Site-directed spin-labeling of DNA by the azide-alkyne 'Click' reaction: Nanometer distance measurements on 7-deaza-2′-deoxyadenosine and 2′-deoxyuridine nitroxide conjugates spatially separated or linked to a 'dA-dT' base pair

Article abstract of DOI:10.1002/chem.201001572

Nucleobase-directed spin-labeling by the azide-alkyne 'click' (CuAAC) reaction has been performed for the first time with oligonucleotides. 7-Deaza-7-ethynyl-2′-deoxyadenosine (1) and 5-ethynyl-2′- deoxyuridine (2) were chosen to incorporate terminal triple bonds into DNA. Oligonucleotides containing 1 or 2 were synthesized on a solid phase and spin labeling with 4-azido-2,2,6,6-tetramethylpiperidine 1-oxyl (4-azido-TEMPO, 3) was performed by post-modification in solution. Two spin labels (3) were incorporated with high efficiency into the DNA duplex at spatially separated positions or into a 'dA-dT' base pair. Modification at the 5-position of the pyrimidine base or at the 7-position of the 7-deazapurine residue gave steric freedom to the spin label in the major groove of duplex DNA. By applying cw and pulse EPR spectroscopy, very accurate distances between spin labels, within the range of 1-2nm, were measured. The spin-spin distance was 1.8A±0. 2nm for DNA duplex 17(dA^7,10)·11 containing two spin labels that are separated by two nucleotides within one individual strand. A distance of 1.4A±0.2nm was found for the spin-labeled 'dA-dT' base pair 15(dA⁷)·16(dT⁶). The 'click' approach has the potential to be applied to all four constituents of DNA, which indicates the universal applicability of the method. New insights into the structural changes of canonical or modified DNA are expected to provide additional information on novel DNA structures, protein interaction, DNA architecture, and synthetic biology. 'Clicked' DNA spin labels: Spin labels (spheres in the figure) have been incorporated by click chemistry into DNA duplexes at spatially separated positions or into a 'dA-dT' base pair with high efficiency. Very accurate distances between spin labels, within a range of 1-2nm, were measured by continuous wave and pulse EPR spectroscopy. Copyright

Full text of DOI:10.1002/chem.201001572

FULL PAPER
DOI: 10.1002/chem.201001572
Site-Directed Spin-Labeling of DNA by the Azide–Alkyne ꢀClickꢁ Reaction:
Nanometer Distance Measurements on 7-Deaza-2’-deoxyadenosine and
2’-Deoxyuridine Nitroxide Conjugates Spatially Separated or Linked to
a ꢀdA-dTꢁ Base Pair
Ping Ding,[a, b] Dorith Wunnicke,[c] Heinz-Jꢂrgen Steinhoff,*[c] and Frank Seela*[a, b]
Abstract: Nucleobase-directed spin-la-
beling by the azide-alkyne ꢀclickꢁ
(CuAAC) reaction has been performed
for the first time with oligonucleotides.
7-Deaza-7-ethynyl-2’-deoxyadenosine
(1) and 5-ethynyl-2’-deoxyuridine (2)
were chosen to incorporate terminal
triple bonds into DNA. Oligonucleo-
tides containing 1 or 2 were synthe-
sized on a solid phase and spin labeling
with 4-azido-2,2,6,6-tetramethylpiperi-
dine 1-oxyl (4-azido-TEMPO, 3) was
performed by post-modification in so-
lution. Two spin labels (3) were incor-
porated with high efficiency into the
DNA duplex at spatially separated po-
sitions or into a ꢀdA-dTꢁ base pair.
Modification at the 5-position of the
pyrimidine base or at the 7-position of
the 7-deazapurine residue gave steric
freedom to the spin label in the major
groove of duplex DNA. By applying
cw and pulse EPR spectroscopy, very
accurate distances between spin labels,
within the range of 1–2 nm, were mea-
sured. The spin–spin distance was 1.8ꢀ
containing two spin labels that are sep-
arated by two nucleotides within one
individual strand. A distance of 1.4ꢀ
0.2 nm was found for the spin-labeled
ꢀdA-dTꢁ base pair 15ACTHNUGRTENNUG ACHTUNGTRENNUNG
(dA*7)·16(dT*6).
The ꢀclickꢁ approach has the potential
to be applied to all four constituents of
DNA, which indicates the universal ap-
plicability of the method. New insights
into the structural changes of canonical
or modified DNA are expected to pro-
vide additional information on novel
DNA structures, protein interaction,
DNA architecture, and synthetic biol-
ogy.
0.2 nm for DNA duplex 17ACTHNUTRGNEUNG
(dA*7,10)·11
Keywords: click chemistry · DEER
spectroscopy · EPR spectroscopy ·
oligonucleotides · spin labeling
Introduction
DNA is a polymorphic molecule that forms a variety of
three-dimensional structures such as A-, B-, or Z-DNA as
well as other motifs. Different base sequences can influence
structural parameters such as groove width, local twist, cur-
vature, and mechanical rigidity.[1] These features help pro-
teins to read and recognize one oligonucleotide sequence in
preference to another, not only through the chemical prop-
erties in the positions of nucleotide residues, but also
through sequence-dependent structural features.[2]
During the last decade electron paramagnetic resonance
(EPR) spectroscopy in combination with site-directed spin
labeling (SDSL) has emerged as a powerful tool for investi-
gating RNA/DNA architectures under biological condi-
tions.[3] Recently, pulse EPR spectroscopy has been applied
to DNA, focusing on B/A conformational transitions[4] and
DNA damage,[5] and a nanometer distance ruler based on
folded DNA/RNA has been developed.[6] To apply EPR
spectroscopic protocols to DNA, unpaired electron spins
[a] P. Ding, Prof. F. Seela
Laboratory of Bioorganic Chemistry and Chemical Biology
Center for Nanotechnology
Heisenbergstrasse 11, 48149 Mꢂnster (Germany)
Fax : (+49)251-53406857
E-mail: frank.seela@uni-osnabrueck.de
seela@uni-muenster.de
[b] P. Ding, Prof. F. Seela
Laboratorium fꢂr Organische und Bioorganische Chemie
Institut fꢂr Chemie, Universitꢃt Osnabrꢂck
Barbarastrasse 7, 49069 Osnabrꢂck (Germany)
Fax : (+49)251-53406857
[c] D. Wunnicke, Prof. H.-J. Steinhoff
Fachbereich Physik, Universitꢃt Osnabrꢂck
Barbarastrasse 7, 49069 Osnabrꢂck (Germany)
Fax : (+49)541-969-2656
E-mail: hsteinho@uni-osnabrueck.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001572.
Chem. Eur. J. 2010, 16, 14385 – 14396
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14385

Figure 1. Structures of the ethynyl-substituted nucleosides and 4-azido-TEMPO conjugates.
have to be introduced and several paramagnetic species can
be chosen for this purpose, for example, metal ions, Cu2+ or
FeS clusters, radical centers, and nitroxide spin labels that
are chemically stable and geometrically fairly rigid.[3] Incor-
poration of spin labels into oligonucleotides should occur at
defined positions with high efficiency causing only small
structural perturbations of the DNA structure. However, the
spin label should be rigid enough to reduce motion that ob-
scures physical data.
Continuous wave (cw) EPR analysis of spin-labeled bio-
molecules provides information about the mobility of the ni-
troxide side chain,[3] the polarity of its microenvironment,[7]
and the intra- or intermolecular distances between two spin-
labeled side chains in the distance range of 1–2 nm.[8]
Double electron electron resonance (DEER) spectroscopy
extends the measurable interspin distances to up to 8 nm.[9]
Several approaches have been applied to the spin labeling
of RNA/DNA, as recently reviewed by Klare and Steinh-
off.[10] The nitroxide side-chains can be introduced on the
monomeric level, into phosphoramidites for solid-phase oli-
gonucleotide synthesis[11] or into triphosphates for enzymatic
polymerization,[12] as well as on the polymeric level by post-
modification of oligonucleotides, RNA, or DNA.[6a,13]
A promising approach to the spin labeling of oligonucleo-
tides is the copper(I)-catalyzed Huisgen–Sharpless–Meldal
alkyne–azide cycloaddition (CuAAC), the so-called ꢀclickꢁ
reaction.[14] It has been proven to be an ideal bio-orthogonal
protocol for ligating functional molecules to biological or
nonbiological materials in organic and aqueous reaction sys-
tems.[15]
zapurines have been used to modify the 7-position (purine
numbering is used throughout the manuscript). Likewise,
the 5-position of the pyrimidine base can be used for the
same purpose.[16] Our laboratory has made many contribu-
tions to this field over the years.[17] Recently, it was shown
that reporter groups of moderate size can be introduced
into the major groove of B-DNA by the CuAAC reac-
tion.[18]
Herein we report on the synthesis of oligonucleotides in-
corporating a derivative of 7-deaza-2’-deoxyadenosine (1) or
2’-deoxyuridine (2) bearing ethynyl side-chains and their
post-modification with nitroxide labels (Figure 1).[19] For this
purpose, phosphoramidites of nucleosides 1 and 2 were syn-
thesized and employed in solid-phase synthesis. The post-
synthetic functionalization of the oligonucleotides by ꢀclickꢁ
chemistry was performed with 4-azido-2,2,6,6-tetramethylpi-
peridine 1-oxyl (3, 4-azido-TEMPO) (Scheme 1). For com-
parison, the reaction was studied on the nucleoside level,
which yielded conjugates 4 and 5. Two TEMPO residues
were introduced at distant positions of a single-stranded oli-
gonucleotide. Furthermore, a TEMPO residue was also
ꢀclickedꢁ to each nucleobase of a modified ꢀdA-dTꢁ base pair
within an oligonucleotide duplex incorporating the nucleo-
sides 1 and 2. For both approaches, distance measurements
were performed by using cw and pulse EPR spectroscopy.
Results and Discussion
Synthesis and properties of the monomers: 7-Deaza-7-eth-
ynyl-2’-deoxyadenosine (1) has already been synthesized[19]
but has not been incorporated into oligonucleotides, where-
as the incorporation of nucleoside 2 in oligonucleotides has
To avoid perturbation of the DNA structure, the click re-
action is performed most efficiently when the ligand is intro-
duced into the major groove of DNA. Consequently, 7-dea-
Scheme 1. CuI-catalyzed alkyne–azide cycloaddition ꢀclickꢁ reaction. TBTA=tris(benzyltriazolylmethyl)amine; DIPEA=N,N-diisopropylethylamine.
14386
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14385 – 14396

