A fluorescent probe for concurrent detection of cysteine, homocysteine, and superoxide anion

Abstract

Redox imbalance is a key factor in the pathogenesis of diseases such as epilepsy and liver injury. Superoxide anion (O₂^•−), cysteine (Cys), and homocysteine (Hcy) play central roles in maintaining redox homeostasis, and their dysregulation drives oxidative stress and disease progression. Here, we report a multifunctional fluorescent probe, BPC, capable of simultaneously and selectively detecting Cys, Hcy, and O₂^•− in complex biological environments. BPC shows high sensitivity, selectivity, and biocompatibility, enabling real-time visualization of redox fluctuations in living cells and zebrafish with minimal cytotoxicity. In pentylenetetrazole (PTZ)– and acetaminophen (APAP)–induced models of epilepsy and liver injury, BPC revealed notable alterations in Cys, Hcy, and O₂^•− levels, providing mechanistic insights into redox dysregulation. Moreover, BPC successfully tracked redox restoration following N-acetylcysteine (NAC) treatment. These findings establish BPC as a versatile tool for redox biology and highlight its promise for diagnostic and therapeutic applications.

Fluorescent probe enables real-time simultaneous tracking of superoxide anion, cysteine, and homocysteine.

INTRODUCTION

Redox imbalance is a critical factor in the pathogenesis of numerous diseases, including epilepsy and liver injury (1–4). Disruptions in redox homeostasis result in the excessive accumulation of reactive oxygen species (ROS), leading to oxidative stress and subsequent cellular damage (5–8). Epilepsy and liver injury are particularly vulnerable to oxidative stress due to their high metabolic activity and dependence on redox-sensitive signaling pathways (9–11). The superoxide anion (O₂^•−), one of the primary ROS, acts as a precursor to other reactive species such as hydrogen peroxide (H₂O₂), hydroxyl radicals (•OH), and singlet oxygen (¹O₂) (12, 13). O₂^•− is capable of directly interacting with metalloproteins to mediate signaling, whereas excessive O₂^•− promotes the generation of more highly ROS that indiscriminately damage biological molecules, thereby exacerbating cellular dysfunction and contributing to disease progression (14–18). Simultaneously, intracellular thiols such as cysteine (Cys) and homocysteine (Hcy) play essential roles in maintaining redox homeostasis by regulating antioxidant defense mechanisms, synthesizing glutathione (GSH), and participating in redox signaling pathways (19–22). Given the close relationship between oxidative stress and thiol metabolism, real-time monitoring of O₂^•−, Cys, and Hcy is essential for understanding their biological functions and their involvement in the pathophysiology of epilepsy and liver injury.

Despite the availability of various analytical methods for detecting O₂^•−, Cys, and Hcy—such as electrochemical analysis, chromatography, and mass spectrometry—these techniques often involve complex sample preparation, lengthy processing times, and are not well suited for real-time monitoring within live biological systems (23–25). In contrast, fluorescence imaging technology has emerged as a powerful tool due to its high sensitivity, excellent spatial and temporal resolution, and noninvasive nature (26–29). While numerous fluorescent probes have been developed to detect O₂^•−, Cys, and Hcy individually or in pairs (table S2) (12, 17, 30–38), the simultaneous and selective detection of all three in biological systems remains a notable challenge (39–43). To the best of our knowledge, there is now no fluorescent probe capable of simultaneously differentiating O₂^•−, Cys, and Hcy. Although the combined use of two different single-analyte probes has the potential to image two distinct biomarkers or molecular events concurrently, differences in their penetration capabilities, pharmacokinetics, and metabolism may affect their behavior and sensing performance in bioimaging applications (31, 36, 44, 45). Therefore, developing a single probe capable of differentiating and visualizing these key redox species in biological environments would provide critical insights into redox regulation and oxidative stress–related disease mechanisms.

To address these challenges, we designed and synthesized a multifunctional fluorescent probe, BPC, integrating benzopyran and coumarin derivatives linked via a piperazine group for the selective detection of Cys, Hcy, and O₂^•−. The unique structural design of BPC enables triple-channel fluorescence responses, allowing simultaneous differentiation of these redox-active species through distinct emission signals. The coumarin moiety serves as a reactive site for thiol-specific recognition via substitution and Michael addition, while the benzopyran unit facilitates O₂^•− detection through oxidative reaction mechanisms. BPC exhibits high sensitivity, selectivity, and biocompatibility, making it a robust tool for real-time visualization of redox fluctuations in living cells and zebrafish models with minimal cytotoxicity. Its specificity ensures minimal interference from endogenous biological molecules, enabling accurate monitoring in complex biological environments. To address the limitations of current redox probes in capturing subcellular redox heterogeneity, we developed a mitochondrial-targeted version (BPC) and a nontargeted cytoplasmic counterpart (BPC-NET). The mitochondrial probe selectively accumulates in the mitochondrial matrix, as confirmed by colocalization with organelle markers (Pearson’s correlation coefficient = 0.91). Using this tool, we investigated oxidative stress at the subcellular level. In paraquat (PQ)–induced oxidative stress in retinoic acid (RA)/12-O-tetradecanoylphorbol-13-acetate (TPA)–differentiated human SH-SY5Y dopaminergic neurons, mitochondrial BPC revealed pronounced mitochondrial O₂^•− accumulation and elevated Hcy levels, accompanied by a marked decrease in Cys. These results indicate a distinct redox imbalance within mitochondria during oxidative insult. In epilepsy and liver injury models induced by pentylenetetrazole (PTZ) and acetaminophen (APAP), BPC detected notable alterations in Cys, Hcy, and O₂^•− levels, providing valuable insights into oxidative stress and redox dysregulation in disease progression. In addition, BPC successfully tracked the restoration of redox balance following N-acetylcysteine (NAC) treatment in APAP-treated cells, highlighting its potential for evaluating therapeutic interventions. These findings highlight the potential of BPC not only as a highly effective fluorescent probe for redox biology research but also as a promising tool for advancing diagnostic and therapeutic strategies related to oxidative stress–related pathologies. The development of this multifunctional probe opens avenues for studying redox homeostasis in both fundamental and applied biomedical research.

RESULTS

Design of probe BPC

To date, numerous fluorescent probes have been developed for the individual detection and imaging of Cys, Hcy, and O₂^•−. However, because of the structural and reactive similarities of Cys and Hcy, their simultaneous discrimination and detection remain challenging. Moreover, a fluorescent probe capable of concurrently discriminating and detecting Cys, Hcy, and O₂^•− has not yet been reported. To address this challenge, we designed and synthesized probe BPC by conjugating a benzopyran fluorescent dye to a coumarin derivative via a piperazine linker. This structural design enables the selective discrimination and detection of Cys, Hcy, and O₂^•−. Inspired by previous studies (46), we used a protection-deprotection strategy for O₂^•− detection by incorporating a trifluoromethanesulfonyl group onto the benzopyran moiety. This strong electron-withdrawing group reduces the electron-donating ability of the phenolic anion, which in turn enhances the energy gap between the highest occupied molecular orbital and lowest unoccupied molecular orbital levels, thereby suppressing intramolecular charge transfer (ICT) and resulting in weak fluorescence. Upon reaction with O₂^•−, the protection group is removed, restoring the ICT process and leading to a pronounced fluorescence enhancement.

Furthermore, the introduction of an n-butylthio group at the four-position of the coumarin moiety enhances nonradiative decay through dynamic molecular motion, effectively reducing background fluorescence and improving the signal-to-background ratio of the probe (43). The reaction mechanism of probe BPC with Hcy and Cys follows the same principles established in our previous studies (Fig. 1) (47). However, it is noteworthy that this probe does not respond to GSH, which may be attributed to the steric hindrance imposed by the benzopyran moiety and the altered reactivity of the coumarin unit within the probe structure.

Fig. 1. Schematic illustration of this work.

(A) Proposed sensing mechanisms of probe BPC for the detection of Cys, Hcy, and O₂^•−; (B) its applications in biological systems.

Spectral response of probe BPC

To evaluate the spectral response of probe BPC, we first investigated its reactivity toward Cys, Hcy, and O₂^•− and determined the optimal testing conditions (fig. S2). The results indicate that the ideal solvent system for Cys and Hcy detection is a dimethyl sulfoxide (DMSO)–to–phosphate-buffered saline (PBS) ratio of 6:4, while a ratio of 4:6 is optimal for O₂^•− (figs. S3 and S4). As shown in Fig. 2, probe BPC exhibits a maximum absorption peak at 509 nm, which shifts upon the addition of Cys, Hcy, and O₂^•− or leads to the appearance of a new absorption peak in the near-infrared region, corresponding to 360, 480, and 650 nm, respectively. In agreement with expectations, probe BPC alone shows negligible fluorescence. However, upon interaction with Cys, Hcy, and O₂^•−, substantial fluorescence enhancements are observed at 456, 561, and 721 nm, respectively, with no spectral overlap, demonstrating the probe’s capability to effectively distinguish and detect these analytes. Further evaluation of the fluorescence response of probe BPC across a range of concentrations revealed a good linear correlation for Cys (0 to 15 μM), Hcy (0 to 15 μM), and O₂^•− (0 to 12 μM) (fig. S5). The calculated detection limits, based on signal-to-noise ratio (S/N) = 3, were 58.8 nM for Cys, 76.6 nM for Hcy, and 82.6 nM for O₂^•− (table S1), highlighting the probe’s high sensitivity for each analyte.

Fig. 2. Ultraviolet-visible and fluorescence properties of probe BPC.

(A) Absorption spectra of probe BPC (10 μM) upon addition of Cys, Hcy and O₂^•− (100 μM) at 25°C for 20 min. (B to D) The corresponding fluorescence spectra of probe BPC to Cys (λ_ex = 360 nm), Hcy (λ_ex = 480 nm), and O₂^•− (λ_ex = 650 nm). Condition: Cys, Hcy in DMSO/PBS (10 mM, pH = 7.4, v/v, 6/4) solution, O₂^•− in DMSO/PBS (10 mM, pH = 7.4, v/v, 4/6) solution, slits (nanometers): Cys, Hcy = 5.0/10.0, O₂^•− = 10.0/10.0. a.u., arbitrary units. Abs, absorption. FL, fluorescence.

To assess the selectivity of probe BPC, we measured its fluorescence response at specific excitation and emission wavelengths in the presence of potential interferents (including 10 mM GSH) (fig. S6). Upon excitation at 360 nm, the fluorescence enhancement signal of BPC toward Cys is substantially stronger than that of other analytes, demonstrating good selectivity. Similarly, under excitation at 480 and 650 nm, probe BPC exhibited fluorescence enhancements at 561 and 721 nm for Hcy and O₂^•−, respectively, with minimal interference from other species. To eliminate potential interference from trace metal ions—particularly iron—in PBS, which may catalyze the degradation of redox-sensitive species such as superoxide and hydrogen peroxide, EDTA was used as a metal chelator. Control experiments using metal-chelated PBS were performed to assess their impact on detection performance. As shown in fig. S7, these controls confirmed that trace metal ions in PBS did not significantly affect the selectivity of detection. Furthermore, to validate the specificity of probe BPC toward biological thiols, we examined its fluorescence response in the presence of various potentially competing species and included interference tests using oxidized forms of thiols (figs. S8 and S9). The results demonstrated that common biological interferents had negligible influence on the fluorescence intensity of BPC, confirming its good selectivity and specificity for the target analytes.

To evaluate the reactivity of probe BPC toward Cys, Hcy, and O₂^•−, we conducted real-time kinetic studies of its interactions with these analytes. As shown in fig. S10, the probe exhibited negligible fluorescence at excitation wavelengths of 360, 480, and 650 nm in the absence of analytes. Upon addition of Cys, fluorescence intensity at 456 nm increased gradually under 360-nm excitation, reaching a plateau within 10 min, with a 12-fold fluorescence enhancement. No appreciable fluorescence changes were observed with Hcy or O₂^•− at this wavelength, confirming the probe’s selective response to Cys without noticeable interference. Similarly, upon introducing Hcy and O₂^•−, fluorescence intensities at 561 and 721 nm increased under 480- and 650-nm excitation, respectively, stabilizing within 15 and 8 min, with 20-fold and 17-fold fluorescence enhancements, respectively. To further assess the practical applicability of probe BPC, we performed stability experiments by incubating the probe with its respective target analytes (fig. S11). The results confirmed that BPC maintains excellent fluorescence stability in the detection of Cys, Hcy, and O₂^•− over a 1-hour period. These favorable properties underscore the probe’s strong potential for reliable use in future biological and biomedical applications.

The pH stability of probe BPC was further examined by evaluating its fluorescence response over a pH range of 3.0 to 10.0 (fig. S12). The probe remained stable across this pH range, indicating its robustness under physiological conditions. Upon the addition of Cys, the fluorescence intensity at 456 nm initially increased and then decreased with pH changes from 5.0 to 10.0. In contrast, the fluorescence intensity at 561 and 721 nm corresponding to Hcy and O₂^•− increased steadily within the pH range of 6.0 to 10.0, indicating that basic conditions are favorable for monitoring Hcy and O₂^•−. These findings suggest that probe BPC exhibits good sensitivity under alkaline conditions, making it well suited for applications in biological environments within the normal physiological pH range. Thus, these comprehensive spectral studies confirm that probe BPC offers good sensitivity, selectivity, and rapid response capabilities for the simultaneous detection and discrimination of Cys, Hcy, and O₂^•−, demonstrating its potential for biomedical applications.

Sensing mechanism

On the basis of the spectroscopic analyses, we propose a detailed reaction mechanism for the probe BPC in response to Cys, Hcy, and O₂^•− (Fig. 1). BPC integrates a benzopyran-based hemicyanine fluorophore and a coumarin derivative through a piperazine linker, enabling highly selective detection of Cys, Hcy, and O₂^•− via the synergistic interplay of multiple reactive sites. For thiol recognition, Cys and Hcy specifically react with the coumarin moiety, yielding distinct fluorescent products. These products exhibit substantially different photophysical characteristics due to their divergent molecular structures and electronic configurations, thereby enabling simultaneous discrimination of Cys and Hcy through two independent fluorescence emission channels. These findings are consistent with our previous studies (41–43, 47, 48). Regarding O₂^•− detection, the trifluoromethane sulfonate group on the benzopyran hemicyanine moiety is selectively cleaved by O₂^•−, generating a phenolic hydroxyl group and restoring the conjugated π-system. This structural transformation activates a strong near-infrared fluorescence signal (12, 15, 17, 30, 36). Given the high reactivity and short lifetimes of the products formed during the reactions with Cys, Hcy, and O₂^•−, conventional purification techniques (e.g., column chromatography) are unsuitable. Instead, we used high-resolution mass spectrometry (HRMS) to successfully identify the Cys/Hcy adducts and confirm the structure of the O₂^•−-induced cleavage product (figs. S50 to S52).

To further validate the specificity of BPC for O₂^•−, we established a xanthine/xanthine oxidase (X/XO) system to generate superoxide under physiological conditions and performed inhibition experiments using superoxide dismutase (SOD) and catalase (CAT). The fluorescence intensity of BPC increased significantly upon treatment with X/XO (300 μM/0.01 U/ml) or KO₂ but was markedly suppressed by SOD (100 U/ml), while CAT (200 U/ml) had no appreciable effect (fig. S13). These results conclusively demonstrate the probe’s high specificity for O₂^•−. Together, these findings establish BPC as a robust, multichannel fluorescent probe capable of selectively detecting key redox-active species. Its ability to concurrently monitor thiol metabolism and oxidative stress renders it a powerful tool for investigating their interplay in complex biological systems and disease contexts.

Fluorescence imaging of Cys, Hcy, and O₂^•− in living cells

The cytotoxicity of probe BPC was evaluated using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The assay results demonstrated high cell viability over 24 hours (fig. S16A), highlighting the biocompatibility of probe BPC for cellular applications. In fluorescence imaging experiments, probe BPC effectively detected endogenous Hcy, Cys, and O₂^•− in HeLa cells, generating distinct fluorescence signals in three separate channels upon coincubation (G2 to G4 in Fig. 3). Specificity validation confirmed negligible interference from GSH, as no appreciable fluorescence signal was detected in the designated GSH detection channel. For the imaging of exogenous Hcy, Cys, and O₂^•−, fluorescence signals corresponding to Hcy and Cys were significantly attenuated upon treatment with N-ethylmaleimide (NEM), a well-established biothiol scavenger, while the fluorescence for O₂^•− remained largely unchanged (A2 to A4 in Fig. 3). Exposure to phorbol 12-myristate 13-acetate (PMA), an inducer of O₂^•− production, led to a marked increase in fluorescence intensity in the corresponding detection channel without interfering with the Cys and Hcy fluorescence channels (B2 to B4 in Fig. 3). Furthermore, after NEM treatment, the reintroduction of Hcy and Cys successfully restored their respective fluorescence signals, confirming the probe’s reversibility and reliability in detecting Hcy and Cys (C2 to D4 in Fig. 3). In addition, other reactive sulfur species did not induce strong fluorescence, demonstrating the probe’s specificity for Cys and Hcy. Treatment with NAC, a known antioxidant, further reduced the fluorescence signal for O₂^•−, reinforcing the probe’s sensitivity in monitoring oxidative stress–related species (E2 to E4 in Fig. 3). The subsequent addition of exogenous O₂^•− (100 μM) restored fluorescence intensity in the O₂^•− detection channel (F2 to F4 in Fig. 3), demonstrating BPC’s ability to dynamically track ROS fluctuations. These results highlight BPC’s effectiveness in distinguishing Cys, Hcy, and O₂^•−, establishing it as a robust tool for real-time oxidative stress analysis.

Fig. 3. Confocal fluorescence imaging of GSH, Hcy, Cys, and O₂^•− in HeLa cells.

(A) (A1 to A5) Cells treated with NEM (100 μM) for 30 min were further incubated with probe BPC (5 μM) for 45 min; (B1 to B5) cells previously treated with PMA (100 μM) for 60 min were further incubated with probe BPC (5 μM) for 45 min; (C1 to D5) cells previously treated with NEM (100 μM) for 30 min were incubated with probe BPC (5 μM) for 45 min, and then (100 μM) exogenous Hcy (C1 to C5) and Cys (D1 to D5) were added to incubate with cells for 30 min, respectively; (E1 to E5) cells treated with NAC (100 μM) for 30 min were further incubated with probe BPC (5 μM) for 45 min; (F1 to F5) cells previously treated with NAC (100 μM) for 30 min were incubated with probe BPC (5 μM) for 45 min, and then exogenous PMA (100 μM) was added to incubate with cells for 30 min; (G1 to G4) only probe BPC (5 μM) was added and incubated with cells for 45 min. (B) Corresponding normalized fluorescence intensities of the three channels in (A). Scale bar, 25 μm.

Similar fluorescence response patterns were observed in SH-SY5Y and HepG2 cells (figs. S17 and S18), demonstrating the probe’s consistent performance across different cell lines. These findings collectively highlight the capability of probe BPC to simultaneously discriminate and visualize endogenous and exogenous Hcy, Cys, and O₂^•− in living cells with exceptional sensitivity and selectivity. Given its robust performance, probe BPC holds great promise as a versatile imaging tool for biological and medical research, particularly in the study of redox biology and thiol-related cellular processes.

Mitochondria are the primary site of intracellular oxidative stress, where excessive generation of ROS and compromised antioxidant defenses within the mitochondrial matrix constitute critical mechanisms of cellular dysfunction in various diseases (6, 18, 21, 49). However, redox signaling and oxidative stress dynamics within mitochondrial microdomains are often masked in whole-cell analyses. Although cationic fluorescent probes can target mitochondria via membrane potential-driven accumulation (50, 51), existing strategies remain inadequate for simultaneous and real-time imaging of key redox regulators—specifically Cys, Hcy, and O₂^•−—within mitochondria. To address this limitation, we incorporated a delocalized cationic moiety into the BPC probe structure to enhance mitochondrial targeting through electrostatic interactions. Localization studies (fig. S19) confirmed that BPC strongly colocalized with the mitochondrial marker MitoTracker Red, with a Pearson correlation coefficient of 0.91. In contrast, moderate colocalization was observed with other organelles such as the endoplasmic reticulum, Golgi apparatus, and lysosomes (fig. S20). This efficient mitochondrial accumulation is attributed to the probe’s positive charge, which facilitates selective retention within the negatively charged mitochondrial matrix.

To rigorously validate the role of the cationic moiety in subcellular targeting, we designed a structurally analogous control probe lacking this feature, termed BPC-NET, and confirmed its structure via HRMS and nuclear magnetic resonance (NMR) spectroscopy (figs. S47 to S49). In vitro assays showed that BPC-NET retained selective recognition for Cys, Hcy, and O₂^•− (fig. S15), providing a reliable control for spatial specificity studies. Compared to BPC, BPC-NET exhibited markedly different intracellular distribution, showing diffuse cytoplasmic localization and negligible mitochondrial enrichment (fig. S20). Together, this complementary probe pair—BPC (mitochondria targeted) and BPC-NET (cytoplasmic)—enables comparative analysis of redox dynamics across distinct cellular compartments, providing a powerful platform for subcellular oxidative stress mapping.

We next applied this platform to oxidative stress studies in SH-SY5Y cells, a human neuroblastoma line capable of differentiating into dopaminergic neurons. Differentiation was induced using a 7-day protocol with sequential treatment of RA (10 μM) and TPA (80 nM), following established methods (52–56). Morphological analysis (fig. S21B) revealed that RA/TPA-treated cells developed extended neurites characteristic of mature neuronal phenotypes, in contrast to the cluster-like morphology of undifferentiated cells. Differentiation was further confirmed by quantitative polymerase chain reaction analysis, which showed clear up-regulation of neuronal markers β-tubulin III and tyrosine hydroxylase (fig. S22 and table S2). MTT assays demonstrated negligible cytotoxicity of BPC and BPC-NET in differentiated neurons within the tested concentration range (0 to 10 μM) (fig. S21, C and D).

To explore subcellular redox dynamics under stress conditions, we treated differentiated SH-SY5Y neurons with PQ (2.5 to 100 μM) and its mitochondria-targeted derivative Mito-PQ (2.5 to 100 μM). Redox changes were monitored using both BPC and BPC-NET (Fig. 4). Our data revealed a dose-dependent increase in mitochondrial O₂^•− and Hcy levels, accompanied by a notable reduction in mitochondrial Cys. Notably, the redox perturbations observed in mitochondria were more pronounced than those in the cytoplasm (Fig. 4, D to F), highlighting the specific oxidative damage induced by PQ-like compounds within mitochondrial compartments and validating the compartment-specific targeting and sensing capabilities of BPC.

Fig. 4. Effects of Mito-PQ and PQ on oxidative stress induction during SH-SY5Y cells differentiation.

(A) In the cases of different concentrations of Mito-PQ and PQ, (A1 to A4) cells were only incubated with probe BPC (5.0 μM) for 45 min; (B1 to F4) cells were incubated with different concentrations of Mito-PQ for 45 min and then incubated with probe BPC (5.0 μM) for another 45 min. (G1 to L4) Cells were only incubated with probe BPC-NET (5.0 μM) for 45 min; (B1 to F4) cells were incubated with different concentrations of PQ for 45 min and then incubated with probe BPC-NET (5.0 μM) for another 45 min. Scale bar, 25 μm. (B and C) Fluorescence signal changes of Hcy, Cys, and O₂^•− of BPC and BPC-NET. (D to F) Dynamic comparison diagrams of the changes of BPC and BPC-NET with different concentrations of Mito-PQ and PQ in each channel. Data are presented as mean ± SD, n = 3. Statistical significance was calculated by one-way analysis of variance and least significant difference (LSD) post hoc test. *P < 0.05, **p < 0.01, ***P < 0.001, and ****P < 0.0001. ns, not significant.

Fluorescence imaging of Cys, Hcy, and O₂^•− in zebrafish

To validate the BPC’s ability to detect redox biomarkers in complex living systems, fluorescence imaging studies in a zebrafish model were conducted. Three-day-old zebrafish were incubated with BPC (5 μM) to monitor endogenous Cys, Hcy, and O₂^•− levels. As shown in fig. S23 (F1 to F4), distinct fluorescence signals were observed in three detection channels (456 nm for Cys, 561 nm for Hcy, and 721 nm for O₂^•−), confirming the probe’s ability to simultaneously differentiate these analytes in vivo. To assess BPC’s specificity for biothiols, zebrafish were pretreated with the thiol-blocking agent NEM (100 μM), leading to a notable reduction in Cys- and Hcy-associated fluorescence signals (A1 to A4, fig. S23), verifying their endogenous thiol origin. Notably, O₂^•− fluorescence remained unchanged, indicating no cross-reactivity between thiol and ROS detection pathways. BPC’s reversibility was further validated by reintroducing exogenous Cys (100 μM) and Hcy (100 μM) into NEM-pretreated zebrafish, which restored fluorescence in their respective channels (C1 to D4, fig. S23). In addition, treatment with the O₂^•− inducer PMA (100 μM) significantly increased O₂^•−-associated fluorescence (B1 to B4, fig. S23). Conversely, pretreatment with the antioxidant NAC (100 μM) suppressed O₂^•− fluorescence, which was restored upon subsequent O₂^•− addition (E1 to E4, fig. S23), confirming the probe’s specificity for ROS detection. Thus, BPC is well suited for the simultaneous in vivo detection of Cys, Hcy, and O₂^•−.

Fluorescence imaging of Hcy, Cys, and O₂^•− dynamics in PTZ-induced epilepsy models

To assess the imaging capability of probe BPC in detecting epilepsy-related biomarkers at the cellular level, a PTZ-induced SH-SY5Y cell injury model was used. PTZ, a well-known proconvulsant drug, is widely used to simulate cellular damage associated with epileptic conditions. Before cellular incubation, we systematically assessed the potential interference of NEM, NAC, PTZ, and APAP on the intrinsic fluorescence properties of probe BPC (fig. S14). Control experiments demonstrated that these compounds did not induce substantial changes in the spectral characteristics or response kinetics of the probe. This validation confirms that BPC maintains high stability and selectivity, ensuring the accuracy of analyte-specific signal detection in subsequent biological applications. Upon incubation with probe BPC, SH-SY5Y cells exhibited notable fluorescence signals in the respective channels for Hcy, Cys, and O₂^•− (A1 to A3 in Fig. 5). With increasing PTZ concentrations (0.1 to 0.5 mM), the fluorescence intensities corresponding to Hcy and O₂^•− exhibited a progressive increase, while the fluorescence signal of Cys showed a gradual decline. This trend may be attributed to PTZ-induced oxidative stress within the cells, leading to an elevated production of ROS, which subsequently promotes the accumulation of Hcy and O₂^•−. Conversely, Cys, known for its antioxidant properties, may be consumed to counteract oxidative stress, resulting in its decreased fluorescence signal. These findings indicate that probe BPC can effectively visualize and quantify endogenous Hcy, Cys, and O₂^•− in PTZ-induced SH-SY5Y cell, offering a powerful imaging tool for investigating the molecular mechanisms underlying PTZ-induced cellular damage.

Fig. 5. Dynamic fluorescence imaging of Hcy, Cys, and O₂^•− in a PTZ-induced epilepsy model of SH-SY5Y cells.

(A) (A1 to A4) Cells incubated with probe BPC only (5.0 μM) for 45 min; (B1 to D4) add 0.1 mM PTZ (B1 to B4), 0.3 mM PTZ (C1 to C4), and 0.5 mM PTZ (D1 to D4), respectively, and incubate with cells for 12 hours and then with probe BPC (5.0 μM) for 45 min. Scale bar, 25 μm. (B) Corresponding normalized fluorescence intensities of the three channels in (A). Data are presented as mean ± SD, n = 3. Statistical significance was calculated by one-way analysis of variance and LSD post hoc test. **P < 0.01, ***P < 0.001, and ****P < 0.0001.

