Synthesis and EPR-spectroscopic characterization of the perchlorotriarylmethyl tricarboxylic acid radical (PTMTC) and its 13C labelled analogue (13C-PTMTC)

Article abstract of DOI:10.1039/c6cp07200c

A hydrophilic tris(tetrachlorotriaryl)methyl (tetrachloro-TAM) radical labelled 50% with ¹³C at the central carbon atom was prepared. The mixture of isotopologue radicals was characterised by continuous wave and pulsed X-band electron paramagne

Full text of DOI:10.1039/c6cp07200c

PCCP
View Article Online
View Journal
PAPER
Synthesis and EPR-spectroscopic characterization of
the perchlorotriarylmethyl tricarboxylic acid radical
Cite this:DOI: 10.1039/c6cp07200c
(PTMTC) and its C labelled analogue (13C-PTMTC)†
13
Marwa Elewa,‡ab Nadica Maltar-Strmecki,‡§c Mohamed M. Said,b
ˇ
Hosam A. El Shihawy,b Mohamed El-Sadek,d Juliane Frank,a Simon Drescher,a
Malte Drescher,e Karsten Mader,a Dariush Hinderberger*c and Peter Imming*a
¨
A hydrophilic tris(tetrachlorotriaryl)methyl (tetrachloro-TAM) radical labelled 50% with 13C at the central
carbon atom was prepared. The mixture of isotopologue radicals was characterised by continuous wave and
pulsed X-band electron paramagnetic spectroscopy (EPS). For the pharmaceutical and medical applications
planned, the quantitative influence of oxygen, viscosity, temperature and pH on EPR line widths was studied
in aqueous buffer, DMSO, water–methanol and water–glycerol mixtures. Under in vivo conditions, pH can
be disregarded. There is a clear oxygen dependence of the width of the 12C isotopologue single EPR line in
aqueous solutions while changes in rotational motion (viscosity) are observable only in the doublet lines
of the central carbon of the 13C isotopologue. The tetrachloro-TAM proved to be very stable as a solid.
Its thermal decay was determined quantitatively by thermal annealing. Towards ascorbic acid as a
reducing agent and towards an oocyte cell extract it had a half-life of approx. 60 and 10 min. Thus for
in vivo applications, 50% 13C tetrachloro-TAMs are suitable for selective and simultaneous oxygen and
macroviscosity measurements in a formulation, e.g. nanocapsules.
Received 20th October 2016,
Accepted 8th February 2017
DOI: 10.1039/c6cp07200c
rsc.li/pccp
usefulness for the exploration of the microenvironment of a spin
probe with respect to molecular oxygen concentrations, micro-
1 Introduction
Electron paramagnetic resonance (EPR/ESR, electron spin polarity, microviscosity, and pH.1–3 EPR-based imaging (EPRI)
resonance) spectroscopy is a magnetic resonance method to detect has been used to observe the distribution of free radicals in
and characterize paramagnetic species with unpaired electrons. biological systems.4,5 However, in vivo EPRI is hampered by the
In the context of (bio)medical applications, EPR has shown its lack of metabolically stable, non-toxic spin probes. Moreover, for
meaningful in vivo measurements the signal needs to be strong
and resolved to a degree that allows distinguishing which of the
various microenvironmental parameters cause the observed
effects in the EPR signal.
a
¨
¨
Institut fur Pharmazie, Martin-Luther-Universitat Halle-Wittenberg,
Wolfgang-Langenbeck-Str. 4, 06120 Halle, Germany.
E-mail: peter.imming@pharmazie.uni-halle.de
b Faculty of Pharmacy, Suez Canal University, P.O. 41522, Ismailia, Egypt
c
The Gomberg radical, viz. the unsubstituted triarylmethyl (TAM)
radical,6 dimerizes in solution7,8 and is quickly oxidized.9–11
The multiple splitting of the EPR signal results from electron
spin-hydrogen nuclear spin hyperfine couplings and reduces
signal intensity pronouncedly. So for in vivo EPR applications, a
stable radical with a single strong line, e.g., a member of the
tris(tetrathiaphenyl)methyl radicals, would be ideal in terms of
intensity. At the same time, the information content of a single
line, i.e. an isotropic spectrum, is low. A change in line width, for
¨
¨
Institut fur Chemie, Physikalische Chemie, Martin-Luther-Universitat
Halle-Wittenberg, Von-Danckelmann-Platz 4, 06120 Halle, Germany.
E-mail: dariush.hinderberger@chemie.uni-halle.de
d Faculty of Pharmacy, Zagazig University, Zagazig, Egypt
e Department of Chemistry and Konstanz Research School Chemical Biology,
University of Konstanz, 78457 Konstanz, Germany
† Electronic supplementary information (ESI) available: Syntheses; ratio of peak-to-peak
intensities of central 12C line and 13C doublet lines; dependence of 1/TM relaxation rate
on water/methanol compositions, influence of temperature on 13C-PTMTC line width;
ascorbic acid reduction assay. See DOI: 10.1039/c6cp07200c
‡ Author contributions: M. E. and N. M. S. contributed equally. M. E. did the instance, cannot be ascribed to a particular microenvironmental
syntheses. N. M. S., M. E., J. F. and S. D. performed the EPR and related
parameter easily, but may result from changes of several para-
meters (e.g. viscosity and oxygen concentration). Tetrathiatriaryl-
methyl radicals were synthesized and thoroughly investigated in
experiments. M. D. performed the oocyte experiment. M. E. and N. M. S. wrote
a first draft. P. I. and D. H. wrote the manuscript with input from K. M., M. M. S.,
H. A. El-E. and M. El-S. edited the manuscript.
-
terms of their applicability for oxygen measurements and other
purposes enlisted above.
ˇ
´
§ Present address: Ruder Boskovic Institute, Division of Physical Chemistry,
ˇ
Bijenicka c. 54, 10000 Zagreb, Croatia.
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Tris(tetrachlorotriaryl)methyl (tetrachloro-TAM) radicals, in superposition of the single strong EPR line of the 12C species,
which the hydrogen atoms on the aromatic ring were replaced accompanied by the high field and low field peaks of the
with bulky chlorine atoms, were reported as early as 1967.12 Steric 13C labelled species, Fig. 1B. Here, we report the synthesis of
shielding of the central methyl carbon by the six ortho-chlorine this mixture of isotopologue TAM radicals and characterize the
atoms is responsible for high chemical and thermal stability of impact of some parameters mimicking physiological conditions,
these radicals.13–15 Some applications were reported for radicals e.g., variation of viscosity, pH, and oxygen content.
of the tetrachloro-TAM family such as imaging of oxygen concen-
While 14N, 15N isotope-enriched or -depleted nitroxide radicals
tration in tissue,16,17 detection of superoxide radical anion,18–20 were studied intensively, reports on 13C-enriched trityl radicals
detection of hydroxyl radical,21 spin labelling of amino acids and are limited to estimates of electron and spin density distribu-
peptides,22 and their usefulness for dynamic nuclear polariza- tions, information about the anisotropic hyperfine interactions,
tion (DNP)23–25 and material science.26–29
and ENDOR studies, using triphenyl[13C]methyl and deuterated
The EPR spectra of the tetrachloro-TAM radicals, e.g., the analogues. We here present not only the spectroscopic and
perchlorotriarylmethyl tricarboxylic acid (PTMTC) radical, show physical-chemical characterization but also explore the capability
the intensive single peak typical of TAM radicals substituted in to employ the PTMTC radical and its isotopologue in functional
such a way as to minimize electron–nuclear coupling. The peak in vitro model systems and from in vivo studies under more complex
is accompanied by three symmetrical pairs of satellite peaks conditions. These radicals are known to be toxic. However, they are
stemming from hyperfine coupling with natural abundance also reduced quickly in cell lysates (see below). If they are released
13C at the ipso, ortho, para and meta positions of the phenyl slower from the lipophilic formulations we published than they are
rings, Fig. 1A. Surprisingly, hyperfine coupling to the central reduced after release, toxicity will probably not become an issue.
carbon displayed low intensity under the conditions of the EPR
measurement (no influence on the central line; below saturation
limit). Paniagua et al.30 attributed this fact to fast relaxation
because of the slow tumbling rate of the radical in the solvent
2 Experimental methods
2.1. Materials and general methods for synthesis and
used (H2O/DMSO = 1/1, V/V).
analytical characterization
We wanted to explore further when and how hyperfine
Commercial chemicals used for the synthesis were used without
coupling of the unpaired electron (formally located at the central
further purification unless stated otherwise. Buffer salts were
carbon) occurs with the central carbon atom nuclear spin. To this
obtained from Gruessing (Filsum, Germany). Glycerol was obtained
end, we needed to prepare a tetrachloro-TAM radical labelled with
from VWR International (Fontenay-sous-Bois, France). 13C labelled
13C at the position in question. At the same time, we planned to
chloroform was purchased from Cambridge Isotope Laboratories
exploit the increased spectral information that is inherent in the
Inc. Water was used in doubly distilled quality.
coupling pattern and its interaction with the microenvironment.
For this, we chose to enrich the methyl radical center by 50%
13C only. Through this, we would simultaneously observe two
2.2. General methods for synthesis and analytical
characterization
radical species, PTMTC and 13C-PTMTC, that behave identically
regarding chemical stability, solubility, chromatographic properties, All organic solvents were distilled and dried before use and
and biodistribution. One would expect to detect the spectral stored over molecular sieves (3 Å). Glassware for reactions
Fig. 1 EPR spectra of PTMTC and 13C-PTMTC radical in phosphate buffer, pH 7.4 (c = 1 mM). (A) PTMTC appears as a single line (12C central peak),
accompanied by natural abundance 13C satellite couplings. (B) The 12C central peak of PTMTC and the doublet of 13C-PTMTC stemming from the
hyperfine coupling to the 13C nuclei (I = 1/2), again accompanied by 13C natural abundance couplings.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
under argon atmosphere was oven-dried at 100 1C for 2 h prior 2.4. Measurements at defined oxygen contents
Measurements were conducted as described previously.31
Briefly, PTMTC and 13C-PTMTC buffer solutions were flushed
with either pure nitrogen or defined mixtures of oxygen and
nitrogen at a flow rate of 2 L minꢀ1 for 3 min using septum vials
and cannulae. An anesthesia gas mixer with flow meter tubes
to use, evacuated, and flushed with argon immediately. The
purity of all compounds and the progress of reactions were
monitored by thin layer chromatography (TLC) using silica gel
60 F254 plates (Merck KGaA, Darmstadt, Germany). Visualiza-
tions were accomplished with a UV lamp (254 nm) or iodine
staining, and the Rf values given are uncorrected. Purification
of the compounds was achieved either by crystallization from
appropriate solvents or flash chromatography. Melting points
(Mp) were determined with a Boethius apparatus. NMR spectra
were recorded on an Agilent Technologies VNMRS 400 MHz
¨
(Drager, Lu¨beck, Germany) provided defined gas mixtures. The
partial pressure of oxygen (mmHg) in the gas above the solution
was confirmed by a needle-type optical oxygen microsensor
(Type PSt1, PreSens GmbH, Regensburg, Germany) directly
after the EPR measurements. Oxygen content (%) in the gas
above the solution was calculated assuming ambient pressure.
¨
(Agilent Technologies, Boblingen, Germany. Chemical shifts (d)
are reported in parts per million (ppm) relative to TMS. The
splitting pattern was assigned as follows: s = singlet, bs =
broad singlet, d = doublet, t = triplet, q = quartet, m =
multiplet and coupling constants ( J) are given in Hertz (Hz).
13C NMR chemical shifts were reported as d values (ppm)
relative to the residual nondeuterated solvent peak in the
corresponding spectra (chloroform d = 77.2, methanol d =
49.0, DMSO d = 39.5). Mass spectrometry (MS): Electron spray
ionization (ESI) mass spectra were recorded on a LCQ
Classic (Thermo Finnigan, San Jose, California, USA). Electron
impact (EI) mass spectra were recorded on an AMD 402 (AMD
Intectra GmbH, Harpstedt, Germany) with a medium ioniza-
tion voltage of 70 eV. High-resolution mass spectra were
recorded on an orbitrap XL mass spectrometer (Thermo Fisher
Scientific) with a resolving power of 100 000 at m/z 400, samples
were introduced to the MS by static nano-electrospray ioniza-
tion (ESI). Mass calculations were done for the isotope peak
with the highest intensity. Infrared spectra (IR) spectra
were recorded on an IFS 28 FTIR spectrometer (Bruker, Bill-
erica, USA) with a Thermo Spectra-Tech ATR unit (Thermo
Scientific).
2.5. Continuous wave EPR (CW EPR) spectra
CW EPR spectra of 13C-PTMTC were recorded on a Miniscope
MS 400 benchtop spectrometer (Magnettech, Berlin, Germany)
working at X-band frequencies (nmw B 9.4 GHz). The microwave
frequency (nmw) was recorded with a Racal-Dana frequency counter,
model 2101 (Racal Instruments, Neu-Isenburg, Germany). The
temperature was set and controlled within ꢁ1 K with a TC H03
(Magnettech, Berlin, Germany) control unit, using nitrogen gas
flow. A manganese standard sample, Mn2+ in ZnS (Magnettech,
Berlin, Germany), was used to calibrate the magnetic field of the
spectrometer. The spectra were simulated with a custom-built
program in MATLAB (The MathWorks Inc., Natick, Massachusetts,
USA) using the EasySpin program package for EPR.32
2.6. Pulse X-band EPR
Pulse X-band EPR studies were performed on an X-band Bruker
ELEXSYS 580 spectrometer. The temperature was controlled by
a closed cycle cryostat (ARS AF204). The phase memory time,
TM were measured using a simple Hahn spin echo sequence:
p/2–t–p–t–echo. A microwave pulse length of 28 ns was used for
a p/2 pulse. An initial interpulse delay, t, of 400 ns and an
increment of 8 ns were used. The standard four-step phase
cycle procedure was applied to eliminate measurement arte-
facts from unwanted echoes. Relaxation curves have been fitted
by monoexponential decay function.
2.3. Sample preparation for EPR spectroscopic
characterization
The hydrophilic radicals perchlorotriarylmethyl tricarboxylic
acid (PTMTC) radical and its 13C analogue (13C-PTMTC) were
dissolved (c = 1 mM) in phosphate buffer (PB 50 mM, pH 7.4).
The effect of pH was studied with 1 mM solutions of 13C-PTMTC
in PB (50 mM, pH range from 0 to 14). The impact of viscosity
was investigated with solutions of radicals PTMTC and 13C-PTMTC
3 Results and discussion
3.1. Synthesis of PTMTC and 13C-PTMTC
(c = 1 mM) in a mixture of absolute glycerol and PB (50 mM, The preparation of PTMTC has been reported before.33 For
pH 7.4), the glycerol content ranging from 0% to 80% (m/m). this study, the preparation of PTMTC and the partially labelled
Solutions of PTMTC and 13C-PTMTC radicals (c = 1 mM) were 13C-PTMTC (Fig. 2) was optimized regarding reaction time
prepared in a mixture of PB (50 mM, pH 7.4) and absolute (45 min instead of 20 h), ease (no oleum, one-step instead of
methanol, the methanol content ranging from 0% to 100% three-step procedure for triarylmethane formation), and yield
(V/V). A solution of 13C-PTMTC in glycerol/PB 50 mM, pH 7.4 (lit. 33 reported no yield; we reproducibly obtained the yields
(40/60, m/m) was selected to examine the relation between line shown).
width and the viscosity/temperature ratio. A solution of 13C-PTMTC
Perchlorotriarylmethanes (Fig. 2, 2a and 2b) were synthe-
(c = 1 mM) in PB (50 mM, pH 7.4) was used to study both the sized by Friedel–Crafts alkylation of 1,2,4,5-tetrachlorobenzene
thermal annealing within the temperature range 323–353 K (1) with chloroform (CHCl3) in the presence of aluminum
and the stability against ascorbic acid. A solution of PTMTC in trichloride (AlCl3).34 A mixture of CHCl3 and 13CHCl3 (1 : 1
PB (50 mM, pH 7.4) was used to perform the cell-lysate study ratio) was used to prepare 2b. Reaction of 2a or 2b with an
(c = 0.5 mM).
excess of n-BuLi and tetramethylethylenediamine (TMEDA) in
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Fig. 2 Synthesis of PTMTC and 13C-PTMTC radicals. Reagents and conditions: (i) CHCl3/13CHCl3 (1 : 1 ratio)/AlCl3/160 1C, 39%; (ii) n-BuLi, TMEDA, ethyl
chloroformate, 79%; (iii) 1. Bu4NOH, 2. p-Chloranil, 83%; (iv) 95% H2SO4, 85%; central carbon C is the 50% 13C labelled atom of 13C-PTMTC.
ꢀ
THF at low temperature gave the corresponding trianion. With recorded at room temperature, and the corresponding simulation
ethyl chloroformate, the perchlorotriarylmethane triethyl esters (dashed red line). The g-value, 2.0028 ꢁ 0.0002, and effective
(3a and 3b) were formed18,31 and were then converted to the isotropic hyperfine couplings were determined by simulation of
triester radicals (PTMTE and 13C-PTMTE) by a two-step process: the experimental spectra.32
Conversion into the triaryl-carbanion with a 1 mM solution
The data for the 13C hyperfine couplings are presented in Table 1.
of tetra-n-butylammonium hydroxide (Bu4NOH) in methanol, The values are in accordance with data for natural abundance
followed by oxidation with p-chloranil.35 This is different from 13C couplings reported by Paniagua et al.30 Sabacky et al.36
the synthesis of the tris(tetrathiaphenyl)methyl triethyl ester reported hyperfine coupling values of two related radicals, viz.
radical that was released through the conversion of the corres- the unsubstituted triarylmethyl and 2,6,20,60,200,600-hexamethoxytri-
ponding methane derivative into the carbocation followed by arylmethyl radicals. They had been enriched with 13C (56% and
reduction with stannous chloride.31
54%, respectively) at the central carbon. The hyperfine coupling
Heating PTMTE or 13C-PTMTE in concentrated sulfuric acid of the central carbon for the unsubstituted triarylmethyl radical
yielded the tricarboxylic acid radicals PTMTC and 13C-PTMTC. was 64.46 MHz and 73.43 MHz for the 2,6,20,60,200,600-
hexamethoxytriarylmethyl radical.36
3.2. EPR spectra
The naturally most abundant isotope of carbon, 12C (B98.9%)
3.3. Influence of pH on 13C-PTMTC line width
has a nuclear spin of I = 0 and, hence, does not interact with the Since the PTMTC radical bears an ionizable carboxyl group, it
unpaired electron spin. The EPR spectrum of PTMTC accordingly can be expected to show pH sensitivity. By varying pH, terminal
showed a single peak, in addition to three symmetrical pairs of groups are ionized from carboxylic acid at low pH to carboxylate
satellites because of hyperfine coupling with naturally abundant at high pH and a mixture of both at pH values around the pK.
13C (1.07%) at three different positions/types of carbon, Fig. 1A. Solutions of 13C-PTMTC (c = 1 mM) in PB (50 mM, pH range
The spectrum of the isotopically enriched analogue, 13C-PTMTC, from 2 to 12) were used to investigate the impact of pH on EPR
displayed three primary EPR lines that are a superposition of the spectral parameters such as line widths. The lowest and highest
single line of 12C-isotopologue PTMTC and the doublet derived pH values were adjusted using pure HCl and NaOH solutions.
from the 13C isotopologue, Fig. 1B.
Remarkably, the observed hyperfine splitting and line widths
Fig. 3A shows an experimental CW-EPR spectrum (solid were found to be independent of the solution pH. Only in very
black line) of a solution of 13C-PTMTC in DMSO (c = 1 mM) acidic solution (pH o 2) the EPR signal strength was reduced,
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
Fig. 3 (A) CW-EPR spectrum of 13C-PTMTC (1 mM in DMSO) recorded at room temperature. The experimental spectrum is shown in black (solid black
line), the simulated spectrum in red (dashed red line). (B) CW EPR X-band spectra of 13C-PTMTC recorded at 37 1C for pH = 5.6 (red line) and pH = 1.0
(black line).
Table 1 13C hyperfine coupling values of 13C-PTMTC
13C hyperfine couplings (MHz)
Central carbon
83.5
34.9
29.0
5.9
1-Phenyl 13C bridgehead (ipso)
2,6-Phenyl 13C in ortho position
3,5-Phenyl 13C in meta position
Hyperfine couplings are measured in absolute values. The estimated
error of hyperfine coupling is ꢁ0.1 MHz.
simply because of precipitation, Fig. 3B. Importantly, at physio-
logical pH values the samples were found to be stable for the
duration of the experiments.
The negligible influence of pH can also be seen by following the
ratio of the central 12C peak and of the 13C doublet line amplitudes
at the low field (LF) and high field (HF) positions, respectively,
Fig. 4. Between pH 2 and 12, the ratio is virtually constant, which is Fig. 4 Dependence of the ratio of peak-to-peak amplitudes of the central
12C line and 13C high field and low field doublet lines (filled circles, I(12C)/
a clear indication that the unpaired electron spin distribution is not
ILF( (
13C); open circles, I(12C)/IHF 13C)) of 13C-PTMTC (c = 1 mM) in PB (50 mM)
altered. A change in the ratio would be expected if changes in the
electronic structure, potentially induced through protonation/
deprotonation by varying pH, took place.
on the pH value ranging from pH 2 to 12. Extreme pH values were adjusted
using pure HCl and NaOH solutions.
Hence in vivo, the effect of pH as such on line width and
shape may be neglected, both for the central peak and the 13C (c = 1 mM). For PTMTC, between 10 and 40% (m/m) glycerol
hyperfine coupling lines. This is advantageous as it means that content, in aerated solutions a slight decrease in the EPR signal
there is one parameter less to be considered, but disadvanta- line width was detected (data not shown). This was most likely
geous since it corroborates that carboxylic acid moieties in trityl caused by the decrease in oxygen solubility with increasing
radicals are poor microsensors for pH.
percentage of glycerol,37 whereas above 60% of glycerol (m/m),
The same negligible effect of pH on the EPR line width of the increase in line width was dominated by the pronounced
tris(tetrathiaphenyl)methyl tricarboxylic acid radicals was increase in viscosity. Apparently, viscosity by itself up to the
observed in our previous study.31
value of about 40% of glycerol (m/m) has no effect on the EPR
signal line width. This fact is important for in vivo applications
as blood has a viscosity of 3–4 mPa s at 37 1C,38 which is only
3.4. Influence of viscosity on 13C-PTMTC line width
Different glycerol–water mixtures (0% to 80% glycerol in water, reached at 44% of glycerol in water (m/m) at 20 1C.39 As we
m/m) were used to prepare solutions of PTMTC and 13C-PTMTC found previously for this type of radical,31 pH and viscosity can
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
be disregarded for in vivo EPR signals of this type of radical. 3.5. Influence of oxygen on PTMTC line width
However, microviscosity may play a role in matrices more
The EPR signal line width increase is directly proportional to
the oxygen concentration.43 A solution of PTMTC (c = 1 mM) in
PB (50 mM, pH 7.4) shows a linear relationship between line
width and oxygen concentration (Fig. 6). Again, we only considered
the physiologically relevant range of oxygen content. It is important
to note that the small slope of the curve is not primarily caused by
an insensitivity of the radicals. It is simply due to the fact that in
aqueous solution, the concentration of dissolved oxygen increases
only very little with increasing oxygen partial pressure.44
complex than glycerol–water mixtures (including proteins, cell
membranes, etc.). Reversible plasma protein binding, not studied
here, will probably play a role, esp. for the very lipophilic esters of
the radicals studied here. For tetrathiasubstituted trityl radicals,
oxidative metabolism even led to covalent protein binding.40
Viscosity-dependent measurements (Fig. 5) were performed
under aerated conditions, hence two contributions to linewidth
are present: with increasing viscosity, rotational motion is
slowed down, which is reflected in the low- and high-field lines
of 13C-PTMTC. Also, with higher viscosity collisional broad-
ening by dissolved oxygen is reduced, which is visible in the
reduction of line width with increasing viscosity in the central
line. Apparently, the reduction of collisional broadening leads
to a reduced linewidth and overcompensates for the small
linewidth increase that was found for deoxygenated glycerol/
water solutions.41,42
In contrast, viscosity did have significant influence on line
width in the spectrum of the 13C enriched radical. As depicted in
Fig. 5, the doublet peak line widths were dominated by viscosity
from the beginning. Indeed, beyond approx. 30% glycerol content,
they became too broad for exact line width determination. So
despite the fact that viscosity has no effect on PTMTC EPR signal
line width up to a value of about 40% of glycerol, viscosity and
viscosity changes of a medium can be observed through the ratio
of the intensity of the central 12C peak and the 13C doublet line.
A linear dependence was observed for line width ratio and
temperature between 20 1C and 90 1C. The dependence for two
temperatures only that are relevant for in vivo applications is shown
in Fig. S1 (ESI†). The observable temperature effect (smaller peak
ratio at higher temperature) is mainly due to viscosity changes and
little contribution of oxygen loss with temperature.
The influence of oxygen was also studied in organic solvents.
To examine the influence of solution parameters on the EPR
signal, methanol was chosen as an example solvent. EPR spectra
of methanol–water mixtures (10 to 100% methanol in water, V/V)
including 1 mM of 13C-PTMTC were investigated. In aerated
solutions (Fig. 7A), with increasing MeOH concentration the
line width of both the 12C central peak and the doublet peak
(13C radical signal) increased. This can be explained as follows:
with increasing MeOH concentration, oxygen solubility increases,45
leading to shorter relaxation times and, thus, broader lines because
of the interaction between the paramagnetic oxygen and the spin
probe. Under deoxygenated conditions (Fig. 7B), it was expected
that the line width would not change because the oxygen effect was
excluded. However, the line width of both 13C-PTMTC doublet
peaks increased up to approx. 50% MeOH and decreased beyond.
This pattern follows the increase and decrease of viscosity of
methanol–water mixtures.43 The (slight) reduction, about 4%, of
the line width of the 12C-PTMTC signal follows the deviations
reported for the water–methanol system such as volume contrac-
tion with its maximum at 56% of MeOH46 as well as a parabolic
entropy profile for both methanol and water in the methanol–water
mixtures compared to the pure liquids.47 It should also be noted
that preferential solvation effects for radicals in methanol–water
mixtures may be attributed to nonlinear behavior of the spectral
parameters.48 Similar trends can be observed by looking at the
Fig. 5 Changes in peak to peak line width at different glycerol concen-
trations in aerated solution: 13C-PTMTC (c = 1 mM). The 12C peak (black
squares) as well as the high field (red circles) and low field (blue triangles)
peak of the 13C doublet peak are shown.
Fig. 6 Line width of the EPR signal of PTMTC (c = 1 mM) in PB (50 mM, pH
7.4) at different oxygen contents in the vapour phase. The straight line is
only meant to guide the eye.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
Fig. 7 Effect of increasing MeOH content on the line width of 13C-PTMTC (c = 1 mM); the central peak (black squares) as well as the high field (red
circles) and low field (blue triangles) peak of the doublet peak are shown either under (A) aerobic conditions (20.9% O2) or (B) deoxygenated conditions
(E0% O2).
dependence of the 1/TM relaxation rate (1/TM B linewidth) on study the changes in spectral parameters during accelerated
water–methanol compositions recorded at room temperature (see aging.51 In the temperature range of 323–353 K, the peak-to-
Fig. S2, ESI†). This effect is not evident for the 1/TM relaxation rates peak amplitudes of the central line (Ipp) were used to follow
of 13C-PTMTC doublet peaks due to the stronger influence of other changes in the signal with time. As expected, the decrease in
nuclear relaxation mechanisms including effects of viscosity.
the EPR signal intensity was more pronounced at higher
annealing temperatures. The EPR data presented in Fig. 9A
were analyzed in terms of first-order kinetics:
3.6. Influence of temperature on 13C-PTMTC line width
The effect of temperature on the intensity of the central PTMTC
EPR peak is consistent with Curie’s law, whereas the intensity
of the 13C doublet shows more complex behaviour as can be
seen in Fig. S3, ESI.†
ln(Ipp/Io) = ꢀk(T)ꢂt,
(1)
where Io is the peak height of the central line obtained at time
t = 0 from the first derivative of the spectrum, and k is the rate
constant, which is a function of the absolute temperature, T.
The kinetic parameters for the process (ko, DE) were derived by
fitting the obtained rate constants ki(T) to a simple Arrhenius-
type equation
The effect of viscosity changes on the 13C-PTMTC signal is
pronounced, leading concomitantly to an increase of the
amplitude of both the HF and LF parts of the doublet by a
factor of approx. 5 (data not shown) and a decrease of their line
widths by a factor of approx. 3. The linear relation between the line
width and the viscosity/temperature ratio as described by the
Stokes–Einstein equation is depicted in Fig. 8. As can be expected,
DE
RT
kðTÞ ¼ ko ꢂ eꢀ
;
(2)
this linear behavior was not observed for the 12C-PTMTC line, where ko is the frequency factor, DE is the activation energy, and R
where the lack of spectral anisotropy effectively leads to EPR spectra is the universal gas constant (R = 8.314 J molꢀ1 Kꢀ1). The Arrhenius
that show no dependence on rotational motion.
plot of the derived rate constants (first order kinetics), from which
The linear viscosity/temperature relation indicates that for the activation energies have been extracted, are shown in Fig. 9B.
13C-PTMTC, the rotational diffusion is the dominant effect on The resulting activation energy (DE) is 226.5 ꢁ 16.9 kJ molꢀ1, and
line width.49,50 Importantly in view of potential applications, the frequency factor (ln ko) is 73.4 ꢁ 6.0 minꢀ1 (r2 = 0.9889). These
this allows for measurements of viscosity (in the local environ- data indeed indicate a long-term stability when radical solutions
ment of the radical) even if other solvent parameters are also are stored properly, i.e. under dark conditions, as the radicals were
changed, e.g., oxygen content.
not stable in solution after 24 h if kept in light (no data given).
Hence, for in vivo applications, long-term stability of the
radicals can be assured when solutions are kept in the dark and
3.7. Radical stability
A solution of 13C-PTMTC (c = 1 mM) in PB (50 mM, pH 7.4) was at moderate temperatures, conditions that will normally be
used to test the long-term stability of the samples. Thermal met. To test for stability against reducing agents, which of
annealing of solutions stored in the dark (to suppress decay course can be seen as a crude measure of in vivo conditions,
pathways based on photochemical reactions) was performed to we investigated the reaction with ascorbic acid as a common
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
Fig. 8 Effect of temperature on line width. Solution of 13C-PTMTC (c = 1 mM) in glycerol/phosphate buffer 50 mM, pH 7.4 (40/60, V/V), measured from
20 to 70 1C: (A) HF peak (red circles); (B) LF peak (blue triangles). Straight lines are meant to guide the eye.
biological reducing agent. This is presented in the Fig. S4, ESI.† intensity. The reaction with superoxide involved a reduction of
After approx. 60 min, half of the signal intensity had decayed. PTMTC radical into aH-PTMTC unlike the reaction with
Despite this rather rapid decay, lifetimes on this order of tetrathia-TAM radicals which was an oxidation process into the
magnitude may still be sufficient for in vivo imaging where cation.19 In general, tetrachlorotritylmethyl radicals are more
other factors determine the time-span of EPR measurements, stable towards oxidation, while tetrathiatritylmethyl radicals
e.g., for how long an animal can be maintained in a fixed are more stable towards reduction.
position.
3.8. Radical stability in cell extracts
It was reported that the tetrathiaphenylmethyl radical Ox063
was stable in the presence of ascorbate, glutathione and
NADPH. PTMTC was reported to be stable towards hydrogen
peroxide (500 mM), alkyl peroxyl radicals, hydroxyl radicals, nitric
oxide, glutathione (1 mM), and ascorbate (100 mM). PTMTC reacts
specifically with superoxide, causing decrease of EPR signal
With the help of cell lysate experiments, in vivo conditions can
be mimicked even better and radical degradation can be
quantified. We used a Xenopus laevis oocyte cell extract. The
Xenopus laevis oocytes trial is an established and easy-to-use
model system to perform microinjection experiments. In our
Fig. 9 (A) Thermal decay of 13C-PTMTC sample stored at different elevated temperatures: 353.15 K, r2 = 0.9922 (black squares); 343.15 K, r2 = 0.9847
(red circles); 333.15 K, r2 = 0.9999 (blue triangles); 323.15 K, r2 = 0.9987 (green stars). The solid lines represent monoexponential fits to eqn (1). (B) Rate of
radical decay as a function of temperature.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017

View Article Online
PCCP
Paper
experiment, 5 ml of the 0.5 mM solution of PTMTC in PB (50 mM, This research was supported by Deutsche Forschungsgemeinschaft
pH 7.4) was mixed with 15 ml of the cell lysate and was vortexed (DFG) (MA 1648/8 to K. M. and DR 1024/1-1 to S. D.), the Egyptian
for 10 seconds and measured in a capillary ring cap. The EPR Ministry of Higher Education and Scientific Research and the
signal of PTMTC has a half-life of less than ten minutes. This Institut fuer Angewandte Dermatopharmazie (IADP) Halle.
