Abstract Bone regeneration in diabetic patients poses a significant clinical challenge due to persistent hyperglycemia and chronic inflammation, which disrupt the crucial interaction between the immune microenvironment and bone marrow-derived mesenchymal stem cells (BMSCs), thereby impairing osteogenesis. To address this limitation, ultrasonic-responsive barium titanate (BTO) nanoparticles were coated with BMSC membranes (B-TNs) and subsequently integrated into a carboxylated modified silk fibroin (CMS) hydrogel. This resulted in a dual-functional B-TNs@CMS composite hydrogel designed to combine targeted BMSC stimulation with immunomodulatory properties. Under diabetic conditions, the composite hydrogel facilitated macrophage polarization from the pro-inflammatory M1 phenotype to the anti-inflammatory M2 phenotype. Proteomic analysis and validation assays demonstrated that CMS reprograms macrophages through the PI3K-Akt-mTORC1 signaling axis, thereby restoring an osteogenesis-supportive immune microenvironment conducive to osteogenesis. Simultaneously, the degradation of CMS released B-TNs, which produced a moderate level of reactive oxygen species (ROS) under optimized extracorporeal ultrasound (US) stimulation. This process activated Wnt/β-catenin signaling, enhancing BMSC proliferation and osteogenic differentiation. This study highlights the significant potential of a dual-modular strategy integrating immune modulation with spatially targeted osteogenic stimulation, offering a promising therapeutic approach for diabetic bone regeneration. Keywords: Diabetes mellitus, Bone healing, Carboxylated modified silk fibroin, Targeted BMSC stimulation, Immune-osteogenic homeostasis Graphical abstract Image 1 [45]Open in a new tab Highlights * • We creatively developed BMSCs-targeting piezoelectric transducer (B-TNs) to stimulate BMSCs precisely. * • By utilizing carboxylated modified silk fibroin hydrogels, our approach overcomes the obstacle to modulate immune microenvironment under diabetic condition. * • Our study offers integrated immunoregulatory and osteogenic strategy to promote bone regeneration, which restore the immune-osteogenic homeostasis under DM condition. 1. Introduction A steadily rising population has been involved in diabetes mellitus (DM) worldwide [[46]1], and this metabolic disorder presents a major challenge for bone healing. DM increases fracture risk by 20–300 % [[47]2], weakens bone strength [[48]3], and extends healing time by 87 % [[49]4], contributing to $966 billion in global healthcare costs [[50]5]. Bone regeneration follows a highly characteristic process driven by the ordered dynamic transitions [[51]6], which involves the overlapping stages of inflammation, repair, and remodeling [[52]7]. BMSCs play a key role in the process [[53]8], exhibiting multidirectional differentiation potential for bone healing [[54]9], which is tightly regulated by the osteoimmune microenvironment [[55]10]. Nevertheless, a significant cause of compromised bone healing is the disruption of osteo-immune homeostasis [[56]11,[57]12]. However, chronic hyperglycemia and persistent inflammation in DM break this balance [[58]13], which impedes bone repair and remodeling [[59]14]. Current therapies targeting either promoting BMSC function or focusing on immune modulation alone have proven ineffective[[60][15], [61][16], [62][17], [63][18]],these methods ignore the coordinated crosstalk between BMSCs and immune microenvironment in bone regeneration [[64][19], [65][20], [66][21]]. A more promising approach involves dual-regulation strategies to achieve greater flexibility in the modulation of both osteo-immune microenvironment and BMSC function simultaneously, offering a potential breakthrough in diabetic bone regeneration (see [67]Scheme 1). Scheme. [68]Scheme [69]Open in a new tab Schematic diagram of composite piezoelectric hydrogel promoting bone defect repair and regeneration in DM environment. The B-TNs@CMS improves the osteogenic immune microenvironment by shifting macrophage phenotypes under DM conditions and precisely stimulates BMSCs through ROS, thereby promoting bone regeneration in DM rats. Due to the inherent electrophysiological properties of nature bone tissue [[70]21], the electrical stimulation are a pivotal factor in regulating cellular behavior and accelerating bone healing [[71][22], [72][23], [73][24], [74][25], [75][26], [76][27]]. As a perovskite-type material, barium titanate (BTO) exhibits exceptional piezoelectricity [[77]28,[78]29], which is highly biocompatible with the piezoelectric properties of bone tissue, making it a promising biomaterial to be applied in bone regeneration [[79]30,[80]31]. However, previous studies indicate non-selective electrical stimulation can activate inflammatory cells, which can exacerbate inflammation and trigger pathological changes in diabetic environments [[81]32]. To prevent this potential side effect, a targeted piezoelectric transducer capable of delivering precise electrical stimulation to BMSCs is essential. Meanwhile recent studies show that piezocatalytic materials can generate reactive oxygen species (ROS) through redox reactions with water by ultrasound irradiation [[82]33,[83]34]. At moderate levels, ROS act as signaling molecules, activating the Akt/Erk pathway to promote cell proliferation[[84][35], [85][36], [86][37]]. Thus, optimizing ultrasonic parameters and modifying BTO biologically can fine-tune piezoelectric stimulation, ensuring controlled ROS production to enhance BMSC osteogenic differentiation effectively. Additionally, BMSC function is also influenced by the immune microenvironment, primarily regulated by macrophages. In diabetes, the disruption of immune-osteogenic homeostasis is mainly caused by imbalance M1/M2 activation due to the delayed phenotypic polarization of macrophages from M1 to M2 [[87][38], [88][39], [89][40]]. Therefore, it could be considered that restoring M2 predominance is critical for reestablishing a regenerative osteoimmune environment in DM. Recent research suggests that chemically modified silk fibroin can modulate immune responses while maintaining biocompatibility of silk fibroin [[90]41], Therefore, it is a promising vehicle for targeted piezoelectric transducer delivery making it a promising vehicle for targeted piezoelectric transducer delivery. In this study, a composite hydrogel with dual functions of BMSC-targeting stimulation and immunomodulation was developed for diabetic bone regeneration. The hydrogel comprises carboxylated modified silk fibroin (CMS) and BTO nanoparticles encapsulated in BMSC membranes (B-TNs). The immunomodulatory capacity of CMS promotes M1-to-M2 macrophage transition, restoring a favorable osteo-immune microenvironment balance in diabetic conditions. Concurrently, its degradation releases B-TNs, facilitating their uptake by BMSCs. Upon ultrasound (US) exposure, B-TNs can generate precise and controlled stimulation by adjusting US parameters, enhancing BMSC proliferation and differentiation through optimized ROS signaling. This integrated immunoregulatory and osteogenic strategy offers a promising solution for diabetic bone healing, advancing regenerative therapy. 2. Materials and methods 2.1. Preparation of the CMS 20 g silkworm cocoons from B. mori were degummed by boiling in 0.02mNa2CO3 (Sigma-Aldrich, USA) solution (4 L) for 30 min and dried overnight, added the obtained silk fibroin to 125g of 1-butyl-3-methylimidazolium chloride (TCI USA) and reacted at 100 °C for 8 h. Then Added 75 ml of dimethylformamide and react at 100 °C for 1 h, 40 g of succinic anhydride was added to the SF mixed solution and reacted for 2h at 100 °C, dialyzing the solution in dialysis bag (MWCO 3.5 kDa, Pierce, USA) against urea for 24h, and then dialyzing in deionized water for 72h. The precipitate obtained was lyophilized,the dried precipitate was dissolved in 9.3 m LiBr solution (Sigma-Aldrich, USA) at 60 °C for 4h,dialyzing the blended solution in deionized water for 72h again. After centrifuging for 2 × 20 min at 12000 r.p.m, the purified CMS solution (concentration ≈ 6 ‰) was injected into a dialysis bag (MWCO 3.5 kDa, Pierce, USA) and concentrated to 30 %. 2.2. Preparation of the B-TNs Synthesized BTO using hydrothermal method, using the MinuteTM Plasma Membrane Isolation Kit (Invent Biotechnologies, USA) to isolate the plasma membrane from BMSCs (BM), P2 BMSCs (50 × 106) were washed with pre-cooled PBS and collected. Centrifugation at 600×g for 5 min at room temperature. The pellet was suspended of ice--cold isolation 200ul buffer A and incubated on ice for 10 min, vortexed for 30 s. Centrifugation in the tube at 16,000×g for 30 s, vortexed for 10 s. Centrifugation at 700×g for 1 min. Transferred the supernatant to a new 1.5 ml centrifuge tube, and centrifuged at 16,000×g for 30 min at 4 °C. Discarded the supernatant and kept the pellet. Added 200ul of Buffer B and resuspended the total membrane fraction by vortexing. centrifuged at 7,800×g for 5 min at 4 °C, and transferred the supernatant to a new 2.0 ml centrifuge tube. Added 1.6 ml of pre-cooled PBS and mixed. Centrifuged at 16,000×g for 30 min, discarded the supernatant and kept the pellet. The plasma membranes of primary rat osteoblasts (PO) and HEK293T cells were also prepared using the aforementioned method. To coat BTO by Plasma Membrane of BMSCs, 100 μl aliquots of the membrane solution, containing 300 μg of BMSC membrane, were prepared and mixed with 100 μl of BTO (100 μg/ml), extrusion through a 400--nm cutoff extruder (Hamilton Company) for 20 cycles to obtain B-TNs. 2.3. Preparation of the B-TNs@CMS The B-TNs were combined with CMS using inorganic-organic hybridization technology to prepare B-TNs@CMS hydrogel. A glass rod was used to stir the concentrated CMS hydrogel in the syringe (20 mL type) for 5 min at a stirring speed of ≈120 r min−1. After sterilization through 60Co γ-irradiation at a dose of 25 kGy, the B-TNs@CMS hydrogel was used for in vitro experiments, and was directly injected into the rat femur for in vivo experiments. 2.4. Characterization of B-TNs@CMS Using transmission electron microscopy (TEM, Hitachi High-Technologies Corporation, Japan) to inspect the BM, B-TNs. Using Malvern Zetasizer (Malvern Instruments, UK) to measure the diameter and the stability of BM and B-TNs. The morphologies of the lyophilized B-TNs@CMS and cells were inspected by scanning electron microscopy (SEM, JSM-7600F, JEOL, Japan). The element distribution of B-TNs@CMS can be detected using scanning electron microscopy (SEM, ZEISS Sigma 300, Germany). Using AFM (Bruker Dimension Icon, Germany) to probe piezoelectricity and structure of the B-TNs@CMS. Using X-ray diffractometry (XRD, Rigaku SmartLab SE, Japan) to identify the phase compositions of B-TNs@CMS. Using Fourier Transform Infrared Spectroscopy, (FTIR, Thermo Fisher Scientific Nicolet iS20, USA) to detect the structures of SF, CMS and B-TNs@CMS. Nuclear magnetic resonance hydrogen spectrum (NMR, MestReNova, USA) were used to detect the carboxyl groups in CMS. X-ray photoelectron spectroscopy (XPS, Thermo Scientific K-Alpha, USA) were used to detect the valence states of elements in B-TNs@CMS. 2.5. Cell culture To obtain and culture of BMSCs, Macrophages, Primary rat osteoblasts (PO), HEK293T and Raw246.7. BMSCs, BMDM and PO were isolated from the femoral medullary canal and calvarias of neonatal Sprague-Dawley(SD) rats. HEK293T and Raw264.7 were obtained from obtain The laboratory of School of Basic Medical Sciences, Air Force Medical University. BMSCs and PO cultured in MEM Alpha medium supplemented with 10 % fetal bovine serum (FBS) and 1 % penicillin/streptomycin, and then incubated in atmosphere of 5 % CO[2] at 37 °C. BMDM were cultured in RPMI 1640 medium supplemented with 10 % FBS, 1 % penicillin/streptomycin, and 20 ng mL^−1 rat macrophage colony-stimulating factor (Pepro Tech, Inc). HEK 293T and Raw246.7 were seed in 10 cm cell culture dish with DMEM medium supplemented with 10 % fetal bovine serum (FBS) and 1 % penicillin/streptomycin. The type 2 diabetes was mimicked by containing 25 mmol L^−1 glucose (high glucose) and 500 μmol L^−1 BSA-conjugated palmitate (high fat). The Experiments were con-ducted according to an animal experimental protocol approved by the Air Force Medical University. 2.6. Transwells BMSCs were cocultured with macrophages using 24-Transwell plates with a 0.4 μm pore size (Corning Life Sciences). BMSCs were seeded on the bottom chamber of the transwell, Macrophages and were seeded on the top chamber, Simultaneously, high-fat and high-glucose medium and B-TNs@CMS were added, and ultrasonic intervention was applied at the bottom. 2.7. Cell viability assays BMSCs proliferation was measured using a Cell Counting Kit-8 (CCK-8) (Beyotime, China). BMSCs were seeded onto the lyophilized B-TNs@CMS located in 96-well plates at a density of 1 × 10^3 cells per well at 37 °C with 5 % CO[2]. After 1, 3, 5 days incubation, discarding original medium, and incubating the cells with medium supplemented with 10 μl of CCK-8 solution for 2h at 37 °C. Subsequently, the supernatant of each well was collected and detected at the 450 nm absorbance using a microplate reader (EnSpire, PerkinElmer). cell viability was evaluated by the Calcein-AM/PI Double Stain Kit (YEASEN, China).Seeding BMSCs and co-cultured on the surface of lyophilized B-TNs@CMS. After incubation for 1, 3, 5 days, discarding original medium, and the cells were incubated with 200uL of dye solution (1 × Assay Buffer containing calcein−AM(2 μmol/mL) and PI (1.5 μmol/mL)). After incubation at 25 °C for 30 min, the cells were captured with a laser scanning confocal microscope (Nikon A1-Si, Japan). 2.8. Targeting BMSCs experiment in vitro After co-culturing vesicles formed by BTO encapsulated in three different cell plasma membranes (BMSCs, HEK293T and PO) with BMSCs for 8h, the samples were fixed and the uptake efficiency of BTO was observed using TEM (Hitachi High-Technologies Corporation, Japan). Preparing three types of cell plasma membranes (BMSCs, HEK293T and PO) and stain them with 3 nM Dil (Thermo Fisher Scientific, USA) for 10 min. Stain the target cells, BMSCs, with 5 nM Dio (Thermo Fisher Scientific, USA) for 10 min, and stain the nuclei of BMSCs with 5 μg/mL Hoechst (MedChemExpress, USA) for 30 min. Co-culture the three types of plasma membranes with BMSCs overnight, and then capture images using laser confocal microscope (Olympus, Japan). Assess the uptake of plasma membranes by BMSCs based on the fluorescence intensity. 