Synthesis and properties of a pH-insensitive fluorescent nitric oxide cheletropic trap (FNOCT)

Article abstract of DOI:10.1002/hlca.200690212

The synthesis and properties of the new fluorescent nitric oxide cheletropic trap (FNOCT) 14, designed for the trapping and quantification of nitric oxide (NO) production in chemical and biological systems, is described (Scheme 3). The dicarboxylic acid 1

Full text of DOI:10.1002/hlca.200690212

Helvetica Chimica Acta – Vol. 89 (2006)
2281
Synthesis and Properties of a pH-Insensitive Fluorescent Nitric Oxide
Cheletropic Trap (FNOCT)
by Frank Stefan Horniga), Hans-Gert Kortha), Ursula Rauenb), Herbert de Grootb), and Reiner
Sustmann*a)
a) Institut für Organische Chemie der Universität Duisburg-Essen, Universitätsstr. 5, D-45117 Essen
(phone: +49-201-1833097; fax: +49-201-1834259; e-mail: reiner.sustmann@uni-duisburg-essen.de)
b) Institut für Physiologische Chemie, Universitätsklinikum Essen, D-45122 Essen
In memoriam Hanns Fischer
The synthesis and properties of the new fluorescent nitric oxide cheletropic trap (FNOCT) 14,
designed for the trapping and quantification of nitric oxide (NO) production in chemical and biological
systems, is described (Scheme 3). The dicarboxylic acid 14 and the corresponding bis[(acetyloxy)methyl]
ester derivative 15 of the FNOCT contain a 2-methoxy-substituted phenanthrene group as fluorophoric
unit. The fluorescence of the reduced NO adduct of this FNOCT (lexc 320 nm, lem 380 nm) is pH-inde-
pendent. Trapping experiments were carried out in aqueous buffer solution at pH 7.4with nitric oxide
being added as a bolus as well as being released from the NO donor compound MAHMA NONOate
(=(1Z)-1-{methyl[6-(methylammonio)hexyl]amino}diazen-1-ium-1,2-diolate), indicating a trapping effi-
ciency of ca. 50%. In a biological application, nitric oxide was scavenged from a culture of lipopolysac-
charide-stimulated rat alveolar macrophages. Under the applied conditions, a production of 11.1ꢀ1.5
nmol of NO per hour and per 105 cells was estimated.
Introduction. – Since nitric oxide (NO) has been recognized as an important mes-
senger molecule in mammalian cells, innumerable studies have been published on its
significance in biological systems [1][2]. A highlight in this development was the
Nobel prize in physiology and medicine awarded to Furchgott [3], Murad [4], and
Ignarro [5] in 1997. The study of the physiological and pathophysiological functions
of this simple molecule requires sensitive detection methods. A variety of methods
are known to determine nitric oxide, most of them, however, are difficult to apply in
living systems or are indirect assays, i.e., are based on the identification of metabolites
of nitric oxide [6].
We have been concerned with the development of a reaction for the direct detection
of NO which we discovered in 1992 [7–10]. Although nitric oxide as a radical is rather
unreactive towards organic molecules under normal conditions, it was found that it
reacts readily with o-quinodimethanes (=5,6-dimethylenecyclohexa-1,3-diene) to
form cyclic nitroxide radicals in a kind of cheletropic reaction. This reaction is shown
in Scheme 1 for the molecule first used for this purpose, tetramethyl-o-quinodimethane
1 [7]. EPR Spectroscopy seemed to be the method of choice for the detection of the
thus formed nitroxide 2.
The reactive o-quinodimethanes were named nitric oxide cheletropic traps
(NOCTs). To apply this method to biological systems, a number of problems had to
ꢀ 2006 Verlag Helvetica Chimica Acta AG, Zürich

2282
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 1
be overcome, the most important of them being the thermal stability and water solubil-
ity of the NOCT compounds. Substituted o-quinodimethanes had to be found which are
stable enough to be applied to cellular systems in aqueous solution. According to the
literature, a certain stability can be given to the o-quinodimethane system when typical
rearrangement reactions, in particular 1,5-sigmatropic H-shifts, are suppressed. One
such example has been reported [11]. In our studies, we followed this strategy and
tested whether this stable, but highly reactive compound traps nitric oxide. A successful
trapping experiment was achieved with NOCT derivative 3, substituted by two carboxy
groups to provide sufficient water solubility and which was sufficiently stable. The cor-
responding nitroxide was characterized by EPR spectroscopy [12]. The trapping kinet-
ics, studied by stopped-flow techniques, unveiled a second-order rate constant k2 =5600
M
ꢁ1 sꢁ1 at 228 in buffer solution at pH 7.4. Unfortunately, the lifetime of the produced
nitroxide radical was only 100 s under these conditions, i.e., too short to apply EPR
detection for monitoring NO production in biological systems [13].
A further problem arose when it was realized that nitroxide radicals may be not sta-
ble in a cytosolic medium. We, therefore, turned away from EPR spectroscopy as
detecting method to the more sensitive fluorescence spectroscopy. We thus developed
ꢁfluorescent NOCTsꢂ (FNOCTs) such as 4 (Scheme 2). By employing a phenanthrene-
based o-quinodimethane system, a radical 6 with a fluorophoric phenanthrene moiety
was formed on reaction with nitric oxide. To avoid self-quenching of fluorescence by
the nitroxide radical [14], a reducing agent, e.g., ascorbic acid, is added to generate
the corresponding hydroxylamine derivative 8. After reduction, the phenanthrene fluo-
rescence can easily be detected. This FNOCT was further modified by adding a dime-
thylamino group in 2-position of the phenanthrene ring (! 5) [15]. Thereby, the exci-
tation wavelength for the hydroxylamine product 9 (from 7) is shifted from 315 (8) to
380 nm (9), which is advantageous for applications in cellular systems because short-
wavelength excitation may damage living cells and hampers also the application due
to interference with fluorescence of typical cell constituents, especially NAD(P)H.
A disadvantage of this otherwise perfect FNOCT 5 is the pH dependence of the
fluorescence intensity because of the presence of the protonable dimethylamino
group [15].
Here we report on the synthesis of a corresponding methoxy-substituted FNOCT
which should display similar fluorescence properties as the dimethylamino-substituted
compound and where the fluorescence properties should be pH-independent.
Results and Discussion. – Synthesis of FNOCTs 14 and 15. The 2-methoxyphenan-
threne-9,10-quinone (10) was prepared in analogy to the literature by nitration of phen-
anthrene-9,10-quinone, which yielded a mixture of the 2-nitro (36%) and the 4-nitro

