1. Introduction

Hydrogen peroxide (H₂O₂), as one of the most important markers for reactive oxygen species (ROS), can influence growth, development, and fitness of living organisms. At low levels (< 0.7 μmol/L), H₂O₂ also functions a ubiquitous intracellular second messenger [1, 2], which can activate signaling pathways to stimulate cell proliferation [3], differentiation [4] and migration [5]. H₂O₂ is produced through a few different pathways, such as NADPH oxidase, the electron transport chain in the mitochondria, various oxidases in peroxisomes and phagosomes in specialized cells of the immune system, and consumed by superoxide dismutase (SOD) or catalase enzymes in living systems. In living cells, the release and compartmentalization of H₂O₂ is precisely controlled and exquisitely regulated. However, under the conditions of stress or stimulation by exogenous chemicals, ROS generated aberrantly may attack cellular biomolecules, such as proteins [6, 7], liposomes [8], and DNA [9, 10] and induce oxidative stress. Indeed, oxidative stress is closely related with many diseases, including cancer [11, 12] and neurodegenerative disorders, such as Alzheimer's [13], Parkinson's [14], and Huntington's [15] diseases.

The involvement of H₂O₂ in cellular signaling and disease states motivated the development of chemical tools that can be employed to monitor the level change in situ. Among them, fluorescence labeling is a powerful technique, because of its nondisruptive, real-time and visualization features [16-31]. Compared with other fluorescent indicators, such as fluorescence proteins, quantum dots and nanoparticles, organic fluorescence probes are widely appreciated by chemists and biologists, due to their unique advantages: small size, easy modification, low toxicity and adjustable emission band.

In recent years, a variety of fluorescent probes have been developed for specifically monitoring H₂O₂ in vitro, ex vivo and in vivo, including five main response strategies: the reaction with arylboronates or phenylboronic acid to phenols [32-37], the conversion of oxonium or diketone to acid [38], oxidation of phenol to quinone [39], carbon-carbon double bond cleavage [40] and tandem Payne/Dakin reaction [41]. However, most of these probes enter cells through free diffusion, randomly distribute in the cytoplasm and lack organelle-targetable or accumulated capacity. However, the function of H₂O₂ is largely dependent on the subcellular compartment. Considering these demands, some targetable H₂O₂ fluorescent probes were designed and synthesized.

In this review, we summarized the reported targetable fluorescent probes that were used for the sensing H₂O₂ in mitochondria, nuclei and lysosomes (Scheme 1). The review is divided into three sections according to above different organelles. In every section, we further classify probes according to different targetable strategies. Finally, we state our perspective on the development of new probes for local visualization H₂O₂ in living cells.

2. Mitochondria-targetable H₂O₂ fluorescent probes

Mitochondria are considered the major source of H₂O₂ and primary cellular compartment of oxygen consumption [42, 43]. Mitochondria-targetable H₂O₂ probes were shown in Scheme 2.

Specifically, the mitochondrial membrane has typically a potential of -180 mV, therefore dyes with cationic ion can easily penetrate the mitochondrial membranes and subsequently accumulate inside it [44]. Positively charged groups, such as triphenylphosphonium (TPP), indolium moiety, quaternarized pyridine group, quaternarized quinoline unit and oxonium ion showed excellent mitochondrial target feature.

2.1. By linking to TPP head as mitochondrial bullet

In 2008, Chang et al. using rhodamine dye as a fluorophore and TPP head as a targetable group, designed and synthesized first mitochondria-directed H₂O₂ fluorescent probe 1 [45]. Reaction of probe 1 with H₂O₂ can trigger an enhanced fluorescence emission (λ_em = 528 nm). Owing to its chemospecific boronate switch, the probe showed highly selectivity for H₂O₂, over ROS like HOCl, O₂^- and NO. 100 μmol/L H₂O₂ addition resulted in a large increase of fluorescence signal in HeLa cells, meaning that probe 1 was able to detect H₂O₂ in living cells (Fig. 1). In co-staining experiments, probe 1 and a commercially available mitochondrial indicator (MitoTracker Deep Red) showed excellent overlap. This co-staining experiment was operated in a wide variety of cell types (HEK293, Cos-7 and CHO.K1 cell lines). Importantly, probe 1 was successfully used to visualize endogenous H₂O₂ increase in the mitochondria of living cells, upon stimulation with paraquat, a small-molecule inducer of Parkinson's-like phenotypes.

