Reactive oxygen species (ROS) are highly reactive and short-lived chemical species in living organism, which play a key role in linking fundamental chemistry to intricate physiological and pathological processes^{[1⇓⇓⇓-5]}. Cancer cells are commonly identified with increased ROS level compared to normal cells. Mounting evidences suggest that such intrinsic ROS stress of cancer cells is highly associated with tumor progression^[6⇓⇓-9]. However, ROS beyond the threshold level, can also act as an anticancer agent to break the cellular redox homeostasis, resulting in apoptosis^[10-11]. Thus, ROS is a double-edged sword for cancer cells. Thereinto, HClO, as one of the most important ROS, can be endogenously generated from hydrogen peroxide and chloride ions with the catalysis of myeloperoxidase enzyme^[12]. Generally, HClO is mainly produced by immune cells (e.g., neutrophils, macrophages) as immune defense against invasive pathogens and bacteria. As is known, the immune cells are somehow disabled of its surveillance feature while tumorigenesis happens, and further turn to aid tumor progression while remaining in the tumor microenvironment^[13]. Surprisingly, they can still be reactivated of its antitumor effect, known as the popular cancer immunotherapy^[14]. Thus, the specific role of HClO in tumor biology or immunotherapy is believed to be vital yet complicated and needs further exploration, which, however, is still in its infancy.

Real-time detection of HClO fluctuation in cancer cells is considered to be the prerequisite for studying its biological effect in malignant tumors. Due to high spatiotemporal resolution of fluorescence imaging, a variety of organic fluorescent probes have been developed for selective imaging of HClO^{[15⇓⇓-18]}, most of which are those molecular probes containing HClO-specific redox group (e.g., C=C double bond, chalcogenide, aldoxime group). The specific HClO-mediated oxidation reaction can lead to fluorescence signal alterations of such molecular probes, including “Turn On”, “Turn Off”, “ratiometric”, etc., thus rendering highly sensitive detection or imaging of HClO. However, the construction of these organic probes is generally cumbersome and the existence of carbocyclic structure endows them with poor water solubility, resulting in inhomogeneous distribution in live cells and incomplete reaction with HClO. In addition, these small molecular organic probes also suffer from cell leakage and need binding with larger substrate to avoid efflux^[19]. To the best of our knowledge, it is rarely explored of using hydrophilic nanoprobe for fluorescence imaging of HClO via the transition metal-mediated redox reaction, which could present a great potential to circumvent the intrinsic drawbacks of organic probes. Prussian Blue (PB), as a typical metal organic frame material, shows excellent performances in wastewater treatment^[20], energy storage^[21], tumor treatment^[22], sensor^[23], and other aspects. Recently, PB stands out in biosensing application^[24⇓-26], owing to its low toxicity, unique framework structure and easy to be synthesized. For example, a high performance fluorescence switching system triggered electrochemically by Prussian Blue was reported for detecting sulfite^[27]. However, there were no more reports for the fluorescent probes based on Prussian Blue for detecting HClO as yet.

Herein, a novel inorganic-organic composite nanoprobe, namely, FITC@HMPB (F@H), was designed by simply loading organic dye into the pores of hollow mesoporous Prussian Blue nanoparticles (HMPB), which presents good dispersion stability without complicated molecular design and covalent attachment in the composite nanoprobe, effectively avoiding disadvantages of pure organic dye probes. Specifically, the HMPB was prepared by polyvinylpyrrolidone (PVP) as protective agent and hydrochloric acid (HCl) as etching agent. The obtained F@H exhibits excellent detection capability of HClO, based on the inner filter effect^{[28⇓⇓⇓-32]}. More details, the excitation or emission light can be absorbed by absorbers when the absorbance spectra of absorbers overlaps with the fluorescence excitation or emission spectra of fluorophores. Owing to the inner filter effect, the fluorescence of FITC within HMPB quenched without HClO. Interestingly, the fluorescence signals could recover accompanied by the degradation of HMPB in the presence of HClO, enabling the real-time detection of HClO in the cancer cells (Fig. 1).

All chemicals including potassium ferricyanide (K₃[Fe(CN)₆]), polyvinyl pyrrolidone (PVP K30), hydrochloric acid (HCl, 12 mol/L), fluorescein isothiocyanate isomer (FITC), sodium hypochlorite (NaClO, 4%), hydrogen peroxide (H₂O₂, 30%), tert-butyl hydroperoxide (TBHP, 70%) were used without further purification. All reagents mentioned above were purchased from Sinopharm Chemical Reagent Co.. 2,2′-azobis-2-methyl-propanimidamide (AAPH) was obtained from Aladdin, and sodium nitroferricyanide dihydrate (SNP) was purchased from Rhawn. 4T1 cells (mouse breast cancer cells) were obtained from National Collection of Authenticated Cell Culture. Distilled water was used for all experiments.

X-ray diffraction (XRD) patterns of all samples were collected in the range 5°-90°(2θ) using a Rigaku D/max 2550 diffractometer (CuKα radiation, λ=0.15406 nm), operated at 40 kV and 40 mA. Transmission electron microscopy (TEM) images were acquired on a JEM-2100F electron microscope. Dynamic light scattering measurement was conducted on Malvern Zetasizer Nano ZS model ZEN3600 (Worcestershire, UK). UV-Vis-NIR spectra were recorded on a UV-3101PC shimadzu spectroscope. X-ray photoelectron spectrum (XPS) was obtained on a Thermo ESCALAB 250Xi X-ray photoelectron spectrometer. The fluorescence spectra were recorded on a Shimadzu RF-5301 PC spectrophotometer. The CLSM images were collected by using FV 1000 system (Olympus Company, Japan).

Prussian Blue nanoparticles and hollow mesoporous Prussian Blue nanoparticles were synthesized via a modified hydrothermal method according to our previous report ^[33]. Afterwards, 5 mg of HMPB were dispersed in 5 mL distilled water via sonication. Then, 1 mg of FITC was added in the above HMPB dispersion, and the obtained mixture was placed in the incubating shaker at a constant temperature of 37 ℃ overnight. Afterwards, the FITC@HMPB (F@H) was obtained via centrifugation and washed with distilled water once. Finally, the obtained F@H was redispersed in 5 mL distilled water for further use.

The F@H was dispersed in Tris-HCl buffer (pH 5.5) to obtain the stock suspension. Then the HClO was added into the stock solution to gain 2 mL of mixed solution, which contains 50 μg/mL of F@H and different concentrations of HClO (0-50 μmol/L). After incubation for 2 min, fluorescence spectra were obtained with excitation wavelength of arround 488 nm. All procedures were conducted at room temperature.

4T1 cells were seeded into a 96-well culture plate at a density of 5000 cells/well and left to adherence in a cell culture incubator for 12 h. The as-prepared F@H with various concentrations (0, 12.5, 25, 50, 100 μg/mL, n=6) in cell culture medium DMEM containing 10% fetal bovine serum (FBS) were added into the above culture plate. The corresponding cell viability was measured by a standard CCK-8 reduction assay (Cell Counting Kit, Dojindo Laboratories, Kumamoto, Japan).

4T1 cells were cultured on the 6-well cell culture plate with 14 mm glass coverslips and allowed to incubate for 24 h, followed by washing with phosphate buffer solution (PBS) three times. After that, cells in blank group were incubated in DMEM without serum while cells in negative control and experimental groups were incubated with F@H (50 μg/mL) in DMEM without serum. All groups were incubated for 2 h at 37 ℃ in a 5%CO₂/95% air incubator in this step. After incubation, cells were washed three times with PBS. For detecting exogenous HClO, the experimental group was treated with 100 μmol/L NaClO for 0.5  h in DMEM without serum at 37 ℃, while the other groups were treated by DMEM without serum and NaClO. For detecting endogenous HClO, the experimental group was treated with 50 μmol/L elescolmol in DMEM without serum at 37 ℃ while the other groups were treated by DMEM without serum and elescolmol for 0.5 h.

TEM image (Fig. 2(a)) shows that the prepared HMPB displays high dispersion with average size of arround 160 nm. The dynamic light scattering result indicates that the suspension containing 50 μg/mL of HMPB in Tris-HCl keeps high dispersion with narrow size distribution of 190 nm in diameter even after 12 h (Fig. S1), indicating that HMPB is of great dispersion stability. XRD pattern (Fig. 2(b)) confirms the single crystal structure of HMPB with good crystallinity. N₂ adsorption-desorption analysis result (the inset of Fig. 2(c)) indicates that the specific surface area and pore volume of HMPB are 124 m²/g and 0.56 cm³/g, respectively, with a homogeneous mesopore of 3.2 nm. This hollow mesoporous structure endows HMPB with high loading efficiency. More details, the maximum loading amount of FITC is 14.4% and encapsulation rate is 83.9% (Fig. S2 and Table S1). Notably, the loading amount of FITC could be controlled via different feed ratios, nevertheless, the loading amount of FITC affects little to the sensitivity or stability of the fluorescent probe.

UV-Vis spectra shown in the Fig.2 (d) indicate that free FITC presents an obvious adsorption peak at 488 nm, and the absorbance peak at around 750 nm is ascribed to HMPB. It is interesting to find that the obtained F@H shows the same absorbance peak at 750 nm, meanwhile, the intensity at 488 nm, was obviously enhanced, implying the successful loading of FITC into HMPB. It is worth noting that the fluorescence intensity at emission peak (520 nm) of FITC sharply decreased after loading into HMPB (Fig. 2(e)). To explore this interesting phenomenon, the fluorescence life was tested. As shown in Fig. 2(f), F@H shows almost the same fluorescence lifetime as FITC, confirming the existence of inner filter effect between HMPB and FITC.

Afterwards, the in vitro detection capability of the probe in aqueous solution was studied. As shown in Fig. 3(a), it is clearly found that the fluorescence intensity at emission peak of 520 nm gradually increases with the addition of HClO to F@H suspension. The detection limit could be calculated to be 2.01 μmol/L according to the linear relationship between fluorescence intensity at 520 nm and HClO concentration within the range of 0 to 50 μmol/L (Fig. 3(b)). Fig. 3(c) shows the change of absorbance of F@H before and after addition of NaClO, indicating that HMPB could react with HClO within 1 min, in accordance with the fluorescence responsive time (within 2 min, Fig. S3), suggesting quick response and high sensitivity of the obtained probe. Furthermore, the electronic states and chemical composition of HMPB before and after addition of NaClO were measured by X-ray photoelectron spectroscope (XPS). As shown in Fig. 3(d), the peaks located at 708.6 and 709.6 eV are assigned to the binding energy of Fe2p_3/2 of HMPB with the valence state of Fe²⁺ and Fe³⁺, respectively. It is noting to find that after addition of NaClO, the ratio Fe²⁺/Fe³⁺ of peak area decreases from 1 to 0.72, which results from that HMPB is oxidized by ClO^-.

Furthermore, the specificity of F@H probe for HClO under the same analytical conditions was tested. As shown in Fig. 3(a), it is found that the fluorescence of F@H presents negligible changes in the presence of other ROS (TBHP, ROO·, NO, H₂O₂, ·OH, ONOO^-), even at high concentration (500 μmol/L). Moreover, the UV-Vis spectra (Fig. 3(b)) also indicate that HMPB could not react with other ROS species. Thus the obtained F@H nanoprobe exhibits outstanding selectivity to HClO.

Before investigation on the applicability of the prepared F@H probe to detect HClO in living cells, the cytotoxicity of F@H was assessed by 2-(2-methoxy-4- nitrophenyl)-3-(4-nitrophenyl)-5-(2, 4-disulfonic acid benzene)-2h-tetrazole monosodium salt (CCK-8) analysis in 4T1 cells. The results indicate that the prepared F@H shows good cytocompatibility with almost no toxicity (Fig. S4).

Afterwards, the confocal fluorescence imaging of different groups were conducted in cellular level for comparison. As shown in Fig. 5, for the control group, i.e., F@H-pretreated 4T1 cells without addition of NaClO or elesclomol, the 4T1 cells incubated with F@H (50 μg/mL) for 2.5 h showed weak fluorescence under excitation. In contrast, for the NaClO group, after further added NaClO (100 μmol/L) into F@H and co-cultured with 4T1 cells for another 30 min, much enhanced fluorescence was clearly observed, indicating that the F@H probe was capable of detecting exogenous HClO in living cells.

To confirm the capability detection of endogenous HClO, elesclomol, an experimental ROS-generating anticancer agent^[34], was used as a model drug to induce the endogenous production of HClO in 4T1 cells in this study. In the elesclomol group, the F@H-pretreated 4T1 cells were co-incubated with 50 μmol/L elesclomol, and the fluorescence imaging was recorded at 0.5 h (Fig. 5(c, d, g-i)). It is found that the fluorescence signal of elesclomol group shows significant increase at 30 min, which is ascribed to the endogenous generation of HClO in 4T1 cells induced by elesclomol. This result also indicates that the F@H can be used to monitor the cellular oxidative stress.

In brief, a novel fluorescent probe based on hollow mesoporous Prussian Blue nanoparticles was developed via a facile method without complex fluorescent molecular design, for the first time, to achieve the rapid and specific detection of HClO with high sensitivity. More details, FITC molecular confined inside the mesoporous channel of HMPB induces the fluorescence quenching owing to the inner filter effect. Significantly, the fluorescence signal of FITC can be restored accompanied by degradation of HMPB in the presence of HClO. The obtained F@H presents a high signal-to-noise ratio with response time within 2 min and the detection limit is calcaulted to be 2.01 μmol/L. More importantly, it can be further used to detect the endogenous HClO of tumor cells and monitor the elevation of HClO in 4T1 cells induced by elesclomol. Overall, this study demonstrated a new fluorescence detection method using inorganic nanoparticles as responsive carrier, providing a promising detection strategy for HClO in tumor tissue.

Supporting Materials related to this article can be found at https://doi.org/10.15541/jim20220119.