Synthesis of a complete Janus-type guanosine-cytosine base and its 2'-deoxyribonucleoside

Article abstract of DOI:10.1246/cl.2011.684

Novel Janus-type guanosinecytosine pyrimido[4,5-d]pyrimidine and its 2'-deoxyribonucleoside have been synthesized in a viable route employing the Vorbruggen method.

Full text of DOI:10.1246/cl.2011.684

doi:10.1246/cl.2011.684
684
Published on the web May 28, 2011
Synthesis of a Complete Janus-type Guanosine-Cytosine Base and Its 2¤-Deoxyribonucleoside
Hang Zhao, Wen Huang, Xiaohua Wu, Yong Qing, Zhihua Xing, and Yang He*
Laboratory of Ethnopharmacology, Institute for nanobiomedical Technology
and Membrane Biology, Regenerative Medicine Research Center, West China Hospital,
West China Medical School, Sichuan University, Chengdu 610041, P. R. China
(Received March 25, 2011; CL-110256; E-mail: heyangqx@scu.edu.cn)
Novel Janus-type guanosine-cytosine pyrimido[4,5-d]py-
rimidine and its 2¤-deoxyribonucleoside have been synthesized
in a viable route employing the Vorbruggen method.
guanosine-cytosine base (4,7-diaminopyrimido[4,5-d]pyrimi-
dine-2,5(1H,6H)-dione (2)), then we adopted direct glycosyla-
tion to obtain the target compound, 4,7-diamino-1-(2-deoxy-¢-
D-erythro-pentofuranoyl)pyrimido[4,5-d]pyrimidine-2,5(1H,6H)-
dione (1) using the Vorbruggen method by employing the
silylated Janus-type base moieties as sugar acceptor and 1-¡-
chloro-2-deoxy-3,5-di-O-p-toluoyl-D-furanose as sugar donor in
the presence of Friedel-Crafts catalysts.8,9
Nucleoside analogs have played an important role in the
treatment of viral disease for many years.1,2 The mechanisms of
these compounds to inhibit viruses are mainly through incorpo-
rating and altering natural DNA/RNA macromolecules, inhibit-
ing various enzymes which participate in the synthesis of nucleic
acid and terminating the synthesis of the genome.3 Apart
from these, Crotty and his colleagues4 have demonstrated
that ribavirin (1-¢-D-ribofuranosyl-1,2,4-triazole-3-carbox-
amide) has an unique antiviral mechanism: its antiviral activity
can be fully attributed to lethal mutagenesis5 due to dual base
pairing which causes transition or transversion mutations (A to
G and C to U). However, the efficiency of incorporating into a
viral genome is relatively low for ribavirin,6 due to its pseudo
base structure being significant different from nature nucleoside.
Inspired by this consideration, we designed a novel Janus
guanine-cytosine (J-GC) nucleoside. The base moiety of this
nucleoside has both a Watson-Crick H-bond donor-donor-
acceptor (DDA) pattern of guanine and a Watson-Crick H-bond
acceptor-acceptor-donor (AAD) pattern of cytosine, which will
base-pair with either guanosine or cytosine when it adopts either
syn or anti conformation upon the rotation of the glycosyl bond.
Our laboratory has synthesized the Janus guanine-cytosine
ribonucleoside by a linear method, and preliminary in vivo
testing shows that it has anti HBV potential.7 On this basis, we
would like to expand this J-GC systems into deoxyribonucleo-
side to see if it can also be incorporated into viral DNA genome
and to investigate their base-pair properties in synthetic DNA
fragments (Figure 1). For this purpose, compound 1 is needed.
Surprisingly, the synthesis of the free Janus GC pyrimido-
[4,5-d]pyrimidine base moiety and its corresponding 2¤-deoxy-
ribonucleoside have not been reported before. Therefore, we will
describe synthesis of them.
To prepare the free Janus-type guanosine-cytosine nucleo-
base 2, we adopted a procedure starting from commercially
available 2-amino-4,6-dihydroxypyrimidine (Scheme 1). The
synthesis of 2-N-phthaloyl-5-cyano-4,6-dichloropyrimidine (3)
has been reported by our group.7 As compound 3 was difficult to
purify by flash chromatography, we attempted to purify it by
recrystallizing from organic solvent. After a series of experi-
ments, ethyl acetate was finally found to be the best choice.