In addition to cellular studies, the imaging performance of probe BPC was further explored using PTZ-induced zebrafish epilepsy models to monitor the dynamic changes of Hcy, Cys, and O₂^•− in vivo. Zebrafish were coincubated with PTZ for varying durations (3, 6, and 12 hours), followed by fluorescence imaging with probe BPC. The results revealed a time-dependent increase in the fluorescence intensities of Hcy and O₂^•−, while the fluorescence signal of Cys progressively weakened over time (Fig. 6). These observations suggest that prolonged PTZ exposure exacerbates oxidative stress and metabolic imbalance in zebrafish, further validating the capability of probe BPC to monitor epilepsy-related biomolecular changes in living systems. Thus, probe BPC demonstrates remarkable sensitivity to PTZ-induced oxidative stress and metabolic alterations both at the cellular and zebrafish levels. Its ability to effectively track the dynamic changes of Hcy, Cys, and O₂^•− in the zebrafish epilepsy model provides a valuable strategy for studying the biochemical mechanisms of epilepsy and related neurological disorders. Furthermore, the high specificity and sensitivity of probe BPC pave the way for its potential applications in epilepsy research and therapeutic development.

Fig. 6. Dynamic fluorescence imaging of Hcy, Cys, and O₂^•− in a PTZ-induced zebrafish epilepsy model.

(A) (A1 to A4) Zebrafish incubated with probe BPC only (5 μM) for 45 min; (B1 to D4) PTZ added (6 mM) was incubated with zebrafish for 3 hours (h; B1 to B4), 6 hours (C1 to C4), and 12 hours (D1 to D4) and then incubated with probe BPC (5.0 μM) for 45 min. (B) Corresponding normalized fluorescence intensities of the three channels in (A). Scale bar, 500 μm. Data are presented as mean ± SD, n = 3. Statistical significance was calculated by one-way analysis of variance and LSD post hoc test. **P < 0.01 and ****P < 0.0001.

Fluorescence imaging of Hcy, Cys, and O₂^•− dynamics in cells and mouse tissues within the DILI model

Following, we used an APAP-induced HepG2 cell model to simulate drug-induced liver injury (DILI) and evaluated the fluorescence imaging capability of probe BPC for detecting dynamic changes in Hcy, Cys, and O₂^•− levels. The experiment was divided into three groups: a control group, an APAP-induced experimental group, and a treatment group supplemented with NAC. In the control group, HepG2 cells were incubated with probe BPC alone, yielding strong fluorescence signals in the Hcy, Cys, and O₂^•− channels (Fig. 7), reflecting baseline biomarker levels under normal physiological conditions. In the APAP-induced experimental group, exposure to varying concentrations of APAP led to a marked decrease in fluorescence intensities corresponding to Hcy and Cys, while a substantial increase in the O₂^•− signal was observed. These changes suggest that APAP-induced oxidative stress depletes Cys and Hcy while promoting excessive accumulation of ROS, particularly O₂^•−, a key contributor to oxidative damage and hepatocellular dysfunction. Notably, these findings differ from those observed in PTZ-induced epilepsy cell models. In contrast, cells in the NAC-treated group exhibited a notable restoration of fluorescence signals for Hcy and Cys, along with a notable reduction in O₂^•− fluorescence intensity. This suggests that NAC, a well-known antioxidant and thiol donor, effectively counteracts APAP-induced oxidative stress by replenishing cellular thiol levels and reducing oxidative damage. These findings highlight the potential of probe BPC in monitoring oxidative stress–related biochemical changes and assessing therapeutic interventions in DILI at the cellular level.

Fig. 7. Dynamic fluorescence imaging of Hcy, Cys, and O₂^•− in an APAP-induced liver injury model of HepG2 cells.

(A) (A1 to A4) Cells incubated with probe BPC only (5.0 μM) for 45 min; (B1 to D4) add 200 μM APAP (B1 to B4), 500 μM APAP (C1 to C4), and 1000 μM APAP (D1 to D4) and incubate the cells for 12 hours and then incubate with probe BPC (5.0 μM) for 45 min. (E1 to E4) First, add 1000 μM APAP to incubate with cells for 12 hours, then add 400 μM NAC to incubate with cells for 1 hour, and lastly incubate with probe BPC for 45 min. (B) Corresponding normalized fluorescence intensities of the three channels in (A). Data are presented as mean ± SD, n = 3. Statistical significance was calculated by one-way analysis of variance and LSD post hoc test. **P < 0.01 and ****P < 0.0001. Scale bar, 25 μm.

To further validate the applicability of probe BPC in vivo, fluorescence imaging was performed on liver tissue sections from mice subjected to APAP-induced acute liver injury and compared to normal mice (Fig. 8). The results revealed that APAP-treated mice exhibited a notable reduction in fluorescence intensities in the Hcy and Cys channels, while the O₂^•− signal was markedly increased, mirroring the findings observed in the HepG2 cell model. Following NAC treatment, fluorescence signals for Hcy and Cys were partially restored, while O₂^•− fluorescence was significantly reduced, confirming that probe BPC can effectively track biomarker fluctuations associated with oxidative stress and antioxidant intervention in a living system. Overall, these findings establish probe BPC as a sensitive and reliable tool for the real-time visualization of redox-related biomarkers in APAP-induced liver injury models at both cellular and tissue levels. Its high selectivity and specificity in detecting dynamic changes in Hcy, Cys, and O₂^•− make it a valuable asset for investigating the molecular mechanisms underlying DILI and evaluating the efficacy of potential therapeutic agents.

Fig. 8. Fluorescence imaging of endogenous Hcy, Cys, and O₂^•− in liver tissue sections of normal mice and APAP-induced acute liver injury model mice.

(A) (A1 to A4) Liver tissue sections from normal mice were coincubated with probe BPC (5 μM) for 45 min. (B1 to B4) Liver tissue sections from mice with acute liver injury were incubated with probe BPC (5 μM) for 45 min. (C1 to C4) Liver tissue sections from mice with acute liver injury were incubated with NAC (400 μM) for 1 hour and then coincubated with probe BPC (5 μM) for 45 min. (B) Corresponding normalized fluorescence intensities of the three channels in (A). Data are presented as mean ± SD, n = 3. Statistical significance was calculated by one-way analysis of variance and LSD post hoc test. **P < 0.01 and ****P < 0.0001. Scale bar, 50 μm.

DISCUSSION

In this study, we successfully designed and developed a multifunctional fluorescent probe, BPC, capable of rapid, sensitive, and selective detection of cysteine (Cys), homocysteine (Hcy), and superoxide anion (O₂^•−) across various biological systems through distinct excitation and emission channels. Probe BPC exhibited excellent biocompatibility and low cytotoxicity, enabling efficient imaging of both endogenous and exogenous Cys, Hcy, and O₂^•− in living cells and zebrafish models. Moreover, probe BPC proved to be a valuable tool for investigating the dynamic changes of these biomarkers in epilepsy and liver injury, effectively monitoring oxidative stress and cellular damage in real time, highlighting its potential for studying redox imbalances in pathological conditions. The versatility of probe BPC offers promising opportunities for further exploration in redox biology, with potential applications in the development of targeted therapeutic strategies. Future studies could extend its use to other oxidative stress–related diseases, such as neurodegenerative disorders and metabolic syndromes, thereby broadening its diagnostic and therapeutic potential.

MATERIALS AND METHODS

Synthesis of the probe BPC

The synthetic route for probe BPC is illustrated in fig. S1. Detailed synthesis procedures, along with the NMR and mass spectra of the intermediates, are provided in the Supplementary Materials.

Compound 1-8 (60.0 mg, 82.21 μmol) was dissolved in 8 ml of anhydrous dichloromethane, followed by the addition of 4-dimethyl-aminopyridine (DMAP) (5.1 mg, 41.11 μmol). The reaction mixture was stirred at room temperature for 10 min. Subsequently, compound 2-3 (57.7 mg, 82.21 μmol) was added, and the mixture was stirred for an additional 5 min. Last, EDCI (20.1 mg, 123.32 μmol) was introduced, and the reaction was allowed to proceed overnight at room temperature. Upon completion, the reaction mixture was concentrated and purified via column chromatography, yielding the probe BPC as a blue-violet solid (39.3 mg, 36.8% yield). ¹H NMR (500 MHz, DMSO-d₆) δ 8.62 (d, J = 15.0 Hz, 1H), 8.36 (d, J = 3.3 Hz, 1H), 8.22 to 8.17 (m, 1H), 8.06 (d, J = 13.6 Hz, 1H), 7.94 to 7.87 (m, 4H), 7.74 to 7.66 (m, 2H), 7.59 to 7.51 (m, 2H), 7.45 to 7.43 (m, 2H), 6.97 (d, J = 9.4 Hz, 1H), 6.76 (d, J = 14.8 Hz, 2H), 4.56 (d, J = 7.7 Hz, 2H), 3.61 to 3.53 (m, 8H), 3.17 (s, 3H), 3.06 to 3.03 (m, 2H), 2.75 to 2.72 (m, 4H), 1.86 to 1.84 (m, 2H), 1.82 (s, 6H), 1.54 to 1.51 (m, 2H), 1.44 to 1.41 (m, 3H), 1.35 to 1.30 (m, 3H), 1.27 to 1.23 (m, 5H), 1.20 to 1.17 (m, 2H), 0.80 to 0.77 (m, 3H). ¹³C NMR (700 MHz, CD₃OD) δ 180.72, 167.18, 166.27, 160.45, 160.42, 157.81, 157.50, 156.52, 155.97, 154.10, 146.01, 145.80, 144.24, 141.13, 140.62, 134.63, 132.49, 131.92, 131.63, 130.37, 130.33, 129.98, 129.95, 129.48, 129.04, 128.91, 126.73, 126.41, 125.16, 125.09, 119.34, 114.00, 113.73, 111.48, 111.43, 111.38, 111.34, 110.11, 97.68, 64.61, 64.40, 64.30, 63.27, 62.82, 61.67, 60.81, 52.09, 52.07, 45.96, 40.42, 37.50, 33.17, 27.41, 27.31, 22.61, 14.25, 13.81, 13.13, 12.78. ¹⁹F NMR (659 MHz, DMSO) δ −72.64. HRMS (ESI⁺), calculated for C₆₂H₆₂F₃N₆O₈S₃⁺ [M-I]⁺ calcd: 1171.3718, found: 1171.3710 (fig. S1).

Cell culture

HeLa, SH-SY5Y, and HepG2 cell lines were obtained from the China Center for Type Culture Collection (Wuhan, China). The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) heat-inactivated fetal bovine serum (FBS) (Gibco BRL, Grand Island, NY, USA) and penicillin-streptomycin (100 U/ml and 100 μg/ml, respectively). The cells were maintained in a humidified incubator at 37°C under an atmosphere of 5% CO₂ and 95% air.

SH-SY5Y cell differentiation protocol

Differentiation of SH-SY5Y cells was initiated by seeding undifferentiated cells at a density of 1 × 10⁵ cells/cm². On day 0, cells were cultured in complete medium containing 10% FBS. On day 1, the medium was replaced with differentiation medium supplemented with 1% FBS and RA (10 μM) to induce neuronal commitment. On day 5, the medium was refreshed with 1% FBS supplemented with RA (10 μM) and TPA (80 nM) to promote neurite outgrowth and maturation. From day 7 onward, cells were maintained in 1% FBS medium containing TPA (80 nM) to sustain differentiation. Morphological characterization was performed on day 8, confirming successful differentiation, as evidenced by extended neuritic processes characteristic of mature neuron-like phenotypes.

Imaging of endogenous and exogenous Cys, Hcy, and O₂^•− in living cells

To monitor endogenous Cys, Hcy, and O₂^•− in living cells, HeLa, SH-SY5Y, and HepG2 cells were incubated with probe BPC (5 μM) in serum-free DMEM containing 0.5% (v/v) DMSO for 60 min. After incubation, the cells were washed three times with PBS and subsequently imaged. For the detection of exogenous Cys, Hcy, and O₂^•−, the cells were pretreated with NEM (a biothiol scavenger, 0.1 mM) for 30 min and PMA (0.1 mM) for 60 min. Following pretreatment, the cells were incubated with Cys/Hcy/O₂^•− (100 μM) for 30 min, followed by incubation with probe BPC (5 μM) for an additional 60 min. After incubation, the cells were washed three times with PBS before imaging.

Fluorescence imaging was performed using a Leica SP8 confocal microscope (Leica, Germany). All experiments were conducted in triplicate. The excitation (λ_ex) and emission (λ_em) wavelengths were as follows: 405-nm excitation with 430- to 480-nm emission for the blue channel, 476-nm excitation with 530- to 580-nm emission for the red channel, and 633-nm excitation with 680- to 730-nm emission for the pink channel.

Imaging of Cys, Hcy, and O₂^•− in PTZ-induced cell

SH-SY5Y cells were cultured in confocal dishes, and when the cell adhesion rate reached 70 to 80%, PTZ was added at final concentrations of 0.1, 0.3, and 0.5 mM. The cells were incubated for 12 hours, washed three times with PBS, and subsequently incubated with probe BPC (5 μM) for 45 min. After incubation, the cells were washed three times with PBS before fluorescence imaging. In the control group, cells were incubated with probe BPC (5 μM) for 45 min, followed by three washes with PBS before imaging. All experiments were conducted in triplicate to ensure reproducibility.

Imaging of Cys, Hcy, and O₂^•− in paracetamol (APAP)–induced cell

HepG2 cells were cultured in confocal dishes for imaging experiments. In the experimental group, when the cell adhesion rate reached 70 to 80%, APAP was added at final concentrations of 200, 500, and 1000 μM, followed by incubation for 12 hours. The cells were then washed three times with PBS and incubated with probe BPC (5 μM) for 45 min, followed by another three washes with PBS before fluorescence imaging. In the treatment group, cells were treated with 1000 μM APAP for 12 hours, followed by the addition of 400 μM NAC for 1 hour. The cells were then washed three times with PBS, incubated with probe BPC (5 μM) for 45 min, and washed again with PBS before fluorescence imaging. In the control group, cells were incubated with probe BPC (5 μM) for 45 min, washed three times with PBS, and imaged directly. All experiments were performed in triplicate to ensure reproducibility.

Imaging of endogenous Cys, Hcy, and O₂^•− in zebrafish

We sincerely thank Y. Deng (Zebrafish Genetics Laboratory, College of Life Science, Hunan Normal University, Changsha, China) for providing the zebrafish, which were constructed according to (57). For in vivo detection of endogenous Cys, Hcy, and O₂^•−, 2- to 4-day-old zebrafish were used. The zebrafish were incubated with probe BPC (5 μM) in E3 embryo medium containing 0.5% (v/v) DMSO at 28°C for 60 min, followed by fluorescence imaging. Before imaging, all zebrafish were terminally anesthetized using MS222. Fluorescence imaging was performed using a Leica SP8 confocal microscope (Leica, Germany). All experiments were conducted in triplicate to ensure reproducibility. The excitation (λ_ex) and emission (λ_em) wavelengths used were 405 nm (430 to 480 nm) for the blue channel, 476 nm (530 to 580 nm) for the red channel, and 633 nm (680 to 730 nm) for the pink channel.

Imaging of Cys, Hcy, and O₂^•− in a zebrafish epileptic model induced by PTZ

Zebrafish were cultured in six-well plates for imaging studies. In the experimental group, zebrafish were treated with 6 mM PTZ and incubated for 3, 6, and 12 hours, respectively. Following incubation, the zebrafish were washed three times with PBS buffer and subsequently incubated with probe BPC (5 μM) for 45 min. After incubation, the zebrafish were washed again three times with PBS, placed on a slide surface, and fixed using tissue fixation solution for fluorescence imaging. In the control group, zebrafish were directly incubated with probe BPC (5 μM) for 45 min, washed three times with PBS, and similarly placed on a slide surface with tissue fixation solution for imaging.

Imaging of Cys, Hcy, and O₂^•− in mouse tissue

Six-week-old male C57BL/6J mice were purchased from Hunan Lacke-Jingda Animal Co. Ltd. Acute liver injury was induced by intraperitoneal injection of APAP (200 μl, 300 mg/kg) for 3 hours. After treatment, liver tissue samples were collected from both normal and APAP-treated mice and then cryosectioned at a thickness of 10 μm using a CM1860UV cryotome. Tissue imaging was performed using a Leica SP8 confocal microscope (Leica, Germany). All experiments were conducted in triplicate to ensure reproducibility.

Ethical approval

All animal procedures were conducted in compliance with the guidelines and approved by the Animal Ethics Committee of Hunan Normal University (no. 2022-161).

Supplementary Materials

This PDF file includes:

Supplementary Text

Tables S1 to S3

Figs. S1 to S52