Dihydroxycoumarins as highly selective fluorescent probes for the fast detection of 4-hydroxy-TEMPO in aqueous solution

Abstract

Due to the biological significance of 4-hydroxy-TEMPO, its detection in biological systems such as blood serum is of great importance in both research and clinical applications. The following paper reports a novel fluorimetric method for the detection and quantitative determination of 4-hydroxy-TEMPO radicals in an aqueous solution using dihydroxycoumarins as fluorescent probes. Among the 17 coumarin derivatives studied, only some dihydroxycoumarins show high sensitivity, specificity and selectivity for 4-hydroxy-TEMPO. Among them, 6,7-dihydroxycoumarin (esculetin) exhibits the strongest fluorescence enhancement under the action of 4-hydroxy-TEMPO. In this assay, esculetin reacts with 4-hydroxy-TEMPO to exclusively yield a dimer. 6,7-Dihydroxycoumarin responds to 4-hydroxy-TEMPO quickly and shows a 2.5-fold fluorescence enhancement with an estimated detection limit of 285 μM.

---

Introduction

4-Hydroxy-TEMPO is a stable, membrane permeable, nitroxide radical, which effectively protects cells and tissues from damage associated with oxidative and nitrative stress conditions. It has been proven that 4-hydroxy-TEMPO interacts with nitrogen dioxide (one of the compounds responsible for nitrative stress) and decreases its concentration in cells up to 500-fold.^1,2 Profluorescent nitroxides have been previously reported as probes for the detection of free radicals, as well as the damage mediated by these species.^3–5 4-Hydroxy-TEMPO, due to its low molecular weight, good superoxide dismutase mimic activity and low toxicity, has been selected as a good candidate for development as a preventive drug. As its infusion appears to prevent pancreatic injury caused by free radicals in experimental cerulean pancreatitis, 4-hydroxy-TEMPO seems to be a promising tool for the treatment of acute pancreatitis.⁶ Additionally, 4-hydroxy-TEMPO has been proven to reduce renal dysfunction, damage by ischemia/reperfusion and intestinal injury in rats subjected to splanchnic artery occlusion.⁷ Furthermore, 4-hydroxy-TEMPO and other derivatives of TEMPO radical are used in a variety of industrial applications as highly selective oxidation catalysts for the production of pharmaceuticals, flavours and fragrances, agrochemicals, and a variety of other specialty chemicals.⁸ Therefore, the development of highly efficient methods for the selective detection of 4-hydroxy-TEMPO radical is extremely desirable.

Many approaches, such as nuclear magnetic resonance, mass spectrometry, electron paramagnetic resonance spectroscopy, electron spin resonance spectroscopy and chemiluminescence techniques have been employed for the study of free radicals and free radical damage.⁹ However, the best known approaches suffer from a couple of drawbacks including time-consuming processes, relatively high costs and lack of temporal and spatial resolution, which significantly limit their potential applications in many biological studies. In view of its simplicity, high sensitivity, high-throughput capability, precise quantitative determination and real-time detection, fluorescent techniques are regarded as one of the most promising methods to detect different species. Furthermore, the approach of fluorescence imaging is the best technique for the determination and measurement of intracellular molecules without the destruction of tissues or cells thus making fluorescence spectroscopy superior to other analytical methods.¹⁰ A number of fluorogenic reagents for the detection and quantitative determination of radicals is commercially available: including hydroethidine and (2-pyridil)-benzothiazoline for superoxide anion,¹¹ (9-anthroyloxy)-2,2,6,6-tetramethylpiperidine-1-oxyl, 1,3-cyclohexanedione, sodium terephthalate and coumarin-3-carboxylic acid for hydroxyl radical,¹¹ cis-parinaric acid, dipyridamole and diphenyl-1-pyrenylphosphine for peroxyl radical,¹¹ 2,3-diaminonaphthalene, diaminorhodamine, 1,2-diaminoanthraquinone, diaminofluoresceins and naphthalimide-based probes for nitric oxide.^12–14 Despite these choices, fluorogenic reagents, combined with new functional mechanisms, are still required for the determination of radicals. On the other hand, there is no literature available on fluorescent sensors (which display a fluorescence enhancement) used for the detection of nitroxides, such as 4-hydroxy-TEMPO. In this work, we demonstrate a novel fluorimetric method for the detection and quantitative determination of 4-hydroxy-TEMPO in aqueous solution using dihydroxycoumarins as fluorescent probes, among which 6,7-dihydroxycoumarin (esculetin) exhibits the strongest response to the radical.

