Triggering of RNA secondary structures by a functionalized nucleobase

Article abstract of DOI:10.1002/anie.200460068

Rearranging RNA: Defined RNA secondary-structure rearrangements can be controlled by chemically functionalized nucleobases. The proof of principle for this previously unrecognized concept for manipulating secondary structures is based on a novel type of f

Full text of DOI:10.1002/anie.200460068

Communications
RNA Structural Rearrangements
Triggering of RNA Secondary Structures by a
Functionalized Nucleobase**
Claudia Höbartner, Harald Mittendorfer,
Kathrin Breuker, and Ronald Micura*
The ability of nucleic acids to undergo conformational
changes is a prerequisite for their immense functional
repertoire. The definition of nucleic acid conformational
changes is rather broad and ranges from local structural
transformations, such as base flipping, to complex reorgan-
izations of the secondary and tertiary structure.[1] The most
prominent examples of the latter are metabolite-binding
riboswitches, which have been recognized as a general means
for gene regulation in many organisms.[2] Other examples
concern ligand-responsive aptamer–ribozyme constructs with
the ribozyme activity directed by allosteric effects upon
binding of the ligand to a receptor domain.[3] The modular
nature of these molecular sensors makes them highly flexible
with respect to the rational design of diagnostic tools, such as
the recently introduced target-assisted signal-amplifying
ribozymes for the detection of single nucleotide polymor-
phism (SNP) or of microRNAs.[4,5]
A crucial feature of the above examples is that an effector
molecule, either a small ligand or another oligonucleotide,
induces the nucleic acid conformational change. As for
natural riboswitches, the conformational changes in the
designed molecules also occur at the secondary-structure
level and lead, for example, to the formation of defined,
mutually dependent stem–loop motifs (terminator–antitermi-
nator) that are responsible for an on/off signal for tran-
scription or translation.[6,7]
With the present work we introduce a novel concept for
the manipulation of nucleic acid secondary structures. The
idea relies on chemically functionalized nucleotides that play
the role of an effector and that have potential as a previously
unrecognized tool for designable secondary-structure modu-
lation.
The underlying basic concept is depicted in Figure 1. A
natural nucleoside is modified by a functional group at the
Watson–Crick base-pairing site. When incorporated into an
oligoribonucleotide this functionalized nucleoside is respon-
sible for the selection of a distinct, defined secondary
structure due to its deficiency of base pairing. However,
[*] Dipl.-Ing. C. Höbartner, Mag. H. Mittendorfer, Dr. K. Breuker,
Prof. Dr. R. Micura
Leopold Franzens Universität
Institut für Organische Chemie
Innrain 52a, 6020 Innsbruck (Austria)
Fax: (+43)512-507-2892
E-mail: ronald.micura@uibk.ac.at
[**] Financial support from the Austrian Science Fund (Grants P15042
to R.M. and P15767 to K.B.) is gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3922
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200460068
Angew. Chem. Int. Ed. 2004, 43, 3922 –3925

Angewandte
Chemie
To prove this concept, we considered O6-(2,2,2-trichloro-
ethyl)-modified guanosine (tceG) to be appropriately func-
tionalized. On the one hand, the O6-tce group alters the
donor–acceptor pattern of the guanosine Watson–Crick base-
pairing site and is therefore expected to significantly lower
the residueꢀs base-pairing probability. On the other hand,
release of this group should be easily achieved in aqueous
solution by zinc, thereby restoring the guanosine as a
potential base-pairing candidate that then represents the
driving force for a secondary-structure rearrangement.
The synthesis of the O6-tce-guanosine phosphoramidite 6
used in our studies is shown in Scheme 1. The 2’-O-TOM-
guanosine derivative 1[8] was treated with phenoxyacetyl
Figure 1. Novel concept for the modulation of nucleic acid secondary
structures. A nucleobase, for example, guanine, is chemically function-
alized at the Watson–Crick base-pairing site. Upon release of the func-
tional group (FG) the base-pairing properties of the nucleobase are
restored; the fold then becomes metastable and reorganizes into a
thermodynamically favorable one.
Figure 2. a) Representative example for the deprotection and purifica-
tion of an O6-tce-containing RNA sequence. HPLC trace of the product
8, after treatment with aqueous NH3/ethanol (1:1) at 558C for 5 h,
treatment with 0.9m TBAF in THF/H2O (9:1) at 258C for 14 h, and
purification by anion-exchange chromatography; MALDI-TOF MS: m/z
calcd for [M+H]+: 6594.3; found: 6597.5ꢀ5. The inset shows the
crude product before purification. b) Representative example for the
release of the functional group, the O6-tce group from 8. HPLC trace
of the crude product 9, after treatment with zincdust and 1 m
NH4OAcat pH 7 and 25 8C; t1/2 =5 min; MALDI-TOF MS: m/z calcd
for [M+H]+: 6462.9; found: 6466.3ꢀ5. Anion-exchange HPLC: Dionex
DNAPaccolumn (4”250 mm), 80 8C, 1 mLminꢁ1, 0!60% B in
45 min (A: 25 mm tris(hydroxymethyl)aminomethane/HCl, 6m urea,
pH 8.0; B: same as A with addition of 0.5m NaClO4).
Scheme 1. Synthesis of O6-(2,2,2-trichloroethyl)guanosine phosphor-
amidite 6. a) Phenoxyacetyl chloride (5.5 equiv), pyridine/CH2Cl2, 08C,
2 h, 74%; b) mesitylenesulfonyl chloride (1.2 equiv), triethylamine
(5.0 equiv), DMAP (0.1 equiv), CH2Cl2, RT, 1 h; c) 2,2,2-trichloroethanol
(2.5 equiv), ethyldimethylamine (5.0 equiv), DBU (1.0 equiv), CH2Cl2,
RT, 2 h, 58% over two steps; d) 0.05m NaOH, THF/methanol/H2O,
48C, 3 min, 97%; e) 2-cyanoethyl-N,N-diisopropylchlorophosphorami-
dite (1.5 equiv), ethyldimethylamine (10 equiv), CH2Cl2, RT, 2 h, 90%.
DBU=1,8-diazabicyclo[5.4.0]undec-7-ene, DMAP=4-dimethylamino-
pyridine, DMT=4,4’-dimethoxytrityl, Pac=phenoxyacetyl, THF=tetra-
hydrofuran, TOM=triisopropylsilyloxymethyl.[9]
chloride in pyridine to give compound 2. Then, mesitylation
of the O6 atom of the lactam moiety and substitution of the
corresponding mesitylenesulfonyl ester 3 with 2,2,2-trichloro-
ethanol in the presence of DBU and ethyldimethylamine
yielded 4. Selective cleavage of the 3’-O-phenoxyacetyl group
with aqueous NaOH in THF/methanol resulted in compound
5, which was readily transformed into the O6-tce-guanosine
cyanoethyl phosphoramidite 6.[9]
upon release of the functional moiety the capability for
Watson–Crick base pairing is restored and the secondary
structure becomes metastable and reorganizes into a thermo-
dynamically favorable one.
Angew. Chem. Int. Ed. 2004, 43, 3922 –3925
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3923

Communications
The incorporation of phosphoramidite 6 into oligoribo-
secondary structures. The RNA 20-mer 5’-rGAAGG-
(GCAA)CC(UUCG)GGUUG (9) is a rationally designed
sequence that exists in a monomolecular equilibrium of two
equally populated secondary structures.[15,16] However, upon
replacement of a single guanosine by O6-tce-guanosine in
strand positions 4 or 17, respectively, only one of the two folds
is selected. The functionalized 20-mer 5’-rGAAtceGGG-
CAACC(UUCG)GGUUG (8) exclusively adopts a UUCG
hairpin conformation, whereas the 20-mer 5’-rGAAGG-
(GCAA)CCUUCGGtceGUUG (7) solely forms the alterna-
tive hairpin with a GCAA tetraloop. Cleavage of the O6-tce
group in 7 or 8 results in sequence 9 and restores the
secondary-structure equilibrium.
The second sequence example was designed to satisfy the
requirements for a complete transformation of the secondary
structure and is represented by the RNA sequence of 5’-
rUGUCUCCG(GACA)CGGAGAC (12). This oligonucleo-
tide forms a highly stable GACA hairpin with a seven-base-
pair stem. However, when a tandem of O6-tce labels is
introduced into this sequence, 5’-rUGUC(UCCG)GACAC
tceGtceGAGAC (11) selects the UCCG hairpin with a four-
base-pair stem as the exclusive conformation. Upon release of
the labels, the secondary structure switches to the thermody-
namically favorable one, 12. All secondary structures dis-
cussed above were ascertained by comparative imino proton
NMR spectroscopy, a simple approach that we have intro-
duced recently for the reliable probing of secondary struc-
tures of small RNAs (see Figure 3 and the Supporting
Information).[16,17]
nucleotides was achieved with coupling yields higher than
99% by using standard DNA/RNA protocols and 2’-O-
triisopropylsilyloxymethyl-protected nucleoside phosphor-
amidites.[10–12] For capping, phenoxyacetic anhydride was
used instead of acetic anhydride. The deprotection of the
synthesized O6-tce-guanosine-containing oligoribonucleoti-
des was optimized thoroughly, as the 2,2,2-trichloroethyl
moiety was unstable under standard RNA-deprotection
conditions based on methylamine. We finally found that
treatment with aqueous ammonia in ethanol (1/1) at 558C for
5–10 hours followed by treatment with 0.9m tetrabutylam-
monium fluoride (TBAF) in THF/water (9/1) at 258C for
14 hours resulted in complete deprotection without strand
degradation (Figure 2). A prerequisite for this procedure was
the use of either N2-methoxyacetylated or N2-unprotected 2’-
O-TOM-guanosine phosphoramidites instead of the com-
monly used N2-acetylated counterparts. Otherwise, cleavage
of the guanosine N2-acetyl groups was incomplete under the
conditions tolerated by the O6-tce group.[13]
Release of the O6-tce group from the oligoribonucleotides
as suggested in our concept (Figure 1) was optimized
experimentally and proceeded smoothly in aqueous ammo-
nium acetate buffer at pH 7 in the presence of zinc dust. Clean
and complete cleavage was achieved at room temperature
and under slight agitation. The half lives of the O6-tce groups
depended on the length of the oligoribonucleotide and ranged
from 0.5 minutes for a pentamer to 5 minutes for the 20-mer
oligoribonucleotides (see the Supporting Information and
Figure 2).[14]
We have demonstrated here that chemical functionaliza-
tion of the Watson–Crick base-pairing site is an efficient tool
for inducing defined rearrangements of RNA secondary
The sequence set in Figure 3 documents the feasibility of
our proposed concept for the modulation of nucleic acid
Figure 3. Secondary-structure modulation of short RNAs containing functionalized nucleosides. a) O6-(2,2,2-trichloroethyl)guanosine (tceG).
b) Incorporation of tceG at positions 17 or 4 of the sequence rGAA(G4)GGCAACCUUCGG(G17)UUG results in conformation selection of either the
GCAA hairpin, 7, or the UUCG hairpin, 8. Release of the O6-tce groups restores the conformational equilibrium of 9 (see also Figure 2). c) rUGU-
C(UCCG)GACACtceGtceGAGAC (11) forms a stable hairpin with a UCCG tetraloop corresponding to reference sequence 10. Upon release of the O6-
tce groups this RNA sequence converts into the thermodynamically favorable GACA hairpin, 12. 1) Zn0, 1m NH4OAc, pH 7, 258C. Structural prob-
ing of the RNA secondary structures by comparative imino proton NMR spectroscopy (500 MHz): 0.2–0.3 mm strand concentration, H2O/D2O
(9:1), 25 mm sodium arsenate buffer, pH 7.4, 268C (see also the Supporting Information).
3924
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 3922 –3925

