The chemicals involved in this study were shown as follows: cetyltrimethylammonium chloride (CTAC), manganese chloride tetrahydrate, triethylamine (MnCl₂·4H₂O), anhydrous toluene, cyclohexane, and tetraethoxysilane (TEOS) (Sinopharm Chemical Reagent Co. Ltd., Shanghai, China); TEOS and dopamine hydrochloride (Aladdin, Shanghai, China); and Ce6 (Frontiersci, Logan, UT, USA).

The prepared MSN-Ce6 was mixed with dopamine hydrochloride (3 mg) and thereafter added into Tris solution (2.5 ml, pH 8.5) and stirred overnight. The mixture was chelated into MnCl₂·4H₂O solution for agitation for 4 h at 40°C. Subsequently, with the mixture rinsed in deionized water and centrifuged thrice (13,000 rpm, 30 min, 4°C). MSN-Ce6@PDA (Mn) (monocalcium phosphate monohydrate nanoparticles (MCPM NPs)) obtained were freeze-dried and stored at 4°C for subsequent experimentation.

The prepared MCPM NPs were physiochemically characterized as the following aspects. With respect to morphology and structure, the nanoparticles were imaged by transmission electron microscopy (TEM, FEI Tecnai 20) and a field emission scanning electron microscopy (Tescan MAIAA3). The Fourier transform infrared (FT-IR) analysis was conducted on a Piketech FTIR spectrometer. The UV-vis absorbance spectra were obtained on NanoDrop, and the fluorescence imaging excitation spectra were acquired on a PerkinElmer spectrophotometer. The zeta potential and size distribution measurements were performed on a Malvern Zetasizer Nano ZS system. The nitrogen adsorption curves were tested on a physical adsorption meter (ASAP 2460). The energy-dispersive spectroscopy (EDS) and X-ray photoelectron spectroscopy (XPS) spectra were determined on an X-ray photoelectron spectrometer (Thermo ESCALAB 250Xi). As for MR relaxivity assessment, several samples of MCPM NP solutions were prepared with an escalation of Mn concentrations in PBS (0, 0.0244, 0.0489, 0.0897, 0.1783, 0.3566, 0.7132 mM). T₁-weighted phantom images and T₁ relaxation duration were obtained with a 3.0-T GE Discovery 750 W RM system. The r₁ relaxivity was thus calculated from the linear slope obtained under a spin-echo sequence and the following parameters: TR = 425 ms, TE = min full, TI = 28.0, matrix size = 328 × 224, field of view = 24 cm × 24 cm, slice thickness = 3.0 mm, slice gap = 1.5 mm.

The photothermal conversion of MCPM NPs was evaluated via a continuous-wave fiber-coupled diode laser (808 nm, FC-808-10W, Changchun New Industries Optoelectronics Technology Co. Ltd., Changchun, China). Aqueous dispersions of MCPM NPs (1 ml) at several concentrations (0, 0.03, 0.1, 0.3, 0.5 mg ml⁻¹) were loaded in a quartz cuvette and irradiated with an 808-nm laser (1 W cm⁻¹) for 10 min. In parallel, the exposure of MCPM NPs (0.3 mg ml⁻¹) to the 808-nm laser was at graded power densities (0.25, 0.5, 1, 1.5 W cm⁻¹), with the temperature range recorded with a FLIR thermal camera at an interval of 30 s. As for the photostability and photothermal stability, MCPM NP dispersion (0.3 mg ml⁻¹) also underwent exposure to the 808-nm laser (1 W cm⁻¹). Thereafter, the hydrodynamic sizes of MCPM NPs with or without laser irradiation were measured in a 7-day period. In addition, with the 808-nm laser powered on and off, the temperature of samples gradually declined and restored. For the evaluation of the photothermal stability of MCPM NPs, eight on/off cycles were conducted with the laser turned on and off interchangeably.

The singlet oxygen behavior of MCPM NPs was assessed via a continuous-wave laser (660 nm, Wuhan Boji Century Technology Co., Ltd., Wuhan, China). The irradiation power was 100 mW cm⁻² (10 min), and the 1,3-diphenylisobenzofuran (DPBF) dye was employed for the detection of singlet oxygen. In the typical experiment, 30 μl of the DPBF (10 mM) was added to the MCPM NPs (0.3 mg ml⁻¹, 0.1 ml). The aqueous dispersion of MCPM NPs (2 ml) was loaded in a quartz cuvette and treated with different lasers (no laser, 808 nm laser, 660 nm laser, 808/660 nm laser). The fluorescence absorbance value of the remaining DPBF was detected by a microplate reader (Thermo Scientific, Waltham, MA, USA). Over the course of 10 min of light exposure, we collected values every minute.

The in vitro biocompatibility tests were performed using Cell Counting Kit 8 (CCK-8) assay on MDA-MB-231 human breast cancer cells and NIH3T3 mouse embryonic fibroblast cells provided by the Cell Bank of the Chinese Academy of Sciences (Shanghai, China).

Briefly, MDA-MB-231 and NIH3T3 cells were seeded into 96-well plates (5 × 10³ cell/well) for 24 h. After coculture with MCPM NPs at a gradient of concentrations (0–0.5 mg ml^-1) for 24 h, 10 μl fresh CCK-8 solution was added for further incubation of 2 h. The absorbance of the MPCP NP solution was determined using a microplate reader at 450 nm. Identical to the manipulation of cell viability assay, the relative cell viability post irradiation was also determined with the standard CCK-8 method.

Subsequently, we also explored the photothermal–photodynamic therapeutic effect on MDA-MB-231 tumor cells. Following overnight culture in 96-wells, MDA-MB-231 cells were treated with MCPM NPs at the end of laser irradiation for 10 min in different conditions (no laser, 808-nm laser, 660-nm laser, 808/660-nm laser). In parallel, standard CCK-8 assay was also conducted to assess the cell viability after photothermal–photodynamic treatment.

MDA-MB-231 cells (8 × 10⁴ cell/well) were seeded on the coverslips in a six-well plate overnight. Thereafter, MCPM NPs were added to each well at a concentration of 0.3 mg ml⁻¹ and incubated for another 4 h. Next, cells were rinsed, fixed, and stained with DAPI, followed by observation with CLSM (LS880, Zeiss, Oberkochen, Germany). With respect to in vitro MR imaging, MDA-MB-231 cells (1 × 10⁵ cell/well) were seeded on a six-well culture dish for 24 h. Thereafter, MCPM NPs at different concentrations (0, 0.03, 0.1, 0.3 mg ml⁻¹) were added in each well. Then, the cells were rinsed, trypsinized, centrifuged, and transferred into Eppendorf tubes for MRI. The in vitro MR images were obtained with the MR system and parameters described previously.

MDA-MB-231 cells (1 × 10⁵ cell/well) were seeded on the coverslips in a six-well plate and kept overnight for cell attachment. Thereafter, MCPM NPs were added in each well at a concentration of 0.3 mg ml⁻¹ and underwent different laser irradiations for 10 min (no laser, 808-nm laser, 660-nm laser, 808/660-nm laser). Thereafter, cells were rinsed thrice in PBS, fixed with 4% paraformaldehyde solution, and mounted with mounting medium containing 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA) and 4′,6-diamidino-2-phenylindole (DAPI), respectively. Subsequently, the samples were evaluated under a CLSM (LS880, Zeiss, Germany).

Calcein-AM (green fluorescence) and propidium iodide (PI, red fluorescence) were employed to assess cell viability before and after laser irradiation (808 nm, 1.0 W cm⁻², 10 min; 660 nm, 100 mW cm⁻², 10 min), with identical manipulations to the CCK-8 assay. With the addition of Calcein-AM and PI into each well, cells were incubated for 30 min without CCK-8 treatment. Thereafter, the cells were observed under CLSM for photography.

The animal experiments were performed in accordance with the protocol approved by the Institutional Animal Care and Use Committee of Xuzhou Medical University (ethical approval number: 202007W001). For the establishment of the mouse tumor model, MDA-MB-231 cells (2 × 10⁶ cells in 150 μl) were subcutaneously injected in the right axilla in BALB/C nude mice (female, 6 weeks, 15–17 g).

For the biodistribution study, BALB/C nude mice were intravenously injected with MCPM NPs (10 mg ml⁻¹, 0.2 ml). The blood was collected via the caudal vein 40 μl/time at time points of 0.5, 1, 2, 3, 4, 6, 8, 12, and 24 h post-injection. The blood samples were digested with 40 μl nitric acid (analytically pure) and then heated in an 80°C water bath for 30 min and diluted to 5 ml by addition of ultrapure water. The Mn content in these samples was determined with inductive coupled plasma mass spectrometry (ICP-OES).

The mouse tumor models were deployed in in vivo MR imaging in the case of the tumor size of 5 mm, wherein MCPM NPs (10 mg kg⁻¹) were systematically administered into the mouse tumor models via the caudal vein. Thereafter, murine MR imaging was performed at different time points pre- and post-administration (0.25, 0.75, 1, 2, 3, 4, 6, 8, 12, and 24 h). For in vivo fluorescent imaging, MCPM NPs (10 mg kg⁻¹) were injected into the mice. The images were obtained on a fluorescence imaging system (Bio-Rad, Hercules, CA, USA) at different time points pre- and post-administration (0.25, 0.75, 1, 2, 3, 4, 6, 8, 12, and 24 h).

The in vivo antitumor efficacy of PTT–PDT combination therapy was investigated in MDA-MB-231 nude mice (with tumor volume ∼100 mm³). The mice were randomized into six groups (three mice/each) and received an injection of 200 μl saline, saline + 808/660 nm laser, MCPM NPs, MCPM NPs + 808 nm laser, MCPM NPs + 660 nm laser, and MCPM NPs + 808/660 nm laser (0.3 mg ml⁻¹), respectively. At 2 h post-injection, the tumor areas of mice were conducted to laser irradiation for 10 min (808 nm, 1.0 W cm⁻²; 660 nm, 100 mW cm⁻², 10 min). The temperature was recorded by a FLIR thermal camera. The body weight and tumor volume in the mice were determined every other day [tumor volume= (tumor length tumor width)²/2]. All the treatments were repeated every 3 days from day 1 to day 14.

The in vivo biosafety of the MCPM NPs was assessed in BALB/C mice with the biochemistry indexes and histology analysis. Healthy BALB/C mice were randomized into control group and three experimental groups. Mice in the experiment group received an intravenous injection of aqueous MCPM NPs (10 mg kg⁻¹), with blood samples collected for blood biochemical analyses at 1, 7, and 21 days post-injection. The major organs such as the heart, liver, spleen, lungs and kidneys of the mice were isolated for histological analysis with hematoxylin and eosin (H and E) staining, followed by histological analyses under a microscope.

All data in this experiment are expressed as mean ± SD. One-way ANOVA was employed for comparison between multiple samples. Student’s t-test was adopted to evaluate the between-group statistical significance, with p < 0.05 considered as statistically significant in all cases.

The strategy for the preparation of MCPM NPs is illustrated in Scheme 1. The TEM images of as-synthesized MSN in Figure 1A demonstrated that the MSN was almost uniform in size with a spherical morphology. The estimated mean diameter was 125 ± 2.46 nm (Figure 1C). As for the MCPM NPs, there were evident changes in morphology, with MCPM NPs presenting a near-spherical appearance and relatively smooth surface (Figure 1B), with an estimated average size of 139 ± 1.70 nm (Figure 1D).

For T₁-weighted MR imaging, Mn ions were chelated inside the structure of PDA by dispersion of MSN-Ce6@PDA in a solution of Tris and manganese (II) chloride tetrahydrate. To explore the element composition of MSN-Ce6@PDA (Mn), all the major elements such as C, N, O, Si, and Mn were detected within the EDS spectrum (Figure 1E) and the elemental mapping images (Figure 1F).

A series of zeta potential measurement is illustrated in Figure 2A. The zeta potential of MSN and MSN-NH₂ was measured at -19.4 ± 0.51 mV and 17.3 ± 0.81 mV attributed to the hydroxyl groups and amino groups, respectively. After conjugation of Ce6, the zeta potential decreased to −23.63 ± 0.73 mV because the Ce6 molecule contains three negatively charged carboxyl groups. After PDA coating and manganese ion coordination, the zeta potential of MCPM NPs increased. In contrast, PDA contains a large amount of amine phenolic hydroxyl (anionic) groups with negative charges, whereas a small quantity of Mn contains positive charges. Thus, after the coordination of Mn²⁺, the MCPM NPs gain a negative surface charge of −16.03 ± 0.12 mV. The change in a series of zeta potential indicated the successful preparation of MCPM NPs. The MSN-Ce6@PDA and MCPM NPs exhibited good dispersity and stability in PBS even for 21 days, whereas MSN-Ce6 and free Ce6 displayed unsatisfactory stability (Supplementary Figures S1, S2). Ce6 has a hydrophobic potency, hence the high dissolution in DMSO rather than PBS. MSN-Ce6 showed good stability in the initial 2 days but precipitated after 7 days (Supplementary Figure S1). Besides, the hydrate particle sizes of MCPM NPs show little change at the 4°C storage condition (Supplementary Figure S3). The laser irradiation treatment also exhibited small change to the hydrodynamic size of MCPM NPs (Supplementary Figure S4). In addition, the MCPM NPs also possess good stability, which enable good T₁-weighted MR stability in PBS for a long term (Supplementary Figure S4).

As illustrated in Figures 2B,C, N₂ sorption isotherms for MSN and MCPM NPs were determined to explore the effect of the Ce6 conjugating and PDA coating on the surface area and pore volume of the MSN. The BET and BJH analyses yielded a highly specific surface area of 728.4 cm²g⁻¹ and a mean pore size of 10.5636 nm for the MSNs. After a series of synthetic procedures, the decreased BET surface area (275.4 cm²g⁻¹) and mean pore size (7.4 nm) of the prepared MCPM NPs confirmed the successful assembly. In addition, the reduced volume of absorption and the diminished pore size manifested the decrease of mesoporous silica/nanoparticle composites, further validating the successful coating of PDA.

To further verify the successful assembly of MSN, a comparison of MSN, MSN-NH₂, MSN-Ce6, and MCPM NPs was performed by the FTIR spectrum of the different formulations. As depicted in Figure 2D, absorbance peaked at approximately 1,076 cm⁻¹ of the MSN, MSN-NH₂, MSN-Ce6, and MCPM NP samples, which could be the stretching of Si–O–Si groups, verifying the successful synthesis of MSN. Compared with the unmodified MSN, the aminated mesoporous silica showed a new peak around 1,556 cm⁻¹, which was related to N–H bending of the amino group, indicating the successful amination. The new absorbance peak appeared at 1,710 cm⁻¹ for MSN-Ce6, and the merged band was formed consequent to the C = O stretching in the amide bond, indicative of the success of the amide reaction. Furthermore, the presence of PDA on the surface of MSN-Ce6 could be validated via the onset of absorbance peaks at 1,619 cm⁻¹ designated to the superposition of N–H bending and C–C stretching of aromatic components, 1,485 cm⁻¹ (N–H scissoring), and 1,282 cm⁻¹ in response to C–O stretching of the phenol of PDA (Hu et al., 2016; Sun et al., 2018). Given that, the MCPM NPs were successfully synthesized.

No absorbance peaks were evidenced in the controlled PDA and MSN groups, with typical UV-vis peaks of free Ce6 at 400 and 660 nm, and the fluorescent emission peak of free Ce6 at 660 nm. Despite the presence of an evident red shift in the UV-vis peak and fluorescence emission peak of MSN-Ce6 and MCPM NPs at 667 nm compared to Ce6, Ce6 was successfully conjugated to MSN (Figure 2E).

The fluorescent efficiency of Ce6 is pivotal to PDT efficacy. As delineated in Figure 2F, compared with the potent fluorescence of free Ce6, the peak in MSN-Ce6 and MCPM NPs was at approximately 660 nm, which is slightly affected owing to the π–π stacking of molecules. We speculated that the MSN-Ce6 and MCPM NPs could still be suitable for PDT.

Given the successful PDA coating on the MSN, we investigated the photothermal performance of the MCPM NP dispersion under the irradiation of the 808-nm laser. The hydrodynamic size of MCPM NPs with or without laser irradiation exhibited little difference (Supplementary Figure S3). As depicted in Figure 3A; Supplementary Figure S5, MCPM NPs were rapidly heated to a threshold in a concentration-dependent manner. At the end of a 10-min exposure to the 808-nm laser (1.0 W cm⁻²), the MCPM NP dispersion of 0.03–0.5 mg ml⁻¹ was quickly heated to 36.2°C–57.3°C from ambient temperature. In contrast, the temperature of the PBS solution was only heated to 29.0°C. Moreover, under irradiation of the 808-nm laser at different intensities (0.25–1.5 W cm⁻²), the MCPM NP solution (0.3 mg ml⁻¹) also presented an evident temperature rise from 32.4°C to 52°C with an obvious NIR power-dependent pattern (Figure 3B), indicating the superior conversion capacity from 808-nm laser energy to thermal energy. In parallel, the photothermal stability was also evaluated via repeated laser irradiation to MCMP NP dispersion for eight on/off cycles. In Figure 3C, the dispersion was steadily heated to the peak temperature in the eight heating/cooling cycles, authenticating its excellent photothermal stability (Figure 3D). Thus, the efficient thermal treatment of the PDA shell validated the stability and continuum of the thermal gradient for the thermophoretic propulsion of the MCPM NPs.

The DPBF kit was employed to evaluate the generation of singlet reaction oxygen species (ROS). DPBF was remarkably degraded by the ROS, with the degree of degradation closely dependent on the ROS generated by MCPM NPs. The 808/660-nm laser group generated the most ROS. This result indicated that the 808-nm laser triggered the release of Ce6, contributing to the greater ROS generation performance than MCPM NPs under the 660-nm laser alone. A previous report also revealed that the degradation of the mesoporous silica nanocarrier could facilitate drug release (Shao et al., 2020). Furthermore, some other studies also indicated that the MSNs with a large mesopore of ∼10 nm revealed a much faster release (Shen et al., 2014). As mentioned in Figure 2C, our MSNs possessed a mean pore size of 10.5636 nm; therefore, the matrix degradation of MSN probably achieved a fast Ce6 release, which probably induced more ROS generation.

Given the successful T₁-weighted MRI of MDA-MB-231 cells by MCPM NPs, we further investigated their possible application for in vivo imaging under a 3.0-T MRI system. The MCPM NPs were intravenously administered into the BALB/c nude mice with heterografted breast cancer. Then, mice were performed with MRI at various time points pre- and post-injection. Prior to injection of MCPM NPs, an MR signal of relatively low intensity was identified across the mice including the tumor region. Nevertheless, following intravenous administration of MCPM NPs, the MR signal was markedly enhanced in comparison to the pre-injection state. A significantly enhanced MR signal was observed in the liver and gallbladder within 4 h post-injection of MCPM NPs, which was attenuated afterward, implying the presence of a possible clearance route of the nanotheranostic agent presumably via hepatobiliary excretion (Figure 7A). Moreover, the gradual brightness in the tumor region was consistent with the prolonged duration after injection, with the onset of the peak at 2–3 h post-injection and decline in an MR signal intensity within 24 h (Figure 7B). Furthermore, we quantified the signal value of the tumor region and identified the presence of enhanced T₁-weighted MR signals after the administration of MCPM NPs (Figure 7C).

A similar approach was utilized by in vivo fluorescence imaging. Accordingly, no contrast enhancement was observed in tumors before i.v. injection of the MCPM NPs. The presence of sensitive brightness in the tumor region appeared at 30 min, with a strong contrast enhancement at 2–3 h, following signal reduction within 24 h (Figures 7C,D). These results confirmed the accumulation and retention of MCPM NPs in tumors. Therefore, the MCPM NPs can be effectively used in MDA-MB-231 nude mice.

Considering the in vivo MRI and fluorescence imaging outcome, the in vivo cancer therapy with MCPM NPs was then evaluated (Figure 7). The temperature of the tumor area rapidly rose to a plateau of 46.5°C and then remained stable (Figure 8A) under laser irradiation. In contrast, the temperature of the saline group barely changed. When the tumor volume grew to 100 mm³, the mice were randomly divided into six groups: (I) saline, (II) MCPM NPs, (III) s plus 808/660 nm laser, (IV) MCPM NPs plus 808 nm laser, (V) MCPM NPs plus 660 nm laser, and (VI) MCPM NPs plus 808/660 nm laser (808 nm laser, 1.0 W cm⁻², 10 min; 660 nm, 100 mW cm⁻², 10 min). Saline and MCPM NPs were injected into each mouse via the tail vein at 10 mg kg⁻¹ for MCPM NPs. Then, 2 h later, those mice were performed with different treatments as described previously. After laser irradiation, the tumor volumes were measured by a vernier caliper every other day for 2 weeks. In the photothermal–photodynamic combination therapy group, the MCPM NPs plus 808/660-nm laser yielded a significant inhibition of tumor growth (Figures 8B,C). MCPM NPs plus laser irradiation (IV, V, and VI), for the initial several days, exerted a slight inhibitory effect on the tumor growth, which however resumed thereafter. No significant effect on tumor growth was observed in the control (I) and the remaining two treatment groups (II, III) (Figure 8D), while the body weight barely changed (Figure 8E).

The biological safety of MCPM NPs in vivo was assessed in healthy mice by means of blood biochemistry and histological analysis of organs at days 1, 7, and 21 after injection. No marked differences were observed with respect to the levels of indicators including blood aspartate aminotransferase, alanine aminotransferase, albumin, total protein, creatinine, urea, erythrocytes, leukocytes, platelets, hemoglobin, mean corpuscular volume, mean corpuscular hemoglobin concentration, and mean corpuscular hemoglobin exhibited compared to the saline group, which confirmed insignificant influence to hepatic and renal functions (Figures 9A,B).

To further explore the potential effect of MCPM NPs on mice, hematoxylin and eosin staining was conducted on the sections from major organs in the control group (saline) and MCPM NP group (days 1, 7, and 21), which showed no appreciable inflammation or lesions (Figure 9C). These experimental results confirmed the non-toxicity of MCPM NPs without laser irradiation. Taken together, these results suggested the negligible toxicity of MCPM NPs in vivo.