A dual sensor spin trap for use with EPR spectroscopy

Article abstract of DOI:10.1021/ol071285o

Redox active metal ions, carbon-centered radicals, and oxygen-centered radicals are important to oxidative stress. A radical detector combining a nitrone spin trap, a phenol, and a cyclopropane radical clocklike unit was prepared and used with EPR spectroscopy to detect and distinguish between hydroxyl radicals, methyl radicals, and iron(III) ions. Iron(III) reacts with the phenol unit inducing opening of the cyclopropane ring and cyclization to generate a stable nitroxyl radical.

Full text of DOI:10.1021/ol071285o

ORGANIC
LETTERS
2007
Vol. 9, No. 18
3499-3502
A Dual Sensor Spin Trap for Use with
EPR Spectroscopy
Stuart T. Caldwell,† Caroline Quin,† Ruth Edge,‡ and Richard C. Hartley*,†
WestCHEM Department of Chemistry, UniVersity of Glasgow, Glasgow, G12 8QQ,
U.K. and EPSRC National EPR SerVice, School of Chemistry, UniVersity of
Manchester, Oxford Road, Manchester, M13 9PL, U.K.
richh@chem.gla.ac.uk
Received May 31, 2007
ABSTRACT
Redox active metal ions, carbon-centered radicals, and oxygen-centered radicals are important to oxidative stress. A radical detector combining
a nitrone spin trap, a phenol, and a cyclopropane radical clocklike unit was prepared and used with EPR spectroscopy to detect and distinguish
between hydroxyl radicals, methyl radicals, and iron(III) ions. Iron(III) reacts with the phenol unit inducing opening of the cyclopropane ring
and cyclization to generate a stable nitroxyl radical.
There is convincing evidence that oxidative stress due to
reactive oxygen species1 (ROS) is a major contributor to the
process of aging.2 ROS are also involved in a wide range of
pathologies including neurodegenerative diseases3 such as
Parkinson’s and Alzheiemer’s diseases and in ischemic
events such as a stroke.4
superoxide to give the superoxide-assisted Fenton reaction
(Scheme 1a and b) and may also act as oxidizing agents
themselves by single electron transfer. Thus, reaction between
iron(III) and phenols, including R-tocopherol and flavonoids
found in the diet, which are generally viewed as antioxi-
dants,6-8 would have a pro-oxidant effect.
The most damaging ROS are hydroxyl (HO•) and peroxyl
radicals (ROO•).1 These can also produce highly reactive
carbon-centered radicals by hydrogen atom abstraction from
biomolecules, particularly polyunsaturated fatty acids,5 which
may then reduce oxygen to regenerate peroxyl radicals.
Redox active metal ions [e.g., copper(I) and iron(II)] are
important to the generation of hydroxyl radicals by the
Fenton reaction (Scheme 1a) and related processes, with their
Central to the study of oxidative stress is the ability to
detect and distinguish between the various chemical con-
tributors to it and to determine their locality in cells and
organs. One approach is to use fluorescent probes, which
are particularly good in vitro where confocal microscopy can
(1) Halliwell, B.; Gutteridge, J. M. C. Free Radicals in Biology and
Medicine, 3rd ed.; OUP: Oxford, 1999.
(2) Schriner, S. E.; Linford, N. J.; Martin, G. M.; Treuting, P.; Ogburn,
C. E.; Emond, M.; Coskun, P. E.; Ladiges, W.; Wolf, N.; Van Remmen,
H.; Wallace, D. C.; Rabinovitch, P. S. Science 2005, 308, 1909-1911.
(3) Lin, M. T.; Beal, M. F. Nature 2006, 443, 787-795.
(4) Adibhatla, R. M.; Hatcher, J. F. Free Radical Biol. Med. 2006, 40,
376-387.
Scheme 1. Superoxide-Assisted Fenton Reaction
(5) Yin, H.; Musiek, E. S.; Gao, L.; Porter, N. A.; Morrow, J. D. J.
Biol. Chem. 2005, 280, 26600-26611.
(6) For a discussion of the importance and limitations of R-tocopherol
as a biological antioxidant, see: Kim, H.-Y.; Pratt, D. A.; Seal, J. R.;
Wijtmans, M.; Porter, N. A. J. Med. Chem. 2005, 48, 6787-6789 and refs
1-4 therein.
(7) Radical scavenging properties of flavonoids are well established: (a)
McPhail, D. B.; Hartley, R. C.; Gardner, P. T.; Duthie, G. G. J. Agric.
Food Chem. 2003, 51, 1684-1690. (b) Bors, W.; Michel, C.; Stettmaier,
K. Methods Enzymol. 2001, 335, 166-180 and refs therein.
(8) The role of flavonoids as dietary antioxidants in vivo is disputed:
Lotito, S. B.; Frei, B. Free Radical Biol. Med. 2006, 41, 1727-1746.
oxidation potential depending on ligands. The higher oxida-
tion states of these metals can be reduced in situ by
†
University of Glasgow
‡
University of Manchester
10.1021/ol071285o CCC: $37.00
© 2007 American Chemical Society
Published on Web 08/01/2007

be employed, and recently great advances have been made
in developing probes that detect different ROS selectively.9
Although simple to use, sometimes results from fluorescent
probes can be misleading; e.g., probes that were at first
believed to be responding to ROS have since been found to
be measuring free cytochrome C instead.10 Electroparamag-
netic resonance (EPR) spectroscopy has the advantage that
it only detects radicals and can be used both in vitro and in
vivo (albeit at different operating frequencies). Although
many of the radicals found in biological systems are too
short-lived for direct detection by EPR spectroscopy, spin
traps have been developed that convert highly reactive
radicals into more stable radicals. Nitrones are generally
employed as these trap highly reactive radicals to give stable
nitroxyl radicals,11 which can be detected, and the adducts
of specific ROS and carbon-centered radicals can have
characteristic spectra.12 Thus, the nitrone can be regarded
as a radical sensor and the resulting nitroxyl as the actuator,
for display by EPR spectroscopy.
atom abstraction or by sequential proton loss-electron
transfer to give phenoxyl radical 3, with the rate and
mechanism depending on the solvent and the species
involved.13 The cyclopropane is related to known radical
clocks14 and should open rapidly to give an unstable primary
radical 4, which will cyclize onto the nitrone to generate a
nitroxyl actuator 5. The different nitroxyl radicals should
have different EPR spectra allowing the processes to be
distinguished. Only radical processes would be detected.
Our strategy for the construction of the double detector
probe involved aldehyde 11 as a key intermediate (Scheme
3). Although benzophenone derivatives bearing masked
Scheme 3. Preparation of Dual Sensor Probe 1
We designed probe 1, which has two different sensor
moieties and the same actuator, to allow the detection and
identification of a wide range of species involved in oxidative
stress. The probe 1 consists of two noncommunicating
aromatic rings, linked by a cyclopropane ring, with the
nitrone and phenol groups acting as sensors for different
types of radicals by different mechanisms (Scheme 2).
Scheme 2. Design of Dual Sensor Probe
aldehydes are easily accessible,15 the choice of protecting
group is problematic as unmasking must tolerate the cyclo-
(10) Burkitt, M.; Jones, C.; Lawrence, A.; Wardman, P. Biochem. Soc.
Symp. 2004, 71, 97-106 and refs 3-5 therein.
(11) Recent examples of nitrones as probes and antioxidants include:
(a) Xu, Y. K.; Kalyanaraman, B. Free Radical Res. 2007, 41, 1-7. (b)
Sklavounou, E.; Hay, A.; Ashraf, N.; Lamb, K.; Brown, E.; MacIntyre, A.;
George, W. D.; Hartley, R. C.; Shiels, P. G. Biochem. Biophys. Res.
Commun. 2006, 347, 420-427. (c) Kamibayashi, M.; Oowada, S.; Kameda,
H.; Okada, T.; Inanami, O.; Ohta, S.; Ozawa, T.; Makino, K.; Kotake, Y.
Free Radical Res. 2006, 40, 1166-1172. (d) Ionita, P. Free Radical Res.
2006, 40, 59-65. (e) Ortial, S.; Durand, G.; Poeggeler, B.; Polidori, A.;
Pappolla, M. A.; Bo¨ker, J.; Hardeland, R.; Pucci, B. J. Med. Chem. 2006,
49, 2812-2820. (f) Hardy, M.; Ouari, O.; Charles, L.; Finet, J.-P.; Iacazio,
G.; Monnier, V.; Rockenbauer, A.; Tordo, P. J. Org. Chem. 2005, 70,
10426-10433. (g) Hay, A.; Burkitt, M. J.; Jones, C. M.; Hartley, R. C.
Arch. Biochem. Biophys. 2005, 435, 336-346. (h) Allouch, A.; Roubaud,
V.; Lauricella, R.; Bouteiller, J.-C.; Tuccio, B. Org. Biomol. Chem. 2005,
3, 2458-2462. (i) Liu, Y. P.; Ji, Y. Q.; Song, Y. G.; Liu, K. J.; Liu, B.;
Tian, Q.; Liu, Y. Chem. Commun. 2005, 4943-4945.
Sterically unhindered electron-rich carbon-centered radicals
and very reactive, and hydroxyl radicals, which are very
reactive, would be expected to form adducts with the nitrone
moiety rapidly, generating the sterically protected nitroxyl
radicals 2.12 On the other hand, electron-poor species might
be expected to react with the phenolic moiety by hydrogen
(12) Rosen, G. M.; Britigan, B. E.; Halpern, H. J.; Pou, S. Free
Radicals: Biology and Detection by Spin Trapping; OUP: Oxford, 1999.
(13) Musialik, M.; Litwinienko, G. Org. Lett. 2005, 7, 4951-4954.
(9) Soh, N. Anal. Bioanal. Chem. 2006, 386, 532-543.
3500
Org. Lett., Vol. 9, No. 18, 2007

tion of 2,6-dimethylphenol gave bromide 6. Lithiation and
reaction with 2-bromobenzaldehyde gave alcohol 7, which
was oxidized to the benzophenone 8. Wittig methylenation
gave alkene 9, and modified Simmons-Smith cyclopro-
panation18 proceeded in good yield to give 1,1-diarylcyclo-
propane 10. Lithiation and reaction with DMF gave the key
intermediate 11, which was converted into a nitrone and
deprotected to give the desired target 1 in modest yield over
three steps.
Nitrone 1 was reacted with hydroxyl radicals generated
from hydrogen peroxide and iron(II) sulfate19 (Scheme 1a),
and the EPR spectrum was recorded (Figure 1a).
This signal slowly declined, and a more complex stable
signal began to appear. Because iron(III) is a byproduct of
the Fenton reaction, we reasoned that this new signal might
arise from the metal ion behaving as an oxidant. Sure enough,
when nitrone 1 was treated with iron(III) chloride, a strong
long-lasting signal was recorded (Figure 1b). Finally, methyl
radicals were generated from hydroxyl radicals by reaction
with DMSO19 and trapped to give a product that had a
different spectrum (Figure 1c).
The hyperfine splittings in spectra a and c shown in Figure
1 were easily derived by simulation. We reasoned that
reaction with the hydroxyl and methyl radicals occurred at
the nitrone to give adducts 12 and 13 (Scheme 4)20 and that
Scheme 4. Reactions of Dual Sensor Probe 1
iron(III) ions caused oxidation of the phenol moiety of
nitrone 1 and production of tricyclic nitroxide 5, where
hyperfine coupling to the γ and/or δ hydrogen atoms is
observed.
To confirm the assignment of tricycle 5 and to assist in
the interpretation of spectrum b (Figure 1), tetradeuterated
Figure 1. EPR experiments with nitrone 1.19
(15) (a) Flanagan, S. R.; Harrowven, D. C.; Bradley, M. Tetrahedron
2002, 58, 5989-6001. (b) Ukita, T.; Nakamura, Y.; Kubo, A.; Yamamoto,
Y.; Moritani, Y.; Saruta, K.; Higashijima, T.; Kotera, J.; Takagi, M.;
Kikkawa, K.; Omori, K. J. Med. Chem. 2001, 44, 2204-2218. (c) Couture,
A.; Deniau, E.; Lebrun, S.; Grandclaudon, P. J. Chem. Soc., Perkin 1 1999,
789-794. (d) Patil, P. A.; Joshi, R. R.; Narasimhan, N. S. Indian J. Chem.
Sect. B 1987, 26, 1025-1029.
propyl unit which is in close proximity.16 Consequently, we
decided to introduce the aldehyde group late in the synthesis
using a cyclopropanation, lithiation, formylation sequence
demonstrated by Jason.17 Thus, bromination and silyl protec-
(16) Preliminary investigation of such routes showed that cyclizations
competed at various stages.
(14) (a) Lemieux, R. P.; Beak, P. J. Org. Chem. 1990, 55, 5454-5460.
(b) Newcomb, M. Tetrahedron 1993, 49, 1151-1176.
(17) Jason, M. E. Tetrahedron Lett. 1982, 23, 1635-1638.
(18) Charette, A. B.; Beauchemin, A. Org. React. 2001, 58, 1-415.
Org. Lett., Vol. 9, No. 18, 2007
3501

cyclopropane 14 was prepared by the same route, but using
Ph3PdCD2 in the alkenation step and CD2I2 for the cyclo-
propanation step. Reaction between nitrone 14 and iron(III)
chloride gave rise to a simplified EPR spectrum correspond-
ing to nitroxide 15 (Figure 2), confirming the original
for nitroxide 15. Coupling to two of the γ and/or δ hydrogen
atoms in nitroxyl radical 5 is observed, and the AH or AH
values of about 1 G derived from simulation are consistent
with those in the literature for N-tert-butyl-N-cyclohexyl
nitroxide.21
γ
δ
In conclusion, we have demonstrated a proof of the
principle that combination of two different radical sensors,
a radical clocklike relay system, and a nitrone allows oxygen-
centered radicals, carbon-centered radicals, and an oxidizing
metal ion to be detected and distinguished by EPR spec-
troscopy.
Acknowledgment. We thank EPSRC and SPARC for
funding, EPSRC National Service for EPR Spectroscopy, and
in particular, Dr. Radoslaw Kowalczyk and Prof. Eric
McInnes for advice and training.
Supporting Information Available: Full experimental
procedures, spectroscopic data, and scanned NMR spectra
for compounds 1, 6-11, and 14 are included. Details of EPR
experiments and simulations are also included. This material
is available free of charge via the Internet at http://pubs.acs.org.
OL071285O
(19) Conditions for generation of hydroxyl and methyl radicals were
adapted from: Rosen, G. M.; Tsai, P.; Barth, E. D.; Dorey, G.; Casara, P.;
Spedding, M.; Halpern, H. J. J. Org. Chem. 2000, 65, 4460-4463. Polyakov,
N. E.; Kruppa, A. I.; Leshina, T. V.; Konovalova, T. A., Kispert, L. D.
Free Radical Biol. Med. 2001, 31, 43-52. (a) 3.3 mM 1, 0.33 mM FeSO4,
0.33 mM H2O2, DMF-H2O (1:2). (b) 3.3 mM 1, 0.33 mM FeCl3, DMF-
H2O (1:2). (c) 3.3 mM 1, 0.33 mM FeSO4, 0.33 mM H2O2, DMSO-H2O
(1:2). g values referenced to strong pitch.
(20) Hydroxyl radicals are very highly reactive, and some reaction at
the phenolic position may have occurred; however, any signal for the
tricyclic nitroxide 5 is initially obscured by the signal from the more
abundant nitroxyl radical 12.
Figure 2. Reaction of deuterated nitrone 14 with FeCl3 and the
EPR spectrum of its adduct 15.22
(21) Wajer, Th. A. J. W.; Mackor, A.; de Boer, Th. J. Recl. TraV. Chim.
assignment of nitroxide 5. The spectrum b of nitroxide 5
was simulated using the values for AN and AH determined
Pays-Bas 1971, 90, 568-576.
â
(22) Conditions: 3.3 mM 14, 0.33 mM FeCl3, DMF-H2O (1:2).
3502
Org. Lett., Vol. 9, No. 18, 2007