[2+2+2]-Co-cyclotrimerization 6-alkynylpurines with diynes: A method for preparation of 6-arylpurines

Article abstract of DOI:10.1016/S0040-4039(02)02643-6

A novel approach to 6-arylpurines based on [2+2+2]-co-cyclotrimerization of 6-alkynylpurines with various α,ω-diynes is described. Co-cyclotrimerization is catalyzed by Ni- and Co-phosphine catalysts and their choice depends on the substitution pattern of the both reactants.

Full text of DOI:10.1016/S0040-4039(02)02643-6

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 785–788
[
2+2+2]-Co-cyclotrimerization 6-alkynylpurines with diynes:
a method for preparation of 6-arylpurines
a
a,
b,
c
Pavel Turek, Martin Kotora, * Michal Hocek * and Ivana C ´ı sa rˇ ov a´
a
Department of Organic and Nuclear Chemistry, Faculty of Science, Charles University, Albertov 2030, 128 43 Prague,
Czech Republic
b
Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Flemingovo n a´ m. 2,
1
66 10 Prague 6, Czech Republic
c
Department of Inorganic Chemistry, Faculty of Science, Charles University, Albertov 2030, 128 43 Prague, Czech Republic
Received 18 September 2002; revised 15 November 2002; accepted 22 November 2002
Abstract—A novel approach to 6-arylpurines based on [2+2+2]-co-cyclotrimerization of 6-alkynylpurines with various a,v-diynes
is described. Co-cyclotrimerization is catalyzed by Ni- and Co-phosphine catalysts and their choice depends on the substitution
pattern of the both reactants. © 2003 Elsevier Science Ltd. All rights reserved.
6
-Arylpurine derivatives were recently found to exhibit
diverse types of biological activity: some substituted
-arylpurine bases are antagonists of corticotropin-
6-Arylpurines have been efficiently prepared by cross-
coupling reactions of 6-halopurines with various
organometallics. However, these methods are limited
by availability, reactivity and stability of the corre-
sponding aryl-organometallic reagent. Therefore, it is
still of general interest to develop alternative methods
for the preparation of some (in particular highly substi-
tuted) 6-arylpurines. One such hitherto unexplored and
6
6
1
releasing hormone or possess antimycobacterial and
2
antibacterial activity, while 6-arylpurine ribonu-
cleosides are potent cytostatics. Moreover, 6-arylpuri-
nes were used as unnatural nucleobases in artificial
base-pairs and as covalent base-pair analogues. So
far, biological activity screening was limited to easily
available purines bearing simple aryl groups, while
those bearing highly substituted and/or functionalized
aryl moieties still remain to be explored.
3
4
5
potentially
interesting
strategy
is
[2+2+2]-co-
cyclotrimerization of 6-alkynylpurines with other alky-
nes or diynes. Herein we report on a novel approach to
the synthesis of 6-arylpurines based on transition-metal
Scheme 1.
Keywords: purines; nucleosides; cycloadditions; cyclotrimerizations; alkynes.
*
Corresponding authors. Tel.: +0420-2-21952277; fax: +0420-2-21952323 (M.K.); tel.: +0420-2-20183324; fax: +0420-2-33331271 (M.H.); e-mail:
kotora@natur.cuni.cz; hocek@marilyn.uochb.cas.cz
0
040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02643-6

7
86
P. Turek et al. / Tetrahedron Letters 44 (2003) 785–788
catalyzed co-cyclotrimerizations of 6-alkynylpurines 1
with a,v-diynes 2.
resulted in the formation of arylpurine 3ab in a low yield
of 10% after 24 h at 80°C. In this instance the use of other
catalysts gave inferior results. Also, its efficiency in the
co-cyclotrimerization of hexynylpurine 1b with diyne 2b
to give arylpurine 3bb was of the same magnitude (yield
10%). When a stoichiometric amount was used product
3bb was isolated in 67% yield. In further experiments the
use of Wilkinson’s catalyst was abandoned because of its
high price and our interest focused to other catalytic
systems.
Although cyclotrimerization can be affected by a number
of transition metal complexes, often the proper choice of
a catalyst determines the successful course of the reac-
tion. Generally speaking, alkynylpurines as a class of
compounds has been prepared only recently, and because
of that there is no information concerning their reactivity
with other alkynes in the presence of transition metal
catalysts. The only exception is the recent report of
homocyclotrimerization7 of 6-ethynylpurine into
tris(purin-6-yl)benzenes. The above mentioned reasons
prompted us to study cyclotrimerization of alkynylpuri-
The most promising catalytic system seems to be the one
based on Ni(0)–phosphine complexes, because Ni(cod)2/
2
PPh was the only catalyst able to homocyclotrimerize
3
7
7
b
1a. To avoid manipulation with rather unstable
nes 1 and diynes 2 to arylpurines 3 (Scheme 1) in the
Ni(cod) , which is sensitive to oxygen and protic solvents,
presence of various transition metal complexes such as
2
8
we decided to generate Ni(0) compounds in situ by the
reduction of Ni(II)–phosphine complexes with excess of
Zn powder. It is noteworthy that the ligation around the
nickel atom played a crucial role in the catalytic activity
of the complexes.
RhCl(PPh ) (Wilkinson’s catalyst), catalytic systems
3
3
9
10
composed of NiX (L) (L=PPh , dppe ) and a reduc-
2
n
3
1
1
tant (Zn), and rarely used CoBr(PPh ) . Some represen-
3
3
tative results are summarized in Table 1. The reactions
were usually carried out on a 0.1 mmol scale in the
presence of 10–20 mol% of the transition metal catalyst.
The best results were obtained when the Rh- and
Co-catalyzed reactions were carried out in toluene and
those with the Ni-catalysts in MeCN.
Two Ni(II) complexes were tested as potential catalysts:
NiX (PPh ) (X=Br or I) and NiBr (dppe). The best
2
3 2
2
result with the former catalyst was obtained in co-
cyclotrimerization of hexynylpurine 1b with diyne 2b,
which resulted in the formation of arylpurine 3bb in 48%.
In comparison with other catalysts its use gave superior
results for co-cyclotrimerizations of ethynylpurine 1a
with diyne 2e to give arylpurine 3ae, and hexynylpurine
The use of Wilkinson’s catalyst, which is known to
catalyze co-cyclotrimerization of various alkynes with
a,v-diynes was not very efficient. In comparison with Ni
and Co catalysts its superiority was demonstrated only
in the reaction of ethynylpurine 1a with diyne 2b that
1
3
b with diynes 2a and 2c to furnish arylpurines 3ba and
bc, respectively, albeit in low yields. The latter showed
highest activity for co-cyclotrimerization of hexynyl-
Table 1. Co-cyclotrimerization of alkynylpurines 1 with
a,v-diynes 2 into 3 catalyzed by various TM-complexes
purine 1b with diyne 2d that afforded arylpurine 3bd in
5
5% yield, and for the reaction of phenylethynylpurine
12
Catalysta
t (h)c
Yield (%)f
1c with diynes 2b and 2d to give arylpurines 3cb and
cd in 64 and 87%, respectively. The reaction of phenyl-
1
1
2
Product
3
a
2b
A
B
D
C
D
E
D
B
E
A
B
D
B
C
D
E
24d
72
72
96
96
72
96
96
96
3ab
3ab
3ab
3ad
3ad
3ad
3ae
3ba
3ba
3bb
3bb
3bb
3bc
3bd
3bd
3bd
3cb
3cb
3cd
3cd
3cd
10
7
0
12
7
55
7
ethynylpurine 1c with diyne 2b was also carried out in
THF under the same conditions with a similar result
(61% yield). It is important to note that when Ni(PPh )
3
no
4
1
3
1
a
2d
complex was prepared from Cp Ni and 4PPh
2
3
cyclotrimerization of ethynylpurine 1a with diyne 2b, or
hexynylpurine 1b with diynes 2a,b,c was observed.
1
1
a
b
2e
2a
11
0
Cyclopentadienylcobalt complexes are very efficient cat-
alysts for cyclotrimerization of alkynes, however, the use
of high reaction temperatures for a successful reaction is
required. Recently, it was shown that the cobalt analogue
d
1
b
2b
24
10
48
35
13
5
18
55
64
61
5
72
72
96
120
120
72
of Wilkinson’s catalyst CoBr(PPh ) can cyclotrimerize
1
1
b
b
2c
2d
3 3
alkynes to benzene derivatives in some cases even at room
11
temperature. This fact sparked our interest to apply this
complex for co-cyclotrimerization of alkynylpurines 1
with diynes 2. Indeed, its use enabled us to achieve the
reaction even in some cases where other catalysts failed
to give any products. The cobalt catalyst was very
efficient in co-cyclotrimerization of alkynylpurines 1a,b,
and c with diynes 2d and afforded arylpurines 3ad, 3bd
1
c
c
2b
2d
D
D
20
20
b
1
C
D
E
96
72
e
87
88
16
14
and 3cd in 55, 55, and 88% yields, respectively. For
comparison, the reaction of ethynylpurine 1a with diyne
a
A=RhCl(PPh ) , B=NiI (PPh ) /Zn, C=NiBr (PPh ) /Zn, D=
3
3
2
3 2
2
3 2
NiBr (dppe) /Zn, E=CoBr(PPh ) .
In THF.
2
2
3 3
b
c
d
e
f
2d catalyzed by NiI (PPh ) gave the product in 12%
2
3 2
yield, and the reaction of hexynylpurine 1b with diyne 2d
At 60°C unless otherwise mentioned.
0°C.
0°C.
Isolated yields.
catalyzed by NiBr (dppe) gave the product in 18% yield.
8
2
2
The structure of arylpurine 3ad was confirmed by a
15
single-crystal X-ray analysis (Fig. 1).

P. Turek et al. / Tetrahedron Letters 44 (2003) 785–788
787
comparable yield (52%) to the catalytic reaction.
Finally, the reaction of phenylethynylpurine 1c with
diyne 2b gave arylpurine 3cb in inferior yield (22%).
In order to check compatibility of the reaction condi-
tions with respect to purine nucleosides, tri-O-acetyl
protected 6-phenylethynylpurine nucleoside 4 was co-
cyclotrimerized with 2b and 2d in the presence of a
catalytic amount of Ni(dppe)Br /Zn catalyst (20 mol%,
2
6
0°C, 72 h) to give the corresponding arylpurines 5a
and 5b in yields of 35 and 47%, respectively (Scheme 2).
This study showed that the [2+2+2]-co-cyclotrimeriza-
Figure 1. View of molecular structure 3ad with the atom
numbering scheme. The thermal ellipsoids are drawn at 50%
probability level.
tion of alkynylpurines with diynes into arylpurines
could be successfully catalyzed by Ni and Co phosphine
12,14
complexes under mild and simple conditions.
The
yields varied from moderate to good and were strongly
dependent on the combination of substituents R and
linkers X and on the nature of catalytic system or
stoichiometric reagent. So far, no generally applicable
conditions were found but, for most combinations of
starting compounds, either Ni or Co catalysts could be
efficiently used. Some intriguing aspects of this reaction
still remain unclear: (i) the effect of ligation around the
central atom in the case of nickel catalysis; (ii) effect of
substitution of the linker in diynes.
Interestingly, for some combinations the use of any of
the catalysts did not result in the formation of benzene
derivatives. This prompted us to compare some results
with the reactions carried with a stoichiometric amount
of a catalyst. For that purpose NiI (PPh ) was chosen
because of its low cost. The results are summarized in
Table 2. Reactions of hexynylpurine 1b with diynes 2a
and 2c under stoichiometric conditions gave higher
yields of the corresponding products 3ba and 3bc than
under catalytic conditions. It is notable that the reac-
tion of ethynylpurine 1a with diynes 2b and
hexynylpurine 1b with 1,7-octadiyne 2f afforded
arylpurines 3ab and 3bf in 9 and 39% yields, respec-
tively. Under catalytic conditions the formation of nei-
ther of these products was detected. Reaction of
hexynylpurine 1b with diyne 2b gave arylpurine 3ab in
2
3 2
In conclusion, this first application of [2+2+2]-co-
cyclotrimerization methodology to purine systems
allows an efficient preparation of highly substituted
6
-phenylpurines (in particular purines bearing 2,4,5-
trisubstituted phenyl moieties) that would not be easily
accessible by standard cross-coupling reactions. Further
extension of this methodology to the synthesis of a
large series of purine bases and nucleosides, and biolog-
ical activity screening of these product is underway.
Table 2. Co-cyclotrimerization of 1 with 2 in the presence
of a stoichiometric amount of NiI (PPh )
2
3 2
1
2
t (h)a
Product
Yield (%)b
Acknowledgements
1
1
1
1
1
1
a
b
b
b
b
c
2b
2a
2b
2c
2f
72
48
48
24
24
48
3ab
3ba
3bb
3bc
3bf
3cb
9
29
52
53
39
22
This work is a part of a research project Z4 055 905. It
was supported by the Grant Agency of the Czech
Republic (grant No. 203/01/0863 to M.K., 203/00/0036
and 203/03/0035 to M.H.). The authors thank Iva
Ti sˇ lerov a´ for the measurement of NMR spectra.
2b
a
At 60°C.
Isolated yields.
b
References
1
. Cocuzza, A. J.; Chidester, D. R.; Culp, S.; Fitzgerald, L.;
Gilligan, P. Bioorg. Med. Chem. Lett. 1999, 9, 1063–1066.
. (a) Bakkestuen, A. K.; Gundersen, L.-L.; Langli, G.; Liu,
F.; Nolsøe, J. M. J. Bioorg. Med. Chem. Lett. 2000, 10,
2
1207–1210; (b) Andresen, G.; Gundersen, L.-L.; Nissen-
Meyer, J.; Rise, F.; Spilsberg, B. Bioorg. Med. Chem.
Lett. 2002, 12, 567–569; (c) Gundersen, L.-L.; Nissen-
Meyer, J.; Spilsberg, D. J. Med. Chem. 2002, 45, 1383–
1386.
3
. (a) Hocek, M.; Hol y´ , A.; Votruba, I.; Dvo rˇ a´ kov a´ , H. J.
Med. Chem. 2000, 43, 1817–1825; (b) Hocek, M.; Hol y´ ,
A.; Votruba, I.; Dvo rˇ a´ kov a´ , H. Collect. Czech. Chem.
Commun. 2000, 65, 1683–1697; (c) Hocek, M.; Hol y´ , A.;
Scheme 2.

7
88
P. Turek et al. / Tetrahedron Letters 44 (2003) 785–788
Votruba, I.; Dvo rˇ a´ kov a´ , H. Collect. Czech. Chem. Com-
heated to 60°C for 20 h. Column chromatography on
mun. 2001, 66, 483–499; (d) Hocek, M.; Hol y´ , A.;
Dvo rˇ a´ kov a´ , H. Collect. Czech. Chem. Commun. 2002, 67,
silica gel (1/1 hexane/EtOAc) afforded 35 mg (64%) of an
1
yellow oil: H NMR (400 MHz, CDCl ) l 1.27 (t, J=7.2
3
3
25–335.
Hz, 6H), 3.70 (s, 2H), 3.72 (s, 2H), 4.23 (q, J=7.2 Hz,
4
. (a) Fujiwara, T.; Kimoto, M.; Sugiyama, H.; Hirao, I.;
Yokoyama, S. Bioorg. Med. Chem. Lett. 2001, 11, 2221–
2
Yokogawa, T.; Okuni, T.; Nakayama, H.; Takio, K.;
Yabuki, T.; Kigawa, T.; Kodama, K.; Yokogawa, T.;
Nishikawa, K.; Yokoyama, S. Nature Biotechnol. 2002,
2
7
8
4
1
1
1
1
H), 5.37 (s, 2H), 7.06–7.14 (m, 5H), 7.14–7.18 (m, 2H),
.28–7.36 (m, 3H), 7.38 (s, 1H), 7.62 (s, 1H), 7.82 (s, 1H),
223; (b) Hirao, I.; Ohtsuki, T.; Fujiwara, T.; Mitsui, T.;
13
.87 (s, 1H); C NMR (100 MHz, CDCl ) l 13.96, 40.14,
3
0.36, 46.99, 60.57, 61.74, 126.21, 126.40, 127.44, 127.61,
28.38, 128.99, 129.15, 132.03, 133.39, 135.15, 139.28,
41.12, 141.31, 142.18, 144.03, 151.39, 152.34, 159.20,
2
0, 177–182.
. Havelkov a´ , M.; Dvo rˇ a´ k, D.; Hocek, M. Tetrahedron
002, 58, 7431–7435.
71.36; IR (CHCl ) w 3037, 3025, 2987, 1731, 1588, 1499,
3
5
6
447, 1369, 1329, 1230, 1217, 1214, 1196, 1159, 1069, 907
2
−
1
+
cm ; m/z (EI) 545 (M , 22), 473 (6), 399 (6), 91 (100);
. Examples; arylmagnesium halides: (a) Estep, K. G.; Josef,
K. A.; Bacon, E. R.; Carabates, P. M.; Rumney, S., IV;
Pilling, G. M.; Krafte, D. S.; Volberg, W. A.; Dillon, K.;
Dugrenier, N.; Briggs, G. M.; Canniff, P. C.; Gorczyca,
W. P.; Stankus, G. P.; Ezrin, A. M. J. Med. Chem. 1995,
HR-MS calcd for C H N O 546.2267, found 546.2274.
33
30
4
4
R (1/2 hexane/EtOAc)=0.40.
f
1
1
3. Leadbeater, L. N. J. Org. Chem. 2001, 66, 7539–7541.
4. Typical experimental procedure for the cobalt-complex
catalyzed
reaction.
9-Benzyl-6-(6-phenyl-1,3-dihy-
3
8, 2582–2595; arylzinc halides: (b) C
M.; Hol y´ , A. Collect. Czech. Chem. Commun. 2000, 65,
357–1373; arylstannanes: (c) Langli, G.; Gundersen, L.-
&
esnek, M.; Hocek,
droisobenzofuran-5-yl)-9H-purine 3cd. In an argon purged
glass vial a mixture of an alkynylpurine 1c (0.1 mmol), a
diyne 2d (0.1 mmol), and CoBr(PPh3)3 (18 mg, 0.02
mmol) in toluene (2 mL) was heated to 60°C for 16 h.
Column chromatography on silica gel (1/4 hexane/
1
L.; Rise, F. Tetrahedron 1996, 52, 5625–5638; arylboronic
acids: (d) Havelkov a´ , M.; Dvo rˇ a´ k, D.; Hocek, M. Syn-
thesis 2001, 1704–1710.
. (a) Hocek, M.; Star a´ , I. G.; Star y´ , I.; Dvo rˇ a´ kov a´ , D.
Tetrahedron Lett. 2001, 42, 519–521; (b) Hocek, M.;
Star a´ , I. G.; Star y´ , I.; Dvo rˇ a´ kov a´ , D. Collect. Czech.
Chem. Commun. 2002, 67, 1223–1235.
1
EtOAc) afforded 36 mg (88%) of a yellow oil: H NMR
7
8
(400 MHz, CDCl
(m, 5H), 7.16–7.22 (m, 2H), 7.32–7.38 (m, 3H), 7.41 (s,
) l 5.22 (s, 4H), 5.40 (s, 2H), 7.08–7.16
3
13
1H), 7.66 (s, 1H), 7.86 (s, 1H), 8.89 (s, 1H); C NMR
. (a) Grigg, R.; Scott, R.; Stevenson, P. J. Chem. Soc.,
Perkin Trans. 1 1988, 1357–1364; (b) McDonald, F. E.;
Zhu, H. Y.; Holmquist, C. R. J. Am. Chem. Soc. 1995,
(100 MHz, CDCl ) l 47.12, 73.39, 73.46, 123.19, 123.32,
126.48, 127.55, 127.75, 128.50, 129.07, 129.21, 132.12,
133.83, 135.09, 138.37, 141.09, 141.16, 141.68, 144.26,
3
1
17, 6605–6606; (c) Witulski, B.; Stengel, T. Angew.
151.49, 152.39, 158.95; IR (CHCl ) w 3021, 2989, 2859,
3
Chem., Int. Ed. Engl. 1999, 38, 2426–2430; (d) Witulski,
B.; Stengel, T.; Fernandez-Hernandez, J. M. Chem. Com-
mun. 2000, 1965–1966; (e) Kotha, S.; Mohanraja, K.;
Durani, S. Chem. Commun. 2000, 1910–1911; (f) Kotha,
S.; Sreenivasachary, N. Eur. J. Org. Chem. 2001, 3375–
1
8
589, 1499, 1455, 1438, 1332, 1216, 1213, 1121, 1048, 902,
55 cm ; m/z (EI) 404 (M , 6), 403 (11), 277 (100), 201
−1
+
(
24), 183 (20), 91 (25), 77 (44), 59 (58); HR-MS calcd for
C H N O 404.1637, found 404.1614. R (EtOAc)=0.32.
26
20
4
f
1
5. Crystal structure determination for 3ad. Single crystals
3
383.
were grown from hexane. Crystal data: C H N O, M=
3
1
U=1596.40(6) A
mm ; 23987 diffractions collected, 3646 unique diffrac-
20
16
4
9
. (a) Bhatarah, P.; Smith, E. H. J. Chem. Soc., Chem.
Commun 1991, 277–278; (b) Sato, Y.; Nishimata, T.;
Mori, M. J. Org. Chem. 1994, 59, 6133–6135; (c) Star a´ , I.
28.37, monoclinic; space group P2 /n; a=9.3510(2), b=
1
2.4700(3),
c=13.9450(3)
A
,
;
i=100.9640(14)°;
3
,
, T=150 K, Z=4, v(Mo Ka)=0.088
G.; Star y´ , I.; Koll a´ rovi cˇ , A.; Tepl y´ , F.; Vysko cˇ il, S
aman, D. Tetrahedron Lett. 1999, 40, 1993–1996.
&
.;
−
1
S
&
2
tions (Rint=0.026) used in all calculations, final R(F )=
0
1
1
1
0. Jeevanandam, A.; Korivi, R. J.; Huang, I.; Cheng, C.-H.
.0433 (for observed diffractions with I>2|(I)) and
Org. Lett. 2002, 4, 807–810.
1. Field, L. D.; Ward, A. J.; Turner, P. Aust. J. Chem. 1999,
2
wR(F )=0.1173 (for all data). Crystallographic data
excluding structure factors) for the structures in this
(
5
2, 1085–1092.
paper, have been deposited with the Cambridge Crystal-
lographic Data Centre as supplementary publication
numbers CCDC 192707. Copies of the data can be
obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (fax: +44(0)-
1223-336033 or e-mail: deposit@ccdc.cam.ac.uk).
2. Typical experimental procedure for nickel-complex cata-
lyzed reaction. 9-Benzyl-6-[6-phenyl-2,2-bis(ethoxycar-
bonyl)indan-5-yl]-9H-purine 3cb. In an argon purged glass
vial a mixture of an alkynylpurine 1c (0.1 mmol), a diyne
2
b (0.1 mmol), a Ni(dppe)Br (12 mg, 0.02 mmol), and
2
Zn powder (32 mg, 0.05 mmol) in MeCN (2 mL) was