Nucleic acid related compounds. 116. Nonaqueous diazotization of aminopurine nucleosides. Mechanistic considerations and efficient procedures with tert-butyl nitrite or sodium nitrite

Article abstract of DOI:10.1021/jo0204101

Nonaqueous diazotization-dediazoniation of two types of aminopurine nucleoside derivatives has been investigated. Treatment of 9-(2,3,5-tri-O-acetyl-β-D-ribofuranosyl)-2-amino-6-chloropurine (1) with SbCl₃/CH₂Cl₂ was examined with benzyltriethylammonium (BTEA) chloride as a soluble halide source and tert-butyl nitrite (TBN) or sodium nitrite as the diazotization reagent. Optimized yields (>80%) of the 2,6-dichloropurine derivative were obtained with SbCl₃. Combinations with SbBr₃/CH₂Br₂ gave the 2-bromo-6-chloropurine product (>60%), and SbI₃/CH₂I₂/THF gave the 2-iodo-6-chloropurine derivative (>45%). Antimony trihalide catalysis was highly beneficial. Mixed combinations (SbX₃/CH₂X′₂; X/X′ = Bt/Cl) gave mixtures of 2-(bromo, chloro, and hydro)-6- chloropurine derivatives that were dependent on reaction conditions. Addition of iodoacetic acid (IAA) resulted in diversion of purine radical species into a 2-iodo-6-chloropurine derivative with commensurate loss of other radical-derived products. This allowed evaluation of the efficiency of SbX₃-promoted cation-derived dediazoniations relative to radical-derived reactions. Efficient conversions of adenosine, 2′-deoxyadenosine, and related adenine nucleosides into 6-halopurine derivatives of current interest were developed with analogous combinations.

Full text of DOI:10.1021/jo0204101

Nu cleic Acid Rela ted Com p ou n d s. 116. Non a qu eou s Dia zotiza tion
of Am in op u r in e Nu cleosid es. Mech a n istic Con sid er a tion s a n d
Efficien t P r oced u r es w ith ter t-Bu tyl Nitr ite or Sod iu m Nitr ite†,1
Paula Francom,‡ Zlatko J aneba, Susumu Shibuya,§ and Morris J . Robins*
Department of Chemistry and Biochemistry, Brigham Young University, Provo, Utah 84602-5700, and
Department of Chemistry, The University of Alberta, Edmonton, Alberta, Canada
morris_robins@byu.edu
Received J une 14, 2002
Nonaqueous diazotization-dediazoniation of two types of aminopurine nucleoside derivatives has
been investigated. Treatment of 9-(2,3,5-tri-O-acetyl-â-D-ribofuranosyl)-2-amino-6-chloropurine (1)
with SbCl3/CH2Cl2 was examined with benzyltriethylammonium (BTEA) chloride as a soluble halide
source and tert-butyl nitrite (TBN) or sodium nitrite as the diazotization reagent. Optimized yields
(>80%) of the 2,6-dichloropurine derivative were obtained with SbCl3. Combinations with SbBr3/
CH2Br2 gave the 2-bromo-6-chloropurine product (>60%), and SbI3/CH2I2/THF gave the 2-iodo-6-
chloropurine derivative (>45%). Antimony trihalide catalysis was highly beneficial. Mixed
combinations (SbX3/CH2X′2; X/X′ ) Br/Cl) gave mixtures of 2-(bromo, chloro, and hydro)-6-
chloropurine derivatives that were dependent on reaction conditions. Addition of iodoacetic acid
(IAA) resulted in diversion of purine radical species into a 2-iodo-6-chloropurine derivative with
commensurate loss of other radical-derived products. This allowed evaluation of the efficiency of
SbX3-promoted cation-derived dediazoniations relative to radical-derived reactions. Efficient
conversions of adenosine, 2′-deoxyadenosine, and related adenine nucleosides into 6-halopurine
derivatives of current interest were developed with analogous combinations.
In tr od u ction
means for manipulation of base substituents. Such ap-
proaches have recently been used for synthesis of ana-
logues of nitrous acid-mediated DNA cross-links,5 DNA-
carcinogen adducts,6 and preparation of oligonucleotides
with site-specific base modifications.7 Continuing interest
in the modification of purines is reflected in recent
reports of synthesis and screening of purine libraries and
nucleoside analogues.4g,i
The versatility of halopurines2 as intermediates for
nucleoside synthesis was recognized by Fischer a century
ago.2a Nucleophilic aromatic substitution reactions3 and
palladium-catalyzed cross-coupling processes4 with 2-, 6-,
and 8-halopurines and nucleosides provide convenient
* Address correspondence to this author at Brigham Young Uni-
versity. Fax: (801) 422-0153.
Diazotization of electron-deficient heterocyclic amines
is problematic,8-12 and coupling of radical species derived
from putative diazo intermediates with arenes gave
† Dedicated to Professor Dr. Wolfgang Pfleiderer on the occasion of
his 75th birthday.
‡ Present address: Biota, Inc., Carlsbad, CA.
§ Present address: Yamasa Corporation, Choshi, J apan.
(1) Paper 115: Robins, M. J .; Miles, R. W.; Samano, M. C.; Kaspar,
R. L. J . Org. Chem. 2001, 66, 8204-8210.
(5) Harwood, E. A.; Sigurdsson, S. T.; Edfeldt, N. B. F.; Reid, B. R.;
Hopkins, P. B. J . Am. Chem. Soc. 1999, 121, 5081-5082.
(6) (a) Lee, H.; Hinz, M.; Stezowski, J . J .; Harvey, R. G. Tetrahedron
Lett. 1990, 31, 6773-6776. (b) Tsarouhtsis, D.; Kuchimanchi, S.;
DeCorte, B. L.; Harris, C. M.; Harris, T. M. J . Am. Chem. Soc. 1995,
117, 11013-11014. (c) DeCorte, B. L.; Tsarouhtsis, D.; Kuchimanchi,
S.; Cooper, M. D.; Horton, P.; Harris, C. M.; Harris, T. M. Chem. Res.
Toxicol. 1996, 9, 630-637.
(7) (a) Seela, F.; Mertens, R.; Kazimierczuk, Z. Helv. Chim. Acta
1992, 75, 2298-2306. (b) Seela, F.; Chen, Y.; Bindig, U.; Kazimierczuk,
Z. Helv. Chim. Acta 1994, 77, 194-202. (c) Lakshman, M. K.; Zajc, B.
Nucleosides Nucleotides 1996, 15, 1029-1039. (d) Adib, A.; Potier, P.
F.; Doronina, S.; Huc, I.; Behr, J .-P. Tetrahedron Lett. 1997, 38, 2989-
2992.
(8) Goodman, L. In Basic Principles in Nucleic Acid Chemistry; Ts’o,
P. O. P., Ed.; Academic: New York, 1974; Vol. 1, pp 93-208.
(9) Srivastava, P. C.; Robins, R. K.; Meyer, R. B., J r. In Chemistry
of Nucleosides and Nucleotides; Townsend, L. B., Ed.; Plenum: New
York, 1988; Vol. 1, pp 113-281.
(10) Bunton, C. A.; Wolfe, B. B. J . Am. Chem. Soc. 1974, 96, 7747-
7752.
(11) McKenzie, T. C.; Epstein, J . W. J . Org. Chem. 1982, 47, 4881-
4884.
(12) Press, J . B.; Eudy, N. H.; Morton, G. O. J . Org. Chem. 1983,
48, 4605-4611.
(2) (a) Fischer, E. Ber. 1899, 32, 435-504. (b) Elion, G. B.; Hitchings,
G. H. J . Am. Chem. Soc. 1956, 78, 3508-3510.
(3) See examples and references cited in: (a) Robins, M. J .; Basom,
G. L. Can. J . Chem. 1973, 51, 3161-3169. (b) van der Wenden, E. M.;
von Frijtag Drabbe Ku¨nzel, J . K.; Mathot, R. A. A.; Danhof, M.;
IJ zerman, A. P.; Soudijn, W. J . Med. Chem. 1995, 38, 4000-4006. (c)
Gupta, V.; Kool, E. T. Chem. Commun. 1997, 1425-1426. (d) Ve´liz, E.
A.; Beal P. A. J . Org. Chem. 2001, 66, 8592-8598.
(4) See examples and references cited in: (a) Matsuda, A.; Shinozaki,
M.; Yamaguchi, T.; Homma, H.; Nomato, R.; Miyasaka, T.; Watanabe,
Y.; Abiru, T. J . Med. Chem. 1992, 35, 241-252. (b) Van Aerschot, A.
A.; Mamos, P.; Weyns, N. J .; Ikeda, S.; De Clercq, E.; Herdewijn, P. A.
J . Med. Chem. 1993, 36, 2938-2942. (c) Langli, G.; Gundersen, L.-L.;
Rise, F. Tetrahedron 1996, 52, 5625-5638. (d) Lakshman, M. K.;
Keeler, J . C.; Hilmer, J . H.; Martin, J . Q. J . Am. Chem. Soc. 1999,
121, 6090-6091. (e) Hocek, M.; Holy´, A.; Votruba, I.; Dvorˇa´kova´, H.