Helvetica Chimica Acta – Vol. 89 (2006)
2283
Scheme 2
isomer (33%). The isomers could be separated due to their different solubility in EtOH
and crystallization of the 2-substituted isomer from AcOH. Treatment of 2-nitrophe-
nanthrene-9,10-quinone with sodium dithionite led to reduction of both the nitro
group and the quinoid system. In situ air oxidation regenerated the quinone system
in 71% yield. Replacement of the NH2 by the OH group via diazotation and subsequent
thermal decomposition of the diazonium salt gave the hydroxy compound (77%).
Reaction with dimethyl sulfate gave 2-methoxyphenanthrene-9,10-quinone (10) in
80% yield.
Condensation of 10 with 1,3-diphenylacetone (=1,3-diphenylpropan-2-one) pro-
vided the corresponding cyclone 11 (yield 77%; Scheme 3). Diels–Alder reaction of
11 with norborn-5-ene-2-endo,3-exo-3-dicarboxylic acid and the corresponding (acety-
loxy)methyl ester yielded the bridged ketone derivatives 12 (67%) and 13 (38%). Com-
pounds 12 and 13 are both mixtures of two diastereoisomers in 1:1 (12) and 4:1 ( 13)
ratio, respectively. Assignment of the isomeric structures was not attempted. The (ace-
tyloxy)methyl ester was prepared, because this type of ester [16][17] is cell-membrane
permeable and is enzymatically hydrolyzed to the corresponding carboxylates [18]. In
principle, the Diels–Alder cycloaddition would provide a mixture of four diaster-
eoisomers. The methoxy group renders phencyclone 11 nonsymmetrical, and both ori-
entations of the diene and dienophile (endo or exo) may be realized. As endo additions
to the norbornene double bond are not known, the number of possible products is
reduced to racemic mixtures of two diastereoisomers of 12. The diastereoisomer mix-
tures 12 and 13 were not separated as their reactivity towards nitric oxide should be
similar. Recently, an X-ray analysis of a related pyrene-based cycloadduct confirmed
1
our H-NMR-based assignment [19].
1H-NMR Spectroscopy supports the assignment of the CO and CH2 bridges of 12 and 13 in an ꢁantiꢂ
position. Due to this configuration, the CH2 group of the norbornane moiety is positioned below the
phenanthrene skeleton. Therefore, the two CH2 protons are differently diamagnetically shielded and
appear at high field as an AB system for both the acid 12 and the ester 13. The two AB systems of the
diastereoisomers are superimposed. In case of acid 12, one of these protons is found at d ꢁ0.47, the
other at d 0.52. The low-field signal comprises two overlapping d with 3J=11.0 and 11.2 Hz, respectively.
The high-field signal appears as a slightly broadened d.
o-Quinodimethanes are normally unstable, reactive molecules [20]. A compound
where the methylene C-atoms of the o-quinodimethane moiety are bound to a norbor-

2284
Helvetica Chimica Acta – Vol. 89 (2006)
Scheme 3
nane system, forming a six-membered ring, proved to be isolable, though remained very
reactive [11]. Photochemically induced removal of carbon monoxide from 12 and 13
makes such bridged o-quinodimethanes accessible (Scheme 3). It was presumed that
compounds 14 and 15 would be isolable, so that they might be used as FNOCTs. To
get FNOCTs which can be applied in biological situations, carbon monoxide extrusion
has to be essentially quantitative, otherwise the residual ketone derivative gives rise to
background fluorescence similar to that of the nitric oxide trapping product (see
below). This situation required optimization of the photolysis conditions. The photoly-
sis was carried out in a Rayonet photochemical reactor. Two wavelengths, 254and 300
nm, were tested, of which only the longer-wavelength irradiation led to the desired
product. In addition, formation of fluorescent by-products made it necessary to opti-
mize the temperature of photolysis and the solvent. Therefore, only small quantities
of ca. 20 mg were photolyzed under Ar in a quartz NMR tube which was immersed
in a thermostated methanol bath (highest commercial purity). THF was found to be
the solvent of choice, but it had to be absolutely dry to avoid side reactions. Deuterated
THF was used to allow monitoring of the progress of the reaction by 1H-NMR spectros-
copy. The optimum photolysis temperature was ꢁ108. In this way, conditions for tem-
perature and reaction time were established. According to 500-MHz 1H-NMR spectros-
copy, the purity of the FNOCTS was ꢂ95%. Attempts to further purify products 14 and
15 after photolysis by chromatography or recrystallization were unsuccessful because of
partial decomposition. Identification and characterization of 14 and 15 was carried out

Helvetica Chimica Acta – Vol. 89 (2006)
2285
by 1H- and 13C-NMR and by high-resolution mass spectrometry (ESI-TOF) (see Exper.
Part).
For the purpose of further studies, in particular nitric oxide trapping experiments,
the solvent of the photolysis solution was removed in vacuo and the dry, orange-red res-
idue was stored under Ar at ꢁ208.
UV/VIS and Fluorescence Spectroscopy. FNOCT 14 displays the typical absorption
spectrum of o-quinodimethanes. Fig. 1 shows the long-wavelength absorption band
with a maximum at lmax 445 nm and log e=3.63. This absorption disappears readily
on reaction with nitric oxide indicating conversion of the quinoid p-system to the aro-
matic phenanthrene system. The disappearance of this band was monitored in stopped-
flow kinetic measurements of the reaction of FNOCT with nitric oxide (see below).
Fig. 1. UV/VIS Spectrum of o-quinodimethane derivative 14 in THF. [14]=50 mM, T 208.
In previous studies on phenanthrene-based FNOCTs, it was observed that the
FNOCT with an unsubstituted phenanthrene ring did not fluoresce [21], whereas
FNOCT 5 carrying a dimethylamino group in 2-position of the fluorophoric unit dis-
played a fluorescence at a longer wavelength, viz. lem 600 nm (lexc 450 nm) [15], than
the NO-trapping product. The intensity of this fluorescence is an order of magnitude
lower than that of the parent ketone or the reduced NO-trapping product. Similarly,
a weak fluorescence was also found for the methoxy-substituted FNOCTs 14 and 15
in THF solution, as shown in Fig. 2 for 14. The spectra of both the acid and the (ace-
tyloxy)methyl ester were essentially identical.
NOTrapping and EPR Spectroscopy. The reaction of FNOCTs 14 and 15 with NO
to give nitroxides 16 and 17, respectively (Scheme 4) was studied by EPR spectroscopy.
For this purpose, a 4m M solution of the FNOCT in THF, freshly prepared by photolysis
of the corresponding bridged ketone derivative, was treated with an excess of NO by
adding an NO-saturated THF solution. The solutions were made oxygen-free to
avoid autoxidation of nitric oxide.
The EPR spectra of 16 and 17 were identical. Fig. 3 shows the spectrum of nitroxide
16 as produced from acid 14. The EPR spectrum is characterized by a large N hyperfine
splitting constant of a(14N)=2.284mTand a g value of 2.00634, both very characteristic

2286
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 2. Excitation and emission spectrum of 14 in THF. [14]=50 mM, lexc 470 nm, lem 580 nm.
Scheme 4
for bridged nitroxides [22]. The N hyperfine lines show an additional q splitting of
a(H)=0.074mT for three equivalent H-atoms. These splittings are attributed to an
interaction of the unpaired spin with norbornane H-atoms, the two bridgehead g-pro-
tons and one proton in e-position. The accidentally identical splitting constants indicate
a double W-arrangement of the e-proton relative to the unpaired electron at the nitro-
xide group. This supports structure 20 (Scheme 4) for the nitroxide radical 16 with the
nitroxide group and the CH2 bridge in ꢁantiꢂ position and establishes the attack of NO at
the less hindered side of the FNOCT plane.

Helvetica Chimica Acta – Vol. 89 (2006)
2287
Fig. 3. Experimental (top) and simulated (bottom) EPR spectrum of nitroxide radical 16 in THF.
g=2.00634, a(14N)=2.284, a(Hg)=0.074, a(He)=0.074mT; T 238.
Fluorescence Spectroscopy of the Reduced NO-Trapping Products 18 and 19. To
overcome the self-quenching of the fluorescence of the nitroxide, NO addition (satu-
rated THF solution) was carried out in the presence of a slight molar excess of ascorbic
acid as reducing agent. By this means radicals 16 and 17 were rapidly reduced to the
corresponding hydroxylamine derivatives 18 and 19, respectively. As expected, the
fluorescence spectrum of hydroxylamine derivative 18, recorded in phosphate buffer
solution at pH 7.2 (Fig. 4), is very similar to that of the parent bridged ketone derivative
12 (spectrum not shown) prior to photolysis to FNOCT 14.
Two excitation wavelengths were probed, lexc 320 and 350 nm, giving rise to the
same emission spectrum (Fig. 4, upper and lower trace, resp.). Compared to the 2-
(dimethylamino)-substituted FNOCT 5 and its NO trapping product 9, the excitation
and emission maxima are located at shorter wavelengths. This suggests that biological
applications might be hampered by the autofluorescence of cell constituents, in partic-
ular by that of NAD(P)H.
Stopped-Flow Measurements. The rate of reaction of FNOCT 14 with nitric oxide
was determined by stopped-flow UV measurements in THF, following the decay of
the maximum of the o-quinoid absorption at 445 nm. The measurements were carried
out at 218 under pseudo-first-order conditions by using a 10-to-100-fold excess of nitric
oxide. A second-order rate constant k2 =750ꢀ11 l molꢁ1sꢁ1 was obtained. This magni-
tude is comparable to that found for the 2-(dimethylamino)-substituted phenanthrene
FNOCT [15]. Similar to the previous FNOCTs it is expected that the rate constant in
aqueous buffer will be somewhat lower due to solvation effects. Although the rate is
several powers of ten lower than the value for reaction of NO with biological molecules
like heme systems, it will be sufficient for the detection of NO in living systems, as has
been demonstrated before [23]. Moreover, a partial trapping of NO may even be advan-
tageous as only a small perturbation of the cellular function of NO will be exerted.

2288
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 4. Excitation (dashed line) and emission spectrum (solid lines) of hydroxylamine derivative 18
from reaction of 14 with NOin the presence of ascorbic acid in phosphate buffer pH 7.20. [phosphate
buffer]=50 mM, [ascorbic acid]=100 mM, [14]=50 mM. Upper and lower trace of emission spectrum
from excitation at lexc 320 and 350 nm, respectively; lem 385 nm.
Properties and Applications of FNOCTs 14 and 15. FNOCTs 14 and 15 were con-
ceived to eliminate the pH dependence of the fluorescence intensity of the (dimethyl-
amino)-substituted FNOCT 5 [15]. The pH insensitivity of the fluorescence was estab-
lished in oxygen-free buffer solution at pH 5.6 (acetate buffer), 7.35 (phosphate buffer),
and 9.0 (borate/KCl/NaOH buffer) with the FNOCT-precursor ketone 12 since its fluo-
rescence characteristics is identical to that of the NO-trapping product 18. Under stan-
dard settings of the fluorimeter (see Exper. Part), the fluorescence intensities from 50
mM solutions was identical at all the three pH values (data not shown).
To check the trapping efficiency for NO, 50 mM solutions of FNOCT 14 in oxygen-
free phosphate buffer pH 7.35 were mixed with increasing aliquots of a saturated (ca. 2
mM) aqueous NO solution. The upper trace in Fig. 5 displays the dependence of the
fluorescence intensity on the concentration of NO in the absence of oxygen, measured
15 min after addition of the NO stock solution. Saturation of the fluorescence intensity,
corresponding to a complete consumption of FNOCT 14, is reached at an approxi-
mately three-fold excess of NO. Hence, under the employed conditions, side reactions
seem to take place which lead to a reduced trapping efficiency for NO. If FNOCT and
NO are present in equimolar amounts ca. 50% of the added NO is trapped. This
amount decreases even further if the trapping experiment is carried out in air-saturated
buffer (Fig. 5, lower trace). A reduced trapping efficiency in the presence of oxygen has
been found for other FNOCTs as well and reflects the competing autoxidation of NO.
In addition to the study of NO trapping with NO being added as a bolus, we assessed
the trapping of continuously generated NO by utilizing the NO donor MAHMA
NONOate (=(1Z)-1-{methyl[6-(methylammonio)hexyl]amino}diazen-1-ium-1,2-dio-
late=6-(2-hydroxy-1-methyl-2-nitrosohydrazino)-N-methylhexan-1-amine). With re-
gard to the envisaged studies in biological systems, these measurements were per-
formed at 378 in air-saturated phosphate buffer (50 mM, pH 7.4) in the presence of
100 mM of the iron chelator diethylenetriaminepentaacetic acid (=N,N-bis{2-[bis(car-