6,7-Dihydroxycoumarin (esculetin) is a naturally occurring coumarin derivative that may be isolated from many plants such as Artemisia capillaries, Citrus limonia and Euphorbia lathyris. It is known to possess antimicrobial properties,¹⁵ as well as pleiotropic biological activities including lipoxygenase-inhibitory activity,¹⁶ free radical and reactive oxygen species scavenging activity,¹⁷ suppressive activity on oxidative damage to DNA,¹⁸ tyrosinase-inhibitory activity,¹⁹ as well as cancer chemopreventive and anti-tumor activities.²⁰ 6,7-Dihydroxycoumarin has no significant cytotoxic effect on normal murine macrophages. It is found to increase the endocytic activity and the mitogenesis of splenic lymphocytes, as well as to augment nitric oxide production and iNOS gene expression in LPS-treated macrophages.¹⁵ Additionally, it has been proven that the presence of 6,7-ortho-dihydroxy functions in the coumarin structure gives it the greatest potency in protection against inactivation of lysozyme by radicals and therefore makes it an attractive candidate for evaluation as a protective agent against disorders in which oxidative stress is implicated.²¹

Coumarin and its derivatives are known to have an ability to reversibly photodimerize (upon irradiation with wavelengths longer than 300 nm in solution) and subsequently photocleave (upon irradiation at wavelengths shorter than 300 nm).²² On the basis of a cis fusion of the cyclobutane ring to the 6-membered pyrone rings, four possible structures may be considered for coumarin dimers (the actual coumarin dimer formed on irradiation depends on the combination of dose, solvent polarity, coumarin concentration and the multiplicity of the excited state molecule undergoing reaction).^23–25 Although studies on photoproducts such as dimers of the coumarin have been previously reported,^26,27 little has been published on the photochemistry of coumarins possessing two oxygen substituents in the phenyl ring or for coumarins in aqueous solution.²⁸ Yu et al. has examined the distribution of isomeric photodimers formed from 6-alkylcoumarins in aqueous solution and in aqueous detergent micelles but did not study the photodimerization of coumarins with oxygen substituents.²⁷ In addition, Moriya studied the fluorescent forms of 7-hydroxycoumarin and 7-hydroxy-4-methylcoumarin in water, as well as the kinetics and mechanism of their tautomerization in photoexcited states at various pH levels but did not examine the formation of photodimers by these coumarins.²⁹

As far as we know, there is no literature available on radical-induced dimerization of naturally occurring coumarins that have oxygen substituents. We have examined the interactions between 4-hydroxy-TEMPO and various fluorescent compounds, such as polycyclic aromatic hydrocarbons,³⁰ coumarins^31,32 and fluoroquinolone antibiotics.³³ In all cases, the interactions were entirely physical and 4-hydroxy-TEMPO acted as a fluorescence quencher.

Herein, we report that 6,7-dihydroxycoumarin rapidly dimerizes under the action of 4-hydroxy-TEMPO. Additionally, for this coumarin derivative, a fluorescence enhancement that is a linear function of 4-hydroxy-TEMPO concentration is observed. The selectivity of the probe for sensing 4-hydroxy-TEMPO was checked. The detection limit of the radical was estimated to be 285 μM.

Materials and methods

Coumarins and 4-hydroxy-TEMPO were purchased from Sigma Aldrich. All the reagents had purity greater than 99.8% and there was no need for further purification. Deionized water and acetonitrile were used as solvents. In order to avoid a self-quenching or inner filter effect, the solutions of 6,7-dihydroxycoumarin and other coumarins were prepared at a fixed concentration (1 × 10⁻⁵ M). The concentration of the stock solution of 4-hydroxy-TEMPO was constant too (0.25 M). All the solutions were prepared just before use.

The UV absorption spectra of 4-hydroxy-TEMPO in aqueous solution were recorded with the use of a Perkin Elmer Lambda 650 UV-VIS spectrometer (equipped with temperature controller) in the absence of 6,7-dihydroxycoumarin and in its presence at different concentrations. All the absorbance measurements were performed at the temperature of 25 °C. The fluorescence emission spectra of 6,7-dihydroxycoumarin in aqueous solution were recorded with the use of a Cary Eclipse Varian spectrofluorimeter (equipped with temperature controller) in the absence of 4-hydroxy-TEMPO and in the presence of increasing amounts of that radical. All the fluorescence measurements were performed under the following conditions: excitation wavelength – 350 nm, temperature – 25 °C, excitation and emission slits – 5 nm.

The high-performance liquid chromatography (HPLC) separation was performed using a Beckman Gold System with UV detector, which was set at 254 nm for monitoring the effluents. The flow rate was set at 0.8 ml min⁻¹. A Wakopak reverse-phase C18 column (4.6 mm × 150 mm; 5 μm in particle size) with a mobile phase consisting of deionized water and acetonitrile (90 : 10, v/v) was used.

An Agilent 1200 Technologies HPLC System was employed for LC-MS/MS experiments. Analysis conditions are described in the previous paragraph. The effluent was coupled with the HCTultra ion-trap mass spectrometer, which was operated in the positive- and negative-ion modes. The mass spectrometer was set up to MS or MS/MS mode. A 20 μl of sample solution was injected in each run. The spray voltage was set to 4.0 kV for all experiments. Each spectrum was obtained by averaging 3 scans, and the time for each scan was 0.1 s.

Results and discussion

Fig. 1 shows absorption spectra of 4-hydroxy-TEMPO in the presence of increasing concentrations (0–35.5 μM) of 6,7-dihydroxycoumarin. In the absence of 6,7-dihydroxycoumarin, 4-hydroxy-TEMPO displayed a strong absorption band around 430 nm. With the addition of 6,7-dihydroxycoumarin, the absorption of 4-hydroxy-TEMPO gradually decreased and the peak at around 350 nm increased simultaneously with a distinct isosbestic point at 406 nm.

Fig. 1 Absorption spectra of 4-hydroxy-TEMPO (0.1 mM) in the presence of increasing concentrations (0–35.5 μM) of 6,7-dihydroxycoumarin in deionized water at 25 °C.

Fig. 2 shows the fluorescence emission spectra of 6,7-dihydroxycoumarin in the presence of an increasing concentration (0–7.09 mM) of 4-hydroxy-TEMPO. Upon excitation at 350 nm, 6,7-dihydroxycoumarin gives a fluorescence emission maximum at 462 nm. Upon addition of 4-hydroxy-TEMPO, a 2.5-fold increase in fluorescence, accompanied by a slightly bathochromic shift (5 nm) in the emission maximum is observed. Our previous studies on the interactions between a group of variously substituted coumarins and 4-hydroxy-TEMPO in aqueous solution showed that all fluorophores studied were sensitive to the presence of the nitroxide radical, which acted as a very effective fluorescence quencher.³¹ Additionally, no changes in absorption spectra due to the presence of 4-hydroxy-TEMPO were observed.

Fig. 2 Increase in fluorescence intensity of 6,7-dihydroxycoumarin (10 μM) with the addition of 4-hydroxy-TEMPO (0–7.09 mM) in deionized water at 25 °C, λ_ex = 350 nm.

Significantly, a linear relationship (R² = 0.9845) was found between the fluorescence enhancement at 462 nm and the 4-hydroxy-TEMPO concentrations up to 9.6 mM (Fig. 3). In an experiment, 10 μM of 6,7-dihydroxycoumarin showed an over 2.5-fold fluorescence increase immediately after an addition of 9.6 mM 4-hydroxy-TEMPO in deionized water. At higher concentrations of 4-hydroxycoumarin, a decrease in fluorescence intensity was observed. The detection limit of the probe for sensing 4-hydroxy-TEMPO was estimated to be 285 μM. These results demonstrate that 6,7-dihydroxycoumarin is sensitive to 4-hydroxy-TEMPO and could be potentially used to quantitatively detect 4-hydroxy-TEMPO concentration.

Fig. 3 4-Hydroxy-TEMPO concentration-dependent fluorescence intensity of 6,7-dihydroxycoumarin changes. 6,7-Dihydroxycoumarin 10 μM, 4-hydroxy-TEMPO 0–9.6 mM in water (λ_ex = 350 nm; λ_em = 462 nm). Concentration dependence data represents an average of 3 independent experiments, while fluorescence intensity values are an average of 15 experiments.

To evaluate the selectivity of 6,7-dihydroxycoumarin towards 4-hydroxy-TEMPO, typical redox agents including glutathione, DL-dithiothreitol, L-cysteine, L-methionine, D-glucose, ammonium nitrate, ammonium perchlorate and hydrogen peroxide were explored. 6,7-Dihydroxycoumarin showed no fluorescence enhancement or the increase in fluorescence was negligible under the action of these species in aqueous solution (Fig. 4). It proves that 6,7-dihydroxycoumarin is a highly selective fluorescence probe for 4-hydroxy-TEMPO in aqueous solution.

Fig. 4 The fluorescence enhancement of 6,7-dihydroxycoumarin (10 μM) in the presence of 4-hydroxy-TEMPO and various redox agents (c = 1 mM) in aqueous solution. λ_ex = 350 nm; λ_em = 462 nm.

To evaluate the selectivity of the radical towards 6,7-dihydroxycoumarin, a group of different coumarins was explored. As shown in Fig. 5, the addition of 4-hydroxy-TEMPO to the solution of 6,7-dihydroxycoumarin induced the most significant enhancement of fluorescence (ca. 2.5-fold). In the case of other dihydroxy-substituted coumarins, a smaller increase in fluorescence intensity was observed, while the fluorescence intensity of other coumarins was quenched by 4-hydroxy-TEMPO. The above results suggest that different dihydroxycoumarins may act as highly selective fluorescent probes for 4-hydroxy-TEMPO in aqueous solution.

Fig. 5 The fluorescence changes of various coumarins (10 μM) in the presence of 4-hydroxy-TEMPO (9.6 mM) in aqueous solution. The experiments were carried out 3 times (λ_ex and λ_em were individually chosen for each coumarin on the basis of its absorption and emission spectrum).

In order to find the optimal sensing conditions, the fluorescence intensity of 6,7-dihydroxycoumarin was determined in buffer solutions of various pH values. It was discovered that there was nearly no change in the fluorescence intensity in the pH range from 5.9 to 9.1, suggesting that 6,7-dihydroxycoumarin could work under physiological conditions (data not shown).

The unique fluorescence enhancement of 6,7-dihydroxycoumarin, and some other dihydroxy-substituted coumarins, under the action of 4-hydroxy-TEMPO is due to the formation of the appropriate dimer. In the proposed sensing mechanism of 4-hydroxy-TEMPO, by initiating the dimerization of dihydroxycoumarins we assume that, as in other, described heteroaromatic systems,³⁴ an electron is transferred from the electron-rich oxygen atom of 4-hydroxy-TEMPO onto the coumarin, with the lowest electron density found on the unpaired electron at C(4) – due to the presence of two highly electron-drawing hydroxyl groups at C(6) and C(7). This radical species can subsequently interact with a second monomer to form the dimeric compound. A decrease in the absorbance at 430 nm observed in Fig. 1 may result from the dimerization of the coumarin group since the level of unsaturation decreases due to the formation of cross-links between cyclobutane rings. It is worth mentioning here that systems are known where the fluorescence intensity increases significantly after dimerization.³⁵ Thus, such a mechanism is in a good agreement with the results presented in Fig. 2 and 5, which show unequivocally that only the presence of two hydroxyl substituents causes an increase in fluorescence intensity. The possible four isomeric structures of 6,7-dihydroxycoumarin dimer are presented in Fig. 6.^25–27,36

Fig. 6 Four possible isomeric structures of 6,7-dihydroxycoumarin dimer.

In order to prove the 4-hydroxy-TEMPO-induced dimerization of 6,7-dihydroxycoumarin, additional experiments were conducted. The chromatogram that presents the product of the interaction between 6,7-dihidroxycoumarin and 4-hydroxy-TEMPO next to peaks corresponding to pure esculetin and pure TEMPO derivative was registered and presented in Fig. 7C. Additionally, the chromatograms of 4-hydroxy-TEMPO and 6,7-dihydroxycoumarinas registered in the same conditions are presented in Fig. 7A and B, respectively.

Fig. 7 (A) The chromatogram of 4-hydroxy-TEMPO; (B) the chromatogram of 6,7-dihydroxycoumarin; (C) the chromatogram of the product of the interaction between 6,7-dihydroxycoumarin and 4-hydroxy-TEMPO; absorbance measured – 254 nm, flow rate – 0.8 ml min⁻¹, time of analysis – 30 minutes, mobile phase – isocratic elution with the use of deionized water and acetonitrile (90 : 10, v/v).

The product of the reaction between 6,7-dihydroxycoumarin and 4-hydroxy-TEMPO was consecutively examined by mass spectrometry. As presented in Fig. 8, the product-ion spectrum of the [M − H]⁻ ion demonstrates the formation of the 6,7-dihydroxycoumarin dimer. It shows a weak parent molecular ion at 354.9, indicating a dimeric product (one mass unit lower than that expected for the parent peak of dimer), as well as a major ion peak at 177, one mass unit lower than that expected for the parent peak for monomeric coumarin. This fission is so efficient that no fragments between the molecular ion and m/z 177 can be detected. In general, the mass spectra of the dimers show a very strong peak corresponding to the monomer ion and in most cases a small peak due to the dimer (molecular ion) is located. It has to be recognized that this process involves cleavage of the molecule into two fragments of equal mass and therefore some structural symmetry must be present in this molecule.^37,38

Fig. 8 The product-ion spectrum of the [M − H]⁻ ion of the product of the reaction between 6,7-dihydroxycoumarin and 4-hydroxy-TEMPO.

Conclusions

The above data presents the development and evaluation of dihydroxycoumarins (namely 6,7-dihydroxycoumarin) for the sensitive and selective detection of 4-hydroxy-TEMPO in aqueous solution. 6,7-Dihydroxycoumarin can detect 4-hydroxy-TEMPO quantitatively with the detection limit of 285 μM. More importantly, esculetin responds to 4-hydroxy-TEMPO quickly and shows a more than 2.5-fold fluorescence enhancement for 9.6 mM nitroxide radical. Although the studies have not been performed in living cells, the probe has the potential application for the fast detection of 4-hydroxy-TEMPO in biological media since the concentrations of 4-hydroxy-TEMPO used in biological experiments are in the range 50 μM to 4 mM.^39,40 The biological significance of this work is proven by the fact that 4-hydroxy-TEMPO is used as a scavenger for different radicals. Its detection and quantitative determination under physiological conditions might help to understand the mechanism of oxidative stress.

Acknowledgements

This work was financially supported by the Polish National Science Centre (NCN) under the Grant No. 2012/07/B/ST5/00753.

References

1. A. Dąbrowska, D. Jacewicz, A. Chylewska, M. Szkatuła, N. Knap, J. Kubasik-Juraniec and L. Chmurzyński, Curr. Pharm. Anal., 2008, 4, 183.
2. A. Dąbrowska, D. Jacewicz, A. Łapińska, B. Banecki, A. Figarski, M. Szkatuła and L. Chmurzyński, Biochem. Biophys. Res. Commun., 2005, 326, 313.
3. J. P. Blinco, D. J. Keddie, T. Wade, P. J. Barker, G. A. George and S. E. Bottle, Polym. Degrad. Stab., 2008, 93, 1613.
4. A. Aspée, L. Maretti and J. C. Scaiano, Photochem. Photobiol. Sci., 2003, 2, 1125.
5. A. Aspée, O. García, L. Maretti, R. Sastre and J. C. Scaiano, Macromolecules, 2003, 36, 3550.
6. Z. Śledziński, M. Woźniak, E. Lezoche, M. Familiari, E. Bertoli, L. Greci and Z. Wajda, Int. J. Pancreatol., 1995, 18, 153.
7. Y. Karataş, M. A. Seçilmiş, I. Karayaylalı, F. Doran, K. Büyükafşar, E. Şingirik and A. Dikmen, Fundam. Clin. Pharmacol., 2004, 18, 79.
8. R. Ciriminna and M. Pagliaro, Org. Process Res. Dev., 2009, 14, 245.
9. B. Chance and G. Gao, Environ. Health Perspect., 1994, 102, 29.
10. Y. Makino, S. Uchiyama, K. I. Ohno and H. Arakawa, Anal. Chem., 2010, 82, 1213.
11. A. Gomes, E. Fernandes and J. L. Lima, J. Biochem. Biophys. Methods, 2005, 65, 45.
12. A. Gomes, E. Fernandes and J. L. Lima, J. Fluoresc., 2006, 16, 119.
13. D. Jacewicz, K. Żamojć and L. Chmurzyński, Curr. Pharm. Anal., 2012, 8, 115.
14. C. B. Huang, J. Huang and L. Xu, RSC Adv., 2015, 5, 13307.
15. L. Jurd, J. Corse, A. D. King Jr, H. Bayne and K. Mihara, Phytochemistry, 1971, 10, 2971.
16. N. Tomohiro, K. Yasuko and M. Sei-Itsu, Biochim. Biophys. Acta, Lipids Lipid Metab., 1983, 753, 130.
17. M. Payá, B. Halliwell and J. R. S. Hoult, Biochem. Pharmacol., 1992, 44, 205.
18. T. Kaneko, S. Tahara and F. Takabayashi, Biol. Pharm. Bull., 2003, 26, 840.
19. Y. Masamoto, H. Ando, Y. Murata, Y. Shimoishi, M. Tada and K. Takahata, Biosci., Biotechnol., Biochem., 2003, 67, 631.
20. K. N. Leung, P. Y. Leung, L. P. Kong and P. K. Leung, Cell. Mol. Immunol., 2005, 2, 181.
21. M. Payá, B. Halliwell and J. R. S. Hoult, Free Radical Res., 1992, 17, 293.
22. W. Li, V. Lynch, H. Thompson and M. A. Fox, J. Am. Chem. Soc., 1997, 119, 7211.
23. R. Anet, Can. J. Chem., 1962, 40, 1249.
24. S. R. Trenor, A. R. Shultz, B. J. Love and T. E. Long, Chem. Rev., 2004, 104, 3059.
25. J. N. Moorthy, K. Venkatesan and R. G. Weiss, J. Org. Chem., 1992, 57, 3292.
26. M. D'Auria and R. Racioppi, J. Photochem. Photobiol., A, 2004, 163, 557.
27. X. Yu, D. Scheller, O. Rademacher and T. Wolff, J. Org. Chem., 2003, 68, 7386.
28. G. J. Smith, R. J. Weston, Y. Tang, Y. Lu and J. M. Dyer, Aust. J. Chem., 2012, 65, 1451.
29. T. Moriya, Bull. Chem. Soc. Jpn., 1983, 56, 6.
30. K. Żamojć, D. Jacewicz and L. Chmurzyński, Anal. Lett., 2013, 46, 349.
31. K. Żamojć, W. Wiczk, B. Zaborowski, D. Jacewicz and L. Chmurzyński, J. Fluoresc., 2014, 24, 713.
32. K. Żamojć, W. Wiczk, B. Zaborowski, D. Jacewicz and L. Chmurzyński, Spectrochim. Acta, Part A, 2015, 136, 1875.
33. K. Żamojć, W. Wiczk, B. Zaborowski, M. Makowski, J. Pranczk, D. Jacewicz and L. Chmurzyński, Spectrochim. Acta, Part A, 2014, 133, 887.
34. D. R. Corbin, G. D. Stucky, W. S. Willis and E. G. Sherry, J. Am. Chem. Soc., 1982, 104, 4298.
35. M. M. Birau and Z. Y. Wang, Tetrahedron Lett., 2000, 41, 4025.
36. K. Gnanaguru, N. Ramasubbu, K. Venkatesan and V. Ramamurthy, J. Org. Chem., 1985, 50, 2337.
37. J. P. Kutney, G. Engendorf, T. Inaba and D. L. Dreyer, Org. Mass Spectrom., 1971, 5, 249.
38. J. P. Kutney, T. Inaba and D. L. Dreyer, J. Am. Chem. Soc., 1968, 90, 813.
39. D. Lejeune, M. Hasanuzzaman, A. Pitcock, J. Francis and I. Sehgal, Mol. Cancer, 2006, 5, 21.
40. A. Sunkaria, D. R. Sharma, W. Y. Wani and K. D. Gill, ACS Chem. Neurosci., 2014, 5, 115.

---