Stereoselective synthesis of 4′-selenonucleosides via seleno-Michael reaction as potent antiviral agents

Article abstract of DOI:10.1021/ol502899b

Based on the hypothesis that the bulky selenium atom, with 4p orbitals, can sterically hinder the approach of a cellular kinase to 5′-OH for phosphorylation, 4′-selenonucleosides with one-carbon homologation were designed and synthesized via a novel selen

Full text of DOI:10.1021/ol502899b

Letter
pubs.acs.org/OrgLett
Stereoselective Synthesis of 4′-Selenonucleosides via Seleno-Michael
Reaction as Potent Antiviral Agents
Pramod K. Sahu,† Gyudong Kim,† Jinha Yu, Ji Yoon Ahn, Jayoung Song, Yoojin Choi, Xing Jin,
Jin-Hee Kim,‡ Sang Kook Lee, Sunghyouk Park, and Lak Shin Jeong*
College of Pharmacy, Seoul National University, Seoul, 151-742, Korea
S
* Supporting Information
ABSTRACT: Based on the hypothesis that the bulky selenium atom, with 4p orbitals, can sterically hinder the approach of a
cellular kinase to 5′-OH for phosphorylation, 4′-selenonucleosides with one-carbon homologation were designed and synthesized
via a novel seleno-Michael reaction, with the stereoselectivity controlled by steric effects. 5′-Homo-4′-selenonucleosides (n = 2)
demonstrated potent antiherpes simplex virus (HSV-1) activity, indicating that the bulky selenium atom might play a key role in
preventing phosphorylation by cellular kinases, resulting in no antiviral activity.
atural DNA and RNA building blocks have served as
valuable resources for the development of new therapeutic
N
agents, such as antiviral and anticancer agents.1 Modifications
of the furanose ring at the 2′- and/or 3′-position have resulted
in many clinically useful drugs, such as AraC, AZT, and
gemcitabine.1,2 However, these 4′-oxo-nucleosides, referred to
as first-generation nucleosides, suffer from many drawbacks, such
as drug resistance, bone marrow and mitochondrial toxicity, and
chemical instability.1 To overcome these side effects, 4′-thio-3
and 4′-carbonucleosides,4 which are bioisosterically related to
4′-oxonucleosides, were developed as second-generation nucleo-
sides. Currently, only abacavir is used clinically as an anti-AIDS
agent, although several second-generation nucleosides are in
clinical trials. The lack of efficacy among second-generation
nucleosides prompted the design of a new template for nucleoside
drug development.
Recently, we5 and another group6 reported 4′-selenonucleosides
as the next generation of nucleosides. However, most of the
synthesized 4′-selenonucleosides showed no significant antiviral or
anticancer activity, most likely due to the absence of phosphoryla-
tion by cellular kinases. Among these, 2′-F-4′-Se-AraC showed
potent anticancer activity;5d however, it was not phosphorylated in
the cell, indicating that it is not a cellular DNA/RNA polymerase
inhibitor, unlike AraC.7
The inability of cellular kinases to phosphorylate 4′-
selenonucleosides may be attributed to a conformation different
from those of 4′-oxo- or 4′-thionucleosides, which was demon-
strated by their X-ray crystal structures.5 We also hypothesized that
the approach of the cellular kinase to the 5′-hydroxyl group for
cellular phosphorylation is hindered by the bulky selenium atom,
which possesses 4p orbitals.
Figure 1. Rationale for the design of 5′-homo-4′-selenonucleosides.
be relieved by one-carbon homologation. Thus, we designed and
synthesized 5′-homo-4′-selenonucleosides (n = 2) using a novel
seleno-Michael reaction as a key step, along with the synthesis of
normal 4′-selenonucleosides (n = 1), and we evaluated them for
antiviral activity (Figure1). Herein, we now report on these efforts.
To synthesize the key homoseleno sugar 4, a novel seleno-
Michael reaction was employed as the key step, as shown in
Scheme 1.
2,3-O-Isopropylidene-L-erythrofuranose (1)8 was treated with
Ph3PCHCO2Et at −10 °C in the presence of trifluoroethanol9 to
As illustrated in Figure 1, it was hypothesized that the absence
of cellular phosphorylation could be due to the presence of the
bulky selenium in 4′-selenonucleosides; it was felt that this might
Received: October 1, 2014
Published: October 23, 2014
© 2014 American Chemical Society
5796
dx.doi.org/10.1021/ol502899b | Org. Lett. 2014, 16, 5796−5799

Organic Letters
Letter
was expected to proceed via 7a over 7b because of the steric
effects of the bulky protecting group, thus affording 8a and 8b as
the major and minor products, respectively. The seleno-Michael
reaction of the Z-isomer, (Z)-6, was not expected to proceed,
since it would have to pass through the unfavored transition
states 7c and 7d. Thus, only trace amounts of 9 were expected to
be produced.
Scheme 1. Synthesis of Key Selenosugar 3 Using a Seleno-
Michael Reaction with E- and Z-Alkenes
Accordingly, we synthesized substrates for the seleno-Michael
reaction with TBS or TBDPS as bulky protecting groups, as
illustrated in Scheme 3. Treatment of L-erythrono-γ-lactone (10)
Scheme 3. Facial Selectivity in the Seleno-Michael Reaction of
E- and Z-Alkenes with Bulky Protecting Groups
avoid an oxa-Michael reaction. This step was followed by
mesylation to afford (Z)-2 as the major isomer in an E/Z = 1/19
ratio. The seleno-Michael reaction of (Z)-2 proceeded smoothly
upon treatment with selenium powder and NaBH4 in EtOH
to afford the seleno-Michael adduct 3 as a separable D/L = 1.1:1
diastereomeric mixture. To examine the difference in reactivity
between the E- and Z-isomers in the seleno-Michael reaction, the
(Z)-isomer 2 was isomerized to the (E)-isomer 2 in the presence
of a Rh catalyst.10 Under the same conditions, the seleno-Michael
reaction of (E)-2 yielded adduct 3 as a separable D/L = 1.1:1
diastereomeric mixture, indicating that the 2,3-acetonide did not
induce facial selectivity in the seleno-Michael reaction. To readily
isolate the target compound from the mixture, 3 was hydrolyzed
with aqueous TFA and then treated with 2,2-dimethoxypropane
in the presence of p-TsOH to yield the desired D-isomer 4 and
the undesired lactone 5, which formed from the L-isomer.
To increase the ratio of the desired D-isomer, we studied the
steric effects in the seleno-Michael reaction. Models indicated
that the transition state of the seleno-Michael reaction might be
shifted to favor the desired D-isomer 8a if a bulky protecting
group such as TBS or TBDPS was introduced, as depicted in
Scheme 2. The seleno-Michael reaction of the E-isomer, (E)-6,
Scheme 2. Changeinthe Transition State in the Seleno-Michael
Reaction To Favor the D-Isomer Based on Steric Effects
with TBSCl followed by reduction with DIBAL-H yielded lactol
11, which exists in its cyclic hemiacetal form (α/β = 14:1). Wittig
reaction of 11 with Ph3PCHCO2Et at 40 °C produced a 1:10
mixture of (Z)-12 and (E)-12 in 94% total yield. Mesylation of
(Z)-12 and (E)-12 yielded the substrates (Z)-13 and (E)-13,
respectively, for use in the seleno-Michael reaction. The seleno-
Michael reaction of (E)-13 under the same conditions as those in
Scheme 2 afforded the desired D-isomer 8a and the undesired
5797
dx.doi.org/10.1021/ol502899b | Org. Lett. 2014, 16, 5796−5799

Organic Letters
Letter
α-isomer 8b in a 4:1 ratio, whereas the same reaction with (Z)-13
failed to give the corresponding adducts. To further study the
steric effects in the seleno-Michael reaction, the TBS groups of
(E)-13 were replaced with bulkier TBDPS groups in (E)-15. As
expected, the ratio of 8a and 8b was greatly increased to 10:1,
but resulted in a low yield because of the remaining starting
material (62%). These results demonstrated that the steric effects
shown in Scheme 2 played a major role in controlling the
transition states and favored the formation of 8a.
Scheme 5. Synthesis of 4′-Selenonucleosides
To synthesize the 5′-homo-4′-selenonucleosides, the seleno-
sugar 4 was reduced with LAH and further protected with TBDPS
ether to give 16 (Scheme 4). Oxidation of 16 with mCPBA
Scheme 4. Synthesis of 5′-Homo-4′-selenonucleosides
The synthesized final nucleosides 19, 22, 26, and 27
were assayed for antiviral activity. As expected, 5′-homo-4′-
selenonucleosides 19 and 22 exhibited antiherpes simplex
virus (HSV-1) activity (EC50 = 2.3 and 2.9 μM, respectively),
but the corresponding 4′-selenonucleosides 27 and 26 exhibited
no antiviral activity up to 100 μM probably because cellular
phosphorylation did not occur.
In conclusion, we have designed and synthesized 5′-homo-4′-
selenonucleosides via a novel seleno-Michael reaction to study
the steric effects of the bulky selenium atom on phosphorylation
by cellular kinases. The facial selectivity of the seleno-Michael
reaction was controlled by steric effects. To our best knowledge,
this seleno-Michael reaction is the first example to be utilized
for the synthesis of carbohydrates. One-carbon homologation
resulted in potent antiviral activity being observed, by allowing
the 5′-homo-4′-selenonucleosides to be phosphorylated by
cellular kinases. These findings indicate that the bulky selenium
played an important role in preventing phosphorylation by
cellular kinases. It is expected that this study can be extended
to other inactive 2′- and/or 3′-modified-4′-selenonucleosides
and greatly contribute to the discovery of new biologically
active nucleosides. The cellular phosphorylation study of
4′-selenonucleosides (n = 1, 2) will be reported elsewhere.
yielded the glycosyl donor 17, which was condensed with uracil
under Pummerer-type conditions5a to afford the protected
nucleoside 18. Removal of the protecting group from 18 yielded
the final 5′-homo-4′-selenouridine 19. To synthesize 5′-homo-4′-
selenoadenosine 22, the glycosyl donor 17 was converted to the
glycosyl donor 20 by heating with acetic anhydride at 100 °C.
Condensation of 20 with silylated 6-chloropurine in the presence
of TMSOTf as a Lewis acid yielded the protected 6-chloropurine
derivative 21. Treatment of 21 with tert-butanolic ammonia
followed by acidic hydrolysis yielded the final 5′-homo-4′-
selenoadenosine 22.
In addition to 5′-homo-4′-selenonucleosides, 4′-selenonucleosides
26 and 27 were synthesized from the same intermediate, 16a
(R = H), as shown in Scheme 5. Compound 16a was converted
to alkene 23 by mesylation followed by an elimination
reaction. Ozonolysis of 23 produced aldehyde 23a; however,
the selenium was also oxidized. Thus, 23a was further treated
with triphenylphosphine to afford the desired aldehyde 24.11
Reduction of 24 followed by protection of the resulting alcohol
with TBDPS afforded the known intermediate 25.5a Intermediate
25 was converted to the desired 4′-selenonucleosides 26 and 27
according to our previously published procedures.5a,f
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures and copies of 1H and 13C NMR spectra
of all new compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: lakjeong@snu.ac.kr.
Present Address
‡College of Pharmacy, Yonsei University, Incheon 406-840, Korea.
5798
dx.doi.org/10.1021/ol502899b | Org. Lett. 2014, 16, 5796−5799

