2H,15N-substituted nitroxides as sensitive probes for electron paramagnetic resonance imaging

Article abstract of DOI:10.1021/jo1011619

Electron paramagnetic resonance imaging (EPRI) using nitroxides is an emergent imaging method for studying in vivo physiology, including O₂ distribution in various tissues. Such imaging capabilities would allow O ₂ mapping in tumors

Full text of DOI:10.1021/jo1011619

pubs.acs.org/joc
2H,15N-Substituted Nitroxides as Sensitive Probes for Electron
Paramagnetic Resonance Imaging
Scott R. Burks,†,‡ Justin Bakhshai,§ Mallory A. Makowsky,§ Sukumaran Muralidharan,†
Pei Tsai,†,§ Gerald M. Rosen,*,†,§ and Joseph P. Y. Kao†,‡
†Center for Biomedical Engineering and Technology, and Center for EPR Imaging In Vivo Physiology,
University of Maryland, Baltimore, Maryland 21201, ‡Department of Physiology, University of Maryland
School of Medicine, Baltimore, Maryland 21201, and §Department of Pharmaceutical Sciences, University of
Maryland School of Pharmacy, Baltimore, Maryland 21201
grosen@umaryland.edu
Received June 14, 2010
Electron paramagnetic resonance imaging (EPRI) using nitroxides is an emergent imaging method for
studying in vivo physiology, including O2 distribution in various tissues. Such imaging capabilities
would allow O2 mapping in tumors and in different brain regions following hypoxia or drug abuse. We
have recently demonstrated that the anion of 3-carboxy-2,2,5,5-tetramethyl-1-pyrrolidinyloxyl (2) can
be entrapped in brain tissue to quantitate O2 concentration in vivo. To increase the sensitivity of O2
measurement by EPR imaging, we synthesized 3-carboxy-2,2,5,5-tetra(2H3)methyl-1-(3,4,4-2H3,1-15N)-
pyrrolidinyloxyl (7). EPR spectroscopic measurements demonstrate that this fully isotopically substi-
tuted nitroxide markedly improves signal-to-noise ratio and, therefore, the sensitivity of EPR imaging.
The new isotopically substituted nitroxide shows increased sensitivity to changes in O2 concentration,
which will enable more accurate O2 measurement in tissues using EPRI.
Introduction
nitroxides.3 Using these spin probes, one can make reliable,
minimally invasive measurements of O2 concentration in
vivo. Besides the significance of O2 generally in brain metab-
olism, the importance of O2 measurement in brain stems
from the suggestion that oxygen-centered free radicals are
responsible for methamphetamine-induced neurotoxicity,4
leading to altered O2 levels in the affected regions of the
brain. Therefore, it is clinically important to map the regions
of decreased O2. A major obstacle to using EPR imaging to
quantify O2 in brain tissue is the difficulty of transporting
O2-sensitive imaging probes across the blood-brain barrier.
As EPRI probes, nitroxides have many desirable qualities,
including ease of preparation, chemical flexibility, and
high stability at physiologic pH and temperature. We have
Electron paramagnetic resonance imaging (EPRI) is an
emergent magnetic resonance imaging modality that can be
used to study a wide range of physiology.1 For example,
molecular oxygen, being paramagnetic, broadens the elec-
tron paramagnetic resonance (EPR) spectral lines of other
paramagnetic species, including trityl radicals2 and
(1) Halpern, H. J.; Spencer, D. P.; van Polen, J.; Bowman, M. K.;
Masssoth, R. J.; Nelson, A. C.; Dowey, E. M.; Teicher, B. A. Rev. Sci.
Instrum. 1989, 60, 1040.
(2) Elas, M.; Williams, B. B.; Parasca, A.; Mailer, C.; Pelizzari, C. A.;
Lewis, M. A.; River, J. N.; Karczmar, G. S.; Barth, E. D.; Halpern, H. J.
Magn. Reson. Med. 2003, 49, 682.
(3) (a) Kutala, V. K.; Parinandi, N. L.; Pandian, R. P.; Kuppusamy, P.
Antioxid. Redox Signal. 2004, 6, 597. (b) Shen, J.; Khan, N.; Lewis, L. D.;
Armand, R.; Grinberg, O.; Demidenko, E.; Swartz, H. Biophys. J. 2003, 84,
1291. (c) Khan, N.; Shen., J.; Chang, T. Y.; Chang, C. C.; Fung, P. C.;
Grinberg, O.; Demidenko, E.; Swartz, H. Biochemistry 2003, 42, 23. (d) Liu,
S.; Timmins, G. S.; Shi, H.; Gasparovic, C. M.; Liu, K. J. NMR Biomed.
2004, 17, 327.
(4) (a) Cadet, J. L.; Brannock, C. Neurochem. Int. 1998, 32, 117. (b) Imam,
S. Z.; el-Yazal, J.; Newport, G. D.; Itzhak, Y.; Cadet, J. L.; Slikker, W., Jr.;
Ali, S. F. Ann. N.Y. Acad. Sci. 2001, 939, 366. (c) Riddle, E. L.; Fleckenstein,
A. E.; Hanson, G. R. AAPSJ 2006, 8, E413. (d) Quinton, M. S.; Yamamoto,
B. K. AAPSJ 2006, 8, E337.
DOI: 10.1021/jo1011619
r
Published on Web 09/09/2010
J. Org. Chem. 2010, 75, 6463–6467 6463
2010 American Chemical Society

Burks et al.
JOCArticle
SCHEME 1
Reagents and conditions: (a) 15ND4Cl, MgO; (b) D2O2, D2O; (c) (i) D2, Pd/C, CH3OD, (ii) DCl, D2O; (d) (i) Br2, CHCl3, (ii) NaNO2, D2O;
(e) (i) KOD, (ii) DCl; (f ) CH3CO2CH2Br, K2CO3, CH3CN.
recently synthesized nitroxide labile esters5 (e.g., 1) that can
cross the blood-brain barrier to enter brain tissue, where
they are converted by enzymatic hydrolysis to 3-carboxy-
2,2,5,5-tetramethyl-1-pyrrolidinyloxyl (2). This compound,
being anionic at physiologic pH, becomes entrapped in the
brain and can report O2 concentration through EPRI. Based
on this success,5 we describe the synthesis of 3-carboxy-
2,2,5,5-tetra(2H3)methyl-1-(3,4,4-2H3,1-15N)pyrrolidinyloxyl
methyl(3,3,5,5-2H4,1-15N)piperidine (3), prior to subsequent
reactions, including ring contraction. We originally considered
using the procedure of Rozantsev,6 which called for the con-
densation of acetone with ammonia. The reported low yield
(19%) for this reaction, however, compelled us to seek an
alternative synthetic pathway. A review of the literature
uncovered the procedure of Lin et al.,7a in which heating
perdeuteroacetone and 14ND4Cl with MgO in a sealed vessel
gave the desired piperidine with 56% yield. 15ND4Cl is not
commercially available. Therefore, it was prepared by isotope
exchange: 15NH4Cl was dissolved in D2O and lyophilized; the
process was repeated four times.
2
(7), wherein isotopic substitution with 15N and H are ex-
pected to improve detection limits and signal-to-noise ratio
(SNR) for EPRI. The rationale for isotopic substitution is
straightforward. In an isotopically unmodified nitroxide,
hyperfine interaction with the spin-1 14N nucleus splits the
electron resonance into three spectral lines, each with roughly
one-third the total signal intensity. Because EPRI measures
the amplitude of only one spectral line, the signal contained in
the other two lines is wasted. The spin-1/2 15N nucleus splits
the nitroxide resonance into only two lines, each of which
contains 50% more signal than any line in the 14N EPR spec-
trum. Additionally, in isotopically unmodified nitroxides,
hyperfine interactions with 1H nuclei broaden the EPR spec-
tral lines, with corresponding reduction of peak amplitudes.
Complete isotopic substitution with 2H, which has a weaker
nuclear magnetic moment, should significantly reduce line
broadening and thereby increase spectral peak amplitude and
SNR. Lastly, O2 concentrations are estimated from the width
of the EPR spectral lines. We demonstrate that in comparison
with the unmodified nitroxide 2, the isotopically substituted
nitroxide 7 has a relative line width dependence on O2 con-
centration that is 2.6-fold steeper, which makes the latter a
more sensitive probe for O2 measurement in vivo.
In our hands, heating perdeuteroacetone and 15ND4Cl
with MgO in a sealed vessel yielded 3in ∼40%yield(Scheme 1).
Whether 3 was purified by distillation made little difference
in the yield of the next reaction; therefore, crude 3 was used
directly in the next step to ensure efficient use of isotopically
substituted material. The oxidation of 3 with D2O2 in D2O
afforded nitroxide 4 in reasonable yield.
Although it has been reported8 that addition of HCl to
isotopically unmodified nitroxide 4 followed by bromination
gave the corresponding 6 in good yield via the intermediate 5,
Marc and Pecar9 found that reducing 4 by catalytic hydro-
genation prior to bromination gave superior results. For the
isotopically unmodified 4, this procedure should give a
nearly quantitative yield of 5. In the case of 2H-substituted
nitroxide 4, however, the procedure could lead to hydrogen-
deuterium exchange, owing to keto-enol tautomerization
(Scheme 2). To eliminate this possibility, we initially con-
sidered using perdeuterated ammonium formate as the elec-
tron source for reduction10 but ultimately decided to reduce
piperidinyloxyl 4 to the hydroxylamine 5 with D2 over Pd/C
in CH3OD.
Subsequent bromination of 5 at a single position R to
the carbonyl, achieved by dropwise addition of Br2 in
(6) Rozantsev, E. G. Free Nitroxyl Radicals; Plenum Press: New York,
1970; p 203.
Results and Discussion
(7) (a) Lin, Y. J.; Teicher, B. A.; Halpern, H. J. J. Labelled Comp. Rad.
1990, 28, 621. (b) Yamada, K.-I.; Kinoshita, Y.; Yamasaki, T.; Sadasue, H.;
Mito, F.; Nagai, M.; Matsumoto, S.; Aso, M.; Suemune, H.; Sakai, K.;
Utsumi, H. Arch. Pharm. Chem. Life Sci. 2008, 341, 548.
(8) Sosnovsky., G.; Cai, Z. W. J. Org. Chem. 1995, 60, 3414.
(9) Marc, G.; Pecar, S. Synth. Commun. 1995, 25, 1015.
In preparing isotopic-substituted nitroxide 7, 15N must be
introduced in the first step, forming 4-oxo-2,2,6,6-tetra(2H3)-
(5) (a) Rosen, G. M.; Burks, S. R.; Kohr, M. J.; Kao, J. P. Y. Org. Biomol.
Chem. 2005, 3, 645. (b) Burks, S. R.; Ni, J.; Murakidharan, S.; Coop, A.;
Kao, J. P. Y.; Rosen, G. M. Bioconjugate Chem. 2008, 19, 2068.
ꢀ
(10) (a) Hideg, K.; Hankovszky, H. O.; Halasz, H. A. J. Chem. Soc.,
Perkin Trans. 1 1988, 2905. (b) Derdau, V. Tetrahedron Lett. 2004, 45, 8889.
6464 J. Org. Chem. Vol. 75, No. 19, 2010

Burks et al.
JOCArticle
SCHEME 2
chloroform over 45 min, followed by NaNO2 oxidation,
afforded the bromopiperidinyloxyl 6,8 with no evidence of
dibromination.11 Importantly, piperidinyloxyl 4, which was
recovered by chromatography, was recycled through the
reaction to bring the total yield of 6 to 64%. Favorskii
rearrangement of 6 with KOD8 gave the desired carboxylic
acid 7.
Figure 1 shows EPR spectra for the Kþ salts of nitroxides
2 and 7, each at 100 μM in air-equilibrated H2O ([O2] =
0.25 mM). Although the samples are at the same concentra-
tion, nitroxide 7 exhibits significantly larger and narrower
EPR spectral peaks. Using relatively low magnetic fields
permits EPR imaging with low-frequency electromagnetic
radiation, which penetrates tissue well but reduces SNR.
Nitroxide 7 remedies this deficit with much narrower, and
thus larger, spectral peaks. Using nitroxide 7 improves the
limit of detection and enables imaging with higher contrast.
Figure 2 shows the dependence of the EPR line width on
O2 concentration for nitroxides 7 and 2. As O2 concentration
varies, the relative change in line width for nitroxide 7 is
much larger than for nitroxide 2. Linear least-squares fits of
the data show that the line-broadening effect of O2 is 2.74-
fold greater for nitroxide 7 than for nitroxide 2. Thus,
nitroxide 7 is more O2-sensitive and capable of resolving
smaller changes in O2 than nitroxide 2. In EPR imaging,
overmodulation is often used to improve SNR, at the cost of
spectral line broadening. The threshold modulation ampli-
tude at which significant line broadening occurs is lower for 7
than for 2. Therefore, overmodulation could reduce the
difference in the line-broadening effect of O2 on the two
nitroxides. The spectra in Figure 2 were acquired at a
modulation amplitude of 0.125 G, at which line broadening
was negligible for both nitroxides (Figure S1, Supporting
Information). We measured the dependence of the line-
widths of 7 and 2 on O2 at an increased modulation ampli-
tude of 0.5 G (Figure S2, Supporting Information). At this
larger modulation amplitude, the difference in the line-
broadening effect of O2 is reduced from 2.74-fold to 2.60-
fold. Thus, even under typical conditions of overmodula-
tion, nitroxide 7 is still far superior to nitroxide 2 as an O2
sensor.
FIGURE 1. EPR spectra of nitroxides 2 and 7 (Kþ salt, each at
100 μM in H2O equilibrated with air). Both spectra were acquired
with identical spectrometer settings (see the Experimental Section)
and are represented on the same intensity scale. The hyperfine
splittings and linewidths are indicated.
FIGURE 2. Relative linewidths of nitroxides 2 and 7 at different O2
concentrations in water. For each nitroxide, the linewidths were
normalized to the value measured at 0.003 mM O2 (in N2-sparged
water). Solid lines are least-squares fits of the data; the slope of each
line is indicated on the graph.
labile ester 8, like its isotopically unmodified counterpart, is
expected to cross the blood-brain barrier. When introduced
into live animals, ester 8 will enable us to determine the actual
extent to which isotopic substitution improves O2 imaging in
vivo. These studies are underway.
Isotopic substitution of 15N and 2H in the pyrrolidinyloxyl
ring is a significant advancement in using nitroxides as
oxygen-sensitive probes in EPR imaging. We have shown
previously that nitroxide 1, the isotopically unmodified
analogue of labile ester 8, is a pro-imaging agent that can
cross the blood-brain barrier and be converted to nitroxide
2 in brain tissue.12 Therefore, the isotopically substituted
Experimental Section
General Materials and Methods. Reagents and solvents from
commercial vendors were used without further purification.
Silica gel (230-400 mesh) was used for column chromatog-
raphy.
4-Oxo-2,2,6,6-tetra(2H3)methyl-(3,3,5,5-2H4,1-15N)piperidine (3).
This compound was prepared following the general procedure
of Lin et al.7a with minor modifications. To generate 15ND4Cl,
15NH4Cl (10 g) was dissolved in D2O (99.9%; 15 mL); the
solution was evaporated to dryness under vacuum. This proce-
dure was repeated a total of four times.
(11) Alcock, N. W.; Golding, B. T.; Ioannou, P.; Sawyer, J. F. Tetra-
hedron 1977, 33, 2969.
(12) (a) Shen, J.; Liu, S.; Miyake, M.; Liu, W.; Pritchard, A.; Kao, J. P.;
Rosen, G. M.; Tong, Y.; Liu, K. J. Magn. Res. Med. 2006, 55, 1433. (b)
Miyake, M.; Shen, J.; Liu, S.; Shi, H.; Liu, W.; Yuan, Z.; Pritchard, A.; Kao,
J. P.; Liu, K. J.; Rosen, G. M. J. Pharmacol. Exp. Ther. 2006, 318, 1187. (c)
Shen, J.; Sood, R.; Weaver, J.; Timmins, G. S.; Schnell, A.; Miyake, M.; Kao,
J. P.; Rosen, G. M.; Liu, K. J. Cereb. Blood Flow Metabol. 2009, 29, 1695.
In a glovebox under positive internal N2 pressure, 15ND4Cl
(3.5 g, 60 mmol) was added to a 250-mL round-bottom flask
J. Org. Chem. Vol. 75, No. 19, 2010 6465

Burks et al.
JOCArticle
containing oven-dried anhydrous Na2CO3 (3.18 g, 30 mmol)
and MgO (3.0 g, 75 mmol). Thereafter, perdeuteroacetone (12.5
mL, 150 mmol; 99.9%) was introduced into the flask by canula
under N2 pressure. While still under N2 atmosphere, the flask
was sealed with a rubber septum. The reaction mixture was kept
for 3 days at 50 °C in an oil bath and then allowed to cool.
Perdeuterocetone (20 mL) was added to the flask, and the
resulting mixture was filtered. The filter cake was crushed into
a fine powder, washed with dry ether and perdeuteroacetone
(1:1 mixture, 20 mL), and again filtered; this procedure was
repeated three more times. The combined filtrates were concen-
trated on a rotary evaporator to give a red liquid (5.5 g), a
portion of which was distilled to yield a yellow liquid (bp 60-
64 °C at 12 mmHg),7 which solidified upon cooling. In trial runs,
we found that distillation of 3 did not significantly affect the
yield of 4; therefore, crude 3 was used for the next reaction
without further purification.
with D2 over 5% Pd/C (25 mg) in CH3OD (30 mL) followed by
bromination and then oxidation with NaNO2 led to a 64%
overall yield of piperidinyloxyl 6 (2.0 g). Recrystallization from
hexane yielded light orange crystals, whose purity was con-
firmed by HPLC (Figure S3, Supporting Information): mp=
81-82 °C; IR (CHCl3) 1730 cm-1 (CdO); HRMS (ESI) calcd
1
for C9 H2H1515NO279Br [MþH]þ 265.12773, found 265.11910;
calcd for C9 H2H1515NO281Br [M þ H]þ 267.12558, found
1
267.12953.
3-(2H)Carboxy-2,2,5,5-tetra(2H3)methyl-1-(3,4,4-2H3,1-15N)-
pyrrolidinyloxyl (7). The general procedure of Sosnovsky and
Cai8 was used, with minor modifications. KOD (1 M in D2O,
5 mL) was added to 3-bromo-4-oxo-2,2,6,6-tetra(2H3)methyl-
1-(3,5,5-2H3,1-15N)piperidinyloxyl (6) (0.52 g, 2 mmol). With
stirring over the next 2 h, nitroxide 6 dissolved completely. The
alkaline solution was extracted with Et2O (3 ꢀ 20 mL), cooled in
an ice bath, and adjusted to pH 3 with dilute DCl (10% in D2O).
This acidic solution was extracted with Et2O (3 ꢀ 20 mL); the
combined extract was dried over anhydrous MgSO4, filtered,
and evaporated under reduced pressure to yield 3-carboxy-
2,2,5,5-tetra(2H3)methyl-1-(3,4,4-2H3,1-15N)pyrrolidinyloxyl (7)
as a yellow solid (0.25 g, 61%). Recrystallization from CHCl3/
hexane gave a yellow powder: mp =190 - 194 °C dec;13 IR
(CHCl3) 3500 cm-1 (broad peak, OH), 1711 cm-1 (CdO).
3-Acetoxymethoxycarbonyl-2,2,5,5-tetra(2H3)methyl-1-(3,4,4-
2H3,1- 15N)pyrrolidinyloxyl (8). The synthesis followed our reported
procedure.14 Bromomethyl acetate (0.044 g, 0.03 mL, 0.33 mmol)
was added to a solution of 3-carboxy-2,2,5,5-tetra(2H3)methyl-
1-(3,4,4-2H3,1-15N)pyrrolidinyloxyl (7) (0.050 g, 0.25 mmol) and
K2CO3 (0.070 g, 0.50 mmol) in acetonitrile (dried over CaH2,
10 mL); the mixture was stirred overnight at room temperature.
Thereafter, this mixture was filtered through Celite, and the filtrate
was evaporated to dryness. The oily residue was chromatographed
(hexane/ethyl acetate, 5:1) to yield a thick oil, which was crystallized
from hexane to yield 3-acetoxymethoxycarbonyl-2,2,5,5-tetra(2H3)-
methyl-1-(3,4,4-2H3,1-15N)pyrroldinyloxyl (8) as a yellow solid
(0.050 g; 74%): mp=77-78 °C; IR (CHCl3) 1763 cm-1 (CdO);
4-Oxo-2,2,6,6-tetra(2H3)methyl-1-(3,3,5,5-2H4,1-15N)piperi-
dinyloxyl (4). To a solution of crude 4-oxo-2,2,6,6-tetra(2H3)-
methyl-(1,3,3,5,5-2H5,1-15N)piperidine (3) (5.5 g, 34 mmol) dis-
solved in D2O (60 mL) were added oven-dried Na4EDTA (0.55 g,
1.5 mmol) and oven-dried Na2WO4 (0.55 g, 1.7 mmol). Upon
dissolution of the salts, D2O2 (30% in D2O, 6 mL) was added, and
the reaction was allowed to proceed in the dark for 10 days. The
reaction mixture was filtered and extracted with ether (3 ꢀ 50 mL).
The ether extract was first washed with cold dilute DCl (10% in
D2O, 2 ꢀ 20 mL) and then with saturated Na2CO3 in D2O(10 mL).
Thereafter, the solution was dried over anhydrous MgSO4, filtered,
and reduced to dryness on a rotary evaporator. This residue was
chromatographed (hexane/Et2O, 2:1) to yield 4-oxo-2,2,6,6-tetra-
(2H3)methyl-1-(3,3,5,5-2H4,1-15N)piperidinyloxyl 4 as a red oil,
which solidified in the cold (2.8 g, 51%): IR (CHCl3) 1720 cm-1
2
2
(CdO). Anal. Calcd for C9 H1615NO2: C, 57.69; H, 8.61; 15N,
7.48. Found: C, 57.57; 2H, 8.58; 15N, 7.40.
4-Oxo-2,2,6,6-tetra(2H3)methyl-1-(2H)hydroxy-(3,3,5,5-2H4,
1-15N)piperidine (2H)Hydrochloride (5). The general procedure
of Marc and Pecar9 was used with minor modifications. 4-Oxo-
2,2,6,6-tetra(2H3)methyl-1-(3,3,5,5-2H4,1-15N)piperidinyloxyl (4)
(2.8 g, 15 mmol) was dissolved in CH3OD (30 mL), and 5%
Pd/C (50 mg) was added. Deuterium gas (99%) was gently
bubbled into the reaction mixture for several min and the flask
was sealed. The flask was periodically recharged with D2 over
the next several hours. After being stirred overnight, the reaction
mixture was filtered through Celite. The colorless filtrate was
acidified with 4 M DCl (in D2O, 2.5 mL) and reduced to dryness
on a rotary evaporator. The residue was washed with dry ether
(2 ꢀ 20 mL) to remove any remaining nitroxide and dried, in
vacuo, to yield 4-oxo-2,2,6,6-tetra(2H3)methyl-1-(2H)hydroxyl-
(3,3,5,5-2H4,1-15N)piperidine (2H)hydrochloride (5) as a white
solid (2.7 g, 80%). Compound 5 is very hygroscopic and was
therefore used immediately in the monobromination reaction to
yield 6, as described below.
HRMS (ESI) calcd for C121H6 H1515NO5 [M þ H]þ 275.23316,
2
2
found 275.23085. Anal. Calcd for C12 H151H515NO5: C, 52.52;
1
1
2H þ H, 7.35; 15N, 5.10. Found: C, 52.76; 2H þ H, 7.42; 15N, 5.12.
EPR Spectroscopy. EPR spectra were recorded on an X-band
spectrometer at the following settings: microwave power,
20 mW; microwave frequency, 9.55 GHz; field set, 3335 G for
14N or 3324 G for 15N; modulation frequency, 1 kHz; modulation
amplitude, 0.125 G; field sweep, 4 G at 13.3 G min-1. These
settings encompassed the central spectral peak of the 14N spec-
trum, or the first spectral peak of the 15N spectrum. For recording
the complete spectrum of either nitroxide, the field set was 3335 G,
and field sweep was 50 G at 13.3 G min-1. Digital acquisition of
EPR spectra was through EWWIN software.
EPR Spectral Linewidth Measurements. To assess the effect of
O2 on the EPR linewidths of the nitroxides, deionized H2O (18.3
3-Bromo-4-oxo-2,2,6,6-tetra(2H3)methyl-1-(3,5,5-2H3,1-15N)-
piperidinyloxyl (6). The procedure of Sosnovsky and Cai8 was
used with minor modifications. Br2 (2.14 g, 11.9 mmol) in CHCl3
(10 mL) was added dropwise over 45 min at room temperature
to a stirred solution of 4-oxo-2,2,6,6-tetra(2H3)methyl-1-(2H)-
hydroxyl-(3,3,5,5-2H4,1-15N)piperidine (2H)hydrochloride (5)
(2.7 g, 11.9 mmol) in CHCl3 (25 mL); thereafter, the reaction
mixture was stirred for 2.5 h. Thereafter, a solution of NaNO2
(1.85 g, 27 mmol) in D2O (10 mL) was added dropwise over
10 min to the vigorously stirred reaction mixture; stirring was
continued for another 15 min. The organic phase was washed
with D2O, dried over anhydrous MgSO4, filtered, and evapo-
rated to dryness under reduced pressure. Chromatography
(hexane/Et2O, 2:1) yielded two fractions: (1) 3-bromo-4-oxo-
2,2,6,6-tetra(2H3)methyl-1-(3,5,5-2H3,1-15N)piperidinyloxyl (6)
and (2) nitroxide 4. Rereducing the recovered piperidinyloxyl 4
MΩ cm resistivity) was sparged with N2, equilibrated with air,
3
or sparged with O2 at 24 °C for 30 min, to yield solutions
containing O2 at 0.003, 0.25, and 1.25 mM (O2-saturated15),
respectively. Submillimolar O2 concentrations were determined
using a dissolved O2 meter. Stock solutions (10 mM) of the Kþ
salt of nitroxide 2 or 7 were diluted 500-fold into the gas-
equilibrated H2O samples to a final nitroxide concentration of
20 μM. Each solution was transferred into a flat quartz EPR cell
previously purged with the appropriate gas. The quartz cell was
sealed, and immediately positioned in the EPR spectrometer.
Duplicate spectroscopic measurements on the samples were
(13) Rozantsev, E. G. Free Nitroxyl Radicals; Plenum Press: New York,
1970; p 206.
(14) Kao, J. P. Y.; Rosen, G. M. Org. Biomol. Chem. 2004, 2, 99.
(15) Steen, H. Limnol. Oceanogr. 1958, 3, 423.
6466 J. Org. Chem. Vol. 75, No. 19, 2010

Burks et al.
JOCArticle
performed in random order. Reported linewidths are the peak-
to-peak width of the central spectral line of 2 and the first
spectral line of 7.
Professor Slavko Pecar for helpful suggestions on the synthe-
sis described in this paper.
Acknowledgment. This research was supported in part by
grants from the National Institutes of Health, EB2034 and
DA023473 (G.M.R.) and GM056481 (J.P.Y.K.). We thank
Supporting Information Available: Supplemental methods
and figures. This material is available free of charge via the
Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 75, No. 19, 2010 6467