An efficient unnatural base pair for a base-pair-expanded transcription system

Article abstract of DOI:10.1021/ja0425280

For the site-specific incorporation of artificial components into RNA by transcription, an efficient, unnatural base pair between 2-amino-6-(2-thiazolyl) purine (denoted as v) and 2-oxo(1H)pyridine (denoted as y) was developed. The substrates of y and 5-s

Full text of DOI:10.1021/ja0425280

Published on Web 05/25/2005
An Efficient Unnatural Base Pair for a Base-Pair-Expanded
Transcription System
Tsuneo Mitsui,† Michiko Kimoto,‡ Yoko Harada,‡ Shigeyuki Yokoyama,*,‡,§,| and
Ichiro Hirao*,†,‡
Contribution from the Research Center for AdVanced Science and Technology, The UniVersity of
Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8904, Japan, Protein Research Group, RIKEN
Genomic Sciences Center, 1-7-22 Suehiro-cho, Tsurumi-ku, Yokohama, Kanagawa 230-0045,
Japan, Department of Biophysics and Biochemistry, Graduate School of Science, The UniVersity
of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, and RIKEN Harima Institute at
SPring-8, 1-1-1 Kohto, Mikazuki-cho, Sayo, Hyogo 679-5148, Japan
Received December 13, 2004; E-mail: ihirao@riken.jp; yokoyama@biochem.s.u-tokyo.ac.jp
Abstract: For the site-specific incorporation of artificial components into RNA by transcription, an efficient,
unnatural base pair between 2-amino-6-(2-thiazolyl)purine (denoted as v) and 2-oxo(1H)pyridine (denoted
as y) was developed. The substrates of y and 5-substituted y were site-specifically incorporated into RNA
by T7 RNA polymerase opposite v in templates. The efficiency and fidelity of the v-y pairing in transcription
were as high as those of the natural A-T(U) and G-C pairings. Furthermore, RNAs containing two adjacent
y bases were also transcribed from DNA templates containing two v bases. This specific transcription
allows the large-scale preparation of artificial RNAs and can be combined with other systems to
simultaneously incorporate several different components into a transcript.
Transcription mediated by unnatural base pairs enables the
site-specific incorporation of artificial components into RNA
for generating novel RNA molecules with increased functional-
ity.1,2 In addition, the combination of this specific transcription
with conventional translation systems can expand the genetic
code toward the synthesis of proteins containing amino acid
analogues.3,4 For these purposes, the creation of unnatural base
pairs that are recognized and replicated by polymerases has been
studied on the basis of several different concepts, relying on
nonstandard hydrogen-bonding patterns,5,6 shape comple-
mentarity,7-9 and hydrophobic interactions.10-12 However, a few
unnatural base pairs, such as isoguanine (isoG)-isocytosine
(isoC)5,13 and xanthosine-diaminopyrimidine,6 have been tested
in transcription. Only one example of the incorporation of a
functional component into RNA, using N6-(6-aminohexyl)isoG
and 5-methyl-isoC, has been reported, and the yield of the
transcript containing N6-(6-aminohexyl)isoG was approximately
50% of that for transcription using natural templates and
substrates.14 There was no other extra base pair for the efficient,
site-specific incorporation of a variety of artificial components
into RNA.
Recently, we developed unnatural base pairs of 2-amino-6-
(N,N-dimethylamino)purine (denoted by x) and 2-oxo(1H)-
pyridine (denoted by y)15 and of 2-amino-6-(2-thienyl)purine
(denoted by s) and y16 (Figure 1). These unnatural base pairs
were designed to employ different shape fitting and hydrogen-
bonding patterns from those of the natural base pairs, and they
exhibit high specificity in transcription. In particular, the fidelity
of the y incorporation into RNA opposite s in templates by T7
RNA polymerase is as high as that of the natural base
incorporations opposite the complementary bases. Using the s-y
pair, we have created an extra codon-anticodon interaction and
achieved the site-specific incorporation of an amino acid
analogue, 3-chlorotyrosine, into a protein.16 In addition, chemi-
cally synthesized 5-modified y, such as 5-iodo-y (I-y)17 and
5-phenylethynyl-y (Ph-y),18 can also be incorporated into
† Research Center for Advanced Science and Technology, The University
of Tokyo.
‡ RIKEN Genomic Sciences Center.
§ Graduate School of Science, The University of Tokyo.
| RIKEN Harima Institute at SPring-8.
(1) Benner, S. A.; Burgstaller, P.; Battersby, T. R.; Jurczyk, S. In The RNA
World, 2nd ed.; Gesteland, R. F., Cech, T. R., Atkins, J. F., Eds.; Cold
Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1999; p 163.
(2) Bergstrom, D. E. Chem. Biol. 2004, 11, 18.
(3) Wang, L.; Schultz, P. G. Chem. Commun. 2002, 1.
(4) Hendrickson, T. L.; de Cre´cy-Lagard, V.; Schimmel, P. Annu. ReV. Biochem.
2004, 73, 147.
(5) Switzer, C.; Moroney, S. E.; Benner, S. A. J. Am. Chem. Soc. 1989, 111,
8322.
(6) Piccirilli, J. A.; Krauch, T.; Moroney, S. E.; Benner, S. A. Nature 1990,
343, 33.
(7) Morales, J. C.; Kool, E. T. Nat. Struct. Biol. 1998, 5, 950.
(8) Matray, T. J.; Kool, E. T. Nature 1999, 399, 704.
(9) Rappaport, H. P. Biochemistry 1993, 32, 3047.
(10) McMinn, D. L.; Ogawa, A. K.; Wu, Y.; Liu, J.; Schultz, P. G.; Romesberg,
F. E. J. Am. Chem. Soc. 1999, 121, 11585.
(11) Wu, Y.; Ogawa, A. K.; Berger, M.; McMinn, D. L.; Schultz, P. G.;
Romesberg, F. E. J. Am. Chem. Soc. 2000, 122, 7621.
(12) Mitsui, T.; Kitamura, A.; Kimoto, M.; To, T.; Sato, A.; Hirao, I.; Yokoyama,
S. J. Am. Chem. Soc. 2003, 125, 5298.
(13) Switzer, C. Y.; Moroney, S. E.; Benner, S. A. Biochemistry 1993, 32, 10489.
(14) Tor, Y.; Dervan, P. B. J. Am. Chem. Soc. 1993, 115, 4461.
(15) Ohtsuki, T.; Kimoto, M.; Ishikawa, M.; Mitsui, T.; Hirao, I.; Yokoyama,
S. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 4922.
(16) Hirao, I.; Ohtsuki, T.; Fujiwara, T.; Mitsui, T.; Yokogawa, T.; Okuni, T.;
Nakayama, H.; Takio, K.; Yabuki, T.; Kigawa, T.; Kodama, K.; Yokogawa,
T.; Nishikawa, K.; Yokoyama, S. Nat. Biotechnol. 2002, 20, 177.
(17) Kimoto, M.; Endo, M.; Mitsui, T.; Okuni, T.; Hirao, I.; Yokoyama, S. Chem.
Biol. 2004, 11, 47.
9
8652
J. AM. CHEM. SOC. 2005, 127, 8652-8658
10.1021/ja0425280 CCC: $30.25 © 2005 American Chemical Society

An Unnatural Base Pair for Specific Transcription
A R T I C L E S
of conformation 2, because the function of the C-H moiety of
s is similar to that of the methyl group in the 6-dimethylamino
group of x (Figure 1b), and the efficiency and selectivity of the
s-y pairing in transcription and replication are higher than those
of the x-y pairing.16,19 Thus, the replacement of the C-H
moiety of s with other electronegative atoms could improve the
incorporation efficiency of y into RNA by transcription.
Here, we have designed an unnatural base, 2-amino-6-(2-
thiazolyl)purine (denoted by v) (Figure 1c), instead of s, as a
pairing partner of y. The two conformations of the v base,
resulting from the free rotation of the 6-thiazolyl group, exhibit
similar electronegative effects on the pairing partners, thus
enhancing the transcription efficiency of the v-y pairing over
that of the s-y pairing.
The 2′-deoxyribonucleoside of v (3) was synthesized by
coupling the 2-amino-6-toluenesulfonylpurine nucleoside de-
rivative (1) and 2-tributylstannylthiazole20,21 (47% overall yield
from 2′-deoxyguanosine (dG) in four steps, Scheme 1). It is
noteworthy that the 2′-deoxyribonucleoside 3 exhibited a
fluorescence emission centered at 457 nm, characterized by two
major excitation maxima (295 and 364 nm), and its fluorescence
quantum yield (excitation at 363 nm) was 0.41 in ethanol. The
fluorescence emission of v was stronger than that of s (the
quantum yield was 0.23).22 For the amidite synthesis, the
2-amino group was protected by a phenoxyacetyl group, and
the 5′- and N2-protected nucleoside 5 was converted to the
phosphoramidite 6. The 2′-deoxyribonucleoside of v was slightly
decomposed by treatments with concentrated ammonia for 6 h
at 55 °C. Thus, for DNA synthesis, base-labile protection, using
phenoxyacetyl for A, p-isopropylphenoxyacetyl for G, and acetyl
groups for C, was used,23 and deprotection of DNA fragments
was carried out for 2-3 h at 55 °C with concentrated ammonia.
The v nucleotide in DNA fragments was more stable than that
expected from the stability of the 2′-deoxyribonucleoside, and
no decomposition of the DNA fragments was observed under
the deprotection conditions. The coupling efficiency of the
amidite of v was more than 98% on a DNA synthesizer (Applied
Biosystems, CA).
Before testing the base pairs in transcription, to assess the
selectivity and efficiency of the v-y pairing in polymerase
reactions, we determined the kinetic parameters of the single-
nucleotide insertion of y and natural substrates into DNA
opposite v or s in templates, using the exonuclease-deficient
Klenow fragment of Escherichia coli DNA polymerase I. This
is because the assessment of unnatural base pairings by the
kinetic parameters in replication is much easier than that in
transcription.24,25 The single-nucleotide insertion experiments
were carried out using the nucleoside triphosphate of y (dyTP)
or the natural bases and a partially double-stranded template
(35-mer, Table 1) containing v or s, with a primer labeled with
6-carboxyfluorescein (20-mer), in which various bases in the
template were adjacent to the 3′ end of the primer. The insertion
Figure 1. Unnatural base pair structures. Two conformations of the s-y
pair (a), the x-y pair (b), and two conformations of the v-y pair (c). The
substrates of 5-modified y bases can be synthesized chemically.
RNA molecules; I-y functions as a photocrosslinking component
or a rational phasing tool for X-ray crystallography, and Ph-y
stabilizes higher-ordered RNA structures. In addition, various
functional substrates can also be derived chemically from the
nucleoside of I-y. Thus, the method for the site-specific
incorporation of y and modified y bases into RNA would be a
powerful tool for the creation of novel RNA molecules with
increased functionality.
Although the s-y pairing exhibits high selectivity with T7
RNA polymerase, the transcription efficiency involving one s-y
pair is 50-60% lower than that of native transcription mediated
by only the natural base pairs. In addition, this transcription
cannot synthesize RNA molecules containing two adjacent y
bases. Thus, to develop more efficient base-pair-expanded
transcription systems, we have continued to improve the
unnatural base pair. One of the problems of the s-y pair is the
two possible conformations of s, caused by the free rotation of
the thienyl group at position 6 of s, as shown in Figure 1a.
Either the electronegative sulfur atom of conformation 1 or the
C-H moiety of conformation 2 is located on the pairing surface
with y, but both groups sterically prevent the pairing with the
natural bases. The sulfur atom of conformation 1 may be more
efficient for the specific pairing with y than the C-H moiety
(19) Fujiwara, T.; Kimoto, M.; Sugiyama, H.; Hirao, I.; Yokoyama, S. Bioorg.
Med. Chem. Lett. 2001, 11, 2221.
(20) Peters, D.; Ho¨rnfeldt, A.-B.; Gronowitz, S. J. Heterocycl. Chem. 1990, 27,
2165.
(21) Gutierrez, A. J.; Terborst, T. J.; Matteucci, M. D.; Froehler, B. C. J. Am.
Chem. Soc. 1994, 116, 5540.
(22) Hirao, I.; Harada, Y.; Kimoto, M.; Mitsui, T.; Fujiwara, T.; Yokoyama, S.
J. Am. Chem. Soc. 2004, 126, 13298.
(23) Schulhof, J. C.; Molko, D.; Teoule, R. Tetrahedron Lett. 1987, 28, 51.
(24) Vaught, J. D.; Dewey, T.; Eaton, B. E. J. Am. Chem. Soc. 2004, 126, 11231.
(25) Goodman, M. F.; Creighton, S.; Bloom, L. B.; Petruska, J. Crit. ReV.
Biochem. Mol. Biol. 1993, 28, 83.
(18) Endo, M.; Mitsui, T.; Okuni, T.; Kimoto, M.; Hirao, I.; Yokoyama, S.
Bioorg. Med. Chem. Lett. 2004, 14, 2593.
9
J. AM. CHEM. SOC. VOL. 127, NO. 24, 2005 8653

A R T I C L E S
Mitsui et al.
Scheme 1. Synthesis of the Nucleoside and the Phosphoramidite of va
a Conditions: (a) tert-butyldimethylsilyl chloride, imidazole, DMF, 90%; (b) p-toluenesulfonyl chloride, Et3N, 4-(dimethylamino)pyridine, CH2Cl2, 94%;
(c) 2-tributylstannylthiazole, Pd(PPh3)4, LiCl, dioxane, 70%; (d), TBAF, THF, 80%; (e) (i) trimethylsilyl chloride, pyridine, (ii) 1-hydroxybenzotriazole,
phenoxyacetyl chloride, pyridine, CH3CN, 90%; (f) 4,4′-dimethoxytrityl chloride, pyridine, 99%; (g) ClP(N-iPr2)(OCH2CH2CN), iPr2EtN, THF, 83%.
Abbreviations: TBDMS, tert-butyldimethylsilyl; Ts, p-toluenesulfonyl; Pac, phenoxyacetyl; DMT, 4,4′-dimethoxytrityl.
104) (Table 1, entry 7). Although the efficiency of the C
Table 1. Steady-State Kinetic Parameters for Insertion of Single
Nucleotides into a Template-Primer Duplex by the
incorporation opposite v (Vmax/KM ) 5.7 × 104) (Table 1, entry
Exonuclease-Deficient Klenow Fragmenta
3) was slightly increased relative to that opposite s (Vmax/KM
)
primer 1
template 1
template 2
5
′
-ACTCACTATAGGGAGGAAGA
-TATTATGCTGAGTGATATCCCTCCTTCTNTCTCTT
′-TATTATGCTGAGTGATATCCCTCCTTCTNTCTCGA
4.2 × 104) (Table 1, entry 8), the selectivity of the cognate
v-y pairing was 3.3-fold higher than that of the v-C pairing,
and the selectivity of the s-y pairing was only 1.5-fold higher
than that of the s-C pairing. Although the efficiency of the y
incorporation opposite v was still 15-fold lower than that of
the T incorporation opposite A (Table 1, entry 11), the v-y
pair was more effective than the representative unnatural base
pair between isoG and isoC;5 the y incorporation opposite v
(Vmax/KM ) 1.9 × 105) was 2.7-fold higher than the isoG
incorporation opposite isoC (Vmax/KM ) 7.0 × 104).12
3
′
3
template
nucleoside
KM
M)
Vmax
(% min-
efficiency
1
d
entry
(N)
triphosphate
(
µ
)
(Vmax/KM)
1
2
3
4
5
6
7
8
9
template 1 (v)
template 1 (v)
template 1 (v)
template 1 (v)
template 1 (v)
template 2 (s)
template 2 (s)
template 2 (s)
template 2 (s)
y
210 (140)b 40 (20)
1.9 × 105
6.8 × 103
5.7 × 104
T
C
A
G
y
T
C
A
G
T
C
280 (90)
370 (70)
89 (25)
n.d.c
1.9 (0.7)
21 (5)
0.37 (0.08) 4.2 × 103
n.d.c
170 (40)
270 (40)
430 (180)
87 (27)
n.d.c
11 (2)
6.5 × 104
1.1 × 104
4.2 × 104
3.1 (0.2)
18 (9)
0.31 (0.07) 3.6 × 103
The different selectivity and efficiency between the v-y and
s-y pairings suggest the existence of both conformations (1
and 2) of v and s in the single-stranded DNA templates. The
improvement of the v-y pairing indicates that the nitrogen of
conformation 2 of v is favorable for the C-H moiety of
conformation 2 of s, for the selective pairing with y. To exclude
the pairing with T, the electronegative nitrogen atom of the
thiazolyl group is more effective than the C-H moiety of the
thienyl group, and to sterically fit with y, the nitrogen of the
thiazolyl group may be favorable for the C-H moiety. In
addition, the relative abundances of conformations 1 and 2 of
v might differ from those of s, and this difference would affect
the selectivity and the efficiency of these base pairs in
replication. If conformation 2 of v and conformation 1 of s are
dominant in solution, then the efficiencies of the y incorporation
opposite v (Vmax/KM ) 1.9 × 105) and s (Vmax/KM ) 6.5 ×
104) would correlate with the van der Waals radii of the
heteroatoms in the 6-groups of these bases opposite the H6 of
y: the nitrogen (1.55 Å) in the thiazolyl group of v and the
sulfur (1.80 Å) in the thienyl group of s. This correlation is
supported by analyses of another unnatural base, 2-amino-6-
(2-furanyl)purine (denoted by o);27 conformation 1 of o has the
smaller oxygen (1.52 Å) in the furanyl group opposite the H6
10 template 2 (s)
11 template 1 (A)
12 template 1 (A)
n.d.c
1.2 (0.3)
3.5 (1.3)
2.9 × 106
1600 (600) 2.5 (1.3)
1.6 × 103
a Assays were carried out at 37 °C for 1-20 min using 5 µM template-
primer duplex, 3-66 nM enzyme, and 0.6-3000 µM nucleoside triphos-
phate in a solution (10 µL) containing 50 mM Tris-HCl (pH 7.5), 10 mM
MgCl2, 1 mM DTT, and 0.05 mg/mL bovine serum albumin. b Standard
deviations are given in parentheses. c No inserted products were detected
after an incubation for 20 min with 1500 µM nucleoside triphosphate and
50 nM enzyme. d The units of this term are % min-1 M-1
.
of each substrate opposite each base in the template was
analyzed with an automated ABI 377 DNA sequencer with the
GeneScan software26 (Table 1).
The incorporation efficiency and selectivity of the v-y pairing
were improved in comparison to those of the s-y pairing. The
incorporation efficiency of dyTP opposite v in the template
(Vmax/KM ) 1.9 × 105) was higher than that opposite s (Vmax
/
KM ) 6.5 × 104) (Table 1, entries 1 and 6). In contrast, the
misincorporation efficiencies of the natural dNTPs opposite v
were mostly similar to those opposite s (Table 1, entries 2-5
and 7-10). However, the efficiency of the T incorporation
opposite v (Vmax/KM ) 6.8 × 103) (Table 1, entry 2) was slightly
decreased, in comparison to that opposite s (Vmax/KM ) 1.1 ×
(27) Hirao, I.; Fujiwara, T.; Kimoto, M.; Yokoyama, S. Bioorg. Med. Chem.
Lett. 2004, 14, 4887.
(26) Kimoto, M.; Yokoyama, S.; Hirao, I. Biotechnol. Lett. 2004, 26, 999.
9
8654 J. AM. CHEM. SOC. VOL. 127, NO. 24, 2005

An Unnatural Base Pair for Specific Transcription
A R T I C L E S
the 32P-labeled transcripts were analyzed on a gel (Figure 2b).
The relative yields of the full-length transcripts (17-mer)
containing one y base from the s template (N ) s, temp35s-1)
were ∼50% (Figure 2b, lanes 5 and 6). In contrast, the relative
yields of the 17-mer transcripts from the v template (N ) v,
temp35v-1) (Figure 2b, lanes 2 and 3) were greatly increased
(∼100%). Interestingly, even though abortive transcripts (12-
mer) were observed for lanes 2 and 3 with the v-y pair, the
relative yields of the full-length transcripts were as high as that
obtained from the natural template (N ) A, temp35A-1) with
the natural NTPs (Figure 2b, lane 7).
Although the transcription of the v template (temp35v-1)
without yTP also yielded the 17-mer transcript by the incorpora-
tion of the natural substrates, mainly CTP, opposite v (Figure
2b, lane 1; Table 2, entry 4), this misincorporation was
completely excluded by the addition of yTP. The faithful y
incorporation opposite v was confirmed by a nucleotide-
composition analysis15,16 of the 17-mer transcripts. For the
nucleotide-composition analysis, transcripts labeled with [R-32P]-
ATP were fully digested to nucleoside 3′-phosphates. Afterward,
the nucleotides were analyzed by 2D-TLC (Figure 2c), and the
amount of each nucleoside 3′-phosphate was quantified (Table
2, entries 1-4). Since [R-32P]ATP was used for labeling in
transcription, the 3′-phosphates of the nucleotides that became
the 5′-neighbor of A in the transcripts were labeled. Thus, the
digestion of the transcripts by RNase T2 generated one labeled
Ap, two labeled Gp’s, and a labeled nucleotide at position 13.
As shown in Figure 2c, among the pyrimidine substrates, only
a spot corresponding to y was observed in the transcription of
the v template (N ) v), and thus no misincorporation of CTP
and UTP opposite v occurred in the presence of yTP. In addition,
in the transcription of the natural template (N ) A) with yTP,
no misincorporation of yTP opposite the natural bases was
observed (Figure 2c, N ) A). The quantification of each spot
on the 2D-TLC also confirmed the high fidelity of the v-y
pairing (Table 2, entries 1 and 2). Each nucleotide composition
agreed well with the theoretical number.
Figure 2. T7 transcription mediated by the v-y pairing. (a) Schemes of
the experiments. (b) Gel electrophoresis of transcripts using the templates
(N ) v, s, or A) with the natural NTPs (1 mM) in the presence (0.5 or 1
mM) or absence of the substrate of y (yTP). Transcripts were internally
labeled with [R-32P]ATP. The relative yields of each transcript were
determined by comparison to the yields of native transcripts from templates
consisting of the natural bases, and each yield was averaged from three
data sets. (c) 2D-TLC analysis of the labeled ribonucleoside 3′-phosphates
obtained from the nuclease digestion of the transcripts (17-mer) (Figure
2b). The spots on the TLC were obtained from the 17-mer fragment
transcribed in the presence of 1 mM yTP.
