A water-soluble NIR-II fluorescent probe for non-invasive real-time detection of blood ATP via optoacoustic and fluorescence imaging

Abstract

Adenosine triphosphate (ATP) is a critical biomolecule in cellular energy metabolism, with abnormal levels in the bloodstream linked to pathological conditions such as ischemia, cancer, and inflammatory disorders. Accurate and real-time detection of ATP is essential for early diagnosis and disease monitoring. However, conventional biochemical assays and other techniques suffer from limitations, including invasive sample collection, time-consuming procedures, and the inability to provide dynamic, in vivo monitoring. To address these challenges, we present a water-soluble near-infrared-II (NIR-II) fluorescent probe based on a heptamethine-cyanine/Zn[II] complex for the dual-modal detection of ATP via NIR-II fluorescence and optoacoustic imaging. The probe is designed with polyethylene glycol-functionalized benzindole groups for enhanced water solubility and biocompatibility as well as a dipicolylamine–Zn[II] complex that selectively binds ATP. Upon interaction with ATP, the probe exhibits a distinct absorption band (700–850 nm), enhanced NIR-II fluorescence (900–1200 nm, peak at 924 nm), and strong optoacoustic signals, enabling non-invasive and real-time ATP monitoring. This approach offers significant advantages over existing detection methods by combining high sensitivity and dynamic imaging capabilities. Our findings demonstrate that the selective responsiveness of the probe to ATP renders it highly suitable for real-time in vivo monitoring of ATP levels.

---

Introduction

Adenosine triphosphate (ATP) is an indispensable molecule in cellular energy metabolism, playing vital roles in energy transfer, cellular signalling, and numerous biochemical processes. It is the primary energy source for various cellular functions, including protein synthesis, muscle contraction, and active transport.^1,2 However, elevated levels of ATP in the bloodstream can be indicative of various pathological conditions, such as ischemia, cancer, and inflammatory disorders. Excess ATP in the blood can result in the activation of purinergic receptors, leading to inflammation, oxidative stress, and disruption of cellular homeostasis. Furthermore, abnormal ATP levels can impair tissue and organ function, contributing to the progression of diseases like sepsis, myocardial infarction, and cancer.^3–7 Consequently, the accurate monitoring and quantification of ATP levels in the bloodstream are essential for the early diagnosis and monitoring as well as treatment of these diseases.

Currently, the detection of elevated ATP levels predominantly relies on biochemical assays or other techniques, such as liquid chromatography, mass spectrometry, and enzyme-linked immunosorbent assays (ELISA).^8–11 While these methods provide valuable insights into molecular concentrations, they generally suffer from several limitations. Biochemical assays often require extensive sample preparation, complex instrumentation, and time-consuming processes, which can delay clinical decision-making. Additionally, these techniques typically involve invasive procedures or the extraction of biological samples, precluding real-time monitoring in living organisms. Moreover, these methods lack the ability to dynamically track ATP fluctuations in response to physiological changes or therapeutic interventions. Therefore, there is a pressing need for the development of more efficient, sensitive, and non-invasive approaches for the real-time detection of ATP in vivo in a visualizing way.

Near-infrared (NIR) fluorescence and optoacoustic imaging have emerged as promising technologies for non-invasive, in vivo molecular imaging due to their high spatial resolution and minimal background interference. Specifically, fluorescence imaging in the NIR-II range (900–1700 nm) has garnered increasing attention because of its reduced autofluorescence and good tissue penetration,^12–15 which enables deeper tissue imaging with enhanced signal-to-noise ratios, making it ideal for tracking biomolecular changes in vivo.^16–20 On the other hand, optoacoustic imaging, which combines the tissue-penetrating capabilities of ultrasound with the high spatial resolution of optical imaging, provides a unique and versatile approach for molecular and functional imaging.^21–28 The complementary nature of these techniques allows for the simultaneous acquisition of both high-resolution optical and deep-tissue ultrasound data, offering substantial advantages over traditional imaging modalities. Together, NIR-II fluorescence and optoacoustic imaging hold great potential for real-time, non-invasive visualization and monitoring of biomolecular targets with high sensitivity, offering significant improvements over current detection methods.

