Strong red fluorescent probes suitable for detecting hydrogen peroxide generated by mice peritoneal macrophages

Abstract

This paper reports the synthesis, fluorescence properties, and biological applications of naphthofluorescein disulfonate (NFDS-1), as a red fluorescence imaging probe to detect intracellular H₂O₂.

---

The far-visible and the near-IR spectral regions (600–1000 nm), where only a few classes of molecules exhibit significant absorption, and they do not contribute to the fluorescence signal, are areas of low background fluorescence interference in biological systems.^1,2 The features of the particular spectral region make it ideal for using a fluorogenic probe to detect reactive oxygen species (ROS) in living cells. However, as far as we know, only one probe in this region has been reported so far, for detecting nitric oxide.²

In the current work, we aim to develop molecular probes that could be used to detect reactive oxygen species in this spectral region. Reactive oxygen species such as superoxide radical anion, hydrogen peroxide, hydroxyl radical, nitric oxide and peroxynitrite play a vital role in physiology.³ A rapid rise in intracellular oxidant levels under oxidative stress could cause damage to biological molecules and result in various diseases.^4–6 Hydrogen peroxide is the precursor to other ROS, and its homeostasis can have diverse physiological and pathological consequences.⁷ However, how this damage occurs is insufficiently understood even in the simplest eukaryotic organisms.⁸

In order to explore the role of hydrogen peroxide in toxicology and human diseases, it is necessary to establish an accurate way of detecting the ROS, especially in living cells. Many probes such as dihydro-analogues of fluorescent dyes [e.g. dichlorofluorescin diacetate (DCFDA), dihydrorhodamine 123,^9,10 phosphine-based fluorophores,^11,12 lanthanide coordination complexes,¹³ and chromophores with ROS-cleavable protecting groups^14,15] have been developed in recent years. DCFDA is the most popular one, and has been used frequently to detect cell-derived H₂O₂,⁹ but it suffers from a major drawback in that it is poorly selective toward H₂O₂ owing to its autoxidation and reaction with other ROS or peroxidase.

Recently, an interesting method for H₂O₂ detection was developed based on the selective H₂O₂-mediated transformation of monopentafluorobenzenesulfonyl fluorescein to fluorescein.¹⁵ The method relies on simple deprotection, not on oxidation, which allows the highly specific and peroxidase-independent detection of H₂O₂ under the complicated oxidative circumstances. However, there are two disadvantages in the fluorescein-based probe: the weak fluorescence of the probe itself and the high background fluorescence interference in biological systems reduced measurable sensitivity (the detection limit is 92.3 nmol or higher). In vivo, hydrogen peroxide concentrations are usually considered to be in the lower nanomolar range. Therefore, it is necessary to develop probes that have higher selectivity and enough sensitivity for H₂O_2.

We designed and synthesized naphthofluorescein disulfonates (NFDS-1, NFDS-2, Scheme 1) as fluorescence imaging probes for intracellular H₂O₂, which were characterized with elemental analysis, IR, and ¹H NMR. NFDS are closed, colorless, and non-fluorescent lactones. Upon treatment with H₂O₂, hydrolytic deprotection of NFDS would subsequently generate open, colored, and fluorescent products.

Scheme 1 The synthesis of fluorescent probes and their reaction with H₂O₂.

The NFDS-2 was chosen for the following reasons: First, owing to the effect of perfluorooctanesulfonic acid, a surfactant generated from the hydrolysis of NFDS-2 in the reactive circumstances, the excitation and emission spectra of the product, naphthofluorescein, generated from reaction of NFDS-2 with H₂O₂ would have a red shift compared with NFDS-1, theoretically. Second, the perfluorooctane chain would enhance the reactivity of the NFDS-2 toward H₂O₂ owing to the lipophilic characteristics of fluorine atoms and hydrogen peroxide.

NFDS-1 and NFDS-2 were evaluated under chemical circumstances. A solution of NFDS-1 (50 µM) or NFDS-2 (50 µM) in DMSO was diluted 10 times with 2-[4-(hydroxyethyl)-1-piperazinyl]ethanesulfonic acid (HEPES) buffer (0.1 M, pH 7.4). Treatment of each of the solutions (5 µM) containing the probe compounds with H₂O₂ (1 µM) at 37 °C for 40 min gave fluorescence responses, Fig. 1. With different concentrations of H₂O₂ we obtained the NFDS-1 linear calibration curve from 6.0 × 10⁻⁹ to 4.0 × 10⁻⁶ M and the detection limit (81.5 pM) which was lower than for reported probes (92.3 nM).¹⁵ We believe that the lower detection limit of NFDS-1 for H₂O₂ was based on its low background interference and the non-fluorescence of the probe itself. The experiments also showed that the fluorescence property of the product generated from reaction of NFDS-2 with H₂O₂ is not stable enough owing to the concentration change of the surfactant, perfluorooctanesulfonic acid, generated from the hydrolysis of NFDS-2, as shown in Fig. 1(b). Therefore, in the following experiments, we have focused our attention on NFDS-1.

Fig. 1 (a) Emission spectra of 5 µM NFDS-1 with 1 µM H₂O₂ (A) and blank (B) in HEPES (pH = 7.4) at 37 °C for 40 min (λ_em = 662 nm). (b) Emission spectra of 5 µM NFDS-2 with 1 µM H₂O₂ (A) at 37 °C for 40 min, 45 min and 50 min, and blank (B) at 37 °C for 40 min (λ_em = 692 nm).

The reactivity of NFDS-1 toward various ROS, reductant, glutathione (GSH), and esterase was studied in detail. The fluorescent response from the solution of NFDS-1 (10 µM, in HEPES buffer) with H₂O₂ (2 µM) at 662 nm with excitation at 602 nm after incubation at 37 °C for 40 min was compared to those of reactions with NaOCl, t-BuOOH, esterase, 1,4-hydroquinone, GSH, 3-(aminopropyl)-1-hydroxy-3-isopropyl-2-oxo-1-triazene (NOC-5), 3-morpholinosydnonimine (SIN-1), and ascorbic acid (Vc) (final 10 µM each), and with superoxide radical anion or hydroxyl radical. Superoxide radical anions were generated by the enzymatic reaction of hypoxanthine (HPX, 1 ml, 10 µM) with xanthine oxidase (XO, 0.2 ml, 1 U ml⁻¹) or KO₂ (10 µM). The Fenton reaction between H₂O₂ (5 µM) and Fe²⁺ ion (10 µM) was used as the source of ˙OH. All results are summarized in Table 1. The experiments showed that NFDS-1 provided a highly specific fluorescent response toward H₂O₂, while giving a small response to ascorbic acid, glutathione, esterase, and other ROS, especially to hydroxyl radical. We suggest that the observed selectivity of NFDS-1 for H₂O₂ over more oxidizing ROS is based on simple deprotection, not on an oxidative mechanism.

Table 1 Relative fluorescence intensity (RFI) observed upon reaction of NFDS-1 with various ROS, and biological compounds

Compounds	RFI^a	Compounds	RFI^a
a All data obtained at 662 nm after incubation at 37 °C for 40 min.
Blank	10	ONOO^− (SIN-1)	8
H2O2	155	t-BuOOH	35
O2^−˙ (XO/HPX)	20	1,4-hydroquinone	8
O2^−˙ (KO2)	42	GSH	8
˙OH	5	Ascorbic acid	11
^−OCl	34	Esterase	10
NO˙ (NOC-5)	7

---

We next assessed the reaction of NFDS-1 for H₂O₂ generated in living cells. Separated mice peritoneal macrophages (PM) were seeded onto a glass slide. Then the cells were loaded with NFDS-1 (10 µM, DMSO–HEPES buffer, pH 7.4) by incubation at 37 °C for 30 min and showed negligible intracellular background fluorescence, Fig. 2(a). Probe-loaded macrophages were stimulated with phorbol myristate acetate (PMA: 2 ng ml⁻¹) at 37 °C for 10 min, and a strong signal was observed, Fig. 2(b). In addition, prompt fluorescent increases in probe-loaded macrophages treated with exogenous H₂O₂ (10 or 100 nM, respectively) were also observed, Fig. 2(c) and (d). Brightfield transmission measurements after NFDS-1 incubation and 100 nM H₂O₂ addition confirm that the cells are viable throughout the imaging experiments, Fig. 2(e). These data establish that NFDS-1, a small lipophilic molecule, is membrane-permeable, and can respond to nanomolar change in H₂O₂ concentration within living cells, while native cellular species such as GSH and ascorbic acid, as shown in Fig. 2(a), do not contribute to the fluorescence imaging.

Fig. 2 Confocal fluorescence and phase contrast images of live PM. (a) Fluorescence image of PM incubated with 10 µM NFDS-1 at 37 °C for 30 min. (b) Fluorescence image of probe-stained PM stimulated with PMA (2 ng ml⁻¹) at 37 °C for 10 min. (c) Fluorescence image of probe-stained PM treated with H₂O₂ (10 nM) at 37 °C for 10 min. (d) Fluorescence image of probe-stained PM treated with H₂O₂ (100 nM) at 37 °C for 10 min. (e) Brightfield image of live PM after NFDS-1 incubation and 100 nM H₂O₂ addition to confirm viability (Scale bar = 10 µM).

In conclusion, we have presented in this article the synthesis, fluorescence properties, and biological applications of NFDS-1, as a fluorescence imaging probe to detect intracellular H₂O₂. This naphthofluorescein-based reagent features high selectivity for H₂O₂ over other intracellular ROS and some biological compounds, a wide dynamic response range and low detection limit owing to its nonredox mechanism, and far-visible excitation and near-IR fluorescence emission profiles to minimize cell and tissue damage while avoiding native fluorescence from native cellular species. Furthermore, we have demonstrated the value of this probe by measuring living cell-derived H₂O₂ and the nanomolar concentration of exogenous H₂O₂ within living macrophages. We found that NFDS-1 is an excellent fluorescence reagent in detecting intracellular H₂O₂ and can respond to nanomolar change in H₂O₂ concentrations within living cells. Current efforts are devoted toward applying NFDS-1 in different biological systems to explore its potential applications. We believe that such a naphthofluorescein-based fluorescent dye will have an important application in detecting oxidative stress through direct intracellular imaging.

We thank Dr Chun-ming Liu from Plant Research International, Wageningen, The Netherlands for critical comments on the manuscript. This work was supported by research grants from the Important Project of National Natural Science Foundation of P. R. China (Nos. 20335030 and 90401019) and the Important Project of Natural Science Foundation of Shandong Province in China (No. Z2003B01).

Notes and references

1. G. Patonay and M. D. Antoine, Anal. Chem., 1991, 63(6), 321.
2. E. Sasaki, H. Kojima, H. Nishimatsu, Y. Urano, K. Kikuchi, Y. Hirata and T. Nagano, J. Am. Chem. Soc., 2005, 127, 3684.
3. T. Finkel and N. J. Holbrook, Nature, 2000, 408, 239.
4. H. Wiseman and B. Halliwell, Biochem. J., 1996, 313, 17.
5. J. M. McCord, Science, 1974, 185, 529.
6. K. Dobashi, B. Ghosh, J. K. Orak, I. Singh and A. K. Singh, Mol. Cell. Biochem., 2000, 205, 1.
7. B. Halliwell and J. M. C. Gutteridge, Free Radicals in Biology and Medicine, 3rd edn.; Clarendon Press: Oxford, UK, 1999.
8. T. Finkel, Curr. Opin. Cell Biol., 2003, 15, 247.
9. A. S. Keston, Anal. Biochem., 1965, 11, 6.
10. S. L. Hempel, G. R. Buettner, Y. Q. O'Malley, D. A. Wessels and D. M. Flaherty, Free Radical Biol. Med., 1999, 27, 146.
11. K. Akasaka, T. Suzuki, H. Ohrui and H. Meguro, Anal. Lett., 1987, 20, 731.
12. M. Onoda, S. Uchiyama, A. Endo, H. Tokuyama, T. Santa and K. Imai, Org. Lett., 2003, 5, 1459.
13. O. S. Wolfbeis, A. Dürkop, M. Wu and Z. Lin, Angew. Chem., Int. Ed., 2002, 41, 4495.
14. M. C. Y. Chang, A. Pralle, E. Y. Isacoff and C. J. Chang, J. Am. Chem. Soc., 2004, 126, 15392.
15. H. Maeda, Y. Fukuyasu, S. Yoshida, M. Fukuda, K. Saeki, H. Matsuno, Y. Yamauchi, Y. Yoshida, K. Hirata and K. Miyamoto, Angew. Chem., Int. Ed., 2004, 43, 2389.

---

Footnote

† Electronic supplementary information (ESI) available: detailed synthetic procedures and confocal fluorescence imaging. See DOI: 10.1039/b512440a

---