Modulation of RNA tertiary folding by incorporation of caged nucleotides

Article abstract of DOI:10.1002/anie.200502928

(Figure Presented) Knock it off: Individual RNA nucleotides with a caged nucleobase were incorporated, through their phosphoramidites, into a 160-mer RNA molecule that adopts a specific tertiary structure (see picture). Position-dependent local and global

Full text of DOI:10.1002/anie.200502928

Angewandte
Chemie
RNA Folding
DOI: 10.1002/anie.200502928
Modulation of RNA Tertiary Folding by
Incorporation of Caged Nucleotides**
Claudia Höbartner and Scott K. Silverman*
Photochemical caging groups[1] are often used to study
biomolecular interactions, such as those between cofactors
and enzymes or between neurotransmitters and their recep-
tors.[2] In a caging experiment, a functionally important part of
a molecule is modified by the strategic introduction of a non-
natural moiety that disrupts key contacts or interferes with a
critical reaction step. When the non-natural moiety is
removed by photolysis, the native biomolecular interactions
are restored. Photochemical caging has been used to study
protein folding,[3] RNA catalysis,[4] secondary structure
changes in RNA,[5] and other RNA-related phenomena,[6]
but not higher-order tertiary folding of RNA.[7] When a
caging group is appended at any specific nucleotide position,
the effect on RNA tertiary folding is difficult to predict. This
is unlike the situation of RNA catalysis (for which caging of
an active-site hydroxy-group nucleophile is certain to disrupt
reactivity) or of RNA secondary structure changes (for which
reliable Watson–Crickinteractions provide a strong predic-
tive tool). Herein we report how an RNA molecule with a
well-defined tertiary structure responds to the introduction of
individual caging groups. Our results, which demonstrate
varying local and global structural effects, have several
implications for ongoing efforts to understand RNA folding.
We appended the photochemically removable (S)-1-(2-
nitrophenyl)ethyl (NPE) group onto the nucleobase of each
of the four standard RNA nucleotides (Figure 1).[8] The four
corresponding phosphoramidites (NPE-U, NPE-C, NPE-A,
and NPE-G) were prepared for conventional solid-phase
RNA synthesis by using the 2’-O-triisopropylsilyloxymethyl
(TOM) methodology.[9] The preparative route for the caged
guanosine phosphoramidite was based on a recent report,[5]
and the routes for the other three phosphoramidites are
detailed in the Supporting Information. For each of the four
phosphoramidites, solid-phase synthesis was used to prepare
Figure 1. Structures of the (S)-NPE-cagedRNA nucleotides.
15-mer and 24-mer RNA oligonucleotides, each of which
incorporate a single caged RNA nucleotide at a specific
internal position. In all cases, denaturing polyacrylamide gel
electrophoresis (PAGE), anion-exchange HPLC analysis, and
ESI mass spectrometry revealed that decaging proceeds to
completion in less than 10 min by using a 300-W Xe-arc-lamp
source (300–400 nm; Figure 2) or in approximately 30–60 min
by using a handheld UV illuminator (365 nm; data not
shown).[10]
We then placed individual caged nucleotides within a
large RNA molecule, P4–P6. The 160-nucleotide P4–P6
domain of the Tetrahymena group I intron RNA adopts a
characteristic tertiary structure in the presence of millimolar
concentrations of Mg2+.[11,12] As revealed by X-ray crystallog-
raphy, the folded domain adopts a three-dimensional struc-
ture that resembles a candy cane (Figure 3).[13,14] The two
largely helical domains of the RNA are held together, in part,
by a key tertiary interaction between the GAAA tetraloop
and its 11-nucleotide receptor.[15] Nucleotide and functional-
group alterations within this tetraloop–receptor interaction
have deleterious effects on the P4–P6 thermodynamic stabil-
ity.[11,12,15–17]
[*] Dr. C. Höbartner, Prof. S. K. Silverman
Department of Chemistry
We sought to probe the effects on P4-P6 folding of caging
specific nucleotides with the NPE group. For this purpose, we
synthesized modified P4–P6 variants by using a ligation
strategy that joins a caged 15-mer or 24-mer oligonucleotide
with an unmodified 145-mer or 136-mer T7 RNA polymerase
transcript (see Supporting Information). We prepared 14
caged P4–P6 derivatives in this fashion. The caging sites were
distributed throughout the RNA and were chosen on the basis
of the X-ray crystal structure (Figure 3a). Caging sites were
placed in positions where the NPE group was anticipated to
have a combination of three effects: 1) alteration of the
hydrogen-bonding interactions within either secondary- or
University of Illinois at Urbana-Champaign
600 South Mathews Avenue, Urbana, IL 61801 (USA)
Fax: (+1)217-244-8024
E-mail: scott@scs.uiuc.edu
[**] This research was supportedby an Erwin Schrödinger fellowship
from the Austrian Science Fundto C.H. andby the UIUC
Department of Chemistry. S.K.S. is a Fellow of The DavidandLucile
PackardFoundation. We thank Prof. RonaldMicura (Innsbruck) for
use of the HPLC system, Mary Smalley and Chandra Miduturu for
assistance with RNA transcriptions, andKimberly Peterson for
preliminary synthetic experiments.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 7305 –7309
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7305

Communications
Figure 2. Photodeprotection of caged RNA oligoribonucleotides.
