A concise and traceless linker strategy toward combinatorial libraries of 2,6,9-substituted purines

Full text of DOI:10.1021/jo016010f

J . Org. Chem. 2001, 66, 8273-8276
8273
Several solid- and solution-phase approaches for the
synthesis of purine analogues have been reported in the
literature over the past 5 years.4 One limitation of these
approaches is that one substituent is held invariant in
order to anchor the purine ring to the solid phase
(Scheme 1). To avoid this limitation, a “traceless” strategy
was desired that would be compatible with production-
scale library synthesis in spatially separate or divide-
recombine formats.
A Con cise a n d Tr a celess Lin k er Str a tegy
tow a r d Com bin a tor ia l Libr a r ies of
2,6,9-Su bstitu ted P u r in es
Sheng Ding,† Nathanael S. Gray,*,‡,§ Qiang Ding,‡ and
Peter G. Schultz*,†,‡
Department of Chemistry and the Skaggs Institute for
Chemical Biology, The Scripps Research Institute,
10550 North Torrey Pines Road, La J olla, California 92037,
and Genomics Institute of the Novartis Research
Foundation, 3115 Merryfield Row,
Another limitation of previous synthetic approaches4g
is the low reactivity of the 2-fluoro group once an amino
substituent has been installed at C6. For example,
complete displacement at C2 of a 2-fluoro-6-benzylami-
nopurine in solution requires heating at over 100 °C for
12 h using n-butanol as solvent. Complete aromatic
substitution of 2-fluoro or 2-chloro purine compounds on
solid support requires even higher temperatures and
often results in significant side reactions. This limits the
range of functional groups that can be installed at C2
and also creates difficulties in library production.
San Diego, California 92121
schultz@scripps.edu
Received August 6, 2001
In tr od u ction
Despite concerns that it would be extremely difficult
to design specific ATP competitive inhibitors of kinases,
there have been a number of success stories including
the p38 Map kinase, tyrosine kinases, and cyclin-depend-
ent kinases.1,3 Selective inhibitors of each of these kinases
are in various stages of clinical testing. The ability to
discriminate between extremely homologous kinases such
as CDK1 vs CDK2 has been demonstrated by the
development of novel thioflavopiridol derivatives that
display enhanced selectivity for CDK1 relative to CDK2.2
To date, a variety of heterocyclic scaffolds including
pyrimidines, indolines, pyrrolopyrimidines, indirubins,
purines, quinazolines, trisubstituted imidazoles, pyra-
zolopyrimidines, flavones, and anilinoquinolines have
been developed as kinase inhibitors.1,3 As each scaffold
presents unique opportunities for the presentation of
functional groups to the kinase active site, there is a need
for efficient and flexible methods for preparing libraries
of each of these inhibitor classes. We have chosen to focus
our development efforts toward the purine nucleus for
several reasons: (1) solid-phase and solution-phase pu-
rine chemistries have been sufficiently explored such that
unified schemes toward the derivatization of the 2-, 6-,
7-, 8-, and 9-positions4 should be possible; (2) purines
have been demonstrated to provide high-affinity ligands
for a variety of proteins; and (3) solid-phase methods used
to prepare purine libraries such as resin capture, nu-
cleophilic aromatic substitution reactions, Mitsunobu
alkylations, and palladium coupling reactions are readily
generalized to other heterocyclic systems of interest.
Resu lts a n d Discu ssion
We found that 6-amino-2-fluoro-9-alkylpurines react
with primary amines in methanol at room temperature.
With slightly more forcing conditions (in refluxing metha-
nol), sterically hindered amines such as the R-amino
group of arginine can be successfully introduced at the
C2-position (Scheme 2) with good yields. Unfortunately,
these conditions failed to translate to solid support,
presumably due to resin swelling problems in methanol.
Despite testing a range of solvent systems (NMP, DMF,
dioxane, DMSO, THF, and their combinations such as
DMFV/MeOHV 1/1), no solvent was found that allowed
complete substitution below 100 °C.