Collect. Czech. Chem. Commun. 2001, 66, 483-499. (f) Ve´liz, E. A.;
Stephens, O. M.; Beal, P. A. Org. Lett. 2001, 3, 2969-2972. (g) Brun,
V.; Legraverend, M.; Grierson, D. S. Tetrahedron Lett. 2001, 42, 8169-
8171. (h) Lakshman, M. K.; Hilmer, J . H.; Martin, J . Q.; Keeler, J . C.;
Dinh, Y. Q. V.; Ngassa, F. N.; Russon, L. M. J . Am. Chem. Soc. 2001,
123, 7779-7787. (i) Ding, S.; Gray, N. S.; Ding, Q.; Schultz, P. G.
Tetrahedron Lett. 2001, 42, 8751-8755.
10.1021/jo0204101 CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/29/2002
6788
J . Org. Chem. 2002, 67, 6788-6796

Nonaqueous Diazotization of Aminopurine Nucleosides
F IGURE 1. Proposed intermediate structure B14b for con-
verion of a putative purine 2-diazonium ion A into the oxazine
ring of an isolated oxanosine derivative17 C.
F IGURE 2. Conversion of a purine-6-cation species D into
the 6-chloropurine product F via a solvent-derived chloronium
intermediate18,19 E.
variable yields of aryl derivatives.11,12 Nonaqueous dia-
zotization is useful for incorporation of halogens at the
(2 or 6)-positions of purine nucleosides,4b,13 but purinedia-
zonium ions and their fragmentations have not been
investigated in detail. Experimental10 and theoretical
studies14 indicate that purinediazonium species have very
short lifetimes and lose N2 much more readily than
common arenediazonium ions.15 In fact, purinediazonium
ions have not been observed directly.10,14 Glaser et al.
proposed a ring-opened ketene-carbodiimide species14b
(Figure 1) to rationalize formation of the 2′-deoxyox-
anosine that results from nitrous acid-mediated dedia-
zoniation16 of 2′-deoxyguanosine. Isolation of this rear-
rangement product17 and theoretical analyses14b indicate
that multiple decomposition pathways are possible for
transient purinediazonium species.
F IGURE 3. Generation of an aryl radical I via electron
transfer (G f H) from an anion (X-) with an appropriate redox
potential to a (photoexcited) diazonium ion.15,22
Nair et al.21 studied dediazoniation with aminopurine
derivatives. They employed pentyl nitrite in THF or
halocarbon solvents with both photoactivation and ther-
mal activation for hydro- or halo-dediazoniation of 6-ami-
nopurine derivatives.21a,b Purinyl radicals were detected
(ESR) during photolysis of 6-iodo-9-ethylpurine,21c and
it was postulated that homolysis of purinediazonium
intermediates was followed by abstraction of hydrogen
or halogen atoms from solvent.21b However, photolytic
cleavage of a carbon-iodine bond to form a purinyl
radical is quite different from photochemical activation
for dediazoniation. Evidence exists that photoinitiated
dediazoniation proceeds via electron transfer from a
counterion to a photoexcited arenediazonium ion within
a charge-transfer complex, followed by homolytic cleavage
of an arenediazenyl radical15,22 (Figure 3). Photoirradia-
tion was not used in later procedures for hydro- and halo-
dediazoniation.21d,e We and others23 have observed ∼50%
of protected inosine derivatives (hydroxy-dediazoniation
characteristic of carbocationic processes) upon halo-
dediazoniation21b of acetylated adenosine derivatives with
alkyl nitrites in halocarbon solvents, and product analysis
cannot exclude halide abstraction by a cationic mecha-
nism18 such as that illustrated in Figure 2.
White and co-workers presented arguments for pre-
dominance of heterolytic decomposition pathways18,19
resulting from protonation of 9-methylpurine-6-diazoate
salts19 with acetic acid in CHCl3 or deuteriobenzene.
Products of 6-(acetoxy, chloro, hydro, and hydroxy)-
dediazoniation16 were detected with CHCl3, and 6-(ac-
etoxy, hydroxy, and phenyl)purine derivatives were
formed in C6D6.19 Chloro-dediazoniation might occur via
formation of a purine carbocation, which might abstract
chloride via a chloronium ion species18 (Figure 2) rather
than by radical mechanisms. Product analysis cannot
distinguish between homolytic and heterolytic pro-
cesses,15,20 and radical abstraction of chlorine from solvent
might also have occurred. Hydro-dediazoniation is thought
to occur by homolytic processes, except in strongly
alkaline media or with sodium borohydride.20 In solvents
of low electron-donating ability (CHCl3 or C6D6) and in
the absence of an exogenous source of electrons, high
levels of homolytic dediazoniation products are not
expected.15,20
(13) (a) Robins, M. J .; Uznanski, B. Can. J . Chem. 1981, 59, 2608-
2611. (b) Matsuda, A.; Satoh, K.; Tanaka, H.; Miyasaka, T. Synthesis
1984, 963-965.
We observed mixtures of 2-(chloro and fluoro)purine
nucleoside derivatives upon treatment of protected 2-ami-
no precursors with TBN and BF3‚OEt2 in CHCl3 or CH2-
Cl2.13a Montgomery and co-workers24 had reported 2-(chlo-
ro and fluoro) substitution with diazotization of a
2-aminopurine nucleoside derivative in 48% HBF4/H2O//
CHCl3. They detected minor quantities of a 2-chloro
(14) (a) Glaser, R.; Lewis, M. Org. Lett. 1999, 1, 273-276. (b) Glaser,
R.; Rayat, S.; Lewis, M.; Son, M.-S.; Meyer, S. J . Am. Chem. Soc. 1999,
121, 6108-6119.
(15) (a) Zollinger, H. Angew. Chem., Int. Ed. Engl. 1978, 17, 141-
150. (b) Zollinger H. Diazo Chemistry I. Aromatic and Heteroaromatic
Compounds; VCH: New York, 1994.
(16) Bunnett’s nomenclature15 is used throughout the present paper.
Replacement of a diazonium species by another group is termed
dediazoniation regardless of mechanism. The name of the entering
group is added as a prefix (i.e., chloro-dediazoniation, hydro-dediazo-
niation, etc.).
(21) (a) Nair, V.; Richardson, S. G. Tetrahedron Lett. 1979, 1181-
1184. (b) Nair, V.; Richardson, S. G. J . Org. Chem. 1980, 45, 3969-
3974. (c) Nair, V.; Richardson, S. G.; Coffman, R. E. J . Org. Chem.
1982, 47, 4520-4524. (d) Nair, V.; Richardson, S. G. Synthesis 1982,
670-672. (e) Nair, V.; Chamberlain, S. D. Synthesis 1984, 401-403.
(22) Becker, H. G. O.; Israel, G.; Oertel, U.; Vetter, H.-U. J . Prakt.
Chem. 1985, 327, 399-410.
(23) Gao, X.; J ones, R. A. J . Am. Chem. Soc. 1987, 109, 1275-1278.
(24) Montgomery, J . A.; Clayton, S. D.; Shortnacy, A. T. J . Hetero-
cycl. Chem. 1979, 16, 157-160.
(17) Suzuki, T.; Yamaoka, R.; Nishi, M.; Ide, H.; Makino, K. J . Am.
Chem. Soc. 1996, 118, 2515-2516.
(18) White, E. H.; McGirk, R. H.; Aufdermarsh, C. A., J r.; Tiwari,
H. P.; Todd, M. J . J . Am. Chem. Soc. 1973, 95, 8107-8113.
(19) Song, F.; St. Hilaire, V. R.; White, E. H. Org. Lett. 1999, 1,
1957-1959 and references therein.
(20) Galli, C. Chem. Rev. 1988, 88, 765-792.
J . Org. Chem, Vol. 67, No. 19, 2002 6789

Francom et al.
SCHEME 1
product upon diazotization of a 2-aminopurine derivative
in concentrated HCl/H2O,25 and Gerster and Robins
synthesized 2-chloropurine nucleoside derivatives from
2-amino precursors with concentrated HCl/H2O.26 Sha-
piro and Pohl had isolated 2-nitroinosine from treatment
of guanosine with nitrous acid and excess nitrite in an
acetate buffer.27
Electron-deficient nucleobase analogues should give
diazonium ions with high reduction potentials, which
would facilitate electron transfer and homolytic cleav-
age.15,20 However, the low electron-donating potential of
halohydrocarbon solvents would favor heterolytic frag-
mentation mechanisms.15,20 Interplay between these fac-
tors could make halo-dediazoniation of purine nucleosides
particularly sensitive to changes in reaction conditions.
Ionic catalysis of aryl halo-dediazoniation is known, and
in situ generation of NOBr or NOCl from Br- or Cl- can
occur.15 Electron transfer also might facilitate dediazo-
niation. Galli has noted that electron transfer to ben-
zenediazonium salts can occur from substrates with
appropriate redox potentials, and electron transfer from
iodide is well-known.28 Increased yields of 2-iodopurine
nucleosides were obtained from iodo-dediazoniation of
2-aminopurine nucleosides with isoamyl nitrite in CH3-
CN when both I- and CuI were added.4a Bromide is a
borderline case, and might transfer an electron to an
arenediazonium ion with a compatible reduction poten-
tial.28,29 Nitrite also can transfer an electron to electron-
deficient arenediazonium ions to give aryl radicals and
a nitrogen dioxide radical.15,20 Effects of different sources
of halide and nitrite on product distributions resulting
from nonaqueous dediazoniation of purines have not been
studied systematically.