In this study, we present the development of a water-soluble NIR-II fluorescent probe based on a heptamethine-cyanine/Zn[II] complex (referred to as HCD9–Zn[II]), designed specifically for monitoring blood ATP levels via both NIR-II fluorescence and optoacoustic imaging (Scheme 1). The probe is strategically engineered with two terminal benzindole groups, each attached to a polyethylene glycol (PEG9) chain to ensure solubility in aqueous environments. At the core of the probe lies a dipicolylamine–Zn[II] complex that facilitates specific interaction with ATP. In the absence of ATP, weak fluorescence in the NIR-II range along with weak absorption around 750 nm can be found for the probe. However, upon interaction of ATP with the probe, the zinc ion in the probe forms a stronger coordination bond with ATP, thus generating free HCD9 and consequently resulting in a marked change in the optical properties. This interaction leads to a pronounced absorption peak at 750 nm, accompanied by enhanced fluorescence in the NIR-II range (900–1200 nm, with a peak at 924 nm) and the generation of strong optoacoustic signals. The probe's ability to selectively respond to ATP levels makes it an ideal candidate for real-time monitoring of ATP concentrations in vivo. Importantly, this system allows for the non-invasive, simultaneous visualization of ATP levels through both NIR-II fluorescence and optoacoustic imaging, providing a powerful tool for biomedical research and clinical diagnostics.

Scheme 1 Schematic presentation of the application of HCD9–Zn[II] (the probe) for detecting blood ATP in mice via NIR-II fluorescence (NIR-II FL) and multispectral optoacoustic tomography (MSOT) imaging in real time. Magenta dotted-line square: scanning region covering the jugular veins in MSOT imaging; aqua blue dotted-line circle: the region covering the jugular veins in NIR-II fluorescent image.

Results and discussion

Preparation of the dye HCD9 and the probe HCD9–Zn[II] and their optical spectral properties

To synthesize the dye HCD9, two poly(ethylene glycol) (PEG9) chains were integrated into the terminal benzindole groups of heptamethine cyanine, significantly enhancing its water solubility and biocompatibility, as PEG is a well-known hydrophilic and biocompatible polymer. Subsequently, dipicolylamine (DPA) was conjugated to the central heptamethine cyanine core, serving as the Zn[II] complexation site. The whole preparation process is shown in Fig. S1. The DPA group, a multidentate ligand, provides strong selectivity for Zn[II] binding over biologically prevalent metal ions such as Mg²⁺, Na⁺, K⁺, and Ca²⁺.^29–33

The successful synthesis of HCD9 was confirmed through proton and carbon nuclear magnetic resonance spectroscopy (¹H and ¹³C NMR) and high-resolution mass spectrometry (HRMS) (Fig. S2–S5). Finally, the HCD9–Zn[II] complex was prepared by dissolving HCD9 and zinc chloride (1 : 1.2 in mol) in water under stirring for 3 h. The high-resolution mass (HRMS) spectrum indicates the formation of the HCD9–Zn[II] complex (Fig. S6).

As shown in Fig. S7(a), the dye HCD9 exhibits an absorption peak at approximately 750 nm and a fluorescence emission in the NIR-II region (900–1200 nm), with a primary emission peak at around 924 nm. Upon binding with Zn[II] ions, the HCD9–Zn[II] complex shows significantly reduced absorption and emission intensities compared to free HCD9. This attenuation is attributed to the strong coordination between Zn[II] and HCD9, which induces partial fluorescence quenching through metal–ligand interactions (Fig. 1(a) and (b)). Fig. 1(c) and (d) demonstrate that, apart from minor interference from Cu²⁺, other metal ions, including Na⁺, Li⁺, Mg²⁺, K⁺, Al³⁺, Ca²⁺, Fe³⁺, Co²⁺, Cr³⁺, Ni²⁺, and Ba²⁺, exhibit negligible impact on the binding of Zn[II] to HCD9. These findings highlight the high specificity of HCD9 as a ligand for Zn[II] ions.

Fig. 1 (a) Normalized absorption and (b) normalized emission spectra of the HCD9 and HCD9 in the presence of Zn[II]. (c) Normalized emission intensity of HCD9 at 924 nm in the absence (Blank) or presence of different metal ions. (n = 3) (d) Normalized emission intensity of HCD9 solution at 924 nm without Zn[II] (Blank) or under coexistence of Zn[II] and different metal ions. (n = 3) (HCD9 : 20 μM).

To investigate the binding stoichiometry between HCD9 and Zn[II] ions, the total molar concentration of HCD9 and Zn[II] was maintained constant at 30 μM, while the Zn[II] to HCD9 molar ratio was systematically varied by adjusting Zn[II] concentrations from 0 to 30 μM in increments of 3 μM. Fluorescence intensity variations were analysed using Job's plot to determine the precise binding stoichiometry.³⁴ As shown in Fig. 2(a), at 924 nm, the fluorescence intensity of the HCD9 solution decreases progressively as the Zn[II] molar fraction increases from 0 to 0.5. Beyond this point, further Zn[II] additions result in a slower quenching trend. Linear regression analysis of the fluorescence data indicates a 1 : 1 binding ratio, suggesting the formation of a 1 : 1 HCD9–Zn[II] coordination complex (Fig. 2(a)). High-resolution mass spectrometry (HRMS) results further confirm this stoichiometry. The characteristic ion peak observed at m/z = 543.6178 (Fig. S6) closely matches the theoretical m/z of 543.6177 for HCD9–Zn[II], validating the 1 : 1 binding interaction. Moreover, HCD9 demonstrates a high binding affinity for Zn[II], with a binding constant (Ka) of 1.64 × 10⁴ M⁻¹ (Fig. S7(b)), as calculated using the Benesi–Hildebrand equation.^35,36 The probe HCD9–Zn[II] exhibits a response time of approximately 100 s toward ATP, as evidenced by the emission intensity ratios (I/I₀) in Fig. S7(c). The pH stability of both HCD9 and its Zn[II] complex was also evaluated (Fig. 2(b)), revealing minimal variation across the pH range of 5.0 to 8.0. This pH insensitivity suggests strong potential for physiological applications.

Fig. 2 (a) Complexation ratio of Zn[II] with HCD9. (b) Normalized fluorescence intensity of HCD9 and HCD9–Zn[II] at different pH values (n = 3, HCD9 : 20 μM). (c) Normalized emission spectra of the HCD9–Zn[II] in the presence of different ions (HCD9 : 20 μM). (d) Normalized emission intensity of HCD9–Zn[II] at 924 nm in the presence of different ions (n = 3, HCD9 : 20 μM).

Fluorescence and optoacoustic response of HCD9–Zn[II] toward ATP

Complexes formed between dipicolylamine (DPA) and Zn[II] ions are known for their selective affinity toward phosphate groups, in particular organic phosphate groups, with formation constants reaching up to 10⁶ M⁻¹ for Zn[II]–organic phosphate interactions.^37–42 Adenosine triphosphate (ATP), a biologically important molecule, consists of adenine, a five-carbon ribose sugar, and a triphosphate chain composed of three phosphate groups. The binding between ATP and Zn[II] usually involves three phosphate oxygen atoms and one nitrogen atom (namely N7, from adenine), with the binding constant reaching up to 10⁵–10⁶ M⁻¹, which is much stronger than Zn[II]–inorganic phosphate (binding constant 10²–10³ M⁻¹).^37,41,42 Based on this structural feature, our probe HCD9–Zn[II] is expected to selectively recognize ATP. Experimental results confirm strong binding interactions between HCD9–Zn[II] and ATP. As shown in Fig. 2(c) and (d), the addition of ATP to the HCD9–Zn[II] complex induces a significant fluorescence enhancement in emission intensity compared to HCD9–Zn[II] alone. Among the various analytes tested, only AMP and ADP, which also contain phosphate groups, produce modest fluorescence increases. In contrast, other anions, including HPO₄²⁻ and H₂PO₄⁻, do not elicit any notable fluorescence change. In blood, phosphate primarily exists as inorganic phosphate species, with HPO₄²⁻ and H₂PO₄⁻ being the dominant forms at physiological pH. This selectivity ensures that inorganic phosphate in blood will not interfere with the probe's response toward ATP. These results highlight the high selectivity of HCD9–Zn[II] for ATP recognition.

To further explore this interaction, fluorescence spectra were recorded with varying ATP concentrations in HCD9–Zn[II] solution. As illustrated in Fig. 3(a), the fluorescence intensity increases proportionally with ATP concentrations. Notably, a linear relationship was observed between fluorescence intensity at 924 nm and ATP concentrations within the 0–15 μM range (Fig. 3(b)). The limit of detection (LOD) for ATP is determined to be 1.09 μM using the equation LOD = 3σ/κ.⁴³

Fig. 3 (a) Fluorescence spectra of the HCD9–Zn[II] after addition of ATP. (b) Fluorescence intensity versus ATP level (for determination of the detection limit of HCD9–Zn[II] for ATP). (n = 3). (c) Determination of formation constant of competitive binding between ATP and Zn[II] in the HCD9–Zn[II] complex. (d) Relative optoacoustic intensity at 750 nm for HCD9–Zn[II] upon reaction in water with varied concentrations of ATP. (n = 3). (HCD9: 20 μM).

ATP, as an organic phosphate, exhibits strong affinity for Zn[II], enabling it to displace HCD9 from the probe (HCD9–Zn[II]) and competitively binds with Zn[II]. The binding affinity between ATP and Zn[II] in the HCD9–Zn[II] complex was quantified using the Benesi–Hildebrand equation,^35,36 yielding a binding constant (K_a) of 1.27 × 10⁵ M⁻¹ (Fig. 3(c), y = 3.29 × 10⁻⁸x + 0.0042). Correspondingly, the optoacoustic intensity of HCD9–Zn[II] also rises with increasing ATP levels, as shown in Fig. 3(d). Moreover, after HCD9–Zn[II] was reacted with ATP, the HR mass spectrum was recorded for the reaction solution (Fig. S8). It is obvious that the m/z peak corresponding to HCD9 can be found, indicating that ATP binds with Zn[II] and consequently HCD9 is released. These results confirm that HCD9–Zn[II] responds effectively to ATP, producing enhanced NIR-II fluorescence and optoacoustic signals suitable for bioimaging applications.

Biosafety of the probe HCD9–Zn[II].

Before conducting in vivo imaging, the cytotoxicity of HCD9 and its Zn[II] complex (HCD9–Zn[II]) was evaluated using the MTT assay in L929 cells. As shown in Fig. S9(a), cell viability remains above 90% after 24 hours of incubation with HCD9 or HCD9–Zn[II] at concentrations up to 70 μM, indicating low cytotoxicity. Additionally, hemolysis tests reveal that neither HCD9 (degree of hemolysis 0.8%) nor HCD9–Zn[II] (degree of hemolysis 2.6%) causes significant red blood cell damage at a concentration of 60 μM (Fig. S9(b)).

To further assess biosafety, healthy mice were intravenously injected with either saline (100 μL, control) or HCD9–Zn[II] (20 μM in 100 μL saline). Seven days post-injection, major organs were collected for histological examination via H&E staining analysis. No noticeable histopathological abnormalities could be observed in the tissues of either group (Fig. S10), further confirming the excellent biosafety of HCD9–Zn[II] for in vivo applications.

In vivo imaging of blood ATP in mice.

ATP serves as the primary energy currency of cells and is predominantly produced and utilized intracellularly (e.g., in the mitochondria and cytoplasm). Under normal physiological conditions, most cells do not release significant amounts of ATP into the bloodstream. However, extracellular ATP levels can increase in specific pathological states. To simulate elevated blood ATP levels for experimental purposes, ATP dissolved in saline was intravenously administered to mice. This approach allowed evaluation of the potential of HCD9–Zn[II] for non-invasive, real-time imaging of ATP in the bloodstream in vivo. All in vivo experiments were approved by and conducted in accordance with the guidelines of the Ethics Committee of Laboratory Animal Center of South China Agricultural University (Approval no. 2025D002). This study was conducted in conformity with the Regulations on the Administration of Laboratory Animals of Guangdong Province and the National Regulations on the Management of Laboratory Animals of China. For the experiments, healthy mice were randomly divided into 3 mice per group. The control group received no ATP, while three experimental groups were administered different doses of ATP via intravenous injection: Group 1 : 0.5 mg kg⁻¹, Group 2 : 1.1 mg kg⁻¹, Group 3 : 2.2 mg kg⁻¹. Imaging was performed and focused on the neck region, where the jugular veins – relatively large and accessible – facilitate signal acquisition. One minute after ATP administration, HCD9–Zn[II] was intravenously injected, and NIR-II fluorescence images were acquired at various time points post-injection (Fig. S11 and Fig. 4(a)).

Fig. 4 (a) NIR-II fluorescence images of various groups of mice at 35 min after the HCD9–Zn[II] injection and after ATP pre-injection. (b) 3D-MSOT images (vertical view) at 35 min after the HCD9–Zn[II] injection and after ATP pre-injection. Scale bar: 5 mm. The magenta dotted-line square in the mouse picture shows the scanning region in MSOT imaging. (HCD9–Zn[II]: 3.2 mg kg⁻¹).

5 minutes after probe injection, a clear NIR-II fluorescence signal appears in the vascular region of the neck, with intensity continuing to increase and peaking around 35 minutes. This allows for effective visualization of ATP levels in the jugular veins, as illustrated in Fig. 4(a). In contrast, control mice that did not receive ATP exhibit negligible fluorescence signals in the same region, since under normal physiological conditions, there is very little free ATP in the bloodstream. Quantitative analysis of the average fluorescence intensity (within the region of interest (ROI) encompassing the neck veins) at each time point is provided in Fig. S12.

Following this, multispectral optoacoustic tomography (MSOT) was employed to image the jugular veins of mice. The animals were divided into four groups, with three groups receiving tail vein injections of ATP at varying doses (group 1 :  0.5 mg kg⁻¹, group 2 :  1.1 mg kg⁻¹, and group 3 :  2.2 mg kg⁻¹), and the control group received no ATP. Prior to MSOT imaging, all mice were intravenously injected with the HCD9–Zn[II] probe (3.2 mg kg⁻¹). MSOT images acquired at different times are shown in Fig. S13–S15, and Fig. 4(b).

Notably, mice administered with ATP exhibit significantly enhanced MSOT signals in the jugular veins compared to the control group, which received no ATP. Moreover, the MSOT signal intensity increases in a dose-dependent manner with ATP concentration and peaks around 35 minutes post-injection, enabling clear visualization of ATP levels in the blood vessels. These findings are consistent with the NIR-II fluorescence imaging results (Fig. 4a and Fig. S11, S12).

Importantly, as shown in Fig. 4(b), MSOT imaging provides higher spatial resolution and clearer visualization of the jugular veins than NIR-II fluorescence imaging. These results demonstrate that HCD9–Zn[II] is a promising probe for real-time, non-invasive monitoring of blood ATP levels with dual-modality MSOT and NIR-II fluorescent imaging.

To assess the metabolic pathway of the probe, the dye HCD9 was intravenously injected into healthy mice, followed by NIR-II fluorescence imaging. As displayed in Fig. S16, detectable fluorescence signals persist for up to 24 hours post-injection but nearly disappear by 36 hours, indicating efficient clearance of the dye from the body. This suggests that HCD9—the response product formed from the interaction between ATP and HCD9–Zn[II]—can be effectively eliminated without posing a risk of long-term in vivo accumulation.

Additionally, twelve hours after injection, both NIR-II fluorescence and MSOT imaging were performed on major organs to examine biodistribution. As shown in Fig. 5, strong fluorescence and optoacoustic signals can be observed in the liver and kidneys, with the kidneys displaying the most intense signals. These results confirm that HCD9 is primarily excreted via the renal pathway, with partial hepatic clearance, supporting its suitability for biological applications with minimal long-term retention.

Fig. 5 (a) The organs’ (1: heart, 2: liver, 3: kidneys, 4: spleen and 5: lungs) NIR-II fluorescence images at 12 h after the HCD9–Zn[II] injection and after ATP pre-injection. (b) Mean NIR-II fluorescence intensity for the organs in (a) (n = 3). (c) Cross-sectional MSOT images of main organs from various groups at 12 h after the HCD9–Zn[II] injection and after ATP pre-injection. (d) Mean MSOT intensities of major organs in (c) (n = 3).

Conclusions

In summary, we have developed a dual-mode ATP-responsive probe, HCD9–Zn[II], capable of real-time, non-invasive imaging of ATP levels through both NIR-II fluorescence and optoacoustic modalities. Compared to previously reported optical probes for ATP (Table S1), this probe is the first NIR-II fluorescent/optoacoustic dual-mode sensor with a rapid response, enabling real-time in vivo imaging. In the absence of ATP, the probe exhibits minimal fluorescence in the NIR-II region. In the presence of ATP, Zn[II] in the probe coordinates much strongly with ATP, resulting in a distinct absorption peak at 750 nm, a marked enhancement in NIR-II fluorescence, and a strong optoacoustic signal.

This ATP-triggered response enables effective dual-mode imaging using NIR-II fluorescence and 3D multispectral optoacoustic tomography (3D-MSOT), providing a powerful platform for dynamic ATP detection in blood in vivo. Moreover, the probe is efficiently cleared from the body post-imaging, mitigating concerns about long-term toxicity or accumulation. Collectively, this system offers a non-invasive, sensitive, and biocompatible strategy for visualizing blood ATP dynamics, with significant potential for applications in biomedical research and clinical diagnostics.

Author contributions

Yonghe Zhang: conceptualization, investigation, methodology, formal analysis, and writing – original draft. Fang Zeng: conceptualization, project administration, formal analysis, funding acquisition, supervision, resources, and writing – review & editing. Shuizhu Wu: conceptualization, formal analysis, project administration, funding acquisition, supervision, resources, and writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the SI. Supplementary information: experimental procedures, synthetic route, ¹H/¹³C NMR and HR mass spectra, absorption and emission spectra, NIR-II and MSOT images, and biosafety data. See DOI: https://doi.org/10.1039/d5tb01208b

Acknowledgements

This work was supported by NSFC (52373209 and 22274057) and the Fund from the Guangdong Provincial Key Laboratory of Luminescence from Molecular Aggregates (2023B1212060003).

References

1. W. Ryu, T. S. Fung, D. C. Baker, M. Saoi, J. Park, C. A. Febres-Aldana, R. G. Aly, R. Cui, A. Sharma, Y. Fu, O. L. Jones, X. Cai, H. A. Pasolli, J. R. Cross, C. M. Rudin and C. B. Thompson, Nature, 2024, 635, 746–754.
2. D. Bennett, E. H. Kimball, M. Gao, R. Osterhout, S. J. Van Dien and J. D. Rabinowitz, Nat. Chem. Biol., 2009, 5, 593–599.
3. S. E. Radzikowska-Cieciura, A. Chworos and R. Pawlowska, Curr. Med. Chem., 2023, 30, 1232.
4. T. V. Zhao, Y. Li, X. Liu, S. Xia, P. Shi, L. Li, Z. Chen, C. Yin, M. Eriguchi, Y. Chen, E. A. Bernstein, J. F. Giani, K. E. Berstein and X. Z. Shen, Sci. Immunol., 2019, 4, eaau6426.
5. Y. Li, J. Zhou, I. Burkovskiy, P. Yeung and C. Lehmann, Med. Hypotheses, 2019, 124, 84–86.
6. A. W. Lohman, M. Billaud and B. E. Isakson, Cardiovasc. Res., 2012, 95, 269–280.
7. W. J. Ganesan, T. Müller, C. Durr, M. Grimm, A. Beilhack, C. D. Krempl, S. Sorichter, U. V. Gerlach, E. Juttner, A. Zerweck, F. Gartner, P. Pellegatti, F. D. Virgilio, D. Ferrari, N. Kambham, P. Fisch, J. Finke, M. Idzko and R. Zeiser, Nat. Med., 2010, 16, 1434–1438.
8. M. Menegollo, I. Tessari, L. Bubacco and G. Szabadkai, Methods Mol. Biol., 2019, 1925, 223–232.
9. M. Rajendran, E. Dane, J. Conley and M. Tantama, Biol. Bull., 2016, 231, 73–84.
10. H. Xu, J. Liang, Y. Wang, B. Wang, T. Zhang, X. Liu and L. Gong, PLoS One, 2019, 14, e0221665.
11. Y. Li, M. E. Weese, M. T. Cryan and A. E. Ross, Anal. Methods, 2021, 13, 2320–2330.
12. R. Li, F. Zeng and S. Wu, Anal. Chem., 2025, 97, 13681–13690.
13. G. S. Hong, J. C. Lee, J. T. Robinson, U. Raaz, L. M. Xie, N. F. Huang, J. P. Cooke and H. J. Dai, Nat. Med., 2012, 18, 1841–1846.
14. X. Chen, J. Y. Li, S. Roy, Z. Ullah, J. S. Gu, H. Y. Huang, C. Yu, X. J. Wang, H. Wang, Y. H. Zhang and B. Guo, Adv. Healthcare Mater., 2024, 13, 2304506.
15. F. F. Wang, Y. T. Zhong, O. Bruns, Y. Y. Liang and H. J. Dai, Nat. Photonics, 2024, 18, 535–547.
16. C. B. Zhang, Y. L. Wu, F. Zeng, Y. B. Wen, J. W. Chen, G. W. Deng, L. L. Zhang, S. L. Zhao, S. Z. Wu and Y. L. Zhao, Angew. Chem., Int. Ed., 2024, 34, e202406694.
17. S. Liu, W. H. Dong, H. Q. Gao, Z. R. Song and Z. Cheng, Acc. Chem. Res., 2025, 58, 543–554.
18. S. X. Lyu, S. Y. Lu, C. H. Gui, C. Y. Guo, J. J. Han, Y. L. Xiao, R. P. Zhang and X. C. Hong, J. Med. Chem., 2024, 67, 1861–1871.
19. Y. F. Tang, Y. Y. Li, C. X. He, Z. Wang, W. Huang, Q. L. Fan and B. Liu, Nat. Commun., 2025, 16, 278.
20. H. J. Chen, L. Wang, H. Zhu, Z. G. Wang and S. L. Liu, ACS Appl. Mater. Interfaces, 2024, 22, 28011–28028.
21. Y. L. Wu, F. Zeng, Y. L. Zhao and S. Z. Wu, Chem. Soc. Rev., 2021, 50, 7924.
22. J. J. Chen, L. Q. Chen, Y. L. Wu, Y. C. Fang, F. Zeng, S. Z. Wu and Y. L. Zhao, Nat. Commun., 2021, 12, 6870.
23. J. Ouyang, L. H. Sun, Z. Zeng, C. Zeng, F. Zeng and S. Wu, Angew. Chem., Int. Ed., 2020, 25, 10111–10121.
24. L. H. Sun, J. Ouyang, Y. Q. Ma, Z. Zeng, C. Zeng, F. Zeng and S. Z. Wu, Adv. Healthcare Mater., 2021, 24, 2100867.
25. J. Ouyang, L. H. Sun, Z. P. She, R. Li, F. Zeng, Z. C. Yao and S. Z. Wu, ACS Appl. Mater. Interfaces, 2024, 16, 20315–20328.
26. J. G. Huang and K. Y. Pu, Angew. Chem., Int. Ed., 2020, 59, 11717.
27. Y. L. Wu, S. L. Huang, J. Wang, L. H. Sun, F. Zeng and S. Z. Wu, Nat. Commun., 2018, 9, 3983.
28. Z. P. She, R. Li, F. Zeng and S. Z. Wu, Adv. Healthcare Mater., 2024, 19, 2400791.
29. A. Krinochkin, M. Valieva, E. Starnovskaya, N. Slovesnova, A. Minin, A. Belousova, L. Sadieva, O. Taniya, A. Khasanov, A. Novikov, V. Bruskov, S. Vatolina, D. Kopchuk, P. Slepukhin, V. Sharutin and G. Zyryanov, J. Fluoresc., 2024, 35, 1–17.
30. D. Paderni, E. Macedi, G. E. Giacomazzo, M. Formica, L. Giorgi, B. Valtancoli, P. Rossi, P. Paoli, L. Conti, V. Fusi and C. Giorgi, Dalton Trans., 2024, 22, 9495–9509.
31. H. T. Ngo, X. Liu and K. A. Jolliffe, Chem. Soc. Rev., 2012, 14, 4928–4965.
32. A. Waheed, T. Ahmad, M. Haroon and N. Ullah, ChemistrySelect, 2020, 17, 5300–5305.
33. J. Feng, J. Z. Li, X. M. Mao, Q. Wang, S. P. Li and C. Y. Wang, Analyst, 2021, 146, 3971–3976.
34. J. X. Fu, Y. X. Chang, B. Li, X. H. Wang, X. M. Xie and K. X. Xu, Spectrochim. Acta, Part A, 2020, 225, 117493.
35. I. D. Kuntz, F. P. Gasparro, M. D. Johnston and R. P. Taylor, J. Am. Chem. Soc., 1968, 90, 4778–4781.
36. J. S. Heo, D. Gil and C. Kim, J. Fluoresc., 2022, 32, 825–833.
37. A. Ojida, Y. Mito-oka, M. Inoue and I. Hamachi, J. Am. Chem. Soc., 2002, 124, 6256–6258.
38. Y. B. Zhang, B. L. Wang, X. Q. Rong, J. Liu, X. Y. Qiu, L. Sun and Y. T. Cheng, Tetrahedron Lett., 2023, 129, 154713.
39. A. Ojida, Y. Mito-oka, K. Sada and I. Hamachi, J. Am. Chem. Soc., 2004, 126, 2454–2463.
40. A. S. Rao, D. Kim, H. Nam, H. Jo, K. H. Kim, C. Ban and H. Ahn, Chem. Commun., 2012, 48, 3206–3208.
41. H. T. Ngo, X. J. Liu and K. A. Jolliffe, Chem. Soc. Rev., 2012, 41, 4928–4965.
42. A. Krężel and W. Maret, Arch. Biochem. Biophys., 2016, 611, 3–19.
43. M. Subbaiah, K. K. Mandrekar, A. K. Raees Ahmed, A. K. Rahiman, K. Muthu and S. G. Thangavelu, Appl. Organomet. Chem., 2025, 39, e70093.

---