One possible solution to this problem involves C2
substition prior to substitution at C6. This requires
reversing the natural reactivity which favors initial
substitution at C6. We found that a C6 sulfenylpurine,
such as 2-fluoro-6-phenylsulfenyl (or a 6-benzylsulfenyl)
purine, directs quantitative and selective substitution by
an amine to the C2-position at 80 °C (Scheme 3). To
develop this as a combinatorial scheme, we envisioned
that we could subsequently substitute C6 after oxidation
of the thioether to the sulfone as has been demonstrated
in the synthesis of 2,4-diaminopyrimidine.5 The 2-fluoro-
6-thiophenylpurine was easily prepared by reacting
excess thiophenol with 2-fluoro-6-chloropurine in metha-
nol at 0 °C and then purifying by recrystallization.
We have previously demonstrated that the N9-position
can be alkylated on solid support under Mitsunobu
† The Scripps Research Institute.
‡ Genomics Institute of the Novartis Research Foundation.
(4) For 2-, 6-, 8-, or 9-substituted purine analogues, please see the
following. (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. (b) Chang, Y.-T.; Gray, N. S.; Chang, Rosania,
G. R.; Sutherlin, D. P. Kwon, S.; Norman, T. C.; Sarohia, R.; Leost,
M.; Meijer, L.; Schultz, P. G. Chem. Biol. 1999, 6, 361. (c) Lucrezia, R.
D.; Gilbert, I. H.; Floyd, C. D. J . Comb. Chem. 2000, 2, 249. (d) Nolsoe,
J . M. J .; Gundersen, L.-L.; Rise, F. Synth. Commun. 1998, 28, 4303.
For 7-substituted purine analogues, please see the following. (e) Dalby,
C.; Bleasdale, C.; Clegg, W.; Elsegood, M. R. J .; Golding, B. T. Angew.
Chem., Int. Ed. Engl. 1993, 32, 1696. (f) Zaitseva, G. V.; Sivets, G. G.;
Kazimierczuk, Z.; Vilpo, J . A.; Mikhailopulo, I. A. Bioorg. Med. Lett.
1995, 5, 2999. (g) Dorff, P. H.; Garigipati, R. S. Tetrahedron Lett. 2001,
42, 2771.
§ E-mail: gray@gnf.org.
(1) (a) McMahon, G.; Sun, L.; Liang, C.; Tang, C. Curr. Opin. Drug
Discovery Dev. 1998, 1, 131. (b) Adams, J . L.; Lee, D. Curr. Opin. Drug
Discovery Dev. 1999, 2, 96. (c) Cohen, P. Curr. Opin. Chem. Biol. 1999,
3, 459. (d) Garcia-Echeverria, C.; Traxler, P.; Evans, D. B. Med. Res.
Rev. 2000, 20, 28 and references therein.
(2) Kim, K. S.; Sack, J . S.; Tokarski, J . S.; Qian, L.; Chao, S. T.;
Leith, L.; Kelly, Y. F.; Misra, R. N.; Hunt, J . T.; Kimball S. D.;
Humphreys, W. G.; Wautlet, B. S.; Mulheron, J . G.; Webster, K. R. J .
Med. Chem. 2000, 43, 4126.
(3) (a) Druker, B. J .; Tamura, S.; Buchdunger, E.; Ohno, S.; Segal,
G. M.; Fanning, S., Zimmermann, J .; Lydon, N. B. Nature Med. 1996,
2, 561 (b) Taylor, S. S.; Radzio-Andzelm, E. Curr. Opin. Chem. Biol.
1997, 1, 2219 (c) Schindler, T.; Bornmann, W.; Pellicena, P.; Miller,
W. T.; Clarkson, B.; Kuriyan, J . Science 2000, 289, 1938.
(5) Gayo, L. M.; Suto, M. J . Tetrahedron Lett. 1997, 38, 211.
10.1021/jo016010f CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/31/2001

8274 J . Org. Chem., Vol. 66, No. 24, 2001
Notes
Sch em e 1. P r eviou sly Descr ibed Com bin a tor ia l Str a tegies for 2,6,9-Tr isu bstitu ted P u r in es
Sch em e 2. Cou p lin g Am in o Acid s to P u r in e C2-P osition
Sch em e 3. Su lfen yl Gr ou p P r otectin g P u r in e
C6-P osition a
fenyl-N9-alkylpurine directly from the crude reaction
mixture. This allows the moisture-sensitive Mitsunobu
reaction to be performed as the first combinatorial step
in solution, making the overall scheme more convergent.
To achieve a “traceless” linkage to solid support,
primary amines were coupled by reductive amination
using sodium triacetoxyborohydride to a 4-formyl-3,5-
dimethoxyphenoxymethyl-functionalized polystyrene resin
(PAL).6 The purine ring was then captured at the C2-
position by reacting the PAL-amine resin with 1.5 equiv
of the crude N9-alkylated 2-fluoro-6-phenylsulfenylpurine
and 3 equiv of diisopropylethylamine in n-butanol at 80
°C. The C6-position could then be substituted following
oxidation-activation of the thioether to the sulfone
(Scheme 4). The use of m-chloroperbenzoic acid to oxidize
the thioether linkage resulted in the premature cleavage
from solid support, presumably as a result of acid and
a
(a) 2 equiv of 4-methoxybenzylamine, 3 equiv of DiEA, BuOH,
80 °C.
conditions with a variety of alcohols. However, perform-
ing the N9-modification reaction on support had several
drawbacks, including incomplete alkylation with second-
ary alcohols, consumption of large excesses of reagent,
and inconvenient handling of reaction in a 96-well format.
To circumvent this problem, we devised a scheme whereby
a resin-bound amine is used to capture a C6-phenylsul-
Sch em e 4. Tr a celess Com bin a tor ia l Ap p r oa ch tow a r d 2,6,9-Tr isu bstitu ted P u r in e Libr a r ya
a
(a) 5 equiv of R2-NH2, 3 equiv of NaBH(OAc)3, 1% HOAc, THF; (b) 1.5 equiv of R2OH, 1.8 equiv of PPh3, 1.3 equiv of DiAD, THF, rt;
(c) 0.5 equiv of 1, 1.5 equiv of DiEA, BuOH, 80 °C; (d) 10 equiv of m-CPBA/NaOH (1:1), 1,4-dioxane with 10% H2O; (e) 2 equiv of R3-NH2,
anhydrous dioxane, 80 °C; (f) CH2Cl2; TFA:Me2S:H2O 45:45:5:5.

Notes
J . Org. Chem., Vol. 66, No. 24, 2001 8275
Ta ble 1. N-9 Su bstitu en t (R2) In tr od u ced by Alk yla tion ;
C-2 Su bstitu en t (R1) In tr od u ced th r ou gh Resin -Bou n d
Am in es; C-6 Su bstitu en t (R3) In tr od u ced by
Disp la cem en t w ith Am in esa
tamination from the N7 regioisomer. Alkylation can also
be performed with active alkyl bromides (such as tert-
butyl bromoacetate) or N9 can be temporally protected
with the THP group to provide additional diversity. C2
substituents are limited to primary amines (Table 1) due
to the need for a reaction site to attach the purine to solid
support. Immoblized anilines only result in partial
capture of the purine from solution. Because m-chloro-
perbenzoic acid is used to convert the C6 thioether to a
sulfone in the final activation step, substituents having
functional groups prone to oxidation cannot be used in
the first two derivatization steps. C6 displacement of the
sulfone works for primary amines, cyclic secondary
amines (such as piperazines), and electron-rich anilines
(Table 1). The displacement reaction fails for many
acyclic, sterically hindered secondary amines and electron-
poor anilines.
Con clu sion s
In summary, we report a concise and traceless linker
strategy toward making 2,6,9-trisubstituted combinato-
rial purine libraries. Resin-bound amines are used to
capture a C6-phenylmecapto-9-alkylated purine directly
from the crude Mitsunobu alkylation reaction mixture.
The C6-position is then subsituted following oxidation-
activation of the thioether to the sulfone. This strategy
overcomes the difficulty of C2 substitution by “directing”
the first substitution to C2 by using a phenylsulfenyl
group to protect the C6-position. A 1000-compound
combinatorial purine library has been synthesized using
this approach in a 96-well format and will be reported
elsewhere.
