Engineering of Reversible Luminescent Probes for Real-Time Intravital Imaging of Liver Injury and Repair

Open AccessCCS ChemistryRESEARCH ARTICLE1 Jan 2022Engineering of Reversible Luminescent Probes for Real-Time Intravital Imaging of Liver Injury and Repair Xiao Liu, Linhui He, Xiangyang Gong, Yue Yang, Dan Cheng, Juanjuan Peng, Lu Wang, Xiao-Bing Zhang and Lin Yuan Xiao Liu State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 , Linhui He State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 , Xiangyang Gong State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 , Yue Yang State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 , Dan Cheng State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 , Juanjuan Peng State Key Laboratory of Natural Medicines, School of Basic Medical Sciences and Clinical Pharmacy, China Pharmaceutical University, Nanjing, Jiangsu 211198 , Lu Wang Department of Chemical Biology, Max Planck Institute for Medical Research, Heidelberg 69120 , Xiao-Bing Zhang State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 and Lin Yuan *Corresponding author: E-mail Address: [email protected] State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082 https://doi.org/10.31635/ccschem.021.202000679 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail As the major organ for drug metabolism and detoxification, the liver is prone to damage and severely impaired functionality. The treatment of liver diseases is based on a clear understanding of the process underlying liver injury and repair. However, intravital real-time imaging of liver injury and repair is still limited due to the lack of in vivo reversible visualization methods. To this end, we proposed a rational design strategy for the development of a reversible upconversion luminescence nanoprobe that allows real-time and in vivo imaging of liver injury and repair processes. As a proof of concept, we first developed a small molecule probe NB3 which can reversibly respond to related analytes of early liver injury [peroxynitrite (ONOO−)] and liver repair [glutathione (GSH)]. The small molecule probe was then integrated with a core–shell upconversion nanoparticle to form a sophisticated nanoprobe. Compared with traditional small molecule probes, this nanoprobe exhibited a higher selectivity to ONOO−, longer retention time in liver, and wider dynamic response range to GSH after oxidation by ONOO−. The novel nanoprobe facilitated the successful monitoring and discrimination among the different degrees of liver injury and repair in a mouse model. Download figure Download PowerPoint Introduction Liver injury and the associated repair mechanisms constitute a prominent area of study in clinical medicine.1–5 A clear understanding of the processes underlying liver injury and repair forms the theoretical basis for the treatment of hepatic disease.6 Liver injuries of different degrees often require different repair methods.7–10 However, it is difficult to obtain real-time information on the process of liver injury and repair. Several traditional methods, including tissue imaging by hematoxylin and eosin (H&E), immunohistochemistry staining, and blood analysis by alanine aminotransferase (ALT) and aspartate aminotransferase (AST) tests, are often applied to measure liver injury.11,12 Nevertheless, limited information is provided for such end-point detection techniques, which severely hampers the intensive investigations of liver injury and repair mechanisms. To solve this problem, luminescent imaging technologies with excellent sensitivity and high spatial–temporal resolution,13–15 which has been widely used in the imaging of molecules associated with various disease models,16–18 are promising in the early diagnosis of hepatic conditions. Early investigation involves the use of luminescent probes for in vivo assessment of liver injury or repair using imaging-related analytes.19–29 However, those probes reported can only facilitate one-way imaging of either liver injury or repair due to irreversible characteristics. To date, reversible probes for in vivo real-time monitoring of liver injury and repair have not been reported. Previously, several kinds of reversible fluorescent probes have been reported,30–32 but application to in vivo liver imaging studies remains difficult. In fact, optical probes that could be used for in vivo and real-time monitoring of liver injury and repair processes should fulfill the following requirements: (1) rapid enrichment and long-term retention in liver; (2) long excitation and emission wavelengths at the near-infrared (NIR) region to ensure sufficient penetration depth in vivo;33 (3) most importantly, reversible response to the biomarkers associated with liver injury and repair processes; and (4) broad dynamic range for different targets.34 Previous investigations have uncovered that the peroxynitrite (ONOO−) levels increase drastically during early liver injury, whereas the glutathione (GSH) levels increase significantly during liver repair.35–38 Therefore, it is possible to evaluate the specific liver injury and repair status by detecting the relative changes of ONOO− and GSH levels in liver. However, it remains difficult to design a probe that can meet the response interval of two analytes simultaneously, since ONOO− is present in the micromolar range and GSH in the millimolar range.39–41 To date, only a few probes have been described for reversible detection of ONOO−/GSH.30–32 Unfortunately, the short interval of response to GSH or the short wavelength of these probes limits their applications for in situ analysis of liver injury and repair. Therefore, it is necessary to design a reversible probe that can meet the above requirements to explore the process underlying liver injury and repair in vivo. In this study, we report a rational design strategy for the development of a reversible nanoprobe that enables real-time detection of ONOO−/GSH in the process of liver injury and repair in vivo. The upconversion nanoprobes (UCNP) NaYF4,Yb50%@NaYF4,Ho1%@[email protected] [ (UCNPs)[email protected]@NB3] were constructed using a small molecule probe NB3 and a core–shell UCNP coated with polyetherimide (PEI) with an activator present in the shell (Scheme 1a). The NB3 probe reversibly responded to ONOO− and GSH. The nanoprobe exhibited high sensitivity and selectivity to ONOO− and required millimolar GSH to reduce the product oxidized by ONOO− at the micromolar level, which are well matched with the ONOO− and GSH levels in liver. Moreover, this nanoprobe had a relatively long retention time in the liver, which laid the foundation for its application for monitoring liver injury and repair in vivo. When the degree of liver injury was aggravated, the fluorescence of the nanoprobe increased significantly. However, if the liver was repaired, the corresponding fluorescence decreased, indicating the ( UCNPs)[email protected]@NB3 probe's potential for reversible real-time monitoring of liver injury and repair (Scheme 1b). With this novel probe, we demonstrated the reversible detection of ONOO− and GSH in live cell and animals and achieved the dynamic analysis of different degrees of liver injury and repair. It was found that the time required for liver repair is directly correlated to the severity of liver injury, and often, the liver cannot be repaired completely. To the best of our knowledge, this is the first study to report the reversible real-time imaging of liver injury and repair in mice. Scheme 1 | (a) Design and response mechanism of reversible nanoprobe. (1) Reversible redox reaction of Ref and Rsr to ONOO− and GSH. (2) Reversible response mechanism of NB3 to ONOO− and GSH. (3) Design and synthesis of [email protected]@NB3 used for reversible detection of ONOO− and GSH. (b) Schematic diagram of real-time reversible imaging of liver injury and repair with luminescent probe. Liver injured by CCl4 and repaired by NAC. Download figure Download PowerPoint Experimental Methods Materials and instruments Details of the materials and instruments used are provided in the Supporting Information. Synthetic of NB3 First, a mixture of m-aminophenol (109.0 mg, 1.0 mmol) and acrylic acid (94.7 mg, 1.1 mmol) was dissolved in EtOH/H2O (1/4, 25.0 mL) with stirring at 70 °C for 4 h. The solvent was removed under vacuum, and the crude product was purified using column chromatography on silica (CH2Cl2/EtOH, 20/1, v/v) to yield the pure product 5 (138.0 mg, 76.2%). Next, the compound 5 (181.2 mg, 1.0 mmol) and iodomethane (156.1 mg, 1.1 mmol) were dissolved in acetonitrile (10.0 mL), followed by addition of K2CO3 (276.0 mg, 2.0 mmol). The resulting mixture was heated and stirred at 80 °C for 8 h. When the mixture was cooled to room temperature, the solvent was removed under vacuum, the product was obtained by extraction with CH2Cl2, and the crude product was purified using column chromatography on silica (CH2Cl2/EtOH, 20/1, v/v) to yield the pure product 6 (127.8 mg, 61.1%). Next, the compound p-nitroaniline (138.1 mg, 1.0 mmol) and NaNO2 (69.0 mg, 1.0 mmol) were dissolved in 10 mL of 1/1 HCl/H2O (10.0 mL) and stirred in an ice bath for 30 min. Following this, compound 6 (251.1 mg, 1.2 mmol) dissolved in acetonitrile (5.0 mL) was added to the above solution and stirred for 30 min at room temperature. The product 7 (324.3 mg, 90.5%) was obtained by removing the solvent under reduced pressure, followed by washing three times with water before purification using column chromatography on silica. In the last step, 1-naphthylamine and compound 7 (358.4 mg, 1.0 mmol) were dissolved in 15.5 mL mixed solution of HClO4 and dimethylformamide (DMF, 1:30) and stirred at 160 °C for 30 min. Upon cooling to room temperature, the solvent was removed by freeze-drying, and the crude product was purified using column chromatography on silica (CH2Cl2/EtOH, 25:1, v/v) to obtain the pure product NB3 (117.8 mg, 32.5%). 1H NMR (400 MHz, DMSO-d6, δ): 12.53 (s, 1H), 9.92 (s, 1H), 9.84 (s, 1H), 8.67 (d, J = 7.9 Hz, 1H), 8.38 (d, J = 7.9 Hz, 1H), 7.94 (t, J = 7.6 Hz, 1H), 7.84 (t, J = 7.3 Hz, 1H), 7.73 (d, J = 9.3 Hz, 1H), 7.19 (d, J = 8.9 Hz, 1H), 6.92 (s, 1H), 6.74 (s, 1H), 5.77 (s, 1H), 3.82 (s, 2H), 3.67 (d, J = 6.5 Hz, 2H),2.69 (t, J = 6.4 Hz, 2H), 1.24 (t, J = 6.4 Hz, 3H). 13C NMR (100 MHz, DMSO-d6, δ): 173.0, 161.6, 153.9, 151.5, 147.8, 134.5, 133.0, 132.8, 131.6, 130.2, 129.3, 124.6, 124.4, 123.0, 115.2, 97.2, 96.6, 46.9, 46.2, 32.5, 12.8. Electrospray ionization mass spectrometry (ESI-MS): M+ calcd for C21H20N3O3+, 362.20; found 362.15. Synthesis of the nanoprobes [email protected]@NB3 ([email protected] and [email protected]) The UCNPs (UCNPs-1 and UCNPs-2) were synthesized using a typical thermal decomposition process, and [email protected] ([email protected] and [email protected]) were synthesized using the ligand exchange method. The detailed synthesis methods are outlined in the Supporting Information. During the assembly of NB3 and [email protected], the dye (NB3) (0.05 mmol) was added to an 2-(N-morpholino) ethanesulfonic acid (MES) solution (30 mmol) of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC; 0.06 mmol) and N-hydroxysuccinimide (NHS; 0.12 mmol). The mixture was stirred for approximately 30 min at room temperature. Next, [email protected] (5 mg/mL, 2.0 mL) was added to the solution, and the pH was adjusted to 8 before stirring overnight. The excess dye was removed by centrifugation. The precipitate was washed with ethanol or deionized water and collected after centrifugation. The obtained nanoprobe ([email protected]@NB3) was re-dispersed in deionized water.27 Intracellular endogenous reversible sensing of ONOO− and GSH using [email protected]@NB3 in RAW264.7 cells RAW264.7 cells were seeded on Petri dishes and incubated for 24 h. Next, the cells were treated with [email protected]@NB3 (200 μg/mL) for 3 h and washed with phosphate-buffered saline (PBS) three times to remove the residual nanoparticles. Cells