Synthesis of 6-Se-guanosine RNAs for structural study

Article abstract of DOI:10.1021/ol401698n

6-Se-guanosine phosphoramidite and RNAs have been synthesized by selenium substitution of the 6-oxygen atom, and it is revealed that the Se-derivatization is relatively stable and that bulge and wobble structures can better accommodate a large Se atom than a duplex. This Se-modification is useful in the structural study of RNAs and their protein complexes.

Full text of DOI:10.1021/ol401698n

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of 6‑Se-Guanosine RNAs
for Structural Study
Jozef Salon, Jianhua Gan, Rob Abdur, Hehua Liu, and Zhen Huang*
Department of Chemistry, Georgia State University, Atlanta, Georgia 30303, United States
huang@gsu.edu
Received June 14, 2013
ABSTRACT
6-Se-guanosine phosphoramidite and RNAs have been synthesized by selenium substitution of the 6-oxygen atom, and it is revealed that
the Se-derivatization is relatively stable and that bulge and wobble structures can better accommodate a large Se atom than a duplex. This
Se-modification is useful in the structural study of RNAs and their protein complexes.
RNA secondary structures and motifs help to diversify
structures and functions of noncoding RNAs (ncRNA).
Bulges and U/G wobble pairs are frequent structural
motifs in ncRNAs.1,2 Functions of these motifs are re-
vealed by structural study, such as X-ray crystallography,
one of the most powerful tools for structure determination
of biomacromolecules and complexes.3,4 Structures of pro-
teins and proteinꢀnucleic acid complexes are commonly
determined by selenium-derivatized (i.e., selenomethionyl)
proteins via multiwavelength or single-wavelength anoma-
lous diffraction (MAD or SAD) phasing.5ꢀ7 Inspired by the
Se-derivatized protein approach, our research group has
pioneered and developed selenium-derivatized nucleic acids
(SeNA) for the X-ray crystal structure study of nucleic acids
and proteinꢀnucleic acid complexes.8ꢀ10
To facilitate the RNA structure study, the Se-atom-
specific modification has been developed by us. Herein
we report the first synthesis of the 6-Se-guanosine (SeG)
phosphoramidite and SeG-containing RNAs. Since ncRNAs
often contain bulges and U/G wobble pairs in the duplex
regions,1,6 the stability of the G/C pairs close to bulge and
wobble-pair structures has been investigated, by using the
selenium atomic probe. The UV-melting study indicates
that, in the presence ofthe bulge and wobble structures, the
native and Se-modified G/C structures have similar stabil-
ity. This Se atomic probe can be used in studying RNA
secondary structures. Moreover, theSeG-RNA wasused as
the transient Se-modification to enhance crystallization,
although the Se-derivatization is normally used for phase
determination. Since crystallization is a major challenge
in X-ray crystallography, a Se-RNA/DNA/protein com-
plex was used to explore the crystallization and crys-
tal quality. Particularly, we intended to investigate the
crystallization property of RNA transiently modified with
(1) Cate, J. H.; Gooding, A. R.; Podell, E.; Zhou, K.; Golden, B. L.;
Kundrot, C. E.; Cech, T. R.; Doudna, J. A. Science 1996, 273, 1678–1685.
(2) Macrae, I. J.; Zhou, K.; Li, F.; Repic, A.; Brooks, A. N.; Cande,
W. Z.; Adams, P. D.; Doudna, J. A. Science 2006, 311, 195–198.
(3) Liu, Q.; Dahmane, T.; Zhang, Z.; Assur, Z.; Brasch, J.; Shapiro,
L.; Mancia, F.; Hendrickson, W. A. Science 2012, 336, 1033–1037.
(4) Nakanishi, K.; Weinberg, D. E.; Bartel, D. P.; Patel, D. J. Nature
2012, 486, 368–374.
(5) Hendrickson, W. A. Trends Biochem. Sci. 2000, 25, 637–643.
(6) Ferre-D’Amare, A. R.; Zhou, K. H.; Doudna, J. A. Nature 1998,
395, 567–574.
