Two simple protocols for the preparation of diallylaminoethyl-substituted nucleic bases: A comparison

Article abstract of DOI:10.1016/j.tetlet.2006.01.035

The syntheses of pyrimidine and purine nucleic bases substituted with diallylaminoethyl groups are reported following two different protocols. A comparison is made between the yield, expense, and difficulty of each route.

Full text of DOI:10.1016/j.tetlet.2006.01.035

Tetrahedron Letters 47 (2006) 1767–1770
Two simple protocols for the preparation of
diallylaminoethyl-substituted nucleic bases: a comparison
Rania S. Shatila and Kamal H. Bouhadir*
Department of Chemistry, American University of Beirut, Beirut 11-0236, Lebanon
Received 12 October 2005; revised 5 January 2006; accepted 11 January 2006
Available online 26 January 2006
Abstract—The syntheses of pyrimidine and purine nucleic bases substituted with diallylaminoethyl groups are reported following
two different protocols. A comparison is made between the yield, expense, and difficulty of each route.
Ó 2006 Elsevier Ltd. All rights reserved.
The importance of synthetic methods for the prepara-
tion of synthetic oligodeoxynucleotides (ODNs) has
increased remarkably during the last decade,1–3 because
of their potential use in therapeutic applications, such as
antisense and antigene, and diagnostic applications,
such as biosensors and microarrays.4 One critical
requirement for synthetic ODNs is their stability in bio-
logical environments, and hence, reasonable half-life
in vivo.5 Exo- and endonucleases cleave the phosphodi-
ester backbone in RNA and DNA; as a result, carbo-
cyclic ODN analogs should be less susceptible to
enzymatic fission than the furanosyl heterocycles.2,6
One possible approach to synthetic neutral ODNs is
the copolymerization of non-conjugated dienes contain-
ing nucleic bases with sulfur dioxide. Non-conjugated
dienes such as diethyl diallylmalonate have been copoly-
merized with sulfur dioxide to form polysulfones.7–9 We
have also demonstrated in an earlier publication10 that
1,6-heptadienes with nucleic bases attached directly to
C-4 can be copolymerized with sulfur dioxide to form
carbocyclic polynucleotides with sulfone groups replac-
ing the phosphodiester backbone and cyclopentane rings
substituting the furanose ring of nucleic acids. In addi-
tion, polysulfones have been formed via the copolymer-
ization of several N-substituted diallylamines with sulfur
dioxide.11–13 Sulfones are neutral, achiral, and isoelec-
tronic analogues of phosphodiesters and are stable to
both chemical and biochemical degradations.14 We report
herein, the preparation of a new series of nucleoside
analogs bearing a diallylaminoethyl group as potential
precursors to carbocyclic polynucleotide analogs.
Two protocols (Scheme 1) have been investigated for the
preparation of 1-(2-diallylaminoethyl)pyrimidine nucleic
bases starting with uracil and thymine. In the first pro-
tocol, uracil and thymine were reacted with ethylene car-
bonate in dry dimethylformamide and catalytic amounts
of sodium hydroxide following reported procedures15–18
to from 1-(2-hydroxyethyl)uracil (2a) and 1-(2-hydroxy-
ethyl)thymine (2b) in 30% and 33% yields respectively.
The yields of these reactions were low due to the forma-
tion of the 1,3-bis(2-hydroxyethyl) and 3-(2-hydroxy-
ethyl) derivatives, which were isolated from the desired
compounds via lengthy and time consuming ion-
exchange column chromatography.19 Compounds 2a
and 2b were reacted with thionyl chloride to afford
1-(2-chloroethyl)uracil (3a) and 1-(2-chloroethyl)thymine
(3b) in 91% and 72% yields.20 Finally, nucleophilic
displacement of the chloro substituent with diallylamine
resulted in the formation of 1-(2-diallylaminoethyl)ura-
cil (4a) and 1-(diallylaminoethyl)thymine (4b) in 62%
and 49% yields. The overall isolated yields of com-
pounds 4a and 4b from uracil and thymine were 14%
and 12%, respectively.28
An alternative route for the synthesis of compounds 4a
and 4b has been investigated utilizing the Mitsunobu
reaction (Scheme 1).21 The N3-protected uracil and thy-
mine intermediates 5a and 5b were prepared from uracil
(1a) and thymine (1b).22,23 Uracil and thymine were
allowed to react with benzoyl chloride to form the
N1,N3-dibenzoyl derivatives that were subsequently
hydrolyzed with aqueous potassium carbonate to yield
Keywords: Uracil; Thymine; Adenine; Nucleosides; Diallylamine;
Mitsunobu.
*
Corresponding author. Tel.: +961 1 350000; fax: +961 1 365217;
e-mail: kb05@aub.edu.lb
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.01.035

1768
R. S. Shatila, K. H. Bouhadir / Tetrahedron Letters 47 (2006) 1767–1770
O
O
O
O
N
R
R
R
R
HN
HN
HN
HN
ethylene carbonate
NaOH, dioxane
SOCl2
diallylamine
dioxane
O
N
H
O
N
O
N
O
pyridine
1a, R = H
1b, R = CH3
OH
Cl
N
2a, R = H, 30%
3a, R = H, 91%
2b, R = CH3, 33%
3b, R = CH3, 72%
(a) Benzoyl chloride,
pyridine, CH3CN
(b) K2CO3, dioxane
4a, R = H, 52%
4b, R = CH3, 49%
O
diallylamine,
dioxane,
R
∆
BzN
O
4a, 55%
4b, 46%
O
N
BrCH2CH2OH, DIAD,
Ph3P, dioxane
R
BzN
O
N
H
Br
6a, R = H
5a, R = H, 75%
6b, R = CH3
5b, R = CH3, 82%
Scheme 1.
the N3-benzoyluracil and N3-benzoylthymine in 75%
and 82% yields, respectively. The Mitsunobu reaction
was employed to synthesize intermediates 6a and 6b by
coupling 5a and 5b to 2-bromoethanol with triphenyl-
phosphine and diisopropyl azodicarboxylate (DIAD)
in dry dioxane.24 The crude intermediates 6a and 6b
were refluxed with diallylamine in dry dioxane to yield
4a and 4b in 55% and 46% yields, respectively (Scheme
1). Hence, the overall yields of 4a and 4b from uracil
and thymine were 41% and 38%, respectively. In com-
parison, the first route was relatively more economical
than the Mitsunobu route, which utilized the relatively
expensive DIAD reagent; however, the overall yields
of the Mitsunobu route were threefold for 4a and 4b.
Moreover, the first route engaged the use of an ion-
exchange chromatography, which necessitated the elution
of the desired products with large amounts of aqueous
solutions and resulted in prolonged work-up periods
for both compounds.
We have applied these simple protocols to the synthesis
of the cytosine derivative as well (Scheme 2). In the first
protocol, cytosine was reacted with ethylene carbonate
to give 1-(2-hydroxyethyl)cytosine (8) in 70% yield.25
Compound 8 was then reacted with thionyl chloride to
give 1-(2-chloroethyl)cytosine (9) in 80% yield.26 React-
ing compound 9 with diallylamine resulted in the forma-
tion of 1-(2-diallylaminoethyl)cytosine (10) in 33% yield.
The overall yield of compound 10 from cytosine was
18%.28 We have also prepared compound 10 following
the second protocol. First, N6-isobutyrylaminocytosine
(11) was prepared from cytosine (7) via refluxing with
NH2
NH2
N
NH2
N
NH2
N
N
N
N
N
SOCl2
diallylamine
dioxane
ethylene carbonate
NaOH, dioxane
O
N
H
O
O
O
pyridine
7
OH
8, 70%
Cl
9, 80%
N
isobutyric anhydride
10, 33%
DMF,
∆
NaOMe
MeOH
O
O
(10, 85%)
HN
HN
a) BrCH2CH2OH, DIAD,
Ph3P, dioxane
N
N
b) diallylamine, dioxane,
∆
O
N
O
N
H
11, 80%
N
12, 37%
Scheme 2.

R. S. Shatila, K. H. Bouhadir / Tetrahedron Letters 47 (2006) 1767–1770
1769
NH2
NH2
N
NH2
NH2
N
N
N
N
N
N
N
Ethylene carbonate
NaOH, dioxane
SOCl2
N
N
N
diallylamine
dioxane
N
pyridine
N
H
N
N
N
N
OH
Cl
13
14, 87%
15, 83%
16, 55%
Isobutyric anhydride
DMF,
NaOMe
MeOH
∆
(16, 77%)
O
O
HN
O
HN
HN
N
N
N
diallylamine
BrCH2CH2OH
DIAD, Ph3P, dioxane
N
N
N
N
dioxane, ∆
N
N
N
N
N
H
N
Br
17, 66%
18
19, 38%
Scheme 3.
isobutyric anhydride followed by selective hydrolysis
following a reported procedure.22 Afterwards, the Mitsu-
nobu reaction was utilized to couple intermediate 11 to
2-bromoethanol with triphenylphosphine and DIAD in
dry dioxane followed by nucleophilic substitution with
diallylamine to yield 1-(2-diallylaminoethyl)isobutyryl-
amidocytosine (12) in 37% yield.