2.9. Immunofluorescence staining Using CD86 and CD206 to label M1 and M2 macrophages, respectively. After treating the Raw264.7 cell line, the cells were fixed with 4 % paraformaldehyde and permeabilized with 0.1 % Triton X-100, followed by blocking with BSA for 2 h. The cell samples were then incubated with mouse monoclonal primary antibody CD86 and mouse polyclonal primary antibody CD206, respectively. Subsequently, they were incubated with Alexa Fluor 647 (red) anti-mouse IgG (Abcam) and Alexa Fluor 488 (green) anti-rabbit IgG (Abcam) secondary antibodies, respectively. Finally, the samples were observed and imaged using laser confocal microscope (Olympus, Japan). 2.10. Quantitative real-time PCR To measure gene expression levels, RNA was extracted from intervened BMSCs and macrophages, and then converted into cDNA using the PrimeScript RT Kit (Takara). Quantitative PCR (qPCR) was performed using SYBR Premix Ex TaqII (TaKaRa) on the CFX96 PCR System (Bio-Rad) to determine the gene expression levels. 2.11. Western blot analysis For Western blot analysis, BMSCs and macrophages were washed three times with cold PBS and then lysed on ice for 10 min using RIPA buffer (Beyotime, Jiangsu, China) containing a protease inhibitor cocktail (Roche, Basel, Switzerland). The lysates were then centrifuged at 4 °C and 12,000 rpm for 15 min. The supernatants were collected and boiled in a metal bath. Then the proteins were subjected to SDS-PAGE electrophoresis, followed by transfer to polyvinylidene fluoride (PVDF) membranes (Merck Millipore, Billerica, MA, USA). The membranes were incubated with BSA at room temperature for 1 h and then incubated with the corresponding primary and secondary antibodies. The bands were detected using an ECL substrate kit (Mishushengwu, Xi'an, China). The antibodies used in this experiment included: Phospho-PI3K (Y1054 + Y1059) (1:1000, Abcam, ab5473), and Phospho-Akt (Ser473) (1:1000, Cell Signaling Technology, #4060). 2.12. Measurement of ROS and H[2]O[2] levels Using the DCFH-DA (MedChemExpress, USA) probe to detect intracellular ROS levels in BMSCs, the cells were co-cultured with B-TNs overnight to ensure sufficient uptake of BTO by BMSCs. Subsequently, a ROS inhibitor NCA was added to some samples, and others were exposed to US at power densities ranging from 0.1 to 0.5 w/cm^2. After these treatments, the culture medium was aspirated, and the cells were rinsed with PBS to remove excess B-TNs. DCFH-DA (20 μM) dissolved in serum-free DMEM was then added to the culture medium, and the cells were incubated for 30 min. Intracellular ROS levels were subsequently measured using a flow cytometer. For the detection of intracellular H[2]O[2] content, the Amplex™ Red (Thermo Fisher, USA) kit was used, with the same cell intervention methods as described above. Standards were prepared by serially diluting the standard solution provided in the kit to 20 μM, 10 μM, 5 μM, 2.5 μM and 1.25 μM using 1 × Reaction Buffer. Then, 0.5 μL of 10 mM Amplex Red reagent (from the H2O2 detection kit), 1 μL of 10 U/mL HRP, and 48.5 μL of 1 × Reaction Buffer were mixed thoroughly. Fifty microliters of each sample and standard were added to a 96-well plate, followed by the addition of 50 μL of the previously prepared reaction mixture. The plate was then incubated at room temperature for 30 min. The enzyme-linked immunosorbent assay (ELISA) plate reader was set with an excitation wavelength of 545 nm and an emission wavelength of 590 nm to measure the OD values, and a standard curve was plotted based on these values. 2.13. Cell proliferation experiment Using Edu (MedChemExpress, USA) and Hoechst MedChemExpress, USA) staining to assess cell proliferation. After subjecting BMSCs to different interventions, Edu (10 μM) was added and the cells were incubated for 2 h. Subsequently, the cells were rinsed with PBS, fixed with 4 % paraformaldehyde, and then permeabilized with Triton X-100. The reaction mixture was added, followed by nuclear staining with Hoechst. Finally, the cells were observed and imaged using a laser confocal microscope (Olympus, Japan). 2.14. Cell morphology staining Using vinculin (ABCAM, UK), F-actin (proteintech, USA), and DAPI (Abcam, UK) for cell morphology staining to observe cell shape. After completing the cell interventions, the cells were washed with PBS to remove the culture medium. The cells were then fixed with 4 % paraformaldehyde and washed again with PBS. Next, 0.1 % Triton X-100 was added for permeabilization, followed by another PBS wash to remove the permeabilization agent. The cells were then incubated with vinculin primary antibody (1:500) overnight, followed by PBS washing. Subsequently, the secondary antibody (Alexa Fluor 647, red, anti-mouse IgG, Abcam) was added, and the cells were washed again. After that, F-actin primary antibody was added for incubation. Following washing, the secondary antibody (Alexa Fluor 488, green, anti-rabbit IgG, Abcam) was added, and the cells were washed with PBS. DAPI was then added for nuclear staining. After removing DAPI, the cells were washed with PBS, and an appropriate amount of quencher was added. Finally, the results were observed using a laser confocal microscope (Olympus, Japan). 2.15. Cell differentiation and mineralization experiment To evaluate the ability of B-TNs@CMS to promote the differentiation and mineralization of BMSCs in vitro, ALP staining was performed using the BCIP/NBT Alkaline Phosphatase Color Development Kit (Beyotime Biotechnology Institute) and observed under an optical microscope. Alizarin red staining was utilized to assess calcium deposition. A solution of 10 % cetylpyridinium chloride in sodium phosphate (pH = 7.0) was added to each plate to dissolve the stain for 15 min, and the absorbance of the samples was measured at 450 nm for statistical analysis. 2.16. Animal experiment Construction of Diabetic Rats: Using high-fat, high-glucose diet and streptozotocin (STZ)-induced diabetic rats, male SD rats (200–250 g, 8 weeks old) were purchased from the Animal Experiment Center of Air Force Medical University (Xi'an, Shaanxi, China). The rats were housed in an air-conditioned room with a temperature of 22 ± 1 °C, humidity of 50 ± 10 %, and a constant light-dark cycle. All animal experiments were supervised and approved by the Institutional Animal Care and Use Committee. The rats were fed with a high-fat, high-glucose diet for 4 weeks, followed by intraperitoneal injections of STZ, administered once every other day for two weeks to ensure the survival rate of the rats. Subsequently, blood glucose levels were measured using a glucometer via the tail vein, and rats with non-fasting blood glucose levels >300 mg dl^−1 were considered diabetic. Six rats were selected as controls for both 4 and 8 weeks (n = 3), and the remaining 18 diabetic rats were randomly divided into three groups: the diabetes group (4 and 8 weeks, n = 3), the CMS group (4 and 8 weeks, n = 3), and the B-TNs@CMS group (4 and 8 weeks, n = 3). The rats were anesthetized using sevoflurane. Subsequently, the right hind limb was shaved and disinfected with ethanol, and a 10 mm incision was made on the medial side of the knee joint. The ligaments and muscles were separated, and a hole was drilled into the femoral condyle using a rotary drill. The CMS group received an implantation of CMS hydrogel, while the B-TNs@CMS group received an implantation of B-TNs@CMS hydrogel. Afterward, the tissue structure was restored, and the incision was closed with layered sutures. The rats were allowed to move freely postoperatively and were euthanized at 4 and 8 weeks, respectively, for specimen collection and subsequent testing and evaluation. The animal experiment was conducted in strict accordance with the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health and was approved by the Animal Care Committee of Air Force Military Medical University. 2.17. Micro-CT analysis After 4 and 8 weeks of material implantation, the rats were euthanized, and femoral specimens were collected and fixed in 10 % neutral buffered formalin for 7 days. Then, the specimens were scanned using micro CT system (AX-2000, Aoying, China) to assess bone regeneration. The bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), trabecular number (Tb.N) and average gray value (AGV) was calculated using VG Studio MAX 3.5. 2.18. Histological staining Rats were euthanized at 4 and 8 weeks post-implantation, and then harvested the specimens. The specimens are fixed in 4 % paraformaldehyde for 1 week, followed by decalcification using EDTA for three weeks, with regular changes of the solution. The specimens are then dehydrated using a gradient of ethanol, and subsequently cleared in xylene. Afterward, they are embedded in paraffin and sectioned. Finally, the sections undergo HE and Masson staining. Additionally, immunofluorescence staining is performed using CD86, CD206, and RUNX2 antibodies. 