Organic Letters
Letter
Author Contributions
†P.K.S. and G.K. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by intramural Seoul National
University funds for the Research Resettlement and the Next
Generation Leaders Training for the new faculty.
REFERENCES
■
(1) Jordheim, L. P.; Durantel, D.; Zoulim, F.; Dumontet, C. Nat. Rev.
Drug Discovery 2013, 12, 447.
(2) De Clercq, E. J. Clin. Virol. 2004, 30, 115.
(3) Gunaga, P.; Moon, H. R.; Choi, W. J.; Shin, D. H.; Park, J. G.; Jeong,
L. S. Curr. Med. Chem. 2004, 11, 2585.
(4) (a) Piperno, A.; Chiacchio, M. A.; Iannazzo, D.; Romeo, R. Curr.
Med. Chem. 2006, 13, 3675. (b) Schneller, S. W. Curr. Top. Med. Chem.
2002, 2, 1087. (c) Ferrero, M.; Gotor, V. Chem. Rev. 2000, 100, 4319.
(d) Jeong, L. S.; Lee, J. A. Antiviral Chem. Chemother. 2004, 15, 235.
(e) Marquez, V. E.; Lim, M.-I. Med. Res. Rev. 1986, 6, 1.
(5) (a) Jeong, L. S.; Tosh, D. K.; Kim, H. O.; Wang, T.; Hou, X.; Yun,
H. S.; Kwon, Y.; Lee, S. K.; Choi, J.; Zhao, L. X. Org. Lett. 2008, 10, 209.
(b) Tosh, D. K.; Choi, W. J.; Kim, H. O.; Lee, Y.; Pal, S.; Hou, X.; Choi,
J.; Choi, S.; Jeong, L. S. J. Org. Chem. 2008, 73, 4259. (c) Jeong, L. S.;
Choi, Y. N.; Tosh, D. K.; Choi, W. J.; Kim, H. O.; Choi, J. Bioorg. Med.
Chem. 2008, 16, 9891. (d) Jeong, L. S.; Tosh, D. K.; Choi, W. J.; Lee, S.
K.; Kang, Y.-J.; Choi, S.; Lee, J. H.; Lee, H.; Lee, H. W.; Kim, H. O. J.
Med. Chem. 2009, 52, 5303. (e) Varughese, A.; Choi, W. J.; Chun, J.;
Kim, H. O.; Jeon, J. H.; Tosh, D. K.; Lee, H. W.; Chandra, G.; Choi, J.;
Jeong, L. S. Org. Lett. 2010, 12, 2242. (f) Yu, J.; Kim, J.-H.; Lee, H. W.;
Alexander, V.; Ahn, H.-C.; Choi, W. J.; Choi, J.; Jeong, L. S. Chem.Eur.
J. 2013, 19, 5528.
(6) (a) Inagaki, Y.; Minakawa, N.; Matsuda, A. Nucleic Acids Symp. Ser.
2007, 51, 139. (b) Jayakanthan, K.; Johnston, B. D.; Pinto, B. M.
Carbohydr. Res. 2008, 343, 1790. (c) Watts, J. K.; Johnston, B. D.;
Jayakanthan, K.; Wahba, A. S.; Pinto, B. M.; Damha, M. J. J. Am. Chem.
Soc. 2008, 130, 8578.
(7) (a) Plunkett, W.; Iacoboni, S.; Keating, M. J. Scand. J. Haematol.
1986, No. Suppl 44, 51. (b) Grant, S. Adv. Cancer Res. 1998, 72, 197.
(8) Krishna, U. M.; Deodhar, K. D.; Trivedi, G. K. Tetrahedron 2004,
60, 4829.
(9) (a) Arjona, O.; Medel, R.; Rojas, J.; Costa, A. M.; Vilarrasa, J.
Tetrahedron Lett. 2003, 44, 6369. (b) Carrillo, J.; Costa, A. M.; Sidera,
M.; Vilarrasa, J. Tetrahedron Lett. 2011, 52, 5153.
(10) Yagupsky, M.; Wilkinsin, G. J. Chem. Soc. (A) 1970, 941.
(11) Clive, D. L. J.; Postema, M. H. D. J. Chem. Soc., Chem. Commun.
1994, 235.
5799
dx.doi.org/10.1021/ol502899b | Org. Lett. 2014, 16, 5796−5799