Expanding the diversity of purine libraries

Article abstract of DOI:10.1016/S0040-4039(01)01925-6

In recent years, there has been a resurgence of interest in the synthesis of purine derivatives due to the discovery of purine-derived ligands for a variety of nucleotide dependent enzymes. The majority of chemistry has focused on substitution of the purine core structure by alkylation at N9 and nucleophilic-aromatic substitution reactions at C2 and C6. Here we report the syntheses of aryl, N-aryl, O-aryl substituted purine libraries by the palladium-mediated coupling of boronic acids, anilines or phenols at the C2 position, and copper(II)-mediated N-arylation with boronic acids at the N9 position. The chemistry described here greatly expands our ability to introduce different functionality and create new purine scaffolds.

Full text of DOI:10.1016/S0040-4039(01)01925-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 8751–8755
Expanding the diversity of purine libraries
Sheng Ding,a Nathanael S. Gray,b,* Qiang Dingb and Peter G. Schultza,b,
*
aDepartment of Chemistry and the Skaggs Institute for Chemical Biology, The Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, CA 92037, USA
bGenomics Institute of the Novartis Research Foundation, 3115 Merryfield Row, San Diego, CA 92121, USA
Received 9 August 2001; revised 25 September 2001; accepted 27 September 2001
Abstract—In recent years, there has been a resurgence of interest in the synthesis of purine derivatives due to the discovery of
purine-derived ligands for a variety of nucleotide dependent enzymes. The majority of chemistry has focused on substitution of
the purine core structure by alkylation at N9 and nucleophilic–aromatic substitution reactions at C2 and C6. Here we report the
syntheses of aryl, N-aryl, O-aryl substituted purine libraries by the palladium-mediated coupling of boronic acids, anilines or
phenols at the C2 position, and copper(II)-mediated N-arylation with boronic acids at the N9 position. The chemistry described
here greatly expands our ability to introduce different functionality and create new purine scaffolds. © 2001 Published by Elsevier
Science Ltd.
1. Introduction
Previous syntheses of purine libraries have relied on
nucleophilic–aromatic substitution and alkylation
chemistry to derivatize the 2-, 6- and 9-positions of the
purine ring. One of the primary limitations of this
chemistry is the inability to access a large number of
pharmacologically relevant derivatives bearing aryl,
anilino or phenolic substituents. In addition, the slug-
gish aromatic substitution of 2-fluoro or 2-chloro sub-
stituted purine compounds precludes the introduction
of sterically hindered amines or anilines.3c,3d Here we
report our efforts to expand the types of reactions that
can be used to modify the purine ring to include
transition metal-catalyzed coupling reactions.
The purine ring is a ubiquitous recognition motif in
biological systems. Guanosine and adenosine, two of
the most common purines, serve as key recognition and
anchoring elements in a variety of cofactors and signal-
ing molecules (e.g. ATP, GTP, cAMP, cGMP, adoMet,
adenosine and NADH). Correspondingly, an enormous
number of proteins have evolved to recognize the
purine motif including reductases, polymerases, G-
proteins, methyltransferases, and protein kinases.
Despite the abundance of protein kinases1 (estimated to
be encoded by 2–5% of the eukaryotic genome) and the
high degree of conservation of active site residues,
ATP-binding site directed inhibitors have been designed
that are highly specific. For example, STI5712 has been
2. Results and discussion
developed as
a potent and selective Abl kinase
inhibitor, and is in use for the treatment of chronic
myelogenous leukemia (CML). Screens of purine
libraries3 have resulted in the identification of diverse
purines that inhibit mitosis, alter cellular morphology,
and induce apoptosis. By constructing new purine
derivatives, we hope to develop inhibitors of different
ATP-dependent proteins, which will be useful for eluci-
dating function and potentially provide starting points
for the development of new therapeutics.
2.1. Derivatization at the C2 position with boronic
acids, anilines or phenols by palladium-catalyzed cross-
coupling reactions using a ‘carbene’ ligand
Recently, there have been significant advances in
methodology for performing palladium-catalyzed CꢀC,
CꢀN and CꢀO bond formation reactions with a wide
variety of substrates. For example, new phosphine
ligands4 have allowed palladium-mediated functional-
ization of inexpensive chloroarenes with boronic acids
and amines at room temperature. 1,3-Dimesityl-imida-
zolin-2-ylidene and its saturated analog, originally
developed by Grubbs as carbene ligands for ruthenium-
based olefin metathesis catalysts,5 have also been found
to be highly effective ligands. We sought to investigate
* Corresponding authors. Tel.: 858-812-1624; fax: 858-812-1583;
e-mail: gray@gnf.org; schultz@scripps.edu
0040-4039/01/$ - see front matter © 2001 Published by Elsevier Science Ltd.
PII: S0040-4039(01)01925-6

8752
S. Ding et al. / Tetrahedron Letters 42 (2001) 8751–8755
whether a 2-chloropurine would react with boronic
acids, anilines and phenols under palladium-catalyzed
cross-coupling conditions using the sterically demand-
ing N-heterocyclic ‘carbene’ ligands 1 and 2. Nolan
(ligand 3)6 and Hartwig (ligand 4)7 have also explored
the use of similar carbene ligands system for a variety
of cross-coupling reactions.
propylpurine was coupled with boronic acids, anilines
and phenols (Table 1) with isolated final products over
90% yields. For the substrates bearing strong electron-
withdrawing groups or sterically very hindered ortho
substituents, the yields are relatively low. Not surpris-
ingly, substrates with halogens, especially those with
chloro or bromo substituents, usually result in complex
We found that all three substrates can be efficiently
coupled to the purine core using either ligand 1 or 2.
The typical conditions involved reaction of the 2-
chloropurine substrate with 1.5 equiv. of arylboronic
acid in the presence of Pd2(dba)3 (1.5 mol%), carbene
ligand 1 (3 mol%), and 2.0 equiv. of Cs2CO3 in anhy-
drous 1,4-dioxane under argon. The reactions are typi-
cally complete after stirring at 80°C for 12 h. For
coupling reactions to anilines and phenols potassium
t-butoxide was used as the base. Using these reaction
conditions8 2-chloro-6-(4-methoxybenzylamino)-9-iso-
mixtures. Only fluoro-substituted substrates can be cou-
pled to purine in good yield.
While we chose to validate the palladium-catalyzed
coupling reaction at the most unreactive position of the
purine ring, namely C2, it was also of interest to
determine whether boronic acids can be selectively cou-
pled with 2,6-dichloropurine at the C6 position. Indeed,
with 1 equiv. of boronic acid the C6 position can be
selectively functionalized with only a trace quantity
bis-functionalized product detected by LC–MS. The
Table 1. Palladium-catalyzed C2 substitution of purine
R3
OCH3
OCH3
HN
HN
HN
O
a
N
N
N
N
N
N
R3
R3
Cl
N
R
N
R=
a. 1.5 equiv of boronic acids, 2.0 equiv of Cs2CO3 / 1.5 equiv of anilines,
2.0 equiv of KOtBu, / 1.5 equiv of phenols, 2.0 equiv of KOtBu, 1.5% of
Pd2(dba)3, 3.0% of carbene ligand 1 or 2 , anhydrous dioxanes
R3
R1
R2
Entry
Entry
Entry
Yield (%)
96
Yield (%)
OCH3
97
Yield (%)
HN
HN
OCH3
O
CH3
5
90
90
1
9
H3CO
H3C
6
95
10
O
93
2
CH3
NH
O
7
8
94
93
94
91
11
12
3
4
90
91
O
H3C
N
1 via cross coupling with boronic acids; 2 via cross coupling with anilines; 3 via cross coupling with phenols

S. Ding et al. / Tetrahedron Letters 42 (2001) 8751–8755
8753
bulky phosphine ligands, 2-(di-t-butylphosphino)-
biphenyl and 2-(di-cyclohexylphosphino)biphenyl, were
also tested according to the conditions reported by
Buchwald4 and resulted in similar or lower yields for
certain substrates compared to the ‘carbene’ ligand.
afford the 9-arylpurine.9 While this procedure offers the
opportunity to simultaneously derivatize the C8 posi-
tion with select alkyl groups, it would be difficult to
prepare a N9-focused combinatorial library by this
method. The copper(II) mediated N-arylation of
heterocycles10 such as imidazole offers the possibility of
direct arylation of the purine N9 position. Here, we
demonstrate that 9-arylpurines can be prepared by
cupric acetate-mediated coupling of purine N9 with
arylboronic acids.
The choice of base has significant effects on the yield of
the coupling reaction.4 Although the precise details of
the reaction are unclear,6,7 it has been proposed that the
imidazolium salts are first deprotonated to form to a
carbene species which is then complexed with the tran-
sition metal. However, even a strong base such as
potassium tert-butoxide cannot deprotonate the imida-
zolium ligand 2 at room temperature and is instead
believed to add to the 2-position of the imidazolium
salt to form the 2-alkoxy-adduct.5b It would be even
more surprising if a much weaker base such as Cs2CO3
or CsHCO3, which are employed for the Suzuki cou-
pling reaction, can deprotonate the imidazolium salt at
room temperature. Certainly, further mechanistic stud-
ies are required to clarify the real active catalyst for this
palladium-catalyzed cross coupling reaction employing
imidazolium salt as ligands.
Reaction of 2,6-dichloropurine with boronic acids in
the presence of cupric acetate and triethylamine
resulted in the desired N9-aryl product11 (Table 2).
While the reaction appears to proceed to completion as
judged by LC–MS, isolation of the product from the
reaction mixture is difficult, presumably due to the
strong surface absorption of product to the crude reac-
tion materials. Although the purine is capable of form-
ing regioisomers (N9/N7), it was found in each case
that the N9 arylated product is the major regioisomer
(>90%) with the N7 regioisomer being produced in low
yield.
2.2. Derivatization of the purine N9 position with
boronic acids via cupric acetate catalyzed cross-
coupling reaction
In summary, we have demonstrated the versatility of
palladium and copper(II)-mediated coupling reactions
to expand the diversity of purine libraries. Focused
collections of purine compounds bearing CꢀC(sp2),
CꢀN-aryl and CꢀO-aryl bonds were synthesized by
palladium-mediated coupling of boronic acids, anilines
and phenols at the C2 position, and copper(II)-medi-
ated N-arylation with boronic acids at the N9 position.
The chemistry described here greatly expands our abil-
ity to introduce different functionality and create
hybrid aryl-purine scaffolds.
The most commonly used methods to modify the
purine N9 position include anionic alkylation with an
alkyl halide and strong base, Mitsunobu alkylation,
and various glycosylation reactions. None of these
methods allow synthesis of N9-aryl substituted purines.
Previously, N9-aryl purines were constructed by aniline
addition to 5-amino-6-chloro-pyrimidine followed by
cyclization with triethylorthoformate (or equivalent) to
Table 2. N9-arylation of purine
Cl
Cl
N
N
a
N
N
R'
N
Cl
N
N
H
B(OH)2
Cl
N
R'
a. anhydrous cupric acetate, 4 Å activated molecular
sieves, triethylamine,DCM.
Yield (%) Entry N9 aryl substituents Yield (%)
Entry N9 aryl substituents
3
4
45
43
1
47
44
CH3
H3CO
2
O
OCH3

8754
S. Ding et al. / Tetrahedron Letters 42 (2001) 8751–8755
evacuated and backfilled with argon and charged with
Acknowledgements
anhydrous 1,4-dioxane (2.0 mL). The reaction was stirred
under argon at 80°C and monitored by TLC. When the
reaction was complete after 8 h, the solvent was removed
in vacuo and the reaction crude was purified by flash
column chromatography (3% methanol in dichloro-
methane) to afford desired 2-(2,4-dimethoxyphenyl)-6-(4-
methoxybenzylamino)-9-isopropylpurine (207 mg, 96%).
1H NMR (400 MHz, CDCl3) l 1.59 (d, 6H, J=6.8 Hz),
3.79 (s, 3H), 3.84 (s, 3H), 3.85 (s, 3H), 4.87–4.96 (m, 3H),
6.14 (br, 1H), 6.57–6.60 (m, 2H), 6.85 (d, 2H, J=8.6 Hz),
7.34 (d, 2H, J=8.6 Hz), 7.73 (s, 1H), 7.78 (d, 1H, J=8.2
Hz); HRMS (MALDI-FTMS) [MH+] C24H28N5O3
434.2187, found: 434.2168.
(b) Aniline coupling: A 10 mL flame-dried Schlenk flask
equipped with a magnetic stir bar was charged with
2-chloro-6-(4-methoxybenzylamino)-9-isopropylpurine
(0.193 g, 0.5 mmol, 1.0 equiv.), 4-methoxyaniline (0.092
g, 0.75 mmol, 1.5 equiv.), Pd2(dba)3 (0.0069 g, 0.0075
mmol, 0.015 equiv.), ligand 1 (0.0051 g, 0.015 mmol, 0.03
equiv.) and KOtBu (0.112 g, 1.0 mmol, 2.0 equiv.). The
Schlenk flask was evacuated and backfilled with argon
and charged with anhydrous 1,4-dioxane (2.0 mL). The
reaction was stirred under argon at 80°C and monitored
by TLC. When the reaction was complete after 8 h, the
solvent was removed in vacuo and the reaction crude was
purified by flash column chromatography (3% methanol
in dichloromethane) to afford desired 2-(4-methoxy-
phenylamino) - 6 - (4 - methoxybenzylamino) - 9 - isopropyl-
purine (203 mg, 97%): 1H NMR (500 MHz, CDCl3) l
1.54 (d, 6H, J=6.6 Hz), 3.76 (s, 3H), 3.78 (s, 3H), 4.66
(m, 1H, J=6.6 Hz), 4.72 (br, 2H), 6.06 (br, 1H), 6.60 (br,
1H), 6.82 (d, 2H, J=13 Hz), 6.84 (d, 2H, J=13 Hz), 7.09
(s, 1H), 7.26 (d, 2H, J=8.8 Hz), 7.42 (s, 1H), 7.56 (d, 2H,
J=8.8 Hz); HRMS (MALDI-FTMS) [MH+] C23H27N6O2
419.2195, found: 419.2209.
Funding was provided by the Skaggs Institute for
Chemical Biology (P.G.S.), the Novartis Research
Foundation (N.S.G., Q.D.) and a predoctoral fellow-
ship from Howard Hughes Medical Institute (S.D.).
References
1. Venter, J. C.; et al. Science 2001, 291, 1304.
2. (a) Druker, B. J.; Tamura, S.; Buchdunger, E.; Ohno, S.;
Segal, G. M.; Fanning, S.; Zimmermann, J.; Lydon, N.
B. Nat. Med. 1996, 2, 561; (b) Zimmermann, J.; Buch-
dunger, E.; Mett, H.; Meyer, T.; Lydon, N. B. Bioorg.
Med. Chem. Lett. 1997, 7, 187; (c) Schindler, T.; Born-
mann, W.; Pellicena, P.; Miller, W. T.; Clarkson, B.;
Kuriyan, J. Science 2000, 289, 1938.
3. (a) Gray, N. S.; Wodicka, L.; Thunnissen, A.-M. W. H.;
Norman, T. C.; Kwon, S.; Espinoza, F. H.; Morgan, D.
O.; Barnes, G.; LeClerc, S.; Meijer, L.; Kim, S.-H.;
Lockhart, D. J.; Schultz, P. G. Science 1998, 281, 533 and
references cited therein; (b) Rosania, G. R.; Merlie, Jr.,
J.; Gray, N. S.; Chang, Y.-T.; Schultz, P. G. Proc. Natl.
Acad. Sci. USA 1999, 96, 4797; (c) Chang, Y.-T.; Gray,
N. S.; Chang, R. G. R.; Sutherlin, D. P.; Kwon, S.;
Norman, T. C.; Sarohia, R.; Leost, M.; Meijer, L.;
Schultz, P. G. Chem. Biol. 1999, 6, 361; (d) Amino
substituent at the purine C6 position significantly
decreases the reactivity of the 2-fluoro or 2-chloro group.
For example, complete displacement at C2 of a 2-fluoro-
6-benzylaminopurine with non-hindered primary amines
in solution requires heating at over 100°C for 12 h using
n-butanol as solvent. Displacement with acyclic sec-
ondary amines requires more forcing conditions. Under
similar substitution conditions, displacement with anilines
with reasonable yields was never achieved.
(c) Phenol coupling: A 10 mL flame-dried Schlenk flask
equipped with a magnetic stir bar was charged with
2-chloro-6-(4-methoxybenzylamino)-9-isopropylpurine
(0.193 g, 0.5 mmol, 1.0 equiv.), 4-methylphenol (0.081 g,
0.75 mmol, 1.5 equiv.), Pd2(dba)3 (0.0069 g, 0.0075 mmol,
0.015 equiv.), ligand 1 (0.0051 g, 0.015 mmol, 0.03 equiv.)
and KOtBu (0.112 g, 1.0 mmol, 2.0 equiv.). The Schlenk
flask was evacuated and backfilled with argon and
charged with anhydrous 1,4-dioxane (2.0 mL). The reac-
tion was stirred under argon at 90°C and monitored by
TLC. When the reaction was complete after 8 h, the
solvent was removed in vacuo and the reaction crude was
purified by flash column chromatography (2% methanol
in dichloromethane) to afford desired 2-(4-methylphen-
oxy)-6-(4-methoxy-benzylamino)-9-isopropylpurine (180
mg, 90%): 1H NMR (400 MHz, CDCl3) l 1.53 (d, 6H,
J=6.8 Hz), 2.37 (s, 3H), 3.79 (s, 3H), 4.54 (br, 2H), 4.71
(m, 1H, J=6.8 Hz), 6.35 (br, 1H), 6.81 (d, 2H, J=8.4
Hz), 7.10–7.18 (m, 6H), 7.65 (s, 1H); HRMS (MALDI-
FTMS) [MH+] C23H26N5O2 404.2081, found: 404.2080.
9. Cocuzza, A. J.; Chidester, D. R.; Culp, S.; Fitzgerald, L.;
Gilligan, P. Bioorg. Med. Chem. Lett. 1999, 9, 1063.
10. (a) Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.;
Winters, M. P.; Chan, D. M. T.; Combs, A. P. Tetra-
hedron Lett. 1998, 39, 2941; (b) Chan, D. M. T.; Monaca,
K. L.; Wanag, R.-P.; Winters, M. P. Tetrahedron Lett.
1998, 39, 2933; (c) Combs, A. P.; Saubern, S.; Rafalski,
4. (a) Wolfe, J. P.; Buchwald, S. L. Angew. Chem., Int. Ed.
1999, 38, 2413; (b) Wolfe, J. P.; Singer, R. A.; Yang, B.
H.; Buchwald, S. L. J. Am. Chem. Soc. 1999, 121, 9550;
(c) Stu¨rmer, R. Angew. Chem., Int. Ed. 1999, 38, 3307
and references cited therein; (d) Wolfe, J. P.; Tomori, H.;
Sadighi, J. P.; Yin, J.; Buchwald, S. L. J. Org. Chem.
2000, 65, 1158.
5. (a) Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs, R.
H. Tetrahedron Lett. 1999, 40, 2247; (b) Scholl, M.; Ding,
S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1, 953.
6. (a) Zhang, C.; Huang, J.; Trudell, M. L.; Nolan, S. P. J.
Org. Chem. 1999, 64, 3804; (b) Huang, J.; Grasa, G.;
Nolan, S. P. Org. Lett. 1999, 1, 1307.
7. Stauffer, S. R.; Lee, S.; Stambuli, J. P.; Hauck, S. I.;
Hartwig, J. F. Org. Lett. 2000, 2, 1423.
8. Representative experimental procedure of CꢀC, CꢀN and
CꢀO bond formation at C2/C6 of purine via palladium-
catalyzed cross-coupling reactions:
(a) Boronic acid coupling reactions: A 10 mL flame-dried
Schlenk flask equipped with a magnetic stir bar was
charged with 2-chloro-6-(4-methoxybenzylamino)-9-iso-
propylpurine (0.193 g, 0.5 mmol, 1.0 equiv.), 2,4-
dimethoxyphenylboronic acid (0.136 g, 0.75 mmol, 1.5
equiv.), Pd2(dba)3 (0.0069 g, 0.0075 mmol, 0.015 equiv.),
ligand 1 (0.0051 g, 0.015 mmol, 0.03 equiv.) and Cs2CO3
(0.326 g, 1.0 mmol, 2.0 equiv.). The Schlenk flask was

S. Ding et al. / Tetrahedron Letters 42 (2001) 8751–8755
8755
M.; Lam, P. Y. S. Tetrahedron Lett. 1999, 40, 1623; (d)
Lam, P. Y. S.; Deudon, S.; Averill, K. M.; Li, R.; He,
M. Y.; DeShong, P.; Clark, C. G. J. Am. Chem. Soc.
2000, 122, 7600; (e) Lam, P. Y. S.; Clark, C. G.;
Subern, S.; Adams, J.; Averill, K. M.; Chan, D. M. T.;
Combs, A. Synlett 2000, 5, 674.
(0.500 g), triethylamine (0.443 mL, 3.18 mmol, 3.0
equiv.) and dichloromethane (5.0 mL). The reaction
was stirred under air at ambient temperature and moni-
tored by TLC. When the reaction was complete after
24 h, it was filtered through Celite, washed with
methanol and purified by flash column chromatography
(1% methanol in dichloromethane) to afford desired
2,6-dichloro-9-(4-methylphenyl)-purine (0.136 g, 47%).
1H NMR (400 MHz, CDCl3) l 2.47 (s, 3H), 7.41 (d,
2H, J=8.1 Hz), 7.54 (d, 2H, J=8.1 Hz), 8.35 (s, 1H);
HRMS (MALDI-FTMS) C12H8Cl2N4 [MH+] 279.0199,
found: 279.0208.
11. Purine N9 arylation via boronic acids/cupric acetate: A
20 mL glass vial equipped with a magnetic stir bar was
charged with 2,6-dichloropurine (0.200 g, 1.06 mmol,
1.0 equiv.), 4-methylphenylboronic acid (0.288 g, 2.12
mmol, 2.0 equiv.), anhydrous cupric acetate (0.384 g,
,
2.12 mmol, 2.0 equiv.), 4 A activated molecular sieves