A Highly Stable Two-Photon Ratiometric Fluorescence Probe for Real-Time Biosensing and Imaging of Nitric Oxide in Brain Tissues and Larval Zebrafish

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE6 Jun 2022A Highly Stable Two-Photon Ratiometric Fluorescence Probe for Real-Time Biosensing and Imaging of Nitric Oxide in Brain Tissues and Larval Zebrafish Zhiwen Gong, Zhichao Liu, Zhonghui Zhang, Yuxiao Mei and Yang Tian Zhiwen Gong Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200241 Google Scholar More articles by this author , Zhichao Liu Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200241 Google Scholar More articles by this author , Zhonghui Zhang Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200241 Google Scholar More articles by this author , Yuxiao Mei Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200241 Google Scholar More articles by this author and Yang Tian *Corresponding author: E-mail Address: [email protected] Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200241 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202101038 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Because nitric oxide (NO) plays important roles in nerve conduction, signal regulation, and immune protection, analysis of NO is of great significance for understanding the physiological and pathological processes related to neurological diseases. Herein, a highly stable and selective two-photon ratiometric fluorescent probe was developed for real-time sensing and imaging of NO in neurons, brain tissues, and larval zebrafish, in which a Rhodamine B derivative (RBD) was designed for specific recognition of NO and gold nanoclusters (AuNCs) were synthesized as reference element. The developed organic–inorganic nanoprobe exhibited high stability against biological thiol compounds and high selectivity against other reactive oxygen and nitrogen species, metal ions, and acids. In addition, the response time of the present nanoprobe was less than ∼55 s. By using the developed nanoprobe, we proved that hypoxia-induced neuronal death was regulated by NO. Moreover, it was found that the hypoxia-induced NO increase in different brain regions was various and that the NO burst contributed to hypoxia-induced death of zebrafish. Download figure Download PowerPoint Introduction Nitric oxide (NO) is an important signal molecule in living systems and plays important roles in nerve conduction, signal regulation, and immune protection.1–3 The disorder of NO concentration has a close relationship with physiological and pathological processes, especially in the brain, including Parkinson’s disease, Alzheimer’s disease, focal cerebral ischemia, and stroke.4–9 Since NO has high affinity for interaction with other biological species, such as oxygen (O2), superoxide anion and peroxides,10 and the concentration of NO changes dynamically in living organisms, it is challenging to monitor the fluctuations of NO in real time in living organisms with high selectivity and accuracy. So far, there are many elegant methods for determination of NO, such as electron paramagnetic resonance spectroscopy, electrochemical methods, Raman spectroscopy, and fluorescent methods.11–14 Our group is very interested in the development of analytical methods for biosensing of biological substances in the brain, including reactive oxygen and nitrogen species (ROS/RNS),15–18 proteins,19,20 and metal ions.21–27 Especially, we have developed a trisoctahedral gold nanostructures-based Raman probe for selectively biosensing NO in live cells,14 but it is limited in real-time imaging of NO in living cells because Raman mapping of cells takes a long time. Fluorescent methods have attracted great attention because of their unique advantages in noninvasive and real-time sensing and imaging of biological samples.28–31 The fluorescent probes currently used for NO detection are mostly organic fluorescent molecules based on