**1. Introduction**

346 Nitroxides – Theory, Experiment and Applications

293, No 3, pp. H1442-50.

5703-7.

Zhu, X., Liu, B., Zhou, S., Chen, Y. R., Deng, Y., Zweier, J. L. and He, G. (2007). Ischemic preconditioning prevents *in vivo* hyperoxygenation in postischemic myocardium with preservation of mitochondrial oxygen consumption. *Am J Physiol Heart Circ Physiol,* Vol.

Zweier, J. L. and Kuppusamy, P. (1988). Electron paramagnetic resonance measurements of free radicals in the intact beating heart: a technique for detection and characterization of free radicals in whole biological tissues. *Proc Natl Acad Sci U S A,* Vol. 85, No 15, pp.

> Reactive oxygen species (ROS) are the prize we are paying for an energy-efficient life under oxygen. They play a role in many diseases such as atherosclerosis, hypertension, ischemiareperfusion injury, inflammation, type-2 diabetes, certain neurodegenerative diseases, even cancer and, certainly, aging (Kohen & Nyska, 2002). Among these ROS are several radicals such as the hydroxyl, peroxy, alkoxy as well as the superoxide anion radical (Boveris, 1977). Direct measurement of the radicals is hampered by their short half lifes in the range of nanoto microseconds. They can, however, be trapped by addition to nitrones leading to relatively stable nitroxides with t½ of minutes (Janzen, 1971). Respiring mitochondria are a major source of ROS, particularly of the superoxide anion radical, which is formed in complexes I and III (Cadenas & Davies, 2000; Dröge, 2002; Inoue et al., 2003; Turrens, 2003).

> A couple of years ago we have described the synthesis and first application of a fluorescent nitrone composed of *tert-*butyl-nitrone and a *p*-nitro-stilbene moiety, which can be used for the detection of ROS with subcellular resolution. Short-lived radicals add to the nitrone under formation of a nitroxide radical which then quenches the fluorescence of the *p*-nitrostilbene (Hauck et al., 2009).

> Similar double labels have previously been synthesized or at least suggested, partially in collaboration with us. Likhtenshtein and Hideg suggested in the eighties to couple nitroxides to a fluorophore which will become fluorescent only after reduction of the nitroxide moiety in viable biological systems (Bystryak et al., 1986). Hideg used fluorescent pyrrolines which can be oxidized to nitroxides mainly by singlet oxygen (Kalai et al., 1998). Rosen and collaborators investigated a fluorescent nitrone much like ours composed of nitrobenzene instead of nitrostilbene. The authors did not observe a decrease in the fluorescence upon reaction of the nitrone with α-hydroxyethyl radicals which was probably due to an only very small concentration of the nitroxide formed in this reaction (Pou et al.,

© 2012 Trommer et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2012 Trommer et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1995). To the best of our knowledge, this work has not been continued. Somewhat different approaches for the detection and trapping of ROS were taken by Bottle et al., 2003, Heyne et al., 2007 and Blough et al., 1988.

Here we describe some extended studies with the *p*-nitrostilbene-*tert*-butyl-nitrone (**1**) under a variety of physiological conditions *i.e.*, employing inhibitors of components of the respiratory chain, the F1-F0-ATP synthase and the membrane potential as well as similar studies with a corresponding coumaryl-styryl-*tert*-butyl-nitrone derivative (**2**) and finally, a third compound based on 4-pyrrolidine-1,8-naphthylimido-methylbenzene as fluorophore (**3**).

## **2. Results and discussion**

### **2.1.** *p***-Nitrostilbene-***tert***-butyl-nitrone, 1**

Fig. 1 shows the reaction of the previously employed fluorescent nitrone **1** with the hydroxyl radical under formation of the corresponding nitroxide. The hydroxyl radical was generated by the Fenton reaction and the product extracted from the aqueous phase with degassed ethyl acetate yielding an EPR spectrum composed of 6 lines typical for hydroxyl radical adducts (Hauck et al., 2009).

When cultures of various cell lines, *i.e.*, HeLa, COS 7 and CHO, were incubated with **1** and subsequently washed, it had almost exclusively accumulated in mitochondria as shown by co-staining with tetramethylrhodamine ethylester (TMRE, Farkas et al., 1989) and confocal laser spectroscopy (Fig. 2). Without any further addition the fluorescence slowly decreased to almost none within about 20 min. However, generation of hydroxyl radicals by the Fenton reaction (Walling, 1975) reduced this time to 20 sec (Fig. 3, Hauck et al., 2009). Similar results were obtained in presence of the complex I and III inhibitors, rotenone and antimycin A, but within a timeframe of 40 to 60 sec. Under these conditions both complexes are known to produce substantial amounts of the superoxide anion radical (Dlaskova et al., 2008; Han et al., 2001). In these early studies a quasi-confocal microscope was used (Axiovert 440, equipped with an ApoTome, Carl Zeiss, Jena) which has broad bandwidth filters, only. Thus, not allowing for monitoring small shifts in the emission spectra of **1** upon addition of the radical as had been observed in cell-free controls.

**Figure 1.** Structural formula of the *p*-nitrostilbene-*tert*-butyl-nitrone **1** and its reaction with the hydroxyl radical

Fluorescent Nitrones for the Study of ROS Formation with Subcellular Resolution 349

348 Nitroxides – Theory, Experiment and Applications

al., 2007 and Blough et al., 1988.

**2. Results and discussion** 

radical adducts (Hauck et al., 2009).

radical

**2.1.** *p***-Nitrostilbene-***tert***-butyl-nitrone, 1** 

addition of the radical as had been observed in cell-free controls.

O H O2N <sup>H</sup>

O2N <sup>H</sup>

1995). To the best of our knowledge, this work has not been continued. Somewhat different approaches for the detection and trapping of ROS were taken by Bottle et al., 2003, Heyne et

Here we describe some extended studies with the *p*-nitrostilbene-*tert*-butyl-nitrone (**1**) under a variety of physiological conditions *i.e.*, employing inhibitors of components of the respiratory chain, the F1-F0-ATP synthase and the membrane potential as well as similar studies with a corresponding coumaryl-styryl-*tert*-butyl-nitrone derivative (**2**) and finally, a third compound based on 4-pyrrolidine-1,8-naphthylimido-methylbenzene as fluorophore (**3**).

Fig. 1 shows the reaction of the previously employed fluorescent nitrone **1** with the hydroxyl radical under formation of the corresponding nitroxide. The hydroxyl radical was generated by the Fenton reaction and the product extracted from the aqueous phase with degassed ethyl acetate yielding an EPR spectrum composed of 6 lines typical for hydroxyl

When cultures of various cell lines, *i.e.*, HeLa, COS 7 and CHO, were incubated with **1** and subsequently washed, it had almost exclusively accumulated in mitochondria as shown by co-staining with tetramethylrhodamine ethylester (TMRE, Farkas et al., 1989) and confocal laser spectroscopy (Fig. 2). Without any further addition the fluorescence slowly decreased to almost none within about 20 min. However, generation of hydroxyl radicals by the Fenton reaction (Walling, 1975) reduced this time to 20 sec (Fig. 3, Hauck et al., 2009). Similar results were obtained in presence of the complex I and III inhibitors, rotenone and antimycin A, but within a timeframe of 40 to 60 sec. Under these conditions both complexes are known to produce substantial amounts of the superoxide anion radical (Dlaskova et al., 2008; Han et al., 2001). In these early studies a quasi-confocal microscope was used (Axiovert 440, equipped with an ApoTome, Carl Zeiss, Jena) which has broad bandwidth filters, only. Thus, not allowing for monitoring small shifts in the emission spectra of **1** upon

**Figure 1.** Structural formula of the *p*-nitrostilbene-*tert*-butyl-nitrone **1** and its reaction with the hydroxyl

C N O

C N O

OH

C

C

With kind permission from Springer Science + Business Media: Appl. Magn. Reson., p-Nitrostilbene-*tert*-butyl-nitrone: A novel fluorescent spin trap for the detection of ROS with subcellular resolution, vol. 36 (2009), p. 143, Stefan Hauck, Yvonne Lorat, Fabian Leinisch, Wolfgang E. Trommer, Fig. 8

**Figure 2.** COS cells incubated with **1** (**A**) or with TMRE (**B**). The same cells were stained for DNA in the nuclei with DAPI (**C**, λabs = 340 nm, λem = 504 nm), and also with **1** (λabs = 388 nm, λem = 504 nm)

The accumulation of **1** in mitochondria came unexpected but is possibly due to an effect previously observed with so-called Skulachev ions, derivatives of ubiquinone or plastoquinone with long hydrophobic side chains composed of up to 10 isoprene units and a positively charged triphenylphosphonium group at the end, *e.g.*, SkQ1 as introduced by Vladimir Skulachev. The positive charge is not localized but spread over the aromatic rings, thus allowing for membrane permeation due to the already negative potential of cells with respect to their outside and even more so of mitochondria, the only human organelle with a negative potential with respect to the cytoplasm. Once inside, they are trapped by the opposite potential outside (Severin et al., 2010). Similar effects have been observed with certain amphiphilic dipolar compounds to which **1** could belong (M.V. Skulachev, personal communication).

With kind permission from Springer Science + Business Media: Appl. Magn. Reson., p-Nitrostilbene-*tert*-butyl-nitrone: A novel fluorescent spin trap for the detection of ROS with subcellular resolution, vol. 36 (2009), p. 145, Stefan Hauck, Yvonne Lorat, Fabian Leinisch, Wolfgang E. Trommer, Fig. 10

**Figure 3.** Fluorescence decay of *p*-nitrostilbene-*tert*-butyl-nitrone **1** in CHO cells: **A**: 0 sec, **B**: 5 sec, **C**: 10 sec, **D**: 20 sec after addition of 20 µl each of 10 mM Fe2+ and 10 mM H2O2 (Fenton reaction)

The fluorescence half-life of 1 in mitochondria was studied under a variety of conditions. Data are summarized in Table I together with those from the coumaryl derivative **2**. But why look for another double label? **1** has an absorption maximum at 383 nm which is well separated from the emission at 568 nm. However, only recently have confocal laser microscopes been equipped with lasers of 405 nm, most commercial instruments operate at 480 nm and above, rather outside the absorption range of **1**. Moreover, stilbenes as fluorophores pose yet another problem, *cis-trans* isomerization upon irradiation leading to a substantial shift in emission wavelength which is accompanied by photobleaching and also recovery rendering interpretation of data more complex (Fig. 6).

### **2.2. Coumaryl-styryl-***tert***-butyl-nitrone, 2**

2-(4-Trifluoromethyl-2-oxo-2H-chromen-7-yl)-*E*-vinyl-1-(N-(1,1-dimethyl))-phenyl-4-nitrone **2** was synthesized according to the scheme shown in Fig. 4.

Unexpectedly, the absorbance and fluorescence properties of **1** and **2** do not differ very much as shown in Fig. 5, A & B.

350 Nitroxides – Theory, Experiment and Applications

allowing for membrane permeation due to the already negative potential of cells with respect to their outside and even more so of mitochondria, the only human organelle with a negative potential with respect to the cytoplasm. Once inside, they are trapped by the opposite potential outside (Severin et al., 2010). Similar effects have been observed with certain amphiphilic dipolar compounds to which **1** could belong (M.V. Skulachev, personal communication).

With kind permission from Springer Science + Business Media: Appl. Magn. Reson., p-Nitrostilbene-*tert*-butyl-nitrone: A novel fluorescent spin trap for the detection of ROS with subcellular resolution, vol. 36 (2009), p. 145, Stefan Hauck,

**Figure 3.** Fluorescence decay of *p*-nitrostilbene-*tert*-butyl-nitrone **1** in CHO cells: **A**: 0 sec, **B**: 5 sec, **C**: 10

The fluorescence half-life of 1 in mitochondria was studied under a variety of conditions. Data are summarized in Table I together with those from the coumaryl derivative **2**. But why look for another double label? **1** has an absorption maximum at 383 nm which is well separated from the emission at 568 nm. However, only recently have confocal laser microscopes been equipped with lasers of 405 nm, most commercial instruments operate at 480 nm and above, rather outside the absorption range of **1**. Moreover, stilbenes as fluorophores pose yet another problem, *cis-trans* isomerization upon irradiation leading to a substantial shift in emission wavelength which is accompanied by photobleaching and also

2-(4-Trifluoromethyl-2-oxo-2H-chromen-7-yl)-*E*-vinyl-1-(N-(1,1-dimethyl))-phenyl-4-nitrone

sec, **D**: 20 sec after addition of 20 µl each of 10 mM Fe2+ and 10 mM H2O2 (Fenton reaction)

recovery rendering interpretation of data more complex (Fig. 6).

**2** was synthesized according to the scheme shown in Fig. 4.

Yvonne Lorat, Fabian Leinisch, Wolfgang E. Trommer, Fig. 10

**2.2. Coumaryl-styryl-***tert***-butyl-nitrone, 2** 

**Figure 5.** Absorption and emission spectra of **A**, *p*-nitrostilbene-*tert*-butyl-nitrone **1** and **B**, cumarylstyryl-*tert*-butyl-nitrone **2,** excitation at 410 nm

Whereas **1** undergoes substantial photobleaching upon irradiation at 310 nm for 1 min (Fig. 6 B), the fluorescence of **2** is almost completely stable under these conditions (Fig. 6 A). However, **1** recovers rather quickly reaching about 40 % of the initial fluorescence within 4 min (Fig. 6 C).

Unfortunately, **2** turned out to be far more toxic to cells in comparison with **1**. Growing adherent MCF-7 cells, a human breast adenocarcinoma cell line, were exposed to these compounds for three days. After fixation of viable cells with trichloroacetic acid, the damaged cells were removed by washing and cell proteins of the remaining cells were stained with sulforhodamine and measured photometrically at 570 nm (Skehan et al., 1990). In a rough estimate assuming sigmoidal behavior in a semi-logarithmic plot the half maximal dosis, IC50, of **1** was about 300 µM as compared to 40 µM for **2**. Simultaneous irradiation at 366 nm for 3 min further reduced this value to 30 µM. The known toxicity of coumarin may play a role here (Fig. 7, A & B; Oodyke, 1974). Therefore spin-trapping experiments in the presence of inhibitors were primarily carried out with **1**. However, **2** may be employed as well by using lower concentrations of 10 to 20 µM.

**Figure 6.** Photobleaching of **2** (CSN, **A**) und **1** (NSN, **B**) upon irradiation at 310 nm for 1 min. The recovery of **1** with time is shown in panel **C**.

**Figure 7.** Cytotoxicity of **1** (**A**) and **2** (**B**) as determined in MCF-7 cells after a three days exposure. **B**, lower trace upon additional irradiation at 366 nm.

The effect of rotenone as inhibitor of complex I of the respiratory chain on the formation of the superoxide anion radical formation in HeLa cells (Hauck et al., 2009) was followed with **2** (Fig. 8). The initial fluorescence and after 100 or 90 sec, respectively, is shown in the absence and presence of rotenone. The time course of this decay as measured in 0.5 sec intervals is shown as well. Clearly, inhibition of complex I significantly increases ROS formation. The figure also reveals that the intracellular distribution of **2** is not as enriched in mitochondria as had been found with **1**. Evidence that quench did not result, or at least not largely, from photobleaching came from experiments in which the shutter was closed during measurements. However, some non-specific redox reactions cannot be excluded.

352 Nitroxides – Theory, Experiment and Applications

**A B**


I / a.u.

400 600 800 1000

/ nm

**C**

I / a.u.


**A B**

recovery of **1** with time is shown in panel **C**.

lower trace upon additional irradiation at 366 nm.

**Figure 6.** Photobleaching of **2** (CSN, **A**) und **1** (NSN, **B**) upon irradiation at 310 nm for 1 min. The

400 600 800 1000

/ nm

0.0 0.2 0.4 0.6 0.8 1.0

> NSN NSN (h) NSN (h) +4 min

I / a.u.

 CSN CSN (h)

400 600 800 1000

 NSN NSN (h)

/ nm

**Figure 7.** Cytotoxicity of **1** (**A**) and **2** (**B**) as determined in MCF-7 cells after a three days exposure. **B**,

**Figure 8.** False color representation of the fluorescence decay of **2** upon trapping of ROS in HeLa cells in the absence (upper panels) and presence of 10 µM rotenone (lower panels), 0 and 100 sec after addition of the inhibitor as well as the time course of these changes (bottom panel).
