A semi-rigid isoindoline-derived nitroxide spin label for RNA

Article abstract of DOI:10.1039/c7ob02870a

A new isoindoline-derived benzimidazole nitroxide spin label, ^ImUm, was synthesized and incorporated into RNA oligoribonucleotides. ^ImUm is the first example of a conformationally unambiguous spin label for RNA, in which the nitroxide N-O bond lies on the same axis as the single bond used to attach the rigid isoindoline-based spin label to a uridine base. This results in minimal displacement of the nitroxide upon rotation of this single bond, which is a useful property for a label to be used for distance measurements. Continuous-wave (CW) EPR measurements of RNA duplexes containing ^ImUm indicate a restricted rotation around this single bond, presumably due to an intramolecular hydrogen bond between the benzimidazole N-H and O4 of the uracil. Orientation-selective pulsed electron-electron double resonance (PELDOR, also called double electron-electron resonance, or DEER) distance measurements between two spin labels in two RNA duplexes showed in one case a strong orientation dependence, further confirming the restricted motion of the spin labels in RNA duplexes.

Full text of DOI:10.1039/c7ob02870a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
A semi-rigid isoindoline-derived nitroxide spin
label for RNA†
Cite this: DOI: 10.1039/c7ob02870a
Dnyaneshwar B. Gophane,a Burkhard Endeward,b Thomas F. Prisner*b and
a
Snorri Th. Sigurdsson
*
A new isoindoline-derived benzimidazole nitroxide spin label, ImUm, was synthesized and incorporated
into RNA oligoribonucleotides. ImUm is the first example of a conformationally unambiguous spin label
for RNA, in which the nitroxide N–O bond lies on the same axis as the single bond used to attach the
rigid isoindoline-based spin label to a uridine base. This results in minimal displacement of the nitroxide
upon rotation of this single bond, which is a useful property for a label to be used for distance measure-
ments. Continuous-wave (CW) EPR measurements of RNA duplexes containing ImUm indicate a restricted
rotation around this single bond, presumably due to an intramolecular hydrogen bond between the ben-
zimidazole N–H and O4 of the uracil. Orientation-selective pulsed electron–electron double resonance
(PELDOR, also called double electron–electron resonance, or DEER) distance measurements between
two spin labels in two RNA duplexes showed in one case a strong orientation dependence, further confi-
rming the restricted motion of the spin labels in RNA duplexes.
Received 21st November 2017,
Accepted 5th January 2018
DOI: 10.1039/c7ob02870a
rsc.li/obc
during the oligonucleotide synthesis.24 This approach has the
advantage that spin labels with limited flexibility can be incor-
Introduction
RNA is a ubiquitous family of biopolymers that have multiple porated into the oligoribonucleotides. One such label is Çm,24
vital roles in the coding, decoding, regulation and expression a ribonucleotide derivative of the rigid spin label Ç 25–28
of genes.1–3 The understanding of RNA, DNA and protein func- (Fig. 1), which has been used to probe mono- and bimolecular
tion, including their interactions with other molecules, RNA structures and measurements of distances and orien-
requires knowledge of their respective molecular structures tation within RNA.29
and conformational dynamics. EPR spectroscopy is a useful
We have previously reported the conformationally restricted
technique to extract such information.4–10 It requires small isoindoline-derived benzimidazole spin label ImU (Fig. 1) for
amounts of material and can be used to study biopolymers DNA, which showed EPR-based spectroscopic properties
under biologically relevant conditions. Since RNA is diamag- similar to that of the rigid spin label Ç.30 The restricted mobi-
netic, paramagnetic groups, or spin labels, must be incorpor- lity was believed to arise, at least in part, from an intra-
ated at specific sites, a technique referred to as site-directed molecular hydrogen bond between the NH of the benzimida-
spin labeling (SDSL).11–14 Apart from a recent report of non- zole and O4 of the uracil base.30 ImU has been used for dis-
covalent and site-specific spin-labeling of RNA, based on binding tance measurements in duplex DNAs and showed a strong
of a spin-labeled guanine to an abasic site in duplex RNA,15
there are two strategies that have been used to incorporate
spin labels into RNA using covalent bonds. One is via post-
synthetic labeling of pre-functionalized sites.16–23 The other
covalent-labeling approach is the phosphoramidite method, in
which the spin label is incorporated into the desired oligomer
aUniversity of Iceland, Department of Chemistry, Science Institute, Dunhaga 3,
107 Reykjavik, Iceland. E-mail: snorrisi@hi.is
bInstitute of Physical and Theoretical Chemistry, J. W. Goethe University,
Max-von-Laue-Str. 7, D-60438 Frankfurt, Germany.
E-mail: prisner@chemie.uni-frankfurt.de
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
Fig. 1 Ç and Çm (A), ImU and ImUm (B), and base-pairing of ImUm with
c7ob02870a
adenine (C).
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
orientation dependence, another indication of limited mobi- subjected
to
dihydroxylation
using
OsO4
and
lity.31 Aside from the orientation dependence of ImU, the N-methylmorpholine N-oxide (NMO) to obtain the corres-
single bond that connects the benzimidazole moiety to the ponding diol,34 which was further treated with NaIO4 to give
base lies on an axis that runs through the N–O bond of the the desired aldehyde 4 in good yield.35 Aldehyde 4 was treated
nitroxide (Fig. 1), which makes ImU useful for precise distance with diamino-tetramethylisoindoline 5 and K3Fe(CN)6, yielding
measurements.
the imidazole derivative 6. Compound 6 was subjected to
Herein, we report the synthesis of ImUm, a nitroxide deriva- m-CPBA-mediated oxidation in the presence of NaN3,30 which
tive of uridine for RNA spin labeling. The phosphoramidite of presumably adds transiently to the 6-position of the pyrimi-
ImUm was prepared and incorporated into different RNA struc- dine base,36 generating compound 7. The TBDMS protecting
tural contexts by solid-phase synthesis. The new label was well groups of 7 were removed using TBAF to afford the nucleoside
tolerated in the A-form helices when paired with adenosine ImUm, followed by tritylation and phosphitylation to give phos-
(Fig. 1), as judged by its small effect on the thermodynamic phoramidite 9 in good yield.
stability of the labeled RNAs. ImUm showed very limited mobi-
Syntheses and purification of RNA oligonucleotides
containing ImUm
lity in duplex RNA, indicating that rotation around the single
bond, linking the spin label to the uracil, is indeed restricted.
CW-EPR spectroscopy was used to show that the ImUm was
able to report on the local environment of the labeling site.
PELDOR distance measurements on RNA duplexes using
ImUm were in close agreement with distances derived from
molecular modeling. Furthermore, PELDOR measurements of
ImUm-labeled RNA duplexes showed orientation dependence,
further confirming the limited motion of this spin label in
RNA duplexes.
ImUm was incorporated into RNA oligoribonucleotides by solid
phase synthesis using the previously reported protocol, with a
slight modification.24,26 Upon completion of the synthesis, the
oligoribonucleotides were cleaved from the solid support and
the nucleobases and phosphodiesters were deprotected in a
1 : 1 mixture of 33% aqueous NH3 and 8 M MeNH2 in EtOH.
The 2′-O-TBDMS groups were removed by treatment with
Et3N·3HF, after which water was added and the oligoribonu-
cleotides precipitated with n-butanol. The crude oligoribonu-
cleotides were purified by DPAGE and quantified using UV
absorbance spectroscopy.
Results and discussion
Synthesis and incorporation of ImUm into RNA
The effect of ImUm on RNA duplex stability
The synthesis of ImUm (Scheme 1) began with iodination of The incorporation of ImUm into RNA was first demonstrated by
2′-O-methyluridine (Um).32 The 3′- and 5′-hydroxyl groups of 1 synthesis of the 14-mer oligoribonucleotide 5′-(CACGAImUm-
were protected by TBDMS to give 2,33 which was treated with GCGAGGUC). After annealing to a complementary RNA, a
vinyl acetate in the presence of Pd(OAc)2 to yield the 5-vinyl thermal denaturation experiment of the duplex 5′-(CACGA
functionalized compound 3 in good yield.33 Compound 3 was ImUmGCGAGGUC)·5′-(GACCUCGCAUCGUG) (II) showed
a
Scheme 1 Synthesis of the nucleoside ImUm and its corresponding phosphoramidite (9).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 1 Sequences and TMs of RNA hairpins and duplexes
RNA
Sequence
TM
ΔTM
−6.1
I
73.4 0.6
II
67.3 0.3
85.2 1.0
III
IV
V
84.7 0.3
76.0 0.5
79.7 0.3
−0.5
VI
+3.7
VII
VIII
IX
74.5 1.0
73.8 0.8
77.2 0.8
−0.7
+2.7
4 mM duplex in 10 mM sodium phosphate, 100 mM NaCl, 0.1 mM
Na2EDTA, pH 7.0. TM is the melting temperature and ΔTM is the differ-
ence in TM between unmodified and modified duplexes.
Fig. 2 EPR spectra of ImUm (A), the 14-mer RNA single strand
5’-(CACGAImUmGCGAGGUC) (B) and the duplex 5’-(CACGAImUmGC-
GAGGUC)·5’-(GACCUCGCAUCGUG) (II) (C). EPR spectra were recorded
single melting transition, lowering the melting temperature at 20 °C in a phosphate buffer (10 mM, pH 7.0) containing NaCl
(100 mM) and Na2EDTA (0.1 mM).
(TM) by 6.1 °C compared to the unmodified duplex (I) (Table 1
and Fig. S1A†). We have previously observed a lowering in TM
by 4.0 °C when ImU was incorporated into the center of a
14-mer DNA.30 CD spectra of the ImUm-modified and unmodi-
14-mer RNA single strand and the corresponding 14-mer RNA
duplex II (Fig. 2). The nucleoside showed three sharp lines
(Fig. 2A) that broadened upon incorporation into the 14-mer
fied oligoribonucleotides were almost identical, confirming
that ImUm is well tolerated in an A-form RNA helix (Fig. S1B†).
The spin-label ImUm was also incorporated into the stem
oligoribonucleotide 5′-(CACGAImUmGCGAGGUC) (Fig. 2B).
region of an RNA hairpin 5′-(GImUmCGACGGAAGUC-
Upon annealing to its complementary strand 5′-(GACCUCGC-
AUCGUG), the CW-EPR spectrum showed a splitting of the
GACAGUA) (IV), which contains a six base-pair helix and a
stable GGAA tetraloop. In hairpin IV, the location of the label
is close to the end of the stem and as expected, it showed only
high- and low-field components (Fig. 2C), similar to the pre-
viously reported rigid spin labels Ç 25 and Çm,24 as well as to
a minor effect on the TM (−0.5 °C), compared to the corres-
ponding unmodified hairpin (III) (Table 1 and Fig. S1C†).
ImUm-modified hairpin IV was annealed to its complementary
RNA strand, 3′-(CAGCUGCCUUGAGCUGUCAU), yielding
duplex VIII, thus placing the label close to the end of the
duplex and resulting in only minor destabilization (−0.7 °C)
(Table 1 and Fig. S1E†). ImUm was also incorporated into the
overhang region of the oligoribonucleotide hairpin VI, where it
stabilized the hairpin by 3.7 °C, compared to the unmodified
hairpin V (Table 1 and Fig. S1D†). When hairpin VI was
annealed to its complementary strand, the resulting duplex
(IX) was stabilized by 2.7 °C (Table 1 and Fig. S1F†), relative to
the unmodified sequence (VII).
the semi-rigid ImU.30 Such broadening is characteristic for the
slow-motion regime of nitroxide radicals.24
To investigate the mobility of the ImUm in different struc-
tural contexts, it was incorporated into RNA hairpins and
duplexes (Fig. 3). The ImUm-labeled hairpins IV (Fig. 3A) and
VI (Fig. 3B) exhibited broadened CW-EPR spectra, compared to
single strands. However, spectral line-broadening of hairpin VI
is less pronounced than hairpin IV, since the label is located
in the overhang region of hairpin VI. The ImUm-labeled
duplexes VIII and IX showed additional broadening of the
spectra, compared to the hairpins. This is presumably due to
the increased size of the RNA and the associated slower
rotational correlation times, as has been observed with the
rigid spin label Çm.24 The EPR spectrum of IX at 20 °C is
slightly broader than for duplex VII (Fig. 3), presumably
CW-EPR analyses of ImUm in duplexes and hairpins
To evaluate the mobility of ImUm in RNA, we recorded the because the spin label in the latter is closer to the duplex end,
CW-EPR spectra of the nucleoside ImUm, the ImUm-labeled where base pairs are more dynamic.37,38
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 3 EPR spectrum of the hairpin IV and the spectrum after annealing
IV to the complementary hairpin V to form duplex VIII (A). EPR spectrum
of the hairpin VI and the spectrum after annealing V with the comp-
lementary hairpin III to form duplex IX (B). EPR spectra were recorded at
20 °C in 10 mM phosphate buffer (pH 7.0) containing NaCl (100 mM)
and Na2EDTA (0.1 mM).
Fig. 4 Spartan-derived molecular models of RNA duplexes containing
two ImUm spin labels, placed either nine base-pairs apart (duplex XI,
5’-(GImUmCGACGGAAGImUmCGACAGUA)·3’-(CAGCUGCCUUGAGCUG-
UCAU), left) or 13 base-pairs apart (duplex XII, 5’(GImUmCGACGG-
AAGUCGACAGUA)·3’(CAGCUGCCUUGAGCUGImUmCAU), right).
Distance measurements of ImUm-labeled RNAs by PELDOR
Two doubly ImUm-labeled RNA duplexes were prepared for dis-
tance measurements, one containing nine base pairs (XI) and
the other fourteen base pairs (XII) between the spin labels
(Fig. 4). Both the distances between the two spin-label pairs
and their relative orientations were different in these duplexes.
Orientation selective PELDOR27 was performed at X-Band fre-
quencies in which the pump frequency coincided with the
cavity resonance maximum as well as the maximum in spectral
density of the nitroxide spectrum. The detection was per-
formed with different frequency offsets on the high frequency
side, relative to the pump frequency. For both RNAs, the
PELDOR time traces of each offset were different from each
other (Fig. 5). This is a strong indication of orientation depen-
dence resulting from limited mobility of the spin labels within
the duplexes. The orientation dependence is very clear, in par-
ticular for RNA XII. However, in the case of RNA XI, the orien-
tation dependence is not as clear, but the differences between
the time traces are evident. For each duplex, the different time
traces were added to get an average time trace, which removes
the orientation effects. DeerAnalysis39 of these time traces
yielded the distance distribution shown in Fig. 5. The
measured distances between the spin labels in RNAs XI
(32.2 Å) and XII (38.5 Å) was close to that of the modelled dis-
tances (29.9 Å and 36.2 Å, respectively).
Fig. 5 PELDOR time traces at different frequency offsets for duplex XI
5’(GImUmCGACGGAAGImUmCGACAGUA)·3’(CAGCUGCCUUGAGCUGU-
CAU) (A) and duplex XII 5’-(GImUmCGACGGAAGUCGACAGUA)·3’-(CA
GCUGCCUUGAGCUGImUmCAU) (B). Evaluation of interspin distances in
doubly spin-labeled RNA duplexes from average over all PELDOR time
traces by DeerAnalysis39 for duplex XI (C) and duplex XII (D).
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
ing solution was heated at 80 °C for 1 h, followed by cooling to
22 °C. The solid was filtered, washed with H2O (100 mL) and
Conclusions
The new spin label ImUm, a derivative of uridine, was prepared CH3CN (50 mL), and dried in vacuo to yield compound 1 as a
1
for RNA spin-labeling. It was incorporated into different RNA white solid (6.4 g, 86%). H NMR (d6-DMSO): δ 11.69 (s, 1H),
structural contexts by solid-phase oligonucleotide synthesis. 8.53 (s, 1H), 5.79 (d, J = 3.9 Hz, 1H), 5.30 (s, 1H), 5.12 (d, J =
The spin-labeled RNAs were analyzed by thermal denaturation 6.3 Hz, 1H), 4.11 (d, J = 5.5 Hz, 1H), 3.86 (d, J = 5.5 Hz, 1H),
experiments, CD- and EPR-spectroscopy. The new label ImUm 3.79 (d, J = 4.4 Hz, 1H), 3.70 (ddd, J = 12.0, 4.6, 2.8 Hz, 1H),
was fairly well tolerated in A-form helices, as judged by 3.57 (ddd, J = 12.1, 4.3, 2.4 Hz, 1H), 3.38 (s, 3H). 13C NMR (d6-
thermal denaturation experiments. CW-EPR spectra of ImUm- DMSO): δ 160.46, 150.09, 144.82, 86.55, 84.76, 83.02, 69.31,
labeled RNA duplexes showed limited mobility, indicating that 67.79, 60.25, 59.64, 57.57. HRMS-ESI: m/z calcd for
rotation around the single bond, linking the spin label to the C10H12IN2O6 (M − H)− 382.9746, found 382.9741.
uracil, is restricted. ImUm also reported on its local environ-
Compound 2
ment by CW-EPR and provided information on the transition
from RNA hairpins to duplexes. Distances between pairs of To a solution of 2′-O-methyl-5-iodouridine (6.4 g, 0.017 mol) in
ImUm spin labels in RNA duplexes, measured by PELDOR, DMF (19.2 mL) and pyridine (19.2 mL) was added tert-butyldi-
were in close agreement with distances derived from molecular methylsilyl chloride (7.53 g, 0.050 mol) and imidazole (3.40 g,
models. The PELDOR measurements showed a strong orien- 0.050 mol). After stirring at 22 °C for 16 h, the solvent was
tation-dependence, in particular for RNA XII, indicating removed under reduced pressure and the residue was dissolved
limited motion of the spin label. Therefore, ImUm is a promis- in EtOAc (200 mL) and H2O (100 mL) was added. The EtOAc
ing RNA spin label for extracting accurate distances and layer was separated and washed with H2O (2 × 100 mL). The
obtaining information about orientations using PELDOR. organic layer was dried over Na2SO4, filtered and the solvent
Such information gives insights into both structure and con- evaporated in vacuo. The crude product was purified by flash
formational dynamics of the RNA under investigation.