The transcription of templates containing two adjacent v bases
(temp35v-21) or two v bases separated by one T (temp35v-22)
also yielded 17-mer transcripts containing two y bases (Figure
3a,b, lanes 2 and 3). The relative yields of the transcripts were
71% for those containing two adjacent y bases and 90% for
those containing two y bases separated by one A. In contrast,
the 17-mer transcripts containing two y bases were rarely
observed in the transcription of s templates (Figure 3b, lanes 5
and 6). The high fidelity of this multiple y incorporation was
also confirmed by the nucleoside-composition analysis (Figure
3c and Table 2, entries 5 and 6). In the transcription of temp35v-
22, containing two v bases separated by one T, the transcripts
were internally labeled with [R-32P]ATP or [R-32P]UTP, and
thus, after RNase T2 digestion, each nucleoside 3′-phosphate
of y was detected on 2D-TLC; the y base at position 13 was
labeled with [R-32P]ATP, and the y base at position 15 was
labeled with [R-32P]UTP (Figure 3c). The quantification of these
spots confirmed the site-specific insertion of both y bases (Table
2, entries 5 and 6). The multiple incorporations of y or modified
y bases into RNA will be useful for the creation of RNA
molecules with increased functionality and the expansion of the
genetic code. For example, the multiple labeling of RNA with
fluorescent-yTP would enhance the detection sensitivity, and
of y, and the efficiency of the y incorporation opposite o (Vmax
/
KM ) 2.4 × 105) is slightly higher than that opposite v. This
suggests that the nitrogen of v is still somewhat large to fit well
with the H6 of y, and thus, the efficiency of the v-y pair is
also lower than that of the natural A-T pair. However, the
smaller oxygen of o cannot effectively exclude the misincor-
poration of T (Vmax/KM ) 4.3 × 104).27 Thus, the combination
of the size and the electronegativity of the nitrogen in the
thiazolyl group of v might be suitable for conferring high
selectivity. Conformational analyses of the nucleosides of v and
s are in progress.
Since the v-y pair exhibited higher selectivity and efficiency
in replication than the s-y pair, we next examined the
incorporation of yTP into RNA opposite v in templates by T7
RNA polymerase. Transcription was carried out by using DNA
templates containing v or s, in which the unnatural base was
located at a complementary site corresponding to position 13
in the transcripts (Figure 2a). The sequences of the DNA
templates (temp35N-1, N ) v, s, or A) were optimized to reduce
the addition of one or more nontemplated nucleotides at the
3′-terminus of the nascent transcript.28 After 3 h of transcription,
(28) Nacheva, G. A.; Berzal-Herranz, A. Eur. J. Biochem. 2003, 270, 1458.
9
J. AM. CHEM. SOC. VOL. 127, NO. 24, 2005 8655

A R T I C L E S
Mitsui et al.
Table 2. Nucleotide-Composition Analysis of T7 Transcripts
composition of nucleotides incorporated as 5
′
neighbor of A or Ua
Up
entry
template
[
R-
32P] NTP
yTP (mM)
Ap
Gp
Cp
yp
1
temp35v-1
ATP
ATP
ATP
ATP
ATP
UTP
ATP
ATP
ATP
ATP
ATP
ATP
ATP
ATP
ATP
ATP
yTP
(1)
yTP
(1)
yTP
(0)
yTP
(0)
yTP
(1)
yTP
(1)
I-yTP
(0.25)
I-yTP
(0.5)
I-yTP
(1)
I-yTP
(0.25)
I-yTP
(0.5)
I-yTP
(1)
Ph-yTP
(0.5)
Ph-yTP
(1)
Ph-yTP
(0.5)
Ph-yTP
(1)
1.01b [1]c
(0.01)d
1.01 [1]
(<0.01)
1.00 [1]
(0.02)
1.08 [1]
(0.02)
1.00 [1]
(<0.01)
1.03 [1]
(0.01)
1.02 [1]
(0.03)
1.00 [1]
(0.01)
1.00 [1]
(<0.01)
1.00 [1]
(0.03)
1.01 [1]
(0.02)
1.01 [1]
(<0.01)
1.02 [1]
(0.01)
1.02 [1]
(0.01)
1.00 [1]
(0.03)
1.01 [1]
(0.01)
1.97 [2]
(0.06)
2.00 [2]
(0.02)
1.99 [2]
(0.03)
2.02 [2]
(0.01)
1.97 [2]
(0.01)
n.d. [0]
(-)
1.95 [2]
(0.04)
1.95 [2]
(0.01)
1.95 [2]
(0.01)
2.01 [2]
(0.04)
2.02 [2]
(0.02)
2.01 [2]
(0.01)
1.98 [2]
(0.03)
2.00 [2]
(0.05)
2.00 [2]
(0.03)
n.d.e [0]
(-)
n.d. [0]
(-)
1.02 [1]
(0.07)
2
temp35A-1
temp35A-1
temp35v-1
temp35v-22
temp35v-22
temp35v-1
temp35v-1
temp35v-1
temp35A-1
temp35A-1
temp35A-1
temp35v-1
temp35v-1
temp35A-1
temp35A-1
n.d. [0]
(-)
1.00 [1]
(0.02)
n.d. [0]
(-)
3
n.d. [0]
(-)
0.79 [0]
(0.01)
1.01 [1]
(0.01)
0.11 [0]
(0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.01 [0]
(0.01)
0.01 [0]
(<0.01)
0.01 [0]
(0.01)
0.98 [1]
(0.02)
0.95 [1]
(0.01)
0.93 [1]
(0.01)
0.01 [0]
(0.01)
n.d. [0]
(-)
4
n.d. [-]
(-)
5
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.01 [0]
(0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.03 [0]
(0.01)
1.02 [1]
(0.01)
6
0.95 [1]
(0.01)
1.01 [1]
(0.06)
1.03 [1]
(0.01)
1.02 [1]
(0.01)
0.01 [0]
(<0.01)
0.02 [0]
(<0.01)
0.04 [0]
(<0.01)
0.96 [1]
(0.01)
0.94 [1]
(0.06)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
7
8
9
10
11
12
13
14
15
16
0.03 [0]
(0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.01 [0]
(<0.01)
0.99 [1]
(0.05)
0.97 [1]
(0.01)
2.01 [2]
(0.01)
a Composition of nucleotides incorporated as 5′ neighbor of A (entries 1-5 and 7-16) or U (entry 6), as shown in Figures 2, 3, and 4. b The values were
determined using the following formula: (radioactivity of each nucleotide)/[total radioactivity of all nucleotides (3′-monophosphates)] × (total number of
nucleotides at 5′ neighbor of [R-32P]NTP). c The theoretical number of each nucleotide is shown in brackets. d Standard deviations are shown in parentheses.
e Not detected.
an artificial codon containing two unnatural bases would
increase the specificity of the codon-anticodon interaction,
relative to codons containing one unnatural base.
from the I-y nucleoside. Actually, we succeeded in the syntheses
and the site-specific incorporataion of other modified yTPs
linked with biotin or fluorescent reagents (unpublished data by
Hirao and Yokoyama), which will enable the site-specific
immobilization and fluorescent labeling of RNA molecules.
In this study, we have described the specificity and efficiency
of the unnatural v-y pair for a base-pair-expanded transcription
system, for the site-specific incorporation of y or modified y
bases into RNA. In T7 transcription, the v-y pair exhibits the
highest efficiency among the unnatural base pairs developed
thus far, such as the s-y and isoG-isoC pairs. The DNA
templates containing the v bases can be prepared and amplified
by PCR, using 3′-primers containing the v bases. Since this
transcription system is quite efficient, it can be combined with
other unnatural base pairs, such as isoG and isoC13,14 or 2-amino-
6-(2-thienyl)purine (s) and imidazolin-2-one (z).22 The fluores-
cent s base can be site-specifically incorporated into RNA
opposite z in templates. By using these unnatural base pairs,
various different modified bases derived from y,17,18 isoG,14 and
s could be simultaneously incorporated into a transcript. In
addition, several different modified nucleotides of the natural
bases, such as 2′-modified nucleotides,31,32 modified bases,24
and phosphorothioates,33 can be applied to the specific transcrip-
tion mediated by the unnatural base pairs by a T7 RNA
The modified y substrates, I-yTP17 and Ph-yTP,18 were also
incorporated into RNA opposite v with high efficiency (Figure
4). These transcription efficiencies mediated by the v-y pairing
(Figure 4b, lanes 1 and 4) were more than 2-fold higher than
those mediated by the s-y pairing (Figure 4b, lanes 2 and 5).
The quantification of the nucleotide-composition analysis of the
transcripts (Figure 4c and Table 2, entries 7-12) indicated that
the incorporation efficiency of I-yTP was higher than that of
yTP, and the slight misincorporation of I-yTP opposite A was
observed in the transcripts by using 0.5 or 1 mM I-yTP. This
tendency was shown in the T7 transcription using s as a template
base and I-yTP as a substrate.17 This may happen because, as
shown with 5-substituted uracil derivatives,29,30 the introduction
of the electron-withdrawing iodine into y stabilizes the hydrogen
bonding of the H-N1 position of y with the N1 positions of
both v and A. Thus, for the site-specific incorporation of I-yTP
into RNA, the combination of 0.25 mM I-yTP with 1 mM
natural NTPs was sufficient (Table 2, entry 7). The site-specific
incorporation of Ph-yTP was also confirmed by the nucleotide-
composition analysis (Figure 4c and Table 2, entries 13-16).
One of the advantages of I-y is that various functional substrates
(5-modified yTPs, such as Ph-yTP) can be derived chemically
(31) Jellinek, D.; Green, L. S.; Bell, C.; Lynott, K.; Gill, N.; Vargeese, C.;
Kirshenheuter, G.; McGee, D. P. C.; Abesinghe, P.; Pieken, W. A.; Shapiro,
R.; Rifkin, D. B.; Moscatelli, D.; Janjiæ, N. Biochemistry 1995, 34, 11363.
(32) Davis, K. A.; Lin, Y.; Abrams, B.; Jayasena, S. D. Nucleic Acids Res.
1998, 26, 3915.
(29) Sanghvi, Y. S.; Hoke, G. D.; Freier, S. M.; Zounes, M. C.; Gonzalez, C.;
Cummins, L.; Sasmor, H.; Cook, P. D. Nucleic Acids Res. 1993, 21, 3197.
(30) Kawahara, S.; Wada, T.; Kawauchi, S.; Uchimaru, T.; Sekine, M. Nucleic
Acids Symp. Ser. 1998, 39, 123.
9
8656 J. AM. CHEM. SOC. VOL. 127, NO. 24, 2005

An Unnatural Base Pair for Specific Transcription
A R T I C L E S
Figure 4. Transcription with modified y bases. (a) Schemes of the
experiments. (b) Gel electrophoresis of transcripts containing modified y,
I-y, and Ph-y, employing the v-y or s-y pairing. Transcription was carried
out using 1 mM natural NTPs and 0.25, 0.5, or 1 mM I-yTP or Ph-yTP,
and transcripts were labeled at the 5′-termini with [γ-32P]GTP. (c) 2D-
TLC analysis of the labeled ribonucleoside 3′-phosphates obtained from
the nuclease digestion of the 17-mer RNA fragments transcribed with 0.25
mM I-yTP or 1 mM Ph-yTP.
Figure 3. Transcription of the templates containing two v or s bases. (a)
Schemes of the experiments. (b) Gel electrophoresis of transcripts using
the templates containing two v or s bases with the natural NTPs (1 mM)
and yTP (1 mM). Transcripts were labeled with [γ-32P]GTP. (c) Nucleotide-
composition analysis of the transcript containing two y bases. The transcripts
were internally labeled with [R-32P]ATP or [R-32P]UTP. Each transcript
was digested by RNase T2, and the labeled nucleotides were detected by
2D-TLC.
4.6 mm, Eichrom Technologies). High-resolution mass spectra (HRMS)
and electrospray ionization mass spectra (ESI-MS) were recorded on a
JEOL HX-110 or JM 700 mass spectrometer and a Waters micromass
ZMD 4000 equipped with a Waters 2690 LC system, respectively.
Fluorescence measurements were made on a FP-6500DS spectrofluo-
rometer (JASCO).
2-Tributylstannylthiazole.20,21 To a solution of 2-bromothiazole (901
µL, 10 mmol) in ether (50 mL) was added n-butyllithium (1.58 M in
hexane solution, 6.3 mL, 10 mmol) at -78 °C. The solution was stirred
at -78 °C for 30 min. To the solution was added tributyltin chloride
(2.7 mL, 10 mmol), and the reaction mixture was stirred at room
temperature for 30 min. The mixture was poured into a saturated NaCl
solution. The organic phase was washed twice with the saturated NaCl,
dried with MgSO4, and evaporated in vacuo. The resulting brown
solution (2-tributylstannylthiazole, 10 mmol) was used for the Pd-
mediated coupling reaction without further purification.
polymerase variant.34 These unnatural base and base-pair
systems could overcome the serious limitation of the functional-
ity of RNAs, as compared to that of proteins, and will facilitate
advances in the creation of novel RNA molecules as diagnostic
and therapeutic agents.
Experimental Section
General. Reagents and solvents were purchased from standard
suppliers without further purification. Reactions were monitored by thin-
layer chromatography (TLC) using 0.25-mm silica gel 60 plates
impregnated with 254-nm fluorescent indicator (Merck). 1H NMR (270
MHz), 13C NMR (68 MHz), and 31P NMR (109 MHz) spectra were
recorded on a JEOL EX270 magnetic resonance spectrometer. Purifica-
tion of nucleosides was performed on a Gilson HPLC system with a
preparative C18 column (Waters Microbond Sphere, 150 × 19 mm).
The triphosphate derivatives were purified with a DEAE-Sephadex A-25
column (300 × 15 mm) and a C18 column (Synchropak RPP, 250 ×
2-Amino-6-(2-thiazolyl)-9-[2-deoxy-3,5-di-O-(tert-butyldimethyl-
silyl)-â-D-ribofuranosyl]purine (2). A mixture of 2-amino-6-O-(p-
toluenesulfonyl)-9-[2-deoxy-3,5-di-O-(tert-butyldimethylsilyl)-â-D-
ribofuranosyl]purine (1)35 (1.3 g, 2.0 mmol), Pd(PPh3)4 (116 mg, 0.1
mmol), 2-tributylstannylthiazole (3.7 g, 10 mmol), and LiCl (170 mg,
4.0 mmol) in dioxane (20 mL) was refluxed for 3.5 h. The reaction
(33) Jhaveri, S.; Olwin, B.; Ellington, A. D. Bioorg. Med. Chem. Lett. 1998, 8,
2285.
(34) Chelliserrykattil, J.; Ellington, A. D. Nat. Biotechnol. 2004, 22, 1155.
9
J. AM. CHEM. SOC. VOL. 127, NO. 24, 2005 8657

A R T I C L E S
Mitsui et al.
mixture was evaporated in vacuo, and the product was purified by silica
gel column chromatography (1% methanol in CH2Cl2) to give 2 (790
mg, 70%).
The amount of enzyme used (3-66 nM), the reaction time (1-20 min),
and the gradient concentration of dNTP (0.6-3000 µM) were adjusted
to give reaction extents of 25% or less. Reactions were quenched by
adding 10 µL of a stop solution containing 95% formamide and 20
mM EDTA, and the mixtures were immediately heated at 75 °C for 3
min. The diluted products were analyzed on an automated ABI 377
DNA sequencer equipped with the GeneScan software (version 3.0).
Relative velocities (V0) were calculated as the extents of the reaction
divided by the reaction time and were normalized to the enzyme
concentration (20 nM) for the various enzyme concentrations used. The
kinetic parameters (KM and Vmax) were obtained from Hanes-Woolf
plots of [dNTP]/V0 against [dNTP]. Each parameter was averaged from
three to six data sets.
2-Amino-6-(2-thiazolyl)-9-(2-deoxy-â-D-ribofuranosyl)purine (3).
A solution of 2 (790 mg, 1.4 mmol) in tetrahydrofuran (THF) (10 mL)
was treated with a solution of 1 M tetrabutylammonium fluoride
(TBAF) in THF (4.2 mL, 4.2 mmol) at room temperature for 30 min.
The solvent was evaporated in vacuo, and the product was purified by
silica gel column chromatography (5% methanol in CH2Cl2) and then
purified by RP-HPLC (10-50% CH3CN in H2O, 15 min) to give 3
(372 mg, 80%) as a yellow solid.
2-N-Phenoxyacetyl-6-(2-thiazolyl)-9-(2-deoxy-â-D-ribofuranosyl)-
purine (4). A solution of 3 (167 mg, 0.5 mmol) and trimethylsilyl
chloride (460 µL, 3.7 mmol) in pyridine (2.5 mL) was stirred at room
temperature for 25 min (solution A). A solution of phenoxyacetyl
chloride (104 µL, 0.8 mmol) and 1-hydroxybenzotriazole (120 mg, 0.9
mmol) in CH3CN (240 µL) and pyridine (240 µL) was stirred at 0 °C
for 5 min (solution B). Solution A, precooled to 0 °C, was added to
solution B at 0 °C, and then the reaction mixture was stirred at room
temperature overnight. The solution was cooled to 0 °C and treated
with 14% NH4OH (440 µL) for 10 min. The reaction mixture was
separated with EtOAc and H2O, and the organic phase was dried with
Na2SO4 and evaporated in vacuo. The product was purified by silica
gel column chromatography (5% methanol in CH2Cl2) to give 4 (211
mg, 90%) as a white solid.
2-N-Phenoxyacetyl-6-(2-thiazolyl)-9-[2-deoxy-5-O-(4,4′-dimethoxy-
trityl)-â-D-ribofuranosyl]purine (5). Compound 4 (211 mg, 0.45
mmol) was coevaporated with dry pyridine three times and was
dissolved in pyridine (4.3 mL). To the solution was added 4,4′-
dimethoxytrityl chloride (159 mg, 0.47 mmol), and the mixture was
stirred at room temperature for 3 h. The reaction mixture was poured
into 5% NaHCO3 in H2O and extracted with EtOAc. The organic phase
was washed with saturated NaCl three times, dried with Na2SO4, and
evaporated in vacuo. The product was purified by silica gel column
chromatography (CH2Cl2/EtOAc ) 1:1, v/v) to give 5 (346 mg, 99%).
