Synthesis of (6-13C)pyrimidine nucleotides as spin-labels for RNA dynamics

Article abstract of DOI:10.1021/ja302148g

We present a ¹³C-based isotope labeling protocol for RNA. Using (6-¹³C)pyrimidine phosphoramidite building blocks, site-specific labels can be incorporated into a target RNA via chemical oligonucleotide solid-phase synthesis. This la

Full text of DOI:10.1021/ja302148g

Article
pubs.acs.org/JACS
Synthesis of (6-13C)Pyrimidine Nucleotides as Spin-Labels for RNA
Dynamics
Christoph H. Wunderlich, Romana Spitzer, Tobias Santner, Katja Fauster, Martin Tollinger,
and Christoph Kreutz*
Institute of Organic Chemistry and Center for Molecular Biosciences Innsbruck (CMBI), University of Innsbruck, Innrain 80/82,
6020 Innsbruck, Austria
S
* Supporting Information
ABSTRACT: We present a 13C-based isotope labeling
protocol for RNA. Using (6-13C)pyrimidine phosphoramidite
building blocks, site-specific labels can be incorporated into a
target RNA via chemical oligonucleotide solid-phase synthesis.
This labeling scheme is particularly useful for studying milli- to
microsecond dynamics via NMR spectroscopy, as an isolated
spin system is a crucial prerequisite to apply Carr−Purcell−
Meiboom−Gill (CPMG) relaxation dispersion type experi-
ments. We demonstrate the applicability for the character-
ization and detection of functional dynamics on various time
scales by incorporating the (6-13C)uridine and -cytidine labels into biologically relevant RNAs. The refolding kinetics of a bistable
terminator antiterminator segment involved in the gene regulation process controlled by the preQ1 riboswitch class I was
investigated. Using 13C CPMG relaxation dispersion NMR spectroscopy, the milli- to microsecond dynamics of the HIV-1
transactivation response element RNA and the Varkud satellite stem loop V motif was addressed.
by applying the techniques to the isolated amide 15N−1H spin
1. INTRODUCTION
systems.10 In principle, nucleic acids would provide a similar
Besides the structural characterization of biologically relevant
spin system, namely, the NH protons of uridine (N3H) and
macromolecules, solution NMR spectroscopy is perfectly suited
guanosine (N1H), which are involved in the formation of base
to study dynamic features over a wide range of time scales, from
nano- to milliseconds and beyond.1−4 Fast fluctuations in the
pair hydrogen bonds. However, these protons often suffer from
fast exchange with solvent protons, especially in single-stranded
nanosecond time regime can be characterized by the analysis of
regions, where dynamic phenomena are commonly located.
longitudinal and transverse relaxation rates and the hetero
nuclear Overhauser effect (hetero-NOE) of 13C and 15N spins
The broadening effect due to this chemical exchange can render
in the biomolecule of interest.4,5 Alternatively, residual dipolar
the nitrogen-bound imino proton resonances unobservable.
Alternatively, nonexchangeable 13C−1H spin systems would
couplings (RDCs) can be used to extract dynamic information
on a biomolecule.4,6 The interpretation of the relaxation and
provide information on the structural and dynamic changes
occurring in single-stranded sequence regions of a nucleic
RDC data to delineate quantitative information on the
structural flexibility of a biomolecule can be a challenging
task, e.g., if a motional coupling between local fluctuations and
acid.11
A first mandatory step to study both structural and dynamic
the overall tumbling rate of the molecule occurs.7 Order
features of biomacromolecules is the introduction of an
appropriate isotope labeling pattern.12−15 Currently, enzymatic
methods to introduce uniformly 13C- and/or 15N-labeled
nucleotides into RNA and DNA are state-of-the-art. Using
these methods, nucleotide-specific labeling patterns (i.e., only
adenosines/uridines or guanosines/cytidines) can be applied in
a straightforward manner by mixing labeled and unlabeled
(deoxy)ribonucleotide triphosphates in polymerase-catalyzed
reactions.16 This labeling pattern, however, does impair to some
extent the applicability of relaxation dispersion techniques.3,11
Very recently, to bypass problems caused by the uniform
labeling pattern, site-specific 13C isotope labeling of RNA was
parameters (0 < S2 < 1) are extracted via the model-free
approach, giving information on the flexibility of a certain
residue.8,9
Motions in the micro- to millisecond time regime can be
addressed using relaxation dispersion (RD) techniques.1,3,4
Structural adaptations of biomolecules in this time regime are
often involved in functionally important processes, such as
ligand binding or enzymatic reactions. Two experiments, the
Carr−Purcell−Meiboom−Gill (CPMG) relaxation dispersion
method and the determination of the power dependence of
R1ρ, are suitable to extract the exchange lifetimes between
multiple states, the populations of the states, and the chemical
shift difference between the states.1 In proteins, both methods
are routinely used to obtain information on backbone dynamics
Received: March 4, 2012
Published: April 11, 2012
© 2012 American Chemical Society
7558
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
addressed.17−21 This is a crucial prerequisite to apply NMR
methods for studying micro- to millisecond dynamics, as strong
scalar coupling effects (e.g., 13C−13C couplings in ribose) or
dipolar interactions with adjacent 13C or 15N spins are absent.
These technical issues make relaxation dispersion experiments
difficult to perform for nucleic acids as documented by the
relatively small number of such studies so far.22−24 Besides
enzymatic RNA/DNA synthesis, significant progress was made
in the chemical synthesis of oligonucleotides. Commercially
available DNA sequences comprising up to 100 nucleotides can
be synthesized for biophysical investigations. RNA constructs,
on the other hand, with up to 50 nucleotides can be produced
in amounts suitable for NMR spectroscopic investigations. This
approach offers the highest flexibility on placing isotope
labeling and on the number of labels, but depends on the
(labor-intensive) chemical synthesis of appropriate isotope 13C-
modified phosphoramidites prior to sequence assembly.16 Our
group recently introduced a near-native isotope labeling
protocol relying on a 2′-O-13CH3-modified uridine phosphor-
amidite precursor.24 The 2′-O-13CH3 spin-label was use to
detect and quantify structural heterogeneity on a secondary
structure level and further to probe milli- to microsecond
dynamics by CPMG relaxation dispersion experiments. Within
this work we present a fully native site-specific 13C labeling
protocol for RNA. The proposed method uses (6-13C)-
pyrimidine precursors to introduce isolated 13C−1H spin
systems into the target RNA via chemical synthesis.25 The
(6-13C)uridine and -cytidine residues proved to be appropriate
labels to monitor functional micro- and millisecond dynamics
of RNAs, as well as the kinetics of a secondary structure
refolding reaction of a mRNA terminator antiterminator
sequence involved in a gene regulation process.26
evaporated, and the remaining yellow 1 was dried in high
vacuum. Yield: 3.96 g (92%). H NMR (300 MHz, DMSO-d6,
1
25 °C): δ 3.46 (d, 2JCH = 9.5 Hz, 2H, H2C13C) ppm. 13C NMR
(75 MHz, DMSO-d6, 25 °C): δ 168.1 (COOH); 117.3 (CN);
2
26.0 (CH2, d, JCC = 59.4) ppm.
2.2.2. [(3-13C)Cyanoacetyl]urea (2). Compound 1 (2.0 g, 23.3
mmol) was treated with urea (1.5 g, 25 mmol) in 5 mL of acetic
anhydride The reaction mixture was heated to 90 °C for 30 min. After
5 min a white precipitate was formed. Then 8 mL of water was added,
and stirring was continued for another 5 min. After the solution was
cooled to room temperature, the precipitated 2 was filtered off and
1
dried in high vacuum overnight. Yield: 2.68 g (90%). H NMR (300
MHz, DMSO-d6, 25 °C): δ 10.38 (s, 1H, NH); 7.35 (s, 2H, NH2);
2
3.90 (d, JCH = 10.0 Hz, 2H, H2C13C) ppm. 13C NMR (75 MHz,
DMSO-d6, 25 °C): δ 168.08 (HNCOCH2); 165.72 (H2NCON);
1
115.6 (CN); 24.65 (CH2, d, JCC = 61.5 Hz) ppm.
2.2.3. (6-13C)Uracil (3). Pd/BaSO4 (5%, 800 mg) was suspended in
10 mL of 50% aqueous acetic acid in a 100 mL three-necked round-
bottom flask. The palladium catalyst was treated several times with
hydrogen by evacuating the flask followed by spilling it with hydrogen.
Simultaneously, compound 2 (1.6 g, 12.5 mmol) was dissolved in 40
mL of boiling aqueous 50% acetic acid and then added to the reduced
palladium catalyst. The reaction was stirred at room temperature
under a hydrogen atmosphere for 36 h. Before being filtered through
Celite, the mixture was heated to 70 °C for 1 h. The solvent was
evaporated until a white precipitate was formed. Then 3 was
precipitated by storing the suspension at 4 °C overnight. Compound
3 was filtered off and dried in high vacuum overnight. Yield: 1.15 g
1
(82%). TLC (CH2Cl2/MeOH = 9/1): Rf = 0.3. H NMR (300 MHz,
1
DMSO-d6, 25 °C): δ 10.64 (s, 2H, NH); 7.38 (dd, 1H JCH = 180.6
Hz, 3JHH = 7.5 Hz, C6H); 5.44 (dd, 1H, 2JCH = 4.0 Hz, 3JHH = 7.5 Hz,
C5H) ppm. 13C NMR (75 MHz, DMSO-d6, 25 °C): δ 164.31 (C4);
1
151.50 (C2); 142.17 (C6); 100.18 (d, JCC = 65.8 Hz, C5) ppm.
2.2.4. (2′,3′,5′-O-Tribenzoyl)(6-13C)uridine (4). Compound 3 (560
mg, 5 mmol) was mixed with 1-O-acetyl-(2′,3′,5′-O-tribenzoyl)-β-D-
ribofuranose (ATBR; 2.5 g, 5 mmol) and suspended in 20 mL of dry
acetonitrile. Then bis(trimethylsilyl)acetamide (BSA; 3.7 mL, 15
mmol) was added, and the mixture was heated to 60 °C for 30 min,
resulting in a clear solution. After the addition of trimethylsilyl
trifluoromethanesulfonate ((TMS)OTf; 3.2 mL, 17.5 mmol), stirring
at 60 °C was continued for another 30 min. The reaction was
monitored by TLC, which showed complete conversion after the 1/2
h of stirring. The solvent was evaporated and the oily residue dried in
high vacuum. The oil was dissolved in methylene chloride, and the
organic phase was washed with saturated sodium bicarbonate solution,
dried over sodium sulfate, and evaporated to give a yellow solid.
Compound 4 was used without further purification in the next
reaction. Yield: 2.73 g (98%). TLC (ethyl acetate/hexanes = 1/1): Rf =
2. EXPERIMENTAL SECTION
2.1. General Procedures. NMR spectra were acquired on a
Varian 500 MHz Unity Plus instrument or a Bruker 300 MHz DRX
instrument. Chemical shifts are reported relative to TMS and
referenced to the residual proton solvent signal: CDCl3 (7.26 ppm)
1
for H NMR spectra and CDCl3 (77.0 ppm) for 13C spectra, DMSO-
1
d6 (2.50 ppm) for H NMR spectra and DMSO-d6 (39.52 ppm) for
1
13C spectra, and D2O (4.79 ppm) for H NMR spectra. 31P shifts are
1
reported relative to external phosphoric acid (85%). H assignments
are based on gradient-selected correlation spectroscopy (COSY)
experiments. 13C shifts are assigned from gradient-selected phase-
sensitive heteronuclear single-quantum coherence (HSQC) and
magnitude heteronuclear multiple-bond correlation (HMBC) experi-
ments. Silica 60F-254 plates were used for TLC (thin-layer
chromatography). For FC (flash column chromatography) silica gel
60 (230−400 mesh) was used. All reactions were conducted under an
argon atmosphere. Reagents and solvents were purchased from Sigma-
Aldrich and used without further purification. Organic solvents were
extensively dried using freshly activated molecular sieves (4 Å).
2.2. Synthesis of the (6-13C)Uridine tert-Butyldimethylsilyl
(TBDMS)- and [(Triisopropylsilyl)oxy]methyl (TOM)-Protected
Phosphoramidites 8 and 10. 2.2.1. (3-13C)Cyanoacetic Acid (1).
2-Bromoacetic acid (6.99 g, 50.3 mmol) was dissolved in 20 mL
of water. Then sodium carbonate (Na2CO3; 3.4 g, 32.1 mmol)
dissolved in 10 mL of water was added to reach pH 9. 13C-
labeled potassium cyanide (3.24 g, 49 mmol) was dissolved in
10 mL of water and added to the deprotonated 2-bromoacetic
acid. The reaction mixture was heated to 80 °C for 3 h and then
stirred at room temperature for 20 h. The reaction was stopped
by the addition of concentrated hydrochloric acid until pH 1.
The solvent was evaporated and the residue dried in high
vacuum for 30 min. Then (2-13C)cyanoacetic acid was extracted
from the salt cake with 500 mL of diethyl ether. The ether was
1
0.8. H NMR (300 MHz, CDCl3, 25 °C): δ 8.66 (s, NH); 8.12−7.3
(m, 15H, CH(ar)); 7.4 (dd, 1H 1JCH = 180.5 Hz, 3JHH = 8.3 Hz, C6H);
6.31 (triplettoid, 1H, CH1′); 5.89 (dd, 1H, CH3′); 5.75 (dd, 1H,
2
3
CH2′); 5.61 (dd, 1H, JCH = 4.5 Hz, JHH = 7.7 Hz, C5H); 4.84 (dd,
1H, CH5″); 4.70 (m, 1H, CH4′); 4.68 (dd, 1H, CH5′) ppm. 13C NMR
(75 MHz, CDCl3, 25 °C): δ 139.69 (C6); 133.98, 133.83, 130.04,
129.80, 128.92, 128.70 (6 C(ar)); 104.66 (C5); 88.27 (C1′); 81.20
(C4′); 73.88 (C3′); 71.31 (C2′); 64.10 (C5′) ppm.
2.2.5. (6-13C)Uridine (5). Crude compound 4 (2.5 g, 4.5 mmol) was
treated with 15 mL of methylamine in ethanol (8 M) for 16 h under an
argon atmosphere. After evaporation of the solvent, the oily residue
was dissolved in water and the N-methylbenzamide was removed by
extensive washing of the aqueous phase with methylene chloride. The
aqueous phase evaporated to dryness, and the resulting yellow foam
was recrystallized from ethanol to give 5. Yield: 1.1 g (90% referring to
(6-13C)uracil). TLC (CH2Cl2/MeOH = 9/1): Rf = 0.1. 1H NMR (300
MHz, D2O, 25 °C): δ 7.75 (dd, 1H 1JCH = 184.14 Hz, 3JHH = 8.0 Hz,
C6H); 5.79 (m, 1H, CH1′); 5.77 (m, C5H); 4.11 (triplettoid, 1H,
CH2′); 4.23 (triplettoid, 1H, CH3′); 4.01 (dd, 1H, CH4′); 3.80 (dd,
1H, CH5″); 3.68 (dd, 1H, CH5′) ppm. 13C NMR (75 MHz, D2O, 25
1
°C): δ 165.6 (C4); 141.38 (C6); 151.27 (C2); 101.79 (d, JCC = 67.2
7559
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
Hz, C5); 88.92 (C1′); 83.74 (C4′); 73.19 (C2′); 68.96 (C3′); 60.28
(C5′) ppm.
C6H); 7.45−7.15 (m, 9 CH(ar)); 6.90−6.77 (m, 4 CH(ar)); 5.99
(triplettoid, 1H, CH1′); 5.31 (m, 1H, C5H); 4.44 (triplettoid, 1H,
CH2′); 4.33 (m, 1H, CH3′); 4.22 (m, 1H, CH4′); 4.12 (ddd, 1H, 2JHH
2.2.6. 5′-O-(4,4′-Dimethoxytrityl)(6-13C)uridine (6). 5 (1.0 g, 4.1
mmol) was coevaporated three times with anhydrous pyridine and
then dissolved in 5 mL of anhydrous pyridine. Then (4,4′-
dimethoxytriphenyl)methyl chloride ((DMT)Cl; 1.7 g, 5.0 mmol)
was added in three portions within 1 h. The reaction mixture was
stirred for 4 h at room temperature. TLC showed nearly complete
conversion at that time point. The solvent was evaporated and the
residue coevaporated three times with toluene. The orange foam was
dried in high vacuum and then dissolved in methylene chloride. The
organic phase was washed with 5% citric acid, water, and saturated
sodium bicarbonate solution, dried over sodium sulfate, and
evaporated to dryness. The crude product was further purified by
column chromatography (CC) on silica (CH2Cl2/methanol = 99/1 →
95/5 + 1% NEt3) to give compound 6. Yield: 1.5 g (67%). TLC
2
= 7.2 Hz, −CH2OP); 3.80 (s, 6H, 2 OCH3); 3.75−3.60 (m, 1H, JHH
= 7.2 Hz, −CH2OP); 3.58 (m, 2H, 2 CH(iPr)); 3.45−3.35 (m, 2H,
CH5″ and CH5′); 2.64 (t, 1H, 2JHH = 6.4 Hz, −CH2CN); 2.40 (t, 1H,
2JHH = 6.4 Hz, −CH2CN); 1.33−1.12 (duplettoid, 9H, 3JHH = 6.7 Hz,
3
3 CH3(iPr)); 1.11−0.98 (d, 3H, JHH = 6.7 Hz, CH(iPr)); 0.95−0.77
(s, 9H, SiC(CH3)3); 0.02−0.2 (m, 6H, Si(CH3)2) ppm. 13C NMR (75
MHz, CDCl3, 25 °C): δ 140.43 (C6) ppm. 31P NMR (121 MHz,
CDCl3, 25 °C): δ 150.27, 150.74 ppm. ESI-MS (m/z): [M + H]+ calcd
for (13C1)C44H61N4O9PSi, 862.04; found, 862.40.
2.2.9. 5′-O-(4,4′-Dimethoxytrityl)-2′-O-[[(Triisopropylsilyl)oxy]-
methyl](6-13C)uridine (9). 6 (616 mg, 1.13 mmol) was dissolved
with DIPEA (765 μL, 4.5 mmol) in 10 mL of 1,2-dichloroethane
(DCE). To the clear solution was added dibutyltin dichloride
(Bu2SnCl2; 616 mg, 2.02 mmol), and the reaction mixture was stirred
for 1 h at room temperature. After the mixture was heated to 80 °C,
[(triisopropylsilyl)oxy]methyl chloride ((TOM)Cl; 276 mg, 1.24
mmol) was added. After 30 min the reaction mixture was cooled to
room temperature and diluted with methylene chloride. The organic
phase was then washed with saturated sodium bicarbonate solution
and water, dried over sodium sulfate, and filtered over Celite. The
organic phase was then evaporated to dryness. The residual yellow
foam was further purified by CC on silica (ethyl acetate/hexanes = 20/
80 → 60/40) to give compound 9. Yield: 240 mg (30%). TLC (ethyl
1
(CH2Cl2/MeOH = 9/1): Rf = 0.4. H NMR (300 MHz, CDCl3, 25
°C): δ 10.35 (s, NH); 8.02 (dd, 1H 1JCH = 183.94 Hz, 3JHH = 8.0 Hz,
C6H); 7.40−7.18 (m, 9 CH(ar)); 6.88−6.77 (m, 4 CH(ar)); 5.90
(triplettoid, 1H, CH1′); 5.49 (s, C2′OH); 5.35 (triplettoid, 1H, C5H);
4.43 (triplettoid, 1H, CH3′); 4.35 (triplettoid, 1H, CH2′); 4.17 (dd,
1H, CH4′); 3.54 (dd, 1H, CH5″); 3.77 (s, 6H, 2 OCH3); 3.48 (dd, 1H,
CH5′); 3.40 (duplettoid, 1H, C3′OH) ppm. 13C NMR (75 MHz,
CDCl3, 25 °C): δ 167.50 (C4); 158.82 (C(ar)); 151.30 (C2); 140.62
(C6); 135.45 (C(ar)); 130.28, 128.15, 113.44 (C(ar)); 106.82 (C5);
90.14 (C1′); 87.16 (C4′); 75.70 (C2′); 69.40 (C3′); 62.10 (C5′); 55.37
(OCH3) ppm. ESI-MS (m/z): [M + H]+ calcd for (13C1)C29H30N2O8,
547.56; found, 547.29.
1
acetate/hexanes = 5/5): Rf = 0.6. H NMR (300 MHz, CDCl3, 25
1
3
°C): δ 7.93 (dd, 1H JCH = 182.44 Hz, JHH = 8.2 Hz, C6H); 7.41−
2.2.7. 5′-O-(4,4′-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-
(6-13C)uridine (7). Compound 6 (1.5 g, 2.74 mmol) was dissolved
in anhydrous tetrahydrofuran (THF; 12 mL). Silver nitrate (AgNO3;
695 mg, 4.11 mmol) and pyridine (664 μL, 8.22 mmol) were added,
and the suspension was stirred for 1/2 h in the dark. Then tert-
butyldimethylsilyl chloride ((TBDMS)Cl; 620 mg, 4.11 mmol) was
added, and the mixture was stirred for 4 h at room temperature in the
dark. The suspension was filtered over Celite, and the filtrate was
evaporated to dryness. The residue was dried in high vacuum and then
dissolved in methylene chloride. The organic phase was washed with
saturated sodium bicarbonate solution, dried over sodium sulfate, and
evaporated to dryness. The crude product was further purified by CC
on silica (CH2Cl2/ethyl acetate = 100/0 → 80/20) to give compound
7. Yield: 1.0 g (55%). TLC (CH2Cl2/ethyl acetate = 6/4): Rf = 0.71.
1H NMR (300 MHz, DMSO-d6, 25 °C): δ 8.56 (s, NH); 7.75 (dd, 1H,
7.13, 6.89 (m, 13 CH(ar)); 6.02 (triplettoid, 1H, CH1′); 5.28 (m, 1H,
C5H); 5.05 (d, 1H, C3′OH); 5.02−4.97 (dd, 2H, JHH = 4.99 Hz,
2
−OCH2OSi); 4.26 (triplettoid, 1H, CH2′); 4.16 (dd, 1H, CH3′); 3.95
(dd, 1H, CH4′); 3.74 (s, 6H, 2 −OCH3); 3.25 (dd, 1H, CH5″); 3.20
(dd, 1H, CH5′); 1.20−1.17 (m, 3H, SiCH(CH3)2); 1.01−0.95 (m,
18H, SiCH(CH3)2) ppm. 13C NMR (75 MHz, CDCl3, 25 °C): δ
162.28 (C4); 158.30 (C(ar)); 149.56 (C2); 143.90 (C(ar)); 139.73
(C6); 134.7 (C(ar)); 129.71 (C(ar)); 127.59 (C(ar)); 126.75 (C(ar));
112.89 (C(ar)); 102.20 (C5); 101.31 (C1′); 90.24 (CAr3); 87.41
(O2′CH2O); 86.73 (C4′); 83.28 (C2′); 69.00 (C3′); 61.73 (C5′); 54.84
(2 −OCH3); 17.38 (SiCH(CH3)2); 11.45 (SiCH(CH3)2) ppm.
We also isolated the 3′-regioisomer 5′-O-DMT-3′-O-TOM-(6-13C)-
uridine for repeated tomylation reactions. Yield: 231 mg (28%). TLC
(ethyl acetate/hexanes = 5/5): Rf = 0.4. 1H NMR (300 MHz, CDCl3,
1
3
25 °C): δ 8.13 (s, 1H, NH); 7.76 (dd, 1H, JCH = 181.34 Hz, JHH
=
3
1JCH = 181.33 Hz, JHH = 7.8 Hz, C6H); 5.30 (triplettoid, 1H, C5H);
8.11 Hz, C6H); 7.41−7.18 (m, 9H, CH(ar)); 6.87−6.77 (m, 4H,
CH(ar)); 5.95 (triplettoid, 1H, CH1′); 5.39 (m, 1H, C5H); 5.07−4.89
(dd, 2H, 2JHH = 4.91 Hz, −OCH2OSi); 4.32 (m, 1H, CH3′); 4.27 (m,
1H, CH2′); 3.79 (s, 6H, −OCH3); 3.56 (dd, 1H, CH5″); 3.41 (dd, 1H,
CH5′); 3.36 (dd, 1H, CH4′); 1.25 (t, 3H, SiCH(CH3)2); 1.06 (m,
18H, SiCH-(CH3)2) ppm.
7.41−7.15 (m, 9 CH(ar)), 6.95−6.86 (m, 4 CH(ar)); 5.76 (triplettoid,
1H, CH1′); 5.11 (duplettoid, 1H, C3′OH); 4.21 (triplettoid, 1H,
CH2′); 4.05 (triplettoid, 1H, CH3′); 4.00 (m, 1H, CH4′); 3.4−3.2 (m,
2H, CH5″ and CH5′); 3.74 (s, 6H, 2 OCH3); 0.86 (s, 9H,
SiC(CH3)3); 0.06 (s, 6H, Si(CH3)2) ppm. 13C NMR (75 Hz,
DMSO-d6, 25 °C): δ 162.92 (C4); 158.82 (C(ar)); 150.46 (C2);
144.65 (C(ar)); 150.05 (C6); 135.20, 127.70, 126.81, 113.26 (C(ar));
101.43 (d, 1JCC = 67.4 Hz, C5); 88.53 (C1′); 85.93 (C4′); 75.52 (C2′);
69.36 (C3′); 62.71 (C5′) 55.05 (OCH3); 25.62 (SiC(CH3)3); 17.82
(SiC(CH3)3); −5.07 (Si(CH3)2) ppm. ESI-MS (m/z): [M + H]+ calcd
for (13C1)C35H44N2O8Si, 661.82; found, 661.42.
The 3′-regioisomer 5′-O-DMT-3′-O-TOM-(6-13C)uridine was iso-
lated and treated with 2 mL of 1 M tetrabutylammonium fluoride
(TBAF) in acetonitrile. After 3 h TLC indicated complete removal of
the silyl protection group. The solvent was evaporated and the residual
oil dried in high vacuum. The oil was dissolved in methylene chloride
and the organic phase washed with saturated sodium bicarbonate
solution and water, dried over sodium sulfate, and evaporated to
dryness. Crude compound 6 was purified by column chromatography
on silica (CH2Cl2/methanol = 99/1 → 95/5 + 1% NEt3, analytical
data identical to those of compound 6) and was again used for the
tomylation reaction.
2.2.8. 5′-O-(4,4′-Dimethoxytrityl)-2′-O-(tert-butyldimethylsilyl)-
(6-13C)uridine 3′-O-(2′-Cyanoethyl N,N-diisopropylphosphorami-
dite) (8). Anhydrous tetrahydrofuran (4 mL) and N,N-diisopropyle-
thylamine (DIPEA; 523 μL, 3.0 mmol) were mixed. Then compound
7 (400 mg, 0.6 mmol) was added, and the clear solution was stirred for
15 min. Then 2′-cyanoethyl N,N-diisopropylchlorophosphoramidite
((CEP)Cl; 343 mg, 1.2 mmol) was added dropwise. The reaction was
monitored by TLC. After 2 h the reaction mixture was diluted with
methylene chloride and washed with half-saturated sodium bicar-
bonate solution. The organic phase was dried over sodium sulfate and
evaporated to dryness. The crude product was further purified by CC
on silica (ethyl acetate/hexanes = 50/50 + 1% NEt3) to give
compound 8 as a mixture of diastereoisomers. Yield: 422 mg (83%).
2.2.10. 5′-O-(4,4′-Dimethoxytrityl)-2′-O-[[(triisopropylsilyl)oxy]-
methyl](6-13C)uridine 3′-O-(2′-Cyanoethyl N,N-diisopropylphos-
phoramidite) (10). Anhydrous methylene chloride (5 mL) and N,N-
dimethylethylamine (DMEA; 540 μL, 5.0 mmol) were mixed. Then
compound 9 (720 mg, 0.98 mmol) was added, and the clear solution
was stirred for 15 min. Then (CEP)Cl (350 mg, 1.5 mmol) was added
dropwise. The reaction was monitored by TLC. After 2 h the reaction
mixture was diluted with methylene chloride and washed with half-
saturated sodium bicarbonate solution. The organic phase was dried
over sodium sulfate and evaporated to dryness. The crude product was
1
TLC (ethyl acetate/hexanes = 1/1): Rf = 0.48. H NMR (300 MHz,
1
3
CDCl3, 25 °C): δ 7.99 (dd, 1H JCH = 181.73 Hz, JHH = 8.13 Hz,
7560
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
further purified by CC on silica (ethyl acetate/hexanes = 50/50 → 60/
40 + 1% NEt3) to give compound 10 as a mixture of diastereoisomers.
Yield: 784 mg (86%). TLC (ethyl acetate/hexanes = 1/1): Rf = 0.55.
1H NMR (500 MHz, CDCl3, 25 °C): δ 8.02 (s, 1H, NH); 7.86 (dd,
2.3.3. N4 -Acetyl-5′-O-(4,4′-dimethoxytrityl)-2′-O-
[[(triisopropylsilyl)oxy]methyl](6-13C)cytidine (13). Compound 12
(190 mg, 0.26 mmol) was dissolved in 1 mL of anhydrous
dimethylformamide (DMF) followed by the addition of Ac2O (25
μL, 0.26 mmol). The reaction was stirred for 22 h at room
temperature. Then a few drops of methanol were added. The solvent
was evaporated and the oily residue dissolved in methlyene chloride.
The organic phase was washed with saturated sodium bicarbonate
solution, dried over sodium sulfate, and evaporated to dryness. The
crude product was further purified by CC on silica (CH2Cl2/methanol
= 99/1 → 98/2) to give compound 13. Yield: 180 mg (89%). TLC
(ethyl acetate/hexanes = 7/3): Rf = 0.38. 1H NMR (300 MHz, CDCl3,
1
3
1H, JCH = 181.46 Hz, JHH = 8.25 Hz, C6H); 5.36 (dd, 1H, C5H);
7.47−7.21, 6.86−6.80 (m, 13 CH(ar)); 6.12 (m, 1H, CH1′); 5.35 (m,
2H, 2JHH = 4.84 Hz, −OCH2OSi); 4.41 (m, 1H, CH2′); 4.46 (m, 1H,
CH3′); 4.27 (m, 1H, CH4′); 3.60 (dd, 1H, CH5″); 3.98−3.84 (m, 1H,
POCH2); 3.71−3.35 (m, 3 × 1H, POCH2, CH5′, NCH(CH3)2); 3.79
2
(s, 6H, −OCH3); 2.64 (m, 1H, −CH2CN); 2.39 (m, 1H, JHH = 6.35
Hz, −CH2CN); 1.58 (s, 3H, −SiCH(CH3)2); 1.23−1.10 (m, 12H,
−NCH(CH3)2); 1.10−0.99 (m, 18H, SiCH(CH3)2) ppm. 31P NMR
(201 MHz, CDCl3, 25 °C): δ 150.53, 150.11 ppm. ESI-MS (m/z): [M
+ H]+ calcd for (13C1)C48H69N4O10PSi, 934.14; found, 934.37.
2.3. Synthesis of the (6-13C)Cytidine TOM-Protected
Phosphoramidite 14. 2.3.1. 3′-O-Acetyl-5′-O-(4,4′-dimethoxytri-
tyl)-2′-O-[[(triisopropylsilyl)oxy]methyl](6-13C)uridine (11). Com-
pound 9 (350 mg, 0.47 mmol) was dissolved in 700 μL of
anhydrous pyridine. Then 4-(dimethylamino)pyridine (DMAP;
5.7 mg, 0.05 mmol) was added. After the mixture was cooled to
0 °C, acetic anhydride (Ac2O; 48 μL, 0.51 mmol) was added.
The ice bath was removed, and the reaction mixture was stirred
for 2 h at room temperature. The solvent was evaporated and
the residue twice coevaporated with toluene. The residue was
then dissolved in methylene chloride and washed with 5%
citiric acid, water, and saturated sodium bicarbonate solution.
The organic phase was dried over sodium sulfate and then
evaporated. The foam was dried in high vacuum and then
purified by CC on silica (ethyl acetate/hexanes = 50/50) to
give compound 11. Yield: 300 mg (82%). TLC (ethyl acetate/
hexanes = 1/1): Rf = 0.7.
1
3
25 °C): δ 8.58 (s, NH); 8.48 (dd, 1H, JCH = 183.67 Hz, JHH = 7.33
Hz, C6H); 7.47−7.20 (m, 13 CH(ar)); 7.04 (dd, 1H, C5H); 6.02
1
(triplettoid, 1H, CH1′); 5.21 (dd, 2H, JHH = 4.7 Hz, −OCH2OSi);
4.22 (m, 1H, CH2′); 4.37 (m, 1H, CH3′); 4.09 (dd, 1H, CH4′); 3.81
(s, 6H, 2 −OCH3); 3.60 (dd, 1H, CH5″); 3.52 (dd, 1H, CH5′); 3.33
(d, 1H, C3′OH); 2.21 (s, 3H, −CONHCH3); 1.61 (s, 3H, CH(iPr));
1.30−0.98 (m, 18H, 3 CH3(iPr)) ppm. 13C NMR (75 MHz, CDCl3,
25 °C): δ 169.77 (HNCOCH3); 162.68 (C4); 158.83 (C(ar)); 155.09
(C2); 145.05 (C6); 140.27 (C(ar)); 138.28 (C(ar)); 130.31 (C(ar));
128.31 (C(ar)); 127.25 (C(ar)); 113.44 (C(ar)); 98.65 (C5); 90.25
(C1′); 90.29 (O2′CH2O); 87.31 (C(quart)); 83.51 (C4′); 83.29 (C2′);
68.21 (C3′); 61.30 (C5′); 55.38 (−OCH3); 25.11 (NHCOCH3); 17.91
(CH3CHSi); 11.99 (CH3CHSi) ppm.
2.3.4. N4 -Acetyl-5′-O-(4,4′-dimethoxytrityl)-2′-O-
[[(triisopropylsilyl)oxy]methyl](6-13C)cytidine 3′-(2′-Cyanoethyl
N,N-diisopropylphosphoramidite) (14). Anhydrous methylene chlor-
ide (5 mL) and DMEA (250 μL, 2.3 mmol) were mixed. Then
compound 13 (130 mg, 0.23 mmol) was added, and the clear solution
was stirred for 15 min. Then (CEP)Cl (83 mg, 0.35 mmol) was added
dropwise. The reaction progress was monitored by TLC. After 2 h the
reaction mixture was diluted with methylene chloride and washed with
half-saturated sodium bicarbonate solution. The organic phase was
dried over sodium sulfate and evaporated to dryness. The crude
product was further purified by CC on silica (ethyl acetate/hexanes =
70/30 + 1% NEt3) to give compound 14 as a mixture of
diastereoisomers. Yield: 205 mg (91%). TLC (ethyl acetate/hexanes
= 7/3): Rf = 0.46. 1H NMR (300 MHz, CDCl3, 25 °C): δ 8.59 (s, 1H,
Compound 11 was used without further characterization in the next
steps.
2.3.2. 5′-O-(4,4′-Dimethoxytrityl)-2′-O-[[(triisopropylsilyl)oxy]-
methyl](6-13C)cytidine (12). Step a: Anhydrous methylene chloride
(4 mL) was mixed with triethylamine (Et3N; 528 μL, 3.8 mmol). Then
compound 11 (300 mg, 0.38 mmol) and DMAP (6.0 mg, 0.05 mmol)
were added. After 10 min, 2,4,6-triisopropylbenzenesulfonyl chloride
((TPS)Cl; 176 mg, 0.58 mmol) was added in small portions, and the
reaction was stirred for 4 h at room temperature. Then the reaction
mixture was diluted with methylene chloride, and the organic phase
was washed with half-saturated sodium bicarbonate solution, dried
over sodium sulfate, and evaporated to dryness. The resulting foam
was dried in high vacuum.
1
3
NH); 8.47 (dd, 1H, JCH = 182.48 Hz, JHH = 7.41 Hz, C6H); 7.47−
7.19 (m, 13 CH(ar)); 6.98 (dd, 1H, C5H); 6.15 (triplettoid, 1H,
CH1′); 5.20 (m, 2H, JHH = 4.7 Hz, −OCH2OSi); 4.51 (m, 1H,
1
CH3′); 4.36 (m, 1H, CH2′); 4.33 (m, 1H, CH4′); 3.98−3.84 (m, 1H,
POCH2); 3.81 (s, 6H, 2 −OCH3); 3.69 (dd, 1H, CH5″); 3.73−3.40
(m, 1H, POCH2, CH5′, NCHMe2); 2.60 (m, 1H, −CH2CN); 2.40 (t,
1H, 2JHH = 6.4 Hz, −CH2CN); 2.19 (s, 3H, CONHCH3); 1.61 (s, 3H,
CH(iPr)(TOM)); 1.17 (m, 1H, NCHMe2(CEP)); 1.21−1.10 (m,
12H, NCH(CH3)2); 1.10−0.93 (m, 18H, SiCH(CH3)2(TOM)) ppm.
31P NMR (121 MHz, CDCl3, 25 °C): δ 150.84, 150.58 ppm. ESI-MS
Step b: The crude product of the previous step was dissolved in 4
mL of tetrahydrofuran and then treated with 4 mL of 28% aqueous
ammonia. The mixture was stirred for 18 h at room temperature.
Step c: After 18 h, 4 mL of methylamine solution in ethanol (8 M)
was added to the yellow suspension, resulting in a clear solution. The
reaction was continued for 45 min until TLC showed the removal of
the 3′-O-acetyl protection group. The solvent was then evaporated and
the residual oil dried in high vacuum. The oil was dissolved in
methylene chloride, washed with saturated sodium bicarbonate
solution, and dried over sodium sulfate and the solvent evaporated.
The resulting foam was purified by column chromatography on silica
(CH2Cl2/methanol = 99/1 → 95/5) to yield compound 12. Yield: 190
mg (68% referred to compound 11). TLC (CH2Cl2/methanol = 96/
4): Rf = 0.45. 1H NMR (300 MHz, DMSO-d6, 25 °C): δ 7.68 (dd, 1H
1JCH = 178.59 Hz, 3JHH = 7.30 Hz, C6H); 7.41−7.20 (m, 13 CH(ar));
(m/z): [M + H]+ calcd for (13C1)C50H72N5O10PSi, 975.20; found,
975.41.
2.4. Chemical Synthesis of RNA Oligonucleotides on a Solid
Support. The 13C-modified phosphoramidites were used in
combination with standard 2′-O-TOM-protected building blocks
(ChemGenes) to synthesize RNA sequences 15, 16, 17, and 18.
Custom primer support PS 200 (GE Healthcare) with an average
loading of 80 μmol g−1 was used. The sequences were synthesized on
an Applied Biosystems 391 PCR Mate using self-written RNA
synthesis cycles. Amidite (0.1 M) and activator (5-(benzylthio)-1H-
tetrazole, 0.25 M) solutions were dried over freshly activated
molecular sieves overnight.
2
5.94 (triplettoid, 1H, CH1′); 5.54 (dd, 1H, C5H); 5.12 (dd, 2H, JHH
= 4.7 Hz, −OCH2OSi); 4.30 (dd, 1H, CH3′); 4.18 (m, 1H, CH2′);
4.10 (dd, 1H, CH4′); 3.79 (s, 6H, −OCH3); 3.51 (m, 2 × 1H, CH5″
and CH5′); 3.15 (d, 1H, C3′OH); 1.57 (s, 3H, SiCH(CH3)2); 1.30−
1.10 (m, 18H, SiCH(CH3)2) ppm. 13C NMR (75 MHz, DMSO-d6, 25
°C): δ 164.43 (C4); 158.30 (C(ar)); 151.73 (C2); 144.95 (C(ar));
143.83 (C6); 132.64 (C(ar)); 130.05 (C(ar)); 127.5 (C(ar)); 116.13
(C(ar)); 98.65 (C5); 94.8 (C(quart)); 90.40 (C1′); 87.41 (O2′CH2O);
71.66 (C4′); 80.90 (C2′); 71.66 (C3′); 63.68 (C5′); 57.75 (−OCH3);
19.94 (CHSi); 14.45 (CH3CHSi) ppm.
The removal of protecting groups and the cleavage from the solid
support were achieved by treatment with aqueous methylamine (40%,
650 μL) and ethanolic methylamine (8 M, 650 μL) at room
temperature for 6−8 h. After evaporation of the alkaline deprotection
solution, the 2′-O-protecting groups were removed by adding 1 M
TBAF in THF (1200 μL). After 16 h at 37 °C, the reaction was
quenched by the addition of 1 M triethylammonium acetate (TEAA;
pH 7.0, 1200 μL). The volume was reduced to approximately 1 mL,
and then the solution was applied to a HiPrep 26/10 desalting column
7561
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
(GE Healthcare). The crude RNAs were eluted with water, evaporated
to dryness, and dissolved in 1 mL of water.
nmrPipe package. These intensities were used to calculate effective
transverse relaxation rates R2,eff according to
The quality of the crude RNAs was checked via anion exchange
chromatography on a Dionex DNAPac PA-100 column (4 × 250 mm)
using our standard eluents and at elevated temperature (80 °C).24
Purification of the RNA sequences was achieved by applying the crude
RNA on a semipreparative Dionex DNAPac PA-100 column (9 × 250
mm). The fractions containing the desired RNA were pooled and
loaded onto a C18 SepPak cartridge (Waters) to remove HPLC buffer
salts. The RNA triethylammonium salt form was then eluted from the
C18 column with water/acetonitrile (1/1, v/v) and lyophilized. To
obtain the RNA sodium salt form, the RNAs were treated with a
sodium chloride solution (250 mM, 2 × 10 mL) and water (3 × 10
mL) in a Vivaspin 20 ultracentrifugation unit with a molecular weight
cutoff of 3000 (Satorius Stedim). The integrity of the RNAs was
further checked by mass spectrometry on a Finnigan LCQ Advantage
MAX ion trap instrument connected to an Amersham Ettan micro-LC
instrument (GE Healthcare).
I(νCPMG
)
1
τCP
R2,eff = −
ln
I0
(1)
with τCP the relaxation delay, I(νCPMG) the peak intensity at the
CPMG frequency νCPMG, and I0 the peak intensity from the reference
experiment. The transverse relaxation rates were used as the
experimental input for fitting the dispersion profiles. The data were
fit according to a general expression for the intermediate exchange
regime (Carver−Richards equation32):
kex
−1
R2,eff = R2,0
with
D =
+
− νCPMG cosh [D cosh(η ) − D− cosh(η )]
+
+
−
2
(2)
Ψ + 2δω2
⎡
⎢
⎣
⎤
⎥
⎦
1
2
2.5. NMR Spectroscopy. RNA samples were lyophilized as the
sodium salts and dissolved in the appropriate buffer in 9/1 H2O/D2O
or pure D2O. The HIV-1 transactivation response element (TAR)
RNAs 15 and 16 were dissolved in 15 mM sodium phosphate buffer
(pH 6.4) and 25 mM sodium chloride. The 6-13C−1H pyrimidine
resonance assignments for this RNA are reported in the literature.7
The Varkud satellite stem loop V motif 17 was dissolved in 10 mM
Tris·HCl buffer (pH 6.5) and 25 mM NaCl. The 6-13C−1H pyrimidine
resonances of this RNA were previously assigned.27 The bistable RNA
18 involved in the gene regulation process by the preQ1 riboswitch
was dissolved in 25 mM sodium arsenate buffer, pH 6.5.
1 +
(Ψ2 + ξ2)1/2
[ Ψ + (Ψ + ξ2)1/2 1/2
]
η =
2 2 νCPMG
Ψ = kex − δω2
ξ = −2δω(p kex − pbkex)
2
a
where δω is the chemical shift difference between the ground and
excited states, kex (= kforward + kbackward) is the exchange rate, pa is the
population of the ground state, and pb is the population of the excited
state. Errors for the exchange parameters were obtained from 1000
runs of Monte Carlo analysis. Outliers were identified using
Chauvernet’s criterion at a significance level of 96% and omitted
from further analysis.33 Mean values and standard deviations of fitted
parameters are reported.
NMR experiments on 6-13C-modifed RNA sequences were
conducted either on a Bruker 600 MHz Avance II+ instrument or
on Varian Unity instruments operating at 800 or 500 MHz proton
1
Larmor frequency. The H 1D NMR spectra of H2O samples were
acquired using a double-pulsed field gradient spin−echo pulse
sequence.28 The 2D 1H−13C correlation spectra were acquired via
double INEPT (insensitive nuclei enhanced by polarization transfer)
using sensitivity improvement in phase-sensitive mode, echo/antiecho
TPPI (time proportional phase incrementation) gradient selection
with decoupling during acquisition, trim pulses in INEPT transfer, and
shaped pulses for all 180° pulses on the carbon channel and with
gradients in back-INEPT (Bruker pulse program hsqcetgpsisp2.2).
The 2D heteronuclear correlation spectra were processed using
nmrPipe and visualized using nmrDraw.29
2.6.2. Processing and Analysis of Longitudinal Exchange Data.
Peak intensities at the various mixing periods were obtained by
summing over 3 × 3 (t1 × t2) data points as above. Then a least-
squares fitting procedure was applied to determine kAB, kBA, RA1 and R1B
by fitting the following expressions:
1
1
2
MBB(t)/MBB(0) =
[(a11 + λ1)eλ t − (a11 + λ2)eλ t
]
λ1 − λ2
1
1
2
MAA(t)/MAA(0) =
[(a22 + λ1)eλ t − (a22 + λ2)eλ t
]
13C CPMG relaxation dispersion experiments on RNA sequences
15, 16, and 17 were recorded at 11.7 and 18.8 T using the pulse
sequence previously published with CPMG field strengths of 80, 80,
160, 240, 320, 400, 480, 560, 640, 640, 720, 800, 880, and 960 Hz.30
The relaxation delay τCP was set to 25 ms. A total of 1024 × 64 (1024
× 64) complex data points were recorded at 18.8 (11.7) T with the
spectral widths in the proton and carbon dimensions set to 12775 and
3000 Hz (8000 and 1875 Hz), respectively. The number of transients
was set to 40 at 18.8 and 11.7 T, and the interscan delay was set to 1.3
or 1.8 s, yielding total experimental times of 32 or 44 h.
Longitudinal exchange (ZZ) and longitudinal relaxation (T1)
experiments are based on pulse sequences recently published.31 For
a 1 mM sample of sequence 18, arrays of T1 and ZZ exchange spectra
were recorded at 11.7 T at 306 K with mixing periods of 10, 50, 100,
200, 300, 400, 600, and 800 ms. The size of the data matrices for each
spectrum was 5000 × 1500 complex data points, the number of scans
was 64, and the interscan delay was 1.2 s to yield a total measuring
time of 22 h.
λ1 − λ2
(3)
1
MA(t)/MA(0) =
MB(t)/MB(0) =
[(a22 − a21 + λ1)eλ t
1
λ1 − λ2
− (a22 − a21 + λ2)eλ t
]
2
1
[(a11 − a12 + λ1)eλ t
1
λ1 − λ2
− (a11 − a12 + λ2)eλ t
]
2
(4)
MAA and MBB are the intensities of the correlation peaks from the ZZ
exchange experiment. These intensities are normalized by the peak
volumes at zero mixing time (MAA(0)and MBB(0)). MA(0)/MB(0) and
MA(t)/MB(t) are the peak intensities of folds A and B at zero mixing
time or mixing time t from the T1 experiment. The eigenvalues of the
spin density matrix with magnetization transfer effects from chemical
exchange are given by
2.6. NMR Data Analysis. Spectral processing and peak picking
and integration were performed using nmrPipe and the nmrDraw
software package.29 All subsequent steps were performed using in-
house-written software written in Matlab (The MathWorks, www.
mathworks.com).
2.6.1. Processing and Analysis of CPMG RD Data. Peak intensities
at the various CPMG fields were obtained by summing over 3 × 3 or 1
× 1 (t1 × t2) data points using the serT script implemented in the
1
2
λ1/2
=
[−(a11 + a22) [(a11 − a22)2 + 4a12a21]1/2
]
with the elements aij defined as
a11 = kAB + R1A
a21 = −kAB
a22 = kBA + R1B
a12 = −kBA
7562
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
(1) via a Kolbe nitrile synthesis using 2-bromoacetic acid.36
Then [(3-13C)cyanoacetyl]urea (2) was obtained by treating
compound 1 with urea in the presence of acetic anhydride. In
the following step, key compound (6-13C)uracil (3) was
obtained by a reduction of the nitrile and a subsequent ring
closure reaction. The reduction was achieved by the use of a
palladium catalyst (5% palladium on barium sulfate) under a
hydrogen atmosphere. 3 was then coupled to the commercially
The parameters kAB and kBA describe the microscopic rate constants
describing the interconversion between the two folding states of RNA
18, while R1A and R1B are the longitudinal relaxation rates in
conformations A and B, respectively. Experimental uncertainties
were estimated from 1000 Monte Carlo runs in which synthetic data
sets were generated from the best fit values of kAB, kBA, RA1, and R1B by
adding random errors based on the signal-to-noise ratio to the best fit
curves. For details of the analysis please refer to ref 31.
3. RESULTS AND DISCUSSION
available ATBR under Vorbruggen conditions to give the
̈
benzoyl-protected uridine nucleoside 4.37 Removal of the
benzoyl residues to give (6-13C)uridine (5) and subsequent
tritylation gave 5′-O-DMT-(6-13C)uridine (6) (Scheme 1).
3.1. Choice of 13C Modification. Our goal was the
development of a native, site-specific 13C isotope labeling
protocol for the investigation of RNA dynamics by solution
NMR spectroscopy. We opted for (6-13C)pyrimidine as an
appropriate modification (Figure 1a).34 The main advantage of
Scheme 1. Synthesis of Intermediate 5′-O-DMT-
a
(6-13C)uridine
a
Reagents and conditions: (a) sodium carbonate, pH 9, 3 h at 80 °C,
20 h at rt, 92%; (b) urea in acetic anhydride, 30 min at 90 °C, 90%; (c)
5% Pd/BaSO4 in 50% aqueous acetic acid, 36 h at rt, 82%; (d) ATBR,
BSA, (TMS)OTf in acetonitrile, 1 h at 60 °C, 98%; (e) CH3NH2 in
ethanol, 16 h at rt, 90% (related to (6-13C)uracil used in step d); (f)
(DMT)Cl in pyridine, 4 h at rt, 67%. An asterisk indicates 13C.
In the next steps two alternative silyl protection groups for
the 2′-hydroxyl group were used. The TBDMS protection
group was introduced first, whereas the [(triisopropylsilyl)-
oxy]methyl group was established by Pitsch et al. more
recently.38 The latter is reported to be slightly superior in RNA
solid-phase synthesis as the lower steric demand gives higher
coupling yields, routinely higher than 98%. In our hands,
however, both the TOM- and the TBDMS-(6-13C)uridine
building blocks gave similar yields in the respective coupling
step by prolonging the coupling time for the TBDMS building
block. The TBDMS group was attached at the 2′-hydroxyl
group by treating compound 6 with tert-butyldimethylsilyl
chloride, silver nitrate, and pyridine in anhydrous tetrahy-
drofuran.39 Finally, the phosphoramidite building block 8 was
obtained by treatment with 2′-cyanoethyl N,N-diisopropyl-
chlorophosphoramidite (Scheme 2).
Figure 1. (a) (6-13C)Uridine (left) and (6-13C)cytidine (right)
modification. (b) HIV-1 TAR RNAs 15 and 16.35 (c) Varkud satellite
(VS) stem loop V RNA 17.27 (d) Bistable terminator antiterminator
RNA element 18 from the preQ1 riboswitch class I system.26 13C-
labeled residues are highlighted in orange.
1
this labeling protocol is that it represents an isolated H−13C
spin system, for which relaxation dispersion experiments can be
conducted in a straightforward manner. Technical issues,
normally encountered when relaxation dispersion methods
are applied to uniformly 13C/15N-labeled RNA, e.g.,
13C−13C/15N scalar couplings or dipolar couplings and other
complex relaxation pathways, are not present in the tailor-made
RNAs presented here (Figure 1b−d). Thus, we were able to
conduct NMR experiments for studying dynamics in the micro-
to millisecond time regime (and beyond) to give interesting
insights into functionally important dynamics of RNA.
3.2. Synthesis of 6-13C-Modified Uridine and Cytidine
Phosphoramidites 8, 10, and 14. The key compound in the
presented synthesis is the (6-13C)uracil precursor, which was
coupled to an appropriate sugar building block to give the
uridine nucleoside, the starting point for all subsequent
transformations.
Concurrently, we used the standard protocol published by
Pitsch et al. for the synthesis of the 2′-O-TOM-protected
(6-13C)uridine building block 10.38 First, 6 was treated with
(TOM)Cl in the presence of dibutyltin dichloride and DIPEA
to yield both the 2′-O-TOM- and 3′-O-TOM-protected
compounds.40 The latter was deprotected with tetrabutylam-
monium fluoride and again treated with (TOM)Cl. The 2′-O-
TOM-protected uridine derivative 9 was then transformed
under standard conditions to the phosphoramidite building
block 10 (Scheme 2).
Starting from 5′-O-DMT-2′-O-TOM-(6-13C)uridine (9), the
corresponding cytidine derivative 12 was obtained by a
multistep reaction according to published methods.41 The
We started our synthesis from 13C-modified potassium
cyanide, which was converted into (3-13C)cyanoacetic acid
7563
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
Scheme 2. Synthesis of (6-13C)Uridine TOM- and TBDMS-
Protected Phosphoramidites
was used to guarantee coupling yields greater than 98% as in
the case of the 2′-O-TOM-(6-13C)uridine and -cytidine
phosphoramidites. After successful assembly of the target
RNA, standard deprotection conditions using methylamine in
ethanol/water followed by treatment with 1 M tetrabutylam-
monium fluoride in tetrahydrofuran were used. After
purification by anion-exchange HPLC, the homogeneity and
the 13C isotope enrichment were confirmed for all RNAs by
LC−ESI mass spectrometry (Table 1; Supporting Figure 2,
a
Table 1. Sequence Information and Analytical Data for the
13C-Modified RNA Sequences 15, 16, 17, and 18
molecular weight
a
b
c
d
ID
no. of 13C labels
length
yield /nmol
calcd
found
15
16
17
18
7 (U)
8 (C)
4 (U)
27
27
17
21
355
510
750
745
8584.1
8585.1
5406.3
6705.1
8584.4
8585.6
5406.8
6705.1
a
Reagents and conditions: (a) (TBDMS)Cl, AgNO3, pyridine in THF,
1 (C)
4 h at rt, 55%; (b) (CEP)Cl, DIPEA in anhydrous THF, 2 h at rt, 83%;
(c) (TOM)Cl, Bu2SnCl2, DIPEA in 1,2-dichloroethane, 1.5 h at 80 °C,
30%; (d) (CEP)Cl, DMEA in anhydrous methylene chloride, 2 h at rt,
86%. An asterisk indicates 13C.
a
b
Sequence identifier. Nucleotide type C = (6-13C)cytidine and U =
c
d
(6-13C)uridine. Number of nucleotides. Yield on a 2 μmol synthesis
scale.
Supporting Information). We successfully incorporated up to
eight 13C-modified pyrimidine labels into the target RNA on a
synthesis scale of 2 μmol. All syntheses yielded sufficient
amounts of oligoribonucleotides for the NMR experiments
carried out later (on average 500 nmol, i.e., 1 mM, of sample in
500 μL of volume). This makes our approach competitive with
enzymatic methods for which similar yields are obtained.16
At this point we stress some advantages compared to the
standard enzymatic method using T7 RNA polymerase.44,45
The solid-phase synthesis of RNA does not have any demands
on the sequence composition, especially at the 5′-terminus.
Furthermore, labeled and unlabeled phosphoramidites of the
same nucleotide type can be used to give site-specifically 13C-
labeled RNAs on the nucleotide level. Thereby, the problem of
resonance overlap can be eliminated or at least reduced. This
can be of special interest in the dynamic characterization of a
nucleic acid as resonance overlap virtually impedes the data
analysis. The approach can in the future also be used to
introduce resonance assignment starting points by single 6-13C-
modified pyrimidine substitutions into the target RNA. In
combination with enzymatic isotope labeling, the resonance
assignment process for larger RNAs, such as naturally occurring
riboswitch aptamer domains comprising up to 50 nucleotides
(nt), will be simplified by this method. Currently, the upper
limit of the presented approach is set by RNAs comprising
more than 55 nucleotides; e.g., for a 52 nt riboswitch aptamer
domain, we reproducibly obtain about 100 nmol from a
synthesis on a 1.4 μmol scale (unpublished data). Combining
two such synthetic runs yields 0.8 mM samples in restricted
volume NMR tubes, a concentration range where biomolecular
NMR methods to study structure and dynamics can be applied
on a high-field spectrometer ideally equipped with cryogenic
probe technology.
uridine to cytidine transformation was followed by the acylation
of the excocylic amino functionality, yielding compound 13.
Finally, the (6-13C)cytidine phosphoramidite 14 was obtained
using the phosphitylation reaction conditions as above (Scheme
3). NMR spectra of the phosphoramidite building blocks are
included in the Supporting Information (Supporting Figure 1).
Scheme 3. Synthesis of (6-13C)Cytidine TOM-Protected
a
Phosphoramidite
a
Reagents and conditions: (a) Ac2O, DMAP in pyridine, 5 min at 0
°C, 2 h at rt, 82%; (b) NEt3, DMAP, (TPS)Cl in methylene chloride, 4
h at rt, then 28% aqueous NH3 in THF, 18 h at rt, then CH3NH2 in
ethanol, 45 min at rt, 68%; (c) Ac2O in anhydrous DMF, 22 h at rt,
89%; (d) (CEP)Cl, DMEA in anhydrous methylene chloride, 2 h at rt,
91%. An asterisk indicates 13C.
3.3. Chemical Synthesis of RNA Sequences 15, 16, 17,
and 18 Bearing 6-13C-Modified Nucleotides. The
oligoribonucleotides 15, 16, 17, and 18 comprising up to 27
nucleotides were synthesized using RNA solid-phase synthesis
(Figure 1b−d).7,26,27,35,42,43 Both the 2′-O-TOM/TBMDS-
(6-13C)uridine and the 2′-O-TOM-(6-13C)cytidine building
blocks proved to be suitable for the strand assembly procedure
in combination with unlabeled 2′-O-TOM-protected nucleoside
phosphoramidite,s yielding tailor-made site-specifically 13C-
labeled RNAs. Noteworthy, for the solid-phase synthesis cycle
using the 2′-O-TBDMS-(6-13C)uridine building block, a
prolonged coupling time (4 min vs 2 min, TBDMS vs TOM)
3.4. 13C CPMG Relaxation Dispersion Experiments To
Probe Micro- to Millisecond Dynamics. 3.4.1. HIV-1 TAR
RNA. We used CPMG relaxation dispersion experiments to
probe pyrimidine nucleobase dynamics of the TAR from HIV-1
in the millisecond time regime. The HIV-1 TAR RNA is a well-
known paradigm that nucleic acids can adaptively change their
conformation and bind diverse ligands.7,46 It was recently
7564
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
shown that the 27 nt RNA construct (15 and 16) utilizes a
conformational selection mechanism to interact with its binding
partners, such as argininamide.7 These data suggest that the free
HIV-1 TAR RNA cycles through multiple conformations
resembling ligand-bound states with lifetimes of nano- to
picoseconds. For this RNA, we replaced all pyrimidine
nucleotides with the corresponding 6-13C counterpart as
discussed in the previous section, resulting in constructs 15
1
and 16. Via H−13C HSQC spectra, the expected labeling
pattern was confirmed (Figure 2). With these tailor-made
Figure 3. (a) 13C CPMG relaxation dispersion profiles of HIV-1 TAR
RNA 15 at 125 MHz carbon Larmor frequency. Dots represent
experimental data. Residue U23 shows a nonflat dispersion profile.
R2,eff values at νCPMG = 960 Hz were subtracted from the data for
display purposes. (b) CPMG dispersion profiles of U23 at 125 (red)
and 200 (blue) MHz carbon Larmor frequency, which were used for a
quantitative description of the excited state. Dots represent the
experimental data. Black times signs represent repeat experiments. The
dashed blue and red lines represent the best fits for a two-state
equilibrium according to eq 2 in the intermediate exchange regime.
1
Figure 2. (a) H−13C HSQC spectra of the (6-13C)uridine-labeled
reasonably good agreement with the relaxation dispersion data
for U23 (Δδfree‑ARG = 3.3 ppm and ΔδRD = 2.4 0.7 ppm).
This could point toward a transient formation of the base triple
among U38, A27, and U23 found in the excited state. We also
conducted CPMG relaxation dispersion experiments for the
free HIV-1 TAR RNA 15 at lower (17 °C) and elevated (33
°C) temperatures. In both cases the exchange process for
residue U23 was quenched, resulting in flat dispersion profiles
(Supporting Figure 5, Supporting Information). We then
repeated the CPMG RD NMR experiments in the presence of
the ligand argininamide (1/2 and 1 equiv with respect to the
RNA). With a reported dissociation constant for the HIV-1
TAR RNA−argininamide complex of approximately 1 mM, the
CPMG RD technique is expected to be the appropriate
experiment to measure the dynamics (kon and koff rates) of the
ligand binding process. Flat relaxation dispersion profiles for
the majority of the peaks were observed, indicating that the
exchange process either occurs on a time scale not detectable
with the CPMG RD experiment or does not involve sufficiently
large chemical shift differences (Supporting Figure 6,
Supporting Information). Residue U23, however, did show a
scattered (and potentially nonflat) relaxation dispersion profile
in the presence of argininamide, along with higher than average
effective transverse relaxation rates, most due to line broad-
ening arising from the chemical exchange process beween the
free and the ligand-bound conformations. However, the
reduced signal-to-noise ratio of U23 and the concomitant
scattering of the experimental data especially at low νCMPG
values impeded a quantitative analysis of the U23 CPMG RD
data.
HIV-1 TAR RNA 15 and (b) the (6-13C)cytidine-labeled HIV-1 TAR
RNA 16. Assignments were reported previously.7 Conditions: 0.7 (15)
and 1 (16) mM RNA, 15 mM sodium phosphate, pH 6.4, 25 mM
sodium chloride, H2O/D2O = 9/1, 25 C°.
RNAs, we were able to probe the HIV-1 TAR system for
functional dynamics at micro- to milliseconds using 13C CPMG
relaxation dispersion experiments.
