Concept
Scheme 5). This derivative was treated with 2,4,6-triisopropyl-
benzenesulfonyl chloride in the presence of triethylamine and
the desired 2’-O-tBDMS protection. Standard tritylation and
15
phosphitylation gave N-labeled cytidine phosphoramidite
4
N3
DMAP in dichloromethane to give regioselective O -trisylation.
building block C3 in an overall yield of 16% over 10 steps,
15
N3/N4
After workup, the trisylated intermediate (U3·Tris) can be used
and N double-labeled C3
in an overall yield of 10% over
without further purification and is directly converted into the
12 steps. Alternatively, the corresponding 2’-O-TOM-protected
15
N3
protected N(3) cytidine derivative C1 by using aqueous am-
phosphoramidites can be prepared according to Pitsch and co-
[
15]
workers, starting from C2.
Concerning the ribose hydroxyl protection for the synthesis
N1
N1/N4
of the pyrimidine nucleosides U5, C3 , and C3
, it is worth
noting that the tBDMS groups withstand the harsh basic con-
15
ditions of in situ generated NH very well. We prefer this pro-
3
tection to protection with acetyl groups that are partially
cleaved during the extended time that is needed to complete
15
N insertions. Moreover, the protocol we developed for the
deprotection of the tBDMS groups based on the polymer sup-
ported fluoride reagent is much less labor-intensive compared
to workup and purification protocols when acetyl protection is
used.
Hydrogen-Bond Nucleobase Interactions De-
tected by NMR Spectroscopic Experiments
1
5
N3
Scheme 5. Syntheses of N(3) cytidine phosphoramidite C3 and
1
5
15
N3N4
N3
N(3), N(4) cytidine phosphoramidite C3
i) 2.5 equiv mesitylene sulfonylchloride in CH
NH , in THF, room temperature, 16 h, 95%; for C3
lene sulfonylchloride in CH Cl /NEt
bicyclo[2.2.2]octan (DABCO), DMAP, room temperature, overnight, 87%,
. Conditions for C3 : a.
NMR spectroscopy allows the precise identification of base
pairs in RNA based on the detection of hydrogen bonds in-
volving amido protons (often coined “imino protons”), which
are shielded from exchange with bulk water. Indirect evidence
for hydrogen-bonds comes from NOE correlations of amido
protons that establish base pairing in helical regions of
2
Cl
2
/NEt
3
, 3.5 h, ii) 32% aqueous
: a’. i) 2.5 equiv mesity-
3
, 1.5 h, ii) 4 equiv 2-nitrophenol, 1,4-diaza-
N3N4
3
2
2
1
5
iii) 4.5 equiv NH
4
Cl, 4.5 equiv K
2
CO
3
in DMSO, 808C, 2 d, 67%. Conditions
N3
N3N4
for both, C3 and C3
: b. Amberlite IRA 900 (fluoride form) in toluene,
reflux, 4 h, 92%; c. i) 1.1 equiv di-tert-butylsilyl bis(trifluoromethanesulfonate)
in DMF, 08C, 40 min, ii) 1.1 equiv tBDMS-Cl, 4.2 equiv imidazole, 608C, 2 h,
[
4,48]
RNA,
while direct evidence can be obtained from HNN-
iii) 2.4 equiv Ac
dine/CH Cl , 2 h, v) 1.1 equiv 4,4’-dimethoxytrityl chloride in pyridine, room
temperature, overnight, 66% (2 steps); vi) 3.5 equiv CEP-Cl, 2 equiv N,N-diiso-
2
O in DMF/pyridine; overnight, 60% (3 steps), iv) HF in pyri-
COSY experiments that were originally introduced by Dingley
[
7]
2
2
and Grzesiek. The size of these cross hydrogen-bond scalar
2
couplings ( J =6–7 Hz) allows efficient magnetization transfer
15
NN
propylethylamine, room temperature, 4 h, 73%. N indicated in red.
[
7]
to correlate chemical shifts in base pairs. To date, a large
number of experiments are available to measure J-couplings
[
49–58]
monium hydroxide in THF. We point out that the correspond-
across hydrogen bonds in nucleic acids.
A limitation, how-
15
15
N3/N4
ing double-labeled N(3), N-(C4) cytidine C
(see the Sup-
ever, is that most of these approaches detect exchangeable
protons. Hence, regions of stable secondary or tertiary struc-
tures are straightforward to determine, while weak base inter-
actions are less readily detectable because of signal broaden-
ing due to proton exchange with bulk water and conforma-
tional dynamics, in particular at higher temperatures. Recently,
the Sattler team introduced a new experiment termed BEST-
sellr that provides a large sensitivity improvement for hydro-
gen-bond correlation experiments by the optimizations of
magnetization transfers in combination with the detection of
porting Information) is conveniently accessible from the trisy-
lated intermediate (U3·Tris) as well, when first substituted by
4
2
-nitrophenol under the formation of O -(2-nitrophenyl)uridine
1
5
(
U3·o-NO Ph) followed by treatment with NH Cl and K CO in
2 4 2 3
DMSO in a pressure reactor, in analogy to a procedure by Chat-
[17]
topadhyaya and co-workers. The tBDMS protecting groups
are conveniently cleaved with a polymer-supported fluoride re-
agent (Amberlite IRA 900) in boiling toluene to provide C2 in
excellent yield and purity without the requirement for chroma-
tographic purification.
[
59]
non-exchangeable protons. This experiment enables the de-
tection of weak and transient A–U base pairs. The novel devel-
opments in NMR spectroscopy of ribonucleic acids imply an in-
For the attachment of the necessary protecting groups for
[
46]
solid-phase synthesis, we recommend the Beigelman route.
15
Simultaneous protection of 3’- and 5’-hydroxyl with the di-tert-
butylsilylene protecting group, subsequent silylation of the 2’-
hydroxyl using tBDMS-Cl and acetylation of the exocyclic
amino group proceeded in excellent yields. Removal of the
strained cyclic silyl protection group proceeded straightfor-
wardly using hydrogen fluoride in pyridine. This procedure
gives significantly higher yields for 2’-O-tBDMS protection com-
creasing need for site-specifically (beside uniformly) N-labeled
RNA using optimized synthetic pathways as discussed above.
Conformational States of PreQ Riboswitches
Analyzed by Site-Specific N-labeled Base
1
15
Pairs
[41]
pared to the more widely used Ogilvie procedure, which re-
RNA secondary and tertiary structure predictions require exper-
imental validation. Frequently, this is achieved by biochemical
sults in an unfavorable regioselectivity of 4:3 between 3’- and
Chem. Eur. J. 2015, 21, 11634 – 11643
11639
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim