Self-assembled biomimetic nanocomposite integrating defect-enhanced piezoelectricity with NIR-II fluorescence and MR imaging for ultrasound-triggered piezo/chemodynamic therapy of subcutaneous glioma Authors: Ikram Hasan, Zia Ullah, Fei Wang, Li Wang, Yibin Yan, Tingting Gong, Muhammad Madni, Dhananjoy Mondal, Hanyang Qu, Jhilik Roy, Yinghe Zhang, Shubham Roy, Bing Guo, Chunqi Chang Published: 2026 ABSTRACT Multifunctional theranostic nanoplatforms enabling precise targeting and controlled deep tissue therapy are vital. Herein, we present a self-assembled, defect-engineered, biomimetic nanoplatform (CDMBBTO), in which piezoelectric material and downconversion nanoparticles (DCNPs) are co-assembled within a polymer matrix, preventing interactions between both components, thereby allowing each to retain its intrinsic properties while collectively enabling dual second near infrared fluorescence (NIR-II FL)/magnetic resonance (MR) imaging and synergistic piezo/chemodynamic therapy (PZDT/CDT). Mn/Bi codoped piezoelectric BaTiO3 (MBBTO) NPs exhibit a significantly enhanced piezoelectric coefficient (~5 fold higher than pristine BTO) owing to bandgap modulation induced by defect engineering. The incorporated Mn2+ not only catalyzes Fenton like reactions for sustained tumor suppression but also provides strong T2 weighted MR contrast. Self-assembled NIR-II emissive DCNPs enable deep tissue optical imaging, forming an integrated theranostic system. Cloaking of U87 glioma cell membranes imparts homologous tumor targeting capability. In vivo dual modal imaging reveals efficient tumor accumulation, with peak NIR-II FL and MR signal intensities observed 6 h post-injection. Upon ultrasound activation, CDMBBTO elicits potent tumor ablation through synergistic reactive oxygen species generation and immune modulation, characterized by macrophage polarization (M2 → M1) and upregulation of proinflammatory cytokines (TNF-α and IL-6). This work establishes CDMBBTO as a powerful nanoplatform for dual modal imaging-guided, immune potentiated PZDT/CDT toward effective glioblastoma treatment.Graphical abstract Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12951-026-04694-7. Introduction Conventional reactive oxygen species (ROS) based dynamic therapies, including photodynamic therapy (PDT), chemodynamic therapy (CDT), and sonodynamic therapy (SDT), have shown good efficacy in cancer treatment [1–3]. They are often integrated with imaging modalities like magnetic resonance imaging (MRI), computed tomography (CT), positron emission tomography (PET), ultrasound imaging (USI), and fluorescence (FL) imaging for real time monitoring of drug distribution and image guided treatment [4–6]. Among various imaging modalities, MRI offers excellent soft tissue contrast but cannot provide real time monitoring, CT provides rapid imaging but involves radiation, PET enables sensitive metabolic detection but has low spatial resolution, and USI is real time and safe but limited by poor spatiotemporal resolution [7–9]. The FL imaging exhibits high cost effectiveness and high sensitivity, but often suffers from limited penetration and relatively low temporal resolution, and unignorable background interferences [10, 11]. The emerging NIR-II FL imaging could circumvent these problems with improved imaging fidelity, enabling deep tissue penetration, high spatial resolution, reduced light scattering, and real-time visualization for accurate tumor localization and image-guided therapy [12–15]. More importantly, it is straightforward to combine NIR-II FL with other imaging modalities, such as MRI, to obtain complementary diagnostic information, synergistic enhancement of spatial and temporal resolution, precise anatomical and functional correlation, and accurate real-time guidance for theranostic applications [16, 17]. Thus, it is a burgeoning trend to develop multiple imaging (NIR-II FL, CT, MRI, etc.) guided dynamic therapies for a precise theranostic system. ROS based dynamic therapies offer high therapeutic throughput by oxidative stress that effectively destroys tumor cells. Their stimuli responsive nature enables noninvasive, spatiotemporally controlled treatment, reducing systemic toxicity compared to conventional therapies [18, 19]. However, they often suffer from limited ROS diffusion distances and short lifetimes, restricting their therapeutic reach. In addition, the hypoxic tumor microenvironment (TME) limits ROS yield, and non-selective oxidative stress may damage healthy tissues [20]. Therefore, achieving on demand and localized ROS production is essential to enhance therapeutic precision and minimize off target effects. Herein, piezodynamic therapy (PZDT) plays a pivotal role by producing an enormous amount of localized ROS under any mechanical stimulation [21, 22]. It generates cytotoxic ROS in tumor tissues by using piezoelectric materials to transform mechanical energy into localized electric charges [23]. Deep tissue penetration, low invasiveness, and spatially accurate ROS generation are some of the significant benefits of PZDT over traditional modalities [24]. It has shown great advantages in cancer therapy, being effectively applied for tumor eradication, inhibition of metastasis, and enhancement of immune response [25, 26]. The piezoelectric materials regulate the macrophage polarization via localized electric signals at the cell/material interface upon mechanical stimulation [27]. These electric signals modulate the membrane potential and trigger downregulation of NF-кβ, MAPK, PI3K/Akt, and STAT signaling pathways that determine M1 or M2 phenotypes [28, 29]. At the same time, piezoelectric charge separation drives controlled ROS generation, which acts as a secondary messenger for macrophage polarization [30, 31]. Material design plays a critical role in this phenomenon. The surface charge influences protein adsorption and adhesion-mediated mechanotransduction, while crystal structure, defect engineering, and stress responsiveness govern piezoelectric output [32]. Recent advances in nanostructuring, electrical poling, defect engineering, and heteroatom doping have significantly enhanced the piezoelectric response and ROS generating efficiency of such materials, thereby improving therapeutic outcomes [33]. However, PZDT is inherently limited by its dependence on external ultrasound (US) stimulation. The ROS are generated only upon US activation, resulting in a transient therapeutic window [34, 35]. This transient effect means that PZDT alone is insufficient to prevent long term tumor recurrence once the stimulation ceases. As chemodynamic materials would generate ROS in a relatively long lasting span in the TME, which is an ideal partner to work with PZDT to achieve efficient destruction of tumors by PZDT, and long term moderate eradication of residual and infiltrated tumors [22]. Several studies have demonstrated that the dopant used in modulating the band gap (precisely tilting the band edges by introducing defects) may have other therapeutic functions, such as CDT by initiating Fenton-like reaction. Mn, Cu, and Fe ions have intrinsic properties of producing ROS under H2O2 environment and simultaneously modulating the band structure by inducing lattice defects [36, 37]. Therefore, these metal ions doping within the piezoelectric materials could integrate multiple ROS based dynamic therapies in a single nanoplatform. Besides, these dopants may have intrinsic imaging functionalities (such as MRI, CT, etc.). For instance, Mn2+ could provide T2 MR signals under strong magnetic fields, which could aid in combining MRI guidance with PZDT/CDT. However, it is challenging to achieve NIR-II FL guided PZDT due to the optical darkness of piezoelectric materials and ROS induced degradation of NIR-II fluorophores [38]. Therefore, it is recommended to use inorganic fluorophores to achieve NIR-II FL for real time imaging. Most recently, we reported that a defect engineered biomimetic nanoplatform has been designed by coating a NIR-II transparent Gd doped KNbO3 piezoelectric shell onto NIR-II fluorescent NaGdF4@Nd/Yb/Er nanocores [39]. This multifunctional system enabled dual NIR-II FL and MR imaging guidance with US triggered ROS generation for tumor inhibition and macrophage polarization, demonstrating efficient image guided PZDT [38, 39]. It achieved NIR-II FL and MR imaging guided PZDT, which is of great potential for clinical application. However, the synthesis process is rather tedious, involving multistep core-shell construction, and it can only be achieved with the optically transparent piezoelectric shells [40]. Thus, there is a pressing need to develop a facile strategy for constructing nanoplatforms that can not only achieve strong and sustained ROS generation through synergistic PZDT and CDT effects but also provide bright NIR-II FL and MR imaging capabilities, features that remain scarcely reported in the literature. To substantiate this, the fluorophores could be non-covalently self-assembled with the piezoelectric nanoparticles (NPs) in a common polymer matrix, preventing the interaction between both components. Such non-covalent self-assembly restricts charge and electron transfer, making the sample extremely fluorescent with enriched piezodynamic activity. It is anticipated that such an engineered self-assembly would generate dual PZDT/CDT nanoplatform with not only strong and long-term ROS generation, but also bright NIR-II FL and MRI imaging functionalities. To demonstrate the proof of concept, we designed and synthesized a self-assembly driven theranostic system comprising defect-engineered Mn/Bi codoped piezoelectric BaTiO3 (MBBTO) NPs and NIR-II emissive NaY0.58F4@Yb0.4Er0.02 downconversion NPs (DCNPs), co-assembled within a biocompatible Pluronic F-127 matrix to form a multimodal nanoplatform (DMBBTO), which was further camouflaged with U87 glioma cell membrane to achieve homologous tumor targeting (Scheme 1). The design we adopted herein is simpler, having the advantage of using any piezoelectric material with the conjugation of NIR-II fluorescent DCNPs, which makes this strategy more general and advanced. The F-127 polymer matrix is responsible for creating a boundary between MBBTO and DCNPs, maintaining both its piezodynamic ROS production and FL intensity. Initially, the MBBTO NPs were integrated with NIR-II DCNPs via polymer assisted non-covalent self-assembly, followed by comprehensive physicochemical characterization, including piezoelectric performance assessment and dual NIR-II FL and MR imaging evaluation. In the MBBTO system, Bi3+ incorporation induces lattice distortion through microstrain modulation, whereas Mn2+ doping introduces oxygen vacancies due to ionic size and charge mismatch (Ba2+: 135 pm, Bi3+: 103 pm, Mn2+: 80 pm), collectively resulting in defect mediated bandgap widening. The piezoelectric coefficient (d33) is greatly increased by this tailored defect chemistry. DMBBTO has shown an exceptionally high d33 value of 106.91 pmV− 1, resulting in large US-triggered piezocatalytic ROS production. Mn2+ doping simultaneously permits Fenton-like chemodynamic ROS production and produces intrinsic MR contrast with a T2 relaxivity of 11.36 mM−1s− 1. In vivo NIR-II FL and MR dual modal imaging of subcutaneous glioma demonstrated peak signal accumulation at 6 h post-injection, confirming precise tumor homing enabled by U87 membrane camouflage (CDMBBTO). In vitro studies, including cytotoxicity, hemolysis, cellular internalization, ROS staining, and live/dead assays, verified the excellent biosafety and potent therapeutic functionality of CDMBBTO. Furthermore, in vivo synergistic PZDT/CDT achieved pronounced tumor regression, facilitated by macrophage repolarization (M2 → M1) and ROS induced oxidative stress (elevated TNF-α and IL-6). Finally, H&E histology and blood biochemistry confirmed the systemic biocompatibility of CDMBBTO, underscoring its strong translational promise for dual NIR-II FL and MR imaging guided synergistic PZDT/CDT against glioblastoma. Scheme 1Schematic representation of designing a self-assembly based biomimetic nanoplatform (CDMBBTO) containing defect-engineered Mn/Bi codoped piezoelectric MBBTO NPs with ultrahigh d33 coefficient and bright NIR-II fluorescent DCNPs (DMBBTO), further camouflaged with U87 glioma cell membrane for targeted dual NIR-II FL and MR imaging guided PZDT/CDT of subcutaneous glioblastoma and immunoregulation in mice models. This has been achieved by co-assembling the piezoelectric material and DCNPs within a polymer matrix, preventing their interaction, preserving their individual functions, and enabling dual NIR-II FL/MR imaging with synergistic PZDT/CDT Schematic representation of designing a self-assembly based biomimetic nanoplatform (CDMBBTO) containing defect-engineered Mn/Bi codoped piezoelectric MBBTO NPs with ultrahigh d33 coefficient and bright NIR-II fluorescent DCNPs (DMBBTO), further camouflaged with U87 glioma cell membrane for targeted dual NIR-II FL and MR imaging guided PZDT/CDT of subcutaneous glioblastoma and immunoregulation in mice models. This has been achieved by co-assembling the piezoelectric material and DCNPs within a polymer matrix, preventing their interaction, preserving their individual functions, and enabling dual NIR-II FL/MR imaging with synergistic PZDT/CDT Materials and methods Materials Chemicals used in this research include bismuth nitrate (Bi(NO3)3.5H2O; 99%), manganese(II) chloride (MnCl2.4H2O; 98%), barium nitrate (Ba(NO3)2; 99%), titanium(IV) butoxide (Ti(OC4H9)4; 99%), cetyltrimethylammonium bromide (CTAB), and sodium hydroxide (NaOH; >96%) for preparing the piezoelectric material. After that, yttrium acetate (Y(CH3COO)3.6H2O; 99.9%), ytterbium acetate (Yb(CH3COO)3.6H2O; 99.9%), erbium acetate (Er(CH3COO)3.xH2O; 99.5%), oleic acid (C18H34O2; 85%), 1-octadecene (C18H36; >90%), and NH4F (98%) were acquired for the preparation of second near infrared (NIR-II) fluorescent downconversion nanoparticles (DCNPs). Moreover, rhodamine-B (Rh-B, C28H31ClN2O3; 99%), methylene blue (C16H18ClN3S), TMB (C16H20N2), SOSG, and DMPO (C6H11NO) were used to probe ROS experiments. The chemicals were obtained from Aladdin Chemical Technology and Macklin Chemical Reagents. All materials were used without any additional purification. Deionized (DI) water was used for all experiments. All apparatus was first cleaned with chromic acid and subsequently rinsed thoroughly with distilled water. Material preparation Synthesis of Bi0.5Ba0.5TiO3 (BBTO) and Mn-doped BBTO (MBBTO) In a typical synthesis, 0.1 g of CTAB was added to 25 mL of DI water and stirred until a clear solution was obtained. Subsequently, Bi(NO3)3.5H₂O (0.05 mmol) and Ba(NO3)3 (0.95 mmol) were added to the CTAB solution. The resulting mixture was stirred at 500 RPM for 2 h to ensure complete homogenization. Separately, Ti(OC4H9)4 (1 mmol) was added to 5 mL of n-butanol and then dropped into the reaction mixture. After 30 min of continuous stirring, a NaOH solution (500 mg in 5 mL DI water) was slowly introduced, and the solution was kept in the same condition for another 3 h. The reaction mixture was then transferred into an autoclave and heated at 150 °C for 18 h. After cooling, the precipitates were collected from the bottom of the Teflon liner, washed, and dried at 60 °C for 8 h. The dried material was then finely ground using an agate mortar and designated as BBTO NPs. Similarly, MBBTO NPs were synthesized by introducing 0.05 mmol of MnCl2.4H2O following a similar procedure described above. The obtained product was then dried, ground, and labeled as MBBTO. Synthesis of β-NaY0.58F4@Yb0.4Er0.02 DCNPs Accurately weighed Y(CH3COO)3.6H2O (0.58 mmol), Yb(CH3COO)3.6H2O (0.40 mmol), and Er(CH3COO) 3.xH2O (0.02 mmol) were added to a 50 mL three-necked flask containing 6 mL of oleic acid and 15 mL of 1-octadecane and stirred at 30 °C for 30 min. The reaction mixture was then sealed and heated to 150 °C for 45 min. After cooling to 50 °C, NaOH (2.5 mmol) and NH4F (4 mmol) were slowly added dropwise into the flask. Following 45 min of reaction, the reaction mixture was gradually heated to 100 °C and degassed for 30 min to evaporate methanol. Subsequently, the reaction temperature was raised to 300 °C for 1 h under an inert atmosphere. Once cooled, an equal volume of methanol was added, and the precipitate was collected by centrifugation at 10,000 RPM. The collected precipitate was washed three times with ethanol, redispersed in 5 mL of cyclohexane, and stored at 4 °C for further use. Pluronic F-127 mediated self-assembly of MBBTO and DCNPs (DMBBTO) The self-assembly of MBBTO and DCNPs was carried out by dispersing 0.1 g of the synthesized MBBTO NPs in 25 mL of DI water, followed by sonication for 1 h. Separately, 0.3 g of Pluronic F-127 was added to 20 mL of DI water and stirred until the formation of a clear solution. The MBBTO dispersion was then added to the F-127 solution and allowed to stir for an additional 24 h. The product was subsequently washed several times by centrifugation to remove excess F-127, and the collected F-127-coated MBBTO NPs were air-dried. Next, 0.1 g of the F-127-coated MBBTO NPs was redispersed in 25 mL of DI water and sonicated for 1 h to obtain a homogeneous solution. In a separate container, 0.2 g of DCNPs was dissolved in chloroform and sonicated for 20 min to ensure complete dispersion. This DCNP solution was then added dropwise to the F-127-coated MBBTO solution. The mixture was stirred for another 24 h and washed multiple times to remove any residual F-127. The final DMBBTO NPs precipitates were collected and air-dried. For investigating the reproducibility and the effect of concentration, stirring speed, and stirring time on the piezoelectric properties, fluorescence properties, size, and stability of the prepared sample, different batches (3 batches) were synthesized by following the similar procedure by adding different concentrations (1:1 and 1:2) of MBBTO and DCNPs at different stirring speed (300, 500, 700 rpm) for various time intervals (6 to 24 h). Isolation of the U87 glioma cell membrane For cell membrane extraction, U87 glioblastoma cells were first collected by centrifugation and lysed using a hypotonic buffer combined with repeated freeze-thaw cycles. The suspension was subjected to three freeze-thaw cycles at −80 °C, followed by disruption using a 20% ultrasonic cell disruptor with cycles of 2 s ON and 3 s OFF for a total of 5 min. The mixture was then processed via gradient centrifugation using an ultracentrifuge at 700 x g for 10 min and 14,000 x g for 30 min to isolate the cell membranes [41]. The membrane pellet was washed and resuspended in PBS, and the protein content of the membranes was quantified using a BCA Protein Assay Kit (Beyotime, China). Preparation of U87 glioma cell membrane-coated DMBBTO (CDMBBTO) DMBBTO NPs (0.5 mg/mL) and U87 glioma cell membranes were dispersed in PBS and passed through a 0.4 μm filtration membrane using a mini extruder (Avanti, AU) 20 times to achieve uniform membrane coating. The weight ratio of membrane protein to DMBBTO NPs was maintained at 1:1. The resulting product was designated as CDMBBTO. Characterizations of the prepared samples The phase composition of all synthesized samples was analyzed using an X-ray diffractometer (XRD) (D8, Bruker AXS, Wisconsin, USA). The measurements were conducted over a 2θ range of 10°−70°, with the instrument operating at 35 kV and 35 mA at a scan rate of 0.5 s/step [42]. The obtained diffractograms were further refined using the MAUD v2.996 software package. The refined crystal structures and bonding configurations were visualized using VESTA v3.5.2 [43]. Bonding interactions in the synthesized samples were additionally confirmed by Fourier transform infrared (FTIR) spectroscopy. An ITR ATR Diamond accessory (Thermo Fisher Scientific) was used for measurements, performed in the 650-4000 cm− 1 range employing a Nicolet iS50 FTIR spectrometer, to confirm the various functional groups and vibrational modes [38]. The morphology and particle size were examined via transmission electron microscope (TEM) (JEM-1400 microscope, JEOL, Japan) [44]. Elemental distribution was analyzed through energy dispersive X-ray scattering (EDS) mapping (S-4800 microscope, Hitachi, Japan) [45]. The surface charge of the samples was determined in DI water at 7.4 pH using a Zetasizer (Malvern Instruments) via zeta potential measurements. Additionally, X-ray photoelectron spectroscopy (XPS) was performed using a K-ALPHA (Thermo Scientific) instrument equipped with a monochromatic Al source (E = 1486.68 eV) [46]. The protein concentration of U87 glioma cell membranes was determined via the BCA assay. For SDS-PAGE analysis, 25 µg of cell membrane protein for each sample was loaded onto 4–20% gels and separated using a two-step voltage protocol [47]. For Western blotting, the same amount of each protein sample was separated and transferred onto PVDF membranes. Membranes were blocked with 5% BSA in TBST for 2 h at normal temperature [47, 48]. After that, membranes were incubated with primary antibodies overnight at 4 °C, including mouse monoclonal anti-β-actin (1:10,000) and anti-E-cadherin (1:1,000). After washing with TBST, the membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (dilution 1:1000) diluted in TBST for 1 h at room temperature. The enhanced chemiluminescence system was employed to detect the protein signals, and the intensity of protein bands was quantified. Moreover, for demonstration of coating efficiency and stability of the cell membrane, the FITC labelled cell membrane was coated on the DMBBTO, and bright field and fluorescence images were captured. Assessment of the piezoelectric properties of the prepared samples Piezoforce microscopy (PFM) analysis was conducted using an atomic force microscope (AFM) (Bruker Dimension Icon, Germany) integrated with a ferroelectric testing system in a ± 10 V range to assess the piezoelectric behavior of the different batches of DMBBTO NPs. The optical band gap of the prepared samples was evaluated using Tauc’s plot method, described by the following equation: 1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha\mathrm{E=A(E-Eg)}^\mathrm{n}$$\end{document} In this equation, E, Eg, α, and A represent the photon energy, optical band gap energy (eV), absorption coefficient, and a proportionality constant, respectively. The band gap energy was estimated by extrapolating the linear region of the Tauc’s plot, plotted as (αhν)2 vs. photon energy (hν), to its intersection with the energy axis [49]. Additionally, a three-dimensional simulation was conducted using the Piezoelectric Devices (pze) Multiphysics module in COMSOL Multiphysics to investigate the piezoelectric response and surface polarization of MBBTO NPs under mechanical stimulation. The model consisted of an MBBTO nanosphere with an approximate diameter of 5 nm, placed in a vacuum domain to avoid interference from external mechanical or electrical fields [50]. A similar simulation was also performed for glioma membrane-coated MBBTO NPs to assess how the physiological environment influences surface polarization behavior [51]. Essential material parameters, such as density (ρ = 6080 kg m− 3), elastic stiffness matrix (cE), piezoelectric coupling matrix (eES), and relative permittivity (εr), were appropriately assigned for the MBBTO NPs. The key simulation parameters are outlined below:2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{{\upepsilon\:}}_{r}=\left[\begin{array}{ccc}1220&\:0&\:0\\\:0&\:1220&\:0\\\:0&\:0&\:1520\end{array}\right]$$\end{document}3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:cE=\left[\begin{array}{cccccc}166&\:73&\:76&\:0&\:0&\:0\\\:73&\:166&\:76&\:0&\:0&\:0\\\:76&\:76&\:142&\:0&\:0&\:0\\\:0&\:0&\:0&\:42&\:0&\:0\\\:0&\:0&\:0&\:0&\:42&\:0\\\:0&\:0&\:0&\:0&\:0&\:46\end{array}\right]\mathrm{GPa}$$\end{document}4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:eES=\left[\begin{array}{cccccc}0&\:0&\:0&\:0&\:7.6&\:0\\\:0&\:0&\:0&\:7.6&\:0&\:0\\\:-2.7&\:-2.7&\:8.3&\:0&\:4&\:4\end{array}\right]\mathrm{C.m}^{-2}$$\end{document} Polarization was evaluated along all three axes (x, y, z) while keeping the center fixed and electrically grounded. The applied pressure on the NPs was modeled as a sinusoidal oscillation ranging from 105 Pa to 108 Pa, with the force duration on the nanocrystals matching the period of ultrasonic vibrations (frequency = 106 Hz). Assessment of piezodynamic and chemodynamic ROS production In our study, ROS generation occurs via two distinct mechanisms: (1) piezocatalytic ROS production triggered by US stimulation, and (2) Mn2+-mediated Fenton-like reactions [52]. The first mechanism contributes to the rapid PZDT effect, while the second provides a sustained CDT effect. To detect ROS generated by the piezoelectric samples under US activation, a spectrometer equipped with electron spin resonance (ESR) technology was employed. DMPO served as the spin-trapping agent for •OH and •O2− radicals. ESR measurements were performed on three sample groups at standard temperature: a control group (DMPO), an MBBTO group (DMPO+MBBTO), and an MBBTO + US group (DMPO+MBBTO+US) [53]. Two different solvents were used to evaluate ROS production: DMSO for •O2− and water for •OH. The resulting ESR spectra were normalized to allow a clearer comparison of ROS levels, and ROS quantification was carried out via double integration of the spectra for each solvent [54]. Furthermore, piezocatalytic experiments were performed using Rhodamine B (Rh-B) dye as a model substrate [55]. Dye degradation was assessed using a Telesonic ultrasonic device operating at 40 kHz. For each experiment, 20 mL of Rh-B solution was placed in separate glass vials containing different piezocatalysts at a uniform concentration of 1 mg/mL (w/v). The solutions were subjected to US irradiation for up to 80 min, with 10 min interval sampling. At each interval, small aliquots were collected, and the catalysts were removed by centrifugation [22]. The optical density of the supernatants was measured at 554 nm using a UV-vis spectrophotometer to calculate the percentage of dye degradation and to determine the reaction kinetics. The degradation efficiency (Ƞ) and the 1 st -order kinetic constant (k) were calculated using the following equations:5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\upeta=\:\frac{{\mathrm{C}}_{0-}\mathrm{C}}{{\mathrm{C}}_{0}}\mathrm{x}100\mathrm{\%}$$\end{document}6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\mathrm{C=C}_0 \,\mathrm{exp(-kt)}$$\end{document} Here, C0 and C represent the initial and current concentrations of the dye solutions, respectively. The generation of was further evaluated using two specific ROS indicators: TMB for •OH detection and SOSG for 1O2 detection [48]. MBBTO NPs (100 µg/mL) were incubated with 200 µL of 100 µM H2O2 and 200 µL of 2 mM TMB, with TMB serving as a •OH indicator. After 30 min, the UV-vis-NIR spectra were recorded, and the absorbance of TMB at 652 nm was monitored to assess •OH production. Similarly, for 1O2 detection, SOSG (5 µM) and H2O2 (100 µM) were added to separate sample solutions, which were stirred in the dark for 1 h before FL measurements with excitation at 494 nm. Typically, the samples were categorized into three groups: PBS+H2O2, PBS+SOSG+H2O2, and MBBTO+SOSG+H2O2 to evaluate the ROS-generating capability of the MBBTO NPs. Additionally, •OH production was examined using methylene blue in the presence of HCO3− [56]. To investigate the role of Mn2+ in •OH generation, MBBTO (Mn2+: 8 mM, 100 µL) and MB (10 µg/mL) were mixed in 1 mL of 25 mM NaHCO3. Control samples, including MB alone and MB+MBBTO, were also prepared in 1 mL of 25 mM NaHCO3. All samples were incubated at normal temperature in the dark for 24 h. At designated time points, the residual MB was quantified by recording absorption spectra, demonstrating the chemodynamic ROS-generating capacity of the samples. Evaluation of the optical properties of the sample The absorption spectra of the synthesized samples were measured using a UV-vis spectrophotometer over a range of 400–1100 nm to assess the optical properties of DCNPs and DMBBTO [57]. Subsequently, the fluorescence spectra were measured with a spectrofluorometer (Fluoromax-Plus, HORIBA Scientific, USA). The samples were excited using a 980 nm laser, and emission spectra for both DCNPs and DMBBTO were collected in the 1100–1600 nm range [39]. The excitation spectrum (λex) of DCNPs was obtained using the same spectrofluorometer, with the emission monochromator fixed at the emission maximum (λem = 1523 nm) while scanning excitation from 960 to 1000 nm in 5 nm increments. The resulting excitation spectrum was normalized to unity at its peak to allow direct comparison with the UV-vis absorption profile of DCNPs. The NIR-II quantum yield (Φ) of DCNPs was measured using a relative method with IR-783 dye (Φ = 0.11) as a reference standard due to its similar absorption and emission characteristics in the NIR region. A series of dilutions of both DCNPs and IR-783 solutions was prepared, ensuring absorbance at 980 nm excitation remained below 0.1 to minimize reabsorption effects [39]. Absorption spectra were recorded with a UV-vis spectrophotometer, while FL emission spectra were collected under identical conditions using the spectrofluorometer. DCNPs and IR-783 were excited at 980 nm and 808 nm, respectively, and emission was recorded over 1100–1600 nm [58]. The integrated FL intensity for each dilution was plotted against its corresponding absorbance at the excitation wavelength to produce linear calibration curves for both DCNPs and IR-783. The slopes of these curves (SlopeDCNPs and SlopeIR−783) were determined by linear regression, and the relative quantum yield (ΦDCNPs) was calculated using the following equation:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:{\varPhi}_{DCNPs}={\varPhi}_{IR-783}\frac{{Slope}_{DCNPs}}{{Slope}_{IR-783}}{\left(\frac{{n}_{DCNPs}}{{n}_{IR-783}}\right)}^{2}$$\end{document} Here, SlopeDCNPs and SlopeIR−783 refer to the slopes of the FL intensity vs. absorbance plots for DCNPs and IR-783, respectively, while nDCNPs and nIR−783 represent the refractive indices of the solvents used for DCNPs and IR-783. In vitro NIR-II FL imaging was performed using a NIR-II FL imaging system (Teledyne, Princeton Instruments, USA) with a 980 nm excitation by using a 1300 nm long-pass filter with an exposure time of 1000 ms and a power density of 40 mW/cm2. Imaging was conducted for PBS, DCNPs, and DMBBTO samples [39]. The collected images were subsequently processed and analyzed using ImageJ software. In vitro MRI studies To evaluate the MR contrast-enhancing ability of the samples, in vitro MRI measurements were conducted. Aqueous solutions of the samples with varying Mn concentrations (0, 0.2, 0.4, 0.6, 0.8, and 1.0 mM) were prepared in 500 µL microtubes, with DI water serving as the control [39]. Transverse (T2) weighted phantom images were acquired using a Philips Elition 3.0T MRI system. The imaging was performed using a turbo spin echo (TSE) sequence for T2WI (TR/TE = 360/27 ms, NSA = 8, FOV = 31 × 31 × 15 mm, voxel size = 0.2 × 0.198 × 2.0 mm, slice thickness = 2.0 mm, matrix = 152 × 151 × 16 slices). The resulting in vitro T2-weighted images were analyzed using RadiAntViewer and ImageJ software [38]. In vitro biological experiments Cell viability assay U87 glioma cells were added to 96-well plates (1 × 104 cells/well) and incubated for 24 h at 37 °C. After that, cells were treated with CDMBBTO NPs (0 to 100 µg/mL) and cultured for an additional 24 h. Cells without NPs treatment served as the control group. Following treatment, 10 µL of MTT solution was dropped into each well, and the plates were incubated for 3–4 h at 37 °C to allow formazan crystal formation [59]. Then, 100 µL of DMSO was added to dissolve the formazan crystals. The plates were gently shaken, and absorbance was measured at 450 nm. Cell viability was then calculated using the following equation:8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Cell\:viability\:\left(\mathrm{\%}\right)=\left(\frac{{A}_{sample}-\:{A}_{0}}{{A}_{control}-\:{A}_{0}}\right)\times\:100$$\end{document} Here, Asample and Acontrol are the absorbance of the sample and control, respectively. A0 was the background absorbance of the well plate with the medium. Moreover, for evaluating the biocompatibility of F-127, the cell viability assay was repeated at different concentrations of F-127 polymer by following a similar procedure. Determination of PZDT/CDT-induced in vitro cytotoxicity U87 glioma cells were plated in 96-well plates (1 × 104 cells/well), incubated for 24 h at 37 °C. The wells were then randomly designated as: (1) PBS-US, (2) PBS+US, (3) BBTO-US, (4) BBTO+US, (5) CDMBBTO-US, and (6) CDMBBTO+US [60]. Groups designated for US exposure were irradiated under the following conditions: 1.0 MHz frequency, 1.2 W/cm2 intensity, 50% duty cycle, for 10 min. Cytotoxicity of the various samples toward U87 glioma cells was subsequently evaluated using the MTT assay. Hemolysis assay Erythrocyte suspensions (2%) were prepared from adult mouse blood and incubated with different concentrations of CDMBBTO (0 to 100 µg/mL) at 37 °C for 1 h. After incubation, the samples were centrifuged at 800 x g for 10 min, and the absorbance of the supernatant was measured at 540 nm to determine hemoglobin release [61]. The hemolysis rate was then calculated using the following equation:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Hemolysis\:\left(\mathrm{\%}\right)=\left(\frac{{A}_{sample}-\:{A}_{negative}}{{A}_{positive}-\:{A}_{negative}}\right)\times\:100$$\end{document} Erythrocytes treated with Triton X-100 (positive control) and those treated with PBS were used as the negative control. Samples causing less than 10% hemolysis were considered non-toxic. In vitro cellular uptake evaluation U87 glioma cells were seeded in glass-bottom dishes (1 × 105 cells/dish) and incubated for 24 h at 37 °C. After incubation, the glioma cells were added to the fresh medium containing FITC-labeled CDMBBTO and DMBBTO at different concentrations (10 to 100 µg/mL). Cells were then incubated for 1 to 12 h, washed three times, fixed with 4% paraformaldehyde, and washed again. The nuclei were stained with DAPI, and the samples were imaged at 488 nm laser excitation, and emission was collected within the 500–550 nm range [50]. In vitro blood-brain barrier (BBB) crossing assay A dual-chamber Transwell system was employed to simulate the BBB in vitro. The bEnd.3 cells were added into the upper chamber of the Transwell plate (5 × 104 cells/insert). After a 48-h incubation period, U87 cells were plated into the corresponding lower chamber (1 × 105 cells per well). The U87 glioma cells and bEnd.3 cells were then cocultured for a duration of 5 days [62]. After 5 days of incubation, the integrity of the bEnd.3 ell monolayer was verified by confirming a transepithelial electrical resistance (TEER) value exceeding 200 Ω.cm2. Following this, FITC-labeled CDMBBTO NPs were applied to the upper chamber. The system was incubated for a period of 24 h at 37 °C. The samples from the lower chamber were collected at different intervals, and their associated FL intensity was quantified [62]. In vitro evaluation of macrophage polarization For the in vitro assessment of macrophage polarization, RAW264.7 cells were plated in 6-well plates (5 × 105 cells/well) and allowed to incubate for 24 h. After that, the cells were treated with the following groups: (1) PBS-US, (2) DMBBTO-US (100 µg/mL), (3) CDMBBTO-US (100 µg/mL), (4) PBS+US, (5) DMBBTO+US (100 µg/mL), and (6) CDMBBTO+US (100 µg/mL). Following treatment, the macrophage cells were then harvested. At the same time, U87 cells were seeded into 6-well plates and cultured for 24 h until reaching 70–80% confluence [63]. Then, the harvested macrophage cells were treated with U87 cells. After an additional incubation period of 20 h, the cells were collected and subjected to a dual-staining procedure using annexin V-FITC (5 µg/mL) and PI (100 µg/mL) and analyzed by flow cytometry [63]. Intracellular ROS production U87 glioma cells were seeded in confocal glass-bottom dishes (1 × 105 cells/dish) and incubated for 24 h at 37 °C. After that the cells were treated with one of the following groups: (1) PBS-US, (2) F-127-US (100 µg/mL) (3) DMBBTO-US (100 µg/mL), (4) CDMBBTO-US (100 µg/mL), (5) PBS+US, (6) F-127+US (100 µg/mL), (7) DMBBTO+US (100 µg/mL), and (8) CDMBBTO+US (100 µg/mL). Following treatment, the cells were incubated for 4 h at 37 °C. To detect ROS, cells were subsequently incubated with DCFH-DA for 40 min in the dark and washed twice. Intracellular ROS generation was then observed using confocal laser scanning microscopy (CLSM) [39]. Moreover, for comparing the piezodynamic and chemodynamic intracellular ROS production, the experiment was repeated for CDMBBTO under different conditions: (1) CDMBBTO-H2O2, (2) CDMBBTO-US, (3) CDMBBTO+H2O2, and (4) CDMBBTO+US. In vitro live/dead assay U87 glioma cells were cultured in 6-well plates for 24 h until reaching 70–80% confluence. The cells were then treated with the following groups: (1) PBS-US, (2) F-127-US (100 µg/mL) (3) DMBBTO-US (100 µg/mL), (4) CDMBBTO-US (100 µg/mL), (5) PBS+US, (6) F-127+US (100 µg/mL), (7) DMBBTO+US (100 µg/mL), and (8) CDMBBTO+US (100 µg/mL), and co-incubated with the respective treatments for 8 h at 37 °C. After treatment, cells were stained with Calcein-AM and PI for 30 min at 37 °C to distinguish live from dead cells [22]. Excess dye was removed by gently washing with PBS, and the stained cells were then visualized using a FL microscope. In vivo biological experiments Development of in vivo subcutaneous U87 glioma-bearing mice models Female BALB/c nude mice (5–6 weeks old, 17–20 g) were obtained from Guangdong Yuyuan Biotechnology Co., Ltd. (China). All experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of Shenzhen University (Approval No. IACUC-202400090). The animals were maintained under controlled conditions, with a temperature of 25 °C and relative humidity of 55%, and were provided with unrestricted access to standard laboratory chow and water [39]. These conditions ensured a stable, consistent, and humane environment throughout the study. To generate U87 glioma-bearing mouse models, each mouse was subcutaneously injected with 1 × 106 U87 glioma cells in 100 µL of PBS. About 7 days after injection, when tumor volumes reached 50–70 mm3, the mice were considered ready for subsequent in vivo experiments. In vivo NIR-II FL imaging of the glioma-bearing mice models Female BALB/c mice (n = 5) were used for NIR-II fluorescence imaging studies. 7 days after tumor inoculation, the mice were intravenously administered 200 µL of CDMBBTO solution (10 mg/kg body weight). NIR-II FL images were recorded at specified time points before and after injection, up to 12 h, using a NIR-II imaging system (Teledyne, Princeton Instruments, USA; excitation 980 nm, long-pass filter 1300 nm, exposure time 1000 ms, power density 40 mW/cm2) [39]. To assess biodistribution, the mice were sacrificed 12 h post-injection, and major organs, including the heart, liver, spleen, lungs, kidneys, and tumor, were collected for ex vivo FL imaging. All images were then processed and analyzed using ImageJ software. In vivo MRI of the glioma-bearing mice models Female BALB/c mice (n = 5) were selected for in vivo MRI studies. 7 days after tumor cell inoculation, the mice were intravenously injected with CDMBBTO solution at a dose of 10 mg/kg body weight. MRI scans were performed using a 3 T clinical MRI system equipped with a high-performance 3 T/m gradient coil, and a custom-designed animal imaging coil was employed to improve signal-to-noise ratio and resolution for small animal imaging. During imaging, the mice were anesthetized with 2% isoflurane in medical-grade oxygen at a flow rate of 1 L/min and placed prone on an MRI-compatible animal bed [64]. T2-weighted images of the U87 glioma-bearing mice were acquired at multiple time points up to 12 h using a Philips Elition 3.0 T MRI system. The primary imaging sequence was turbo spin echo (TSE) T2WI (TR/TE = 360/27 ms, NSA = 8, FOV = 31 × 31 × 15 mm, voxel size = 0.2 × 0.198 × 2.0 mm, slice thickness = 2.0 mm, matrix = 152 × 151 × 16). Acquired in vivo T2 images were analyzed using RadiAntViewer and ImageJ software. In vivo PZDT/CDT of the glioma-bearing mice models When the U87 tumor-bearing female BALB/c mice reached tumor sizes of approximately 40–60 mm3, they were randomly divided into four groups (n = 5) for therapeutic evaluation: (1) PBS-US, (2) PBS+US, (3) CDMBBTO-US, and (4) CDMBBTO+US. The treatment procedures for each group were as follows: (1) mice in the PBS-US group received an intravenous injection of PBS only; (2) mice in the PBS+US group received PBS intravenously and were exposed to ultrasound (1.5 W/cm2, 50% duty cycle, 1 MHz, 10 min) 6 h post-injection; (3) the CDMBBTO-US group received only an intravenous injection of CDMBBTO (10 mg/kg body weight); (4) the CDMBBTO+US group received intravenous CDMBBTO followed by US treatment (same parameters) 6 h post-injection [65]. For groups 3 and 4, CDMBBTO injections were administered on days 1, 3, and 5, with US applied during PZDT precisely focused on the tumor center. Tumor volumes and body weights were monitored every two days up to day 14.10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:Tumor\:volume\:\left({mm}^{3}\right)=\:\frac{1}{2}\:\times\:length\:\times\:\:{\left(width\right)}^{2}$$\end{document} To assess the biosafety of the treatment, the survival of the injected mice was monitored for up to 30 days. At the end of the study, tumors and major organs, including the heart, liver, spleen, lungs, and kidneys, were harvested and processed for hematoxylin and eosin (H&E) staining. On day 30, all mice were euthanized, and the same organs, along with tumor tissues, were collected for histological examination using H&E staining. H&E staining Tissue samples from the different treatment groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US) were prepared for H&E staining following the standard protocol from BBC Biochemical Company. Frozen tissue Sect. (8 μm thick) were first fixed in 10% formalin at room temperature for 30 min. After fixation, the slides were rinsed under running water for 5 min to remove residual formalin. The sections were then dehydrated through a graded ethanol series, sequentially immersing them in 100%, 95%, and 70% ethanol for 20 s each, to preserve tissue structure [66]. Subsequently, the slides were stained with hematoxylin for 3 min, rinsed in water for 1 min to remove excess dye, and counterstained with eosin for 1 min. After staining, the slides were washed in xylene and mounted with Canada balsam. Finally, the stained tissues were examined and imaged to obtain high-resolution histological images. TUNEL staining Tumor samples collected from the various treatment groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US) were first fixed in 4% paraformaldehyde for 30 min to preserve cellular structures. After washing, the samples were treated with 0.1% Triton X-100 for 5–10 min to enhance cell membrane permeability [52, 66]. The sections were then incubated with the TUNEL reaction mixture, which included the TdT enzyme and fluorescently labeled dUTP, following the manufacturer’s instructions. Samples were incubated at 37 °C for 1–2 h, then washed to remove unbound TdT and dUTP. After gently drying under light refrigeration, the sections were sealed with an anti-fluorescence mounting medium. Finally, the samples were examined and imaged using a fluorescence microscope. Immunofluorescence studies Tumor tissues from the different treatment groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US) were first rinsed with cold PBS to remove residual blood. Approximately 50–100 mg of tumor tissue was homogenized on ice in 9 volumes of PBS containing protease inhibitors using a glass homogenizer, followed by brief sonication. The homogenates were then centrifuged, and the resulting supernatants were collected. Total protein contents were determined using a BCA assay and normalized to ≤ 0.3 mg protein per well. Levels of TNF‑α and IL‑6 were measured using Mouse TNF‑α and IL‑6 uncoated ELISA kits, following the manufacturer’s instructions, with 100 µL of sample or standard per well and an incubation period of 4.5 h [47]. Optical density was read at 450 nm (corrected at 570 nm), and cytokine concentrations (pg/mL) were calculated from standard curves for each group to assess inflammatory upregulation. For immunofluorescence analysis, tumor samples from the same groups were fixed in 4% paraformaldehyde and cryo-sectioned. The sections were permeabilized and blocked with 3% BSA for 1 h, followed by overnight incubation at 4 °C with primary antibodies against iNOS (M1 marker) and CD206 (M2 marker) [50]. HRP-conjugated secondary antibodies were then applied for 50 min at room temperature, and nuclei were counterstained with DAPI. Quantitative analysis was performed in ImageJ to determine mean FL intensity per cell for iNOS and CD206, and the iNOS/CD206 ratio was calculated to evaluate macrophage repolarization from M2 to M1 [39]. Flow cytometry analysis for immunogenic death cell marker expression and macrophage polarization U87 cells were cultured in 24 well plates and left overnight to adhere. After 24 h of treatment with PBS, DMBBTO, and CDMBBTO, the cells were harvested. The flow cytometric analysis was performed to measure surface calreticulin (CRT) expression, with a Fixable Viability Dye 450 (Thermo Fisher, 65–0863- 18), and Calreticulin-PE antibody (CST, 19780). To prepare a cell suspension from tumor tissue, the sample was first minced into small fragments. These fragments were digested for 1 h at 37 °C in a 5 mL culture medium containing a cocktail of collagenase (0.2 mg/mL), DNase (25 U/mL), and hyaluronidase (0.1 mg/mL). Mechanical dissociation was then performed using a Dissociator. The resulting digestate was passed through a 70 μm filter to remove any remaining debris. The filtered cells 100 µL were subsequently incubated in the dark for 15 min at room temperature with a panel of fluorescently conjugated antibodies: CD3-BV650, CD8-APC, PE-conjugated anti-mouse CD206, and FITC-conjugated anti-mouse CD86. This was followed by washing and resuspension in 100 µL PBS. CD3, CD8+, CD86, and CD206 expression levels were determined using a Cyto-FLEX flow cytometer, and the data were analyzed using Flow-Jo software [67]. Blood biochemistry studies Blood samples were collected from the retro-orbital sinus of anesthetized BALB/c U87 glioma-bearing mice in all treatment groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US) into heparinized tubes. The samples were centrifuged to separate plasma [48]. Standard clinical chemistry procedures were followed using commercial reagent kits: creatinine and uric acid levels were determined through enzymatic colorimetric assays, while activities of ALT, AST, and ALP were measured using kinetic UV-transfer methods in accordance with IFCC guidelines [39]. Results are presented in conventional units (µmol/L for creatinine and uric acid; U/L for enzyme activities). In vivo biosafety analysis For the in vivo biodistribution study, mice were intravenously administered a single dose of CDMBBTO (25 mg/kg) and sacrificed after 14 days. Major body organs such as liver, kidney, heart, spleen, and lungs were harvested, weighed, and subjected to acid digestion using a mixture of concentrated HNO3 and H2O2 [68]. The digests were diluted and analyzed by ICP-MS to quantify Ba, Bi, and Mn content. The biosafety data were calculated and expressed as µg of Ba, Bi, and Mn deposited per gram of tissue (µg/g) for each organ [69]. Statistical analysis All experimental data are presented as mean ± SD from a minimum of three replicates. Statistical significance was evaluated by using one-way analysis of variance (ANOVA) followed by Student’s t-test. A significance threshold of α = 0.05 was applied, with p-values denoted as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference. Results and discussion Synthesis and physicochemical characterization of the DMBBTO nanoplatform Self-assembled nanotheranostic platforms have recently emerged as adaptive and intelligent systems, capable of autonomously organizing into well-defined architectures through non-covalent interactions [70]. Their programmable structural design enables the seamless cointegration of diagnostic and therapeutic modalities, presenting an ideal framework for precision oriented interventions such as glioma therapy [71]. In particular, embedding both NIR‐II FL and MR imaging within a self‐assembled construct offers deep tissue penetration, high spatial resolution, and real time intraoperative guidance within the complex TME, thereby allowing accurate tumor delineation, responsive treatment modulation, and dynamic feedback well beyond the capabilities of conventional single modality approaches. In order to create interstitial defect states inside the perovskite lattice, BTO precursors were co-introduced with stoichiometric (5%) quantities of Mn2+ and Bi3+ to create MBBTO NPs using a traditional solvothermal process. Multifunctional DMBBTO nanoplatforms with optimized piezocatalytic output under dual NIR-II FL and MR imaging navigation were next produced by co-assembling the synthesized NIR-II emissive DCNPs with the preformed MBBTO NPs in an amphiphilic Pluronic F-127 matrix. The intrinsically non-emissive piezoelectric component is successfully interfaced with NIR-II fluorophores by using this polymer orchestrated self-assembly strategy. This solves the long-standing problem of optical darkness in piezoelectric materials, which has traditionally required transparent alternatives or limited their compatibility with deep tissue optical imaging. Additionally, the U87 glioma cell membrane is used to camouflage the DMBBTO nanoplatform, giving it homologous targeting capabilities and allowing highly precise accumulation within glioma lesions for precision therapy. The crystallographic structures of DCNPs, BBTO, MBBTO, and DMBBTO were examined using XRD. As shown in Fig. 1A and Figure S1A, all samples have exhibited well defined diffraction peaks that can be indexed to the standard tetragonal perovskite phase of BTO (JCPDS No. 83–1800) and hexagonal phase of DCNPs (JCPDS No. 00–064-0156), indicating the preservation of the crystal structure after sequential doping and self-assembly [72]. No impurity or secondary phase peaks are detected during the Rietveld refinement, validating the successful incorporation of Mn2+ and Bi3+ into the BTO lattice without structural collapse (Fig. 1B and Figure S1B-D). A slight peak broadening and leftward shift have been observed in BBTO, MBBTO, and DMBBTO. This shift can be attributed to Bi3+ induced lattice strain arising from ionic radius mismatch (Bi3+ = 103 pm < Ba2+ = 135 pm), introducing local distortion (Fig. 1D), as well as Mn2+ induced oxygen vacancies (Fig. 1E), altering the Ti-O bond length and generating microstructural disorder (Table S1). These structural perturbations are highly desirable, as defect engineering is known to enhance polarization asymmetry, thereby improving the piezoelectric response, a key prerequisite for efficient PZDT activation [73]. According to the consistent diffraction pattern in DMBBTO, the self-assembly using DCNPs and F-127 did not alter the intrinsic crystallinity of MBBTO. FTIR spectroscopy has been used to further examine the bonding environment of DCNPs, BBTO, MBBTO, DMBBTO, and CDMBBTO (Fig. 1C and Figure S1E) [74]. Every sample showed distinctive absorption bands that matched the BTO perovskite structure. The Ti-O-Ti stretching vibrations have been identified as the source of the strong bands in the 800–1000 cm− 1 area, indicating that the BTO crystal structure was preserved following doping and self-assembly [22]. Notably, local symmetry distortion brought on by microstrain created by Bi3+ doping and oxygen vacancy creation triggered by Mn2+ can be responsible for the small shifts and intensity modulations in these lattice vibration bands. The appearance of additional bands in the 2800–3000 cm− 1 region in DMBBTO corresponds to the C-H stretching vibration of -CH2/-CH3 groups from the PPO block of Pluronic F-127, confirming the successful self-assembly within the polymeric matrix [75]. Mechanistically, the amphiphilic F-127 facilitates assembly by preferential adsorption of its hydrophobic PPO segments onto relatively hydrophobic, defect-rich surface domains of MBBTO and surface-ligand regions of DCNPs through hydrophobic interactions and van der Waals forces, thereby acting as a physical bridge between the two components [75]. On the other hand, the hydrophilic PEO chains extend outward, providing steric stabilization and preventing direct interparticle contact. This interaction is further stabilized by hydrogen bonding between PEO ether oxygens and surface -OH groups on MBBTO, as evidenced by the slight broadening of the O-H stretching band in the 3300–3500 cm− 1 region. Furthermore, the broad absorption band around 3300–3500 cm− 1 is also associated with O-H and N-H stretching vibrations from the membrane proteins and phospholipid groups [39]. In addition, the peak near ~2000–2300 cm− 1 originates from C = O stretching and N-H bending of the membrane proteins, respectively, and the band around ~1630 cm− 1 is associated with C-N stretching and COO− vibrations. Overall, the FTIR spectra corroborate the XRD findings, verifying structural stability of the BTO host, while simultaneously confirming defect induction, polymer mediated self-assembly, and membrane camouflaging. Fig. 1Physicochemical characterizations of the prepared nanoplatform. (A) XRD and (B) Rietveld refinement spectra illustrate the crystallinity and phase purity of BBTO, MBBTO, and DMBBTO. (C) FTIR spectra demonstrate the bonding networks of the prepared samples. Demonstration of the microstructure of (D) BBTO shows Bi-induced microstrain, and (E) MBBTO shows the Mn induced oxygen vacancies in the crystal lattice. (F) TEM images illustrate (i) the spherical shaped DCNPs (scale bar: 20 nm), (ii) high resolution interplanar spacing in hexagonal DCNPs (scale bar: 5 nm), (iii) spherical MBBTO NPs (scale bar: 20 nm), (iv) self-assembly of MBBTO and DCNPs (DMBBTO; scale bar: 20 nm), and (v-vi) U87 glioma cell membrane camouflaged DMBBTO (CDMBBTO). The blue colored particles represent MBBTO, and red colored particles represent DCNPs. The TEM image has been post-processed and pseudo colored to demonstrate the cell membrane thickness marked in red (scale bar: 100 nm). (G) EDS mapping shows the presence of component elements in DMBBTO self-assembly (scale bar: 100 nm). (H) Zeta potential demonstrates the surface charge density of BBTO, MBBTO, DMBBTO, and CDMBBTO (S.D. ±0.61 mV). (I) XPS survey illustrates the composition and chemical states of the component elements in DMBBTO. High-resolution XPS spectra of DMBBTO show the chemical states of metal cations as (J) Mn 2p3/2 and Mn 2p1/2, (K) Ba 3d5/2 and Ba 3d3/2, and (L) Bi 4f7/2 and Bi 4f5/2 Physicochemical characterizations of the prepared nanoplatform. (A) XRD and (B) Rietveld refinement spectra illustrate the crystallinity and phase purity of BBTO, MBBTO, and DMBBTO. (C) FTIR spectra demonstrate the bonding networks of the prepared samples. Demonstration of the microstructure of (D) BBTO shows Bi-induced microstrain, and (E) MBBTO shows the Mn induced oxygen vacancies in the crystal lattice. (F) TEM images illustrate (i) the spherical shaped DCNPs (scale bar: 20 nm), (ii) high resolution interplanar spacing in hexagonal DCNPs (scale bar: 5 nm), (iii) spherical MBBTO NPs (scale bar: 20 nm), (iv) self-assembly of MBBTO and DCNPs (DMBBTO; scale bar: 20 nm), and (v-vi) U87 glioma cell membrane camouflaged DMBBTO (CDMBBTO). The blue colored particles represent MBBTO, and red colored particles represent DCNPs. The TEM image has been post-processed and pseudo colored to demonstrate the cell membrane thickness marked in red (scale bar: 100 nm). (G) EDS mapping shows the presence of component elements in DMBBTO self-assembly (scale bar: 100 nm). (H) Zeta potential demonstrates the surface charge density of BBTO, MBBTO, DMBBTO, and CDMBBTO (S.D. ±0.61 mV). (I) XPS survey illustrates the composition and chemical states of the component elements in DMBBTO. High-resolution XPS spectra of DMBBTO show the chemical states of metal cations as (J) Mn 2p3/2 and Mn 2p1/2, (K) Ba 3d5/2 and Ba 3d3/2, and (L) Bi 4f7/2 and Bi 4f5/2 TEM imaging has been used to analyze the sequential creation and morphological evolution of the nanoplatform. The development of crystalline NPs is confirmed by the consistent spherical shape of DCNPs (~10–15 nm), as seen in Fig. 1Fi. The highly ordered structure responsible for NIR-II FL emission was evident in the distinctive hexagonal crystalline structure of DCNPs with well-resolved lattice fringes (Fig. 1Fii). Additionally, the MBBTO displays spherical-shaped NPs that are about 7–10 nm in size (Fig. 1Fiii). Upon self-assembly, Fig. 1Fiv and Figure S16A-B revealed the formation of hybrid DMBBTO mosaic nanostructures, where MBBTO and DCNPs are closely interfaced within the Pluronic F-127 matrix, confirming spontaneous nanoscale integration mediated by noncovalent interactions [76]. Importantly, the final biomimetic coating shown in Fig. 1Fv-vi and Figure S16C-D has demonstrated a hazy, regular outer shell surrounding the DCNPs and MBBTO self-assembly, characteristic of U87 glioma cell membrane cloaking, indicating successful camouflage. To further confirm the glioma membrane camouflage, the isolated U87 cell membranes and CDMBBTO nanoplatforms have been assessed via SDS-Page analysis (Figure S1 3 A) [76]. Moreover, the Western blot analysis further confirmed the presence of E-cadherin and the absence of β-actin, demonstrating efficient membrane extraction and successful biomimetic camouflaging (Figure S1 3B-C). Furthermore, EDS elemental mapping confirmed the presence of all the component elements and demonstrated their even distribution throughout the DMBBTO nanoplatform in Fig. 1G, Figure S1F, and Figure S2. The surface charge evolution during each construction step has been monitored via zeta potential measurements (Fig. 1H). BBTO displayed a negatively charged surface (~−19 mV), which became less negative (~−17.5 mV) after Mn/Bi codoping (MBBTO) due to the generation of oxygen vacancies and exposed lattice defects. Following self-assembly with Pluronic F-127, the DMBBTO nanoplatform exhibited a partially neutralized surface charge of ~−13.5 mV, confirming polymer adsorption and improved colloidal stability. Finally, U87 glioma cell membrane cloaking (CDMBBTO) resulted in a slight shift in zeta potential (~−12 mV), characteristic of biomimetic stealth behavior, favoring immune evasion and tumor specific homologous targeting. After that, the colloidal stability of the nanostructure in the physiological environments has been evaluated in PBS, FBS, and DMEM over 7 days (Figure S3A). No visible aggregation or precipitation has been observed in any medium throughout the incubation period, indicating good macroscopic dispersion stability. The zeta potential remained nearly unchanged (−13 to −16 mV) across all the media and time points, demonstrating preserved surface charge and electrostatic stability (Figure S3B) [77]. In parallel, DLS analysis further confirmed this behavior, with hydrodynamic size distribution remaining stable over time, as reflected by consistently low PDI values (~0.20 to 0.35) (Figure S3C). Although a slight increase in PDI has been observed at longer time points in serum, this could be attributed to protein adsorption rather than particle aggregation [77]. Moreover, the reproducibility of the nanoplatform has also been systematically evaluated across three independent batches by varying precursor concentration, stirring speed, and stirring duration, with consistent zeta potential and PDI values confirming robust batch-to-batch uniformity and colloidal stability (Figure S6-S7). High resolution XPS spectra have verified the chemical states and defect-engineered composition (Fig. 1I-L). The presence of Mn2+ has been verified by the Mn 2p3/2 and 2p1/2 peaks, confirming its function in creating oxygen vacancies and facilitating CDT/Fenton-like action. The structural stability of the BTO host was indicated by the consistency of Ba 3 d spectra with Ba2+ [74]. Meanwhile, the Bi 4f7/2 and 4f5/2 peaks have confirmed the successful incorporation of Bi3+, supporting the microstrain induced piezoelectric enhancement strategy [78]. Therefore, the XPS results have confirmed successful defect engineering at the electronic level, aligning with the intended multifunctional PZDT/CDT activation. The successful coating of the DMBBTO nanoplatform with the U87 glioma cell membrane has also been confirmed by fluorescence microscopy (Fig. 2A). FITC-labeled membrane has exhibited a clear and continuous green fluorescence surrounding the nanoplatform, indicating efficient membrane camouflaging [79]. Notably, the stable fluorescence intensity up to 12 h has demonstrated the stability and structural integrity of the membrane coating over time. The strong spatial overlap observed in the merged images has further confirmed the uniform membrane coverage rather than nonspecific adsorption. Moreover, to evaluate the targeting ability and BBB transport capability of CDMBBTO, an in vitro transwell BBB model has been employed by using bEnd.3 endothelial cells and U87 glioma cells (Fig. 2B). Quantitative analysis revealed a clear time-dependent increase in the transport efficiency of FITC-labeled CDMBBTO across the BBB model (Fig. 2C). Minimal transport has been observed at early time points (30 min to 3 h), while a marked enhancement occurred after 6 h, reaching a maximum at 12 to 24 h. This behavior indicates that membrane camouflage effectively facilitates transcytosis and cellular interaction with brain endothelial cells [80]. This homologous targeting and improved transport capability are critical for achieving effective glioma accumulation, thereby supporting the suitability of the CDMBBTO nanoplatform for brain tumor-targeted theranostic applications. Fig. 2(A) FITC labelled U87 glioma cell membrane demonstrates the coating efficiency on DMBBTO nanoplatform (scale bar: 25 μm). (B) Graphical representation of BBB crossing of FITC labelled CDMBBTO. (C) BBB transport (%) of FITC labelled CDMBBTO platform at different time intervals (30 min, 1 h, 3 h, 6 h, 12 h, and 24 h). The p-values are calculated using ordinary one-way ANOVA and Student t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference (A) FITC labelled U87 glioma cell membrane demonstrates the coating efficiency on DMBBTO nanoplatform (scale bar: 25 μm). (B) Graphical representation of BBB crossing of FITC labelled CDMBBTO. (C) BBB transport (%) of FITC labelled CDMBBTO platform at different time intervals (30 min, 1 h, 3 h, 6 h, 12 h, and 24 h). The p-values are calculated using ordinary one-way ANOVA and Student t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference Collectively, these physicochemical characterizations unequivocally confirmed the successful construction of a defect-engineered, self-assembled, and biomimetic CDMBBTO nanoplatform with preserved crystallinity, tunable surface charge, and functionally activated electronic states, establishing a solid foundation for its subsequent application in targeted dual modal NIR-II FL and MR imaging guided PZDT/CDT of glioma. Piezoelectric response and PZDT/CDT ROS generation via DMBBTO nanoplatform The intrinsic piezoelectric behavior of DMBBTO has been directly evaluated via PFM analysis [81]. As shown in Fig. 3A, the DMBBTO nanoplatform exhibited a characteristic butterfly shaped amplitude vs. voltage loop, confirming reversible piezoelectric deformation in response to the applied electric field, a hallmark of strong electromechanical coupling. The corresponding 2D amplitude mapping (Fig. 3B) has revealed a homogeneous nanoscale piezoelectric response distribution, which may indicate the uniform defect-mediated polarization within the nanostructure. The d33 is found to be 106.91 pmV− 1, which we measured from the PFM amplitude loop. Such a high d33 value in perovskite structures is very uncommon and could be attributed to the dual defect caused by Bi3+ and Mn2+ codoping (Table S2). Furthermore, a 180° phase hysteresis loop has been observed in Fig. 3C, which confirms the switchable ferroelectric domains arising from stable spontaneous polarization. The high contrast nanoscale phase image in Fig. 3D further validated well defined domain orientation, reflecting strong intrinsic polarization asymmetry. For verifying the reproducibility of the self-assembly, the PFM analysis has been repeated up to three different batches prepared under similar conditions (Figure S8A-C). All the batches demonstrated almost comparable piezoelectric response, confirming the reproducibility of the DMBBTO nanoplatform [40]. The measurable piezoelectric voltage output (~ ± 1.5 V) from the DMBBTO nanoplatform under mechanical stimulation has directly confirmed its intrinsic piezoelectric behavior, demonstrating efficient conversion of external mechanical energy into electrical signals (Figure S8G). To validate the piezoelectric response of the DMBBTO nanoplatform at the nanoscale, a 3D FEM simulation has been performed in COMSOL Multiphysics using the piezoelectric devices module [50]. The multiphysics coupling of elastic stress and electrical polarization revealed a strong local deformation-induced potential distribution (piezopotential) at the curvature edges and crystal facet junctions, consistent with flexoelectric enhancement in ultrasmall nanodomains (Fig. 3E-H). The simulated open circuit piezopotential reached up to ~ ± 1.8 Cm− 2 and ±0.18 Cm− 2 in DMBBTO and CDMBBTO, respectively, which is sufficient to catalytically trigger redox reactions or accelerate interfacial charge carrier (electron-hole) separation for ROS production. Moreover, FEM simulations reveal a pressure-dependent enhancement of piezoelectric polarization in DMBBTO, with progressively stronger polarization generated as the applied stress increases from 0.5 to 2 MPa, confirming efficient mechanical to electrical energy conversion under clinically relevant US pressures (Figure S4). Importantly, these simulations therefore provided quantitative electromechanical evidence that the self-assembled piezoelectric nanotherapeutic system can efficiently harvest mechanical energy and transduce it into bioelectrically active potential in the TME. Further, the EPR spectroscopy has directly confirmed the US triggered ROS generation from the self-assembled piezoelectric DMBBTO nanoplatforms. The spin trapping agent DMPO captured the short-lived free radicals (•OH and •O2−) in water and DMSO, respectively. Upon US irradiation, the characteristic signals corresponding to the DMPO-•OH adduct and DMPO-•O2− have been observed, confirming the generation of hydroxyl (•OH) radicals and superoxide (•O2−) radicals (Fig. 3I-J) [82]. The significantly enhanced EPR signal intensity in the presence of the DMBBTO nanoplatforms compared to the control confirmed that the piezoelectric materials efficiently convert mechanical energy into catalytic redox active charge carriers, verifying the continuous piezocatalytic ROS production under sustained US stimulation. Therefore, the EPR results have provided direct molecular evidence that the designed self-assembled nanoplatforms can act as robust piezoelectric nanoenzymes, capable of generating cytotoxic ROS species in situ. The piezocatalytic ROS production has further been validated by the RhB dye degradation assay to quantitatively evaluate the piezocatalytic ROS production by BTO, BBTO, DMBBTO, and CDMBBTO nanoplatforms under US stimulation. As shown in Fig. 3K and Figure S5A-B, the UV-vis absorption spectra exhibited a progressive decrease in the RhB absorption peak intensity with increasing US time, confirming rapid degradation triggered by US induced piezocatalytic ROS production [83, 84]. The pseudo first order kinetic study for RhB degradation of CDMBBTO has exhibited the highest slope and rate constant compared to the pristine BTO and BBTO, indicating its superior piezocatalytic performance due to enhanced polarization and charge separation (Fig. 3L-M and Figure S5C). After that, the dye degradation comparison of DMBBTO with the cancer cell membrane has demonstrated that the membrane coating does not exhibit any inherent piezocatalytic activity, confirming that the catalytic activity originates from the DMBBTO nanoplatform (Figure S5D-G). Moreover, the chemodynamic ROS production ability of the nanoplatform has been evaluated through Fenton like catalytic reactions. The TMB resulted in a strong absorbance signal over time and directly confirmed the •OH radical generation (Fig. 3N) [85]. Similarly, singlet oxygen (1O2) generation has been analyzed using the SOSG probe as shown in Fig. 3O [86]. The significant signal amplification in both assays indicated that the nanoplatform can efficiently catalyze H2O2 into ROS even in the absence of external energy input, validating its strong chemodynamic capability. Fig. 3Evaluation of the piezoelectric properties and the piezo-chemodynamic ROS production ability of the DMBBTO nanoplatform. (A) Butterfly like amplitude loop with (B) corresponding 2D amplitude micrograph and (C) Phase hysteresis loop with (D) corresponding 2D phase pattern of DMBBTO measured by PFM demonstrates the piezoelectric behavior of DMBBTO. The FEM simulation using COMSOL Multiphysics demonstrates the (E) 100 MPa of applied stress and the resulting piezoelectric surface polarization in (F) DMBBTO and (G) CDMBBTO nanospheres. (H) Demonstration of piezoelectric surface polarization in DMBBTO nanospheres along x-, y-, and z-axis under different periodic forces ranging from 0 to 100 MPa. The EPR spectra of DMBBTO represent the ROS production in (I) water (•OH) and (J) DMSO (•O2−) under US stimulation. (K) Piezocatalytic dye degradation (RhB) by CDMBBTO nanoplatform under US stimulation up to 60 min. (L, M) Dye degradation kinetics for CDMBBTO as compared to BTO and BBTO up to 60 min (S.D. ±0.002 min− 1). Estimation of •OH radicals and 1O2 production via Fenton like reaction to evaluate the chemodynamic efficiency of DMBBTO by (N) TMB assay and (O) SOSG assay, respectively. (I) Estimation of chemodynamic methylene blue dye degradation ability of DMBBTO in H2O2 environment Evaluation of the piezoelectric properties and the piezo-chemodynamic ROS production ability of the DMBBTO nanoplatform. (A) Butterfly like amplitude loop with (B) corresponding 2D amplitude micrograph and (C) Phase hysteresis loop with (D) corresponding 2D phase pattern of DMBBTO measured by PFM demonstrates the piezoelectric behavior of DMBBTO. The FEM simulation using COMSOL Multiphysics demonstrates the (E) 100 MPa of applied stress and the resulting piezoelectric surface polarization in (F) DMBBTO and (G) CDMBBTO nanospheres. (H) Demonstration of piezoelectric surface polarization in DMBBTO nanospheres along x-, y-, and z-axis under different periodic forces ranging from 0 to 100 MPa. The EPR spectra of DMBBTO represent the ROS production in (I) water (•OH) and (J) DMSO (•O2−) under US stimulation. (K) Piezocatalytic dye degradation (RhB) by CDMBBTO nanoplatform under US stimulation up to 60 min. (L, M) Dye degradation kinetics for CDMBBTO as compared to BTO and BBTO up to 60 min (S.D. ±0.002 min− 1). Estimation of •OH radicals and 1O2 production via Fenton like reaction to evaluate the chemodynamic efficiency of DMBBTO by (N) TMB assay and (O) SOSG assay, respectively. (I) Estimation of chemodynamic methylene blue dye degradation ability of DMBBTO in H2O2 environment Furthermore, the methylene blue (MB) degradation test has demonstrated the destructive action of these ROS on organic molecules under H2O2 rich conditions (Fig. 3P). The gradual fading of MB has confirmed its oxidative breakdown via Fenton triggered ROS [87]. Therefore, Mn2+ doping plays a dual functional role in enhancing both piezodynamic and chemodynamic therapeutic performance of BBTO. Incorporation of Mn2+ into the BBTO lattice modifies the electronic structure and expands the bandgap, which reduces electron-hole recombination and promotes efficient piezoelectric charge separation under US stimulation. The enhanced polarization consequently increases ROS generation during the piezodynamic process. Simultaneously, loosely bound surface-associated Mn2+ ions could be released under US mechanical disturbance and within the acidic TME. The released Mn2+ subsequently participates in Mn2+/Mn3+ redox cycling, catalyzing H2O2 decomposition through a Fenton-like reaction and thereby enhancing CDT activity, as confirmed by SOSG and TMB assays. Importantly, the piezoelectric potential does not directly accelerate the Mn2+/Mn3+ redox cycle; instead, PZDT and CDT act as complementary yet mechanistically independent processes, collectively contributing to the superior therapeutic efficacy of the DMBBTO nanoplatform. Evaluation of dual NIR-II FL and MR imaging capability of DMBBTO nanoplatform To investigate the dual modal imaging performance of the self-assembled nanoplatform, the NIR-II FL and T2 weighted MR imaging capabilities of CDMBBTO have been systematically evaluated in vitro and in vivo. As depicted in Fig. 4A, the UV-vis-NIR absorption spectra confirmed that the introduction of MBBTO into DCNPs has not compromised the intrinsic broadband NIR absorption of the DCNPs, an essential prerequisite for deep tissue imaging in the NIR-II biological window. Upon 980 nm laser excitation, CDMBBTO exhibited a strong emission peak at approximately 1525 nm (Fig. 4B), characteristic of Yb3+ (2F7/2 → 2F5/2) to Er3+ (2F5/2 → 4I15/2 → 4I15/2) transitions, which has been further validated by the excitation spectra of DCNPs taken at 1525 nm (Fig. 4C). Additionally, the comparable FL intensity of different batches of DMBBTO prepared under similar conditions has demonstrated the stability and reproducibility of the self-assembly (Figure S8D-F). The results were nearly similar to pristine DCNPs possessing a Φ of 13.76 as compared to the IR783 dye (Figure S9A-B) [88]. This retention of high NIR-II emission intensity has confirmed that F-127 assisted self-assembly optimized the FL quenching, thereby preserving long wavelength emission desirable for low scattering deep tissue visualization. The in vitro NIR-II FL imaging has further verified the high radiative efficiency of CDMBBTO compared to PBS control. The MR imaging capability has next been assessed by quantifying the T2 relaxivity of Mn integrated CDMBBTO nanoplatforms. A linear increase in the 1/T2 relaxation rate as a function of Mn2+ concentration (r2 = 11.36 s− 1 mM− 1) has validated its strong T2 contrast performance (Fig. 4D) [89]. Fig. 4In vitro and in vivo dual NIR-II FL and MR imaging of subcutaneous glioma bearing mice models. (A) UV-vis spectra of DCNPs and DMBBTO demonstrate their respective absorption peaks. (B) FL spectra of DCNPs and DMBBTO upon 980 nm laser excitation demonstrate their respective emission peaks at ~1530 nm (inset: In vitro NIR-II FL images of PBS, DCNPs, and DMBBTO). (C) Excitation spectra at 1530 nm excitation demonstrate the presence of fluorescent centers in DCNPs. (D) T2 relaxation rate (r2) of DMBBTO at different concentrations (0 to 1 mM) of Mn (inset: In vitro T2 MR phantom images of DMBBTO at different concentrations (0 to 1 mM) of Mn). (E) In vivo T2 MR images of tumor bearing mice (white arrows) at different time intervals (up to 12 h) after administration of CMBBTO nanoplatform. (F) Demonstration of T2 contrast intensity at different time points (up to 12 h; n = 3). (G) In vivo NIR-II FL images of tumor bearing mice models at different time points (up to 12 h) after administration of CMBBTO nanoplatform (scale bar: 10 mm). (H) A cross sectional NIR-II FL intensity profile measured along the red dashed line, with its peaks fitted to a Gaussian function for the FL imaging of subcutaneous glioma. (I) Representation of tumor to background T2 contrast intensity at different time points (up to 12 h). (J) Representative tumor to background graph demonstrates the targeting ability of the CDMBBTO nanoplatform (S.D. ±1.0). (K) Ex vivo NIR-II FL imaging of major body organs and tumor isolated from mice. (L) Ex vivo NIR-II FL intensity in different major body organs and tumor isolated from mice (S.D. ±2.36%) In vitro and in vivo dual NIR-II FL and MR imaging of subcutaneous glioma bearing mice models. (A) UV-vis spectra of DCNPs and DMBBTO demonstrate their respective absorption peaks. (B) FL spectra of DCNPs and DMBBTO upon 980 nm laser excitation demonstrate their respective emission peaks at ~1530 nm (inset: In vitro NIR-II FL images of PBS, DCNPs, and DMBBTO). (C) Excitation spectra at 1530 nm excitation demonstrate the presence of fluorescent centers in DCNPs. (D) T2 relaxation rate (r2) of DMBBTO at different concentrations (0 to 1 mM) of Mn (inset: In vitro T2 MR phantom images of DMBBTO at different concentrations (0 to 1 mM) of Mn). (E) In vivo T2 MR images of tumor bearing mice (white arrows) at different time intervals (up to 12 h) after administration of CMBBTO nanoplatform. (F) Demonstration of T2 contrast intensity at different time points (up to 12 h; n = 3). (G) In vivo NIR-II FL images of tumor bearing mice models at different time points (up to 12 h) after administration of CMBBTO nanoplatform (scale bar: 10 mm). (H) A cross sectional NIR-II FL intensity profile measured along the red dashed line, with its peaks fitted to a Gaussian function for the FL imaging of subcutaneous glioma. (I) Representation of tumor to background T2 contrast intensity at different time points (up to 12 h). (J) Representative tumor to background graph demonstrates the targeting ability of the CDMBBTO nanoplatform (S.D. ±1.0). (K) Ex vivo NIR-II FL imaging of major body organs and tumor isolated from mice. (L) Ex vivo NIR-II FL intensity in different major body organs and tumor isolated from mice (S.D. ±2.36%) Encouraged by these results, in vivo T2 weighted MR imaging has been performed on subcutaneous U87 glioma-bearing mice models. As shown in Fig. 4E-F and Figure S17, a time dependent darkening at the tumor site intensified at 6 h and was monitored up to 12 h post-injection, confirming efficient enrichment and stable tumor retention, attributable to biomimetic U87 membrane mediated homologous targeting and prolonged systemic circulation of the CMBBTO nanoplatforms [90]. Simultaneously, NIR-II FL imaging revealed a rapid and precise tumor localizing pattern of CDMBBTO nanoplatforms (Fig. 4G), with detectable accumulation as early as 5 min and peak signal intensity at 6 h post-injection, sustaining beyond 12 h post-injection. Quantitative tumor to background analysis (Fig. 4I-J) and FWHM (full width half maxima, Fig. 4H) corroborated the highly specific tumor targeting behavior of CDMBBTO. Ex vivo biodistribution (Fig. 4K-L) further showed pronounced FL retention in tumor tissue, with secondary accumulation in the liver and spleen [91]. Taken together, these results have demonstrated that CDMBBTO achieves high brightness NIR-II FL and clinically relevant T2 MR contrast with excellent tumor selectivity and temporal persistence, establishing itself as a powerful dual modal imaging probe for precision image guided glioma therapy. In vitro biocompatibility and therapeutic efficiency of CMBBTO nanoplatform To evaluate the biosafety of the CDMBBTO nanoplatform prior to therapeutic application, a systematic in vitro biocompatibility assessment has been conducted. As shown in Fig. 5A, CDMBBTO nanoplatforms exhibited negligible cytotoxicity toward U87 glioma cells across a wide concentration range (0–100 µg/mL), with cell viability consistently maintained above 90%, confirming their outstanding cytocompatibility even at therapeutically relevant doses. Moreover, the cell viability assay has also been conducted for F-127 polymer up to 100 µg/mL concentration against U87 glioma cells to demonstrate the biocompatibility of the polymer matrix employed for self-assembly (Figure S10A). A hemocompatibility study further verified the translational safety of the platform, where the hemolysis ratio remained below 2% even at 100 µg/mL, comparable to PBS and significantly lower than the positive control Triton X-100 (Fig. 5B). These results establish CDMBBTO as a blood safe and biologically inert nanoplatform under non-activated conditions. The therapeutic competency of CDMBBTO under US activation has then been quantitatively assessed. All treatment groups without US had negligible cytotoxicity, as seen in Fig. 5C, supporting the stimulus responsiveness of the nanoplatform. Because of its inherent piezodynamic impact, BBTO+US exposure only caused significant tumor cell killing. On the other hand, CDMBBTO+US considerably outperformed all control groups in terms of cell viability reduction (****p < 0.0001). The enhanced intracellular ROS stress resulting from the synergistic activation of the Mn-driven Fenton-like chemodynamic (CDT) and piezocatalytic (PZDT) pathways is the cause of this higher therapeutic potency. Having established biosafety and therapeutic efficacy, the cellular uptake kinetics of CDMBBTO and DMBBTO have been studied to investigate their tumor targeting behavior [39]. As visualized in Fig. 5D, minimal FL has been observed at 1 h, while a time dependent enhancement in intracellular FL has been clearly detected from 2 to 12 h at different concentrations of DMBBTO and CDMBBTO. This efficient cellular internalization is attributed to biomimetic homologous targeting through U87 glioma cell membrane coating, which enables receptor ligand mediated recognition and accelerated endocytosis compared to non-camouflaged NPs. Such tumor specific uptake is essential for maximizing localized ROS generation while minimizing off target cytotoxicity. Fig. 5In vitro biocompatibility studies of the prepared nanoplatform. (A) Cell viability of U87 glioma cells treated with different concentrations (0 to 100 µg/mL) of CDMBBTO (S.D. ±2.17%). (B) Hemolysis assay for mouse blood at different concentrations (0 to 100 µg/mL) of CDMBBTO demonstrates its biosafety. (C) In vitro PZDT/CDT against the U87 glioma cell line treated with different sample groups (PBS-US, BBTO-US, CDMBBTO-US, PBS+US, BBTO+US, CDMBBTO+US) demonstrates a significant reduction of glioma cells upon US stimulation (****p < 0.0001 is calculated by ordinary one-way ANOVA). (D) Cellular uptake assay at different time intervals (0 to 12 h) demonstrates the cellular internalization of CDMBBTO and DMBBTO at different concentrations (0 to 100 µg/mL; scale bar: 25 μm). The p-values are calculated using ordinary one-way ANOVA and Student t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference In vitro biocompatibility studies of the prepared nanoplatform. (A) Cell viability of U87 glioma cells treated with different concentrations (0 to 100 µg/mL) of CDMBBTO (S.D. ±2.17%). (B) Hemolysis assay for mouse blood at different concentrations (0 to 100 µg/mL) of CDMBBTO demonstrates its biosafety. (C) In vitro PZDT/CDT against the U87 glioma cell line treated with different sample groups (PBS-US, BBTO-US, CDMBBTO-US, PBS+US, BBTO+US, CDMBBTO+US) demonstrates a significant reduction of glioma cells upon US stimulation (****p < 0.0001 is calculated by ordinary one-way ANOVA). (D) Cellular uptake assay at different time intervals (0 to 12 h) demonstrates the cellular internalization of CDMBBTO and DMBBTO at different concentrations (0 to 100 µg/mL; scale bar: 25 μm). The p-values are calculated using ordinary one-way ANOVA and Student t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference The DCFH-DA probe has been used to monitor intracellular ROS formation in order to further assess whether the piezo/chemodynamic cascade might successfully induce oxidative stress inside glioma cells [92]. U87 glioma cells treated with PBS, F-127, DMBBTO, and CDMBBTO without US stimulation showed very little green FL, as seen in Fig. 6A, indicating limited baseline ROS leakage. However, in the DMBBTO and CDMBBTO groups, the administration of US significantly increased intracellular ROS signals, confirming that mechanoresponsive piezoelectric polarization causes a burst of cytotoxic radicals within tumor cells [93]. Nevertheless, after US stimulation, the F-127-treated group did not exhibit significant ROS generation, ruling out its involvement in the therapeutic outcome. In order to compare the piezodynamic and chemodynamic activity of the CDMBBTO in the presence and absence of H2O2 and US stimulation, intracellular ROS generation was then carried out (Fig. 6B). The equal involvement of piezodynamic and Fenton-like processes in tumor inhibition has been confirmed by the considerable ROS generation upon US stimulation and the H2O2 environment. Calcein AM/PI live/dead labeling has been used to further validate the therapeutic effects of this intracellular ROS storm (Fig. 6C and Figure S10B). The untreated and control groups (PBS and F-127) showed primarily green FL, as predicted, suggesting intact cellular membranes and good vitality [84, 94]. However, significant red FL has been seen in cells treated with US-triggered nanoplatforms, indicating permanent cell death and impaired membrane integrity. The superiority of strategy over traditional monotherapies has been demonstrated by the abrupt change from green to red signals under synergistic activation. These results clearly demonstrated that the piezo redox machinery has been successfully activated intracellularly, resulting in strong glioma cell ablation via ROS-mediated necrotic and apoptotic pathways. To establish a clinically relevant therapeutic comparison, the anti-glioma efficacy of Temozolomide (TMZ) has been evaluated against U87 cells using CCK-8 and Calcein AM/PI assays (Figure S15). TMZ treatment produced concentration-dependent cytotoxicity from 20 to 100 µg/mL, with nearly complete inhibition of cell viability observed at 100 µg/mL. Compared with TMZ chemotherapy, the CDMBBTO+US system demonstrated strong anti-tumor activity through a non-invasive piezo-chemodynamic mechanism involving US-triggered ROS generation and enhanced Mn-mediated Fenton-like activity. These findings suggest that CDMBBTO may serve as a promising alternative with dual NIR-II FL and MRI guidance, making it a better theranostic nanoplatform than the conventional TMZ-based chemotherapy for glioblastoma treatment (Table S3). Fig. 6In vitro piezo/chemodynamic ROS production efficiency of CDMBBTO. (A) In vitro ROS production in U87 glioma cells after treatment with PBS, F-127, BBTO, and CDMBBTO with and without US stimulation. The ROS in U87 glioma cells are stained with DCFH-DA and are highlighted by green fluorescence (scale bars: 100 μm). (B) Evaluation of in vitro chemodynamic ROS production in U87 glioma cells in the presence and absence of H2O2, and piezodynamic ROS production in U87 glioma cells with and without US stimulation (scale bar: 100 μm). (C) Live/dead staining assay for U87 glioma cells after treatment with PBS, MBBTO, and CDMBBTO with and without US stimulation. The green fluorescence indicates Calcein stained live tumor cells, while red fluorescence indicates PI-stained dead tumor cells (scale bar: 100 μm). (D) Flow cytometry analysis demonstrates the cellular apoptosis in U87 glioma cells incubated with and without macrophages (MPs) and then stained with Annexin V-FITC and PI upon treatment with different groups (PBS, DMBBTO, and CDMBBTO) with and without US stimulation In vitro piezo/chemodynamic ROS production efficiency of CDMBBTO. (A) In vitro ROS production in U87 glioma cells after treatment with PBS, F-127, BBTO, and CDMBBTO with and without US stimulation. The ROS in U87 glioma cells are stained with DCFH-DA and are highlighted by green fluorescence (scale bars: 100 μm). (B) Evaluation of in vitro chemodynamic ROS production in U87 glioma cells in the presence and absence of H2O2, and piezodynamic ROS production in U87 glioma cells with and without US stimulation (scale bar: 100 μm). (C) Live/dead staining assay for U87 glioma cells after treatment with PBS, MBBTO, and CDMBBTO with and without US stimulation. The green fluorescence indicates Calcein stained live tumor cells, while red fluorescence indicates PI-stained dead tumor cells (scale bar: 100 μm). (D) Flow cytometry analysis demonstrates the cellular apoptosis in U87 glioma cells incubated with and without macrophages (MPs) and then stained with Annexin V-FITC and PI upon treatment with different groups (PBS, DMBBTO, and CDMBBTO) with and without US stimulation The flow cytometry analysis has been conducted by using Annexin V-FITC/PI staining to demonstrate the apoptosis in U87 glioma cells upon treatment with CDMBBTO co-incubated with the macrophages (MPs) and without MPs (Fig. 6D) [88]. The flow cytometry data indicated that the proportion of viable U87 glioma cells is 94.9%, 85%, 78.7% and 87.8% in the PBS+MP, DMBBTO+MP, CDMBBTO+MP, and CDMBBTO-MP without US, respectively. Importantly, the viability of U87 cells is further decreased to 91.1%, 46.9%, 36.2%, and 66.3% upon treatment with PBS+MP, DMBBTO+MP, CDMBBTO+MP, and CDMBBTO-MP under US stimulation [95]. These results further confirmed that the application of US with CDMBBTO in MPs cells has superior polarization efficacy, compared to the DMBBTO and CDMBBTO without US. Moreover, Figure S14 demonstrated that defect-engineered piezoelectric stimulation distinctly enhanced immunogenic cell death (ICD) in U87 cells, as confirmed by flow cytometric analysis of surface calreticulin (CRT) exposure, a key DAMP marker. PBS-US and PBS+US groups retained predominantly viable CRT-positive cells (Q4: 98.1% and 97.7%, respectively). In contrast, DMBBTO+US and CDMBBTO+US treatments significantly shifted CRT-positive cells toward the dead-cell compartment (Q2: 37.7% and 56.7%, respectively), compared to their corresponding non-US groups [96]. Notably, CDMBBTO+US exhibited the strongest ICD induction, supporting that defect-enhanced piezoelectric stimulation promotes anti-tumor immune activation, potentially through dendritic cell maturation and cGAS-STING pathway activation. These results suggest that the piezo-mediated repolarized MPs (M1), together with PZDT/CDT, enhance tumor cell apoptosis, thereby contributing to effective tumor inhibition. In vivo antitumor efficacy, immune modulation, and biosafety of CDMBBTO nanoplatform The in vivo PZDT/CDT efficacy of the CDMBBTO nanoplatform under US activation was assessed by using subcutaneous U87 glioma tumor-bearing mouse models (Fig. 7A). All groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US) maintained consistent body weight over the course of the 14-day treatment period (Fig. 7B), indicating that neither the nanoplatform nor US caused systemic toxicity, metabolic burden, or appetite suppression. On the other hand, tumor progression dynamics showed a clear treatment difference. While CDMBBTO without US stimulation generated very little tumor reduction, the control groups (PBS-US and PBS+US) displayed aggressive and continuous tumor growth, suggesting that passive NPs delivery alone was insufficient for therapeutic response. Notably, the CDMBBTO+US group achieved a significant ablation and showed progressive tumor shrinkage by Day 12 (Fig. 7D). This has confirmed that US-induced ROS production, which results from synergistic piezoelectric polarization-induced redox activation (PZDT) and Mn-mediated chemodynamic Fenton-like ROS formation (CDT), is the only source of therapeutic potential for the CDMBBTO nanoplatform. As a result, all control animals (PBS-US, PBS+US, and CDMBBTO-US) died before Day 30, while the CDMBBTO+US group showed significantly longer life, with a definite survival advantage beyond 30 days (Fig. 7C). Figure 7E provides a visual representation of the spatiotemporally regulated antitumor effectiveness of the CDMBBTO nanoplatform under US activation. From Day 0 to Day 12, the subcutaneous glioma tumors in the PBS-US and PBS+US control groups grew quickly and continuously, demonstrating the normal course of untreated glioma. Only slight tumor suppression was observed in the CDMBBTO-US group, indicating that passive nanoparticle accumulation without US activation is not enough to produce a significant therapeutic advantage. In contrast, the CDMBBTO+US group showed a significant tumor ablation by day 12 and a quick and time-dependent decrease in tumor volume, with noticeable shrinking starting as early as day 2. The progressive tumor regression for the CDMBBTO+US group is clearly visible in the areas indicated by red dashed circles, highlighting the fact that the therapeutic impact is solely dependent on the synergistic activation of the nanoplatform by US stimulation. These findings provide compelling evidence for the precise tumor-selective therapeutic capacity and good in vivo PZDT/CDT performance of the CDMBBTO nanoplatform. Fig. 7In vivo PZDT/CDT of subcutaneous glioma-bearing mice models via CDMBBTO nanoplatform. (A) Schematic illustration of the timeline for PZDT/CDT of subcutaneous glioma bearing mice models. (B) Time-dependent body weight (S.D. ±0.91 g), (C) mice survival curves, and (D) relative tumor volumes for U87 glioma-bearing mice models after treatment with different groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US). (E) Representative images of subcutaneous U87 glioma bearing mice models up to day 12 demonstrate PZDT/CDT upon treatment with different groups (PBS-US, PBS + US, CDMBBTO-US, and CDMBBTO+US; US stimulation at 1 MHz of frequency with 1.2 W/cm2 power and 50% duty cycle for 10 min; scale bar: 10 mm). (F, G) Quantification of proinflammatory markers (TNF-α and IL-6) demonstrates the antitumor activity of different treatment groups. (H) M1/M2 ratio of tumor tissues treated with different groups validates the macrophage repolarization. The p-values are calculated using ordinary one-way ANOVA and Student t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference In vivo PZDT/CDT of subcutaneous glioma-bearing mice models via CDMBBTO nanoplatform. (A) Schematic illustration of the timeline for PZDT/CDT of subcutaneous glioma bearing mice models. (B) Time-dependent body weight (S.D. ±0.91 g), (C) mice survival curves, and (D) relative tumor volumes for U87 glioma-bearing mice models after treatment with different groups (PBS-US, PBS+US, CDMBBTO-US, and CDMBBTO+US). (E) Representative images of subcutaneous U87 glioma bearing mice models up to day 12 demonstrate PZDT/CDT upon treatment with different groups (PBS-US, PBS + US, CDMBBTO-US, and CDMBBTO+US; US stimulation at 1 MHz of frequency with 1.2 W/cm2 power and 50% duty cycle for 10 min; scale bar: 10 mm). (F, G) Quantification of proinflammatory markers (TNF-α and IL-6) demonstrates the antitumor activity of different treatment groups. (H) M1/M2 ratio of tumor tissues treated with different groups validates the macrophage repolarization. The p-values are calculated using ordinary one-way ANOVA and Student t-test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, and n.s. indicating no significant difference The in vivo therapeutic mechanism of CDMBBTO nanoplatform under US activation has been validated by histological and immunological evidence. Only in the CDMBBTO+US group does TUNEL labeling show significantly increased DNA fragmentation, indicating widespread ROS-induced apoptotic tumor cell death (Fig. 8A) [97]. While the PBS-US, PBS+US, and CDMBBTO-US groups maintain intact tumor shape with densely packed viable tumor cells, H&E staining consistently reveals significant cellular damage and necrosis in this group, confirming minimal therapeutic impact without US stimulation (Fig. 8C). The good histocompatibility of the CDMBBTO nanoplatform was confirmed by H&E staining of important organs, such as the heart, liver, lungs, spleen, and kidneys, which revealed no discernible pathological changes. Additionally, a significant change in the tumor immunological microenvironment has been revealed by immunofluorescence labeling. The CDMBBTO + US group demonstrated high expression of M1-like proinflammatory MPs and a notable drop in M2 immunosuppressive MPs by using CD206 and iNOS biomarkers, respectively, suggesting successful macrophage repolarization (M2 → M1) (Figs. 7H and 8B). The development of an anticancer inflammatory response has been further enhanced by this immunological modulation, which has improved therapeutic outcomes in a synergistic way. The histological findings confirm that CDMBBTO-mediated PZDT/CDT modifies the tumor immune milieu toward a proinflammatory, tumor-suppressive phenotype (M1) in addition to directly inducing ROS-driven tumor apoptosis. Pro-inflammatory cytokines TNF-α and IL-6 were significantly upregulated along with macrophage repolarization (Fig. 7F-G), indicating the transformation of the tumor from an immune-silent to an immune-responsive state [98]. Such immunological normalization is crucial for clinical relevance because it establishes CDMBBTO as both an immune-restorative nanotherapeutic platform and a tumor-destructive agent. Fig. 8(A) Representative images of TUNEL staining demonstrate ROS-induced apoptosis in tumor tissue sections for different treatment groups (scale bar: 100 μm). (B) Immunofluorescence staining illustrates the macrophage repolarization (M2 → M1) in tumor tissue sections for different treatment groups (scale bar: 50 μm). (C) H&E staining shows the presence of dead tumor cells and biosafety of the CDMBBTO for the major body organs (heart, liver, lung, spleen, and kidney) from the tumor bearing mice models excised at day 14 (scale bar: 100 μm) (A) Representative images of TUNEL staining demonstrate ROS-induced apoptosis in tumor tissue sections for different treatment groups (scale bar: 100 μm). (B) Immunofluorescence staining illustrates the macrophage repolarization (M2 → M1) in tumor tissue sections for different treatment groups (scale bar: 50 μm). (C) H&E staining shows the presence of dead tumor cells and biosafety of the CDMBBTO for the major body organs (heart, liver, lung, spleen, and kidney) from the tumor bearing mice models excised at day 14 (scale bar: 100 μm) Flow cytometry analysis of the tumor tissues has also revealed a pronounced immune activation in response to CDMBBTO treatment under US stimulation (Fig. 9). The CDMBBTO+US noticeably increases the population of CD3 and CD8+ T cells up to 20.3% as compared to PBS-US (5.38%), PBS+US (6.67%), and CDMBBTO-US (10.3%), indicating enhanced activation of tumor-infiltrating cytotoxic T lymphocytes (CTLs). In parallel, macrophage phenotype analysis shows a clear shift from an immunosuppressive M2 phenotype (CD206) toward a pro-inflammatory M1 phenotype (CD86) in the CDMBBTO+US group [85]. The percentage of M2 MPs (CD206+) in the TME is significantly decreased to approximately 51.5% and 25.1% in the PBS and CDMBBTO without US, respectively, and significantly elevated CD86 expression is observed in approximately 11.9% and 30% in the PBS and CDMBBTO with US [99]. After the US application, the CDMBBTO group (0.31%) showed the most significant decrease in M2 MPs compared to other groups, and significantly enhanced CD86 expression (52.6%) of CDMBBTO with the US. These results indicate that US-activated CDMBBTO not only directly induces tumor cell inhibition but also remodels the tumor immune microenvironment by simultaneously promoting CTLs infiltration and reprogramming MPs toward an M1 phenotype [100, 101]. This dual immune modulation provides mechanistic evidence for the observed enhancement of antitumor efficacy and supports the role of CDMBBTO as an effective immunoregulating PZDT/CDT platform. Fig. 9Flow cytometry analysis of CD3 and CD8 + markers represents the activation of T cells and tumor-infiltrating cytotoxic lymphocytes (CTLs), while analysis of CD206 and CD86 represents the M2 → M1 macrophage repolarization in the tumor tissue after treatment with the different groups (PBS and CDMBBTO) with and without US stimulation Flow cytometry analysis of CD3 and CD8 + markers represents the activation of T cells and tumor-infiltrating cytotoxic lymphocytes (CTLs), while analysis of CD206 and CD86 represents the M2 → M1 macrophage repolarization in the tumor tissue after treatment with the different groups (PBS and CDMBBTO) with and without US stimulation Finally, the blood biochemistry analysis confirmed that liver (ALT, AST, and ALP) and renal (creatinine, uric acid) markers remained within the normal physiological range across all groups, confirming excellent systemic biosafety, absence of off-target organ damage, and high translational suitability (Figure S11A-E) [102]. Moreover, elemental analysis has revealed low levels of Ba, Bi, and Mn in major organs at day 14, with preferential accumulation in the liver and spleen consistent with reticuloendothelial clearance and negligible deposition in the heart and kidneys, indicating efficient clearance and favorable in vivo biosafety of the CDMBBTO nanoplatform (Figure S11F). Therefore, upon US activation, the CDMBBTO nanoplatform achieves potent and selective tumor eradication via synergistic PZDT/CDT while reprogramming the immunosuppressive TME, all with negligible systemic toxicity, highlighting its strong translational potential for precision cancer therapy. Conclusion In summary, defect-engineered, biomimetic CDMBBTO nanoplatform has been rationally developed as a dual NIR-II and MR imaging guided PZDT/CDT platform for efficient glioma suppression. This work presents one of the first demonstrations of integrating NIR-II emissive DCNPs with optically compatible, defect-engineered piezoelectric MBBTO NPs in a self-assembled architecture, thereby overcoming the long-standing challenge of optical darkness in piezoelectric systems. The Mn/Bi dual defect engineering strategy successfully introduced interstitial defects and oxygen vacancies, widening the bandgap and significantly enhancing the piezoelectric coefficient (d33 = 106.91 pmV− 1), which in turn enabled robust US triggered piezocatalytic ROS generation. Simultaneously, Mn2+ imparted intrinsic T2 weighted MR contrast and Fenton like chemodynamic activity, achieving continuous oxidative stress even beyond US exposure. Additionally, the self-assembled Pluronic F-127 matrix preserved the ultrahigh d33 coefficient of the DMBBTO NPs and NIR-II FL integrity of the DCNPs by preventing the direct interaction between both the components of the self-assembly, thereby maintaining the individuality of the self-assembled components. The U87 glioma cell membrane camouflage endowed the nanoplatform with precise homologous tumor targeting, rapid accumulation (peak at 6 h post-injection), and immune evasion. The CDMBBTO nanoplatform achieved potent synergistic PZDT/CDT in vitro and in vivo, inducing apoptosis and necrosis via ROS overload, while also triggering immunomodulation through macrophage repolarization (M2 → M1) and strong pro-inflammatory cytokine activation (TNF-α, IL-6) in subcutaneous glioma-bearing mice model. Importantly, comprehensive histological and blood biochemistry analyses confirmed excellent biocompatibility and absence of systemic toxicity. From a translational perspective, the CDMBBTO nanoplatform combines noninvasive US activation with clinically relevant dual NIR-II FL and MR imaging. It offers precise, controllable, and safe cancer treatment. Its biomimetic targeting, immune modulation, and strong biosafety profile support potential clinical applicability beyond glioma, including other therapy-resistant solid tumors. This strategy could advance imaging-guided PZDT/CDT toward real-world oncology applications by further validation in the orthotopic model and optimization under clinically relevant parameters. Supplementary Information Supplementary Material 1 Supplementary Material 1