indicates that encapsulation or some other means of protection
will be advantageous in vivo, e.g., in polymeric or small molecule
carriers. The encapsulation was described and investigated in
more detail in ref. 31. With PTMTC it had been shown that in
References
human blood or plasma, a 40% decrease of the signal intensity
occurred after 30 min only.19 So while tetrachloro-TAM radicals
are more resistant to oxidation (shelf-life of several years if kept
from light), they are more easily reduced in comparison to
tetrathia-substituted TAMs.
1 Y. Liu, F. A. Villamena and J. L. Zweier, Chem. Commun.,
2008, 4336.
ˆ
2 B. Driesschaert, V. Marchand, P. Leveque, B. Gallez and
J. Marchand-Brynaert, Chem. Commun., 2012, 48, 4049.
¨
3 K. Mader, B. Bittner, Y. Li, W. Wohlauf and T. Kissel, Pharm.
Res., 1998, 15, 787.
4 R. Ahmad and P. Kuppusamy, Multifrequency Electron Para-
magnetic Resonance, Wiley-VCH Verlag GmbH & Co. KgaA,
2011, p. 755.
5 B. B. Williams and H. J. Halpern, Biomedical EPR, Part A:
Free Radicals, Metals, Medicine, and Physiology, Springer,
2005, p. 283.
6 M. Gomberg, J. Am. Chem. Soc., 1900, 22, 757.
7 P. Jacobson, Ber. Dtsch. Chem. Ges., 1905, 38, 196.
8 H. Lankamp, W. T. Nauta and C. MacLean, Tetrahedron
Lett., 1968, 9, 249.
9 J. Schmidlin, Ber. Dtsch. Chem. Ges., 1908, 41, 2471.
10 E. G. Janzen, F. J. Johnston and C. L. Ayers, J. Am. Chem.
Soc., 1967, 89, 1176.
11 M. Gomberg, J. Am. Chem. Soc., 1901, 23, 496.
12 M. Ballester, Pure Appl. Chem., 1967, 15, 123.
13 M. Ballester, J. Riera-Figueras, J. Castaner, C. Badfa and
J. M. Monso, J. Am. Chem. Soc., 1971, 93, 2215.
14 M. Ballester, J. Castaner, J. Riera, A. Ibanez and J. Pujadas,
J. Org. Chem., 1982, 47, 259.
15 M. Ballester, J. Riera, J. Castaner, A. Rodriguez, C. Rovira
and J. Veciana, J. Org. Chem., 1982, 47, 4498.
16 A. Bratasz, A. C. Kulkarni and P. Kuppusamy, Biophys. J.,
2007, 92, 2918.
17 G. Meenakshisundaram, E. Eteshola, A. Blank, S. C. Lee and
P. Kuppusamy, Biosens. Bioelectron., 2010, 25, 2283.
18 V. Dang, J. Wang, S. Feng, C. Buron, F. A. Villamena, P. G. Wang
and P. Kuppusamy, Bioorg. Med. Chem. Lett., 2007, 17, 4062.
19 V. K. Kutala, F. A. Villamena, G. Ilangovan, D. Maspoch,
N. Roques, J. Veciana, C. Rovira and P. Kuppusamy, J. Phys.
Chem. B, 2008, 112, 158.
4 Conclusion
Among stable trityl radicals with a single line in the EPR
spectrum, perchlorinated trityl radicals are relatively easy to
prepare from readily available starting materials in just three
steps, including the possibility of 13C labelling of the central
carbon atom. PTMTC is distinguished by a single rather narrow
EPR line, attractive for in vivo applications including imaging,
where a strong signal is mandatory. However, the lack of
anisotropy of the spectrum (in particular hyperfine interactions)
limits spectroscopic conclusions or makes line width and shape
changes ambiguous as to what causes them. We show that pH
and macroviscosity can be disregarded in vivo, while there is a
clear but—due to small oxygen solubility changes with increasing
partial pressure—weak oxygen dependence of the line width in
aqueous solutions. As long as the radical is not exposed to light, it
is stable even in solution. However, since it quickly reacts with
reducing agents (ascorbic acid or in cell lysate), this type of radical
can only be used in a formulation that will protect it. Partial isotope
labelling of PTMTC leads to its isotopologue 13C-PTMTC, a
radical with practically identical physicochemical parameters
but a different EPR spectrum. In particular, effects of changes
in rotational motion (viscosity) are observable in 13C-PTMTC and
can be analyzed. Since the partial (50%) isotope-labelling allows
for a simultaneous detection of 12C-PTMTC and the pure
13C-PTMTC EPR signals, it combines benefits of in vivo and
imaging applications (12C-PTMTC) with the possibility to
obtain deeper insights into the local, nanoscopic environment
(13C-PTMTC). Hence, this mixture of isotopic radicals could
gain importance as a probe for both types of parameters
(viscosity and oxygen concentration) and applications in, e.g.,
spectral-spatial imaging. The problem of toxicity will not be an
issue when and if the radicals are incorporated in lipophilic
nano- or microcapsules31 with slower release than metabolic
reduction after release.
20 J. Wang, V. Dang, W. Zhao, D. Lu, B. K. Rivera, F. A. Villamena,
P. G. Wang and P. Kuppusamy, Bioorg. Med. Chem., 2010, 18, 922.
´
´
´
21 J. A. Mesa, S. Chavez, L. Fajarı, J. L. Torres and L. Julia, RSC
Adv., 2013, 3, 9949.
22 M. Ballester, J. Riera, J. Castaner, C. Rovira, J. Veciana and
C. Onrubia, J. Org. Chem., 1983, 48, 3716.
23 F. M. Vigier, D. Shimon, V. Mugnaini, J. Veciana, A. Feintuch,
M. Pons, S. Vega and D. Goldfarb, Phys. Chem. Chem. Phys.,
2014, 16, 19218.
Acknowledgements
¨
We would like to thank Dr Dieter Strohl (MLU Halle) for NMR 24 D. Banerjee, J. C. Paniagua, V. Mugnaini, J. Veciana,
analyses, Dr Christian Ihling (MLU Halle) for HRMS measure-
ments, and Ms Heike Rudolf (MLU Halle) for IR measurements.
A. Feintuch, M. Pons and D. Goldfarb, Phys. Chem. Chem.
Phys., 2011, 13, 18626.
This journal is ©the Owner Societies 2017
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
25 C. Gabellieri, V. Mugnaini, J. C. Paniagua, N. Roques, 37 B. B. Williams, H. al Hallaq, G. Chandramouli, E. D. Barth,
M. Oliveros, M. Feliz, J. Veciana and M. Pons, Angew. Chem.,
Int. Ed., 2010, 49, 3360.
J. N. Rivers, M. Lewis, V. E. Galtsev, G. S. Karczmar and
H. J. Halpern, Magn. Reson. Med., 2002, 47, 634.
´
´
26 M. Oliveros, L. Gonzalez-Garcıa, V. Mugnaini, F. Yubero, 38 G. Elert, The Physics Hypertextbook, 2005.
´
N. Roques, J. Veciana, A. R. Gonzalez-Elipe and C. Rovira, 39 CRC handbook of chemistry and physics, (Internet Version
Langmuir, 2011, 27, 5098.
27 D. Maspoch, D. Ruiz-Molina, K. Wurst, C. Rovira and
J. Veciana, Chem. Commun., 2004, 1164.
2009), CRC Press/Taylor and Francis, Boca Raton, FL, 89th
edn, 2009.
40 C. Decroos, J.-L. Boucher, D. Mansuy and Y. Xu-Li, Chem.
Res. Toxicol., 2014, 27, 627.
28 D. Maspoch, N. Domingo, D. Ruiz-Molina, K. Wurst,
J. Tejada, C. Rovira and J. Veciana, C. R. Chim., 2005, 41 L. Yong, J. Harbridge, R. W. Quine, G. A. Rinard, S. S. Eaton,
8, 1213.
G. R. Eaton, C. Mailer, E. Barth and H. J. Halpern, J. Magn.
Reson., 2001, 152, 156.
42 R. Owenius, G. R. Eaton and S. S. Eaton, J. Magn. Reson.,
2005, 172, 168.
29 V. Mugnaini, M. Paradinas, O. Shekhah, N. Roques, C. Ocal,
C. Woll and J. Veciana, J. Mater. Chem. C, 2013, 1, 793.
30 J. C. Paniagua, V. Mugnaini, C. Gabellieri, M. Feliz,
N. Roques, J. Veciana and M. Pons, Phys. Chem. Chem. Phys., 43 W. M. Haynes, CRC handbook of chemistry and physics, CRC
2010, 12, 5824. press, 2012.
31 J. Frank, M. Elewa, M. M. Said, H. A. El Shihawy, M. El-Sadek, 44 R. Ahmad and P. Kuppusamy, Chem. Rev., 2010, 110, 3212.
D. Mu¨ller, A. Meister, G. Hause, S. Drescher, H. Metz, P. Imming 45 I. Kutsche, G. Gildehaus, D. Schuller and A. Schumpe,
¨
and K. Mader, J. Org. Chem., 2015, 80, 6754.
J. Chem. Eng. Data, 1984, 29, 286.
32 S. Stoll and A. Schweiger, J. Magn. Reson., 2006, 178, 42.
33 D. Maspoch, N. Domingo, D. Ruiz-Molina, K. Wurst,
46 L. Dougan, S. Bates, R. Hargreaves, J. Fox, J. Crain, J. Finney,
V. Reat and A. Soper, J. Chem. Phys., 2004, 121, 6456.
G. Vaughan, J. Tejada, C. Rovira and J. Veciana, Angew. 47 T. A. Pascal and W. A. Goddard III, J. Phys. Chem. B, 2012,
Chem., 2004, 116, 1864. 116, 13905.
34 M. Ballester, J. Riera, J. Castaner, C. Rovira and O. Armet, 48 J. Heller, H. Elgabarty, B. Zhuang, D. Sebastiani and
Synthesis, 1986, 64. D. Hinderberger, J. Phys. Chem. B, 2010, 114, 7429.
35 N. Roques, D. Maspoch, K. Wurst, D. Ruiz-Molina, C. Rovira 49 L. Burlamacchi, Mol. Phys., 1969, 16, 369.
and J. Veciana, Chem. – Eur. J., 2006, 12, 9238. 50 P. Atkins and D. Kivelson, J. Chem. Phys., 1966, 44, 169.
36 M. Sabacky, C. Johnson, R. Smith, H. Gutowsky and 51 K. C. Waterman and R. C. Adami, Int. J. Pharm., 2005,
˜
J. Martin, J. Am. Chem. Soc., 1967, 89, 2054.
293, 101.
Phys. Chem. Chem. Phys.
This journal is ©the Owner Societies 2017