Trityl radicals: Spin labels for nanometer-distance measurements

Article abstract of DOI:10.1002/chem.201203014

Spin labelling with trityls: To gather information about the structure and dynamics of trityl radicals, spin-labelled polymers were measured with pulsed electron-electron double resonance (PELDOR) and double-quantum coherence (DQC). This study demonstrate

Full text of DOI:10.1002/chem.201203014

COMMUNICATION
DOI: 10.1002/chem.201203014
Trityl Radicals: Spin Labels for Nanometer-Distance Measurements
Gunnar W. Reginsson,[a, b] Nitin C. Kunjir,[b] Snorri Th. Sigurdsson,*[b] and
Olav Schiemann*[a, c]
Structural biology and material sciences are engaging
ever-larger complexes, either isolated, in composites or
within whole cells. Thus, methods that can provide structural
information for these molecular architectures on the rele-
vant length scale are needed. EPR methods, especially
pulsed electron–electron double resonance (PELDOR or
DEER), have emerged as very powerful tools to quantita-
tively measure nanometer distances in the range of approxi-
mately 1.4 to 8 nm.[1–4] Because many of the systems are dia-
magnetic, site-directed spin labelling is required, usually
Figure 1. The chemical structure of the trityl spin label 1 and itꢁs field-
swept X band EPR spectrum.
with nitroxides.[5,6,7] Although nitroxides are relatively stable
radicals that can be readily incorporated into molecules to
be studied by EPR, they have the disadvantage of rather
low sensitivity for EPR-based nanometer-distance measure-
ments and require measurements at cryogenic temperatures.
Furthermore, they are only stable for minutes under the re-
ducing environment that exists within cells,[8] which makes
in-cell distance measurements with nitroxides very demand-
ing.[9]
Herein, an approach that may overcome some of these
limitations by using carbon-centred triarylmethyl (trityl)
radicals instead of nitroxides for nanometer-distance meas-
urements is introduced. Specifically, the tetrathiatriaryl-
methyl radical 1[10] was used as the spin label (Figure 1),
which gave an EPR spectrum with one line only (plus very
weak 13C satellite lines; Figure 1) and has a transverse relax-
ation time TM in the microsecond regime, even at room tem-
perature in the liquid state.[11,12] The trityl radical is stabi-
lized against dimerization by the substituted aryl groups[13]
and the in-cell survival time is in the range of hours.[14,15]
Poly(para-phenyleneethynylene)s (PolyPPEs) was chosen
to evaluate the potential of trityls as spin labels for nanome-
ter distance measurements. The structure and conformation-
al flexibility of this class of polymers, which have found
wide use in material science,[16,17] have previously been stud-
ied by PELDOR in combination with nitroxide label-
ling.[18,19] Two compounds (2 and 3, Scheme 1) were pre-
pared for distance measurements with trityl spin labels.
Compound 2 contains one trityl and a typical nitroxide,
whereas compound 3 has two trityl groups. The reason for
synthesizing a compound with both a trityl and a nitroxide
had two purposes: the combination of a trityl label with a
nitroxide spin label may be useful for biological heterodim-
ers, and because it enables a direct comparison to PELDOR
measurements on bisnitroxides. In both compounds 2 and 3,
the two radicals are connected by a linear tether consisting
of aryl and acetylene units that were linked through a series
of Sonogashira cross-coupling reactions. Synthesis of the
[a] G. W. Reginsson,+ Prof. O. Schiemann
Biomedical Sciences Research Complex
Centre of Magnetic Resonance
key-linking unit
4 (Scheme 1a), which contains heptyl
North Haugh, St. Andrews, KY16 9ST (UK)
groups to facilitate adequate solubility of the biradicals, is
outlined in Scheme S1 in the Supporting Information. Syn-
thesis of the trityl-nitroxide biradical 2 (Scheme 1a) started
with conjugation of nitroxide 12, prepared by coupling 4-io-
dobenzoic acid and 4-amino TEMPO (see the Supporting
Information), to linker 4 to give compound 13. The monoa-
cid trityl radical 1, prepared by limited alkaline hydrolysis of
trityl alcohol 14 and subsequent treatment with TFA (see
the Supporting Information), was coupled with nitroxide 13
to give the trityl-nitroxide biradical 2. For synthesis of trityl
biradical 3, complex 1 was coupled with linker 4 to give
trityl radical 16 (Scheme 1b), followed by a Pd-catalyzed di-
[b] G. W. Reginsson,+ N. C. Kunjir,+ Prof. S. T. Sigurdsson
Science Institute
University of Iceland
Dunhaga 3, 107 Reykjavꢀk (Iceland)
E-mail: snorrisi@hi.is
[c] Prof. O. Schiemann
Institute of Physical and Theoretical Chemistry
University of Bonn
Wegelerstrasse 12, 53115 Bonn (Germany)
E-mail: schiemann@pc.uni-bonn.de
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203014.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Scheme 1. Synthesis of a) the trityl-nitroxide biradical 2 and b) the trityl–trityl biradical 3. HOBT=hydroxybenzotriazole; BOP=benzotriazol-1-yloxy-
tris(dimethylamino)-phosphonium hexafluorophosphate.
merization in the presence of CuI under atmospheric
oxygen, which gave biradical 3.
quency of the inversion pulse on the centre of the trityl res-
onance. Because the trityl peak has a linewidth of only 2 G,
the inversion pulse with an excitation bandwidth of 11 G
should invert all of the trityl spin centres.
Two-pulse Hahn echo field-swept spectra of compounds 2
and 3 confirmed the presence of the radicals (Figure 2). For
compound 2, the width of the nitroxide spectrum enabled
placement of the detection pulses on the nitroxide spectrum,
at 30–90 MHz higher frequencies (Figure 2a) than the fre-
Experimentally, a modulation depth D of 90% was ob-
served independent of the frequency offset for the detection
pulses (Figure 3a). This is twice the modulation depth ob-
served for bisnitroxide systems at X band[20,21] and indicates
that excitation of the trityl is almost complete. Deviation
from the expected 100% is attributed to the inversion pulse
not having an excitation profile of an ideal p pulse (see the
Supporting Information). Fourier transforms of the time
traces showed that the intensity of the perpendicular compo-
nent (q=908) of the dipolar coupling increases, whereas the
intensity of the parallel component (q=08) decreases, if the
detection frequency is moved from 30 to 90 MHz offset
(Figure 3b). This indicates that orientation-selective
PELDOR measurements can still be performed on such
mixed trityl/nitroxide-labelled systems, but with much im-
proved modulation depth compared to bisnitroxides.[22] To
extract the distance distribution, the time traces of the dif-
Figure 2. The two-pulse Hahn echo field swept EPR spectra of a) com-
pound 2 and b) compound 3.
&
2
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Spin Labels for Nanometer-Distance Measurements
COMMUNICATION
time trace (inset in Figure 3c and Table 1). Distributions in
length and flexibility of segments that resulted in the best fit
between simulations and experiment are summarized in
Table S2 in the Supporting Information. These results are in
excellent agreement with those from Jeschke et al., which
studied similar polymers by nitroxide labelling and
PELDOR.[18] This indicates that the size of the trityl-spin
label has no influence on the dynamics of the PolyPPEs, and
that the spin density distribution within the trityl label can
be neglected.
The narrow spectral width of the trityl centre also
prompted us to perform double-quantum coherence (DQC)
measurements on compound 2 (Figure 4).[25,26] DQC EPR is
Figure 3. PELDOR data obtained from 2. a) Background-subtracted time
traces. b) Fourier transformed time traces in a). c) Orientation-averaged
time trace and distance distributions obtained from Deer analysis (blue
line) and by simulations (red line). d) Background-corrected time traces
(blue line) and simulated time traces (broken red line). Experimental
and simulated time traces have been displaced on the vertical axis for
clarity.
ferent frequency offsets (Figure 3a) were added together,
and the resulting orientation-averaged time trace (Fig-
ure 3c) was analysed with Deer analysis[23] (Figure 3c,
inset). The obtained mean distance fits to the distance calcu-
lated with molecular mechanics (Table 1). The small peak in
Table 1. Interspin distances for 2 and 3 obtained from molecular me-
chanics, simulations and from PELDOR and DQC, both in combination
with Deer analysis. All distances are given as a mean value Æ two stand-
ard deviations.
