Trifluoromethylated cyclic-ADP-ribose mimic: Synthesis of 8-trifluoromethyl-N1-[(5″;-O-phosphorylethoxy)methyl]-5′- O-phosphorylinosine-5′,5″-cyclic pyrophosphate (8-CF 3-cIDPRE) and its calcium release activity in T cells

Article abstract of DOI:10.1039/c0ob00090f

A convenient trifluoromethylation method was firstly applied to the synthesis of 8- CF₃-purine nucleosides. On the basis of this method, new protection and deprotection strategies were developed for the successful synthesis of the trifluoromethylated cyclic-ADP-ribose mimic, 8-CF ₃-cIDPRE 1. Using intact, fura-2-loaded Jurkat T cells compound 1 and 2′,3′-O-isopropylidene 8-CF₃-cIDPRE 14 were characterized as membrane-permeant cADPR agonists. Contrary to the 8-substituted cADPR analogues that mainly act as antagonists of cADPR in cells, 8-substituted cIDPRE derivatives were shown to be Ca²⁺ mobilizing agonists. Here we report that even compound 1, the 8-substituted cIDPRE with the strong electron withdrawing CF₃ group, behaves as an agonist in T cells. Interestingly, also the partially protected 2′,3′-O-isopropylidene 8-CF₃-cIDPRE activated Ca²⁺ signaling indicating only a minor role for the hydroxyl groups of the southern ribose of cADPR for its biological activity. To our knowledge 8-CF₃-cIDPRE 1 is the first reported fluoro substituted cADPR mimic and 8-CF₃-cIDPRE 1 and compound 14 are promising molecular probes for elucidating the mode of action of cADPR.

Full text of DOI:10.1039/c0ob00090f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Trifluoromethylated cyclic-ADP-ribose mimic: synthesis of 8-trifluoromethyl-
N1-[(5¢¢-O-phosphorylethoxy)methyl]-5¢-O-phosphorylinosine-5¢,5¢¢-cyclic
pyrophosphate (8-CF3-cIDPRE) and its calcium release activity in T cells†
Min Dong,‡a Tanja Kirchberger,‡b Xiangchen Huang,a Zhen Jun Yang,a Liang Ren Zhang,a Andreas H. Guse*b
and Li He Zhang*a
Received 7th May 2010, Accepted 28th June 2010
DOI: 10.1039/c0ob00090f
A convenient trifluoromethylation method was firstly applied to the synthesis of 8- CF3-purine
nucleosides. On the basis of this method, new protection and deprotection strategies were developed for
the successful synthesis of the trifluoromethylated cyclic-ADP-ribose mimic, 8-CF3-cIDPRE 1. Using
intact, fura-2-loaded Jurkat T cells compound 1 and 2¢,3¢-O-isopropylidene 8-CF3-cIDPRE 14 were
characterized as membrane-permeant cADPR agonists. Contrary to the 8-substituted cADPR
analogues that mainly act as antagonists of cADPR in cells, 8-substituted cIDPRE derivatives were
shown to be Ca2+ mobilizing agonists. Here we report that even compound 1, the 8-substituted cIDPRE
with the strong electron withdrawing CF3 group, behaves as an agonist in T cells. Interestingly, also the
partially protected 2¢,3¢-O-isopropylidene 8-CF3-cIDPRE activated Ca2+ signaling indicating only a
minor role for the hydroxyl groups of the southern ribose of cADPR for its biological activity. To our
knowledge 8-CF3-cIDPRE 1 is the first reported fluoro substituted cADPR mimic and 8-CF3-cIDPRE
1 and compound 14 are promising molecular probes for elucidating the mode of action of cADPR.
was found to be one of the principal Ca2+-releasing second
Introduction
messengers involved in cellular Ca2+ homeostasis.3 The important
Cyclic ADP-ribose (cADPR) (Fig. 1) was discovered as a Ca2+
mobilizing metabolite of b-nicotinamide adenine dinucleotide
(NAD+) by Lee and coworkers in sea urchin egg homogenates.1
The structure of cADPR was determined by X-ray crystallography
analysis.2 Besides D-myo-inositol 1,4,5-trisphosphate (InsP3) and
nicotinic acid adenine dinucleotide phosphate (NAADP), cADPR
role of cADPR and its Ca2+ signaling pathway is illustrated by
its phylogenetically widespread occurrence from unicellular pro-
tozoa, plants, and animals up to humans.4 In addition, cADPR-
mediated Ca2+ signaling has been found in a variety of cellular
processes, such as fertilization, insulin secretion in pancreatic
islets, hormonal signaling in pancreatic acinar cells, chemotaxis
in neutrophils, bone resorption, muscle contraction, long-term
synaptic depression in hippocampus, lymphocyte activation and
proliferation, abscisic acid signaling in plants and sponges, and in
plant circadian clock.5
Several lines of evidence, both pharmacological and molecular,
argue for activation of ryanodine receptors (RyR) by cADPR.
A number of key proteins is involved in specifically shaping
RyR-dependent Ca2+ signaling.6 However, it is unclear whether
cADPR elicits Ca2+ release by direct binding to RyR or via an
associated protein. A significant approach to explore the molecular
mechanism of Ca2+ release is to investigate the structure–activity
relationship of cADPR and its analogues regarding their calcium
mobilizing activities. Agonists or antagonists of cADPR that
resulted from such studies are valuable tools to study cellular
signal transduction and are also useful as lead structures for drug
discovery.7,8
Many derivatives of cADPR have been synthesized and studied
in terms of biological activity, specifically the analogues with sub-
stitutions at the C-8 position of the adenine ring.7,8 8-Substituted
cADPR, such as 8-Br-cADPR, 8-N3-cADPR, and 8-NH2-cADPR
are antagonists of cADPR in sea urchin egg homogenates. 8-
NH2-cADPR is the most potent one and has been used to
assess cADPR-modulated Ca2+ signaling pathways in a variety
of biological systems, such as sea urchin eggs, smooth muscle and
Fig. 1 cADPR and its analogues.
aState Key Laboratory of Natural and Biomimetic Drugs, School of
Pharmaceutical Sciences, Peking University, Beijing 100191, China. E-mail:
zdszlh@bjmu.edu.cn ; Fax: +86-10-82802724; Tel: +86-10-82801700
bThe Calcium Signaling Group, University Medical Center Hamburg-
Eppendorf, Center of Experimental Medicine, Institute of Biochemistry
and Molecular Biology I: Cellular Signal Transduction, 20246 Hamburg,
Germany. E-mail: guse@uke.de; Fax: +49-40-7410-59880; Tel: +49-40-
7410-52828
† Electronic supplementary information (ESI) available: 1H, 19F, 31P-
NMR Spectra of 13, 14 and 1 and HPLC of 14 and 1. See DOI:
10.1039/c0ob00090f
‡ Equal contribution.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4705–4715 | 4705
©

View Article Online
T cells. 8-Br-cADPR was shown to be cell permeant and became
an invaluable tool in studying cADPR function in intact cells,
although in its antagonist properties not as potent as 8-NH2-
cADPR.7,9 Shuto et al. reported cyclic ADP-carbocyclic-ribose
(cADPcR) (Fig. 1) as a stable mimic of cADPR and determined
a stronger Ca2+ release effect than cADPR in sea urchin egg
homogenates, while in Jurkat T-lymphocytes it is less potent as
compared to cADPR.10 8-Substituted cADPcR analogues are
agonists rather than antagonists of cADPR in sea urchin egg
homogenates. Potter and Guse reported N1-cyclized inosine 5¢-
diphosphoribose (N1-cIDPR) (Fig. 1) to be as potent as cADPR
in mediating Ca2+ release in permeabilized human Jurkat T cells.
A series of 8-substituted N1-cIDPR was also synthesized and
characterized as cADPR agonists. In particular, 8-I-N1-cIDPR
and 8-N3-N1-cIDPR showed even higher activities than the natural
cADPR. 8-I-, 8-phenyl- and 8-Br-N1-cIDPR were reported as
membrane permeant agonists of cADPR in intact Jurkat T cells.11
A series of modifications on the northern and southern ribose
moiety of cIDPR or cADPR was investigated in our group.12
N1-Ether cyclic inosine 5¢-diphosphoribose (cIDPRE) (Fig. 2)
behaved as a stable mimic of cADPR. cIDPRE penetrated
the plasma membrane and released Ca2+ from an intracellular
cADPR-sensitive Ca2+ store, and subsequently initiated Ca2+
release-activated Ca2+ entry in intact Jurkat T cells. Several 8-
substituted derivatives of cIDPRE were also synthesized and
characterized as membrane-permeant agonists of cADPR in
Jurkat T cells.12b We found that different functional groups at
the 8-position resulted in different biological activities: 8-N3- and
8-NH2-derivates showed similar effect in their agonist activity as
compared to cIDPRE, while 8-Br- and 8-Cl-cIDPRE had no effect
on the free cytosolic Ca2+ concentration in Jurkat T cells.12b
it would be interesting to analyze if the loss of calcium release
activity of 8-Br- and 8-Cl-cIDPRE in Jurkat T cells is related to
their electronic properties or the potential of hydrogen bonding
formation of 8-substitution. In this report, 8-CF3-cIDPRE was
designed. The strong electron withdrawing property and the ability
to form the hydrogen bonding of fluorine and its novel role
in medicinal chemistry,13 such as to provide metabolic stability
and membrane permeability, 8-CF3-cIDPRE is an interesting
molecule for the investigation of the SAR of cADPR. In addition,
although a number of studies on the synthesis and SAR have been
reported,9–12,14–16 fluorinated cADPR derivatives have not yet been
investigated.
In this paper, the structural position of 8-CF3-cIDPRE is
numbered as follows: a single prime numbering scheme is used
for the position of the N9-ribosyl moiety and a double prime
numbering scheme is used for the position of the N1-substitution
(Fig. 1).
Results and discussion
Chemistry
As trifluoromethyl substituted compounds play an important
role in medicinal chemistry, the trifluoromethylation reaction
has often been studied.17 However, due to the sensitivity of the
glycoside bond under the reaction conditions, only limited work
has been reported on the trifluoromethylation of nucleosides
and nucleotides. Beal18 reported FSO2CF2CO2Me19 as a reagent
introducing the trifluoromethyl group to the 6-position of purine
nucleosides. The first attempt to prepare 8-CF3 purine nucleosides
was described by Kobayashi et al.20 The trifluoromethylation was
carried out in the presence of the active reagent CF3–Cu which was
prepared from Cu powder, but the product of trifluoromethylation
could only be obtained by using very pure CF3–Cu and traces of
Cu powder interfered with this reaction. Recently, Yamakawa21
reported that the CF3I/FeSO4/H2O2/H2SO4 system could be used
to introduce the trifluoromethyl group at the C-8 position of purine
base in moderate yields. However, when inosine was used as the
substrate, the yield was very low and furthermore, the gaseous
reagent CF3I made the reaction difficult to handle.
We tried to develop a convenient method to introduce CF3
to the C-8 position of purine. Firstly, 8-bromo-2¢,3¢,5¢-tri-O-
acetylinosine (2) was used as a model compound and the
CF3CO2Na/CuI22 system was selected as the reagent. The reaction
was performed in NMP at 150 ◦C or in DMF/HMPA at
135 ◦C and with the aid of microwave, respectively. However, the
desired compound was not detected. When the trifluoromethy-
lation of compound (2) was carried out under the condition of
FSO2CF2CO2Me/CuI/HMPA/DMF,18 the trifluoromethylation
product (3) was obtained (data not shown), but the product (3)
decomposed slightly after storage for several hours in the reaction
solution (as monitored by TLC). However, compound (5) was
synthesized smoothly by trifluoromethylation of compound (4)
under the same conditions. Thus, it seems that the ether chain
at N1 position makes the trifluoromethyl compound more stable
(Scheme 1). To the best of our knowledge, this is the first report
using this convenient method to introduce the trifluoromethyl
group at the 8-position of purine nucleoside.
Fig. 2 Structures of cIDPRE and 8-CF3-cIDPRE (1).
To study the structure–activity relationship and further explore
the importance of the 8-substitution in cADPR, we report here the
synthesis of 8-CF3-cIDPRE (1) (Fig. 2) and its pharmacological
characterization with regard to Ca2+ signaling. 8-Substituted
cADPR analogues, with properties of either an electron with-
drawing group or an electron releasing group, such as 8-Br-
cADPR, 8-N3-cADPR, and 8-NH2-cADPR, are antagonists of
cADPR in sea urchin egg homogenate and mammalian cells.9
In contrast, 8-substitutes cIDPR, 8-I-, 8-N3-, 8-phenyl- and 8-
Br-N1-cIDPR were reported as membrane permeant agonists of
cADPR in intact Jurkat T cells.11 However, in the cIDPRE series,
among the 8-substituted cIDPRE analogues only the 8-N3- and
8-NH2-derivatives showed similar effect in their agonist activity as
compared to cIDPRE, while 8-Br- and 8-Cl-cIDPRE had no effect
on the free cytosolic Ca2+ concentration in Jurkat T cells.12b Thus,
4706 | Org. Biomol. Chem., 2010, 8, 4705–4715
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 1 Model reaction for 8-position trifluoromethylation of purine nucleoside.a
With the convenient trifluoromethylation model reaction in
hand, a new sequence of suitable protection and deprotection
strategies was developed for the synthesis of 8-CF3-cIDPRE.
The special property of the 8-trifluoromethyl group at the purine
moiety renders the nucleoside molecule more sensitive to strong
acidic or basic reagents. In general, TBDMS, DMT, MMT were
used for the protection of hydroxyl groups in the synthesis of
cIDPRE and cADPR analogues,12 but in our case they were
sensitive to the condition for trifluoromethylation. (Scheme 2)
2¢,3¢-O-Isopropylidene-8-bromoinosine23 (6) was used as the
starting material. At first, tert-butyldimethylsilyl chloride (TB-
DMSCl), DMTCl, and MMTCl were used for the protec-
tion of 5¢-OH. (Scheme 2) After the N1 ether chain was
introduced successively, substrate (8) was trifluoromethylated
by the FSO2CF2CO2Me/CuI system. However, the desired 8-
trifluoromethyl substituted product was not obtained, but instead
the 5¢-deprotected compound 912a appeared (Scheme 2). The
protecting group TBDPS is 100–250 times more stable than
TBDMS in acid condition and could be deprotected selectively
by 70% HF·Py.24 After optimization of the conditions, we found
that the desired trifluoromethylated products were obtained only
using Ac, Bz or TBDPS as the protecting groups. On the other
hand, deprotection in strong basic condition also must be avoided
because of the lability of the CF3 group in alkali.25 Thus the
synthesis of 8-CF3-cIDPRE was designed as depicted in Scheme 3.
Compound 6 reacted with TBDPSCl to give compound 7.
Then the N1 substitution was carried out regioselectively with
2-chloromethoxyethyl acetate in the presence of excess DBU to
afford 8 in 60% yield. The structure of the N1-substituted inosine
of the trifluoromethyl on 8-position of inosine renders the glycosyl
bond sensitive to acid conditions. Compound 11 was phosphory-
lated with cyclohexylammonium S,S-diphenylphosphorodithioate
(PSS) in the presence of triisopropylbenzenesulfonyl chloride
(TPSCl) and tetrazole in pyridine to give 12 in 86% yield. As the
S,S-diphenylphosphorodithioate and the trifluoromethyl group
are sensitive to basic conditions, the condition of deacetylation
in 12 was optimized. When 0.6 equivalents of AcCl-MeOH26
was used for the reaction for 24 h at room temperature, 12 was
successfully deacetylated in good yield. The free hydroxyl of 12
was phosphorylated by using POCl3/DIEA in CH3CN at 0 ◦C for
12 h, and the partial deprotection of S,S-diphenylphosphate was
completed successfully in one step by treatment in 1 M TEAB for
6 h.27 After purification using HPLC, the desired compound 13
was obtained as its triethylammonium form.
The intramolecular cyclization of compound 13 was performed
10,12
˚
in the presence of excess I2 and 3 A molecular sieve in pyridine.
The cyclic product 8-CF3-2¢,3¢-O-isopropylidene cIDPRE 14 was
purified by HPLC as its triethylammonium salt in 51% yield.
1
The structure of 14 was confirmed by HR-ESI-MS, H-NMR,
19F NMR and 31P NMR. Finally, the removal of isopropylidene
group of 14 was carried out with 15% HCOOH in water at
room temperature for 48 h to obtain the target compound 8-
CF3-cIDPRE 1. Compound 1 was purified by HPLC as its
triethylammonium salt in 64.5% yield, characterized by HR-ESI-
MS, 1H NMR, 19F NMR and 31P NMR.
Pharmacology
1
8 was confirmed by H NMR, 13C NMR and HMBC. 8 reacted
When added to Fura2-loaded intact Jurkat T cells, compound
1 showed a biphasic type of Ca2+ release activity (Fig. 3B). To
confer solubility, a small concentration of DMSO had to be
used in these experiments. Thus, the corresponding control was
carried out using the same carrier solution (Fig. 3A), but with
negligible Ca2+ signaling activity. Jurkat T cells have been used as
a model for T cell signaling since many years; the biphasic nature
of the Ca2+ response to TCR/CD3 stimulation is well established
with FSO2CF2CO2Me/CuI in HMPA and the 8-CF3-substituted
compound 10 was achieved successfully in 59% yield. The structure
of 10 was confirmed by 1H, 13C, 19F-NMR and HRMS.
