'Reductive Heck reaction' of 6-halopurines

Article abstract of DOI:10.1016/j.tetlet.2003.10.181

Alkenylation of 9-benzyl-6-halopurines with alkenes does not proceed under conventional Heck conditions. However, in the presence of triethylammonium formate a mixture of the corresponding saturated products was obtained. In contrast to the classical Heck reaction, the regioselectivity of this process is low giving a mixture of both α- and β-products regardless of the electronic nature of the substituent on the double bond of the alkene.

Full text of DOI:10.1016/j.tetlet.2003.10.181

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 273–276
ÔReductive Heck reactionÕ of 6-halopurines
*
ꢀꢁ
Tomas Tobrman and Dalimil Dvorak
ꢁꢀ
Department of Organic Chemistry, Prague Institute of Chemical Technology, Technickaꢀ 5, 166 28 Prague 6, Czech Republic
Received 28 August 2003; revised 23 October 2003; accepted 30 October 2003
Abstract—Alkenylation of 9-benzyl-6-halopurines with alkenes does not proceed under conventional Heck conditions. However, in
the presence of triethylammonium formate a mixture of the corresponding saturated products was obtained. In contrast to the
classical Heck reaction, the regioselectivity of this process is low giving a mixture of both a- and b-products regardless of the
electronic nature of the substituent on the double bond of the alkene.
Ó 2003 Elsevier Ltd. All rights reserved.
The Heck reaction, a Pd-catalyzed coupling reaction of
sp2-halides (or triflates) with olefins, has become an
indispensable tool for organic synthesis.1 Numerous
applications of this methodology have been described in
the field of heterocyclic chemistry.2 Somewhat surpris-
ingly, this reaction has not been successfully used for
derivatization of purines. Such a reaction would make
functionalized alkenyl purines with a wide range of
biological effects like cytokinin activity,3 activity against
Mycobacterium tuberculosis4 and antioxidant activity
easily accessible.5 However, to the best of our knowl-
edge, the reaction of 8-bromocaffeine with tert-butyl
acrylate is the only example of this type to have
appeared in the literature.6 Interestingly, the opposite
approach, reactions of 6-vinylpurine with organohalides
were successful.3
We have recently attempted to perform the Heck reac-
tion of 6-halo-9-benzylpurines with activated alkenes
like butyl acrylate and acrylonitrile, however, the reac-
tion did not proceed in any case, and the starting purine
derivatives remained unreacted. However, in the pres-
ence of TlOAc or AgOAc, N0-substituted hypoxanthine
derivatives were obtained.7 Lack of reactivity of
6-halopurines in the Heck reaction is unexpected, since
other Pd-catalyzed reactions, for example, Stille8a;b or
Suzuki8c;d couplings, proceeded smoothly.8e;f Examples
of low reactivity of 2-haloazines in the Heck reaction,
have been reported9a;b and were explained by the for-
mation of a stable dimer9a of the product of oxidative
addition.10 In the case of 9-benzyl-6-chloropurine the
corresponding dimer 1 was isolated and characterized.11
It was shown, that 1 reacts with tributylphenyltin giving
Ph
N
N
Y
Y
Cl
Pd
N
N
N
Ph3P
PdL(X)
PPh3
N
(X)LPd
Pd
Cl
N
N
N
N
N
N
N
N
N
N
Ph
Ph
Ph
2a
1
2b
Keywords: Heck reaction; Alkylpurine; Pd-catalysis.
* Corresponding author. Tel.: +42-022-435-4245; fax: +42-022-435-4288; e-mail: dvorakd@vscht.cz
0040-4039/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.10.181

274
T. Tobrman, D. Dvoꢁrꢀak / Tetrahedron Letters 45 (2004) 273–276
Table 1. Reaction of 6-halopurines with butyl acrylate 5aa (Scheme 2)
Y
Entry
Halopurine Catalyst
Product
(% yield)b
R
N
c
1. Transmetallation
2. Red. elimination
N
1
2
3
4
5
6
7
4a
4a
4a
4a
4a
4a
4a
Pd(PPh3)2Cl2
3a (22)
6a (25)
3a (22)
6a (53)
3a (25)
6a (48)
3a (25)
6a (54)
3a (26)
6a (50)
3a (15)
6a (24)
3a (10)
6a (17)
Traces of 3a and 6af
––f
2a or 2b
+
"M-R"
N
N
Pd(PPh3)2Cl2
Pd(PPh3)2Cl2
Pd(PPh3)4
Ph
d
3
Scheme 1.
e
Pd(OAc)2, PPh3
the product of Stille coupling and might therefore be an
intermediate for Stille reactions of 6-halopurines.11
Another process which would exclude palladium from
the catalytic cycle might be the formation of a stable
chelate either with N7 2a or N1 2b,12 after alkene
insertion.
e
Pd(OAc)2, P(OPh)3
Pd(OAc)2,
e
P(2-furyl)3
Pd(OAc)2, P(Cy)3
Pd(OAc)2, dpppg
Pd(OAc)2
e
8
4a
4a
4a
4a
9
10
11
––f
Pd(OAc)2
P(2-biphenyl)Cy2
Pd(PPh3)2Cl2
3a (35)
6a (15)
3a (27)
6a (46)
––f
In the latter case, reaction with an appropriate trans-
metallating reagent might result in the formation of
saturated derivatives 3 via transmetallation and reduc-
tive elimination (Scheme 1). Herein we report on our
attempts to use this approach for the preparation of
6-alkylpurine derivatives.
e
12
13
4b
4c
Pd(PPh3)2Cl2
a Reaction conditions: 5 mol % of Pd, 4 equiv of alkene, 8 equiv Et3N,
6 equiv HCO2H, DMF, 100 °C.
b Isolated yield.
c HCO2Na was used instead of HCO2H.
d Reaction in CH3CN at 80 °C.
e Pd/phosphine 1:2.
f Almost complete dehalogenation was observed.
g Pd/dppp 1:1.
Reaction of 9-benzyl-6-iodopurine 4a8b with butyl
acrylate 5a under the conditions of the Heck reaction
(Pd(PPh3)2Cl2, DMF, Et3N) in the presence of phenyl-
boronic acid and phenyltributyltin was unsuccessful.13
However, when the same conditions were applied but in
the presence of sodium formate, a mixture of the satu-
rated products 3a and 6a in an approximately 1:1 ratio
was obtained in 47% overall yield (Table 1, entry 1).
With triethylammonium formate formed in situ the ratio
3a to 6a reached 1:2 in 75% overall yield (Scheme 2,
Table 1, entry 2).
purine 4b8b reacted like iodopurine 4a, however, the
reaction was slower (Table 1, entry 12). 7-Benzyl-6-
iodopurine 4c8b gave exclusively the product of dehalo-
genation (Table 1, entry 13).
PdCl2(PPh3)2 was used as the catalyst in the reaction
with other substrates (Scheme 3).14 While substitution of
the double bond in its a-position resulted in a sub-
stantial reduction in yield (methyl methacrylate 5b,
Table 2, entry 1), introduction of a methyl group at the
b-position (methyl crotonate 5c, Table 2, entry 2) led
mainly to dehalogenated product. Among other electron
acceptor substituted alkenes, methyl vinyl ketone 5d
produced only the b-substituted product in a fair yield
(Table 2, entry 3), while reaction of acrylonitrile 5e was
more complicated. Besides the expected products 3e and
6e, the product of Michael addition of 6e to acrylonitrile
7 was formed (Table 2, entry 4). Evidently, the strong
electron deficiency of the purine nucleus taken together
Reaction in acetonitrile gave similar results to DMF
(Table 1, entry 3). A significant effect of the phosphine
ligand on the course of reaction was observed. While
Pd(PPh3)2Cl2, Pd(PPh3)4 and Pd(OAc)2/PPh3 (2 equiv)
in DMF gave almost identical results (Table 1, entries 3–
5), the combination of Pd(OAc)2 and P(OPh)3 (Table 1,
entry 6) or P(2-furyl)3 (Table 1, entry 7) gave lower
overall yields. A mixture of Pd(OAc)2 with the more
electron donating P(Cy)3 (Table 1, entry 8), chelating
dppp (Table 1, entry 9) as well as Pd(OAc)2 without
phosphine ligand (Table 1, entry 10) gave only the
product of dehydrohalogenation. In contrast to other
ligands, 2-(dicyclohexylphosphino)biphenyl gave 3a as
the main product (Table 1, entry 11). 9-Benzyl-6-chloro-
CO2Bu
CO2Bu
I
N
N
N
N
N
Et3N, HCO2H, DMF
Pd(PPh3)2Cl2, 100 °C
N
N
N
+
CO2Bu
5a
+
N
N
N
N
Ph
Ph
Ph
4a
6a (53%)
3a (22%)
Scheme 2.

T. Tobrman, D. Dvoꢁrꢀak / Tetrahedron Letters 45 (2004) 273–276
275
X
R1
N
R1
X
R2
N
R2
I
PdCl2(PPh3)3
Et3N, HCO2H
X
N
N
N
N
N
N
N
+
R2
+
DMF, 100 °C
or
MeCN, 80 °C
N
R1
5b-g
N
N
Ph
Ph
Ph
4a
3b-g
6b-g
Scheme 3.
Table 2. Reaction of 9-benzyl-6-iodopurine 4a with alkenesa
(Scheme 3)
CCRF-CEM cell line (ATCC CCL 119). None of the
compounds tested showed any significant activity.
Entry
1
Alkene
Product
(% yield)b
In conclusion, 6-halopurines react with terminal alkenes
under the conditions of the Heck reaction in the pres-
ence of formate giving saturated analogues of the
expected Heck reaction products.16 In contrast to the
usual outcome of Heck reactions, regioselectivity in this
process is low, and a mixture of linear and branched
products was formed in most cases. Despite the rela-
tively low yields and selectivity this reaction allows rapid
access to purine derivatives that are not easily accessible
by other methods. The scope and mechanistic aspects of
this reaction as well as other activity tests are now under
study in our laboratory.
5b
X ¼ CO2CH3; R1 ¼ H;
3b (16)
6b (9)
R2 ¼ CH3
2
5c
X ¼ CO2CH3; R1 ¼ CH3;
9-Benzylpurine (80)
R2 ¼ H
3
4
5d
X ¼ COCH3; R1 ¼ R2 ¼ H
3d (34)c
3d (25)c;d
3e (22)
6e (23)
7 (30)
5e
X ¼ CN; R1 ¼ R2 ¼ H
5
6
5f
3f (24)
6f (27)
6g (29)
8 (12)
X ¼ Ph; R1 ¼ R2 ¼ H
5g
X ¼ OAc; R1 ¼ R2 ¼ H
a Reaction conditions: 5 mol % of Pd, 4 equiv of alkene, 8 equiv Et3N,
6 equiv HCO2H, DMF, 100 °C.
b Isolated yield.
c NMR yield.
d Reaction in acetonitrile at 80 °C.
Acknowledgements
Financial support from the Grant Agency of the Czech
Republic (grant no. 203/03/0035) is gratefully acknowl-
edged.
with the nitrile group of 6e makes the a-hydrogen in 6e
acidic enough to be deprotonated by triethylamine.
Styrene 5f reacted smoothly giving 51% of a mixture of
both linear 3f and branched 6f products in an approxi-
mately 1:1 ratio (Table 2, entry 5). When vinyl acetate
5g, as an example of an electron rich alkene, was used,
the starting purine 4a was consumed in less then 1 h and
a-substituted product 6g was formed accompanied by
9-benzyl-6-ethylpurine 8 (Table 2, entry 6).15
References and Notes
€
1. For recent reviews, see: (a) Soderberg, B. C. In Compre-
hensive Organometallic Chemistry II; Abel, E. W., Stone,
F. G. A., Wilkinson, G., Eds.; Pergamon: Oxford, UK,
1995; Vol. 12, pp 241–297; (b) Overman, L. E. Pure Appl.
Chem. 1994, 66, 1423–1430; (c) deMeijere, E.; Meyer, F. E.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2379–2411; (d)
Cabri, W.; Candiani, I. Acc. Chem. Res. 1995, 28, 2–7.
2. For examples, see: Li, J. J.; Gribble, G. W. Palladium in
Heterocyclic Chemistry. In Tetrahedron Organic Chemis-
try Series; Baldwin, J. E., Williams, R. M., Eds.;
Pergamon: Oxford, UK, 2000.
CN
NC
C2H5
€
3. Brathe, A.; Gundersen, L.-L.; Rise, F.; Eriksen, A. B.;
Vollsnes, A. V.; Wang, L. Tetrahedron 1999, 55, 211–228.
4. Bakkestuen, A. K.; Gundersen, L.-L.; Langli, G.; Liu, F.;
N
N
N
N
N
N
N
N
€
Nolsoe, J. M. J. Bioorg. Med. Chem. Lett. 2000, 10, 1207–
1210.
Ph
Ph
7
8
€
5. Brathe, A.; Andersen, G.; Gundersen, L.-L.; Malterud, K.
E.; Rise, F. Bioorg. Med. Chem. 2002, 10, 1581–1586.
6. Jacobson, K. A.; Shi, D.; Gallo-Rodriguez, C.; Manning,
€
M.; Muller, C.; Daly, J. W.; Meumayer, J. L.; Kiriasis, L.;
Pfleiderer, W. J. Med. Chem. 1993, 36, 2639–2644.
The compounds prepared were tested for cytostatic
activity by evaluating their in vitro inhibition of cell
growth in mouse leukaemia L1210 cells (ATCC CCL
219); human promyelocytic leukaemia HL60 cells
(ATCC CCL 240); human cervix carcinoma HeLa S3
cells (ATCC CCL 2.2) and human T lymphoblastoid
ꢀ
ꢀ
ꢁꢀ
7. Havelkova, M.; Studenovsky, M.; Dvorak, D. Collect.
Czech. Chem. Commun. 2000, 65, 797–804.
8. (a) Gundersen, L.-L. Tetrahedron Lett. 1994, 35, 3155–
3158; (b) Gundersen, L.-L.; Bakkestuen, A. K.; Aasen, A.
ꢂ
J.; Øveras; Rise, F. Tetrahedron 1994, 50, 9743–9756;

276
T. Tobrman, D. Dvoꢁrꢀak / Tetrahedron Letters 45 (2004) 273–276
ꢁ
(c) Havelkova, M.; Hocek, M.; Cesnek, M.; Dvorak, D.
ꢀ
ꢁꢀ
ethyl acetate 1:2 gave 6a (84 mg, 50%) and further elution
with ethyl acetate/methanol 9:1 afforded 3a (44 mg, 26%).
3a: White amorphous solid; H NMR (500 MHz, CDCl3)
d 0.89(t, J ¼ 7:4 Hz, 3H, CH3), 1.32 (m, 2H, CH2CH3),
1.57 (m, 2H, CH2CH2CH3), 3.00 (t, J ¼ 7:4 Hz, 2H,
PuCH2CH2CO), 3.55 (t, J ¼ 7:4 Hz, 2H, PuCH2CH2CO),
4.08 (t, J ¼ 6:6 Hz, 2H, CO2CH2CH2 CH2), 5.44 (s, 2H,
CH2Ph), 7.28–7.40 (m, 5H, PhH), 8.01 (s, 1H, H-8), 8.90
(s, 1H, H-2); 13C NMR (125 MHz, CDCl3) d 13.6 (CH3),
ꢀ
ꢁꢀ
Synlett 1999, 1145–1147; (d) Havelkova, M.; Dvorak, D.;
Hocek, M. Synthesis 2001, 1704–1710; (e) For a recent
review on Pd and other transition metal mediated synthe-
ses of purine derivatives, see: Hocek, M. Eur. J. Org.
Chem. 2003, 23, 245–254; (f) For a recent review on Pd-
catalyzed syntheses of nucleosides, see: Agrofoglio, L. A.;
Gillaizeau, I.; Saito, Y. Chem. Rev. 2003, 103, 1875–1916.
1
9. (a) Bozell, J. J.; Vogt, C. E.; Gozum, J. J. Org. Chem.
1991, 56, 2584–2587; (b) Schnyder, A.; Aemmer, T.;
Indolese, A. F.; Pittelkow, U.; Studer, M. Adv. Synth.
Catal. 2002, 344, 495–498.
19.0 (CH2CH3), 27.8 (PuCH2
CH2CO), 30.6
(CH2CH2CH3), 31.5 (PuCH2CH2CO), 47.3 (CH2Ph),
64.4 (OCH2CH2CH2CH3), 127.8 (CH-Ph), 128.6 (CH-
Ph), 129.1 (CH-Ph), 132.4 (C5-Pu), 135.1 (C-Ph), 143.6
(C8-Pu), 150.7 (C4-Pu), 152.5 (C2-Pu), 160.6 (C6-Pu),
172.3 (CO2Bu); IR (CHCl3) 3021, 2965, 1730, 1596, 1499,
1407, 1333, 1209, 1190, 699 cmꢀ1. HR-MS (EI) calcd for
C19H22N4O2 338.1743, found 338.1736. 6a: White amor-
phous solid; 1H NMR (500 MHz, CDCl3) d 0.82 (t,
J ¼ 7:4 Hz, 3H, CH2CH3), 1.20 (m, 2H, CH2CH3), 1.52
(m, 2H, CH2CH2CH3), 1.69(d, J ¼ 7:1 Hz, 3H, CHCH3),
4.12 (t, J ¼ 6:3 Hz, 2H, CO2CH2 CH2CH2), 4.59(q,
J ¼ 7:1 Hz, 1H, CHCH3), 5.44 (s, 2H, CH2Ph), 7.29–7.38
(m, 5H, PhH), 8.03 (s, 1H, H-8), 8.95 (s, 1H, H-2); 13C
10. For examples, see: (a) Nakatsu, K.; Kinoshita, K.; Kanda,
H.; Isobe, K.; Nakamura, Y.; Kawaguchi, S. Chem. Lett.
1980, 23, 913–914; (b) Isobe, K.; Kawaguchi, S. Hetero-
cycles 1981, 16, 1603–1612; (c) Yamamoto, Y.; Yanagi, A.
Chem. Pharm. Bull. 1982, 30, 2003–2010; (d) Mantovani,
A.; Crociani, B. J. Organomet. Chem. 1982, 236, C37–C40;
(e) Crociani, B.; Di Bianca, F.; Giovenco, A.; Scrivanti, A.
J. Organomet. Chem. 1983, 251, 393–441; (f) Crociani, B.;
Di Bianca, F.; Giovenco, A.; Scrivanti, A. J. Organomet.
Chem. 1985, 291, 259–271; (g) Bertani, R.; Berton, A.; Di
Bianca, F.; Crociani, B. J. Organomet. Chem. 1986, 303,
283–299; (h) Urata, H.; Tanaka, M.; Fuchikami, T. Chem.
Lett. 1987, 23, 751–754; (i) Benneche, T. Acta Chem.
Scand. 1990, 44, 927–931.
11. Gundersen, L.-L. Acta Chem. Scand. 1996, 50, 462–465.
12. Similar chelates were found to be responsible for the
failure of the carbonylation of aliphatic amines: Hegedus,
L. S.; McKearin, J. M. J. Am. Chem. Soc. 1982, 104, 2444–
2451.
13. With phenyltributyltin, the product of Stille coupling
(9-benzyl-6-phenylpurine) was formed, while no reaction
took place with phenylboronic acid.
14. Representative procedure: To a mixture of 4a (168 mg,
0.5 mmol), Pd(OAc)2 (6 mg, 0.025 mmol) and PPh3 (13 mg,
0.05 mmol), degassed DMF (3 mL) was added under
argon. Then 5a (0.29mL, 2 mmol), Et 3N (0.55 mL,
4 mmol) and HCO2H (0.11 mL, 3 mmol) was successively
added via syringe and the mixture was stirred for 1 h at
100 °C. During this time all starting 4a disappeared (TLC).
The solvent was then evaporated in vacuum and the
residue chromatographed on silica. Elution with heptane/
NMR (125 MHz, CDCl3)
d 13.0 (CH2CH3), 15.9
(CHCH3), 18.9( CH2CH3), 30.4 (CH2CH2CH2CH3), 43.3
(CHCH3), 47.3 (CH2Ph), 65.0 (OCH2CH2CH2CH3), 127.9
(CH-Ph), 128.6 (CH-Ph), 129.1 (CH-Ph), 131.9 (C5-Pu),
135.0 (C-Ph), 144.0 (C8-Pu), 151.3 (C4-Pu), 152.7 (C2-Pu),
160.0 (C6-Pu), 172.3 (CO2Bu); IR (CHCl3) 2991, 2964,
1734, 1594, 1500, 1457, 1406, 1333, 1208, 1196 cmꢀ1. HR-
MS (EI) calcd for C19H22N4O2 338.1743, found 338.1735.
15. Formation of the ethyl derivative 8 remains unclear.
Acetoxy derivative 6g is not an intermediate, since it is
stable under the conditions used. In our opinion, the most
probable way for the formation of 8 involves formation of
a 6-vinylpurine derivative by b-elimination of acetate from
the product of insertion of vinyl acetate. The ethyl
derivative is then formed by Pd-catalyzed transfer hydro-
genation of the vinyl group.
16. 2-Bromopyridine, in accordance with Ref. 9a, did not
react with butyl acrylate under ÔclassicalÕ Heck conditions.
Addition of HCO2H led to the formation of the product
analogous to 3a, but the yield was very low (8%).