Unsaturated fluoroketonucleosides as anticancer agents: the synthesis and biological activity of 5-fluoro-1-(3,4-di-deoxy-3-fluoro-6-O-trityl-beta-d-glycero-hex-3-eno-pyranos-2- ulosyl) uracil.

Article abstract of DOI:10.1081/NCN-120006828

Direct oxidation of 5-fluoro-1-(4-O-acetyl-3-deoxy-3-fluoro-6-O-trityl-beta-D-glucopyranosyl) uracil 9 led to the title compound 10 after a beta-elimination reaction. The formation of the hydrate of ketone 10 due to the highly electronegative fluorine ato

Full text of DOI:10.1081/NCN-120006828

Nucleosides, Nucleotides and Nucleic Acids
ISSN: 1525-7770 (Print) 1532-2335 (Online) Journal homepage: http://www.tandfonline.com/loi/lncn20
UNSATURATED FLUOROKETONUCLEOSIDES
AS ANTICANCER AGENTS: THE SYNTHESIS AND
BIOLOGICAL ACTIVITY OF 5-FLUORO-1-(3,4-DI-
DEOXY-3-FLUORO-6-O-TRITYL-β-D-GLYCERO-HEX-3-
ENO-PYRANOS-2-ULOSYL) URACIL
M. J. Egron , B. I. Dorange , K. Antonakis , J. Herscovici & A. P. Ollapally
To cite this article: M. J. Egron , B. I. Dorange , K. Antonakis , J. Herscovici & A. P. Ollapally (2002)
UNSATURATED FLUOROKETONUCLEOSIDES AS ANTICANCER AGENTS: THE SYNTHESIS AND
BIOLOGICAL ACTIVITY OF 5-FLUORO-1-(3,4-DI-DEOXY-3-FLUORO-6-O-TRITYL-β-D-GLYCERO-
HEX-3-ENO-PYRANOS-2-ULOSYL) URACIL, Nucleosides, Nucleotides and Nucleic Acids, 21:4-5,
327-334, DOI: 10.1081/NCN-120006828
To link to this article: http://dx.doi.org/10.1081/NCN-120006828
Published online: 31 Aug 2006.
Submit your article to this journal
Article views: 10
View related articles
Citing articles: 3 View citing articles
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=lncn20
Download by: [University of Lethbridge]
Date: 06 November 2015, At: 02:00

MARCEL DEKKER, INC. • 270 MADISON AVENUE • NEW YORK, NY 10016
©2002 Marcel Dekker, Inc. All rights reserved. This material may not be used or reproduced in any form without the express written permission of Marcel Dekker, Inc.
NUCLEOSIDES, NUCLEOTIDES & NUCLEIC ACIDS, 21/4&5), 327±334 /2002)
UNSATURATED FLUOROKETONUCLEOSIDES
AS ANTICANCER AGENTS: THE SYNTHESIS
AND BIOLOGICAL ACTIVITY OF 5-FLUORO-1-
ꢀ3,4-DI-DEOXY-3-FLUORO-6-O-TRITYL-b-
D-GLYCERO-HEX-3-ENO-PYRANOS-2-ULOSYL)
URACIL
M. J. Egron,1 B. I. Dorange,2 K. Antonakis,1 J. Herscovici,1
and A. P. Ollapally2
1Ecole Nationale Supzrieure de Chimie, CNRS-UMR 7001;
11, Rue P.et M. Curie-75005 Paris, France
2Department of Chemistry, Florida A&M University,
Tallahassee, FL 32307, USA
ABSTRACT
Direct oxidation of 5-¯uoro-1- /4-O-acetyl-3-deoxy-3-¯uoro-6-O-trityl-b-
D-glucopyranosyl) uracil 9 led to the title compound 10 after a b-elim-
ination reaction. The formation of the hydrate of ketone 10 due to the
highly electronegative ¯uorine atom in the a position to the carbonyl group,
prompted us to carry out a comparative study of di€erent methods of
oxidation and to de®ne the best strategy for the synthesis of such molecules.
Results of in vitro and in vivo biological evaluations are reported.
INTRODUCTION
The introduction of a ¯uorine atom in a molecule is known to confer on
it low toxicity and increased permeability through the cell membrane by
increasing the lipophilicity. Although substitution of a ¯uorine atom for a
hydroxyl group is sterically conservative, the high strength of the C-F bond
327
Copyright # 2002 by Marcel Dekker, Inc.
www.dekker.com

328
EGRON ET AL.
may hinder metabolism pathways and thus increase the e€ective lifetime of
the active molecule.[1±4] The syntheses of several ¯uoroketonucleosides pos-
sessing signi®cant antineoplastic activity and immunosuppressive e€ect have
been reported.[5±7] These studies on various normal and tumor cells showed
that the introduction of a ¯uorine atom in 30-position of an unsaturated
40-ketonucleoside leads to an important di€erence in the toxicity toward
neoplastic and normal cells, [6] the cytotoxicity was shown to be signi®cantly
higher for the neoplastic than the normal cells. In addition, the comparison
of the antineoplastic activity and the immunosuppressive e€ect of this ¯uoro-
40-ketonucleoside with an analog devoid of the ¯uorine atom showed a higher
antineoplastic activity and a lower immunosuppressive e€ect /for steady state
cells) for the ¯uorocompounds.[5±6] To further con®rm this observation with
other analogs, the synthesis and the in vitro antitumor activity of ¯uoro
analogs of unsaturated 20-ketonucleosides has recently been reported.[7]
These observations prompted us to synthesize a keto ¯uoronucleoside
possessing a ¯uoro group in the 30-position and the keto group in the 20-
position. Syntheses of unsaturated 40-ketonucleosides containing ¯uoro
groups in the 30-position possessing either a purine or a pyrimidine base have
been reported.[5, 6] It appears that the crucial problem of these syntheses is the
oxidation of OH groups in the sugar moiety[8] and the nature of the base.
Although the synthesis of the ¯uoro 40-ketonucleosides containing purine as
well as pyrimidine bases could be carried out without many diculties, that
of ¯uoro-20-ketonucleosides of pyrimidines has proved to be delicate, giving
unforeseen results. In a recent note,[7] we reported brie¯y the synthesis of an
unsaturated 30-¯uoro-20-ketonucleoside of 5-¯uorouracil. We now report a
complete comparative study of the di€erent routes to this biologically
important molecule.
RESULTS AND DISCUSSION
The free 30-¯uoro derivative of ¯uorouracil 3 /see the Scheme) has been
prepared from 1,2,4,6-tetra-O-acetyl-3-deoxy-3-¯uoroglucose 1 by a recently
described method.[7] The molecule was protected in 40,60-position with a
isopropylidene group, using 2,2-dimethoxypropane in dry DMF. This par-
tially protected compound 4 was obtained pure as a crystalline material but
the yields were relatively low. The isopropylidene derivative 4 should be
isolated from the organic layer by extraction. However, this method gave
better results than that involving methoxyethylidene as the blocking
group. The protection of the 20-hydroxyl was performed using t-butyl-
chlorodimethylsilane with imidazole to give the 20-O-t-butyldimethylsilyl
derivative 5 in more than 90% yield, which could be deacetalated directly
with IR 120 resin in methanol to give 6 in excellent yield. Some other
methods of acid hydrolysis of acetals have been examined. In particular the

UNSATURATED FLUOROKETONUCLEOSIDES
329
Scheme.
method using catalytic amounts of p-toluenesulfonic acid led to a mixture
containing the starting material. The speci®c protection of 60-hydroxyl group
was accomplished with triphenylmethyl chloride in the presence of dime-
thylaminopyridine /DMAP) in pyridine. The 60-trityl derivative 7 was easily
isolated pure in 60% yield. The following step consisted of the acetylation of
40-hydroxyl with acetic anhydride in pyridine in the presence of DMAP. This
easily gave the desired 8 in 95% yield. The conditions for the speci®c removal
of the 20-t-butyldimethylsilyl group were established after a careful study of
the action of tetrabutylammonium ¯uoride trihydrate /TBAF) on 8 in THF.
After 30 min in dry conditions and under nitrogen, the mixture was extracted

330
EGRON ET AL.
with CH2Cl2 to give pure 9 in excellent yields. The oxidation of nucleoside
structures such as 9 is a crucial problem depending on many factors dicult
to predict.[8±11] Di€erent oxidation methods have been studied. The use of
PDC=3 A molecular sieves[11] gave pure ketonucleoside 10 in poor yields
/10±12%). After a chromatographic study, we were able to ®nd the best
conditions and improved the yields to only up to 20% by stopping the
reaction after 2 h. The low yields are probably due to the long reaction time
and possibly the formation of small quantities of gem-diol. The chromato-
graphic study showed that the ketonucleoside appeared after one hour of
reaction and no degradation of the compound appeared before 2 h. After this
time a little degradation appeared but also the ®ltration of the mixture on
silicagel might have contributed to a further degradation.
An alternative approach consisted of the use of the DMSO=acetic
anhydride at room temperature for 48 h. This method led to an approxi-
mately 1:1 mixture of the desired 10 and its gem-diol. The same method at
40ꢀC for 6 h gave the ketonucleoside and its gem-diol in a 7:3 ratio, but
diculties in the separation of the derivatives led ®nally to low yields of pure
10 /15%) characterized by NMR and physical data.
Another method based on the use of oxalyl chloride, appeared to be
promising since we obtained pure 10 in about 20% yields, without resorting
to tedious puri®cations. We believe that improvements in the technical
conditions of this approach, such as temperature and time of reaction, may
signi®cantly improve the yields.
The in vitro biological activity of ¯uoroketonucleoside 10 was tested
on HT29 human colon carcinoma cells and KB human epidermoid carci-
noma cells in comparison with 9 both possessing a 60-O-trityl group. This
study clearly con®rmed previous results, that the introduction of an enone
system in a nucleosidic structure, increases the anti-tumor properties of the
molecule.
IC50 /lM) Values for Compounds 9 and 10 for KB and HT29
Cells
KB
10
HT29
9
10
15
5
0.7
The in-vivo tests performed on mice bearing RXF 393 renal tumor,
showed moderate activity for high tolerate doses /7.50 mg=kg=dose), the
compound 10 was well tolerated and no loss of body weight was observed in
comparison to the control mice. Further investigations with various tumors
will be undertaken.

UNSATURATED FLUOROKETONUCLEOSIDES
331
EXPERIMENTAL
General methods: Melting points /uncorrected) were recorded using a
Mel-Temp apparatus. TLC was performed on Silica Gel 60 /240±400 mesh,
Merck). NMR spectra were recorded at room temperature with a Bruker
300 MSL spectrometer with internal Me4Si for 1H and internal C6F6 for 19F;
the positions in carbohydrate moieties are designated by primes. Me2SO was
distilled over CaH2 under reduced pressure and stored over 3A molecular
sieves. Oxalyl chloride was freshly distilled under N2 and kept in a sealed
bottle.
5-Fluoro-1-ꢀ3-deoxy-3-¯uoro-2, 4, 6-tri-O-acetyl-b-D-glucopyranosyl)uracil
ꢀ2). In a ¯ask, 3.55 g /27.3 mmol) of 5-¯uorouracil and a small quantity of
saccharin were dried by dissolving in benzene and concentrating, then 53 mL of
dried acetonitrile and 19.1 mL /2 equiv) of hexamethyldisilazane were added.
The solution was heated at 80ꢀC for 30 min under re¯ux, then concentrated.
7.85g /22.4 mmol) of the tetraacetylated ¯uoro- D-glucose 1 in anhydrous
acetonitrile /2 mL) was added to the previous ¯ask containing silylated 5-
¯uorouracil. The mixture was kept at 0ꢀC and then was added a solution of
1.44 mL /1.2 equiv) of trimethylsilyl tri¯uoromethanesulfonate in 2 mL anhy-
drous acetonitrile. It was heated at 85ꢀC for 3 h, then cooled to ambient tem-
perature and stirred overnight. After evaporation of the solvent, cold water
/0ꢀC) was added and the mixture was neutralized with aq. NaHCO3 /10%), at
0ꢀC and then extracted with CH2Cl2 /3650 mL). This product was puri®ed by
¯ash chromatography, eluting with ethyl acetate=heptane, /6:4 v=v).
Crystallization from ethanol, gave 2 in 92% yield: mp and elemental
analysis agreed with the data previously reported.[7]
5-Fluoro-1-ꢀ3-deoxy-3-¯uoro-b-D-glucopyranosyl)uracil ꢀ3). To a solu-
tion of 8.5g /20.23 mmol) of 2 in 80 mL of anhydrous methanol was added
8 mL of 2 N MeONa. After 1.5h, the solution was neutralized with IR 120
resin and concentrated to give 5.64 g of 3 as an oil /yield 95%), which was
used directly in the next step.
5-Fluoro-1-ꢀ3-deoxy-3-¯uoro-4,6-O-isopropylidene-b-D-glucopyranosyl)-
uracil ꢀ4). To 7.69 g /21 mmol) of 3 dissolved in a mixture of 19 mL of dry 2,
2-dimethoxypropane and 50 mL of DMF were added 20 drops of conc. H2SO4.
This mixture was stirred for 5h and then neutralized with NaHCO 3. The
mixture was ®ltered and the liquid phase was concentrated under high vacuum
to remove the DMF. Pure crystalline 4 was obtained from a mixture of heptane
and ethanol, in 60% yield: mp 152±154ꢀC. Anal. Calcd for C13H16O8N2F2
/1=2C7H16): C, 47.59; H, 5.77; N, 6.73; Found: C, 47.92; H, 5.61; N, 7.16.

332
EGRON ET AL.
5-Fluoro-1-ꢀ2-O-t-butyldimethylsilyl-3-deoxy-3-¯uoro-4, 6-O-isopropyli-
dene-b-D-glucopyranosyl)uracil ꢀ5). A mixture of 5.16 g /14.1 mmol) of 4,
4.13 g. /27.53 mmol, 1.2 equiv) of t-butylchlorodimethylsilane and 2.39 g /2.5
equiv) of imidazole, were dried with benzene. DMF /9.6 mL) were then added
and the solution was stirred for 14 h at 35ꢀC. The mixture was concentrated
under high vacuum and the residue extracted with CH2Cl2. Flash chroma-
tography /heptane=ethyl acetate 6:4 v=v), gave 6.09 g of 5 in 90% yield as an
oil which was used directly in the next step.
5-Fluoro-1-ꢀ2-O-t-butyldimethylsilyl-3-deoxy-3-¯uoro-b-D-glucopyr-
anosyl)-uracil ꢀ6). Compound 5 was deacetalated by heating at re¯ux for 1.5h
in methanol containing IR 120 resin. The mixture was then neutralized with
IR 45resin. Pure 6 was isolated as a solid with CH2Cl2 in 90% yield: mp
128±132ꢀC.
Anal. Calcd for C16H26O6SiN2F2: C, 47.05; H, 6.37; N, 6.94: Found: C,
46.88; H, 6.54; N, 6.86.
5-Fluoro-1-ꢀ2-O-t-butyldimethylsilyl-3-deoxy-3-¯uoro-6-O-trityl-b-D-
gluco-pyranosyl)uracil ꢀ7). To 5g /11.11 mmol) of 6 was dissolved in 55 mL
of anhydrous pyridine was added, 4.046 g of triphenylmethylchloride and a
catalytic amount of DMAP. The solution was stirred for 24 h at room
temperature and then extracted with CH2Cl2 and concentrated. The trity-
lated product was puri®ed /60% yield) by ¯ash chromatography /ethyl
acetate=heptane 6:4 v=v) and crystallized from ethanol: mp 224±226ꢀC. Anal.
Calcd for C35H40O6SiN2F2: C, 64.61; H, 6.15; N, 4.30. Found: C, 64.49; H,
1
6.18; N, 4.24. HNMR /CDCl3): d 7.49±7.20 /m, 15H, 3C H5), 5.65 /d, 1H,
6
J1 ,2 ˆ 8.81 Hz, H10), 4.45/dt, 1H,
JF,3 ˆ 51.88 Hz, JF,4 ˆ 16.89 Hz,
0
0
0
0
0
0
0
0
0
0
J3 ,4 ˆ 9.31 Hz, H3 ), 3.90 /m, 1H J4 ,5 ˆ 3.94 Hz, H4 ), 3.63 /m, 2H,
J5 ,6 b ˆ 3.73 Hz, H20, H50), 3.46 /2H, J6 a,6 b, H60).
0
0
0
0
5-Fluoro-1-ꢀ4-O-acetyl-2-O-t-butyldimethylsilyl-3-deoxy-3-¯uoro-6-O-tri-
tyl-b-D-glucopyranosyl)uracil ꢀ8). To 7.97 g /12.25mmol) of 7 in 100 mL of
CH2Cl2 were added 2 mL /1.2 equiv) of pyridine, 2.3 mL /1.2 equiv) of acetic
anhydride and a small quantity of DMAP. After 30 min the solvents were
evaporated in vacuo. Puri®cation by ¯ash chromatography /heptane=ethyl
acetate 6:4 v=v) and crystallization from isopropylalcohol gave pure 8 in 95%
yield: mp 212±214ꢀC. Anal Calcd for 8 C37H42O7N2F2H2O: C, 62.41; H, 5.03;
N, 4.69; Found: C, 62.07; H, 5.35; N, 4.45. 1HNMR /CDCl3): d 7.49±7.20 /m,
0
0
0
0
0
15H, 3 C6H5), 5.66 /d, 1H, J1 ,2 ˆ 8.6 Hz, H1 ), 5.28 /m, 1H, J4 ,5 ˆ 10.5Hz,
0
0
0
0
0
0
0
0
JF,4 ˆ 22.04 Hz, J3 ,4 ˆ 10 Hz, H4 ), 4.48 /t, 1H, JF,3 ,2 ˆ 51.48 Hz, J3 ,2
ˆ
8.78 Hz, H30), 3.77 /m 2H, J
ˆ 4.64 Hz, H20, H50), 3.16 /dm, 2H,
0
0
5 ,6 b
0
0
J6a ,6 b ˆ 10.86 Hz).

UNSATURATED FLUOROKETONUCLEOSIDES
333
5-Fluoro-1-ꢀ4-O-acetyl-3-deoxy-3-¯uoro-6-O-trityl-b-D-glucopyranosyl)-
uracil ꢀ9). Tetrabutylammonium ¯uoride trihydrate was used to cleave the
20-tert-butyldimethylsilyl group. This was done using 692 mg /1.00 mmol) of
8 and 1.039 g /3.3 mmol) of dry tetrabutylammonium ¯uoride trihydrate
/TBAF) in anhydrous THF under nitrogen. After 30 min, the mixture was
concentrated in vacuo and extracted with CH2Cl2. The compound 9 was
obtained in 92% yield as a powder from diisopropyl ether: mp 133±136ꢀC.
Anal. Calcd for CH31H28C7N2F2H2O: C, 62.41; H, 5.03; N, 4.69; Found: C,
62.07; H, 5.35; N, 4.45. 1HNMR /CDCl3): d 7.49±7.20 /m, 15H, 3C6H5), 5.76
0
0
0
0
0
0
/d, 1H, J1 ,2 ˆ 8.6 Hz, H1 ), 5.38 /m, 1H, JF,4 ˆ 21.4 Hz, J4 ,5 ˆ 10.45Hz,
0
0
0
0
0
0
0
J3 ,4 ˆ 9.67 Hz, H4 ), 4.65/dt, 1H, JF,4 ˆ 21.45Hz, JF,3 ˆ 52.46 Hz, J3 ,2
ˆ
ˆ
8.9 Hz, H30), 3.86 /m, 2H, J5 ,6 b ˆ 3.97 Hz, H20, H50), 3.25/m, 2H, J6 a,6 b
0
0
0
0
10.77 Hz, H60), 1.87/1s, 3H, COCH3)
5-Fluoro-1-ꢀ3,4-dideoxy-3-¯uoro-6-O-trityl-b-D-glycero-hex-3-enopyr-
anos-2-ulosyl)uracil ꢀ10). Method 1 ꢀwith PDC=molecular sieves): 3 g
/5.26 mmol) of 9 and 3.9 g of PDC were dried with dry benzene. Then 100 mL
of dry CH2Cl2, freshly activated 3 A molecular sieves and 5drops of acetic
acid were added and the mixture was stirred during 5h at room temperature.
The mixture was then ®ltered through silica gel G, washed twice with
CH2Cl2. The combined ®ltrate and washings were concentrated in vacuo.
Then it was puri®ed by Flash chromatography: The yields were low /10±
12%). A chromatographic study showed that after 2 h the keto compound
was formed in about 60% yield with 35% of the starting material remaining.
Finally stopping the reaction after 2 h we could isolate the 10 in 20% yield:
Anal. Calcd for C29H22O5N2F2: C, 67.31; H, 4.45; N, 5.41; Found: C, 67.24;
1
H, 5.01; N, 4.70. HNMR /CDCl3): d 7.50±7.25 /m, 15H, 3 C6H5), 6.71 /m,
0
0
0
0
0
1H, JF,4 ˆ 11.4 Hz, J4 ,5 ˆ 1.73 Hz, H4 ), 6.34 /s, 1H, H-1 ), 4.88 /dd, 1H,
J5 ,6 a ˆ 4.89 Hz, J5 ,6 b ˆ 5.25 Hz, H50), 3.43 /dq, 2H, J 6 a,6 b ˆ 9.43 Hz, H60).
Method 2 ꢀwith DMSO/Acetic anhydride): The reaction of 0.11 mmol of
nucleoside 9 in DMSO /4 mL) and acetic anhydride /0.25mL), after two days
at room temperature, gave a 1:1 mixture of the ketonucleoside and the
corresponding gem-diol detected by chromatography. This is due to the
formation of hydrogen bonding favored by the method. Improvement:
0.11 mmol of 9, 0.4 mL of DMSO and 0.3 mL of acetic anhydride stirred for
6 h at 40ꢀC, led to a 7:3 mixture of the keto nucleoside 10 and its gem-diol.
The reaction was followed by NMR analysis. The gem-diol was obtained in
semi-crystalline form. Anal. Calcd for C29H24O6N2F2: C, 65.06; H, 4.49; N,
5.24; Found: C, 64.56; H, 5.01; N, 4.57.
0
0
0
0
0
0
0
Method 3 ꢀwith oxalyl chloride): This method was carried out with 578
mg /1 mmole) of nucleoside 9.
In ®rst ¯ask, 3.6 mL of dried CH2Cl2, and 91.68 mL /1.05mmol) of
freshly distilled oxalyl chloride were added and kept at À70ꢀC. In a second
¯ask: 1 mL of dried CH2Cl2, and 156 mL /2.2 mmol) of dried DMSO were

334
EGRON ET AL.
injected. The solution was kept at 40ꢀC for 5min then the contents of the ®rst
¯ask, kept at À70ꢀC, was injected very slowly into the second ¯ask. In a third
¯ask, 1 mmol of 9 was dissolved in 1 mL of CH2Cl2 and kept at À40ꢀC for
15min then the contents of this ¯ask were injected into the second ¯ask.
After 30 min the mixture was neutralized with 557 mL /3.7 equiv) of trie-
thylamine and removing the DMSO and the remaining triethylamine yielded
10 /characterized by NMR) in 15±20% yield. This method was selected for
the oxidation step by virtue of the ease with which the ®nal ¯uoroenone
compound could be isolated.
ACKNOWLEDGMENT
This work was supported by the National Institutes of Health through
the MBRS program GM08111 and by the 3M Foundation, St Paul, Minnesota
for basic research. We thank also the French Association pour la Recherche
sur le Cancer /ARC) for ®nancial support. The authors appreciate the
excellent secretarial assistance rendered by Mrs. Glory Brown of the College
of Pharmacy and Pharmaceutical Sciences.
REFERENCES
1. Van Aerschot, A.; Herdewijn, P.; Balzarini, J.; De Clercq, E. J. Med. Chem.
1989, 32, 1743±1749.
2. Blandin, M.; Son, T.-D.; Catlin, J.; Guschlbauer, J. Biochim. Biophys. Acta
1974, 361, 249±253.
3. Morton, G.; Lancaster, J.; Van Lear, G.; Fulmor, W.; Meyer, W. J. Amer.
Chem. Soc. 1969, 91, 1535±1537.
4. Lipnick, R.; Fissekis, J. Biochm. Biophys Acta 1980, 608, 96±102.
5. Egron, M.J.; Leclercq, F.; Antonakis, K.; Bennani-Baiti, I.; Frayssinet, C. Car-
bohydr. Res. 1993, 248, 143±150.
6. Leclercq, F.; Egron, M.J.; Antonakis, K.; Bennani-Baiti, I.; Frayssinet, C. Car-
bohydr. Res. 1992, 228, 95±102.
7. Ollapally, A.P.; Dorange, B.I.; Whitaker, Jr., T.L. Nucleosides and Nucleotides
1999, 18 /4&5), 703±705.
8. Crich, D.; Neelamkavil, S. J. Am. Chem. Soc. 2001, 123, 7449±7450.
9. Alaoui, M.; Egron, M.J.; Bessodes, M.; Antonakis, K. and Chouroulinkov, I.
Eur. J. Med. Chem. 1987, 22, 305±310.
10. Komiotis, D.; Delatre, S.; Holt, S.; Ollapally, A.P.; Balzarini, J.; De Clercq, E.;
Masuaki, I. Nucleosides and Nucleotides 1991, 110, 431±455.
11. Herscovici, J.; Antonakis, K. J. Chem. Soc. Chem. Commun. 1979, 12,
561±563.
Received November 1, 2001
Accepted April 30, 2002