70% HF·Py24 was used to remove the 5¢-TBDPS group in com-
pound 10 at 0 ◦C in high yield (90%). This reaction should be han-
dled carefully since when HF·Py was added at room temperature
the de-purine product was detected. The strong electrophilic action
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4705–4715 | 4707
©

View Article Online
Scheme 2 Selection of the proper protection group of 5¢-OH for trifluoromethylation.a
Scheme 3 Synthesis of 8-CF3-cIDPRE.a
4708 | Org. Biomol. Chem., 2010, 8, 4705–4715
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 3 Ca2+ mobilization by compounds 1 (8-CF3-cIDPRE) and 14 in intact Jurkat T cells. Jurkat T-cells were loaded with Fura-2AM and subjected
to Ca2+ imaging. A–C Characteristic tracings from a representative experiment are shown. The time points of addition of (A) buffer (containing 1.3%
DMSO), (B) 2 mM 8-CF3-cIDPRE (containing 1.3% DMSO) and (C) 2 mM compound 14 (containing 1.6% DMSO) are indicated by arrows. All
concentrations are final concentrations. The black tracings indicate the mean from individual cells (grey tracings) from a single experiment; n = cell
number analyzed in this experiment. D, E Combined data representing mean S.E.M. (n = 22–77 cells) of single tracings. For control, buffer containing
different amounts of DMSO was added instead of the stimulus as described in the experimental section. (D) Ca2+ peak is expressed as the difference
between the maximal peak amplitude and the basal Ca2+ concentration (D values). (E) Ca2+ plateau is expressed as the difference between the Ca2+ plateau
(mean at 770–820 s) and the basal Ca2+ concentration (D values). Significance was tested against corresponding buffer controls. ***, P <0.001; *, P <0.05
(t test).
when cells are analyzed in suspension. However, it is similarly
well established that individual cells show Ca2+ signaling kinetics
differing very much from each other (Fig. 3B,C).28,29 Interestingly,
subcloning of individual Jurkat T cells does not result in cells
responding in identical manner, but the same variety of signaling
pattern occurs again. The latter indicates that the Ca2+ signaling
tools used by Jurkat T cells allow the generation of different
Ca2+ signaling pattern and that minimal differences in signal
transduction may result in very different signaling pattern over
time.
When compared to the quasi-physiological agonist, the mono-
clonal anti-CD3 antibody OKT3, especially the first phase of Ca2+
signaling was considerably smaller with compound 1 (Fig. 3D,E).
However, the additional Ca2+ releasing second messengers formed
upon OKT3 stimulation, D-myo-inositol 1,4,5-trisphosphate30 and
nicotinic acid adenine dinucleotide phosphate31 likely account for
the increased Ca2+ response upon OKT3 in the first minutes.
Importantly, the sustained phase of Ca2+ signaling upon both
OKT3 or compound 1 appears quite similar indicating ongoing
Ca2+ release via RyR followed by capacitative Ca2+ entry (Fig. 3B).
Interestingly, the partially protected compound 14 also showed
Ca2+ release activity in intact Jurkat T cells (Fig. 3C). Saturation of
Ca2+ signaling was achieved at approx. 1 mM of both compounds
1 and 14 for both the initial Ca2+ peak and the sustained Ca2+
signaling phase (Fig. 3D,E).
For comparison of the magnitude of Ca2+ signaling induced
by compound 1, the parent compound cIDPRE and its partially
protected precursor 2¢,3¢-O-isopropylidene cIDPRE12b were an-
alyzed in the same series of experiments (Fig. 4). Although the
Ca2+ signaling amplitudes were somewhat lower as compared
to compounds 1 and 14 (Fig. 4E,F vs. Fig. 3D,E), similar
biphasic signaling pattern were observed (Fig. 4B,D) while the
corresponding controls with vehicle showed only negligible signals
(Fig. 4A,C).
The 8-substituted cIDPRE analogues reported earlier,12b 8-
NH2- and 8-N3-cIDPRE showed agonist activity in intact human
Jurkat T-lymphocytes, but 8-Cl- and 8-Br-cIDPRE were inactive.
Thus it seems that the electronegative action of halogen atoms may
either directly prevent binding at the binding pocket of the related
cADPR receptor or may influence the overall conformation of
the whole molecule in a way that it does not resemble cADPR
any more. However, the newly synthesized 8-CF3-cIDPRE 1
retained the agonist effect. This indicates that a strong electron
withdrawing substitution at the 8-position does not necessarily
interfere negatively with the interaction at the cADPR receptor.
One may speculate that this phenomenon may also be attributed
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4705–4715 | 4709
©

View Article Online
12b
Fig. 4 Ca2+ mobilization by cIDPRE and 2¢,3¢-O-isopropylidene cIDPRE
in intact Jurkat T cells. Jurkat T-cells were loaded with Fura-2AM and
subjected to Ca2+ imaging. A–D Characteristic tracings from a representative experiment are shown. The time points of addition of (A) buffer, (B)
2 mM cIDPRE, (C) buffer (containing 1.6% DMSO) and (D) 2 mM 2¢,3¢-O-isopropylidene cIDPRE (2¢,3¢-O-ip-cIDPRE; containing 1.6% DMSO) are
indicated by arrows. All concentrations are final concentrations. The black tracings indicate the mean from individual cells (grey tracings) from a single
experiment. n = cell number analyzed in this experiment. E,F Combined data representing mean S.E.M. (n = 30–77 cells) of single tracings. For control,
buffer including different amounts of DMSO was added instead of the stimulus as described in the experimental section. (E) Ca2+ peak is expressed as
the difference between the maximal peak amplitude and the basal Ca2+ concentration (D values). (F) Ca2+ plateau is expressed as the difference between
the Ca2+ plateau (mean at 770 to 820 s) and the basal Ca2+ concentration (D values). Significance was tested against corresponding buffer control. ***,
P <0.001; **, P <0.01; *, P <0.05 (t test).
to potential hydrogen bond formation of trifluoromethyl with
the cADPR binding pocket. Theoretical and crystallographic
evidences support the idea that fluorine is a poor hydrogen bond
acceptor with only a moderate capacity to replace oxygen (or
nitrogen) in this role. The C(sp3)–F ◊ ◊ ◊ H bond strength is half of
that of the O ◊ ◊ ◊ H value.32 However, F ◊ ◊ ◊ H hydrogen bonding has
been shown to widely exist33 and relatively short intermolecular
˚
O–H ◊ ◊ ◊ F–C distances of 2.01 A were observed with the fluorine of
CF3 group.34 There are also examples illustrating that the affinity
of substrate and receptor was strengthened by a F ◊ ◊ ◊ H hydrogen
bond.35
Fig. 5 cADPR and Protonated cADPR.
of antagonist activity and the molecular permeability as well.
cIDPRE derivatives do not form the positively charged structure at
N1 and the ether linkage instead of northern ribose in N1-cIDPRE
renders this molecule more stable against enzymatic hydrolysis.36
Potter and colleagues systematically analyzed the modification
of the 2¢- and 3¢-hydroxyl groups of the southern ribose of cADPR.
They found that 2¢-deoxy-cADPR is almost as potent as cADPR
in mediating Ca2+ release from sea urchin egg homogenates, indi-
cating that 2¢-OH has little effect on the agonist activity of cADPR.
While the deletion of 2¢-OH decreases the antagonist activity of 8-
substituted cADPR, even some 8-substituted 2¢-deoxy analogues
showed agonist activity.14 3¢-deoxy-cADPR is much less potent
than cADPR, indicating that the 3¢-hydroxyl group is essential for
8-Substituted cADPR analogues, such as 8-Br-cADPR, 8-N3-
cADPR, and 8-NH2-cADPR, are antagonists of cADPR in sea
urchin egg homogenate and mammalian cells.9 In contrast, 8-
substituted cIDPRE analogues are agonists in mammalian cells.
Here we report a new member of this agonist family, 8-CF3
substituted cIDPRE. Structurally, N1 ribose substituted adenine
moiety in cADPR shows a positive charge at N1 in the presence of
protons (Fig. 5).
Obviously, the 8-substituted group in cADPR will influence
the electron density of the whole molecule, especially, the N1
positive charge which may play a key role in the mechanism
4710 | Org. Biomol. Chem., 2010, 8, 4705–4715
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Ca2+ releasing activity of cADPR in sea urchin eggs. In contrary,
3¢-O-methyl-cADPR is an antagonist of cADPR-induced Ca2+
release in sea urchin eggs. Two other analogues, 3¢-phospho-
cADPR and 2¢,3¢-cyclic-cADPR phosphate, were inactive with
respect to both agonist and antagonist activities on the cADPR-
sensitive Ca2+ release mechanism in sea urchin egg homogenates.15
In our case, 2¢,3¢-O-isopropylidene protection of 8-CF3-cIDPRE
14 did not much alter the agonist activity of compound 1;
a similar result was obtained for cIDPRE and its protected
precursor 2¢,3¢-O-isopropylidene cIDPRE. This indicates on the
one hand that such lipophilic masking of two polar hydroxyl
groups may render the protected precursors more membrane-
permeant, probably leading to a higher intracellular concentration
as compared to the non-protected final products. Secondly, it
indicates that the southern (and northern) ribose appear to be
structural parts of the cADPR molecule that rather are important
to hold the adenine ring and the pyrophosphate backbone in a
certain distance and orientation than being involved in cADPR
receptor recognition on their own. This has been postulated by
us previously since the cyclic inosine derivative in which both the
northern and the southern ribose were replaced by ether bridges,
termed cIDP-DE,12c showed full biological activity though at
higher concentrations as compared to endogenous cADPR.
CF3COOH). Compounds (13, 14 and 1) were purified twice on
Alltech preparative C18 reversed phase column (2.2 ¥ 25 cm) with
Gilson HPLC by MeCN/TEAB (pH 7.5) buffer systems.
N1 -[(5¢¢-Acetoxyethoxy)methyl]-2¢,3¢,5¢-tri-O-acetyl-8-trifluo-
romethyl-inosine 5. To a solution of 8-bromo derivative 4 (62 mg,
0.105 mmol) and CuI (24 mg, 0.126 mmol) in DMF (2 mL), HMPA
(90 mL, 0.52 mmol) and FSO2CF2CO2Me (134 mL, 1.054 mmol)
were added successively. The mixture was stirred for 27 h at 75 ◦C
under nitrogen. The reaction mixture was then cooled to room
temperature, 6 mL of saturated aq. NH4Cl was added and the
mixture was extracted with 10 mL of EtOAc–hexane (7 : 3). The
organic layer was washed successively with sat. aq. NaHCO3,
water and brine, dried (Na2SO4), filtered, and concentrated under
reduced pressure. Flash chromatography (PE: Acetone = 5 : 1~3 : 1)
afforded the title compound 5 (21 mg, 34%) as a light yellow foam.
1H NMR (300 MHz, DMSO) d 1.94, 1.99, 2.07, 2.12 (each s, each
3H, 4 ¥ Ac), 3.79 (t, 2H, J = 4.5 Hz OCH2), 4.10 (t, 2H, J = 4.5 Hz,
CH2OAc), 4.22 (m, 1H, H-5¢), 4.41–4.45 (m, 2H, H-5¢,H-4¢), 5.49
(s, 2H, 2 ¥ H-1¢¢), 5.68 (m, 1H, H-3¢), 6.09 (m, 2H, H-1¢, H-2¢), 8.73
(s, 1H, H-2); 19F NMR (470 MHz, Acetone) d 24.8; (relative to
CF3COOH). HRMS (ESI-TOF+): calcd for C22H25F3N4O11 [(M +
H)+], 579.1545; found, 579.1550.
5¢-O-TBDPS-2¢,3¢-O-isopropylidene-8-bromoinosine 7. To a
solution of 6 (0.65 g, 1.68 mmol) in DMF (10 mL) imidazole
(1.14 g, 16.8 mmol) and TBDPSCl (2.31 mL, 8.4 mmol) were added
under N2, and the mixture was stirred at room temperature for 12
h. Ice water (10 mL) was added to stop the reaction and the mixture
was evaporated under reduce pressure, the residue was partitioned
between H2O and CH2Cl2. The aqueous phase was extracted again
with CH2Cl2, the organic layers were combined and washed with
brine, dried (Na2SO4), and evaporated. The residue was purified
by silica gel column chromatography (PE:Acetone = 3 : 1) to give
compound 7 (1.047 g, 99%) as a yellow foam.1H NMR (400 MHz,
CDCl3) d 1.02 (s, 9H, (CH3)3C-), 1.40, 1.63 (each s, each 3H,
(CH3)2C), 3.72–3.84 (m, 2H, H-5¢), 4.39 (m, 1H, H-4¢), 5.09 (dd,
1H, J H3¢,H4¢ = 3.6, JH2¢,H3¢ = 6.4 Hz, H-3¢), 5.56 (dd, 1H, JH1¢,H2¢ = 2.0,
JH2¢,H3¢ = 6.4 Hz, H-2¢), 6.15 (d, 1H, JH1¢,H2¢ = 2.0 Hz, H-1¢), 7.24–
7.40 (m, 6H, Ar H), 7.54 (d, JAB = 6.4 Hz, 2H, A of aryl A2B2), 7.60
(d, JAB = 6.4 Hz, 2H, B of aryl A2B2) 7.87 (s, 1H, H-2) 13.07 (s, 1H,
NH); 13C NMR (100 MHz, CDCl3) d,157.9, 149.4, 145.0, 135.5,
133.3, 133.0, 129.8, 129.7, 127.6, 127.5, 126.9, 125.4, 114.4, 91.5,
88.0, 83.2, 81.7, 63.8, 27.2, 26.8, 25.4, 19.2; HRMS (ESI-TOF+):
calcd for C29H33BrN4O5Si [(M + H)+], 625.1476; found, 625.1480.
Conclusions
A convenient reagent and protection strategy is reported for
the synthesis of trifluoromethylated cyclic-ADP-ribose mimic,
8-CF3-cIDPRE 1. Compound 1 and 2¢,3¢-O-isopropylidene 8-
CF3-cIDPRE 14 are cell membrane-permeant cADPR agonists.
To our best knowledge 8-CF3-cIDPRE 1 is the first reported
fluoro substituted cADPR mimic. Importantly, the data shown
here indicate that a strong electron withdrawing substitution
at the 8-position does not necessarily interfere negatively with
the interaction to the cADPR receptor. This phenomenon may
also be attributed to potential hydrogen bond formation of
trifluoromethyl with the cADPR binding pocket. On the another
hand, 2¢,3¢-O-isopropylidene protection of 8-CF3-cIDPRE 14 did
not much alter the agonist activity of compound 1; a similar result
was obtained for cIDPRE and its protected precursor 2¢,3¢-O-
isopropylidene cIDPRE, it may indicate only a minor role for the
both 2¢ and 3¢-hydroxyl groups of the southern ribose of cADPR
for its biological activity.
Experimental
N1-[(5¢¢-Acetoxyethoxy)methyl]-5¢-O-TBDPS-2¢,3¢-O-isopropy-
lidene-8-bromoinosine 8. To the solution of
7 (510 mg,
Chemistry
0.82 mmol) and DBU (1.23 mL, 8.2 mmol) in CH2Cl2 (10 mL)
ClCH2OCH2CH2OAc (0.56 mL, 4.1 mmol) was added at 0 ◦C.
After being stirred for 1 h, the solvent was evaporated in vacuo
and the residue was purified by silica gel column chromatography
(PE:Acetone = 4 : 1) to give compound 8 (365 mg, 60%) as a white
foam. 1H NMR (400 MHz, CDCl3) d 1.02 (s, 9H, (CH3)3C-), 1.39,
1.62 (each s, each 3H, (CH3)2C), 2.03 (s, 3H, AcO), 3.77–3.88 (m,
4H, OCH2, 2 ¥ H-5¢), 4.15–4.17 (m, 2H, CH2OAc), 4.32–4.36 (m,
HR-ESI-MS (electrospray ionization) were performed with
Bruker BIFLEX III. 1H NMR and 13C NMR were recorded with a
Bruker AVANCE III 400 or a JEOL AL300 spectrometer. CDCl3,
DMSO-d6 or D2O were used as a solvent. Chemical shifts are
reported in parts per million downfield from TMS (1H and 13C). 31P
NMR spectra were recorded at room temperature by use of JEOL
AL300 spectrometer (121.5 MHz); Orthophosphoric acid (85%)
was used as an external standard. 19F-NMR spectra were recorded
on a Varian VXR-500 spectrometer. Chemical shifts of 19F-NMR
are reported in ppm with reference to CF3-benzene (-63.7 relative
to CFCl3) as an external standard except compound 5 (relative to
1H, H-4¢), 5.06 (dd, 1H, J
= 4.4, JH2¢,H3¢ = 6.4 Hz, H-3¢), 5.31
H3¢,H4¢
(d, 1H, JH1¢¢a,H1¢¢b = 10.0 Hz, H-1¢¢a), 5.48 (dd, 1H, JH1¢,H2¢ = 2.4,
JH2¢,H3¢ = 6.4 Hz, H-2¢), 5.51 (d, 1H, JH1¢¢a,H1¢¢b = 10.0 Hz, H-1¢¢b),
6.15 (d, 1H, JH1¢,H2¢ = 2.4 Hz, H-1¢), 7.23–7.63 (m, 10H, Ar H),
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4705–4715 | 4711
©

View Article Online
7.63 (s, 1H, H-2); 13C NMR (100 MHz, CDCl3) d170.7, 155.2,
147.9, 146.9, 135.5, 133.5, 132.9, 129.7,127.6, 126.3, 125.2, 114.6,
91.2, 87.5, 83.2, 81.4, 74.9, 68.2, 63.8, 62.9, 27.3, 26.7, 25.5, 20.8,
19.2; HRMS (ESI-TOF+): calcd for C34H41BrN4O8Si [(M + H)+],
741.1950; found, 741.1953.