Figure 3. The 160-nucleotide P4–P6 RNA and sites of caging. a) X-ray
crystal structure[13] with caging sites marked. Spheres are centered on
the nucleobase atoms that bear the NPE caging groups andare
coloredredor green as in the data plots. The sphere marking the N6
atom of A248 is larger than the other spheres. Adenosine nucleotides
A151, A152, andA153 of the GAAA tetraloop are thickenedfor
emphasis. See Supporting Information for a picture of secondary
structure. b) The base triple involving A248.
a) Denaturing PAGE (20%) shows the photolysis time course for each
of four caged15-mer oligonucleotides (5’-GGAAUUGCGGGAAAG-3’,
the underlined positions were individually caged with the NPE group).
Control (ctrl)=oligonucleotide of the same sequence without the NPE
group. b) Representative anion-exchange HPLC traces for the oligonu-
cleotide with NPE-caged adenosine. Photolysis was carried out with a
300-W Xe arc lamp. (Traces for the other three cagedoligonucleotides
are shown in the Supporting Information, along with ESIMS data
before andafter photolysis for all four cagedoligonucleotides.)
which reports on global folding (Figure 4a).[17] The P4–P6
domain has a Mg2+-dependent mobility on native PAGE; that
is, when the RNA folds at higher Mg2+ concentrations, it
adopts a more compact structure and thus migrates faster
through the gel. Thermodynamic disruptions to the P4–P6
structure result in an increased Mg2+ requirement for folding,
tertiary-structure elements; 2) disruption of base-stacking
interactions; and 3) introduction of steric clashes (see Sup-
porting Information for detailed images). In the NPE-A248
P4–P6 derivative, for example, the NPE caging group is
placed on an adenine N6 group that is located in a densely
packed region of RNA structure and at the interface between
the GAAA tetraloop and the tetraloop receptor (Figure 3a).
Because A248 is the central nucleotide of a base triple
(Figure 3b) and because there appears to be no room to
accommodate the bulky NPE group, NPE caging of A248 is
expected to disrupt several hydrogen bonds and introduce a
steric clash directly within the key tetraloop–receptor tertiary
interaction. The determination of the consequences of such
disruptions on RNA tertiary folding was a main experimental
goal of this study.
as quantified by the Mg2+ midpoint ([Mg2+
]
value) of a
1/2
mobility versus [Mg2+] titration curve.[15,17,18] Our data
allowed all 14 of the caged P4–P6 derivatives to be classified
readily into just two types (Figure 4b): 1) those in which the
caging group has a substantially destabilizing thermodynamic
impact on tertiary folding (red), and 2) those in which the
thermodynamic impact is small (green). Regardless of the
effect of the caging group, its photolytic removal before
analysis by native PAGE restored the RNA folding to that of
the unmodified wild-type P4–P6 (Figure 4).
The effects of individual caged nucleotides on P4–P6
folding were analyzed by nondenaturing (native) PAGE,
When the differential effects of caging the various
nucleotides are considered in the context of their locations
7306 www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7305 –7309

Angewandte
Chemie
that the NPE caging group
should be accommodated
without causing a steric clash.
To provide an assessment
of RNA folding that is inde-
pendent of the native gel
experiments, chemical prob-
ing with dimethyl sulfate was
also performed. This reagent
methylates either N1 of ade-
nine or N3 of cytosine but
only if the nucleobase is
locally accessible to the
reagent.[11,12,19] Tertiary fold-
ing of the RNA molecule
buries some of the nucleobase
functional groups within its
interior;[20] this renders these
functional groups inaccessible
to dimethyl sulfate and thus
protects them from methyla-
tion. Because the functional
groups
methylated
by
dimethyl sulfate are located
on the Watson–Crickfaces of
the nucleobases, methylation
inhibits primer extension by
reverse transcriptase and
leads to an abort band when
the RTase reaction assay is
examined by denaturing
PAGE.