o-phenylenediamine,32 which usually suffer from poor water solubility and photobleaching.32,33 It is essential to develop a fluorescence analytical method for real-time bioimaging and quantification of NO in living cells with high stability and selectivity. In this work, a ratiometric two-photon fluorescence (TPF) probe was designed for real-time imaging and biosensing of NO in neurons, brain tissues, and larval zebrafish, in which a Rhodamine B derivative (RBD) was designed for specific recognition of NO, and gold nanoclusters (AuNCs) were synthesized as the reference element (Scheme 1). The developed organic–inorganic nanoprobe showed two clearly separated TPF emissions around ∼540 and ∼580 nm under the excitation of 800 nm. The present nanoprobe displayed good water solubility and high stability against biological sulfhydryl compounds and rapid response time (∼55 s) toward NO. In addition, it displayed a wide detection range of 0.5–120 μM for NO, as well as high selectivity for NO against other ROS/RNS, metal ions, and amino acids. This developed nanoprobe, was applied for real-time imaging of NO in neurons, and it was found that hypoxia-induced neuronal death was regulated by NO. Taking advantage of two-photon imaging with deep penetration, we found that the hypoxia-induced NO increase in different brain regions was different, and the NO burst contributed to hypoxia-induced death of larval zebrafish. Scheme 1 | (a) Synthesis route for RBD. (b) Working mechanism of [email protected] probe for determination of NO. Download figure Download PowerPoint Experimental Methods Synthesis of RBD probe Compound 1 Triethylamine (8.34 mL, 59.71 mmol), 4-iodo-1,2-diaminobenzene (4 g, 17.1 mmol) and (trimethylsilyl) acetylene (4.17 mL, 29.92 mmol) were added into anhydrous tetrahydrofuran (THF; 25 mL). Then N2 was bubbled for a period of time to remove the oxygen of the reaction mixture. Next, CuI (0.16 g, 0.85 mmol) and Pd(PPh3)4 (750 mg, 0.65 mmol) were added to the reaction during the N2 bubbling. The reaction was allowed to take place overnight, and the gray solid product obtained with yield was 85% by purification (ethyl acetate:petroleum ether = 1:4, v/v). Proton nuclear magnetic resonance (1H NMR) [dimethyl sulfoxide (DMSO)-d6, δ]: 6.59 (s, 1H), 6.53–6.48 (m, 1H), 6.45–6.40 (m, 1H), 4.89 (s, 2H), 4.56 (s, 2H), 0.18 (s, 9H). Carbon nuclear magnetic resonance (13C NMR) (DMSO-d6, δ): 137.13, 134.86, 122.16, 117.57, 114.09, 109.96, 108.64, 89.73. Compound 2 The obtained compound 1 (3 g, 14.54 mmol) and K2CO3 (5 g, 36.18 mmol) were added to a mixed solution of MeOH and CH2Cl2 (1:1, v/v), and stirred at 25 °C for 5 h. After the reaction was done, saturated brine (45 mL) was added to the postreaction system, and then extracted with CH2Cl2 (80 mL). Saturated brine was further poured into the final organic phase and dried with sodium sulfate and then rotary-evaporated to a small volume. Finally, it was further purified to obtain an oily product (CH2Cl2:n-hexane = 7:3, v/v) with a yield of 80% (1.54 g, 11.63 mmol). 1H NMR (MeOD, δ): 6.81 (d, J = 1.8 Hz, 1H), 6.74 (dd, J = 8.0, 1.8 Hz, 2H), 6.61 (d, J = 8.0 Hz, 1H), 3.16 (s, 1H). 13C NMR (MeOD, δ): 136.27, 133.96, 123.61, 119.29, 115.16, 111.97, 84.54, 73.76. Compound RBD Through a constant pressure funnel, SOCl2 (0.11 g, 0.9 mmol) was gradually added dropwise to Rhodamine B (0.14 g, 0.30 mmol) in dichloromethane (20 mL) at 25 °C. After the device was cooled, the solvent was spin-dried to obtain Rhodamine B acid chloride. Then, Rhodamine B acid chloride and tetraethylammonium (TEA; 8 mL) were added to anhydrous acetonitrile (20 mL), and compound 2 (0.21 g, 1.57 mmol) dissolved in anhydrous acetonitrile (20 mL) was added dropwise to the above mixture. After stirring for 4 h at 25 °C, the mixture was concentrated in vacuo to a small volume and then further purified (CH2Cl2:MeOH = 80:1, v/v). A yellow-white powder was obtained with yield of 60%. 1H NMR (CDCl3, δ): 8.05 (d, J = 7.1 Hz, 1H), 7.60 (dd, J = 17.8, 9.6 Hz, 2H), 7.33 (d, J = 7.0 Hz, 1H), 7.10 (d, J = 11.6 Hz, 1H), 6.63 (dd, J = 13.2, 8.9 Hz, 2H), 6.46 (d, J = 8.3 Hz, 1H), 6.41–6.26 (m, 4H), 6.11–6.05 (m, 1H), 3.56 (s, 2H), 3.42–3.28 (m, 8H), 2.72 (s, 1H), 1.17 (t, J = 7.0 Hz, 12H). 13C NMR (CDCl3, δ): 166.23, 153.83, 149.07, 145.54, 144.33, 132.81, 132.21, 128.67, 128.63, 128.56, 128.42, 124.43, 120.53, 116.16, 111.06, 108.00, 106.62, 97.98, 83.88, 77.28, 77.03, 76.78, 74.18, 68.21, 44.45, 12.52. Synthesis of [email protected] nanoprobe Bovine serum albumin (BSA)-stabilized AuNCs were synthesized according to the method previously reported in the literature.34 The mixture of HAuCl4 (4 mL, 10 mM) and BSA solution (4 mL, 20 mg/mL) was vigorously stirred at 37 °C for 3 min, and 40 μL of AA (0.35 mg/mL) was added to the mixture, followed by addition of an appropriate amount of NaOH (1 M) under stirring to control the pH of the mixed solution that reached ∼8. The mixture was further stirred at 37 °C for 5 h. AuNCs were obtained by centrifugation at 10,000 rpm for 15 min. Then, RBD solution (1 mL, 10 mM) was bubbled with N2 for half an hour to remove dissolved oxygen, followed by addition of AuNCs, and the mixture reacted at 60 °C for 6 h. Excess RBD was removed by centrifugation treatment (6000 rpm, 3 min). The produced [email protected] probe was resuspended in phosphate-buffered saline (PBS) buffer before being used. Confocal fluorescence imaging All animal experiment protocols strictly complied with the implementation guidelines of Care and Use of Laboratory Animals formulated by the Ministry of Science and Technology of China and have been approved by the Animal Protection and Use Committee of East China Normal University (approval number: m+R20190304, Shanghai, China). To image NO in living cells, neurons were first incubated with the [email protected] probe in Hank’s balanced salt solution (HBSS) buffer for 30 min. Then, the neurons were washed three times by PBS to remove unabsorbed probe. Next, different groups of neurons were stimulated by different concentrations of NO and further used for confocal imaging. Similar experiments were also conducted under hypoxia treatment for different times before imaging. For the control experiment, the neurons were first incubated with the [email protected] probe and 2-(4-carboxyphenyl)-4,4,5,5-tetramethyl-imidazoline-1-oxyl-3-oxide potassium (Carboxy-PTIO) for 30 min, and then the neurons were washed three times by PBS to remove unabsorbed probe and Carboxy-PTIO. Subsequently, the neurons were stimulated by hypoxia for a different time before imaging. For confocal fluorescence imaging of NO in brain slices, different brain slices were obtained from 2-month-old mice treated by decapitation. Then, the brain slices were transferred into artificial cerebrospinal fluid (aCSF) buffer and further incubated with the [email protected] probe for 30 min. Next, the brain slices were washed three times by aCSF to remove unabsorbed probe before imaging. Similar experiments were also conducted under hypoxia treatment for different times before imaging. For confocal fluorescence imaging of NO in larval zebrafish with the stimulation of hypoxia, 3-day-old larval zebrafish were first incubated with [email protected] probe in nutrient solution for 60 min. Then the larval zebrafish were washed three times by nutrient solution to remove unabsorbed probe. Next, N2 was continuously blown into the nutrient solution to remove O2 before imaging. Results and Discussion Design, synthesis, and characterization of [email protected] probe For specific identification of NO, an RBD was designed in three parts: the first part—Rhodamine B as fluorophore, the second part—o-phenylenediamine as recognition site, and the third part—an alkynyl group as anchor group (Scheme 1a). RBD was synthesized and characterized by NMR and mass spectrometry (MS) ( Supporting Information Figures S1–S9). Because a lactam ring in RBD was formed through the reaction between the lactone group and o-phenylenediamine, resulting in the destruction of the conjugate structure of Rhodamine B, absorption was observed around 550 nm for RBD (Figure 1a). However, the diazotization of the amino group induced by the reaction between NO and RBD resulted in the spiral ring opening of RBD (Scheme 1b). Thus, the fluorescence of RBD was recovered, with typical absorption at ∼556 nm and fluorescence emission around 580 nm (Figure 1a). In addition, NO induced an obvious increase of TPF of RBD, with two-photon cross-section values (δ) increased from 10 ± 3 GM to 103 ± 5 GM under the excitation of 800 nm (Rhodamine 6G in methanol as standard, 1 GM = 1 × 10−50 cm4 s) (Figure 1b). Figure 1 | (a) UV–vis absorption spectra and fluorescence emission spectra of RBD in the absence (1 and 2) and in the presence of NO (3 and 4). (b) Two-photon action spectra of RBD in the absence (1) and in the presence of NO (2) (Error bars: S.D., n = 3). (c) TEM image of AuNCs. Inset shows the particle size distribution of AuNCs. (d) FT-IR characterization of [email protected] after exposure to 1 mM thiols (GSH, Cys, and Hcy) for 2 h. Download figure Download PowerPoint To develop a ratiometric fluorescence probe with good water solubility and internal reference, BSA-templated AuNCs were synthesized according to the previously reported method.32 Transmission electron microscopy (TEM) results showed that the synthesized AuNCs were nondispersed with an average size of 1.5 ± 0.3 nm (Figure 1c), and X-ray photoelectron spectroscopy (XPS) spectra of Au 4f region confirmed two typical binding energies of Au(0) at 4f7/2 (84.8 eV) and 4f5/2 (88.5 eV) electrons in AuNCs ( Supporting Information Figure S10d). Moreover, the synthesized AuNCs displayed a broad absorption and apparent TPF emission around 540 nm ( Supporting Information Figure S10). Then, a ratiometric TPF probe was assembled by conjugating RBD onto AuNCs through a robust Au–C≡C bond,35 named as [email protected] The Fourier transform infrared (FT-IR) spectrum of [email protected] exhibited the disappearance of typical vibration peaks of C≡C (3293 cm−1) and ≡C–H (2098 cm−1) ( Supporting Information Figure S11a), strongly proving the successful conjugation of RBD onto AuNCs through the Au–C≡C bond. Meanwhile, the contact angle of [email protected] was 11.4 ± 1.8° (n = 5, S.D.), which was much smaller than that of RBD (73.0 ± 2.0°, n = 5, S.D.) ( Supporting Information Figure S11b), demonstrating that the developed [email protected] showed better water solubility than that of individual RBDs and benefited from the good hydrophility of BSA-templated AuNCs. Since there are many biological sulfhydryl compounds in live cells, the stability of the developed [email protected] probe exposed to a high concentration (1 mM) of glutathione (GSH), cysteine (Cys), and homocysteine (Hcy) for 2 h was further tested. As shown in Figure 1d, no signal ascribed to the alkynyl group was observed from the FT-IR spectra, demonstrating the high stability of the [email protected] probe, which can be attributed to the highly stable Au–C≡C bond. Analytical performance of [email protected] probe for determination of NO To confirm the working principle, [email protected] were further used for the determination of NO in cell lysis. As illustrated in Figure 2a, the individual [email protected] probe showed single TPF emission around 450 nm under the excitation of 800 nm. With the addition of NO from 0 to 132 μM, an obvious increase for TPF emission at ∼580 nm (Fred: 550–600 nm) was observed, while change was seldom observed for the TPF intensity around 450 nm (Fgreen: 430–500 nm), leading to a ratiometric response to NO. The intensity ratio of Fred to Fgreen (Fred/Fgreen) displayed a good linear relationship with NO concentrations in the range of 0.5–120.0 μM (Figure 2b), which was better or comparable with previously reported methods ( Supporting Information Table S2), and the detection limit was estimated to 105.0 ± 3.1 nM (S/N = 3, n = 20). In comparison with the fluorescent NO probes in the literature,36–42 our developed [email protected] probe showed a wider detection linear range. It should be noted that the absorbance intensity of individual RBD apparently increased around 556 nm with the increasing concentrations of NO, accompanied with fluorescence enhancement around 580 nm ( Supporting Information Figure S12). Thus, the increase of TPF emission at ∼580 nm can be attributed to the production of Rhodamine B, which was further confirmed by MS ( Supporting Information Figure S13). In addition, the response time of developed [email protected] toward NO (100 μM) was evaluated to ∼55 s (Figure 2c), which was faster than most reported fluorescence methods for NO sensing,36–42 as summarized in Supporting Information Table S2. Moreover, the signal variation for [email protected] was <3.7% at pH 6.5–8.0, and little signal decrease (<4.2%) was found for [email protected] when exposed to a Xe lamp (90 W) for 150 min ( Supporting Information Figure S14), indicating the high stability of the designed [email protected] nanoprobe. Figure 2 | (a) TPF emission spectra of [email protected] probe with the addition of different concentrations of NO (0, 0.5, 20, 30, 40, 50, 60, 70, 80, 100, 120, 125, and 132 μM). (b) Working calibration plot between Fred/Fgreen (Fgreen: 430–500 nm, Fred: 550–600 nm) of [email protected] probe and NO concentrations (n = 20, S.D.) (c) Response time of [email protected] Probe toward NO (100 μM). (d) Selectivity and competition tests of [email protected] probe in the presence of NO (50 μM) and other ROS/RNS (100 μM) (n = 3, S.D.). Download figure Download PowerPoint Due to the complex environment of organisms in living cells, the selective and competitive tests of developed [email protected] probe were also investigated. No apparent effect (<4.6%) was observed from the common ROS, including H2O2, •OH, ClO−, O2•−, and 1O2 or RNS such as ONOO−, HNO, and NO2− (Figure 2d). Meanwhile, negligible influence (<4.2%) was found from amino acids and metal ions including K+ (100 mM), Na+ (50 mM), Ca2+ (10 mM), Mg2+, Fe3+, Fe2+, Cu2+, Cu+, and Zn2+ ( Supporting Information Figure S15). In addition, no obvious influence (<5.0%) was found for the competitive tests in the presence of NO and other potential interference of ROS/RNS, metal ions, or amino acids. The above tested results indicated the high selectivity of developed [email protected] toward NO, which benefited from the specific recognition part of RBD. Real-time imaging and quantification of NO in neurons Before applying the developed nanoprobe for biological sensing of NO, the cytotoxicity and biocompatibility of [email protected] nanoprobe were estimated. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) results proved that the cell viability was higher than 90% after the cells incubated with [email protected] nanoprobe (90 μg/mL) for 48 h ( Supporting Information Figure S16). In addition, flow cytometry assay results demonstrated that apoptotic cells (<6.2%) were seldom found after the cells incubated with the [email protected] probe (90 μg/mL) for 24 h ( Supporting Information Figure S17). The above results proved that the low cytotoxicity and high biocompatibility of the developed probe. Subsequently, the developed TPF probe [email protected] was further used for real-time biological imaging of NO in neurons. Colocalization imaging experiments proved that the [email protected] probe entered into neurons and primarily located in cytoplasm (Pearson’s correlation coefficient: 0.89) (Figure 3a). With the increasing concentrations of exogenous NO from 0 to 100 μM, the fluorescence of neurons from the red channel (Fred: 550–600 nm) gradually increased, while that from the green fluorescence channel (Fgreen: 430–500 nm) remained unchanged (Figure 3b). The intensity ratio of Fred to Fgreen (Fred/Fgreen) was increased from 0.52 ± 0.02 to 1.63 ± 0.06 (Figure 3c). It is worth noting that endogenous NO concentration in neurons was estimated to 19.5 ± 0.5 μM (n = 10, S.D.) in the absence of exogenous NO based on the standard curve established in vitro, which was consistent with the previously reported result.43 Moreover, no apparent fluorescence increase (<5.4%) was observed from the red channel under the stimulation exogenous NO (100 μM) in the presence of Carboxy-PTIO (0.1 mM), a NO scavenger, further certifying the observed fluorescence increase in the red channel was ascribed to the concentration changes of NO. Furthermore, t-test results proved that there was no significant difference between the determined results obtained from the commercial NO probe (4,5-diaminofluorescein, DAF-2) and those from our developed nanoprobe ( Supporting Information Figure S18 and Table S1). The above experimental results showed that our developed [email protected] probe can be applied to biosensing and imaging of NO in neurons with high accuracy. Figure 3 | (a) Colocalization of neurons treated with [email protected] probe and CellTracker Deep Red. The signal of probe was collected from 430 to 500 nm under the excitation of 405 nm. The CellTracker Deep Red channel was collected from 650 to 700 nm at excitation of 638 nm. The overlay channel is the overlay of the green channel and the CellTracker channel. (b) TPF imaging of neurons incubated with [email protected] probe collected from different channels under the stimulation of different concentrations of NO (0, 10, 50, and 100 μM), and 100 μM NO preincubated with Carboxy-PTIO (0.1 mM). (c) Average fluorescence intensity ratio (Fred/Fgreen) obtained from (b) (n = 35, S.D.). Scale bar: 25 μm. Download figure Download PowerPoint Two-photon imaging of NO fluctuation in neurons and brain tissues under hypoxia Hypoxia is a typical feature of many diseases, including ischemia, stroke, and hypertension. However, the exact mechanism of hypoxic brain injury is still unclear.44–46 To understand the hypoxic signaling in the brain, neuronal activity was first estimated under the stimulation of hypoxia. Decreasing O2 concentration or prolongation of hypoxic time resulted in significant neuronal death ( Supporting Information Figure S19). The cell viability was reduced to 35.7% after acute hypoxia (1% O2) for 60 min. To understand the molecular mechanism of hypoxia in neuronal death, the concentration of NO in neurons was further assessed under acute hypoxia by using our developed [email protected] probe. With the prolongation of hypoxia time (Figure 4a), the fluorescence intensity of the red channel was significantly enhanced, while that of the green channel remained unchanged, indicating that hypoxia caused a significant increase in NO concentration. The concentration of NO increased to ∼100.0 ± 2.4 μM after acute hypoxia for 45 min (Figure 4b). Moreover, flow cytometry assay results showed that after hypoxia for 15 min, ∼18.9% of apoptotic neurons were detected, while ∼46.6% of apoptotic cells were found after hypoxia for 45 min (Figure 4c). However, little apoptotic cells (11.3%) were found after neurons were stimulated by acute hypoxia for 45 min in the presence of Carboxy-PTIO (Figure 4c). The results implied that NO plays an important role as an apoptosis regulator in neuronal death caused by hypoxia. Figure 4 | (a) Two-photon imaging of neurons incubated with [email protected] probe after acute hypoxia for different times. (b) Average fluorescence intensity ratio (Fred/Fgreen) obtained from (a) (n = 35, S.D.). Scale bar: 25 μm. (c) The apoptosis assay of neurons after acute hypoxia for different times (0, 15, 45 min), and hypoxia for 45 min in the presence of Carboxy-PTIO (0.1 mM). Q1, Q2, Q3, and Q4 represent the regions of live, early apoptotic, late apoptotic, and dead neurons, respectively. (d) TPF imaging of NO in different brain regions (S1BF, CA1, LD, and CPu) by using the developed [email protected] probe. (e) Average fluorescence intensity ratio (Fred/Fgreen) obtained from (d) in different brain regions (S1BF, CA1, LD, and CPu) (n= 6, S.D.). Scale bar: 50 μm. Download figure Download PowerPoint On the other hand, taking advantage of two-photon imaging with deep penetration, the concentrations of NO in different brain regions of brain tissues under acute hypoxia were also estimated (4d), including primary somatosensory cortex (S1BF), field CA1 of hippocampus (CA1), laterodorsal thalamic nucleus (LD), and caudate putamen (CPu). As shown in Figure 4e, the concentrations of endogenous NO in those four brain regions were estimated to be 21.4 ± 0.4 μM. However, after the brain tissues were treated with acute hypoxia for 45 min, the concentrations of NO in those four brain regions increased differently. The concentration of NO in S1BF and CA1 increased to ∼94.7 ± 0.3 μM and ∼100.3 ± 0.3 μM, which was obviously higher than that in LD (∼83.7 ± 0.2 μM) and CPu (∼82.4 ± 0.4 μM). The results imply that different brain regions have different sensitivities to hypoxic signaling. Further, all these results prove that our developed nanoprobe can be used for quantitative analysis of NO in neurons and brain tissues. In vivo imaging of larval zebrafish under normoxic and hypoxic conditions To further examine the performance of the developed [email protected] probe for bioimaging in vivo, the 3-day-old larval zebrafish with optical transparency was chosen as living models due to its high degree of genetic homology with mammals.47 As shown in Figure 5a, TPF imaging of larval zebrafish displayed two fluorescence channels with a depth of 50–260 μm, much deeper than the one-photon fluorescence images ( Supporting Information Figure S20). In view of the fact that hypoxia (0.6 mg/L O2) resulted in the apparent decrease of the survival rate of zebrafish (Figure 5b), hypoxia induced an obvious increase of NO in the zebrafish brain (Figure 5c and Supporting Information Figure S21). The concentration of NO in the larval zebrafish brain under hypoxia was estimated to ∼84.2 ± 3.6 μM, which was significantly higher than that in normal conditions (∼18.8 ± 2.5 μM) (Figure 5d). The data implied that hypoxic zebrafish experienced evident oxidative damage. These results further prove our developed [email protected] probe holds great potential in applications of bioimaging and biosensing of NO in vivo. Figure 5 | (a) 3D two-photon confocal imaging of larval zebrafish incubated with [email protected] probe under the excitation of 800 nm. (b) Survival rate of larval zebrafish after hypoxia for different times. (c) Two-photon confocal imaging of larval zebrafish brain incubated with [email protected] probe under normoxic (6 mg/L O2) and hypoxic (0.6 mg/L O2) conditions. (d) Average fluorescence intensity ratio (Fred/Fgreen) obtained from (c) (n = 5, S.D.) Scale bar: 250 μm. Download figure Download PowerPoint Conclusions A highly selective ratiometric TPF nanoprobe was successfully developed for detection of the NO, with high stability and rapid response dynamics. In view of the excellent properties of [email protected] nanoprobe, including high biocompatibility and low cytotoxicity, it was successfully applied for real-time biosensing and bioimaging of NO in neurons, and it was demonstrated that hypoxia-induced neuronal death was regulated by NO. In addition, the fact that hypoxia-induced NO increase in various brain regions was different was also observed. Moreover, we proved that hypoxia-caused death of larval zebrafish has a close relationship with the NO burst. This work not only provides a new perspective for understanding the mechanism of hypoxic injury, but also supplies a methodology to develop a highly stable metal cluster-based inorganic–organic fluorescent nanoprobe for ROS/RNS and other biological