Efficient access to nonhydrolyzable initiator tRNA based on the synthesis of 3'-azido-3'-deoxyadenosine RNA

Article abstract of DOI:10.1002/anie.201003424

Flexibility exercised: Hydrolysis-resistant 3′-aminoacyl-tRNA conjugates that contain a stable amide linkage instead of the natural ester are valuable substrates for biochemical studies of ribosomal processes. In a novel preparation of the stable E. coli

Full text of DOI:10.1002/anie.201003424

Communications
DOI: 10.1002/anie.201003424
Modified RNA Derivatives
Efficient Access to Nonhydrolyzable Initiator tRNA Based on the
Synthesis of 3’-Azido-3’-Deoxyadenosine RNA**
Jessica Steger, Dagmar Graber, Holger Moroder, Anna-Skrollan Geiermann, Michaela Aigner,
and Ronald Micura*
Dedicated to Professor Karl Grubmayr on the occasion of his 60th birthday
3’-Aminoacyl-tRNA conjugates with a hydrolysis-resistant
amide linkage instead of the natural ester represent valuable
substrates for biochemical studies of ribosomal processes.
Access to such conjugates is currently a serious bottleneck for
the investigation and functional characterization of pre- and
post-peptidyl-transfer states,[1] of the tRNA hybrid states,[2] of
translation initiation[2] and termination,[3] as well as of
phenomena like ribosome stalling.[4] For the preparation of
3’-aminoacyl-tRNA with a stable amide linkage, an approach
was originally developed in the 1970s that involved enzymatic
degradation of the tRNA ..CCA76 3’-terminus to yield tRNA
..CC75 intermediates for the enzymatic attachment of 3’-
amino-3’-deoxyadenosine using tRNA nucleotidyl transfer-
ase.[5] The resulting tRNA provided a reactive 3’-NH2 group,
which was charged enzymatically with an amino acid by using
Scheme 1. 3’-Modified E. coli tRNAfMet target structures 1– 4 in this
study. The aim was a flexible synthetic strategy to allow efficient access
to 3’-N3-, 3’-NH2-, and 3’-(N-formylmethionyl)amino-derivatized tRNA
with and without natural nucleoside modifications.
the cognate tRNA synthetase. This method has been applied
occasionally;[1b,5c,d] however, it is rather inefficient. An early
report on the acylation of 3’-NH2-modified tRNA by N-
protected amino acid hydroxysuccinimide esters or anhy-
drides had only faint resonance, most likely because of the
phosphoramidites and indeed found that this matrix
was applicable to standard RNA solid-phase synthesis
using nucleoside phosphoramidites as building blocks
(Scheme 2B).
poor selectivity and incomplete coupling.[6]
Here, we introduce a novel combined chemical and
enzymatic concept for the preparation of the E. coli initiator
tRNA derivatives 3’-(N-formylmethionyl)amino-tRNAfMet
3
The 3’-N3-RNA was quantitatively reduced by overnight
incubation with tris(2-carboxyethyl)phosphine hydrochloride
(TCEP) (Figure 1B) and subsequently charged with fMet
using the activated amino acid Pfp ester. Amide bond
formation proceeded within 15 to 45 min and in 77–94%
yield depending on the length of the RNA (Figure 1C,
Table 1; see the Supporting Information). To show that there
is no unspecific reaction of the Pfp ester with the nucleobases,
control experiments were conducted using RNA oligonucleo-
tides with the genuine 3’-OH; no adducts were obtained
under the conditions used. Also, when free fMet was
incubated with 3’-NH2-RNA, no unspecific aminal or imino
adducts with the formyl group were observed. We further
mention that N-(9-fluorenyl)methoxycarbonyl (Fmoc) amino
acid Pfp esters reacted significantly slower and only in poor
and 4. En route, our flexible approach allows access to the
respective 3’-N3- and 3’-NH2-functionalized tRNAs 1 and 2
(Scheme 1).
The starting point for our undertaking was the 3’-azido-3’-
deoxyadenosine derivative 5, which is readily available
according to a previously introduced synthesis.[7] This com-
pound was transformed into the pentafluorophenyl (Pfp)
adipic acid ester 6 and finally into the functionalized solid
support 7 (Scheme 2A). We suspected the sterically hindered
azide group of 7 to have limited reactivity in reactions with
[*] J. Steger, D. Graber, Dr. H. Moroder, A.-S. Geiermann, M. Aigner,
Prof. Dr. R. Micura
Institute of Organic Chemistry
Center for Molecular Biosciences CMBI
University of Innsbruck, 6020 Innsbruck (Austria)
E-mail: ronald.micura@uibk.ac.at
Table 1: Selection of synthesized 3’-azido-3’-deoxyadenosine RNA.
[**] Funding by the Austrian Science Foundation FWF (P21640, I317)
and the Ministry of Science and Research bm:wf (GenAU III, “Non-
coding RNA” P0726-012-012) is acknowledged. M.A. thanks the
Austrian Academy of Science and the UNESCO for a L’ORꢀAL
fellowship.
Sequence
Amount
[nmol]
Mr(calcd)
[amu]
Mr(obsd)
[amu]
5’-ACCA-3’-N3
128
264
225
370
1231.8
5673.5
5753.5
5424.3
1231.6
5673.4
5752.9
5424.0
5’-AUC2G2C5GCA2C2A-3’-N3
5’-p-AUC2G2C5GCA2C2A-3’-N3
5’-p-UC2G2C5GCA2C2A-3’-N3
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201003424.
7470
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7470 –7472

Angewandte
Chemie
enzymatic ligation using T4 RNA ligase. We synthesized the
5’-phosphorylated 17 nucleotide(nt) fMet-RNA conjugate 8
and the respective 60nt 5’-tRNA fragment 9 (Figure 2A).
These RNAs aligned properly into a sufficiently stable pre-
ligation complex so that the 5’-phosphate of the donor 8 came
into close vicinity of the 3’-OH of the acceptor 9 to allow
efficient ligation (77% yield, Figure 2B). The full-length
tRNA-fMet conjugate was isolated by anion-exchange chro-
matography (1.9 nmol of purified 3), and the correct mass was
confirmed by LC-ESI mass spectrometry (Figure 2C). Like-
wise, when we applied the 5’-terminal tRNAfMet fragment 10
carrying all genuine nucleoside modifications, a satisfying
ligation yield of almost 70% was achieved (0.6 nmol of
purified 4, Figure 2D–F). Fragment 10 was readily obtained
by cleavage of tRNAfMet in the TYC loop using a 10–23 DNA
enzyme and subsequent dephosphorylation (see the Support-
ing Information). We mention that this particular generation
of natural 5’-tRNA fragments with all nucleoside modifica-
tions is applicable also to other tRNA species, as demon-
strated very recently in the context of nonhydrolyzable 3’-
peptidyl-tRNAs.[8]
In this study, we have demonstrated a novel approach for
the efficient access to hydrolysis-resistant fMet-tRNAfMet with
and without the natural modification pattern. Moreover, we
stress that the 3’-N3- and 3’-NH2-modified E. coli tRNAfMet
variants, 1 and 2, respectively, were prepared in equally
efficient manner (see the Supporting Information); thus, this
approach is highly flexible, also for other types of tRNA, and
from different organisms. Many potential applications are
conceivable, since the 3’-amino group can be charged with
other amino acids including nonnatural ones, by using either
an appropriate chemical activation or potentially, also the
flexizyme methodology.[9] Another promising aspect is the use
of 3’-azido-modified tRNA for cellular studies that focus on
the action of tRNA modification enzymes.[10] Since the 3’-
azido group is bioorthogonal and generally does not affect
cellular functions, direct isolation and/or labeling of these
metabolized tRNA derivatives from cell lysates by means of
one of the modern bioconjugation strategies, such as the
Staudinger ligation or click chemistry,[11] are within reach.
Lastly, we mention that these studies have encouraged us to
envisage and realize the synthesis of RNAwith site-specific 2’-
N3 groups as potential siRNA reagents, on which we will
report in near future.
Scheme 2. Synthesis of the 3’-azido-3’-deoxyadenosine-derivatized
solid support 7 and its use in RNA solid-phase synthesis. A) Reaction
conditions: a) 4.7 equiv PfpOOC(CH2)4COOPfp, 1 equiv DMAP, in
DMF/pyridine (1:1), room temperature, 1 h, 74%; b) 3 equiv (w/w)
amino-functionalized support (GE Healthcare, Custom Primer Support
200 Amino), 2 equiv pyridine, in DMF, room temperature, 22 h,
loading: 76 mmolgÀ1. B) Reaction conditions: c) standard RNA solid-
phase synthesis and deprotection; d) cRNA =20 mm, 0.5 mm TCEP,
100 mm Tris·HCl, pH 8, 1 d, À208C, 95–98%; e) cRNA =100 mm, 25 mm
fMet-OPfp, 100 mm Tris·HCl, pH 8, DMSO/H2O (1:1), 378C, 15–
45 min, 77–94%. DMAP=4-(dimethylamino)pyridine, DMF=N,N-
dimethylformamide, DMSO=dimethyl sulfoxide. For details see the
Supporting Information.
Figure 1. Characterization of 3’-modified 18nt oligoribonucleotides;
anion-exchange HPLC traces and LC-ESI mass spectra. A) RNA-3’-N3.
B) RNA-3’-NH2. C) RNA-3’-NH-fMet. For conditions see the Supporting
Information.
Experimental Section
RNA solid-phase synthesis on the azido-modified support 7: All
oligonucleotides were synthesized on a Pharmacia Gene Assembler
Special or Pharmacia Gene Assembler Plus following standard
synthesis protocols. Detritylation (2.0 min): dichloroacetic acid/1,2-
dichloroethane (4:96); coupling (3.0 min): phosphoramidites/aceto-
nitrile (0.1m; 120 mL per coupling) were activated by benzylthiote-
trazole/acetonitrile (0.3m; 360 mL per coupling); capping (3 ꢀ
0.4 min): A: Ac2O/sym-collidine/acetonitrile (2:3:5), B: 4-(dimethyl-
amino)pyridine/acetonitrile (0.5m), A/B = 1:1; oxidation (1.0 min): I2
(10 mm) in acetonitrile/sym-collidine/H2O (10:1:5). Solutions of
amidites, tetrazole solutions, and acetonitrile were dried over
activated molecular sieves (4 ꢁ) overnight. All sequences were
synthesized trityl-off.
yields with 3’-NH2-RNA. We also tested amino acids that
were activated as thioesters and observed slow and incom-
plete coupling at best (see the Supporting Information). We
considered that the electron-withdrawing properties of the N-
formyl group (ÀI effect) can be the reason for the advanced
performance of the fMet-OPfp ester.
With this facile access to 3’-N3-, 3’-NH2-, and 3’-NH-fMet-
modified RNA in our hands, we moved on to the synthesis of
E. coli tRNAfMet targets which we intended to achieve by
Angew. Chem. Int. Ed. 2010, 49, 7470 –7472
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7471

Communications
For experimental procedures for fMet loading onto 3’-amino-
3’-deoxyoligonucleotides and enzymatic ligation to obtain the
corresponding tRNA derivatives 1–4 see the Supporting Informa-
tion.
Received: June 5, 2010
Published online: August 31, 2010
Keywords: azides · phosphoramidites · RNA ·
.
solid-phase synthesis · tRNA
[1] a) D. N. Wilson, K. H. Nierhaus, Angew. Chem. 2003, 115,
3586 – 3610; Angew. Chem. Int. Ed. 2003, 42, 3464 – 3486;
b) R. M. Voorhees, A. Weixlbaumer, D. Loakes, A. C. Kelley,
V. Ramakrishnan, Nat. Struct. Mol. Biol. 2009, 16, 528 – 533;
c) K. Lang, M. Erlacher, D. N. Wilson, R. Micura, N. Polacek,
Chem. Biol. 2008, 15, 485 – 492; d) P. Khade, S. Joseph, FEBS
Lett. 2010, 584, 420 – 426; e) N. Clementi, A. Chirkova, B.
Puffer, R. Micura, N. Polacek, Nat. Chem. Biol. 2010, 6, 344 –
351.
[2] a) W. Zhang, J. A. Dunkle, J. H. D. Cate, Science 2009, 325,
1014 – 1017; b) J. B. Munro, R. B. Altman, N. OꢂConnor, S. C.
Blanchard, Mol. Cell 2007, 25, 505 – 517; c) S. E. Walker, S.
Shoji, D. Pan, B. S. Cooperman, K. Fredrick, Proc. Natl.
Acad. Sci. USA 2008, 105, 9192 – 9197.
[3] A. Weixlbaumer, H. Jin, C. Neubauer, R. M. Voorhees, S.
Petry, A. C. Kelley, V. Ramakrishnan, Science 2008, 322,
953 – 956.
Figure 2. Enzymatic ligation of donor conjugate 8 using T4 RNA ligase to
prepare tRNAfMet derivatives 3 and 4. Acceptor: Synthetic 5’-tRNA fragment
9 (A–C) or natural counterpart 10 containing genuine nucleoside modifi-
cations (D–F). HPLC profiles and LC-ESI mass spectra. For conditions see
the Supporting Information.
[4] S. Bhushan, M. Gartmann, M. Halic, J.-P. Armache, A.
Jarasch, T. Mielke, O. Berninghausen, D. N. Wilson, R.
Beckmann, Nat. Struct. Mol. Biol. 2010, 17, 313 – 317.
[5] a) T. H. Fraser, A. Rich, Proc. Natl. Acad. Sci. USA 1973, 70,
2671 – 2675; b) M. Sprinzl, H. G. Faulhammer, Nucleic Acids
Res. 1978, 5, 4837 – 4840; c) V. Ramesh, C. Mayer, M. R.
Dyson, S. Gite, U. L. RajBhandary, Proc. Natl. Acad. Sci. USA
1999, 96, 875 – 880; d) C. Mayer, U. L. RajBhandary, Nucleic
Acids Res. 2002, 30, 2844 – 2850.
Deprotection of RNA strands synthesized on the azido-modified
support 7: The beads were transferred into an Eppendorf tube, and
equal volumes of CH3NH2 in EtOH (8m, 0.65 mL) and CH3NH2 in
H2O (40%, 0.65 mL) were added. The mixture was kept at room
temperature for 8 h. After the supernatant was filtered and evapo-
rated to dryness, the 2’-O-silyl ethers were removed by treatment with
a 1.0m solution of tetrabutylammonium fluoride (TBAF)·3H2O in
THF (1.0 mL) for 16 h at 378C. The reaction was quenched by the
addition of triethylammonium acetate buffer (1.0m, pH 7.3, 1.0 mL).
The volume of the solution was reduced to 0.5 mL and directly
applied on a HiPrep 26/10 desalting column (GE Healthcare). The
crude oligonucleotide was eluted with H2O and subsequently
evaporated to dryness. For analysis and purification of the 3’-azido-
modified RNA see the Supporting Information.
Reduction of 3’-azido-3’-deoxyoligonucleotides: One equivalent
of 3’-azido-3’-deoxyoligonucleotide (final concentration = 20 mm) and
25 equivalents of TCEP (tris(2-carboxyethyl)phosphine hydrochlo-
ride, final concentration = 0.5 mm) were dissolved in 100 mm Tris·HCl
(pH 8.0). For crude 3’-azido-3’-deoxyoligonucleotides (not purified
before reduction) the yield of the oligonucleotide (1 mmol scale) was
estimated to be 400 nmol. After 24 h at À208C, the reaction solution
was desalted on a C18 SepPak Plus cartridge (Waters). The reduction,
purification, and analysis of 3’-amino-3’-deoxyoligonucleotides were
monitored by anion-exchange chromatography and LC-ESI mass
spectrometry (for details see the Supporting Information).
[6] a) J. Shiloach, Y. Lapidot, N. de Groot, M. Sprinzl, F. Cramer,
FEBS Lett. 1975, 57, 130 – 133; b) Y. Lapidot, N. de Groot, Prog.
Nucleic Acid Res. Mol. Biol. 1972, 12, 189 – 228.
[7] H. Moroder, J. Steger, D. Graber, K. Fauster, K. Trappl, V.
Marquez, N. Polacek, D. N. Wilson, R. Micura, Angew. Chem.
2009, 121, 4116 – 4120; Angew. Chem. Int. Ed. 2009, 48, 4056 –
4060.
[8] D. Graber, H. Moroder, J. Steger, K. Trappl, N. Polacek, R.
Micura, Nucl. Acids Res. 2010, DOI: 10.1093/nar/gkq508.
[9] a) M. Ohuchi, H. Murakami, H. Suga, Curr. Opin. Chem. Biol.
2007, 11, 537 – 542; b) H. Xiao, H. Murakami, H. Suga, A. R.
Ferrꢃ DꢂAmarꢃ, Nature 2008, 454, 358 – 362.
[10] H. Grosjean, DNA and RNA Modification Enzymes, Landes
Biosciences, Austin, TX, 2009.
[11] a) E. M. Sletten, C. R. Bertozzi, Angew. Chem. 2009, 121, 7108 –
7133; Angew. Chem. Int. Ed. 2009, 48, 6974 – 6998; b) P. M. E.
Gramlich, C. T. Wirges, A. Manetto, T. Carell, Angew. Chem.
2008, 120, 8478 – 8487; Angew. Chem. Int. Ed. 2008, 47, 8350 –
8358.
7472
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7470 –7472