Angewandte
Chemie
[14] Deprotection of the O6-tce groups was first evaluated for the
structures. The requirements for the functional labels are
quite stringent. With respect to their incorporation into
RNAs, they have to be orthogonal to the protecting groups
used throughout the solid-phase synthesis. This is effectively
achieved by use of the trichloroethyl group. In the late sixties,
this group was introduced into oligonucleotide chemistry in
the context of phosphate protection.[18] With respect to the
applications of the trichloroethyl group for secondary-struc-
ture manipulation, we had to consider that the Zn2+ ions
generated during the cleavage of the O6-tce group may affect
the secondary structure.[19–21] However, for the relatively small
RNAs investigated so far we have no evidence for a structure-
determining influence of the Zn2+ ions. In extending our
approach towards larger RNAs, nucleobase functionalization
by other chemical groups is conceivable; highly promising
candidates are photolabile groups.
The control of conformational changes of nucleic acid
structures is an important issue in RNA biotechnology. In
particular, great efforts have been made in the development
of allosterically triggered ribozymes for gene therapy and for
nucleic acid diagnostics.[22–24] With the present concept for
secondary-structure manipulation by functionalized nucleo-
bases, a potential tool for applications in these fields has been
created.
sequence rCGtceGUA under conditions with various pH values,
strand concentrations, and buffer compositions, as well as
different metals (Zn, Cd). For details, see the Supporting
Information.
[15] For other examples of short bistable RNAs, see: C. Höbartner,
M.-O. Ebert, B. Jaun, R. Micura, Angew. Chem. 2002, 114, 619 –
623; Angew. Chem. Int. Ed. 2002, 41, 605 – 609.
[16] C. Höbartner, R. Micura, J. Mol. Biol. 2003, 325, 421 – 431.
[17] R. Micura, C. Höbartner, ChemBioChem 2003, 4, 984 – 990.
[18] F. Eckstein, I. Rizk, Angew. Chem. 1967, 79, 939; Angew. Chem.
Int. Ed. Engl. 1967, 6, 949.
[19] E. Ennifar, P. Walter, P. Dumas, Nucleic Acids Res. 2003, 31,
2671 – 2682.
[20] E. B. Khomyakova, H. Gousset, J. Liquier, T. Huynh-Dinh, C.
Gouyette, M. Takahashi, V. L. Florentiev, E. Taillandier, Nucleic
Acids Res. 2000, 28, 3511 – 3516.
[21] M. Wu, I. Tinoco, Jr., Proc. Natl. Acad. Sci. USA 1998, 95,
11555 – 11560.
[22] M. Warashina, T. Kuwabara, H. Kawasaki, J. Ohkawa, K. Taira
in RNA (Eds.: D. Söll, S. Nishimura, P. B. Moore), Pergamon
Elsevier Science, Oxford, 2001, pp. 277 – 308.
[23] B. A. Sullenger, E. Gilboa, Nature 2002, 418, 252 – 258.
[24] M. B. Long, J. P. Jones III, B. A. Sullenger, J. Byun, J. Clin.
Invest. 2003, 112, 312– 318.
Received: March 21, 2004 [Z460068]
Keywords: conformational switches · nucleobase modifications ·
.
oligonucleotides · RNA structures
[1] D. M. Crothers in RNA (Eds.: D. Söll, S. Nishimura, P. B.
Moore), Pergamon Elsevier Science, Oxford, 2001, pp. 61 – 70.
[2] W. C. Winkler, R. R. Breaker, ChemBioChem 2003, 4, 1024 –
1032.
[3] R. R. Breaker, Curr. Opin. Biotechnol. 2002, 13, 31 – 39.
[4] S. Sando, T. Sasaki, K. Kanatani, Y. Aoyama, J. Am. Chem. Soc.
2003, 125, 15720 – 15721.
[5] J. S. Hartig, I. Grüne, S. H. Najafi-Shoushtari, M. Famulok, J.
Am. Chem. Soc. 2004, 126, 722 – 723.
[6] A. S. Mironov, I. Gusarov, R. Rafikov, L. E. Lopez, K. Shatalin,
R. A. Kreneva, D. A. Perumov, E. Nudler, Cell 2002, 111, 747 –
757.
[7] M. Mandal, R. Breaker, Nat. Struct. Biol. 2004, 11, 29 – 35.
[8] A. Stutz, C. Höbartner, S. Pitsch, Helv. Chim. Acta 2000, 83,
2477 – 2503.
[9] For detailed experimental procedures, see the Supporting
Information.
[10] R. Micura, Angew. Chem. 2002, 114, 2269 – 2373; Angew. Chem.
Int. Ed. 2002, 41, 2265 – 2267.
[11] S. Pitsch, P. A. Weiss, L. Jenny, A. Stutz, X. Wu, Helv. Chim. Acta
2001, 84, 3773 – 3795.
[12] C. Höbartner, C. Kreutz, E. Flecker, E. Ottenschläger, W. Pils,
K. Grubmayr, R. Micura, Monatsh. Chem. 2003, 134, 851 – 873.
[13] Oligoribonucleotide synthesis and deprotection protocols were
first elaborated for the sequence rCGtceGUA by using N2-acetyl-,
N2-methoxyacetyl-, or N2-unprotected 2’-O-TOM-guanosine
phosphoramidites in various combinations with N2-acetyl-, N2-
methoxyacetyl-, or N2-phenoxyacetyl-protected O6-tce-2’-O-
TOM-guanosine phosphoramidites. These experiments and the
synthesis of the corresponding building blocks will be described
elsewhere.
Angew. Chem. Int. Ed. 2004, 43, 3922 –3925
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3925