2-N-Phenoxyacetyl-6-(2-thiazolyl)-9-[2-deoxy-5-O-(4,4′-dimethoxy-
trityl)-â-D-ribofuranosyl]purine 2-cyanoethyl-N,N-diisopropylphos-
phoramidite (6). Compound 5 (346 mg, 0.45 mmol) was coevaporated
with pyridine and THF three times each and was dissolved in THF
(2.3 mL). To the solution was added N,N-diisopropylethylamine (DIEA)
(117 µL, 0.67 mmol) and 2-cyanoethyl-N,N-diisopropylamino chloro-
phosphoramidite (110 µL, 0.49 mmol). The mixture was stirred at room
temperature for 1.5 h. The reaction was quenched by adding 50 µL of
methanol. The solution was diluted with EtOAc/triethylamine (20 mL,
20:1, v/v) and then washed with 5% NaHCO3 and saturated NaCl three
times. The organic phase was dried with Na2SO4 and evaporated in
vacuo. The product was purified by silica gel column chromatography
(hexane/CH2Cl2 ) 3:2, v/v containing 2% triethylamine) to give 6 (360
mg, 83%) as a white foam.
T7 Transcription. Templates (10 µM of a 35-mer coding DNA and
a 21-mer noncoding DNA) were annealed in a buffer containing 10
mM Tris-HCl (pH 7.6) and 10 mM NaCl by heating at 95 °C and slow
cooling to 4 °C. Transcription was carried out in a buffer (20 µL)
containing 40 mM Tris-HCl (pH 8.0), 24 mM MgCl2, 2 mM spermidine,
5 mM DTT, 0.01% Triton X-100, 1 mM natural NTPs, 0, 0.5 or 1 mM
modified or unmodified yTP, 2 µCi [γ-32P]GTP, 2 µM template, and
50 units of T7 RNA polymerase (Takara, Kyoto). By the use of [γ-32P]-
GTP, the transcripts were labeled only at the 5′-end, which facilitated
the analyses of the yields. After an incubation at 37 °C for 3 h, the
reaction was quenched by the addition of a dye solution (20 µL)
containing 10 M urea and 0.05% BPB. The mixture was heated at 75
°C for 3 min, and the products were analyzed on a 20% polyacrylamide-
7M urea gel.
Nucleotide-Composition Analysis in T7 Transcription.15,16 Tem-
plates (10 µM) were annealed in 10 mM Tris-HCl buffer (pH 7.6)
containing 10 mM NaCl by heating at 95 °C for 3 min and cooling to
4 °C. Transcription was carried out in 40 mM Tris-HCl buffer (pH
8.0, 20 µL) containing 24 mM MgCl2, 2 mM spermidine, 5 mM DTT,
0.01% Triton X-100, 10 mM GMP, 1 mM natural NTPs, 0-1 mM
yTP or modified yTP, 2 µCi [R-32P]ATP or [R-32P]UTP (Amersham),
2 µM template, and 50 units of T7 RNA polymerase (Takara). After
an incubation for 3 h at 37 °C, the reaction was quenched by the addition
of 20 µL of the dye solution. This mixture was heated at 75 °C for 3
min and then was loaded onto a 15% polyacrylamide-7 M urea gel.
The full-length products were eluted from the gel with water and were
precipitated with ethanol and 0.05 A260 units of E. coli tRNA. The
transcripts were digested by 0.75 units of RNase T2 at 37 °C for 70-
120 min, in 10 µL of 15 mM sodium acetate buffer (pH 4.5). The
digestion products were analyzed by 2D-TLC using a Merck HPTLC
plate (100 × 100 mm) (Merck, Darmstadt, Germany) with the following
developing solvents: isobutyric acid/NH4OH/H2O (66:1:33 v/v/v) for
the first dimension, and isopropyl alcohol/HCl/H2O (70:15:15 v/v/v)
for the second dimension. The products on the gels and the TLC plates
were analyzed by the Bio-imaging analyzer. The quantification of each
spot was averaged from three to five data sets.
Steady-State Kinetics for the Single-Nucleotide Insertion Experi-
ments with the Klenow Fragment. Steady-state kinetic analyses for
single-nucleotide insertions were performed according to the literature.26
A primer (primer 1, 20-mer) labeled with 6-carboxyfluorescein at the
5′-end was annealed with templates (template 1 or template 2, 35-mer)
in a buffer containing 100 mM Tris-HCl (pH 7.5), 20 mM MgCl2, 2
mM DTT, and 0.1 mg/mL bovine serum albumin, by heating at 95 °C
and slow cooling to 4 °C. The primer-template duplex solution (10
µM, 5 µL) was mixed with 2 µL of a solution containing the
exonuclease-deficient Klenow fragment (Amersham USB, Cleveland,
OH), which was diluted in a buffer containing 50 mM potassium
phosphate (pH 7.0), 1 mM DTT, and 50% glycerol. The mixture was
incubated for more than 2 min. Reactions were initiated by adding 3
µL of each dNTP solution to the duplex-enzyme mixture at 37 °C.
Acknowledgment. This work was supported by the RIKEN
Structural Genomics/Proteomics Initiative (RSGI), the National
Project on Protein Structural and Functional Analyses, Ministry
of Education, Culture, Sports, Science and Technology of Japan,
and a Grant-in-Aid for Scientific Research (KAKENHI 15350097)
from the Ministry of Education, Culture, Sports, Science, and
Technology. We are grateful to Dr. Makoto Komiyama, The
University of Tokyo, for his support and helpful discussions.
Supporting Information Available: NMR and MS data for
the nucleoside derivatives of v and MS data for the d(TvT)
trimer. This material is available free of charge via the Internet
at http://pubs.acs.org.
(35) Nagatsugi, F.; Uemura, K.; Nakashima, S.; Maeda, M.; Sasaki, S.
Tetrahedron 1997, 53, 3035.
JA0425280
9
8658 J. AM. CHEM. SOC. VOL. 127, NO. 24, 2005