Synthesis of 2',5'-dideoxy-2-fluoroadenosine and 2',5'-dideoxy-2,5'-difluoroadenosine: potent P-site inhibitors of adenylyl cyclase.

Article abstract of DOI:10.1021/jm0303599

Glycosylation of 2-fluoroadenine with the appropriate protected thioglycoside derivatives, followed by deprotection and anomer separation, produced the alpha- and beta-anomers of 2',5'-dideoxy-2-fluoroadenosine (1), 2',5'-dideoxy-2,5'-difluoroadenosine (2

Full text of DOI:10.1021/jm0303599

J . Med. Chem. 2004, 47, 1207-1213
1207
Syn th esis of 2′,5′-Did eoxy-2-flu or oa d en osin e a n d
2′,5′-Did eoxy-2,5′-d iflu or oa d en osin e: P oten t P -Site In h ibitor s of Ad en ylyl
Cycla se
Song Ye,† Martha M. Rezende,† Wei-Ping Deng,† Brian Herbert,† J ohn W. Daly,† Roger A. J ohnson,‡ and
Kenneth L. Kirk*,†
Laboratory of Bioorganic Chemistry, National Institute of Diabetes and Digestive and Kidney Diseases,
National Institutes of Health, DHHS, Bethesda, Maryland 20892, and Department of Physiology and Biophysics,
State University of New York at Stony Brook, Stony Brook, New York 11794-8661
Received J uly 29, 2003
Glycosylation of 2-fluoroadenine with the appropriate protected thioglycoside derivatives,
followed by deprotection and anomer separation, produced the R- and â-anomers of 2′,5′-dideoxy-
2-fluoroadenosine (1), 2′,5′-dideoxy-2,5′-difluoroadenosine (2), and 2′-deoxy-2-fluoroadenosine
(3). These were examined as P-site inhibitors of adenylyl cyclase. The presence of fluorine on
the purine ring increased potency of inhibition, and the most potent compound, â-2′,5′-dideoxy-
2-fluoroadenosine (1b), was 3 times more potent than â-2′,5′-dideoxyadenosine.
In tr od u ction
Key structural requirements for P-site ligands include
a requirement for an intact adenine moiety, enhanced
inhibitory potency with 2′-deoxy- and especially 2′,5′-
dideoxyribosyl moieties, a â-glycosidic linkage for the
ribosyl moiety, and a notably strong preference for a 3′-
phosphate. For example, examination of adenosine
analogues revealed an order of potency 2′,5′-dd-3′-A4P
> 2′,5′-dd-3′-ATP > 2′,5′-dd-3′-ADP > 2′,5′-dd-3′-AMP
> 2′-d-3′-AMP > 3′-AMP > 2′-d-Ado > Ado.7,8 This SAR
and kinetics of inhibition have provided valuable insight
into the mechanism and modes of binding of P-site
ligands.4,6-9 From structural studies, it is now known
that “P-site” ligands bind within the catalytic active site
at the same locus as substrate 5′-ATP, but with the post-
transition configuration of the enzyme, with the possible
exclusion of divalent cation at one of the two metal
binding sites.10,11
Of the adenosine derivatives studied as P-site inhibi-
tors, those containing 2-F-substituted adenines con-
sistently exhibited increased inhibitory potency.12,13 To
take advantage of this observation, structural features
of other known P-site ligands were combined with this
enhancing 2-fluoro substitution. The initial target was
2′,5′-dideoxy-2-fluoroadenosine (1). With procedures
developed in this synthetic work, we extended this work
to include 2′5′-dideoxy-2,5′-difluoroadenosine (2) and the
known 2′-deoxy-2-fluoroadenosine (3). We report herein
the details of the syntheses, as well as the results of
examination of their effects on adenylyl cyclase activity.
Adenosine 3′:5′-cyclic monophosphate (cAMP) is a
ubiquitous regulatory molecule that plays a critical role
in intracellular signaling pathways through activation
of protein kinases and of cyclic nucleotide-gated ion
channels. The diverse and critical effects on cell function
that result from these downstream activation events
make clear the importance of precise regulation of
intracellular cAMP levels. This regulation is accom-
plished by modulation of the activity of adenylyl cyclase,
the enzyme that catalyzes the synthesis of cAMP from
5′-ATP, most notably by interaction with G-protein R
subunits. These subunits mediate the influence of cell-
surface receptors that are activated by such diverse
moieties as peptide hormones, neurotransmitters, pros-
taglandins, and adenosine.1
The actions of adenosine on adenylyl cyclase are
complex. In 1970 Sattin and Rall2 demonstrated that
adenosine stimulates cAMP accumulation in brain slices
through stimulation of adenylyl cyclase. Later it was
shown that adenosine also could inhibit adenylyl cy-
clase. Subsequently, Londos and Wolff3 proposed two
distinct sites of action for adenosine. These were the so-
called R-sites (ribose moiety of adenosine strictly re-
quired) that are adenosine binding domains of cell-
surface G-protein coupled receptors that activate or
inhibit cyclase activity indirectly through interactions
of Gs or Gi. In contrast, the proposed P-sites (purine
moiety of adenosine required) proved to be directly on
adenylyl cyclase and binding of adenosine to these sites
was responsible for direct inhibition of the enzyme. The
mechanism and possible physiological significance of
modulation of adenylyl cyclase activity by the binding
of endogenous adenosine and by synthetic P-site inhibi-
tors recently have attracted much attention.4-9
Ch em istr y
The development of methods to synthesize nucleosides
and nucleotides fluorinated on the sugar and/or base
moiety has been an active area of research in nucleoside
chemistry. This research reflects the potent biological
properties possessed by many of these fluorinated
analogues. In particular, many members of this class
have proven to be effective anticancer and antiviral
agents. There have been two general approaches to the
syntheses of 2-fluoroadenine nucleosides. Much of the
* To whom correspondence should be addressed. Address: NIH, 8
Center Drive, MSC 0810, Bethesda, MD 20892. Phone: 301 496-2619.
Fax: 301 402-4182. E-mail: kennethk@bdg8.niddk.nih.gov.
† National Institutes of Health, DHHS.
‡ State University of New York at Stony Brook, Stony Brook.
10.1021/jm0303599 CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/29/2004

1208 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 5
Ye et al.
Sch em e 1a
Sch em e 2a
a
Reagents and conditions: (a) Dowex H+, MeOH; (b) I2, Ph3P
(ref 10); (c) H2/Pd-C (ref 10); (d) Ac2O, pyridine; (e) BnBr, NaH,
(n-Br)4N+I-; (f) PhSH, ZnI2, (n-Bu)4N+I-.
a
Reagents and conditions: (a) HMDS, (NH4)2SO4, reflux, 3 h,
48-52%; (b) sugar 7 or 8, CH2Cl2, molecular sieves, 4Å, NBS, room
temp, 72 h; (c) NaOMe/MeOH, reflux, 3 h, ∼100%; (d) (60%) HF/
pyridine, t-BuNO2, -20 to -30 °C, ∼4 h, ∼10%.
early work was based on synthesis of an appropriate
2-aminopurine nucleoside, followed by generation of a
2-diazonium group and displacement of this group with
fluoride. For example, Montgomery and Hewson14 pre-
pared 2-fluoroadenosine from 2-aminoadenosine in 16%
yield by a Schiemann reaction in aqueous solution.
Using KNO2/HF/pyridine to effect this conversion,
Krolikiewicz and Vorbru¨ggen15 improved the conversion
to 80%. Montgomery and Hewson16 initially prepared
2′-deoxy-2-fluoroadenosine by a multistep sequence that
included fluorination of a 2,6-diaminopurine nucleoside
precursor as a key step, but the overall yield was quite
low. Secrist and co-workers17 recently developed an
alternative approach to 2′-deoxy-2-fluoroadenosine based
on free radical removal of the 2′-OH group from the
more readily available 2-fluoroadenosine. The second
general approach involved synthesis of a fluorinated
base followed by coupling to the sugar. However, direct
coupling of such bases with 2-deoxysugars is com-
plicated by a tendency to form mixtures of R- and
â-anomers, a complication lessened in the ribose series
because of the participation of the sugar C-2 OH group
during coupling. However, a recent report18 describes
an NBS-promoted coupling reaction of thioglycosides
with silylated heterocyclic bases that favors formation
of the â-anomer when applied to 2′-deoxy-D-threo-pento-
furanosyl thioglycosides. Although steric factors in the
glycosylation reaction undoubtedly will differ with 5′-
deoxypentofuranosides in the erythro configuration, we
have emphasized this coupling procedure in the present
syntheses.
The condensation reaction of the sugars 7 or 8 with
the silylated derivative of 2,6-diacetamidopurine pro-
ceeded with poor stereoselectivity to give a chromato-
graphically inseparable anomeric mixture of the N-9
regioisomers of 2-acetamido-N6-acetyl-9-(3-O-acetyl-2,5-
dideoxy-R,â-d-erythro-pentofuranosyl)adenine (9) in 52%
yield (R:â ) 1.25:1 by 1H NMR) and 2-acetamido-N6-
acetyl-9-(3-O-benzyl-2,5-dideoxy-R,â-D-erythro-pento-
1
furanosyl)adenine (10) in 48% (R:â ) 4:1 by H NMR)
(Scheme 2). The anomeric configurations of 9 or 10, and
also of 11 (below), were tentatively assigned on the basis
of empirical anomeric proton NMR line shape criteria
described by Robins et al.21 According to these criteria,
the H-1′ proton of the â-anomer will appear as an
“apparent triplet” whereas in the R-configuration H-1′
will appear as a doublet of doublets. The deacyl-
ation of the anomeric mixtures of 9 and 10 in 1 N
methanolic sodium methoxide produced 11 and 12
(quantitative yield), respectively, in an R:â ratio of 1:1
of 11 and an R:â ratio of 4:1 of 12 (Scheme 2) (see
Supporting Information).
The R and â anomers of 11 were separated by
preparative thin-layer chromatography on silica gel for
further characterization. The anomeric configuration of
the higher Rf product was assigned the R-configuration
from a 2D-NOESY experiment (see Supporting Infor-
mation) confirming the assignment based on the signal
shape of the H-1′ proton.
The anomeric mixtures of compounds 11 (R:â ) 1:1)
and 12 (R:â ) 4:1) were then treated with 60% HF/
pyridine and tert-butyl nitrite at -10 °C to produce 1a
and 13 in unsatisfactory yields (ca. 10%). The 3-O-benzyl
derivative 13 was obtained as an anomeric mixture (R:â
) 1:1) instead of the R:â ratio of 4:1 present in 12, and
only the R-anomer 1a was obtained from 11. This may
be due to the difficulties encountered during the separa-
tion/purification of 13 or 1 by preparative chromatog-
raphy.
Cou p lin g of Dia m in op u r in e Der iva tives F ol-
low ed by F lu or in a t ion . We initially attempted the
synthesis of 2′,5′-dideoxy-2-fluoroadenosine (1) based on
initial coupling of a silylated derivative of readily
available 2,6-diacetamidopurine with the appropriate
thioglycoside. Methyl 2,5-dideoxyribofuranoside (4) was
prepared from 2-deoxyribofuranose as described by Soll
and Seitz,19 with minor changes in the procedure
(Scheme 1). For example, we used an acidic resin
(Dowex, hydrogen form, HCR-W2) to catalyze the initial
formation of methyl 2-deoxyribofuranoside from the
furanose, since product could be isolated by simple
filtration from the resin, obviating the neutralization
of acid. Replacement of the 5-hydroxyl group with iodine
occurred under Mitsunobu conditions, and hydrogenoly-
sis produced 4 as a mixture of anomers. This mixture
was converted to the 3-O-acetyl derivative 5 with acetic
anhydride and to the 3-O-benzyl derivative 6 by alkyl-
ation with benzyl bromide. The phenyl 2-deoxy-1-thio-
D-erythro-pentofuranosides (7 and 8) were easily pre-
pared by treating 5 or 6 with benzenethiol, zinc iodide,
and tetrabutylammonium iodide in anhydrous di-
chloromethane (Scheme 1).20
Glycosyla tion Rea ction w ith 2-F lu or oa d en in e.
Syn th esis of 2′,5′-Did eoxy-2-flu or oa d en osin e (1).
The low yield in the fluorination reaction and the poor
stereoselectivity of the glycosylation reaction prompted
us to examine a different approach, based on glycosyl-
ation of the fluorinated purine. This alternative ap-
proach was made more attractive by the commercial
availability of 2-fluoroadenine.
Sugino and Sugimura20 found that reaction of a series
of silylated purines (and silylated pyrimidines) with
2-deoxy thiosugars (threo) produced excellent yields of
nucleosides with good â-selectivity. However, condensa-

Synthesis of Fluoroadenosines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 5 1209
Sch em e 3a
a
Reagents and conditions: (a) HMDS, (NH4)2SO4, reflux; (b)
F igu r e 1. NOEs observed between H-1′ and other protons in
compounds 5a and 5b.
thioglycoside 7, NBS, molecular sieves, 4 Å, solvent, 0°C to room
temperature.
Ta ble 1. Optimization of the Coupling Reaction of Silylated
2-F-adenine with Thioglycoside 7
amount of
(NH4)2SO4, equiv
entry
solvent
yield,a
%
R:â of 14b
1
2
3
4
0.1c
1.0
1.0
4.0
DCM
DCM
toluene
toluene
29
33
47
34
4.6:1
1.4:1
1.3:1
1.2:1
a
b
Isolated yields. Determined by 1H NMR (300 MHz). c Small
F igu r e 2. N-9 and N-7 regioisomers of 1b.
amount of DMF was added (DMF/HMDS, 5:100).
Sch em e 4a
(Figure 1A). These findings indicate that the higher Rf
product is the R anomer 1a . This confirmed the previous
assignment of 1a produced by the alternative synthetic
procedure described above. The 2D-NOESY experiment
of the lower Rf product showed NOEs between H-1′ and
H-2′a and between H-1′ and H-4′, while there was no
NOE between H-1′ and H-5′ (Figure 1B). These results
indicate that this is the â anomer 1b. The spectra are
provided in Supporting Information.
a
Reagents and conditions: (a) 5% Et3N in MeOH, room tem-
perature.
The assignment of the site of the attachment of the
purine ring (N-7 or N-9) of 1b was investigated by the
heteronuclear multiple bond correlation (HMBC). The
cross-peaks in HMBC spectra contain information on
heteronuclear long-range coupling constants between
protons and carbons separated by two or three chemical
tion of 7 (erythro) with silylated 2-fluoroadenine by
NBS-promoted reaction under conditions described by
Sugino and Sugimura (Scheme 3) initially gave modest
yields and undesired R-selectivity (entry 1, Table 1).
This was similar to our results with the diacetamido-
purine. The low solubility of 2-fluoroadenine in hexa-
methyldisilazane seemed perhaps problematic, but ad-
dition of DMF led to lower yields. However, if DMF was
excluded, silylation was not complete even after 24 h.
Fortunately, with addition of 1.0 equiv of ammonium
sulfate and vigorous reflux in hexamethylsilazane, the
reaction was complete after several hours, as evidenced
by the formation of a clear solution. The yield of 14 was
increased only modestly, but there was a more signifi-
cant decrease in R-selectivity (entry 2). The yield was
further improved by replacing dichloromethane with
toluene as solvent (entry 3). When the amount of
ammonium sulfate was increased to 4 equiv, the yield
was decreased with almost the same R-selectivity. We
have used conditions corresponding to entry 3 for the
preparation the R- and â-anomers of 14 (Scheme 3) and
related nucleosides (see below) and consistently have
achieved satisfactory results.
3
3
bonds (2J C-H or J C-H). As shown in Figure 2, a J C-H
can be observed between H-1′and C-4 in the case of the
3
regioisomer N-9, while a J C-H between H-1′and C-5′
can be detected in the N-7 regioisomer (C-4 distin-
3
guished from C-5 with the coupling of J C-F on 13C
NMR). The HMBC spectrum of compound 1b showed a
correlation between H-1′ and C-4, while there was no
correlation between H-1′ and C-5. This suggested that
compound 1b was the N-9 regioisomer. The spectrum
is provided in Supporting Information.
Syn th esis of 2′,5′-Did eoxy-2,5′-d iflu or oa d en osin e
(2). The stronger inhibition of adenylyl cyclases by 2′,5′-
dideoxyadenosine compared with 2′-deoxyadenosine
demonstrates the favorable influence on binding caused
by replacement of the 5′-hydroxyl group with hydrogen.
The ability of the 5′-deoxy sugar to have favorable
hydrophobic interactions with a binding domain has
been proposed as an explanation for the enhanced
potency.9 Fluorine is considered to be a good isosteric
replacement for hydroxyl, but fluorine is unable to
function as a hydrogen bond donor.22 Fluorine substitu-
tion would also increase hydrophobicity relative to a 5′-
OH but presumably would decrease hydrophobicity
relative to the 5′-deoxy analogue.22 To explore the effects
of 5′-substitution, the 5′-deoxy-5′-fluoro analogue was
prepared.
The anomeric mixture of product 14 was deacetylated
to give the corresponding R and â anomers 1a and 1b
in high yield. These were separated by chromatography
on silica gel (Scheme 4).
The anomeric configuration of the higher Rf product
was assigned from a 2D-NOESY experiment. An NOE
between H-1′ and H-5′, H-1′ and H-2′b, and H-1′ and
H-3 was observed, while there were no NOEs observed
between H-1′ and H-4′ and between H-1′ and H-2a′
The synthesis was also based on condensation of the
thioglycoside with silylated 2-fluoroadenine. For this

1210 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 5
Ye et al.
Sch em e 5a
a
Reagents and conditions: (a) ref 26; (b) Ac2O, pyridine, DCM,
0 °C to room temperature, 98%; (c) bis[(2-methoxyethyl)amino]-
sulfur trifluoride, DCM, reflux, 86%; (d) PhSH, ZnI2, n-Bu4N+I-,
DCM, 0 °C to room temperature, 47%.
F igu r e 4. HMBC observed between H-1′ and C-4 of compound
2b.
Sch em e 6a
a correlation between H-1′ and C-4, while there is no
correlation between H-1′ and C-5 (Figure 4). This
confirms that compound 1b was the desired N-9 regio-
isomer. The spectra are provided in Supporting Infor-
mation.
P r ep a r a tion of 2′-Deoxy-2′-flu or oa d en osin e (3).
Previous syntheses of this analogue by Montgomery and
Hewson16 and by Secrist and co-workers17 were dis-
cussed in the Introduction. Because of the current
interest in this analogue as a prodrug for the toxic
2-fluoroadenine,24 we have communicated in an earlier
report our new synthesis of this compound using the
same approach as described above for the synthesis of
1 and 20.25
a
Reagents and conditions: (a) HMDS, (NH4)2SO4, reflux; (b)
thioglycoside 18, NBS, molecular sievers, 4 Å, DCM, 20% (19a )
and 12% (19b); (c) 5% Et3N in MeOH/H2O (10:1), 70% (2a ), 69%
(2b).
Biologica l Resu lts
The efficacy of the nucleoside derivatives described
in this report was tested with a detergent-dispersed
preparation of adenylyl cyclase extracted from whole rat
brain. This particular preparation has been used to
characterize inhibition kinetics and inhibition by nu-
merous ligands over many years and as such provides
a reliable and remarkably reproducible means by which
inhibitory potencies can be compared.4-8 As predicted
from data in an earlier report with 2F-Ado,12 â-2′-d-2-
F igu r e 3. NOEs observed between H-1′ and other protons in
anomers 2a and 2b.
F-Ado (IC50 ≈ 4.6 µM) and â-2′,5′-dd-2-F-Ado (IC50
≈
synthesis we developed a new synthesis of 2′,5′-deoxy-
5′-fluororibose, a compound previously made by aldolase
condensation of 3-fluoro-2-hydroxypropanal with acet-
aldehyde.23 Methyl 2-deoxyribofuranoside was converted
to the 3-O-acetyl-5-O-(tert-butyldimethylsilyl) derivative
16 in two steps. Reaction of 16 with bis[(2-methoxy-
ethyl)amino]sulfur trifluoride gave the 2,5-dideoxy-
5-fluoro riboside 17, from which was prepared the
phenylthioriboside 18 with the usual procedure (Scheme
5).
Reaction of 18 with 2-fluoroadenine under conditions
optimized for the preparation of 14 produced the ano-
meric mixture 19a and 19b. These were separated by
chromatography on silica gel and deacylated to give 2a
and 2b (Scheme 6).
0.9 µM) were noticeably more potent that the corre-
sponding unmodified nucleosides â-2′-d-Ado (IC50 ≈ 15
µM) and â-2′,5′-dd-Ado (IC50 ≈ 2.8 µM) (Figure 5 and
Table 2). The lack of effectiveness of the cognate
R-anomers, R-2′-d-2-F-Ado and R-2′,5′-dd-2-F-Ado, was
consistent with an earlier observation that R-Ado was
not inhibitory.4 Somewhat unexpected were the obser-
vations that â-2′,5′-dd-2,5′-di-F-Ado (IC50 ≈ 1 µM) was
essentially as potent as â-2′,5′-dd-2-F-Ado. The weak
activity detected for the corresponding R-anomer, R-2′,5′-
dd-2,5′-di-F-Ado (IC50 ≈ 29 µM), could be caused by
traces of the â-anomer in the test sample, an issue under
investigation. The data support the concept that 2-F-
substituted adenine derivatives will provide a basis for
the development of yet more potent and selective
inhibitors and of this important family of enzymes.
Experiments on the effects of these compounds on intact
cell and tissue systems are in progress. Given the
central role that adenylyl cyclases play in transmem-
brane signaling mechanisms, the availability of more
potent, selective, and membrane-permeable inhibitors
should be expected to have broad usefulness in research
in areas of cell physiology and intercellular signaling.
The R and â anomers of 2a and 2b (Figure 3) were
assigned by their distinguished NOESY spectra in the
same manner as for 1a and 1b (Figure 2). The spectra
are provided in Supporting Information.
It is necessary to point out that the coupling system
of H-1′ of R-anomer 1a and 2a is doublet-doublet, while
the coupling system of H-1′ of â-anomer 1b and 2b are
1
triplet (2J 1′,2′a ) 2J 1′,2′b) in their H NMR spectra. Thus,
the “line shape” criterion observed by Robins et al. can
be verified by the NOEs observed in these experi-
ments.21
Exp er im en ta l Section
Melting points were determined on a Thomas-Hoover capil-
lary melting point apparatus in sealed tubes and are uncor-
rected. Analytical TLC was performed on Merck silica gel
The N-9 regioisomer 2b was identified by its HMBC
spectrum. The HMBC spectrum of compound 2b showed

Synthesis of Fluoroadenosines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 5 1211
are expressed in δ ppm upfield (minus sign) from CCl3F. Low-
resolution electron-impact (EI) mass spectra were obtained on
a Finnigan 4600 spectrometer with a typical ionization voltage
of 70 eV. High-resolution chemical ionization (CI) mass spectra
were obtained on a VG Analytical 7070 E spectrometer using
ammonia. Data are reported in the form m/e (intensity relative
to base ) 100). Elemental analyses were performed by Atlantic
Microlab. Inc, Norcross, GA, and Galbraith Laboratories, Inc.,
Knoxville, TN. Yields refer to chromatographically and spec-
troscopically pure compounds.
Meth yl 3-O-Acetyl-2,5-d id eoxy-D-er yth r o-p en tofu r a n o-
sid e (5). To an ice-cold solution of methyl 2,5-dideoxy-R,â-D-
erythro-pentofuranoside 4 (1.1 g, 8.32 mmol) in anhydrous
dichloromethane (5 mL) was added pyridine (1.5 mL, 18.5
mmol) and acetic anhydride (1.7 mL, 18 mmol). The reaction
mixture was stirred for 20 h at room temperature under N2.
The solution was concentrated by rotary evaporation, and the
residue was coevaporated with toluene and was chromato-
graphed on silica gel (hexane/EtOAc, 9:1) to give 1.05 g of a
mixture of R and â anomers of 5 as a colorless oil (72% yield).
1H NMR (CDCl3, 300 MHz) (R and â), δ: 1.33 (d, 3H, J 4,5
6.6, H-5), 2.05, (s, 3H, CH3COO-), 2.19 (ddd, 1H, J 2b,3 ) 3.9
Hz, J 1,2b ) 5.5 Hz, J gem ) 14.5, H-2b), 2.35 (ddd, 1H, J 1,2a
2.7 Hz, J 2a,3 ) 6.75 Hz, J ) 14.5 Hz, H-2a), 3.37 (s, 3H,
OCH3), 4.15 (qd, 1H, J 3,4 )2.7 Hz, J 4,5) 6.6 Hz, H-4), 5.03 (ddd,
1H, J 2a,3) 6.75 Hz, J 2b,3 ) 3.9 Hz, J 3,4 )2.7 Hz, J 2b,3 ) 3.9 Hz,
H-3), 5.12 (dd, 1H, J 1,2a ) 2.7 Hz, J 1,2b ) 5.5 Hz, H-1). IR (CCl4),
cm-1: 2970 (s), 1739 (s), 1450 (w), 1368 (w), 1260 (s), 1096 (s),
990 (s), 865 (w), 798 (s), 665 (w), 487 (w). MS (CI, NH3), m/e:
192 ([M + NH4]+, 5.0), 160 (100), 143 (50). Anal. Calcd for
C8H14O4: C, 55.16; H, 8.10. Found: C, 55.47; H, 8.30.
)
)
2a,2b
F igu r e 5. Inhibition of adenylyl cyclase by fluorine deriva-
tives of adenosine. Adenylyl cyclase was extracted from whole
rat brain and assayed as previously described.4-8 The reaction
mixture included 100 µM MnATP, 5 mM MnCl2, 100 µM
forskolin for a 15 min reaction at 37 °C. Data are from
representative experiments, and values are enzyme velocities
(nmol of cAMP formed/(min‚mg protein)) in the presence of
inhibitors relative to velocities in their absence.
P h en yl 3-O-Acet yl-2,5-d id eoxy-1-t h io-r,â-D-er yth r o-
p en tofu r a n osid e (7). Zinc iodide (3.66 g, 11.47 mmol),
benzenethiol (1.25 mL, 11.48 mmol), and tetrabutylammonium
iodide (0.5 g, 1.35 mmol) were added to a solution of 5 (1.0 g,
5.74 mmol) in dry CH2Cl2 (60 mL). This suspension was
refluxed with stirring under N2. After 7 h, the reaction mixture
was filtered. The filtrate was washed with saturated NaHCO3
(10 mL) and was extracted with CH2Cl2. The organic layer was
dried over Na2SO4 and evaporated under reduced pressure.
The residue was chromatographed on silica gel with hexane/
EtOAc (9:1) to give 1.02 g (70%) of a mixture of R and â
anomers (1.8:1) of 7 as a colorless oil. 1H NMR (CDCl3, 300
MHz) (R and â), δ: 1.30 (d, 3H, J 4,5 ) 6.6 Hz, H-5), 1.33 (d,
3H, J 4,5 ) 6.3 Hz, H-5), 1.99-2.05 (m, 4H, CH3COO-, H-2a),
Ta ble 2. Inhibition of Adenylyl Cyclase by Adenine
Nucleosidesa
nucleoside
IC50, µM
â-Ado
82b
>300b
15b
R -Ado
â-2′-d-Ado
â-2′-d-2-F-Ado
4.6
R-2′-d-2-F-Ado
>100
â-2′,5′-dd-Ado
2.8
â-2′,5′-dd-2-F-Ado
R-2′,5′-dd-2-F-Ado
â-2′,5′-dd-2,5′-di-F-Ado
R-2′,5′-dd-2,5′-di-F-Ado
0.89
>100
0.98
29
2.10 (s, 3H, CH3COO-), 2.28 (ddd, 1H, J 2a,3 ) 5.7 Hz, J 1,2a
8.7 Hz, J 2a,2b ) 14.4 Hz, H-2b), 2.39 (ddd, 1H, J 1,2a ) 6.3 Hz,
J 2a,3 ) 1.8 Hz, J 2a,2b ) 14.4 Hz, H-2a), 2.88 (ddd, 1H, J 1,2b
)
a
Experiments were conducted as described for Figure 5. IC50
values are averages from at least two experiments and were
determined from graphical analyses of data as shown in Figure
)
7.8 Hz, J 2b,3 ) 8.0 Hz, J 2a,2b ) 14.7 Hz, H-2b), 4.17 (qd, 1H,
J 3,4 ) 1.8 Hz, J 4,5 ) 6.6 Hz, H-4), 4.31-4.40 (m, 1H, H-4), 4.77
(ddd, 1H, J ) 3.9 Hz, J ) 5.0, J 2b,3 ) 8.0 Hz, H-3), 4.93 (dt,
1H, J 3,4 ) J 2b,3 ) 1.8 Hz, J 2a,3 ) 5.7 Hz, H-3), 5.54 (dd, 1H,
b
2. Variation in IC50 values is less than (20%. Values from ref 4.
plates (10 cm) with QF-254 indicator. Reactions were moni-
tored by TLC, and spots were visualized with UV light or by
spraying with a phosphomolybdic acid (20 wt % in ethyl
alcohol) reagent. Preparative TLC were performed on Uniplate
silica gel GF plates (20 cm × 20 cm, 2000 µm) purchased from
Analtech. Flash column chromatography was performed with
230-400 mesh silica gel-60 from Merck. Solvents for extraction
and chromatography were technical grade and were used as
received. Reaction solvents were distilled from the indicated
drying agents: dichloromethane (P2O5), methanol (magne-
sium), tetrahydrofuran (THF) (sodium benzophenone ketyl),
and hexamethyldisilazane (CaH2). Infrared spectra (IR) were
obtained on an Easy IR 314 FTS/45 spectrophotometer from
the BIO-RAD Digilab Division. Peaks are reported in cm-1
with the following relative intensities: s (strong, 67-100%),
m (medium, 34-66%), w (weak, 0-33%). 1H and 13C NMR
spectra were recorded on a Varian Gemini-300 FT spectrom-
eter. Tetramethylsilane (TMS) was used as an internal
J 1,2b ) 6.3 Hz, J 1,2a ) 8.7 Hz, H-1), 5.69 (dd, 1H, J 1,2a
)
3.3 Hz, J 1,2b ) 7.8 Hz, H-1), 7.20-7.54 (m, 5H, H-arom). 13C
NMR (CDCl3, 75 MHz), δ: 7.96 and 19.81 (C-5), 20.84
(CH3COO-, ), 37.75 and 38.97 (C-2), 78.67 and 77.91 (C-4),
81.79 (C-3), 85.32 and 86.35 (C-1), 126.96-135.92 (C-arom),
170.94 (-CO-). IR (CCl4), cm-1: 2962 (s), 2906 (w), 1737 (s),
1583 (w), 1481 (w), 1439 (w), 1242 (s), 1178 (w), 1091 (s), 1020
(s), 937 (w), 798 (s), 741 (w), 691 (w), 513 (w). MS (EI, 70 eV),
m/e: 252 (M+, 10), 208 (10), 143 (100), 109 (100), 83 (100), 55
(100). MS (CI, NH3), m/e: 270 ([M + NH4]+, 90), 160 (90), 143
(100). HRMS (CI, NH3) Calcd for (C13H16O3S): 252.33.
Found: 252.08.
2-Flu or o-9-(3-O-acetyl-2, 5-dideoxy-r,â-D-er yth r o-pen to-
fu r a n osyl)a d en in e (14). A suspension of 2-fluoroadenine
(766 mg, 5.0 mmol) and (NH4)2SO4 (661 mg, 5.0 mmol) in
hexamethyldisilazane (10 mL) was heated at rigorous reflux
under N2 until the solution became clear. The excess of
hexamethyldisilazane was removed under reduced pressure.
The residue and thioglycoside 7 (1.26 g, 5.0 mmol) were
dissolved in 20 mL of dry toluene under N2, and then 2.5 g of
powered molecular sieves, 4 Å, was added. After being stirred
at room temperature for 30 min, the reaction mixture was
1
standard for H and 13C NMR spectra and CCl3F for 19F NMR
spectra. Spectra were recorded in the following solvents:
CDCl3 (7.26 ppm for 1H, 77.0 ppm for 13C) and dimethyl
sulfoxide (δ 2.50 ppm for 1H, 39.5 ppm for 13C). All 19F NMR
spectra were recorded in Me2SO-d6, and 19F chemical shifts

1212 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 5
Ye et al.
cooled to 0 °C and NBS (979 mg, 5.5 mmol) was added. The
resulting mixture was allowed to warm to room temperature
and stirred for 18 h. The addition of aqueous Na2S2O3 was
followed by filtration and then extraction with CH2Cl2. The
organic layer was dried over MgSO4 and concentrated under
reduced pressure. The residue was purified by chromatography
on silica gel (dichloromethane/methanol, 40:1) to give 694 mg
(47%) of 14 as a mixture of R and â anomers (R:â ) 1.3:1) as
Hz, C3), 55.35 (OMe), 39.18 (C2), 21.09 (CH3CO). 19F NMR
(CDCl3, 282 MHz), δ -226.7 (dt, J ) 47.4, 18.3 Hz).
1
â An om er . H NMR (CDCl3, 300 MHz), δ: 5.16-5.11 (2H,
m, H-1 and H-3), 4.62 (2H, dm, 1J HF ) 47.4 Hz, H-5), 4.21 (1H,
2
dm, J HF ) 28.8 Hz H-4), 3.41 (3H, s, OMe), 2.42-2.00 (2H,
m, H-2), 2.10 (3H, s, CH3CO). 13C NMR (CDCl3, 75 MHz), δ:
171.24 (CO), 105.42 (C1), 83.93 (1J CF ) 171.3 Hz, C5), 82.34
(2J CF ) 18.2 Hz, C4), 73.60 (3J CF ) 6.8 Hz, C3), 55.32 (OCH3),
39.16 (C2), 21.18 (CH3CO). 19F NMR (CDCl3, 282 MHz), δ:
-226.7 (dt, J ) 47.1, 27.3 Hz).
1
a white solid. H NMR (CDCl3, 300 MHz), δ: 8.06 (1H, s, H-8
(R)), 7.94 (1H, s H-8 (â)), 6.39 (1H, dd, 7.2, 2.1 Hz H-1′ (R)),
6.30 (1H, dd, 8.1, 6.0 Hz, H-1′ (â)), 6.03 (br, NH2 (R and â)),
5.15 (1H, dt, 6.0, 2.1 Hz, H-3′ (â)), 5.09 (1H, dt, 6.6, 1.5 Hz,
H-3′ (R)), 4.56 (1H, dq, 2.1, 6.6 Hz, H-4′ (R)), 4.28 (1H, dq, 2.4,
6.6 Hz, H-4′ (â)), 3.0-2.5 (4H, m, H-2′ (R and â)), 2.13 (3H, s,
OAc (â)), 2.00 (3H, s, OAc (R)), 1.47 (3H, d, 6.6 Hz, H-5′(â)),
1.36 (3H, d, 6.6 Hz, H-5′ (R)). MS (CI, NH3), m/e: 296 (MH+).
P h en yl 3-O-Acetyl-5-flu or o-2-deoxy-1-th io-r,â-D-er yth r o-
p en tofu r a n osid e (18). To the solution of furanoside (17) (2.46
g, 12.8 mmol) in dry dichloromethane (100 mL) at 0 °C were
added benzenethiol (2.82 g, 25.6 mmol), zinc iodide (7.30 g,
22.9 mmol), and tetrabutylammonium iodide (0.946 g, 2.56
mmol). After being stirred for 1 h at 0 °C, the reaction mixture
was allowed to warm to room temperature and stirred for 4
h. The reaction mixture was filtered, and the filtrate was
washed with saturated aqueous NaHCO3 and was extracted
with dichloromethane. The organic layer was dried over
sodium sulfate and concentrated under reduced pressure. The
residue was chromatographed on silica gel (hexane/EtOAc, 9:1)
to give 1.62 g (47%) of 18 as a colorless oil (small amounts of
pure R and â anomers were isolated for characterization).
2-F lu or o-9-(2,5-d id e oxy-D-er yt h r o-p e n t ofu r a n osyl)-
a d en in e (1a a n d 1b). To a solution of 14 (694 mg, 1.89 mmol)
in 15 mL of methanol was added triethylamine (0.75 mL). The
solution was stirred at room temperature for 48 h, and then a
small amount of silica gel was added to quench the reaction.
The solvent was removed under reduced pressure and the
residue was purified by chromatography on silica gel (di-
chloromethane/methanol, 9:1) to give 1a (210 mg, 44%) as a
white solid and then 1b (171 mg, 36%) as a white solid. The
1H NMR spectrum of the R anomer 1a was identical to that of
1a as prepared above (see Supporting Information). The
following data are for the the â-anomer 1b. 1H NMR (CD3OD,
300 MHz), δ: 8.17 (1H, s, H-8), 6.26 (t, 6,6 Hz H-1′), 4.4-4.2
(1H, m, H-4′), 4.1-4.0 (1H, m, H-3′), 2.9-2.8 (1H, m, H-2′b),
2.5-2.4 (1H, m, H-2′a), 1.37 (3H, d, 6.3 Hz, H-5). 13C NMR
1
R An om er . H NMR (CDCl3, 300 MHz), δ: 7.60-7.50 (2H,
m), 7.4-7.2 (3H, m), 5.80 (1H, dd, 7.8, 2.4 Hz, H-1), 5.21-
1
5.17 (1H, m, H-3), 4.78-4.51 (2H, dm, J HF ) 46.8 Hz, H-5),
4.48-4.35 (1H, dm, 2J HF ) 29.1 Hz, H-4), 2.9-2.7 (1H, m, H-2),
2.32-2.1 (1H, m, H-2′), 2.12 (3H, s, CH3CO). 13C NMR (CDCl3,
75 MHz), δ: 170.91 (CO), 135.57, 130.95, 129.05, 127.17, 87.78
(C1), 82.56 (1J CF ) 171.9 Hz, C5), 82.27 (2J CF ) 18.2 Hz, C4),
73.60 (3J CF ) 6.3 Hz, C3), 39.47 (C2), 21.10 (CH3CO).
1
(DMSO, 75 MHz), δ: 158.1 (d, J CF ) 202.1 Hz, C2), 157.6 (d,
3
3J CF ) 21.1 Hz, C6), 150.4 (d, J CF ) 20.5 Hz, C4), 139.9 (C8),
â An om er . 1H NMR (CDCl3, 300 MHz), δ 7.55-7.50 (2H,
m), 7.52-7.25 (3H, m), 5.58 (1H, dd, 9.6, 6.3 Hz, H-1), 5.25-
4
117.6 (d, J CF ) 4.0, C5), 82.92 (C1′), 82.54 (C4′), 74.58 (C3′),
38.40, (C5′), 19.07, (C2′). MS (CI, NH3), m/e: 254 (MH+).
1
5.22 (1H, m, H-3), 4.67-4.38 (2H, dm, J HF ) 46.8 Hz, H-5),
Meth yl 3-O-Acetyl-5-O-(ter t-bu tyldim eth ylsilyl)-2-deoxy-
r,â-D-er yth r o-p en tofu r a n osid e (16). Pyridine (5.78 g. 73.16
mmol), acetic anhydride (7.48 g, 73.16 mmol), and 4-dimethyl-
aminopyrdine (0.22 g, 1.8 mmol) were added to a solution of
methyl 5-O-(tert-butyldimethylsilyl)-2-deoxy-R,â-D-erythro-
pentofuranoside26 (9.66 g, 36.58 mmol) in anhydrous dichloro-
methane (100 mL) at 0 °C. The reaction mixture was stirred
for 6 h at room temperature under N2. The solution was
concentrated under reduced pressure and then was coevapo-
rated with toluene. The residue was purified by chromatog-
raphy on silica gel (hexane/EtOAc, 9:1) to give 10.9 g (98%) of
a mixture of R and â anomers of 16 as a colorless oil. 1H NMR
(CDCl3, 300 MHz), δ: 5.3-5.1 (2H, m, H-3 and H-1), 4.1-4.0
(1H, m, H-4), 4.0-3.6 (2H, m, H-5), 3.39 and 3.36 (3H, s, OMe),
2.2-2.0 (2H, m, H-2), 2.09 and 2.07 (3H, m, CH3CO), 0.89 (9H,
s, Si(C(CH3)3), 0.08 (6H, s, Si(CH3)2). 13C NMR (CDCl3, 75
MHz), δ: 170.91 and 170.26 (CH3CO), 105.48 and 105.37 (C1),
84.38 and 84.21 (C5), 75.39 and 74.64 (C4), 64.29 and 63.45
(C3), 55.17 and 54.97 (OMe), 39.38 and 39.09 (C2), 25.86 and
25.84 (SiC(CH3)3), 21.36 and 20.99 (CH3CO), 18.27 (SiC(CH3)3),
-5.4 (Si(CH3)2).
4.30-4.10 (1H, dm, 2J HF ) 24.6 Hz, H-4), 2.5-2.0 (2H, m, H-2),
2.08 (3H, s, CH3CO). 13C NMR (CDCl3, 75 MHz), δ: 170.58
(CO), 133.95, 131.96, 129.02, 127.63, 86.31 (C1), 84.14 (2J CF
) 19.4 Hz, C4), 82.78 (1J CF ) 172.5 Hz, C5), 75.04 (3J CF ) 4.5
Hz, C3), 39.49 (C2), 21.06 (CH3CO).
2-F lu or o-9-(3-O-a cet yl-2-d eoxy-5-flu or o-D-r-er yth r o-
p en tofu r a n osyl)a d en in e (19a ) a n d 2-Flu or o-9-(3-O-a cetyl-
2-d e oxy-5-flu or o-D-â-er yt h r o-p e n t ofu r a n osyl)a d e n in e
(19b). A suspension of 2-fluoroadenine (918 mg, 6.0 mmol)
and ammonium sulfate (793 mg, 6.0 mmol) in hexamethyl-
disilazane (6 mL) was heated at rigorous reflux under N2 until
the solution became clear. The excess of hexamethyldisilazane
was removed under reduced pressure. The residue and thio-
glycoside 18 (1.62 g, 6.0 mmol) were dissolved in 20 mL of dry
toluene under N2, and then 3 g of powered molecular sieves, 4
Å, was added. After being stirred at room temperature for 30
min, the reaction mixture was cooled to 0 °C and NBS (1.17 g,
6.6 mmol) was added. The resulting mixture was allowed to
warm to room temperature and stirred for 15 h. The addition
of aqueous Na2S2O3 was followed by filtration and then
extraction with dichloromethane. The organic layer was dried
over MgSO4 and concentrated under reduced pressure. The
residue was purified by chromatography on silica gel (100%
ethyl acetate) to give 218 mg (12%) of â-anomer 19b and then
317 mg (20%) of R-anomer 19a .
Meth yl 3-O-Acetyl-5-flu or o-2-deoxy-r,â-D-er yth r o-pen to-
fu r a n osid e (17). To a solution of furanoside 16 (4.57 g, 15.0
mmol) in dichloromethane (10 mL) was added bis[(2-meth-
oxyethyl)amino]sulfur trifluoride (6.64 g, 30.0 mmol) at room
temperature. After the reaction mixture was heated at reflux
for 16 h, it was cooled to 0 °C and methanol (5 mL) was added
dropwise. Triethylamine (5 mL) was added to neutralize the
resulting hydrofluoride. The solvent was removed under
reduced pressure, and the residue was purified by chroma-
tography on silica gel (hexane/EtOAc, 4:1) to give 2.48 g of 17
as a yellow oil (86% yield). Small amounts of pure R and â
anomers were isolated for characterization).
1
R An om er 19a . H NMR (CD3OD, 300 MHz), δ: 8.24 (1H,
s, H-8), 6.42 (1H, dd, J ) 7.2, 2.1 Hz, H-1′), 4.39 (1H, dt, J )
6.6, 1.8 Hz, H-3′), 4.9-4.5 (3H, m, H-4′ and H-5′), 3.0-2.8 (2H,
m, H-2′), 1.98 (3H, s, OAc). 19F NMR (CD3OD, 282 MHz), δ:
-53.01(s, F-2), -234.60 (dt, J ) 30.7, 46.0 Hz, F-5′). FAB-
MS, m/e: 314.2 (MH+).
1
â An om er 19b. H NMR (CD3OD, 300 MHz), δ: 8.18 (1H,
s, H-8), 6.38 (1H, t, J ) 6.0 Hz H-1′), 5.52-4.48 (1H, m, H-3′),
4.69 (2H, dm, 1J HF ) 47.1 Hz, H-5′), 4.34 (1H, dm, 2J HF ) 27.6
Hz, H-4′), 3.0-2.6 (2H, m, H-2′), 2.13 (3H, s, OAc). 19F NMR
(CD3OD, 282 MHz), δ: -52.81 (s, F-2), -232.52 (dt, 27.4, 45.7
Hz, F-5′). FAB-MS, m/e: 314.2 (MH+).
1
R An om er . H NMR (CDCl3, 300 MHz) 5.25-5.19 (1H, m,
H-3), 5.18-5.15 (1H, m, H-1), 4.65-4.35 (2H, dm, 1J HF ) 46.8
2
Hz, H-5), 4.30-4.20 (1H, dm, J HF ) 18.0 Hz, H-4), 3.37 (3H,
s, OMe), 2.42-2.10 (2H, m, H-2), 2.07 (3H, s, CH3CO). 13C
NMR (CDCl3, 75 MHz),δ 170.76 (CO), 105.69 (C1), 83.76 (1J CF
) 171.8 Hz, C5), 82.58 (2J CF ) 19.9 Hz, C4), 74.26 (3J CF ) 5.7
2-F lu or o-9-(2-d eoxy-5-flu or o-D-r-er yth r o-p en tofu r a n o-
syl)a d en in e (2a ). To a suspension of 19a (230 mg, 0.734

Synthesis of Fluoroadenosines
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 5 1213
(8) De´saubry, L.; J ohnson, R. A. Adenine nucleoside 3′-tetraphos-
phates are novel and potent inhibitors of adenylyl cyclases. J .
Biol. Chem. 1998, 273, 24972-24977.
(9) For a review, see the following. Dessauer, C. W.; Tesmer, J . J .
G.; Sprang, S. R.; Gilman, A. G. The interactions of adenylate
cyclases with P-site inhibitors. TIPS 1999, 20, 205-210.
(10) Tesmer, J . J . G.; Sunahara, R. K.; J ohnson, R. A.; Gosselin, G.;
Gilman, A. G.; Sprang, S. R. Two-metal-ion catalysis in adenylyl
cyclases. Science 1999, 285, 756-760.
(11) Tesmer, J . J . G.; Dessauer, C. W.; Sunahara, R. K.; Murray, L.
D.; J ohnson, R. A.; Gilman, A. G.; Sprang, S. R. Molecular basis
for P-site inhibition of adenylyl cyclase. Biochemistry 2000, 39,
14464-14471.
(12) Nimit, Y.; Law, J .; Daly, J . W. Binding of 2′,5′-dideoxyadenosine
to brain membranes: comparison to P-site inhibition of adenyl-
ate cyclase. Biochem. Pharmacol. 1982, 31, 3279-3287.
(13) De´saubry, L. D.; J ohnson, R. A. Unpublished observations, 1996.
(14) Montgomery, J . A.; Hewson, K. Synthesis of potential anticancer
agents. XX. 2-Fluoropurines. J . Am. Chem. Soc. 1960, 82, 463-
468.
mmol) in a mixture of methanol and water (30 mL, 10:1) was
added triethylamine (1.5 mL). The mixture was stirred at room
temperature for 24 h, and then a small amount of silica gel
was added to quench the reaction. The solvent was removed
under reduced pressure. The residue was purified by chroma-
1
tography to give 2a (140 mg, 70%) as a white solid. H NMR
(Me2SO-d6, 300 MHz), δ: 8.36 (1H, s, H-8), 7.84 (br, NH2), 6.24
(1H, dd, J ) 7.2, 3.9 Hz H-1′), 5.70 (1H, d, J ) 3.0 Hz, OH),
1
4.51 (2H, dm, J HF ) 47.1 Hz, H-5′), 4.4-4.3 (2H, m, H-3′ and
H-4′), 2.8-2.6 (1H, m, H-2′), 2.5-2.4 (1H, m, H-2′). 13C NMR
1
(Me2SO-d6, 75 MHz), δ: 158.7 (d, J CF ) 202.6 Hz, C2), 157.7
3
3
(d, J CF ) 21.1 Hz, C6), 150.4 (d, J CF ) 20.0 Hz, C4), 139.9
(C8), 117.4 (C5), 85.3 (d, 17.6 Hz, C4′), 83.5 (C1′), 82.9 (d, 167.9
Hz, C5′), 69.6 (d, J ) 6.3 Hz, C3′), 39.5 (C2′). FAB-MS, m/e:
272.1 (MH+). HRMS (FAB) Calcd for (MH+, C10H12F2N5O2):
272.0959. Found: 272.0959.
2-F lu or o-9-(2-d eoxy-5-flu or o-D-â-er yth r o-p en tofu r a n o-
syl)a d en in e (2b). The same procedure as above used 19b to
give 2b (54 mg, 69%) as a white solid. 1H NMR (DMSO-d6,
300 MHz), δ: 8.25 (1H, s, H-8), 7.86 (br, NH2), 6.27 (1H, t, J
) 6.8 Hz H-1′), 5.53 (1H, d, J ) 4.5 Hz, OH), 4.58 (2H, dm,
1J HF ) 47.7 Hz, H-5′), 4.5-4.4 (1H, m, H-3′), 4.03 (1H, dm, J
) 22.2 Hz, H-4′), 2.8-2.6 (1H, m, H-2′), 2.5-2.4 (1H, m, H-2′).
(15) Krolikiewicz, K.; Vorbru¨ggen, H. The synthesis of 2-fluoropurine
nucleosides. Nucleosides Nucleotides 1994, 13, 673-678.
(16) Montgomery, J . A.; Hewson, K. Nucleosides of 2-fluoroadenine.
J . Med. Chem. 1969, 12, 498-504.
(17) Hassan, A. E. A.; Shortnacy-Fowler, A. T.; Montgomery, J . A.;
Secrist, J . A., III. A convenient synthesis of 2′-deoxy-2-fluoro-
adenosine, a potential prodrug for suicide gene therapy. Nucleo-
sides, Nucleotides, Nucleic Acids 2000, 19, 559-565.
(18) Sugimura, H.; Osumi, K.; Kodaka, Y.; Sujino, K. Stereoselective
synthesis of 2′-deoxy-â-D-threo-pentofuranosyl nucleosides by the
NBS-promoted coupling reaction of thioglycosides with silylated
heterocyclic bases. J . Org. Chem. 1994, 59, 7653-7660.
(19) Soll, R. M.; Seitz, S. P. Synthetic studies on boron containing
antibiotics. Tetrahedron Lett. 1987, 28, 5457-5460.
(20) Sujino, K.; Sugimura, H. Stereocontrolled synthesis of â-D-2′-
deoxyribonucleotides by intramolecular glycosylation. Chem.
Lett. 1993, 1187-1190.
(21) Robins, M. J .; Wood, S. G.; Dalley, N. K.; Herdewijn, P.;
Balzarini, J .; De Clercq, E. Nucleic acid related compounds. 57.
Synthesis, X-ray crystal structure, lipophilic partition properties,
and antiretroviral activities of anomeric 3′-azido-2′,3′-dideoxy-
2,6-diaminopurine ribosides. J . Med. Chem. 1989, 32, 1763-
1768.
(22) Smart, B. E. Characteristics of C-F Systems. In Organofluorine
Chemistry, Principles and Commercial Applications; Banks, R.
E., Smart, B. E., Tatlow, J . C., Eds.; Plenum Press: New York,
1994; pp 57-88.
(23) Chen, L.; Dumas, D. P.; Wong, C.-H. Deoxyribose-5-phosphate
aldolase as a catalyst in asymmetric aldol condensation. J . Am.
Chem. Soc. 1992, 114, 741-748.
(24) Parker, W. B.; Allan, P. W.; Hassan, A. E. A.; Secrist, J . A., III;
Sorscher, E. J .; Waud, W. R. Antitumor activity of 2-fluoro-2′-
deoxyadenosine against tumors that express Escherichia coli
purine nucleoside phosphorylase. Cancer Gene Ther. 2003, 10,
23-29.
(25) Ye, S.; Rezende, M. M.; Deng, W.-P.; Kirk, K. L. Convenient
synthesis of 2′-deoxy-2-fluoroadenosine from 2-fluoroadenine.
Nucleosides, Nucleotides Nucleic Acids 2003, 22, 1899-
1905.
1
13C NMR (DMSO-d6, 75 MHz), δ: 158.6 (d, J CF ) 202.6 Hz,
3
3
C2), 157.7 (d, J CF ) 21.0 Hz, C6), 150.4 (d, J CF ) 20.0 Hz,
2
C4), 139.6 (C8), 117.6 (C5), 84.9 (d, J CF ) 18.2 Hz, C4′), 83.4
1
3
(C1′), 82.9 (d, J CF ) 167.4 Hz, C5′), 69.7 (d, J CF ) 6.3 Hz,
C3′), 39.7 (C2′).
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails for the synthesis and data for characterization of com-
pounds 6, 8-13, and 1a ; NOESY spectra for 1a , 1b, 2a , and
2b; and HMBC spectra of 1b and 2b. This material is available
free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) For a recent review, see the following. Patel, T. B.; Du, Z.; Pierre,
S.; Cartin, L.; Scholich, K. Molecular biological approaches to
unravel adenylyl cyclase signaling and function. Gene 2001, 269,
13-25.
(2) Sattin, A.; Rall, T. W. The effect of adenosine and adenine
nucleotides on the cyclic adenosine 3′, 5′-phosphate content of
guinea pig cerebral cortex slices. Mol. Pharmacol. 1970, 6, 13-
23.
(3) Londos, C.; Wolff, J . Two distinct adenosine-sensitive sites on
adenylate cyclase. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 5482-
5486.
(4) J ohnson, R. A.; Yeung, S.-M. H.; Stu¨bner, D.; Bushfield, M.;
Shoshani, I. Cation and structural requirements for P site-
mediated inhibition of adenylate cyclase. Mol. Pharmacol. 1989,
35, 681-688.
(5) Bushfield, M.; Shoshani, I.; J ohnson, R. A. Tissue levels, source,
and regulation of 3′AMP: An intracellular inhibitor of adenylyl
cyclases. Mol. Pharmacol. 1990, 38, 848-853.
(26) Ichikawa, Y.; Kubota, H.; Fujita, K.; Okauchi, T.; Narasaka, K.
Stereoselective â-C- and â-S-glycosylation of 2-deoxyribofuranose
derivatives controlled by the 3-hydroxy protective group. Bull.
Chem. Soc. J pn. 1989, 62, 845-852.
(6) J ohnson, R. A.; Shoshani, I. Kinetics of “P”-site-mediated inhibi-
tion of adenylyl cyclase and the requirements for substrate. J .
Biol. Chem. 1990, 265, 11595-11600.
(7) De´saubry, L.; Shoshani, I.; J ohnson, R. A. Inhibition of adenylyl
cyclase by a family of newly synthesized adenine nucleoside 3′-
polyphosphates. J . Biol. Chem. 1996, 271, 14028-14034.
J M0303599