Convenient synthesis of (E)-5-aminoallyl-2′-deoxycytidine and some related derivatives

Article abstract of DOI:10.1016/j.tetlet.2010.10.137

Preparation via the Heck reaction of 5-(E)-(3-trifluoroacetamidoallyl)- 2′-deoxycytidine (3a) and some N4-formamidine protected derivatives is reported. Difficulties with coupling N-allyl-trifluoroacetamide to 5-iodo-2′-deoxycytidine were overcome by using Pd₂(dba) ₃ as catalyst and temporary dimethylformamidine protection of the N4-amine of the nucleoside. Compound 3a was isolated with 60% overall yield by crystallization.

Full text of DOI:10.1016/j.tetlet.2010.10.137

Tetrahedron Letters 52 (2011) 181–183
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Convenient synthesis of (E)-5-aminoallyl-20-deoxycytidine and some
related derivatives
⇑
Mark V. Reddington , Daniel Cunninghan-Bryant
Biosearch Technologies Inc., 81 Digital Drive, Novato, CA 94949, USA
a r t i c l e i n f o
a b s t r a c t
Preparation via the Heck reaction of 5-(E)-(3-trifluoroacetamidoallyl)-20-deoxycytidine (3a) and some
N4-formamidine protected derivatives is reported. Difficulties with coupling N-allyl-trifluoroacetamide
to 5-iodo-20-deoxycytidine were overcome by using Pd2(dba)3 as catalyst and temporary dimethylform-
amidine protection of the N4-amine of the nucleoside. Compound 3a was isolated with 60% overall yield
by crystallization.
Article history:
Received 15 September 2010
Revised 25 October 2010
Accepted 26 October 2010
Available online 31 October 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Heck coupling
Palladium
20-Deoxycytidine
Nucleosides
Allylamine
H
Aminopropargyl and aminoallyl modified nucleosides are
important intermediates in the preparation of amino functional-
ized nucleic acids (NAs).1–7 They can be used to prepare phospho-
ramidites for use with automated NA synthesizers1–3 or to prepare
nucleoside triphosphates (NTPs) for enzymatic NA synthesis.4–7
Typically, the aminopropargyl group is introduced via the Sono-
gashira reaction8 of the halogenated nucleoside with protected
propargylamine.4,7 Similarly, aminoallyl substituted nucleosides
are prepared using the Heck reaction.4,9,10 Trifluoroacetate (TFA)
is a preferred protecting group for propargylamine and allylamine
because it is stable to the conditions of the coupling reactions,
automated NA synthesis, and NTP synthesis. The TFA group is eas-
ily removed under basic conditions once NA or NTP synthesis has
been completed. However, not all halogenated nucleosides react
efficiently with these TFA-protected amines.11 While there are
examples of Heck coupling of 5-iodo-20-deoxycytidine (2a) to allyl
compounds4 and despite numerous claims in the patent literature,
a method to couple N-allyl-trifluoroacetamide (1) to 2a by the
Heck reaction is conspicuous by its absence.
N
CF3
O
1
O
X
N
X
N
I
N
N
N
H
CF3
O
O
HO
HO
O
O
OH
OH
2a, X = NH2
2b, X = NHAc
3a, X = NH2
3b, X = NHAc
2c, X = N=CHN-iBu2
2d, X = N=CHNMe2
3c, X = N=CHN-iBu2
3d, X = N=CHNMe2
followed by palladium or palladium-copper mediated coupling
with the allyl substituent.12,13 An alternative route to these com-
pounds involves coupling of the halogenated nucleoside to the allyl
substituent using almost stoichiometric amounts of palladium
reagent.14 Coupling by these routes produces the (E)-isomer. We
prepared 5-trifluoroacetyl-amidoallyl-20-deoxycytidine (3a) by the
first of these routes and observed its formation by the second. How-
ever, we considered the first route unsatisfactory due to the mer-
cury hazard, and the second because it uses large quantities of
palladium reagent and, in our hands, generated a mixture of prod-
ucts. The (Z)-isomer of 3a has been prepared by partial reduction of
In addition to the Heck reaction, C5-allyl-substituted pyrimi-
dine nucleosides are accessible through routes involving mercura-
tion of the halogenated nucleoside
⇑
Corresponding author. Tel.: +1 415 883 8400; fax: +1 415 883 8488.