(51 mg, 0.732 mmol) were added, and the mixture was stirred at
room temperature for 12 h. The mixture was evaporated, and the
residue was partitioned between H2O and CH2Cl2. The organic
layer was washed with brine, dried (Na2SO4), and evaporated.
The residue was purified by silica gel column chromatography
(PE:Acetone = 3 : 1) to give compound 12 (132 mg, 86%). 1H NMR
(400 MHz, DMSO) d 1.34, 1.56 (each s, each 3H, (CH3)2C), 1.94
(s, 3H, AcO), 3.72 (m, 2H, OCH2), 4.05 (m, 2H, CH2OAc), 4.40–
N1-[(5¢¢-Acetoxyethoxy)methyl]-5¢-O-TBDPS-2¢,3¢-O-isopropy-
lidene-8-trifluoromethyl inosine 10. To
a solution of 8-
bromo derivative 8 (300 mg, 0.405 mmol) and CuI (93 mg,
0.486 mmol) in DMF (15 mL) HMPA (0.36 mL, 2.025 mmol) and
FSO2CF2CO2Me (258mL, 2.025 mmol) were added successively.
The reaction mixture was stirred for 12 h at 70 ◦C under nitrogen,
then cooled to room temperature. 10 mL of saturated aq. NH4Cl
were added and the mixture was extracted with 40 mL of
EtOAc–hexane (7 : 3). The organic layer was washed successively
with sat. aq. NaHCO3, water and brine, dried (Na2SO4), filtered,
and concentrated under reduced pressure. Flash chromatography
(PE:Acetone = 5 : 1) afforded the title compound 10 (174 mg,
59%) as a light yellow foam.1H NMR (400 MHz, CDCl3) d 1.02
(s, 9H, (CH3)3C-), 1.38, 1.62 (each s, each 3H, (CH3)2C), 2.03 (s,
3H, AcO), 3.75 (t, 2H, J = 4.6 Hz OCH2), 3.82–3.93 (m, 2H, 2 ¥
H-5¢), 4.17 (t, 2H, J = 4.6 Hz, CH2OAc), 4.35–4.38 (m, 1H, H-4¢),
4.50 (m, 3H, H-4¢, 2 ¥ H-5¢), 5.09 (dd, 1H, JH3¢,H4¢ = 2.8 Hz, JH2¢,H3¢
=
6.4 Hz, H-3¢), 5.47 (q, 2H, J = 10.4 Hz, 2 ¥ H-1¢¢), 5.62 (dd, 1H,
JH1¢,H2¢ = 2.0, JH3¢,H2¢ = 6.4 Hz, H-2¢), 6.10 (d, 1H, JH1¢,H2¢ = 2.0 Hz,
H-1¢), 7.34–7.46 (m, 10H, Ar H), 8.66 (s, 1H, H-2); 13C NMR
(100 MHz, DMSO) d 170.2, 155.6, 151.2, 148.4, 143.1, 136.1 (q,
JCF = 39 Hz), 135.0, 134.9, 134.8, 129.8, 129.6, 125.5, 125.4, 125.3,
125.2, 123.0, 118.2 (q, 1JCF = 269 Hz), 114.0, 90.3, 85.6, 83.3, 80.5,
75.0, 67.2, 66.7, 66.6, 62.8, 26.9, 25.2, 20.6; 19F NMR (470 MHz,
DMSO) d-56.6; 31P NMR (D2O, 121.5 MHz, decoupled with 1H)
d 51.27 ppm (s). HRMS (ESI-TOF+): calcd for C31H32F3N4O9PS2
[(M + H)+], 757.1373; found, 757.1375.
N1 -[(5¢¢-Phosphonoxyethoxy)methyl]-5¢-O-[(phenylthio)phos-
phoryl]-2¢,3¢-O-isopropylidene-8-trifluoromethylinosine 13. To a
solution of 12 (79 mg, 0.104 mmol) in MeOH (4 mL) acetyl
chloride (4.4 mL, 0.062 mmol) was added at 0 ◦C. The mixture
was stirred at the same temperature for 30 min and raised to
room temperature for 24 h, then quenched by addition of a sat.
aq NaHCO3 solution. The mixture was extracted with CH2Cl2,
dried (Na2SO4), filtered and concentrated under reduced pressure.
Flash chromatography (PE:EA = 1 : 2) afforded the deacetylate
product (44 mg) as a white foam. The deacetylate product (44 mg,
0.061 mmol) was dissolved in CH3CN (8 mL). DIEA (63.7 mL,
0.369 mmol) and POCl3 (34.5 mL, 0.302 mmol) was added to the
5.05 (dd, 1H, J
= 4.4, JH2¢,H3¢ = 6.4 Hz, H-3¢), 5.31 (d, 1H,
H3¢,H4¢
JH1¢¢a,H1¢¢b = 10.4 Hz, H-1¢¢a), 5.42 (dd, 1H, JH1¢,H2¢ = 2.4, JH2¢,H3¢
=
6.4 Hz, H-2¢), 5.52 (d, 1H, JH1¢¢a,H1¢¢b = 10.4 Hz, H-1¢¢b), 6.13 (d,
1H, JH1¢,H2¢ = 2.0 Hz, H-1¢), 7.21–7.42 (m, 10H, Ar H), 7.44 (s,
1H, H-2); 13C NMR (100 MHz, CDCl3) d 170.7, 156.1, 148.5,
148.1, 138.3 (d, JCF = 41 Hz) 135.6, 133.5, 132.9, 129.7,127.5,
123.6, 118.1 (d, 1JCF = 270 Hz) 114.8, 90.5, 87.7, 83.7, 81.3, 74.9,
68.3, 63.8, 62.9, 27.3, 26.7, 25.5, 20.8, 19.2; 19F NMR (470 MHz,
CDCl3) d -62.2; HRMS (ESI-TOF+): calcd for C35H41F3N4O8Si
[(M + H)+], 731.2719; found, 731.2720.
◦
solution successively at 0 C, and the mixture was stirred at the
N1 -[(5¢¢-Acetoxyethoxy)methyl]-2¢,3¢-O-isopropylidene-8-tri-
fluoromethylinosine 11. A solution of 10 (165 mg, 0.226 mmol) in
THF (30 mL) was added to 70% HF·Py 1 ml at 0 ◦C. The mixture
same temperature for 14 h, then 5 mL of TEAB (1 M, pH 7.5) was
added at 0 ◦C and stirred for 6 h at r.t. After evaporation in vacuo,
the residue was partitioned between H2O and ethyl acetate, and the
aqueous layer was washed with EA (5 mL ¥ 3) and evaporated in
vacuo. The residue was dissolved in 2 mL of TEAB buffer (0.05 M,
pH 7.5), then applied to a C18 reversed-phase column (2.2 cm ¥
25 cm) developed by a linear gradient of 0–50% CH3CN in TEAB
buffer (0.05 M, pH 7.5) within 30 min to give 13 (26 mg, 36% for
two steps) as a triethylammonium salt. 1H NMR (400 MHz, D2O)
d1.31, 1.52 (each s, each 3H, (CH3)2C), 3.70 (m, 2H, OCH2), 3.84
(m, 2H, CH2OP), 4.02–4.57 (m, 3H, H-4¢, 2 ¥ H-5¢), 5.16 (dd, 1H,
JH3¢,H4¢ = 2.8 Hz, JH2¢,H3¢ = 6.4 Hz, H-3¢), 5.42 (q, 2H, J = 10.8 Hz,
2 ¥ H-1¢¢), 5.67 (dd, 1H, JH1¢,H2¢ = 1.6, JH3¢,H2¢ = 6.4 Hz, H-2¢), 6.10 (s,
1H, H-1¢), 6.92–7.03 (m, 5H, Ar H), 8.46 (s, 1H, H-2). 19F NMR
(470 MHz, D2O) d-60.0; 31P NMR (D2O, 121.5 MHz, decoupled
with 1H) d2.80 ppm (s), 17.74 ppm (s). HRMS (ESI-TOF-) calcd
for C23H27N4O12P2S1F3 [(M - H)-], 701.0701; found, 701.0705.
◦
was stirred at 0 C for 1 h and at room temperature over night.
The reaction mixture was quenched with saturated aq. NaHCO3
at 0 ◦C and diluted with ethyl acetate, then separated and the water
layer was washed with ethyl acetate again. The organic layer was
combined, washed with brine, dried, filtered and concentrated
under reduced pressure. Flash chromatography (PE:EA = 1 : 2)
afforded the title compound 11 (100 mg, 90%) as a white foam.
1H NMR (400 MHz, DMSO) d 1.31, 1.53 (each s, each 3H,
(CH3)2C-), 1.95 (s, 3H, AcO), 3.50–3.58 (m, 2H, 2 ¥ H-5¢), 3.76 (t,
2H, J = 4.6 Hz OCH2), 4.09 (t, 2H, J = 4.6 Hz CH2OAc), 4.16 (dd,
1H, JH3¢,H4¢ = 3.6, JH4¢,H5¢ = 6.0 Hz, H-4¢), 4.94 (t, 1H, JOH,H5¢ = 5.6,
OH), 4.98 (dd, 1H, JH3¢,H4¢ = 3.6 Hz, JH2¢,H3¢ = 6.4 Hz, H-3¢), 5.47
(q, 2H, J = 10.4 Hz, H-1¢¢), 5.54 (dd, 1H, JH1¢,H2¢ = 2.4, JH3¢,H2¢
=
6.4 Hz, H-2¢), 6.01 (d, 1H, JH1¢,H2¢ = 2.4 Hz, H-1¢), 8.66 (s, 1H,
H-2); 13C NMR (100 MHz, DMSO) d 170.2, 155.6, 151.1, 148.5,
N1-[(5¢¢-O-Phosphorylethoxy)methyl]-2¢,3¢-O-isopropylidene-5¢-
O-phosphoryl-8-trifluoromethylinosine 5¢,5¢¢-cyclicpyrophosphate
14. A solution of 13 (26 mg, 37 mmol) in pyridine (4 mL) was
added slowly over 20 h, using a syringe pump, to a mixture of I2
1
136.1 (q, JCF = 40 Hz), 123.0, 118.1 (q, JCF = 270 Hz), 113.7,
90.3, 87.7, 82.9, 81.2, 75.0, 67.2, 62.8, 61.3, 27.0, 25.2, 20.5; 19F
NMR (470 MHz, DMSO) d -56.6; HRMS (ESI-TOF+): calcd for
C19H24F3N4O8 [(M + H)+], 493.1541; found, 493.1542.
˚
(254 mg, 1 mmol) and MS 3 A (1.5 g), in pyridine (40 mL) at room
N1 -[(5¢¢-Acetoxyethoxy)methyl]-5¢-O-[bis(phenylthio)phosphor-
yl]-2¢,3¢-O-isopropylidene-8-trifluoromethylinosine 12. To a solu-
tion of 11 (100 mg, 0.203 mmol) in pyridine (10 mL) TPSCl
(184 mg, 0.609 mmol), PSS (278 mg, 0.732 mmol), and tetrazole
temperature in the dark. The MS 3 A was filtered off with Celite
˚
and washed with H2O. The combined filtrate was evaporated,
and the residue was partitioned between CHCl3 and H2O. The
aqueous layer was evaporated, and the residue was dissolved in
4712 | Org. Biomol. Chem., 2010, 8, 4705–4715
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
0.05 M TEAB buffer (1.0 mL), which was applied to C18 reversed
phase column (2.2 ¥ 25 cm). The column was developed using
a linear gradient of 0–80% CH3CN in TEAB buffer (0.05 M,
pH 7.5) within 30 min to give 14 as a triethylammonium salt.
1.6% DMSO was applied instead of 1 mM or 2 mM 2¢,3¢-O-
isopropylidene cIDPRE. Ratiometric Ca2+ imaging was performed
as described recently36 We used an improvision imaging system
(Tu¨bingen, Germany) built around the Leica microscope usually
at 40-fold magnification. Illumination at 340 and 380 nm was
carried out using a monochromator system (Polychrom II, TILL
Photonics, Gra¨felfing, Germany). Images were taken with a gray-
scale CCD-camera (type C 4742–95; Hamamatsu, Enfield, United
Kingdom) operated in 8-bit mode. The spatial resolution was 512 ¥
640 pixels. The acquisition rate was adjusted to ~14 ratios per
minute. Raw data images were stored on a hard disk and ratio
images (340/380 nm) were constructed pixel by pixel. Finally,
ratio values were converted into Ca2+ concentrations by external
calibration.28 Data processing was performed using Openlab
software, version 3.5.2 (Improvision, Tu¨bingen, Germany).
1
(11.2 mg, 51%). H NMR (400 MHz, D2O) d 1.37, 1.54 (each s,
each 3H, 2 ¥ CH3), 3.73–3.98 (m, 6H, CH2O, 2 ¥ H-5¢, CH2OP),
4.50 (m, 1H, H-4¢), 5.19 (d, 1H, JH1¢¢a,H1¢¢b = 10.8 Hz, H-1¢¢a)
5.41 (dd, 1H, JH2¢,H3¢ = 2.4 Hz, JH3¢,H4¢ = 6.0 Hz, H-3¢), 5.90–5.94
(m, 2H, H-1¢¢b, H-2¢), 6.29 (s, 1H, H-1¢), 8.56, (s, 1H, H-2); 19F
NMR (470 MHz, D2O) d-60.1; 31P NMR (D2O 121.5 MHz,
1
decoupled with H) d-9.77 (d, JP,P = 14.64 Hz), -10.71 (d, JP,P
=
14.64 Hz). HRMS (ESI-TOF-) Calcd for C17H21N4O12P2F3 [(M
-H)-], 591.0511; found, 591.0510.
N1 -[(5¢¢-O-Phosphorylethoxy)methyl]-5¢-O-phosphoryl-8-tri-
fluoromethylinosine 5¢,5¢¢-Cyclic pyrophosphate 1. The solution
of 14 (9 mg, 15.2 mmol) in 15% HCOOH (6 mL) was stirred for 48
h, and then 9 mL TEAB (1M, pH 7.5) was added. The solution was
evaporated under reduced pressure. The residue was dissolved in
0.05 M TEAB buffer (2.0 mL), which was applied to C18 reversed
phase column (2.2 ¥ 25 cm). The column was developed using
a linear gradient of 0–30% CH3CN in TEAB buffer (0.05 M,
pH 7.5) within 30 min to give the designed compound 1 as a
Abbreviations.
Ac
acetyl
Bz
benzoyl
cADPR
cADPcR
cIDPR
cIDPRE
cyclic adenosine 5¢-diphosphoribose
cyclic adenosine 5¢-diphosphocarboribose
cyclic inosine 5¢-diphosphoribose
N1-[(5¢¢-O-phosphorylethoxy)methyl]-5¢-O-
phosphoryl inosine 5¢¢,5¢¢-cyclicpyrophosphate
nicotinamide adenosine dinucleotide
1,8-diazabicyclo[5.4.0]undec-7-ene
dimethylformamide
1
triethylammonium salt. (5.4 mg. 64.5%). H NMR (400 MHz,
D2O) d 3.70–3.74 (m, 2H, CH2O-), 3.86–3.90 (m, 2H, OCH2OP),
3.96–3.99 (m, 1H, H-5¢a), 4.23 (m, 1H, H-5¢b), 4.29–4.31 (m, 1H,
H-4¢), 4.69–4.71 (m, 1H, H-3¢), 5.17 (d, 1H, JH1¢¢a,H1¢¢b = 11.2 Hz,
H-1¢¢a), 5.59 (t, 1H, JH1¢,H2¢ = 5.2 Hz, JH3¢,H2¢ = 5.2 Hz, H-2¢), 5.99
(d, 1H, JH1¢¢a,H1¢¢b = 11.2 Hz, H-1¢¢b), 6.00 (d, 1H, JH1¢,H2¢ = 5.2 Hz,
H-1¢), 8.52 (s, 1H, H-2); 31P NMR (D2O, 121.5 MHz, decoupled
NAD+
DBU
DMF
DMSO
DMTrCl
dimethyl sulfoxide
dimethoxytrityl chloride 1H NMR, proton nuclear
magnetic resonance
1
with H) d-9.65 (d, JP,P = 14.76 Hz), -10.23 (d, JP,P = 14.76 Hz);
19F NMR (470 MHz, D2O) d-60.1; HRMS (ESI-TOF-) Calcd for
C14H17N4O12P2F3 [(M - H)-] 551.0198; found, 551.0216.
13C NMR
19F NMR
carbon-13 nuclear magnetic resonance
fluorine-19 nuclear magnetic resonance. 31P NMR,
phosphorus nuclear magnetic resonance
heteronuclear multiple bond correlation
Hexamethyl phosphoric triamide
high performance liquid chromatography
Pharmacology
HMBC
HMPA
HPLC
Cell culture. Jurkat T-lymphocytes (subclone JMP) were cul-
tured as described previously37 at 37 ◦C in the presence of 5%
CO2 in RPMI 1640 medium containing Glutamax I and HEPES
(25 mM) and supplemented with 7.5% (v/v) newborn calf serum
(NCS), 100 units/ml penicillin and 100 mg ml-1 streptomycin.
HR-ESI- MS high-resolution electro spray ionization mass spec-
trometry
˚
˚
3 A MS
3 A molecular sieve
MMTrCl
PE
PSS
monomethoxytrityl chloride
petroleum ether
Ratiometric Ca2+ imaging. Jurkat T cells were loaded with
Fura-2/AM as described37 and kept in the dark at room tem-
perature until use. For Ca2+ imaging of Jurkat T cells thin glass
coverslips (0.1 mm) were coated first with BSA (5 mg ml-1) and
subsequently with poly-L-lysine (0.1 mg ml-1). Silicon grease was
used to seal small chamber slides consisting of a rubber O-ring
on the glass coverslip. Then, 30 ml buffer A containing 140 mM
NaCl, 5 mM KCl, 1 mM MgSO4, 1 mM CaCl2, 1 mM NaH2PO4,
5.5 mM glucose, 20 mM HEPES, pH 7.4 and 30 ml Fura-2 loaded
T cell suspension were added into the small chamber. The loaded
coverslip was mounted on the stage of a fluorescence microscope
(Leica DMIRBE). For control buffer or buffer containing a small
amount of DMSO were used. In detail, for control (i) pure buffer
was added instead of OKT3 or cIDPRE, (ii) buffer containing
0.1% DMSO was used instead of 1 mM, 10 mM or 100 mM 8-
CF3-cIDPRE, compound 14 or 2¢,3¢-O-isopropylidene cIDPRE,
(iii) buffer containing 1.3% DMSO was used instead of 1 mM or
2 mM 8-CF3-cIDPRE and compound 14, or (iv) buffer containing
cyclohexylammonium
odithioate
S,S-diphenylphosphor-
Py
pyridine
TBDMSCl
TBDPSCl
TEAB
THF
T.L.C.
TPSCl
TMS
tert-butyldimethylsilyl chloride
tert-butyldiphenylsilyl chloride
triethylammonium bicarbonate
tetrahydrofuran
Thin Layer Chromatography
triisopropylbenzenesulfonyl chloride
tetramethyl silicane;.
Acknowledgements
This study was supported by the National Natural Sciences
Foundation of China (grant no. 20332010) and the Deutsche
Forschungsgemeinschaft (GU 360/13-1 to AHG).
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4705–4715 | 4713
©

View Article Online
ribose-mediated Ca2+ release., Biochemistry, 1997, 36, 9509–
9517.
References
1 D. L. Clapper, T. F. Walseth, P. J. Dargie and H. C. Lee, Pyridine
nucleotide metabolites stimulate calcium release from sea urchin egg
microsomes desensitized to inositol trisphosphate., J Biol Chem., 1987,
262, 9561–9568.
2 H. C. Lee, R. Aarhus and D. Levitt, The crystal structure of cyclic
ADP-ribose., Nat. Struct. Biol., 1994, 1, 143–144.
3 H. C. Lee, A. Galione and T. F. Walseth, Cyclic-ADP-Ribose –
metabolism and calcium mobilizing function., Vitam. Horm., 1994,
48, 199–257.
4 A. H. Guse, Regulation of calcium signaling by the second messenger
cyclic adenosine diphosphoribose (cADPR)., Curr. Med. Chem., 2004,
4, 239–248.
5 (a) H. C. Lee, Physiological functions of cyclic ADP-ribose and
NAADP as calcium messengers., Annu. Rev. Pharmacol. Toxicol., 2001,
41, 317–345; (b) A. H. Guse, Biochemistry, biology, and pharmacology
of cyclic adenosine diphosphoribose (cADPR), Curr. Med. Chem.,
2004, 11, 847–855.
6 D. M. Bers, Macromolecular complexes regulating cardiac ryanodine
receptor function, J. Mol. Cell. Cardiol., 2004, 37, 417–429.
7 B. V. Potter and T. F. Walseth, Medicinal chemistry and pharmacology
of cyclicADP-ribose., Curr. Mol. Med., 2004, 4, 303–311.
8 A. H. Guse, Second messenger function and the structure–activity
relationship of cyclic adenosine diphosphoribose (cADPR)., FEBS J.,
2005, 272, 4590–4597.
9 (a) H. C. Lee, R. Aarhus and T. F. Walseth, Calcium mobilization by
dual receptors during fertilization of sea-urchin eggs., Science, 1993,
261, 352–355; (b) A. H. Guse, I. Berg, C. P. da Silva, B. V. L. Potter
and G. W. Mayr, Ca2+ entry induced by cyclic ADP-ribose in intact
T-lymphocytes., J. Biol. Chem., 1997, 272, 8546–8550.
10 S. Shuto, M. Fukuoka, M. Manikowsky, T. Ueno, T. Nakano, R.
Kuroda, H. Kuroda and A. Matsuda, Total synthesis of cyclic ADP-
carbocyclic-ribose, a stable mimic of Ca2+-mobilizing second messenger
cyclic ADP-ribose., J. Am. Chem. Soc., 2001, 123, 8750–8759.
11 (a) C. Moreau, G. K. Wagner, K. Weber, A. H. Guse and B. V. L.
Potter, Structural determinants for N1/N7 cyclization of nicotinamide
hypoxanthine 5¢-dinucleotide (NHD+) derivatives by ADP-ribosyl cy-
clase from Aplysia californica: Ca2+-mobilizing activity of 8-substituted
cyclic inosine 5¢-diphosphoribose analogues in T-lymphocytes., J. Med.
Chem., 2006, 49, 5162–5176; (b) G. K. Wagner, S. Black, A. H. Guse
and B. V. L. Potter, 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.
16 (a) T. Kudoh, M. Fukuoka, S. Ichikawa, T. Murayama, Y. Ogawa,
M. Hashii, H. Higashida, S. Kunerth, K. Weber, A. H. Guse, B. V.
L. Potter, A. Matsuda and S. Shuto, Synthesis of stable and cell-type
selective analogues of cyclic ADP-Ribose, a Ca2+-mobilizing second
messenger. structure–activity relationship of the N1-ribose moiety., J.
Am. Chem. Soc., 2005, 127, 8846–8855; (b) T. Kudoh, T. Murayama,
A. Matsuda and S. Shuto, Substitution at the 8-position of 3¢¢-deoxy-
cyclic ADP-carbocyclic-ribose, a highly potent Ca2+-mobilizing agent,
provides partial agonists., Bioorg. Med. Chem., 2007, 15, 3032–3040;
(c) N. Sakaguchi, Takashi Kudoh, T. Tsuzuki, T. Murayama, T. Sakurai,
A. Matsuda, M. Arisawa and S. Shuto, Synthesis of 5¢¢-branched
derivatives of cyclic ADP-carbocyclic-ribose, a potent Ca2+-mobilizing
agent: The first antagonists modified at the N1-ribose moiety., Bioorg.
Med. Chem. Lett., 2008, 18, 3814–3818; (d) T. Kudoh, K. Weber, A. H.
Guse, B. V. L. Potter, M. Hashii, H. Higashida, M. Arisawa, A. Matsuda
and S. Shuto, Design and synthesis of 4¢¢,6¢¢-unsaturated cyclic ADP-
carbocyclic ribose as a Ca2+-mobilizing agent., Tetrahedron Lett., 2008,
49, 3976–3979.
17 (a) H. Urata and T. Fuchikami, A novel and convenient method
for trifluoromethylation of organic halides using CF3SiR¢3/KF/Cu(I)
system., Tetrahedron Lett., 1991, 32, 91–94; (b) Y. Kobayashi and I.
Kumadaki, Studies on organic fluorine compounds. Part 27. Abnormal
reactions in the trifluoromethylation of aromatic compounds with
trifluoromethyl iodide and copper powder., J. Chem. Soc., Perkin
Trans. 1, 1980, 661–664; (c) P. Lin and J. L. Jiang, Synthesis of
monotrifluoromethyl-substituted saturated cycles., Tetrahedron, 2000,
56, 3635–3671.
18 E. A. Veliz, O. M. Stephens and P. A. Beal, Synthesis and Analysis of
RNA Containing 6-Trifluoromethylpurine Ribonucleoside., Org. Lett.,
2001, 3, 2969–2972.
19 Q. Y. Chen and S. W. Wu, Methyl fluorosulfonyldifluoroacetate, a new
trifluoromethylating agent., J. Chem. Soc., Chem. Commun., 1989, 705–
706.
20 Y. Kobayashi, K. Yamamoto, T. Asai, M. Nakano and I. Kumadaki,
Studies on organic fluorine compounds. Part 35. Trifluoromethylation
of pyrimidine- and purine-nucleosides with trifluoromethyl-copper
complex., J. Chem. Soc., Perkin Trans. 1, 1980, 2755–2761.
21 T. Yamakawa, K. Yamamoto, D. Uraguchi, K. Tokuhisa, Process for
the preparation of nucleic acid base having perfluoroalkyl group. PCT
Int. Appl., 2007055170, 2007-5-18.
22 K. Matsui, E. Tobita, A. Midori and K. Kondo, A convenient
trifluoromethylation of aromatic halides with sodium trifluoroacetate.,
Chem. Lett., 1981, 1719–1720.
12 (a) L. J. Huang, Y. Y. Zhao, L. Yuan, J. M. Min and L. H.
Zhang, Syntheses and calcium mobilizing evaluations of N1-glycosyl
substituted stable mimics of Cyclic ADP-ribose., J. Med. Chem.,
2002, 45, 5340–5352; (b) X. F. Gu, Z. J. Yang, L. R. Zhang and L.
H. Zhang, Synthesis and biological evaluation of novel membrane-
permeant cyclic ADP-ribose mimics: N1-[(5¢¢-O-phosphoryl-ethoxy)-
methyl]-5¢-O-phosphorylinosine 5¢,5¢¢-cyclicpyrophosphate (cIDPRE)
and 8-substituted derivatives., J. Med. Chem., 2004, 47, 5674–5682;
(c) A. H. Guse, X. F. Gu, L. R. Zhang, K. Weber and L. H. Zhang,
A minimal structural analogue of cyclic ADP-ribose., J. Biol. Chem.,
2005, 280, 15952–15959; (d) J.-F. Xu, Z.-J. Yang, W. Dammermann, L.-
R. Zhang, A. H. Guse and L.-H. Zhang, Synthesis and agonist activity
of Cyclic ADP-Ribose analogues with substitution of the northern
ribose by ether or alkane chains., J. Med. Chem., 2006, 49, 5501–5512.
13 (a) Y. Sugimoto, K. Konoki, M. Murata, M. Matsushita, H. Kanazawa
and T. Oishi, Design, Synthesis, and biological evaluation of fluorinated
analogues of salicylihalamide., J. Med. Chem., 2009, 52, 798–806; (b) J.-
X. Yu, V. D. Kodibagkar, W. Cui and R. P. Mason, 19F: a versatile
reporter for non-invasive physiology and pharmacology using magnetic
resonance., Curr. Med. Chem., 2005, 12, 819–848; (c) C. Dalvit, Ligand-
and substrate-based 19F NMR screening: principles and applications
to drug discovery., Prog. Nucl. Magn. Reson. Spectrosc., 2007, 51, 243–
271.
23 S. Shuto, M. Shirato, Y. Sumita and Y. Ueno, Matsuda A. Nu-
cleosides and nucleotides 173. Synthesis of cyclic IDPcarbocyclic-
ribose, a stable mimic of cyclic ADP-ribose, significant facilitation of
the intramolecular condensation reaction of N-1-(carbocyclic-ribosyl)
inosine 5¢,6¢-diphosphate derivatives by an 8-bromo-substitution at the
hypoxyanthine moiety., J. Org. Chem., 1998, 63, 1986–1994.
24 K. C. Nicolaou and S. E. Webber, Stereocontrolled total synthesis of
lipoxins B., Synthesis, 1986, 453–461.
25 C. Heidelberger, David Parsons and D. C. Remy, Syntheses of 5-
trifluoromethyluracil and 5-trifluoromethyl-2¢¢-deoxyuridine, J. Am.
Chem. Soc., 1962, 84, 3597–3598.
26 C. E. Yeom, S. Y. Lee, Y. J. Kim and B. M. Kim, Mild and
chemoselective deacetylation method using a catalytic amount of acetyl
chloride in methanol., Syn Lett., 2005, 10, 1527–1530.
27 L. J. Li, B. C. Lin, Z. J. Yang, L. R. Zhang and L. H. Zhang, A concise
route for the preparation of nucleobase-simplified cADPR mimics by
click chemistry., Tetrahedron Lett., 2008, 49, 4491–4493.
28 S. Kunerth, G. W. Mayr, F. Koch-Nolte and A. H. Guse, Analysis of
subcellular calcium signals in T-lymphocytes., Cell Signal., 2003, 15(8),
783–92.
29 T. Kirchberger, C. Moreau, G. K. Wagner, R. Fliegert, C. C. Siebrands,
M. Nebel, F. Schmid, A. Harneit, F. Odoardi, A. Flu¨gel, B. V. Potter
and A. H. Guse, 8-Bromo-cyclic inosine diphosphoribose: towards a
selective cyclic ADP-ribose agonist., Biochem. J., 2009, 422(1), 139–49.
30 A. H. Guse, A. Goldwich, K. Weber and G. W. Mayr, 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(2), 189–98.
14 B. Zhang, G. K. Wagner, K. Weber, C. Garnham, A. J. Morgan, A.
Galione, A. H. Guse and B. V. L. Potter, 2¢-Deoxy cyclic adenosine
5¢-diphosphate ribose derivatives: importance of the 2¢-hydroxyl motif
for the antagonist activity of 8-substituted cADPR derivatives., J. Med.
Chem., 2008, 51, 1623–1636.
15 G. A. Ashamu, J. K. Sethi, A. Galione and B. V. L. Potter, Roles for
adenosine ribose hydroxyl groups in cyclic adenosine 5¢-diphosphate
4714 | Org. Biomol. Chem., 2010, 8, 4705–4715
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
31 A. Gasser, S. Bruhn and A. H. Guse, Second messenger function of
nicotinic acid adenine dinucleotide phosphate revealed by an improved
enzymatic cycling assay., J. Biol. Chem., 2006, 281, 16906–16913.
32 D. O’Hagan and H. S. Rzepa, Some influences of fluorine in bioorganic
chemistry., Chem. Commun., 1997, 645–652.
33 (a) J. Parsch and J. W. Engels, C–F ◊ ◊ ◊ H–C Hydrogen bonds in
ribonucleic acids., J. Am. Chem. Soc., 2002, 124, 5664–5672; (b) E.
Carosati, S. Sciabola and G. Cruciani, Hydrogen bonding interactions
of covalently bonded fluorine atoms: From crystallographic data to
a new angular function in the GRID force field., J. Med. Chem.,
2004, 47, 5114–5125; (c) R. Frohlich, T. C. Rosen, O. G. J. Meyer, K.
Rissanen and G. Haufe, New indications for the potential involvement
of C–F bonds in hydrogen bonding., J. Mol. Struc., 2006, 78750–
78762.
34 T. J. Barbarich, C. D. Rithner, S. M. Miller, O. P. Anderson and S.
H. Strauss, Significant inter- and intramolecular O–H ◊ ◊ ◊ FC hydrogen
bonding., J. Am. Chem. Soc., 1999, 121, 4280–4281.
35 L. H. Takahashi, R. Radhakrishnan, R. E. Rosenfield, Jr., E. F. Meyer,
Jr. and D. A. Trainor, Crystal structure of the covalent complex
formed by a peptidyl a,a-difluoro-b-keto amide with porcine pancreatic
˚
elastase at 1.78-A resolution., J. Am. Chem. Soc., 1989, 111, 3368–3374.
36 T. Kirchberger, G. Wagner, J. F. Xu, C. Cordiglieri, P. Wang, A. Gasser,
L. Fliegert, S. Bruhn, A. Flu¨gel, F. E. Lund, L. H. Zhang, B. V. L.
Potter and A. H. Guse, Cellular effects and metabolic stability of N-1
cyclic inosine diphosphoribose and its derivatives., Br. J. Pharmacol.,
2006, 149, 337–344.
37 A. H. Guse, E. Roth and F. Emmrich, Intracellular Ca2+ pools in Jurkat
T-lymphocytes., Biochem. J., 1993, 291, 447–451.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 4705–4715 | 4715
©