2.19. Statistical analysis All experimental data were obtained with at least three samples per group (n = 3), and the results are expressed as mean ± standard deviation. The samples were analyzed for differences using one-way ANOVA and Student's t-test. Statistical analyses were performed using SPSS Stats 20 software (SPSS, IBM, USA) and GraphPad Prism 9 (GraphPad Software, USA) 3. Results and discussion 3.1. Preparation and characterization of B-TNs@CMS An injectable piezoelectric hydrogel (B-TNs@CMS) was developed by synthesizing CMS from degummed silk fibroin and succinic anhydride. BMSC-derived membranes were extracted and used to encapsulate barium titanate nanoparticles (BTO-NPs), forming targeted piezoelectric nanoparticles (B-TNs). These were then integrated into CMS via inorganic-organic assembly technology to fabricate the composite hydrogel. Transmission electron microscopy (TEM) ([91]Fig. 1A and B) confirmed successful membrane extraction and nanoparticle encapsulation. Zeta potential and particle size analysis ([92]Fig. 1C and D) revealed that BMSC-derived vesicles measured approximately 130 nm with a Zeta potential of −35 mV, while BTO encapsulation increased the size to ∼150 nm and shifted the Zeta potential to −30 mV, indicating hydrogel stability. Macroscopic imaging ([93]Fig. 1E and F) demonstrated that B-TNs@CMS maintained excellent injectability and fluidity. Scanning electron microscopy (SEM) of freeze-dried B-TNs@CMS ([94]Fig. 1G and H) displayed a porous structure with an average pore size of ∼200 μm, suitable for BMSC adhesion and proliferation. Elemental mapping ([95]Fig. 1I) confirmed a uniform distribution of key elements (C, O, Ba, Ti, and N), with additional N mapping in [96]Fig. S1. Piezoresponse force microscopy (PFM) ([97]Fig. 1J and K) validated the hydrogel's piezoelectric properties, with amplitude butterfly and phase hysteresis loops further confirming its piezoelectric effect ([98]Fig. S2). X-ray diffraction (XRD) analysis ([99]Fig. 1L–S3) identified the tetragonal phase structure of BTO, indicating that CMS and BMSC vesicles did not alter its conformation. Fourier transform infrared spectroscopy (FTIR) ([100]Fig. 1M) detected carboxyl groups at 1750 cm^−1 and a characteristic BMSC membrane peak at 532 cm^−1. Nuclear magnetic resonance (NMR) hydrogen spectroscopy ([101]Fig. 1N–S4) confirmed the presence of methylene groups linked to carboxyl groups at a chemical shift of 2.5 ppm, verifying successful CMS conjugation with silk fibroin. X-ray photoelectron spectroscopy (XPS) ([102]Fig. S5) further confirmed the presence of Ti and C in B-TNs@CMS. Fig. 1. [103]Fig. 1 [104]Open in a new tab Characterization of the B-TNs@CMS. (A) TEM of BM (BMSCs membrane)preparation, scale bars:100 nm. (B) TEM image of B-TNs (BMSCs-targeted nanoparticles)preparation, scale bars:100 nm. (C, D) Particle size analysis and Zeta potential of BM and B-TNs, respectively. (E, F) The injectability of the B-TNs@CMS. (G) porous morphology of the lyophilized B-TNs@CMS. (H) SEM of co-culture of BMSCs with the B-TNs@CMS, scale bars: 100 μm. (I) Elemental analysis of the B-TNs@CMS, scale bars: 250 μm. (J, K) PFM test curves of the B-TNs@CMS. (L) XRD spectrum of the B-TNs@CMS. (M) FITR spectra of SF, CMS and B-TNs@CMS. (N) NMR spectrum of B-TNs@CMS. To evaluate the targeting efficiency of B-TNs in BMSCs, a series of validation experiments were performed. [105]Fig. 2A illustrates the role of B-TNs@CMS in promoting osteogenesis within a DM in vitro. In a transwell co-culture system, the upper layer contained macrophages exposed to CMS under DM conditions. Upon degradation of CMS, the released B-TNs precisely stimulate underlying BMSCs by US stimulation ([106]Fig. S6). TEM confirmed that B-TNs efficiently and selectively delivered BTO into BMSCs in vitro ([107]Fig. 2B). The biocompatibility of B-TNs@CMS was assessed using live/dead cell staining ([108]Fig. 2C–S7), CCK-8 assays ([109]Fig. S8) and Hemolysis assay ([110]Fig. S9). After 1, 3, and 5 days of co-culture, both CMS and B-TNs@CMS demonstrated superior biocompatibility compared to the control group. To further verify the targeting ability of B-TNs@CMS, TEM imaging ([111]Fig. 2D) analyzed BTO uptake in BMSCs following co-culture with BTO encapsulated by membranes from BMSCs, engineered 293T cells, and primary rat calvarial osteoblasts (PO). BMSC-derived membranes exhibited the highest uptake efficiency. To visualize targeting specificity, different cell membranes were labeled with Dil (red fluorescence), BMSCs with DiO (green fluorescence), and nuclei with Hoechst (blue fluorescence) ([112]Fig. 2F). Confocal laser scanning microscopy revealed the strongest red fluorescence signal in BMSCs, further confirming the precise targeting capability of B-TNs. Flow cytometric quantification reveals enhanced fluorescence intensity in BMSCs group indicating superior uptake efficiency ([113]Fig. S10). Fig. 2. [114]Fig. 2 [115]Open in a new tab Targeting specificity validation of B-TNs@CMS in vitro. (A) Schematic diagram of the in vitro co-culture model with hydrogel. (B) TEM of B-TNs targeting and entering BMSCs, scale bars: 400 nm (left), 100 nm (right). (C) Live/dead cell staining results of BMSCs treated with Control, CMS and B-TNs@CMS at 1, 3, and 5 days, respectively, scale bars: 200 μm (D) Comparison of TEM images showing the uptake efficiency of 293T-TNs, PO-TNs, and B-TNs by BMSCs, scale bars: 400 nm (up), 100 nm (down). (E) Comparison of fluorescence images showing the uptake efficiency of different group by BMSCs, scale bars: 20 μm. Compared to conventional piezoelectric stimulation, this BMSC-targeted strategy significantly reduces the required BTO dosage while enhances stimulation efficiency. Additionally, it minimizes off-target effects associated with non-selective piezoelectric stimulation. 3.2. Immunomodulatory effects of CMS in a DM environment via the PI3K/AKT/mTORC1 pathway To investigate the regulatory effects of B-TNs@CMS on the immune microenvironment, we treated macrophages with CMS and conducted proteomic analysis. Heatmap and volcano plot analyses revealed differential protein expression across interventions ([116]Figs. S11 and S12). GO analysis ([117]Fig. 3A and B) showed that, compared to the control group, the DM influenced macrophage inflammatory responses, cytokine secretion, and chemotaxis. CMS treatment under DM conditions significantly enriched pathways associated with macrophage chemotaxis and inflammatory regulation, suggesting potential effects on macrophage proliferation and polarization. To validate this, we performed immunofluorescence ([118]Fig. 3C) and qPCR analyses ([119]Fig. 3D). Under DM conditions, the M1 macrophage marker CD86 (red) and its related genes were upregulated, indicating enhanced M1 polarization and inflammation. CMS treatment significantly increased the expression of the M2 marker CD206 (green) and associated genes, demonstrating a shift from the pro-inflammatory M1 phenotype to the anti-inflammatory M2 phenotype. To explore the underlying mechanisms, KEGG analysis ([120]Fig. 3E and F) identified alterations in the PI3K-Akt pathway in both groups. GSEA further confirmed significant inhibition of the PI3K-Akt-mTORC pathway under DM conditions ([121]Fig. 3G–S13), while CMS treatment partially restored its activity ([122]Fig. 3H–S13). Western blot (WB) analysis ([123]Fig. 3I) corroborated these findings, showing reduced PI3K-Akt pathway protein expression under DM conditions, leading to suppressed cell proliferation and M1 polarization through mTORC1-LXR pathway inhibition. CMS treatment reversed these effects, restoring PI3K-Akt activity and upregulating mTORC1-LXR expression, promoting M2 polarization. These results suggest that CMS regulates macrophage proliferation and polarization via the PI3K-Akt-mTORC1-LXR pathway, thereby improving the osteogenic immune microenvironment under DM conditions. To assess whether immune microenvironment modulation alone enhances osteogenesis, we treated BMSCs under DM conditions with supernatant from CMS-treated macrophages and conducted CCK8 ([124]Fig. S14), qPCR ([125]Fig. S15), and WB analyses ([126]Fig. 3J). Compared to controls, DM conditions significantly impaired BMSC proliferation and osteogenic gene and protein expression. While CMS treatment did not fully restore osteogenic markers, it partially improved BMSC proliferation. A schematic diagram illustrating CMS-mediated macrophage regulation is shown in [127]Fig. 3K. Fig. 3. [128]Fig. 3 [129]Open in a new tab CMS alters macrophage phenotype and influences osteogenesis in DM environment. (A, B) GO functional enrichment analysis results. (C) fluorescence staining results of macrophage marker proteins CD86 and CD206, scale bars: 100 μm. (D) qPCR results of M1 and M2 macrophage polarization genes. (E, F) KEGG pathway enrichment analysis results. (G, H) GSEA analysis results of PI3K-Akt-mTORC pathway. (I, J) WB and quantification analysis of PI3K-Akt-mTORC pathway protein expression and osteogenic protein expression. (K) Schematic diagram illustrating the mechanism of CMS-induced macrophage polarization switching. Data presented as mean ± SD, n = 3 per group. ∗p < 0.05, ∗∗p < 0.01,by one-way ANOVA with Tukey's post-hoc test. Amino acids regulate macrophage polarization through the PI3K-Akt-mTORC pathway [[130]42,[131]43]. Since silk fibroin degrades into natural amino acids, CMS may influence macrophage polarization via this mechanism. While CMS preserves the biocompatibility of native silk fibroin, it also acquires immunomodulatory properties. However, our findings suggest that immune microenvironment modulation alone is insufficient to restore BMSC osteogenic differentiation under diabetic conditions. This limitation likely stems from irreversible DM-induced dysregulation of key signaling pathways controlling BMSC proliferation and differentiation—pathways CMS therapy cannot fully reactivate. Although CMS shifts macrophage polarization from the pro-inflammatory M1 to the anti-inflammatory M2 phenotype, reducing inflammation ([132]Fig. S16), this alone does not support bone regeneration. qPCR and WB analyses reveal persistent deficiencies in osteogenic gene expression and protein synthesis, emphasizing the need for additional osteogenic stimulation. 