Fluorogenic Autophagic Vesicle-Engaging Probe Enables High-Contrast and Homogeneous Imaging of Autophagic Flux in Live Cells

Abstract

Autophagic flux is a highly dynamic process essential for cellular homeostasis, yet its reliable visualization in live cells remains challenging due to the limitations of transfection-based LC3 reporters. Here, we present ATP3, a fluorogenic and ratiometric chemical probe that directly engages autophagic vesicles to enable the high-contrast and homogeneous imaging of autophagic flux without genetic manipulation. ATP3 constitutes a guanine targeting moiety and a smart fluorophore, with the former facilitating autophagic vesicle engagement, while the latter is intrinsically quenched and environment-responsive. ATP3 exhibits minimal background fluorescence and robust signal amplification upon autophagy-dependent engagement. Side-by-side comparisons demonstrate that ATP3 outperforms the conventional mRFP–GFP–LC3 assay in imaging contrast and staining uniformity. Furthermore, ATP3 enables dynamic monitoring of autophagic responses in an oxygen–glucose deprivation model, revealing progressive enhancement of autophagic flux under ischemic stress. Together, ATP3 provides a robust and broadly applicable chemical tool for visualizing autophagy dynamics in physiologically and disease-relevant contexts.

Introduction

Autophagy is an essential intracellular degradation pathway that maintains cellular homeostasis by orchestrating the turnover of proteins, organelles, and invading pathogens. − Dysregulation of autophagy has been implicated in a broad spectrum of pathological conditions, − including neurodegenerative diseases, cancer, metabolic disorders, and infection. Consequently, the reliable assessment of autophagic activity is of fundamental importance for both basic research and translational studies. However, autophagy is inherently a highly dynamic, multistep process that encompasses autophagosome formation, maturation, and fusion with lysosomes, posing substantial challenges for accurate detection and quantification. , Conventional biochemical approaches, such as Western blotting analysis of LC3 lipidation or p62 turnover, as well as transmission electron microscopy, remain widely used standards for autophagy evaluation. Despite their utility, these methods suffer from intrinsic limitations, including poor temporal resolution, population-averaged readouts, and labor-intensive sample preparation. More importantly, they cannot directly capture the dynamic progression of autophagic flux in living systems. The transient and highly regulated nature of autophagy, therefore, necessitates the development of analytical tools that can monitor autophagic processes in real time with high spatial and temporal resolution.

Fluorescence imaging has emerged as a powerful approach for interrogating autophagy dynamics in living cells. , Genetically encoded fluorescent-protein reporters, such as LC3-based dual-color constructs, , enable visualization of autophagosome maturation and lysosomal fusion (Figure A). Nevertheless, these methods require genetic manipulation, which can lead to variable expression levels, perturbation of endogenous pathways, and limited applicability in primary cells or difficult-to-transfect systems. Small-molecule fluorescent probes offer complementary advantages, including ease of use and compatibility with diverse biological models. However, most small-molecule probes reported to date do not directly recognize components of the autophagy machinery and are solely responsive to microenvironmental cues such as viscosity, pH, oxidative stress, etc (Figure B), − which makes achieving high specificity particularly challenging in the complex intracellular environments associated with autophagy progression. Recent advances in selective autophagy, particularly the recognition of conserved interactions between ATG8 family proteins and autophagy receptors, − have opened new avenues for probe design. Building on this principle, we and others have mimicked ATG8–receptor interactions by labeling LC3-interaction region (LIR)-like peptides with fluorophores, enabling selective imaging of autophagy-related structures. , The success of these approaches highlights the feasibility of directly harnessing components of autophagy machinery for imaging purposes. Another important advance in selective autophagy is the development of autophagy-targeting chimeras (AUTACs), which demonstrated that 8-substituted guanine derivatives can function as chemical signals to engage the autophagy machinery and promote cargo-selective degradation (Figure C). , These studies established a direct link between guanine-based motifs and autophagy engagement, providing a chemically tractable handle for modulating autophagic processes. Although AUTACs were originally conceived for targeted degradation, this concept also suggests the potential of guanine-based motifs as targeting units for dynamic autophagy imaging.

1.

Conceptual framework for autophagy imaging strategies and the design rationale of this study. (A) Conventional autophagy imaging approaches based on transfection-dependent fluorescent-protein-tagged LC3 reporters. (B) Previously reported small-molecule probes that rely on microenvironment-sensitive fluorogenic responses without direct engagement of the autophagy machinery. (C) AUTAC strategy for chemically mediated selective autophagy, in which 8-substituted guanine derivatives act as autophagy-engaging signals and promote cargo degradation. (D) Design concept of this work, which repurposes the AUTAC-inspired guanine motif to engage autophagic vesicles and integrates an environment-responsive fluorophore to enable smart and dynamic optical reporting of autophagy.

Herein, inspired by this chemical framework, we report a rationally designed small-molecule probe (ATP3) for highly specific and dynamic imaging of the autophagy process. By functionalizing the 8-position of guanine with a cysteine-based linker, we conjugated a smart fluorogenic reporter that exhibits a hydrophobic environment and restricted intramolecular rotation–induced fluorescence activation upon membrane association, along with pH-responsive spectral modulation during autophagosome-to-autolysosome transition. The guanine-based motif may be nonspecifically interactive with cellular proteins, subsequently engaging with the autophagy machinery. In this context, the smart fluorophore is exposed to the hydrophobic autophagy-related membranes, and its free rotation is inhibited. The dual-responsive design then enables sensitive visualization of autophagic vesicles and real-time reporting of autophagy progression (Figure D). Genetic knockout experiments further confirm the high specificity of the probe toward autophagosomes, validating its autophagy-dependent activation mechanism. Compared with classical fluorescent-protein-based reporters, this probe exhibits markedly enhanced imaging performance, including higher signal contrast and improved staining homogeneity across cell populations. Leveraging these advantages, we further demonstrate the utility of this probe for monitoring autophagic flux under oxygen–glucose deprivation conditions. Collectively, this work establishes guanine-based targeting as an effective chemical strategy for selective autophagy imaging and provides a versatile analytical tool for investigating autophagy dynamics in complex biological settings.

Design of the Autophagy Imaging Probe

The molecular design of the autophagy imaging probe was inspired by recent advances in chemically mediated selective autophagy, particularly the development of autophagy-targeting chimeras (AUTACs), which established that 8-substituted guanine derivatives can function as effective chemical signals to engage the autophagy machinery at a functional level (Figure C). , In the AUTAC framework, modification at the C8 position of guanine was shown to be chemically tolerated while retaining autophagy-targeting capability, thereby providing a synthetically accessible and modular handle for molecular engineering. We reasoned that this chemical motif, originally exploited for cargo degradation, could be repurposed as a targeting unit for dynamic autophagy imaging when coupled to an appropriate fluorescent reporter.

Guided by this concept, we selected guanine as the autophagy-engaging core and introduced a cysteine-based linker at the C8 position to enable site-specific conjugation (Figure A). The use of cysteine offers several advantages for probe construction, including well-defined connectivity, synthetic versatility, and compatibility with diverse functional modules. Importantly, this linker strategy spatially decouples the engaging motif from the reporter, thereby minimizing the potential interference of the fluorophore with the autophagy-guiding function of the guanine motif.

2.

Molecular design of ATP3 and its environment-responsive fluorescent behavior. (A) Chemical structure of ATP3 and schematic illustration of its environment-sensitive fluorescence activation and pH-dependent spectral modulation. (B) Fluorescence emission spectra of ATP3 in PBS at pH 7.0 and pH 4.0, in the absence or presence of hexadecyltrimethylammonium bromide (CTAB, 10 mg mL^–1), illustrating hydrophobic surface–induced fluorescence activation and acidification-triggered spectral shift. (C) Fluorescence emission spectra of ATP3 in PBS at pH 7.0 and 4.0, with or without 50% (v/v) acetonitrile (AcCN), demonstrating fluorogenic activation in hydrophobic solvent environments and pH-dependent emission red shift. (D) Fluorescence emission spectra of ATP3 in PBS at pH 7.0 and 4.0, with or without 90% (v/v) glycerol, revealing fluorescence enhancement upon restriction of intramolecular rotation and concomitant acidification-induced spectral modulation. ATP3 was used at a concentration of 10 μM, and fluorescence measurements were performed with an excitation at 405 nm.

