) 6.2 Hz, HN6), 8.72 (dt, 1H, J4,5
)
8.2 and J4,6 ) 1.4 Hz, HN4), 8.10 (dd, 1H, J5,4 ) 8.2 and J5,6 ) 6.2
Hz, HN5), 7.93 (s, 1H, H-2), 7.32 (d, 1H, J8,7 ) 3.8 Hz, H-8), 6.55
(d, 1H, J7,8 ) 3.8 Hz, H-7), 6.06 (d, 1H, J1′,2′ ) 6.5 Hz, H-1′′),
6.04 (d, 1H, J1′′,2′′ ) 5.2 Hz, H-1′), 4.53 (dd, 1H, J2′,1′ ) 6.5 and
J2′,3′ ) 5.4 Hz, H-2′), 4.48-4.46 (m, 1H, H-4′), 4.40 (app t, J2′′, 1′′
) J2′′,3′′ ) 5.2 Hz, 1H, H-2′′), 4.37-4.35 (m, 1H, H-3′), 4.33 (dd,
1H, J3′′,2′′ ) 5.2 and J3′′,4′′ ) 3.8 Hz, H-3′′), 4.27-4.25 (m, 1H,
H-5′a), 4.23 (app q, 1H, J ) 2.7 Hz, H-4′′), 4.14-4.08 (m, 3H,
H-5′′a, H-5′b and H-5′′b). 31P (109 MHz, D2O) δ -10.5 (d, J )
18.8 Hz), and -10.9 (d, J ) 18.8 Hz). MS (ES-) m/z 662.3 [(M -
H)-, 100%]. HRMS (ES-) calcd for C22H26N5O15P2 662.0895 [(M
- H)-], found 662.0897. UV (H2O, pH 5.8) λmax 259 nm (ꢀ 12 200).
t
1.34 (3H, s, CH3), 0.85 (9H, s, Bu), and 0.008 (6H, s, 2 × CH3).
MS (FAB+) m/z 436.5 [(M + H)+, 10%]. HRMS (ES+) calcd for
C21H34N3O5Si 436.2262 (MH)+, found 436.2263.
4-Methoxy-7-â-D-ribofuranosyl)pyrrolo[2,3-d]pyrimidine (22).
4-Methoxy-7-(5′-O-tert-butyldimethylsilyl-2′,3′-O-isopropylidene-
â-D-ribofuranosyl)pyrrolo[2,3-d]pyrimidine 21 (300 mg, 0.686
mmol) was dissolved in a 75% aqueous TFA solution (10 mL) and
stirred at 0 °C for 30 min, after which TLC analysis showed
completion of the reaction (DCM/MeOH, 9:1, Rf ) 0.48). It was
then evaporated to dryness, and the residue was coevaporated three
times with MeOH to remove traces of TFA and purified by column
chromatography with elution with DCM/MeOH (95:5) to give the
deprotected nucleoside 22 as a white solid (190 mg, 98%). 1H NMR
(270 MHz, CDCl3) δ 8.43 (s, 1H, H-2), 7.67 (d, 1H, J8,7 ) 3.6 Hz,
H-8), 6.59 (d, 1H, J7,8 ) 3.6 Hz, H-7), 6.15 (d, 1H, J1′,2′ ) 6.2 Hz,
H-1′), 4.41 (app t, 1H, J2′,3′ ) J2′,1′ ) 6.2 Hz, H-2′), 4.11 (m, 1H,
H-3′), 4.04 (s, 3H, OMe), 3.90 (m, 1H, H-4′), 3.62 (dd, 1H, J5′a,5′b
) 11.9 and J5′a,4′ ) 3.7 Hz, H-5′a), 3.53 (dd, 1H, J5′b,5′a ) 11.9 and
J5′b,4′ ) 3.9 Hz, H-5′b). MS (FAB+) m/z 282.3 [(M + H)+, 80%].
HRMS (ES+) calcd for C12H16N3O5 282.1084 (MH)+, found
282.1084.
