A new dual probe for hydrogen abstraction

Article abstract of DOI:10.1016/j.tet.2009.05.073

A new dual (fluorescent and spin) probe is described, where a N-aryl-2,4,6-triphenylpyridinium fluorophore is attached to a TEMPO fragment through an amide link. The resulting sensor 4 was evaluated as a hydrogen-abstracting agent in acetonitrile and in a

Full text of DOI:10.1016/j.tet.2009.05.073

Tetrahedron 65 (2009) 6025–6028
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
A new dual probe for hydrogen abstraction
*
Carolina Aliaga , Marcos Caroli Rezende, Cristian Tirapegui
Facultad de Qu´ımica y Biolog´ıa, Universidad de Santiago, Casilla 40, Correo 33, Santiago, Chile
a r t i c l e i n f o
a b s t r a c t
Article history:
A new dual (fluorescent and spin) probe is described, where a N-aryl-2,4,6-triphenylpyridinium fluo-
rophore is attached to a TEMPO fragment through an amide link. The resulting sensor 4 was evaluated as
a hydrogen-abstracting agent in acetonitrile and in an aqueous solution of reduced Triton-X 100, being as
resistant to hydrolysis as quinoline-TEMPO 1, but more hydrophobic than this probe.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 23 April 2009
Accepted 26 May 2009
Available online 6 June 2009
.
O
1. Introduction
N
The use of fluorescence as a method to sense different processes
at molecular level is well established.1–6 Recently, pre-fluorescent
probes obtained by attaching a nitroxide moiety to a fluorophore
were successfully employed to monitor hydrogen-transfer
reactions.6
The mechanism of action of these pre-fluorescent probes is well
described.7–9 The paramagnetic nitroxide fragment attached to the
quenched fluorophore selectively abstracts a hydrogen atom from
a good hydrogen donor6 or efficiently traps carbon-centered radi-
cals.10,11 The resulting diamagnetic hydroxylamine or alkoxyamine
thereby restores the fluorescence of the chromophore providing
experimental evidence for the reaction and an easily measurable
way of following its kinetics.
Pre-fluorescent probes are very sensitive and appear less prone
to suffer from the problems of electron-deficient probes such as
DPPH or of mechanisms of complex stoichiometry, as when ABTS
radicals are used.12–14
Another remarkable advantage of these sensors is their high
selectivity. Hydrogens in polyphenols are abstracted exclusively
from the most reactive hydroxyl group, leading to clean and easily
identifiable processes.6,15
.
O
N
O
O
O
O
O
O
OH
N
N
1
2
Substituted pyridinium salts seemed to us an interesting family
to be explored, because of their synthetic versatility and accessi-
bility. Their preparation starts from substituted pyrylium salts,
available with a variety of substituents.19–21
Their fluorescence is well documented.22–27 Although there is
one report in the literature of a pyridinium derivative incorporating
a TEMPO fragment28 the resulting probe was not pre-fluorescent,
a limitation that restricts its use to measurements employing
electron paramagnetic resonance spectroscopy.
In previous reports, we have employed quinoline-TEMPO 1 and
coumarin-TEMPO 2 as pre-fluorescent probes to specifically eval-
uate the hydrogen-transfer reaction rates of antioxidants.6,15–17
As pointed out by us in a recent study18 when measuring anti-
oxidant activities in micro-heterogeneous media, the lipophilicity
of the assessing probe is a key factor to be considered. For this
reason, we have been interested in developing new easily available
pre-fluorescent probes with different lipophilicities.
In the present report we describe the preparation of the dual
pre-fluorescent probes 3 and 4, and evaluate them as hydrogen-
abstractors, comparing their stability and lipophilicity with that of
quinoline-TEMPO 1.
2. Results and discussion
The pyridinium TEMPO probe 3 was prepared in two steps from
the 2,4,6-triphenylpyrylium perchlorate, which was converted into
the N-(4-carboxyphenyl)pyridinium salt,29 and esterified with 4-
hydroxyTEMPO. The fluoroborate salt 4 was prepared by reaction of
the N-(4-carboxyphenyl)pyridinium salt with 4-amino-TEMPO.30
* Corresponding author.
E-mail address: carolina.aliaga@usach.cl (C. Aliaga).
0040-4020/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2009.05.073

6026
C. Aliaga et al. / Tetrahedron 65 (2009) 6025–6028
O
N
O
N
1.68
1.56
1.44
A
B
O
O
NH
O
-
-
ClO4
Ph
BF4
1.32
+
+
Ph
N
Ph
Ph
N
1.20
1.08
0.96
Ph
Ph
3
4
0
500
1000
Time, sec
1500
2000
2500
Ph
3
Ph
ii
Figure 1. (A) Variation of the fluorescence of 3 (10 mM), evaluated at 466 nm, after
addition of TROLOX (10 mM) in acetonitrile (excitation wavelength lexc¼355 nm) (-);
i
(B) control experiment for probe 3 in acetonitrile, in the absence of TROLOX (,).
Ph
N
Ph
iii
+
Ph
O
Ph
+
derivative 3, we compared the obtained rate value in acetonitrile
with the rate constant obtained in methanol-d6. In Figure 2 the
exponential decay curves for the two pseudo-first order reactions
are compared. It is seen that the reaction in methanol-d6 is much
slower than in acetonitrile. In contrast with the reaction in aceto-
nitrile, practically finished after 1500 s, in deuterated methanol,
only about 20% of the TEMPO derivative had been consumed. Cal-
culation of the second-order constants from these plots led to
a deuterium isotope effect of 8.9, arising from the fully deuterated
phenolic group. The observed value of 0.019 Mꢁ1 sꢁ1 in the deu-
terated solvent (Fig. 2, curve B) reproduced the value obtained for
the reaction of probe 1 with a phenolic compound in methanol-
d6.35 This kinetic result was another indication that, for both
TEMPO derivatives 1 and 3, the same hydrogen/deuterium ab-
straction from the hydroxyl group of the phenol to the TEMPO
derivative was operating, with the formation of analogous hy-
droxylamine derivatives. Further evidence that no hydrogen ab-
straction by probe 4 took place in MeCN in the absence of TROLOX
was obtained from a blank experiment, by measuring the fluores-
cence of a solution of derivative 4 in acetonitrile (Fig. 3, curve C).
Under these circumstances, no variation of the fluorescence in-
tensity of the sample was observed after 30 min.
4
X
X
CO2H
X = ClO4 or BF4
(i) 4-HO CC6H4NH2, EtOH, reflux 48 h;
2
(ii) 4-hydroxy-TEMPO, DMAP, EDAC;
(iii) 4-amino-TEMPO, DMAP, DCC.
In order to test probes 3 and 4 as dual sensors by using fluo-
rescence and EPR techniques, we evaluated their ability to monitor
the hydrogen-transfer reaction from a well-known hydrogen donor
such as TROLOX.
2.1. Hydrogen abstraction in a pure solvent
Abstraction of a hydrogen atom from TROLOX by the pre-fluo-
rescent probes 3 or 4 led to the formation of the corresponding
hydroxylamines. Several examples of such a conversion for a vari-
ety of TEMPO derivatives are found in the literature.6,31–33 Indirect
evidence for the formation of these fluorescent hydroxylamines
was obtained by comparison of the emission spectrum of the
product of the reaction of 3 and TROLOX with the fluorescence
spectrum of the N-(4-carboxyphenyl)-2,4,6-triphenylpyridinium
fluoroborate. This was based on the fact that the restored emission
of the fluorophore in these nitroxide-fluorophore adducts is un-
affected by the nitroxide substituent.34 Both spectra exhibited the
same emission at 466 nm, when excited at 355 nm, indicating that
the formation of a hydroxylamine from 3 had restored the emission
of the carboxyphenylpyridinium fluorophore.
We next investigated the stability of our newly synthesized
probes and the possible hydrolysis of the ester or amide functions
1.0
B
0.8
0.6
0.4
The reactions of probes 3 or 4 with TROLOX in acetonitrile were
followed by fluorescence and EPR. Both compounds showed
a similar kinetic behavior.
A plot of the fluorescence intensity I/I0 vs time for the reactions
of probe 3 with TROLOX is shown in curve A of Figure 1. An
analogous plot for probe 4 is shown in Figure 3, curve A. The cal-
culated rate constants in MeCN were 0.24ꢀ0.01 Mꢁ1 sꢁ1 for probe 3
and 0.17ꢀ0.01 Mꢁ1 sꢁ1 for probe 4. Figure 2, curve A, and Figure 3,
curve B, represent the course of the same reaction, as followed by
A
0.2
0.0
0
300
600
900
1200
1500
EPR. Addition of 10 mM of TROLOX to a 20 mM solution of the probe
3 or 4 in acetonitrile led to a gradual decrease of the radical triplet
Time, sec
signal with time. The calculated rate for hydrogen abstraction in
Figure 2. (A) Variation with time of the intensity of the first signal of the EPR triplet of
3 (20
M) by reaction with excess TROLOX (10 mM) in MeCN at 25 ꢂC (-); (B) decrease
both cases was 0.19ꢀ0.01 Mꢁ1 sꢁ1
.
m
In order to confirm that the mechanism involved a hydrogen
transfer from the hydroxyl group of TROLOX to the TEMPO
of the EPR signal of 3 (20
25 ꢂC (,).
mM) in methanol-d6 in the presence of TROLOX (11 mM) at

C. Aliaga et al. / Tetrahedron 65 (2009) 6025–6028
6027
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
1.00
0.98
0.96
0.94
0.92
0.90
0.88
0.86
1.0
0.8
0.6
0.4
A
B
4
1
C
0
500
1000
1500
Time, s
2000
2500
3000
0
10
20
30
Time, min
40
50
60
Figure 3. (A) Fluorescence of 4, evaluated at 465 nm, after addition of TROLOX
(10 mM) in acetonitrile (excitation wavelength lexc¼355 nm) (-); (B) variation with
Figure 4. Variation with time of the intensity of the first signal of the EPR triplet of 1 or
4 (20 M), by reaction with excess TROLOX (20 mM) in an aqueous micellar solution of
reduced Triton-X 100 (20 mM).
m
time of the intensity of the first signal of the EPR triplet of 4 (10 mM) by reaction with
excess TROLOX (10 mM) in acetonitrile at 25 ꢂC; (C) variation of the fluorescence of 4
(10 M) in acetonitrile, in the absence of TROLOX (,).
m
signal of 1 decayed by nearly 65% at the end of the reaction. By
contrast, in the case of compound 4, the observed decay was only
35% (Fig. 4). Assuming that the hydrophilic TROLOX, responsible for
quenching radicals 1 and 4, was overwhelmingly present in the
aqueous phase, we conclude that 65% of probe 1 and 35% of probe 4
were present in water. Thus, this simple method allowed us to
estimate partitioning ratios between the hydrophobic micelle and
water of 0.35/0.65–0.5 and 0.35/0.65–1.9 for compounds 1 and 4,
respectively, indicative of the greater hydrophobic character of
compound 4 when compared with 1.
that linked the TEMPO fragment with the fluorophore. Blank exper-
iments with compound 3 or 4 in acetonitrile at 25 ꢂC, in the absence
of TROLOX, indicated that, under these conditions, probe 3 un-
derwent some degree of hydrolysis by the water present in MeCN
(Fig. 1, curve B). This concomitant hydrolysis of 3 would account for
the larger value obtained by fluorescence measurements for the rate
of hydrogen abstraction from TROLOX (0.24 Mꢁ1 sꢁ1), when com-
pared with the rate for probe 4 (0.17 Mꢁ1 sꢁ1) and the value reported
for 1 in MeOH (0.19 Mꢁ1 sꢁ1).6 Under the same conditions, the amide
4 was stable (Fig. 3, curve C). The greater stability of probe 4 was
further confirmed by running blank experiments in water, at pH 7
and 4. In both cases, the amide 4 showed no signs of hydrolysis,
proving to be as stable as quinoline-TEMPO 1. The agreement be-
tween the results from the two techniques when probe 4 was
employed was also an indication that the rise in fluorescence and
decay in EPR are due solely to the hydrogen abstraction from TROLOX.
These results made us concentrate our interest in probe 4.
3. Conclusion
In summary, attachment of a TEMPO fragment to a tetraphe-
nylpyridinium skeleton leads to two new, dual (fluorescent and
spin) probes for hydrogen abstraction processes. Like quinoline-
TEMPO probe 1, in the pyridinium system 3, the TEMPO fragment is
attached to the chromophore by an ester function. Blank experi-
ments indicated that this ester functionality was more resistant to
hydrolysis in the case of probe 1 than of probe 3. Replacement of
the ester in 3 by an amide function led to the more stable probe 4,
which was not hydrolyzed under the same experimental conditions
nor in water at pH 7 or 4. Compound 4 was also more hydrophobic
than probe 1, as shown by estimates of relative partitioning in an
aqueous micellar solution of reduced Triton-X 100. Thus, the stable
pyridinium salt 4 should complement the quinoline-TEMPO de-
rivative 1 as a more hydrophobic hydrogen-abstracting probe, in
those cases where the antioxidant activity of hydrophobic sub-
strates in micro-heterogeneous systems is evaluated.
2.2. Hydrogen abstraction in a micellar environment
Previous work had shown that compound 1 does not undergo
any self-quenching by the presence of a phenolic hydroxyl group in
the molecule.6 This might be attributed to the conditions under
which kinetics of hydrogen abstraction are normally performed. In
the presence of a 1000-fold excess of a phenol, the competing in-
termolecular abstraction of a hydrogen atom from the phenolic
hydroxyl group of 1 should not be significant. A second, reasonable
explanation for the absence of self-quenching might be that com-
pound 1 exists as a zwitterion, and its OH group is deprotonated to
form a quinolinium cation. The resulting zwitterionic form of the
quinoline-TEMPO 1 should then be more hydrophilic than its un-
charged isomer.
Development of new dual probes with variously substituted
pyridinium fluorophores is presently under way in our laboratories.
4. Experimental section
The relative hydrophobicities of probes 1 and 4 were evaluated
in terms of their partitioning in an aqueous neutral micellar solu-
tion of reduced Triton-X 100. Their reactions with TROLOX in this
medium were followed by EPR and are shown in Figure 4.
Hydrogen abstraction from the hydrophilic TROLOX by either
probe can be assumed to take place in the aqueous phase of the
micellar system. The rates of hydrogen abstraction do not vary with
the nature of the TEMPO derivative, as confirmed by the first-order
exponential decay curves for both probes, which yielded very
similar rate constants. However, their different plateau values at
the end of the reactions point to different final concentrations of 1
and 4 in the aqueous phase. The intensity of the normalized EPR
The melting points of compounds 3 and 4 were recorded with
a capillary Electrothermal apparatus and were not corrected. Steady-
state fluorescence measurements were performed in a Spex Fluo-
rolog 1681 spectrofluorimeter. EPR experiments were carried out
with a Bruker EMX 1572 spectrometer that operates in the X-band
(9.2–9.9 GHz) equipped with a thermostatized cavity. UV–visible
spectra were collected with a HP-8453 diode array spectrometer.
Chemicals were purchased from Sigma–Aldrich and used as
received.
The N-(4-carboxyphenyl)-2,4,6-triphenylpyridinium per-
chlorate was prepared by refluxing equimolar amounts of

6028
C. Aliaga et al. / Tetrahedron 65 (2009) 6025–6028
2,4,6-triphenylpyrylium perchlorate and 4-carboxyaniline in
ethanol for 48 h.25 The fluoroborate pyridinium salt was pre-
pared following the same procedure. Excitation of this salt
The reactions between probes 3 or 4 and TROLOX in acetonitrile
were followed by EPR under pseudo-first-order conditions at 25 ꢂC,
by monitoring the decrease with time of the first signal of the
nitroxyl radical triplet. In an analogous way, the kinetics of hy-
drogen abstraction from TROLOX of probes 1 or 4 in an aqueous
solution of reduced Triton-X 100 was followed by EPR.
(10 mM) in MeCN at 355 nm led to an emission band at 466 nm.
4.1. 4-{4-[(2,4,6-Triphenylpyridinio)benzoyloxy]}-2,2,6,6-
tetramethylpiperidine-1-oxyl perchlorate 3
Acknowledgements
A solution of N-(4-carboxyphenyl)-2,4,6-triphenylpyridinium
perchlorate (400 mg, 0.76 mmol), 4-hydroxy-2,2,6,6-tetramethyl-
piperidine-1-oxyl (217 mg, 1.26 mmol), 4-dimethylaminopyridine
(60 mg, 0.49 mmol), and N-(3-dimethylaminopropyl)-N0-ethyl-
carbodiimide hydrochloride (160 mg, 0.83 mmol) in acetonitrile
(20 mL) was warmed to 50 ꢂC for 2 h. The solvent was then rotary
evaporated, to the residue was added diethyl ether (30 mL) and the
precipitate was filtered to give 410 mg (79% yield) of the product,
purified by flash chromatography (silica 60H, acetone as eluent).
The light yellow product 3 melted at 239–240 ꢂC. Anal. Found: C,
68.13; H, 5.72; N, 3.81%; required for C39H38ClN2O7: C, 68.66; H,
5.61; N, 4.11%. IR (KBr) nmax 1719 (CO), 1610, 1600, 1280, 1250, 1090
C.A. thanks PBCT-Conicyt and FONDECYT (#1080234).
References and notes
1. Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluewer Aca-
demic/Plenum: New York, NY, 1999; Chapter 3.
2. Kalai, E.; Hideg, E.; Vass, I.; Hideg, K. Free Radical Biol. Med. 1998, 24, 649–652.
3. Setsukinai, K.; Urano, Y.; Kakinuma, K.; Majima, H. J.; Nagano, T. J. Biol. Chem.
2003, 278, 3170–3175.
4. Setsukinai, K.; Urano, Y.; Kikuchi, K.; Huguchi, T.; Nagano, T. J. Chem. Soc., Perkin
Trans. 2 2000, 2453–2457.
5. Ueno, T.; Urano, Y.; Kojima, Y. H.; Nagano, T. J. Am. Chem. Soc. 2006, 128,
10640–10641.
(broad, ClO4) cmꢁ1
.
6. Aliaga, C.; Aspe´e, A.; Scaiano, J. C. Org. Lett. 2003, 5, 1515–1518.
7. Green, S. A.; Simpson, D. J.; Zhou, G. P.; Ho, S.; Blough, N. V. J. Am. Chem. Soc.
1990, 112, 7337–7346.
8. Herbelin, S. E.; Blough, N. V. J. Phys. Chem. B 1998, 102, 8170–8176.
9. Lozinsky, E.; Shames, A. I.; Likhtenshtein, G. I. Recent Res. Dev. Photochem.
Photobiol. 2001, 5, 41–55.
4.2. 4-{4-[(2,4,6-Triphenylpyridinio)benzamido]}-2,2,6,6-
tetramethylpiperidine-1-oxyl tetrafluoroborate 4
A solution of N-(4-carboxyphenyl)-2,4,6-triphenylpyridinium
fluoroborate (644 mg, 1.25 mmol), 4-amino-2,2,6,6-tetramethylpi-
peridine-1-oxyl (258 mg, 1.42 mmol), 4-dimethylaminopyridine
(178 mg, 2.15 mmol), and N,N0-dicyclohexycarbodiimide (298 mg,
1.45 mmol) in CHCl3 (50 mL) was stirred at 25 ꢂC for 24 h. The
separated solid was filtered off, the filtrate was rotary evaporated,
and diethyl ether was added to the residue to precipitate the light
yellow product that was filtered to give 636 mg (76% yield) of crude
4, purified by column chromatography (silica gel, 60H, acetone), mp
180–183 ꢂC. IR (KBr) nmax 1656 (CO), 1620, 1598, 1551, 1497, 1279,
1240, 1100 (broad, BF4) cmꢁ1. Anal. Found: C, 69.97; H, 5.98; N,
6.98%; required for C39H39BF4N3O2: C, 70.06; H, 5.88; N, 6.29%.
´
10. Aspee, A.; Garcia, O.; Maretti, L.; Sastre, R.; Scaiano, J. C. Macromolecules 2003,
36, 3550–3556.
11. Coenjarts, C.; Garcia, O.; Llauger, L.; Palfreyman, J.; Vinette, A. L.; Scaiano, J. C.
J. Am. Chem. Soc. 2003, 125, 620–621.
12. Goupy, P.; Dufour, C.; Loonis, M.; Dangles, O. J. Agric. Food Chem. 2003, 51,
615–622.
13. Miller, N. J.; Rice-Evans, C. A. Redox Report 1996, 2, 161–171.
14. Aliaga, C.; Lissi, E. A. Can. J. Chem. 2000, 78, 1052–1059.
15. Aliaga, C.; Juac´rez-Ruiz, J. M.; Scaiano, J. C.; Aspec´e, A. Org. Lett. 2008, 10,
2147–2150.
16. Aspe´e, A.; Aliaga, C.; Scaiano, J. C. Photochem. Photobiol. 2007, 83, 481–485.
17. Scaiano, J. C.; Aliaga, C.; Chre
´tien, M. N.; Frenette, M.; Focsaneanu, K. S.;
Mikelsons, L. Pure Appl. Chem. 2005, 77, 1009–1018.
18. Aliaga, C., Rezende, C. M and Arenas, A. Food Chem. doi:10.1016/j.foodchem.
2008.08.075
19. Katritzky, A. R.; Thind, S. S. J. Chem. Soc. Perkin Trans. 1 1980, 1895–1900.
20. Katritzky, A. R.; Lloyd, J. M.; Patel, R. J. Chem. Soc. Perkin Trans. 1 1981, 117–123.
21. Katritzky, A. R.; Rezende, M. C. J. Chem. Res. (S) 1980, 312–313.
22. Kovalska, V. B.; Losytskyy, Y. M.; Kryvorotenko, D. V.; Balanda, A. O.; Tokar, V. P.;
Yarmoluk, S. M. Dyes Pigments 2006, 68, 39–45.
23. Gawinecki, R.; Dobosz, R.; Zakrzewska, A. Dyes Pigments 2004, 60, 143–145.
24. Bajorek, A.; Trzebiatowska, K.; Jedrzejewska, B.; Pietrzak, M.; Gawinecki, R.;
Paczkowski, J. J. Fluoresc. 2004, 14, 295–307.
25. Jedrzejewska, B.; Kabatc, J.; Paczkowski, J. Dyes Pigments 2007, 74, 262–268.
26. Kharlanov, V. A.; Knyazhansky, M. I. J. Photochem. Photobiol. A: Chem. 1999, 125,
21–27.
27. Kharlanov, V. A.; Rettig, W. J. Photochem. Photobiol. A: Chem. 2001, 141, 127–132.
28. Rahman, B.; Kanbara, K.; Akutsu, H.; Yamada, J.; Nakatsuji, S. Polyhedron 2007,
26, 2287–2290.
4.3. Kinetic measurements
The kinetics of the reactions between probes 3 or 4 and a 100-
fold excess TROLOX (6-hydroxy-2,5,7,8-tetramethylchroman-2-
carboxylic acid) was followed under pseudo-first-order conditions,
by monitoring the increase of the fluorescence band of the probe at
466 nm after excitation at 355 nm in acetonitrile. The product band
was identical to the emission of the N-(4-carboxyphenyl)-2,4,6-
triphenylpyridinium fluoroborate, after excitation at 355 nm in
acetonitrile. The observed rate constant was obtained by applying
Eq. 1,
29. Katritzky, A. R.; Krutsikova, A.; Ramsden, C. A.; Lewis, J. Coll. Czechoslov. Chem.
Commun. 1978, 43, 2046–2053.
!
30. Hassner, A.; Alexanian, V. Tetrahedron Lett. 1978, 46, 4475–4478.
31. Aliaga, C.; Stuart, D. R.; Aspe´e, A.; Scaiano, J. C. Org. Lett. 2005, 7, 3665–3668.
32. Okazaki, M.; Kuwata, K. J. Phys. Chem. 1985, 89, 4437–4440.
33. Blough, N. V.; Simpson, D. J. J. Am. Chem. Soc. 1988, 110, 1915–1917.
34. Green, S. A.; Simpson, D. J.; Zhou, G.; Ho, P. S.; Blough, N. V. J. Am. Chem. Soc.
1990, 112, 7337–7346.
IN ꢁ I0
IN ꢁ It
ln
¼ kt
(1)
where IN, I0, and It represent the fluorescence intensities in the
plateau region, initially, and at time t, respectively.
35. Howard, J. A.; Ingold, K. U. Can. J. Chem. 1962, 40, 1851–1864.