were then stimulated with lipopolysaccharide (LPS; 5 μg/mL) and interferon gamma (IFN-γ; 250 ng/mL) for 4 h and phorbol 12-myristate 13-acetate (PMA; 5 μg/mL) for 30 min. In another dish, 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO; 1.5 mM) and GSH reduced ethyl esters (GREs; 5 mM) were used to treat the cells simultaneously for 30 min after the addition of the abovementioned substances. The cells were washed with PBS three times and subjected to confocal fluorescence microscopy (TI-E+A1 SI; Nikon, Japan) for fluorescence imaging under a 980 nm excitation laser at a power intensity of 1.2 W. The upconversion luminescence (UCL) emission was recorded between 500–590 nm (green channel) and 593–676 nm (red channel). Real-time upconversion imaging of different degrees of liver injury and repair in mice All animal operations were performed in compliance with protocol no. SYXK (Xiang) 2018-0006, approved by Laboratory Animal Center of Hunan. Female BALB/c homozygous athymic mice (∼3 weeks old, 18–20 g) were obtained from Hunan SJA Laboratory Animal Co., Ltd. and used under protocols approved by Laboratory Animal Center of College of Biology, Hunan University. The BALB/c mice were divided into four large groups: control group with subcutaneous injection of 50 μL PBS and liver injury group with subcutaneous injection of 50 μL CCl4 (containing 60% vegetable oil) one, two, or three times, and the corresponding treatment groups (n = 4 in each group). CCl4 was injected every 3 days. Fluorescence imaging commenced 12 h after the last injection of CCl4. Meanwhile, 100 μL of the 1.5 mg/mL nanoprobe was injected intravenously into the mice 1 h before fluorescence imaging. Subsequently, 450 mg/kg N-acetyl-l-cysteine (NAC) was injected into the abdominal cavity of the mice in each of the three treatment groups after imaging was performed for 2 h, which was followed by imaging every 30 min. Other routine experimental steps are outlined in the Supporting Information. Results and Discussion Design and synthesis of a nanoprobe for the reversible detection of ONOO−/GSH As a physiological reducing agent, GSH plays an important role in multiple biological processes.42,43 The concentration of GSH in the liver under normal physiological conditions is approximately 3 mM.44 Therefore, as the response range of most probes to GSH is considerably narrow, only a few probes45 can be used to detect GSH in the liver. Previous studies46–48 and our preliminary experimental results indicate that resazurin (Rsr) can be reduced to resorufin (Rsf) by GSH present at physiological concentrations (0–3 mM) ( Supporting Information Figures S2 and S3). In addition, the reduced product, Rsf, can be oxidized to Rsr by ONOO− ( Supporting Information Figures S2, S4, and S5). Hence, Rsf has potential for application in the reversible detection of ONOO− and GSH (Scheme 1a1). However, Rsf is not an ideal probe for the intravital imaging of ONOO−/GSH because (1) it is sensitive to pH changes and its signal quality is compromised at an acidic pH ( Supporting Information Figures S6a and S6b) and (2) the excitation/emission wavelengths of Rsf are beyond the bioimaging window (<650 nm). To overcome these limitations, we synthesized the NIR-excited pH-insensitive amino derivatives NB1 and NB2 of Rsf ( Supporting Information Figures S1 and S34–S36). Titration experiments showed that NB1 and NB2 can be reversible oxidized by ONOO− and reduced by GSH ( Supporting Information Figures S7a–S7d). Unfortunately, NB1 and NB2 are not specific for ONOO−, as it also responded to hypochlorite (HOCl) ( Supporting Information Figures S7e and S7f). This could be attributed to the oxidation of the amino group to nitroso group by HOCl.49 To overcome these limitations, an NB3 (a carboxylated derivative of NB1, Supporting Information Figures S1, S34, and S37) was synthesized to act as the sensing component. Moreover, optically inactive PEI was used as the secondary component to prepare the nanoprobe [email protected]@NB3 (Scheme 1a3), which connected with NB3 through an amide bond and acted as a masking group for HOCl ( Supporting Information Figure S10).50 Such a combined approach consequently transformed NB3 into a water-soluble reversible NIR nanoprobe with high ONOO− selectivity ( Supporting Information Figures S8 and S9), suitable GSH response range, and pH inertness, which made it more suitable for liver injury and repair imaging in vivo. Spectroscopic properties of NB3 and the UCNP Before covalently grafting NB3 onto the surface of the nanomaterials, we first investigated the reversible response of NB3 to ONOO− and GSH. Figure 1a shows that the absorption peak of NB3 decreased gradually upon the addition of ONOO−, indicating that NB3 was gradually oxidized to NB3 oxidation product (NBO) by ONOO−. However, when GSH was added to the above solution, the diminished absorption peak recovered (Figure 1b), indicating that the oxidation product NBO was reduced back to NB3. The above results showed that NB3 reversibly responded to ONOO− and GSH. The mechanism underlying NB3 oxidation by ONOO− was like that of Rsf oxidation (Scheme 1a2), and its oxidation product NBO was confirmed by ESI-MS analysis ( Supporting Information Figure S11). Next, we tested the reversible response cycle of NB3 to ONOO− and GSH. The results showed that the reversible cycle could be repeated at least twice without obvious attenuation (Figure 1c). Moreover, to test the stability of NB3, the absorption response of NB3 toward physiological pH range (from 4.0 to 9.0) and serum were measured. The results showed that NB3 was stable in physiological pH range ( Supporting Information Figure S12), as well as in serum at 37 °C for 9 h ( Supporting Information Figure S13). In addition, the enrichment and metabolism of NB3 in the liver were also studied. As shown in Supporting Information Figure S14, the metabolism of NB3 in the liver reached 30% within 3 h, indicating that NB3 is not suitable for real-time reversible imaging of liver injury and repair. Figure 1 | (a) The UV absorption spectrum of 5 μM NB3 reacting with ONOO− (0–8 μM) in 25 mM PBS (contains 1% ethanol) at pH 7.4. (b) The absorption response of NB3 to GSH (0–1.5 mM) after treatment with 8 μM ONOO− in 25 mM PBS (contains 1% ethanol) at pH 7.4. (c) Reversible cycle response of 5 μM NB3 to 8 μM ONOO− and 1.5 mM GSH at absorption wavelength of 630 nm in 25 mM PBS (contains 1% ethanol) at pH 7.4. Download figure Download PowerPoint The application of core–shell UCNPs is often limited by a low signal-to-background ratio,51,52 which is primarily caused by the luminescence in the center of the nanomaterials in the thick core layer. Reducing the distance between the two luminescent centers of nanomaterials and dye molecules can improve the energy transfer efficiency between them,53 which could improve the signal-to-background ratio of nanoprobes.54 To verify this, we first synthesized two types of nanomaterials: NaYF4,Yb50%@NaYF4,Ho1% (UCNPs-1) and NaYF4,Yb20%/Ho1%@NaYF4 (UCNPs-2), with the activator present in the shell or core, respectively, which had emission peaks at approximately 650 nm to ensure that the particles had good spectral overlap with NB3 ( Supporting Information Figure S15). Transmission electron microscopy (TEM) patterns showed that the two kinds of nanoparticles had similar hexahedral shapes and sizes (Figures 2a1–2a4 and Supporting Information Figures S17a and S17b). X-ray diffraction patterns further confirmed that the core and core–shell structures of the nanoparticles had β-type hexahedral structures ( Supporting Information Figure S16). Next, the PEI polymer was coated on the surface of the core–shell nanoparticles using the ligand exchange method,28 which not only made the nanoparticles water-soluble, but also provided an amino group that could covalently bind to NB3 and act as a masking agent for HOCl ( Supporting Information Figure S10). TEM, Fourier transform infrared (FTIR) spectroscopy, and the zeta potential spectra (Figure 2a3 and Supporting Information Figures S17–S19) showed that PEI was successfully coated on the surfaces of the two nanoparticles. Finally, we attached NB3 to the nanomaterials. The UV absorption spectra of the nanoparticles with equal concentrations showed that the grafting rate of NB3 on the surfaces of [email protected] and [email protected] were similar (4.27% and 5.3%, respectively) ( Supporting Information Figure S20). In addition, the FTIR and zeta potential spectra further confirmed that NB3 was successfully connected to the surfaces of the nanoparticles ( Supporting Information Figures S18 and S19). Figure 2 | (a) TEM images of NaYF4,Yb50% (1); NaYF4,Yb50%@NaYF4,Ho1% (2); NaYF4,Yb50%@NaYF4,Ho1%@PEI (3); and NaYF4,Yb50%@NaYF4,Ho1%@PEI @NB3 (4). Scale bar: 50 nm. (b) The upconversion fluorescence spectrum of [email protected]@NB3 (300 μg/mL) treated with ONOO− (0–20 μM) in 10 mM PBS at pH 7.4. (c) The fluorescence response of [email protected]@NB3 (300 μg/mL) to GSH (0–3 mM) after treatment with 10 μM ONOO− in 10 mM PBS at pH 7.4. (d) Reversible cycle response of [email protected]@NB3 (300 μg/mL) to 10 μM ONOO− and 3 mM GSH in 10 mM PBS at pH 7.4. (e) Fluorescence intensity ratio of [email protected]@NB3 (300 μg/mL) response to ONOO− (10 μM) and other analytes (100 μM, 50 μM for HOCl, 1 mM for GSH, Cys) in 10 mM PBS at pH 7.4. (1) Cys, (2) GSH, (3) Hcy, (4) Fe2+, (5) Fe3+, (6) H2O2, (7) H2S, (8) H2S2, (9) HOCl, (10) HSO3−, (11) CO32−, (12) SO32−, (13) Cl−, (14) HCO3−, (15) NO2−, (16) O2•−, (17) •OH, (18) SO42−, (19) t-BuOO−, (20) t-BuOOH, (21) Probe, (22) ONOO−, (23) ONOO− + HOCl, (24) ONOO− + H2O2, (25) ONOO− + O2•−, (26) ONOO− + •OH, (27) ONOO− + NO2−. Download figure Download PowerPoint Next, we tested the response of the two kinds of nanoprobes to ONOO−, respectively ( Supporting Information Figure S21). As expected, the results showed that the nanoprobe [email protected]@NB3 with the activator in the shell had a higher signal-to-background ratio than that of the nanoprobe [email protected]@NB3 with the activator in the core. This is because the luminescent centers of [email protected] were concentrated in the thin shell (∼3 nm), which reduced the distance between the luminescent centers and NB3. Thus, the energy transfer efficiency, which is closely related to the distance, was higher. In contrast, the luminescent centers of [email protected] were concentrated in the thicker core (approximately 22 nm), which is far from NB3. Therefore, [email protected]@NB3 was selected for further investigation. Figure 3 | Real-time reversible upconversion fluorescence imaging of exogenous ONOO− and GSH by nanoprobe [email protected]@NB3 in HepG2 cells. (a) The nanoprobe (200 μg/mL) was cultured with cells for 3 h, then treated with SIN-1 (400 μM) for another 1 h, and finally treated with 10 mM GSH for real-time imaging. The relative upconversion fluorescence intensity of green channel (b), red channel (c), and R/G (d) were acquired from (a). Scale bar: 50 μm. Download figure Download PowerPoint Figure 4 | Reversible upconversion fluorescence imaging of endogenous ONOO− and GSH by nanoprobe [email protected]@NB3 in RAW264.7 cells. The nanoprobe (200 μg/mL) was cultured with cells for 3 h (a), and then stimulated with 5 μg/mL LPS, 250 ng/mL IFN-γ for 4 h, and then PMA (5 μg/mL) for 30 min (b). TEMPO (1.5 mM) and GREs (5 mM) were then added to cells for 30 min after adding the above substances (c). The relative upconversion fluorescence intensity of green channel (d), red channel (e), and R/G (f). Scale bar: 50 μm. Download figure Download PowerPoint Spectral properties of [email protected]@NB3 responsive to ONOO− and GSH We investigated the reversible response of [email protected]@NB3 to ONOO− and GSH. With the addition of ONOO−, the fluorescence of the nanoprobe at 650 nm gradually increased, and the color of the solution gradually changed from blue to purple because of NB3 oxidation (Figure 2b). However, upon the addition of GSH, NBO was gradually reduced, which weakened the nanoprobe fluorescence at 650 nm, and the color of the solution changed from purplish red to blue (Figure 2c). The reversible response cycle of [email protected]@NB3 to ONOO−/GSH could be repeated at least four times ( Supporting Information Figure S22), and Figure 2d shows the reversible response of the nanoprobe to ONOO− and GSH without attenuation between two cycles. Therefore, the reversible nanoprobe can meet the requirements of subsequent cell and in vivo experiments. The titration spectrograms indicate that the response range of the nanoprobe to ONOO− was at the micromolar level but required millimolar GSH to reduce the product (Figures 2b and 2c). In a word, the reversible response range of the nanoprobe to ONOO−/GSH are well matched with the ONOO− and GSH levels in liver,39–41 which is the key to realize real-time imaging of liver injury and repair. Finally, the response rate of the nanoprobe to ONOO− and GSH were measured by time-dependent luminescent intensity. As shown in Supporting Information Figure S23, the response of the nanoprobe to ONOO− and GSH can be finished within seconds, which indicated that our probe has the potential for reversible real-time imaging. Next, we investigated the selectivity of [email protected]@NB3 to various reactive oxygen species (ROS), reactive nitrogen species (RNS), reactive sulfur species (RSS), and other inorganic ions. As shown in Figure 2e, the nanoprobe exhibited limited response to other analytes, including 50 μM HOCl ( Supporting Information Figure S9). Meanwhile, it exhibited good selectivity for ONOO−. In addition, the nanoprobe was stable at pH 4–11 ( Supporting Information Figure S24). Based on these results, we concluded that [email protected]@NB3 was suitable for the imaging of liver injury and repair. Reversible sensing of intracellular exogenous and endogenous ONOO− and GSH using [email protected]@NB3 To determine whether the nanoprobe could be used for biological imaging, we first explored the cytotoxicity of the nanoprobes using standard colorimetric thiazoyl blue (MTT) experiments. As shown in Supporting Information Figure S25, the cell viability remained over 85% when human hepatocyte carcinoma (HepG2) cells were incubated with 300 μg/mL nanoprobe for 24 h, indicating that our nanoprobe could be used as a bioimaging reagent. Next, the reversible responses to exogenous ONOO− and GSH in cells that were close to in vitro concentrations were investigated using the nanoprobe [email protected]@NB3. As shown in Figures 3a–3d, when the ONOO− donor 3-morpholinosydnonimine (SIN-1) was added to the cells preincubated with the nanoprobe,55 the upconversion fluorescence in the red channel (593–676 nm) was enhanced significantly, whereas that in the green channel (510–590 nm) was marginally enhanced. This led to an increase in the fluorescence intensity ratio Fred/FGreen (FR/G). However, after 10 mM GSH was added to the culture, the fluorescence of both channels gradually decreased over time, particularly in the red channel. After treatment with GSH for 60 min, the fluorescence of the two channels recovered to normal levels. The experimental results showed that the nanoprobe could be used for the reversible detection of exogenous ONOO− and GSH in HepG2 cells. Next, we investigated the endogenous ONOO−-sensing capacity of the nanoprobe. The HepG2 cells were cultured with the nanoprobe and then stimulated with LPS, IFN-γ, and PMA, which induce oxidative stress in cells.56,57 As shown in Supporting Information Figure S26, the fluorescence in the red channel was enhanced significantly compared with that in the green channel, and the FR/G value increased (see Supporting Information). However, when TEMPO, a ONOO− production inhibitor,58 was added to the cells treated with LPS, IFN-γ, and PMA, the fluorescence of the red and green channels decreased to levels similar to those in cells cultured only with the probe. These