Detection of ferrous ions in extracellular vesicles at the single-particle level by nano-flow cytometry

Abstract

Iron, particularly redox-active ferrous ions (Fe²⁺), is essential for biological processes. Despite their pivotal roles, analysis of Fe²⁺ ions within individual extracellular vesicles (EVs) has been hindered by the ultralow Fe²⁺ content and substantial heterogeneity of EVs. To address this, we developed a novel approach by integrating an Fe²⁺-specific fluorescent chemosensor (Ac-FluNox) with nano-flow cytometry (nFCM) for precise single-EV Fe²⁺ mapping. Method specificity to Fe²⁺ was validated via Fe²⁺-loaded liposomal models at the single-particle level. Comprehensive profiling of Fe²⁺ distributions in HT-1080-derived EVs under varying ferroptotic stress conditions revealed the striking heterogeneity in Fe²⁺ loading among EVs and a strong positive correlation between EV Fe²⁺ levels and their parental cells. Notably, we identified an EV-mediated Fe²⁺ export mechanism that functionally parallels to ferroportin (FPN)-dependent iron efflux, suggesting EVs may serve as a compensatory iron-release pathway during FPN inhibition. The nFCM platform achieved superior detection sensitivity with high throughput (up to 10⁴ particles per min), providing a powerful analytical tool for investigating EV heterogeneity and Fe²⁺-mediated regulatory networks in iron homeostasis and ferroptosis-related pathologies.

---

Introduction

Iron, the most abundant transition metal in human physiology, serves as a pivotal redox cofactor in oxygen-dependent metabolic pathways, spanning mitochondrial respiration, hemoglobin-mediated oxygen transport, and ATP synthesis, mediated by its unique electron-shutting capacity via Fe²⁺/Fe³⁺ interconversion.^1–3 Cellular ion homeostasis is precisely controlled by iron-regulatory proteins that maintain the dynamic equilibrium between Fe²⁺ and Fe³⁺.^4–6 Due to the high cellular abundance of reductants such as glutathione, intracellular labile iron consists of Fe²⁺ primarily, which participates in regulating iron metabolism through binding to small molecules or proteins.^7–9 Therefore, elucidating Fe²⁺-mediated regulatory mechanisms provides critical insights into metabolic homeostasis and iron-regulatory networks. Although cells have evolved sophisticated systems for the uptake, transport, utilization, storage, and export of Fe²⁺,^10–12 the intricacies of intercellular Fe²⁺ regulation remain poorly understood.^13–15 Extracellular vesicles (EVs), nanoscale lipid-bilayer-delimited particles that mediate intercellular communication through selective cargo transfer,^16–18 have emerged as key players in iron-related bioactive substance trafficking.^2,19,20 The biogenesis of Fe²⁺-containing EVs occurs through coordinated interactions between multiple organelles (including endosomes, lysosomes, and mitochondria) and cytosolic Fe²⁺ pools, enabling active loading of redox-active iron during vesicle maturation. Through this mechanism, EVs critically regulate parental cell iron homeostasis and ferroptosis propagation.^13,14,21 Consequently, quantification of intravesicular Fe²⁺ is indispensable for deciphering EV-mediated biological functions and their underlying regulatory networks.

Iron exhibits diverse chemical speciation and functional states within EVs, encompassing free Fe²⁺, Fe³⁺, and ferritin-bound iron.² While inductively coupled plasma mass spectrometry (ICP-MS) has enabled quantification of total iron content in EVs, this approach lacks specificity for discriminating between distinct iron species.²² Current methodologies, including advanced fluorescent chemosensors and commercial assay kits, require EV disruption for Fe²⁺ detection,^23,24 thereby providing only ensemble-averaged measurements that obscure the inherent heterogeneity of EVs. Single-EV characterization technologies, such as super-resolution fluorescence microscopy, droplet digital PCR (ddPCR), and nano-flow cytometry (nFCM), have unequivocally demonstrated the marked heterogeneity of EVs in terms of size, composition, and cargo loading.^25–29 Consequently, single-particle-resolution analysis of Fe²⁺ within EVs represents a fundamental prerequisite for investigating the heterogeneity and regulatory mechanisms of EVs in iron metabolism. Building on these technological advances, nFCM stands out as the optimal platform for single-EV Fe²⁺ analysis, uniquely combining: (i) high-throughput detection (up to 10⁴ particles per min) essential for capturing population heterogeneity; (ii) quantitative multiparameter measurement of size, fluorescence intensity, and concentration; and (iii) sizing accuracy comparable to cryo-TEM, which collectively enabling unprecedented resolution of Fe²⁺ distribution patterns in EV subpopulations.^30,31

In this study, we developed a novel method for detecting Fe²⁺ in individual EVs by integrating the Fe²⁺-specific fluorescent chemosensor (Ac-FluNox) with nFCM (Scheme 1). Upon esterase activation within EVs, Ac-FluNox generated m̲orpholinor̲hodol (MR) through its reaction with Fe²⁺, exhibiting detectable green fluorescence. The specificity and reliability of this Fe²⁺ labeling approach were validated using Fe²⁺-loaded liposomes as model systems. Leveraging this platform, we performed systematic profiling of Fe²⁺ distributions and dynamic changes in EVs derived from HT-1080 cells under various ferroptosis-inducing conditions. Significantly, our results revealed a strong correlation of Fe²⁺ levels between EVs and their parental cells, providing mechanistic insights into EV-mediated regulation of iron homeostasis during ferroptosis.

Scheme 1 Single-EV Fe²⁺ profiling via Ac-FluNox labeling coupled with nFCM detection.

Results and discussion

Synthesis and performance evaluation of Fe²⁺ chemosensor Ac-FluNox

As illustrated in Scheme S1, the fluorescent chemosensor Ac-FluNox was synthesized according to the previously reported procedure,^32,33 with the target compound characterized by ¹H NMR, ¹³C NMR (Fig. S1 and S2). Leveraging N-oxide chemistry for fluorescence switching, the N–O bond in Ac-FluNox isolates the nitrogen atom from p-conjugation, yielding a fluorescence turn-off state. Subsequential esterase hydrolysis followed by Fe²⁺-mediated deoxygenation converts Ac-FluNox to m̲orpholinor̲hodol (MR), which exhibits strong green fluorescence (Fig. 1a). Owing to its inherent lipophilicity, Ac-FluNox readily traverses lipid bilayers via passive diffusion, achieving efficient loading into both natural EV membranes and artificial liposomes. This dual-enzyme/metal-responsive mechanism establishes Ac-FluNox as a promising probe for Fe²⁺ detection in EVs via nFCM.

Fig. 1 Fluorescence responses of Ac-FluNox for Fe²⁺ detection in HEPES buffer (50 mM, pH 7.4, 0.2% DMSO). (a) The mechanism of Fe²⁺-specific fluorescence activation of Ac-FluNox via sequential esterase-catalyzed hydrolysis (generating H-FluNox) and Fe²⁺-mediated deoxygenation to yield m̲orpholinor̲hodol (MR), a green-emitting fluorophore. (b) Fluorescence emission spectra of Ac-FluNox (1.0 μM) upon reaction with different concentrations of Fe²⁺ (0–10 μM) in the presence of esterase (0.1 mg mL⁻¹). (c) Dose–response curves of Ac-FluNox (1.0 μM) fluorescence intensity (l_em = 535 nm) with increasing Fe²⁺ (0–16 μM), with or without of esterase (0.1 mg mL⁻¹). (d) Time-dependent fluorescence enhancement (recorded at 5 min intervals) of Ac-FluNox (1.0 μM) upon addition of Fe²⁺ (10 μM) in the presence of esterase (0.1 mg mL⁻¹). (e) Metal selectivity test against various metal ions. The error bar represents the standard deviation (s.d.) of three replicate experiments (n = 3, mean ± s.d.).

The fluorescence response of Ac-FluNox was systematically evaluated for Fe²⁺ sensing in HEPES buffer containing esterase (0.1 mg mL⁻¹). Upon gradual increase of Fe²⁺ (0–10 μM), a 75.2-fold fluorescence enhancement in green fluorescence emission (λ_ex = 490 nm, λ_em = 535 nm) was observed at 10 μM Fe²⁺ within 60 min (Fig. 1b–d), demonstrating rapid and efficient Fe²⁺ detection, a feature that leverages the ubiquitous presence of esterase in EVs for precise intravesicular Fe²⁺ analysis.^34,35 Furthermore, the selectivity of Ac-FluNox for Fe²⁺ was investigated. As demonstrated in Fig. 1e, negligible fluorescence was observed in the presence of biologically relevant interferents, confirming the probe's specificity and stability for Fe²⁺ detection in complex biological environments. Collectively, these results rigorously validated that Ac-FluNox not only quantitatively detects Fe²⁺ in solutions, but also serves as a highly selective sensor for monitoring Fe²⁺ changes.

Single-liposome Fe²⁺ profiling

To validate the single-particle Fe²⁺ labeling method, we synthesized liposomes (LPs) encapsulating Fe²⁺ (0–4.0 μM) and esterase (0.1 mg mL⁻¹) via microfluidics (Fig. 2a(i) and b). During microfluidic assembly, Fe²⁺ and esterase were successfully co-encapsulated within the liposomes' aqueous lumen, leveraging their characteristic phospholipid bilayer-enclosed aqueous core architecture (Fig. 2b). The liposomes were stained with Ac-FluNox and analyzed by nFCM. Analysis of liposome size distribution following refractive index correction using Mie theory revealed that variations in encapsulated Fe²⁺ concentration had negligible effects on particle size distribution (Fig. S3 and S4). The observed heterogeneity in size and fluorescence distributions reflects the size-dependent loading variability inherent to liposome populations (Fig. 2a(ii)). Both the percentage ratio and median FL intensity of Fe²⁺-positive-liposomes (Fe²⁺-LPs) showed concentration-dependent responses to Fe²⁺ (Fig. 2c and d). The strong linear correlation (R² = 0.9812) between Fe²⁺-LPs' median FL intensity and Fe²⁺ loading concentration demonstrates Ac-FluNox's capability for quantifying Fe²⁺ in single particles via nFCM (Fig. 2d). These results establish the nFCM-based single-particle Fe²⁺ detection method as a robust platform for intravesicular Fe²⁺ analysis with direct applicability to EV studies.

Fig. 2 Analysis of Fe²⁺ in individual Fe²⁺-encapsulated liposomes using nFCM. (a) Schematic diagram of liposome formulations with graded Fe²⁺ concentrations (0–4.0 μM) (i) and representative bivariate dot-plots of Fe²⁺ fluorescence (FL) versus particle size for liposomes labeled with Ac-FluNox (ii). (b) Microfluidic device schematic for liposomes synthesis. (c) Scatter plot presenting percentage ratios of Fe²⁺-LPs. (d) Linear regression correlating FL median with intra-liposomal Fe²⁺ concentration. Initial concentration of liposome: ∼5.0 × 10¹⁰ particles per mL. The concentrations of Fe²⁺-LPs were calculated by multiplying the positive ratios with the initial liposome concentration. Error bars represent standard deviation (s.d.) of three replicate experiments (n = 3, mean ± s.d.).

Property characterization of EVs derived from ferroptotic cells

The nFCM-based single-particle Fe²⁺ detection method was applied to analyze EVs derived from HT-1080 cells, a well-established human fibrosarcoma cell line for ferroptosis research. EVs were isolated through sequential centrifugation (1000g, 5 min; 2000g, 10 min) and ultracentrifugation (100 000g, 2 h, twice) (Fig. 3a). TEM imaging confirmed typical EV morphology (Fig. 3b and S5), while Western blotting verified the presence of classical EV markers CD9 and TSG101 and the absence of the endoplasmic reticulum (ER) marker calnexin in the EV preparation (Fig. 3c). Triton X-100 treatment, developed in our previous study,^31,36 was conducted, affirming a purity (defined as the detergent sensitivity) of 88.6% for the HT-1080 EVs isolate (Fig. 3d).

Fig. 3 Characterization of EVs released during ferroptosis. (a) Workflow of EV isolation and purification from HT-1080 cells. (b) Representative TEM micrograph of EVs isolated from the conditioned cell culture medium (CCCM) of untreated HT-1080 cells. (c) Immunoblots comparing a cell lysate with an EV preparation (10 μg protein per lane). (d) Side scatter (SSC) distribution histograms of EVs before and after Triton X-100 treatment. (e) Quantitative comparison of secreted particle concentration of EVs under different treatment conditions in per mL of conditioned medium. (f) Comparison of EV size distribution under different treatment conditions. Error bars represent the standard deviation (s.d.) of three replicate experiments (n = 3, mean ± s.d.).

Three ferroptosis inducers were used to treat HT-1080 cells: ammonium ferric citrate (FAC, 0–200 μg mL⁻¹), RSL3 (0–2 μM), and erastin (0–4 μM). FAC serves as a direct iron source that elevates intracellular labile iron pool (LIP) levels through cellular uptake and subsequent reduction.^2,37 Both RSL3 and erastin are classical ferroptosis inducers that indirectly increase intracellular Fe²⁺ levels through promoting ferritin degradation, RSL3 through GPX4 inhibition and erastin via system x_c⁻ blockade.¹ All treatments (48 h incubation) maintained >80% cell viability (Fig. S6). Iron overload conditions significantly enhanced EV secretion (Fig. 3e), suggesting a stress-responsive regulatory mechanism in parental cells. Notably, all EV populations exhibited similar size distributions (Fig. 3f), confirming the preservation of EV integrity under different iron metabolic states. Taken together, the isolated EVs met all quality criteria for subsequent Fe²⁺ profiling.

Single-particle analysis of Fe²⁺ in EVs during ferroptosis via nFCM

Previous studies have established the association between Fe²⁺ fluctuations and ferroptosis, with distinct cellular Fe²⁺ levels observed across different induction pathways.¹ However, whether ferroptosis drives Fe²⁺ accumulation in EVs, and whether EVs serves as an iron export mechanism parallel to ferroportin FPN, remains unclear. Capitalizing on the single-particle resolution of nFCM for intravesicular Fe²⁺ analysis, we performed quantitative mapping of Fe²⁺ in individual EVs. Initial investigations focused on FAC-induced iron overload (Fig. 4a). Flow cytometry and confocal microscopy confirmed dose-dependent intracellular Fe²⁺ accumulation in HT-1080 cells after 48-hour FAC treatment (Fig. 4b and c and S7). To compare the correlation between parental cells and EVs in Fe²⁺ content, Ac-FluNox was used to label Fe²⁺ in EVs. nFCM analysis revealed FAC concentration-dependent increases in Fe²⁺-positive EV ratios and their median fluorescence intensity, which were attenuated by the iron chelator deferoxamine (DFO), confirming the iron-dependent nature of these responses (Fig. S8). These results, combined with the FAC-induced increase in EV secretion (Fig. 3e) demonstrate synchronized intracellular–extracellular Fe²⁺ dynamics, suggesting a regulatory link between cellular iron status and EV biogenesis.

Fig. 4 Analysis of Fe²⁺ in EVs derived from ferroptotic HT-1080 cells treated with different ferroptosis inducers. (a) Schematic diagram of the regulatory mechanisms of FAC, RSL3, and erastin on iron metabolism. (b) Fe²⁺ level in HT-1080 cells treated with different inducers by flow cytometric analysis. (c) Confocal microscope images of Ac-FluNox-stained HT-1080 cells treated with different inducers. (d) Bivariate dot-plots of FL versus particle size of EVs derived from different treatments. (e) Absolute concentrations of Fe²⁺-positive EVs calculated by multiplying the positive ratios from panel (d) with corresponding total particle concentrations. (f) Bar graphs presenting FL median of Fe²⁺ positive EVs in (d). The labels i–iv represent different treatments (i: control, ii: 200 μg mL⁻¹ FAC, iii: 2.0 μM RSL3, and iv: 2.0 μM erastin). Scale bar: 20 μm. The error bar represents the standard deviation (s.d.) of three replicate experiments (n = 3, mean ± s.d.).

We extended these observations to ferroptosis inducers that indirectly alter iron homeostasis through promoting ferritin degradation (Fig. 4a). Flow cytometry and fluorescence microscopy imaging (Fig. 4b and c) demonstrated RSL3- and erastin-induced iron overload in HT-1080 cells, mediated by ferritin degradation followed by Fe³⁺ release and reduction. EVs generated under ferroptosis induction exhibited Fe²⁺ accumulation patterns similar to those of FAC-treated EVs (Fig. 4b–f and S8), with increases in both particle concentration and median fluorescence intensity of Fe²⁺-positive populations. Together, these results demonstrate that iron-overloaded cells actively release Fe²⁺-enriched EVs, with their Fe²⁺ content closely linked to parental cell iron metabolism—supporting the involvement of an active transport process.

Exploring the relationship between FPN and EVs in Fe²⁺ export

Building on nFCM evidence of EV-mediated iron homeostasis regulation during stress (Fig. 4b–f and S8), we postulated that EVs might functionally compensate for FPN deficiency by providing an alternative Fe²⁺ efflux pathway. To test this, we employed VIT-2763, a specific FPN inhibitor, to pharmacologically block canonical iron efflux.³⁸ Consistent with FPN inhibition, quantitative analysis revealed significant intracellular Fe²⁺ accumulation in HT-1080 cells (Fig. 5a and b), while parallel detection of EV-associated Fe²⁺ confirmed compensatory extracellular export (Fig. 5c–e). Notably, under combined FPN inhibition and iron overload induced by FAC, RSL3, or erastin, elevated particle concentrations and Fe²⁺ FL intensities of Fe²⁺-EVs were observed (Fig. 5c–e, S9 and S10). These results further demonstrate that EV-mediated Fe²⁺ export increases proportionally with both pharmacological FPN inhibition and pathological iron overload. Collectively, these findings establish a dual-phase iron export system wherein iron-overloaded cells not only export Fe²⁺ via FPN but also release Fe²⁺ into the extracellular microenvironment in an EV-dependent manner. Importantly, EV-meditated efflux is further activated as an adaptive response to FPN dysfunction (Fig. 5f schematic). This spontaneous compensatory mechanism likely represents a cellular safeguard against iron toxicity when FNP-meditated export pathway is compromised.

Fig. 5 Analysis of Fe²⁺ in EVs derived from ferroptotic HT-1080 cells under FPN inhibition. (a and b) Flow cytometric analysis (a) and confocal microscope images (b) of intracellular Fe²⁺ level in HT-1080 cells treated with different conditions upon Ac-FluNox staining. (c) Bivariate dot-plots of FL versus particle size of Fe²⁺-EVs derived from different treatments. (d and e) Bar graphs of the particle concentration (d) and FL median intensity (e) of Fe²⁺-EVs in (c). Scale bar: 20 μm. (f) Schematic diagram of the dual-phase iron export systems. The labels i–iii represent different treatments (i: control, ii: 1.0 μM VIT-2763, and iii: 1.0 μM VIT-2763 + 200 μg mL⁻¹ FAC). The error bar represents the standard deviation (s.d.) of three replicate experiments (n = 3, mean ± s.d.).

Conclusions

In conclusion, this study elucidates the regulatory role of EVs in iron homeostasis, with specific focus on Fe²⁺. By developing a novel analytical platform combining Fe²⁺-specific fluorescent chemosensor (Ac-FluNox) with nFCM, we achieved single-EV resolution for quantitative Fe²⁺ characterization. This approach enabled the precise characterization of Fe²⁺ distribution and content in EVs, revealing a strong correlation between Fe²⁺ levels in EVs and their parental cells. Moreover, we discovered the capacity of EVs to function as an auxiliary Fe²⁺ export system complementing FPN-mediated efflux. This study not only advances the understanding of iron regulatory networks by demonstrating EV-mediated Fe²⁺ transport but also establishes a versatile platform for investigating the heterogeneity and functional significance of EVs in iron metabolism and associated pathological conditions. While the current methodology specifically detects free Fe²⁺, future studies incorporating nFCM's multiparametric capabilities could extend to iron-containing cargos (e.g., ferritin), offering more comprehensive insights into EV-mediated iron communication. This direction would further unravel the multifaceted roles of EVs in iron regulation under physiological and pathological conditions.

Conflicts of interest

The authors declare the following competing financial interest(s): X. Y. declares competing financial interest as a cofounder of NanoFCM Inc., a company committed to commercializing the nano-flow cytometry (nFCM) technology.

Data availability

Supplementary information is available: Including experimental details and supplementary results. See DOI: https://doi.org/10.1039/D5SD00060B.

The data supporting this article have been included as part of the SI.

Acknowledgements

We thank the National Key R&D Program of China (2021YFA0909400 and 2024YFA1108700) the National Natural Science Foundation of China (32450337 and 21934004) for the support.

References

1. S. J. Dixon and J. A. Olzmann, Nat. Rev. Mol. Cell Biol., 2024, 25, 424–442.
2. B. Galy, M. Conrad and M. Muckenthaler, Nat. Rev. Mol. Cell Biol., 2024, 25, 133–155.
3. K. B. Muchowska, S. J. Varma and J. Moran, Nature, 2019, 569, 104–107.
4. M. W. Hentze, M. U. Muckenthaler, B. Galy and C. Camaschella, Cell, 2010, 142, 24–38.
5. P. Matak, A. Matak, S. Moustafa, D. K. Aryal, E. J. Benner, W. Wetsel and N. C. Andrews, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 3428–3435.
6. F. X. Sun, Z. Z. Zhao, M. M. Willoughby, S. Q. Shen, Y. Zhou, Y. Y. Shao, J. Kang, Y. T. Chen, M. Y. Chen, X. J. Yuan, I. Hamza, A. R. Reddi and C. Y. Chen, Nature, 2022, 610, 768–774.
7. K. Kawai, T. Hirayama, H. Imai, T. Murakami, M. Inden, I. Hozumi and H. Nagasawa, J. Am. Chem. Soc., 2022, 144, 3793–3803.
8. R. C. Hider and X. L. Kong, BioMetals, 2011, 24, 1179–1187.
9. S. J. Patel, A. G. Frey, D. J. Palenchar, S. Achar, K. Z. Bullough, A. Vashisht, J. A. Wohlschlegel and C. C. Philpott, Nat. Chem. Biol., 2019, 15, 872–881.
10. L. C. Montemiglio, C. Testi, P. Ceci, E. Falvo, M. Pitea, C. Savino, A. Arcovito, G. Peruzzi, P. Baiocco, F. Mancia, A. Boffi, A. d. Georges and B. Vallone, Nat. Commun., 2019, 10, 1121.
11. J. Gruszczyk, U. Kanjee, L. J. Chan, S. Menant, B. Malleret, N. T. Y. Lim, C. Q. Schmidt, Y. F. Mok, K. M. Lin, R. D. Pearson, G. Rangel, B. J. Smith, M. J. Call, M. P. Weekes, M. D. W. Griffin, J. M. Murphy, J. Abraham, K. Sriprawat, M. J. Menezes, M. U. Ferreira, B. Russell, L. Renia, M. T. Duraisingh and W. H. Tham, Science, 2018, 359, 48–55.
12. H. H. Jabara, S. E. Boyden, J. Chou, N. Ramesh, M. J. Massaad, H. Benson, W. Bainter, D. Fraulino, F. Rahimov, C. Sieff, Z. J. Liu, S. H. Alshemmari, B. K. Al-Ramadi, H. Al-Dhekri, R. Arnaout, M. Abu-Shukair, A. Vatsayan, E. Silver, S. Ahuja, E. G. Davies, M. Sola-Visner, T. K. Ohsumi, N. C. Andrews, L. D. Notarangelo, M. D. Fleming, W. Al-Herz, L. M. Kunkel and R. S. Geha, Nat. Genet., 2016, 48, 74–78.
13. I. Yanatori, D. R. Richardson, H. S. Dhekne, S. Toyokuni and F. Kishi, Blood, 2021, 138, 1490–1503.
14. C. W. Brown, J. J. Amante, P. Chhoy, A. L. Elaimy, H. Liu, L. J. Zhu, C. E. Baer, S. J. Dixon and A. M. Mercurio, Dev. Cell, 2019, 51, 575–586.
15. M. Truman-Rosentsvit, D. Berenbaum, L. Spektor, L. A. Cohen, S. Belizowsky-Moshe, L. Lifshitz, J. Ma, W. Li, E. Kesselman, I. Abutbul-Ionita, D. Danino, L. Gutierrez, H. H. Li, K. Y. Li, H. F. Lou, M. Regoni, M. Poli, F. Glaser, T. A. Rouault and E. G. Meyron-Holtz, Blood, 2018, 131, 342–352.
16. R. Kalluri and V. S. LeBleu, Science, 2020, 367, 640.
17. M. Tkach and C. Théry, Cell, 2016, 164, 1226–1232.
18. G. van Niel, G. D'Angelo and G. Raposo, Nat. Rev. Mol. Cell Biol., 2018, 19, 213–228.
19. G. S. Bi, J. Q. Liang, Y. Y. Bian, G. Y. Shan, Y. W. Huang, T. Lu, H. Zhang, X. Jin, Z. C. Chen, M. N. Zhao, H. Fan, Q. Wang, B. Y. Gan and C. Zhan, Nat. Commun., 2024, 15, 2461.
20. C. W. Brown and A. M. Mercurio, Mol. Cell. Oncol., 2020, 7, e1730144.
21. H. K. C. Co, C. C. Wu, Y. C. Lee and S. H. Chen, Nature, 2024, 631, 654–662.
22. A. Purnianto, C. Mawal, M. M. Kulkarni, H. Su, T. F. Koukoulis, P. Wongsodirdjo, Y. H. Hung, S. Ayton, A. I. Bush, K. J. Barnham and L. J. Vella, J. Extracell. Biol., 2024, 3, e70012.
23. C. Alarcón-Veleiro, R. Mato-Basalo, S. Lucio-Gallego, A. Vidal-Pampín, M. Quindós-Varela, T. Al-Qatarneh, G. Berrecoso, A. Vizoso-Vázquez, M. C. Arufe and J. Fafián-Labora, Antioxidants, 2023, 12, 183.
24. K. Grover, A. Koblova, A. T. Pezacki, C. J. Chang and E. J. New, Chem. Rev., 2024, 124, 5846–5929.
25. K. Lee, K. Fraser, B. Ghaddar, K. Yang, E. Kim, L. Balaj, E. A. Chiocca, X. O. Breakefield, H. Lee and R. Weissleder, ACS Nano, 2018, 12, 494–503.
26. Y. N. Guo, J. Tao, Y. R. Li, Y. M. Feng, H. X. Ju, Z. F. Wang and L. Ding, J. Am. Chem. Soc., 2020, 142, 7404–7412.
27. J. L. Zhang, J. J. Shi, H. L. Zhang, Y. F. Zhu, W. Liu, K. X. Zhang and Z. Z. Zhang, J. Extracell. Vesicles, 2020, 10, e12025.
28. B. Q. Lin, T. Tian, Y. Z. Lu, D. Liu, M. J. Huang, L. Zhu, Z. Zhu, Y. L. Song and C. Y. Yang, Angew. Chem., Int. Ed., 2021, 60, 7582–7586.
29. H. S. Liu, Y. Tian, C. F. Xue, Q. Niu, C. Chen and X. M. Yan, J. Extracell. Vesicles, 2022, 11, e12206.
30. S. B. Zhu, L. Ma, S. Wang, C. X. Chen, W. Q. Zhang, L. L. Yang, W. Hang, J. P. Nolan, L. N. Wu and X. M. Yan, ACS Nano, 2014, 8, 10998–11006.
31. Y. Tian, L. Ma, M. F. Gong, G. Q. Su, S. B. Zhu, W. Q. Zhang, S. Wang, Z. B. Li, C. X. Chen, L. H. Li, L. N. Wu and X. M. Yan, ACS Nano, 2018, 12, 671–680.
32. T. Peng and D. Yang, Org. Lett., 2010, 12, 496–499.
33. T. Hirayama, H. Tsuboi, M. Niwa, A. Miki, S. Kadota, Y. Ikeshita, K. Okuda and H. Nagasawa, Chem. Sci., 2017, 8, 4858–4866.
34. G. Adamo, S. Picciotto, P. Gargano, A. Paterna, S. Raccosta, E. Rao, D. P. Romancino, G. Ghersi, M. Manno, M. Salamone and A. Bongiovanni, J. Extracell. Vesicles, 2025, 14, e70030.
35. K. Wardhani, A. Levina, G. E. R. Grau and P. A. Lay, Chem. Soc. Rev., 2024, 53, 6779–6829.
36. N. G. Cai, X. Z. Zhan, Y. Chen, J. W. Xue, C. Chen, Y. R. Li, Y. Tian and X. M. Yan, Anal. Chem., 2024, 96, 12718–12728.
37. J. A. Pan, W. Y. Xiong, A. L. Zhang, H. Zhang, H. Lin, L. Gao, J. H. Ke, S. Y. Huang, J. F. Zhang, J. Gu, A. C. Y. Chang and C. Q. Wang, Adv. Sci., 2023, 10, 2206007.
38. V. Manolova, N. Nyffenegger, A. Flace, P. Altermatt, A. Varol, C. Doucerain, H. Sundstrom and F. Dürrenberger, J. Clin. Invest., 2020, 130, 491–506.

---