interlocked boat-like 6-membered rings for the DABCO sys-
tem. The free nitrogen atom of the DABCO is hydrogen-
bonded to a water molecule (the presence of which was also
indicated by elemental analysis), which acts as a bridge to a
chloride anion. The latter is bonded to the hydrogen atom at
N-9 of another purine molecule, to give continuousؒؒؒcation ؒؒؒ
anion ؒ ؒ ؒ water ؒ ؒ ؒ hydrogen-bonded chains.
Compounds 1a–d undergo facile displacement reactions with
alkoxides in dimethyl sulfoxide, with a typical example being
shown in Scheme 1. Results for 1a leading to 6-substituted
purines 2a–h† are summarised in Table 1. The order of
displacement of groups from the 6-position of purines is
100:10:5:1). The slower displacement of DABCO vs. Me N is
probably a steric effect (more crowded transition state for
3
DABCO in the addition–elimination pathway). However, we
have not observed competing attack of the alkoxide on a
DABCO methylene group, and the only disadvantage to the use
of DABCO instead of trimethylamine is the slower rate of reac-
tion of the DABCO derivatives. There are several advantages of
using DABCO over chloro and trimethylammonio moieties,
besides those already mentioned. Reactions with DABCO-
purines are easy to monitor by thin layer chromatography
(TLC) because they exhibit fluorescent spots at relatively low
R , which disappear as the reaction proceeds. Furthermore, the
f
Me N > DABCO > quinuclidine > Cl (relative rates ca.
desired product is very easy to separate from the starting
DABCO-purine, because of the latter’s greater polarity and
solubility in water. In contrast, chloropurines often have a simi-
3
lar R (TLC on silica) to the derived 6-substituted purine (e.g.
f
with C alkoxy groups), which makes separation difficult by
4
normal phase chromatography.
In a previously reported preparation of 2,4-dinitrophenyl
ethers from carbohydrates with a free hydroxy group, the
alcohol (e.g. a mixture of the anomers of 2,3,4,6-tetra-O-
acetyl--galactopyranose) was treated with 2,4-dinitro-
fluorobenzene and an excess of DABCO in dimethylform-
Scheme 1
1
0
amide. The mechanism of this reaction was not discussed,
but it seems likely that it proceeds via 1-(1-azonia-4-
azabicyclo[2.2.2]oct-1-yl)-2,4-dinitrobenzene fluoride. Indeed,
such adducts have been shown to be intermediates in reactions
Table 1 Reactions of DABCO-purine 1a with alcohols leading to
a
products 2a–h
1
1
of DABCO with 2,4-dinitrochlorobenzene and quinuclidine
1
2
with picryl chloride.
We thank ZENECA, EPSRC (for provision of an X-ray
diffractometer) and the North of England Cancer Research
Campaign for support of this research.
Reaction
ROH
time/h
Product
Yield (%)
References
MeOH
12
2a
2b
2c
2d
88
87
72
81
1
2
3
A. E. Pegg, M. E. Dolan and R. C. Moschel, Prog. Nucleic Acid Res.
Mol. Biol.,1995, 51, 162.
M. Y. Chae, M. G. McDougall, M. E. Dolan, K. Swenn, A. E. Pegg
and R. C. Moschel, J. Med. Chem., 1994, 37, 342.
T. B. H. McMurry, R. S. McElhinney, J. McCormick, R. H. Elder,
J. Kelly, G. P. Margison, J. A. Rafferty, A. J. Watson and M. A.
Willington, Int. Pat. Appl., WO 94/29312.
C. Arris, C. Bleasdale, A. H. Calvert, N. J. Curtin, C. Dalby, B. T.
Golding, R. J. Griffin, J. M. Lunn, G. N. Major and D. R. Newell,
Anti-cancer Drug Des., 1994, 9, 401.
H C᎐CHCH OH
12
48
48
2
2
HC᎐CCH OH
BnOH
2
7
4
4
4
2
8
8
8
2e
51
74
82
70
4
5
2f
L. R. Lewis, F. H. Schneider and R. K. Robins, J. Org. Chem., 1961,
2
6, 3837.
6
7
8
J. Kiburis and J. H. Lister, J. Chem. Soc. (C), 1971, 3942.
B. L. Gaffney and R. A. Jones, Tetrahedron Lett., 1982, 23, 2253.
M. Ashwell, C. Bleasdale, B. T. Golding and I. K. O’Neill, J. Chem.
Soc., Chem. Commun., 1990. 955.
J. Kiburis and J. H. Lister, J. Chem. Soc. (C), 1971, 1587.
0 H. J. Koeners, A. J. de Kok, C. Romers and J. H. van Boom, Recl. J.
R. Neth. Chem. Soc., 1980, 99, 355.
2g
2h
9
1
1
1
1 S. D. Ross, J. J. Bruno and R. C. Peterson, J. Am. Chem. Soc., 1963,
8
5, 3999.
a
The alcohol (5.5 equiv.) was added to a stirred suspension of sodium
hydride (2.0 equiv.) in anhydrous dimethyl sulfoxide under nitrogen at
room temperature. After 1 h the DABCO-purine 1a (1 equiv.; 1 in
DMSO)wasadded and themixturewasstirred at room temperatureuntil
the reaction was complete (an aliquot was assayed by TLC and UV spec-
troscopy: shift from λmax 316 to 280 nm). The base was neutralised by
addition of glacial acetic acid and the solvent was removed in vacuo. The
residue was absorbed onto silica and the product was purified by medium
pressure chromatography (elution with 8% v/v MeOH in CH Cl ).
2 J. R. Gandler, I. U. Setiarahardjo, C. Tufon and C. Chen, J. Org.
Chem., 1992, 57, 4169.
Paper 6/08207F
Received 4th December 1996
Accepted 4th December 1996
2
2
1
86
J. Chem. Soc., Perkin Trans. 1, 1997