Spin Trapping with 5-Methyl-5-phenylpyrroline N-Oxide. A Replacement for 5,5-Dimethylpyrroline N-Oxide

Full text of DOI:10.1021/jo961476x

1176
J . Org. Chem. 1997, 62, 1176-1178
toxic compound detrimental to the longevity of cells,9 and
Sp in Tr a p p in g w ith
(c) a “dormant” compound (perhaps a hydroxylamine)
which exposes itself to EPR spectroscopy only after free
radicals are formed.10 For best results pure DMPO must
be kept under N2 in the dark and at dry ice temperature
even during shipping, and used immediately after warm-
ing to room temperature.
5-Meth yl-5-p h en ylp yr r olin e N-Oxid e. A
Rep la cem en t for 5,5-Dim eth ylp yr r olin e
N-Oxid e
Nagaraju Sankuratri, Edward G. J anzen,*,†
Melinda S. West, and J . Lee Poyer
4
5
3
2
4
5
3
2
Free Radical Biology and Aging Research Program,
Oklahoma Medical Research Foundation,
Oklahoma City, Oklahoma 73104
Me
Me
C6H5
Me
1
N
1
N
O
O
DMPO
5,5-dimethylpyrroline
MPPO
5-methyl-5-phenylpyrroline N-oxide
Received August 1, 1996
N-oxide
5,5-Dimethylpyrroline N-oxide (DMPO)1 has become
the most commonly used spin trap in biological systems
to date. This is because the hydroxyl radical spin adduct
gives a simple and distinctive EPR spectrum2 consisting
of four peaks with relative intensity 1:2:2:1, and the
superoxide/hydroperoxyl radical spin adduct gives a
different but also distinctive multiline EPR spectrum2
composed of 6 major peaks plus additional partially
resolved doublets. These unique signatures are readily
recognizable in the presence of each other and thus
provide a convenient approach to the study of systems
where both the hydroxyl and superoxide/hydroperoxyl
radicals are produced simultaneously. Moreover, since
EPR spectroscopy is a very sensitive technique the spin
trapping method is unequaled in its capability to detect
low concentrations of these two species when in situ
experiments can be arranged. It has been shown that
DMPO is better than cytochrome c in detecting super-
oxide,3,4 and alternate methods for detecting hydroxyl
radical by salicylate or phenylalanine hydroxylation5 also
bring along with them problems associated with meta-
bolic production of similar derivatives. Cytochrome c and
salicylate are the “benchmark” methods for detecting
superoxide and hydroxyl radicals, respectively.
However, a major concern in the use of DMPO is the
presence of artifacts some of which appear to be due to
compounds also produced in the synthesis of DMPO.6
Small molecular weight nitrones are very reactive mol-
ecules and self-reaction is usually prevented by attaching
substituents with sterically hindering groups or conjuga-
tion.7,8 DMPO is relatively unprotected and contains no
aryl groups. The melting point is low, and the colorless
crystalline state is difficult to achieve. Most preparations
result in a liquid which acquires a yellow color on
standing. This product often contains (a) an EPR signal
which confounds the analysis of an EPR spectrum,6 (b) a
We have synthesized a DMPO replacement which gives
similar hydroxyl radical and superoxide/hydroperoxyl
radical spin adduct signatures as DMPO but which can
be readily prepared in pure form and appears to be stable
under normal shelf-life conditions indefinitely. Here we
describe the first synthesis of 5-methyl-5-phenylpyrroline
N-oxide (MPPO). Derivatives of DMPO for the purpose
of spin trapping have been synthesized before, but all of
them suffer from serious drawbacks ranging from com-
plex EPR spectra of spin adducts to expensive pathways
for synthesis. Moreover, simply placing a phenyl or
methyl group in the 3- or 4-position leaves the 3- or
4-tertiary carbon-hydrogen bond vulnerable to reaction
with the same radicals the nitronyl function is designed
to trap. MPPO has the carbon position on carbon-5
protected with a methyl group.
The prime reason for seeking improvements in spin
trapping methodology is for use in biological applica-
tions.11 Although in vivo detection of free radicals by
magnetic resonance is making progress,12 there is still a
problem with sensitivity. The sensitivity of these meth-
ods is directly dependent on the number of EPR lines
attributed to any spin adduct. Thus, the most desirable
spin traps are those which give the least number of EPR
hyperfine lines and which produce the most persistent
spin adducts with no artifacts.6
The approach we have used to synthesize MPPO is a
textbook preparation of synthon I, namely, 2-methylpyr-
roline N-oxide by Michael addition of nitromethane to
methyl vinyl ketone followed by reductive cyclization with
zinc:
(9) Ueno, I.; Kohno, M.; Mitsuta, K.; Mizuta, Y.; Kanegasaki, S. J .
Biochem. 1989, 105, 905-910.
(10) Makino, K.; Imaishi, H.; Morinishi, S.; Hagiwara, T.; Takeuchi,
T.; Murakami, A. Free Radical Res. Commun. 1989, 6, 19.
(11) (a) J anzen, E. G. Free Radical Biol. 1980, 4, 115. (b) McCay, P.
B.; Noguchi, T.; Fong, K.-L.; Lai, E. K.; Poyer, J . L. Free Radical Biol.
1980, 4, 153. (c) J anzen, E. G. In Oxygen Radicals in Biological
Systems, Methods in Enzymology; Packer, L., Ed.; Academic Press,
Inc.: New York, NY, 1984; Vol. 105, p 188. (d) Mason, R. P. In Spin
Labeling in Pharmacology; Holtzman, J . L., Ed.; Academic Press: New
York, 1984; p 87. (e) Rosen, G. M. Adv. Free Radical Biol. Med. 1985,
1, 345. (f) Mason, R. P.; Maples, K. R.; Knecht, K. T. Electron Spin
Resonance; Symons, M. C. R., Ed.; Royal Society of Chemistry: London,
1989, Vol. 11B, p 1. (g) Dodd, N. S. F. Electron Spin Resonance;
Symons, M. C. R. Ed.; Royal Society of Chemistry: London, 1990, Vol.
12A, p 136.
† Alternate address: Departments of Clinical Studies and Biomedi-
cal Sciences Ontario Veterinary College, University of Guelph, Guelph,
Ontario, Canada N1G 2W1.
(1) J anzen, E. G.; Liu, J . I.-P. J . Magn. Reson. 1973, 9, 510.
(2) Harbour, J . R.; Chow, V.; Bolton, J . R. Can. J . Chem. 1974, 52,
3549.
(3) Kotake, Y.; Reinke, L. A.; Tanigawa, T.; Koshida, H. Free Radical
Biol. Med. 1994, 17, 215.
(4) Sanders, S. P.; Harrison, S. J .; Kuppusamy, P.; Sylvester, J . T.;
Zweier, J . L. Free Radical Biol. Med. 1994, 16, 753.
(5) Powell, S. R. Free Radical Res. 1994, 21, 355.
(6) Tomasi, A.; Iannone, A. In EMR of Paramagnetic Molecules -
Biol. Magn. Reson. Plenum Press: New York, NY, 1993; Vol. 13,
Chapter 9, p 353.
(7) J anzen, E. G.; Haire, D. L. In Advances in Free Radical
Chemistry; Tanner, D. D., Ed.; J AI Press, Inc.: Greenwich, CT, 1990,
Chapter 6, p 253.
(12) (a) Lurie, D. L.; McLay, J .; Nicholson, I.; Mallard, J . R. J . Magn.
Reson. 1991, 95, 191. (b) Halpern, H. J .; Pou, S.; Peric, M.; Yu, C.;
Barth, E.; Rosen, G. M. J . Chem. Soc. 1993, 115, 218. (c) Halpern, H.
J .; Yu, C.; Barth, E.; Peric, M.; Rosen, G. M. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 796. Recently phosphorylated spin traps have been
published with more persistent spin adducts but additional hyperfine
splittings are produced in this case: Zeghdaoui, A.; Tuccio, B.; Finet,
J .-P.; Cerri, V.; Tordo, P. J . Chem. Soc., Perkin Trans. 2. 1995, 2087
and references therein.
(8) J anzen, E. G. In Bioradicals Detected by ESR Spectroscopy;
Ohya-Nishiguchi, Packer, L., Eds.; Birkhau¨ser Verlag: Switzerland,
1995; p 113.
S0022-3263(96)01476-4 CCC: $14.00 © 1997 American Chemical Society

Notes
J . Org. Chem., Vol. 62, No. 4, 1997 1177
KF
basic
O2NCH3 +
O2N
alumina
O
O
Zn
C6H5
CH3
C6H5MgCl
[O]
N
H
N
O–
O–
MPPO
I
Addition of phenyl magnesium chloride to I followed
by oxidation of the produced hydroxylamine with cupric
acetate provides MPPO in an overall yield of 28%. With
a melting point of 68 °C, MPPO is easier to purify than
DMPO (mp 25-30 °C).
F igu r e 1. EPR spectrum of the hydroxyl radical spin adduct
of MPPO produced in water from photolysis of 1% H2O2 in the
presence of 50 mM MPPO.
A key point in the synthesis of MPPO is the temper-
ature of Grignard addition to I.
If addition is performed at room temperature most
EPR spectra of MPPO spin adducts contain a major
impurity signal consisting of a 14 G triplet. This artifact
is completely removed by low-temperature addition of
Grignard, and final purification of MPPO by sublimation.
The importance of removing all EPR detectable impu-
rities or precursors to EPR detectable impurities cannot
be underestimated. Most literature references to the
preparation of nitrones describe methods of synthesis
which are rarely tested by EPR. Persistent nitroxides
are always produced in these schemes and never removed
because the concentrations are too low to influence the
outcome of the usual organic synthesis. EPR spectrom-
eters have high sensitivity, and good spin adduct spectra
can be obtained from 1 × 10-7 M solutions of 0.5 mL
volume samples. If the nitrone is typically 10-100 mM
in concentraion a 0.001% EPR active impurity can give
a spectrum intensity equal to that of the spin adduct of
interest. Commercial sources of DMPO have failed to
recognize this problem, and hence the large number of
reported artifactual determinations of false spin trapping
results.6
Typical hydroxyl radical generating systems produced
a 1:2:2:1 EPR spectrum (since aN ) aâ-H) assigned to the
HO• adduct of MPPO (see Figure 1).
F igu r e 2. (a) EPR spectrum of hydroperoxyl/superoxide
radical spin adduct of MPPO produced in water from photoly-
sis of 30% H2O2 in the presence of 50 mM MPPO. (b) Computer
simulation of the hydroperoxyl/superoxide radical spin adduct
shown in a. Parameters: scan width 100 G; (component A)
line width 1.1 G, relative amount 0.81, aN ) 14 G, aâ-H ) 12.1
G, aγ-H ) 1.1 G; (component B) line width 1.2 G, relative
amount 0.19, aN ) 14.1 G, aâ-H ) 7.6 G, aγ-H ) 0.7 G.
OH
C6H5
CH3
C6H5
CH3
•
+ HO
N
O
H
N
O
H
•
At this time we do not know whether the 1:2:2:1
pattern is due to a mixture of diastereoisomeric spin
adducts where each isomer happens to give the same
hyperfine splitting constants, and hence the same EPR
spectral pattern, or whether only one isomer (hydroxyl
group cis or trans to the phenyl group) is formed. For
usual hydroxyl radical spin trapping experiments where
only the question of detection is considered, knowledge
of the stereochemistry of addition is not necessary.
The addition of superoxide to MPPO produces an EPR
spectrum very similar to that of DMPO (see Figure 2).
smaller â-H hyperfine splitting: aN ) 14.1, aâ-H ) 7.6,
aγ-H ) 0.7 G.
In response to a reviewer’s comment it is fair to say
that the possible presence of diastereomeric spin adduct
spectra does complicate the issue. The spectra are more
complex and some sensitivity is lost. However, for some
problems this feature may serve as an advantage in
sorting out real spin trapping events from imposter spin
adduct products. Other spin adduct spectra of MPPO
have been described and illustrated elsewhere.13
Preliminary in vivo studies in the rat show that MPPO
is successful at trapping trichloromethyl radicals pro-
duced in the liver metabolism of carbon tetrachloride.
When 13CCl4 is used the spectrum shown in Figure 3 is
OOH
H
H+
C6H5
CH3
C6H5
CH3
–
•
+ O2
N
H
N
O
O–
•
In this case careful inspection indicates the presence
of two EPR spectra. The stronger spectrum has EPR
parameters very similar to those of DMPO: aN ) 14.0,
aâ-H ) 12.1, aγ-H ) 1.1 G. The weaker spectrum has a
(13) J anzen, E. G.; Sankuratri, N.; Kotake, Y. J . Magn. Reson. B.
1996, 111, 254-261.

1178 J . Org. Chem., Vol. 62, No. 4, 1997
Notes
mL of dry THF was added 20 g of KF/basic alumina containing
3.5 mol equiv KF/g of alumina with vigorous stirring for 1.5 h.
After the solution was filtered and washed with ethyl acetate,
rotoevaporated, and distilled, 13.2 g (83% yield) of a colorless
oil (5-nitro-2-pentanone) was isolated (bp ) 37 °C at 0.03 Torr)
and used in the next step. 5-Nitro-2-pentanone (10 g, 0.07 mol)
and ammonium chloride (4.4 g, 0.82 mol) in 40 mL of water were
cooled to -10 °C and treated with zinc dust added in small
portions over 3 h with vigorous stirring (temperature less than
5 °C). After this solution was stirred (30 min) and filtered,
washed with methanol, concentrated, extracted with CHCl3,
dried with MgSO4, and rotoevaporated, the nitrone was isolated
by distillation [(bp 56 °C at 0.02 Torr; 4.62 g (61% yield)]. 1H-
NMR (300 MHz, CDCl3): δ 3.91 (2H, br t, J ) 6.5 Hz, H-5),
2.69 (2H, br t, J ) 6.7 Hz, H-3), 2.13 and 2.07 (2H, AB q, J )
7.7 Hz, H-4), 2.01 ppm (3H, s, methyl). 13C-NMR (75 MHz,
CDCl3): δ 143.6 (C-2), 60.6 (C-5), 31.4, 15.0, 10.8 ppm.
F igu r e 3. EPR spectrum in CHCl3 obtained from Folch
extraction (CHCl3/CH3OH 2:1) of homogenized liver from a rat
(226 g) subjected to 13CCl4 (66.9 µL) and MPPO (27 mg) in 1
mL each of phosphate buffer (pH 7.5) and corn oil suspension
5-Meth yl-5-p h en ylp yr r olin e N-Oxid e (MP P O). To a pre-
cooled (-20 °C) solution of a phenylmagnesium chloride (25 mL,
0.075 mol) in dry ether was added 2-MPO (4.0 g, 0.04 mol) in
ether. After stirring, the reaction was quenched with wet ether
followed by NH4Cl solution. The Mg(OH)2 was filtered and
washed with chloroform, and the solvent was evaporated under
reduced pressure to yield a yellow oil (4.1 g, 57%). This oil was
treated with copper acetate (520 mg) in methanol (15 mL)
containing aqueous NH4OH and bubbled with air until the
solution remained blue. After evaporation and dissolution in
chloroform, the solution was chromatographed on a silica column
and evaporated as a waxy solid. MPPO was isolated and
sublimed at 40 °C (0.02 Torr), yield 2 g (28%) mp 68 °C. IR
(thin film): νmax 1573, 1501, 1447, 1375, 1276, 1233, 1193, 1072,
13
by intraperitoneal injection after 1 h. The â-H and â-
C
hyperfine splittings are accidently equal: aN ) 13.0, aâ-H
)
13
16.3, aâ- ) 16.3 in CHCl3.
C
obtained from a chloroform extraction of homogenized
liver. Because the hydrogen and â-13C hyperfine splitting
constant in the spin adduct are acccidentally equal the
pattern becomes a 1:2:1 triplet of 1:1:1 triplets. It should
be noted that DMPO is ineffective in trapping radicals
from rat liver metabolism of CCl4.14
Exp er im en ta l Section
1025, 822, 765 cm-1 1H-NMR (300 MHz, CDCl3): δ 7.30 (5H,
.
2-Meth ylp yr r olin e N-Oxid e (2-MP O). To nitromethane (50
mL, 0.9 mol) and methyl vinyl ketone (10 mL, 0.1 mol) in 500
m, aromatic), 7.16 (1H, t, J ) 2.4 Hz, H-2), 2.61 (2H, m, H-3),
2.50 and 2.36 ppm (2H, m, H-4). 13C-NMR (75 MHz, CDCl3): δ
141.7 (C-2), 133.9, 128.8, 127.7, 125.3, 79.2, 36.6, 24.6, 24.3 ppm.
Anal. Calcd. for C11H13NO: C, 75.50; H, 7.50; N, 8.00. Found:
C, 75.51; H, 7.56; N, 7.89.
(14) Results from unpublished experiments with DMPO indicate
variable outcomes in the in vivo metabolism of CCl4 in the rat (work
with Dr. J . L. Poyer and Ms. M. S. West in these laboratories). When
DMPO is used, usually no spin adduct is obtained. Very occasionally
a weak EPR signal is recorded but this result is not reproducable.
J O961476X