The
adenosine
Figure 4. Native PAGE analysis of cagedP4–P6 RNAs. a) Representative gel images. wt =unmodified wild-
type P4–P6; unf=unfolded control that cannot fold even at high [Mg2+].[11,17,23] b) Mg2+ dependence of
native gel mobility, computed for each caged derivative relative to the unfolded control. The wild-type P4–
nucleotides of the GAAA
tetraloop in unfolded P4–P6
(for example, P4–P6 in the
absence of Mg2+) are methy-
lated readily by dimethyl sul-
fate. These nucleotides, how-
ever, are substantially pro-
tected from methylation
when P4–P6 is folded prop-
*
P6 is in black ( ). Data in red represent positions where an unambiguous thermodynamic disruption is
~
!
&
^
observed( A246; U247; A248; G250). Data in green represent positions where little effect is
observed( ” C109; the remaining symbols represent other positions from panel (a)). Photolysis of samples
before native PAGE restored folding to that equivalent to the wild-type P4–P6 (brown). See the Supporting
Information for experimental details.
within P4–P6, a clear pattern emerges. Caging of a nucleotide
far from the tetraloop–receptor interaction has little effect on
global folding. In sharp contrast, caging of a nucleotide that is
located within (or very close to) the tetraloop–receptor
interaction, such as A248, introduces a distinct thermody-
namic destabilization. Nevertheless, native PAGE indicates
that each of the latter derivatives eventually adopts the native
global tertiary structure when the Mg2+ concentration is
increased sufficiently (ꢀ 20 mm). For the NPE-caged A248
derivative, this finding is surprising because the tetraloop–
receptor interaction must distort considerably to accommo-
date the NPE group, although the precise nature of the
distortion cannot be discerned from these data. The only
exception to the clear pattern is that the NPE-caged U249
derivative is not destabilized even though it is located within
the receptor. Significantly, U249 is a bulged nucleotide in
which the nucleobase does not form any tertiary hydrogen
bonds, and inspection of the X-ray crystal structure suggests
erly owing to their sequestration in the tetraloop–receptor
interaction.[11,12]
For several of the caged P4–P6 RNAs, dimethyl sulfate
probing was performed at varying Mg2+ concentrations, and
the accessibility of the GAAA tetraloop was monitored
(Figure 5a). These experiments provide information on local
structure that cannot be determined with native PAGE.
Depending on the caging site, the dimethyl sulfate probing
results revealed a difference in the accessibilities of the
GAAA tetraloop at high Mg2+ concentrations (Figure 5b).
The tested caging sites that are remote from the tetraloop–
receptor interaction (U107, C109) had essentially no effect on
Mg2+-dependent accessibility of the tetraloop. In contrast,
caging of nucleotides within or near the receptor (A246,
U247, A248, and G250) significantly suppressed protection of
the tetraloop from dimethyl sulfate modification upon an
increase in the Mg2+ concentration. As with the native gel
assays, U249 located within the receptor provides a predict-
Angew. Chem. Int. Ed. 2005, 44, 7305 –7309
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7307

Communications
contacts, such as the tetraloop–receptor interaction, are not
strictly required for RNA tertiary folding. It also suggests that
a chemically more sophisticated strategy (e.g., one that
involves the attachment of a covalent tether at more than
one position) is required to induce qualitative changes in
global RNA structure that cannot be recovered simply by an
increase in the Mg2+ concentration.[22] Second, comparisons of
the native PAGE and dimethyl sulfate probing data indicate a
clear differentiation between local and global structural
effects upon the introduction of a chemical perturbation.
This relates to the ability of individual caging groups to induce
misfolded RNA states. A future direction for these studies
will be to combine individual caging groups or photocleavable
covalent tethers with time-resolved decaging to monitor RNA
tertiary-folding processes in real time.
Received: August 18, 2005
Published online: October 17, 2005
Keywords: caging · nucleobases · oligonucleotides · photolysis ·
.
RNA structures
Figure 5. Chemical probing experiments with dimethyl sulfate of caged
P4–P6 RNAs. a) Representative gel images for reverse transcriptase
(RTase) assays of cagedP4–P6 treatedwith dimethyl sulfate. The
presence of an RTase abort bandindicates methylation of the
corresponding nucleobase with dimethyl sulfate, which must have
been accessible to the reagent. The GAAA tetraloop includes nucleo-
tides A151–A153. b) Mg2+ dependence of fraction methylated for the
tetraloop adenosine nucleotides when various nucleotides are caged
[1] J. H. Kaplan, B. Forbush III, J. F. Hoffman, Biochemistry 1978,
17, 1929 – 1935.