In summary, we have synthesized three pyrimidine and
one purine nucleic bases with 2-diallylaminoethyl
groups attached following two different procedures.
The first procedure involved the reaction of the nucleic
bases with ethylene carbonate followed by chlorination
and nucleophilic displacement with diallylamine
whereas the second procedure employed protection of
the heterocyclic ring, Mitsunobu reaction and nucleo-
philic displacement with diallylamine followed by depro-
tection of the heterocyclic ring. The overall yields of the
uracil and thymine derivatives were more than tripled in
the second procedure, which rendered it more appropri-
ate for these compounds. In comparison, the cytosine
derivative was formed in comparable yields following
both protocols. In contrast, the adenine derivative was
obtained in higher yields following the first protocol
due to eliminating the ion-exchange chromatography
step.
Compound 12 was hydrolyzed in a methanolic solution
of sodium methoxide to yield 10 in 85% yield (Scheme
2). Hence, the overall yield of 10 in the second protocol
was 25% starting from cytosine. Both synthetic routes
were simple and convenient to perform and resulted in
comparable isolated yields; however, the first synthetic
protocol is more economical especially for large-scale
production of this compound.
In addition to the pyrimidine analogs, we have also pre-
pared 9-(2-diallylaminoethyl)adenine 16 following both
protocols (Scheme 3). In the first route, adenine was
refluxed with ethylene carbonate to yield 9-(2-hydroxy-
ethyl)adenine 14 in 87% yield. Intermediate 14 was
then chlorinated utilizing thionyl chloride in dry dioxane
to afford the chloro derivative 15 in 83% yield. Com-
pound 15 was reacted with diallylamine to give 9-(2-di-
allylaminoethyl)adenine in 55% yield (overall yield of 16
from adenine was 40%).28 Alternatively, adenine was
protected following reported procedures by reaction
with isobutyric anhydride followed by selective hydrol-
ysis to afford 6-isobutyryl adenine 17 in 66% yield.27
Compound 17 was coupled with bromoethanol via the
Mitsunobu reaction to form the bromoethyl derivative
18, which was not isolated. Compound 18 was refluxed
with diallylamine in dry dioxane to yield 6-isobutyryl-
9-(2-bromoethyl)adenine 19 in 38% yield. Hydrolysis
of the isobutyryl group with sodium methoxide gave
the 9-(2-diallylaminoethyl)adenine 16 in 77% yield
(19% overall yield from adenine). As a result, the overall
yield of 16 was more than twofold from the first proce-
dure in comparison to the Mitsunobu route.
Acknowledgements
The authors are grateful to the University Research
Board (URB) at the American University of Beirut
and the Lebanese National Council for Scientific
Research (LNCSR) for financial support.
References and notes
1. Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 543–584.
2. Agrawal, S. Biochim. Biophys. Acta 1999, 1489, 53–68.
3. Han, M. J.; Chang, J. Y. Adv. Polym. Sci. 2000, 153, 1–36.
4. Milligan, J. F.; Matteucci, M. D.; Martin, J. C. J. Med.
Chem. 1993, 36, 1923–1937.
5. Miller, K. J.; Das, S. D. Pharm. Sci. Technol. Today 1998,
1, 377–386.
6. Miller, P. S.; Hamma, T. Biochemistry 1999, 38, 15333–
15342.
7. Butler, G. B. Cyclopolymerization and Cyclocopolymeriza-
tion; Marcel Dekker: New York, 1992.

1770
R. S. Shatila, K. H. Bouhadir / Tetrahedron Letters 47 (2006) 1767–1770
8. Amemiya, Y.; Katayama, M.; Harada, S. Makromol.
(75 MHz, CDCl3) d 12.2 (q), 46.6 (t), 51.4 (t), 57.2 (t),
109.4 (d), 135.0 (d), 117.9 (t), 141.8 (d), 151.1 (s), 164.8 (s);
FTIR (KBr disk, cmÀ1) 3163, 3072, 3024, 2811, 1700,
1653, 1484.8, 1423, 915; HRMS (EI) calcd for
C13H19N3O2 (M+), 249.14773, found 249.14791.
Chem. 1977, 178, 2499–2505.
9. Tsai, J.; Shevlin, P. B. J. Org. Chem. 1998, 63, 3230–3234.
10. Tsai, J.; Bouhadir, K. H.; Zhou, J.; Webb, T. R.; Sun, Y.;
Shevlin, P. B. J. Org. Chem. 2003, 68, 1235–1241.
11. Harada, S.; Arai, K. Makromol. Chem. 1967, 107, 78–93.
12. Asrof, A.; Al-Muallem, H.; Mazumder, M. Polymer 2003,
44, 1671–1679.
13. Umar, Y.; Al-Muallem, H.; Abu-Sharkh, B. F.; Asrof, A.
Polymer 2004, 45, 3651–3661.
14. Huang, Z.; Schneider, K. C.; Benner, S. A. J. Org. Chem.
1991, 56, 3869–3882.
15. Tuppy, H.; Kuechler, E. Monatsh. Chem. 1964, 95, 1698–
1701.
16. Prystas, M.; Gut, J. Collect. Czech. Chem. Commun. 1962,
27, 1054.
17. Lapienis, G.; Penczek, S. J. Polym. Sci. Polym. Chem.
1990, 28, 1519–1990.
18. Ustyuzhanin, G. E.; Kolomeitseva, V. V.; Tichomirova-
Sidrova, N. S. Khim. Geterotsikl. Soedin. 1978, 684–689.
19. Pitha, J.; Ts’o, P. O. P. J. Org. Chem. 1968, 33, 1341–
1344.
20. Ueda, N.; Kondo, K.; Kono, M.; Takemoto, K.; Imoto,
M. Makromol. Chem. 1968, 120, 13–20.
21. Mitsunobu, O. Synthesis 1981, 1–28.
22. Brown, D. M.; Todd, A. R.; Varadarajan, S. J. Chem. Soc.
1956, 2384–2387.
23. Cruickshank, K. A.; Jirrinc, J.; Reese, C. B. Tetrahedron
Lett. 1984, 25, 681–684.
24. Jenny, T. F.; Benner, S. A. Tetrahedron Lett. 1992, 33,
6619–6620.
25. Shibata, M.; Takenaka, A.; Sasada, Y.; Ohki, M. Acta
Cryst. 1985, C41, 1354–1356.
26. Cowie, J. M. G.; Love, C. Polymer 2001, 42, 4783–4789.
27. Zhou, J.; Bouhadir, K. H.; Webb, T. R.; Shevlin, P. B.
Tetrahedron Lett. 1997, 38, 4037–4038.
28. Compounds 2a, 2b, 3a, 3b, 5a, 5b, 8, 9, 11, 14, 15 and 17
were characterized through comparison with the literature
data. Selected data for 4a: mp 91–92 °C; 1H NMR
(300 MHz, CDCl3) d 2.63 (t, J = 5.8 Hz, 2H), 3.03 (d,
J = 6.4 Hz, 4H), 3.71 (t, J = 5.8 Hz, 2H), 5.08 (m, 4H),
5.69 (d, J = 7.8 Hz, 1H), 5.66 (m, 2H), 7.19 (d, J = 7.8 Hz,
1H), 10.33 (br s, NH); 13C NMR (75 MHz, CDCl3) d 46.6
(t), 51.1 (t), 57.2 (t), 100.9 (d), 146.0 (d), 151.1 (s), 164.6
(s); FTIR (KBr disk, cmÀ1) 3154, 3024, 2805, 1694, 1648,
1467, 1419, 1363, 1242, 920; HRMS (EI) calcd for
C12H17N3O2 (M+), 235.13208, found 235.13160.
Selected data for 4b: mp 117–118 °C; 1H NMR (300 MHz,
CDCl3) d 1.91 (s, 3H), 2.71 (t, J = 5.9 Hz, 2H), 3.12 (d,
J = 6.2 Hz, 4H), 3.76 (t, J = 5.9 Hz, 2H), 5.12 (m, 4H),
5.75 (m, 2H), 7.08 (s, 1H), 10.29 (br s, 1H); 13C NMR
Selected data for 10: mp 103–194 °C; 1H NMR (300 MHz,
CDCl3) d 7.24 (d, J = 7.1 Hz, 1H), 5.74 (d, J = 7.4 Hz,
1H), 3.70 (t, J = 5.6 Hz, 2H), 2.65 (t, J = 5.6 Hz, 2H), 3.02
(d, J = 6.0 Hz, 2H), 5.60 (m, 1H), 5.06 (m, 2H), 6.06 (br s,
NH2); 13C NMR (75 MHz, CDCl3) d 156.0 (s), 165.2 (s),
93.3 (d), 146.9 (d), 51.6 (t), 48.1 (t), 57.2 (t), 135.1 (d),
117.8 (t); FTIR (KBr disk, cmÀ1) 3352, 3139, 3005, 2976,
2924, 2799, 1656, 1614, 1523, 1485.9, 1456, 1383, 1268.8,
916, 808, 788, 677.8, 619; HRMS (EI) calcd for
C12H18N4O (M+), 234.14806, found 234.14771.
Selected data for 12: mp >157 °C (decomposition); 1H
NMR (300 MHz, CDCl3) d 1.28 (d, J = 6.8 Hz, 6H), 2.60
(h, J = 6.8 Hz, 1H), 7.34 (d, J = 7.0 Hz, 1H), 7.62 (d,
J = 7.1 Hz, 1H), 3.90 (d, J = 5.1 Hz, 1H), 3.90 (d,
J = 5.7 Hz, 1H), 2.78 (d, J = 5.2 Hz,1H), 2.78 (d, J =
5.6 Hz, 1H), 3.09 (d, J = 6.1 Hz, 4H), 5.67 (m, 2H), 5.14
(m, 4H), 8.30 (br s, NH); 13C NMR (75 MHz, CDCl3) d
19.1 (q), 36.40 (d), 177.4 (s), 156.0 (s), 162.4 (s), 95.50 (d),
150.0 (d), 51.10 (t), 48.80 (t), 57.10 (t), 134.8 (d), 118.0 (t);
FTIR (KBr disk, cmÀ1) 3181, 3076, 2976, 2931, 2815,
1712, 1654, 1555, 1491, 1467, 1426, 1361, 1311, 1210, 919;
HRMS (ESI) calcd for C16H24N4O26 (M+1)+, 305.197203,
found 305.1959.
Selected data for 16: mp 142–144 °C; 1H NMR (300 MHz,
CDCl3) d 7.89 (s, 1H), 8.35 (s, 1H), 4.23 (d, J = 5.8 Hz,
1H), 4.23 (d, J = 6.0 Hz, 1H), 2.85 (t, J = 5.9 Hz, 2H),
3.10 (d, J = 6.4 Hz, 4H), 5.95 (br s, NH2), 5.08 (m, 4H),
3.63 (m, 2H); 13C NMR (75 MHz, CDCl3) d 118.4 (t),
134.4 (d), 57.1 (t), 41.9 (t), 52.0 (t), 141.3 (d), 119.3 (s),
155.5 (s), 152.5 (d), 149.9 (s); FTIR (KBr disk, cmÀ1) 3290,
3119, 2979, 2801.48, 2679, 1670, 1644, 1602, 1574, 1514,
1478, 1414, 1352, 1322, 1205, 909; HRMS (EI) calcd for
C13H18N6 (M+), 258.15929, found 258.15860.
1
Selected data for 19: mp 88–90 °C; H NMR (300 MHz,
CDCl3) d 1.20 (d, J = 6.8 Hz, 6H ), 3.20 (m, J = 6.80 Hz,
1H), 8.20 (s, 1H), 8.70 (s, 1H), 4.30 (d, J = 5.5 Hz, 1H),
4.30 (d, J = 6.0 Hz, 1H), 2.80 (d, J = 5.7 Hz, 1H), 2.80 (d,
J = 5.8 Hz, 1H), 3.10 (d, J = 6.3 Hz, 4H), 5.60 (m, 2H),
5.10 (d, 4H), 10.00 (br s, NH); 13C NMR (75 MHz,
CDCl3) d 117.8 (t), 56.9 (t), 134.7 (d), 42.1 (t), 51.8 (t),
151.6 (s), 143.7 (d), 121.9 (s), 149.3 (s), 152.0 (d), 176.6 (s),
35.5 (d), 19.1 (q); FTIR (KBr disk, cmÀ1) 3544, 3304,
3172, 3090, 3034, 2970, 2925, 2806, 1709, 1675, 1611, 1579,
1542, 1489, 1458, 1436, 1401, 1349, 1316, 1275, 1216, 921;
HRMS (ESI) calcd for C17H25N6O (M+1)+, 329.208436,
found 329.2091.