Helvetica Chimica Acta – Vol. 89 (2006)
2289
Fig. 5. NOTrapping by FNOCT 14 as a function of added NOat pH 7.35. [14]=50 mM, [phosphate
*
buffer]=50 mM, [ascorbic acid]=100 mM; oxygen-free buffer, ¤ air-saturated buffer.
boxymethyl)amino]ethyl}glycine; dtpa) and the reductant 6-hydroxy-2,5,7,8-tetrame-
thylchroman-2-carboxylic acid (=3,4-dihydro-6-hydroxy-2,5,7,8-tetramethyl-2H-1-
benzopyran-2-carboxylic acid=Trolox), a water-soluble analogue of vitamin E, and
with a lower concentration of FNOCT 14, i.e., 10 mM (see below). The time evolution
of the fluorescence of the reduced NO adduct 18 was followed for a range (0–100
mM) of MAHMA NONOate concentrations (Fig. 6). For all concentrations of
MAHMA NONOate, the time dependence of fluorescence obeys a clean first-order
rate law, reflecting the rate of decomposition of MAHMA NONOate. From the
least-squares fits displayed in Fig. 6, an average rate constant for NO release from
MAHMA NONOate of k310K =(7.8ꢀ0.8)·10ꢁ3 sꢁ1 as evaluated, corresponding to a
half-life of 1.5 min under these conditions. This value agrees well with the half-life of
MAHMA NONOate of 2.7 min at 228 as given in the literature [24]. Hence, FNOCTs
14 and 15 can advantageously be used for kinetic studies of NO-releasing compounds.
As in the experiments with NO solution, fluorescence intensity (as determined after
complete decomposition of the NO donor after >15 min) increased with increasing
MAHMA NONOate concentrations (Fig. 7). Again, saturation was observed, albeit
a larger excess of NO was required to reach saturation: even with 40 mM MAHMA
NONOate, i.e., an eightfold excess of NO, saturation was not yet complete.
As to be expected, the trapping efficiency increased with increasing concentrations
of FNOCT 14 (Fig. 8). Maximal trapping of NO generated by 0.5–5 mM MAHMA
NONOate (=1–10 mM NO) was observed with 50 mM FNOCT 14, at higher concentra-
tions of the trap, i.e., 100 mM, the fluorescence intensity was slightly decreased, probably
due to photophysical quenching effects.
With 10 mM FNOCT 14, i.e., with the concentration of 14 later applied in a cellular
system (see below), the NO release by 0.2–2 mM MAHMA NONOate (=0.4–4 mM
NO) could be readily measured with a microplate reader as often used for biological
applications (sample volume 300 ml); monitoring of lower NO levels proved to be
too strongly obscured by background fluorescence to allow reliable measurements.

2290
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 6. Time evolution of NOrelease from MAHMA NONOate as monitored by trapping with
FNOCT 14 at pH 7.4. [14]=10 mM, [phosphate buffer]=50 mM, [Trolox]=100 mM, [dtpa]=100 mM; lexc
320 nm, lem 380 nm; T 378. The arrow marks the point of addition of MAHMA NONOate; numbers
indicate the initial concentration of MAHMA NONOate. Solid lines are least-squares fits to a first-
order rate law.
Fig. 7. Efficiency of NOtrapping by FNOCT 14 at pH 7.4 as a function of the concentration of the
NOdonor
MAHMA NONOate. [14]=10 mM, [phosphate buffer]=50 mM, [Trolox]=100 mM,
[dtpa]=100 mM; lexc 320 nm, lem 380 nm; T 378.
For 0.2–0.6 mM MAHMA NONOate, trapping efficiency was above 50% (up to 65%)
and with 1–2 mM MAHMA NONOate between 40 and 50%.
Trapping of NOProduced from Isolated Macrophages. As an example of NO-pro-
ducing cells, cultured rat alveolar macrophages, i.e., nonspecific immune cells isolated
from the rat lung, were used. These cells can be activated in vitro by bacterial lipopo-
lysaccharides (LPS) to express the enzyme inducible nitric oxide synthase (iNOS) and
then to produce considerable amounts of NO. Unfortunately, FNOCT 14 displayed
cytotoxicity when applied to the alveolar macrophages in the optimal trapping concen-

Helvetica Chimica Acta – Vol. 89 (2006)
2291
Fig. 8. Efficiency of NOtrapping at pH 7.4 as a function of the concentration of FNOCT
14.
[MAHMA NONOate]=1 mM, [phosphate buffer]=50 mM, [Trolox]=100 mM, [dtpa]=100 mM; lexc 320
nm, lem 380 nm; T 378.
tration of 50 mM (see above). Therefore, a concentration of 10 mM FNOCT 14 was used
in all cellular applications. At this concentration, no significant cell death was observed
within the duration of the experiments. As a reductant to be used in the cellular system,
the H2O-soluble vitamin-E analogue Trolox proved to be advantageous because reduc-
tion of nitroxide radical 16 to the fluorescent hydroxylamine 18 was found to be much
more rapid with Trolox (100 mM) than with glucose (10 mM), the reductant used in pre-
vious biological studies [23]. Ascorbate, the preferred reductant in most of the chemical
experiments (see above), was found to increase the background fluorescence over time,
thus turned out to be unfavorable for cellular systems where NO is produced rather
slowly and at low levels, requiring prolonged incubation periods for NO trapping.
The (extracellular) iron chelator diethylentriaminepentaacetic acid (dtpa, 100 mM)
was found to suppress the slow increase of the background fluorescence observed dur-
ing prolonged incubation periods and was, therefore, added to all buffers used for cel-
lular measurements. Under these conditions, a strong increase in fluorescence intensity
could be observed in the supernatant of LPS-activated alveolar macrophages following
1 h of incubation with FNOCT 14. This increase in fluorescence intensity was undoubt-
edly due to trapping of NO, as it was not observed in the supernatant of LPS-activated
cells incubated with FNOCT 14 in the presence of the NOS inhibitor NG-monomethyl-
L-arginine (=N5-[imino(methylamino)methyl]-L-ornithine; NMA). Similarly, nonacti-
vated cells did not increase fluorescence intensity to a considerable extent. Additional
experiments were performed in which a large excess of MAHMA NONOate (100 mM)
was added to the supernatant samples after cellular NO production had been meas-
ured. Comparison of the thus obtained fluorescence intensities with those from similar
experiments performed with cell lysates (cells were lysed with sodium dodecyl sulfate
(SDS), final concentration 0.4%, at the end of the experiments) revealed that after 1 h
of incubation, 32ꢀ4% of the applied FNOCT 14 was strongly associated with the cells,
i.e., incorporated in the cells and/or bound to cellular membranes. The data shown in
Fig. 9 are total fluorescence, i.e., fluorescence from the supernatant plus fluorescence

2292
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 9. Trapping of NOproduced by rat alveolar macrophages. LPS+Arg: Cells activated with LPS
(0.5 mg/ml, 18 h) in the presence of L-arginine (0.5 mM); LPS+NMA: Cells activated with LPS (0.5
mg/ml, 18 hrs) in the presence of the NO-synthase inhibitor NMA (0.1 mM); ꢁLPS+Arg: Cells incu-
bated with L-arginine (0.5 mM), no pretreatment with LPS; ꢁLPS+NMA: Cells incubated with
NMA (0.1 mM), no pretreatment with LPS. All experiments were performed in Krebs–Henseleit buf-
fer, [D-glucose]=10 mM, [dtpa]=100 mM, [Trolox]=100 mM, [14]=10 mM; incubation time 1 h, T 378.
Fluorescence intensity (lexc 320, lem 380 nm) shown is the sum of the fluorescence from the superna-
tant and cell-associated fluorescence as determined in the cell lysates (corrected for volume, back-
ground, and cell density).
from the lysate, corrected for background fluorescence, volume, and cell density. The
increase in fluorescence due to NO production from activated cells (in the absence
of the NOS inhibitor) was observed in the supernatant as well as in the cell lysate.
Estimation of the amount of NO produced by the cells by using the calibration
curve shown in Fig. 7 yielded a NO production of 11.1ꢀ1.5 nmol hꢁ1 (105 cells)ꢁ1 in
LPS-activated, noninhibited cells, 1.3ꢀ0.1 nmol hꢁ1 (105 cells)ꢁ1 in LPS-activated,
NOS-inhibited cells, 1.6ꢀ0.5 nmol hꢁ1 (105 cells)ꢁ1 in nonactivated cells, and 1.8ꢀ0.3
nmol hꢁ1 (105 cells)ꢁ1 in nonactivated, NOS-inhibited cells. The NO produced in the
activated, non-inhibited cells was trapped to ca. 30% in the cells and to ca. 70% in
the cellular supernatant, a finding that fits well with the distribution of the NO trap
(see above).
The foregoing experiments demonstrate that FNOCT 14 can be successfully used in
cellular/biological systems to trap endogenously produced NO and to estimate cellular
NO production, although there are still some limitations with regard to sensitivity and
toxicity (use of higher concentrations of the trap would be desirable to achieve higher
trapping efficiencies).
Biological applications of the (acetyloxy)methyl ester FNOCT 15 are currently in
progress.
The excellent technical assistance of Ms N. Boschenkov is gratefully acknowledged.

Helvetica Chimica Acta – Vol. 89 (2006)
2293
Experimental Part
General. The following compounds were prepared according to published procedures: 2-nitrophen-
anthrene-9,10-quinone [25], 2-aminophenanthrene-9,10-quinone [26], 2-hydroxyphenanthrene-9,10-qui-
none [27], 2-methoxyphenanthrene-9,10-quinone (10) [28], chloromethyl acetate [29], dimethyl norborn-
5-ene-2-endo,3-exo-dicarboxylate (=dimethyl bicyclo[2.2.1]hept-5-ene-2-endo,3-exo-dicarboxylate)
[30], norborn-5-ene-2-endo,3-exo-dicarboxylic acid (=bicyclo[2.2.1]hept-5-ene-2-endo,3-exo-dicarbox-
ylic acid) [31]. M.p.: uncorrected. UV/VIS: Varian Cary 300 Bio; lmax [nm] (log e). Fluorescence:
J&M FL3095. Fluorescence microplate reader: BMG Labtech FLUOstar Optima. Stopped flow:
HITech-Scientific Ltd SF41. IR: Bio-Rad Series FTS 135 FT-IR; in cmꢁ1 1H- and 13C-NMR Spectra:
.
Bruker DRX 500 and Varian Gemini 200; d in ppm, I in Hz; an asterisk (*) denotes signals of a second
isomer. EPR: Bruker ESP 300E. MS: Bruker BioTof II. Elemental analyses: Elektrothermal 9100.
5-Methoxy-1,3-diphenyl-2H-cyclopenta[l]phenanthren-2-one (11). A mixture of 10 (1.50 g, 6.30
mmol) and 1,3-diphenylpropan-2-one (1.30 g, 6.18 mmol) in EtOH (100 ml) was heated to 358. To this
suspension was added within 30 min 3.5M KOH in EtOH (20 ml). Five minutes after complete addition,
the black precipitate was filtered and washed once with a small amount of EtOH: 2.00 g (77% of 11).
M.p. 2378. UV (THF): 632 (3.19), 499 (2.99), 442 (3.30), 305 (4.75), 264 (4.68). IR (KBr): 3070, 3002,
2900, 2830, 1693, 1599. 1H-NMR (500 MHz, CDCl3): 7.74( d, 3J=7.57, 1 H); 7.67 (d, 3J=8.51, 1 H);
7.45 (d, 3J=7.73, 1 H); 7.46–7.27 (m, 10 H); 7.21 (t, 3J=7.88, 1 H); 7.00 (d, 4J=2.37, 1 H); 6.85 (t,
3
4
3J=7.72, 1 H); 6.80 (dd, J=8.52, J=2.22, 1 H); 3.33 (s, MeO). 13C-NMR (125 MHz, CDCl3): 200.3;
159.1; 148.5; 148.3; 137.7; 133.8; 131.4; 125.8; 119.3; 119.0; 107.6; 111.9; 128.5; 128.1; 127.7; 127.3;
126.4; 126.7; 127.2; 54.7 (MeO). HR-ESI-MS: 413.1567 (M+, C30
N
N
Bicyclo[2.2.1]hept-5-ene-2-endo,3-exo-dicarboxylic Acid Bis[(acetyloxy)methyl] Ester. To norborn-5-
ene-2,3-dicarboxylic acid (4.55 g, 25.0 mmol) in dimethylformamide (50 ml), dried by filtration through
neutral Al2O3, was added in 10 portions NaH suspension in paraffine (2.40 g, 100 mmol). Within 30 min,
chloromethyl acetate (5.42 g, 50.0 mmol), kept at ꢁ108, was added dropwise into this soln. After 3 h at
408, the mixture was freed of surplus NaH by filtration. After evaporation of the solvent at r.t./10 Pa, the
1
residue was bulb-to-bulb distilled at 1308/10 Pa: 3.83 g (47%). H-NMR (200 MHz, CDCl3): 6.31 (m, 1
3
H); 6.05 (m, 1 H); 5.85–5.30 (m, 2 OCH2O); 3.45 (t, J=4.2, 1 H), 3.32 (br. s, 1 H); 3.15 (br. s, 1 H);
2.74( dd, 3J=4.3, 1 H); 2.14–2.05 (d, 2 Me); 1.70–1.49 (m, 2 H).
9,9a,10,11,12,13,13a,14-Octahydro-2-methoxy-15-oxo-9,14-diphenyl-9,14:10,13-dimethanobenzol[b]-
triphenylene-11,12-dicarboxylic Acid (12). A suspension of 11 (1.00 g, 2.42 mmol) and bicyclo[2.2.1]hept-
5-ene-2-endo,3-exo-dicarboxylic acid (0.459 g, 2.52 mmol; ca. 4% excess) in chlorobenzene (25 ml) was
refluxed for 8 h in the dark under Ar. After filtration at r.t., the residue was dissolved in THF (ca. 130
ml) at slightly elevated temp. To the soln., charcoal (0.3 g) was added. After stirring for 30 min, the char-
coal was filtered off, the solvent evaporated, and the product precipitated by addition of a 10-fold excess
of heptane: 0.95 g (67%) of 12 as a 1:1 diastereoisomer mixture. M.p. 2938. UV (THF): 350 (2.78), 343
(3.00), 313 (3.94), 267 (4.77). IR (KBr): 3318, 3063, 3028, 2835, 2968, 2911, 1768, 1736, 1703, 1615, 1173,
1
3
1093. H-NMR (500 MHz, (D8)THF): 11.29 (br. s, 2 H, COOH); 8.67 (t, J=9.0, 1 H); 8.19–7.89 (dd,
3J=7.89, 1 H); 7.71–7.57 (m, 2 H); 7.48–7.06 (m, 11 H); 6.63 (d, 4J=3.0, 1 H); 3.27, 3.26* (s, 3 H,
2
MeO); 3.26–3.23 (m, 2 H); 3.06 (m, 1 H); 3.23–3.13 (m, 2 H); 2.95–2.83 (m, 1 H); 0.52 (dd, J=11.8,
1 H); ꢁ0.47 (dd, 3J=11.5, 1 H). 13C-NMR (125 MHz, (D8)THF): 199.0; 174.8; 174.1; 158.9; 138.8;
138.5; 138.4*; 138.1; 136.3; 136.1*; 135.0; 134.8*; 132.8; 132.6*; 129,6*; 127,2; 126.2; 130.1; 130.0;
129.9*; 129.8; 129.4; 129.3*; 129.1; 129.0*; 128.9; 128.8*; 128.7; 127.8; 127.7*; 126.1; 126.0*; 125.9;
125.8*; 125.7; 123.8; 118.6; 118.5*; 106.1; 106.0*; 64.6; 64.4; 64.1; 64.0; 54.8 (MeO); 51.9; 51.8; 51.2;
48.4; 48.2*; 45.5; 44.4; 44.3; 43.1; 34.7 (C(10ꢁ)). HR-ESI-MS: 617.1945 ([M+Na]+, C39
617.1935).
A
N
9,9a,10,11,12,13,13a,14-Octahydro-2-meethoxy-15-oxo-9,14-diphenyl-9,14:10,13-dimethanoben-
zo[b]triphenylene-11,12-dicarboxylic Acid Bis[(acetyloxy)methyl] Ester (13). Under Ar and in the dark,
11 (200 mg, 0.48 mmol) and bicyclo[2.2.1]hept-5-ene-2-endo,3-exo-dicarboxylic acid bis[(acetyloxy)-
methyl] ester (158.2 mg, 0.48 mmol) were heated under reflux in chlorobenzene (5 ml) for 4 h. A residual
black solid was filtered off and discarded. Chlorobenzene was evaporated. The remaining brownish oil
was dissolved in THF (10 ml) and the product was precipitated from the soln. by the slow addition of