For the signal-reporting module, we incorporated a smart fluorophore that was previously demonstrated to undergo fluorescence activation upon restriction of intramolecular rotation or exposure to hydrophobic environments and to exhibit a pronounced spectral shift in response to environmental acidification (Figure A). This fluorophore remains weakly emissive in aqueous environments due to both rotational freedom and efficient nonradiative quenching by high-energy O–H vibrations of water. However, upon preferential partitioning into autophagy-related membrane environments, which is promoted by the guanine-based autophagy-engaging motif, the fluorophore becomes strongly emissive as a result of increased hydrophobicity and restricted molecular motion. Furthermore, as autophagosomes mature and fuse with lysosomes, the accompanying acidification induces a distinct spectral modulation, enabling differentiation between the autophagic stages. By integrating these dual-responsive features (fluorogenic and ratiometric) into a single fluorophore framework, the probe is capable of reporting both the presence of autophagy-related structures and the dynamic progression of the autophagic process.

Collectively, this design integrates an autophagy-engaging guanine motif, a modular cysteine linker, and an environmentally responsive fluorogenic reporter into a unified small-molecule architecture (Figure A). The resulting probe, termed ATP3, is thus not merely a fluorescently labeled autophagy tag but a chemically engineered analytical tool capable of translating autophagy-associated microenvironmental changes into distinct optical signals, thereby enabling highly specific and dynamic imaging of autophagy in living cells.

Environment-Responsive Fluorescence Behavior of Probe ATP3

After synthesizing ATP3 following the procedures described in the Supporting Information (Scheme S1–S3), we systematically evaluated its photophysical behavior in solution to validate the rational design of its environment-responsive fluorogenic reporter. A series of model systems was employed to dissect the contributions of hydrophobic surface association, solvent polarity, and the restriction of intramolecular rotation to fluorescence activation.

In phosphate-buffered saline (PBS), ATP3 exhibited weak fluorescence emission, consistent with efficient nonradiative decay arising from unrestricted intramolecular rotation in aqueous environments. This weakly emissive ground state is highly desirable for imaging applications, as it minimizes background fluorescence prior to probe enrichment at target sites. To assess hydrophobic surface-enhanced fluorescence, we compared the emission of ATP3 in PBS with that in PBS containing 10 mg/mL hexadecyltrimethylammonium bromide (CTAB), which serves as a well-established model for hydrophobic and membrane-like interfaces. In the presence of CTAB, a pronounced fluorescence enhancement was observed (Figure B), indicating that surface association and increased hydrophobicity effectively activate the fluorogenic reporter. Moreover, decreasing the pH of the CTAB-containing solution induced a clear bathochromic shift in the emission spectrum, demonstrating the probe’s sensitivity to environmental acidification. Further verification in PBS (pH 7) containing Triton X-100 or sodium dodecyl sulfate (SDS) confirmed this observation, with increasing Triton X-100 or SDS proportion intensifying its emission (Figure S1). Noteworthy, in the PBS/Triton X-100 system, ATP3 demonstrated its maximum emission centered at 480 nm, while it centered at 555 nm in the PBS/SDS system, presumably due to local pH microenvironment change incurred by the charge property of the surfactants.

To further evaluate the influence of solvent polarity, the fluorescence of ATP3 was examined in aqueous/acetonitrile (AcCN) mixed solvents. Increasing the AcCN content to 50% resulted in a dramatic enhancement of fluorescence intensity relative to pure aqueous buffer, consistent with a reduced solvent polarity and diminished nonradiative quenching (Figure C). Notably, when the PBS component of the mixed solvent was adjusted to pH 4, a substantial red shift in the emission maximum was observed, further confirming the acid-responsive nature of the fluorophore under hydrophobic conditions. The solvatochromism of ATP3 was further verified by measuring its emission in various solvents, wherein its maximum emission wavelength linearly correlated with the solvent polarity parameter E _T (30) (Figure S2).