In order to construct the second pyrimidine ring, a suitable
protected amine needs to be introduced into the first pyrimidine
counterpart. At the beginning, we tried to use benzylamine, but
actually the removal of the benzyl group proved to be problem-
atic and unsuccessful, such as treatment with Lewis acids or
hydrogenation. To overcome this problem, p-methoxybenzyl-
amine (PMB) was selected as a solution due to its removability
under mild conditions. When compound 3 was treated with
PMB (2 M) in THF at 50 °C for about 1 h, compound 4 was
obtained with a yield of 76% after flash column chromatog-
raphy. Initially, we chose N-(chlorocarbonyl) isocyanate (CCI)
as a reagent to obtain the the urea intermediate [5] for the
subsequent cyclization which offers compound 5 as the fully
constructed pyrimido[4,5-d]pyrimidine. However, the total yield
was not ideal (20%) and the CCI was neither stable nor
Cl
Cl
CN
N
Cl
N
O
N
O
NH2
O
CN
NH
N
O
N
O
CN
Cl
(b)
(a)
N
N
N
O
N
O
N
MeO
OMe
There are two major parts for the synthesis of J-GC 2¤-
deoxyribonucleoside. First, we successfully synthesized the free
3
4
5
OMeNH2
N
OMeNH2
N
N
O
NH2
(e)
(c)
(d)
N
H2N
N
N
O
HN
N
G
N
C
H
N
G
H
N
O
H
H N
NH2
H2N
N
N
H
O
H2N
N
O
O
H
O
N
H2N
N
N
H
C
O
O
O
O
O
O
MeO
N
C
G
C
N
N
N
H
H
N
N
G
N
N
N
5
6
2
O
O
N
O
O
O
N
H
H
O
O
H N
O
O
H
Scheme 1. Synthesis of Janus-type guanosine-cytosine base:
(a) p-methoxybenzylamine (PMB), THF, 50 °C, 1 h, 76%; (b)
i-Pr2EtN, trichloroacetyl isocyanate, dichloromethane (DCM), rt,
1 h; (c) 0.5 M NaOMe, 75 °C, 2 h, 65%; (d) CH3CN-H2O (3:1,
v/v), CAN, rt, 16 h, 73%; (e) CH3CN, NaI, TMSCl, rt, 16 h,
94%.
J-GC (1)
guanosine
J-GC (1) cytidine
Figure 1. The designed J-GC deoxyribonucleoside is sup-
posed to pair with either guanosine or cytidine through the
specific pattern of hydrogen-bonding patterns.
Chem. Lett. 2011, 40, 684-686
© 2011 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

685
OMeNH2
economical. For this reason, we tried to find a better route to
obtain 5. After several failed attempts, we finally found
trichloroacetyl isocyanate to be the best choice. In this case,
compound 4 was treated with trichloroacetyl isocyanate and
i-Pr2EtN as base catalyst in dry DCM for about 1 h at rt to give
the urea intermediate [5], which was used directly in the
cyclization reaction (refluxed in 0.5 M NaOMe for about 2 h) to
lead to the formation of the key compound pyrimido[4,5-d]-
pyrimidine 5. During the cyclization the phthaloyl protecting
group of the 2-amino group was removed and the remaining
chloride at 4-position of the first pyrimidine ring was replaced
by a methoxy group simultaneously. Compared to CCI, prep-
aration of 5 with trichloroacetyl isocyanate was more econom-
ical and produced a higher yield (total yield of 65% from 4).
Next, the PMB group was removed under mild conditions
by treating compound 5 with AlCl3 in anisole10,11 or ceric
ammonium nitrate (CAN) in 3:1 CH3CN and H2O.12 However,
the AlCl3 was difficult to remove completely when compound 6
was purified, and it may affect the glycosylation in the later
stages because AlCl3 is a strong Lewis acid. Therefore, we
treated 5 with CAN (4 equiv) in 3:1 CH3CN and H2O, and the
product precipitated from the reaction solution. Pure compound
6 was obtained by washing with MeOH thoroughly (73%). The
removal of the methoxy group was as follows. Compound 6 was
treated with NaI and trimethylchlorosilane (TMSCl) in aceto-
nitrile, the mixture was stirred overnight at rt, and the white
precipitate was collected and washed with methanol to get
compound 2 (94%).
Synthesis of the J-GC 2¤-deoxyribonucleoside (1) was
achieved by the route shown in Scheme 2. Compound 6 was
reacted with hexamethyldisilazane (HMDS) to provide the
intermediate 7 which was used directly in the later glycosylation
reaction without any purification due to its sensitivity to
moisture. Compared with the pyrimidine system, the silylation
step was rather slow due to the additional exocyclic amino and
oxy groups. Therefore, we attempted to find a catalyst to
overcome this problem. After several experiments (ammonium
sulfate,13,14 trimethylchlorosilane,8 and pyridine + TMSCl15),
HMDS and TMSCl was finally found to be a suitable
combination for this reaction. As mentioned before, attention
should be paid not to introduce moisture when evaporating
HMDS due to the sensitivity of compound 7 to moisture.
The silylated base 7 was used as sugar acceptor and 1-¡-
chloro-2-deoxy-3,5-di-O-p-toluoyl-D-furanose was used as sugar
donor.16 In the case of riboside, because of the steric hindrance
of the 2¤-acyloxy group, only ¢-anomer nucleoside was formed.
But in the case of deoxyribonucleoside, without the neighboring
group participation, usually, 1:1 mixtures of ¡:¢ anomers will
be formed. It has been reported that CuI can facilitate the ¢-
configuration selectivity with an electronic push-pull processs in
CHCl3 for the normal purine and pyrimidine system.17 However,
this method did not give satisfactory results in this pyrimido-
[4,5-d]pyrimidine system. We tried the reaction of silylated base
7 treating with 1-¡-chloro-2-deoxy-3,5-di-O-p-toluoyl-D-fura-
nose and with CuI as catalyst, however, the yield was low.
Therefore, we attempted to obtain 8a by reaction of 7 with 1-¡-
chloro-2-deoxy-3,5-di-O-p-toluoyl-D-furanose by employing
either SnCl4 or TMSOTf as catalyst. Finally, we found tri-
methylsilyl trifluoromethanesulfonate (TMSOTf) was superior
to SnCl4 as catalyst for the synthesis of deoxyribonucleoside 8a.
N
N
H2N
HO
N
N
O
O
OMeNH2
N
(c)
OH
9a
N
OMeNH2
N N
H2N
N
N
H
O
H2N
RO
N
N
O
6
O
(d)
(a)
OR
8a
RO
O
NH2
O
HN
N
C
N
G
N
OMeNHTMS
N
Cl
OR
(b)
H2N
O
R=p-CH3C6H4CO
N
HO
O
TMSHN
N
N
OTMS
OH
1
RO
O
R=p-CH3C6H4CO
7
OR
O
N
N
NH2
N
N
NH2 OMe
8b R=p-CH3C6H4CO
9b R=H
(e)
Scheme 2. Synthesis of J-GC 2¤-deoxyribonucleoside: (a)
HMDS, TMSCl, 140 °C; (b) TMSOTf, CH3CN-1,2-dichloro-
ethane (1:1, v/v), rt, 1 h; (c) 0.5 M NaOMe, 70 °C, 92%; (d) 2 M
NaOH, 1,4-dioxane, 70 °C, 6 h, 90%; (e) 0.5 M NaOMe, 70 °C,
90%.
Silylated 7 was treated with 1-¡-chloro-2-deoxy-3,5-di-O-p-
toluoyl-D-furanose and TMSOTf in anhydrous CH3CN and 1,2-
dichloroethane. Subsequently, the reaction was stirred at room
temperature for about 1 h, saturated sodium bicarbonate was
added to quench the reaction at 0 °C, then extracted with DCM
and washed with saturated saline, H2O, dried with anhydrous
Na2SO4, and the compound 8a (¢, 37%) and 8b (¡, 11%) was
obtained by flash chromatography. Compound 8a and 8b was
treated with 0.5 M NaOMe at 70 °C until the solution become
clear, the reaction was neutralized with acetic acid, then the
precipitate was collected and washed with MeOH to get
compound 9a (92%) and 9b (90%), the ¡ and ¢ configuration
was clearly established by their NMR spectra. 1¤-H of 9a
produced a strong NOE at 2¤-Ha and 2¤-Hb while 1¤-H of 9b
showed an NOE only at 2¤-Ha, simultaneously, 1¤-H of 9b show
an NOE at 3¤-H. It is worthwhile to mention that initially we
tried the same method as in the case of compound 6 to replace
the methoxy group of compound 9a by treating it with NaI and
TMSCl in CH3CN. Unfortunately, this reaction did not proceed,
instead of compound 1 we finally obtained compound 2.
Obviously, the glycosyl bond was unstable under such con-
ditions. Consequently, we found a way to deprotect the Tol
group and to replace the methoxyl group as follows. 8a was
treated with 2 M NaOH and 1,4-dioxane at 70 °C for about 6 h,
then neutralized with 2 M HCl, the white precipitate was
collected and washed with MeOH to give pure compound 1
(90%).
In conclusion, we have developed a viable route to obtain
the complete Janus-type guanosine-cytosine base and its
2¤-deoxyribonucleoside. Our next step is to investigate their
biological activity, base-pairing properties, and enzymatic
reactions in the context of DNA. These experiments are under
investigation currently and will be published in the near
future.
Chem. Lett. 2011, 40, 684-686
© 2011 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

686
This study was financially supported by the National Natural
Science Foundations of China (document No. 20772087).
10 T. Tsuji, T. Kataoka, M. Yoshioka, Y. Sendo, Y. Nishitani, S.
Hirai, T. Maeda, W. Nagata, Tetrahedron Lett. 1979, 20,
2793.
References
11 T. Akiyama, M. Kumegawa, Y. Takesue, H. Nishimoto, S.
Ozaki, Chem. Lett. 1990, 339.
12 A. L. LaFrate, J. A. Katzenellenbogen, J. Org. Chem. 2007,
72, 8573.
13 M. Okabe, R.-C. Sun, S. Y. K. Tam, L. J. Todaro, D. L.
Coffen, J. Org. Chem. 1988, 53, 4780.
14 J.-Q. Wang, Q.-H. Zheng, X. Fei, X. Liu, T. A. Gardner, C.
Kao, S. P. Raikwar, B. E. Glick-Wilson, M. L. Sullivan,
B. H. Mock, G. D. Hutchins, Synth. Commun. 2004, 34, 917.
15 U. Niedballa, H. Vorbrueggen, J. Org. Chem. 1974, 39,
3672.
1
2
3
T. Cihlar, A. S. Ray, Antiviral Res. 2010, 85, 39.
E. D. Clercq, Med. Res. Rev. 2009, 29, 571.
C. M. Galmarini, J. R. Mackey, C. Dumontet, Lancet Oncol.
2002, 3, 415.
4
S. Crotty, C. Cameron, R. Andino, J. Mol. Med. 2002, 80,
86.
5
6
7
J. D. Graci, C. E. Cameron, Future Virol. 2008, 3, 553.
J. D. Graci, C. E. Cameron, Rev. Med. Virol. 2006, 16, 37.
H.-Z. Yang, M.-Y. Pan, D.-W. Jiang, Y. He, Org. Biomol.
Chem. 2011, 9, 1516.
8
9
U. Niedballa, H. Vorbrueggen, J. Org. Chem. 1974, 39,
3654.
H. Vorbrüggen, K. Krolikiewicz, B. Bennua, Chem. Ber.
1981, 114, 1234.
16 V. Rolland, M. Kotera, J. Lhomme, Synth. Commun. 1997,
27, 3505.
17 J. N. Freskos, Nucleosides, Nucleotides Nucleic Acids 1989,
8, 549.
Chem. Lett. 2011, 40, 684-686
© 2011 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/