Spin-Labeling of DNA by ꢀClickꢁ Reaction
FULL PAPER
Scheme 2. Synthesis of phosphoramidite building block 8. DMT-Cl=4,4’-dimethoxytrityl chloride.
previously been described.[16] Herein, we report the conver-
sion of nucleoside 1 into the phosphoramidite building
block 8, which was employed in solid-phase oligonucleotide
synthesis. For amino group protection of nucleoside 1, the
isobutyric residue was chosen[17a] and transient protection
was employed to give 6 in a yield of 72% (Scheme 2). Com-
pound 6 was converted into the DMT derivative 7 under
standard conditions in a yield of 65%. Phosphitylation with
2-cyanoethyl N,N-diisopropylchlorophosphoramidite fur-
nished the phosphoramidite 8 in a yield of 83% (Scheme 2).
completion of the reaction within 4 h. CuI has been used as
the copper(I) source instead of the CuIISO4/ascorbic acid
system to avoid reduction of the nitroxide radical by ascor-
bic acid to the nonparamagnetic hydroxylamine derivative
during the click reaction.[10,21] The spin-labeled 1,2,3-triazolyl
nucleoside conjugate 4 was obtained in a yield of 64%,
whereas 5 was formed in a yield of 55%.
All compounds were characterized by UV, 1H, and
13C NMR spectroscopy, mass spectrometry, and elemental
analysis (see the Experimental Section, Table 2, and the
The phosphoramidite
9 was
prepared according to a pub-
lished procedure.[16]
Next the click reaction was
performed on nucleosides 1 and
2 with 4-azido-2,2,6,6-tetrame-
thylpiperidine 1-oxyl (3, 4-
azido-TEMPO). The paramag-
netic radical, which contains an
unpaired electron and a ꢀclicka-
bleꢁ azido function, was synthe-
sized from 4-hydroxy-2,2,6,6-
tetramethylpiperidine 1-oxyl ac-
cording to a literature proce-
dure.[20] Then the 7-ethynyl nu-
cleosides 1 and 2 were function-
alized with 3 by the click reac-
tion to give the conjugates 4
and 5, respectively, in the
presence of CuI in a 3:1:1 mix-
ture
(Scheme 3).
of N,N-diisopropylethylamine
of
THF/tBuOH/H2O
The addition
(DIPEA) was essential for the Scheme 3. Functionalization of nucleosides 1 and 2 with 4-azido-TEMPO (3).
Chem. Eur. J. 2010, 16, 14385 – 14396
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14387

H.-J. Steinhoff, F. Seela et al.
Supporting Information). All the 1H NMR signals of the
spin-labeled compounds are broadened and furthermore
some atoms of the TEMPO residue do not appear in either
Table 1. Tm values of oligonucleotide duplexes containing ethynylated
nucleosides and spin-labeled conjugates.[a]
Duplex
Tm
[8C]
DTm DG8310
[b]
1
[8C] [kcalmolꢁ1
]
the H or the 13C NMR spectra. The signal intensities of the
5’-d(TAG GTC AAT ACT) (10) 43(47)
3’-d(ATC CAG TTA TGA) (11)
–
ꢁ9.7
ꢁ9.9
ꢁ9.5
ꢁ9.2
peaks obtained for the triazole moiety and partially for the
nucleobase are also affected.[22] The monomeric spin-labeled
conjugates 4 and 5 were analyzed by EPR spectroscopy to
confirm the presence of the intact nitroxide label (data not
shown).
5’-d(TAG GTC 1AT ACT) (12) 44(47) +1
3’-d(ATC CAG TTA TGA) (11)
5’-d(TAG GTC AAT ACT) (10) 43(46)
3’-d(ATC CAG 2TA TGA) (13)
0
0
Synthesis and characterization of the oligonucleotides for
spin labeling: A series of oligonucleotides were synthesized
(see Table 1). After cleavage from the solid support, the oli-
gonucleotides were deprotected according to the standard
procedure (25% aq. NH3, 608C, 14–16 h). However, in the
case of oligonucleotides containing 5-ethynyl-2’-deoxyuri-
dine (2), significant amounts of byproducts were formed
during the deprotection with ammonia at elevated tempera-
ture (558C, 14–16 h). As the impurity shows a similar mobi-
lity in HPLC as the target molecule, it was necessary to
change the deprotection conditions to room temperature
(25% aq. NH3, 12 h). Consequently, tBPA-protected phos-
phoramidites were used.[23] The oligonucleotides were detri-
tylated and purified by reversed-phase HPLC. The homoge-
neity of the oligonucleotides was confirmed by reversed-
phase HPLC as well as by MALDI-TOF mass spectrometry
(see the Supporting Information). The base compositions of
the oligonucleotides containing 1 and 2 were determined by
enzymatic hydrolysis with snake venom phosphodiesterase
followed by alkaline phosphatase and subsequent reversed-
phase HPLC chromatography (see the Supporting Informa-
tion).
5’-d(TAG GTC 1AT ACT) (12) 43
3’-d(ATC CAG 2TA TGA) (13)
5’-d(TAG GTC 1AT 1CT) (14)
3’-d(ATC CAG TTA TGA) (11)
47
+4 ꢁ10.6
5’-d(TAG GTC 4AT ACT) (15) 40[c]
3’-d(ATC CAG TTA TGA) (11)
ꢁ3
ꢁ1
ꢁ6
ꢁ2
ꢁ8.6
ꢁ9.3
ꢁ8.0
ꢁ8.7
5’-d(TAG GTC AAT ACT) (10) 42[c]
3’-d(ATC CAG 5TA TGA) (16)
5’-d(TAG GTC 4AT ACT) (15) 37
3’-d(ATC CAG 5TA TGA) (16)
5’-d(TAG GTC 4AT 4CT) (17)
3’-d(ATC CAG TTA TGA) (11)
41[c]
[a] Measured at 260 nm in 0.1m NaCl, 10 mm MgCl2, and 10% glycerol
(pH 7.0) at 5 mm single-strand concentration. Data in parentheses were
measured in 0.1m NaCl and 10 mm MgCl2 buffer (pH 7.0) without glycer-
ol. [b] DG8310 values were determined from the melting curves by using
the software MELTWIN, version 3.0 (J. A. McDowell, 1996). The DG8310
values are given to within an error of ꢀ15%. [c] 10% excess of the un-
modified sequence.
Previously it was reported by our laboratory that the vari-
ous 5-alkynylpyrimidines and 7-alkynylated 7-deazapurines
increase duplex stability by 1–38C per modification.[18,24] In
this work we studied the impact of the ethynyl side-chain of
nucleosides 1 and 2 on duplex stability (Table 1). For this,
the oligonucleotide duplex 5’-d(TAGGTCAATACT)-3’
(10)·3’-d(ATCCAGTTATGA)-5’ (11) was used as reference.
To be consistent with the conditions used in low-tempera-
ture EPR experiments, 10% glycerol (cryoprotectant) was
added to the buffer solution for all the Tm measurements.
From Table 1 it can be seen that the replacement of a dA
residue by 7-deaza-7-ethynyl-2’-deoxyadenosine (1) has a
positive effect on duplex stability (Tm =448C for 12·11). This
effect is more pronounced when multiple positions are
modified with nucleoside 1 (14·11, DTm =28C per modifica-
tion). According to Table 1, the replacement of one dT resi-
due by 5-ethynyl-2’-deoxyuridine (2) has a marginal influ-
ence on duplex stability (Tm =438C for duplex 10·13). A
similar result was obtained with the duplex 12·13, which
contains one ꢀdA-dTꢁ base pair modified with ethynyl-substi-
tuted residues.
in which it was found that glycerol reduces the thermal sta-
bility of duplexes by decreasing the number of water mole-
cules interacting with oligonucleotide solvation sites. This
alters the electrostatic interactions within the polynucleotide
chain and its surrounding counter-ion. The influence of the
spin label on duplex stability (oligonucleotide duplexes con-
taining 4 or 5) will be discussed later.
Introduction of nitroxide labels into the oligonucleotides:
Of the synthetic strategies available for introducing spin
labels into oligonucleotides, spin-labeled phosphoramidites
have already been employed in solid-phase oligonucleotide
synthesis.[11] Other approaches use backbone labeling. Either
hydrogen phosphonates or phosphorothioates are incorpo-
rated into the phosphodiester chain selectively and are then
labeled afterwards with appropriately functionalized spin-la-
beled derivatives.[13b,26] Because phosphorothioates prepared
under standard conditions are diastereomeric, spin-labeled
conjugates with an RP or SP configuration located in differ-
ent environments of the DNA chain are formed. Another
possibility is the application of thiooxo nucleosides as spin-
label targets, for example, 4-thiouridine or 6-thioguanosine
as target sites that are functionalized with methanethiosulfo-
nate spin labels.[27] However, this approach is restricted to
The thermal stability of the duplexes in 0.1m NaCl buffer
with or without glycerol (data in parentheses) was compared
and revealed significant destabilization (10·11, 12·11, and
10·13). These results are in agreement with previous work[25]
14388
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14385 – 14396

