Synthesis and characterization of sterically and electrostatically shielded pyrrolidine nitroxide radicals

Article abstract of DOI:10.1016/j.molstruc.2019.01.015

Two new pyrrolidine nitroxide radicals, cis/trans-2,5-bis(carboxymethyl)-2,5-diethylpyrrolidine 1-oxyl and 2-(carboxymethyl)-2,5,5-triethylpyrrolidine 1-oxyl, for potential applications as spin probes and labels are reported. Carboxymethyl and ethyl groups have been introduced in the α-positions of the nitroxide group in order to improve the stability of the radicals through steric and electrostatic shielding. The compounds were structurally characterized by X-ray crystallography and EPR spectroscopy. An ascorbic acid reduction assay proves that the newly synthesized radicals exhibit higher reductive stability than the well-known and commercially available nitroxide radicals 3-carboxy-PROXYL and 4-carboxy-TEMPO.

Full text of DOI:10.1016/j.molstruc.2019.01.015

Accepted Manuscript
Synthesis and characterization of sterically and electrostatically shielded pyrrolidine
nitroxide radicals
Lisa Lampp, Uwe Morgenstern, Kurt Merzweiler, Peter Imming, Rüdiger W. Seidel
PII:
S0022-2860(19)30021-3
DOI:
https://doi.org/10.1016/j.molstruc.2019.01.015
MOLSTR 26066
Reference:
To appear in: Journal of Molecular Structure
Received Date: 15 November 2018
Revised Date: 3 January 2019
Accepted Date: 4 January 2019
Please cite this article as: L. Lampp, U. Morgenstern, K. Merzweiler, P. Imming, Rü.W. Seidel, Synthesis
and characterization of sterically and electrostatically shielded pyrrolidine nitroxide radicals, Journal of
Molecular Structure (2019), doi: https://doi.org/10.1016/j.molstruc.2019.01.015.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Synthesis and characterization of sterically and electrostatically shielded
#
pyrrolidine nitroxide radicals
a
b
b
a
a
Lisa Lampp , Uwe Morgenstern , Kurt Merzweiler , Peter Imming* , Rüdiger W. Seidel*
a
Institut für Pharmazie, Martin-Luther-Universität Halle-Wittenberg, Wolfgang-Langenbeck-
Str. 4, 06120 Halle (Saale), Germany
b
Institut für Chemie, Martin-Luther-Universität Halle-Wittenberg, Kurt-Mothes-Straße 2,
0
6120 Halle (Saale), Germany
_
_________________________
#
Dedicated to Professor Martin Feigel on the occasion of his 70th birthday.
*
Corresponding authors
E-Mail: peter.imming@pharmazie.uni-halle.de, ruediger.seidel@pharmazie.uni-halle.de
Abstract
Two
new
pyrrolidine
nitroxide
radicals,
cis/trans-2,5-bis(carboxymethyl)-2,5-
diethylpyrrolidine 1-oxyl and 2-(carboxymethyl)-2,5,5-triethylpyrrolidine 1-oxyl, for potential
applications as spin probes and labels are reported. Carboxymethyl and ethyl groups have
been introduced in the α-positions of the nitroxide group in order to improve the stability of
the radicals through steric and electrostatic shielding. The compounds were structurally
characterized by X-ray crystallography and EPR spectroscopy. An ascorbic acid reduction
assay proves that the newly synthesized radicals exhibit higher reductive stability than the
well-known and commercially available nitroxide radicals 3-carboxy-PROXYL and 4-
carboxy-TEMPO.
Keywords: spin label; nitroxide; radical; crystal structure, EPR spectroscopy
1

ACCEPTED MANUSCRIPT
Introduction
Electron paramagnetic resonance (EPR) spectroscopy is a useful analytical method for the
elucidation of the structure and dynamic behavior of paramagnetic compounds and their
microenvironment [1]. Most polymeric materials and biomolecules are diamagnetic and thus
EPR silent. The absence of an EPR signal in most (bio)materials provides analytical and
diagnostic possibilities through artificial introduction of a paramagnetic centre via spin
probing and labelling [2-4].
•
Radicals based on the nitroxide group (N−O ) are by far still the most widely used
compounds for spin probing and labeling [5]. Typical nitroxide radicals are based on
piperidine, pyrrolidine or other nitrogen heterocycles. In these compounds, the nitroxide
moiety, where the unpaired electron is mainly located, is flanked by two quaternary carbon
atoms, which provide steric shielding. The absence of α-hydrogen atoms prevents nitrone
formation. 3-Carboxy-PROXYL (3-CP) and 4-carboxy-TEMPO (4-CT) are two well-known
and commercially available nitroxide radicals (Scheme 1). In vivo, common nitroxide radicals
are, however, reduced within minutes, which limits biological applications. Thus, attempts
have been made to increase the reductive stability [6, 7].
Scheme 1 Chemical diagrams of 3-carboxy-PROXYL (3-CP) and 4-carboxy-TEMPO (4-
CT).
We have synthesized and studied two new pyrrolidine nitroxide radicals for spin
labelling and potential in vivo use as spin probes. Carboxymethyl and ethyl groups were
tethered to the α-positions of the nitroxide group to achieve higher reductive stability.
Electrostatic shielding through ionizable groups, such as carboxymethyl, could increase the
stability of the radicals further [7, 8]. We report the synthesis and structural elucidation of 2,5-
bis(carboxymethyl)-2,5-diethylpyrrolidine
1-oxyl
and
2-(carboxymethyl)-2,5,5-
triethylpyrrolidine 1-oxyl. The reductive stability of the new radicals was assessed through an
ascorbic acid reduction assay.
2

ACCEPTED MANUSCRIPT
Experimental section
General
Starting materials were obtained from commercial sources and used as received. The nitroxide
radicals 3-CP and 4-CT were purchased from Sigma-Aldrich. Solvents were of reagent grade.
Diethyl ether used for Grignard reactions was dried over sodium metal and freshly distilled
before use. Methanol used for the preparation of a sodium methoxide solution was dried over
molecular sieve. Grignard reactions and reactions with sodium metal were carried out under
argon using standard Schlenk techniques. 6-Nitrooctane-3-one (1) was prepared from 1-
nitropropane and ethyl vinyl ketone following the method described by McMurry and Melton
[
9]. The spectral properties agreed with those reported in the literature [10]. 3-Nitropentane
was synthesized from pentane-3-amine using meta-chloroperoxybenzoic acid as oxidizing
agent, as described by Gilbert and Borden [11]. The spectral properties were in accord with
those published previously [12]. Column chromatography was performed on silica gel 60 (70-
2
30 mesh or 230-400 mesh). The purity of all compounds and the progress of the reactions
were monitored by thin layer chromatography using silica gel 60 F254 plates (Merck KGaA,
Darmstadt, Germany). Visualizations were accomplished with an UV lamp (254 nm) or
1
13
iodine staining. The R values given are uncorrected. H and C NMR spectra were recorded
f
on an Agilent Technologies VNMRS 400 MHz or a Varian Inova 500 MHz spectrometer.
Chemical shifts (δ) are reported relative to the residual solvent peak of CDCl as internal
3
standard: δ = 7.26 ppm, δ = 77.0 ppm. Infrared (IR) spectra were recorded on a Bruker IFS
H
C
2
8 FTIR spectrometer equipped with a Thermo Spectra-Tech attenuated total reflection
(ATR) unit with a 20 mm ZnSe-Fresnel crystal. High resolution mass spectra (HRMS) were
measured on a LTQ-Orbitrap-XL (ESI source) of Thermo Scientific. Samples were dissolved
in chloroform / methanol.
Synthesis
2
,5-Diethyl-3,4-dihydro-2H-pyrrole 1-oxide (2): Zinc powder (33.0 g, 505 mmol) was
added in small portions to a stirred solution of 6-nitrooctane-3-one (22.0 g, 127 mmol) and
ammonium chloride (7.4 g, 138 mmol) in 170 mL of water cooled with an ice bath, ensuring
that the temperature of the reaction mixture did not exceed 10 °C. Subsequently, the mixture
was continued to stir for 12 h while allowing to warm up to room temperature. The reaction
3

ACCEPTED MANUSCRIPT
mixture was filtered through celite and the residue was rinsed with a small amount of
methanol. The filtrate was concentrated in vacuum and extracted seven times with 20 mL of
chloroform. The combined organic layers were dried over magnesium sulfate and the solvent
was removed was removed under reduced pressure. The crude product was purified by
distillation under reduced pressure (0.17 mbar, b.p. 94-96 °C). Yield: 14.60 g (103 mmol, 72
1
%
). H NMR(400 MHz, CDCl ): δ 3.94 (m, J = 10.4, 5.1, 3.4, 1.8 Hz, NCH, 1H), 2.60 (m, J =
3
7
.6, 3.3, 1.7 Hz, CH , 2H), 2.51 (m, J = 12.5, 7.8, 6.2, 3.2, 1.5 Hz, CH 2H), 2.28 – 2.15 (m,
2 2,
CH 1H), 2.09 (m, J = 13.7, 7.6, 3.4 Hz, CH 1H), 1.82 – 1.61 (m, CH 2H), 1.09 (t, J = 7.7
2
,
2,
2,
1
3
Hz, CH , 3H), 0.90 (t, J = 7.5 Hz, CH , 3H). C NMR(101 MHz, CDCl ): δ 147.87, 73.50,
3
3
3
+
2
8.75, 25.16, 22.00, 19.95, 9.35, 8.79. HRMS(ESI): calcd. for C H NO [M+H] 142.1232,
8 16
found 142.1221.
2
-Allyl-2,5-diethyl-3,4-dihydro-2H-pyrrole 1-oxide (3): compound 2 (3.0 g, 21.24 mmol)
dissolved in 7 mL of diethyl ether was added dropwise to a stirred solution (31.9 mL) of
allylmagnesium bromide (1.0 M) in diethyl ether at −10 °C. After stirring for 12 h while
allowing the mixture to warm up to room temperature, 2.5 ml of a saturated aqueous solution
of ammonium chloride and 2.5 mL of water were successively added. After filtration through
a frit, the diethyl ether phase was separated, washed with brine and evaporated. The residue
was taken up with 43 mL of methanol containing 1.1 mL of conc. ammonia and anhydrous
copper(II) acetate (0.12 g, 0.045 mmol). Oxygen was bubbled through the mixture with
stirring until the color turned dark blue. Afterwards, the solvent was removed under reduced
pressure and the residue was taken up with 20 mL of chloroform. After washing successively
with a saturated solution of sodium bicarbonate (3 × 20 mL) and brine (20 mL) and drying
over magnesium sulfate, the solvent was removed under reduced pressure. The crude product
was purified by column chromatography (ethyl acetate/methanol 9:1, R = 0.19). Yield: 2.43 g
f
1
(
13.41 mmol, 60 %). H NMR (400 MHz, CDCl ) δ 5.66 (m, J = 17.1, 10.1, 8.5, 6.1 Hz,
3
CH =CHCH , 1H), 5.18 – 5.06 (m, CH =CHCH , 2H), 2.72 – 2.60 (m, CH =CHCH , 1H),
2
2
2
2
2
2
2
1
.50 (m, J = 7.9 Hz, CH CH , 4H), 2.25 (m, J = 13.8, 8.6, 0.8 Hz, CH =CHCH , 1H), 2.10 –
2 3 2 2
1
3
.50 (m, CH CH , 4H), 1.09 (t, J = 7.6 Hz, CH , 3H), 0.84 (t, J = 7.4 Hz, CH , 3H). C NMR
2
2
3
3
(
101 MHz, CDCl ) δ 147.42, 132.73 , 119.16 , 79.31 , 41.85 , 30.28 , 27.78 , 24.32 , 19.99 ,
3
+
9
2
2
.49 , 7.66. HRMS(ESI): calcd. for C H NO [M+H] 182.1545, found 182.1533.
1
1
20
,5-Diallyl-2,5-diethylpyrrolidine 1-oxyl (4): compound 3 (0.98 g, 5.41 mmol) dissolved in
mL of diethyl ether was added dropwise to a stirred solution (8.1 mL) of allylmagnesium
bromide (1.0 M) in diethyl ether at −10 °C. After stirring for 1.5 h, a saturated solution of
ammonium chloride (1 mL) and water were (1 mL) were successively added. The mixture
4

ACCEPTED MANUSCRIPT
was filtered through a frit, and the residue was rinsed successively with small amounts of
saturated ammonium chloride solution and diethyl ether. The diethyl ether phase was
separated and washed with brine. The aqueous phase was extracted once with diethyl ether
and the organic layers were combined and the solvent was removed under reduced pressure.
The residue was taken up with 11 mL of methanol containing 0.3 mL of conc. ammonia and
anhydrous copper(II) acetate (26 mg, 0.146 mmol). Oxygen was bubbled through the mixture
with stirring, which was stopped 30 min after the color of the mixture had turned dark blue.
Subsequently, the solvent was removed under reduced pressure. The crude product was
purified by column chromatography (ethyl acetate/n-heptane 1:9, R = 0.38), to give 5 as
f
+
orange liquid. Yield: 0.85 g (3.84 mmol, 71 %). HRMS (ESI): calcd. for C H NO [M]
1
4
24
2
1
22.1851, found 222.1851. IR(ATR): 3076, 3005, 2965, 2938, 2880, 1639, 1462, 1441, 1434,
−1
407, 1380, 1312, 1292, 1218, 996, 965, 912, 797 cm .
2
,5-Bis(carboxymethyl)-2,5-diethylpyrrolidine 1-oxyl (5): compound 4 (0.7 g, 3.15 mmol)
was added to a suspension of potassium permanganate (2.96 g, 18.89) and 18-crown-6 (0.33
g, 1.26 mmol) in 25 mL of benzene and stirred for 48 h at room temperature. Subsequently,
the mixture was filtered and successively with a 5 % aqueous sodium hydroxide solution and
water. The combined aqueous layers were acidified with hydrochloric acid and extracted
several times with chloroform. The combined organic layers were dried over magnesium
sulfate and the solvents were removed under reduced pressure. The crude product was
purified and the cis- and trans-diastereomers partly separated by repeated column
chromatography [chloroform/methyl tert-butyl ether/acetic acid 4:6:0.2, R (cis) = 0.40,
f
R (trans) = 0.55]. Yield: 0.10 g (0.39 mmol, 12 %, cis/trans 1:1). IR(ATR): 3682-2222, 3018,
f
2
1
7
979, 2969, 2935, 2884, 2745, 2666, 2634, 2570, 1694, 1460, 1447, 1408, 1346, 1329, 1312,
274, 1255, 1232, 1203, 1156, 1123, 1092, 1068, 1015, 990, 977, 941, 916, 889, 916, 799,
-1
14 cm . Crystals of cis-5 suitable for single-crystal X-ray analysis were grown from a
solution in chloroform/toluene (1:1) by the slow-evaporation method. trans-5: HRMS(ESI):
+
calcd. for C H NO [M+H] 259.1420, found 259.1410. IR (ATR): 3585-2295, 2970, 2942,
1
2
21
5
2
8
6
883, 1776, 1704, 1463, 1412, 1385, 1309, 1188, 1177, 1120, 955, 931, 905, 878, 846, 825,
-1
01, 736 cm .
-Ethyl-6-nitrooctan-3-one (6): To prepare a solution of sodium methoxide, sodium metal
(0.49 g, 21.26 mmol) was placed in a flask and 15 mL of methanol were added dropwise, so
that the solution boiled gently. Subsequently, 3-nitropentane (3.0 g, 25.61 mmol) was added
with stirring. To the stirred solution, ethyl vinyl ketone (1.88 g, 22.35 mmol) was added
dropwise. After stirring for 4.5 h, 2 mL of glacial acetic acid were added dropwise and the
5

ACCEPTED MANUSCRIPT
solvent was removed under reduced pressure. The residue was partioned between water and
dichloromethane. The organic layer was separated and washed successively with 10 %
aqueous solution of sodium carbonate and brine. After drying over magnesium sulfate, the
solvent was removed under reduced pressure. The compound was purified by distillation
under reduced pressure (0.14 mbar, b.p. 110-113 °C) to give a pale yellow liquid. Yield: 2.37
1
g (11.78 mmol, 53 %). H NMR(500 MHz, CDCl ): δ 2.47-2.39 (q, CH , 2H), 2.38 – 2.14 (m,
3
2
CH , 4H), 2.06-1.83 (m, CH , 4H), 1.06 (t, J = 7.4, 1.0 Hz, CH , 3H), 0.86 (t, J = 7.5, 1.0 Hz,
2
2
3
1
3
CH , 6H). C NMR(101 MHz, CDCl ): δ 209.27, 94.65, 36.32, 36.04, 28.80, 27.70, 8.05,
3
3
+
7
.75. HRMS(ESI): calcd. for C H NO [M+H] 202.1443, found 202.1438.
10 20 3
2
,2,5-Triethyl-3,4-dihydro-2H-pyrrole 1-oxide (7): compound 6 (1.80 g, 8.94 mmol) was
added to a solution ammonium chloride (0.52 g, 9.66 mmol) in 12 mL of water. The mixture
was cooled to ca. −10 °C and zinc powder (2.34 g, 35.8 mmol) was added in small portions,
ensuring that the temperature of the reaction mixture did not exceed 10 °C. Afterwards, the
®
mixture was filtered through Celite 545 and the residue was rinsed with a small amount of
methanol. The filtrate was concentrated in vacuum and extracted with chloroform (5 × 10
mL). The combined organic layers were dried over magnesium sulfate and, subsequently, the
solvent was reduced under reduced pressure. The crude product was purified by column
chromatography (chloroform/methanol 10:0.2, R = 0.13) to give 7 as an orange liquid. Yield:
f
1
1
1
.49 g (8.80 mmol, 98 %). H NMR (400 MHz, CDCl ): δ 2.59-2.44 (m, CH CH , 4H), 2.00-
3
2
2
13
.49 (m, CH CH , 6H), 1.09 (t, J = 7.7, 1.4 Hz, 3H), 0.81 (t, J = 7.4, 1.6 Hz, 6H) ppm. C
3
2
NMR (101 MHz, CDCl ): δ 147.47, 80.00, 77.32, 77.21, 77.01, 76.69, 30.31, 27.86, 24.19,
3
+
1
2
2
9.97, 9.45, 7.67 ppm. HRMS(ESI): calcd. for C H NO [M+H] 170.1545, found 170.1535.
1
0
20
-Allyl-2,5,5-triethylpyrrolidine 1-oxyl (8): compound 7 (1.00 g, 5.91 mmol) dissolved in
mL of diethyl ether was added dropwise to a solution (8.9 mL) of allylmagnesium bromide
(1.0 M) in diethyl ether with stirring at −10 °C. After stirring for 1.5 h, successively 1 mL of a
saturated aqueous solution of ammonium chloride and 1 mL of water were added dropwise.
Subsequently, the reaction mixture was filtered through a frit and the residue was rinsed
successively with small amounts of diethyl ether and a saturated aqueous solution of
ammonium chloride. The diethyl ether phase was separated and washed with brine. After
evaporation of the solvent, the residue was taken up with 11 mL of methanol containing
0
.3 mL of conc. ammonia and anhydrous copper(II) acetate (0.03 g, 0.16 mmol). Oxygen was
bubbled through the mixture with stirring, which was stopped 30 min after the color of the
mixture had turned dark blue. Subsequently, the solvent was removed under reduced pressure.
The crude product was purified by column chromatography (ethyl acetate/n-heptane 1:9, R =
f
6

ACCEPTED MANUSCRIPT
0
.48) to give 7 as an orange liquid. Yield: 0.83 g (3.95 mmol, 67 %). HRMS(ESI): calcd. for
+
C H NO [M] 210.1858, found 210.1848. IR(ATR): 3076, 2965, 2937, 2879, 1639, 1462,
13
24
1
8
2
444, 1406, 1380, 1347, 1331, 1314, 1295, 1219, 1554, 1112, 997, 968, 953, 913, 888, 866,
−1
49, 800, 734 cm .
-(Carboxymethyl)-2,5,5-triethylpyrrolidine 1-oxyl (9): compound 8 (0.6 g, 2.85 mmol)
was added to a suspension of potassium permanganate (2.71 g, 17.12 mmol) and 18-crown-6
0.30 g, 1.14 mmol) in 23 mL of benzene and stirred for 24 h at room temperature.
(
Subsequently, the mixture was filtered and successively with a 5 % aqueous sodium
hydroxide solution and water. The combined aqueous layers were acidified with hydrochloric
acid and extracted once with chloroform. The combined organic layers were dried over
magnesium sulfate and the solvents were removed under reduced pressure. The crude product
was purified by column chromatography (chloroform/acetic acid 10:0.2, R = 0.50) to give 9
f
as a pale yellow solid. Yield: 0.19 g (0.83 mmol, 29 %). HRMS (ESI): calcd. for C H NO :
1
2
22
3
+
[
M] 228.1600, found 228.1600. IR(ATR): 3408-2242, 3089, 3024, 2965, 2939, 2924, 2881,
2
1
740, 2658, 2633, 2546, 1703, 1459, 1455, 1426, 1407, 1376, 1339, 1328, 1290, 1260, 1234,
−1
218, 209, 1160, 1140, 975, 961, 926, 890, 880 cm .
Single-crystal X-ray analysis
The X-ray intensity data were collected on a STOE IPDS 2T diffractometer for cis-5 and on a
STOE IPDS II for 9, using graphite-monochromated Mo-K radiation. The crystal structures
α
were solved with SHELXT [13] and refined with SHELXL-2018/3 [14]. The OLEX2
software was used as a graphical interface [15]. Anisotropic atomic displacement parameters
were introduced for all non-hydrogen atoms. With the exception of the methyl and carboxy
groups in cis-5, hydrogen atoms were placed at geometrically calculated positions and refined
with the appropriate riding model, with U (H) = 1.2 U (C, O) (1.5 for methyl groups). The
iso
eq
O−H distances in cis-5 were restrained to a target value of 0.84(2) Å and with U (H) = 1.2
iso
U (O). Due to the rather low quality of the crystals of 9, the R1 and wR2 values are higher
eq
than in the case of cis-5. Representations of the crystal and molecular structures were drawn
with DIAMOND [16]. Crystal data and refinement details for cis-5 and 9 are listed in Table 1.
CCDC 1876526 (cis-5) and 1876527 (9) contains the supplementary data for this
paper. These data can be obtained free of charge from The Cambridge Crystallographic Data
Centre via http://www.ccdc.cam.ac.uk/structures.
7

ACCEPTED MANUSCRIPT
EPR measurements
1
mM solutions of cis-5, trans-5, 9 and 3-CP in 2.3 M actetate buffer pH 4.5 (Ph. Eur.), 0.08
M phosphate buffer pH 7.4 (Ph. Eur.) and 1.5 M tris-HCl buffer pH 8.8 (Ph. Eur.) were
prepared. EPR spectra were measured in 50 µl capillaries using an X-band EPR spectrometer
at 9.30 – 9.55 GHz (Miniscope MS 400, Magnettech, Berlin, Germany) at 25 °C under
atmospheric oxygen. General settings were as follows: microwave power, 3.162 mW; sweep,
1
1.715 mT; scan time, 60 s; modulation frequency, 100 kHz; modulation amplitude, 100 µT.
EPR spectra were analyzed using the MATLAB toolbox EasySpin. Determination of the
hyperfine coupling constants a and peak-to-peak linewidths (ꢀB ) was carried out by
pp
simulation of each spectrum, using EasySpin‘s function garlic.
Ascorbic acid reduction assay
1
00 µl radical solution (cis-5, trans-5, 9, 3-CP, 4-CT, c = 1 mM) in phosphate buffer pH 7.4
(50 mM, 2 mM EDTA) were mixed with 100 µl ascorbic acid solution (c = 10 mM in
phosphate buffer pH 7.4, 50 mM with 2 mM EDTA). After mixing EPR spectra were
measured every 2 minutes in 50 µl capillaries using an X-band EPR spectrometer at 9.30 –
9
.55 GHz (Miniscope MS 400, Magnettech, Berlin, Germany) at 25 °C under atmospheric
oxygen. General settings were as follows: microwave power, 3.162 mW; sweep, 11.715 mT;
scan time, 60 s; modulation frequency, 100 kHz; modulation amplitude, 30 µT. Second-order
-1 -1
rate constants k (M s ) for the initial rates of reduction were obtained from the decay of the
low-field EPR peak height by using a literature protocol [17].
8

ACCEPTED MANUSCRIPT
Table 1 Crystal data and refinement details for cis-5 and 9.
cis-5
9
Empirical formula
C H NO
C H NO
12 22 3
1
2
20
5
Mr
258.29
228.30
λ
(Å)
0.71073
0.71073
Crystal size (mm)
Crystal system
Space group
T (K)
0.50 × 0.18 × 0.15
0.246 × 0.221 × 0.164
Monoclinic
Monoclinic
P2 /n
P2 /n
1
1
200(2)
213(2)
a (Å)
8.4588(8)
9.5452(6)
16.2549(14)
97.014(7)
1302.61(19)
4
8.9620(7)
9.5965(5)
15.0929(12)
93.166(6)
1296.07(16)
4
b (Å)
c (Å)
β
(°)
3
V (Å )
Z
−3
ρcalc (g cm )
1.317
1.170
−
1
µ
(mm )
0.102
0.083
F(000)
556
500
θ
range (°)
4.243-29.278
14319 / 3503
0.0386
2.516-25.000
8125 / 2288
0.0420
Reflections collected / unique
Rint
Observed reflections [I > 2
σ
(I)] 2166
1691
2
Goodness-of-fit on F
0.858
1.036
Parameters / restraints
171 / 2
145 / 0
R1 [I > 2
σ
(I)]
0.0390
0.0697
wR2 (all data)
0.0940
0.2226
−3
Residuals (eÅ )
0.239 / −0.153
0.698 / −0.286
9

ACCEPTED MANUSCRIPT
Results and discussion
Synthesis
The synthetic route to pyrrolidine nitroxide radicals (Scheme 2) was adapted from Hideg and
Lex [18, 19]. The γ-nitroketones 1 and 6 were prepared by Michael addition of the respective
nitroalkane to an α,β-unsaturated ketone in the presence of a base [9, 10]. 1 and 6 were
reduced to the corresponding hydroxylaminoketones, which cyclized in situ to yield the
nitrones 2 and 7, using zinc powder and ammonium chloride [20]. The nitrones 2 and 7 were
treated with the Grignard reagent allyl magnesium bromide to give the corresponding N-
hydroxy derivatives, which were not isolated but directly oxidized to obtain the N-oxyl
II
radicals 4 and 8, using Cu as catalyst. The latter two steps had to be carried out twice to
obtain 4 via 3. Successful quantitative oxidation of the N-hydroxy intermediates to the N-oxyl
radicals 4 and 8 was confirmed by the absence of O−H stretching bands in the IR spectra,
−1
−1
which would appear in range 3500-3600 cm (e. g. 3580 cm was reported for 2,2,6,6-
−1
tetramethylpiperidine N-hydroxylamine [21]). The IR bands observed at 3076 cm for both 4
and 8 are characteristic of the C−H stretching vibrations of the terminal alkenyl groups.
Compounds 4 and 8 were converted to the carboxymethyl derivatives 5 and 9, respectively,
by oxidative cleavage of the terminal double bonds with potassium permanganate. The broad
−1
bands in the range of 3500-2500 cm in the IR spectra of 5 and 9 are typical of carboxy O−H
stretching vibrations. The cis- and trans- diastereomers of 5 were separated by column
−1
chromatography. The positions of the carbonyl stretching bands for cis-5 (1694 cm ) and
−1
−1
trans-5 (1705 cm ) differ in the IR spectra within spectral resolution (4 cm ),.
1
0

ACCEPTED MANUSCRIPT
Scheme 2 Synthetic route to the pyrrolidine nitroxide radicals 5 and 9.
Structural descriptions of cis-5 and 9
Compounds cis-5 and 9 were structurally characterized by single-crystal X-ray analysis. The
molecular structures are depicted in Figure 1 and selected bond lengths and angles are listed
in Table 2. The molecular geometry parameters of the pyrrolidine N-oxyl moieties are within
expected ranges [22-27]. The conformation of the five-membered rings in cis-5 and 9 is best
described as envelope with respectively C3 and C4 on the flap. The conformations of the
peripheral substituents about the Cpyrrolidine−Cmethylene bonds in cis-5 are antiperiplanar with
respect to the pyrrolidine nitrogen atom, except for the ethyl group attached to C2, which
adopts a synclinal conformation. In 9, the two trans-related 2- and 5-ethyl groups show
synclinal conformations about the Cpyrrolidine−Cmethylene bonds with respect to the nitrogen atom,
whereas the carboxymethyl group and the remaining 5-ethyl group exhibit antiperiplanar
conformations.
1
1

ACCEPTED MANUSCRIPT
cis-5
Figure 1 Molecular structures of cis-5 and 9, respectively showing the (2R,5S) form and the
2R) form in the centrosymmetric crystal structures. Displacement ellipsoids are drawn at the
0 % probabillity level. Hydrogen atoms are represented by small spheres of arbitrary radii.
9
(
5
Table 2 Selected bond lengths (Å) and angles (°) for cis-5 and 9.
cis-5
9
O1−N1
N1−C2
N1−C5
C2−C3
C3−C4
C4−C5
1.2735(14)
1.4854(15)
1.4844(16)
1.5297(18)
1.5281(18)
1.5298(18)
1.273(3)
1.475(3)
1.485(3)
1.528(4)
1.491(4)
1.549(4)
O1−N1−C2
O1−N1−C5
C2−N1−C5
N1−C2−C3
C4−C3−C2
C3−C4−C5
N1−C5−C4
121.86(10)
122.33(10)
115.41(10)
100.80(9)
105.22(10)
106.24(11)
101.32(9)
121.7(2)
123.1(2)
115.2(2)
102.0(2)
107.1(2)
106.2(3)
99.9(2)
1
2

ACCEPTED MANUSCRIPT
In the crystal structures of cis-5 and 9, the molecules are joined through hydrogen
ͦ
bonds of the O−H⋅⋅⋅O type between carboxy goups with centrosymmetric ͌ (8) motifs [28].
ͦ
This results in polymeric zigzag chains, extending in the [10-1] direction, in cis-5, and in
dimers in 9 (Figure 2). The hydrogen bond geometry parameters (Table 3) indicate strong
O−H⋅⋅⋅O hydrogen bonds [29]. The hydrogen bonding patters observed in cis-5 and 9 are
consistent with Etter’s third rule for hydrogen bonding, which states that the best hydrogen
bond donors and the best hydrogen bond acceptors form hydrogen bonds to one another [30].
The nitroxide oxygen atom can be considered a weaker hydrogen bond acceptor than the
hydrogen bond acceptor site of the carboxy group and, thus, the nitroxide groups in cis-5 and
9
only exhibit C−H⋅⋅⋅O contacts that are shorter than the sum of the corresponding van der
Waals radii [31], but do not form classical hydrogen bonds.
Figure 2 Hydrogen-bonded chains in cis-5 (top) and hydrogen-bonded dimers in 9 (bottom).
Hydrogen bonds are represented by dashed lines. Carbon-bound hydrogen atoms are omitted
for clarity. Symmetry codes: (a) −x+1, −y, −z; (b) −x, −y, −z+1.
1
3

ACCEPTED MANUSCRIPT
a
Table 3 Hydrogen bond geometry (Å, °) for cis-5 and 9.
D−H⋅⋅⋅A
d(D−H)
d(H⋅⋅⋅A)
cis-5
d(D⋅⋅⋅A)
<(DHA)
O2−H2⋅⋅⋅O3a
O4−H4⋅⋅⋅O5b
0.858(14)
0.878(15)
1.789(15)
1.748(15)
2.6450(14)
2.6248(15)
175.0(18)
176.3(19)
9
O2−H2⋅⋅⋅O3a
0.83
1.83
2.657(3)
177.9
a
Symmetry codes: (a) −x+1, −y, −z; (b) −x, −y, −z+1.
EPR spectroscopy
The EPR spectra of cis-5, trans-5 and 9, 3-CP were measured in buffer solutions at different
pH levels. Figure 3 shows the EPR spectrum of each radical at pH 7.4. The intensities of the
three lines are almost equal, which means that the anisotropies of hyperfine coupling and g
factor are almost averaged [1]. The spectra are therefore close to the isotropic limit. The
hyperfine coupling constants a(N) (Table 4) are similar for cis- and trans-5 and increase in
the following order: cis-/trans-5 < 9 < 3-CP. The nitrogen hyperfine signals are accompanied
1
3
13
by C satellites. The spectra of 3-CP show two different groups of C satellites; the spectra
of cis-/trans-5 and 9 only one. The lineshape of the spectra is mainly Gaussian. The EPR
signals of cis-5, trans-5 and 9 are 50 to 70 µT broader than the signal of 3-CP.
Nitroxide radicals with a hydrogen acceptor or donor site are sensitive to the pH level
in their environment. Table 5 shows the pH sensitivity ꢀa(N)max and pKa values of the
nitroxide radicals derived from titration curves (see Supporting Information). A pH
dependence of the hyperfine coupling constant a(N) is observed in acidic medium between
pH 2.5 and 6.5. Within physiological pH range, the influence of pH on a(N) is negligible. The
exact pH range depends on the pKa of the radical. pH sensitivity and pKa values vary notably
depending on the molecular structure of the nitroxide radical. The carbon hyperfine coupling
is not influenced by the pH.
1
4

ACCEPTED MANUSCRIPT
Figure 3 EPR spectra (black solid line: experiment, red dotted line: simulation) of 1 mM
solutions of cis-5, trans-5, 9 and 3-CP in 1 mM phosphate buffer (pH 7.4).
Table 4 Hyperfine coupling constants a and peak-to-peak linewidths (ꢀB ) of 3-CP, cis-5,
pp
trans-5 and 9 at different pH levels. The number of carbon atoms is given in parenthesis.
pH
a(N)
mT]
a(C1)/a(C2)
[mT]
∆Bpp
(Gaussian/
Lorentzian)
[
[
mT]
cis-5
trans-5
9
4.5
1.527
1.565
1.559
1.529
1.552
1.552
1.556
1.572
1.572
1.615
1.625
1.624
0.653 (6)
0.653 (6)
0.653 (6)
0.653 (6)
0.653 (6)
0.653 (6)
0.669 (6)
0.669 (6)
0.185/0.015
0.210/0.010
0.210/0.010
0.180/0.015
0.190/0.015
0.190/0.015
0.190/0.015
0.200/0.015
0.195/0.015
0.135/0.012
0.140/0.012
0.140/0.012
7
8
.4
.8
4.5
7
8
.4
.8
4.5
7
8
.4
.8
0.669 (6)
3
-CP
4.5
0.962 (5)/0.530 (2)
0.962 (5)/0.530 (2)
0.962 (5)/0.530 (2)
7
8
.4
.8
1
5

ACCEPTED MANUSCRIPT
Table 5 pH sensitivity ꢀa(N)max and pKa values.
∆
a(N)max [µT]
pKa
4.75
4.34
4.32
3.69
cis-5
trans-5
9
41
30
16
20
3
-CP
The nitrogen hyperfine coupling constant a(N) depends on extrinsic and intrinsic
factors like the polarity and the hydrogen bonding capability of the solvent and the overall
•
+
•−
constitution of the radical. In the resonance structures [R N−O ↔ R N −O ], polar solvents
2
2
such as water stabilize the ionic structure of the nitroxide group, leading to increased spin
density on the nitrogen atom and larger hyperfine coupling constant [1, 32]. Since the
nitroxide radicals studied showed different a(N) values (Table 4) in the same solvent, these
differences were attributed to the different molecular structures. The hyperfine coupling
•
constant a(N) generally increases with an increasing out-of-plane angle of the N−O moiety
[
33]. For five-membered rings, this angle is usually small [34]. In the crystal structures, the
angle between the N1−O1 bond and the C2−N1−C5 plane is 6.01(12)° for cis-5, 0.8(3)° for 9
and 3.24(14)° for (R)-3-CP [23]. Since a(N) increases from 5 to 3-CP, this observation may
•
be ascribed to factors other than the out-of-plane angle of the N−O moiety − bearing in mind
that the out-of-plane angles were determined for the molecular structures in the solid-state.
The a(N) values also depends on the torsional oscillation of the nitroxide, which is influenced
by the substituents in the α positions. It has been shown that substitution in the α-position
hinders the torsional oscillation and thereby lowers a(N) [33]. Therefore, it is reasonable to
assume that the values for a(N) decreasing from 3-CP to cis-/trans-5 result from decreased
conformational flexibility of the pyrrolidine ring.
The spectral linewidth of nitroxide radicals depends on different factors: unresolved
proton hyperfine couplings, anisotropic interactions and interactions with other paramagnetic
species [32, 35]. Exchange broadening by molecular collisions between nitroxides is
negligible under the experimental conditions. Exchange broadening due to interaction with
oxygen occurs, but does not explain the smaller linewidth of 3-CP compared with nitroxide
radicals bearing ethyl/carboxymethyl substituents in 2- and 5-positions. The main
contributions are therefore unresolved proton hyperfine couplings and anisotropic hyperfine
interactions. Hyperconjugation and long-range coupling of ethyl- and carboxymethyl-protons
1
6

ACCEPTED MANUSCRIPT
might cause additional line-broadening in cis-/trans-5 and 9 [35]. The smaller linewidth of 3-
CP accompanies the larger nitrogen hyperfine coupling constant a(N). Torsional oscillations
and a high flexibility of the ring structure increases a(N) and decreases a(H), owing to
dynamic averaging of the proton coupling constants of different conformations [35, 36].
The pH-dependency of the hyperfine coupling constant a(N) in the nitroxide radicals
is due to the presence of the carboxy groups. The pH range in which sensitivity is observed
depends on the pKa value of nitroxide radical and the medium. The pKa values can be
estimated from the titration curves (see Supporting Information) at the pH of the half-
(2)-
equivalence point: a(N) + a(N) /2 [32]. The pKa value estimated for 3-CP is comparable
with experimental values from other literature (pKa = 3.89 [32]; 3.40 [37]). pKa and pKa of
1
2
the 5 isomers cannot be distinguished on the basis of the experimental data. The average pKa
values obtained for cis-/trans-5 and the pKa found for 9 are within the expected range for
carboxylic acids (Table 5). The higher pH sensitivity of cis-/trans-5 as compared with 9 and
3
-CP is most likely a result of an increased electrostatic influence of the second carboxy
group in the former compounds. The observed differences between cis- and trans-5 can be
ascribed to a different extent of stabilization of the ionic form of the nitroxides.
Reductive stability
The reductive stability of cis/trans-5 and 9 was compared with the two commercial nitroxide
radicals 3-CP and 4-CT in an ascorbic acid reduction assay. The rates of reduction were
studied assuming pseudo first-order conditions using a 10-fold excess of ascorbic acid in
phosphate buffer pH 7.4 (50 mM, 2 mM ETDA). The decay curves are shown in Figure 4.
Second order rate constants for the initial reduction rates were obtained from the decay of the
low-field peak height by analysis of the linear part of the decay curves (Table 6). The results
indicate an increased stability of the pyrrolidine nitroxides as compared with the piperidine
nitroxide 4-CT. For the pyrrolidine nitroxides, the stability increases in the order 3-CP < cis-5
≈
trans-5 < 9.
1
7

ACCEPTED MANUSCRIPT
Figure 4 Reduction profiles of the tested nitroxides. Nitroxide 0.5 mM with tenfold excess of
ascorbic acid in phosphate buffer 50 mM pH 7.4 at 298 K. Mean ± standard deviation of three
runs.
Table 6 Second-order rate constants k for initial rates of reduction.
-
1 -1
a
avg k [M s ]
time [s] radical left [%]
cis-5
trans-5
9
5400
5400
5400
840
56.7 ± 2.1
62.7 ± 2.3
78.7 ± 1.7
29.7 ± 0.9
0.0 ± 0.0
0
0
.02074 ± 0.002
± 0.002
± 0.0005
± 0.001
± 0.140
.01737
0.0086
0.0946
3.6253
3
-CP
-CT
4
150
a
residual radical after 90 min
The reductive stability of cyclic nitroxides depends mainly on the ring size. It is well-
known that the stability dramatically increases from piperidine to pyrrolidine derivatives. In
contrast to five-membered ring nitroxides, piperidine derivatives can undergo conformational
changes which lead to better accessibility of the nitroxide moiety [38]. This explains the
difference between 4-CT and the pyrrolidine nitroxides.
1
8

ACCEPTED MANUSCRIPT
The stability is further influenced by steric, electrostatic and field or inductive effects
of the ring substituents [38]. Bulky alkyl groups adjacent to nitroxyl decrease the reduction
+
rate by steric shielding [39, 40]. Ionizable substituents lead to a reduced (e. g. –NH ) or
3
–
increased (e. g. –COO ) reductive stability due to electrostatic attraction or repulsion of the
ascorbic acid anion [38, 41]. These observations, especially the steric shielding, explains the
increased stability of compounds cis/trans-5 and 9 compared with 3-CP. Inductive effects of
the substituents have to be considered, too. Literature reports indicate that electron-
withdrawing groups increase the accessibility of the nitroxide moiety for reductive agents,
whereas electron-donating groups have the opposite effect [6, 8, 38]. Attempts have been
made to correlate the stability of nitroxides with field or inductive constants of the
substituents like the Swain/Lupton F-parameter [8]. The F-values [42] for the substituents of
the nitroxides studied in this work (Table 7) indeed correlate well with the decreased
reductive stability of cis/trans-5 compared with 9. Substituents with positive F-values exhibit
an electron-withdrawing effect leading to reduced stability of the nitroxide group. In
cis/trans-5, the influence of the electron-withdrawing carboxymethyl group is doubled
compared with 9. The reductive stability is thus decreased. The reduction rate of 9 is about
eleven times lower than the reduction rate of 3-CP. The reduction rate of one the of the most
stable nitroxides reported, the tetraethyl derivative of 3-CP, is, however, about 63 times lower
than the reduction rate of 3-CP [7]. This suggests that introduction of carboxymethyl groups
in the α-positions of the nitroxide group is not favorable for inducing additional reductive
stability.
Table 7 Swain/Lupton F-parameters [42].
cis/trans-5
2 × 0.19 + 2 × 0.00
0.19 + 3 × 0.00
4 × 0.01
9
3
4
-CP
-CT
4 × 0.01
1
9

ACCEPTED MANUSCRIPT
Conclusions
The nitroxide radicals cis/trans-5 and 9 exhibit higher reductive stability against ascorbic acid
than the common nitroxide radicals 3-CP and 4-CT, which is ascribed to steric and
electrostatic shielding imparted by the tethered ethyl and carboxymethyl groups in the α-
positions. Compound 9, bearing one carboxymethyl group exhibits a higher reductive stability
than cis/trans-5, bearing two carboxymethyl groups. This can be explained by the
destabilizing electron-withdrawing effect of these functional groups, which compensates the
stabilization through electrostatic shielding. While the reductive stability of the tetraethyl
derivative 3-CP is still higher, 5 and 9 are more versatile spin labels because of the carboxy
groups which impart hydrophilicity - important for biological fluids - and pronounced pH
sensitivity of the EPR signal, and they allow labelling of (bio)polymers.
Acknowledgements
We would like to thank Professor Dariush Hinderberger (Institut für Chemie, Martin-Luther-
Universität Halle-Wittenberg) for providing access to the EPR spectrometer. RWS is grateful
to Dr Richard Goddard for helpful discussions.
References
[1] E. Bordignon (2017). EPR spectroscopy of nitroxide spin probes. In eMagRes, (Eds. R. K. Harris and
R.L. Wasylishen). doi: 10.1002/9780470034590.emrstm1513
[
[
2] J.P. Klare, H.-J. Steinhoff, Spin labeling EPR, Photosynthesis Research 102 (2009) 377-390.
3] D. Bardelang, M. Hardy, O. Ouari, P. Tordo (2012). Spin labels and spin probes. In Encyclopedia of
Radicals in Chemistry, Biology and Materials (Eds. C. Chatgilialoglu and A. Studer). doi:
0.1002/9781119953678.rad081
4] M.M. Haugland, J.E. Lovett, E.A. Anderson, Advances in the synthesis of nitroxide radicals for use
in biomolecule spin labelling, Chemical Society Reviews 47 (2018) 668-680.
5] M.M. Haugland, E.A. Anderson, J.E. Lovett (2017). Tuning the properties of nitroxide spin labels
1
[
[
for use in electron paramagnetic resonance spectroscopy through chemical modification of the
nitroxide framework, Electron Paramagnetic Resonance: Volume 25, The Royal Society of Chemistry,
pp. 1-34. doi: 10.1039/9781782629436-00001
[6] I.A. Kirilyuk, A.A. Bobko, S.V. Semenov, D.A. Komarov, I.G. Irtegova, I.A. Grigor’ev, E.
Bagryanskaya, Effect of sterical shielding on the redox properties of imidazoline and imidazolidine
nitroxides, The Journal of Organic Chemistry 80 (2015) 9118-9125.
2
0

ACCEPTED MANUSCRIPT
[7] J.T. Paletta, M. Pink, B. Foley, S. Rajca, A. Rajca, Synthesis and reduction kinetics of sterically
shielded pyrrolidine nitroxides, Organic Letters 14 (2012) 5322-5325.
8] S. Huang, H. Zhang, J.T. Paletta, S. Rajca, A. Rajca, Reduction kinetics and electrochemistry of
tetracarboxylate nitroxides, Free Radical Research 52(3) (2018) 327-334.
9] J.E. McMurry, J. Melton, Conversion of nitro to carbonyl by ozonolysis of nitronates - 2,5-
heptandione, Organic Syntheses 56 (1977) 36.
10] R. Ballini, P. Marziali, A. Mozzicafreddo, Amberlyst A-27, an efficient heterogeneous catalyst for
[
[
[
the Michael reaction of nitroalkanes with beta-substituted alkene acceptors, The Journal of Organic
Chemistry 61 (1996) 3209-3211.
[
11] K.E. Gilbert, W.T. Borden, Peracid oxidation of aliphatic amines - general synthesis of
nitroalkanes, The Journal of Organic Chemistry 44 (1979) 659-661.
12] S. Miele, P. Nesvadba, A. Studer, 1-tert-Butyl-3,3,5,5-tetraalkyl-2-piperazinon-4-oxyls: Highly
Efficient Nitroxides for Controlled Radical Polymerization, Macromolecules 42 (2009) 2419-2427.
13] G.M. Sheldrick, SHELXT - integrated space-group and crystal-structure determination, Acta
Crystallographica A71 (2015) 3-8.
[
[
[
[
14] G.M. Sheldrick, Crystal structure refinement with SHELXL, Acta Crystallographica C71 (2015) 3-8.
15] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann, OLEX2: a complete
structure solution, refinement and analysis program, Journal of Applied Crystallographica 42 (2009)
39-341.
3
[
[
[
16] K. Brandenburg, DIAMOND, Crystal Impact GbR, Bonn, Germany, 2016.
17] J.F. Corbett, Pseudo first-order kinetics, Journal of Chemical Education 49 (1972) 663.
18] K. Hideg, L. Lex, A versatile new method for the synthesis of various pyrrolidin-1-oxyl fatty acids,
Journal of the Chemical Society, Chemical Communications (1984) 1263-1265.
19] K. Hideg, L. Lex, Synthesis of various new nitroxide free-radical fatty-acids, Journal of the
Chemical Society, Perkin Transactions 1 (1986) 1431-1438.
20] W. Rundel (1968). Methoden zur Herstellung und Umwandlung von Nitronen. In Houben-Weyl:
Methoden der Organischen Chemie (Ed. E.Müller), Georg Thieme Verlag, Stuttgart, pp. 309-448.
21] W.D. Morris, J.M. Mayer, Separating proton and electron transfer effects in three-component
concerted proton-coupled electron transfer reactions, Journal of the American Chemical Society 139
[
[
[
(
2017) 10312-10319.
[
22] J.F.W. Keana, K. Hideg, G.B. Birrell, O.H. Hankovszky, G. Ferguson, M. Parvez, New mono- and
3
difunctionalized 2,2,5,5-tetramethylpyrrolidine and ∆ -pyrroline-1-oxyl nitroxide spin labels,
Canadian Journal of Chemistry 60 (1982) 1439-1447.
[
23] J.B. Wetherington, S.S. Ament, J.W. Moncrief, Structure and absolute configuration of the spin-
label R-(+)-3-carboxy-2,2,5,5-tetramethyl-1-pyrrolidinyloxy, Acta Crystallographica B30 (1974) 568-
73.
24] Y. Uchida, R. Tamura, N. Ikuma, M.A. Kazuyoshi, T.A. Hiroki, S. Shimono, J. Yamauchi,
5
[
Antiferromagnetic interactions arising from a close contact between nitroxyl oxygen and β-methyl
carbon atoms carrying an α-spin in the solid state, Mendeleev Communications 16 (2006) 69-71.
[
25] B. Chion, J. Lajzerowicz, Structures cristallographiques de formes racémique et optiquement
active du tétraméthyl-2,2,5,5 pyrrolidine-3 carboxamide oxyle-1, Acta Crystallographica 31 (1975)
430-1435.
26] I.A. Kirilyuk, Y.F. Polienko, O.A. Krumkacheva, R.K. Strizhakov, Y.V. Gatilov, I.A. Grigor'ev, E.G.
1
[
Bagryanskaya, Synthesis of 2,5-bis(spirocyclohexane)-substituted nitroxides of pyrroline and
pyrrolidine series, including thiol-specific spin label: an analogue of MTSSL with long relaxation time,
The Journal of Organic Chemistry 77 (2012) 8016-8027.
[27] B. Chion, J. Lajzerowicz, A. Collet, J. Jacques, Etude des mélanges d'antipode optiques. XI. Un
exemple de solution solide entre énantiomères: le nitroxyde de la tétraméthyl-2,2,5,5 hydroxy-3
pyrrolidine. Etude cristallographique, Acta Crystallographica B32 (1976) 339-344.
[28] J. Bernstein, R.E. Davis, L. Shimoni, N.L. Chang, Patterns in hydrogen bonding - functionality and
graph set analysis in crystals, Angewandte Chemie International Edition 34 (1995) 1555-1573.
2
1

ACCEPTED MANUSCRIPT
[29] R. Thakuria, B. Sarma, A. Nangia (2017). Hydrogen bonding in molecular crystals. In,
Comprehensive Supramolecular Chemistry II (Ed. J.L. Atwood), Elsevier, Oxford, pp. 25-48.
30] M.C. Etter, Encoding and decoding hydrogen-bond patterns of organic-compounds, Accounts of
Chemical Research 23 (1990) 120-126.
31] A. Bondi, Van der Waals volumes and radii, The Journal of Physical Chemistry 68(3) (1964) 441-
51.
32] G.A.A. Saracino, A. Tedeschi, G. D'Errico, R. Improta, L. Franco, M. Ruzzi, C. Corvaia, V. Barone,
[
[
4
[
Solvent polarity and pH effects on the magnetic properties of ionizable nitroxide radicals:ꢀ a
combined computational and experimental study of 2,2,5,5-tetramethyl-3-carboxypyrrolidine and
2
,2,6,6-tetramethyl-4-carboxypiperidine nitroxides, The Journal of Physical Chemistry A 106(44)
(
2002) 10700-10706.
[33] A. Rockenbauer, M. Györ, H.O. Hankovszky, K. Hideg, E.S.R. of the conformation of 5- and 6-
membered cyclic nitroxide (aminoxyl) radicals, in: M.C.R. Symons (Ed.), Electron Spin Resonance:
Volume 11A, The Royal Society of Chemistry 1988, pp. 145-182.
[34] L.J. Berliner, Spin Labeling: Theory and Applications, Elsevier Science 1976.
[35] J.A. Weil, J.R. Bolton, Electron Paramagnetic Resonance: Elementary Theory and Practical
Applications, Wiley 2007.
36] S.R. Burks, M.A. Makowsky, Z.A. Yaffe, C. Hoggle, P. Tsai, S. Muralidharan, M.K. Bowman, J.P.Y.
[
Kao, G.M. Rosen, The effect of structure on nitroxide EPR spectral linewidth, The Journal of Organic
Chemistry 75 (2010) 4737-4741.
[
37] R.N. Schwartz, M. Peric, S.A. Smith, B.L. Bales, Simple test of the effect of an electric field on the
N-14-hyperfine coupling constant in nitroxide spin probes, J. Phys. Chem. B 101 (1997) 8735-8739.
38] S. Morris, G. Sosnovsky, B. Hui, C.O. Huber, N.U.M. Rao, H.M. Swartz, Chemical and
electrochemical reduction rates of cyclic nitroxides (nitroxyls), Journal of Pharmaceutical Sciences 80
[
(
1991) 149-152.
[39] A.P. Jagtap, I. Krstic, N.C. Kunjir, R. Hänsel, T.F. Prisner, S.T. Sigurdsson, Sterically shielded spin
labels for in-cell EPR spectroscopy: Analysis of stability in reducing environment, Free Radical
Research 49 (2015) 78-85.
[
3
40] S.A. Dobrynin, Y.I. Glazachev, Y.V. Gatilov, E.I. Chernyak, G.E. Salnikov, I.A. Kirilyuk, Synthesis of
,4-Bis(hydroxymethyl)-2,2,5,5-tetraethylpyrrolidin-1-oxyl via 1,3-dipolar cycloaddition of
azomethine ylide to activated alkene, The Journal of Organic Chemistry 83 (2018) 5392-5397.
41] J.F.W. Keana, S. Pou, G.M. Rosen, Nitroxides as potential contrast enhancing agents for MRI
[
application: Influence of structure on the rate of reduction by rat hepatocytes, whole liver
homogenate, subcellular fractions, and ascorbate, Magnetic Resonance in Medicine 5 (1987) 525-
5
[
36.
42] C. Hansch, A. Leo, R.W. Taft, A survey of Hammett substituent constants and resonance and field
parameters, Chemical Reviews 91 (1991) 165-195.
2
2

ACCEPTED MANUSCRIPT
•
•
•
Two new pyrrolidine nitroxide radicals were synthesized and structurally
characterized.
Carboxymethyl and ethyl groups in the α-positions of the nitroxide group improve the
stability of the radicals.
The new radicals exhibit higher reductive stability than 3-carboxy-PROXYL and 4-
carboxy-TEMPO.