column chromatography (silica gel) using a gradient elution
(EtOAc : pet. ether; 10 : 90 to 15 : 85) to give compound 2 as a
1
white solid (8.9 g, 87% yield). H NMR (CDCl3): δ 8.64 (s, 1H),
8.06 (s, 1H), 5.99 (d, J = 4.5 Hz, 1H), 4.23 (t, J = 4.8 Hz, 1H),
4.09–4.01 (m, 1H), 3.97 (dd, J = 11.7, 1.8 Hz, 1H), 3.76 (dd, J =
11.7, 1.8 Hz, 1H), 3.67 (t, J = 4.7 Hz, 1H), 3.46 (s, 3H), 0.98 (s,
9H), 0.91 (s, 9H), 0.19 (d, J = 1.6 Hz, 6H), 0.10 (d, J = 7.9 Hz,
6H). 13C NMR (CDCl3): δ 159.93, 149.99, 144.40, 87.50, 85.65,
84.15, 69.84, 69.00, 62.26, 58.52, 26.56, 25.92, 18.91, 18.35,
−4.38, −4.55, −4.71, −4.81. HRMS-ESI: m/z calcd for
C22H40IN2O6Si2 (M − H)− 611.1475, found 611.1443.
Experimental section
All commercially available reagents were purchased from
Sigma-Aldrich or Acros Organics and used without further
purification. 2′-O-Methyluridine was purchased from Rasayan
Inc. USA. All moisture or air sensitive reactions were performed
in flame-dried glasswares under a positive pressure of nitro-
gen. Solvents were distilled and stored over activated 4 Å mole-
cular sieves under nitrogen. Water was purified on a Milli-Q
water purification system. Analytical thin layer chromatography
(TLC) was performed on silica gel glass plates (Silicycle, ultra-
pure silica gel, 60 Å, F254), TLC visualisation was performed
with UV light. Flash-column chromatography was performed
on silica gel (Silicycle, 230–400 mesh, 60 Å). 1H-Spectra were
recorded using deuterated solvents as internal standards on a
Bruker Advance 400 spectrometer and are reported in ppm.
Residual proton signals from the deuterated solvents were
used as references [D2O (4.81 ppm), d6-DMSO (2.50 ppm),
CDCl3 (7.26 ppm), for 1H spectra]. 13C NMR chemical shifts
are reported in reference to undeuterated residual solvent
(CDCl3 (77.0 ppm), d6-DMSO (39.43 ppm)). 31P NMR chemical
shifts were reported relative to 85% H3PO4 as an external stan-
dard. Commercial grade CDCl3 was passed over basic alumina
shortly before use. Mass spectrometric analyses of all organic
compounds were performed on an ESI-HRMS (Bruker,
microTOF-Q) in positive or negative mode.
Compound 3
Palladium(II) acetate (74 mg, 0.33 mmol), PPh3 (154 mg,
0.590 mmol) and anhydrous triethylamine (3.64 ml,
26.11 mmol) were combined in anhydrous DMF (10.8 mL) and
stirred at 64 °C until an intense red colour developed. A solu-
tion of compound 5 (2 g, 3.26 mmol) and vinyl acetate
(15.17 g, 176.25 mmol) in anhydrous DMF (16.50 mL) was
added and the stirring maintained at 64 °C for 17 h. The
resulting precipitate of palladium was removed by filtration,
the filtrate evaporated to dryness and H2O (75 mL) was added
to the residue. The aqueous mixture was extracted with CH2Cl2
(3 × 100 mL). The organic layer was dried over Na2SO4, filtered,
the solvent evaporated in vacuo and the crude product was pur-
ified by flash column chromatography (silica gel) using a gra-
dient elution (CH2Cl2 : pet. ether; 5 : 95 to 10 : 90) to give com-
pound 3 as a white solid (903 mg, 54% yield). 1H NMR
(CDCl3): δ 8.64 (s, 1H), 8.06 (s, 1H), 5.99 (d, J = 4.5 Hz, 1H),
4.23 (t, J = 4.8 Hz, 1H), 4.09–4.01 (m, 1H), 3.97 (dd, J = 11.7, 1.8
Hz, 1H), 3.76 (dd, J = 11.7, 1.8 Hz, 1H), 3.67 (t, J = 4.7 Hz, 1H),
Compound 1
To a solution of 2′-O-methyluridine (5.0 g, 0.019 mol) in 3.46 (s, 3H), 0.98 (s, 9H), 0.91 (s, 9H), 0.19 (d, J = 1.6 Hz, 6H),
CH3CN (18 mL) was added iodine (2.94 g, 0.012 mol) and 0.10 (d, J = 7.9 Hz, 6H). 13C NMR (CDCl3): δ 13C NMR (CDCl3)
ammonium cerium(IV) nitrate (5.30 g, 0.0097 mol). The result- δ 161.91, 159.99, 150.03, 149.56, 144.39, 137.03, 128.18,
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
117.13, 112.96, 87.49, 85.64, 85.22, 84.15, 83.91, 69.80, 69.03, NMR (d6-DMSO): δ 12.04 (s, 1H), 8.59 (s, 1H), 7.29 (s, 1H), 7.20
62.28, 58.52, 58.46, 26.56, 26.30, 25.93, 18.91, 18.82, 18.37, (s, 1H), 5.93 (d, J = 4.4 Hz, 1H), 4.28 (t, J = 5.0 Hz, 1H),
−4.36, −4.57, −4.71, −4.82, −4.98, −5.10. HRMS-ESI: m/z calcd 4.04–3.75 (m, 4H), 3.39 (d, J = 5.0 Hz, 3H), 1.39 (s, 13H), 0.89
for C24H43N2O6Si2 (M − H)− 511.2665, found 511.2643.
(d, J = 4.6 Hz, 19H), 0.13 (dd, J = 17.6, 2.4 Hz, 12H). 13C NMR
(d6-DMSO): δ 161.83, 149.56, 145.62, 143.70, 143.45, 142.45,
139.88, 134.09, 109.73, 104.78, 104.12, 87.23, 84.69, 82.16,
Compound 4
Compound 3 (3.72 g, 0.00725 mol), N-methylmorpholine-N- 69.66, 62.25, 61.74, 61.60, 57.69, 45.67, 32.59, 32.56, 30.64,
oxide (2.13 g, 0.0181 mol), and OsO4 (0.0276 g, 0.000109 mol) 25.96, 25.58, 18.09, 17.76, 11.49, −4.79, −5.01, −5.37, −5.38.
were stirred in a solution of acetone-H2O-tBuOH (4 : 1 : 1, HRMS-ESI: m/z calcd for C35H58N5O6Si2 (M + H)+ 700.3920,
30 mL) at 22 °C for 22 h. The reaction mixture was partitioned found 700.3952.
between EtOAc (300 mL) and H2O (150 mL). The aqueous layer
was extracted with EtOAc (2 × 100 mL). The combined organic
Compound 7
layers were dried over Na2SO4, filtered, the solvent evaporated To a suspension of 6 (270.0 mg, 0.39 mmol) in CH3CN (18 mL)
in vacuo and the crude product was purified by flash column was added NaN3 (100.0 mg, 1.54 mmol) and the reaction
chromatography (silica gel) using
a
gradient elution stirred at 22 °C. After 30 min, mCPBA (133.0 mg, 0.77 mmol)
(MeOH : CH2Cl2; 0 : 100 to 2 : 98) to give compound 4A, the was added, the reaction mixture stirred for 3 h and concen-
diol intermediate, as a white solid (3.17 g, 80% yield). 1H NMR trated in vacuo. The residue was purified by silica gel column
(CDCl3): δ 9.07 (s, 1H), 7.65 (d, J = 7.9 Hz, 1H), 6.07–5.89 (m, chromatography using a gradient elution (CH2Cl2 : MeOH;
1H), 4.77–4.44 (m, 1H), 4.23 (dd, J = 10.0, 5.0 Hz, 1H), 100 : 0 to 98 : 2) to give compound 7 as a yellow solid
1
4.07–3.97 (m, 1H), 3.91 (ddd, J = 11.5, 5.5, 2.8 Hz, 1H), (162.0 mg, 59% yield). H NMR (CDCl3): δ 11.45 (br s), 9.90 (br
3.87–3.79 (m, 1H), 3.79–3.63 (m, 3H), 3.46 (dd, J = 14.5, 2.8 Hz, s), 8.92 (br s), 8.08 (br s), 7.98 (br s), 7.26 (br s), 6.13 (br s),
3H), 1.11–0.75 (m, 18H), 0.27 to −0.05 (m, 12H). 13C NMR 4.37 (br s), 4.11 (br s), 4.00 (br s), 3.97 (br s), 3.86 (br s), 3.46
(CDCl3): δ 163.58, 163.45, 149.92, 149.87, 138.32, 138.19, (br s), 2.18 (br s), 2.13 (br s), 1.04 (br s), 0.93 (br s), 0.89 (br s),
113.76, 87.84, 87.71, 85.43, 83.46, 83.37, 70.20, 70.07, 69.98, 0.21 (br s), 0.19 (br s), 0.14 (br s), 0.13 (br s). 13C NMR (CDCl3):
69.73, 65.79, 62.70, 62.56, 58.48, 58.45, 26.27, 25.93, 18.77, δ 170.34, 162.59, 149.30, 146.90, 141.72, 136.68, 136.01,
18.36, −4.40, −4.43, −4.55, −5.06, −5.10, −5.13. HRMS-ESI: 133.89, 131.40, 129.73, 129.15, 127.73, 104.30, 88.24, 86.11,
calcd for C24H46N2O8Si2Na (M
569.2688.
+
Na)+ 569.2685, found 83.15, 77.55, 77.23, 76.91, 73.48, 69.94, 62.49, 58.17, 31.52,
29.54, 26.03, 25.91, 25.74, 25.63, 25.56, 18.35, 17.97, 8.29,
To a solution of 4A (790.0 mg, 1.444 mmol) in CH2Cl2 −4.79, −4.82, −5.34, −5.45. HRMS-ESI: m/z calcd for
(50 mL) and H2O (5 mL) at 22 °C was added NaIO4 (4.015 g, C35H57N5O7Si2 (M + H)+ 715.3791, found 715.3776.
18.77 mmol) and the resulting mixture was stirred vigorously
ImUm
for 16 h. The reaction was diluted with H2O (25 mL) and
extracted with CH2Cl2 (3 × 50 mL). The combined organic To a solution of compound 7 (320 mg, 0.4475 mmol) in THF
layers were dried over Na2SO4, filtered and the solvent evapor- (32 mL) was added tert-butyl ammonium fluoride (1 M in THF,
ated in vacuo. The crude product was purified by flash column 1.03 mL, 0.985 mmol) and the solution was stirred at 22 °C for
chromatography (silica gel) using
a
gradient elution 16 h. The solvent was removed under reduced pressure to give
(CH2Cl2 : pet. ether; 50 : 50 to 100 : 0) to give compound 4 as a a sticky reddish oil. The crude product was purified by flash
1
white solid (610 mg, 82% yield). H NMR (CDCl3): δ 10.01 (s, column chromatography (silica gel) using a gradient elution
1H), 8.90 (s, 1H), 8.48 (s, 1H), 6.03 (d, J = 4.6 Hz, 1H), 4.25 (t, (CH2Cl2 : MeOH; 100 : 0 to 95 : 5) to give compound ImUm as a
1
J = 4.7 Hz, 1H), 4.08 (dt, J = 4.6, 2.4 Hz, 1H), 3.96 (dd, J = 11.7, yellow solid (175 mg, 80% yield). H NMR (d6-DMSO): δ 12.45
2.3 Hz, 1H), 3.86–3.69 (m, 2H), 3.46 (s, 3H), 0.92 (t, J = 6.9 Hz, (br s), 12.00 (br s), 8.77 (br s), 7.90 (br s), 7.71 (br s), 7.54 (br
19H), 0.13 (dd, J = 23.0, 6.1 Hz, 12H). 13C NMR (CDCl3): s), 6.00 (br s), 5.74 (br s), 5.22 (br s), 4.17 (br s), 3.95 (br s),
δ 185.69, 161.74, 149.40, 145.65, 111.81, 88.08, 86.18, 84.24, 3.74 (br s), 3.65 (br s), 3.42 (br s), 3.32 (br s), 3.17 (br s). 13C
69.91, 62.31, 58.57, 26.34, 25.91, 18.82, 18.33, −4.41, −4.52, NMR (d6-DMSO): δ 161.86, 149.30, 132.60, 130.54, 128.65,
−5.21, −5.22. HRMS-ESI: m/z calcd for C23H42N2O7Si2Na 127.76, 86.68, 85.14, 82.82, 68.27, 60.53, 57.57, 55.17.
(M + Na)+ 537.2423, found 537.2432.
HRMS-ESI: m/z calcd for C23H29N5O7 (M + H)+ 487.2061, found
487.2068.
Compound 6
Compound 8
To a solution of compound 4 (876.0 mg, 1.701 mmol) and
1,1,3,3-tetramethylisoindoline-5,6-diamine (5) (349.0 mg, ImUm (150.0 mg, 0.31 mmol), DMTCl (189.0 mg, 0.56 mmol)
1.701 mmol) in MeOH (9 mL) was added K3Fe(CN)6 (672.0 mg, and DMAP (4.0 mg, 0.032 mmol) were weighed into a round
2.041 mmol) and the reaction stirred for 16 h at 22 °C. The bottom flask and kept in vacuo for 12 h. Pyridine (2.0 mL) was
reaction mixture was concentrated in vacuo and the crude added and the solution was stirred at 22 °C for 3 h, after which
product was purified by flash column chromatography (silica MeOH (100 μL) was added. The solvent was removed in vacuo
gel) using a gradient elution (CH2Cl2 : MeOH; 98 : 02 to 90 : 10) and the residue was purified by flash column chromatography
to give compound 6 as a yellow solid (590.0 mg, 50% yield). 1H (silica gel) using a gradient of (CH2Cl2 : MeOH; 100 : 0 to
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
97 : 2.5 + 0.5% Et3N); the column was prepared in 0.5% Et3N After manual coupling, the column was remounted on the
in CH2Cl2. Compound 8 was obtained as a yellow solid synthesizer and the synthesis cycle completed. Upon com-
1
(195.0 mg, 80.0%). H NMR (CDCl3): δ 11.32 (br s), 8.91 (br s), pletion of the synthesis, the oligoribonucleotides were cleaved
8.69 (br s), 7.56 (br s), 7.49 (br s), 6.78 (br s), 6.09 (br s), 5.33 from the solid support and the nucleobases and the phospho-
(br s), 4.50 (br s), 4.50 (br s), 4.20 (br s), 4.13 (br s), 3.71 (br s), diesters deprotected in a 1 : 1 mixture of conc. aqueous NH3
3.60 (br s), 2.98 (br s). 13C NMR (CDCl3): δ 161.97, 157.71, and 8 M MeNH2 in EtOH (2 mL) at 65 °C for 40 min. The
157.66, 149.08, 148.58, 143.83, 141.07, 135.35, 135.15, 129.52, supernatant was collected, the beads washed three times with
129.41, 127.59, 127.11, 126.05, 112.41, 88.12, 86.13, 83.36, a mixture of EtOH : H2O (1 : 1, 300 µL), and the combined
82.81, 68.44, 61.75, 58.59, 54.88, 45.07, 8.81. HRMS-ESI: m/z washings were dried. The 2′-O-TBDMS groups were removed by
calcd for C44H49N5O7 (M + H)+ 789.3368, found 789.3396.
treatment with a mixture of Et3N·3HF : DMF (3 : 1800 µL) at
55 °C for 1.5 h, followed by addition of H2O (200 µL). This
mixture was transferred to a 50 mL Falcon tube and n-butanol
Compound 9
Diisopropyl ammonium tetrazolide (33.0 mg, 0.19 mmol) and (40 mL) was added and stored at −20 °C for 12 h, centrifuged
compound 8 (100.0 mg, 0.13 mmol) were dissolved in pyridine and the solvent decanted from the RNA pellet. The crude RNA
(2 mL), the solvent evaporated in vacuo and residue kept under was subsequently purified by 20% denaturing polyacrylamide
vacuum for 17 h. CH2Cl2 (3.0 mL) and 2-cyanoethyl N,N,N′,N’- gel electrophoresis. The RNA oligonucleotide bands were visu-
tetraisopropyl phosphoramidite (115.0 mg, 0.38 mmol) were alized under UV light, excised from the gel, crushed, and
added. The reaction mixture was stirred at 22 °C for 16 h, eluted from the gel with a Tris buffer (2 × 10 mL; Tris (10 mM,
diluted with CH2Cl2 (10 mL) and washed successively with pH 7.5), NaCl (250 mM), Na2EDTA (1 mM)). The RNA elutions
saturated aq. NaHCO3 (3 × 10 mL) and saturated aq. NaCl (2 × were filtered through a 0.45 mm cellulose acetate membrane
10 mL). The organic layer was dried over anhydrous Na2SO4, (Whatman) and desalted using a Sep-Pak cartridge (Waters
filtered and concentrated in vacuo. The crude solid was puri- Corporation). The dried oligoribonucleotides were dissolved in
fied by precipitation by first dissolving it in CH2Cl2 (0.5 mL), sterile H2O (400 µL) and their final concentrations were calcu-
followed by addition of pet. ether (50 mL). The liquid was dec- lated according to Beer’s law based on UV absorbance of oli-
anted and the operation repeated three times to furnish phos- goribonucleotides at 260 nm. Extinction coefficients were
phoramidite 9 as a pale yellow solid (95 mg, 76%). 1H NMR determined by using the UV WinLab oligoribonucleotide cal-
(CDCl3): δ 11.29 (br s), 9.02 (br s), 8.95 (br s), 7.61 (br s), 7.53 culator (V2.85.04; PerkinElmer). Molecular weights of oligori-
(br s), 6.82 (br s), 6.19, 6.10 (br s), 4.70 (br s), 4.54 (br s), 4.39 bonucleotides were determined by MALDI-TOF analysis
(br s), 4.25 (br s), 3.66 (br s), 3.54 (br s), 2.71 (br s), 2.45 (br s), (Bruker, Autoflex III) after calibration with an external stan-
1.47 (br s), 1.35 (br s), 1.24 (br s), 1.13 (br s). 13P NMR (CDCl3): dard. UV/vis spectra were recorded on a PerkinElmer Lambda
δ 150.73. HRMS-ESI: m/z: calcd for C53H64N7O10P (M + H)+ 25 UV/vis spectrometer. CD spectra were recorded on a JASCO
989.4446, found 989.4411.
J-810 spectropolarimeter at 20 °C with path length of 1 mm
(Hellma), 10 scans, scanned from 500 to 200 nm with response
of 1 s, data pitch of 0.1 nm, and bandwidth of 1.0 nm.
Oligonucleotide synthesis
RNA oligonucleotides were synthesized on an automated
ASM800 DNA/RNA synthesizer (Biosset, Novosibirsk, Russia)
CW-EPR measurements
by using a trityl-off protocol and phosphoramidites with stan- CW-EPR spectra were recorded on a MiniScope MS200 spectro-
dard protecting groups on a 1.0 mmol scale, using 1000 Å CPG meter using 100 kHz modulation frequency, 1.0 G modulation
columns. All commercial phosphoramidites, CPG columns, amplitude, and 2.0 mW microwave power. Each spectrum was
and solutions were purchased from ChemGenes Corporation scanned 100–120 times. The temperature was regulated by a
(Wilmington, MA). ImUm was incorporated into RNA oligoribo- Magnettech temperature controller M01 with an error of
nucleotides by solid phase synthesis using the previously
0.5 °C. The sample was prepared by dissolving spin-labeled,
reported protocol,24 with a slight modification. The activator single-stranded RNA (2.0 nmol) and its complementary strand
5-(benzylthio)-1H-tetrazole that is normally used for RNA syn- (2.4 nmol) in phosphate buffer (10 mM phosphate, 100 mM
thesis was not suitable for ImUm, as we observed a coupling NaCl, 0.1 mM Na2EDTA, pH 7.0; 10 mL, oligonucleotide final
efficiency of less than 5% (as judged by the change of color conc. 200 mM). The resulting sample was annealed by using
during trityl deprotection after coupling of ImUm phosphora- the following protocol: 90 °C for 2 min, 60 °C for 5 min, 50 °C
midite) when using this reagent. Therefore, 5-(ethylthio)-1H- for 5 min, 40 °C for 5 min, 22 °C for 15 min. The samples
tetrazole, the activator generally used in our laboratory for (10 µL) were placed in a quartz capillary (BLAUBRAND
DNA synthesis, was used instead. The spin-labeled phosphora- intraMARK) prior to EPR measurements.
midite was incorporated manually into the oligonucleotides by
pausing the synthesizer program after completion of the prior
PELDOR measurements
cycle, removing the column from the synthesizer, and running The RNA samples for PELDOR measurement were prepared by
the standard activator solution (200 μL) and a solution of the annealing 10 nmol of each strand with 10 nmol of its comp-
spin-labeled phosphoramidite (0.05 M, 200 μL) in 1,2-dichlor- lementary strand in phosphate buffer (100 µL, 10 mM, pH 7.0),
oethane back and forth through the column for 10–12 min. NaCl (100 mM), and EDTA (0.1 mM), followed by evaporation
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
of the water. The annealed and dried samples were dissolved
in 20% ethylene glycol/H2O (100 µL) before the PELDOR
measurements. The dead-time free four-pulse PELDOR
sequence was used for all experiments40 and carried out on a
Bruker Elexsys E580 X-band spectrometer equipped with
8 B. Endeward, A. Marko, V. P. Denysenkov, S. T. Sigurdsson
and T. F. Prisner, in Methods Enzymol., ed. P. Z. Qin and
K. Warncke, Academic Press, 2015, vol. 564, pp. 403–425.
9 T. F. Prisner, A. Marko and S. T. Sigurdsson, J. Magn.
Reson., 2015, 252, 187–198.
Flexline MS-3 probe in an Oxford CF935 cryostat and a 10 G. Jeschke, Annu. Rev. Phys. Chem., 2012, 63, 419–446.
PELDOR frequency unit. Microwave pulses were amplified by a 11 G. Z. Sowa and P. Z. Qin, Prog. Nucleic Acid Res. Mol. Biol.,
1 kW TWT amplifier (ASE 117x). Typical pulse lengths were 32
2008, 82, 147–197.
ns (π/2 and π) for the probe pulses and 12 ns (π) for the pump 12 S. T. Sigurdsson, Pure Appl. Chem., 2011, 83, 677–686.
pulse. The delay between the first and second probe pulses 13 S. A. Shelke and S. T. Sigurdsson, in Struct. Bonding, ed.
was varied between 132 and 196 ns in 8 ns steps to reduce con-
tributions from proton modulations. The pulse separation
C. Timmel and J. Harmer, Springer, Berlin, 2011, vol. 152,
pp. 121–162.
between the second and third probe pulses was 3.0 μs. The fre- 14 S. A. Shelke and S. T. Sigurdsson, Eur. J. Org. Chem., 2012,
quency of the pump pulse was fixed to the central maximum 2291–2301.
of the nitroxide powder spectrum to obtain maximum 15 N. R. Kamble, M. Granz, T. F. Prisner and S. T. Sigurdsson,
pumping efficiency. The probe frequency was chosen Chem. Commun., 2016, 52, 14442–14445.
40–85 MHz above this frequency. This range corresponds to 16 C. R. Allerson, S. L. Chen and G. L. Verdine, J. Am. Chem.
the smallest frequency offset that avoids a strong pump/probe Soc., 1997, 119, 7423–7433.
frequency overlap, and therefore large proton modulation arti- 17 A. Ramos and G. Varani, J. Am. Chem. Soc., 1998, 120,
facts. The 85 MHz offset is the frequency offset that excites the 10992–10993.
edge of the nitroxide spectrum. All PELDOR experiments were 18 T. E. Edwards, T. M. Okonogi, B. H. Robinson and
carried out at 50 K.
S. T. Sigurdsson, J. Am. Chem. Soc., 2001, 123, 1527–1528.
19 P. Z. Qin, S. E. Butcher, J. Feigon and W. L. Hubbell,
Biochemistry, 2001, 40, 6929–6936.
20 N. K. Kim, A. Murali and V. J. DeRose, Chem. Biol., 2004,
11, 939–948.
Conflicts of interest
There are no conflicts to declare.
21 G. Sicoli, F. Wachowius, M. Bennati and C. Hobartner,
Angew. Chem., Int. Ed., 2010, 49, 6443–6447.
22 M. Kerzhner, D. Abdullin, J. Więcek, H. Matsuoka,
G. Hagelueken, O. Schiemann and M. Famulok, Chem. –
Eur. J., 2016, 22, 12113–12121.
Acknowledgements
S. Th. S. acknowledges financial support from the Icelandic 23 T. E. Edwards and S. T. Sigurdsson, Nat. Protocols, 2007, 2,
Research Fund (141062-051). T. F. P. acknowledges financial 1954–1962.
support by the SFB 902 – Molecular Principles of RNA-based 24 C. Hobartner, G. Sicoli, F. Wachowius, D. B. Gophane and
Regulation. We thank Dr S. Jonsdottir for assistance in collect- S. T. Sigurdsson, J. Org. Chem., 2012, 77, 7749–7754.
ing analytical data for structural characterization of the new 25 N. Barhate, P. Cekan, A. P. Massey and S. T. Sigurdsson,
compounds and Dr Subham Saha for assistance with prepa-
ration of this manuscript.
Angew. Chem., Int. Ed., 2007, 46, 2655–2658.
26 P. Cekan, A. L. Smith, N. Barhate, B. H. Robinson and
S. T. Sigurdsson, Nucleic Acids Res., 2008, 36, 5946–5954.
27 O. Schiemann, P. Cekan, D. Margraf, T. F. Prisner and
S. T. Sigurdsson, Angew. Chem., Int. Ed., 2009, 48, 3292–
3295.
References
1 S. R. Eddy, Nat. Rev. Genet., 2001, 2, 919–929.
2 L. He and G. J. Hannon, Nat. Rev. Genet., 2004, 5, 522–531.
3 M. V. Rockman and L. Kruglyak, Nat. Rev. Genet., 2006, 7,
862–872.
4 O. Schiemann and T. F. Prisner, Q. Rev. Biophys., 2007, 40,
1–53.
28 A. Marko, V. Denysenkov, D. Margraf, P. Cekan,
O. Schiemann, S. T. Sigurdsson and T. F. Prisner, J. Am.
Chem. Soc., 2011, 133, 13375–13379.
29 I. Tkach, S. Pornsuwan, C. Höbartner, F. Wachowius,
S. T. Sigurdsson, T. Y. Baranova, U. Diederichsen, G. Sicoli and
M. Bennati, Phys. Chem. Chem. Phys., 2013, 15, 3433–3437.
5 J. P. Klare and H.-J. Steinhoff, Photosynth. Res., 2009, 102, 30 D. B. Gophane and S. T. Sigurdsson, Chem. Commun., 2013,
377–390. 49, 999–1001.
6 Y. Ding, P. Nguyen, N. S. Tangprasertchai, C. V. Reyes, 31 D. B. Gophane, B. Endeward, T. F. Prisner and
X. Zhang and P. Z. Qin, Electron Paramagn. Reson., 2014, 24,
122–147.
7 O. Duss, M. Yulikov, F. H. Allain and G. Jeschke, in
S. T. Sigurdsson, Chem. – Eur. J., 2014, 20, 15913–15919.
32 J. Asakura and M. J. Robins, J. Org. Chem., 1990, 55, 4928–
4933.
Methods Enzymol, ed. S. A. Woodson and F. H. T. Allain, 33 K. Fujimoto, S. Matsuda, N. Takahashi and I. Saito, J. Am.
Academic Press, 2015, vol. 558, pp. 279–331.
Chem. Soc., 2000, 122, 5646–5647.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
34 A. Kittaka, T. Kuze, M. Amano, H. Tanaka, T. Miyasaka, 38 D. Andreatta, S. Sen, J. L. Pérez Lustres, S. A. Kovalenko,
K. Hirose, T. Yoshida, A. Sarai, T. Yasukawa and S. Ishii,
N. P. Ernsting, C. J. Murphy, R. S. Coleman and M. A. Berg,
Nucleosides Nucleotides, 1999, 18, 2769–2783.
J. Am. Chem. Soc., 2006, 128, 6885–6892.
35 J. J. Topczewski, J. G. Kodet and D. F. Wiemer, J. Org. 39 G. Jeschke, V. Chechik, P. Ionita, A. Godt, H. Zimmermann,
Chem., 2011, 76, 909–919.
36 R. Kumar, L. I. Wiebe and E. E. Knaus, Can. J. Chem., 1994,
J. Banham, C. R. Timmel, D. Hilger and H. Jung, Appl.
Magn. Reson., 2006, 30, 473–498.
72, 2005–2010.
40 R. E. Martin, M. Pannier, F. Diederich, V. Gramlich,
M. Hubrich and H. W. Spiess, Angew. Chem., Int. Ed., 1998,
37, 2833–2837.
37 S. R. Holbrook and S.-H. Kim, J. Mol. Biol., 1984, 173, 361–
388.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem.