New RNA purine building blocks, 4'-selenopurine nucleosides: First synthesis and unusual mixture of sugar puckerings

Article abstract of DOI:10.1002/chem.201300741

Writer's blocks: The first synthesis of RNA purine building blocks, 4'-selenoadenosine and 4'-selenoguanosine was achieved from D-ribose by regioisomeric rearrangement, which was confirmed by X-ray crystallography. 4'-Selenoadenosine exists in an unusual mixture of north and south conformers in the solid state.

Full text of DOI:10.1002/chem.201300741

COMMUNICATION
DOI: 10.1002/chem.201300741
New RNA Purine Building Blocks, 4’-Selenopurine Nucleosides: First
Synthesis and Unusual Mixture of Sugar Puckerings
Jinha Yu,[a] Jin-Hee Kim,[a] Hyuk Woo Lee,[a] Varughese Alexander,[a] Hee-Chul Ahn,[b]
Won Jun Choi,[a, b] Jungwon Choi,[c] and Lak Shin Jeong*[a]
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Modified nucleosides have been generally used as biologi-
cal tools as well as chemotherapeutic agents.[1] DNA or
RNA building blocks have extensively been utilised as ex-
cellent templates for the development of modified nucleo-
sides. Extensive modifications have been done on the fura-
nose ring of DNA or RNA, resulting in the 4’-oxonucleo-
sides, referred to as the first-generation nucleosides
(Figure 1).[2] On the basis of bioisoteric rationale, 4’-thionu-
formidable obstacle in the synthesis of purine nucleosides is
the correct identification of the N3, N7, or N9 isomers.
Given the inherent difficulties, whether or not the N7
isomer and/or the N3 isomer is rearranged to the desired N9
isomer during the condensation reactions remains a contro-
versial issue.[7,8] Thus, it is also meaningful to establish the
rearrangement of the isomers formed during the synthesis of
the 4’-selenopurine nucleosides.
Recently the synthesis of N9-4’-selenoadenosine (1a,
Figure 1), which may serve as a new RNA purine building
block was reported, using a Pummerer-type condensation of
4-selenoxide with 6-chloropurine.[5c] It was concluded that
the initially formed N3- or N7-6-chloropurine derivative re-
arranged to the N9-6-acetoxypurine derivative on being
heated with acetic acid, which further converted to 1a after
being treated with methanolic ammonia.[5c] However, from
our study it was concluded the N7-4’-selenoinosine had been
synthesised (13b, without the expected rearrangement) in-
stead of the desired N9-4’-selenoadenosine (Supporting In-
formation, Figures S1 and S2). Herein, we report the first
synthesis of new RNA purine building blocks, 4’-selenoade-
nosine and 4’-selenoguanosine (1b) and the structural revi-
sion, and unveil our insights to confirm the regioisomeric re-
arrangement along with the studies on the unusual dynamic
mixture of sugar puckerings.
Figure 1. Rationale for the development of new RNA purine building
blocks.
Our synthesis of 4’-selenoadenosine (1a) and 4’-seleno-
guanosine (1b) commenced with the synthesis of 4-seleno-
ACHUTNGERNsNUG ugAHCTUNGTRNENaUGN r 6 starting from d-ribose (Scheme 1). d-Ribose was
cleosides[3] and carbocyclic nucleosides,[4] referred to as the
second-generation nucleosides, have been developed. Re-
cently, new 4’-selenopyrimidine nucleosides, also called the
third-generation nucleosides, were synthesised using a Pum-
merer-type condensation of the 4-selenoxide and pyrimidine
bases.[5] From the X-ray crystallographic study, it was discov-
ered that 4’-selenouridine adopted the C2’-endo/C3’-exo
twist (south) conformation, which is different from that of
uridine, which takes the C2’-exo/C3’-endo (north) conforma-
tion.[5a] Thus, it is interesting to synthesise the 4’-selenopur-
ine nucleosides and to compare their conformations with
those of the 4’-selenopyrimidine nucleosides.
converted to 2,3-O-isopropylidene-l-lyxono-1,4-lactone (2)
by a known method.[9] The primary hydroxyl group of 2 was
protected with TBDPS group to give 3, which on reduction
with NaBH4 produced diol 4. Mesylation of 4 followed by
selenenation gave the desired 4-selenosugar 6. This was effi-
ciently oxidised to 4-selenoxide 7 with mCPBA at À788C.