2294
Helvetica Chimica Acta – Vol. 89 (2006)
heptane (100 ml). The yellow suspension was stirred for 5 h, after which the product was isolated by fil-
tration: 0.203 g (58%) of 13, 4:1 diastereoisomer mixture (*: minor component). M.p. 209 8. UV (THF):
361 (3.13), 344 (3.18), 313 (3.93), 266 (4.69). IR (KBr): 3056, 3028, 2835, 2968, 2927, 1768, 1732, 1614,
1
3
3
1170. H-NMR (500 MHz, (D8)THF): 8.68 (d, J=8.1, 1 H); 8.66 (d, J=7.7,1 H); 8.05–7.87 (m, 2 H);
3
7.75–7.64( m, 2 H); 7.49–7.07 (m, 10 H); 6.61, 6.57* (d, J=3.0, 1 H); 5.96–5.90* (m, 1 H, OCH2O);
5.68–5.61 (m, 2 H, OCH2O); 3.31–3.25 (m, 3 H); 3.28, 3.26* (s, 3 H, MeO); 3.25, 3.23* (d, J=4.1, 1
3
H); 3.05, 2.98* (d, 3J=3.8, 1 H); 2.86, 2.82* (s, 1 H); 2.01, 1.98*, 1.97*, 1.95 (s, 6 H, Me); 0.52 (d,
2J=11.7, 1 H); ꢁ0.43 (d, 2J=10.1, 1 H). 13C-NMR (125 MHz, (D8)THF): 198.7; 172.4; 171.6; 169.8;
169.5; 159.03; 158.99*; 138.5; 138.2*; 138.1*; 137.8; 136.2*; 135.9; 135.0; 134.7*; 132.8*; 132.7; 130.2;
130.06*; 130.05; 130.1*; 130.0; 129.5*; 129.4; 129.2; 129.1*; 129.0; 128.8; 128.7; 128.6; 128.42; 128.37;
128.32; 128.27; 127.8*; 127.6; 127.3; 126.2*; 126.0; 125.9*; 125.8; 123.9; 118.6; 106.0; 80.59; 80.62;
80.68; 80.75 (OCH2O); 64.0; 64.2; 64.3; 64.5; 54.9 (MeO); 51.7*; 51.1; 48.1*; 47.9; 45.2; 44.4; 44.3*;
43.2; 20.5; 20.3* (Me). HR-ESI-MS: 761.2356 ([M+Na]+, C45
N
N
9,9a,10,11,12,13,13a-Octahydro-2-methoxy-9,14-diphenyl-10,13-methanobenzo[b]triphenylene-11,12-
dicarboxylic Acid (14). Acid 12 (22.7 mg, 3.18·10ꢁ2 mmol) was dissolved in (D8)THF (1.0 ml) in a quartz
NMR tube. Photolysis was carried out in a Rayonet photochemical reactor at 300 nm and at ꢁ108. After
24h, the conversion of 12 was 100% (checked by 500-MHz 1H-NMR). The orange soln. was transferred
to a Schlenk vessel and the solvent evaporated at 10 Pa. The orange 14, a 1:1 diastereoisomer mixture,
was stored in a freezer (ꢁ208) or at ꢁ788 (dry ice). M.p. 160–165. UV (THF): 214 (4.51), 245 (4.50), 311
(4.12), 449 (3.71). IR (KBr): 3317 (OH), 2821 (MeO), 1731, 1704, 1599, 1179, 942. 1H-NMR (500 MHz,
(D8)THF): 7.65 (d, 3J=8.5, 1 H); 7.48 (d, 3J=8.5, 1 H); 7.44–7.37 (d, 3J=8.7, 1 H); 7.41 (d, 3J=7.1, 1 H);
7.36 (d, 3J=7.2, 1 H); 7.24–7.03 (m, 8 H); 6.71–6.64( m, 3 H); 3.21, 3.18* (s, 3 H, MeO); 3.31–3.22 (m, 3
H); 2.97–2.94( m, 1 H); 2.86–2.84( m, 1 H); 2.74( d, J=3.6, 1 H); 2.14( m, 1 H); 1.60 (dd, 3J=10.5, 1 H).
13C-NMR (125 MHz, (D8)THF): 175.4; 175.3; 159.1; 159.2*; 144.4; 144.2*; 143.3*; 143.2; 135.7*; 135.5;
135.4; 135.3; 135.2; 135.1*; 133.90; 133.86*; 133.2; 132.8; 132.5; 130.8*; 130.7; 130.6*; 130.5; 129.7*;
129.6; 129.2; 129.0*; 128.4; 128.3; 128.1; 126.3; 126.2*; 125.4; 123.6; 116.5; 115.7; 68.2; 67.9; 53.0
(MeO); 54.9*; 54.7; 52.4*; 52.3; 50.7*; 50.5; 48.6; 48.2*; 47.5; 34.9*; 34.75. HR-ESI-MS: 566.2046 (M+,
C
N
G
11,12-dicarboxylic Acid Bis[(acetyloxy)methyl] Ester (15). As described for 14, with 13 (20.2 mg,
2.78·10ꢁ2 mmol) in (D8)THF (1.0 ml) for 18 h. The orange 15, a 4:1 diastereoisomer mixture (*:
minor component) was stored in a freezer (ꢁ208) or at ꢁ788 (dry ice). M.p. 75–778. UV (THF): 246
1
(4.63), 309 (4.24), 446 (3.78). IR (KBr): 3055, 3024, 2835, 2968, 2927, 1745, 1613, 1163. H-NMR (500
MHz, (D8)THF): 7.66–7.63 (m, 2 H); 7.48–7.44 (d, 3J=8.5, 1 H); 7.44–7.39 (d, 3J=8.4, 1 H);
7.36–7.29 (m, 3 H); 7.26–7.16 (m, 4H); 7.09–7.02 ( m, 2 H); 6.73–6.65 (s, 2 H); 6.64, 6.63* (s, 1 H);
5.82–5.58 (m, 4H, OCH 2O); 3.36–3.32 (m, 1 H); 3.32–3.25 (m, 3 H); 3.18, 3.23* (s, 3 H, MeO);
3
2
3.01–3.03 (m, 1 H); 2.79 (d, J=4.1, 1 H); 2.17 (d, J=10.6, 1 H); 2.03, 2.01*, 1.99, 1.97* (s, 6 H, Me);
1.60 (d, J=10.9, 1 H). 13C-NMR (125 MHz, (D8)THF): 172.7; 172.0; 169.5; 169.4; 159.2*; 159.1; 144.2;
2
144.1*; 143.1; 135.2; 134.9; 134.5; 133.6; 133.2; 132.8; 132.6*; 130.74; 130.67; 130.57*; 130.3; 129.8;
129.7*; 129.4; 128.5; 128.3*; 126.4; 126.3*; 125.4; 123.7; 116.6; 115.8; 115.7; 80.5; 80.4* (OCH O; 68.0;
2
67.9 (C(3ꢁ), C(6ꢁ); 54.9; 54.8 (MeO); 52.1; 52.0; 50.4*; 50.3; 49.8*; 49.7; 48.5; 48.0; 34.7; 20.4; 20.3*
(Me). HR-ESI-MS: 733.2420 ([M+Na]+, C44
N
N
Fluorescence Measurements. Fluorescence spectra were recorded at r.t. in 1-cm cuvettes. Stock solns.
of the compounds (5 mM) in oxygen-free DMSO were added to the buffer to give a 50 mM final concen-
tration.
ESR Measurements. X-Band ESR spectra were recorded in Ar-flushed, septum-capped 4-mm o.d.
quartz tubes by injecting 30 ml of a sat. NO soln. in THF (30 ml) to 1 mM FNOCT in oxygen-free THF.
Microwave frequency 9.48 GHz, microwave power 2 mW, modulation amplitude 0.03 mT, sweep range
7 mT, sweep time 15 min. Spectra simulation was performed by the WinSim program [32].
NO Trapping from MAHMA NONOate. These experiments were performed in nondeoxygenated
sodium phosphate buffer (50 mM, pH 7.4) supplemented with 100 mM dtpa, 100 mM Trolox, and 1–100
mM FNOCT 14 at 378. Fluorescence intensity at lexc 320 nm, lem 380 nm was followed over time (sampling
interval 60–120 s) using a fluorescence microplate reader. After establishing baseline fluorescence,

Helvetica Chimica Acta – Vol. 89 (2006)
2295
MAHMA NONOate (20 nm–100 mM) was added from a freshly prepared stock soln. (in dilute NaOH).
Fluorescence given in the figures was assessed after a plateau was reached following complete decompo-
sition of the MAHMA NONOate and was corrected for background fluorescence (same buffer with the
same additives and the same FNOCT concentration without added MAHMA NONOate).
Cell Isolation and Cell Culture. Alveolar macrophages were isolated from male Wistar rats (350–380
g) by bronchoalveolar lavage as described previously [33]. The cells were seeded on to two 4-well cell
culture plates (Nunc) and cultured in Dulbeccoꢀs modified Eagles medium (DMEM) supplemented
with L-glutamine (2 mM) and gentamicin (50 mg/ml). One hour after seeding, cells were washed with
warm Hanksꢂ balanced salt soln. (HBSS; 137 mM NaCl, 5.4m M KCl, 1.0 mM CaCl2, 0.5 mM MgCl2, 0.4
mM KH2PO4, 0.3 mM Na2HPO4, 25 mM Hepes (=4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid),
pH 7.4) and supplied with fresh medium (DMEM) supplemented with L-glutamine (2 mM), gentamicin
(50 mg/ml), and 10% heat-inactivated fetal calf serum.
NOTrapping from Alveolar Macrophages. Four to five hours after cell isolation and 18 h prior to the
experiments, cells in half of the wells were primed for NO production by the addition of lipopolysacchar-
ides (LPS; 0.5 mg/ml) for 18 h. For experiments, cells were washed with HBSS, counted, and supplied with
Krebs–Henseleit buffer (115 mM NaCl, 25 mM NaHCO3, 5.9 mM KCl, 1.2 mM MgCl2, 1.2 mM NaH2PO4,
1.2 mM Na2SO4, 2.5 mM CaCl2, 20 mM Hepes, pH 7.4) supplemented with 10 mM D-glucose, 100 mM
dtpa, 100 mM Trolox, and 10 mM FNOCT 14; cell density in the experiments was 5.4ꢀ0.5·104 cells cmꢁ2
(1.0ꢀ0.1·105 cells/well). To half of the wells with activated and half of the wells with nonactivated
cells, L-arginine (0.5 mM) was added, to the other wells NMA (0.1 mM) was added. Cells were incubated
at 378 in a humidified atmosphere containing 95% air/5% CO2 for 1 h. After 1 h of incubation, the super-
natant was removed and assessed in the microplate reader at lexc 320 nm, lem 380 nm by using the same
instrument settings as for the cell-free experiments. Cells were lysed in HBSS containing sodium dodecyl
sulfate (SDS) in a final concentration of 0.4% (w/v), dtpa, and Trolox. Cell lysates were centrifuged to
remove cell debris, and supernatants were then assessed in the same way as the samples of the superna-
tant Krebs–Henseleit buffer. Blanks of all solns. with all additives including FNOCT 14 were incubated
(in the absence of cells) for the same time periods and measured at the same time as samples. Blank val-
ues were subtracted from all sample values. Cell blanks were incubated in the absence of FNOCT; how-
ever, cellular autofluorescence proved to be negligible in the experimental setting used here. Fluores-
cence values obtained are either given as arbitrary units (corrected for the fluorescence of blanks, for vol-
ume, and cell density) or were corrected for blanks, converted to NO concentrations based on the cali-
bration curve shown in Fig. 7, and then corrected for volumes and cell density. After fluorescence of cel-
lular supernatants and cellular lysates was determined, MAHMA NONOate was added in large excess
(100 mM) to all samples to obtain maximal FNOCT 14 fluorescence; these values were used to determine
FNOCT 14 distribution (supernatant vs. cell-associated).
REFERENCES
[1] ꢁThe Biology of Nitric Oxideꢂ, Eds. S. Moncada, M. Feelisch, R. Busse, and E. A. Higgs, Vol. 3 and 4,
Portland Press, London, 1994.
[2] M. Marzinzig, A. K. Nüssler, J. Stadler, E. Marzinzig, W. Barthlen, N. C. Nüssler, H. G. Beger, S. M.
Morris, Jr., U. B. Brückner, Nitric Oxide: Biol. Chem. 1997, 1, 177.
[3] R. F. Furchgott, Angew. Chem., Int. Ed. 1999, 38, 1870.
[4] F. Murad, Angew. Chem., Int. Ed. 1999, 38, 1856.
[5] L. J. Ignarro, Angew. Chem., Int. Ed. 1999, 38, 1882.
[6] ꢁNitric Oxide, Part A: Sources and Detection of NO; NO Synthaseꢂ, Ed. L. Packer, in ꢁMethods in
Enzymologyꢂ, Vol. 268, Academic Press, 1996.
[7] H.-G. Korth, K. U. Ingold, R. Sustmann, H. de Groot, H. Sies, Angew. Chem., Int. Ed. Engl. 1992, 31,
891.
[8] H.-G. Korth, K. U. Ingold, R. Sustmann, H. de Groot, H. Sies, Angew. Chem. 1992, 104, 915.
[9] H.-G. Korth, R. Sustmann, P. Lommes, T. Paul, A. Ernst, H. de Groot, L. Hughes, K. U. Ingold, J.
Am. Chem. Soc. 1994, 116, 2767.

2296
Helvetica Chimica Acta – Vol. 89 (2006)
[10] H.-G. Korth, R. Sustmann, P. Lommes, T. Paul, A. Ernst, K. U. Ingold, L. Hughes, H. de Groot, H.
Sies, in ꢁThe Biology of Nitric Oxideꢂ, Vol. 4, Eds. S. Moncada, M. Feelisch, R. Busse, and E. A.
Higgs, Portland Press, London, 1994, p. 220–224.
[11] D. W. Jones, A. Pomfret, J. Chem. Soc. Perkin, Trans. 1 1991, 263.
[12] M. Bätz, Ph. D. Thesis, Universität Essen, 1997.
[13] I. Ioannidis, M. Bätz, T. Paul, H.-G. Korth, R. Sustmann, H. de Groot, Biochem. J. 1996, 318, 789.
[14] S. A. Green, D. J. Simpson, G. Zhou, P. S. Ho, N. V. Blough, J. Am. Chem. Soc. 1990, 112, 7337.
[15] P. Meineke, U. Rauen, H. de Groot, H.-G. Korth, R. Sustmann, Chem.–Eur. J. 1999, 5, 1738.
[16] A. B. A. Jansen, T. J. Russell, J. Chem. Soc. 1965, 2127.
[17] H. Ferres, Chem. Ind. 1980, 435.
[18] R. Y. Tsien, Nature (London) 1981, 290, 527.
[19] P. Talbiersky, Diploma Thesis, Universität Duisburg-Essen, 2006.
[20] J. L. Segura, N. Martin, Chem. Rev. 1999, 99, 3199.
[21] M. Bätz, H.-G. Korth, R. Sustmann, Angew. Chem., Int. Ed. 1997, 36, 1501.
[22] A. R. Forrester, in ꢁLandolt-Bçrnstein New Series, Magnetic Properties of Free Radicalsꢂ, Vol. II/9,
Part c1, Eds. H. Fischer and K. H. Hellwege, Springer Verlag, Berlin, 1979.
[23] P. Meineke, U. Rauen, H. de Groot, H.-G. Korth, R. Sustmann, Biol. Chem. 2000, 381, 575.
[24] J. A. Hrabie, J. R. Klose, D. A. Wink, L. K. Keefer, J. Org. Chem. 1993, 58, 1472.
[25] A. Werner, Ber. Dtsch. Chem. Ges. 1904, 37, 3083.
[26] K. Brass, E. Ferber, Chem. Ber. 1922, 55, 541.
[27] J. Schmidt, O. Spoun, Chem. Ber. 1922, 55, 1194.
[28] W. Dilthey, I. ter Horst, W. Schommer, J. Prakt. Chem. N. F. 1935, 143, 189.
[29] M. Neuenschwander, R. Iseli, Helv. Chim Acta 1977, 60, 1061.
[30] H. Koch, Monatsh. Chem. 1962, 93, 1343.
[31] H. Koch, L. Katlan, K. Markut, Monatsh. Chem. 1965, 96, 1646.
[32] D. R. Duling, J. Magn. Reson. B 1994, 104, 105.
[33] J.-F. Wang, P. Komarov, H. de Groot, Arch. Biochem. Biophys. 1993, 304, 189.
Received April 25, 2006