Resu lts a n d Discu ssion
Halopurine nucleosides are valuable synthetic inter-
mediates.2-9 Conflicting opinions on the SNAr reactivity
of 6-(bromo versus chloro)purine compounds have ap-
peared recently.3d,4d We undertook a systematic study of
reagents and reaction conditions to provide reliable and
convenient procedures for introduction of halogens at the
2- and/or 6-positions of purine nucleosides. There is
uncertainty regarding the intermediacy of cation versus
radical (and other) intermediates in dediazoniation reac-
tions of electron-deficient rings such as purines.14,15,19-21
Iodoacetic acid (IAA) has been reported33 to be a selective
trapping agent for radical species, and detection of aryl
iodide products provides evidence for the transient pres-
ence of aryl radicals. Concomitant decreases in yields of
products otherwise derived from aryl radicals occur as
IAA diverts these species into iodo products.33
We first examined product distributions resulting from
dediazoniation of 9-(2,3,5-tri-O-acetyl-â-D-ribofuranosyl)-
2-amino-6-chloropurine34,35 (1) with tert-butylnitrite (TBN)
and benzyltriethylammonium (BTEA) bromide at ambi-
ent temperature in CH2Cl2 for 3 h with or without IAA
(Scheme 1). Ratios of 2-[chloro (2), bromo (3), and iodo
(4)]purine nucleosides were determined by 1H NMR
analysis of purified dediazoniation mixtures [H8 signals
at δ 8.27 (2), 8.25 (3), and 8.18 (4)]. Products were
separated by TLC (radial, RTLC, Chromatotron) or
preparative thin-layer chromatography (PTLC). Yields
were quantitated by UV spectroscopy, and structures
were confirmed by MS. The hydro-dediazoniation product
5 was isolated (PTLC or RTLC) and quantitated (UV),
and its structure was verified (MS and NMR). Results
are summarized in Table 1, entries 1 and 2.
Antimony trichloride has been employed for stibono-
dediazoniation of diazonium salts.30 Our introduction of
SbCl3 and SbBr3 for nonaqueous halo-dediazoniation of
2-aminopurine nucleoside derivatives provided high yields
of 2-halo products, and no 2-(hydro, hydroxy, or stibono)-
dediazoniation byproducts were detected.13a,31 These pro-
cedures have been used for preparation of 2-halopurine9,31
and related nucleosides,32 and a recent application of our
SbBr3 catalysis for syntheses of some 2-bromonucleoside
analogues has been reported.5 We observed both 2-(bromo
and chloro)-dediazoniation products upon treatment of
acetylated 2-amino-6-chloropurine nucleosides with SbX3/
CH2X′2 mixtures.13a We now report studies on dediazo-
niation of (2 and 6)-aminopurine nucleosides with dif-
ferent antimony trihalides, nitrite and halide salts, and
solvents at different temperatures.
No 5 was detected with IAA present, which indicated
that 5 had been derived from a purinyl radical. Yields of
3 were markedly diminished with IAA present, which
suggested that purinyl radicals had been diverted from
3 into iodo compound 4. The 3 (24%) observed with IAA
indicated that heterolytic bromo-dediazoniation occurred,
and the amounts of chloro-dediazoniation product 2 were
similar with or without IAA (11 or 8%). It has been noted
that radicals usually abstract hydrogen much more
rapidly than chlorine.19 Our observation that IAA ef-
fectively interrupted hydrogen atom abstraction without
(25) Montgomery, J . A.; Hewson, K. J . Am. Chem. Soc. 1960, 82,
463-468.
(26) Gerster, J . F.; Robins, R. K. J . Org. Chem. 1966, 31, 3258-
3262.
(27) Shapiro, R.; Pohl, S. H. Biochemistry 1968, 7, 448-455.
(28) Galli, C. J . Chem. Soc., Perkin Trans. 2 1981, 1459-1461.
(29) Barbero, M.; Degani, I.; Dughera, S.; Fochi, R. J . Org. Chem.
1999, 64, 3448-3453.
(30) Doak, G. O.; Steinman, H. G. J . Am. Chem. Soc. 1946, 68,
1987-1989.
(33) Wassmundt, F. W.; Kiesman, W. F. J . Org. Chem. 1997, 62,
(31) Robins, M. J .; Wnuk, S. F. In Encyclopedia of Reagents for
Organic Synthesis; Paquette, L. A., Ed.; Wiley: Chichester, 1995; Vol
1, pp 205-206.
8304-8308.
(34) Gerster. J . F.; J ones, J . W.; Robins, R. K. J . Org. Chem. 1963,
28, 945-948.
(32) Ramasamy, K.; Ugarkar, B. G.; McKernan, P. A.; Robins, R.
K.; Revankar, G. R. J . Med. Chem. 1986, 29, 2231-2235.
(35) Robins, M. J .; Uznanski, B. Can. J . Chem. 1981, 59, 2601-
2607.
6790 J . Org. Chem., Vol. 67, No. 19, 2002

Nonaqueous Diazotization of Aminopurine Nucleosides
TABLE 1. P r od u cts fr om Ded ia zon ia tion Rea ction s of 1
products (yield %)
entry
(reagents/solvents)
(BTEA-Br/TBN/CH2Cl2)
(BTEA-Br/TBN/CH2Cl2)
(BTEA-Cl/TBN/CH2Br2)
(BTEA-Cl/TBN/CH2Br2)
(SbBr3/TBN/CH2Cl2)
(SbBr3/TBN/CH2Cl2)
(SbCl3/TBN/CH2Br2)
IAA
2
3
4
5
1
2
3
4
5
6
7
8
9
10
11
12
13a
14a
15a
16a
-
+
-
+
-
+
-
+
-
-
-
-
-
-
-
-
8
58
24
30
21
30
25
35
32
44
40
28
27
4
11
22
<1
40
43
48
46
4
29
32
33
>80
52
28
2
8
5
(SbCl3/TBN/CH2Br2)
4
(BTEA-Br/SbBr3/TBN/CH2Cl2)
(BTEA-Br/SbBr3/NaNO2/DCA/CH2Cl2)
(BTEA-Cl/SbCl3/TBN/CH2Br2)
(BTEA-Cl/SbCl3/NaNO2/DCA/CH2Br2)
(BTEA-Cl/SbCl3/NaNO2/DCA/CH2Cl2)
(BTEA-Br/SbBr3/NaNO2/DCA/CH2Br2)
(SbI3/TBN/CH2I2/THF)
1
<1
2
1
>60
>45
(NaNO2/DCA/THF)
>80
a
See the Results and Discussion and Experimental Section for conditions for preparative scale reactions.
significant alteration in yields of the chloro-dediazonia-
tion product is consistent with White’s suggestion that
overall abstraction of Cl- from solvent might occur by a
heterolytic process18 (Figure 2). Our results are compat-
ible with competing homolytic and heterolytic dediazo-
niation at ambient temperature.
(entries 5-8). Only minor quantities of the 2-iodo product
4 were generated with IAA/SbBr3 (2%, entry 6) or IAA/
SbCl3 (4%, entry 8), in contrast with the reactions with
BTEA-Br (52%, entry 2) or BTEA-Cl (28% entry 4). Ratios
of 2/3 (SbX3 to solvent-derived halo-dediazoniation prod-
ucts) were within experimental error for entries 5 and
6, or 7 and 8. With SbX3 present, purinediazonium ion
decompositions depart from radical pathways.
Because SbCl3 induces DNA damage and is toxic,36 we
evaluated BTEA-Cl37 versus SbCl3 for chloro-dediazonia-
tion of 1 with TBN in CH2Cl2. One equivalent of BTEA-
Cl or SbCl3 was added to 1 and TBN in CH2Cl2 at
ambient temperature. Reactions proceeded rapidly, and
starting material was absent after 10 min. However,
chloro-dediazoniation of 1 with BTEA-Cl resulted in a
significantly reduced yield of 2 (46%) relative to the
reaction with SbCl3 (70%), and byproduct 5 (7%) was
formed with BTEA-Cl. Compounds 2 and 5 are difficult
to separate. Treatment of 1 with excess TBN and BTEA-
Cl (10 equiv) in CH2Cl2 at -10 °C increased the yield of
2 from ∼50 to 62%, but did not eliminate formation of 5
(∼4%). Addition of 20 mol % of SbCl3 to 1/TBN/BTEA-Cl
(1 equiv each) in CH2Cl2 at -10 °C did not diminish
formation of 5. Equimolar SbCl3 and BTEA-Cl resulted
in lower yields of 2 (42%), and did not suppress formation
of 5. Clearly, BTEA-Cl was not a viable substitute for
SbCl3.