Although the intermediate 7 served as an efficient glycosyl
donor during the synthesis of 4-selenoribofuranosyl pyrimi-
dines under the Pummerer-type conditions,[5a,b] similar at-
tempts towards the condensation reaction of 7 with 6-chlo-
Purine nucleosides are generally synthesised by condens-
ing the glycosyl donor with a purine base in the presence of
a Lewis acid. It is believed that glycosylation of purines is
rarely regioselective and usually produces the unnatural N7
or N3 isomer along with the natural N9 isomer.[6] Another
ACHUTGTNRENrNUG oACHTGNUTRENNUpG urine failed (see the Supporting Information). There-
after, an alternative strategy involving the conversion of 7 to
the corresponding 4-selenoacetate 8 was envisioned. This
could be then condensed under classical Vorbrꢁggen condi-
tions with silylated 6-chloropurine in the presence of
TMSOTf.
Condensation of 8 with silylated 6-chloropurine in the
presence of TMSOTf afforded a mixture of N9 and N7 re-
gioisomers through the rearrangement of the N7 to N9 re-
gioisomer. Attempts to optimise this reaction in favour of
the N9 regioisomer were made (Supporting Information,
Table S1). The best result (N9 isomer=49%, N7 isomer=
15%) was obtained in toluene at 958C after 15 h. The struc-
tures of regioisomers 9 and 10 were confirmed by the analy-
[a] J. Yu, Dr. J.-H. Kim, Dr. H. W. Lee, Prof. V. Alexander,
Prof. W. J. Choi, Prof. L. S. Jeong
Department of Bioinspired Science and College of Pharmacy
Ewha Womans University, Seoul 120-750 (Korea)
Fax : (+82)2-3277-2851
E-mail: lakjeong@ewha.ac.kr
[b] Prof. H.-C. Ahn, Prof. W. J. Choi
College of Pharmacy, Dongguk University
Gyeonggi-do 410-820 (Korea)
1
sis of the H and 13C NMR spectra, in which the signals for
[c] Prof. J. Choi
H-8, C-4, C-8, and H-1’ are shielded whereas the signal for
C-5 is deshielded in the N9 isomer 10 when compared with
the N7 isomer 9 (Supporting Information, Table S2). Their
structures were finally confirmed by the X-ray crystallogra-
Department of Chemistry, Univeristy of Suwon
Gyeonggi-do 445-743 (Korea)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201300741.
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

4’-Selenopurine Nucleosides
COMMUNICATION
the concomitant formation of
the hydrolysed product 13b.
On the other hand, the N9
isomer 10 gave the 6-chloro-
purine derivative 15 exclusive-
ly, which was then converted
to the 4’-selenoinosine ana-
logue 16 by treatment with 2-
mercaptoethanol and sodium
methoxide.[10] Finally, the treat-
ment of 13a and 15 with am-
monia in tBuOH afforded the
N7-4’-selenoadenosine
(14)
and N9-4’-selenoadenosine
(1a), respectively. Using a sim-
ilar strategy, the 2-amino-6-
chloropurine derivative 12 was
also converted to the corre-
sponding N9-4’-selenoguano-
sine.
The structure of the N9-4’-
selenoadenosine (1a) was also
confirmed by X-ray crystallog-
raphy (Figure 2). Based on our
experimental results and spec-
troscopic evidences, it is quite
clear that the previously re-
Scheme 1. The rearrangement of the N7 isomers 9 and 11 to the N9 isomers 10 and 12 under the Vorbrꢁggen
condensation conditions, respectively.
ed[5c] structure of the N9-
ACHUTNGRENpNUG ortACHTUNGTRENNUGN
4’-selenoadenosine is incorrect
and in fact it is the N7-4’-sele-
noinosine (13b); this indicates
phy of 13a and 15 obtained after the removal of protective
groups of 9 and 10 with 50% TFA, respectively (Scheme 2).
Interestingly, the X-ray crystal structures of 13a and 15
(Supporting Information, Figure S3) showed different con-
formations, although they possess the same 4-selenoribose
sugar moiety. The N7 isomer 13a took the 2’-exo/3’-endo
(north) conformation, whereas the N9 isomer 15 adopted
the 2’-endo/3’-exo (south) conformation. These conforma-
tional differences may provide further insights towards un-
derstanding the differences in biological activity of the two
isomers. However, during all condensation experiments, no
traceable N3 isomer was observed.
Further, in order to establish the observed rearrangement
of the N7 to the N9 isomers during the condensation reac-
tions, several attempts were made. During the screening ex-
periments it was realised that the initially formed kinetic N7
isomer gradually rearranged to the thermodynamically
stable N9 isomer. To confirm this transformation, the isolat-
ed N7 isomer 9 was treated with TMSOTf in toluene at
958C. As expected, 9 immediately rearranged to 10 in 50%
yield. This is the first example for which the rearrangement
of the N7 to the N9 isomer has been confirmed with the aid
of X-crystal structures.
that the rearrangement of the N7 isomer into the N9 isomer
did not occur on heating with AcOH (see the Supporting In-
formation, Figure S1 and S2 for comparison of 1H and
13C NMR spectroscopy data).
The X-ray crystal structure of N9-4’-selenoadenosine (1a)
has two molecules (1a and 1a’) in one asymmetric unit, in
which surprisingly they show different sugar conformations
(Figure 2). The conformation of the glycosidic bond is anti
in both structures. The major differences between the two
molecules are on the sugar ring puckering and the confor-
À
mation about the C4’ C5’ bond. The sugar conformation of
the crystal of 1a is 2’-endo/3’-exo (south), with pseudorota-
tion parameters[11] of P=177.48 and tm =54.08, whereas the
sugar part of crystal 1a’ has a 2’-exo/3’-endo (north) confor-
mation with pseudorotation parameters P=0.548 and tm =
44.98. These two molecules also show different conforma-
À
tions about the C4’ C5’ bond; the values of torsion angle
g(C3’-C4’-C5’-O5’)2 are À77.4(3)8 and 57.8(4)8 for 1a and
1a’, respectively. This result indicates that the base and the
hydroxymethyl group undergo the same disrotatory motion
so that the Coulombic repulsion between atoms C8 and O5’
is minimised (the (Se4’-C1’-N9-C4) angle is À135.2(3)8 and
À173.5(2)8 for 1a and 1a’, respectively). This result is quite
surprising in that the dynamic equilibrium of the conformers
existing in solution state was thus confirmed in the solid
state.
The final synthesis of N9-4’-selenoadenosine (1a) and N9-
4’-selenoguanosine (1b) is illustrated in Scheme 2. The N7
isomer 9, on treatment with 50% TFA, gave 13a along with
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

L. S. Jeong et al.
packing of 4’-selenoadenosine
was investigated (Figure 3). It
was suggested that the ade-
nine–adenine base pairing (hy-
drogen bonds between N6 and
N1B (xÀ1,y,z) and N7 and
N6B (xÀ1,y,z) and p–p stack-
ing effect of adenine rings, led
to the stable DNA-like interac-
tions in this crystal packing.
This type of crystal packing
was also observed for S-adeno-
syl-l-homocysteine.[12]
It
should be also noted that the
two conformers have quite dif-
ferent thermal motions in the
crystal. The 1a molecule has
an average equivalent isotropic
displacement parameter of
3.076ꢂ10À2 ꢃ2 whereas the
value for 1a’ is 3.562ꢂ10À2.
