ABSTRACT
-
Background/Aims
Ferroptosis, recently emerged as a new cell death modality characterized by iron-dependent peroxidation of lipids, has been explored in various diseases. However, detection of ferroptosis, particularly in chronic liver disease models, is hampered by the lack of universal ferroptosis markers and limited number of fluorescence sensors for in vivo ferroptosis.
-
Methods
In this study, we developed TTM-4 as a highly sensitive near-infrared (NIR) fluorescent probe to detect ferroptosis.
-
Results
TTM-4 exhibited turn-on fluorescence upon viscosity change, enabling visualization of lipid peroxidation (LPO) in ferroptotic hepatocytes and liver tissue samples with greater sensitivity than BODIPY 581/591 C11. Timelapse live-cell imaging of erastin-treated cells revealed real-time LPO dynamics involving cytosolic lipid droplets (cLDs), endoplasmic reticulum, and nuclear LDs in a chronological order. Further gene expression analysis of 216 liver tissue samples from the NCBI GEO database showed a significant increase in CIDEC concurrent with TTM-4 fluorescence during progression to metabolic dysfunction-associated steatotic hepatitis (MASH). TTM-4, with its low toxicity and turn-on NIR emission during ferroptosis, also enabled in vivo visualization of ferroptosis in liver injury and metabolic dysfunction-associated steatotic liver disease (MASLD) models.
-
Conclusions
Our findings suggest that TTM-4 enables monitoring of ferroptosis in MASLD and would aid in early MASH diagnosis.
-
Keywords: Near-infrared emissive probe; Viscosity; Nuclear lipid droplets; Ferroptosis; Metabolic dysfunction-associated steatotic liver disease
Study Highlights
• We developed TTM-4 as a highly sensitive NIR fluorescent probe to detect ferroptosis. TTM-4 exhibited turn-on fluorescence upon viscosity change, enabling visualization of lipid peroxidation in ferroptotic hepatocytes and liver tissues with greater sensitivity than BODIPY 581/591 C11. TTM-4, with its low toxicity and turn-on fluorescence during ferroptosis, also enabled in vivo visualization of ferroptosis in liver injury and MASLD models.
Graphical Abstract
INTRODUCTION
Ferroptosis is a type of iron-dependent regulated cell death characterized by the accumulation of lipid-based reactive oxygen species (lipid ROS) generated via lipid peroxidation (LPO) [
1]. Two major antioxidant systems, involving glutathione peroxidase 4 (GPx4), which reduces toxic lipid peroxides (R-OOH) to their corresponding alcohols (R-OH) in a glutathione-dependent reaction, and ferroptosis suppressor protein-1 (FSP1), which catalyzes the regeneration of non-mitochondrial Coenzyme Q10 using NAD(P)H [
2,
3], have been demonstrated to suppress ferroptosis. Any perturbation related to these protective systems can cause a failure to protect cells against oxidative damage, eventually leading to ferroptotic cell death. Recent studies have revealed the involvement of ferroptosis in a variety of physiological processes. Disorders of iron and lipid homeostasis lead to various diseases, including anemia [
4], neurodegenerative diseases [
5], cardiovascular diseases [
6], diabetes [
7], liver diseases [
8], and cancers [
9]. Ferroptosis promotes the progression of chronic liver diseases to hepatocellular carcinoma [
10]. Moreover, ferroptosis has emerged as a critical driver of metabolic dysfunction-associated steatotic liver disease (MASLD) progression, as iron-dependent LPO in hepatocytes not only induces cell death but also amplifies inflammatory and fibrotic signaling, thereby promoting the transition from simple steatosis to metabolic dysfunction-associated steatohepatitis (MASH) [
11,
12]. Therefore, gaining an in-depth insight into ferroptosis could offer an alternative therapeutic or diagnostic strategy for chronic liver diseases. In particular, the development of chemical probes for detecting ferroptosis at an early stage is clinically essential.
Changes in the cell microenvironments, including viscosity, polarity, and pH, play pivotal roles in regulating physiological processes [
13-
20]. Over the past few years, the development of fluorescent probes for detecting such changes in ferroptotic cells has significantly expanded, greatly enhancing our understanding of ferroptosis [
21,
22]. Viscosity plays a critical role in diffusion-dependent reactions by controlling reaction rates across the cell, thereby influencing signal transduction, metabolite diffusion, and biomolecular interactions. Anomalous cellular viscosity is linked to various ferroptosis-related pathologies, including neurodegenerative diseases and MASLD [
23,
24]. Organic fluorescent probes designed to monitor ferroptosis based on molecular rotors, wherein the suppression of nonradiative intramolecular rotations enhances emission under high-viscosity conditions, have served as highly valuable tools for understanding ferroptosis [
25,
26]. These viscosity-sensitive fluorescent probes have been used to visualize alterations in diverse subcellular organelles, including the endoplasmic reticulum (ER) [
27-
29], mitochondria [
30-
33], lipid droplets (LDs) [
34-
36], lysosomes, and the plasma membrane [
37], during ferroptosis. Nevertheless, such single-organelle-targeted probes fail to provide information regarding the interplay between these organelles. Therefore, the development of new fluorescent probes to explore interorganelle interactions during ferroptosis would offer significant advantages in understanding the physiological mechanisms and regulation of ferroptosis in disease-related contexts. To date, only a few ferroptosis sensors with direct clinical applications are available. Moreover,
in vivo imaging of ferroptosis remains challenging. Chemical probes capable of detecting ferroptosis at an early stage of chronic liver disease, including MASLD, are essential for early diagnosis of ferroptosis in such patients and are expected to prevent the progression of irreversible liver fibrosis.
In this study, we developed a highly sensitive near-infrared (NIR) turn-on fluorescent probe, TTM-4, to monitor the spatiotemporal emergence of LPO during ferroptosis. This probe was used to investigate the role of ferroptosis in the progression of MASLD using
in vitro and
in vivo models (
Fig. 1). Our findings provide a robust foundation for the application of this probe in early diagnosis and medical intervention of chronic liver diseases.
MATERIALS AND METHODS
Optical characterization
The UV-Vis absorption and fluorescence emission spectra of the synthesized compounds were measured at room temperature using an Orion AquaMate 8100 UV-VIS spectrophotometer (Thermo Scientific, WI, USA) and a FP-8350 fluorescence spectrometer (JASCO, Tokyo, Japan), respectively. The photostability of the synthesized compounds was assessed by SpectraMax iD3 Multi-Mode Microplate Reader (Molecular Devices, CA, USA) under glycerol, MeOH, and DW conditions. Samples were irradiated every 25 seconds and photostability was monitored for 90 minutes.
Human liver tissue samples and mouse experiments
This study was conducted in accordance with the principles outlined in the Declaration of Helsinki regarding research involving human participants. It was approved by the Institutional Review Board (IRB) of Boramae Hospital (IRB No. 16-2013-45), and written informed consent was obtained from all participants in the study cohort. Human liver specimens were obtained using a 16-gauge disposable needle, immediately immersed in liquid nitrogen, embedded in optimal cutting temperature compound, and cryosectioned. The sections were subsequently fixed in 4% paraformaldehyde. Appropriate samples were selected for confocal microscopy, and BODIPY 581/591 C11 and TTM-4 fluorescence were visualized. In all animal experiments, 7–9 weeks old male C57BL/6J and ICR mice (Narabio, Seoul, Korea) were maintained in the specific-pathogen-free facility of Seoul National University Institute of Laboratory Animal Resources (Seoul, Korea). All animal experiments were conducted with appropriate welfare measures throughout the study period. All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University (SNU-241021-1-1, SNU-250122-2) and were conducted in accordance with the guidelines specified for the management and use of laboratory animals.
Statistical analysis
The significance of differences between two groups were analyzed for significance using two-tailed Student’s t-tests. The P-values are indicated as *P<0.05, **P<0.01, ***P<0.001, and ****P<0.0001.
RESULTS
Design and synthesis of probes
The structures of the developed NIR probes are shown in
Figure 1A. For the donor-π bridge-acceptor framework, 1
H-indene-1,3(2
H)-dione or its derivatives with malononitrile, which possess strong electron-withdrawing ability, were used as acceptor units, and dimethylamino- or methoxyphenyl groups were adapted as donor units. An additional double bond was inserted between the thiophene and phenyl ring to contribute to a bathochromic shift of the spectra due to elongated π conjugation. As evident from density functional theory (DFT) calculations (
Supplementary Fig. 1), the lowest unoccupied molecular orbital (LUMO) was mainly located on the indandione moiety, whereas the highest occupied molecular orbital (HOMO) was mainly located on the dimethylaminophenyl ring, indicating a considerable intramolecular charge transfer effect. The synthetic procedures for these compounds, along with the characterization data, are provided in Supporting Information.
The optical properties of the synthesized compounds are shown in
Supplementary Figures 2 and
3,
Supplementary Tables 1 and
2. The absorption maxima of TTM-4 and TTO-6 were detected at approximately 570 nm. The emission maxima for TTM-4 range from 670 to 840 nm in various solvents with large Stoke shifts (>150 nm). We investigated the fluorescence response of the synthesized probes to the addition of glycerol to samples dissolved in water (
Fig. 1C). When the solvent was changed from water to glycerol, the emission intensities of TTM-4, TTO-4, and TTO-6 increased significantly by 508-, 1650-, and 95-fold, respectively (
Fig. 1D,
Supplementary Fig. 2G,
2H, and
3G,
3H). However, the increase in fluorescence from TTM-6 was much lower (
Supplementary Fig. 3C,
3D). We also measured the stability of fluorescence from TTM-4 and TTO-4 by irradiation every 25 seconds. Enhanced fluorescence of these compounds in glycerol remained the same over time, in contrast to the decrease in fluorescence detected in methanol (
Supplementary Fig. 5). Based on these results, we selected TTM-4, exhibiting the most bathochromic-shifted fluorescence emission with an emission maximum at 812 nm, for further
in vivo imaging. The time-correlated single-photon counting measurements confirmed that the lifetime of TTM-4 in glycerol was greater than in water (
Supplementary Fig. 6).
To confirm the viscosity sensitivity of TTM-4, HepG2 cells treated with 10 mM H
2O
2 were imaged via confocal microscopy. TTM-4 fluorescence was markedly increased in treated cells compared to controls, showing over 80-fold signal-to-noise enhancement (
Fig. 2A1,
2A2). This response was attenuated by co-treatment with liproxstatin-1 (Lip-1), β-mercaptoethanol, or N-acetyl-L-cysteine, suggesting redox dependency. Temperature-dependent fluorescence further confirmed viscosity sensitivity (
Supplementary Fig. 7). Since ferroptosis involves LPO and redox imbalance [
3], we assessed TTM-4 in cells treated with ferroptosis inducers erastin and RSL3. Strong red signals were observed (
Fig. 2B1,
2B2), which were suppressed by ferrostatin-1 (Fer-1) and Lip-1, indicating TTM-4’s utility in monitoring ferroptosis via lipid-ROS.
To localize initial LPO, cells were treated with various TTM-4 concentrations. At 200 nM, fluorescence was concentrated in LDs, while higher doses revealed spread to other organelles (
Fig. 2C). Compared to BODIPY 581/591 C11, TTM-4 showed stronger red-shifted “turn-on” emission, better for
in vivo use, and required lower concentrations (<1 μM vs. 5 μM). MTT assays confirmed minimal cytotoxicity (
Supplementary Fig. 8), supporting 1 μM as an optimal dose. We also compared the sensitivity of TTM-4 in ferroptotic cells with that in apoptotic or pyroptotic cells. The TTM-4 fluorescence intensity in staurosporine (an apoptosis-inducing agent)- and lipopolysaccharide (LPS; a pyroptosis-inducing agent)-treated cells was significantly lower than that in RSL3-treated cells, indicating that the increase in viscosity in organelles occurred more rapidly in ferroptotic cells than in apoptotic or pyroptotic cells (
Fig. 2D1,
2D2,
Supplementary Fig. 9). Taken together, these data demonstrated that TTM-4 selectively detects LPO during ferroptosis.
To assess TTM-4’s organelle-specific detection of LPO, ferroptotic HepG2 cells were co-stained with BODIPY 493/503 or endoplasmic reticulum (ER) Tracker. TTM-4 signals showed high colocalization with LDs and ER (Pearson’s r=0.85 and 0.75;
Fig. 3A,
3B), indicating that ferroptosis induces LPO primarily in these compartments. Control experiments with other markers, including MitoTracker, LysoTracker, and Golgi-Tracker, revealed weaker fluorescence signals of TTM-4 in these organelles (
Supplementary Fig. 10). Notably, nuclear lipid droplets (nLDs) were detected in RSL3- or erastin-treated cells using TTM-4 (
Fig. 3C and
Supplementary Video 1). 3D imaging with Hoechst dye confirmed their nuclear localization, with nLDs (white arrows) clearly distinct from nucleoli (yellow arrowheads) (
Supplementary Fig. 11). Time-lapse imaging over 4 hours of erastin-treated cells showed dynamic TTM-4 signal progression from cytosolic LDs (cLDs) to the nucleus via ER (
Fig. 3D,
3E,
Supplementary Fig. 12A,
Supplementary Videos 2,
3). Unlike BODIPY 493/503 detecting neutral lipids, TTM-4 marked oxidized LDs. Aggregation and fusion of cLDs preceded nLD formation, suggesting early LPO occurs in cLDs, followed by oxidation in nLDs during later ferroptosis stages. Sołtysik et al. [
38] reported enhanced nLD formation in hepatocytes under ER stress. Our data suggest that in ferroptosis, nLD formation may represent an adaptive response to ER stress, potentially aiding early MASLD detection. To investigate regulatory mechanisms, we analyzed nLDs under ferroptosis inhibition (
Supplementary Fig. 13). Co-treatment with Fer-1 or deferoxamine (DFO; an iron chelator) significantly reduced oxidized LD and nLD populations (
Supplementary Fig. 13C), indicating their effective suppression of lipid ROS. Post-treatment with Fer-1 further decreased oxidized LDs and nLDs, supporting its therapeutic potential. However, DFO post-treatment lowered total oxidized LDs but not nLDs, suggesting nLDs may arise from cytoplasmic or ER-derived oxidized lipids, rather than direct nuclear Fe
²⁺-driven peroxidation.
MASH is the most severe manifestation of MASLD that can progress to cirrhosis and hepatocellular carcinoma [
39]. Recent studies highlight lipotoxicity and oxidative stress as central drivers of MASH pathogenesis [
40,
41]. To explore this, we tested whether lysosomal acid lipase (LAL) inhibition by lalistat-2 modulates TTM-4 signals in HepG2 cells. LAL inhibition, which mimics lipid accumulation by blocking autophagy-mediated lipid clearance, significantly reduced TTM-4 signals in the ER surrounding cLDs and decreased oxidized nLDs (
Fig. 4A1,
4A2), while cLD oxidation remained unchanged. These findings suggest that ER-mediated transfer of peroxidized lipids from cLDs to nLDs is disrupted by lalistat-2, implicating LD breakdown as a key step in nLD formation during ferroptosis. Given that LDs form contacts with organelles via membrane contact sites (MCSs), we further modeled MASLD
in vitro using palmitic acid (PA)-treated HepG2 cells, which showed increased TTM-4 fluorescence and elevated oxidized LDs and nLDs (
Fig. 4B1,
4B2), consistent with clinical findings of nLD accumulation in MASH patients [
42]. Since ferroptosis is increasingly recognized as a contributor to MASH, we examined whether suppression of ACSL3, a key enzyme in ferroptosis and LD biogenesis, affects lipid oxidation. Eicosapentaenoic acid (EPA) treatment suppressed ACSL3, and significantly reduced oxidized LDs and nLDs in PA-treated cells, indicating the protective role of EPA against lipotoxic ferroptosis. To validate these findings in a more physiological setting, we used AML12 mouse hepatocytes and induced ferroptosis using RSL3, which elicited strong TTM-4 fluorescence suppressed by DFO (
Fig. 4C1,
4D), confirming ferroptotic specificity. Notably, TTM-4 outperformed BODIPY 581/591 C11 in detecting oxidized lipids in both RSL3- and PA-treated cells (
Fig. 4E,
4F). These results position TTM-4 as a superior probe for monitoring ferroptosis and oxidative lipid remodeling in MASH models and highlight nLDs as a potential biomarker for ferroptotic stress in steatohepatitis.
Given the promising capability of TTM-4 for ferroptosis detection, we applied it to liver tissue samples. In liver biopsy samples from MASLD patients (human liver fibrosis stage 3), ferroptosis was frequently detected using TTM-4 in regions with high expression of the profibrotic marker α-SMA (
Fig. 5A). To mimic the pathological stages observed in human MASLD samples, C57BL/6J mice were fed a choline-deficient, L-amino acid-defined, high-fat diet (CDAHFD) for 1, 6, or 12 weeks. Ferroptosis was detected at each time point using TTM-4 to evaluate its effectiveness. At the 12-week time point, which corresponded to a stage similar to that shown in Figure 5A, TTM-4 successfully detected ferroptosis in the mouse liver tissue. Additionally, ferroptosis was observed in the same regions as the neutral lipids, detected using BODIPY 493/503, which indicated that TTM-4 can detect oxidized lipids in damaged liver tissues (
Fig. 5B). Similar to the findings in cells, in the experiment evaluating detection at different time points after CDAHFD feeding, TTM-4 exhibited fluorescence earlier in the week 1 samples than did BODIPY 581/591 C11, which underscores the higher sensitivity of TTM-4 to ferroptosis in the liver tissue samples than that of BODIPY 581/591 C11 (
Fig. 5C). As mentioned earlier, this was believed to be due to the characteristic detection of TTM-4 over a longer wavelength range (
Fig. 2E). Similar to the
in vitro findings, detection using TTM-4 confirmed that oxidized LDs, initially observed in the cytoplasm at week 1, were abundant in the nucleus at week 6. This pattern persisted through week 12 (
Fig. 5D). Overall, these results indicated that TTM-4 can track the translocation of oxidized lipids in liver tissue samples.
As shown in
Supplementary Videos 2 and
3, cLDs underwent oxidation and clustering prior to the detection of lipid peroxides in the ER, with CIDE proteins implicated in this process. While CIDEB is predominant under normal conditions, CIDEA and CIDEC are elevated during metabolic disorders such as MASLD, promoting LD–LD fusion and lipid storage, a process that may reduce lipotoxicity [
43]. We analyzed CIDE gene expression in a MASLD patient cohort from Seoul National University Boramae Medical Center. Among CIDE family genes, CIDEC expression consistently increased with disease progression, correlating with both steatosis and fibrosis severity (
Fig. 6A1,
6A2,
Supplementary Fig. 14A). Public dataset GSE135251 (n=216) confirmed this trend, as CIDEC showed the most significant upregulation from healthy controls to advanced NAFLD (
Fig. 6A3–
6A5). This trend was recapitulated in a mouse CDAHFD model, especially after 12 weeks (
Fig. 6B1). To further assess this, primary hepatocytes treated with 100 or 300 μM PA showed significant CIDEC induction at 300 μM (
Fig. 6B2). Similarly, AML12 cells treated with increasing PA concentrations demonstrated a dose-dependent CIDEC elevation, unlike CIDEA (
Fig. 6B3,
Supplementary Fig. 14B). TTM-4 and BODIPY 493/503 imaging revealed that 250 μM PA increased cLDs, while 1,000 μM PA reduced neutral lipids (BODIPY 493/503 signal) but enhanced oxidized lipid accumulation (TTM-4 signal), coinciding with increased CIDEC expression (
Fig. 6C). In human liver biopsies, BODIPY 493/503 signal decreased and TTM-4 signal increased from steatosis to MASH, suggesting progression from lipid accumulation to LPO (
Fig. 6D). In advanced MASH (aMASH), oxidized nLDs were elevated (
Fig. 6E). Finally, confocal colocalization of TTM-4 and CIDEC in aMASH liver tissue confirmed overlapping signals, indicating that CIDEC upregulation contributes to oxidized LD formation and ferroptotic stress (
Fig. 6F).
Ferroptosis was induced in mice using either chemical (tetrachloride; CCl
4) or dietary (methionine- and choline-deficient; MCD) models. A single intraperitoneal injection of 20% CCl
4 followed by 24 hours incubation led to increased TTM-4 fluorescence in liver tissue, indicating enhanced ferroptosis (
Fig. 7A1,
Supplementary Fig. 15A,
15B). Similarly, mice fed with MCD diet for 3 weeks also exhibited stronger TTM-4 signals compared to controls (
Fig. 7A2,
Supplementary Fig. 15C,
15D), confirming diet-induced ferroptosis. Given its long emission wavelength, TTM-4 was suitable for noninvasive IVIS imaging. To ensure safety, 5 mg/kg of TTM-4 was administered via tail vein, and serum markers of liver (AST, ALT, and GGT) and kidney function (BUN, and CRE) were measured 2 hours later, revealing no acute toxicity (
Fig. 7B). For
in vivo live imaging, mice underwent fasting (24 hours for CCl
4; final day for MCD diet) to minimize autofluorescence, followed by TTM-4 injection and IVIS imaging 2 hours later (
Fig. 7C1). Liver damage was confirmed by gross morphology (
Fig. 7C2,
7C3) and elevated AST and ALT levels in serum (
Fig. 7D1,
7D2). Strong IVIS fluorescence was consistently detected in ferroptosis-induced mice. To pinpoint the signal source, five major organs were dissected, revealing highest fluorescence in the liver (
Fig. 7E). Overall, both CCl
4 and MCD diets effectively induced ferroptosis
in vivo, and TTM-4 enabled real-time, noninvasive imaging of ferroptosis, as validated by quantitative IVIS analysis (
Fig. 7F1,
7F2).
DISCUSSION
In this study, we synthesized TTM-4, a viscosity-sensitive probe capable of selectively detecting lipid peroxides across various organelles during ferroptosis. Unlike conventional lipid dyes, TTM-4 responded specifically to peroxidized lipids, enabling the visualization of dynamic changes in lipid localization during disease progression. Live-cell imaging demonstrated that lipid peroxides initially accumulated in cLDs, then clustered and translocated to the nucleus, likely mediated by ER stress pathways. This spatiotemporal pattern of lipid peroxide migration was recapitulated in liver tissues from MASH models, suggesting the potential of TTM-4 as an early diagnostic tool for ferroptotic liver injury.
Chemically, TTM-4 possesses a donor-π bridge-acceptor framework in the structure, classifying it as a push–pull dye and thus conferring sensitivity to solvent polarity, as indicated in
Supplementary Figure 2. While solvatochromic dyes exhibit strong fluorescence in lipid membranes, TTM-4 showed only minimal fluorescence under basal cellular conditions (
Supplementary Fig. 2A1,
2B1,
2C), indicating that polarity alone could not drive its activation in the cellular context. LPO is known to increase cellular polarity due to the formation of polar functional groups, including hydroperoxide (-OOH), aldehyde (-CHO), carboxylic acid (-COOH). Previously reported polarity-sensitive probes [
44,
45] have shown decreased fluorescence upon treatment with ferroptosis inducers, reflecting the increased polarity by oxidized lipids. Consistently, we also found that TTM-4 exhibited reduced fluorescence in the enhanced polarity condit ion (e.g., higher water contents in methanol) (
Supplementary Fig. 4). However, TTM-4 showed a net increase in fluorescence intensity under LPO-inducing conditions, which is contrary to the expectation based solely on polarity effects. This finding indicates that viscosity changes outweigh the opposing polarity effect. After accounting for the polarity-induced contributions, the residual fluorescence enhancement could be predominantly driven by viscosity changes associated with LPO.
Upon erastin treatment, prolonged glutathione depletion led to excessive LPO in the ER, supporting the idea of lipid transfer from cLDs to the ER (
Supplementary Fig. 12A,
Supplementary Videos 2,
3). This lipid relay, likely mediated by oxidized PUFAs, may propagate ferroptosis. Interestingly, TTM-4-labeled nLDs suggest lipid peroxides reach the nucleus, although the exact route of ER-to-nLD lipid transfer remains unclear. Notably, nLD formation occurred independently of iron, indicating redistribution of ER- or cLD-derived lipids rather than direct nucleoplasmic peroxidation. Given previous reports linking nLDs with ER stress mitigation, these results hint at a compensatory role for nLDs in MASH under lipotoxic conditions.
To explore potential regulators of this process, we investigated the role of CIDE proteins and identified a significant correlation between MASLD and CIDEC expression using public GEO datasets. Confocal microscopy confirmed colocalization of CIDEC with TTM-4-labeled lipid droplets. Given that CIDEC has been previously reported to promote lipid droplet fusion and enlargement [
46,
47], this finding provides supportive evidence for the spatial distribution of LPO during ferroptosis. Unexpectedly, CIDEC knockdown exacerbated ferroptosis (
Supplementary Fig. 16), implying a potentially protective function of CIDEC. Several studies have demonstrated that CIDEC facilitates the stabilization of lipid storage within hepatocytes, thereby reducing lipotoxic stress and protecting against cellular injury [
48-
50]. In contrast, other reports indicate that excessive CIDEC expression contributes to hepatic lipid overload and exacerbates liver injury [
51,
52]. These conflicting findings imply that maintaining an optimal level of CIDEC is critical to homeostasis of hepatocytes. Despite growing interest, the role of CIDEC in hepatocytes remain relatively underexplored compared to adipocytes. Additional studies are needed to clarify the mechanistic basis and pathological functions of hepatocyte CIDEC.
While FibroScan remains the non-invasive standard tool to assess liver fibrosis, it cannot detect early-stage ferroptotic liver damage. In contrast, TTM-4 offers a novel imaging-based strategy for detecting ferroptotic injury prior to fibrosis or irreversible cirrhosis. Although its in vivo selectivity for liver-specific ferroptosis remains to be refined, our study is the first to propose TTM-4 as a diagnostic imaging agent for MASH/MASLD and to visualize ferroptosis in vivo using IVIS. Collectively, our findings establish TTM-4 as a promising ferroptosis-sensitive probe with potential applications in early diagnosis, mechanistic studies, and therapeutic monitoring of MASH/MASLD-related liver injury.
In conclusion, we developed an NIR emissive probe, TTM-4, to monitor viscosity changes associated with LPO during ferroptosis. TTM-4 exhibited sensitive fluorescence responses to cellular viscosity changes. Formation of oxidized nLDs detected using TTM-4 indicated that lipid peroxides migrated from the cLDs to the nucleus through the ER, which was also confirmed in the liver tissue of mice and human samples. Owing to the enhanced NIR emission during ferroptosis, along with its low cytotoxicity, our in vivo imaging with TTM-4 revealed real-time detection of ferroptosis in live mice with chemically induced acute liver injury and in diet-induced chronic liver disease models. Taken together, our findings provide valuable insights into the progression of ferroptosis and its implications in related liver diseases and highlight the potential application of TTM-4 as a diagnostic agent for the clinical detection of MASH/MASLD in patients.
FOOTNOTES
-
Authors’ contributions
Conceptualization: Le Bich Hang Pham, Taeeung Kim, Keon Wook Kang, Jeeyeon Lee. Data curation: Le Bich Hang Pham, Taeeung Kim, Yun Seok Kim, Wan Seob Shim. Formal analysis: Le Bich Hang Pham, Taeeung Kim, Seoyoung Kim, Yun Seok Kim, Jiyeon Kim, Kyeongseon Kim, Hyeonwoo Lim, So-Yeol Yoo, Jae-Young Lee, Murim Choi, Won Kim. Funding acquisition: Keon Wook Kang, Jeeyeon Lee. Investigation: Le Bich Hang Pham, Taeeung Kim, Seoyoung Kim, Yun Seok Kim, Jiyeon Kim, Kyeongseon Kim, Hyeonwoo Lim, So-Yeol Yoo, Jae-Young Lee, Murim Choi, Won Kim. Methodology: Le Bich Hang Pham, Taeeung Kim, Seoyoung Kim, Yun Seok Kim, Wan Seob Shim, Byoungmo Kim. Supervision: Keon Wook Kang, Jeeyeon Lee. Writing-original draft: Le Bich Hang Pham, Taeeung Kim. Writing-review & editing: Keon Wook Kang, Jeeyeon Lee.
-
Acknowledgements
This study was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (RS-2021-NR059390 to J. Lee), and the Bio&Medical Technology Development Program of the NRF funded by the Korea government (MSIP) (RS-2022NR067269 to KW. Kang).
-
Conflicts of Interest
J. Lee, KW. Kang, L. B. H. Pham, T. Kim, K. Kim, J. Kim, and S. Kim are inventors on a pending patent application related to the technology described in this work, filed by the SNU R&DB Foundation.
SUPPLEMENTARY MATERIAL
Supplementary material is available at Clinical and Molecular Hepatology website (
http://www.e-cmh.org).
Supplementary Figure 2.
(A) Absorption spectra of 10 μM TTM-4 in various solvents. (B) Fluorescent spectra of 2 μM TTM-4 in various solvents. λex=576 nm. (C) Fluorescent spectra of 10 μM TTM-4 in glycerol-water mixed solvents. λex=576 nm. (D) Linear relationship of TTM-4 between log(I812.5) and log η. (E) Absorption spectra of 10 μM TTO-4 in various solvents. (F) Fluorescent spectra of 2 μM TTO-4 in various solvents. λex=504 nm. (G) Fluorescent spectra of 10 μM TTO-4 in glycerol-water mixed solvents. λex=504 nm. (H) Linear relationship of TTO-4 between log(I659.5) and log η.
Supplementary Figure 3.
(A) Absorption spectra of 10 μM TTM-6 in various solvents. (B) Fluorescent spectra of 2 μM TTM-6 in various solvents. λex=576 nm. (C) Fluorescent spectra and (D) the quantification of 10 μM TTM-6 in glycerol-water mixed solvents. λex=664 nm. (E) Absorption spectra of 10 μM TTO-6 in various solvents. (F) Fluorescent spectra of 2 μM TTO-6 in various solvents. λex=504 nm. (G) Fluorescent spectra and (H) the quantification of 10 μM TTO-6 in glycerol-water mixed solvents. λex=561 nm.
Supplementary Figure 4.
Fluorescent spectra and the quantification of 10 μM TTM-4 in (A) glycerol-methanol and (B) water-methanol mixed solvents. λex=576 nm.
Supplementary Figure 7.
TTM-4 fluorescence in HepG2 cells measured at different temperatures. Live cell fluorescence imaging of 1 μM TTM-4 (λex=561 nm, λem=635–700 nm) in HepG2 cells incubated at 37 C, 20 C, and 4 C for 1 hour. The scale bar represents 50 µm.
Supplementary Figure 8.
Cytotoxicity of TTM-4 in various cell lines. Effect of TTM-4 on cell viability in (A) HeLa, (B) HepG2, (C) AML12 cells, and (D) mouse primary hepatocytes (mPH). Cells were treated with TTM-4 at various concentrations for 24 hours.
Supplementary Figure 9.
Fluorescence of TTM-4 in pyroptosis or ferroptosis-induced primary mouse bone marrow-derived macrophages (BMDMs). (A–D) BMDM cells were isolated from mice and subjected to either LPS (100 ng/mL, 4 h) followed by ATP (1 mM, 1 h) to induce pyroptosis, or IL4 (20 ng/mL, 24 h) followed by RSL3 (10 µM, 5 h) to induce ferroptosis. (A) Representative Incucyte fluorescence images after treatment with TTM-4 (1 μM) under each condition. (B) Quantification of red fluorescence intensity among treatment groups. (C) Measurement of IL-1β levels using ELISA. (D) Western blot analysis of pyroptosis markers (NLRP3, CASPASE1_p20) in LPS/ATP-treated BMDMs. Data are presented as the mean±SD, ****P<0.0001, as determined using one-way ANOVA.
Supplementary Figure 10.
Colocalization of TTM-4 and different organelle-targeted probes in HepG2 cells treated with erastin. (A–C) Detection of lipid peroxidation in organelles of HepG2 cells treated with 10 μM erastin for 6 hours to induce ferroptosis, using 1 μM TTM-4 (λex=561 nm; λem=635–700 nm). Subcellular organelles were detected by costaining with (A) BioTracker 405 Blue Mitochondria Dye, (B) LysoTracker blue (λex=405 nm; λem=410–490 nm), or (C) BODIPY FL C5-Ceramide (λex=488 nm; λem=493–550 nm). Results of colocalization analysis are displayed as histograms. Scale bar represents 10 µm.
Supplementary Figure 11.
Nuclear LDs detected by TTM-4 in HepG2 cells treated with 10 µM of erastin. Nuclear LDs were induced by 10 µM of erastin and revealed by 1 μM TTM-4 (λex=561 nm, λem=635–700 nm). Subcellular organelle was determined by co-staining with
(A) BODIPY 493/503 (λex=488 nm, λem=493–550 nm) or (B) Hoechst (λex=405 nm, λem=410–490 nm). The scale bar is 5 µm. White arrows and yellow arrowheads indicated nuclear LDs and nucleoli, respectively.
Supplementary Figure 12.
Sequential emergence of lipid peroxidized organelles during ferroptosis in HepG2 cells. Timelapse imaging of HepG2 cells induced by (A) 10 μM erastin for 4 hours or (B) 10 mM H2O2 for 15 minutes, and revealed by 1 μM TTM-4 (λex=561 nm, λem=635–700 nm). White arrows indicate nuclear LD. The scale bars are 5 µm and 10 µm.
Supplementary Figure 13.
Visualization of changes in the number of oxidized nuclear LDs in ferroptotic HepG2 cells by TTM-4. (A) Scheme of sequence of treatment with erastin and Fer-1 or DFO to HepG2 cells. (B) Lipid peroxidation in LDs induced by 10 μM of erastin for 4 hours in HepG2 cells treated with or without 5 μM of Fer-1 or 50 µM of DFO and visualized by 1 μM TTM-4. The scale bar represents 10 µm. (C) The changes in the number of oxidized LDs and nuclear LDs versus total LDs. The data are presented as the mean±SD, n>7 regions, >25 cells per region. ***P<0.001, and *P<0.05 as determined by ANOVA one way with post-hoc Tukey’s test.
Supplementary Figure 14.
RNA expression level of CIDEA in human samples and AML12 cells. (A) Expression level of CIDEA at various stages of metabolic dysfunction-associated steatotic liver disease in samples from patients. (B) Quantitative real-time PCR for Cidea mRNAs in AML12 cells.
Supplementary Figure 15.
Increase in ferroptosis-related markers in the CCl4 single-dose injection model and the 3-week MCD diet
model. (A, B) C57BL6/J mice were administered 20% CCl4 intraperitoneally for one day, and corn oil was used as a vehicle (n=3). (A) Western blot analysis for 4HNE, GPX4, and GAPDH. (B) Quantitative real-time PCR of the mouse liver tissue mRNAs for Acsl4, Gpx4, Tfrc, and Hmox1 which are known ferroptosis markers. (C, D) C57BL6/J mice were fed MCD diet for 3 weeks, and normal chow diet was used as a control (n=3). (C) Western blot analysis for 4HNE, GPX4, and GAPDH. (D) Quantitative real-time PCR of the mouse liver tissue mRNAs for Acsl4, Gpx4, Tfrc, and Hmox1 which are known ferroptosis markers. Data are presented as mean±SD; *P<0.05, ***P<0.001. Statistical significance of the differences was determined by two-tailed Student’s t-test.
Supplementary Figure 16.
(A) Cidec mRNA levels after 36 hours knockdown with control or Cidec siRNA. (B) Left: representative IncuCyte images. After Cidec knockdown, AML12 cells were treated with or without 1 mM palmitate for 24 hours, and then fluorescence was monitored with 1 μM TTM-4 (red); Right: Quantification of TTM-4 fluorescence. (C) mRNA levels of ferroptosis markers (Trp53, Hmox1, and Gpx4). After Cidec knockdown, AML12 cells were treated with or without 500 μM palmitate for 6 hours, and then total RNA samples were subjected to quantitative real-time PCR analyses. (D) Western blot analyses of ferroptosis markers (4-HNE and GPX4). After Cidec knockdown, AML12 cells were treated with or without 1 mM palmitate for 9 hours, and then total cell lysates were subjected to immunoblotting. Data are presented as the mean±SD, *P<0.05, **P<0.01, ***P<0.001, and ****P<0.0001, as determined using one-way ANOVA.
Figure 1.Near-infrared (NIR) emission from TTM-4 based on changes in viscosity. (A) Structures of TTM-4, TTO-4, TTM-6, and TTO-6. (B) Turn-on fluorescence based on change in viscosity. (C) Fluorescence enhancement and (D) quantification of 10 μM TTM-4 in glycerol-water mixed solvents with λex =576 nm. (E) Fluorescent intensity of TTM-4 (10 μM) in the presence of various analytes (200 μM). (F) Visualization of lipid peroxidation by TTM-4 using in vitro and in vivo MASLD models.
Figure 2.Sensitivity and selectivity of intracellular lipid peroxidation detection using TTM-4. Live-cell fluorescence imaging of lipid peroxides in HepG2 cells treated with (A1) 10 mM H2 O2 or (B1) 10 μM erastin and 5 μM RSL3 in the presence or absence of ferrostatin-1 (Fer-1), liproxstatin-1 (Lip-1), β-mercaptoethanol (β-ME), and N-acetyl-L-cysteine (NAC) using 1 μM TTM-4. (A2) and (B2) The intensity of fluorescence emitted from TTM-4 was shown in the graph with statistical analysis (n=3). ***P<0.001, as determined using one-way ANOVA with post-hoc Tukey’s test. (C) Detection of intracellular lipid peroxidation in HepG2 cells with ferroptosis induced by 5 μM RSL3 using various concentrations of TTM-4 (λex =561 nm; λem =635–700 nm). Control images were obtained using 5 μM BODIPY 581/591 C11 probe (λex =488 nm and λem =493–550 nm for oxidized BODIPY C11, and λex =561 nm and λem =570–620 nm for reduced BODIPY C11). Scale bar represents 10 µm. (D1) Live-cell fluorescence images of HepG2 cells visualized using 1 μM TTM-4 after induction of different cell death pathways using 10 μM erastin, 5 μM RSL3, 10 μg/mL LPS, or 10 μM staurosporine for different incubation periods. The scale bar represents 50 µm. (D2) Quantification of fluorescence from TTM-4 under different conditions. Data are presented as the mean±standard deviation, n=4 regions. **P<0.01 and ***P<0.001, as determined using two-way ANOVA.
Figure 3.Specific detection of lipid peroxidation in organelles using TTM-4. (A–C) Detection of lipid peroxidation in organelles of HepG2 cells, treated with 5 μM RSL3 for 24 hours to induce ferroptosis, using 1 μM TTM-4 (λex =561 nm; λem=635–700 nm). Subcellular organelles were detected by costaining with (A) BODIPY 493/503 (λex=488 nm; λem=493–550 nm) or (B) ER Tracker Blue-White (λex=405 nm; λem=410–490 nm). Results of colocalization analysis are displayed as histograms. Scale bar represents 10 µm. (C) Nuclear lipid droplets are indicated by white arrows. (D, E) Assessment of the level of lipid peroxidation in specific organelles during 10 μM erastin-induced ferroptosis in HepG2 cells. White arrowhead indicates the ER. Scale bars represent 2 or 0.5 µm.
Figure 4.Visualization of lipid peroxidation in an in vitro metabolic dysfunction-associated steatotic hepatitis (MASH) model using T-TM4. (A1) Detection of nuclear lipid droplets (LDs) induced by treatment of cells with 10 µM of erastin for 6 hours using 1 μM TTM-4 (λex=561 nm; λem=635–700 nm). HepG2 cells were pretreated with 50 μM lalistat-2 for 24 hours before incubation with erastin. Subcellular organelles were detected by costaining with BODIPY 493/503 (λex=488 nm; λem=493–550 nm) and Hoechst (λex=405 nm; λem=410–500 nm). (B1) Visualization of lipid peroxidation in LDs induced by treatment of HepG2 cells with 500 µM palmitic acid for 24 hours, with or without 30 µM eicosapentaenoic acid (EPA) using 1 μM TTM-4. Scale bars represent 10 µm. (A2) and (B2) Changes in the percentage of oxidized LDs versus total LDs and oxidized nLDs versus total nLDs in (A1) and (B1) conditions, respectively. Data are presented as the mean±standard deviation, n>4 regions, >100 cells per region. ****P<0.0001, ***P<0.001, and *P<0.05, as determined using Student’s t-test, ns, not significant. (C) 1) Representative IncuCyte images of AML12 cells treated with 5 μM RSL3 for 24 hours, with or without 100 μM deferoxamine (DFO) using 1 μM TTM-4; 2) Cellular fluorescence intensity of TTM-4 was represented in the graph. (D) TTM-4-positive cell population data obtained using FACS analysis. (E, F) FACS analysis was performed after treating AML12 cells with (E) 5 µM RSL3 or (F) palmitate (250 or 1,000 µM) for 24 hours, followed by incubation with 1 µM BODIPY 581/591 C11 or 1 µM TTM-4 for 1 hour.
Figure 5.TTM-4 can detect ferroptosis in the liver tissue of mice and humans with MASH. (A) Fluorescence signals for DAPI (blue), α-SMA (green), and TTM-4 (red) in liver biopsy samples from patients with MASLD were analyzed. The merged data represent the combined fluorescence of αSMA and TTM-4. (B) Representative images of liver tissue of 12-week CDAHFD-fed C57BL/6J mice stained with BODIPY 493/503 (green), TTM-4 (red), and DAPI (gray). The scale bar is 0.5 mm. (C) Comparison of BODIPY 581/591 C11 and TTM-4 fluorescence in liver tissue samples from mice at weeks 1, 6, and 12 of CDAHFD feeding using confocal microscopy with statistical analysis (n=3). The scale bar is 50 µm. (D) Confocal microscopy of nLDs in liver tissue samples at weeks 1, 6, and 12 of CDAHFD using T-TM4. MASH, metabolic dysfunction-associated steatotic hepatitis; MASLD, metabolic dysfunction-associated steatotic liver disease. The scale bar is 10 µm. Data are presented as mean±standard deviation, n=5 regions, >100 cells per region. **P<0.01 and ***P<0.001, as determined using two-tailed Student’s t-test.
Figure 6.Detection of lipid toxicity induced by CIDEC-mediated fusion of lipid droplets using TTM-4. (A1, A2) Expression levels of CIDEA, CIDEB, and CIDEC at various stages of metabolic dysfunction-associated steatotic liver disease in samples from patients. (A3–A5) Analysis of public data (GSE135251) for con-early-moderate NAFLD related genes. (A3) Heatmap (false-discovery rate 0.05, fold-change 1.5). (A4) Volcano plot for CIDEA and CIDEC. (A5) Violin plot for CIDEC mRNA level. (B) Quantitative real-time PCR for Cidec mRNA 1) in the mouse liver tissue in choline-deficient, L-amino acid-defined, high-fat diet (CDAHFD) 1, 6, and 12-week mouse model; 2) in mouse primary hepatocytes following palmitic acid (PA; 100 and 300 μM) treatment; 3) in AML12 cells following PA treatment at 250 and 1,000 µM. (C) Comparison of BODIPY 493/503 and TTM-4 fluorescence in PA (250 and 1,000 μM)-treated AML12 cells with statistical analysis (n=5). (D) Comparison of BODIPY 493/503 and TTM-4 fluorescence in human liver biopsy samples with statistical analysis (n=3). The scale bar is 50 µm. (E) Detection of nLDs using confocal microscopy (n=4 regions, >70 cells per region). eMASH, early MASH; aMASH, advanced MASH. Data are presented as mean±standard deviation, *P<0.05, **P<0.01, ***P<0.001, and ****P<0.0001. Statistical significance of the differences was determined using one-way ANOVA followed by Tukey’s test. (F) Detection of lipid toxicity induced by CIDEC and TTM-4 in human MASH sample. Colocalization analysis is displayed as a histogram. Scale bar represents 10 µm.
Figure 7.TTM-4 enables the measurement of ferroptosis in live mice. (A1, A2) Ferroptosis was induced in C57BL/6J mice with a single intraperitoneal injection of CCl4 or by feeding a methionine- and choline-deficient (MCD) diet for 3 weeks. Fluorescence expression in the liver tissue was observed using TTM-4. (B) ICR mice were administered a single dose of TTM-4 (5 mg/kg) via the tail vein injection. Toxicity evaluation was performed 2 hours post-injection. Liver toxicity was assessed by measuring serum levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyl transferase (GGT), whereas kidney function was evaluated by comparing the levels of blood urea nitrogen (BUN) and creatinine (CRE). Results are presented as comparative graphs. (C1–F2) Ferroptosis was induced in ICR mice through two distinct experimental approaches. In one experiment, mice received a single intraperitoneal injection of 20% CCl4 at a dose of 100 µl/20 g body weight during a 24-hour fasting period (n=4). In another experiment, mice were fed an MCD diet for 3 weeks. On the final day, following a 24-hour fasting period (n=5), TTM-4 (5 mg/kg) was administered via the tail vein injection, and fluorescence expression was evaluated 2 hours later using the IVIS. (C1) Illustration of animal experiments. (C2, C3) Photographs of the mouse liver. (D1, D2) Serum AST and ALT levels. (E) Measurement of TTM-4 fluorescence expression in each organ 24 hours after oral administration of CCl4 at the same concentration. (C1–F2) Fluorescence images were captured using the IVIS. (F1, F2) Fluorescence intensity was quantified using the ImageJ software based on IVIS fluorescence images, and the results are presented as comparative graphs. Data are presented as mean±standard deviation. ns, not significant; **P<0.01, ***P<0.001, ****P<0.0001. Statistical significance of the differences was determined using the two-tailed Student’s t-test.
Abbreviations
metabolic dysfunction-associated steatotic hepatitis
metabolic dysfunction-associated steatotic liver disease
methionine- and choline-deficient
REFERENCES
- 1. Dixon SJ, Stockwell BR. The role of iron and reactive oxygen species in cell death. Nat Chem Biol 2014;10:9-17.
- 2. Dixon SJ, Lemberg KM, Lamprecht MR, Skouta R, Zaitsev EM, Gleason CE, et al. Ferroptosis: an iron-dependent form of nonapoptotic cell death. Cell 2012;149:1060-1072.
- 3. Bersuker K, Hendricks JM, Li Z, Magtanong L, Ford B, Tang PH, et al. The CoQ oxidoreductase FSP1 acts parallel to GPX4 to inhibit ferroptosis. Nature 2019;575:688-692.
- 4. Li Y, Feng D, Wang Z, Zhao Y, Sun R, Tian D, et al. Ischemia-induced ACSL4 activation contributes to ferroptosis-mediated tissue injury in intestinal ischemia/reperfusion. Cell Death Differ 2019;26:2284-2299.
- 5. Ou M, Jiang Y, Ji Y, Zhou Q, Du Z, Zhu H, et al. Role and mechanism of ferroptosis in neurological diseases. Mol Metab 2022;61:101502.
- 6. Wang K, Chen XZ, Wang YH, Cheng XL, Zhao Y, Zhou LY, et al. Emerging roles of ferroptosis in cardiovascular diseases. Cell Death Discov 2022;8:394.
- 7. Miao R, Fang X, Zhang Y, Wei J, Zhang Y, Tian J. Iron metabolism and ferroptosis in type 2 diabetes mellitus and complications: mechanisms and therapeutic opportunities. Cell Death Dis 2023;14:186.
- 8. Chen J, Li X, Ge C, Min J, Wang F. The multifaceted role of ferroptosis in liver disease. Cell Death Differ 2022;29:467-480.
- 9. Zhou Q, Meng Y, Li D, Yao L, Le J, Liu Y, et al. Ferroptosis in cancer: from molecular mechanisms to therapeutic strategies. Signal Transduct Target Ther 2024;9:55.
- 10. He F, Zhang P, Liu J, Wang R, Kaufman RJ, Yaden BC, et al. ATF4 suppresses hepatocarcinogenesis by inducing SLC7A11 (xCT) to block stress-related ferroptosis. J Hepatol 2023;79:362-377.
- 11. Peleman C, Hellemans S, Veeckmans G, Arras W, Zheng H, Koeken I, et al. Ferroptosis is a targetable detrimental factor in metabolic dysfunction-associated steatotic liver disease. Cell Death Differ 2024;31:1113-1126.
- 12. Yu Q, Song L. Unveiling the role of ferroptosis in the progression from NAFLD to NASH: recent advances in mechanistic understanding. Front Endocrinol (Lausanne) 2024;15:1431652.
- 13. Yin J, Huang L, Wu L, Li J, James TD, Lin W. Small molecule based fluorescent chemosensors for imaging the microenvironment within specific cellular regions. Chem Soc Rev 2021;50:12098-12150.
- 14. Wang S, Ren WX, Hou JT, Won M, An J, Chen X, et al. Fluorescence imaging of pathophysiological microenvironments. Chem Soc Rev 2021;50:8887-8902.
- 15. Hu W, Chai L, Chen X, Chen J, Ren H, Li C, et al. Unveiling dynamic alterations of lipid droplet polarity during NAFLD-triggered lipophagy utilizing a far-red fluorescent probe with large stokes shift. Sens Actuator B: Chem 2025;422:136509.
- 16. Hu W, He Y, Ren H, Chai L, Li H, Chen J, et al. Near-infrared imaging for visualizing the synergistic relationship between autophagy and NFS1 protein during multidrug resistance using an ICT-TICT integrated platform. Chem Sci 2024;15:6028-6035.
- 17. Xu SL, Guo FF, Xu ZH, Wang Y, James TD. A hemicyanine-based fluorescent probe for ratiometric detection of ClO- and turn-on detection of viscosity and its imaging application in mitochondria of living cells and zebrafish. Sens Actuator B: Chem 2023;383:133510.
- 18. Wu WN, Chen X, Liu SS, Bie HY, Fan YC, Wang Y, et al. A dual-response fluorescent hemicyanine probe for the detection of mitochondrial hypochlorite and viscosity based on ESIPT/AIE and TICT. Sens Actuator B: Chem 2025;423:136695.
- 19. Lai HJ, Wu H, Wang Y, Bie HY, Fan YC, Wu WN, et al. A Golgitargeted AIE-active probe for monitoring peroxynitrite and viscosity during drug-induced liver injury. Sens Actuator B: Chem 2026;447:138723.
- 20. Hu W, Chai L, Chen X, Zhang H, Chen J, James TD, et al. Solvent-tunable rotational barrier-based polarity-responsive probe: From molecular design to cellular autophagy applications. Sens Actuator B: Chem 2026;447:138837.
- 21. Zeng F, Nijiati S, Tang L, Ye J, Zhou Z, Chen X. Ferroptosis detection: from approaches to applications. Angew Chem Int Ed Engl 2023;62:e202300379.
- 22. Wang B, Ren L, Li M, Zhang X, Hu W, Cheng F, et al. Advances in ferroptosis: Synergistic therapies and fluorescent probes as arrows to target cancer’s achilles heel. Coord Chem Rev 2025;541:216779.
- 23. Wang S, Zhao X, Liu M, Yang L, Yu M, Li Z. A dual-responsive crimson fluorescent probe for real-time diagnosis of alcoholic acute liver injury. Biosens Bioelectron 2023;239:115596.
- 24. Zheng Z, Yang Y, Wang P, Gou X, Gong J, Wu X, et al. A bright two-photon lipid droplets probe with viscosity-enhanced solvatochromic emission for visualizing lipid metabolic disorders in deep tissues. Adv Funct Mater 2023;33:2303627.
- 25. Su D, Teoh CL, Wang L, Liu X, Chang YT. Motion-induced change in emission (MICE) for developing fluorescent probes. Chem Soc Rev 2017;46:4833-4844.
- 26. Liang T, Qiang T, Ren L, Cheng F, Wang B, Li M, et al. Near-infrared fluorescent probe for hydrogen sulfide: high-fidelity ferroptosis evaluation in vivo during stroke. Chem Sci 2022;13:2992-3001.
- 27. Wei H, Tang X, Chen Q, Yue T, Dong B. An endoplasmic reticulum-targeting fluorescent probe for the visualization of the viscosity fluctuations during ferroptosis in live cells. Anal Chim Acta 2022;1232:340454.
- 28. Song W, Zhang W, Yue L, Lin W. Revealing the effects of endoplasmic reticulum stress on ferroptosis by two-channel real-time imaging of pH and viscosity. Anal Chem 2022;94:6557-6565.
- 29. Zeng S, Wang Y, Chen C, Kim H, Liu X, Jiang M, et al. An ER-targeted, Viscosity-sensitive hemicyanine dye for the diagnosis of nonalcoholic fatty liver and photodynamic cancer therapy by activating pyroptosis pathway. Angew Chem Int Ed Engl 2024;63:e202316487.
- 30. Yin J, Xu Q, Mo X, Dai L, Ren M, Wang S, et al. Construction of a novel mitochondria-targeted near-infrared (NIR) probe for detection of viscosity changes in cancer cells ferroptosis process. Dyes and Pigments 2022;200:110184.
- 31. Wei H, Yu Y, Wu G, Wang Y, Duan S, Han J, et al. Dual-responsive fluorescent probe for imaging NAD(P)H and mitochondrial viscosity and its application in cancer cell ferroptosis. Sens Actuator B: Chem 2022;350:130862.
- 32. Zheng F, Ding J, Huang S, Bi A, Liu S, Zhang K, et al. Real-time monitoring viscosity during ferroptosis with a novel mitochondria-specific fluorescent probe. Dyes and Pigments 2023;217:111424.
- 33. Hu W, Wang C, Chen X, Zhang Y, Jin T, Ye Y. Mitochondrial viscosity detection using a large stokes shift near-infrared fluorescent probe: an innovative tool for precise monitoring of oxidative stress in stroke and early diagnosis. Bioorg Chem 2025;163:108710.
- 34. Dong B, Song W, Lu Y, Sun Y, Lin W. Revealing the viscosity changes in lipid droplets during ferroptosis by the real-time and in situ near-infrared imaging. ACS Sens 2021;6:22-26.
- 35. Li J, Tian M, Yu F, Zhang J, Zheng G, Yan M. Rational design of a symmetric double rotor-based fluorescent probe for revealing LDs viscosity changes during ferroptosis. Dyes and Pigments 2022;207:110779.
- 36. Wei D, Dai Y, Cao J, Fu N. A novel fluorescent probe for visualizing viscosity changes in lipid droplets during chemotherapy-induced ferroptosis. Anal Chim Acta 2024;1299:342422.
- 37. Zheng Y, Ding Y, Zheng X, Zhang C, Zhang Y, Xiang Y, et al. Long-term dynamic imaging of cellular processes using an AIE lipid order probe in the dual-color mode. Anal Chem 2021;93:10272-10281.
- 38. Sołtysik K, Ohsaki Y, Tatematsu T, Cheng J, Fujimoto T. Nuclear lipid droplets derive from a lipoprotein precursor and regulate phosphatidylcholine synthesis. Nat Commun 2019;10:473.
- 39. Llovet JM, Willoughby CE, Singal AG, Greten TF, Heikenwälder M, El-Serag HB, et al. Nonalcoholic steatohepatitis-related hepatocellular carcinoma: pathogenesis and treatment. Nat Rev Gastroenterol Hepatol 2023;20:487-503.
- 40. Machado MV, Diehl AM. Pathogenesis of nonalcoholic steatohepatitis. Gastroenterology 2016;150:1769-1777.
- 41. Marra F, Svegliati-Baroni G. Lipotoxicity and the gut-liver axis in NASH pathogenesis. J Hepatol 2018;68:280-295.
- 42. Imai N, Ohsaki Y, Cheng J, Zhang J, Mizuno F, Tanaka T, et al. Distinct features of two lipid droplets types in cell nuclei from patients with liver diseases. Sci Rep 2023;13:6851.
- 43. Lee SW, Rho JH, Lee SY, Chung WT, Oh YJ, Kim JH, et al. Dietary fat-associated osteoarthritic chondrocytes gain resistance to lipotoxicity through PKCK2/STAMP2/FSP27. Bone Res 2018;6:20.
- 44. Wang W, Chai L, Chen X, Li Z, Feng L, Hu W, et al. Imaging changes in the polarity of lipid droplets during NAFLD-Induced ferroptosis via a red-emitting fluorescent probe with a large Stokes shift. Biosens Bioelectron 2023;231:115289.
- 45. Wang KN, Liu LY, Mao D, Xu S, Tan CP, Cao Q, et al. A polarity-sensitive ratiometric fluorescence probe for monitoring changes in lipid droplets and nucleus during ferroptosis. Angew Chem Int Ed Engl 2021;60:15095-15100.
- 46. Puri V, Konda S, Ranjit S, Aouadi M, Chawla A, Chouinard M, et al. Fat-specific protein 27, a novel lipid droplet protein that enhances triglyceride storage. J Biol Chem 2007;282:34213-34218.
- 47. Lyu X, Wang J, Wang J, Yin YS, Zhu Y, Li LL, et al. A gel-like condensation of Cidec generates lipid-permeable plates for lipid droplet fusion. Dev Cell 2021;56:2592-2606.e2597.
- 48. Xu L, Li L, Wu L, Li P, Chen FJ. CIDE proteins and their regulatory mechanisms in lipid droplet fusion and growth. FEBS Lett 2024;598:1154-1169.
- 49. Li H, Chen A, Shu L, Yu X, Gan L, Zhou L, et al. Translocation of CIDEC in hepatocytes depends on fatty acids. Genes Cells 2014;19:793-802.
- 50. Zhou L, Yu M, Arshad M, Wang W, Lu Y, Gong J, et al. Coordination among lipid droplets, peroxisomes, and mitochondria regulates energy expenditure through the CIDE-ATGL-PPARα pathway in adipocytes. Diabetes 2018;67:1935-1948.
- 51. Langhi C, Arias N, Rajamoorthi A, Basta J, Lee RG, Baldán Á. Therapeutic silencing of fat-specific protein 27 improves glycemic control in mouse models of obesity and insulin resistance. J Lipid Res 2017;58:81-91.
- 52. Sans A, Bonnafous S, Rousseau D, Patouraux S, Canivet CM, Leclere PS, et al. The differential expression of CIDE family members is associated with NAFLD progression from steatosis to steatohepatitis. Sci Rep 2019;9:7501.
, Jeeyeon Lee1
