Tetrathiatriarylmethyl radicals conjugated to an RGD-peptidomimetic

Article abstract of DOI:10.1002/ejoc.201403049

Tetrathiatriarylmethyl (TAM) radicals are favourite spin labels for electron paramagnetic resonance imaging (EPRI). In this work, we report a straightforward synthesis of new TAM radicals bound to an Arg-Gly-Asp (RGD)-peptido-mimetic known to be a cell-targeting agent. These new radicals are stable in solution, and sensitive to oxygen, and their experimental EPR data is consistent with density functional theory (DFT) calculations.

Full text of DOI:10.1002/ejoc.201403049

FULL PAPER
DOI: 10.1002/ejoc.201403049
Tetrathiatriarylmethyl Radicals Conjugated to an RGD-Peptidomimetic
Benoît Driesschaert,[a,b] Philippe Levêque,[b] Bernard Gallez,[b] and
Jacqueline Marchand-Brynaert*[a]
Keywords: Imaging agents / Radicals / EPR spectroscopy / Peptidomimetics / Density functional calculations
Tetrathiatriarylmethyl (TAM) radicals are favourite spin
labels for electron paramagnetic resonance imaging (EPRI).
In this work, we report a straightforward synthesis of new
TAM radicals bound to an Arg-Gly-Asp (RGD)-peptido-
mimetic known to be a cell-targeting agent. These new radi-
cals are stable in solution, and sensitive to oxygen, and their
experimental EPR data is consistent with density functional
theory (DFT) calculations.
Introduction
Electron paramagnetic resonance imaging (EPRI) is an
emerging technique that allows researchers to map the dis-
tribution of paramagnetic species.[1] Compared to the well-
known nuclear magnetic resonance imaging (MRI) tech-
nique extensively used in clinics, the EPRI technique has
the advantages of a higher intrinsic sensitivity and the ab-
sence of background. Indeed, as there is no endogenous
paramagnetic species in sufficient concentration and with a
long enough half-life, with the exception of melanin,[2] an
exogenous paramagnetic spin label has to be injected for
EPRI applications.[3] Two classes of water-soluble paramag-
netic probes are currently used, namely nitroxide (NR) radi-
cals and tetrathiatriarylmethyl (TAM) radicals. The latter,
exemplified by CT-03 or the more hydrophilic compound
Ox063 (Figure 1), are better spin probes than nitroxide rad-
icals because of their higher stability in biological media
and the lower intrinsic linewidth of their single EPR line,
which results in a higher signal-to-noise ratio (signal inten-
sity is inversely proportional to the square of the line-
width).[4]
Figure 1. Structures of CT-03 and Ox063.
acid moieties to target liver cells,[6] and to peptides for
targeting integrins.[7]
The highly challenging chemistry of TAM[8] radicals
could explain this lack of synthetic investment. In this
paper, we report the first synthesis of TAM radicals coupled
to a peptidomimetic molecule to target the αvβ3 integrin.
Integrins are transmembrane receptors that play crucial
roles in cell-to-cell and cell-to-matrix adhesion phenom-
ena.[9] Specifically, the αvβ3 integrin is overexpressed on the
spreading endothelial cells of new capillaries in the angio-
genic process linked to tumour growth and development.[10]
In addition, it is also expressed on certain invasive tumour
cells.[11] Therefore the αvβ3 integrin has been considered as
an appealing target for anticancer therapy (drugs) and can-
cer diagnosis (probes).[12] The αvβ3 integrin interacts with
several extracellular matrix proteins through recognition of
the RGD (Arg-Gly-Asp) tripeptide sequence.[13] RGD-con-
taining cyclic peptides, such as Cilengitide[14] (in clinical
trials for treatment of glioblastoma) and peptidomimetics
of this sequence have been intensively developed as potent
antagonists of αvβ3 integrin for therapeutic purposes.[15]
Some molecules bearing a spacer arm for grafting onto
devices or matrices are also used in various biotechnologi-
cal applications.[16]
To the best of our knowledge, TAM radicals conjugated
to a targeting moiety have not been described in the litera-
ture except by ourselves,[5] although a few examples of
similar conjugates based on nitroxide radicals exist. For in-
stance, NRs have been coupled to arabinogalactan or fatty
[a] Institute of Condensed Matter and Nanosciences (IMCN),
Molecules, Solids and Reactivity (MOST)
Université catholique de Louvain
Place Louis Pasteur 1, L4.01.02, 1348 Louvain-la-Neuve,
Belgium
E-mail: jacqueline.marchand@uclouvain.be
http://www.uclouvain.be/en-most.html
[b] Louvain Drug Research Institute (LDRI), Biomedical
Magnetic Resonance Section (REMA)
Guided by the X-ray structure[17] of the extracellular part
of αvβ3 integrin in complex with Cilengitide, our group has
previously designed RGD-peptidomimetics based on the l-
tyrosine scaffold (Figure 2). The parent compound (i.e., 1)
Université catholique de Louvain
Avenue Mounier 73, B1.73.08, 1200 Bruxelles, Belgium
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201403049.
Eur. J. Org. Chem. 2014, 8077–8084
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8077

B. Driesschaert, P. Levêque, B. Gallez, J. Marchand-Brynaert
FULL PAPER
showed an excellent ability to bind isolated human αvβ3 the reaction of 2 with an activated ester of the TAM radical,
integrin, with an IC50 value of 0.1 nm.[18] The derivatives namely pentafluorophenol triester derivative 3.[5] This acti-
(i.e., 2) equipped with oligoethylene glycol (OEG) spacer vated ester 3 was treated with 4 equiv. of peptidomimetic 2
arms retained the antagonist activity (IC50 = 0.3–0.7 nm).[18] (bis TFA salt) and Na2CO3 in degassed DMSO (Scheme 1).
Moreover, the anchorage of 2 onto the surface of culture A change of colour from red to green occurred during the
substrates allowed a strong adhesion of human endothelial reaction as a result of the coupling of 2 through the
cells and their retention under shear stress.[19]
terminal primary amine of its OEG spacer. The product
(i.e., 4), featuring three peptidomimetic moieties, was de-
tected in the crude mixture by mass spectrometry (calcd.
for [(M + 2)/2]+ 1821.4931; found 1821.2750). However, we
failed to purify this compound by flash chromatography on
silica gel, crystallization, or semi-preparative reverse-phase
HPLC. This was probably due to the high molecular weight
of the compound, and the presence of several acidic and
basic functionalities giving polyionic species for all the
values of pH we screened for HPLC purification.
To overcome the purification difficulties, an alternative
synthetic strategy was envisaged with a double objective.
Firstly, to limit the number of ionizable groups, and to al-
low purification on silica gel, we decided to attach only one
peptidomimetic moiety onto the TAM core, and to use an
acid-protected form of peptidomimetic 2 (n = 3). Secondly,
to keep the possibility of analysing the product by nuclear
magnetic resonance (NMR) spectroscopy, we chose to use
the TAM alcohol instead of the radical for the coupling
reaction. Starting from bis-protected intermediate 5,[18] se-
lective Boc (tert-butoxycarbonyl) deprotection was carried
out by treatment with 3 equiv. of sulfuric acid in tert-butyl
acetate (Scheme 2).[21] Under these particular conditions,
amine deprotection was complete, because the resulting
amine is stabilised by protonation, while the deprotection
Figure 2. Structures of Cilengitide and (bottom) peptidomimetic
developed by our group.[18]
Results and Discussion
The graftable RGD-peptidomimetic 2 (n = 6) was synthe- of the carboxylic acid is reversible. Indeed, the tert-butyl
sized as described previously.[20] Since all the polar func- cation, produced from the reaction between tert-butyl acet-
tionalities are unprotected, we envisaged as a first strategy ate and sulfuric acid, can re-esterify the carboxylic acid to
Scheme 1. First strategy envisaged for the synthesis of TAM radical bound to peptidomimetic 2.
Scheme 2. Selective N-Boc deprotection of diprotected pepidomimetic 5.
8078
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 8077–8084

Tetrathiatriarylmethyl Radical Conjugates
give the tert-butyl ester. After a basic work-up, the selec-
tively Boc-deprotected compound (i.e., 6) was isolated in
50% yield by flash chromatography.
The triethyl ester of TAM alcohol (i.e., 7) was synthe-
sized using a procedure described in the literature[22] with
slight modifications; some side-products were identified
that have never been reported before (see Supporting Infor-
mation). Treatment of 7 with just 1 equiv. of aqueous LiOH
in refluxing 1,4-dioxane led to the partial, statistical hydrol-
ysis of 7 (Scheme 3). Flash chromatography allowed the re-
covery of about 40% of the starting material (i.e., 7), 30%
of the desired monohydrolysed compound (i.e., 8) and 10%
of the dihydrolysed compound (i.e., 9). Due to the high po-
larity of the trihydrolysed compound (i.e., 10), it was not
recovered from the silica gel column.
Scheme 3. Partial hydrolysis of TAM alcohol 7.
Activation of the monohydrolysed compound was It is worth noting that no reaction was observed between
achieved by treating 8 with an excess of oxalyl chloride and the trityl alcohol and oxalyl chloride, probably due to the
1 equiv. of diisopropylethylamine (DIEA) in dichlorometh- steric protection of the hydroxy group.
ane (Scheme 4). The acyl chloride (i.e., 11) was recovered in
In the next step, selectively deprotected peptidomimetic
96% yield by simple filtration through a short silica gel pad. 6 was treated with activated TAM compound 11 in the pres-
Scheme 4. Activation of TAM alcohol 8, coupling with peptidomimetic 6, and deprotection of the tert-butyl ester with concomitant
formation of the TAM radical, followed by hydrolysis of the ethyl esters.
Eur. J. Org. Chem. 2014, 8077–8084
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8079

B. Driesschaert, P. Levêque, B. Gallez, J. Marchand-Brynaert
FULL PAPER
ence of pyridine in dichloromethane. The resulting coupling
product (i.e., 12) was purified by flash chromatography and
isolated in 57% yield (Scheme 4). The assignment of the
NMR signals of this highly complex molecule was achieved
using 2D sequences such as COSY, HMBC, and HMQC
(see Supporting Information). Then, treatment of 12 with
neat trifluoroacetic acid (TFA) led, simultaneously, to the
deprotection of the tert-butyl ester and the conversion of
the alcohol into the TAM radical to give 13 (Scheme 4).
The formation of a tetrathiatriarylmethyl radical by treat-
ment of a tetrathiatriarylmethyl alcohol with neat TFA has
been previously reported,[23] and the mechanism has been
fully explained in a recent publication.[8c] The reaction
works in the absence of a reductant (like SnCl2) because
trityl cations can participate in electron-transfer reactions.
The production of the radical was accompanied by the for-
mation of an oxidative decarboxylation side-product (the
quinone methide).[8c] Purification by semi-preparative
reverse-phase HPLC allowed the isolation of the pure trityl
spin probe bound to the RGD-peptidomimetic moiety (i.e.,
13) in 39% yield. To enhance the hydrophilicity of this new
spin label, the two remaining ethyl esters were hydrolysed
using excess LiOH. Bis-acid 14 was purified by semi-pre-
parative reverse-phase HPLC to give the final compound in
29% yield (Scheme 4). The relatively low yields of the last
two steps can be explained by a loss of material during the
HPLC purification (the product fraction was collected only
at the maximum of the elution peak to avoid contamination
by side-products[8c]).
Figure 3. Chomatograms and UV spectra of TAM radicals 13 (up-
per) and 14 (lower) obtained under gradient elution (see Support-
ing Information for complete conditions).
The two new radicals (i.e., 13 and 14) were isolated in
high purity, as shown by their chromatograms obtained un-
der gradient elution (Figure 3). In addition, the radicals are
stable in methanol solution at room temperature for at least
48 h (no trace of degradation products could be seen in the
HPLC analyses). The pure radicals can be stored for
months in the freezer.
For biomedical EPR applications, spin label 14 is more
interesting because of its higher hydrophilicity. Its EPR
spectrum was recorded at X-band (≈ 9.4 GHz) under anoxic
conditions (Figure 4). The spectrum mainly shows a triplet
pattern resulting from the hyperfine coupling between the
odd electron and the 14N (I = 1) of the amide group. In
addition, small peaks corresponding to molecules naturally
labelled with 13C (natural abundance of 1.1%, I = 1/2) are
also visible.
For highly accurate determination of the hyperfine
splitting constant (hfc), the spectrum was fitted using the
Levenberg–Marquard algorithm, as implemented in Easy-
Spin.[24] The results are summarized in Table 1.
Figure 4. X-band EPR spectrum of 14 in PBS (phosphate-buffered
saline)/MeOH (50:50), 0.5 mm under nitrogen. The acquisition set-
tings were: field centre: 3375.36 G, scan range: 15 G, data points:
1024, modulation amplitude: 0.05 G, modulation frequency:
100 KHz, time constant: 20.48, power: 1.272 mW. The fitting pa-
rameters are given in the Supporting Information. Exp = experi-
mental spectrum, Sim = simulated spectrum.
carbon 13C labelled was not visible, and the spectrum was
The assignment of the 13C satellites was achieved with focussed on the region of the other hyperfine splittings. The
the help of the intensity ratio and calculation of the iso- hyperfine splitting with the aromatic 13C was well resolved.
tropic hfc (Aiso). The structure of model radical 15 (Fig- The calculated data using the 6-311+G** basis set are in
ure 5) was optimized at the UB3LYP/6-31G* level of good agreement with the measured data, and are more
theory, and Aiso was calculated at the UB3LYP level of accurate than those obtained with the 6-31G* basis set. The
theory with two different basis sets: in the gas phase, and in difference between the calculated and the experimental 14N
water using the PCM model (polarizable continuum model; Aiso values could be due to rotation of the amide group at
Table 1). Due to its low intensity (1.1% of the total signal), room temperature in solution, which could change the spin
the signal corresponding to the molecule with the central density on the nitrogen, and thus the hyperfine coupling
8080
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 8077–8084

Tetrathiatriarylmethyl Radical Conjugates
Table 1. Measured and calculated isotropic coupling constant (Aiso). The values are given in MHz and in parentheses in μT.
Nucleus
X-Band
6-31G*[a]
6-311+G**[a]
6-311+G** in water[a]
Degeneracy
Assignment
[b]
13C
13C
13C
13C
13C
14N
/
90.97
53.98
53.80
1
3
6
6
3
1
Central C
1-Aryl
2,6-Aryl
3,5-Aryl
4-Aryl
31.49 (1123.6)
25.43 (907.4)
6.41 (228.7)
9.38 (334.7)
0.62 (22.1)
30.85Ϯ0.14
29.23Ϯ1.31
7.32Ϯ1.12
12.16Ϯ0.19
0.22
31.64Ϯ0.10
25.744Ϯ1.74
6.21Ϯ1.10
7.60Ϯ0.11
0.15
31.57Ϯ0.11
25.54Ϯ1.47
6.32Ϯ1.07
7.71Ϯ0.12
0.23
Amide
[a] The absolute value. [b] The Aiso value measured for CT-03 was 67.1 MHz.[25]
1
constant. Couplings with the H (I = 1/2) of the amide or amount of dissolved molecular oxygen in the solution. As
the methylene amide were not resolved at X-Band.
shown in Figure 6, there is a linear relationship between the
ΔLp–p linewidth and the partial pressure of oxygen (pO2)
(0.8 μT/%O2), as is the case for soluble spin labels (trityls,
nitroxides).[3] Thus, compound 14 could be used as a spin
Figure 5. Structure of model radical 15.
Oxygen is a free radical containing two unpaired elec-
trons (in its ground state). It can interact with a free radical
to induce relaxation and a line broadening of the EPR
signal of the radical. This phenomenon is used to assay oxy-
genation in vitro and in vivo by EPR.[3] The EPR spectrum
of 14 was analysed as a Voigt function,[26] i.e., the convol-
ution of a Lorentzian function and a Gaussian function
resulting from unresolved hyperfine splitting. The Lorentz
Figure 7. X-band EPR spectrum of 13 in PBS/MeCN (50:50) 5 mm
under nitrogen. The acquisition settings were: field centre:
3374.38 G, scan range: 15 G, data points: 1024, modulation ampli-
tude: 0.05 G, modulation frequency: 100 KHz, time constant:
20.48, power: 1.269 mW. The fitting parameters are given in the
peak-to-peak (ΔLp–p) linewidth of radical 14 depends on the Supporting Information.
Figure 6. Calibration of the peak-to-peak Lorentz linewidth vs. pO2.
Eur. J. Org. Chem. 2014, 8077–8084
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8081

B. Driesschaert, P. Levêque, B. Gallez, J. Marchand-Brynaert
FULL PAPER
(ppm), referenced to NMR solvents, [D]chloroform [δ(1H) =
7.27 ppm, δ(13C) = 77.2 ppm], [D6]dimethyl sulfoxide [δ(1H) =
2.50 ppm, δ(13C) = 39.52 ppm]. The following abbreviations are
used: s = singlet, br. s = broad singlet, d = doublet, t = triplet,
m = multiplet. The atom numbering used for the NMR spectral
assignment is given in the Supporting Information. Coupling con-
stants (J) are reported in Hertz (Hz). HRMS analysis was carried
out at University College London. Analytical HPLC was carried
out with a Waters Alliance 2690 separation module equipped with
a Waters 2998 photodiode array (PDA) detector. Semi-preparative
reverse-phase HPLC was carried out with a Waters 600 Pump
equipped with a Waters 486 UV detector, a Waters fraction collec-
tor III, and a Hitachi L-7200 HPLC autosampler. EPR spectra
were recorded with an X-Band Bruker EMX spectrometer
equipped with an ER4119HS resonator; the data were recorded
under nonsaturating conditions. Calculations were performed with
Gaussian 09.[29] Geometry optimization was performed at the UB3-
LYP/6-31G* level, and hfc was calculated with a single point analy-
sis at the UB3LYP/6-311+G** level. Simulations of the EPR spec-
tra were performed with Easy Spin software for Mathlab.
label for EPR oximetry. The less hydrophilic structure 13
shows a similar EPR pattern (Figure 7). However, the reso-
lution of the spectrum is lower, due to additional hyperfine
splitting with the methylene protons of the esters. Com-
pound 13 also shows a similar sensitivity to oxygen (data
not shown).
Conclusions
We have synthesized new TAM radicals conjugated to an
RGD-peptidomimetic molecule whose antagonist activity
against αvβ3 integrin has been demonstrated in previous
publications.[18,19] This work represents the first example of
EPR spin labels linked to a specific recognition motif for
a receptor, although probes conjugated to ligands of αvβ3
integrin have already been described in the context of other
imaging techniques such as MRI, SPECT (single photon
emission tomography), PET (positron emission tomogra-
phy), and FMT (fluorescence-mediated tomography).[27]
Generally, RGD-peptides and cyclopeptides are used;
RGD-peptidomimetics have been considered in a few cases,
as exemplified by the coupling of 2 (n = 6) to ultra-small
particles of iron oxide (USPIO).[20]
The stability of radicals 13 and 14 and their EPR charac-
teristics are adequate for the further development of im-
aging applications. The success of our synthetic strategy is
built on several chemical tricks that can be useful in general
for the difficult TAM radical chemistry. By keeping the
alcohol group on the trityl core for most of the reaction
sequence, we allow accurate characterization of the interme-
diates by NMR spectroscopy. Moreover, activation of the
carboxyl group with oxalyl chloride is an efficient method
to produce the acyl chloride in high yield with no alteration
of the TAM hydroxy group. In contrast, all attempts to
carry out the coupling reaction with in situ activation
reagents {BOP [(benzotriazol-1-yloxy)tris(dimethylamino)-
phosphonium hexafluorophosphate], carbodiimides} failed
(data not shown), even though they are reported to be ef-
ficient with the TAM radical.[28] Finally, partial hydrolysis
of TAM 7 allows isolation of monoacid 8 in 52% yield
[based on the recovery of starting material (SM)], and bis-
acid 9 in 17% yield (based on the recovery of SM). While
8 was used here to bind only one carrier molecule, com-
pound 9 could be similarly used to link the trityl core to
two dedicated entities. Radicals 13 and 14 are sensitive to
oxygen, and can be used as spin labels for EPR oximetry.
In addition, density functional theory (DFT) calculation of
the hfc for radical 15 (model) was in good agreement with
the experimental data for radical 14. The new spin labels
are currently being investigated in in vitro assays.
Synthesis of 6: A solution of H2SO4 (70 μL, 1.299 mmol, 3 equiv.)
in tBuOAc (4 mL) was added dropwise over 10 min to a pale yellow
solution of 5 (400 mg, 0.433 mmol, 1 equiv.) in tBuOAc (30 mL) at
0 °C. A white precipitate formed during the addition of the sulfuric
acid. The heterogeneous solution was stirred at 0 °C for 10 min,
and then for 1 h at room temp. The reaction was quenched with
NaHCO3 (saturated aq.), and the mixture was vigorously stirred
until the white-yellow solid had completely dissolved. EtOAc
(15 mL) was added, and the layers were separated. The aqueous
layer was extracted with EtOAc (3ϫ 20 mL). The combined or-
ganic layers were dried with MgSO4, and filtered, and the solvent
was removed under reduced pressure. The residue was purified by
flash chromatography on deactivated silica gel. Starting material
and impurities were removed with CH2Cl2/MeOH, 9:1, then the
product was eluted with CH2Cl2/MeOH/Et3N, 89:10:1 to give 6
1
(169 mg, 47%) as a pale yellow oil. H NMR (500 MHz, CDCl3):
δ = 1.27 (s, 9 H, 11-H), 1.90 (m, 2 H, 28-H), 2.19 (m, 2 H, 20-H),
2.69 (t, J = 6.3 Hz, 2 H, 29-H), 2.74 (t, J = 7.5 Hz, 2 H, 21-H),
2.80 (m, 2 H, 37-H), 2.92 (dd, J = 6.6, 14.0 Hz, 1 H, 7-H), 2.99
(dd, J = 5.4, 14.0 Hz, 1 H, 7-H), 3.30 (br. s, 3 H, NH), 3.40 (m, 2
H, 27-H), 3.44 (t, J = 5.2 Hz, 2 H, 36-H), 3.50–3.90 (m, 8 H, 32-
H to 35-H), 4.00 (t, J = 6.4 Hz, 2 H, 19-H), 4.09 (dd, J = 5.6,
6.4 Hz, 1 H, 8-H), 4.14 (br. s, 2 H, 31-H), 5.17 (br. s, 1 H, NH),
6.35 (d, J = 7.3 Hz, 1 H, 23-H), 6.71 (d, J = 8.4 Hz, 1 H, 3-H),
6.81 (dd, J = 2.1, 8.3 Hz, 1 H, 4-H), 7.06 (d, J = 7.2 Hz, 1 H, 24-
H), 7.54 (t, J = 7.8 Hz, 1 H, 14-H), 7.72 (d, J = 7.7 Hz, 1 H, 15-
H), 7.94 (d, J = 7.9 Hz, 1 H, 13-H), 8.02 (s, 1 H, 17-H), 8.10 (d, J
= 2.0 Hz, 1 H, 6-H), 9.00 (s, 1 H, NH) ppm. 13C NMR (125 MHz,
CDCl3): δ = 21.5 (C-28), 26.5 (C-29), 27.8 (C-11), 29.2 (C-20), 33.8
(C-21), 38.8 (C-7), 41.7 (C-27, C-37), 57.5 (C-8), 68.0 (C-19), 70.3–
71.3 (C-31 to C-35), 73.3 (C-36), 82.9 (C-10), 111.1 (C-3), 111.4
(C-23), 113.9 (C-25), 121.1 (C-6), 123.4 (q, J = 272.9 Hz, C-18),
124.2 (q, J = 3.9 Hz, C-17), 125.3 (C-4), 127.0 (C-1), 127.9 (C-5),
129.0 (q, J = 3.4 Hz, C-15), 129.7 (C-14), 130.6 (C-13), 131.4 (q, J
= 33.4 Hz, C-16), 136.9 (C-24), 141.9 (C-12), 147.0 (C-2), 156.0 (C-
26), 156.6 (C-22), 167.8 (C-30), 170.0 (C-9) ppm. 19F NMR
(282 MHz, CDCl3): δ = –63.30 (s, 3 F, 18-F) ppm. HRMS (TOF
ES+): calcd. for C39H53F3N5O9S 824.3516; found 824.3500.
Experimental Section
Synthesis of 8 and 9: LiOH (10.6 mg, 0.444 mmol, 1 equiv.) in water
(5 mL) was added to a stirred solution of 7 (490 mg, 0.444 mmol,
1 equiv.) in dioxane (15 mL). The mixture was heated at reflux for
20 h. The solvent was removed under reduced pressure, and EtOAc
(20 mL) and H2SO4 (1 m; 5 mL) were added. The layers were sepa-
General Remarks: All commercially available reagents were used as
received without further purification. NMR spectra were recorded
with a Bruker Avance DPX instrument (1H: 500 MHz; 13C:
125 MHz). Chemical shifts (δ) are reported in parts per million
8082
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 8077–8084

Tetrathiatriarylmethyl Radical Conjugates
rated, and the aqueous layer extracted with EtOAc (3ϫ 10 mL).
The combined organic layers were dried with MgSO4. The solvent
was removed under reduced pressure, and the residue was purified
pressure. The residue was purified by column chromatography on
silica gel (acetone/EtOAc, 1:4 to 1:1) to give 12 (139 mg, 57%) as an
1
orange solid. Rf: 0.25 (EtOAc/acetone, 5:1). H NMR (500 MHz,
by column chromatography on silica gel (CH2Cl2 to CH2Cl2/ CDCl3): δ = 1.24 (s, 9 H, 30-H), 1.44 (t, J = 7.0 Hz, 6 H, 10-H),
MeOH, 7:3) to give recovered starting material (196 mg, 40%), and 1.61–1.76 (several singlets, 36 H, 7-H), 1.87 (m, 2 H, 46-H), 2.17
desired compound 8 (147 mg, 30%) as an orange solid. Rf: 0.5 (m, 2 H, 39-H), 2.66 (t, J = 6.1 Hz, 2 H, 45-H), 2.73 (t, J = 7.3 Hz,
1
(CH2Cl2/MeOH, 9:1). H NMR (500 MHz, CDCl3): δ = 1.45 (t, J 2 H, 40-H), 2.86 (dd, J = 7.0, 13.9 Hz, 1 H, 27-H), 2.98 (dd, J =
= 7.10 Hz, 6 H, 10-H), 1.65–1.77 (several singlets, 36 H, 7-H), 4.44
5.2, 13.9 Hz, 1 H, 27-H), 3.37 (m, 2 H, 47-H), 3.54–3.77 (m, 12 H,
(m, 4 H, 9-H), 6.79 (s, 1 H, OH) ppm. 13C NMR (125 MHz, 17-H to 12-H), 3.98 (t, J = 6.1 Hz, 2 H, 38-H), 4.05 (m, 1 H, 27-
CDCl3): δ = 14.4 (C-10), 28.4 (C-7), 28.8 (C-7), 28.9 (C-7), 29.1 H), 4.11 (s, 2 H, 18-H), 4.42 (m, 4 H, 9-H), 5.44 (s, 1 H, NH), 6.31
(C-7), 29.1 (C-7), 29.8 (C-7), 31.4 (C-7), 32.2 (C-7), 32.3 (C-7), 33.6 (d, J = 7.2 Hz, 1 H, 42-H), 6.68 (d, J = 8.3 Hz, 1 H, 24-H), 6.73
(C-7), 34.0 (C-7), 34.3 (C-7), 60.9 (C-6), 60.9 (C-6), 61.0 (C-6), 61.0 (s, 1 H, OH), 6.8 (dd, J = 1.7, 8.3 Hz, 1 H, 23-H), 7.05 (d, J =
(C-6), 61.0 (C-6), 61.4 (C-6), 62.5 (C-9), 62.5 (C-9), 84.4 (C-1), 7.2 Hz, 1 H, 43-H), 7.53 (t, J = 7.9 Hz, 1 H, 33-H), 7.71 (d, J =
121.4 (3 C, C-Ar), 133.8 (C-Ar), 134.1 (C-Ar), 134.7 (C-Ar), 138.9
(C-Ar), 139.5 (C-Ar), 139.6 (C-Ar), 140.3 (C-Ar), 140.4 (C-Ar),
7.7 Hz, 1 H, 34-H), 7.91 (d, J = 8.0 Hz, 32-H), 7.99 (s, 1 H, 36-H),
8.12 (s, 1 H, 21-H), 8.94 (s, 1 H, NH) ppm. 13C NMR (125 MHz,
140.6 (C-Ar), 141.6 (C-Ar), 141.7 (C-Ar), 141.9 (C-Ar), 142.0 (C- CDCl3): δ = 14.4 (C-10), 21.0 (C-46), 26.3 (C-45), 27.7 (C-7), 27.8
Ar), 142.4 (C-Ar), 143.0 (C-Ar), 166.3 (C-8), 171.2 (C-11) ppm.
HRMS (TOF ES–): calcd. for C44H47O7S12 1070.9970; found
1070.9915.
(C-30), 29.3 (C-7), 29.4 (C-7), 29.7 (C-7), 30.5 (C-7), 31.8 (C-7),
32.7 (C-7), 32.7 (C-40), 33.0 (C-7), 33.8 (C-7), 34.4 (C-7), 34.9 (C-
7), 38.9 (C-26), 39.8 (C-12), 41.5 (C-47), 57.3 (C-27), 60.9 (C-6),
61.1 (C-6), 61.3 (C-6), 61.9 (C-6), 62.0 (C-6), 62.6 (C-9), 67.6 (C-
38), 69.5–71.5 (6 peaks, C18 to C13), 83.1 (C-29), 84.3 (C-1), 111.1
(C-24), 111.1 (C-42), 120.8 (C-21), 121.4 (3 C-Ar), 123.3 (q, J =
273.1 Hz, C-37), 124.3 (q, J = 3.2 Hz, C-36), 125.3 (C-23), 127.0
(C-20), 127.7 (C-22), 129.2 (q, J = 2.8 Hz, C-34), 129.8 (C-33),
130.7 (C-32), 131.5 (q, J = 33.6 Hz, C-35), 131.8 (C-Ar), 134.5 (C-
Ar), 136.7 (C-Ar), 137.2 (C-Ar), 137.8 (C-43), 138.1 (C-Ar), 139.3
(C-Ar), 140.4–142.2 (9 peaks, 8 C-Ar, C-31), 146.7 (C-44), 155.1
(C-41, C-48), 166.3 (C-8), 166.3 (C-8), 167.1 (C-11), 167.6 (C-19),
169.9 (C-28) ppm. HRMS (TOF ES+): calcd. for
C83H99N5O15F3S13 1878.3459; found 1878.3258.
And finally dihydrolysed compound 9 (46 mg, 10%) as an orange
solid. Rf: 0.3 (CH2Cl2/MeOH, 7:3). 1H NMR (500 MHz, [D6]
DMSO): δ = 1.34 (t, J = 7.08 Hz, 3 H, 10-H), 1.54 (s, 3 H, 7-H),
1.54 (s, 3 H, 7-H), 1.57 (s, 3 H, 7-H), 1.58 (s, 6 H, 7-H), 1.62 (s, 3
H, 7-H), 1.63 (s, 3 H, 7-H), 1.64 (s, 9 H, 7-H), 1.69 (s, 3 H, 7-H),
1.70 (s, 3 H, 7-H), 4.33 (m, 2 H, 9-H), 6.76 (s, 1 H, OH) ppm. 13C
NMR (125 MHz, [D6]DMSO): δ = 14.0 (C-10), 27.4 (C-7), 27.5
(C-7), 27.9 (C-7), 28.0 (C-7), 28.7 (C-7), 29.7 (C-7), 30.5 (C-7), 31.8
(C-7), 32.1 (C-7), 34.2 (C-7), 34.2 (C-7), 24.4 (C-7), 59.2 (C-6), 59.5
(C-6), 59.7 (C-6), 59.9 (C-6), 60.4 (C-6), 60.7 (C-6), 62.1 (C-9), 83.5
(C-1), 120.6 (3 C, C-Ar), 130.0–145.0 (15 C, C-Ar), 165.3 (C-8),
169.6 (C-11) ppm. HRMS (TOF ES–): calcd. for C42H43O7S12
1042.9657; found 1042.9702.
Synthesis of 13: Compound 12 (70 mg, 0.037 mmol) was dissolved
in TFA (10 mL). The solution was stirred for 2 h at room temp.,
and then the TFA was removed under reduced pressure. The crude
product was purified by semi-preparative HPLC to give pure 13
(27 mg, 40%). The conditions of purification were as follows: Col-
umn XBridge 10ϫ100 mm 5 μm, equipped with a guard column
10ϫ20 mm 5 μm. UV detection at 485 nm. Flow rate 5 mL/min.
Isocratic elution with MeCN/H2O/H2O containing TFA (1%),
70:20:10. The fraction corresponding to the peak maximum was
collected. HRMS (TOF ES+): calcd. for C79H90N5O14F3S13
1805.28001; found 1805.27957.
Synthesis of 11: DIEA (23 μL, 0.137 mmol, 1 equiv.) was added to
a stirred solution of 8 (147 mg, 0.137 mmol, 1 equiv.) in CH2Cl2
(10 mL). The solution was cooled to –78 °C, and oxalyl chloride
(59 μL, 0.685 mmol, 5 equiv.) was added. The red solution was
stirred at –78 °C for 15 min, and then overnight at room temp. The
solvent was removed under reduced pressure, and the residue was
purified by elution through a short (5 cm) pad of silica gel
(CH2Cl2) to give 11 (143 mg, 96%) as a red solid. 1H NMR
(500 MHz, CDCl3): δ = 1.47 (t, J = 8.10 Hz, 3 H, 10-H), 1.47 (t, J
= 8.10 Hz, 3 H, 10-H), 1.66 (s, 3 H, 7-H), 1.67 (s, 6 H, 7-H), 1.68
(s, 3 H, 7-H), 1.70 (s, 3 H, 7-H), 1.72 (s, 3 H, 7-H), 1.75 (s, 3 H, 7-
H), 1.77 (s, 9 H, 7-H), 1.79 (s, 3 H, 7-H), 1.80 (s, 3 H, 7-H), 1.80
(s, 3 H, 7-H), 4.44 (m, 4 H, 9-H), 6.78 (s, 1 H, OH) ppm. 13C NMR
(125 MHz, CDCl3): δ = 14.4 (C-10), 28.2 (C-7), 28.3 (C-7), 28.5
(C-7), 29.1 (C-7), 29.6 (C-7), 30.5 (C-7), 30.8 (C-7), 32.3 (C-7), 32.8
(C-7), 33.3 (C-7), 34.0 (C-7), 34.3 (C-7), 61.0 (C-6), 61.1 (C-6), 61.2
(C-6), 61.8 (C-6), 62.6 (C-9), 63.3 (C-6), 63.6 (C-6), 84.3 (C-1),
121.5 (3 C, C-Ar), 128.0 (C-Ar), 133.7 (C-Ar), 134.1 (C-Ar), 137.5
(C-Ar), 138.4 (C-Ar), 138.5 (C-Ar), 139.5 (C-Ar), 140.2 (C-Ar),
140.4 (C-Ar), 141.1 (C-Ar), 141.4 (C-Ar), 141.8 (C-Ar), 142.0 (C-
Ar), 142.3 (C-Ar), 166.3 (C-8), 167.5 (C-11) ppm.
Synthesis of 14: 13 (15 mg, 0.038 mmol) was dissolved in dioxane
(2 mL). LiOH (1 m aq.; 1 mL) was added, and the solution was
stirred overnight. HCl (1 m aq.; 2 mL) and EtOAc (3 mL) were
added, and the layers were separated. The aqueous layer was ex-
tracted with EtOAc (3ϫ 6 mL). The combined organic layers were
dried with MgSO4, and filtered, and the solvent was removed under
reduced pressure. The crude product was purified by semi-prepara-
tive HPLC to give 14 (9.3 mg, 29%). The conditions of purification
were as follows: Column XBridge 10ϫ100 mm 5 μm, equipped
with a guard column 10ϫ20 mm 5 μm. UV detection at 486 nm.
Gradient elution: see Table 2. The fraction corresponding to the
Synthesis of 12: Compound 6 (170 mg, 0.206 mmol, 1.6 equiv.) was
dissolved in CH2Cl2 (6 mL), and pyridine (16 μL, 0.206 mmol,
1.6 equiv.) and 11 (140 mg, 0.125 mmol, 1 equiv.) in CH2Cl2 (1 mL)
were added. The solution was stirred at room temp. for 24 h, then
NH4Cl (saturated aq.; 3 mL) was added, and the layers were sepa-
rated. NaHCO3 (saturated aq.; 3 mL) was added to the organic
phase, and the layers were separated. The aqueous phase was ex-
tracted with EtOAc (2ϫ 5 mL). The combined organic layers were
dried with MgSO4, and the solvent was removed under reduced
Table 2. Gradient elution.
Time
[min]
Flow
[mL/min]
MeOH
[%]
H2O
[%]
H2O (1% TFA)
[%]
0
5
5
5
5
5
68
68
90
90
68
22
22
0
0
22
10
10
10
10
10
10
11
14
14.1
Eur. J. Org. Chem. 2014, 8077–8084
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8083

B. Driesschaert, P. Levêque, B. Gallez, J. Marchand-Brynaert
FULL PAPER
peak maximum was collected. HRMS (TOF ES+): calcd. for
C75H82N5O14F3S13 1749.21741; found 1749.21683.
[12] a) D. Cox, M. Brennan, N. Moran, Nat. Rev. Drug Discovery
2010, 9, 804–820; b) A. Perdih, M. Sollner Dolenc, Curr. Med.
Chem. 2010, 17, 2371–2392.
[13] M. Barczyk, S. Carracedo, D. Gullberg, Cell Tissue Res. 2010,
339, 269–280.
[14] M. A. Dechantsreiter, E. Planker, B. Mathä, E. Lohof, G.
Hölzemann, A. Jonczyk, S. L. Goodman, H. Kessler, J. Med.
Chem. 1999, 42, 3033–3040.
Supporting Information (see footnote on the first page of this arti-
cle): Synthetic procedure for compound 7; analytical HPLC data
for radicals 13 and 14; simulation parameters of calculated EPR
spectra; and NMR spectra of nonparamagnetic compounds.
[15] K.-E. Gottschalk, H. Kessler, Angew. Chem. Int. Ed. 2002, 41,
3767–3774; Angew. Chem. 2002, 114, 3919–3927.
[16] K. Chen, X. Chen, Theranostics 2011, 1, 189–200.
[17] J.-P. Xiong, T. Stehle, R. Zhang, A. Joachimiak, M. Frech, S. L.
Goodman, M. A. Arnaout, Science 2002, 296, 151–155.
[18] V. Rerat, G. Dive, A. A. Cordi, G. C. Tucker, R. Bareille, J. l.
Amédée, L. Bordenave, J. Marchand-Brynaert, J. Med. Chem.
2009, 52, 7029–7043.
[19] M. Remy, R. Bareille, V. Rerat, C. Bourget, J. Marchand-Bry-
naert, L. Bordenave, J. Biomater. Sci. Polym. Ed. 2013, 24, 269–
286.
[20] V. Rerat, S. Laurent, C. Burtea, B. Driesschaert, V. Pourcelle,
L. Vander Elst, R. N. Muller, J. Marchand-Brynaert, Bioorg.
Med. Chem. Lett. 2010, 20, 1861–1865.
Acknowledgments
This work was supported by the Université catholique de Louvain
(UCL) and the Fond de la Recherche Scientifique (FRS-FNRS,
Belgium). J. M.-B. is a senior research associate of the FRS-FNRS.
The authors gratefully acknowledge Laurent Collard for his help
with HPLC purifications, as well as Dr. Cecile Le Duff, Prof.
Raphaël Robiette, and Martin Poncelet for helpful discussions.
Sabrina Devouge is acknowledged for the preparation of the manu-
script, and Keyton L. Clayson for proof reading. Computational
resources were provided by the supercomputing facilities of the
Université catholique de Louvain (CISM/UCL) and the Consor-
tium des Équipements de Calcul Intensif en Fédération Wallonie
Bruxelles (CÉCI), funded by the Fond de la Recherche Scientifique
(FRS-FNRS, Belgium) under convention 2.5020.11.
[21] L. S. Lin, T. Lanza Jr., S. E. de Laszlo, Q. Truong, T. Ka-
menecka, W. K. Hagmann, Tetrahedron Lett. 2000, 41, 7013–
7016.
[22] T. J. Reddy, T. Iwama, H. J. Halpern, V. H. Rawal, J. Org.
Chem. 2002, 67, 4635–4639.
[23] a) I. Dhimitruka, M. Velayutham, A. A. Bobko, V. V.
Khramtsov, F. A. Villamena, C. M. Hadad, J. L. Zweier, Bi-
oorg. Med. Chem. Lett. 2007, 17, 6801–6805; b) B.
Driesschaert, R. Robiette, C. S. Le Duff, L. Collard, K. Rob-
eyns, B. Gallez, J. Marchand-Brynaert, J. Org. Chem. 2012, 77,
6517–6525.
[24] S. Stoll, A. Schweiger, J. Magn. Reson. 2006, 178, 42–55.
[25] M. K. Bowman, C. Mailer, H. J. Halpern, J. Magn. Reson.
2005, 172, 254–267.
[1] M. C. Krishna, N. Devasahayam, J. A. Cook, S. Subramanian,
P. Kuppusamy, J. B. Mitchell, ILAR J. 2001, 42, 209–218.
[2] E. Vanea, N. Charlier, J. DeWever, M. Dinguizli, O. Feron, J.-
F. Baurain, B. Gallez, NMR Biomed. 2008, 21, 296–300.
[3] B. Gallez, C. Baudelet, B. F. Jordan, NMR Biomed. 2004, 17,
240–262.
[4] J. H. Ardenkjær-Larsen, I. Laursen, I. Leunbach, G. Ehnholm,
L. G. Wistrand, J. S. Petersson, K. Golman, J. Magn. Reson. [26] A. K. Hui, B. H. Armstrong, A. A. Wray, J. Quant. Spectrosc.
1998, 133, 1–12.
Radiat. Transfer 1978, 19, 509–516.
[27] M. Schottelius, B. Laufer, H. Kessler, H.-J. Wester, Acc. Chem.
Res. 2009, 42, 969–980.
[5] B. Driesschaert, P. Levêque, B. Gallez, J. Marchand-Brynaert,
Tetrahedron Lett. 2013, 54, 5924–5926.
[6] a) B. Gallez, V. Lacour, R. Demeure, R. Debuyst, F. Dejehet,
J. L. De Keyser, P. Dumont, Magn. Reson. Imaging 1994, 12,
61–69; b) B. Gallez, R. Debuyst, R. Demeure, F. Dejehet, C.
Grandin, B. Van Beers, H. Taper, J. Pringot, P. Dumont, Magn.
Reson. Med. 1993, 30, 592–599.
[7] a) J. Liu, M. Zhao, C. Wang, S. Peng, Bioorg. Med. Chem. Lett.
2003, 13, 4065–4069; b) J. Liu, M. Zhao, C. Wang, S. Peng,
Bioorg. Med. Chem. 2008, 16, 4019–4028.
[8] a) A. A. Bobko, I. Dhimitruka, J. L. Zweier, V. V. Khramtsov,
J. Am. Chem. Soc. 2007, 129, 72440–72441; b) I. Dhimitruka,
A. A. Bobko, C. M. Hadad, J. L. Zweier, V. V. Khramtsov, J.
Am. Chem. Soc. 2008, 130, 10780–10787; c) O. Y. Rogozhni-
kova, V. G. Vasiliev, T. I. Troitskaya, D. V. Trukhin, T. V. Mikh-
alina, H. J. Halpern, V. M. Tormyshev, Eur. J. Org. Chem. 2013,
3347–3355.
[28] Z. Yang, Y. Liu, P. Borbat, J. L. Zweier, J. H. Freed, W. L. Hub-
bell, J. Am. Chem. Soc. 2012, 134, 9950–9952.
[29] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09,
Revision B.01, Gaussian, Inc., Wallingford CT, 2009.
Received: August 7, 2014
[9] J. S. Desgrosellier, D. A. Cheresh, Nat. Rev. Cancer 2010, 10,
9–22.
[10] S. D. Robinson, K. M. Hodivala-Dilke, Curr. Opin. Cell Biol.
2011, 23, 630–637.
[11] J. F. Marshall, S. A. Nesbitt, M. H. Helfrich, M. A. Horton,
K. Polakova, I. R. Hart, Int. J. Cancer 1991, 49, 924–931.
Published Online: November 14, 2014
8084
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 8077–8084