Carbazole-based two-photon fluorescent probe for selective imaging of mitochondrial hydrogen peroxide in living cells and tissues

Abstract

This paper reported a two-photon fluorescent probe for mitochondrial H₂O₂ detection and imaging based on the incorporation of a sensing boronate ester and an organelle-targeting triphenylphosphonium moiety onto the carbazole fluorophore. The probe exhibits a fast “turn on” response, good selectivity toward H₂O₂ and high specificity for mitochondria.

---

As a typical ROS, hydrogen peroxide (H₂O₂) can mediate diverse physiological responses and serve as the second messenger in normal cellular signal transduction.¹ In mammalian cells, mitochondria have been demonstrated as main venues for the production of H₂O₂ and the generated H₂O₂ can participate in redox signaling through reversible oxidation of cysteine residues in target proteins.² Excessive production of H₂O₂ in mitochondria will induce oxidative damage and cause physiological changes and diseases, such as cancer, cardiovascular disorders, and Alzheimer's disease.³ Therefore, specific, localizable and readily deployable tools for reliable and precise intracellular measurement of H₂O₂ are imperative. Up to now, a variety of analytical means, such as spectrophotometry, electrochemistry, and MRI⁴ together with recognition principles have been used to develop effective hydrogen peroxide sensors. The fluorescence assay for H₂O₂ detection in human physiology and pathology, which employs H₂O₂-triggered oxidative cleavage of boronate to phenol, sulfonate group hydrolysis,⁵ triphenylphosphine oxidation,⁶ Baeyer–Villiger type reaction⁷ and metal complex mediated reaction,⁸ has shown an extensive prospect because of the high sensitivity and specificity as well as capability of real-time and in situ imaging analysis. Although fluorescent probes were well competent to image the spatial and temporal distribution of cellular H₂O₂, how to maintain the cell viability following prolonged exposure to excitation illumination was still a challenge.

Two-photon microscopy (TPM) can record the most subtle internal structure deep into tissue with low cell damage by using two near-infrared photons for the excitation, which provides greater tissue penetration depth and higher spatial resolution.⁹ A variety of TP probes equipped with an organelle-targeting moiety such as tertiary amine, pyridinium or triphenyl-phosphonium group for labeling lysosomes and mitochondria have been developed which provided organelle-specific imaging with good spatial and temporal resolution. The superior tracing and imaging functions of TPM can be used for determining the spatial and temporal dynamics of reactive oxygen metabolites in a living system, especially for elucidating the diverse roles of H₂O₂ in a complex biological environment. It is particularly beneficial to use TPM for reactive oxygen species (ROS) and reactive nitrogen species (RNS) detections in a complex biological system. Some TP fluorescent probes have been shown to image hypochlorous acid,¹⁰ superoxide anion¹¹ and nitric oxide¹² in cellular environment. Up to date, there are only a handful of TP probes designed for intracellular H₂O₂ imaging,¹³ especially for mitochondrial or lysosomal H₂O₂.¹⁴ Because of the rapid metabolic process and constantly changed concentration of intracellular H₂O₂, there is a great demand for developing TPM probes to monitor the H₂O₂ production and consumption in mitochondria with fast response time and low detection limit to better understand the cellular and physiological roles of mitochondrial H₂O₂.

In this work, a TP fluorescent probe CBZ-H₂O₂ was synthesized by incorporating a boronate functionality as a H₂O₂ recognition site and the triphenylphosphonium group as mitochondrial targeting moiety into the fluorescent carbazole framework. The presence of boronic ester group at the 3-position of a carbazole scaffold would quench the TP fluorescence of CBZ-H₂O₂. After treatment with H₂O₂, a highly fluorescent product would be subsequently yielded due to the chemospecific deprotection of boronate to hydroxyl functionality. Compared with the existing TP probes for mitochondrial H₂O₂, CBZ-H₂O₂ exhibited a fast response time (∼15 min), improved sensitivity and low interference from other ROS. Importantly, it can be applied for direct imaging of mitochondrial H₂O₂ in living cells and frozen slice of rat liver with satisfactory imaging results.

The TP probe, CBZ-H₂O₂ was developed for localized detection of H₂O₂ concentration variations in mitochondrial. The synthetic routes and recognition principle were shown in Schemes 1 and 2, respectively. The molecular structure of CBZ-H₂O₂ was fully characterized with ¹H NMR and ¹³C NMR as well as ³¹P NMR (Fig. S1–S4 in the ESI†). We then investigated the absorption and fluorescence properties of CBZ-H₂O₂ in PBS/DMSO (4/1 v/v, 20 mM, pH 7.4) solution at room temperature. As shown in Fig. S5,† the maximum absorbance at 340 nm of CBZ-H₂O₂ was gradually shifted to 360 nm after adding H₂O₂, with slightly enhanced absorbance. The same trend was followed for the maximum fluorescence intensity at 396 nm. In the presence of H₂O₂, the emission band exhibited an apparent 35 nm red shift and the fluorescence intensity dramatically increased nearly four times. The sensitive spectral response confirmed the specific reaction between CBZ-H₂O₂ and H₂O₂.

Scheme 1 Synthesis of probe CBZ-H₂O₂. Reagents and conditions: (a) NBS, DMF, 88%; (b) NaOH, 1,4-dibromobutane, toluene/H₂O, 86%; (c) Pd(dppf)Cl₂, dioxane, 50%; (d) triphenylphosphine, acetonitrile, 90%.

Scheme 2 The recognition mechanism of CBZ-H₂O₂ to H₂O₂ in mitochondria.

In order to optimize the pH value for the best response, we then investigated the fluorescence response of CBZ-H₂O₂ toward H₂O₂ in buffer solution at different pH ranging from 4.0 to 10.0. The intensity of CBZ-H₂O₂ remained consistent at different pH values. Upon treatment with H₂O₂, compared with the acidic region, the fluorescence intensity displayed a more noticeable enhancement in the alkaline region between 7.0 and 10.0 as shown in Fig. S6.† Therefore, it is advantageous to carry out the experiments in the buffer solution with pH value of 7.4, which is promising for biological applications.

Fluorescence titration of H₂O₂ was conducted using 1.0 μM of CBZ-H₂O₂ and different concentration of H₂O₂ in PBS (20 mM, pH 7.4) containing 20% DMSO. As shown in Fig. 1A, the fluorescence response of probe was highly sensitive to the amount of H₂O₂. When the concentration of H₂O₂ went to 80 μM, the fluorescence enhancement reached a plateau, and further increase in the concentration would not boost the signal. Moreover, the dose-dependent fluorescence enhancement showed a good linearity between the F/F₀ and the concentrations of H₂O₂ in a wide range from 0 to 30 μM with a detection limit (3σ/slope) of 1.96 μM (Fig. 1B, inset), where F₀ and F corresponded to the fluorescence intensities at 430 nm of CBZ-H₂O₂ before and after adding H₂O₂, respectively. Compared with the reported fluorescent probes for H₂O₂, CBZ-H₂O₂ presents an improved detection sensitivity, which exhibited the potential to be used in monitoring physiological concentration of mitochondrial H₂O₂.

Fig. 1 (A) Fluorescence emission spectra of CBZ-H₂O₂ (1.0 μM) in the presence of different concentrations of H₂O₂ (0–80 μM) in PBS (20 mM, pH 7.4) containing 20% DMSO. (B) Calibration curve of CBZ-H₂O₂ as a function of H₂O₂ concentrations. The curve was plotted with the fluorescence intensity vs. H₂O₂ concentration after incubation for 15 min. (C) Real-time fluorescence intensity records of CBZ-H₂O₂ (1.0 μM) upon addition of H₂O₂ (50 μM) in PBS (20 mM, pH 7.4). λ_ex/λ_em = 350 nm/430 nm. (D) Fluorescence responses of 1.0 μM CBZ-H₂O₂ to various reactive oxygen species (O₂˙⁻, HO˙, ClO⁻, ¹O₂), reactive nitrogen species (NO, NO₂⁻, ONOO⁻), TBHP, O^tBu, and GSH.

The real-time kinetic response of CBZ-H₂O₂ toward H₂O₂ was measured by recording the fluorescence emission as a function of time. As shown in Fig. 1C, in the first three minutes, the fluorescence intensity of CBZ-H₂O₂ remained steady and constant under irradiation. Upon the addition of H₂O₂, a drastic fluorescence enhancement was observed and the intensity reached plateaus within 15 minutes, indicating that the reaction proceeded faster than a majority of reported H₂O₂ probes. Considering that ROS and RNS are involved in a wide range of physiological and pathological processes, the selectivity of CBZ-H₂O₂ over ROS, RNS and other potentially competing species was investigated. Treatment of CBZ-H₂O₂ with other ROS (O₂˙⁻, HO˙, ClO⁻, ¹O₂), RNS (NO, NO₂⁻, ONOO⁻), TBHP, O^tBu, and GSH afforded negligible effects on the fluorescence intensity of CBZ-H₂O₂ as shown in Fig. 1D. Only the presence of H₂O₂ could induce dramatically intensity enhancement, suggesting that CBZ-H₂O₂ exhibited a good selectivity toward H₂O₂ in complex physiological conditions.

To investigate the reaction mechanism of CBZ-H₂O₂ with H₂O₂, the differences of ¹H NMR spectra of CBZ-H₂O₂ and the corresponding H₂O₂-reaction product were carefully analyzed and compared as shown in Fig. S7.† Upon the addition of H₂O₂, the resonance signal of CBZ-H₂O₂ at 8.44 ppm gradually moved to the high field, while a new and distinct resonance signal appeared and increased in intensity at 9.04 ppm over a period of 30 min. These changes of NMR signals corresponded to the phenolic proton of the reaction product, supporting the transformation of carbazole boronate to its corresponding phenol in the presence of H₂O₂.

With a good understanding of the sensitivity and selectivity of spectroscopic responses of the CBZ-H₂O₂ probe for detecting H₂O₂in vitro, we sought to apply it for the detection of H₂O₂ in living cells. Before carrying out the cell imaging experiments, the cytotoxicity of CBZ-H₂O₂ with concentrations from 0 to 20 μM and its reaction product with H₂O₂ were determined using MTT viability assay. As shown in Fig. S8,† CBZ-H₂O₂, and its reaction product with H₂O₂ showed low toxicity to HeLa cells, with relatively high cell viability greater than 85% after incubation for 24 h. The results suggested that CBZ-H₂O₂ probe showed superior biocompatibility for cell imaging. To further investigate the application of CBZ-H₂O₂ in bio-imaging, the TP action cross-section (Φδ_max) was determined from the two-photon excited fluorescence (TPEF) spectra of CBZ-H₂O₂ in the absence and presence of H₂O₂ by using rhodamine-6G in methanol as the reference, as shown in Fig. S9,† the Φδ_max of CBZ-H₂O₂ is 4.8 GM at 720 nm, but Φδ_max increased to 65 GM after addition of H₂O₂. Such a drastic enhancement in TPEF indicates that CBZ-H₂O₂ is a potential two-photon probe for imaging H₂O₂ in living specimens.

To investigate the mitochondria-targeting function of CBZ-H₂O₂ and its application as a TP probe for detecting H₂O₂ level changes in mitochondria, a co-localization experiment was carried out to confirm mitochondria-localization specificity of CBZ-H₂O₂ in HeLa cells by co-staining it with Mito Tracker red (1.0 μM), a commercially available mitochondria tracker. As shown in Fig. 2, the red fluorescent from Mito Tracker (2c, red channel, λ_ex = 590 nm; λ_em = 620–660 nm) and the blue fluorescent from CBZ-H₂O₂ in H₂O₂ solution (2b, blue channel, λ_ex = 720 nm; λ_em = 420–460 nm) overlapped very well, and the Pearson's co localization coefficient was calculated at 0.925. The results indicated that CBZ-H₂O₂ was distributed mainly in the mitochondria with mitochondria-positioning capability.

Fig. 2 Localization of CBZ-H₂O₂ in the mitochondrial of HeLa cells. Cells were stained with (a) bright field, (b) 10.0 μM CBZ-H₂O₂ with H₂O₂ solution (50.0 μM) for 30 min at 37 °C (λ_ex = 720 nm, λ_em = 420–460 nm). (c) 1.0 μM Mito Tracker red (λ_ex = 590 nm, λ_em = 620–660 nm). (d) Overlay of images (b) and (c).

To demonstrate the potential bio-applications of CBZ-H₂O₂, the ability of probe to track endogenous burst of H₂O₂ produced within HeLa cells was evaluated. Firstly, CBZ-H₂O₂ was incubated with live HeLa cells for 30 min to guarantee completely entry in cells. As shown in Fig. 3, two-photon confocal fluorescence images taken at an excitation of 720 nm for HeLa cells incubated with 10.0 μM CBZ-H₂O₂ indicated that the probe was cell-permeable. When HeLa cells were stained with CBZ-H₂O₂ only, the image displayed extremely weak fluorescence induced by the intrinsic H₂O₂ produced by cell metabolism. Further more, the endogenous H₂O₂ was produced by stimulating HeLa cells with phorbol myristate acetate (PMA), which induced a phagocytosis-associated H₂O₂ generation.¹⁵ And remarkable fluorescence enhancement from the cells was observed when the cells was stimulated with PMA (1.0 μg mL⁻¹) before introducing CBZ-H₂O₂. It is evident that the obvious signal enhancement is mainly due to the reaction between CBZ-H₂O₂ and endogenous H₂O₂. In order to further validate the main functions of H₂O₂ for signal increases, HeLa cells were then treated with DMSO (0.5%) as a ROS scavenger before loading CBZ-H₂O₂ for 30 min.¹⁶ The presence of CBZ-H₂O₂ in cells did not show any fluorescence response due to the clearance of intracellular H₂O₂.

Fig. 3 Fluorescence images of HeLa cells using CBZ-H₂O₂ under different conditions. (a) Cells incubated with CBZ-H₂O₂ only; (b) cells pretreated with PMA (1.0 μg mL⁻¹) and then loading CBZ-H₂O₂ for 30 min; (c) cells followed treated with DMSO (0.5%) and then loading CBZ-H₂O₂ for 30 min; (d) cells incubated with CBZ-H₂O₂ and adscititious H₂O₂.

Encouraged by the success of cell imaging with CBZ-H₂O₂ for mitochondrial H₂O₂, we further investigated whether it could be used for H₂O₂ imaging in deep-tissue. The images were obtained from 1.0 mm-thick rat liver frozen slice which was treated with CBZ-H₂O₂ before and after incubating with 50 μM H₂O₂ at 37 °C. As shown in Fig. 4, the slice treated with CBZ-H₂O₂ only exhibited feeble fluorescence signal, whereas the addition of H₂O₂ to the slice distinctly intensified the fluorescence signal, clearly demonstrating the capability of CBZ-H₂O₂ for H₂O₂ detection in tissue. Subsequently, to further investigate the utility for deep tissue imaging, the 3D two-photon confocal fluorescence imaging along the z-direction were taken by treating the slice with CBZ-H₂O₂ in the absence and presence of H₂O₂. The Z-scanning confocal imaging showed that luminescence emission of CBZ-H₂O₂ present in the penetration depth of 40 μm, while the presence of H₂O₂ in tissue made the emission deeper to 105 μm, indicating that CBZ-H₂O₂ has excellent capacity of monitoring H₂O₂ at a greater depth of tissue using two-photon microscopy.

Fig. 4 TP confocal microscopy bright-field images (left) and fluorescence images (middle) of rat liver frozen slice after incubation with CBZ-H₂O₂ and CBZ-H₂O₂ + H₂O₂, respectively. And 3D TP confocal fluorescence images (right) of CBZ-H₂O₂ and CBZ-H₂O₂ + H₂O₂ accumulated along the z direction at a depth of 0–40 μm and 0–105 μm with 40× magnification, respectively.

Conclusions

In summary, we have designed and synthesized a new two-photon fluorescent “turn-on” probe for mitochondrial H₂O₂ detection. The probe consists of a boronate ester functionality as a sensing site, a carbazole derivative as a fluorophore and a triphenylphosphonium moiety as a mitochondrial-targeting localizer. It exhibits not only fast response and high selectivity over other ROS, RNS and related biomolecules but also low detection limit of 1.96 μM. Co-localization experiments clearly demonstrated that CBZ-H₂O₂ selectively localizes to and accumulates in the mitochondria. CBZ-H₂O₂ is also sensitive to both exogenous and endogenous H₂O₂. By using two-photon fluorescence microscopy, it can detect H₂O₂ in living cells and tissues at a depth of 105 μm for a long period of time with excellent imaging quality. This probe will allow in-depth investigation of the roles and functions of H₂O₂ in complex biological systems.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21305036, 21507135), Hunan Provincial Natural Science Foundation of China (2015JJ3035), China Postdoctoral Science Foundation funded project (2015M580685) and Anhui Province Natural Science Foundation of China (1608085MB30).

Notes and references

1. B. Halliwell and J. M. C. Gutteridge, Free Radicals in Biology and Medicine, Clarendon Press, Oxford, UK, 3rd edn, 1999.
2. L. J. Chen, R. Na, M. J. Gu, A. B. Salmon, Y. H. Liu, H. Y. Liang, W. B. Qi, H. V. Remmen, A. Richardson and Q. T. Ran, Aging Cell, 2008, 7, 866.
3. (a) H. K. Seitz and F. Stickel, Nat. Rev. Cancer, 2007, 7, 599; (b) H. Ohshima, M. Tatemichi and T. Sawa, Arch. Biochem. Biophys., 2003, 417, 3; (c) A. M. Shah and K. M. Channon, Heart, 2004, 90, 486; (d) M. T. Lin and M. F. Beal, Nature, 2006, 443, 787.
4. (a) S. L. Hempel, G. R. Buettner, Y. Q. O'Malley, D. A. Wessels and D. M. Flaherty, Free Radical Biol. Med., 1999, 27, 146; (b) M. Y. Hua, H. C. Chen, R. Y. Tsai, Y. L. Leu, Y. C. Liu and J. T. Laie, J. Mater. Chem., 2011, 21, 7254.
5. (a) A. R. Lippert, K. Y. Keshari, J. Kurhanewicz and C. J. Chang, J. Am. Chem. Soc., 2011, 133, 3776; (b) G. C. Van de Bittner, C. R. Bertozz and C. J. Chang, J. Am. Chem. Soc., 2013, 135, 1783; (c) M. C. Chang, A. Pralle, E. Y. Isacoff and C. J. Chang, J. Am. Chem. Soc., 2004, 126, 15392; (d) B. C. Dickinson, C. Huynh and C. J. Chang, J. Am. Chem. Soc., 2010, 132, 5906; (e) E. W. Miller, A. E. Albers, A. Pralle, E. Y. Isacoff and C. J. Chang, J. Am. Chem. Soc., 2005, 127, 16652; (f) H. Maeda, Y. Fukuyasu, S. Yoshida, M. Fukuda, K. Saeki, H. Matsuno, Y. Yamauchi, K. Yoshida, K. Hirata and K. Miyamoto, Angew. Chem., Int. Ed., 2004, 43, 2389; (g) K. Xu, B. Tang, H. Huang, G. Yang, Z. Chen, P. Li and L. An, Chem. Commun., 2005, 48, 5974.
6. N. Soh, O. Sakawaki, K. Makihara, Y. Odo, T. Fukaminato, T. Kawai, M. Irie and T. Imato, Bioorg. Med. Chem., 2005, 13, 1131.
7. M. Abo, Y. Urano, K. Hanaoka, T. Terai, T. Komatsu and T. Nagano, J. Am. Chem. Soc., 2011, 133, 10629.
8. (a) T. Zhang, H. Fan, G. Liu, J. Jiang, J. Zhou and Q. Jin, Chem. Commun., 2008, 5414; (b) O. S. Wolfbeis, A. Dü rkop, M. Wu and Z. Lin, Angew. Chem., Int. Ed., 2002, 41, 4495.
9. (a) H. M. Kim and B. R. Cho, Acc. Chem. Res., 2009, 42, 863; (b) F. Helmchen and W. Denk, Nat. Methods, 2005, 2, 932; (c) J. M. Squirrell, D. L. Wokosin, J. G. White and B. D. Bavister, Nat. Biotechnol., 1999, 17, 763.
10. (a) W. R. Zipfel, R. M. Williams and W. W. Webb, Nat. Biotechnol., 2003, 21, 1369; (b) L. Li, X. Shen, Q. Xu and S. Q. Yao, Angew. Chem., Int. Ed., 2013, 52, 424; (c) H. Ahn, K. E. Fairfull-Smith, B. J. Morrow, V. Lussini, B. Kim, M. V. Bondar, S. E. Bottle and K. D. Belfield, J. Am. Chem. Soc., 2012, 134, 4721; (d) H. C. Heo, K. H. Kim, H. J. Kim, S. H. Baik, H. Song, Y. S. Kim, J. Lee, I. Mook-jung and H. M. Kim, Chem. Commun., 2013, 49, 1303; (e) Q. L. Xu, C. H. Heo, G. Kim, H. W. Lee, H. M. Kim and J. Yoon, Angew. Chem., Int. Ed., 2015, 54, 4890.
11. P. Li, W. Zhang, K. X. Li, X. Liu, H. B. Xiao, W. Zhang and B. Tang, Anal. Chem., 2013, 85, 9877.
12. (a) X. H. Dong, C. H. Heo, S. Y. Chen, H. M. Kim and Z. H. Liu, Anal. Chem., 2014, 86, 308; (b) H. B. Yu, Y. Xiao and L. J. Jin, J. Am. Chem. Soc., 2012, 134, 17486.
13. (a) C. Chung, D. Srikun, C. S. Lim, C. J. Chang and B. R. Cho, Chem. Commun., 2011, 47, 9618; (b) K. M. Zhang, W. Dou, P. X. Li, R. Shen, J. X. Ru, W. Liu, Y. M. Cui, C. Y. Chen, W. S. Liu and D. C. Bai, Biosens. Bioelectron., 2015, 64, 542; (c) H. C. Guo, H. Aleyasin, S. S. Howard, B. C. Dickinson, V. S. Lin, R. E. Haskew-Layton, C. Xu, Y. Chen and R. R. Ratan, J. Biomed. Opt., 2013, 18, 106002.
14. (a) G. Masanta, C. H. Heo, C. S. Lim, S. K. Bae, B. R. Cho and H. M. Kim, Chem. Commun., 2012, 48, 3518; (b) J. Jing and J. L. Zhang, Chem. Sci., 2013, 4, 2947.
15. (a) H. R. Petty, Biochim. Biophys. Acta, 1989, 1012, 284; (b) P. Bellavite, Free Radical Biol. Med., 1988, 4, 225.
16. M. Y. Song, L. Z. Zeng, X. J. Hong, Z. H. Meng, J. F. Yin, H. L. Wang, Y. Liang and G. B. Jiang, Environ. Sci. Technol., 2013, 47, 2886–2891.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra21260c

‡ These author contributed equally to this work.

---