Spin-Labeling of DNA by ꢀClickꢁ Reaction
FULL PAPER
Scheme 4. Oligonucleotides labeled with 4-azido-TEMPO (3) by the click reaction. The abbreviations dA*7, dT*6, and dA*7,10 correspond to the modified
nucleoside (4 or 5) and its position within the oligonucleotide sequence indicated by the superscript number.
nucleobases in which oxygen can be replaced by sulfur, for
example, 2’-deoxyuridine or -guanosine. The resulting S–S
bridges were obtained in high yields, but cleavage of the di-
sulfide was observed over longer periods of time.[27] Another
common protocol uses the Sonogashira cross-coupling reac-
tion performed on a CPG resin used for solid-phase oligonu-
cleotide synthesis. In this case, fairly rigid spin labels were
introduced into oligonucleotides.[6a,13d]
were carried out in aqueous THF/H2O/tBuOH (3:1:1) at
room temperature in the presence of the chelate ligand of
CuI–TBTA [tris(benzyltriazolylmethyl)amine] and N,N-di-
AHCTUNGTREGiNNNU sopropylethylamine (DIPEA; Scheme 4). Owing to the del-
icate nature of the nitroxide, no reducing reagent was used
during the ꢀclickꢁ reactions to form the oligonucleotide con-
jugates. Monitoring the click reactions by HPLC showed
that the starting materials were completely consumed after
4 h (Figure 2 and the Supporting Information). Labeled oli-
gonucleotides were further purified by reversed-phase
HPLC (RP-18 column). The formation of click products was
confirmed by MALDI-TOF mass spectrometry and enzy-
matic hydrolysis (see the Supporting Information).
As we wanted to develop an efficient and universal proto-
col for spin labeling, this manuscript reports for the first
time the use of the CuAAC ꢀclickꢁ reaction for direct oligo-
nucleotide spin labeling using nucleobases as target sites.
This approach has the following advantages: 1) The simulta-
neous incorporation of two spins or even more labels in a
highly efficient way, 2) spin labeling can be accomplished at
different nucleobases within one oligonucleotide strand,
3) the target sites for spin labeling (nucleosides 1 and 2)
within one particular oligonucleotide can also be used for
the introduction of other reporter groups, for example, for
FRET studies, and 4) accumulated information relating to
structural changes can be collected by direct comparison of
data obtained from the various labels introduced at identical
positions by using one particular oligonucleotide for post-
modification. The click method, which is presented below,
can be performed on free oligonucleotides in solution, but is
also applicable to solid-support-bound oligonucleotides.
Oligonucleotides with ethynyl-substituted nucleosides 1 or
2 were functionalized with 4-azido-TEMPO (3) in solution
to produce paramagnetic oligonucleotides. The reactions
As discussed above, the ethynyl substituents of nucleo-
sides 1 and 2 have a negligible influence on DNA stability.
The short alkynyl side-chain (2–3 ꢅ), which is shorter than
the depth of the B-DNA major groove (8.8 ꢅ),[2] increases
base-pair stability only slightly (Table 1). To evaluate the in-
fluence of the bulky spin-label modification on duplex sta-
bility, the oligonucleotides incorporating the functionalized
residues 4 and 5, both carrying a spin label, were investigat-
ed (Table 1). A single incorporation of the 7-deazapurine
conjugate 4 at a central position [15
bilizes the ꢀdA-dTꢁ base pair by 38C, whereas the incorpora-
tion of two conjugates of 4 [17
(dA*7,10)·11] does not destabi-
(dA*7)·11] slightly desta-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
lize the duplex further (ꢁ18C per modification). A similar
trend was observed with the 5-substituted 2’-deoxyuridine
conjugate
5
(10·16
ACHTUNGTRENNUNG
duplex 15
N
ACHTUNGTRENNUNG
Chem. Eur. J. 2010, 16, 14385 – 14396
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14389

H.-J. Steinhoff, F. Seela et al.
Figure 3. Room-temperature (T=298 K) cw EPR spectra (9.4 GHz) of
single- and double-labeled samples measured in 0.1m NaCl, 10 mm
MgCl2, and 10% glycerol (pH 7.0). All plots are normalized by ampli-
tude.
Figure 2. HPLC profiles of the CuI-catalyzed ꢀclickꢁ reaction of the oligo-
nucleotide 5’-d(TAG GTC 1AT ACT) (12) with 4-azido-TEMPO (3)
monitored at intervals by HPLC: 20 mL reaction mixture per injection
after a) 0, b) 60, c) 120, and d) 240 min. Gradient: 0–30 min 0–60% B in
A, 30–40 min 60% B in A, 60–0% B in A, flow rate 0.7 mLminꢁ1 (A=
0.1m (Et3NH)OAc (pH 7.0)/MeCN (95:5), B=MeCN).
and double-labeled oligonucleotides are compared in
Figure 3. All the spectra indicate a high mobility of the spin-
labeled side-chains and the presence of only one distinct
motional component; the reorientational freedom of the
spin-labeled side-chains is not restricted by their immediate
microenvironment. Interestingly, for the modified duplex 15-
pair, the Tm value is lowered by 68C. The above results
clearly demonstrate that a single bulky spin label at the 7-
position of a 7-deazapurine or the 5-position of a pyrimidine
does not disturb the DNA duplex structure due to the non-
critical position of the spin label (C-7 of 7-deazapurine and
C-5 of pyrimidine). However, when two spin labels are in-
troduced at both sides of a ꢀdA-dTꢁ base pair, the helix
structure is slightly perturbed as reflected by the decrease in
Tm (Table 1). The base pair is not destroyed as such a phe-
nomenon would lead to a more significant decrease in the
value of Tm (DTm =10–158C).
ACHUTNGRENNUG CAHTUNGTRENNUGN
(dA*7)·16(dT*6), the amplitudes of the signals decrease from
the low-field to the high-field peaks. In addition, considera-
ble broadening of the lines of the spectra of the double-la-
beled strand relative to the spectra of the single-labeled
strand is evident and is a result of spin–spin interactions. Al-
though the double-labeled duplex 15
this anisotropic behavior, the single-labeled duplexes 15-
(dA*7)·11 and 10·16(dT*6) do not display such properties.
(dA*7)·16(dT*6) reveals
ACHUTGTNRENNUG CAHTUNGTRENNUGN
A
ACHTUNGTRENNUNG
Given that the anisotropic behavior is not caused by the
constraints of neighboring nucleobases, the anisotropy pro-
vides evidence of steric interactions between the two spin-
labeled side-chains or of a slightly perturbed ꢀdA-dTꢁ base
pair. The perturbation of the DNA structure has already
been indicated by the lower Tm value (DTm =ꢁ68C;
Table 1). In contrast, the room-temperature spectrum of the
EPR analysis of the spin-labeled oligonucleotides: As melt-
ing temperature data cannot answer questions relating to
nanoscale changes, EPR distance measurements were under-
taken. In this regard, oligonucleotide duplexes with spatially
isolated TEMPO residues or with nitroxides linked to each
site of a base pair were investigated by cw and pulse EPR
spectroscopy. The room-temperature spectra of the single-
modified duplex with distant spin labels 17ACTHNUTRGNEUNG
(dA*7,10)·11 does
not exhibit any anisotropic behavior. No perturbation of the
DNA structure caused by spin labeling and no interaction of
14390
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14385 – 14396

Spin-Labeling of DNA by ꢀClickꢁ Reaction
FULL PAPER
Figure 4. Low-temperature (T=50 K) DEER spectra of spin-labeled duplex 17
ACHTUNGTRENNUNG
F(t). b) Comparison of the distance distributions P(r) obtained by Tikhonov regularization[33] of the pulse EPR spectrum (solid line) and simulated inter-
spin distances using YASARA Dynamics (dotted line). c) Low-temperature (T=160 K) cw EPR powder spectra recorded at the X-band frequency. Ex-
(dA*7)·16(dT*6) (black solid line) and single-labeled oligonucleotide 15(dA*7) (grey solid line). The
perimental spectra of double-labeled DNA duplex 15CAHTUNGTRENUNNG ACHTUNREGTNNUNG ACHTUNGTRENNUGN
corresponding simulated EPR spectra are represented by dashed lines. All plots are normalized by spin number. d) Comparison of the distance distribu-
tions P(r) obtained by Tikhonov regularization (ShortDistances)[31] of the cw EPR spectrum (solid line) and interspin distances using YASARA Dynam-
ics[34] (dotted line). All experiments were performed in 0.1m NaCl, 10 mm MgCl2, and 10% glycerol (pH 7.0). Distance distributions are normalized by
amplitude.
the two spin labels are observable in the spectrum, which is
not unexpected because the two spin labels are at a dis-
tance.
The results of the cw and DEER measurements per-
formed at 160 and 50 K, respectively, are depicted in
Figure 4. Tikhonov regularization of the DEER data for the
data not shown). Therefore we used cw EPR spectroscopy
at 160 K to determine the interspin distance for 15-
ACHUTNGRENNUG CAHTUNGTRENNUGN
(dA*7)·16(dT*6). No dipolar broadening was observed for
the single-stranded oligonucleotide 15ACTHNGUTERNNUG
(dA*7), which is taken
as a reference for the simulation of dipolar broadening. In
contrast, the experimentally observed spectrum of 15-
double-labeled duplex 17
N
(dA*7)·16(dT*6) exhibits significant dipolar broadening (Fig-
ACHUTGTNRENNUG CAHTUNGTRENNUGN
single distance population with an interspin distance of 1.8ꢀ
0.2 nm (Figure 4b). The distance distribution width of
0.2 nm is narrow, half of the width found for an analogous
labeled DNA duplex reported by Flaender et al.[28] In fact, it
is similar to a DNA duplex system with 4-amino-TEMPO
attached to 2-fluorohypoxanthine bases.[5] Hence the pre-
sented DNA system, spin-labeled by the click reaction, is a
powerful tool for measuring accurate interspin distances and
is sensitive even to small structural changes. Furthermore
the rigid spin label will allow orientation-selective studies to
obtain the relative orientation of both spin labels as demon-
strated by Schiemann et al.[29]
ure 4c). Fitting of the simulated dipolar broadened EPR
spectrum (Figure 4c, dashed line) to the experimental spec-
trum by using the Tikhonov regularization approach provid-
ed by the ShortDistances program[31] yielded a mean inter-
spin distance of 1.4ꢀ0.3 nm and a fraction of the single
spin-labeled component of 11% (Figure 4d, solid line). In
addition, the data were fitted with the DipFit program[32] in
which a sum of the Gaussian distributions of interspin dis-
tances is assumed. This results in a distance distribution cen-
tered at 1.4ꢀ0.1 nm and an Azz value of 3.7 mT. The frac-
tion of the single spin-labeled component amounts to 10%.
The high Azz value indicates a high polarity in the immedi-
ate microenvironment of the spin-labeled side-chains. Hence
the spin-labeled side-chains are accessible to water mole-
cules, which is in agreement with the DNA model
The data obtained for the double-labeled duplex 15-
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
(dA*7)·16(dT*6) reveal an interspin distance below the ac-
cessible distance range of DEER spectroscopy (<1.5 nm;[30]
Chem. Eur. J. 2010, 16, 14385 – 14396
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14391

H.-J. Steinhoff, F. Seela et al.
Figure 5. Molecular dynamics (MD) simulation snapshots of the DNA duplexes 15
formed by using the AMBER99 force field implemented in YASARA Dynamics.[34] a) MD simulation snapshot of DNA duplex 15
b) Modified base pair of 4 and 5 and the spin-label distance, determined for the oxygen atoms of the nitroxides. c) MD simulation snapshot of the spin-
labeled ꢀdA-dTꢁ base pair. d) MD simulation snapshot of DNA duplex 17
(dA*7,10)·11. e) Spin–spin distance for 17(dA*7,10)·11 containing two spin labels
within one of the strands, determined for the oxygen atoms of the nitroxides. f) MD simulation snapshot of duplex 17
(dA*7,10)·11.
(dA*7)·16(dT*6) and 17
ACHTUNGTRNEUNGN ACHTNUGTRENNUNG ACTUHNGTRENNUGN
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(Figure 5). The high spin-labeling efficiency is reflected by
the minor single-labeled fraction of 10–11%.
YASARA Dynamics (see the Experimental Section for de-
tails).[34]
From the point of view of physical measurements, a spin
label should be rather rigid and not show significant inde-
pendent movement with regard to the DNA duplex. Short
alkynyl linkers[6a,11a] or phenoxazine moieties[22b] fulfill these
requirements. However, rigid and bulky spin labels affect
the DNA structure. They can stabilize or destabilize base
pairs and provide a ꢀmore or lessꢁ altered picture of the
ꢀrealꢁ DNA duplex structure. On the other hand, a more
flexible spin label is prone to independent movements but
has a smaller influence on the duplex structure. Hence a
suitable compromise has to be found between the ease of
synthesis, the universal character of the protocol, the rigidity
of the label, and the steric perturbation of the DNA struc-
ture induced by the label. We believe that spin labeling
using CuAAC to afford spin labels linked through an s-
1,2,3-triazole moiety fulfills most of these requirements. This
is supported by the very small linewidths of the signals ob-
tained from DEER measurements.
For duplex 15ACTHNGUTRENNUG ACHTGNUTRENNNUG
(dA*7)·16(dT*6), the spin label leads to a
marginal perturbation of the spin-labeled ꢀdA-dTꢁ base pair
during the MD simulation. At the same time, the overall
structure and in particular the base pair remains stable
during the whole MD simulation of 10 ns, as depicted in
Figure 5. The experimental cw EPR spectrum of 15-
ACHUTNGRENNUG CAHTUNGTRENNUGN
(dA*7)·16(dT*6) at room temperature indicates either a
steric interaction between the two spin-labeled side-chains
or a disturbed ꢀdA-dTꢁ base pair. On the basis of the MD
simulation, we can exclude a disturbed base pair and con-
clude that steric interactions lead to the anisotropic charac-
teristics of the spectrum. For 15ACTHNUTRGENNUG CAHTUNGTRENNUGN
(dA*7)·16(dT*6), the simulat-
ed mean interspin distance is centered at 1.3ꢀ0.2 nm, as
shown in Figure 4d (dotted line) and Figure 5b. Thus, there
is reasonable agreement between the experimental and si-
mulated interspin distance distributions with a slight shift of
the simulated interspin distance to shorter distances. Analy-
sis of the data for duplex 17ACHTNUGRTNEUNG
(dA*7,10)·11 exhibits an experi-
mentally determined interspin distance of 1.8ꢀ0.2 nm and a
simulated mean interspin distance of 1.8ꢀ0.2 nm (Figure 4b
and Figure 5e). The agreement between the simulated and
experimental interspin distance distributions is excellent,
both for the interspin distances and the widths of the distri-
butions.
Comparison with molecular dynamics simulation: To ana-
lyze the effects induced by spin labeling on the duplex struc-
ture and to provide a more comprehensive description, mo-
lecular dynamics (MD) simulations were performed on the
DNA duplexes 15ACHTUNGTRENNUNG ACHTUNRTGENNUNG CAHTNUGTRENNUGN
(dA*7)·16(dT*6) and 17(dA*7,10)·11. As a
starting structure both models were built with spin labels ad-
justed to their experimentally determined distance by using
14392
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14385 – 14396

Spin-Labeling of DNA by ꢀClickꢁ Reaction
FULL PAPER
spectrophotometer. lmax and e values are given in nm and dm3 molꢁ1 cmꢁ1
,
Conclusion
respectively. NMR spectra were recorded on a Bruker DPX 300 spec-
1
trometer at 300 MHz for H and 75 MHz for 13C NMR spectroscopy (see
In this study the TEMPO spin label was introduced into
DNA nucleobases 1 and 2 by the copper-assisted azide–
alkyne ꢀclickꢁ reaction. In this way a ꢀdA-dTꢁ base pair was
generated with one spin label linked to each nucleobase. In
addition, one strand of a DNA duplex was modified with
two spin labels at distant positions. Both post-modification
reactions occurred with high efficiency. The spin-labeling
protocol performed on the two ethynylated DNA constitu-
ents 1 (dA*) and 2 (dU*) can be extended to dG and dC de-
rivatives, making all four DNA constituents, two pyrimidine
bases and two purine bases in the form of 7-deazapurines,
accessible. No steric interaction between the two distant
spin labels was observed but a negligible perturbation of the
Table 2). d values are in ppm relative to Me4Si as the internal standard
for 1H and 13C NMR or relative to 85% H3PO4 as the external standard
for 31P NMR; J values are given in Hz. For NMR spectra recorded in
DMSO, the chemical shift of the solvent peak was set to 2.50 ppm for
1H NMR and 39.50 ppm for 13C NMR. Elemental analyses were per-
formed by the Mikroanalytisches Laboratorium Beller, Gçttingen (Ger-
many). Reversed-phase HPLC was carried out on a 4ꢆ250 mm RP-18
(10 mm) LiChrospher 100 column (VWR International) with a Merck-
Hitachi HPLC pump (Model L-6250) connected with a variable wave-
length monitor (model 655A), a controller (model L-500), and an inte-
grator (model D-2500).
Oligonucleotide synthesis, purification, and characterization: The oligo-
nucleotides were synthesized with a DNA synthesizer, model 392–08
(Applied Biosystems, Weiterstadt, Germany), on a 1 mmol scale starting
from the phosphoramidite 8 following the synthesis protocol for 3’-(2-cy-
anoethyl phosphoramidites) (Userꢁs Manual for the 392 DNA synthesiz-
er, Applied Biosystems, Weiterstadt, Germany). Oligonucleotides con-
taining 5-ethynyl-2’-deoxyuridine (2) were prepared by using the corre-
sponding phosphoramidite 9 and a fast deprotection procedure.[16] For
this purpose, 4-tert-butylphenoxyacetyl-protected canonical phosphorami-
dites (Millipore) and the capping reagent 4-tert-butylphenoxyacetic anhy-
dride instead of acetic anhydride were used. The coupling efficiency was
always higher than 95%. After cleavage from the support, the oligomers
were incubated in a 25% aq. NH3 solution. Reactions were performed at
608C for 14–16 h for oligonucleotides containing 1 and at room tempera-
ture for 12 h for oligonucleotides containing 2. The 5’-O-dimethoxytrityl
oligomers were purified by reversed-phase HPLC (Merck-Hitachi
HPLC; RP-18 column; gradient system [A=0.1m (Et3NH)OAc (pH 7.0)/
MeCN 95:5, B=MeCN]: 3 min 20% B in A, 12 min 20–50% B in A, and
25 min 20% B in A; flow rate= 1.0 mLminꢁ1). The purified ꢀtrityl-onꢁ
oligonucleotides were treated with 2.5% CHCl2COOH/CH2Cl2 for 5 min
at 08C to remove the 4,4’-dimethoxytrityl residues. The detritylated oligo-
mers were purified again by reversed-phase HPLC (gradient: 0–20 min
0–20% B in A; flow rate=1 mLminꢁ1). The oligomers were desalted on
a short column (RP-18, silica gel) and lyophilized in a Speed-Vac evapo-
rator to yield colorless solids that were frozen at ꢁ248C. The melting
temperature curves were measured with a Cary-100 Bio UV-VIS spectro-
photometer (Varian, Australia) equipped with a Cary thermoelectric con-
troller. The temperature was measured continuously in the reference cell
with a Pt-100 resistor at a heating rate of 18Cminꢁ1. The thermodynamic
data for duplex formation were calculated by using the Meltwin 3.0 pro-
gram. The oligonucleotides containing 1, 2, 4, and 5 were enzymatically
hydrolyzed with snake venom phosphodiesterase (EC 3.1.15.1, Crotallus
adamanteus) and alkaline phosphatase (EC 3.1.3.1, Escherichia coli from
Roche Diagnostics GmbH, Germany) in 0.1m Tris–HCl buffer (pH 8.3)
at 378C, as described previously.[17a] The hydrolysis product was analyzed
by reversed-phase HPLC (RP-18). The constituents were quantified from
the peak areas divided by the extinction coefficients e260 of the nucleo-
sides: dA 15400, dC 7300, dG 11700, dT 8800, 1 11100 (MeOH), 2
10500 (MeOH), 4 10500 (MeOH), 5 7600 (MeOH). MALDI-TOF mass
spectra were recorded in the linear negative mode with an Applied Bio-
systems Voyager DE PRO spectrometer with 3-hydroxypicolinic acid (3-
DNA structure was noted [duplex 17ACHTUNGTRNEUNG
(dA*7,10)·11]. However,
steric interactions between the spin labels and a small DNA
structure perturbation were observed when two spin labels
were linked to the ꢀdA-dTꢁ base pair. By cw and pulse EPR
spectroscopy, we determined mean interspin distances of
(1.4ꢀ0.3) nm for 15
(dA*7)·16
E
17
ACHTUNGTRENNUNG
tions, the determined distance distribution is exceptionally
narrow and suitable for the identification of even small
structural changes. Hence the spin-labeled DNA system ob-
tained by the ꢀclickꢁ reaction will allow us to obtain detailed
insights into the structural changes caused to DNA struc-
tures by mispairing, DNA damage, and/or lesions. More-
over, the target sites for spin labeling (ethynylated nucleo-
sides 1 and 2) within one particular oligonucleotide can also
be used for the introduction of other reporter groups. Accu-
mulated information regarding structural DNA/RNA
changes can be collected by direct comparison of data ob-
tained from the various labels. For this reason, the click ap-
proach has advantages over other protocols described earli-
er.
Experimental Section
General: All chemicals were purchased from Acros, Aldrich, Sigma, or
Fluka (Sigma–Aldrich Chemie GmbH, Deisenhofen, Germany). Solvents
were of laboratory grade. Thin-layer chromatography (TLC) was per-
formed on TLC aluminium sheets covered with silica gel 60 F254
(0.2 mm, VWR International, Germany). Flash column chromatography
(FC) was performed on silica gel 60 (VWR International, Darmstadt,
Germany) at 0.4 bar. UV spectra were recorded on a Hitachi U-3000
Table 2. 13C NMR chemical shifts (d) of the alkynylated nucleosides and spin-labeled derivatives.[a,b]
C-2[c] C-2[d] C-4[c] C-6[d] C-4a[c] C-5[d] C-5[c] C-7[d] C-6[c] C-8[d] C-7a[c] C-4[d]
C C
Triazole
C-1’ C-3’ C-4’ C-5’ C=O ACHTUNGTRENNUNG(MeO)2Tr
ꢂ
1[19] 152.7
157.5
144.5
157.7
135.8
151.6
151.6
102.3
97.5
93.9
–
127.0
–
149.1
161.6
150.3
160.8
151.1
151.0
82.9, 77.3
83.6, 76.4
–
–
–
83.2 70.9 87.5 61.8
84.7 69.9 87.5 60.7
–
–
–
–
–
–
–
–
–
2[35] 149.5
4
5
6
7
152.3
149.3
151.6
151.6
106.0
105.2
111.3
111.3
99.6
–
96.2
96.2
119.1
–
131.2
131.1
142.7, 118.4[e] 82.3 70.8 87.1 61.9
–
139.6, 119.8[e] 84.5 70.3 87.3 61.2
82.1, 77.3
82.1, 77.1
–
–
83.1 70.9 87.6 61.7 175.8
83.1 70.6 85.5 64.1 175.8 55.0
[a] Measured in [D6]DMSO at 298 K. [b] The signal of C-2ꢁ is superimposed by DMSO. [c] Systematic numbering. [d] Purine numbering. [e] Decreased
signal intensity by spin label.
Chem. Eur. J. 2010, 16, 14385 – 14396
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14393

H.-J. Steinhoff, F. Seela et al.
HPA) as matrix. The masses determined were identical to the calculated
values (see the Supporting Information).
0.51 and 0.42 mT widths, respectively. The fraction of the single spin-la-
beled component is variable. The interspin distance distribution is calcu-
lated by Tikhonov regularization using the ShortDistances program.[31]
The interspin distance as well as the width and the fraction of the single
spin-labeled component are adjustable during the fitting procedure.
Preparation of oligonucleotides for EPR measurements: Lyophilized
DNA samples were dissolved in 0.1m NaCl, 10 mm MgCl2, and 10% glyc-
erol (pH 7.0) to yield a concentration of 60 mm for the single-stranded oli-
gonucleotides and 60 mm for the DNA duplexes, respectively. Glycerol
was added as a cryoprotectant during all low-temperature EPR experi-
ments. For low-temperature cw and pulse EPR experiments, 30–40 mL of
the sample solutions were used in EPR quartz capillaries with 4 and
3 mm outer diameters, respectively, and frozen in liquid nitrogen before
insertion into the resonator. Sample volumes of 10 mL were loaded into
EPR quartz capillaries with a 0.9 mm inner diameter for cw EPR meas-
urements at room temperature.
Analysis of the DEER data: Analysis of the DEER data revealed inter-
spin distances in the range of 1.5–8 nm based on the dipolar coupling fre-
quency of dipolar coupled spins. The lower limit of the DEER experi-
ments depends on the excitation bandwidth of the pump pulse, which has
to be larger than the dipolar coupling of the spins and is in our case
1.5 nm.[30] To elucidate only interspin distances within one nanoscopic
object, the intermolecular background contribution has to be separated
from the intramolecular contribution. Therefore the experimental echo
decay was background-corrected by using a homogeneous three-dimen-
sional spin distribution followed by normalization of the function. Finally,
the interspin distance distributions were calculated by fitting the correct-
ed dipolar evolution function using Tikhonov regularization as imple-
mented in DEERAnalysis2006.[33]
EPR measurements: Room-temperature cw EPR spectra were recorded
at X-band frequency with a Magnettech Miniscope MS200 X-band spec-
trometer equipped with a rectangular TE102 resonator. To avoid satura-
tion and to obtain EPR spectra with a high signal-to-noise ratio, the mi-
crowave power was set to 10 mW and the B-field modulation amplitude
was adjusted to 0.15 mT. Low-temperature cw EPR spectra for the deter-
mination of the interspin distance in the range of 1–2 nm were recorded
at 160 K using a homemade X-band EPR spectrometer equipped with a
Super High Sensitivity Probehead (Bruker). Temperature stabilization
was achieved by a continuous flow helium cryostat (ESR 900, Oxford In-
struments) in combination with a temperature controller (ITC 503S,
Oxford Instruments). The microwave power was set at 0.2 mW and the
B-field modulation amplitude adjusted to 0.25 mT. A B-NM 12 B-field
meter (Bruker) allowed measurement of the magnetic field. The DEER
experiments were performed at 50 K and X-band frequencies (9.4 GHz)
with a Bruker Elexsys 580 spectrometer equipped with a 3 mm split ring
resonator (ER 4118X-MS3, Bruker). The resonator was overcoupled at
Qꢃ100 as measured by using the Xepr software (Bruker). A continuous
flow cryostat (ESR900, Oxford Instruments) in combination with a tem-
perature controller (ITC 503S, Oxford Instruments) was used for temper-
ature stabilization. The following four-pulse DEER sequence was
used:[36]
Molecular dynamics simulations: The molecular dynamics (MD) simula-
tions were performed by using the AMBER99 force field implemented
in YASARA Dynamics[34] with periodic cell boundaries. A simulation
cell was constructed with 12 ꢅ real space around the DNA model and
filled with water molecules and Na+/Clꢁ counter-ions at locations of the
lowest/highest electrostatic potential during a cell neutralization and pKa
prediction experiment (pH 7.0). After an energy minimization run, the
final 10 ns MD simulations were performed at 298 K with 1 fs time steps
under constant pressure with intermolecular forces being calculated
every 2 fs. Periodic boundary crossing of solute atoms was prevented by
function solute drift. The interspin distances were extracted by using the
macro MD analyses provided by YASARA Dynamics.
7-(2-Deoxy-b-d-erythro-pentofuranosyl)-4-(isobutyrylamino)-5-ethynyl-
7H-pyrroloACHTUNGRTENUNG[2,3-d]pyrimidine (6): Me3SiCl (1.16 mL, 9.10 mmol) was
added to a stirred solution of compound 1[19] (274 mg, 1 mmol) in anhy-
drous pyridine (5 mL) at room temperature. After 45 min, isobutyric an-
hydride (1.16 mL, 7.38 mmol) was introduced and the solution was stirred
for another 2 h. The mixture was cooled to 08C, diluted with H2O
(5 mL), and stirred for 10 min. After the addition of 12% aq. NH3
(5 mL), stirring was continued for 1 h at room temperature. The solvent
was evaporated and the residue purified by flash chromatography (FC;
silica gel, column 10ꢆ3 cm, CH2Cl2/CH3OH, 95:5). Compound 6 was iso-
lated as a colorless solid (248 mg, 72%). Rf =0.42 (CH2Cl2/CH3OH,
90:10); 1H NMR (300 MHz, [D6]DMSO, 258C): d=1.14 (s, 6H; 2 CH3),
2.26 (m, 1H; 2’-Ha), 2.58 (m, 1H; CH), 2.78 (m, 1H; 2’-Hb), 3.43 (m, 2H;
p=2ðuobsÞꢁt1ꢁpðuobsÞꢁt0ꢁpðupumpÞꢁðt1 þ t2ꢁt0ÞꢁpðuobsÞꢁt2ꢁecho
A two-step phase cycling (+<X>, ꢁ<X>) was realized on p/2
ACHTNUGRTEN(NGNU uobs),
whereas for all pulses at the observer frequency the <X> channels were
applied. The pump pulse had a length of 12 ns and the pump frequency
upump was positioned at the center of the resonator dip. This frequency
corresponds to the maximum of the echo-detected nitroxide EPR absorp-
tion spectrum. The observer frequency uobs was set to the low-field local
maximum of the absorption spectrum, which resulted in a 65 MHz offset
with observer pulse lengths of 16 ns for p/2 and 32 ns for p pulses. Time
tꢀ was varied and t1 and t2 were kept constant. Deuterium modulation
was averaged by adding traces at eight different t1 start values starting at
t1,0 =200 ns and incrementing at Dt1 =8 ns. The dipolar evolution time is
given by t=tꢀꢁt1 and data with t >0 were analyzed.
ꢂ
5ꢁ-H), 3.84 (m, 1H; 4ꢁ-H), 4.08 (s, 1H; C CH), 4.37 (m, 1H; 3ꢁ-H), 5.04
(t, 3J
G
G
3
6.62 (t, J
(H,H)=6.4 Hz, 1H; 1ꢁ-H), 8.12 (s, 1H; 6-H), 8.62 (s, 1H; 2-H),
10.21 ppm (s, 1H; NH); UV/Vis (MeOH): lmax (e)=238 (53000), 278 nm
(26000 molꢁ1 m3 cmꢁ1); elemental analysis calcd (%) for C17H20N4O4: C
59.29, H 5.85, N 16.27; found: C 59.03, H 5.84, N 16.35.
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-d-erythro-pentofuranosyl]-4-(iso-
butyrylamino)-5-ethynyl-7H-pyrroloACTHNUGTRNEUNG[2,3-d]pyrimidine (7): Compound 6
Fitting of experimental cw EPR spectra: Fitting of the simulated dipolar
broadened cw EPR spectra to the experimental spectra recorded at tem-
peratures below 200 K revealed average interspin distances in the range
of 1–2 nm, as described before by using the DipFit[32] or ShortDistances
program.[31] The Heisenberg exchange interaction does not lead to signifi-
cant distance errors as long as the through-space distances exceed
1.0 nm.[8,32] However, for distances below 1.0 nm or in case exchange is
facilitated through bonds connecting the nitroxides, the effects of ex-
change interactions have to be considered. DipFit determines the best-fit
parameters for the interspin distances and the distance distributions on
the basis of a Gaussian distribution of distances. During the fitting proce-
dure, the g tensor values, the Axx and Ayy values of the hyperfine tensor,
and the Lorentzian and Gaussian linewidth parameters are fixed to the
(241.1 mg, 0.7 mmol) was dried by repeated coevaporation with anhy-
drous pyridine (3ꢆ5 mL) before dissolving in anhydrous pyridine (5 mL).
4,4’-Dimethoxytrityl chloride (DMT-Cl; 284.6 mg, 0.84 mmol) was added
in three portions to this solution. The remaining solution was stirred for
3 h at room temperature. The reaction was quenched by the addition of
MeOH (2 mL) and the mixture was stirred for another 30 min. The reac-
tion mixture was diluted with CH2Cl2 (2ꢆ10 mL), extracted with 5%
aqueous NaHCO3 (30 mL) followed by H2O (40 mL), dried over Na2SO4,
and then evaporated to dryness. Purification by FC (silica gel, column
15ꢆ3 cm, CH2Cl2/acetone, 95:5!90:10) gave
a colorless foam of 7
(295 mg, 65%). Rf =0.64 (CH2Cl2/CH3OH, 90:10); 1H NMR (300 MHz,
[D6]DMSO, 258C): d=1.15 (s, 6H; 2 CH3), 2.31 (m, 1H; 2’-Ha), 2.67 (m,
1H; CH), 2.81 (m, 1H; 2’-Hb), 3.16 (m, 2H; 5ꢁ-H), 3.72 (s, 6H; OCH3),
values found for the reference spectra 16ACTHNUTRGNEUNG
(dA*7). In detail, Axx and Ayy
3.96 (m, 1H; 4ꢁ-H), 4.07 (s, 1H; C CH), 4.38 (m, 1H; 3ꢁ-H), 5.40 (d, 3J-
ꢂ
were fixed to 0.68 and 0.66 mT, respectively, whereas Azz is variable. The
g tensor values are set to gxx =2.0082, gyy =2.0070, and gzz =2.0022. The
EPR spectra are convoluted with a field-independent line-shape function
composed of a superposition of 28% Lorentzian and 72% Gaussian of
A
R
(m, 4H; Ar-H), 7.16–7.40 (m, 9H; Ar-H), 7.95 (s, 1H; 6-H), 8.60 (s, 1H;
2-H), 10.17 ppm (s, 1H; NH); UV/Vis (MeOH): lmax (e)=235 (56000),
14394
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14385 – 14396

Spin-Labeling of DNA by ꢀClickꢁ Reaction
FULL PAPER
276 nm (18000 molꢁ1 m3 cmꢁ1); elemental analysis calcd (%) for
C38H38N4O6: C 70.57, H 5.92, N 8.66; found: C 70.32, H 5.82, N 8.70.
spectra, Dr. R. Thiele from Roche Diagnostics GmbH, Penzberg, Germa-
ny for the measurement of the MALDI-MS, and Nhat Quang Tran for
oligonucleotide synthesis. Financial support by ChemBiotech, Mꢂnster,
Germany (P.D.) and by the Volkswagenstiftung (I/79 950-952; D.W.) is
gratefully acknowledged.
7-[2-Deoxy-5-O-(4,4’-dimethoxytrityl)-b-d-erythro-pentofuranosyl]-4-(iso-
butyrylamino)-5-ethynyl-7H-pyrroloACTHNUTRGENUGN[2,3-d]pyrimidin-3’-(2-cyanoethyl)-
N,N-diisopropylphosphoramidite (8): Compound 7 (259 mg, 0.4 mmol)
was dissolved in anhydrous CH2Cl2 (3.0 mL) under argon and 2-cya-
noethyl N,N-diisopropylchlorophosphoramidite (185 mL, 0.78 mmol) and
N,N-diisopropylethylamine (150 mL, 0.87 mmol) were added at room tem-
perature. After stirring for 20 min, the reaction mixture was diluted with
CH2Cl2 and washed with 5% aq. NaHCO3 followed by brine. The organ-
ic layer was separated, dried over anhydrous Na2SO4, filtered, and evapo-
rated to dryness. The residue was dissolved in CH2Cl2 (1 mL) and puri-
fied by FC (column 4ꢆ10 cm, CH2Cl2/acetone, 9:1) to yield 8 (287.8 mg,
83%) as a colorless foam; Rf =0.44 (CH2Cl2/acetone, 70:10); 31P NMR
(121.5 MHz, CDCl3, 258C): d=148.87, 148.70 ppm.
[1] a) W. Saenger, Principles of Nucleic Acid Structure, Springer-Verlag,
New York, 1984; b) V. A. Bloomfield, D. M. Crothers, I. Tinoco, Jr.,
Nucleic Acids: Structures, Properties and Functions, University Sci-
ence Books, Sausalito, CA, 1999, pp. 79–108.
[2] G. M. Blackburn, M. J. Gait, D. Loakes, D. M. Williams, Nucleic
acids in chemistry and biology, Vol. 5, The Royal Society of Chemis-
try, Cambridge, 2006, pp. 383–422.
[3] a) R. S. Keyes, A. M. Bobst in Biological Magnetic Resonance: Spin
labeling. The Next Millennium, Vol. 14 (Eds.: L. J. Berliner), Plenum
Press, New York 1989, pp. 283–334; b) O. Schiemann, T. F. Prisner,
Q. Rev. Biophys. 2007, 40, 1–53; c) A. Marko, D. Margraf, P. Cekan,
S. T. Sigurdsson, O. Schiemann, T. F. Prisner, Phys. Rev. E 2010, 81,
021911.
[4] G. Sicoli, G. Mathis, O. Delalande, Y. Boulard, D. Gasparutto, S.
Gambarelli, Angew. Chem. 2008, 120, 747–749; Angew. Chem. Int.
Ed. 2008, 47, 735–737.
[5] G. Sicoli, G. Mathis, S. Aci-Sꢇche, C. Saint-Pierre, Y. Boulard, D.
Gasparutto, S. Gambarelli, Nucleic Acids Res. 2009, 37, 3165–3176.
[6] a) O. Schiemann, N. Piton, Y. Mu, G. Stock, J. W. Engels, T. F. Pris-
ner, J. Am. Chem. Soc. 2004, 126, 5722–5729; b) O. Schiemann, A.
Weber, T. E. Edwards, T. F. Prisner, S. T. Sigurdsson, J. Am. Chem.
Soc. 2003, 125, 3434–3435.
[7] H.-J. Steinhoff, A. Savitsky, C. Wegener, M. Pfeiffer, M. Plato, K.
Mçbius, Biochim. Biophys. Acta Bioenerg. 2000, 1457, 253–262.
[8] H.-J. Steinhoff, Biol. Chem. 2004, 385, 913–920.
[9] P. P. Borbat, J. H. Freed, Chem. Phys. Lett. 1999, 313, 145–154.
[10] J. P. Klare, H.-J. Steinhoff, Photosynth. Res. 2009, 102, 377–390.
[11] a) A. Spaltenstein, B. H. Robinson, P. B. Hopkins, Biochemistry
1989, 28, 9484–9495; b) C. Giordano, F. Fratini, D. Attanasio, L.
Cellai, Synthesis 2001, 565–572.
[12] S. Obeid, M. Yulikov, G. Jeschke, A. Marx, Angew. Chem. 2008, 120,
6886–6890; Angew. Chem. Int. Ed. 2008, 47, 6782–6785.
[13] a) F. Hansske, K. Watanabe, F. Cramer, F. Seela, Hoppe-Seylerꢀs Z.
Physiol. Chem. 1978, 359, 1659–1665; b) P. Z. Qin, S. E. Butcher, J.
Feigon, W. L. Hubbell, Biochemistry 2001, 40, 6929–6936; c) N.-K.
Kim, A. Murali, V. J. DeRose, Chem. Biol. 2004, 11, 939–948; d) N.
Piton, Y. Mu, G. Stock, T. F. Prisner, O. Schiemann, J. W. Engels,
Nucleic Acids Res. 2007, 35, 3128–3143; e) Q. Cai, A. K. Kusnetzow,
W. L. Hubbell, I. S. Haworth, G. P. Gacho, N. Van Eps, K. Hideg,
E. J. Chambers, P. Z. Qin, Nucleic Acids Res. 2006, 34, 4722–4730.
[14] a) H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew. Chem. 2001, 113,
2056–2075; Angew. Chem. Int. Ed. 2001, 40, 2004–2021; b) M.
Meldal, C. W. Tornøe, Chem. Rev. 2008, 108, 2952–3015; c) F. Am-
blard, J. H. Cho, R. F. Schinazi, Chem. Rev. 2009, 109, 4207–4220;
d) R. Huisgen, Angew. Chem. 1963, 75, 604–637; Angew. Chem. Int.
Ed. Engl. 1963, 2, 565–598; e) R. Huisgen in 1,3-Dipolar Cycloaddi-
tion Chemistry, Vol. 2 (Eds: A. Padwa), Wiley, New York, 1984;
f) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708–2711; Angew. Chem. Int. Ed. 2002,
41, 2596–2599.
4-Amino-7-(2-deoxy-b-d-erythro-pentofuranosyl)-5-[1-(2,2,6,6-tetrameth-
yl-1-ylooxypiperidin-4-yl)-1,2,3-triazol-4-yl]-7H-pyrroloACTHNUTRGNE[UNG 2,3-d]pyrimidine
(4): Compound 1 (137 mg, 0.5 mmol) and the 4-azido-2,2,6,6-tetramethyl-
piperidine 1-oxyl radical (3; 118 mg, 0.6 mmol) were dissolved in THF/
H2O/tBuOH (3:1:1, v/v, 5 mL) and then N,N-diisopropylethylamine
(80 mL, 0.46 mmol) was added followed by the addition of copper(I)
iodide (143 mg, 0.75 mmol). The reaction mixture was stirred for 4 h at
room temperature. The solvent was evaporated and the residue was puri-
fied by FC (silica gel, column 4ꢆ10 cm, CH2Cl2/MeOH, 10:1) to give 4 as
a pink solid, which was crystallized from H2O to yield pink crystals
(151 mg, 64%). Rf =0.53 (CH2Cl2/CH3OH, 90:10); 1H NMR (300 MHz,
[D6]DMSO, 258C): d=1.16–1.23 (m, 17H; 4 CH3, 2 CH2, CH, low inten-
sity), 2.41 (br, 2H; 2’-Ha, 2’-Hb), 3.48 (br, 2H; 5ꢁ-H), 3.84 (br, 1H; 4ꢁ-H),
4.38 (br, 1H; 3ꢁ-H), 5.04 (s, 1H; 5ꢁ-OH), 5.31 (s, 1H; 3ꢁ-OH), 6.57 (s,
1H; 1ꢁ-H), 7.39 (br, H; NH2), 7.88 (s, 1H; 6-H), 8.10 (s, 1H; 2-H),
9.03 ppm (br, H; NH2); UV/Vis (MeOH): lmax (e)=245 (14400), 279 nm
(10800 molꢁ1 m3 cmꢁ1); MS (ESI): calcd for C22H31N8O4: 472.25; found:
472.25 [M+H]+.
1-(2-Deoxy-b-d-erythro-pentofuranosyl)-5-[1-(2,2,6,6-tetramethyl-1-
ylooxypiperidin-4-yl)-1,2,3-triazol-4-yl]uracil (5): As described for 4, com-
pound 2 (126 mg, 0.5 mmol) in THF/H2O/tBuOH (3:1:1, v/v, 5 mL) was
treated with the 4-azido-2,2,6,6-tetramethylpiperidine 1-oxyl radical (3)
(118 mg, 0.6 mmol) in the presence of N,N-diisopropylethylamine (80 mL,
0.46 mmol) and copper(I) iodide (143 mg, 0.75 mmol). Purification by FC
(silica gel, column 4ꢆ10 cm, CH2Cl2/MeOH, 10:1) gave 5 as a pink solid,
which was crystallized from H2O to yield light pink crystals (124 mg,
55%). Rf =0.46 (CH2Cl2/CH3OH, 90:10); 1H NMR (300 MHz,
[D6]DMSO, 258C): d=1.13–1.24 (m, 17H; 4 CH3, 2 CH2, CH, low inten-
sity), 2.19 (br, 2H; 2’-Ha, 2’-Hb), 3.59 (m, 2H; 5ꢁ-H), 3.85 (m, 1H; 4ꢁ-H),
4.28 (br, 1H; 3ꢁ-H), 5.03 (s, 1H; 5ꢁ-OH), 5.30 (s, 1H; 3ꢁ-OH), 6.24 (s,
1H; 1ꢁ-H), 8.59 (br, 1H; 6-H), 11.71 ppm (s, 1H; NH); UV/Vis (MeOH):
lmax (e)=291 (14200), 232 nm (18400 molꢁ1 m3 cmꢁ1); MS (ESI): calcd for
C20H29N6O6: 472.20; found: 472.20 [M+Na]+.
General procedure for the cycloaddition of oligonucleotides with 4-
azido-2,2,6,6-tetramethylpiperidine 1-oxyl (3): TBTA ligand (100 mL of a
20 mm stock solution in THF/H2O/tBuOH, 3:1:1), N,N-diisopropylethyla-
mine (2 mL of a 1% stock solution in water), CuI (10 mL, 20 mm stock so-
lution in THF/H2O/tBuOH, 3:1:1), and the 4-azido-2,2,6,6-tetramethylp-
piperidine 1-oxyl radical (3; 50 mL of a 20 mm stock solution in THF/
H2O/tBuOH, 3:1:1) were added to a ss-oligonucleotide (5 A260 units) and
the reaction mixture was stirred at room temperature for 4 h. The reac-
tion mixture was concentrated in a Speed-Vac, dissolved in bidistilled
water (500 mL), and centrifuged for 30 min at 14000 rpm. The superna-
tant solution was collected and further purified by reversed-phase HPLC
with the gradient 0–30 min 0–60% B in A, 30–40 min 60% B in A, 40–
[15] P. M. E. Gramlich, C. T. Wirges, A. Manetto, T. Carell, Angew.
Chem. 2008, 120, 8478–8487; Angew. Chem. Int. Ed. 2008, 47, 8350–
8358.
[16] D. Graham, J. A. Parkinson, T. Brown, J. Chem. Soc. Perkin Trans. 1
1998, 1131–1138.
50 min 60–0% B in A, flow rate=0.7 cm3 minꢁ1
.
[17] a) F. Seela, M. Zulauf, Chem. Eur. J. 1998, 4, 1781–1790; b) X.
Peng, H. Li, F. Seela, Nucleic Acids Res. 2006, 34, 5987–6000; c) F.
Seela, X. Peng, S. Budow, Curr. Org. Chem. 2007, 11, 427–462.
[18] F. Seela, V. R. Sirivolu, P. Chittepu, Bioconjugate Chem. 2008, 19,
211–224.
Acknowledgements
We thank Dr. S. Budow for continuous support throughout the prepara-
tion of the manuscript. We also thank S. S. Pujari for recording the NMR
[19] F. Seela, M. Zulauf, Synthesis 1996, 726–730.
[20] N. G. Bushmakina, A. Y. Misharin, Synthesis 1986, 966–966.
Chem. Eur. J. 2010, 16, 14385 – 14396
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14395

H.-J. Steinhoff, F. Seela et al.
ˇ
[21] M. Kveder, G. Pifat, S. Pecar, M. Schara, P. Ramos, H. Esterbauer,
[30] a) G. Jeschke, Y. Polyhach, Phys. Chem. Chem. Phys. 2007, 9, 1895–
1910; b) J. E. Banham, C. M. Baker, S. Ceola, I. J. Day, G. H. Grant,
E. J. J. Groenen, C. T. Rodgers, G. Jeschke, C. R. Timmel, J. Magn.
Reson. 2008, 191, 202–218.
[31] C. Altenbach, W. L. Hubbell, Biophys. J. 2008, 94, Supplement 1,
826–832.
[32] H.-J. Steinhoff, N. Radzwill, W. Thevis, V. Lenz, D. Brandenburg, A.
Antson, G. Dodson, A. Wollmer, Biophys. J. 1997, 73, 3287–3298.
[33] G. Jeschke, V. Chechik, P. Ionita, A. Godt, H. Zimmermann, J.
Banham, C. R. Timmel, D. Hilger, H. Jung, Appl. Magn. Reson.
2006, 30, 473–489.
Chem. Phys. Lipids 1997, 85, 1–12.
[22] a) P. Cekan, S. T. Sigurdsson, J. Am. Chem. Soc. 2009, 131, 18054–
18056; b) M. Sajid, G. Jeschke, M. Wiebcke, A. Godt, Chem. Eur. J.
2009, 15, 12960–12962.
[23] N. D. Sinha, P. Davis, N. Usman, J. Pꢈrez, R. Hodge, J. Kremsky, R.
Casale, Biochimie 1993, 75, 13–23.
[24] M. Ahmadian, P. Zhang, D. Bergstrom, Nucleic Acids Res. 1998, 26,
3127–3135.
[25] D. Liang, L. Song, Z. Chen, B. Chu, J. Chromatogr. A 2001, 931,
163–173.
[26] S. Nagahara, A. Murakami, K. Makino, Nucleosides Nucleotides
1992, 11, 889–901.
[27] P. Z. Qin, K. Hideg, J. Feigon, W. L. Hubbell, Biochemistry 2003, 42,
6772–6783.
[34] a) E. Krieger, T. Darden, S. B. Nabuurs, A. Finkelstein, G. Vriend,
Proteins Struct. Funct. Bioinf. 2004, 57, 678–683; b) E. Krieger, G.
Koraimann, G. Vriend, Proteins Struct. Funct. Bioinf. 2002, 47, 393–
402.
[28] M. Flaender, G. Sicoli, T. Fontecave, G. Mathis, C. Saint-Pierre, Y.
Boulard, S. Gambarelli, D. Gasparutto, Nucleic Acids Symp. Ser.
2008, 52, 147–148.
[29] O. Schiemann, P. Cekan, D. Margraf, T. F. Prisner, S. T. Sigurdsson,
Angew. Chem. 2009, 121, 3342–3345; Angew. Chem. Int. Ed. 2009,
48, 3292–3295.
[35] V. Borsenberger, M. Kukwikila, S. Howorka, Org. Biomol. Chem.
2009, 7, 3826–3835.
[36] M. Pannier, S. Veit, A. Godt, G. Jeschke, H. W. Spiess, J. Magn.
Reson. 2000, 142, 331–340.
Received: June 4, 2010
Published online: November 29, 2010
14396
www.chemeurj.org
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 14385 – 14396