Product distributions resulting from dediazioniation of
1 with CH2Br2 as solvent with or without IAA were next
determined (Table 1, entries 3 and 4). Marked reductions
in yields of 2 occurred with IAA. Diminished yields of
solvent-derived 3 (30 f 21%) and 5 (8 f 5%) occurred
with IAA, and these results are similar to those for
dediazoniation of 1 with CH2Cl2. Stabilities of the 2-[chlo-
ro (2), bromo (3), or iodo (4)]purine nucleosides were
evaluated by their treatment with TBN (20 equiv) and
BTEA-Cl (2 equiv) in CH2Br2 at ambient temperature for
3 h with or without IAA (2 equiv). Only minor decompo-
sition was observed (recovered 2, 3, or 4, 90-99%). No
apparent differences were seen in reactions with or
without IAA, and no nucleobase or nucleoside byproducts
were detected. However, treatment of 2 or 3 with TBN
(20 equiv) plus BTEA-Br and IAA (2 equiv each) in CH2-
Cl2 at reflux for 3 h resulted in significant glycosyl bond
cleavage and other decomposition processes (∼33% loss
of either 2 or 3).
The effect of ionic halide on dediazonation at elevated
temperature was probed by treatment of 1 with excess
TBN plus IAA and BTEA-Br (2 equiv each) in CH2Cl2 at
reflux for 15 min. Combined yields of 57% of 2-[chloro
(2, 2%), bromo (3, 21%), and iodo (4, 34%)]purine dedia-
zoniation products were obtained, but 5 was not detected.
Competition between homolytic and heterolytic dediazo-
niation mechanisms apparently occurs also at ∼40 °C.
We observed that halo-dediazoniation of 1 with TBN
and SbX3 at reduced temperature (-10 °C) eliminated
formation of 2-(hydro and hydroxy)-dediazoniation prod-
ucts. Halo-dediazoniation of 1 with TBN and SbBr3/CH2-
Cl2 or SbCl3/CH2Br2 at ambient temperature with IAA
(2 equiv) proceeded rapidly [1 was not detected (TLC)
after 15-30 min]. The results with SbX3 (Table 1, entries
5-8) differed markedly from those with BTEA-X salts
(entries 1-4). Yields of radical-derived products were
drastically reduced with SbX3, and the hydro-dediazo-
niation compound 5 was not detected with or without IAA
Heterolytic dediazoniation is known to be favored
under acidic conditions,15,39 and acetic acid has been used
for enhancement of diazotization of heteroaromatic
amines.40 We next probed addition of organic acids and
alternative sources of nitrite. Sodium nitrite is inexpen-
sive and a safer alternative to the potentially hazardous41
TBN. Experiments were performed with addition of an
equimolar amount of acetic, chloroacetic, dichloroacetic,
or trichloroacetic acid (or Dowex 50 [H+] resin) to 1,
(36) Huang, H.; Shu, S. C.; Shih, J . H.; Kuo, C. J .; Chiu, I. D.
Toxicology 1998, 129, 113-123.
(37) BTEA-X salts are easier to use than Et4NX35 or Me4NX.
BTEA-X has been used previously in halodedizoniation procedures.38
(38) Lee, J . G.; Cha, H. T. Tetrahedron Lett. 1992, 33, 3167-3168.
(39) Broxton, T. J .; Bunnett, J . F.; Paik, C. H. J . Org. Chem. 1977,
42, 643-649.
(40) Butler, R. N. Chem. Rev. 1975, 75, 241-257.
(41) Lopez, F. Chem. Eng. News 1992, 70 (51), 2.
J . Org. Chem, Vol. 67, No. 19, 2002 6791

Francom et al.
SCHEME 2
excess NaNO2, and BTEA-Cl (10 equiv) in CH2Cl2. Only
the experiment with dichloroacetic acid (DCA) gave
significant amounts of 2.
Treatment of 1 with either TBN (20 equiv) or excess
powdered NaNO2, BTEA-Br and SbBr3 (1 equiv each) in
CH2Cl2 (or BTEA-Cl and SbCl3 in CH2Br2), and DCA (1.5
equiv) at ambient temperature for 3 h was examined
(Table 1, entries 9-12). Unfavorable interaction(s) among
TBN/SbBr3/BTEA-Br/CH2Cl2 (entry 9) resulted in re-
duced yields of 2 + 3 (48% combined), but the yield of 5
was reduced to ∼1%. Addition of DCA to such a mixture
(DCA/NaNO2/SbBr3/BTEA-Br/CH2Cl2) (entry 10) gave 2
+ 3 (69% combined) with <1% of 5. Analogous treatment
of 1 with SbCl3/BTEA-Cl/CH2Br2 gave the same combined
yields (60%) and proportions of 2 and 3 with either TBN
or DCA/NaNO2 (entries 11 and 12).
earth (Celite) to the reaction mixtures followed by rapid
filtration of the suspensions with a short column of silica
gel layered with activated charcoal and Celite, and yields
comparable to our prior results13a were obtained. Opti-
mized conditions for chlorodediazoniation of 1 (Table 1,
entry 13) were applied to bromo-dediazoniation with
analogous bromine reagents. Brief examination of SbBr3
and BTEA-Br ratios, solvent volume, and temperature
showed that good yields of 3 (>60%) were obtained with
1 (0.025 M), DCA (1.5 equiv), excess NaNO2, and SbBr3/
BTEA-Br (1 equiv each) in CH2Br2 (entry 14). SbI3/TBN/
CH2I2/THF (45%, dark-colored reaction mixture, not
optimized) was used for conversion of 1 f 4.
Effects of reagent proportions, temperature, and con-
centration were evaluated with 1. DCA (1.5 equiv),
NaNO2 (20 equiv), and BTEA-Cl (1 equiv) in CH2Cl2 at
ambient temperature for 3 h were held constant. Opti-
mum yields were obtained with ∼1 equiv of SbCl3. Yields
of 2 declined with >1 equiv, and 5 was detected in
addition to diminished yields of 2 with <0.5 equiv of
SbCl3. Amounts of DCA used impacted quantities of 5
produced. Treatment of 1 with NaNO2 (20 equiv) plus
SbCl3 and BTEA-Cl (1 equiv each) in CH2Cl2 at ambient
temperature for 6 h with variable amounts of DCA (0-2
equiv) gave optimum yields of 2 (>80%) with ∼1.5 equiv
of DCA. Lower quantities of DCA (0.1-1 equiv) resulted
in increasing amounts of 5. Best yields were obtained
with 2 equiv of DCA in ∼3.5 h, 1.5 equiv in 3.5-6 h, or
1-1.5 equiv in g12 h (reactions did not proceed with <0.1
equiv). Treatment of 1 with excess NaNO2 plus SbCl3 and
BTEA-Cl (1 equiv each) in CH2Cl2 at ambient tempera-
ture with increasing amounts of HOAc (1.5-20 equiv)
resulted in negligible suppression of the formation of 5,
and required extended reaction times. The acidic proper-
ties of DCA were most effective for reactions with 1.
Inversion of the order of addition of SbCl3 and DCA
produced dramatic changes. Addition of DCA to the
reaction mixture prior to the addition of SbCl3 resulted
in reduced yields of 2, extended reaction times, a shift
in the optimum ratio from 1 (39%) to 2 equiv (87%) of
SbCl3, formation of 5 at the higher ratio of SbCl3, and
an increase in the quantities of polar decomposition
products. Addition of DCA after the addition of SbCl3
gave higher yields of 2 in shorter times with lower ratios
of SbX3, and resulted in less byproduct formation.
Treatment of 1 with excess NaNO2 plus SbCl3 and
BTEA-Cl (1 equiv each) and DCA (1.5 equiv) in CH2Cl2
gave 2 at ambient temperature (82%), 0 °C (77%), and
-15 to -20 °C (61%). No 5 was detected at these three
temperatures. The same reagent ratios at concentrations
of 1/CH2Cl2 of 0.033 M (1.0 mmol/30 mL), 0.025 M (1.0
mmol/40 mL), and 0.017 M (1.0 mmol/60 mL) gave
crystalline 2 (56, 74, and 64%, respectively). Nucleoside
concentrations of 0.025 M in CH2X2 at ambient temper-
ature were subsequently employed. It is clear that SbX3
has a pronounced positive effect on these halo-dediazo-
niations. Yields of halopurine products were increased
substantially, and byproduct formation was suppressed.
SbX3 compounds hydrolyze to form white precipitates
that produce emulsions during extraction workup.13a This
problem was eliminated by addition of diatomaceous
Effects of SbX3 on product distributions from halo-
dediazoniations at C6 of purine nucleosides were evalu-
ated (Scheme 2). Treatment of 2′,3′,5′-tri-O-acetylade-
nosine (6) with excess TBN/CH2Cl2 and 2 equiv each of
SbBr3 and IAA (or without IAA) at ambient temperature
gave products which were separated (TLC), quantitated
(UV), and verified by NMR and MS. Ratios of the
6-[chloro (5), bromo (7), and iodo (8)] compounds in
dedizoniation mixtures were determined by 1H NMR [H8
singlets at δ 8.76 (5), 8.71 (7), and 8.62 (8)]. The hydroxy-
dediazoniation product [2′,3′,5′-tri-O-acetylinosine (10)]
also was quantitated (Table 2, entries 1 and 2), but no
hydro-dediazoniation product [2′,3′,5′-tri-O-acetylnebu-
larine (9)] was detected. Ratios of 7/5 were similar with
or without IAA, as were yields of 10 (6 or 4%, respec-
tively). Formation of 8 and 9 was suppressed to very low
levels with SbBr3, as was observed with analogous
reactions of 1 at C2 (Scheme 1, Table 1). Treatment of 6
with excess TBN/CH2Cl2, BTEA-Br (1 equiv) with or
without SbBr3 (1 equiv) at ambient temperature resulted
in ∼20-fold enhancement of bromo- (7, 79%) relative to
chloro- (5, 5%) dediazoniation with SbBr3 and BTEA-Br
or with BTEA-Br only [7 (61%), 5 (3%)] (Table 2, entries
3 and 4). A higher yield of 7 (79 versus 61%) was obtained
with the SbBr3/BTEA-Br combination, but hydroxy-
dediazoniation to give 10 (6%) also occurred. Treatment
of 6 with excess TBN/CH2Br2 and SbBr3 (2 equiv) at
reflux significantly increased yields of 7, and suppressed
formation of even traces of 9. Analogous treatment of 11
or 13 also gave good yields of the protected arabino, 12,
or 3′-deoxy, 14, 6-bromopurine nucleosides (Scheme 3).
NaNO2/DCA also was an effective reagent combination.
Treatment of 6 with SbBr3 and BTEA-Br (1 equiv each),
NaNO2 (20 equiv), and DCA (1.5 equiv) in CH2Br2 at
ambient temperature for 18 h gave 7 (69%), whereas
dediazoniation of 6 with TBN/CH2Br2 and SbBr3 (2 equiv)
was not complete after 72 h. Addition of DCA (1.5 equiv)
6792 J . Org. Chem., Vol. 67, No. 19, 2002

Nonaqueous Diazotization of Aminopurine Nucleosides
TABLE 2. P r od u cts fr om Ded ia zon ia tion Rea ction s of 6
products (yield %)
entry
(reagents/solvents)
(SbBr3/TBN/CH2Cl2)
(SbBr3/TBN/CH2Cl2)
(BTEA-Br/TBN/CH2Cl2)
IAA
5
7
8
10
1
2
3
4
-
+
-
-
-
-
-
20
11
3
32
17
61
79
>70
>60
4
6
7
6
b
b
b
1
(BTEA-Br/SbBr3/TBN/CH2Cl2)
(BTEA-Br/SbBr3/NaNO2/DCA/CH2Br2)
(SbBr3/TBN/CH2Br2/∆)
5
5a
6a
7a
(SbI3/TBN/CH2I2/THF/∆)
>45
a
b
See the Results and Discussion and Experimental Section for conditions for preparative scale reactions. Trace quantities (<10%).
SCHEME 3
the sugar can effect SNAr displacement of 6-halo substit-
uents to produce insoluble polymers. Dilute solutions in
alcoholic ammonia at -5 to 0 °C or dilute solutions in
0.01 M NaOH/H2O gave good yields (61-92%) of the
analytically pure deacetylated products.
In summary, our results are consistent with fragmen-
tations of purinediazonium species at C2 or C6 via
competing homolytic and heterolytic processes. Antimony
trihalides are efficient catalysts. Nonaqueous halo-de-
diazoniation of 2-amino-6-chloro-9-(2,3,5-tri-O-acetyl-â-
D-ribofuranosyl)purine (1) with SbX3/BTEA-X/NaNO2
gave good yields (>70%) of 2-(bromo or chloro)-6-chloro-
purine products. The 6-chloro-2-iodopurine compound 4
(45%) was obtained with SbI3, and 6-chloropurine deriva-
tive 5 (82%) was prepared by hydro-dediazoniation of 1.
Bromo-dediazoniations at C6 of 2′,3′,5′-tri-O-acetylad-
enosine (6) and analogues were effected in good yields
(>60%). A modified procedure gave the 6-bromopurine
2′-deoxynucleoside 16 (74%) from 3′,5′-di-O-acetyl-2′-
deoxyadenosine (15). Analogous SbI3 catalysis gave the
6-iodopurine analogue 17 (47%), and mild deacetylation
conditions were employed to provide crystalline 6-(bromo
or iodo)purine 2′-deoxynucleosides (88%). Optimized
procedures are noted in the general material of the
Experimental Section.
SCHEME 4
to 6/SbBr3/TBN/CH2Br2 at ambient temperature did not
enhance the yield of 7. Dediazoniations with NaNO2/
DCA/SbBr3/BTEA-Br proceeded equally well at ambient
temperature or ∼60 °C (CH2Br2 at reflux), whereas lower
temperatures (-10 °C) retarded reaction rates. Efficient
bromo-dediazoniations at C6 were parallel with those at
C2 except lower ratios of DCA were required. Constant
proportions of SbBr3, BTEA-Br, NaNO2, and CH2Br2 gave
better yields of 7 (72%) with ∼0.5 equiv than with 1 (62%)
or 1.5 (58%) equiv of DCA.
Substitution of HOAc (0.5 equiv) for DCA allowed
efficient bromo-dediazoniation of the acid-labile 2′-deoxy
derivative. Treatment of 15 (Scheme 4) with SbBr3,
BTEA-Br, NaNO2, and HOAc in CH2Br2 under optimized
conditions gave 16 (74%). An alternative method with
SbX3 (2 equiv), TBN, and CH2X2 gave 16 (60%) or its
6-iodo analogue 17 (47%).
Finally, treatment of 1 (Scheme 1) with DCA and
powdered NaNO2 (15 equiv each) in THF at ambient
temperature for 2 h gave high yields of 5 (82%) (Table 1,
entry 16). This demonstrated that hydro-dediazoniation
at C2 of 1 can be enhanced dramatically by alteration of
acid and nitrite reagents. This procedure provides ef-
ficient access to the 6-chloropurine derivative 5 starting
from guanosine, a naturally occurring 2-aminopurin-6-
one nucleoside.
Exp er im en ta l Section
Uncorrected melting points were determined with a hot-
stage apparatus. UV spectra were recorded with solutions in
MeOH unless otherwise indicated. 1H NMR spectra (solutions
in TMS/DMSO-d6) were recorded at 100 or 200 MHz unless
otherwise indicated. “Apparent” peak shapes are in quotation
marks when first-order splitting should be more complex, or
when peaks were poorly resolved. High-resolution mass spec-
tra (MS) were determined with FAB (glycerol) or CI (CH4).
All chemicals and solvents were of reagent quality. THF, CH2-
Cl2, and CH2Br2 were dried by reflux over and distillation from
CaH2. SbCl3 (Fisher Scientific) and SbBr3 (Alfa Inorganics)
were commercially available; SbI3 was prepared as described;42
antimony trihalides were purified by sublimation (90 °C, 50
mmHg). Benzyltriethylammonium bromide (BTEA-Br) was
prepared from BTEA-Cl by ion exchange [Dowex 1 × 2 (Br-)].
Column chromatography was performed with silica gel (230-
400 mesh). Radial TLC (Chromatotron) was performed with
silica gel (Merck, TLC grade 7749 with gypsum binder).
Substrates 1,35 6,43 and 153b were prepared as described.
Meth od s 1 (SbX3/BTEA-X/NaNO2/HOAc/CH2X2), 2 (SbX3/
BTEA-X/NaNO2/DCA/CH2X2), and 3 (SbX3/TBN/CH2X2) are
described for conversions of 15 f 16. Analogous treatment
(42) Schenk, P. W. In Handbook of Preparative Inorganic Chemistry;
Brauer, G., Ed.; Academic: New York, 1963; Vol. 1, p 614.
(43) Bredereck, H.; Martini, A. Chem. Ber. 1947, 80, 401-405.
Deacylation of 6-halopurine nucloside derivatives can
be problematic, because liberated oxygen nucleophiles on
J . Org. Chem, Vol. 67, No. 19, 2002 6793

Francom et al.
with equivalent molar proportions of other nucleosides gave
the indicated products and quantities. “Column A” (7-cm
diameter) contained (top to bottom) silica gel (10 g), decoloriz-
ing carbon (0.4 g), Celite filter aid (0.4 g), and silica gel (20 g).
Protected nucleosides were deacetylated by methods 4 or 5.
Meth od 4: The acetylated nucleoside (∼1 mmol) was dissolved
in NH3/MeOH (50 mL, saturated at 0 °C) and the solution was
stored at -5 to 0 °C for 18-24 h. Volatiles were evaporated
under reduced pressure (0-5 °C), and 98% EtOH was added
to and evaporated from the residue. The residue was recrystal-
lized (from 98% EtOH or MeOH), filtered, washed with cold
EtOH, and dried over Drierite (CaSO4). Meth od 5: The
acetylated nucleoside (∼1 mmol) was dissolved in CH2Cl2 (100
mL) in a 2000-mL round-bottom flask, and volatiles were
evaporated to give a thin, dilute film (this procedure retards
polymerization/decomposition). NaOH/H2O (0.01 M, 380 mL,
3.8 mmol) was added, and the solution was stirred at ambient
temperature for 1-2 h (TLC, CHCl3/MeOH, 9:1). The solution
was transferred to a clean flask, the 2000-mL flask was rinsed
with H2O (3 × 10 mL), and Amberlite XAD-4 resin (145 mL)
was added to the combined aqueous solution. The mixture was
stirred for 10 min at ambient temperature (UV of the super-
natant indicated that product was adsorbed), and the resin
was filtered and washed with H2O (2 × 100 mL). The resin
was then suspended in CH3CN (100 mL), the mixture was
stirred for 10 min at ambient temperature, and the resin was
filtered. This extraction was repeated (CH3CN, 2 × 100 mL),
and volatiles were evaporated (0-5 °C) from the combined
CH3CN filtrates. The residue was lyophilized (oil pump), dried
Et2O was added and evaporated (3 × 5 mL), and the solid was
dried over Drierite (CaSO4) at ambient temperature. Drying
some samples over P2O5 caused decomposition.
and 1 (488 mg, 1.0 mmol) were added to dried THF (40 mL)
in a 250-mL round-bottom flask stirred with a heavy magnetic
stirring bar or a mechanical stirrer. Dichloroacetic acid (1.25
mL, 1.93 g, 15 mmol) was added, and the flask was flushed
with dried N2 and sealed. The mixture was stirred vigorously,
and a thick gel was formed. This became a thick slurry after
stirring for 1.5-2 h (TLC; MeOH/CHCl3, 1:9, indicated conver-
sion of 1 into a major and a minor product). The slurry was
transferred (dropwise with a large-bore pipet) from the flask
into a vigorously stirred mixture of saturated NaHCO3/H2O
(200 mL) and CHCl3 (200 mL) that was cooled in an ice-water
bath. The phases were separated, and the organic layer was
washed [H2O (100 mL) and then brine (100 mL)], dried
(MgSO4), and filtered. The filtrate was passed through a layer
of silica gel (20 g) in a fritted-glass funnel, and product was
eluted with MeOH/CHCl3 (0.1:99.9, 1 L). Volatiles were
evaporated to give a slightly yellow oil. Dried Et2O was added
and evaporated several times to give 5 (327 mg, 82%) as a
slightly yellow TLC-homogeneous solid foam. UV and NMR
data were in agreement with published values.21b P r oced u r e
B: Treatment of 6 (1.97 g, 5.0 mmol) by method 3 gave 5 (965
mg, 47%) as a light-yellow oil with TLC migration and spectral
data the same as product from procedure A.
6-Ch lor o-9-(â-D-r ibofu r a n osyl)p u r in e (5a ). Treatment of
5 by method 4 gave 5a (61%) (two crops from EtOH) with mp
193-195 °C dec (lit.44 mp 180-182 °C dec); UV max 265 nm
1
(ꢀ 8700); H NMR δ 3.67 (m, 2H), 4.01 (m, 1H), 4.22 (q, 1H),
4.62 (q, 1H), 5.10 (t, 1H), 5.26 (d, 1H), 5.58 (d, 1H), 6.08 (d,
1H), 8.86, 9.00 (2 × s, 2 × 1H); MS m/z 286.0463/288.0420
(M+ 35/37Cl) ) 286.0468/288.0439). Anal. Cacd for C10H11
( -
ClN4O4: C, 41.90; H, 3.87; N, 19.54. Found: C, 42.14; H, 4.12;
N, 19.45.
9-(2,3,5-Tr i-O-a cetyl-â-D-r ibofu r a n osyl)-2,6-d ich lor op u -
r in e (2). Treatment of 1 (428 mg, 1.0 mmol) by method 2 gave
2 (329 mg, 74%) as slightly yellow crystals (from BuOH) with
mp 161-163 °C (lit.26 mp 158 °C); UV (H2O, pH ∼7) max 273
nm (ꢀ 10 100); 1H NMR (200 MHz; CDCl3) δ 2.07, 2.13, 2.15 (3
× s, 3 × 3H), 4.38 (“d”, J ) 3.4 Hz, 2H), 4.44-4.48, 5.53-5.58
(2 × m, 2 × 1H), 5.77 (“t”, J ) 5.5 Hz, 1H), 6.20 (d, J ) 5.6
9-(2,3,5-Tr i-O-a ce t yl-â-D-r ib ofu r a n osyl)-6-b r om op u -
r in e (7). Treatment of 6 (393 mg, 1.0 mmol) by method 2 gave
1
721b (331 mg, 72%) as a white solid foam with H NMR δ 2.02,
2.06, 2.14 (3 × s, 3 × 3H), 4.10-4.50 (m, 3H), 5.65 (“dd”, 1H),
6.03 (“t”, J ) 5.0 Hz, 1H), 6.37 (d, J ) 5.0 Hz, 1H), 8.80, 8.90
(2 × s, 2 × 1H); MS m/z 456.0281/458.0268 (M+ [C16H17
-
79/81BrN4O7] ) 456.0281/458.0260). Treatment of 6 (1.97 g, 5.0
mmol) by method 3 gave 7 (1.33 g, 58%) as a white solid foam
with TLC migration and spectral data the same as the product
from method 2.
Hz, 1H), 8.27 (s, 1H); LRMS (CI) m/z 447 (MH+ [C16H17
35Cl2N4O7] ) 447).
-
9-(2,3,5-Tr i-O-acetyl-â-D-r ibofu r an osyl)-2-br om o-6-ch lo-
r op u r in e (3). Treatment of 1 (428 mg, 1.0 mmol) by method
2 gave 3 (308 mg, 63%) as white needles (from EtOH) with
mp 162-163 °C (lit.21d mp 155-156 °C); UV max 275 nm
(ꢀ 9100); 1H NMR (200 MHz, CDCl3) δ 2.08, 2.13, 2.15 (3 × s,
3 × 3H), 4.42 (“d”, J ) 3.4 Hz, 2H), 4.41-4.50, 5.56-5.61
(2 × m, 2 × 1H), 5.79 (“t”, J ) 5.5 Hz, 1H), 6.22 (d, J )
6-Br om o-9-(â-D-r ibofu r a n osyl)p u r in e (7a ). Treatment of
7 by method 4 gave 7a (65%) (from EtOH) with mp 192 °C
dec (lit.34 mp 181-182 °C); UV max 267 nm (ꢀ 9800); 1H NMR
δ 3.62 (q, 1H), 3.68 (m, 2H), 4.01 (m, 1H), 4.22 (q, 1H), 5.09 (t,
1H), 5.25 (d, 1H), 5.57 (d, 1H), 6.07 (d, 1H), 8.81, 9.00 (2 × s,
2 × 1H); MS m/z 329.9950 (M+ [79Br] ) 329.9963). Anal. Calcd
for C10H11BrN4O4: C, 36.27; H, 3.35; N, 16.92. Found: C, 36.44;
H, 3.52; N, 17.07. Treatment of 7 by method 5 gave 7a (69%)
(from EtOH) with the same TLC migration and spectral data
as the product from method 4.
6-Iod o-9-(â-D-r ibofu r a n osyl)p u r in e (8a ). TBN (11.9 mL,
10.3 g, 100 mmol) was added to a stirred solution of 6 (1.97 g,
5.0 mmol) and SbI3 (5.0 g, 10 mmol) in CH2I2/THF (1:1, 100
mL) at 60 °C (oil bath temperature), and stirring was contin-
ued for 10 min. Volatiles were evaporated, and the residue
was diluted (CHCl3). The solution was washed (5% NaHSO3/
H2O and then H2O) and dried (MgSO4). Volatiles were evapo-
rated, and a small volume of CHCl3 was added. Chromatog-
raphy (silica gel, 100 mL, CHCl3) gave 9-(2,3,5-tri-O-acetyl-
â-D-ribofuranosyl)-6-iodopurine21b (8) as a light-yellow oil (1.16
g, 46%) with 1H NMR δ 2.02, 2.06, 2.14 (3 × s, 3 × 3H), 4.29
(m, 1H), 4.43 (m, 2H), 5.66 (dd, 1H), 6.04 (t, 1H), 6.34 (d, 1H),
8.70, 8.88 (2 × s, 2 × 1H); MS m/z 504.0158 (M+ [C16H17IN4O7]
) 504.0142).
5.4 Hz, 1H), 8.25 (s, 1H); LRMS (FAB) m/z 491 (MH+ [C16H17
79Br35ClN4O7] ) 491).
-
9-(2,3,5-Tr i-O-a ce t yl-â-D-r ib ofu r a n osyl)-6-ch lor o-2-
iod op u r in e (4). TBN (9.5 mL, 8.2 g, 80 mmol) was added
with stirring at ambient temperature to a solution of 1 (1.71
g, 4.0 mmol) and SbI3 (4 g, 8 mmol) in a mixture of dried THF
(15 mL) and CH2I2 (35 mL). Reaction was complete after 2 h
(TLC; MeOH/CHCl3, 9:91), and volatiles were evaporated
under reduced pressure. The residue was diluted (CHCl3), and
the solution was washed (5% NaHSO3/H2O and then H2O) and
dried (MgSO4). Volatiles were evaporated, and the residue was
diluted (CHCl3) and chromatographed (silica gel, 100 mL,
CHCl3). Volatiles were evaporated, the residual oil was dis-
solved (CHCl3/PrOH), and the solution was concentrated to a
small volume and cooled (∼4 °C) overnight. Light yellow
crystals were filtered and washed (PrOH) to give 4 (965 mg,
45%) with mp 182-183 °C (lit.21d mp 181-183 °C); UV max
255, 281 nm (ꢀ 5700, 8800); 1H NMR δ 2.03, 2.08, 2.14 (3 × s,
3 × 3H), 4.24-4.44 (m, 3H), 5.66 (dd, 1H), 5.92 (t, 1H), 6.33
(d, 1H), 8.86 (s, 1H); MS m/z 539.9717 (M+ [37Cl] ) 539.9722).
Anal. Calcd for C16H16ClIN4O7: C, 35.68; H, 2.99; N, 10.40.
Found: C, 35.99; H, 3.08; N, 10.35.
Treatment of 8 by method 4 gave 8a (76%) (from EtOH) with
mp 173-175 °C (lit.34 mp 173 °C dec); UV max 276 nm (ꢀ
1
10 800); H NMR δ 3.62 (q, 1H), 3.67 (m, 2H), 4.01 (m, 1H),
9-(2,3,5-Tr i-O-a ce t yl-â-D-r ib ofu r a n osyl)-6-ch lor op u -
r in e (5). P r oced u r e A: Powdered NaNO2 (1.04 g, 15 mmol)
(44) Baker, B. R.; Hewson, K.; Thomas, H. J .; J ohnson, J . A., J r. J .
Org. Chem. 1957, 22, 954-959.
6794 J . Org. Chem., Vol. 67, No. 19, 2002

Nonaqueous Diazotization of Aminopurine Nucleosides
4.22 (q, 1H), 5.09 (t, 1H), 5.25 (d, 1H), 5.56 (d, 1H), 6.03 (d,
(m, 1H), 5.08 (t, J ) 5.5 Hz, 1H), 5.75 (d, J ) 4.0 Hz, 1H),
6.01 (d, J ) 1.5 Hz, 1H), 8.75, 8.94 (2 × s, 2 × 1H); MS m/z
315.0098/317.0077 (MH+ [C10H1279/81BrN4O3] ) 315.0094/
317.0073). Anal. Calcd for C10H11BrN4O3: C, 38.11; H, 3.52;
Br, 25.36; N, 17.78. Found: C, 38.11; H, 3.54; Br, 25.19; N,
17.85.
1H), 8.70, 8.95 (2 × s, 2 × 1H); MS m/z 377.9825 (M+ [C10H11
-
IN4O4] ) 377.9824).
9-(â-D-Ribofu r a n osyl)p u r in e (Nebu la r in e) (9a ). TBN
(23.8 mL, 20.6 g, 200 mmol) was added to a stirred solution of
6 (3.94 g, 10 mmol) in dried THF (120 mL) at 60 °C (oil bath
temperature). After 20 min, TLC (MeOH/CHCl3, 9:91) showed
the less polar 9-(2,3,5-tri-O-acetyl-â-D-ribofuranosyl)purine (9)
and more polar 2′,3′,5′-tri-O-acetylinosine (10). Volatiles were
evaporated, and the residue was dissolved (CHCl3) and chro-
matographed (silica gel, CHCl3) to give 921b (1.16 g, 31%) as a
light-yellow oil with 1H NMR δ 2.02, 2.06, 2.14 (3 × s, 3 ×
3H), 4.20-4.60 (m, 3H), 5.70 (dd, 1H), 6.10 (t, 1H), 6.40 (d,
1H), 8.86, 9.04, 9.28 (3 × s, 3 × 1H); MS m/z 378.1155 (M+
[C16H18N4O7] ) 378.1175).
Treatment of 9 by method 4 gave 9a (65%) (from EtOH) with
mp 178-180 °C (lit.45 mp 176-178 °C); UV (H2O, pH ∼7) max
262 nm (ꢀ 5600); 1H NMR δ 3.21 (q, 1H), 3.65 (m, 2H), 3.66 (q,
1H), 4.00 (m, 1H), 5.10 (t, 1H), 5.25 (d, 1H), 5.55 (d, 1H), 6.08
(d, 1H), 8.89, 9.00, 9.24 (3 × s, 3 × 1H); MS m/z 252.0862 (M+
) 252.0859). Anal. Calcd for C10H12N4O4: C, 47.62; H, 4.80;
N, 22.21. Found: C, 47.64; H, 4.98; N, 22.14.
9-(3,5-Di-O-acetyl-2-deoxy-â-D-er yth r o-pen tofu r an osyl)-
6-br om op u r in e (16). Meth od 1: SbBr3 (361 mg, 1.0 mmol)
in CH2Br2 (4.5 mL) was added to a solution of 15 (335 mg, 1.0
mmol), BTEA-Br (272 mg, 1.0 mmol), and NaNO2 (1.4 g, 20
mmol) in CH2Br2 (40 mL). AcOH (29 µL, 30 mg, 0.50 mmol)
was added, and the flask was flushed with dried N2 and sealed.
The mixture was stirred vigorously with a heavy magnetic
stirring bar or mechanical stirrer at ambient temperature (25
( 5 °C) until 15 had been converted into a major and minor
product (∼3 days; TLC, MeOH/CHCl3, 1:9). Celite (3 g) and
CHCl3 (120 mL) were added, and the suspension was stirred
for 10 min. The mixture was applied to “column A” and product
was eluted (MeOH/CHCl3, 0.2:100, ∼600 mL). Volatiles were
evaporated, and dried Et2O was added and evaporated several
times to give 163d (294 mg, 74%) as a yellow glass with UV
1
(EtOH) max 267 nm (ꢀ 7300); H NMR) δ 1.97, 2.09 (2 × s, 2
2′,3′,5′-Tr i-O-a cetylin osin e (10). TBN (11.9 mL, 10.3 g,
100 mmol) was added to a stirred solution of 6 (1.97 g, 5.0
mmol) in DME/H2O (1:1, 100 mL) preheated to 60 °C (oil bath
temperature). After 15 min, the solution was concentrated to
one-half volume, neutralized (NaHCO3/H2O), and extracted
(CHCl3, 3 × 50 mL). Volatiles were evaporated from the
combined organic phase, and the residue was recrystallized
(from EtOH) to give 10 (1.36 g, 69%) with mp 239-243 °C (lit.43
mp 241 °C); UV max 244 (ꢀ 11 500); 1H NMR δ 2.03, 2.05,
2.12 (3 × s, 3 × 3H), 4.10-4.50 (m, 3H), 5.55 (dd, 1H), 5.91 (t,
1H), 6.20 (d, 1H), 8.11, 8.32 (2 × s, 2 × 1H), 12.48 (br, 1H).
9-(â-D-Ar a bin ofu r a n osyl)-6-br om op u r in e (12a ). Treat-
ment of 1146 (4.0 g, 10 mmol) by method 3 gave 9-(2,3,5-tri-
O-acetyl-â-D-arabinofuranosyl)-6-bromopurine (12) (2.9 g, 63%)
× 3H), 2.53-2.67, 3.09-3.22 (2 × m, 2 × 1H), 4.15-4.32 (m,
3H), 5.42-5.45 (m, 1H), 6.49 (“t”, J ) 7.0 Hz, 1H), 8.76, 8.89
(2 × s, 2 × 1H); LRMS (CI) m/z 399/401 (MH+
[
79/81Br] ) 399/
401). Anal. Calcd for C14H15BrN4O5: C, 42.12; H, 3.79; N,
14.04. Found: C, 42.15; H, 4.06; N, 14.23.
Meth od 2: SbBr3 (361 mg, 1.0 mmol) in CH2Br2 (4.5 mL)
was added to a mixture of 15 (335 mg, 1.0 mmol), BTEA-Br
(272 mg, 1.0 mmol), and NaNO2 (1.4 g, 20 mmol) in CH2Br2
(40 mL). Cl2CHCO2H (41 µL, 64 mg, 0.5 mmol) was added,
and the flask was flushed with dried N2 and sealed. The
mixture was stirred at ambient temperature until nearly all
of 15 was converted into a major and three minor products
(∼2 days, TLC, MeOH/CHCl3, 1:9). Celite (3 g) and CHCl3 (120
mL) were added, and the suspension was stirred for 10 min.
The mixture was applied to “column A” and product was eluted
(MeOH/CHCl3, 0.2:100, ∼500 mL). Volatiles were evaporated,
and dried Et2O was added and evaporated several times to
give 16 (264 mg, 66%) as a yellow glass that had the same
TLC migration and spectral data as the product from method
1.
1
as a slightly yellow oil with H NMR (400 MHz, CDCl3) δ 1.93,
2.18, 2.22 (3 × s, 3 × 3H), 4.32 (“dt”, J ) 6.0, 4.5 Hz, 1H), 4.50
(2 × dd, J ) 12.0, 6.0, 4.5 Hz, 2H), 5.48 (dd, J ) 4.5, 3.5, Hz,
1H), 5.56 (dd, J ) 4.5, 3.5 Hz, 1H), 6.67 (d, J ) 4.5 Hz, 1H),
8.37, 8.76 (2 × s, 2 × 1H); MS m/z 456.0277/458.0264 (M+
[C16H1779/81BrN4O7] ) 456.0281/458.0261).
Treatment of 12 by method 4 gave 12a (80%) (from MeOH)
Meth od 3: A stirred solution of 15 (1.8 g, 5.3 mmol) and
SbBr3 (3.84 g, 10.6 mmol) in CH2Br2 (100 mL) was heated at
60 °C for 15 min. TBN (12.6 mL, 10.9 g, 106 mmol) was added,
and heating was continued for 20 min. Volatiles were evapo-
rated (to half the original volume), and this solution was
washed (5% NaHCO3/H2O and then H2O) and dried (MgSO4).
Volatiles were evaporated, and the residual oil was chromato-
graphed (100 mL silica gel, CHCl3). TLC-homogeneous frac-
tions were combined and volatiles were evaporated. Dried Et2O
was added and evaporated several times to give 16 (1.3 g, 60%)
1
with mp >350 °C dec; UV max 266 nm (ꢀ 11 000); H NMR
(400 MHz, DMSO-d6) δ 3.71 (m, J ) 12.0, 5.0 Hz, 2H), 3.85
(q, J ) 5.0 Hz, 1H), 4.18 (q, J ) 5.0 Hz, 1H), 4.27 (q, J ) 5.0
Hz, 1H), 5.15 (t, J ) 5.0 Hz, 1H), 5.61, 5.67, 6.42 (3 × d, 3 ×
1H), 8.77, 8.80 (2 × s, 2 × 1H). Anal. Calcd for C10H11BrN4O4:
C, 36.27; H, 3.35; Br, 24.13; N, 16.92. Found: C, 36.42; H, 3.42;
Br, 24.19; N, 16.95.
6-Br om o-9-(3-d eoxy-â-D-er yth r o-p en t ofu r a n osyl)p u -
r in e (14a ). Treatment of 1347 (5.0 g, 15 mmol) by method 3
gave 9-(2,5-di-O-acetyl-3-deoxy-â-D-erythro-pentofuranosyl)-6-
bromopurine (14) (3.7 g, 61%) as a slightly yellow oil with 1H
NMR (400 MHz, DMSO-d6) δ 1.98, 2.13 (2 × s, 2 × 3H), 2.28
(ddd, J ) 14.0, 6.0, 1.0 Hz, 1H), 2.64 (ddd, J ) 14.0, 10.5, 6.0
Hz, 1H), 4.22 (dd, J ) 12.0, 6.0 Hz, 1H), 4.30 (dd, J ) 12.0,
3.0 Hz, 1H), 4.58 (m, 1H), 5.76 (br d, 1H), 6.30 (d, J ) 1.5 Hz,
1H), 8.81, 8.86 (2 × s, 2 × 1H); MS m/z 398.0214/400.0204
(M+ [C14H1579/81BrN4O5] ) 398.0227/400.0207).
Treatment of 14 by method 4 gave 14a (92%) (from MeOH)
with mp 155-156 °C; UV max 266 nm (ꢀ 11 200); 1H NMR
(400 MHz, DMSO-d6) δ 1.90, (ddd, J ) 13.5, 9.5, 2.5 Hz, 1H),
2.24 (ddd, J ) 13.5, 6.0, 5.5 Hz, 1H), 3.55 (dd, J ) 12.0, 4.0
Hz, 1H), 3.74 (dd, J ) 12.0, 3.5 Hz, 1H), 4.42 (m, 1H), 4.62
1
as a slightly yellow solid foam with H NMR (400 MHz, DMSO-
d6) δ 2.13, 2.14 (2 × s, 2 × 3H), 2.66 (ddd, J ) 14.5, 6.5, 3.0
Hz, 1H), 3.21 (“quint”, J ) ∼6.5 Hz, 1H), 4.26 (dd, J ) 12.5,
7.5 Hz, 1H), 4.33 (m, 1H), 4.34 (dd, J ) 4.5, 3.0 Hz, 1H), 5.49
(“quint”, J ) ∼3.5 Hz, 1H), 6.54 (t, J ) 6.5 Hz, 1H), 8.83, 8.96
(2 × s, 2 × 1H); MS m/z 398.0221/400.0189 (M+
[
79/81Br] )
398.0227/400.0207). Anal. Calcd for C14H15BrN4O5: C, 42.12;
H, 3.79; Br, 20.01; N, 14.04. Found: C, 42.28; H, 3.90; Br,
19.71; N, 13.98.
6-Br om o-9-(2-d eoxy-â-D-er yth r o-p en t ofu r a n osyl)p u -
r in e (16a ). Treatment of 16 by method 4 gave 16a (78%) (from
MeOH/EtOAc) with mp 150-151 °C; UV max 266 nm (ꢀ
10 700); 1H NMR (400 MHz, DMSO-d6) δ 2.39 (ddd, J ) 13.5,
6.5, 4.5 Hz, 1H), 2.78 (“quint”, J ) 6.5 Hz, 1H), 3.55 (dd, J )
12.0, 4.5 Hz, 1H), 3.63 (dd, J ) 12.0, 4.5 Hz, 1H), 3.91 (“q”, J
) 4.5 Hz, 1H), 4.47 (m, 1H), 4.99 (t, J ) 6.5 Hz, 1H), 5.39 (d,
J ) 4.5 Hz, 1H), 6.54 (“t”, J ) 6.5 Hz, 1H), 8.83, 8.96 (2 × s,
(45) Schaeffer, H. J .; Thomas, H. J . J . Am. Chem. Soc. 1958, 80,
4896-4899.
(46) Reist, E. J .; Calkins, D. F.; Fisher, L. V.; Goodman, L. J . Org.
Chem. 1968, 33, 1600-1603.
2 × 1H); LRMS (CI) m/z 315/317 (MH+
[
79/81Br] ) 315/317).
(47) Shibuya, S.; Kodama, K.; Kusakabe, H.; Fujiyama, K.; Kuni-
naka, A. J pn. Kokai Tokkyo Koho 1979, 55,596; Chem. Abstr. 1979,
91, 91933z.
Anal. Calcd for C10H11BrN4O3: C, 38.11; H, 3.52; Br, 25.36;
N, 17.78. Found: C, 38.00; H, 3.55; Br, 25.51; N, 18.04.
J . Org. Chem, Vol. 67, No. 19, 2002 6795

Francom et al.
Treatment of 16 by method 5 gave 16a (88%) (from MeOH/
EtOAc) with mp 150 °C. Its TLC migration and spectral data
were the same as the product from method 4.
(“quint”, J ) 7.2 Hz, 1H), 4.26 (m, 3H), 5.44 (“quint”, J ) 3.3
Hz, 1H), 6.47 (dd, J ) 6.9, 7.2 Hz, 1H), 8.66, 8.85 (2 × s, 2 ×
1H).
Treatment of 17 by method 4 gave 17a 3c (89%) (from EtOH)
with mp 152-153 °C; UV max 275 nm (ꢀ 10 200); 1H NMR
(300 MHz, DMSO-d6) δ 2.36 (m, 1H), 2.76 (“quint”, J ) 6.6
Hz, 1H), 3.52 (m, 1H), 3.60 (m, 1H), 3.89 (“q”, J ) 4.2 Hz, 1H),
4.44 (m, 1H), 4.99 (m, 1H), 5.39 (m, 1H), 6.43 (“t”, J ) 6.6 Hz,
1H), 8.63, 8.86 (2 × s, 2 × 1H); LRMS m/z 362 (M+ ) 362).
Anal. Calcd for C10H11IN4O3: C, 33.17; H, 3.06; N, 15.47.
Found: C, 33.17; H, 3.12; N, 15.30.
9-(2-Deoxy-â-D-er yth r o-p en t ofu r a n osyl)-6-iod op u r in e
(17a ). TBN (1.2 mL, 1.04 g, 10 mmol) was added to a stirred
solution of 15 (167 mg, 0.5 mmol) and SbI3 (0.5 g, 1.0 mmol)
in CH2I2/THF (1:1, 10 mL) at 60 °C (oil bath temperature),
and stirring was continued for 10 min. Volatiles were evapo-
rated, and the residue was diluted (CHCl3). The solution was
washed (5% NaHSO3/H2O and then H2O) and dried (MgSO4).
Volatiles were evaporated, and a small volume of CHCl3 was
added. Chromatography (silica gel, 15 mL, CHCl3) gave
9-(3,5-di-O-acetyl-2-deoxy-â-D-erythro-pentofuranosyl)-6-iodo-
purine (17) as a white solid foam (105 mg, 47%) with UV max
275 nm (ꢀ 10 400); 1H NMR (300 MHz, DMSO-d6) δ 1.98, 2.10
(2 × s, 2 × 3H), 2.60 (ddd, J ) 14.7, 6.6, 3.3 Hz, 1H), 3.18
Ack n ow led gm en t. We thank Brigham Young Uni-
versity, the National Cancer Institute of Canada, and
The University of Alberta for financial support and Mrs.
J eanny K. Gordon for assistance with the manuscript.
J O0204101
6796 J . Org. Chem., Vol. 67, No. 19, 2002