In addition to solvent polarity, the effect of restricted intramolecular motion was investigated using high-viscosity glycerol/water mixtures. In 90% glycerol, ATP3 displayed a strong fluorogenic response (Figure D), indicating that the restriction of intramolecular rotation alone is sufficient to activate fluorescence emission. As observed in the other model systems, acidification of the medium also led to a bathochromic shift of the emission spectrum, consistent with protonation-induced modulation of the electronic structure of the fluorophore.

To quantify the acid-environment-sensitive emission of ATP3, we measured the pK _a of the fluorophore by recording its weak emission in PBS of various pH (Figure S3). Acidification from pH 10 to pH 2 switched its maximum emission from 472 to 578 nm, and the pKa value was determined to be 5.4, which falls into the autolysosome-relevant range.

Collectively, these solution-phase spectroscopic studies confirm that ATP3 integrates three key photophysical features as designed: (i) minimal background emission in aqueous environments, (ii) strong fluorescence activation upon exposure to hydrophobic or membrane-associated environments and/or restriction of intramolecular rotation, and (iii) distinct acidification-induced spectral modulation. When combined with the autophagy-engaging capability conferred by the guanine motif, these properties enable ATP3 to translate autophagy-associated microenvironmental changes into distinct optical signals. This environment-responsive fluorescence behavior provides the analytical foundation for the probe’s ability to dynamically image autophagosomes and autolysosomes in living cells.

ATP3 Visualizes Autophagic Flux in Living Cells

Given the ability of ATP3 to report autophagy-associated microenvironmental changes through dual-channel fluorescence, we next evaluated its performance in monitoring autophagic flux in living cells. Autophagy was pharmacologically modulated using rapamycin (Rapa) to induce autophagy in combination with either wortmannin (Wort) or bafilomycin A1 (BafA1) as inhibitors. Wort, a class III phosphatidylinositol 3-kinase (PI3K) inhibitor, suppresses autophagosome formation at an early stage, whereas BafA1 inhibits vacuolar H⁺-ATPase activity and blocks autophagosome–lysosome fusion.

Cells were treated with Rapa alone, with Rapa in combination with Wort or BafA1, or left untreated as controls. Following ATP3 staining, dual-channel confocal imaging was performed by using a green channel and a red channel. Owing to the protonation-dependent red-shift emission of ATP3 in acidic compartments, we established the following imaging criteria: puncta exhibiting both green and red fluorescence (Green⁺Red⁺, yellow) were assigned as autophagosomes, whereas puncta displaying red fluorescence only (Green^–Red⁺) were classified as autolysosomes. Autophagic flux was quantitatively evaluated by counting yellow- and red-only puncta per cell.

Treatment with Rapa alone resulted in a marked increase in punctate fluorescence in both channels, with substantial colocalization consistent with enhanced autophagosome formation under stimulated autophagy. Notably, discrete puncta were also detected in untreated cells, indicating that ATP3 is sufficiently sensitive to visualize basal autophagy. In contrast, Wort treatment led to a pronounced reduction in fluorescence intensity and puncta number in both channels, consistent with the effective inhibition of autophagosome biogenesis. Importantly, in Rapa-treated cells, BafA1-mediated blockade of autophagic flux was clearly captured by ATP3, as evidenced by increased accumulation of autophagosomes and a concomitant decrease in autolysosome formation compared with Rapa treatment alone (Figure A–C). We also confirmed ATP3′s photostability under continuous imaging (Figure S4).

3.

ATP3 visualizes autophagic flux in living cells under pharmacological modulation. (A) Representative confocal images of HeLa cells treated with rapamycin (Rapa, 10 μM, 4 h) alone or in combination with wortmannin (Wort, 1 μM, 6 h) or bafilomycin A1 (BafA1, 100 nM, 2 h), followed by staining with ATP3. Cells without any drug treatment were used as controls. Scale bars: 10 μm; enlarged views: 2 μm. (B, C) Quantitative analysis of ATP3-labeled puncta in panel A. Yellow puncta (Red⁺Green⁺) were classified as autophagosomes, while red-only puncta (Red⁺Green^–) were classified as autolysosomes, enabling assessment of autophagic flux under different pharmacological conditions. Quantification was performed from n = 30 cells pooled from three independent experiments. Data are presented as mean ± SEM. Statistical significance was determined relative to control (*p < 0.05, **p < 0.01, ***p < 0.001) or rapamycin-alone treatment (#p < 0.05, ###p < 0.001).

Genetic Validation of ATP3 Specificity toward Autophagy-Related Membranes

While pharmacological perturbations demonstrated that ATP3 sensitively reports autophagic flux, small-molecule probes are often challenged by concerns regarding nonspecific responses to acidic or hydrophobic intracellular compartments. To rigorously validate the autophagy specificity of ATP3, we next employed a genetic approach that disrupts key components of the autophagy machinery that are essential for autophagosome membrane formation and maturation.

ATG5 and ATG7 are required for LC3 lipidation and autophagosomal membrane elongation, FIP200 is a core component of the ULK1 initiation complex, and LC3B represents a defining structural marker of autophagosome membranes. CRISPR–Cas9-mediated knockout of these genes is therefore expected to impair autophagosome biogenesis and disrupt autophagy-dependent membrane structures. Consistent with this expectation, ATP3 staining in ATG5, ATG7, FIP200, and LC3B knockout HEK293 cells (Figure A,C) revealed a pronounced reduction in punctate fluorescence compared with that of wild-type controls. Both green and red channel signals were markedly diminished (Figure B), indicating a loss of accumulation of ATP3 in autophagy-related compartments. Quantitative analysis confirmed a significant decrease in the puncta number per cell across all knockout lines (Figure D). These results demonstrate that ATP3 fluorescence activation requires an intact autophagy machinery and is tightly coupled to autophagy-dependent membrane structures, rather than arising from nonspecific labeling of acidic or hydrophobic intracellular compartments. This genetic validation establishes ATP3 as a highly specific probe for imaging autophagy-related membranes in living cells.

4.

Genetic validation of ATP3 specificity toward autophagy-related membranes. (A) Western blot analysis showing the successful knockout of the indicated genes. (B) Representative confocal images of wild-type (WT) and ATG5, ATG7, FIP200, or LC3B knockout (KO) HEK293 cells following incubation with ATP3, showing the impact of genetic disruption of core autophagy components on ATP3-labeled puncta formation. Scale bars: 10 μm; enlarged views: 2 μm. (C) Quantitative analysis confirming efficient CRISPR–Cas9-mediated knockout of ATG5, ATG7, FIP200, and LC3B in HEK293 cells. Quantification was performed from n = 3 independent experiments. (D) Quantification of ATP3-labeled puncta in panel B, where yellow puncta (Red⁺Green⁺) were classified as autophagosomes and red-only puncta (Red⁺Green^–) as autolysosomes. Quantification was performed from 26 to 27 cells per group, pooled from three independent experiments. Data are presented as mean ± SEM. Statistical significance was determined relative to WT cells (**p < 0.01, ***p < 0.001).

ATP3 Outperforms the mRFP–GFP–LC3 Assay in Analytical Performance

Following validation of ATP3 for visualizing autophagic flux, we next benchmarked its performance against the conventional mRFP–GFP–LC3 assay, a widely used transfection-based method for monitoring autophagy in living cells. Despite its broad adoption, the mRFP–GFP–LC3 assay is subject to inherent limitations, including elevated background fluorescence caused by overexpression of fluorescent-protein-tagged LC3 and substantial cell-to-cell variability arising from uneven transfection efficiency. In contrast, ATP3 operates through an intrinsically fluorogenic and ratiometric mechanism and does not require genetic manipulation, suggesting potential advantages in imaging performance and experimental robustness. To systematically compare these two approaches, we performed a series of side-by-side experiments.

We first compared their imaging contrast. In HeLa cells expressing mRFP–GFP–LC3, pronounced background fluorescence was observed, manifested as diffuse cytoplasmic signals following adenoviral infection (Figure A,B). By comparison, ATP3 exhibited minimal background fluorescence owing to its quenched ground state, resulting in well-defined punctate signals corresponding to autophagy-related structures (Figure C,D). Quantitative analysis of randomly selected intracellular regions showed that ATP3 achieved a markedly higher signal-to-background ratio (SBR), with an ∼15-fold improvement in both the green and red channels relative to the fluorescent-protein-based assay (Figure E). These results demonstrate that ATP3 provides substantially enhanced imaging contrast for autophagy visualization.

5.

ATP3 outperforms the mRFP–GFP–LC3 assay in autophagy imaging performance. (A) Representative confocal images of HeLa cells expressing mRFP–GFP–LC3 following transfection, showing both punctate and diffuse cytoplasmic fluorescence. Scale bars: 10 μm. (B) Line-scan analysis of fluorescence intensity profiles across representative regions in panel A, illustrating signal and background distributions for the mRFP–GFP–LC3 assay. (C) Representative confocal images of HeLa cells stained with ATP3, exhibiting discrete punctate fluorescence with a minimal cytoplasmic background. Scale bars: 10 μm. (D) Line-scan analysis of fluorescence intensity profiles across representative regions in panel C, highlighting the enhanced contrast achieved by ATP3. (E) Quantitative comparison of signal-to-background ratios for mRFP–GFP–LC3 and ATP3 in both green and red channels. Quantification was performed on 10 fields of view from three independent experiments. (F) Representative confocal images of HeLa cells expressing mRFP–GFP–LC3, with nuclei counterstained by DAPI (blue). GFP and mRFP signals are shown as green and red channels, respectively. Scale bars: 10 μm. (G) Representative confocal images of HeLa cells stained with ATP3, where yellow and red puncta in merged images correspond to autophagosomes and autolysosomes, respectively. Scale bars: 10 μm. (H) Quantitative comparison of labeling efficiency between the mRFP–GFP–LC3 assay and ATP3 staining, expressed as the percentage of cells exhibiting discernible autophagy-related puncta. Quantification was performed on 13 fields of view from three independent experiments.

Then their staining uniformity was compared. The mRFP–GFP–LC3 assay relies on genetic transfection, which frequently leads to heterogeneous protein expression and variable signal intensity across cells, thereby limiting quantitative reproducibility. In contrast, ATP3 is a cell-permeable small-molecule probe that requires no transfection, enabling a uniform probe delivery throughout the cell population. To directly assess staining homogeneity, basal autophagy in HeLa cells was monitored by using either mRFP–GFP–LC3 or ATP3. Cells infected with adenovirus encoding mRFP–GFP–LC3 were cultured for 36 h to allow protein expression, yet only ∼ 12% of cells exhibited discernible punctate fluorescence (Figure F,H). In striking contrast, a short 25 min incubation with ATP3 labeled nearly all cells with bright and well-defined puncta (Figure G,H). These results highlight ATP3′s superior staining uniformity, operational simplicity, and rapid applicability for live-cell autophagy imaging.

Collectively, these comparative studies establish ATP3 as a more effective analytical tool than the conventional mRFP–GFP–LC3 assay for monitoring the autophagic flux. By offering both markedly reduced background fluorescence and enhanced imaging contrast and highly uniform and reproducible staining across cell populations without the need for genetic manipulation, ATP3 provides a robust and broadly applicable alternative for studying autophagy under diverse experimental conditions.

Monitoring OGD-Induced Autophagic Flux in Live Cells

Autophagy is increasingly recognized as a highly dynamic process whose functional outcome depends on the disease context and stage. To determine whether ATP3 can capture such dynamic changes in autophagic flux under pathophysiologically relevant conditions, we applied this probe to an oxygen–glucose deprivation (OGD) model, a well-established in vitro paradigm of ischemia–reperfusion injury.

Cerebral ischemia has been shown to activate autophagy, and accumulating evidence suggests that excessive or dysregulated autophagy in brain microvascular endothelial cells may impair blood–brain barrier integrity and aggravate neuronal damage. , However, how autophagic flux evolves during ischemic stress remains incompletely understood, in part due to the lack of sensitive and homogeneous tools for live-cell flux monitoring. We therefore employed ATP3 to visualize temporal changes in autophagic activity following OGD exposure.

ATP3-based live-cell imaging in human brain microvascular endothelial cells (HBMECs) subjected to OGD for increasing durations (3, 6, and 12 h) revealed a progressive accumulation of red puncta, corresponding to autolysosomes, accompanied by a reduction in yellow puncta representing autophagosomes (Figure A,B). This shift in puncta composition indicates a sustained progression of autophagic flux with prolonged OGD treatment, rather than simple autophagosome accumulation. To independently validate these imaging results, we examined canonical autophagy markers by immunoblotting. Consistent with the ATP3 readouts, LC3-II levels increased in a time-dependent manner following the exposure to OGD, while the autophagy substrate p62 exhibited a concomitant decrease (Figure C,D), supporting the presence of ongoing and functional autophagic flux.

6.

ATP3 monitors the OGD-induced autophagic flux in living cells. (A) Representative confocal images of human brain microvascular endothelial cells subjected to oxygen–glucose deprivation (OGD) for the indicated durations (3, 6, and 12 h), followed by staining with ATP3. Scale bars: 10 μm, enlarged views: 2 μm. (B) Quantitative analysis of ATP3-labeled puncta in panel A, where yellow puncta (Red⁺Green⁺) were classified as autophagosomes and red-only puncta (Red⁺Green^–) as autolysosomes, revealing time-dependent changes in autophagic flux under OGD conditions. Quantification was performed from 38 cells per group, pooled from three independent experiments. (C) Immunoblot analysis of LC3 and SQSTM1 (p62) levels in HBMECs subjected to the indicated OGD treatments. (D) Densitometric quantification of LC3-II and SQSTM1 levels shown in panel C. Quantification was performed from n = 4 independent experiments. Data are presented as mean ± SEM. Statistical significance was determined relative to control (**p < 0.01, ***p < 0.001).

Taken together, these data demonstrate that ATP3 enables robust, real-time visualization of OGD-induced autophagic dynamics in live endothelial cells. Beyond validating ATP3 in a disease-relevant stress model, our results reveal that autophagic flux in HBMECs is progressively enhanced during ischemic challenges, underscoring the context-dependent role of autophagy in ischemic stroke pathology.

Conclusion

In summary, we have developed ATP3, a fluorogenic and ratiometric autophagy vesicle-engaging probe that enables sensitive, homogeneous, and high-contrast imaging of autophagic flux in live cells. By circumventing the need for genetic manipulation, ATP3 overcomes major limitations inherent to transfection-based LC3 reporters, including elevated background fluorescence and cell-to-cell variability. Through systematic benchmarking against the widely used mRFP–GFP–LC3 assay, we demonstrate that ATP3 provides a superior imaging performance, improved staining uniformity, and rapid usability. Importantly, ATP3 enables direct visualization of dynamic autophagic responses in physiologically and pathologically relevant settings, as illustrated by its application in ischemia-mimicking OGD models. Beyond serving as an alternative to conventional transfection-dependent LC3 reporters, ATP3 establishes a generalizable strategy for chemical imaging of autophagic flux based on the direct engagement of autophagy machinery and its fluorogenic activation. While ATP3 preferentially labels autophagic structures, caution is advised under conditions where other cellular compartments (e.g., late endosomes and lysosomes under certain perturbations) may exhibit similar microenvironmental features. Therefore, parallel control experiments, such as genetic or pharmacological inhibition of autophagy, are essential to confirm the autophagic origin of the observed signals. We anticipate that this approach, in combination with well-set control experiments, will facilitate more accurate dissection of autophagy dynamics across diverse biological systems and disease models and accelerate the development of chemical tools for studying complex, transient cellular processes.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/cbmi.5c00285.

• General chemistry experimental, chemistry synthesis and structure characterization, confocal imaging methods, Western blotting methods, and spectra traces (PDF)

The authors declare no competing financial interest.