[2] a) H. A. Lester, J. M. Nerbonne, Annu. Rev. Biophys. Bioeng.
1982, 11, 151 – 175; b) A. M. Gurney, H. A. Lester, Physiol. Rev.
1987, 67, 583 – 617; c) J. A. McCray, D. R. Trentham, Annu. Rev.
Biophys. Biophys. Chem. 1989, 18, 239 – 270; d) S. R. Adams,
R. Y. Tsien, Annu. Rev. Physiol. 1993, 55, 755 – 784; e) B. E.
Cohen, B. L. Stoddard, D. E. Koshland, Jr., Biochemistry 1997,
36, 9035 – 9044; f) J. C. Miller, S. K. Silverman, P. M. England,
D. A. Dougherty, H. A. Lester, Neuron 1998, 20, 619 – 624; g) K.
Curley, D. S. Lawrence, Curr. Opin. Chem. Biol. 1999, 3, 84 – 88;
h) A. P. Pelliccioli, J. Wirz, Photochem. Photobiol. Sci. 2002, 1,
441 – 458; i) J. E. T. Corrie in Dynamic Studies in Biology (Eds.:
M. Goeldner, R. Givens), Wiley-VCH, Weinheim, 2005, pp. 1 –
28.
~
~
&
*
(
^
wild-type RNA; U107, ” C109; & U249; A246; U247;
!
A248; G250). A high fraction methylatedimplies high accessibility
of the tetraloop to dimethyl sulfate. Photolysis of the samples before
dimethyl sulfate probing restored protection from methylation (see
Supporting Information).
able exception in this regard. These data show that the
tetraloop–receptor interaction can be substantially perturbed
on the local structural level, even when the global structure is
essentially native. This conspicuous difference between local
and global structural effects due to caging of specific
nucleotides increases our understanding of how ꢀkey inter-
actionsꢁ contribute to RNA tertiary structure. Our results may
also relate to the concept of compact, misfolded RNA
states.[21] Some of the caged P4–P6 derivatives such as NPE-
caged A248 adopt globally compact structures (as revealed by
native PAGE) that nevertheless lackat least one native
tertiary interaction (as indicated by dimethyl sulfate probing).
In summary, we have shown herein that nucleotides with
caged nucleobases are readily incorporated into large RNA
molecules that have complex tertiary structures, and the
effects of the caged nucleotides are assessable with appro-
priate biochemical assays. From our results, we drew two
conclusions that relate to the control of RNA folding by using
caged nucleotides. First, incorporation of a bulky caging
group can thermodynamically disrupt global RNA tertiary
structure, but in all tested cases the disruption is relatively
modest because it can be overcome by a sufficient increase in
the Mg2+ concentration. This indicates that even ꢀkeyꢁ tertiary
[3] a) H. S. M. Lu, M. Volk, Y. Kholodenko, E. Gooding, R. M.
Hochstrasser, W. F. DeGrado, J. Am. Chem. Soc. 1997, 119,
7173 – 7180; b) K. C. Hansen, R. S. Rock, R. W. Larsen, S. I.
Chan, J. Am. Chem. Soc. 2000, 122, 11567 – 11568.
[4] a) S. G. Chaulk, A. M. MacMillan, Nucleic Acids Res. 1998, 26,
3173 – 3178; b) S. G. Chaulk, A. M. MacMillan, Angew. Chem.
2001, 113, 2207 – 2210; Angew. Chem. Int. Ed. 2001, 40, 2149 –
2152.
[5] P. Wenter, B. Fürtig, A. Hainard, H. Schwalbe, S. Pitsch, Angew.
Chem. 2005, 117, 2656 – 2659; Angew. Chem. Int. Ed. 2005, 44,
2600 – 2603.
[6] a) L. Kröck, A. Heckel, Angew. Chem. 2005, 117, 475 – 477;
Angew. Chem. Int. Ed. 2005, 44, 471 – 473; b) S. Shah, S.
Rangarajan, S. H. Friedman, Angew. Chem. 2005, 117, 1352 –
1356; Angew. Chem. Int. Ed. 2005, 44, 1328 – 1332.
[7] a) P. Brion, E. Westhof, Annu. Rev. Biophys. Biomol. Struct.
1997, 26, 113 – 137; b) I. Tinoco, Jr., C. Bustamante, J. Mol. Biol.
1999, 293, 271 – 281.
[8] The S isomer of the NPE group was chosen to be consistent with
the report of Wenter et al.[5]
[9] a) S. Pitsch, P. A. Weiss, L. Jenny, A. Stutz, X. Wu, Helv. Chim.
Acta 2001, 84, 3773 – 3795; b) R. Micura, Angew. Chem. 2002,
114, 2369 – 2373; Angew. Chem. Int. Ed. 2002, 41, 2265 – 2269.
[10] The caged uridine derivative with the NPE group at N3 instead
of at O4 was also prepared. However, its photocleavage is
inefficient (only ꢁ 50% deprotection in 10 min with the arc
7308
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 7305 –7309

Angewandte
Chemie
lamp) and results in a mixture of two cleavage products, only one
of which has an unmodified uracil nucleobase. Therefore, we did
not pursue the N3-NPE-caged uridine derivative any further. See
Supporting Information for details.
Doudna, Nat. Struct. Biol. 2001, 8, 339 – 343; f) S. Matsumura, Y.
Ikawa, T. Inoue, Nucleic Acids Res. 2003, 31, 5544 – 5551.
[19] D. A. Peattie, Proc. Natl. Acad. Sci. USA 1979, 76, 1760 – 1764.
[20] D. W. Celander, T. R. Cech, Science 1991, 251, 401 – 407.
[21] a) S. A. Woodson, Nat. Struct. Biol. 2000, 7, 349 – 352; b) K. L.
Buchmueller, A. E. Webb, D. A. Richardson, K. M. Weeks, Nat.
Struct. Biol. 2000, 7, 362 – 366; c) R. Russell, I. S. Millett, S.
Doniach, D. Herschlag, Nat. Struct. Biol. 2000, 7, 367 – 370; d) R.
Russell, I. S. Millett, M. W. Tate, L. W. Kwok, B. Nakatani, S. M.
Gruner, S. G. Mochrie, V. Pande, S. Doniach, D. Herschlag, L.
Pollack, Proc. Natl. Acad. Sci. USA 2002, 99, 4266 – 4271;
e) X. W. Fang, P. Thiyagarajan, T. R. Sosnick, T. Pan, Proc.
Natl. Acad. Sci. USA 2002, 99, 8518 – 8523; f) K. Takamoto, Q.
He, S. Morris, M. R. Chance, M. Brenowitz, Nat. Struct. Biol.
2002, 9, 928 – 933; g) K. L. Buchmueller, K. M. Weeks, Biochem-
istry 2003, 42, 13869 – 13878; h) K. Takamoto, R. Das, Q. He, S.
Doniach, M. Brenowitz, D. Herschlag, M. R. Chance, J. Mol.
Biol. 2004, 343, 1195 – 1206.
[11] F. L. Murphy, T. R. Cech, Biochemistry 1993, 32, 5291 – 5300.
[12] F. L. Murphy, T. R. Cech, J. Mol. Biol. 1994, 236, 49 – 63.
[13] J. H. Cate, A. R. Gooding, E. Podell, K. Zhou, B. L. Golden,
C. E. Kundrot, T. R. Cech, J. A. Doudna, Science 1996, 273,
1678 – 1685.
[14] K. Juneau, E. Podell, D. J. Harrington, T. R. Cech, Structure
2001, 9, 221 – 231.
[15] B. T. Young, S. K. Silverman, Biochemistry 2002, 41, 12271 –
12276.
[16] D. L. Abramovitz, A. M. Pyle, J. Mol. Biol. 1997, 266, 493 – 506.
[17] S. K. Silverman, T. R. Cech, Biochemistry 1999, 38, 8691 – 8702.
[18] a) S. K. Silverman, T. R. Cech, Biochemistry 1999, 38, 14224 –
14237; b) S. K. Silverman, M. Zheng, M. Wu, I. Tinoco, Jr., T. R.
Cech, RNA 1999, 5, 1665 – 1674; c) S. K. Silverman, M. L. Deras,
S. A. Woodson, S. A. Scaringe, T. R. Cech, Biochemistry 2000,
39, 12465 – 12475; d) S. K. Silverman, T. R. Cech, RNA 2001, 7,
161 – 166; e) E. A. Doherty, R. T. Batey, B. Masquida, J. A.
[22] a) C. R. Allerson, G. L. Verdine, Chem. Biol. 1995, 2, 667 – 675;
b) G. D. Glick, Biopolymers 1998, 48, 83 – 96; c) S. B. Cohen,
T. R. Cech, J. Am. Chem. Soc. 1997, 119, 6259 – 6268.
[23] A. A. Szewczak, T. R. Cech, RNA 1997, 3, 838 – 849.
Angew. Chem. Int. Ed. 2005, 44, 7305 –7309
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7309