Figure 4. Six-pulse DQC data obtained from 2. a) Background-corrected
time trace (solid line) and simulated time trace (broken line). b) Fourier
transformation of the time traces in a). c) Distance distributions from the
DQC time trace by using Deer analysis (solid line) and the simulation
model (broken line).
rsim [ꢃ]
rPELDOR [ꢃ][a]
rDQC [ꢃ][a]
rMM [ꢃ]
2
3
34.8Æ1.1
48.8Æ1.3
34.8Æ1.0
34.8Æ1.2
48.9Æ1.4
35
51
–
[a] The mean distance Æ two standard deviations is given for the most
probable peak.
an elegant way to solely detect the dipolar electron–electron
coupling in a time-domain experiment by introducing a
double-quantum coherence filter into the pulse sequence. It
was first introduced by the lab of Freed,[27] and it only re-
quires one microwave source. In contrast to bisnitroxide bir-
adicals, for which we were not able to obtain DQC time
traces on a commercial EPR spectrometer (data not shown),
DQC gave very good time traces for compound 2 with
almost complete excitation of the whole Pake pattern (Fig-
ure 4b). This indicates that it might be sufficient for DQC
to completely excite the dipolar coupling on one of the two
spin centres. The distance distribution obtained from Deer
analysis has a mean value and a standard deviation that is in
agreement with the result obtained by PELDOR (Table 1).
In addition, the DQC time trace and distance distribution
were simulated with the same dynamics model as was used
for the PELDOR time traces and gave an excellent fit (Fig-
ure 4a and c).
the distance distribution from the orientation averaged time
trace is attributed to incomplete orientation averaging, be-
cause its peak value corresponds to the over weighted paral-
lel component of the dipolar spectrum.
To obtain more information on the conformational dy-
namics of the system, the PELDOR time traces for 2 were
also simulated with a home-written Matlab program, which
uses a conformational model and takes orientation selectivi-
ty into account.[24] The conformational model for 2 was gen-
erated by treating the backbone and spin labels as a chain
of rigid segments linked by joints (see the Supporting Infor-
mation).[18,19] This dynamics model and one set of EPR pa-
rameters were used to reproduce the orientation-selected
PELDOR time traces (Figure 3d) and provided an excellent
fit to the distance distribution from the orientation-averaged
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

O. Schiemann et al.
Bistrityl radical 3 gave an EPR spectrum with a width of
only 2 G (Figure 2b), which is about a factor 30 narrower
compared to a typical bisnitroxide spectrum with a width of
about 70 G. This reduced width translates into increased
signal intensity for the bistrityl system by roughly the same
factor. DQC measurements on 3 gave a time trace with ex-
cellent signal-to-noise ratio that is a factor of two better
than for a PELDOR time trace obtained for a bisnitroxide
under the same conditions (Figure 5a and the Supporting
served for a bisnitroxide radical at 100 K, making PELDOR
measurements at this temperature for nitroxide labels with
gem-dimethyl structure or in matrices others than ortho-ter-
phenyl unfeasible.[28,29]
In summary, we have demonstrated that trityl radicals can
be successfully used as spin labels for nanometer-distance
measurements by using pulsed EPR. They provided deep
PELDOR modulations in experiments with mixed trityl/ni-
troxide labels and improved sensitivity in trityl/trityl DQC
experiments. The trityl label is certainly more bulky than a
nitroxide, but whether this leads to structural distortions, for
example, in biological systems will depend on the specific
molecular structure and will have to be checked in each in-
dividual case as for nitroxides. Please note that a paper from
the laboratories of Freed and Hubbell on trityl–trityl dis-
tance measurements in the liquid state on proteins ap-
peared, while this paper was being reviewed.[30]
Acknowledgements
G.W.R. acknowledges the School of Biology, University of St. Andrews
for a SORS (Scottish Overseas Research Students) scholarship. S.Th.S.
acknowledges financial support from the Icelandic Research Fund (No.
090026021 and 120001021). O.S. acknowledges the DFG for a grant in
the Priority Research Program SPP1601 and the research councils of the
U.K. for an RCUK fellowship.
Keywords: dipolar coupling · double-quantum coherence ·
EPR spectroscopy
· nanostructures · pulsed electron–
electron double resonance
Figure 5. Six-pulse DQC data obtained from 3. a) Background-corrected
time trace (solid line) and simulated time trace (broken line). b) Fourier
transform of the time traces in a). c) Distance distributions from the
DQC time trace in a) by using Deer analysis (solid line) and the model-
based simulation (broken line). d) Background-corrected DQC time
trace obtained from 3 at 100 K.
[1] G. W. Reginsson, O. Schiemann, Biochem. J. 2011, 434, 353–363.
[2] O. Schiemann, T. F. Prisner, Q. Rev. Biophys. 2007, 40, 1–53.
[3] G. Jeschke, Y. Polyhach, Phys. Chem. Chem. Phys. 2007, 9, 1895–
1910.
[4] J. H. Freed, Annu. Rev. Phys. Chem. 2008, 59, 655–689.
[5] Biological Magnetic Resonance, Vol. 19, Distance Measurements in
Biological Systems by EPR, (Eds.: L. J. Berliner, S. S. Eaton, G. R.
Eaton), Kluwer academic/Plenum Publishers, New York, 2000.
[6] S. A. Shelke, S. T. Sigurdsson, Eur. J. Org. Chem. 2012, 2291–2301.
[7] C. Altenbach, T. Marti, H. G. Khorana, W. L. Hubbell, Science 1990,
248, 1088–1092.
[8] H. Swartz, N. Khan in Biological Magnetic Resonance, Vol. 23, Bio-
medical EPR - Part A: Free Radicals, Metals, Medicine, and Physiol-
ogy (Eds.: S. S. Eaton, G. R. Eaton, L. J. Berliner), Kluwer Academ-
ic, New York, 2005, pp. 197–224.
Information). Measurement on 3 with PELDOR gave only
electron-spin echo envelope modulation (ESEEM) artefacts
due to a strong pulse overlap (see the Supporting Informa-
tion). The DQC time trace for 3 was simulated by using the
same segmented dynamics model as for 2 (Figure 5a and
Table S2 in the Supporting Information).
The distance distributions, obtained from Deer analysis
and from the model-based simulation are in very good
agreement (Figure 5c). The average distance agrees with the
distance obtained from molecular-mechanics modelling
(Table 1).
´
[9] I. Krstic, R. Hꢄnsel, O. Romainczyk, J. W. Engels, V. Dçtsch, T. F.
Prisner, Angew. Chem. 2011, 123, 5176–5180; Angew. Chem. Int.
Ed. 2011, 50, 5070–5074.
The high signal-to-noise ratio and the TM relaxation be-
haviour of 3 prompted us to perform DQC measurements
also at higher temperatures. Because the spin system has to
be immobile on the time scale of the coupling to be ob-
served, the solvent (toluene) still had to be frozen. There-
fore, a temperature of 100 K, at which a time trace with an
excellent S/N of 60:1 could be obtained within 20 h, was
chosen (Figure 5d). Compared to 50 K, the TM was reduced
by a factor of 1.5 and the signal-to-noise ratio was reduced
by a factor of 2.7 only. In contrast, no spin echo was ob-
[10] T. Reddy, T. Iwama, H. Halpern, V. Rawal, J. Org. Chem. 2002, 67,
4635–4639.
[11] A. Fielding, P. Carl, G. Eaton, S. Eaton, Appl. Magn. Reson. 2005,
28, 231–238.
[12] R. Owenius, G. Eaton, S. Eaton, J. Magn. Reson. 2005, 172, 168–
175.
[13] S. Grimme, P. R. Schreiner, Angew. Chem. 2011, 123, 12849–12853;
Angew. Chem. Int. Ed. 2011, 50, 12639–12642.
[14] A. Bobko, I. Dhimitruka, J. Zweier, J. Am. Chem. Soc. 2007, 129,
7240–7241.
[15] Y. Liu, F. Villamena, J. Sun, Y. Xu, I. Dhimitruka, J. Org. Chem.
2008, 73, 1490–1497.
&
4
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Spin Labels for Nanometer-Distance Measurements
COMMUNICATION
[16] W. R. Browne, B. L. Feringa, Nat. Nanotechnol. 2006, 1, 25–35.
[17] Y. Shirai, A. J. Osgood, Y. Zhao, K. F. Kelly, J. M. Tour, Nano Lett.
2005, 5, 2330–2334.
[24] G. W. Reginsson, R. I. Hunter, P. A. S. Cruickshank, D. R. Bolton,
S. T. Sigurdsson, G. M. Smith, O. Schiemann, J. Magn. Reson. 2012,
216, 175–182.
[18] G. Jeschke, M. Sajid, M. Schulte, N. Ramezanian, A. Volkov, H.
Zimmermann, A. Godt, J. Am. Chem. Soc. 2010, 132, 10107–10117.
[19] A. Godt, M. Schulte, H. Zimmermann, G. Jeschke, Angew. Chem.
2006, 118, 7722–7726; Angew. Chem. Int. Ed. 2006, 45, 7560–7564.
[20] B. E. Bode, D. Margraf, J. Plackmeyer, G. Dꢅrner, T. F. Prisner, O.
Schiemann, J. Am. Chem. Soc. 2007, 129, 6736–6745.
[21] G. Hagelueken, W. J. Ingledew, H. Huang, B. Petrovic-Stojanovska,
C. Whitfield, H. Elmkami, O. Schiemann, J. H. Naismith, Angew.
Chem. 2009, 121, 2948–2950; Angew. Chem. Int. Ed. 2009, 48, 2904–
2906.
[25] P. Borbat, J. Freed, Chem. Phys. Lett. 1999, 313, 145–154.
[26] P. Borbat, H. Mchaourab, J. Freed, J. Am. Chem. Soc. 2002, 124,
5304–5314.
[27] S. Saxena, J. Freed, Chem. Phys. Lett. 1996, 251, 102–110.
[28] A. Rajca, V. Kathirvelu, K. K. Roy, M. Pink, S. Rajca, S. Sarkar,
S. E. Eaton, G. R. Eaton, Chem. Eur. J. 2010, 16, 5778–5782.
[29] G. Jeschke, A. Bender, H. Paulsen, H. Zimmermann, A. Godt, J.
Magn. Reson. 2004, 169, 1–12.
[30] Z. Yang, Y. Liu, P. Borbat, J. L. Zweier, J. H. Freed, W. L. Hubbell,
J. Am. Chem. Soc. 2012, 134, 9950–9952.
[22] 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.
[23] G. Jeschke, V. Chechik, P. Ionita, A. Godt, H. Zimmermann, J.
Banham, C. R. Timmel, D. Hilger, H. Jung, Appl. Magn. Reson.
2006, 30, 473–498.
Received: August 24, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

O. Schiemann et al.
Spin labelling with trityls: To gather
information about structure and
dynamics of trityl radicals, spin-label-
led polymers were measured with
pulsed electron–electron double reso-
nance (PELDOR) and double-quan-
tum coherence (DQC). This study
demonstrates that trityl radicals have
great potential as spin labels that elim-
inate some limitations of nitroxide spin
labels (see figure).
EPR Spectroscopy
G. W. Reginsson, N. C. Kunjir,
S. T. Sigurdsson,*
O. Schiemann* ................ &&&& — &&&&
Trityl Radicals: Spin Labels for Nano-
meter-Distance Measurements
&
6
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!