E-mail address: mark@biosearchtech.com (M.V. Reddington).
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.10.137

182
M. V. Reddington, D. Cunninghan-Bryant / Tetrahedron Letters 52 (2011) 181–183
Table 1
Summary of Heck reactions of 1 with dialkylformamidine protected 5-iodo-20-deoxycytidine derivatives in DMF under argon
Entry
Starting material
Reagentsa
Temperature (°C)
Time (h)
% Coupled productb
1
2
3
4
5
6
7
8
9
2c
2c
2c
2c
2c
2d
2d
2d
2d
Pd(OAc)2, PPh3, TEA
100
55
55
55
55
100
55
55
1
18
1
1
1
1
1
3
3
55
0
Pd(PPh3)4, LiOAc(H2O)2, Bu4NCl, LiCl
Pd2(dba)3, LiOAc(H2O)2, Bu4NCl, LiCl
Pd2(dba)3, LiOAc(H2O)2
Pd2(dba)3, TEA
80
53
55
60
>90
85
70
Pd(OAc)2, PPh3, TEA
Pd2(dba)3, LiOAc(H2O)2, Bu4NCl, LiCl
Pd2(dba)3, LiOAc(H2O)2
Pd2(dba)3, TEA
55
a
10 mol % Pd for Pd(OAc)2, 5 mol % Pd for Pd2(dba)3, and Pd(PPh3)4, 2.5 equiv base.
By RP HPLC at 330 nm.
b
5-trifluoroacetylamidopropargyl-20-deoxycytidine.15 The (E)/(Z)-
geometry of aminoallyl-substituted dNTPs can have a significant
influence on the rate of dNTP incorporation during NA synthesis.6
In the case of dUTP and dATP the (E)-isomer is incorporated more
efficiently than the (Z)-isomer.6 Similarly, the properties of the final
NA can be influenced by the (E)/(Z)-geometry.1,6 To meet the need
for an efficient, hazard-free method for the synthesis of the (E)-3a
we explored a number of published conditions for the Heck reaction
(Table 1). Here we report that 3a can be prepared by the Heck reac-
tion and can be isolated in good yield without the need for
chromatography.
salts, which elute from the column after 3c. In the absence of the
TBA and lithium salts, with either lithium acetate or triethylamine
as base, the coupling reaction was less efficient and impurities sim-
ilar to those observed with palladium acetate were seen (entries 4
and 5).
This success led us to investigate the coupling of 1 to the less
sterically hindered dimethylformamidine (dmf) analog (2d). Again,
coupling with palladium acetate generated the desired compound
(3d) along with numerous minor by-products (entry 6). Coupling
proceeded efficiently with the Pd2(dba)3 catalyst system (entry 7)
but isolation of 3d proved difficult because the compound parti-
tions with some of the reaction mixture salts into aqueous media.
Continued manipulation of the material in aqueous and methano-
lic solutions led to some loss of the dmf group. Omission of the TBA
and lithium salts, using either lithium acetate or triethylamine as
base, lowered the coupling efficiency without ameliorating this
work up problem (entries 8 and 9). Consequently, we decided to
remove the dmf group and isolate 3a. Addition of dilute hydrochlo-
ric acid to the crude reaction mixture after the coupling step re-
moved the dmf group while leaving the TFA group intact. At first
3a was isolated by a combination of filtration, washing, and RP col-
umn chromatography but later we found that the column step
could be omitted and that 3a could be crystallized from an aqueous
solution of the crude reaction mixture. Using a similar crystalliza-
tion method, 3d was isolated from the post coupling reaction mix-
ture but with poor yield due to hydrolysis of the protecting group
during the crystallization step.
We coupled 1 to 5-iodo-20-deoxyuridine using palladium ace-
tate under published conditions for the Heck reaction4 but with
these, and other conditions that were reported16 to enable the cou-
pling with 2a, we observed no reaction. We assumed that the N4-
amine of 2a was inhibiting the reaction and that protecting it
would permit coupling. N4-Acetyl-5-iodo-20-deoxycytidine (2b)
was prepared analogously to N4-acetyl-20-deoxycytidine17 and
was subjected to different Heck coupling conditions. In all cases,
the acetylated nucleoside converted cleanly into N4-acetyl-
20-deoxycytidine, although at rates that depended on the nature
of the palladium catalyst and on the base.
Dialkylformamidines, and in particular diisobutyl-formamidine
(dbf), are commonly used to protect exocyclic amine groups of
nucleosides.18,19 dbf-Protected 5-iodo-20-deoxycytidine (2c) was
prepared and subjected to the Heck coupling. Using palladium ace-
tate (entry 1), the starting material was consumed within 1 h and
the coupled product (3c) was generated in approx. 55% yield. How-
ever, a number of by-products also formed. One of these (approx.
15%) had very similar mobility as 3c on reverse phase (RP) HPLC
and on TLC (SiO2 and Al2O3), and proved difficult to separate from
3c by silica gel column chromatography. Pd(PPh3)4 is an ineffective
catalyst for this reaction (entry 2). Pd2(dba)3 in combination with
tetrabutylammonium chloride (TBACl) and lithium chloride effi-
ciently coupled 1 to 2c to afford 3c (entry 3) without the difficult
to remove impurity. Compound 3c was isolated in 73% yield by a
combination of washing and silica gel column chromatography.
The main difficulty with the purification was removal of the TBA
Next we simplified the practical aspects of the synthesis20
(Scheme 1). 2a was mixed with the dmf protecting group reagent
and all of the other coupling reaction reagents except for the cata-
lyst, and the mixture was heated at 55 °C under a stream of argon
for 1 h. The catalyst was then added and heating was continued for
a further hour before addition of hydrochloric acid to remove the
dmf-protecting group. After work up, 3a was isolated with 60%
yield by crystallization.
The TFA group can be removed from 3a to afford 5-aminoallyl-
20-deoxycytidine by treatment with concentrated ammonia but we
found it more convenient to use a solution of methylamine in
NH2
I
NH2
N
CF3
N
1) Me2NCH(OMe)2, LiOAc(H2O)2,
N
H
O
N
LiCl, Bu4NCl, DMF, Ar, 55 °C, 1
O
N
O
2) Pd2(dba)3
HO
HO
O
O
3) 1.0 M HCl(aq)
OH
OH
2a
3a
Scheme 1. One-pot synthesis of (E)-5-trifluoroacetamidoallyl-20-deoxycytidine.

M. V. Reddington, D. Cunninghan-Bryant / Tetrahedron Letters 52 (2011) 181–183
183
13. Langer, P. N.; Waldrop, A. A.; Ward, D. C. Proc. Natl. Acad. Sci. 1981, 78, 6633.
14. Sakthivel, K.; Barbas, C. F., III Angew. Chem., Int. Ed. 1998, 37, 2872.
15. Lee, S. E.; Vyle, J. S.; Williams, D. M.; Grasby, J. A. Tetrahedron Lett. 2000, 41, 267.
16. Beckvermit, J. T.; Tu, C. US Patent 6 020 483, 2000.
ethanol because the product crystallizes directly from the solution
in high yield.
17. Reddy, M. P.; Hanna, B. N.; Farooqui, F. US Patent 5 726 301, 1998.
18. Froehler, B. C.; Matteucci, M. D. Nucleic Acids Res. 1983, 11, 8031.
19. Seela, F.; Zulauf, M. Helv. Chim. Acta 1999, 82, 1878.
Acknowledgment
20. To a 250 mL round-bottomed flask were added 5-iodo-deoxycytidine (3.55 g,
10.0 mmol), DMF (100 mL), dimethylformamide dimethethylacetal (2.0 mL,
15.0 mmol), tetrabutylammonium chloride (2.8 g, 10.0 mmol), lithium chloride
(0.43 g, 10.0 mmol), lithium acetate dihydrate (2.5 g, 24.5 mmol), and 1 (3.8 g,
24.8 mmol). The flask was fitted with a septum and was flushed with argon
while heating at 55 °C for 1 h. Tris(dibenzylideneacetone)dipalladium (0)
(0.43 g, 0.47 mmol) was added and heating under argon was continued for 1 h.
The septum was removed and 1.0 M hydrochloric acid (50 mL) was added.
Heating was continued for 1 h and then the mixture was cooled to rt. Sodium
bicarbonate (4.2 g, 50 mmol) was added in portions taking care to avoid
excessive frothing. The mixture was vacuum filtered and then evaporated
under vacuum to leave a waxy solid. Water (100 mL) was added and the
mixture was stirred until the emulsion separated into a brown solid in a clear
yellow solution. The solid was filtered off and then the flask and solid were
washed with water (2 ꢀ 60 mL). The combined aqueous solutions were washed
with DCM (4 ꢀ 60 mL), filtered, and evaporated under vacuum without heating
to leave an off-white solid. Water (20 mL) was added and the mixture was
heated to approx. 85 °C to dissolve the solid. The solution was left to cool to rt
and then was placed in a refrigerator at 4 °C overnight. The solid was filtered
off, washed with ice water (3 ꢀ 10 mL), and dried under vacuum with
phosphorus pentoxide to afford 3a (2.33 g, 61%): Rf 0.15 (SiO2, DCM/
methanol, 17:3); 1H NMR (DMSO-d6) d 2.0–2.06 (m, 1H), 2.11–21.7 (m, 1H),
3.54–3.63 (m, 2H), 3.77–3.80 (m, 1H), 3.91 (d, J = 5.8 Hz, 2H), 4.21–4.24 (m,
1H), 5.07 (t, J = 4.9 Hz, 1H), 5.20 (d, J = 4.2 Hz, 1H), 5.84 (t, J = 6.2 Hz, 1H), 5.88
((t, J = 6.1 Hz, 1H), 6.14 (t, J = 6.5 Hz, 1H), 6.42 (t, J = 15.7 Hz, 1H), 7.02 (br s, 1H)
7.38 (br s, 1H), 8.09 (s, 1H), 9.56 ((s, 1H); 13C NMR d 40.6, 41.4, 61.0, 70.1,
85.2, 87.3, 103.2, 111.6, 115.9 (q, JC–F = 288.2 Hz), 123.4, 123.6, 137.8,
156.0 (q, JC–F = 35.4 Hz), 163.3; ESI-MS obsd (+) 401.39 (M+Na), Calcd 401.10
(M = C14H17F3N4O5); UV–vis (water, nm) kmax 209, 242, 290; Anal. Calcd
for C14H17F3N4O5: C, 44.45; H, 4.53; N, 14.81. Found: C, 44.05; H, 4.73; N,
15.17.
The authors thank Mr. Alex Herrault for assistance in acquiring
mass spectra.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.10.137.
References and notes
1. Brazier, J. A.; Shibata, T.; Townsley, J.; Taylor, B. F.; Trary, E.; Williams, N. H.;
Williams, D. M. Nucleic Acids Res. 2005, 33, 1362.
2. Lermer, L.; Roupioz, Y.; Ting, R.; Perrin, D. M. J. Am. Chem. Soc. 2002, 124, 9960.
3. Cook, A. F.; Vuocolo, E.; Brakel, C. L. Nucleic Acids Res. 1988, 16, 4077.
4. Jäger, S.; Rasched, G.; Kornreich-Leshman, H.; Engeseer, M.; Thum, O.; Famulok,
M. J. Am. Chem. Soc. 2005, 127, 15071.
5. Lee, S. E.; Sidorov, A.; Gourlain, T.; Mignet, N.; Thorpe, S. J.; Brazier, J. A.;
Dickman, M. J.; Hornby, D. P.; Grasby, J. A.; Williams, D. M. Nucleic Acids Res.
2001, 29, 1565.
6. Vaghefi, M. Nucleoside Triphosphates and their Analogs; CRC Press, Taylor and
Francis Group: Boca Raton, 2005.
7. Hobbs, F. W. J. Org. Chem. 1989, 54, 3420.
8. Chinchilla, R.; Najera, C. Chem. Rev. 2007, 107, 874.
9. Wellington, K. W.; Benner, S. A. Nucleos. Nucleot. Nucleic. Acids 2006, 25, 1309.
10. Knowles, J. P.; Whiting, A. Org. Biomol. Chem. 2007, 5, 31.
11. Gourlain, T.; Sidorov, A.; Mignet, N.; Thorpe, S. J.; Lee, S. E.; Grasby, J. A.;
Williams, D. M. Nucleic Acids Res. 2001, 29, 1898.
12. Ruth, J. L.; Bergstrom, D. E. J. Org. Chem. 1978, 43, 2870.