In 2012, Kim and coworkers developed first two-photon and ratiometric H₂O₂ fluorescent probe 2 [46] by introducing fluorophore based on 6-(benzo[d]thiazol-2'-yl)-2-(N, N-dimethylamino)naphthalene and TPP head. The boronate and TPP were modified at both ends of the molecule to minimize the interactions between them. After response to H₂O₂, the fluorescence peak of probe 2 in MOPS buffer (30 mmol/L MOPS, 100 mmol/L KCl, pH 7.4) changed from 470 nm to 545 nm. The ratio of the emission intensities at 400–470 nm (F_blue) and 530–600 nm (F_yellow) increased by 75-fold. The probe exhibited high sensitivity with 4.6 μmol/L detection limit. A co-localization cells imaging with commercially available MitoTracker Red in Raw 264.7 and high Pearson's colocalization coefficient (0.91) indicated that probe 2 accumulated predominantly in mitochondria. Importantly, this probe was capable of detecting changes in H₂O₂ levels in tissues at depths ranging from 100 to 180 mm using two-photon microscopy.

In 2016, Wong et al. also reported a two-photon fluorescent probe 3 [47] for mitochondrial H₂O₂ imaging based on carbazole fluorophore and TPP head. After treatment with H₂O₂, probe 3 changed into a highly fluorescent product (majored at 430 nm) due to the chemospecific deprotection of boronate to hydroxyl functionality. Compared with other H₂O₂ probes, probe 3 exhibited a fast response time (~15 min), improved sensitivity and low interference from another ROS. In HeLa cells, the red fluorescence from MitoTracker Red and the blue signal from probe 3 overlapped very well with Pearson's co-localization coefficient 0.925, confirming mitochondria-localization specificity of probe 3.

In 2019, an ICT-based fluorescent probe 4 [48] was synthesized for the detection of mitochondrial H₂O₂, with radical fluorescence intensity changes at 553 nm (up to 120-fold) and a large Stokes shift (93 nm). Probe 4 presented excellent cell permeability and mitochondria targeting ability. Furthermore, this probe was used for monitoring the change in H₂O₂ levels upon stimulation by m-chlorophenylhydrazone (CCCP), an inducer of apoptosis through a mitochondria-dependent pathway.

2.2. Including indolium moiety

For biological applications, TPP head is issue of concern due to the toxicity. Thus, other targeting moieties successively were developed and applied. In 2016, the Peng group, using a hemicyanine dye modified a borate ester moiety, developed a "turn-on" H₂O₂ fluorescent probe 5 [49] with near-infrared (NIR) emission. Upon the addition of 200 μmol/L H₂O₂, the absorption peak at 554 nm of probe 5 decreased and a new peak at 685 nm appeared. When excited with 685 nm and 554 nm respectively, the ratio of the fluorescence intensity at 709 nm (I685/I554) showed a significant increase from 3.0 to 30.8. High sensitivity and good specificity enabled probe 5 to be suitable for sensing endogenous H₂O₂ generated by autophagy with the stimulation of rapamycin (Fig. 2) in MCF7 cells. Fluorescence signal of probe 5 and Rhodamine 123 (a commercially available mitochondrial tracker) in MCF-7 cells overlapped well with a Pearson's coefficient of 0.98, which confirmed that probe 5 was localized in the mitochondria.

In the same year, Tang et al. also reported a "turn-on" H₂O₂ fluorescent probe 6 [50] based on hemicyanine dye. Unlike probe 5, probe 6 introduced the α-ketoamide moiety as a novel reaction switch specific for H₂O₂. Upon reaction with H₂O₂, probe 6 showed significant fluorescence enhancement (λ_ex/λ_em = 670/704 nm) and favorable characteristics, in terms of high specificity, high sensitivity and rapid response. In HepG2 cells, the red channel fluorescencesignal(probe 6)and that of green channel(MitoTracker Green) merged well, with 0.96 of Pearson's colocalization coefficient, indicating that probe 6 was organelle-specifically trapped in mitochondria. With the aid of probe 6, the visualization of H₂O₂ upregulation was realized in ischemiareperfusion injury models.

In 2016, Tang et al. also developed other mitochondriatargetable merocyanine-based H₂O₂ fluorescence probe 7 [51]. Upon addition of H₂O₂, the fluorescence intensity of probe 7 at 530 nm increased gradually with the fluorescence quantum yield increasing from 0.043 to 0.15. They also synthesized a novel zinc ion (Zn²⁺) fluorescence probe (emission peak at 779 nm) for imaging Zn²⁺ inside mitochondria in living cells and zebrafish. By simultaneous two-color fluorescence imaging utilizing these two probes, it was discovered that exogenous and endogenous H₂O₂ can stimulate an increase in the concentration of free Zn²⁺ within mitochondria (Fig. 3), which indicated for the first time that H₂O₂ and Zn²⁺ levels changed synergistically.

In 2016, Tang et al. developed probe 8 [52] to detect H₂O₂ in mitochondria. Probe 8 is comprised merocyanine as a fluorescence reporter and boronic acid as a specific response group. After reaction with H₂O₂, probe 8 showed an enhanced fluorescence emission (λ_em = 555 nm). For co-staining experiments in HepG2 and 4T1 cells, the fluorescence of probe 8 was well co-localized with that of MitoTracker Deep Red (overlap coefficient 0.90). At the same time, authors synthesized another probe for monitoring the level change of H₂O₂ in Endoplasmic Reticulum. Upon addition of H₂O₂, the maximum emission peak of this ER-targetable sensor exhibited a 100 nm red shift from 458 nm to 558 nm. By using the two probes for simultaneous fluorescence imaging of H₂O₂ levels, the results of variations changes of H₂O₂ levels in mitochondria and ER was shown, during diverse apoptotic stimuli, such as Lbuthionine sulfoximine (BSO), CCCP and Tunicamycin (Tm). This is the first example of simultaneous fluorescence imaging of H₂O₂ levels in different organelle during apoptosis.

2.3. Including pyridinium salt

Lin group introduced a carbazole-based fluorophore as a novel fluorescent dye for developing a mitochondra-targetable fluorescence probe 9 [53] for visualization the change of H₂O₂. Under excited at 405 nm, probe 9 showed a "turn-off" response, and the fluorescence intensity evidently decreased at 575 nm, with the addition of H₂O₂. It possessed a large Stokes shift and achieved the highly sensitive detection of H₂O₂ in aqueous solution. Co-location tests were employed to prove the mitochondrial target property.

Tang group [54] developed a mitochondria-targetable ratiometric probe 10 with a NIR emission and a large Stokes shift (357 nm), almost at the same time. The probe exhibited a weak emission band centered at 539 nm. Treatment with H₂O₂ induced the increase of a new emission band centered at 669 nm and disappearance of the emission at 539 nm. The intensity ratio (F₆₆₉/F₅₃₉) of probe 10 exhibited a good linear correlation with the concentration of H₂O₂. The mitochondria-targeting ability of probe was verified through co-localization assay with probe 10 and MitoTracker Green, with high intensity correlation analysis and Pearson's correlation coefficient (0.948) in A549 cells.

Zhang et al. constructed a mitochondria-targeted ratiometric fluorescent probe 11 [55] based on coumarin as the fluorophore and boronate moiety as the reaction site. In the absence of H₂O₂, probe 11 exhibited very strong red fluorescence due to pyridinium cation as the strong electron withdrawing group. After reactionwith H₂O₂, the fluorescence intensity at 640 nm a gradually decreased and a bright green-colored fluorescence was observed. The probe 11 can be quantitative, highly selective and sensitive recognition to H₂O₂ with lowdetectionlimit (10 nmol/L) in the solution, and can be used for determining H₂O₂ in living cells by ratiometric fluorescence imaging with excellent mitochondrial targeting.

In 2016, a mitochondria-targetable fluorescence probe 12 [56] was reported to dual-detection H₂O₂ and viscosity in mitochondria with a large turn-on fluorescence signal around 510 and 604 nm. The probe had high selectivity of H₂O₂ over other ROS and high sensitivity (around 67-fold). In addition, with the change of solution viscosity, the spectra of the probe showed obvious changes (around 16-fold fluorescence enhanced). Fluorescence imaging showed that the staining of probe 12 fitted well with that of MitoTracker Deep Red. The intensity scatter plot of probe 12 channel and MitoTracker channel was high correlated with a high overlap Pearson's coefficient of 0.93. These results indicated that the probe can be used to monitor the level of H₂O₂ and viscosity changes in mitochondria.

Ge's group developed a rosamine-based H₂O₂ probe 13 [57] with pyronine-pyridinium skeleton. Upon addition with H₂O₂, the probe exhibited "turn-on" response accompanied by 33-fold fluorescent enhancement majored at 604 nm. Confocal fluorescent imaging was performed in HeLa cells and Ges-1 cells with probe 13. Its emission matched well with the MitoTracker Green fluorescence, suggesting that the probe can be used for detecting mitochondrial H₂O₂.

Yi et al. synthesized a mitochondria-targeted H₂O₂ fluorescence probe 14 [58] based on positively charged styryl dye tagging with a DNA groove binding peptide and a 1, 8-naphthalimide-based H₂O₂ reporter. The binding peptide can facilitate the styryl group of probe 14 to intercalate into the stacks of DNA base pairs, and then release a red fluorescence emission (majored at 646 nm, λ_ex = 455 nm). The red fluorescence signal showed an excellent colocation with MitoTracker Green in HeLa (Fig. 4) and HepG2 cells, suggesting probe binding to mitochondrial DNA (mtDNA). In the presence of H₂O₂, probe 14 emitted green fluorescence centered at 555 nm. Thus, probe 14 can be used to reflect the change of the H₂O₂ level near mtDNA in living cells.

2.4. Others

In 2016, Shao and co-workers reported a novel mitochondriatargetable cationic probe 15 [59], by introducing carbazole group as the fluorogen; p-pinacolborylbenzyl moiety as the reaction site for H₂O₂; quaternarized quinoline unit as mitochondria-targeted carrier. After the addition with H₂O₂, the solution color of probe 15 changed from light-red to colorless. And, treatment with H₂O₂ triggers a dramatic increase of fluorescence intensity at 527 nm, with 33-fold increase. The "turn-on" fluorescence response showed high sensitivity, appreciable water solubility and rapid response time. In living cells, the probe signal fitted very well with the fluorescence of MitoTracker Deep Red, which confirmed that the dye was in the mitochondria. And the probe can respond to localized exogenous and endogenous changes of H₂O₂ levels in living cells.

Based on quaternarized quinoline unit as a mitochondriatargeted carrier, in 2018, Song's group developed another mitochondria-targeting H₂O₂ fluorescence probe 16 [60] in ratiometric manner. In the design of probe 16, 2-(2'-hydroxyphenyl)benzothiazoles (HBT) dye was attached to the nitrogen atom in quinoline moiety. Upon the addition of H₂O₂ to the solution of probe 16, the H₂O₂-triggered oxidation of borate made the emission peak changing from 666 nm to 594 nm. By confocal imaging experiments, it was verified that probe 16 could detect both endogenous and exogenous H₂O₂ in living cells. In addition, HeLa cells were treated with probe 16 and commercially available mitochondria probe Rhodamine 123. The finely merged image with a Pearson's colocalization coefficient of 0.94 clearly demonstrated that the probe was specific for mitochondria staining.

Zhou's group employed Nile-Red and DCM derivative to develop a new two-photon and near-infrared fluorescent probe 17 [61] for detecting mitochondrial H₂O₂. The fluorescence intensity was quenched by the boric acid moiety through suppress the intramolecular charge transfer (ICT) effect. Upon addition of H₂O₂, boric acid moiety be oxidized to the OH group, resulting in the recovery of ICT effect and turning on the fluorescence. The probe exhibited a more than ~105-fold fluorescence intensity enhancement at emission wavelength of 665 nm after reaction with H₂O₂. HeLa cells were co-incubated with probe 17 and Rhodamine 123, followed by treatment with H₂O₂, the fluorescent intensity obviously enhanced in the red channel (probe 17) and a good overlap with Rhodamine 123 (green channel). In addition, the probe was capable for image mitochondrial H₂O₂ in one-photo and two-photo manner.

3. Nuclei-targetable H₂O₂ fluorescent probes

Cell nucleus, the most critical organelle in living cell, houses the vast majority of DNA. Excess reactive oxygen speces can attack DNA and induce oxidative modifications of DNA bases (including oxidation of purines and pyrimidines, strand breaks and so on), finally forming mutations and even cancer (if unrepaired in a timely manner). However, nuclei-targeting fluorescent probes were extremely scarce. Currently, limited methods were used to construct nucleus-directed probes (Scheme 3).

3.1. Serendipitously discovered nucleus-targeted probe

In 2011, Chang's group serendipitously discovered a nucleustargeted and H₂O₂-specific fluorescent probe 18 [62] (Scheme 3). The probe features two major visible region absorptions (λ_abs = 468 nm, ε = 27, 300 L mol^-1 cm^-1; λ_abs = 490 nm, ε = 26, 000 L mol^-1 cm^-1) and a weak emission (λ_em = 530 nm). Reaction of probe 18 with H₂O₂ triggers a fluorescence increase. Confocal imaging of both HEK293 and HeLa cells loaded with probe 18 and the nuclear stain Hoechst 33342 show good colocalization of the two dyes, establishing that the probe can selectively accumulate in the nuclei. Similar results were observed in a wide variety of cell types, including HeLa, CHO.K1, NIH 3T3, Swiss 3T3, COS-7, A431, and Raw 264.7 Macrophages, demonstrating that probe 18 can visualize changes in localized nuclear H₂O₂ pools in a range of mammalian cell systems. With the aid of probe 18, a reduction in nuclear H₂O₂ levels in worms overexpressing sir-2.1 was revealed in in vivo imaging analysis.

3.2. Selective protein labeling technique

Another commonly used method of nuclear targeting is protein labeling. Selective protein labeling in living cells with synthetic compounds is feasible in either a covalent (such as HaloTag, SNAPTag, iarsenic compounds to tetracysteine haiprins) or noncovalent manner. In 2010, Chang and co-workers exploited a new class of organelle-targeted reporters for cellular H₂O₂ through the versatility of the SNAP-tag labeling technology. They synthesized SNAP-tag substrates for coupling to a permeable and selective H₂O₂ probe to obtain probe 19 [63] (Scheme 3). The probe exhibited a prominent absorption at 495 nm and fluorescent emission maximum at 513 nm. Localization of probe 19 to nuclei is achieved by fusion of SNAP-tag to the C-terminus of the histone H₂B protein (SNAP-H₂B). Imaging experiments using scanning confocal microscopy established nucleus target of the SNAP-tag probe and their fluorescence turn-on in response to changes in local H₂O₂ levels.

3.3. Nuclear localization signal peptide strategy

The third nucleus-targetable methodology is linking H₂O₂ dyes to the nuclear localization signal (NLS) peptides. The NLS peptides can transport cargo into the nuclei by specific recognition and binding to importin α, forming NLS-importin α-importin β trimer [64, 65], which mediates interactions with the pore complex and finally enters nuclei. In 2014, Yi's groupdeveloped a nucleus-directed H₂O₂ fluorescent probe 20 [66] (Scheme 3) based on VQRKRQKLMP peptide, which derived from the transcription factor NF-κB and can internalize NF-κB into the nuclei. Yi et al. linked a ratiometric and highly sensitive H₂O₂ fluorescence probe to the NLS peptide. The combined probe was allowed visualization of changes in localized nuclear H₂O₂ pools in ratiometric manner (Fig. 5).

4. Lysosomes-targetable H₂O₂ fluorescent probes

Lysosomes, as the major digestive compartments, are membrane-bound acidic vesicles (pH 4.5–5.0) and used for the degradation of cell constituents [67]. The functions of lysosomes are reportedly associated with fluctuations intracellular levels of H₂O₂. Once cellular H₂O₂ diffuse into lysosome, it may react with labile iron(II) to form hydroxyl radicals and iron(Ⅲ), accelerate the formation of oxidative stress, and finally result in depressed autophagy and subsequently cellular dysfunction [68].

Traditionally, morpholine can significantly induce the probe distribution in lysosomes. Using pH-switchable probes is also novel lysosomes-targetable strategy. These lysosomes-targetable H₂O₂ fluorescence probes were shown in Scheme 4.

4.1. Morpholine as the target group

In 2015, Yoon et al. developed a lysosome-targetable fluorescent probe 21 [69] (Scheme 4) by introducing naphthalimide as fluophore, morpholine as targeting group and p-dihydroxyphenyl benzyloxycarbonyl as a transponder. After reaction with H₂O₂, fluorescence emission (λ_em = 528 nm) of probe 21 increased. Owing to its chemospecific boronate switch, the probe showed highly selective response for H₂O₂, over other ROS. In co-staining experiments, probe 21 and a LysoTracker Blue DND-22 showed well overlap in RAW 264.7, HeLa and NIH-3T3 cells. In addition, probe 21 was used to monitor the lysosomal endogenous H₂O₂ in the RAW 264.7 cells.

Based on naphthalimide fluophore, in 2016, Lin group developed another fast-responsive and lysosome-targeted two-photon H₂O₂ fluorescent probe 22 [70] (Scheme 4). Upon excitation at 450 nm, the free probe presented a weak fluorescent emission. By contrast, addition of H₂O₂ induced the fluorescence intensity a drastic increase in the emission of 550 nm. In confocal fluorescence images, probe 22 showed the capacity to accumulate into lysosomes and imaging of H₂O₂ in living RAW264.7 cells. The probe also performed a two-photon fluorescence image in a brand new rat section with a tissue penetration depth of 0–150 μm.

In 2017, Kumar et al. designed and synthesized a lysosometargetable naphthalimide-based H₂O₂ fluorescent probe 23 [39] (Scheme 4) with catechol and morpholine moieties. Upon addition of H₂O₂ to the aqueous solution of probe 23, an emission band at 537 nm is observed. Endogenous imaging of H₂O₂ was performed using C6 and BV-2 cells, upon LPS stimulation. Co-localization experiments by co-staining C6 glial cells with a commercial LysoTracker or MitoTracker showed its excellent lysosome-targetable capacity. Importantly, the probe was also capable of detecting H₂O₂ in brain tissues and living nematodes.

In 2017, Xu et al. reported a H₂O₂-responsive and lysosometargetable two-photon fluorescent probe 24 [71] (Scheme 4) based on pyridazinone scaffold conjugation of electronic-donating aromatic ring and precise terminal modification of the alkyl linker. In co-staining experiments, probe 24 presented high colocalization coefficient with LysoTracker Red in SH-SY5Y cells. Probe 24 was applied for detecting the endogenous generation of H₂O₂ in SH-SY5Y cells induced by oxygen and glucose deprivation (OGD) or OGD/reoxygenation (OGD/R) injury (Fig. 6). It was found that both OGD and OGD/R-induced cell injury caused a timedependent increase of H₂O₂ production within lysosomes.

4.2. pH-Switchable fluorophore

Lysosomes are more acidic (pH 4–6) environment than other places within cells. The feature can be used as an endogenous stimulus for the activation the fluorescence signals. As this principle, Zhao group developed a pH-switchable H₂O₂ probe 25 [72] (Scheme 4) for visualization of lysosomal H₂O₂, which is composed of a boronate moiety as H₂O₂ responsive moiety and a pH-switchable spirobenzopyran as fluorophore. Reaction of probe 25 with H₂O₂ induced an enhanced fluorescence emission (λ_em = 584 nm) in acetate buffer at pH 5. Co-localization experiments were conducted using the probe 25 and commercial LysoTracker Green, which showed their excellent overlap. Only signals from lysosomes can be obviously detected, and that of other regions are invisible, which are due to the pH-switchable property of the probe (Fig. 7).

In 2017, Zhao et al. developed another acidic-pH-activatable fluorescent probe 26 [73] (Scheme 4) for the detection and visualization of H₂O₂ in lysosomes. The probe was constructed by a H₂O₂-responsive boronate unit, a lysosome-targetable morpholine group, and a pH-activatable benzorhodol fluorophore. Upon treatment with H₂O₂, the fluorescence intensities (λ_em = 606 nm) of probe 26 at acidic lysosomal pH (4.5–5.0) are significantly higher than that at neutral pH. Cell imaging experiments further demonstrated that 26 can efficiently accumulate in lysosomes and be used for sensing and visualizing endogenous H₂O₂ in the lysosomes of living cells without external stimulations.

4.3. Others

In 2012, Nam et al. developed a metal complexation-based method (probe 27 [74]) for the detection of intracellular H₂O₂ through fluorescence response. The PIPES buffer containing probe 27 (Scheme 4) displayed a weak fluorescent emission. After the reaction with H₂O₂, the solution provoked fluorescence turn-on with a 22-fold increase in the fluorescence quantum yield. The probe showed specifically response for H₂O₂, over ROS like O₂^-, OCl^-, t-BuOOH, NO, and ¹O₂. Other chemical oxidants, such as (NH₄)2Ce^Ⅳ(NO₃)₆ (CAN) and 2, 3-dichloro-5, 6-dicyano-p-benzoquinone (DDQ), also failed to trigger fluorescence enhancement. Living HeLa cells were used for co-localization staining, and the fluorescence signal of probe 27 and a LysoTracker overlapped perfectly.

In 2013, Zhang et al. combined myeloperoxidase (MPO) and compound 28 [75] (Scheme 4) to construct an "MPO-28" probe. MPO can selectivity to convert "H₂O₂+Cl^-" into HOCl. Compound 28 showed good selectivity and fast response rate toward HOCl over other ROS, with 236-fold fluorescence enhancement at emission peak 610 nm. Thus, "MPO-28" probe was capable of detection of H₂O₂ in the presence of MPO in the PBS buffer (pH 4.5 - 6). The signal from the "MPO-28" system displayed good colocalization with that of LysoTracker Green, with a Pearson's correlation coefficient of 0.92 and an overlap coefficient of 0.94. To more precisely characterize "MPO-28" localization, the lysosomalassociated membrane protein 2, biomarker LAMP2-EGFP, was employed to localization of the probe within the lumen of lysosomes (Fig. 8).

In 2018, Ge et al. constructed a near-infrared and lysosomedirected H₂O₂ fluorescent probe 29 [76] (Scheme 4) with phenoxazinium as the fluorophore receptor and methyl (phenyl) sulfone as the recognition group. After reaction with H₂O₂, a fluorescence emission (λ_em = 676 nm) and turn-on response was shown. In co-staining experiments, probe 29 showed good overlap with a commercially LysoTracker Green merged well and can be used to detect H₂O₂ in HeLa cells.

5. Summary and outlook

In summary, we summarized mitochondria, nuclei lysosomestargetable H₂O₂ fluorescent probes and discussed their target strategies, fluorescence behavior and application in bio-imaging. Undoubtedly, the probes for H₂O₂ local detection have witnessed a great progress.

However, much progress may be made with organelletargetable fluorescent H₂O₂ probes: 1) The emission wavelengths of the most reported organelle-targetable fluorescent H₂O₂ probes are generally distributed in the visible light region. The targetable fluorescent H₂O₂ probe with NIR emission or NIR-II emission can expand its application in biological research. 2) These probes may escape from the organelle, due to the physiological state changing of the specific organelle caused by probes entering. Thus, the development of the fluorescence probes that are capable of binding to a specific cellular structure (including, but not limited to, proteins, liposomes, and DNA) will provide accurate localization and more refined research.

We firmly believe, these fluorescent probes will act as powerful molecular tools for diseases diagnosis, physiological and pathological mechanism research of chemists, biologists and medical scientists.

Acknowledgments

We thank the National Natural Science Foundation of China (Nos. 21705102, 21775096), the Shanxi Province Science Foundation for Youths (No. 201701D221061), Shanxi Province Foundation for Returnees (No. 2017-026), and Scientific Instrument Center of Shanxi University (No. 201512).