English

• Singlet oxygen (¹O₂), the lowest excited electronic state of molecular oxygen, exhibits significantly enhanced reactivity in comparison to oxygen in its triplet ground state [1-3]. In clinical photodynamic therapy (PDT), this heightened reactivity of ¹O₂ facilitates the oxidation of nearby biomolecules, triggering inflammatory responses or immune reactions, ultimately resulting in tumor ablation or bacteria eradication [4-10]. Hence, real-time visualization of ¹O₂ in living cells and pathological tissue is of paramount importance to elucidate its mechanism of action in medical processes and optimize therapy protocols [11-15]. Currently, the gold standard for ¹O₂ determination is the measurement of the phosphorescence of ¹O₂ at 1270 nm [16,17]. However, the short luminescence lifetime and extremely low quantum yield of ¹O₂ lead to weak phosphorescence, severely limiting its biological application. Reaction-based optical probes offer an alternative strategy for ¹O₂ detection under biological condition due to their high sensitivity and non-invasive features [18-20]. Particularly, chemiluminescent probes, which emit light through specific chemical reactions with ¹O₂, does not require real-time excitation light irradiation, effectively eliminating interference from light scattering and autofluorescence with fluorescent probes. This characteristic enables unprecedented sensitivity and selectivity [21-24].

  Despite the significant advantages of chemiluminescence for biomarker sensing, the development of chemiluminescent probes is still in its infancy, with merely a few cases enabling imaging ¹O₂ in living cells and animals [25,26]. In general, several factors including photo-induced self-oxidation during photodynamic therapy (PDT), poor cell permeability, and short emission wavelengths greatly limit the in vivo application of the ¹O₂ imaging probes: (ⅰ) Photo-induced self-oxidation: our previous work revealed that chemiluminescent precursors containing electron-rich enolether unit are prone to undergo photo-oxidation reaction accompanied by chemiluminescence generation, potentially leading to an overestimation of singlet oxygen [27]; (ⅱ) Poor cell permeability: despite efforts to enhance stability of dioxane through the incorporation of adamantane units, the rigidity and increased steric hindrance reduce the cell membrane permeability of Schaap’s dioxetane. This limitation has necessitated modifications such as the incorporation of membrane-penetrating peptides like polyarginine to enhance the cell permeability of the probes [28]; (ⅲ) Short emission wavelengths: conventional chemiluminescent probes for ¹O₂ sensing emit visible light, which has limited tissue penetration capabilities [29,30]. To the best of our knowledge, current singlet oxygen probes cannot resolve all of these challenges simultaneously. Consequently, the development of a self-photooxidation resistant near-infrared (NIR) chemiluminescent (CL) probe for the in vivo ¹O₂ detection is highly demand.

  Herein, we presented two CL probes, DCM-Ad and DCM-Cy, for in vivo ¹O₂ imaging in living cells, viruses, and mice models during PDT. Our "step-by-step" strategy focuses on addressing the inherent limitations of traditional CL probes for ¹O₂ determination and improving their biological applicability (Fig. 1). We hypothesized that reducing the electron cloud density of enolether unit could effectively mitigate the photo-oxidation of chemiluminescent probes. Thus, compared the conventional CL probes, we removed the methoxy group to minimize the photoinduced background signals during PDT. Subsequently, we replaced the rigid and bulky adamantane units by compact cyclobutane building blocks to improve its cell permeability capabilities. In the presence of ¹O₂, the probe DCM-Cy undergo oxidation to form an unstable strained phenol-dioxetane intermediate, which rapidly spontaneous decomposes to generate an excited carbonyl species. This species then decays to its ground state, accompanied by the emission of NIR chemiluminescence. DCM-Cy exhibits remarkable resistance to photo-oxidation with minimal background signal interference, with an approximately 6-fold enhancement in cell permeability capabilities. Consequently, based on our step-by-step molecular design strategy, the probe DCM-Cy possesses the capability to monitor in vivo levels of ¹O₂ in bacteria, living cells, and mice during photodynamic therapy.

  Figure 1

  

  Figure 1.  Schematic illustration of "step-by-step" design strategy for CL probes for ultra-high signal-to-ratio ¹O₂ detection: (ⅰ) removing the methoxy moiety from enolether unit to suppress the potential self-photooxidation reaction; (ⅱ) replacing the rigid and bulky adamantane units with a compact cyclobutane moiety significantly enhanced the cell permeability.

  DownLoad: Full-Size Img PowerPoint

  The general synthesis procedure of DCM-Ad and DCM-Cy are depicted in Scheme S1 (Supporting information). 2-Chloro-3-hydroxybenzaldehyde was chosen to undergo four steps to synthesize key intermediates compounds 1–5 which contains of ethyl phosphonates. In the ¹H NMR spectrum, we observed a characteristic methylene signal (δ = 3.39 ppm, d, J = 16.0 Hz), indicative of the proton attached to ethyl phosphonates. Subsequently, reacted with cyclobutanone or adamantanone to form a double bond, through which phenoxy was attached to adamantane or cyclobutene without methyl acrylate. After two synthetic steps, lastly, such structure backbone was finally attached to an acceptor, dicyanomethylchromone in a yield of about 30%. Characteristic protons on adamantane of DCM-Ad (δ = 1.84–2.01 ppm, m, 12H) and cyclobutene of DCM-Cy (δ = 2.15–3.10 ppm, 6H) were found in the ¹H NMR spectrum. The mass spectra of DCM-Ad and DCM-Cy were also confirmed their molecular weight value of 455.1526 and 375.0898, respectively, completely matching with the calculated value 455.1521 and 375.0895. All the key compounds were characterized by ¹H NMR, ¹³C NMR spectra, and high-resolution mass spectrometry (HRMS).

  With these probes in hand, we firstly evaluated their chemiluminescent response ability toward ¹O₂. The mixed solutions of H₂O₂/NaClO were chosen as the ¹O₂ donor, and the ¹O₂ concentration was almost the equivalent to that of NaClO (the conversion efficiency was near 100%). As shown in Fig. 2A, with the presence of 70 equiv. ¹O₂, the CL signals of DCM-Cy and DCM-Ad exhibited remarkable growth at the primary emission peak around 650 nm, while negligible CL signal was found in the absence of ¹O₂. To investigate the capability of these probes to sense ¹O₂, clinically photosensitizer Rose Bengal (RB) was chosen to produce ¹O₂ upon light irradiation in PDT [31]. In the presence of 10 µmol/L RB, the intensity of DCM-Cy and DCM-Ad increased with the extension of illumination time (LED white light, 96 mW/cm²) and attained a maximum CL signal at 70 s (Fig. 2B and Fig. S1 in Supporting information). Notably, there was a good linear correlation between the CL intensity of these probes and the concentration of RB, indicating that the ¹O₂ generated by RB is intuitively detected and quantified by CL signal changes (Fig. S2 in Supporting information). Furthermore, compared to commercially available probe DCFH, DCM-Cy had a higher signal-to-noise ratio for the detection of ¹O₂ (Fig. S3 in Supporting information). These results indicated that DCM-Cy and DCM-Ad has the capability of ultra-sensitive detection of ¹O₂ levels in aqueous solutions.

  Figure 2

  

  Figure 2.  Chemiluminescent response of DCM-Ad and DCM-Cy towards ¹O₂. Test condition: Tris–HCl buffer solution, containing 4 mg/mL bovine albumin (BSA), pH 7.5. (A) Chemiluminescence spectrum of probes (20 µmol/L) in the absence or presence of 70 equiv. ¹O₂ (H₂O₂/NaClO system). (B) Time dependent CL intensity changes of DCM-Cy mixed with equal RB in presence and absence of light irradiation. (C) Selectivity of DCM-Ad and DCM-Cy towards various ROS, ¹O₂, H₂O₂, ClO⁻, O₂^•−, ONOO⁻, ^•OH (100 µmol/L), in aqueous system. (D) The chemiluminescence of DCM-Ad, DCM-Cy and previous CL probes DCM-Cl-OH in the presence or absence of RB under light exposure (10 µmol/L, 96 mW/cm² white LED light for 60 s). Femtosecond time-resolved transient absorption spectra of 0.1 mmol/L DCM-Cy (E) and DCM-Cl-OH (F) in Tris–HCl solutions (containing 20 mg/mL BSA) pumped at 500 nm. Data are presented as mean ± standard deviation (SD) (n = 3).

  DownLoad: Full-Size Img PowerPoint

  The selectivity of the CL probe is an essential factor and serves as a critical criterion for assessing potential future applications [32,33]. Thus, we assessed the capability of DCM-Cy and DCM-Ad to specifically sense of ¹O₂ with other biological reactive oxygen species (ROS) [34]. As expected, when treated with ¹O₂, an reamarkable enhancement of CL signals was observed (Fig. 2C). However, there were only negligible increases in CL intensity upon the incubation with other ROS (10 equiv. of H₂O₂, ClO⁻, ^•OH, O₂^•−, ONOO⁻, respectively). These results demonstrate the excellent selectivity of DCM-Cy and DCM-Ad toward ¹O₂ over other potential ROS.

  Undesirable photo-oxidation reactions are potential contributor to the background noise signal generated by conventional ¹O₂ detector when exposed to light. Our previous investigations have demonstrated that CL precursors containing electron-rich enolether units are susceptible to self-photooxidation, leading to chemiluminescent emission. In DCM-Cy and DCM-Ad, the methoxy groups were removed from enolether building block to enhance the resistance to self-photooxidation. Comparing with reported DCM-Cl-OH [35], these probes show much low background CL signals under light irradiation (Fig. S4 in Supporting information), suggesting significant improvements in resistance to self-photooxidation.

  In a simulated PDT test, both DCM-Ad and DCM-Cy exhibited ultra-high signal-to-noise ratio (the CL signals ratio in the presence and absence of photosensitizer) of ¹O₂ sensing, with 433- and 466-fold CL signals enhancement, respectively. The distinct change is significantly higher than that of DCM-Cl-OH (Fig. 2D). Moreover, we recorded the CL kinetic profile of DCM-Cy and DCM-Ad, which exhibited a typical glow-type CL behavior: The CL half-time of DCM-Cy and DCM-Ad was about 38 s and 53 s, respectively (Fig. S5 in Supporting information). Thus, the CL signal from these probes remained to be detectable even within 5 min. The above results indicate that DCM-Cy and DCM-Cy possess a high contrast detection potentiality for ¹O₂ during clinical photodynamic therapy.

  To gain insight into the self-photooxidation resistant ability of DCM-Cy and DCM-Ad for ¹O₂ detection in PDT, we investigated the photoinert mechanism of these probes by femtosecond transient spectroscopy (fs-TA) [36]. As shown in Fig. 2E, the TA spectra of DCM-Cy show two distinct bands with time-wavelength dependence. The negative excited state absorption (ESA) signals, observed at wavelengths exceeding 630 nm, is attributed to stimulated emission (SE). On the other hand, the band observed at wavelengths below 530 nm is attributed to excited state absorption. In contrast, the TA spectra of our reported DCM-Cl-OH displayed two positive excited state absorption signals peaking at 545 and 575 nm in the water solution (Fig. 2F and Fig. S6 in Supporting information). The excited state species at the 575 nm should potentially play a crucial role in the photooxidation reaction with oxygen, resulting in a high background CL upon light irradiation. The elimination of methoxy groups from the enolether units effectively prevents the formation of similar excited state species in the probes, thereby signally reducing the background signal.

  The aforementioned experiments demonstrated the high selectivity, and minimized background signals of DCM-Ad and DCM-Cy, encouraging us to further investigate the ability of these probes for intracellular ¹O₂. Prior to employing these probes in living cells, it was imperative to validate their biocompatibility. The cytotoxicity of the probes towards HeLa cells was assessed using the methyl thiazolyl tetrazolium (MTT) assay. After incubation with different concentrations of DCM-Ad and DCM-Cy for 24 h, the cell survival rates were estimated to be over 79% and 81%, respectively, indicating their favorable biocompatibility. Additionally, an extra irradiation process (96 mW/cm², 5 min) is added after incubation with probes for 2 h to evaluate their phototoxicity. After 24 h of incubation, the cell survival rates were more than 75%, demonstrating the negligible toxicity of the probes exposed under light during PDT (Fig. S7 in Supporting information).

  Previous studies have demonstrated that the poor permeability of CL probes is one of the major bottlenecks impeding intracellular probe in vivo sensing. Specifically, the rigid and bulky adamantane units are identified as potential factors resulting into poor permeability in these conventional reported CL probes [30,37]. Hence, we suppose that the presence of a compact cyclobutane building block could greatly improve cell permeability of the probes. Subsequently, the confocal microscopy was employed to validate the cell permeability of the DCM-Ad and DCM-Cy. DCM-Ad and DCM-Cy exhibit fluorescent signals in the near-infrared under 500 nm excitation light (Fig. S8 and Table S1 in Supporting information). As shown in Fig. 3A, after incubation with DCM-Cy for 1 min, a distinct NIR fluorescent signal at 650 nm was observed in the HeLa cells, as well as the intracellular fluorescence signal reaching maximum at approximately 4 min. By contrast, DCM-Ad required nearly 25 min to reach the maximum intracellular fluorescence intensity, which is almost six times longer than that of required by DCM-Cy (Fig. 3B). Additionally, the maximum intracellular brightness of DCM-Cy was about 10% higher than that of DCM-Ad, indicating a superior potential for intracellular sensing (Fig. 3C). These results suggest that DCM-Cy not only showed a dramatic enhancement in cell permeability but also an increase in cellular uptake.

  Figure 3

  

  Figure 3.  Confocal Microscopy imaging the cellular internalization of DCM-Cy and DCM-Ad. (A) Fluorescence imaging of DCM-Cy and DCM-Ad in HeLa cells over time. Scale bar: 30 µm. (B) Quantification of intracellular fluorescence signals with different incubation times. (C) Comparison of intracellular fluorescence intensity of DCM-Ad and DCM-Cy at equilibrium state. Error bars represent standard deviations (n = 6).

  DownLoad: Full-Size Img PowerPoint

  Then, we evaluated the intracellular ¹O₂ sensing capabilities of these probes during photodynamic therapy. Given its excellent cell permeability and cellular uptake capability, DCM-Cy was chosen for in vitro biological testing. HeLa cells were incubated with DCM-Cy and RB for 2 h. Subsequently, these cells were irradiated by white LED light for 3 min and then their chemiluminescence signals was captured by ImageQuant LAS4000 system. As shown in Fig. 4A, the cells that incubated with DCM-Cy and RB, followed by light irradiation, exhibited a pronounced CL signals. Conversely, cells that were merely incubated with DCM-Cy (lacking ¹O₂ donor) or RB (lacking ¹O₂ probe), did not produce any detectable chemiluminescence signal upon light irradiation. Furthermore, cells that were maintained in a dark environment did not exhibit any chemiluminescence signals (Fig. 4B).

  Figure 4

  

  Figure 4.  Monitoring of endogenous ¹O₂ in HeLa cells and Staphylococcus aureus. (A) Chemiluminescence images acquired in HeLa cells after incubation with RB, DCM-Cy (10 µmol/L) alone, or simultaneously for 2 h upon exposure to LED light or in darkness. (B) Quantification of the CL intensity in figure A. (C) Confocal fluorescence imaging of DCM-Ad and DCM-Cy in Staphylococcus aureus and the quantitative fluorescence intensity. Scale bar: 30 µm. (D) Staphylococcus aureus was co-incubated with DCM-Cy and RB for 1 h, and then the chemiluminescence intensity was measured with or without white LED light irradiation. Insert: CL imaging of DCM-Cy in Staphylococcus aureus. Error bars represent standard deviations (n = 3).

  DownLoad: Full-Size Img PowerPoint

  Eradicating pathogenic bacteria or inhibiting bacterial proliferation represents a crucial aspect of photodynamic therapy [38-43]. Hence, we further explored the capacity of these probes to detect the endogenous ¹O₂ in bacteria. Initially, we investigated the bacterial internalization capabilities of these probes using fluorescence confocal microscopy. As shown in Fig. 4C, remarkable fluorescent signals were observed in Staphylococcus aureus following incubation with DCM-Ad or DCM-Cy, indicating efficient uptake of the probes by the bacteria. Subsequently, we tested whether these probes can detect ¹O₂ produced in bacteria during the PDT process. Upon exposure to LED light, bright chemiluminescence signals were detected in Staphylococcus aureus following co-incubation with DCM-Cy and RB for 1 h (Fig. 4D), in which the chemiluminescent intensity was approximately 10-fold higher than that of the control group without light irradiation. These results provide solid evidence for the ability of DCM-Cy to detect endogenous ¹O₂ in living cells and pathogenic bacteria produced during the PDT process.

  Encouraged by the excellent capacity of DCM-Cy to image intercellular ¹O₂, we expect utilization DCM-Cy to image ¹O₂ production during PDT in vivo. All animal experiments were performed according to the guidelines of the Care and Use of Laboratory Animals formulated by the Ministry of Science and Technology of China and were approved by the Animal Care and Use Committee of East China University of Science and Technology (2021–07, 001). The nude mice bearing HeLa xenograft tumor were selected as in-vivo animal models. After Intratumoral injection of the DCM-Cy or a mixture of DCM-Cy and RB for 30 min, and subsequent exposure to white light for 1 min, the chemiluminescence signals were immediately acquired by an IVIS living imaging system. As shown in Fig. 5A, the conventional chemiluminescent probe, DCM-Cl-OH, exhibited a noticeable background signal caused by self-photooxidation even in the absence of the photosensitizer. Upon the addition of the photosensitizer, the chemiluminescence signal was merely enhanced 4.4-fold. In contrast, the DCM-Cy group displayed minimal background signals in the absence of the photosensitizer, which is consistent with solution test results (Fig. 5B). Consequently, a remarkable 95-fold increase in signal-to-noise ratio was achieved following the introduction of the photosensitizer. Collectively, all these tests indicate that DCM-Cy can be fully applied to cellular, bacterial, and even in vivo models, providing precise sensing information of ¹O₂ for the visualization of PDT treatment.

  Figure 5

  

  Figure 5.  In vivo imaging of intratumor ¹O₂ during PDT. (A) CL imaging of tumor bearing mice after intratumor injection of DCM-Cl-OH and DCM-Cy with or without RB after 1 min irradiation. (B) Quantification of the chemiluminescence intensity at the same condition. Error bars represent standard deviations (n = 3).

  DownLoad: Full-Size Img PowerPoint

  In summary, we have successfully developed an ultra-high signal-to-ratio near-infrared chemiluminescent probe DCM-Cy for high contrast monitoring ¹O₂ during photodynamic therapy. By removing the methoxy unit to reduce the electron cloud density of the enolether unit, these CL probes can effectively suppress the self-photooxidation reaction, thereby minimizing the photoreaction induced background signals. Additionally, the substitution of the rigid and bulky adamantane units with a compact cyclobutane moiety significantly enhanced the cell permeability of the probe by approximately 6-fold. Consequently, DCM-Cy enables highly sensitive detection of endogenous ¹O₂ in both living cells and pathogenic bacteria. Specifically, DCM-Cy demonstrated 95-fold ultra-high signal-to-noise imaging of intratumor ¹O₂ in living mice. Indeed, our "step-by-step" strategy addressed the limitations of traditional CL probes for ¹O₂ determination and improving their biological applicability. We anticipate this straightforward design strategy will provide new insight into detecting of in vivo ¹O₂ and pave the way for high-performance chemiluminescent probes.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Meiling Zhao: Writing – original draft, Software, Investigation, Formal analysis, Data curation, Conceptualization. Yao Lu: Validation, Formal analysis, Data curation, Conceptualization. Yutao Zhang: Writing – review & editing, Writing – original draft, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Haoyun Xue: Validation, Methodology, Investigation. Zhiqian Guo: Writing – review & editing, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

  Acknowledgments

  This work was supported by National Natural Science Foundation of China (Nos. 32121005, 22225805, 22308101, and 32394001), Shanghai Science and Technology Innovation Action Plan (No. 23J21901600), Innovation Program of Shanghai Municipal Education Commission, Shanghai Frontier Science Research Base of Optogenetic Techniques for Cell Metabolism (Shanghai Municipal Education Commission, No. 2021 Sci & Tech 03–28), the China Postdoctoral Science Foundation (No. 2021M701199), Natural Science Foundation of Shanghai (No. 23ZR1416600).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110105.

  
  
• 1. [1]

     M. Garcia-Diaz, Y.Y. Huang, M.R. Hamblin, Methods 109 (2016) 158–166. doi: 10.1016/j.ymeth.2016.06.025

  2. [2]

     Y. Wang, Y. Lin, S. He, et al., J. Hazard. Mater. 461 (2024) 132538. doi: 10.1016/j.jhazmat.2023.132538

  3. [3]

     Z.Y. Jin, H. Fatima, Y. Zhang, et al., Adv. Ther. 5 (2022) 2100176. doi: 10.1002/adtp.202100176

  4. [4]

     H. Kim, M. Yang, N. Kwon, et al., Bull. Korean Chem. Soc. 44 (2023) 236–255. doi: 10.1002/bkcs.12655

  5. [5]

     X. Li, J.F. Lovell, J. Yoon, et al., Nat. Rev. Clin. Oncol. 17 (2020) 657–674. doi: 10.1038/s41571-020-0410-2

  6. [6]

     J.L. Wang, Y. Chen, J.X. Song, et al., Adv. Funct. Mater. 34 (2024) 2312162. doi: 10.1002/adfm.202312162

  7. [7]

     G.Q. Jin, C.V. Chau, J.F. Arambula, et al., Chem. Soc. Rev. 51 (2022) 6177–6209. doi: 10.1039/d2cs00275b

  8. [8]

     H. Wei, X. Li, F. Huang, et al., Chin. Chem. Lett. 34 (2023) 108564. doi: 10.1016/j.cclet.2023.108564

  9. [9]

     E. Kozma, M. Bojtár, P. Kele, Angew. Chem. Int. Ed. 62 (2023) e202303198. doi: 10.1002/anie.202303198

  10. [10]

      Y. Jiang, Z. Zeng, J. Yao, et al., Chin. Chem. Lett. 34 (2023) 107966. doi: 10.1016/j.cclet.2022.107966

  11. [11]

      S. Hackbarth, W. Islam, J. Fang, et al., Photochem. Photobiol. Sci. 18 (2019) 1304–1314. doi: 10.1039/c8pp00570b

  12. [12]

      H. Wang, Z. Wang, Y. Li, et al., Small 15 (2019) 1902185. doi: 10.1002/smll.201902185

  13. [13]

      A. Looft, M. Pfitzner, A. Preuß, et al., Photodiagn. Photodyn. Ther. 23 (2018) 325–330. doi: 10.1016/j.pdpdt.2018.07.006

  14. [14]

      K. Murotomi, A. Umeno, M. Shichiri, et al., Int. J. Mol. Sci. 24 (2023) 2739. doi: 10.3390/ijms24032739

  15. [15]

      F. Zou, W. Zhou, W. Guan, et al., Anal. Chem. 88 (2016) 9707–9713. doi: 10.1021/acs.analchem.6b02611

  16. [16]

      Y. Lang, S. Wu, Q. Yang, et al., Anal. Chem. 93 (2021) 9737–9743. doi: 10.1021/acs.analchem.1c01160

  17. [17]

      S. Pfitzner, J.C. Schlothauer, B. Röder, et al., Laser Phys. 28 (2018) 085702. doi: 10.1088/1555-6611/aac5cc

  18. [18]

      J.T. Ping, J.L. Qin, C. Zhou, et al., ChemNanoMat 6 (2020) 173-173. doi: 10.1002/cnma.201900655

  19. [19]

      A. Prasad, M. Sedlářová, P. Pospíšil, Sci. Rep. 8 (2018) 13685. doi: 10.1038/s41598-018-31638-5

  20. [20]

      T. Entradas, S. Waldron, M. Volk, J. Photochem. Photobiol. B 204 (2020) 111787. doi: 10.1016/j.jphotobiol.2020.111787

  21. [21]

      H. Shen, F. Sun, X. Zhu, et al., J. Am. Chem. Soc 144 (2022) 15391–15402. doi: 10.1021/jacs.2c07443

  22. [22]

      C. Chen, H. Gao, H. Ou, et al., J. Am. Chem. Soc 144 (2022) 3429–3441. doi: 10.1021/jacs.1c11455

  23. [23]

      X. Duan, G.Q. Zhang, S. Ji, et al., Angew. Chem. Int. Ed. 61 (2022) e202116174. doi: 10.1002/anie.202116174

  24. [24]

      Q. Xiao, C. Xu, TrAC, Trends Anal. Chem. 124 (2020) 115780. doi: 10.1016/j.trac.2019.115780

  25. [25]

      O. Green, T. Eilon, N. Hananya, et al., ACS Cent. Sci. 3 (2017) 349–358. doi: 10.1021/acscentsci.7b00058

  26. [26]

      S. Joseph, S.K. Ashok Kumar, Coord. Chem. Rev. 496 (2023) 215408. doi: 10.1016/j.ccr.2023.215408

  27. [27]

      Y. Zhang, C. Yan, C. Wang, et al., Angew. Chem. Int. Ed. 59 (2020) 9059–9066. doi: 10.1002/anie.202000165

  28. [28]

      R. Blau, O. Shelef, D. Shabat, et al., Nat. Rev. Bioeng. 1 (2023) 648–664. doi: 10.1038/s44222-023-00074-0

  29. [29]

      M. Yang, J. Huang, J. Fan, et al., Chem. Soc. Rev. 49 (2020) 6800–6815. doi: 10.1039/d0cs00348d

  30. [30]

      N. Hananya, O. Green, R. Blau, et al., Angew. Chem. Int. Ed. 56 (2017) 11793–11796. doi: 10.1002/anie.201705803

  31. [31]

      J.C. Peterson, E. Arrieta, M. Ruggeri, et al., Biomed. Opt. Express 12 (2021) 272–287. doi: 10.1364/boe.405601

  32. [32]

      Y. Wen, Z. Long, F. Huo, et al., Org. Chem. Front. 8 (2021) 1302–1314. doi: 10.1039/d0qo01260b

  33. [33]

      M. Fadel, D.A. Fadeel, A. Tawfik, et al., J. Drug Deliv. Sci. Technol. 69 (2022) 103095. doi: 10.1016/j.jddst.2022.103095

  34. [34]

      K.C. Yan, A.C. Sedgwick, Y. Zang, et al., Small Methods 3 (2019) 1900013. doi: 10.1002/smtd.201900013

  35. [35]

      M. Yang, J. Zhang, D. Shabat, et al., ACS Sens. 5 (2020) 3158–3164. doi: 10.1021/acssensors.0c01291

  36. [36]

      P.A. Rose, J.J. Krich, J. Phys. Chem. Lett. 14 (2023) 10849–10855. doi: 10.1021/acs.jpclett.3c02491

  37. [37]

      M.E. Roth-Konforti, C.R. Bauer, D. Shabat, Angew. Chem. Int. Ed. 56 (2017) 15633–15638. doi: 10.1002/anie.201709347

  38. [38]

      X. He, S. Koo, E. Obeng, et al., Coord. Chem. Rev. 492 (2023) 215275. doi: 10.1016/j.ccr.2023.215275

  39. [39]

      M. Hirose, Y. Yoshida, K. Horii, et al., Arch. Oral Biol. 122 (2021) 105024. doi: 10.1016/j.archoralbio.2020.105024

  40. [40]

      S. Sabbahi, L. Ben Ayed, M. Jemli, Appl. Water Sci. 8 (2018) 56.

  41. [41]

      F. Hu, G. Qi, Kenry, et al., Angew. Chem. Int. Ed. 59 (2020) 9288–9292. doi: 10.1002/anie.201910187

  42. [42]

      Q. Jia, Q. Song, P. Li, et al., Adv. Healthc. Mater. 8 (2019) 1900608. doi: 10.1002/adhm.201900608

  43. [43]

      Q. Zhao, G. Qing, J. Yu, et al., Chin. Chem. Lett. 35 (2024) 108535. doi: 10.1016/j.cclet.2023.108535

• 

  Figure 1  Schematic illustration of "step-by-step" design strategy for CL probes for ultra-high signal-to-ratio ¹O₂ detection: (ⅰ) removing the methoxy moiety from enolether unit to suppress the potential self-photooxidation reaction; (ⅱ) replacing the rigid and bulky adamantane units with a compact cyclobutane moiety significantly enhanced the cell permeability.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Chemiluminescent response of DCM-Ad and DCM-Cy towards ¹O₂. Test condition: Tris–HCl buffer solution, containing 4 mg/mL bovine albumin (BSA), pH 7.5. (A) Chemiluminescence spectrum of probes (20 µmol/L) in the absence or presence of 70 equiv. ¹O₂ (H₂O₂/NaClO system). (B) Time dependent CL intensity changes of DCM-Cy mixed with equal RB in presence and absence of light irradiation. (C) Selectivity of DCM-Ad and DCM-Cy towards various ROS, ¹O₂, H₂O₂, ClO⁻, O₂^•−, ONOO⁻, ^•OH (100 µmol/L), in aqueous system. (D) The chemiluminescence of DCM-Ad, DCM-Cy and previous CL probes DCM-Cl-OH in the presence or absence of RB under light exposure (10 µmol/L, 96 mW/cm² white LED light for 60 s). Femtosecond time-resolved transient absorption spectra of 0.1 mmol/L DCM-Cy (E) and DCM-Cl-OH (F) in Tris–HCl solutions (containing 20 mg/mL BSA) pumped at 500 nm. Data are presented as mean ± standard deviation (SD) (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Confocal Microscopy imaging the cellular internalization of DCM-Cy and DCM-Ad. (A) Fluorescence imaging of DCM-Cy and DCM-Ad in HeLa cells over time. Scale bar: 30 µm. (B) Quantification of intracellular fluorescence signals with different incubation times. (C) Comparison of intracellular fluorescence intensity of DCM-Ad and DCM-Cy at equilibrium state. Error bars represent standard deviations (n = 6).

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Monitoring of endogenous ¹O₂ in HeLa cells and Staphylococcus aureus. (A) Chemiluminescence images acquired in HeLa cells after incubation with RB, DCM-Cy (10 µmol/L) alone, or simultaneously for 2 h upon exposure to LED light or in darkness. (B) Quantification of the CL intensity in figure A. (C) Confocal fluorescence imaging of DCM-Ad and DCM-Cy in Staphylococcus aureus and the quantitative fluorescence intensity. Scale bar: 30 µm. (D) Staphylococcus aureus was co-incubated with DCM-Cy and RB for 1 h, and then the chemiluminescence intensity was measured with or without white LED light irradiation. Insert: CL imaging of DCM-Cy in Staphylococcus aureus. Error bars represent standard deviations (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 5  In vivo imaging of intratumor ¹O₂ during PDT. (A) CL imaging of tumor bearing mice after intratumor injection of DCM-Cl-OH and DCM-Cy with or without RB after 1 min irradiation. (B) Quantification of the chemiluminescence intensity at the same condition. Error bars represent standard deviations (n = 3).

  下载: 全尺寸图片 幻灯片
•