Since both molecules are in-
volved in a similar number of
hydrogen bonds, the different
thermal motions may indicate
that 1a (south conformation in
sugar puckering) is more stable
than 1a’, which has a north
conformation in sugar pucker-
ing. This analysis is in com-
plete agreement with the
NMR spectroscopy result in
that the populations of the
Scheme 2. Synthesis of 4’-selenoadenosine (1a) and 4’-selenoguanosine (1b).
north and south conformers
were found to be 35 and 65%
for 4’-selenoadenosine in solution (Supporting Information,
Tables S3 and S4).
We have documented the first synthesis and consequent
structure revision of a new chemical moiety 4’-selenoadeno-
sine (1a) from d-ribose. Also, under identical conditions, we
successfully synthesised 4’-selenoguanosine (1b), which is of
significant interest for biological evaluation. For the first
time, we successfully confirmed the rearrangement of the
N7 to N9 isomer with the aid of X-ray structures. This study
is significant, as the long controversy can be put to rest at
least in the case of 4’-selenopurine nucleosides. Also, the ob-
served difference in sugar puckering of the N7 and N9 re-
gioisomers can provide better insights towards their inherent
biological activities. The X-ray crystal structure of 4’-sele-
noadenosine revealed an unusual mixture of north and
south conformers through A–A base pairing and p–p stack-
ing interactions. It is expected that these new RNA purine
building blocks and their unique conformations will be ex-
tensively useful for the design of therapeutically as well as
biochemically useful antimetabolites and oligonucleotides.
Figure 2. The X-ray crystal structure of 1a showing the unusual mixture
of the south (1a) and north (1a’) conformations through A–A base pair-
ing and p–p stacking interactions.
In order to elucidate why compound 1a exists as an un-
usual mixture of south and north conformations, the crystal
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

4’-Selenopurine Nucleosides
COMMUNICATION
2000, 100, 4319–4348; c) J. A.
Lee, L. S. Jeong, Antiviral Chem.
Chemother. 2004, 15, 235–250;
d) V. E. Marquez, M.-I. Lim,
Med. Res. Rev. 1986, 6, 1–40.
[5] a) L. S. Jeong, D. K. Tosh, H. O.
Kim, T. Wang, X. Hou, H. S.
Yun, Y. Kwon, S. K. Lee, J.
Choi, L. X. Zhao, Org. Lett.
2008,
10,
209–212;
b) V.
Alexander, W. J. Choi, J. Chun,
H. O. Kim, J. H. Jeon, D. K.
Tosh, H. W. Lee, G. Chandra, J.
Choi, L. S. Jeong, Org. Lett.
2010, 12, 2242–2245; c) K. Jaya-
kanthan, B. D. Johnston, B. M.
Pinto, Carbohydr. Res. 2008, 343,
1790–1800; d) J. K. Watts, B. D.
Johnston, K. Jayakanthan, A. S.
Wahba, B. M. Pinto, M. J.
Damha, J. Am. Chem. Soc. 2008,
130, 8578–8579.
[6] a) H. Vorbrꢁggen, C. Ruh-
Polenz, Org. React. 2000, 55, 1–
162; b) H. Vorbrꢁggen, K. Kroli-
kiewicz, B. Bennua, Chem. Ber.
1981, 114, 1234–1255.
Figure 3. A crystal packing projection of 4’-selenoadenosine (1a) along an axis. Hydrogen bonds are shown as
dotted lines.
[7] a) H. O. Kim, R. F. Schinazi, S.
Nampalli, K. Shanmuganathan,
D. L. Cannon, A. J. Alves, L. S.
Acknowledgements
Jeong, J. W. Beach, C. K. Chu, J. Med. Chem. 1993, 36, 30–37;
b) L. S. Jeong, R. F. Schinazi, J. W. Beach, H. O. Kim, S. Nampalli,
K. Shanmuganathan, A. J. Alves, A. McMillan, C. K. Chu, J. Med.
Chem. 1993, 36, 181–195.
This work was supported by grants from Midcareer Research Program
(2010-0026203) through the National Research Foundation, Korea and
Seoul R&BD Program (ST100039). J. Yu. is thankful to the Hi-Seoul
(Humanities) Fellowship funded by Seoul Scholarship Foundation.
[8] T. Naka, N. Minakawa, H. Abe, D. Kaga, A. Matsuda, J. Am. Chem.
Soc. 2000, 122, 7233–7243.
[9] H. Batra, R. M. Moriarty, R. Penmasta, V. Sharma, G. Stanciuc, J. P.
Staszewski, S. M. Tuladhar, D. A. Walsh, Org. Process Res. Dev.
2006, 10, 484–486.
[10] G. Y. Song, V. Paul, H. Choo, J. Morrey, R. N. Sidewell, R. F. Schina-
zi, C. K. Chu, J. Med. Chem. 2001, 44, 3985–3993.
Keywords: nucleic acids
selenonucleosides · X-ray crystallography
·
nucleosides
·
RNA
·
[11] IUPAC-IUB Joint Commission on Biochemical Nomenclature, Eur.
J. Biochem. 1983, 131, 9–15.
[12] a) H. S. Shieh, H. M. Berman, Acta Crystallogr. B 1982, 38, 1513–
1517; b) T. Ishida, A. Tanaka, M. Inuoe, T. Fujiwara, K. Tomita, J.
Am. Chem. Soc. 1982, 104, 7239–7248.
[1] R. T. Walker, E. De Clercq, F. Eckstein, in Nucleoside Analogues:
Chemistry, Biology, and Medical Applications (Eds.: R. T. Walker,
E. De Clercq, F. Eckstein), Plenum, New York, 1979, pp. 1–459.
[2] M. Nasr, C. Litterst, J. McGowan, Antiviral Res. 1990, 14, 125–148.
[3] P. Gunaga, H. R. Moon, W. J. Choi, D. H. Shin, J. G. Park, L. S.
Jeong, Curr. Med. Chem. 2004, 11, 2585–2637.
[4] a) A. Piperno, M. A. Chiacchio, D. Iannazzo, R. Romeo, Curr. Med.
Chem. 2006, 13, 3675–3695; b) M. Ferrero, V. Gotor, Chem. Rev.
Received: February 25, 2013
Published online: && &&, 0000
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

L. S. Jeong et al.
Writerꢀs blocks: The first synthesis of
RNA purine building blocks, 4’-sele-
noadenosine and 4’-selenoguanosine
Nucleic Acids
J. Yu, J.-H. Kim, H. W. Lee,
V. Alexander, H.-C. Ahn, W. J. Choi,
J. Choi, L. S. Jeong* ......... &&&& — &&&&
was achieved from d-ribose by re
ACHTUNGTRENNUNGgio-
A
R
ACHTUNGTRENNUNG
confirmed by X-ray crystallography. 4’-
Selenoadenosine exists in an unusual
mixture of N and S conformers in the
solid state.
New RNA Purine Building Blocks, 4’-
Selenopurine Nucleosides: First
Synthesis and Unusual Mixture of
Sugar Puckerings
RNA
Next-generation RNA purine building blocks, 4’-se
aden sine and 4’-selenoguanosine, were synthesised from
ACHUTNGRENlNUG eACTHNGUTRENNUnG o-
A
ACTHUNGTRENNUNoG ACHTUNGTRENNNUG
d-ribose by the rearrangement of the N7 to the N9 isomer.
The unambiguous confirmation of the rearrangement and
the co-existence of C2’-exo/C3’-endo (north) and C2’-endo/
C3’-exo (south) conformers in the solid state were demon-
strated. These RNA building blocks will open new
avenues in nucleoside, nucleotide, and nucleic acids
research. For more details, see the Communication by
L. S. Jeong et al. on page && ff.
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!