Exp er im en ta l Section
Gen er a l. N9 Mitsunobu alkylation reactions and C6 sulfone
displacement reactions were carried out under anhydrous condi-
tions under an argon atmosphere. C2 resin capture reactions
a
Except c, which was made by reacting purine with 3 equiv of
DHP and catalytic PPTS, and b, which was made by alkylation
with 3 equiv of tert-butyl bromoacetate and 4 equiv of K2CO3 in
DMF, all other substituents (R2) were introduced by Mitsunobu
alkylation. It should be noted that after final TFA cleavage, a, b,
and c gave rise to the corresponding deprotected forms.
were carried out in
4 mL scintillation glass vials unless
otherwise noted. Anhydrous tetrahydrofuran and 1,4-dioxane
were obtained by passing them through commercially available
alumina columns. All other reagents, resins, and solvents were
purchased at highest commercial quality and used without
further purification. Purity of compounds was assessed by
reverse-phase liquid chromatography-mass spectrometry (Agi-
lent Series 1100 LC-MS) with a UV detector at λ ) 255 nm
(reference at 360 nm) and an API-ES ionization source. NMR
spectra were recorded on Bruker 400 and 500 MHz instruments
and calibrated using residual undeuterated solvent as an
internal reference. The following abbreviations were used to
designate the multiplicities: s ) singlet, d ) doublet, t ) triplet,
q ) quartet, m ) multiplet. LC elution methods (using a
Phenomenex Luna 50 × 2.00 mm 5 µm C18 column): (1) 10 min
method, starting from 5% solvent A (acetonitrile) in solvent B
(water with 0.5% acetic acid) and running the gradient to 95%
A in 8 min, followed by 2 min elution with 95% A; (2) 16 min
method, starting from 5% solvent A (acetonitrile) in solvent B
(water with 0.5% acetic acid) and running the gradient to 95%
A in 14 min, followed by 2 min elution with 95% A.
oxidant sensitivity of the benzylic PAL-amine linkage.
This problem was overcome by performing the oxidation
in buffered solution with m-chloroperbenzoic acid that
had been neutralized with a stoichiometric amount of
sodium hydroxide.
This combinatorial scheme has the following merits:
(1) the difficulty of C2 substitution is overcome by
“directing” the first substitution to C2 by using a phen-
ylsulfenyl group as a “protective” group at the C6-
position; (2) N9 modification is accomplished in solution
and purified by resin-capture; (3) the PAL linker allows
traceless cleavage; and (4) construction of focused C6
purine libraries can readily be accomplished because this
is the last step in the combinatorial synthesis scheme.
The scope of the chemistry has been validated for a
range of substituents. Analysis of the final products
following the general protocol described above by LC-MS
revealed greater than 95% conversion of the starting
materials with crude HPLC purities over 85%. Most
primary and secondary alcohols lacking additional acidic
hydrogens work well in the Mitsunobu reaction at N9
(Table 1). As the Mitsunobu is performed on the C6-
phenylsulfenylpurine there is very little detectable con-
Gen er a l P r oced u r e for th e Com bin a tor ia l Syn th esis of
2,6,9-Tr isu bstitu ted P u r in es. (a ) Red u ctive Am in a tion for
th e Syn th esis of P AL-Resin -Bou n d Am in e (1). To a suspen-
sion of 4-formyl-3,5-dimethoxyphenoxymethyl-functionalized poly-
styrene resin (PAL) (10.0 g, 11.3 mmol) in DMF (350 mL) was
added a primary amine (56.5 mmol), followed by addition of
(6) (a) Albericio, F.; Kneib-Cordonier, N.; Biancalana, S.; Gera, L.;
Masada, R. I.; Hudson, D.; Barany, G. J . Org. Chem. 1990, 55, 3730.
(b) Boojamra, C. G.; Burow, K. M.; Thompson, L. A.; Ellman, J . A. J .
Org. Chem. 1997, 62, 1240. (c) J in, J .; Graybill, T. L.; Wang, M. A.;
Davis, L. D.; Moore, M. L. J . Comb. Chem. 2001, 3, 97.

8276 J . Org. Chem., Vol. 66, No. 24, 2001
Notes
sodium triacetoxyborohydride (7.18 g, 33.9 mmol) and acetic acid
(6.52 mL, 113 mmol). The mixture was shaken gently at room
temperature for 12 h and then washed with methanol (300 mL
× 4) and dichloromethane (300 mL × 4) and dried under
vacuum. The complete conversion of PAL aldehyde to resin-
bound amine was confirmed by disappearance of the aldehyde
stretch.
(b) 2-F lu or o-6-p h en ylsu lfen ylp u r in e (2). To a solution of
2-fluoro-6-chloropurine (10.0 g, 57.9 mmol) in methanol (200 mL)
was added diisopropylethylamine (25.2 mL, 144.7 mmol). The
mixture was cooled to 0 °C and followed by slow addition of
thiophenol (11.9 mL, 115.8 mmol) via an addition funnel over 1
h. The reaction was stirred at 0 °C for 12 h. The solvent was
then removed under reduced pressure, and the solid was
collected by filtration and washed twice with hexanes. The
collected solid was further purified by recrystallization from
methonal to afford the desired product (11.8 g, 83% yield). 1H
NMR (400 MHz, (CD3)2SO) δ 7.54 (m, 3H), 7.66(m, 2H), 8.49(s,
1H); MS C11H7FN4S [MH+] 246.04, found 247.05
under vacuum. The complete conversion of secondary amine
(PAL-amine) to tertiary amine was confirmed using the bro-
mophenol blue test.7
Activa tion of C6 by Oxid a tion of Th ioeth er to Su lfon e
(5). To a solution of m-CPBA (0.23 g, 75%, 1.0 mmol) in 1,4-
dioxane (9 mL) cooled to 0 °C was added an NaOH (1 mL, 1M,
1.0 mmol) aqueous solution, followed by addition of resin 4 (0.10
mmol). The suspension was shaken gently at room temperature.
After 8 h the resin was washed with methanol (3 mL × 4) and
dichloromethane (3 mL × 4) and dried under vacuum.
C6 Disp la cem en t w ith Am in es (6) a n d P r od u ct Clea v-
a ge (7). The resin 5 (0.05 mmol) was suspended in anhydrous
1,4-dioxane (0.6 mL), followed by addition of an amine (0.1
mmol). After overnight shaking at 80 °C, the resin was washed
with methanol (1 mL × 4) and dichloromethane (1 mL × 4) and
dried under vacuum to afford resin 6. Resin 6 was subsequently
cleaved using CH2Cl2:TFA:Me2S:H2O 45:45:5:5/v:v:v:v (0.5 mL)
to afford desired product 7 (in average >85% HPLC purity, 80%
purified yield).
(c) 2-F lu or o-6-p h en ylsu lfen yl-9-a lk ylp u r in e (3). To a
flame-dried round-bottom flask (500 mL) were added 2-fluoro-
6-phenylsulfenylpurine (10.0 g, 40.6 mmol), triphenylphosphine
(19.2 g, 73.1 mmol), and alcohol (52.8 mmol), followed by
dissolution in THF (anhydrous, 350 mL). The solution was cooled
to -30 °C, and diisopropyl azodicarboxylate (12.0 mL, 60.9
mmol) was added dropwise. The reaction was allowed to warm
to room temperature and stirred under argon. After overnight
stirring, the solvent was removed under reduced pressure and
the crude material was directly used in the next step without
further purification.
Resin Ca p tu r e of 3 a t C2 fr om Cr u d e Mitsu n obu Rea c-
tion 4. To a solution of crude 2-fluoro-6-phenylsulfenyl-9-
alkylpurine (0.15 mmol) in n-butanol (1.0 mL) was added PAL-
resin-bound amine 1 (0.10 mmol), followed by addition of
diisopropylethylamine (0.30 mmol). The suspension was heated
to 80 °C under argon. After 12 h, the resin was washed with
methanol (3 mL × 4) and dichloromethane (3 mL × 4) and dried
Ack n ow led gm en t. Funding was provided by the
Skaggs Institute for Chemical Biology (P.G.S.), the
Novartis Research Foundation (N.S.G., X.W., and Q.D.),
and a predoctoral fellowship from Howard Hughes
Medical Institute (S.D.).
Su p p or tin g In for m a tion Ava ila ble: Detailed experi-
mental procedures and spectra data of the compounds dis-
closed. This material is available free of charge via the Internet
at http://pubs.acs.org.
J O016010F
(7) (a) Krchn˜a´k, V.; Va´gner, J .; Lebl, M. Int. J . Petp. Protein Res.
1988, 32, 415-416. (b) Krchn˜a´k, V.; Va´gner, J .; Safa´r, P.; Lebl, M.
Collect. Czech. Chem. Commun. 1988, 53, 2542-2548