Synthesis, Characterization, and Nanoencapsulation of Tetrathiatriarylmethyl and Tetrachlorotriarylmethyl (Trityl) Radical Derivativesí-A Study to Advance Their Applicability as in Vivo EPR Oxygen Sensors

Article abstract of DOI:10.1021/acs.joc.5b00918

Tissue oxygenation plays an important role in the pathophysiology of various diseases and is often a marker of prognosis and therapeutic response. EPR (ESR) is a suitable noninvasive oximetry technique. However, to reliably deploy soluble EPR probes as oxygen sensors in complex biological systems, there is still a need to investigate and improve their specificity, sensitivity, and stability. We reproducibly synthesized various derivatives of tetrathiatriarylmethyl and tetrachlorotriarylmethyl (trityl) radicals. Hydrophilic radicals were investigated in aqueous solution mimicking physiological conditions by, e.g., variation of viscosity and ionic strength. Their specificity was satisfactory, but the oxygen sensitivity was low. To enhance the capability of trityl radicals as oxygen sensors, encapsulation into oily core nanocapsules was performed. Thus, different lipophilic triesters were prepared and characterized in oily solution employing oils typically used in drug formulations, i.e., middle-chain triglycerides and isopropyl myristate. Our screening identified the deuterated ethyl ester of D-TAM (radical 13) to be suitable. It had an extremely narrow single EPR line under anoxic conditions and excellent oxygen sensitivity. After encapsulation, it retained its oxygen responsiveness and was protected against reduction by ascorbic acid. These biocompatible and highly sensitive nanosensors offer great potential for future EPR oximetry applications in preclinical research.

Full text of DOI:10.1021/acs.joc.5b00918

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
Synthesis, characterization, and nanoencapsulation of tetrathiatriarylmethyl
and tetrachlorotriarylmethyl (trityl) radical derivatives – a study
to advance their applicability as in vivo EPR oxygen sensors
Juliane Frank, Marwa Elewa, Mohamed M. Said, Hosam A. El Shihawy, Mohamed El-Sadek, Diana
Mueller, Annette Meister, Gerd Hause, Simon Drescher, Hendrik Metz, Peter Imming, and Karsten Mäder
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.5b00918 • Publication Date (Web): 28 May 2015
Downloaded from http://pubs.acs.org on June 1, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Synthesis, characterization, and nanoencapsulation
of tetrathiatriarylmethyl and tetrachlorotriarylmethyl
(trityl) radical derivatives – a study to advance their
applicability as in vivo EPR oxygen sensors
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Juliane Frank,#,† Marwa Elewa,#,†,‡ Mohamed M. Said,‡ Hosam A. El Shihawy,‡ Mohamed Elꢀ
Sadek,§ Diana Müller,† Annette Meister,¶ Gerd Hause,$ Simon Drescher,† Hendrik Metz,† Peter
Imming,*,† and Karsten Mäder*,†
†
Institut fuer Pharmazie, MartinꢀLutherꢀUniversitaet (MLU) HalleꢀWittenberg, Wolfgangꢀ
‡
LangenbeckꢀStr. 4, 06120 Halle (Saale), Germany; Faculty of Pharmacy, Suez Canal Universiꢀ
§
¶
ty, P.O. 41522, Ismailia, Egypt; Faculty of Pharmacy, Zagazig University, Zagazig, Egypt;
Center for structure and dynamics of proteins (MZP), MLU HalleꢀWittenberg, Biocenter, Weinꢀ
$
bergweg 22, 06120 Halle (Saale), Germany; Biocenter MLU HalleꢀWittenberg, Weinbergweg
22, 06120 Halle (Saale), Germany.
# These authors contributed equally (J.F. and M.E.).
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 45
1
2
3
4
ABSTRACT GRAPHIC
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS
Trityl radials, TAM radicals, Synthesis, EPR oximetry, ESR Oximetry, Oxygen sensor, Encapsuꢀ
lation, Biocompatibility, Sensitivity, Specificity
2
ACS Paragon Plus Environment

Page 3 of 45
The Journal of Organic Chemistry
1
2
3
4
ABSTRACT
5
6
7
8
9
Tissue oxygenation plays an important role in the pathophysiology of various diseases and is
often a marker of prognosis and therapeutic response. EPR (ESR) is a suitable nonꢀinvasive
oximetry technique. However, to reliably deploy soluble EPR probes as oxygen sensors in comꢀ
plex biological systems, there is still a need to investigate and improve their specificity, sensitiviꢀ
ty, and stability. We reproducibly synthesized various derivatives of tetrathiatriarylmethyl and
tetrachlorotriarylmethyl (trityl) radicals. Hydrophilic radicals were investigated in aqueous soluꢀ
tion mimicking physiological conditions by, e.g., variation of viscosity and ionic strength. Their
specificity was satisfactory; but, the oxygen sensitivity was low. To enhance the capability of
trityl radicals as oxygen sensors, encapsulation into oilyꢀcore nanocapsules was performed. Thus,
different lipophilic triꢀesters were prepared and characterized in oily solution employing oils
typically used in drug formulations, i.e., middleꢀchain triglycerides and isopropyl myristate. Our
screening identified the deuterated ethyl ester of DꢀTAM (radical 13) to be suitable. It had an
extremely narrow single EPR line under anoxic conditions and excellent oxygen sensitivity.
After encapsulation, it retained its oxygenꢀresponsiveness and was protected against reduction by
ascorbic acid. These biocompatible and highly sensitive nanosensors offer great potential for
future EPR oximetry applications in preclinical research.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 45
1
2
3
4
INTRODUCTION
5
6
7
8
9
The quantification of oxygen levels in vitro and in vivo is crucial not only for the understandꢀ
ing of physiological processes but also in the assessment and therapy of numerous pathological
conditions, such as cancer, peripheral vascular disease, and wounds.1ꢀ3 Various oxygen measureꢀ
ment techniques have been developed, but especially in vivo oximetry is still challenging. EPR
(ESR) oximetry offers several advantages as it enables direct, nonꢀinvasive, and repeatable oxyꢀ
gen measurements.4,5 Spinꢀspin interactions with the paramagnetic oxygen molecule decrease the
relaxation times of EPR spin probe radicals and, therefore, result in a quantifiable broadening of
their spectral line width.6 Two classes of oxygenꢀsensitive EPR spin probes have been developed.
Both have certain advantages and drawbacks. Particulate materials (e.g., carbonꢀbased probes
and lithium salts) attracted much attention as they are very promising oxygen sensors, but they
measure oxygen contents only at the implantation site and are often deficient in terms of the
reproducibility of their preparation and properties.7 In contrast, soluble spin probes, e.g., nitroxꢀ
ides and triarylmethyl (trityl, TAM) radicals, are chemically well defined and can distribute
evenly within samples.7 Especially deuterated trityl radicals and Dꢀ,15Nꢀsubstituted nitroxides
exhibit favorably narrow single EPR lines.8,9 However, their oxygen sensitivities are dependent
on the solvent and their signal characteristics are also affected by changes of viscosity and pH
value.10,11 Additionally, when EPR measurements are conducted in biological systems, the spin
probes might lose their signal intensity and their oxygen responsiveness due to interactions with
tissues, such as chemical reactions (e.g., reduction or oxidation), protein binding, or fibrous
capsule formation.12,13 These aspects in mind, we intend to shed more light on EPR oximetry
using soluble spin probes, in particular trityl radicals. Their main limitations might be overcome
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
by appropriate formulations: By encapsulation, their oxygen sensitivities can be improved,14ꢀ17
a
4
ACS Paragon Plus Environment

Page 5 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
defined microenvironment is created, which ensures specificity of the sensors to oxygen, and the
capsule shell might provide protection against oxidoreductants,17ꢀ19 and prevent biocompatibility
concerns.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The first triarylmethyl radical was prepared by Gomberg in 1900.20 To eliminate hyperfine
coupling with hydrogen nuclei and achieve sharp single EPR lines, the phenyl rings were substiꢀ
tuted with alkylthio moieties by Nycomed Innovation AB in the 1990s leading to the family of
tetrathiaꢀTAM radicals (see Figure 1A).21,22 Salts of these trityl radicals showed good water
solubility as well as stability in the presence of reducing reagents such as ascorbate and glutathiꢀ
one.18,23 Another family of trityl radicals, the tetrachloroꢀTAM radicals, was first introduced in
1967 (see Figure 1B).24 Upon substitution of the six ortho positions with chlorine atoms, the
central methyl carbon was sterically shielded providing high chemical and thermal stability.25,26
TetrachloroꢀTAM radicals show broader EPR lines due to coupling with the chlorine nuclei in
close vicinity, but are distinguished by better synthetic accessibility than tetrathiatrityl radicals.
Trityl radicals have not only been employed in EPR oximetry,14,27,28 but also in many other apꢀ
plications, e.g., specific detection of superoxide radical anions,29ꢀ31 pH measurements,32,33 as well
as analysis of redox status.34 Tritylꢀbased spin labels were used for distance measurements, e.g.,
in nucleic acids.35 Their long relaxation times made trityl radicals also attractive for (pulsed)
EPR imaging36 and dynamic nuclear polarization (DNP).37 To achieve intracellular permeability,
lipophilic tetrathiaꢀTAM ester derivatives were developed.38
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 45
1
2
3
4
A
tetrathiaꢀTAM radicals
B
tetrachloroꢀTAM radicals
5
6
7
8
3
3
C
C
S
S
S
Cl
Cl
Cl
Cl
R
R
R
R
S
9
R
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
OH
R = CH (radical 10)
PTMTE: R = COOCH2CH3
(radical 18)
C-TAM:
3
D-TAM: R = CD3 (radical 9)
Ox063: R = (CH2)2OH
PTMTC: R = COOH
PTMTS--3K+: R = SO3ꢀ K+
Ox031: R = CH2O(CH2)2OH
Figure 1. Chemical structures and common abbreviations of different derivatives of trityl radiꢀ
cals: (A) tetrathiaꢀTAM and (B) tetrachloroꢀTAM radicals.
In the current work, trityl radicals are studied in order to promote their applicability as EPR
oxygen sensors. The study starts with their accessibility, describing reproducible syntheses to
obtain various chemically wellꢀdefined radicals. Hydrophilic trityls, which can be directly used
as molecular oxygen reporters, were investigated in aqueous media mimicking physiological
conditions. Lipophilic trityls and pharmaceutical relevant oils were selected for encapsulation to
develop biocompatible nanosensors with high oxygen sensitivity, specificity, and stability for in
vitro and in vivo EPR oximetry.
RESULTS AND DISCUSSION
Synthesis of tetrathia-TAM radicals, their deuterated analogs, and tetrachloro-TAM esters
The criteria for trityl radicals that will be useful in pharmaceutical analysis and medical diagnoꢀ
sis are stringent, as detailed in the Introduction. We chose to evaluate the preparation of tetrathiaꢀ
and tetrachloroꢀTAM radicals. Both classes and some individual compounds that were of interest
6
ACS Paragon Plus Environment

Page 7 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
to us were known. This is indicated in the Experimental section. However, trityl radicals are
known to be difficult to prepare, and we present herein protocols that worked reproducibly in our
hands for known and new derivatives. The synthesis of tetrathiaꢀTAM radicals is shown in
Scheme 1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OH
3
C
S
S
S
S
R1
R1
R1
R1
R1
R1
R1
R1
S
S
S
S
S
S
S
ii
i
S
1
1
1
2a: R = CH
3a: R = CH
3
3
1
1
2b: R = CD
3b: R = CD
3
3
iii
OH
3
3
CH2 3
C
C
R1
R1
R1
R1
R1
R1
R1
R1
R1
R1
R1
R1
S
S
S
S
S
S
S
S
S
S
for 9
v ꢀ viii
iv
S
S
R2
R2
R2
1
2
1
2
1
2
7: R = CH , R = COOCH CH
4: R = CH , R = COOCH CH
11: R = CD , R = COOCH CH (via
v
)
3
2
3
)
3
2
3
3
2
3
1
2
1
2
1
2
8: R = CH , R = COOC(CH )
5: R = CH , R = COOC(CH )
3 3 3
12: R = CD , R = COCl (via vi
13: R = CD , R = COOCD CD (via vii or viii
3
3
3 3
1
2
1
2
1
2
9: R = CD , R = COOH
6: R = CD , R = COOC(CH )
3 3 3
)
3
2
3
3
1
2
10: R = CH , R = COOH
3
Scheme 1. Synthesis of tetrathiaꢀTAM radicals 7ꢀ13. Reagents and conditions: (i) HBF4, acetone
or acetoneꢀd6; (ii) nꢀBuLi, methyl chloroformate; (iii) nꢀBuLi, TMEDA, and diethyl carbonate or
DiBoc; (iv) BF3×Et2O/SnCl2, TFA; (v) 9, TEA, acetonitrile, ethyl chloroformate, DMAP; (vi) 9,
SOCl2, TEA; (vii) 12, ethanolꢀd6, reflux; (viii) 9, TEA, acetonitrile, ethyl chloroformateꢀd5,
DMAP.
The main hurdle to obtain tetrathiaꢀTAM radicals lies right at the beginning of the synthesis,
i.e., the preparation of molar amounts of tetrakis(tertꢀbutylthio)benzene (1) because of the necesꢀ
sity to use the very odoriferous tertꢀbutyl thiol that is added to city gas to detect leakages. We
previously published a manageable technique to obtain 1.39 The subsequent steps towards the
preparation of the tetrathiaꢀTAM radicals shown in Scheme 1 mainly followed published proceꢀ
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 45
1
2
dures.9,10,21ꢀ23,37,40ꢀ42 Modifications regarding the synthetic procedures are detailed in the Experiꢀ
mental Section. Compound 1 was converted into the cyclized arene (2a, b) by tetrafluoroboric
acid (HBF4) 54 % in Et2O, toluene, and acetone or acetoneꢀd6. The triarylmethanols (3a, b) were
prepared by treatment of 2a or 2b with nꢀBuLi and the subsequent addition of methyl chloroforꢀ
mate. Reaction of 3a or 3b with ten equivalents of nꢀBuLi and tetramethylethylenediamine
(TMEDA) resulted in the formation of the corresponding trianion. Different derivatives (comꢀ
pounds 4-6) were prepared through reaction of the appropriate anion with diethyl carbonate (4)
or diꢀtertꢀbutyldicarbonate (DiBoc) (5,6). Esters 5 and 6 were subsequently hydrolyzed with
trifluoroacetic acid (TFA) overnight at room temperature (RT) to give the tricarboxylic acid
(hydrophilic) radicals 9 and 10. Thin layer chromatography (TLC) of radical 9 revealed the
presence of two products: A green spot identified as the target radical 9 and a violet side product
identified by MS to be the quinone methide (20, see Figure 2) that was recently reported.41 In
contrast to the report and according to MPLC isolation, only 3 % of compound 20 had formed.
This low amount showed no effect on the line shape of the EPR signal of the trityl radical in the
solution. Other reports on the preparation of 9 do not mention the formation of 20. It seems to be
formed invariably, its amount increasing on exposure of 9 to light and air.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
ACS Paragon Plus Environment

Page 9 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Chemical structure of the quinone methide (20) – the product of oxidative decarboxyꢀ
lation reaction of radical 9.
The conversion of compounds 4 and 5 by boron trifluoride diethyl etherate into the carboꢀ
cation followed by reduction with stannous chloride gave the lipophilic trityl radicals 7 and 8. A
mixture of radical 9 and triethylamine (TEA) in acetonitrile was kept at 0 °C. Ethyl chloroforꢀ
mate was added followed by the addition of 4ꢀdimethylaminopyridine (DMAP) to afford the
ethyl ester (11) in 80 % yield. Formation of the fully deuterated ethyl ester analogue 13 was
achieved through two methods. Method A involved the esterification of 9 through a twoꢀstep
reaction: The formation of the corresponding acid chloride 12 with the help of thionyl chloride
and TEA followed by reflux with ethanolꢀd6 to give radical 13 with 73 % yield. Method B inꢀ
cluded the synthesis of ethyl chloroformateꢀd5 and its reaction with a mixture of 9 and TEA
followed by addition of DMAP as described before.
The synthesis of tetrachloroꢀTAM radicals is shown in Scheme 2. Compound 15 was syntheꢀ
sized by FriedelꢀCrafts alkylation of 1,2,4,5ꢀtetrachlorobenzene (14) with CHCl3 in the presence
of AlCl3.43 Reaction of 15 with ten equivalents of nꢀBuLi and TMEDA at low temperature gave
the corresponding trianion. Again, a tenfold excess of the reagent, in this case ethyl chloroforꢀ
mate, had to be used to form compound 16 in 81 % yield.29 Most likely due to the bulkiness of
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 45
1
2
3
4
5
6
7
8
9
the tertꢀbutyl group, not less than 50 equivalents of DiBoc were necessary to form compound 17
in an acceptable yield of 49 %. Finally, the radicals 18 and 19 were achieved from the correꢀ
sponding methane derivatives 16 and 17 through reaction with tetrabutylammonium hydroxide
and pꢀchloranil in tetrahydrofuran (THF).44
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2. Synthesis of tetrachloroꢀTAM radicals 18,19. Reagents and conditions: (i)
CHCl3/AlCl3; (ii) nꢀBuLi, TMEDA, and ethyl chloroformate or DiBoc; (iii) Bu4NOH/pꢀ
chloranil, THF.
Characterization of hydrophilic trityl radicals in aqueous solution by EPR spectroscopy
The line width of soluble EPR spin probes, such as trityl radicals, is directly proportional to the
concentration of dissolved oxygen, rendering them suitable for EPR oximetry.29,45 In this section,
we investigate the specificity of hydrophilic trityl radicals for oxygen in aqueous solution which
is related with the impact of other environmental parameters, e.g., the viscosity and pH value on
the EPR signal of radical 9.
Influence of ionic strength, osmolarity, and pH value:
The hydrophilic radicals were studied under several conditions to mimic environments in drug
formulations or in vivo. Solutions of the radical 9 (c = 50 µM) in phosphate buffer (PB; pH 6.2
10
ACS Paragon Plus Environment

Page 11 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
and 7.4) as well as phosphate buffered saline (PBS; pH 6.2 and 7.4) were used to investigate the
impact of ionic strength and pH value on EPR line widths under aerated (20.9 % O2) and deoxyꢀ
genated (~ 0 % O2) conditions. PBS has ionic strengths and osmolarities which are similar to
physiological values, whereas PB has lower ionic strengths/osmolarities. The pH values were
chosen to cover the pertinent range occurring in human tissues, namely 7.0 to 7.4 in normal
tissues and 6.2 to 7.4 in tumors, respectively.46
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Aerated solutions of compound 9 in the different buffer systems showed similar line widths
(ꢁBPP = (13.7 ± 0.2) µT) as well as similar EPR line width narrowing (ꢁBPP = (3.1 ± 0.1) µT)
after deoxygenation of the samples (Figure 3). The oxygen solubility of aqueous solutions deꢀ
creases with increasing salt content.47 PBS buffers have a higher ionic strength than PB buffers.
This difference was not reflected in a change of line width. The pKa of the undeuterated derivaꢀ
tive (radical 10) was determined to be approx. 4.10 Thus, at both pH values investigated, the
radical should mainly exist in its deprotonated form, without any impact on its EPR properties.
In summary, Figure 3 shows that within physiologically relevant limits of ionic
strength/osmolarity and pH value, there was negligible effect of these parameters on the EPR line
width of radical 9.
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 45
1
2
3
4
5
6
7
8
9
PBS 7.4
PB 7.4
PBS 6.2
PB 6.2
14
12
10
8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
4
2
0
0
20.9
Oxygen content (%)
Figure 3. EPR line widths of radical 9 (c = 50 µM) in PB (pH 6.2 and 7.4) and PBS (pH 6.2 and
7.4) under aerated (20.9 % O2) and deoxygenated (~ 0 % O2) conditions (mean ± SD, n = 3).
Impact of viscosity:
To investigate the effect of viscosity on the apparent EPR line width of hydrophilic trityl radiꢀ
cals, radical 9 was dissolved (c = 50 µM) in different glycerolꢀwater mixtures (0 % – 90 % glycꢀ
erol in water, m/m). In absolute glycerol, radical 9 was insoluble. With increasing percentage of
glycerol, oxygen solubility and, hence, concentration of dissolved oxygen are decreased.48 Acꢀ
cordingly, under aerobic conditions, between 0 % and 60 % (m/m) glycerol content, a decrease in
the EPR line width was detected. It was followed by a sharp increase caused by the strong inꢀ
crease of viscosity of the glycerolꢀwaterꢀmixtures above 50 % (m/m) glycerol content (Figure 4).
To support this assumption, the measurements were repeated under deoxygenated conditions. It
was found that up to 40 % (m/m) of glycerol, the impact of viscosity on the EPR line width was
negligible if compared to the effect of oxygen (see section Oxygen calibration). A sharp increase
in the EPR line width followed, similar to the aerated samples. Our findings are in agreement
12
ACS Paragon Plus Environment

Page 13 of 45
The Journal of Organic Chemistry
1
2
with the literature.11 These results are important with regard to biomedical applications as blood
3
4
5
6
7
8
has a dynamic viscosity of
η
= 3ꢀ4 mPa⋅s at 37 °C,49 which is only reached at 40 % (m/m) of
glycerol in water at 20 °C.50 So provided that there is no specific interaction with biological
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
structures, viscosity by itself should have hardly any effect on EPR line width when using the
radicals dissolved in plasma or blood.
A
20.9 % Oxygen
0 % Oxygen
25
20
15
10
5
0
0
20
40
60
80
Glycerol content (%, m/m)
Figure 4. Change in the apparent EPR line width of radical 9 (c = 50 µM) on different glycerolꢀ
water mixtures.
EPR spectroscopic characterization of lipophilic trityl radicals in oily solution
The triꢀesters of the trityl radicals investigated in this work (7, 8, 11, 13, 18, and 19; see Scheme
1 and Scheme 2) proved to be very lipophilic. When these radicals were distributed between
octanol and water or middleꢀchain triglycerides (MCT) and PBS (pH = 7.4) for 2 h at 37 °C, the
EPR signal resided in the lipophilic solvent (octanol or MCT) exclusively.
Hence, these trityl radicals were analyzed in oily solution. MCT was chosen because it is genꢀ
erally considered to be biologically inert and it already has a long tradition as an ingredient in
medicinal products, such as parenteral nutrition nanoemulsions.51 However, for our purposes, it
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 45
1
2
3
4
has a relatively high viscosity of about
η
= 25ꢀ33 mPa·s at 20 °C. As a consequence, the EPR
5
lines are broadened. This may affect the resolution of the EPR spectra and complicate EPR studꢀ
ies. In addition, signalꢀtoꢀnoise ratios are generally lower in highly viscous media since the EPR
amplitude is inversely proportional to the square of the line width.52 As an alternative, the less
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
viscous isopropyl myristate (IPM;
η = 5ꢀ6 mPa·s at 20 °C) was tested. It is also nonꢀtoxic and
commonly used, e.g., in topical pharmaceutical formulations.53
Solutions of the lipophilic radicals 7, 8, 11, 13, 18, and 19 in MCT and IPM (c = 1 mM) were
investigated in air (20.9 % O2) and after flushing with nitrogen (~ 0 % O2). Figure 5 shows the
EPR spectra of several trityl radicals in IPM under anoxic conditions. The estimated line widths
are listed in
. As predicted from the higher viscosity, the EPR lines under anoxic conditions were broader
in MCT. In air, however, the lines were broader in IPM. This is attributed to the higher oxygen
solubility and correspondingly higher oxygen content, overruling the effect of viscosity (see next
section).
14
ACS Paragon Plus Environment

Page 15 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
Radical 7
10 ꢀT
(7 hfs lines)
Radical 8
9 ꢀT
Radical 11
7 ꢀT
(7 hfs lines)
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Radical 13
12 ꢀT
Radical 18
Radical 19
23 ꢀT
23 ꢀT
100 ꢀT
Figure 5. EPR signals of radicals 7, 8, 11, 13, 18, and 19 and line widths under anaerobic condiꢀ
tions (~ 0 % O2) in IPM (c = 1 mM) (hfs = hyperfine splitting). The small side bands caused by
hyperfine couplings with 13C were not used in this analysis.
Under anaerobic conditions, the signal of radical 11 displayed superhyperfine coupling with
the six equivalent protons of the three methylene groups of the ethyl ester moieties resulting in
seven equidistant lines with relative intensities of 1 : 6 : 15 : 20 : 15 : 6 : 1, a coupling constant of
a = 11.3 µT. The coupling pattern was much better resolved in the less viscous IPM. The same
ester, but with protonated instead of deuterated methyl groups in the ketal moiety (radical 7),
resonated with slightly broader EPR line widths compared to radical 11. The additional superhyꢀ
perfine coupling with the protons of the methyl groups affected the signal pattern in deoxygenatꢀ
ed solutions (Figure 5). In MCT, only the envelope of the coupling pattern was visible.
Deuterium with a nuclear spin of 1 of course also couples with the electron, but the coupling
constant is much smaller compared to the coupling constant of hydrogen.
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 45
1
2
3
4
5
6
7
8
9
The tertꢀbutyl ester of the nonꢀdeuterated trityl (radical 8) displayed the narrowest single EPR
line we observed for the lipophilic esters investigated. As expected, due to the small unresolved
hfs with the protons of the tertꢀbutyl moieties, the line width in anoxic IPM was slightly broader
when compared to the hfs lines of radical 11. Fully deuterated ethyl ester and methyl ketal
groups (radical 13) also led to exceptionally narrow single lines. Compared to radicals 8 and 11,
the lines were slightly broader in anoxic solutions caused by the unresolved coupling with deuꢀ
terium nuclei. However, because radical 13 had a better synthetic accessibility than 8, it was
chosen for the following oxygen calibration measurements and encapsulation investigations.
Due to the chlorine splitting, the tetrachloroꢀTAM triꢀesters generally had broader EPR lines
than the tetrathiaꢀTAM derivatives. The tertꢀbutyl ester (radical 19) displayed the same line
width as the triꢀethyl ester (radical 18) in IPM, while in MCT, the line widths were slightly
broader. The extremely narrow anoxic line width of 18 and 19 in IPM is remarkable. To our
knowledge, only in deoxygenated DMSO (viscosity: 2 mPa·s at 20 °C), a line width < 30 µT
(namely approx. 28 µT) was reported for tetrachloroꢀTAM radicals in the literature.29 Even when
dissolved in deoxygenated hexafluorobenzene, the line width of radical 18 was approx. 38 µT,
although the viscosity is only 1.2 mPa·s (at 20 °C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Oxygen calibration
In this section, the applicability of the selected trityl radicals as oxygenꢀsensitive spin probes was
evaluated. For atmospheric pressure and in vivo applications, only the data between approx. 0
and 20.9 % (0–156 mmHg) oxygen are of interest. Solutions of the hydrophilic radicals 9 and 10
(c = 50 µM) in PBS (pH = 7.4) showed a linear relationship between the EPR line width and
oxygen concentration with similar oxygen sensitivities as reported23,45 (Figure 6A). As expected,
radical 9 had a narrower line width (ꢁBPP = (13.6 ± 0.2) µT at 20.9 % O2) than its undeuterated
16
ACS Paragon Plus Environment

Page 17 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
analogue 10 (ꢁBPP = (16.5 ± 0.1) µT at 20.9 % O2). The slope of the curves, i.e., the oxygen
sensitivity, was with about 0.5 µT/% O2 (0.07 µT/mmHg) quite small. As mentioned before, the
EPR lines of soluble spin probes are broadened by Heisenberg exchange between the probes and
molecular oxygen in solution. According to the Smoluchowski equation, the higher the concenꢀ
tration of dissolved oxygen, the higher is the bimolecular collision rate. Due to the low solubility
of oxygen in water (the oxygen content in water is only approx. 0.6 % (v/v) at 22 °C and 0.213
bar oxygen),47 the concentration of dissolved oxygen increases only slightly with increasing
oxygen partial pressure. Hence, there is only small line broadening, a physicochemical fact
affecting any oxygen determination by soluble EPR spin probes in water that is rarely stated.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
B
IPM
MCT
A
PBS 7.4
Radical 10
Radical
Radical 13
Radical 18
Radical 19
9
100
80
60
40
20
0
15
10
5
0
0
5
10
15
20
0
5
10
15
20
Oxygen content (%)
Oxygen content (%)
Figure 6. Oxygen calibration curves of (A) hydrophilic trityl radicals in aqueous solution (c = 50
µM) and (B) lipophilic triꢀesters dissolved in MCT and IPM (c = 1 mM) (n=3).
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 45
1
2
3
4
The oxygen sensitivities of the lipophilic radicals 13, 18, and 19 (c = 1 mM) were about 1.7
5
6
7
8
µT/% O2 (0.2 µT/mmHg) in MCT and 3.7 µT/% O2 (0.5 µT/mmHg) in IPM (Figure 6B and
).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. EPR line widths of different lipophilic TAM triꢀesters dissolved in MCT and IPM (c = 1
mM) in air (20.9 % O2) as well as after flushing with nitrogen (~ 0 % O2) and their correspondꢀ
ing oxygen sensitivities reported as mean ± SD (n = 3; n.m. = not measured).
MCT
IPM
Oxygen
sensitivity
Oxygen
sensitivity
EPR line width (µT)
EPR line width (µT)
Radical
20.9 %
~ 0 % O2 20.9 % O2 µT/% O2
µT/mmHg
n.m.
~ 0 % O2
O2
µT/% O2
µT/mmHg
15
57
10
94
n.m.
n.m.
n.m.
7
8
(7 hfs lines)
(7 hfs lines)
n.m.
10
(7 hfs lines)
n.m.
53
n.m.
n.m.
9
7
88
91
11
(7 hfs lines)
15
51
53
44
82
90
1.47 ± 0.01 0.20 ± 0.01
1.67 ± 0.12 0.23 ± 0.02
1.84 ± 0.08 0.25 ± 0.02
12
23
23
83
98
98
3.59 ± 0.13 0.48 ± 0.02
3.83 ± 0.19 0.52 ± 0.03
3.75 ± 0.10 0.51 ± 0.02
13
18
19
Oxygen solubility in oils is generally higher than in water at a given temperature and presꢀ
sure,54 such that the concentration of dissolved oxygen rises noticeably with increasing oxygen
partial pressure. Accordingly, the slopes were steeper than in water. By gas chromatography, we
determined oxygen contents of 2.2 % (v/v) in MCT and 2.9 % (v/v) in IPM at 22 °C and 0.213
bar oxygen, both markedly above the oxygen content of water. The different slopes in MCT and
IPM were probably caused by the different polarity and especially viscosity of the oils. Remarkꢀ
ably, chemically different radicals, e.g., radicals 13 and 18, had similar oxygen sensitivities when
dissolved in the same solvent. Smaller deviations might be attributed to resolved or unresolved
hfs. So obviously, the oxygen sensitivity was rather affected by the nature of the solvent than the
18
ACS Paragon Plus Environment

Page 19 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
properties of the radical. However, the yꢀintercepts, i.e., line widths under anoxic conditions,
were quite unlike for different radicals as discussed in the previous section.
7
8
9
Of all investigated spin probes and solvents, a solution of radical 13 in IPM was considered
best for oxygen measurements: It had the steepest slope with the smallest yꢀintercept. Hence, it
combined both excellent oxygen sensitivity and good signalꢀtoꢀnoiseꢀratio. The high oxygen
sensitivity in IPM is comparable with particulate oxygenꢀsensitive spin probes, such as Lithium
phthalocyanine (LiPc)55 or Lithium octabutoxynaphthalocyanine (LiNcꢀBuO)56 and, thus, very
promising for future applications in the field of EPR oximetry as an alternative to particulate
materials. For instance, trityl radicals can be employed in lipophilic formulations as it was reꢀ
ported in the literature, e.g., with microspheres,17 hexafluorobenzene nanoemulsions or soluꢀ
tions,14,15 or polydimethyl siloxane chips.16 The present work shows the incorporation into
nanocapsules (see next section).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Encapsulation of trityl esters and properties of the resulting nanocapsules (NCs)
Our studies had shown that to some extent the EPR properties of different water soluble trityl
radicals were independent of their environment, excepting the parameter of interest, viz. oxygen.
However, during in vitro or in vivo experiments, conditions can change drastically. Therefore, it
was assumed that formulations such as NCs might be a helpful approach. If dissolved in the oily
core of NCs, the spin probes are located in a constant microenvironment without being affected
by changes of the outer pH value or viscosity. Thereby, NCs can also be used in acidic conditions
where hydrophilic trityl radicals would precipitate. Moreover, oxygen molecules can penetrate
the capsule shell, turning NCs into oxygen sensors. Since the oxygen solubility in oils is higher
than in water, the oxygen sensitivity of the spin probes would be improved by encapsulation (see
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 45
1
2
3
4
5
6
7
8
9
section above). Highly lipophilic encapsulated probes would stay inside the NCs without being
partitioned to the outer aqueous phase, shielding them from oxidoreductants. It is known that
suitable formulations are needed to prevent spin probeꢀtissue interactions in longꢀterm in vivo
studies.12 So not only the oxygenꢀresponsiveness of the spin probes would be preserved but also
their bioꢀcompatibility improved.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of the NCs:
The nonꢀbiodegradable polymer poly(vinyl acetate) (PVAc), which already had been successꢀ
fully used as a polymer matrix for particulate spin probes,57 was chosen for encapsulation of the
oily spin probe solutions. NCs with different concentrations (varying from 0.1 to 2 %) of PVAc
(m/m) and MCT (v/m) in the organic phase were prepared. As a hydrophilic stabilizer, poloxꢀ
amer 188 (0.25 % (m/m) in the aqueous phase) was used, and acetone as the organic solvent. In
order to obtain the best signalꢀtoꢀnoise ratio of the EPR spectra, the amount of oil inside the NCs
needed to be high. On the other hand, the polymer shell surrounding the oily core should be as
thin as possible since oxygen has to permeate through the shell. Therefore, NCs should have
lowest possible PVAc and highest possible oil contents. It was found that a concentration of 0.2
% (m/m) PVAc and 1.4 % (v/m) MCT was optimal. In contrast to MCT, it was not possible to
obtain stable NCs containing IPM using poloxamer 188. Therefore, other stabilizers were tested
and polysorbate 80 proved to be suitable. The best formulation contained 0.2 % (m/m) PVAc and
1.2 % IPM (v/m) in the organic phase and 0.2 % (m/m) polysorbate 80 in the aqueous phase.
Particle size, zeta potential (ZP), morphology and stability of the NCs:
20
ACS Paragon Plus Environment

Page 21 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
The size of the NCs was determined by photon correlation spectroscopy (PCS). The hydrodyꢀ
namic diameter (Zꢀaverage) typically was about 160 nm for IPM NCs and about 150 nm for
MCT NCs. The particle size distributions were monomodal and monodisperse, indicated by a
very low polydispersity index (PdI) < 0.1. Figure 7A shows typical particle size distributions of
MCT and IPM NCs. The morphology and shell thickness of the NCs were investigated using
cryoꢀ and freezeꢀfracture transmission electron microscopy (TEM). The capsules were spherical
in shape, their size varied between approx. 50 and 180 nm (Figure 7B). The deviation from the
PCS data might be explained by a particle selection, which usually occurs during sample prepaꢀ
ration for cryoTEM. Therefore, the PCS data are considered to represent particle size better.
Further, the PVAc shell thickness was determined. Fractured NCs clearly showed a structure with
core and shell (Figure 7C). The shell thickness was with about 8 nm very thin, which is in acꢀ
cordance to the literature, where a shell thickness of 10 nm had been found for poly(lacticꢀcoꢀ
glycoliv acid) (PLGA) NCs.58 The ZP, as measured in double distilled water by laser Doppler
electrophoresis, was found to be about 19–22 mV, providing relatively good colloidal stability.
However, it must be stated that due to the stabilizers used, the main stabilization mechanism
probably was static repulsion. The pH value of the dispersions was about 8, which was in a
physiologically well tolerated range.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 45
1
2
3
4
A
MCT
IPM
20
15
10
5
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
1
10
100
Size (nm)
1000
Figure 7. (A) Typical particle size distributions of MCT and IPM NCs. (B) CryoꢀTEM image of
MCT NCs and (C) Freezeꢀfracture TEM image of an IPM NC with an estimated shell thickness
of about 8 nm (bar size: 50 nm).
In order to examine the stability of the NCs, their size and ZP as well as the pH value of the
dispersion were not only measured on the day of preparation but also up to three months of
storage at 2–8 °C (see Figure 8A,B).
Size PdI
A
pH
B
ZP
IPM
IPM
MCT
MCT
200
150
100
50
0.9
0.6
0.3
0.0
ꢁ25
ꢁ20
ꢁ15
ꢁ10
ꢁ5
10
9
8
7
0
0
6
0
4
7
14
Day
30
90
0
7
14
30
90
Day
Figure 8. Change in (A) Zꢀaverage and PdI of the NCs as well as (B) ZP of the NCs and the pH
value of the dispersion on duration of storage at 2–8 °C (n = 3; median and range are shown).
22
ACS Paragon Plus Environment

Page 23 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Whereas MCT NCs were stable over the whole measurement period, IPM NCs showed instaꢀ
bilities. The Zꢀaverage as well as the PdI changed over time of storage. Especially the increased
PdI (from < 0.1 to > 0.2 after 30 days of storage) indicates aggregation. After one week of storꢀ
age, creaming was observed for the IPM NCs, but not for MCT NCs. The reason might be the
lower density of IPM with about 850 kg/m3 compared to MCT with a density of about 950
kg/m3. The NCs were redispersable by gentle shaking. However, it is assumed that creaming is
the reason why IPM NCs tend to aggregate more because of the spatial proximity of the partiꢀ
cles. Hence, IPM is less suitable for the preparation of longꢀterm stable NCs. These findings are
consistent with the literature.59 For both oils used, the ZP was stable, whereas the pH value
slightly decreased over time, which may have been due to a partial hydrolysis of PVAc and,
therefore, release of acetic acid.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Oxygen calibration of the NCs:
The NCs prepared had similar oxygen sensitivities as the unencapsulated solutions (Figure 9).
Thus, the encapsulation of the spin probes did not alter their oxygen sensitive properties. Oxygen
obviously diffuses through the polymer shell without any hindrance. The process was reversible
(data not shown), meaning that no oxygen was accumulated inside the NCs.
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 45
1
2
3
4
5
6
7
8
9
100
80
60
40
20
0
IPM NCs
MCT NCs
IPM solutions
MCT solutions
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
5
10
15
20
Oxygen content (%)
Figure 9. Oxygen calibration curves of oily solutions of radical 13 both incorporated into NCs
and unencapsulated. The calculated oxygen sensitivities were (0.46 ± 0.02) µT/mmHg (IPM) and
(0.22 ± 0.01) µT/mmHg (MCT) for the NCs compared to (0.48 ± 0.02) µT/mmHg (IPM) and
(0.20 ± 0.01) µT/mmHg (MCT) for the unencapsulated solutions (n = 3).
Ascorbic acid reduction assay:
Since the lipophilic triꢀesters (e.g., radical 13) are uncharged, they may be less protected from
reduction by negatively charged reductants since there is no electric repulsion.18,23 To investigate
the stability of encapsulated radicals against reduction, an ascorbic acid reduction assay was
carried out. For this, NCs containing solutions of either tempol benzoate (TB) or radical 13 in
MCT (c = 1 mM) were prepared. TB is a nitroxide with a log P of approx. 2.60 It was chosen for
comparison because even though it is lipophilic enough for encapsulation, TB is still soluble in
water. It is able to partly diffuse through the polymer shell and partition between the two phases.
Once outside the NC, it can be reduced by ascorbic acid. Hence, it serves as a positive control.
As can be seen in Figure 10, TB was reduced by ascorbic acid. The EPR signal intensity deꢀ
creased to about 11 % of the start value within 240 min. Comparable results had also been found
for PLGA NCc.61 In contrast, radical 13 seemed to stay shielded inside the NCs as it was not
reduced by ascorbic acid after 240 min. The result proves that the capsule shell is capable to
24
ACS Paragon Plus Environment

Page 25 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
sterically shield the incorporated radical from reductants. Moreover, it shows that a very high
lipophilicity and correspondingly low water solubility is crucial for effective protection of encapꢀ
sulated radicals.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A
B
Radical 13 NC
100
80
60
40
20
0
0 min
30 min
60 min
120 min
TB NC
180 min
240 min
ꢁ2
ꢁ1
0
1
2
0
50
100
150
200
250
Time (min)
Relative magnetic field (mT)
Figure 10. (A) EPR spectra of NCs containing TB during reduction by ascorbic acid. Note that
the dublett in central position arises from ascorbic acid. (B) Change of the EPR signal intensity
of NCs with incorporated TB compared to NCs with radical 13.
CONCLUSION
Reproducible protocols for the synthesis of different derivatives of tetrathiaꢀ and tetrachloroꢀ
TAM radicals are described. A new readily accessible trityl derivative (radical 13) with deuteratꢀ
ed core and deuterated ethyl ester groups was devised. The EPR properties of the synthesized
radicals were investigated in either aqueous media (hydrophilic radicals) or MCT and IPM (lipoꢀ
philic triꢀesters). It was shown that within physiological limits of osmolarity, viscosity, and pH
value, there was, for practical purposes, no impact on the EPR line widths of the hydrophilic
radicals. The nonꢀdeuterated lipophilic triꢀethyl esters of CꢀTAM or DꢀTAM (radicals 7 and 11)
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 45
1
2
3
4
5
6
7
8
9
showed hyperfine splitting in deoxygenated solutions. Other triꢀesters exhibited single EPR lines
with line widths directly proportional to the oxygen concentration, rendering them candidates for
oxygen measurements. The fully deuterated triꢀethyl ester of DꢀTAM (radical 13) had the most
promising EPR properties, especially when dissolved in IPM. It had a very narrow EPR line
under anoxic conditions and high oxygen sensitivity (~ 0.5 µT/mmHg). Therefore, solutions of
radical 13 in MCT and IPM were encapsulated into NCs. The oxygen responsiveness of the
incorporated radical was retained. Using IPM for encapsulation, a high oxygen sensitivity of the
NCs could be ensured, especially suitable for measuring low oxygen contents. MCT provided
stable NCs for longꢀterm measurements. In addition, despite being uncharged, encapsulated
lipophilic trityl radicals were protected against reducing agents such as ascorbic acid. Hydroꢀ
philic trityl radicals were shown to be useful EPR oximetry probes, but since the concentration of
dissolved oxygen in water does not change much with oxygen partial pressure, lipophilic radicals
in oily core NCs led to higher oxygen sensitivity than any water soluble probe can provide.
Encapsulation of lipophilic trityl radicals offers potential for nanosensors with high oxygen
sensitivity, specificity, and stability, particularly suitable for EPR oximetry in complex biological
systems.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
Materials.
All chemicals used for synthesis were purchased and used without further purification. Mediumꢀ
chain triglycerides (Pionier® MCT) and isopropyl myristate (Pionier® IPM) were purchased from
26
ACS Paragon Plus Environment

Page 27 of 45
The Journal of Organic Chemistry
1
2
Hansen & Rosenthal KG (Hamburg, Germany), Poloxamer 188 (Lutrol® F 68) from BASF SE
(Ludwigshafen, Germany), and Polysorbate 80 from Caesar & Loretz GmbH (Hilden, Germany).
BASF SE kindly provided Poly(vinyl acetate) (PVAc) dispersion 30 % (Kollicoat® SR 30 D).
Water was used in doubly distilled quality.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
General Methods for Synthesis.
All organic solvents were purified and dried before use and stored over molecular sieves (3 Å).
Glassware for reactions under argon atmosphere was ovenꢀdried at 100 °C for 2 h prior 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). Visualizations were accomplished with an 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 by flash chromatography.
Chemical shifts (δ) are reported in parts per million (ppm) relative to the residual nonꢀdeuterated
solvent peak in the corresponding spectra (chloroform δ = 7.26, methanol δ = 3.31, DMSO δ =
2.49). 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 δ values (ppm) relative to the residual nonꢀdeuterated solvent
peak in the corresponding spectra (chloroform δ = 77.2, methanol δ = 49.0, DMSO δ = 39.5).
The samples were analyzed on an orbitrap XL mass spectrometer with a resolving power of
100000 at m/z 400, samples were introduced to the MS by static nanoꢀelectrospray ionization.
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 45
1
2
2,2,6,6ꢀTetramethylbenzo[1,2ꢀd:4,5ꢀd']bis[1,3]dithiole (2a).10,40,42
3
4
5
6
7
8
HBF4 (54 % in Et2O, 10 mL, 37.0 mmol) was added to a solution of 1,2,4,5ꢀtetra(tertꢀ
butylthio)benzene (1) (16.0 g, 37.2 mmol) in toluene (500 mL). The mixture was stirred for 30
min at RT. Acetone (10 mL, 136 mmol) was added and the reaction mixture was stirred for 4 h at
RT, and then heated to reflux overnight. After cooling, a saturated NaHCO3 solution (100 mL)
was added carefully, the organic layer was separated and the aqueous layer extracted three times
with ethyl acetate (EA). The combined organic layers were dried over MgSO4 and concentrated
in vacuum. Ethanol (50 mL) was added to the brown solution. The pure product was precipitated
and collected by filtration, washed several times with EtOH and dried to give 7.2 g (68 % yield)
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
of white crystals, mp 145–147 °C, Rf = 0.32 (heptane). H NMR (400 MHz, CDCl3): δ 7.02 (s,
2H), 1.88 (s, 12H). 13C NMR (100 MHz, CDCl3): δ 135.7, 116.8, 65.7, 31.3. HRMS (ESI): calcd
for C12H14S4 [M]+ 285.998; found 285.997.
2,2,6,6ꢀTetra(2H3ꢀmethyl)benzo[1,2ꢀd:4,5ꢀd']bis[1,3]dithiole (2b).9,21,22 The rocedure used for
2a was applied for the synthesis of 2b using HBF4 (54 % in Et2O, 10 mL, 37.0 mmol), comꢀ
pound 1 (16.0 g, 37.2 mmol) in toluene (500 mL) and acetoneꢀd6 (10 mL, 148 mmol) to afford
1
compound 2b 7.8 g (70 % yield) as white solid, mp 143–145 °C, Rf = 0.35 (heptane). H NMR
(400 MHz, CDCl3): δ 7.02 (s, 2H). 13C NMR (100 MHz, CDCl3): δ 135.7, 116.8, 65.7, 31.3.
HRMS (ESI): calcd for C12H2D12S4 [M]+ 298.073; found 298.073.
Tris[2,2,6,6ꢀtetramethylbenzo[1,2ꢀd:4,5ꢀd']bis([1,3]dithiole)ꢀ4ꢀyl]methanol (3a).10,40,42 Comꢀ
pound 2a (3.5 g, 12.2 mmol) was dissolved in dry diethyl ether (Et2O) (150 mL) under argon
atmosphere. A solution of 2.5 M nꢀBuLi in hexanes (4.88 mL, 12.2 mmol) was added dropwise,
and the reaction mixture was stirred for 2 h at RT. Methyl chloroformate (0.32 mL, 4.0 mmol)
was mixed with Et2O (40 mL) and the mixture was added slowly with a perfusion pump (flow
28
ACS Paragon Plus Environment

Page 29 of 45
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
rate 1 mL/h). Saturated NaHCO3 solution (100 mL) was added and the reaction mixture was
allowed to stir till the formed precipitate completely dissolved. The organic layer was separated
and the aqueous layer extracted with EA. The combined organic layers were dried over MgSO4
and solvent was evaporated to dryness in a vacuum. The resulting solid was heated at reflux in a
mixture of CCl4 and hexane (1/1, v/v) for 15 min. After cooling, the yellow solid was collected,
washed with CCl4/hexane (1/1, v/v), and dried in a vacuum to give 1.7 g (47 % yield) of yellowꢀ
ish solid, mp 250–255 °C, Rf = 0.32 (heptane). 1H NMR (400 MHz, CDCl3): δ 7.17 (s, 3H), 6.20
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
(s, 1H), 1.82 (s, 9H), 1.80 (s, 9H), 1.72 (s, 9H), 1.68 (s, 9H). C NMR (100 MHz, CDCl3): δ
139.2, 138.3, 137.8, 137.2, 131.8, 118.1, 83.6, 64.0, 63.3, 34.8, 32.2, 29.1, 27.6. IR (KBr): ν =
3364, 2954, 2921, 1451, 1147, 756 cmꢀ1. HRMS (ESI): calcd for C37H40OS12 [M]+ 883.973;
found 883.973.
Tris[2,2,6,6ꢀ(2H3ꢀtetramethyl)benzo[1,2ꢀd:4,5ꢀd']bis([1,3]dithiole)ꢀ4ꢀyl]methanol (3b).9,21,22
The procedure for the synthesis of 3a was applied for the synthesis of 3b. Compound 2b (3.49 g,
11.7 mmol), nꢀBuLi (4.7 mL, 11.7 mmol) and methyl chloroformate (0.3 mL, 3.9 mmol) in dry
Et2O (150 mL) were used to give 3b 1.62 g (45 % yield) as white to yellow solid, mp 250–255
°C, Rf = 0.35 (heptane). 1H NMR (400 MHz, CDCl3): δ 7.17 (s, 3H), 6.21 (s, 1H). 13C NMR (100
MHz, CDCl3): δ 139.5, 138.6, 138.1, 137.6, 132.2, 118.5, 83.9, 63.9, 63.6, 63.2. IR (KBr): ν =
3360, 2953, 2922, 2217, 1374, 1247, 1182, 1003, 861, 754 cmꢀ1. MS (ESI): m/z 943.12 [M +
Na]+.
Tris[8ꢀethoxycarbonylꢀ2,2,6,6ꢀtetramethylbenzo[1,2ꢀd:4,5ꢀd']bis([1,3]dithiole)ꢀ4ꢀyl]methanol
(4).23,40 Compound 3a (500 mg, 0.57 mmol) and TMEDA (0.85 mL, 5.7 mmol) were mixed in
dry nꢀhexane (5 mL) under argon atmosphere and cooled to 0 °C. A solution of 2.5 M nꢀBuLi in
hexane (2.3 mL, 5.7 mmol) was added dropwise over 30 min and the mixture was stirred at RT
29
ACS Paragon Plus Environment