Briefly, we could identify only one residue with a nonflat
relaxation dispersion profile, namely, uridine 23 (U23; Figure
3a). All the other pyrimidine nucleobases did show flat
dispersion profiles in the CPMG RD experiments at 125
and/or 200 MHz carbon Larmor frequency (Figure 3a;
Supporting Figure 3, Supporting Information). For the residue
U23, data were acquired at two magnetic field strengths (11.7
and 18.8 T), and fitting the data to a two-site exchange process
in the intermediate NMR chemical shift time scale according to
eq 2 resulted in an excited-state population of 1.0% 0.6%, an
exchange rate kex (= kforward + kbackward) of 2200 1510 s−1, and
a chemical shift difference ΔδRD between the ground and
excited states of 2.4 0.7 ppm.
This two-state equilibrium fits qualitatively to the previously
proposed conformational selection mechanism as U23 under-
goes a rigorous structural rearrangement on the addition of the
ligand argininamide. In detail, a base triple, U38−A27−U23, is
formed on the addition of the ligand, resulting in a drastic
chemical shift change for the 6-13C atom of U23 (Δδfree‑ARG
=
3.3 ppm; Supporting Figure 4, Supporting Information).35,42
The carbon shift difference induced by the binding partner is in
7565
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
3.4.2. Varkud Satellite Stem Loop V Motif. To further
demonstrate the applicability of the (6-13C)pyrimidine labels in
relaxation dispersion experiments, we extended our study on
the 17 nt Varkud satellite (VS) stem loop V segment 17
(Figure 4a).27,47 The solution structure of this RNA was solved
Figure 4. (a) Secondary structure representation of the VS stem loop
V motif 17. Uridine residues are color coded as in (b) and (c). (b)
Solution structure bundle of 17 with uridines highlighted in red and
orange (PDB ID 1TBK).27 (c) 1H−13C HSQC spectrum of the
(6-13C)uridine-labeled VS stem loop V motif 17. Assignments were
reported previously.27 Conditions: 1 mM RNA, 10 mM Tris·HCl, pH
6.5, 25 mM sodium chloride, H2O/D2O = 9/1, 15 C°.
previously (Figure 4b). The observed 1H−13C HSQC spectrum
of (6-13C)uridine-modified sequence 17 is in good accordance
with that reported (Figure 4c).27 We conducted relaxation
dispersion experiments at various temperatures. At 25 °C
residue U696 exhibited an exchange process, which was too fast
to be quantified by the CPMG relaxation dispersion technique.
This was manifested by a very strong dependence of the
effective transverse relaxation rate (R2,eff) on the CPMG pulse
frequency (νCPMG), virtually impairing a reliable analysis of the
exchange data. All other residues showed flat or near-flat
(U695) dispersion profiles (Supporting Figure 7, Supporting
Information).
We then lowered the temperature to 15 °C and repeated the
RD experiments. The temperature change led to observable
nonflat dispersion profiles for the loop-closing residue U695
and the loop residues U696 and U700 (Figure 5a). The data
could be fitted to a collective motion of these residues for a
two-site exchange in the intermediate exchange regime
according to the Carver−Richards equation (eq 2) (Figure
5b). We found a population of the excited state of 3.8% 0.6%,
an exchange rate kex (= kforward + kbackward) of 3080 840 Hz,
and chemical shift differences between the ground and excited
states of 0.7 0.4 ppm (U695), 3.1 0.5 ppm (U696), and 1.0
0.2 ppm (U700). Residue U703 did not show any detectable
micro- to millisecond dynamics as expected for a nucleotide
locked up in a stable Watson−Crick base pair. The observed
dynamics in the millisecond time regime of residues U695 and
U696 and especially U700 could be crucial for the function of
the ribozyme, as the stem loop V motif is involved in a loop−
loop interaction with stem loop I, leading to a shifted, active
Figure 5. (a) 13C CPMG relaxation dispersion profiles of VS stem
loop V motif 17 at 125 MHz carbon Larmor frequency and 15 °C.
Dots represent experimental data. Residues U695, U696, and U700
show nonflat dispersion profiles. R2,eff values at νCPMG = 960 Hz were
subtracted from the data for display purposes. (b) CPMG dispersion
profiles of U695, U696, and U700 at 125 (red) and 200 (blue) MHz
carbon Larmor frequency, which were used for a quantitative analysis
of the excited state. Dots represent the experimental data. Black times
signs represent repeat experiments. The dashed and dotted blue and
red lines represent the individual and global fits according to eq 2 for a
two-state equilibrium in the intermediate exchange regime.
conformation of the substrate hairpin I.27,48 A detailed analysis
of the potentially functional dynamics underlying the loop−
loop kissing tertiary interaction in the presence of magnesium
ions and the stem loop I motif using (6-13C)uridine and
(6-13C)cytidine labels is currently being carried out in our
laboratory and will be reported elsewhere.
3.5. ZZ Exchange Experiments To Probe Secondary
Structure Refolding Kinetics. So far we have shown the
applicability of (6-13C)pyrimidine spin-labels as probes of
micro- to millisecond nucleic acid dynamics. It was recently
shown that slower dynamics with folding times of a few
milliseconds to seconds can be of special importance for the
biological function of riboswitch RNAs.2,49−51 Riboswitches are
a class of nucleic acids for which structural flexibility is
inherently linked to their biological function. These RNAs
7566
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
switch their tertiary/secondary structure in response to the
presence or absence of a low-molecular-weight metabolite.52−54
We and others previously investigated the mode of action of
the preQ1 class I riboswitch system in detail.26,55,56 In the
previous work a bistable terminator antiterminator sequence,
18, was identified (Figure 6a).26
Figure 7. Normalized partial peak volumes of MA(t) and MAA(t) (red
and black rectangles) and MB(t) and MBB(t) (red and black circles)
corresponding to the correlation peaks of spin-label C3 of the ZZ
exchange (MAA/BB(t)) and T1 (MA/B(t)) experiments, together with
best fit curves (dashed red and black lines). All four curves were
obtained by simultaneously fitting the experimental data to eqs 3 and
4. Rate constants kAB = 3.3 0.1 s−1 and kBA = 1.4 0.3 s−1 and
longitudinal relaxation rates RA1 = 5.78 0.08 s−1 and R1B = 4.93 0.13
Figure 6. (a) Secondary structure equilibrium of terminator
antiterminator RNA 18 with a population of 18A/18B of
approximately 35/65. (b) ZZ exchange spectrum of 18 at a mixing
time of 200 ms and at 33 °C. The resonance presumably arising from
the homoduplex is marked with an asterisk. Conditions: 1 mM RNA,
25 mM sodium arsenate, pH 6.5, D2O, 33 C°.
s
−1 for the refolding reaction 18A to 18B and back for residue C3 were
We now were interested in the kinetics underlying the two-
state secondary structure equilibrium between folds 18A and
18B (Figure 6a). For that purpose, we introduced a single
(6-13C)cytidine substitution into the RNA and used 13C
longitudinal exchange NMR spectroscopy to probe the
refolding kinetics. Peak integration yielded the equilibrium
constant, which was in favor of fold 18B (KAB = 3). As expected
we found two correlation peaks (C3A and C3B) for C3 in the
13C ZZ exchange spectrum of RNA 18 for which exchange
peaks could be identified (Figure 6b). Fold assignment was
achieved using a truncated reference sequence (Supporting
Figure 8, Supporting Information). The forward and backward
folding rates of the secondary structure equilibrium were
determined at 33 °C using an approach recently introduced by
analyzing longitudinal relaxation rates with and without
exchange contribution (Figure 7). We found a forward rate
constant kAB of 3.3 0.1 s−1 and a rate constant kBA for the
folding process from state B to state A of 1.4 0.3 s−1, giving a
good agreement with the equilibrium constant obtained from
peak integration (KAB = 3 and KAB = kAB/kBA = 2.4). Using our
13C labeling method approach, we were thus able to
characterize the refolding kinetics of the terminator anti-
terminator RNA 18 in unperturbed equilibrium under near
physiological conditions.
obtained.
13C-isotope-modified pyrimidines were introduced into two
biologically relevant RNAs, the HIV-1 TAR RNAs 15 and 16,
and the Varkud satellite stem loop V motif 17. For both nucleic
acid fragments, residues exhibiting dynamics on the milli- to
microsecond time scale could be identified.
The dynamics of U23 of the HIV-1 TAR RNA could arise
from an excited state, where a base triple including U38, A27,
and U23 is transiently formed resembling the argininamide-
bound state.35 This further corroborates the conformation
selection mechanism of the HIV-1 TAR RNA previously
proposed by solution and solid-state NMR.7,46 Within the 17 nt
long Varkud satellite stem loop motif V 17, three residues,
U695, U696, and U700, exhibited a nonflat CPMG relaxation
dispersion profile. The data could be fit to collective motion of
the three residues, which could be important for a loop−loop
kissing tertiary interaction.47,48
To further illustrate the fields of application of our
(6-13C)pyrimidine labeling protocol, the kinetics of a secondary
structure refolding reaction was characterized. A bistable
terminator antiterminator RNA element (18) of the preQ1
class I riboswitch was labeled with a single (6-13C)cytidine
residue, and longitudinal exchange spectroscopy yielded the
forward and backward rates of the secondary structure
equilibrium.
4. CONCLUSIONS
Within this work we established a fully native 13C labeling
approach for RNA relying on (6-13C)pyrimidine amidite
building blocks suitable for oligonucleotide solid-phase syn-
The (6-13C)pyrimidine labels have proven to be generally
applicable to study conformational dynamics and the refolding
kinetics in RNAs comprising up to 30 nucleotides using
standard NMR techniques. Currently, the limit by the
presented approach is set by RNAs comprising more than 55
1
thesis. This isolated H−13C spin was used to address RNA
nucleobase dynamics in the milli- to microsecond time regime
by applying the CPMG relaxation dispersion technique. The
7567
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
(7) Zhang, Q.; Sun, X.; Watt, E. D.; Al-Hashimi, H. M. Science 2006,
311, 653.
(8) Lipari, G.; Szabo, A. J. Am. Chem. Soc. 1982, 104, 4546.
(9) Lipari, G.; Szabo, A. J. Am. Chem. Soc. 1982, 104, 4559.
(10) Korzhnev, D. M.; Religa, T. L.; Banachewicz, W.; Fersht, A. R.;
Kay, L. E. Science 2010, 329, 1312.
(11) Shajani, Z.; Varani, G. Biopolymers 2007, 86, 348.
(12) Quant, S.; Wechselberger, R. W.; Wolter, M. A.; Woerner, K.
H.; Schell, P.; Engels, J. W.; Griesinger, C.; Schwalbe, H. Tetrahedron
Lett. 1994, 35, 6649.
nt as the yield of the solid-phase synthesis methodology drops
significantly with the increasing size of the oligonucleotide. For
such larger 50 nt RNAs pulse sequences comprising a TROSY
element will be will be particularly useful.57,58 Therefore, the
13C label can be shifted to position C5, which shows more
favorable TROSY properties,59 simply by using commercially
available (2-13C)-2-bromoacetic acid and unlabeled potassium
cyanide as starting materials. These 5-13C-labeled pyrimidine
building blocks could further be deuterated at position C6,
resulting in even more advantageous relaxation properties,60
a
(13) Batey, R. T.; Cloutier, N.; Mao, H.; Williamson, J. R. Nucleic
crucial prerequisite for the NMR spectroscopic investigation of
large biomolecular complexes. The labeling protocol will in the
near future also be expanded to (8-13C)purine building blocks
and will be used to site-specifically modify functional nucleic
acids, such as riboswitch aptamer domains or ribozymes, to give
interesting insights into the functional dynamics of these
RNAs.61,62
Acids Res. 1996, 24, 4836.
(14) Batey, R. T.; Inada, M.; Kujawinski, E.; Puglisi, J. D.;
Williamson, J. R. Nucleic Acids Res. 1992, 20, 4515.
(15) Nikonowicz, E. P.; Pardi, A. J. Mol. Biol. 1993, 232, 1141.
(16) Lu, K.; Miyazaki, Y.; Summers, M. J. Biomol. NMR 2010, 46,
113.
(17) Thakur, C.; Sama, J.; Jackson, M.; Chen, B.; Dayie, T. J. Biomol.
NMR 2010, 48, 179.
(18) Johnson, J. E.; Julien, K. R.; Hoogstraten, C. G. J. Biomol. NMR
2006, 35, 261.
(19) Thakur, C.; Sama, J.; Jackson, M.; Chen, B.; Dayie, T. J. Biomol.
NMR 2010, 48, 179.
(20) Schultheisz, H. L.; Szymczyna, B. R.; Scott, L. G.; Williamson, J.
R. ACS Chem. Biol. 2008, 3, 499.
(21) Schultheisz, H. L.; Szymczyna, B. R.; Scott, L. G.; Williamson, J.
R. J. Am. Chem. Soc. 2010, 133, 297.
(22) Johnson, J. E.; Hoogstraten, C. G. J. Am. Chem. Soc. 2008, 130,
16757.
(23) Nikolova, E. N.; Gottardo, F. L.; Al-Hashimi, H. M. J. Am. Chem.
Soc. 2012, 134, 3667.
(24) Kloiber, K.; Spitzer, R.; Tollinger, M.; Konrat, R.; Kreutz, C.
Nucleic Acids Res. 2011, 39, 4340.
(25) Micura, R. Angew. Chem., Int. Ed. 2002, 41, 2265.
(26) Rieder, U.; Kreutz, C.; Micura, R. Proc. Natl. Acad. Sci. U.S.A.
2010, 107, 10804.
(27) Campbell, D. O.; Legault, P. Biochemistry 2005, 44, 4157.
(28) Hwang, T. L.; Shaka, A. J. J. Magn. Reson., Ser. A 1995, 112, 275.
(29) Delaglio, F.; Grzesiek, S.; Vuister, G. W.; Zhu, G.; Pfeifer, J.;
Bax, A. J. Biomol. NMR 1995, 6, 277.
(30) Tollinger, M.; Skrynnikov, N. R.; Mulder, F. A.; Forman-Kay, J.
D.; Kay, L. E. J. Am. Chem. Soc. 2001, 123, 11341.
(31) Kloiber, K.; Spitzer, R.; Grutsch, S.; Kreutz, C.; Tollinger, M. J.
Biomol. NMR 2011, 51, 123.
ASSOCIATED CONTENT
* Supporting Information
■
S
NMR spectra of phosphoramidite building blocks 8, 10, and 14,
mass spectra of 13C-modified RNAs 15, 16, 17, and 18, CPMG
dispersion profiles of (6-13C)cytidine-modified HIV-1 TAR
RNA 16, titration of (6-13C)uridine-modified HIV-1 TAR RNA
15 with argininamide, CPMG dispersion profiles of (6-13C)-
uridine-modified HIV-1 TAR RNA 15 at 17 and 33 °C and in
the presence of L-argininamide, CPMG dispersion profiles of
(6-13C)uridine-modified VS stem loop V RNA 17 at 25 °C, fold
assignment of RNA 18 using a truncated reference sequence
and a H2O NOESY experiment, and table summarizing the
NMR experiments applied to obtain data on the dynamics of
RNAs 15, 16, 17, and 18 and the respective results. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
christoph.kreutz@uibk.ac.at
■
Notes
The authors declare no competing financial interest.
(32) Carver, J. P.; Richards, R. E. J. Magn. Reson. (1969−1992) 1972,
ACKNOWLEDGMENTS
6, 89.
■
(33) Taylor, J. R. An Introduction to Error Analysis: The Study of
Uncertainties in Physical Measurements, 2nd ed.; University Science
Books: Sausalito, CA, 1997.
(34) SantaLucia, J.; Shen, L. X.; Cai, Z.; Lewis, H.; Tinoco, I. Nucleic
Acids Res. 1995, 23, 4913.
(35) Puglisi, J. D.; Tan, R.; Calnan, B. J.; Frankel, A. D.; Williamson.
Science 1992, 257, 76.
(36) Triplett, J. W.; Mack, S. W.; Smith, S. L.; Digenis, G. A. J.
C.K. thanks Ronald Micura (Innsbruck), Bernhard Kraeutler
(Innsbruck), and Robert Konrat (Vienna) for continuous
support and scientific discussions. We thank Karin Kloiber for
setting up some of the relaxation dispersion experiments. This
work was supported by the Tyrolean Science Fund TWF
(Project 95421 to C.K.), a Young Talents Grant from the
University of Innsbruck (to C.K.), and the Austrian Science
Fund FWF (Grant I844-N17 to C.K.).
Labelled Compd. Radiopharm. 1978, 14, 35.
(37) Vorbruggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber. 1981,
114, 1234.
̈
REFERENCES
■
(38) Pitsch, S.; Weiss, P. A. Current Protocols in Nucleic Acid
Chemistry; John Wiley & Sons, Inc.: New York, 2001.
(39) Milecki, J.; Zamaratski, E.; Maltseva, T. V.; Foldesi, A.; Adamiak,
R. W.; Chattopadhyaya, J. Tetrahedron 1999, 55, 6603.
(40) Wenter, P.; Reymond, L.; Auweter, S. D.; Allain, F. H. T.;
Pitsch, S. Nucleic Acids Res. 2006, 34, e79.
(1) Palmer, A. G.; Kroenke, C. D.; Loria, J. P. Methods Enzymol.
2001, 339, 204.
(2) Al-Hashimi, H. M.; Walter, N. G. Curr. Opin. Struct. Biol. 2008,
18, 321.
(3) Latham, M. P.; Brown, D. J.; McCallum, S. A.; Pardi, A.
ChemBioChem 2005, 6, 1492.
(41) Hoebartner, C.; Micura, R. J. Am. Chem. Soc. 2004, 126, 1141.
(42) Pitt, S. W.; Majumdar, A.; Serganov, A.; Patel, D. J.; Al-Hashimi,
H. M. J. Mol. Biol. 2004, 338, 7.
(4) Mittermaier, A. K.; Kay, L. E. Trends Biochem. Sci. 2009, 34, 601.
(5) Akke, M.; Fiala, R.; Jiang, F.; Patel, D.; Palmer, A. G. RNA 1997,
3, 702.
(43) Aboul-ela, F.; Karn, J.; Varani, G. Nucleic Acids Res. 1996, 24,
(6) Getz, M.; Sun, X.; Casiano-Negroni, A.; Zhang, Q.; Al-Hashimi,
H. M. Biopolymers 2007, 86, 384.
3974.
7568
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569

Journal of the American Chemical Society
Article
(44) Milligan, J. F.; Groebe, D. R.; Witherell, G. W.; Uhlenbeck, O.
C. Nucleic Acids Res. 1987, 15, 8783.
(45) Milligan, J. F.; Uhlenbeck, O. C. Methods Enzymol. 1989, 180,
51.
(46) Olsen, G. L.; Bardaro, M. F.; Echodu, D. C.; Drobny, G. P.;
Varani, G. J. Am. Chem. Soc. 2009, 132, 303.
(47) Campbell, D. O.; Bouchard, P.; Desjardins, G.; Legault, P.
Biochemistry 2006, 45, 10591.
(48) Bouchard, P.; Lacroix-Labonte, J.; Desjardins, G.; Lampron, P.;
Lisi, V.; Lemieux, S.; Major, F.; Legault, P. RNA 2008, 14, 736.
(49) Schwalbe, H.; Buck, J.; Furtig, B.; Noeske, J.; Wohnert, J. Angew.
̈
̈
Chem., Int. Ed. 2007, 46, 1212.
(50) Rinnenthal, J.; Buck, J.; Ferner, J.; Wacker, A.; Furtig, B.;
Schwalbe, H. Acc. Chem. Res. 2011, 44, 1292.
(51) Al-Hashimi, H. M. Biopolymers 2007, 86, 345.
(52) Coppins, R. L.; Hall, K. B.; Groisman, E. A. Curr. Opin.
Microbiol. 2007, 10, 176.
(53) Serganov, A. Curr. Opin. Struct. Biol. 2009, 19, 251.
(54) Tucker, B. J.; Breaker, R. R. Curr. Opin. Struct. Biol. 2005, 15,
342.
(55) Rieder, U.; Lang, K.; Kreutz, C.; Polacek, N.; Micura, R.
ChemBioChem 2009, 10, 1141.
(56) Kang, M.; Peterson, R.; Feigon, J. Mol. Cell 2009, 33, 784.
(57) Brutscher, B.; Boisbouvier, J.; Pardi, A.; Marion, D.; Simorre, J.-
P. J. Am. Chem. Soc. 1998, 120, 11845.
́ ̌
(58) Fiala, R.; Czernek, J.; Sklenar, V. J. Biomol. NMR 2000, 16, 291.
(59) Boisbouvier, J.; Bryce, D.; O’Neil-Cabello, E.; Nikonowicz, E.;
Bax, A. J. Biomol. NMR 2004, 30, 287.
(60) Huang, X.; Yu, P.; LeProust, E.; Gao, X. Nucleic Acids Res. 1997,
25, 4758.
́ ́
(61) Edwards, T. E.; Klein, D. J.; Ferre D’Amare, A. R. Curr. Opin.
Struct. Biol. 2007, 17, 273.
(62) Furtig, B.; Buck, J.; Richter, C.; Schwalbe, H. In Methods in
̈
Molecular Biology; Hartig, J. S., Ed.; Humana Press: Totowa, NJ, 2012;
Vol. 848, p 185.
7569
dx.doi.org/10.1021/ja302148g | J. Am. Chem. Soc. 2012, 134, 7558−7569