4-Hydroxy-7â-D-ribofuranosyl)pyrrolo[2,3-d]pyrimidine (7-
Deazainosine, 23). A solution of 4-methoxy-7â-D-ribofuranosyl)-
pyrrolo[2,3-d] pyrimidine 22 (195 mg, 0.448 mmol) in MeCN (15
mL) was treated with sodium iodide (69 mg, 0.461 mmol) and
trimethylsilyl chloride (59 µL, 0.461 mmol). The reaction mixture
was heated to reflux for 4 h, after which TLC analysis showed the
reaction was completed (DCM/MeOH, 9:1, Rf ) 0.12). The mixture
was then cooled to 0 °C, and the precipitate was filtered and
recrystallized from aqueous MeOH to yield the title compound 23
as a white solid (85 mg, 71%). 1H NMR (270 MHz, DMSO-d6) δ
11.95 (s, 1H, NH), 7.90 (d, 1H, J2,NH ) 3.2 Hz, H-2), 7.36 (d, 1H,
J8,7 ) 3.6 Hz, H-8), 6.51 (d, 1H, J7,8 ) 3.6 Hz, H-7), 5.99 (d, 1H,
J1′,2′ ) 6.1 Hz, H-1′), 5.29 (d, 1H, JOH,2′ ) 6.1 Hz, 2′-OH), 5.10
(d, 1H, JOH,3′ ) 4.6 Hz, 3′-OH), 4.99 (t, 1H, JOH,5′ ) 4.8 Hz, 5′-
OH), 4.30 (app q, 1H, J2′,1′ ) J2′,3′ ) J2′,OH ) 6.1 Hz, H-2′), 4.05
(m, 1H, H-3′), 3.87 (dd, 1H, J4′,5′a ) 7.4 and J4′,3′ ) 3.9 Hz, H-4′),
and 3.63-3.46 (2H, m, H-5a′ and H-5′b). Mp 240-243 °C. MS
(FAB+) m/z 268.0 [(M + H)+, 10%]. HRMS (FAB+) calcd for
C11H14N3O5 268.0933 (MH)+, found 268.0955. Anal. (C11H13N3O5‚
0.5NaI) C, H, N.
Enzymatic Hydrolysis of 17. 7-Deaza-NHD+ sodium salt 17
(35 mg, 0.055 mmol) was incubated with Aplysia cyclase (150 µL)
in a 25 mM HEPES buffer (125 mL, pH 4) at room temperature.
After 4 days at room temperature, HPLC analysis showed comple-
tion of the reaction (tR(product) ) 11.1 min). The mixture was then
applied to a Q Sepharose ion-exchange column with elution with
a lineat gradient of 1 M TEAB buffer against MilliQ water. The
appropriate fractions were collected, evaporated under vacuum, and
coevaporated with MeOH to afford the hydrolyzed product 7-deaza-
IDPR 25 as a glassy solid in the triethylammonium form. 1H NMR
(400 MHz, D2O) δ 8.05 (s, 1H, H-2), 7.51 (d, 1H, J8,7 ) 3.8 Hz,
H-8), 6.76 (d, 1H, J7,8 ) 3.8 Hz, H-7), 6.22 (d, 1H, J1′,2′ ) 6.7 Hz,
H-1′), 5.28 (d, 0.5H, J1′′,2′′ ) 4.0 Hz, Hâ-1′′), 5.16 (d, 0.5H, J1′′,2′′
) 2.2 Hz, HR-1′′), 4.64 (dd, 1H, J2′,1′ ) 6.7 and J2′,3′ ) 5.4 Hz,
H-2′), 4.47-4.45 (m, 1H, H-4′), 4.29-4.21 (m, 2H), 4.12-4.09
(m, 2H), 4.05-3.94 (m, 3H), and 3.86-3.82 (m, 1H). 31P (109
MHz, D2O) δ -10.5 (br s) and -10.7 (br s). MS (ES-) m/z 558.2
[(M - H)-, 100%]. HRMS (ES-) calcd for C16H22N3O15P2
558.0532 [(M - H)-], found 558.0538. UV (H2O, pH 6.1) λmax
260 nm (ꢀ 14 900).
Acknowledgment. We thank the Wellcome Trust for Project
Grant 55709 (to B.V.L.P.), VIP support, Biomedical Research
Collaboration for Grant 068065 (to B.V.L.P. and A.H.G.), and
the Deutsche Forschungsgemeinschaft for Grants GU 360/9-1
and 9-2 (to A.H.G.). We also thank the EPSRC National Mass
Spectrometry service at the University of Wales Swansea.
Supporting Information Available: Elemental analysis data
1
7-Deazainosine 5′-Monophosphate (24). Compound 24 was
obtained from 7-deazainosine 23 following the same procedure as
for compounds 8, 9, and 23; H NMR and HPLC data for targets
compounds 1e-h and 25; and 13C NMR data for all compounds.
This material is available free of charge via the Internet at http://
pubs.acs.org.
1
for 10a (82 mg, 85%). H (270 MHz, D2O) δ 7.98 (s, 1H, H-2),
7.40 (d, 1H, J8,7 ) 3.7 Hz, H-8), 6.68 (d, 1H, J7,8 ) 3.7 Hz, H-7),

Cyclization of NHD+ DeriVatiVes
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17 5175
(23) Hayakawa, H.; Haraguchi, K.; Tanaka, H.; Miyasaka, T. Direct C-8
lithiation of naturally-occurring purine nucleosides. A simple method
for the synthesis of 8-carbon-substituted purine nucleosides. Chem.
Pharm. Bull. 1987, 35, 72-79.
(24) Hayakawa, H.; Tanaka, H.; Haraguchi, K.; Mayumi, M.; Nakajima,
M.; Sakamaki, T.; Miyasaka, T. Preparation of 8-chloropurine
nucleosides through the reaction between their C-8 lithiated species
and para-toluenesulfonyl chloride. Nucleosides Nucleotides 1988, 7,
121-128.
References
(1) Berridge, M. J.; Lipp, P.; Bootman, M. D. The versatility and
universality of calcium signalling. Nat. ReV. Mol. Cell Biol. 2000,
1, 11-21.
(2) Berridge, M. J.; Bootman, M. D.; Roderick, H. L. Calcium signal-
ling: Dynamics, homeostasis and remodelling. Nat. ReV. Mol. Cell
Biol. 2003, 4, 517-529.
(3) Guse, A. H. Cyclic ADP-ribose (cADPR) and nicotinic acid adenine
dinucleotide phosphate (NAADP): novel regulators of Ca2+-signaling
and cell function. Curr. Mol. Med. 2002, 2, 273-282.
(4) Guse, A. H. Biochemistry, biology, and pharmacology of cyclic
adenosine diphosphoribose (cADPR). Curr. Med. Chem. 2004, 11,
847-855.
(25) Mamos, P.; Vanaerschot, A. A.; Weyns, N. J.; Herdewijn, P. A.
Straightforward C-8 alkylation of adenosine analogs with tetraalkyltin
reagents. Tetrahedron Lett. 1992, 33, 2413-2416.
(26) Ikemoto, T.; Haze, A.; Hatano, H.; Kitamoto, Y.; Ishida, M.; Nara,
K. Phosphorylation of nucleosides with phosphorus oxychloride in
trialkyl phosphate. Chem. Pharm. Bull. 1995, 43, 210-215.
(27) 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.
(28) Imai, J.; Torrence, P. F. Expedient chemical synthesis of sequence-
specific 2′,5′-oligonucleotides. J. Org. Chem. 1985, 50, 1418-1426.
(29) Prabahar, K. J.; Cole, T. D.; Ferris, J. P. Effect of phosphate activating
group on oligonucleotide formation on Montmorillonite. The regio-
selective formation of 3′,5′-linked oligoadenylates. J. Am. Chem. Soc.
1994, 116, 10914-10920.
(30) Zhao, Y. X.; Thorson, J. S. A methodological comparison: The
advantage of phosphorimidates in expanding the sugar nucleotide
repertoire. J. Org. Chem. 1998, 63, 7568-7572.
(31) Armstrong, J. I.; Verdugo, D. E.; Bertozzi, C. R. Synthesis of a
bisubstrate analogue targeting estrogen sulfotransferase. J. Org. Chem.
2003, 68, 170-173.
(32) Roelen, H.; Veldman, N.; Spek, A. L.; Kunzel, J. V. D.; Mathot, R.
A. A.; IJzerman, A. P. N-6,C-8-disubstituted adenosine derivatives
as partial agonists for adenosine A(1) receptors? J. Med. Chem. 1996,
39, 1463-1471.
(33) Davoll, J. Pyrrolo[2,3-d]pyirimidines. J. Chem. Soc. 1960, 131-138.
(34) Seela, F.; Richter, R. Aminomethylation of 3,7-dihydropyrrolo[2,3-
d]pyrimidines at C-5. Method for synthesis of aglycone analogs of
nucleoside Q. Chem. Ber. 1978, 111, 2925-2930.
(35) Ramasamy, K.; Imamura, N.; Robins, R. K.; Revankar, G. R. A facile
synthesis of tubercidin and related 7-deazapurine nucleosides via the
stereospecific sodium-salt glycosylation procedure. Tetrahedron Lett.
1987, 28, 5107-5110.
(36) Ramasamy, K.; Imamura, N.; Robins, R. K.; Revankar, G. R. A facile
and improved synthesis of tubercidin and certain related pyrrolo-
[2,3-d]pyrimidine nucleosides by the stereospecific sodium-salt
glycosylation procedure. J. Heterocycl. Chem. 1988, 25, 1893-1898.
(37) Rosemeyer, H.; Seela, F. Stereoselective synthesis of pyrrolo[2,3-
d]pyrimidine R-D-ribonucleosides and â-D-ribonucleosides from
anomerically pure D-ribofuranosyl chlorides. Solid-liquid phase-
transfer glycosylation and N-15-NMR spectra. HelV. Chim. Acta
1988, 71, 1573-1585.
(38) Bailey, V. C.; Sethi, J. K.; Fortt, S. M.; Galione, A.; Potter, B. V. L.
7-Deaza cyclic adenosine 5′-diphosphate ribose: First example of a
Ca2+-mobilizing partial agonist related to cyclic adenosine 5′-
diphosphate ribose. Chem. Biol. 1997, 4, 51-61.
(39) Sethi, J. K.; Empson, R. M.; Bailey, V. C.; Potter, B. V. L.; Galione,
A. 7-Deaza-8-bromo-cyclic ADP-ribose, the first membrane-
permeant, hydrolysis-resistant cyclic ADP-ribose antagonist. J. Biol.
Chem. 1997, 272, 16358-16363.
(40) Saenger, W. Principles of Nucleic Acid Structure; Springer-Verlag:
Berlin, Germany, 1984.
(41) Ikehara, M.; Uesugi, S.; Yoshida, K. Studies on conformation of
purine nucleosides and their 5′-phosphates. Biochemistry 1972, 11,
830-836.
(42) Sarma, R. H.; Lee, C. H.; Evans, F. E.; Yathindra, N.; Sundaralingam,
M. Probing inter-relation between glycosyl torsion, sugar pucker, and
backbone conformation in C(8) substituted adenine-nucleotides by
H-1 and H-1-P-31 fast Fourier-transform nuclear magnetic-resonance
methods and conformational energy calculations. J. Am. Chem. Soc.
1974, 96, 7337-7348.
(43) Stolarski, R.; Dudycz, L.; Shugar, D. NMR-studies on the syn-anti
dynamic equilibrium in purine nucleosides and nucleotides. Eur. J.
Biochem. 1980, 108, 111-121.
(44) Rosemeyer, H.; Toth, G.; Golankiewicz, B.; Kazimierczuk, Z.;
Bourgeois, W.; Kretschmer, U.; Muth, H. P.; Seela, F. Syn anti-
conformational analysis of regular and modified nucleosides by 1D
H-1 NOE difference spectroscopy. A simple graphical-method based
on conformationally rigid molecules. J. Org. Chem. 1990, 55, 5784-
5790.
(5) Guse, A. H. Second messenger function and the structure-activity
relationship of cyclic adenosine diphosphoribose (cADPR). FEBS
J. 2005, 272, 4590-4597.
(6) Lee, H. C. Multiplicity of Ca2+ messengers and Ca2+ stores: A
perspective from cyclic ADP-ribose and NAADP. Curr. Mol. Med.
2004, 4, 227-237.
(7) Zhang, F. J.; Gu, Q. M.; Sih, C. J. Bioorganic chemistry of cyclic
ADP-ribose (cADPR). Bioorg. Med. Chem. 1999, 7, 653-664.
(8) Potter, B. V. L.; Walseth, T. F. Medicinal chemistry and pharmacol-
ogy of cyclic ADP-ribose. Curr. Mol. Med. 2004, 4, 303-311.
(9) Shuto, S.; Matsuda, A. Chemistry of cyclic ADP-ribose and its
analogs. Curr. Med. Chem. 2004, 11, 827-845.
(10) Ashamu, G. A.; Sethi, J. K.; Galione, A.; Potter, B. V. L. Roles for
adenosine ribose hydroxyl groups in cyclic adenosine 5′-diphosphate
ribose-mediated Ca2+-release. Biochemistry 1997, 36, 9509-
9517.
(11) Guse, A. H.; Cakir-Kiefer, C.; Fukuoka, M.; Shuto, S.; Weber, K.;
Bailey, V. C.; Matsuda, A.; Mayr, G. W.; Oppenheimer, N.; Schuber,
F.; Potter, B. V. L. Novel hydrolysis-resistant analogues of cyclic
ADP-ribose: Modification of the “northern” ribose and calcium
release activity. Biochemistry 2002, 41, 6744-6751.
(12) Cakir-Kiefer, C.; Muller-Steffner, H.; Schuber, F. Unifying mecha-
nism for Aplysia ADP-ribosyl cyclase and CD38/NAD(+) glyco-
hydrolases. Biochem. J. 2000, 349, 203-210.
(13) Bailey, V. C.; Sethi, J. K.; Galione, A.; Potter, B. V. L. Synthesis of
7-deaza-8-bromo cyclic adenosine 5′-diphosphate ribose: The first
hydrolysis resistant antagonist at the cADPR receptor. Chem.
Commun. 1997, 695-696.
(14) Walseth, T. F.; Lee, H. C. Synthesis and characterization of
antagonists of cyclic-ADP-ribose-induced Ca2+-release. Biochim.
Biophys. Acta 1993, 1178, 235-242.
(15) Zhang, F. J.; Sih, C. J. Novel enzymatic cyclizations of pyridine
nucleotide analogs: Cyclic-GDP-ribose and cyclic-HDP-ribose.
Tetrahedron Lett. 1995, 36, 9289-9292.
(16) Graeff, R. M.; Walseth, T. F.; Hill, H. K.; Lee, H. C. Fluorescent
analogs of cyclic ADP-ribose: synthesis, spectral characterization,
and use. Biochemistry 1996, 35, 379-386.
(17) Shuto, S.; Shirato, M.; Sumita, Y.; Ueno, Y.; Matsuda, A. Synthesis
of cyclic IDP-carbocyclic-ribose, a stable mimic of cyclic ADP-
ribose. Significant facilitation of the intramolecular condensation
reaction of N-1-(carbocyclic-ribosyl)inosine 5′,6′′-diphosphate deriva-
tives by an 8-bromo-substitution at the hypoxanthine moiety. J. Org.
Chem. 1998, 63, 1986-1994.
(18) Shuto, S.; Fukuoka, M.; Kudoh, T.; Garnham, C.; Galione, A.; Potter,
B. V. L.; Matsuda, A. Convergent synthesis and unexpected Ca2+
-
mobilizing activity of 8-substituted analogues of cyclic ADP-
carbocyclic-ribose, a stable mimic of the Ca2+-mobilizing second
messenger cyclic ADP-ribose. J. Med. Chem. 2003, 46, 4741-
4749.
(19) Gu, X.; Yang, Z.; Zhang, L.; Kunerth, S.; Fliegert, R.; Weber, K.;
Guse, A. H.; Zhang, L. Synthesis and biological evaluation of novel
membrane-permeant cyclic ADP-ribose mimics: N1-[(5′′-O-phos-
phorylethoxy)methyl]-5′-O-phosphorylinosine 5′,5′′-cyclicpyrophos-
phate (cIDPRE) and 8-substituted derivatives. J. Med. Chem. 2004,
47, 5674-5682.
(20) Galeone, A.; Mayol, L.; Oliviero, G.; Piccialli, G.; Varra, M.
Synthesis of a new N-1-pentyl analogue of cyclic inosine diphosphate
ribose (cIDPR) as a stable potential mimic of cyclic ADP ribose
(cADPR). Eur. J. Org. Chem. 2002, 4234-4238.
(21) Wagner, G. K.; Black, S.; Guse, A. H.; Potter, B. V. First enzymatic
synthesis of an N1-cyclised cADPR (cyclic-ADP ribose) analogue
with a hypoxanthine partial structure: discovery of a membrane
permeant cADPR agonist. Chem. Commun. 2003, 1944-1945.
(22) Wagner, G. K.; Guse, A. H.; Potter, B. V. L. Rapid synthetic route
toward structurally modified derivatives of cyclic adenosine 5′-
diphosphate ribose. J. Org. Chem. 2005, 70, 4810-4819.
(45) Evans, F. E.; Kaplan, N. O. 8-Alkylaminoadenyl nucleotides as probes
of dehydrogenase interactions with nucleotide analogs of different
glycosyl conformation. J. Biol. Chem. 1976, 251, 6791-6797.

5176 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 17
Moreau et al.
(46) Jordan, F.; Niv, H. C8-Amino purine nucleosides. Well-defined steric
determinant of glycosyl conformational preferences. Biochim. Bio-
phys. Acta 1977, 476, 265-271.
(47) Love, M. L.; Szebenyi, D. M. E.; Kriksunov, I. A.; Thiel, D. J.;
Munshi, C.; Graeff, R.; Lee, H. C.; Hao, Q. ADP-ribosyl cyclase:
Crystal structures reveal a covalent intermediate. Structure 2004, 12,
477-486.
(48) Yamamoto-Katayama, S.; Ariyoshi, M.; Ishihara, K.; Hirano, T.;
Jingami, H.; Morikawa, K. Crystallographic studies on human BST-
1/CD157 with ADP-ribosyl cyclase and NAD glycohydrolase activi-
ties. J. Mol. Biol. 2002, 316, 711-723.
(49) Liu, Q.; Kriksunov, I. A.; Graeff, R.; Munshi, C.; Lee, H. C.; Hao,
Q. Crystal structure of human CD38 extracellular domain. Structure
2005, 13, 1331-1339.
(50) Chenon, M. T.; Pugmire, R. J.; Grant, D. M.; Panzica, R. P.;
Townsend, L. B. Carbon-13 Magnetic-resonance. 26. Quantitative
determination of tautomeric populations of certain purines. J. Am.
Chem. Soc. 1975, 97, 4636-4642.
(51) Gasser, A.; Glassmeier, G.; Fliegert, R.; Langhorst, M. F.; Meinke,
S.; Hein, D.; Kruger, S.; Weber, K.; Heiner, I.; Oppenheimer, N.;
Schwarz, J. R.; Guse, A. H. Activation of T-cell calcium influx by
the second messenger ADP-ribose. J. Biol. Chem. 2006, 281, 2489-
2496.
(52) Guse, A. H.; Goldwich, A.; Weber, K.; Mayr, G. W. Non-radioactive,
isomer-specific inositol phosphate mass determinations: High-
performance liquid chromatography-micro-metal-dye detection strongly
improves speed and sensitivity of analyses from cells and micro-
enzyme assays. J. Chromatogr., B: Biomed. Sci. Appl. 1995, 672,
189-198.
(53) Guse, A. H.; da Silva, C. P.; Berg, I.; Skapenko, A. L.; Weber, K.;
Heyer, P.; Hohenegger, M.; Ashamu, G. A.; Schulze-Koops, H.;
Potter, B. V. L.; Mayr, G. W. Regulation of calcium signalling in
T-lymphocytes by the second messenger cyclic ADP-ribose. Nature
1999, 398, 70-73.
(55) Guse, A. H.; Berg, I.; daSilva, C. P.; Potter, B. V. L.; Mayr, G. W.
Ca2+-entry induced by cyclic ADP-ribose in intact T-lymphocytes.
J. Biol. Chem. 1997, 272, 8546-8550.
(56) Cancela, J. M.; Van Coppenolle, F.; Galione, A.; Tepikin, A. V.;
Petersen, O. H. Transformation of local Ca2+ spikes to global Ca2+
transients: The combinatorial roles of multiple Ca2+-releasing
messengers. EMBO J. 2002, 21, 909-919.
(57) Clapper, D. L.; Walseth, T. F.; Dargie, P. J.; Lee, H. C. Pyridine-
nucleotide metabolites stimulate calcium release from sea-urchin egg
microsomes desensitized to inositol trisphosphate. J. Biol. Chem.
1987, 262, 9561-9568.
(58) Moreau, C.; Woodman, T. J.; Potter, B. V. L. Unusual entry to the
novel 8-halo-N1-ribosyl hypoxanthine system by degradation of a
cyclic adenosine-5′-diphosphate ribose analogue. Chem. Commun.
2006, 1127-1129.
(59) Hellmich, M. R.; Strumwasser, F. Purification and characterization
of a molluscan egg-specific NADase, a 2nd-messenger enzyme. Cell
Regul. 1991, 2, 193-202.
(60) Briggs, A. P. A modification of the Bell-Doisy phosphate method.
J. Biol. Chem. 1922, 53, 13-16.
(61) Guse, A. H.; Roth, E.; Emmrich, F. Intracellular Ca2+ pools in Jurkat
T-lymphocytes. Biochem. J. 1993, 291 (Part 2), 447-451.
(62) Lin, T. S.; Cheng, J. C.; Ishiguro, K.; Sartorelli, A. C. Purine and
8-substituted purine arabinofuranosyl and ribofuranosyl nucleoside
derivatives as potential inducers of the differentiation of the friend-
erythroleukemia. J. Med. Chem. 1985, 28, 1481-1485.
(63) Moon, H. R.; Kim, H. O.; Lee, K. M.; Chun, M. W.; Kim, J. H.;
Jeong, L. S. Stereoselective synthesis of a novel apio analogue of
neplanocin A as potential S-adenosylhomocysteine hydrolase inhibi-
tor. Org. Lett. 2002, 4, 3501-3503.
(64) Kane, P. D.; Mann, J. The preparation and utility of ethyl 2-(5′-O-
tert-butyldimethylsilyl-2′,3′-O-isopropylidene-â-D-ribofuranosyl)pro-
penoate as a key intermediate for C-nucleoside synthesis. J. Chem.
Soc., Perkin Trans. 1 1984, 657-660.
(54) Guse, A. H.; Gu, X. F.; Zhang, L. R.; Weber, K.; Kramer, E.; Yang,
Z. J.; Jin, H. W.; Li, Q.; Carrier, L.; Zhang, L. H. A minimal structural
analogue of cyclic ADP-ribose. Synthesis and calcium release activity
in mammalian cells. J. Biol. Chem. 2005, 280, 15952-15959.
JM060275A