(7) Schirle, N. T.; MacRae, I. J. Science 2012, 336, 1037–1040.
(8) Carrasco, N.; Ginsburg, D.; Du, Q.; Huang, Z. Nucleosides
Nucleotides Nucleic Acids 2001, 20, 1723–1734.
(9) Lin, L.; Sheng, J.; Huang, Z. Chem Soc Rev 2011, 40, 4591–4602.
(10) Sun, H.; Sheng, J.; Hassan, A. E.; Jiang, S.; Gan, J.; Huang, Z.
Nucleic Acids Res. 2012, 40, 5171–5179.
r
10.1021/ol401698n
XXXX American Chemical Society

the Se-functionality. Surprisingly, the transient Se-RNA
complexed with DNA and RNase H has offered high-
quality crystals and resulted in high-resolution structure
determination.
steps) by the substitution of the sulfonyl-activating group
with sodium 2-cyanoethylselenide generated by the reduction
of di-(2-cyanoethyl) diselenide with NaBH4. In a satisfactory
yield, the 6-Se-guanosine derivative (3) was converted to the
Se-phosphoramidite (4).10,13 The characterization of these
Se-nucleoside derivatives is presentedin Figures S1ꢀS10 in
the Supporting Information (SI).
Scheme 1. Synthesis of SeG-RNAs (5)
Despite the synthesis of the 6-selenoguanine and its
derivatives five decades ago,11 synthesis of RNA contain-
ing the Se-guanine has not been accomplished because of
the synthetic challenge, even though 6-S-purines have been
introduced into nucleic acids.12 Recently our laboratory
has successfully developed a novel strategy to incorporate
the selenium functionality to the 6-position of deoxygua-
nosine in DNA.13 This successful strategy has encouraged
us to introduce the selenium functionality to the 6-position
of guanosine in RNA. Thus the synthesis of the 6-seleno-
guanosine (SeG) phosphoramidite and SeG-RNAs is devel-
oped and reported.
Our development of the 2-cyanoethyl-seleno protection
and deprotection for the Se-nucleobase nucleic acids10,14
has encouraged us to protect the 6-Se-functionality on
guanosine with the 2-cyanoethyl-seleno group. In addi-
tion, the 2-cyanoethyl protecting group can be removed
under ultramild conditions (0.05 M K2CO3 in methanol).
Since strong basic conditions can cause deselenization,
phenoxyacetyl (Pac) was used as the protecting group for
the 2-NH2 of this 6-Se-guanosine phosphoramidite (4),
which can also be removed under the ultramild conditions.
Our synthesis (Scheme 1) started from the partially pro-
tected guanosine derivative (1). This commercial com-
pound was activated at the 6-position of the guanine
with a selective sulfonylation. To avoid protection of the
30-hydroxyl group, 2,4,6-triisopropylbenzenesulfonyl chlo-
ride (TIBS-Cl) was explored.13 Without purification of the
TIBS-activated intermediate (2), the protected 6-Se-
guanosine derivative (3) was obtained (82% yield in two
Figure 1. HPLC analysis and thermostability study. The purified
RNA samples [SeG-RNA: r(50-G-SeG-UAUUGCGGUACC-30);
3 μM each] were heated at 60 °C in 100 mM sodium phosphate
buffer (pH 7.6) for 1, 2, and 3 h. (A) RP-HPLC analysis at
360 nm, no heat; (B) 1-h heating (360 nm); (C) 2-h heating (360 nm);
(D) 3-h heating (360 nm); (E) 3-h heating (260 nm) in blue line; 3-h
heating (360 nm) in red line. HPLC conditions: Welchrom XB-C18
column (4.6 mm ꢁ 250 mm); flow rate, 1 mL/min; column tem-
perature, 25 °C; gradient, 5 to 35% B in 10 min. Buffer A: 10 mM
TEAAc (pH 7.4). Buffer B: 10 mM TEAAc (pH 7.4) in 60%
acetonitrile dissolved in water (6:4 in v/v).
Finally, the SeG-phosphoramidite was incorporated
into RNAs by solid-phase synthesis. The SeG-RNAs were
synthesized using 4, the ultramild CE phosphoramidites
(A, C, and G),10 and BTT activator. Since strong basic
conditions (such as NH3ꢀH2O) for the deprotection cause
the deselenization of SeG, the K2CO3 deprotection con-
ditions were used to remove all ultramild protecting
groups.10,13 If RNAs have any guanosine nucleotides,
phenoxyacetic anhydride (Pac2O, instead of Ac2O) is used
in the capping step to avoid the guanosine acetylation,
which is difficult to remove under K2CO3 treatment. After
the K2CO3 deprotection and fluoride treatment to remove
TBDMS groups, the synthesized SeG-RNAs (5) were
purified by HPLC twice (with and without the 50-DMTr
group). Interestingly, the deprotected Se-RNAs are yellow
colored, which is the first observation of the colored RNAs
after a single atom replacement, while natural RNAs are
colorless. In addition, to measure the coupling efficiency of
Se
the G-phosphoramidite (4), 50-DMTr-SeGG dinucleo-
(11) Mautner, H. G.; Chu, S. H.; Jaffe, J. J.; Sartorelli, A. C. J. Med.
Chem. 1963, 6, 36–39.
(12) Coleman, R. S.; Arthur, J. C.; McCary, J. L. Tetrahedron 1997,
53, 11191–11202.
(13) Salon, J.; Jiang, J.; Sheng, J.; Gerlits, O. O.; Huang, Z. Nucleic
Acids Res. 2008, 36, 7009–7018.
(14) Salon, J.; Sheng, J.; Jiang, J.; Chen, G.; Caton-Williams, J.;
tide was synthesized, analyzed by RP-HPLC, and com-
pared with the synthesis and analysis of the native
50-DMTr-GG, which indicated a high coupling yield (over
96%). The purified Se-RNAs were confirmed by HPLC and
MS (Table 1). The typical HPLC, MS, and UV profiles of the
SeG-RNAs are shown in Figures 1, S11, and S12.
Huang, Z. J. Am. Chem. Soc. 2007, 129, 4862–4863.
B
Org. Lett., Vol. XX, No. XX, XXXX

Table 1. MALDI-TOF MS Data of the SeG-RNAs
Table 2. UV-Melting Temperature of the SeG-RNAs
Tm (°C)
measured
entry
1
Se-RNAs
50-UUC-SeG-CG-30
(calcd) m/za
entry
RNAs
(ΔTm per Se)
[M þ H]þ: 1915
(1915.1)
1
2
3
4
5
6
(50-UACUAAC---G-UA---G-UA-30)2
(50-UACUAAC-SeG-UA---G-UA-30)2
(50-UACUAAC---G-UA-SeG-UA-30)2
(50-G---G-UAUU-G---CGGUACC-30)2
(50-G---G-UAUU-GSe-CGGUACC-30)2
(50-G-SeG-UAUU-G---CGGUACC-30)2
39.0 ( 0.1
C56H71N20P5O41Se: FW 1914.1
50-CCG CGC-SeG-CG dG-30
C95H121N40P9O66Se: FW 3236.4
50-UAC UAA C-SeG-U AGU A-30
C124H153N49P12O87Se: FW 4171.5
50-UAC UAA CGU A-SeG-U A-30
C124H153N49P12O87Se: FW 4171.5
50-G-SeG-U AUU GCG GUA CC-30
C133H165N52P13O97Se: FW 4524.5
50-GGU AUU-SeG-CG GUA CC-30
C133H165N52P13O97Se: FW 4524.5
0
38.5 ( 0.2 (0.3)
36.5 ( 0.3 (1.3)
54.0 ( 0.1
47.2 ( 0.2 (3.4)
40.1 ( 0.2 (6.9)
2
3
4
5
6
7
8
[M þ H]þ: 3237
(3237.4)
[M þ H]þ: 4173
(4172.5)
[M þ H]þ: 4173
(4172.5)
[M þ H]þ: 4526
(4525.5)
heated at 60 °C for 3 h, and no significant deselenization
of the Se-RNA was observed. The UV-melting tempera-
tures of the SeG-RNAs were measured (Tm, Table 2) to
examine the impact of SeG on the G/C pairs close to
various secondary structures (Figure 2A and 2B), such as
the bulge, U/G wobble pair, and duplex. Three typical
melting curves are shown in Figure 2C. The melting tem-
peratures indicate that the SeG/C pairs next to the two
A-bulges do not cause significant destabilization (0.3 °C
per Se modification) of the bulged structure (Figure 2A),
while SeG/C in the duplex structure causes more destabi-
lization (1.3 °C per Se) in the same sequence. In this RNA
crystal structure, these two A-nucleotides flip out,15 and
the space generated by the bulges can better accommodate
the large Se atoms. Thus, these two Se atoms barely cause
any structural destabilization (0.3 °C per Se), which is
consistent with the crystal structure study. Similarlly, the
SeG/C close to the U/G wobble pair (Figure 2B) causes less
destabilization (3.4 °C per Se) than that in a duplex (6.9 °C
per Se). Significant destabilization of RNA duplex by the
SeG/C pair has also been observed in a DNA duplex
(decreased by up to 11 °C per Se).13 These results suggest
that the Se-accommodation is dependent on the locations
and secondary structures of nucleic acids. Thus, the sele-
nium atomic probe can be utilized to study the secondary
structures of RNAs and RNAꢀprotein complexes.
[M þ H]þ: 4526
(4525.5)þ
50-G-SeG-U2 SeMe-AUUGCGGUACC2 OMe-30 [M þ H] : 4602
0
C135H169N52P13O95Se2: FW 4600.7
(4601.7)þ
50-G-SeGUA-U2 SeMe-UGCGGUACC2 OMe-30 [M þ H] : 4602
0
0
C135H169N52P13O95Se2: FW 4600.7
(4601.7)
a 3-Hydroxypicolinic acid (HPA) matrix was used.
Furthermore, we have found that the selenium derivati-
zation can help reduce the local conformational flexibility
of the modified molecules and duplexes.17,18 Thus we
hypothesizedthattheSeheavyatommay reduce molecular
dynamics, thereby facilitating molecular packing and crys-
tallization. Thus, in addition to using Se-derivatization for
phase determination, Se-modification was used for crystal
growth. To demonstrate the proof of principle, the crystal-
lization of the RNA/DNA/RNase H complex was exam-
ined with the Se-derivatized RNA. Initially, crystals were
yellow and generated by the yellow SeG-RNA. The crystal
color can be conveniently used to indicate whether RNA is
present in a crystal. As shown in Figure 1, the SeG-RNA
is relatively stable. However, during the crystallization
over weeks in buffer, the Se-modification was naturally
Figure 2. RNA secondary structures and normalized UV-
melting curves of RNA 1-6 in Table 2. (A) Secondary structures
of RNAs: 1 (native in blue), 2 (with black SeG), and 3 (with red
SeG). (B) Secondary structures of RNAs: 4 (native in blue), 5
(with black SeG), and 6 (with red SeG). The underlined sequences
are U/G wobble pairs. (C) Blue curve: melting curve of the
native RNA containing two adenosine bulges (1; Tm = 39.0 °C).
Black curve: melting curve of 2 (Tm = 38.5 °C). Red curve:
melting curve of 3 (Tm = 36.5 °C). The melting temperatures of
the duplexes (2.5 μM) were measured in 10 mM phosphate
buffer (pH 6.5) containing 50 mM NaCl. The absorbance of the
samples was monitored at 260 nm, and the duplexes were headed
from 5 to 70 °C with a rate of 1 °C/min.
(15) Berglund, J. A.; Rosbash, M.; Schultz, S. C. RNA 2001, 7, 682–691.
(16) Nowotny, M.; Yang, W. EMBO J. 2006, 25, 1924–1933.
(17) Jiang, J.; Sheng, J.; Carrasco, N.; Huang, Z. Nucleic Acids Res.
2007, 35, 477–485.
(18) Sheng, J.; Zhang, W.; Hassan, A. E.; Gan, J.; Soares, A. S.;
Geng, S.; Ren, Y.; Huang, Z. Nucleic Acids Res. 2012, 40, 8111–8118.
Moreover, the HPLC analysis indicated that this
Se-functionality in RNA is relatively stable (Figure 1).
Se
The G-RNA (50-G-SeG-UAUUGCGGUACC-30) was
Org. Lett., Vol. XX, No. XX, XXXX
C

˚
and RNase H did not offer high resolution (only 2.70 A).
These two native structures from directly and transiently
grown crystals are virtually identical (Figure 3). The high
resolution offered by the transient Se-derivatization can
probably be attributed to the reduced molecular dynamics
and bettermolecularpackingassistedby the Se heavyatom
initially. This Se-modification may have great potential in
investigating structures of RNAs and proteinꢀRNA
complexes.
In summary, the 6-Se-guanosine (SeG) phosphoramidite
and SeG-derivatized RNAs have been synthesized. The Se-
modification is relatively stable, and the bulge and wobble
structures can better accommodate the large Se atom than
a duplex structure. This Se atomic probe can be used in
studying RNA secondary structures, and the yellow color
of the SeG-RNA is useful for proteinꢀRNA cocrystalliza-
tion visualization. Moreover, we have demonstrated the
proof of principle that this novel Se-derivatization can be
used for transiently Se-assisted molecular packing and
crystallization forhigh-resolutionstructure determination.
Our exciting discoveries will open a new research avenue in
structure and function studies of ncRNAs. This Se probe
also offers a useful tool for the structural study of RNAs as
well as their protein complexes.
Figure 3. Structure comparison and crystal picture of the RNA/
DNA/RNase H complexes. (A) The Se-transient native struc-
˚
ture (resolution: 1.60 A; PDB: 3ULD) is in cyan (protein) and
red (RNA/DNA duplex), while the direct native structure
˚
(resolution: 2.70 A; PDB: 2G8U) is in gray (protein) and green
(RNA/DNA duplex).16 The sequences of the RNA (50-UCGA-
CA-30) and DNA (50-ATGTCG-30) are the same in these two
structures. (B) Picture of the crystal (0.2 ꢁ 0.2 ꢁ 0.2 mm3; slightly
yellow) of the transient Se-RNA/DNA/RNase H complex.
removed (by hydrolysis) from the Se-modified RNA. In
addition to the time, the crystallization pH (6.0; SI)
probably also played a critical role in the deselenization.
Due to the deselenization, the crystal’s yellow color gra-
dually became colorless. The replacement of the selenium
atoms by oxygen generated the native crystals eventually.
The loss of selenium was confirmed by MS analysis of the
crystals of the Se-RNA [50-UC(SeG)ACA-30] complexed
with DNA (50-ATGTCG-p-30, one-base overhang at
50-end) and RNase H. Surprisingly, the transient Se-RNA
complexed with DNA and RNase H has offered high
quality crystals. The crystal structure was determined by
the molecular replacement, and the structure assisted by
the transient Se-modification offeredhigh resolution(PDB
Acknowledgment. We would like to thank the beamline
staff at Advanced Light Source beamline 8.2.1, in Lawrence
Berkeley National Laboratory, for their help in the data
collection. Plasmids expressing RNase H proteins were
kindly given to us by Dr. Wei Yang at the NIH. This work
was financially supported by the NIH (R01GM095881)
and the Georgia Cancer Coalition (GCC) Distinguished
Cancer Clinicians and Scientists program.
Supporting Information Available. Experimental pro-
cedures, 1H and 31P NMR, HRMS and MALDI-TOF MS
analytical data, HPLC profiles. These materials are avail-
able free of charge via the Internet at http://pubs.acs.org.
˚
ID: 1.60 A; PDB: 3ULD), while the native crystals grown
with the corresponding native RNA complexed with DNA
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX