Efficient synthesis of purines and purine nucleosides via an inverse electron demand Diels-Alder reaction

Full text of DOI:10.1021/ja9842316

J. Am. Chem. Soc. 1999, 121, 5833-5834
5833
Table 1. Tandem Decarboxylation/Diels-Alder Reactions of 2a-c
Efficient Synthesis of Purines and Purine Nucleosides
via an Inverse Electron Demand Diels-Alder
Reaction
entry dienophile
conditiona
product yield (%)
1
2
3
2a
80 °C/2 h/DMF
25 °C/7 days/DMF
80 °C/2 h
tBuOH/buffer (pH ) 4.8)
80 °C/30 h
tBuOH/buffer (pH ) 7)
80 °C/30 h
tBuOH/1N HCl
3a
3a
3a
83
50
49
Qun Dang,* Yan Liu, and Mark D. Erion
4
5
6
3a
3a
3a
10
Department of Medicinal Chemistry, Metabasis
Therapeutics, Inc., 9390 Towne Centre DriVe,
San Diego, California 92121
0b
0c
80 °C/42 h
ReceiVed December 7, 1998
tBuOH/(sat) NaHCO3
110 °C/16 h/DMF-AcOH
100 °C/20 h/DMF-AcOH
90 °C/20 h/DMF-AcOH
80 °C/24 h/DMF-AcOH
90 °C/6 h/DMSO
7
8
9
10
11
2bd
2c
3b
3b
3b
3b
3c
45
Purine nucleosides represent an important class of therapeuti-
cally active agents and consequently a central focus of many drug
discovery efforts.1 Recently we became interested in the synthesis
of 2- and 6-substituted purine nucleosides on the basis of
theoretical calculations suggesting that certain substituents at these
positions might promote hydration of the 1,6-double bond and
consequently enable inhibition of adenosine deaminase with high
potency via transition state mimicry.2 Synthesis of these com-
pounds and purine nucleosides, in general, is frequently achieved
by synthesis of the base and sugar units separately, followed by
a coupling reaction. Unfortunately, the coupling reaction can
produce complex diastereomeric mixtures and proceed in poor
overall yields, especially when applied to the synthesis of 2′-
deoxynucleoside analogues. Accordingly, efforts to develop new
synthetic methodologies that circumvent these limitations continue
to be important. For example, Trost and Shi recently reported a
chemoselective glycosylation reaction that enables enantioselective
alkylation of 6-chloropurine at the 9-position with 2,5-dibenzoy-
loxy-2,5-dihydrofuran under palladium-catalyzed reaction condi-
tions in the presence of chiral ligands.3 Alternatively, the coupling
reaction is avoided in the synthesis of inosine analogues by using
ribosylated imidazoles such as 5-amino-1-(â-D-ribofuranosyl)-4-
imidazolecarboxamide and various reagents that result in pyri-
midine ring closure.4 These strategies, however, are not readily
adapted to the synthesis of purine nucleosides containing electron-
withdrawing carbon substituents at the 6-position. Compounds
such as 6-cyano-9-â-D-ribofuranosylpurine and 6-carbamoyl-9-
â-D-ribofuranosylpurine are prepared from 6-thioinosine and
6-cyanopurine in low yield,5a,5b and 6-formylpurine is prepared
from 6-methylpurine in five steps.5c
70
75e
81
75
a Reactions were conducted with 1 equiv of 1 and 2 equiv of 2a-c.
b Decomposition of 1 was observed. c Compound 1 was recovered.d The
potassium salt of 2b was used. e Three equivalents of 2b were used.
reaction is complicated by the instability of the dienophile, i.e.,
the electron-rich 5-aminoimidazole analogue.8 Since 5-amino-4-
imidazolecarboxylic acids are known to undergo decarboxylation
under relatively mild conditions,9 we chose to study the propensity
of 5-aminoimidazoles, generated in situ via decarboxylation, to
be trapped by 1,3,5-triazines via a [4 + 2] cycloaddition reaction.
Reaction of 5-amino-1-benzyl-4-imidazolecarboxylic acid (2a)10
with 1 at 80 °C in DMF led to 9-benzyl-2,6-bis(ethoxycarbonyl)-
purine (3a) in 83% yield (Table 1). Failure of the reaction to
produce 3a under basic conditions suggests that the decarboxy-
lation of 2a precedes the [4 + 2] cycloaddition reaction and that
2a itself is not reactive enough to participate in the [4 + 2]
reaction with 1 (Table 1, entry 6). In contrast, under conditions
known to induce decarboxyaltion, e.g., slightly acidic conditions,
the reaction is quite facile. The results also suggest that 5-amino-
1-benzylimidazole, generated in situ from 2a, is a more reactive
dienophile compared with 5-aminopyrazole7 on the basis of the
shorter reaction times at similar temperatures. For example,
previously reported 1,3,5-triazine Diels-Alder reactions6 required
extensive heating, whereas the current reaction is capable of
generating 3a at room temperature (Table 1, entry 2). Overall,
the results support the reaction sequence shown in Scheme 1
wherein decarboxylation of 2a produces the highly reactive
5-aminoimidazole which in the presence of 1 is rapidly trapped
as the [4 + 2] cycloadduct. This cycloadduct then spontaneously
undergoes a retro Diels-Alder reaction with the loss of ethyl
cyanoformate followed by the loss of ammonia and aromatization
to produce 3a in a regioselective manner.
In an effort to find a more efficient synthesis of purine
analogues, we explored the inverse electron demand Diels-Alder
reaction between 5-aminoimidazoles and 1,3,5-triazines. Previ-
ously, Boger and co-workers showed that 2,4,6-tris(ethoxycar-
bonyl)-1,3,5-triazine (1) is a useful diene for the synthesis of
pyrimidines,6 and we reported that reaction of 1 with 5-aminopy-
razoles yields pyrazolopyrimidines.7 In contrast to these reactions,
synthesis of purine analogues using the analogous Diels-Alder
Purines prepared using 1 have ester functionalities at the 2-
and 6-positions which can be modified in a regiospecific manner
to produce various 6-substituted and 2,6-disubstituted purine
analogues (Scheme 2). The 2,6-unsubstituted analogue of 3a, i.e.,
* Corresponding author. (tel.) 619-622-5517; (fax) 619-622-5573; (e-mail)
dang@mbasis.com.
(1) Reviews: Appleman, J. R.; Erion, M. D. Exp. Opin. InVest. Drugs 1998,
7, 225. Jacobson, K. A.; Jarvis, M. F., Eds. Purinergic Approaches in
Experimental Therapeutics; Wiley: New York, 1997. Chu, C. K.; Baker, D.
C., Eds. Nucleosides and Nucleotides as Antitumor and AntiViral Agents;
Plenum: New York, 1993. Bonnet, P. A.; Robins, R. K. J. Med. Chem. 1993,
36, 635. Huryn, D. M.; Okabe, M. Chem. ReV. 1992, 92, 1745. Jacobson, K.
A.; van Galen, P. J. M.; Williams, M. J. Med. Chem. 1992, 35, 407.
(2) Erion, M. D.; Reddy, M. R. J. Am. Chem. Soc. 1998, 120, 3295.
(3) Trost, B. M.; Shi, Z. J. Am. Chem. Soc. 1996, 118, 3037.
(4) Srivastava, P. C.; Robins, R. K.; Meyer, R. B., Jr. In Chemistry of
Nucleosides and Nucleotides; Townsend, L. B., Ed.; Plenum: New York,
1988; Vol. 1, pp 150-156.
(5) (a) Yamane, A.; Matsuda, A.; Ueda, T. Chem. Pharm. Bull. 1980, 28,
150. (b) Westover, J. D.; Revankar, G. R.; Robins, R. K.; Madsen, R. D.;
Ogden, J. R.; North, J. A.; Mancuso, R. W.; Rousseau, R. J.; Stephen, E. L.
J. Med. Chem. 1981, 24, 941. (c) Giner-Sorolla, A.; McCravey, M.; Bendich,
A. In Nucleic Acid Chemistry; Townsend, L. B., Tipson, R. S., Eds.; Wiley:
New York, 1986; Part 3, pp 11-15.
(6) For a review on hetero Diels-Alder reactions, see: Boger, D. L.;
Weinreb, S. M. Hetero Diels-Alder Methodology in Organic Synthesis;
Academic: San Diego, 1987. For reports on Diels-Alder reactions with
amidines see: Boger, D. L.; Dang, Q. J. Org. Chem. 1992, 57, 1631. Boger,
D. L.; Menezes, R. F.; Dang, Q. J. Org. Chem. 1992, 57, 4333. Boger, D. L.;
Kochanny, M. J. J. Org. Chem. 1994, 59, 4950.
(7) Dang, Q.; Brown, B. S.; Erion, M. D. J. Org. Chem. 1996, 61, 5204.
(8) Al-Shaar, A. H. M.; Gilmour, D. W.; Lythgoe, D. J.; McClenaghan, I.;
Ramsden, C. A. J. Chem. Soc., Perkin Trans. 1 1992, 2779, and references
therein.
(9) Cusack, N. J.; Shaw, G.; Litchfield, G. J. J. Chem. Soc. C 1971, 1501,
and references therein.
(10) Ryckman, D.; Casillas, S.; Erion, M. D. The synthesis of 2a will be
published elsewhere. For alternative syntheses of 2a and similar compounds,
see: Mackenzie, G.; Wilson, H. A.; Shaw, G.; Ewing, D. J. Chem. Soc., Perkin
Trans. 1 1988, 2541. Hunt, J. T.; Bartlett, P. A. Synthesis 1978, 741. Wong,
J. L.; Fuchs, D. S. J. Chem. Soc., Perkin Trans. 1 1974, 1284.
10.1021/ja9842316 CCC: $18.00 © 1999 American Chemical Society
Published on Web 06/06/1999

5834 J. Am. Chem. Soc., Vol. 121, No. 24, 1999
Communications to the Editor
Scheme 1a
This result is analogous to Boger’s early observations12 that
substituents at the 5-position of 2,6-bis(ethoxycarbonyl)pyrim-
idines provide a steric bias that enables selective reduction of
the 2-ester group.
The successful synthesis of purine analogues via 2a prompted
us to investigate the use of 5-amino-1-(â-D-ribofuranosyl)-4-
imidazolecarboxylic acid (2b)13 for the synthesis of purine
nucleoside analogues. Using similar reaction conditions, the 2,6-
diethyl carboxylate ester of purine riboside (3b) was prepared in
good yield and without significant cleavage of the glycosidic bond
(Table 1, entries 7-11). Reaction times, however, were longer
compared with those of 2a possibly because of the insolubility
of the potassium salt of 2b in DMF. The utility of this reaction
was further demonstrated by the one-step synthesis of nebularine14
from reaction of 1,3,5-triazine (8)15 with the unprotected 2c and
the triacetyl-protected nebularine (9)16 from reaction of 8 with
2b (Scheme 3). Efficient synthesis of nebularine and its congeners
is of interest17 because of the unique biological activity predicted2
and observed for these compounds.18 However, few nebularine
analogues with carbon substituents at either the 2- or 6-positions
are described.5b Earlier efforts synthesized 6-substituted purine
analogues using either Eschenmoser’s sulfide contraction reac-
tion19 or a carbanion substitution reaction with 6-methylsulfonyl
purine nucleosides.5a More recently, analogues with carbon
substituents were prepared from 6-halopurines via palladium-
catalyzed coupling reactions.20 Our results suggest that a variety
of purine and purine nucleoside analogues can be prepared using
an efficient and versatile strategy entailing 5-amino-4-imidazole-
carboxylic acids and 1,3,5-triazines in a tandem decarboxylation/
Diels-Alder reaction.
a R groups: (a) Bn; (b) â-D-ribofuranosyl; (c) 2,3,5-tri-O-acetyl-â-D-
ribofuranosyl.
Scheme 2a
a Reagents and conditions: (a) NaOH (2 equiv), THF, EtOH, H2O;
(b) Ac2O, AcOH, 130 °C; (c) NH3, MeOH; (d) NaOH (1 equiv), THF,
EtOH, H2O.
Acknowledgment. We thank Mr. Joe Kopcho for providing 2b and
Ms. Sandra Casillas for communicating the synthesis of 2a.
Supporting Information Available: Experimental details and char-
acterization data. This material is available free of charge via the Internet
at http://pubs.acs.org.
Scheme 3
JA9842316
(12) Boger, D. L.; Honda, T.; Menezes, R. F.; Colletti, S. L.; Dang, Q.;
Yang, W. J. Am. Chem. Soc. 1994, 116, 82.
(13) Srivastava, P. C.; Mancuso, R. W.; Rousseau, R. J.; Robins, R. K. J.
Med. Chem. 1974, 17, 1207. Srivastava, P. C.; Ivanovics, G. A.; Rousseau,
R. J.; Robins, R. K. J. Org. Chem. 1975, 40, 2920. The potassium salt of 2b
was used because of the instability of the free acid.
(14) Isono, K.; Suzuki, S. J. Antibiot., Ser. A 1960, 13A, 270. Lofgren, N.;
Luning, B.; Hedstrom, H. Acta Chem. Scand. 1954, 8, 670.
(15) Boger, D. L.; Schumacher, J.; Mullican, M. D.; Patel, M.; Panek, J.
S. J. Org. Chem. 1982, 47, 2673.
(16) Iwamura, H.; Hashizume, T. J. Org. Chem. 1968, 33, 1796. Nair, V.;
Richardson, S. G. J. Org. Chem. 1980, 45, 3969.
(17) Secrist, J. A., III.; Shortnacy-Fowler, A.; Bennett, L. L., Jr.;
Montgomery, J. A. Nucleosides Nucleotides 1994, 13, 1017. Lonnberg, H.;
Lehikoinen, P. J. Org. Chem. 1984, 49, 4964. Nair, V.; Buenger, G. S.
Synthesis 1988, 848.
(18) Kati, W. M.; Wolfeden, R. Science (Washington, D.C.) 1989, 243,
1591. Lynch, T. P.; Paran, J. H.; Paterson, A. R. P. Cancer Res. 1981, 41,
560. Nair, V.; Wiechert, R. J. Bioorg. Chem. 1980, 9, 423. Miller, R. L.;
Adamczyk, D. L.; Miller, W. H.; Koszalka, G. W.; Rideout, J. L.; Beacham,
L. M., III.; Chao, E. Y.; Haggerty, J. J.; Krenitsky, T. A.; Elion, G. B. J.
Biol. Chem. 1979, 254, 2346. Janion, C.; Shugar, D. Acta Biochim. Pol. 1973,
20, 271. Simon, L. N.; Bauer, R. J.; Tolman, R. L.; Robins, R. K. Biochemistry
1970, 9, 573.
(19) Roth, M.; Dubs, P.; Gotschi, E.; Eschenmoser, A. HelV. Chim. Acta
1971, 54, 710. Yamane, A.; Inoue, H.; Ueda, T. Chem. Pharm. Bull. 1980,
28, 157. Vorbruggen, H.; Krolikiewicz, K. Angew. Chem., Int. Ed. Engl. 1976,
15, 689.
(20) Gundersen, L. L.; Bakkestuen, A. K.; Aasen, A. J.; Overas, H.; Rise,
F. Tetrahedron 1994, 50, 9743. Gundersen, L. L. Tetrahedron Lett. 1994, 35,
3155.
16
9-benzylpurine (4),11 was prepared by hydrolysis of both esters
with sodium hydroxide (THF/EtOH/H2O, 25 °C, 1 h) followed
by decarboxylation (AcOH/Ac2O, 130 °C, 24 h, 55%). Alterna-
tively, 3a was converted to 9-benzyl-2,6-bis(carbamoyl)purine (5)
via aminolysis (NH3/MeOH, 25 °C, 24 h, 99%). Selective
functionalization of the 2- or 6-ethoxycarbonyl group can be
readily achieved because of the large difference in reactivity
between the ester groups. For example, treatment of 3a with one
equivalent of sodium hydroxide followed by decarboxylation gave
9-benzyl-6-ethoxycarbonylpurine (6, 65%) with no detectable
9-benzyl-2-ethoxycarbonylpurine or 9-benzylpurine. Aminolysis
of 6 under the above-mentioned conditions gave 9-benzyl-6-
carbamoylpurine (7) in 72% yield. The regioselective modification
of the 2-ester group is attributed to the presence of the imidazole
ring annulated to the 4- and 5-positions of the pyrimidine nucleus
which sterically hinders nucleophilic attack to the 6-ester group.
(11) Montgomery, J. A.; Temple, C., Jr. J. Am. Chem. Soc. 1961, 83, 630.