3.3. Targeted piezoelectric stimulation enhances ROS-mediated osteogenesis BMSC proliferation and differentiation are essential for bone regeneration. While piezoelectric materials have been used to modulate macrophage phenotypes in the DM to enhance BMSC activity, our findings suggest that macrophage transformation alone, though beneficial for proliferation, is insufficient to restore osteogenic differentiation. Thus, we shifted our focus to direct BMSC regulation. ROS at appropriate levels are critical for initiating cell proliferation and differentiation. As a biocompatible piezoelectric material, BTO generates intracellular ROS upon ultrasonic stimulation in vitro. Although widely used to induce ROS-mediated tumor cell destruction, its potential role in promoting cell proliferation and differentiation remains unclear. Previous studies have shown that ROS generation correlates with ultrasonic power. Building on CMS-mediated immunomodulation of the osteogenic microenvironment, we designed a strategy using B-TNs to precisely deliver piezoelectric nanoparticles to BMSCs via homotypic targeting[[133][44], [134][45], [135][46]]. We then applied ultrasound at varying power levels to induce intracellular ROS generation and evaluated its effects on proliferation and differentiation. Flow cytometry ([136]Fig. S17) and statistical analysis ([137]Fig. 4A) showed a power-dependent increase in intracellular ROS levels from 0.1 to 0.5 W/cm^2, with hydrogen peroxide levels following a similar trend ([138]Fig. 4B). To determine the optimal ultrasonic power, we conducted Edu staining ([139]Fig. S18) and PCR analyses ([140]Fig. 4C and D, S19). Cell proliferation peaked at 0.2 W/cm^2 but declined at power levels above 0.3 W/cm^2. Addition of the ROS inhibitor NAC significantly suppressed proliferation, confirming the essential role of ROS in this process. qPCR further demonstrated that osteogenic gene expression (Runx2, OPN, and ALP) was significantly upregulated at 0.2 W/cm^2 compared to the control. However, NAC treatment or ultrasonic power exceeding 0.3 W/cm^2 led to a marked decline in gene expression. These results indicate that ROS generated at 0.2 W/cm^2 promotes BMSC proliferation and differentiation. Based on these findings, we selected 0.2 W/cm^2 as the optimal ultrasonic stimulation power. Fig. 4. [141]Fig. 4 [142]Open in a new tab B-TNs@CMS improves BMSCs proliferation, differentiation and mineration in DM environment by generating an appropriate piezoelectric stimulation. (A) Flow cytometry results statistic chart for ROS content. (B) Statistic chart for H[2]O[2] content test results. (C, D) qPCR statistic analysis for the expression of osteoblast-related genes RUNX2 and OPN. (E) Edu image of BMSCs among different groups, scale bars: 100 μm. (F, G) Morphological staining results of BMSCs, scale bars: 20 μm. (H–K) ALP (7 days) staining, ARS (14 days) staining and quantitative analysis, scale bars: 20 μm. (L) Heatmap of osteoblast-related gene expression by qPCR. (M) WB results and quantitative analysis of osteoblast-related protein: RUNX2, OPN, OCN expression. Data presented as mean ± SD, n = 3 per group. ∗p < 0.05, ∗∗p < 0.01,by one-way ANOVA with Tukey's post-hoc test. B-TNs@CMS treatment was evaluated using Edu staining ([143]Fig. 4E–S20). Under DM conditions, cell death increased, and proliferation was significantly impaired compared to the control. While CMS treatment partially restored proliferation by modulating the immune microenvironment, subsequent B-TNs@CMS treatment, which provided controlled ROS levels, led to a significant recovery. Similarly, cell morphology staining ([144]Fig. 4F and G) revealed abnormal BMSC morphology under DM conditions, which improved with CMS and returned nearly to normal after B-TNs@CMS stimulation. These findings demonstrate that B-TNs@CMS effectively enhances BMSC proliferation and restores morphology under DM conditions. Our results preliminarily validate B-TNs@CMS as a promising strategy for bone regeneration in DM environments. Traditional piezoelectric materials have shown limited osteogenic efficacy, primarily targeting non-pathological bone repair, and their lack of targeted delivery increases off-target risks post-implantation. In contrast, the B-TNs@CMS system mitigates excessive inflammation while utilizing BMSC membrane encapsulation for precise targeting. Under US guidance, this bioengineered platform enables spatiotemporally controlled ROS release, mimicking physiological signaling and restoring BMSC function impaired by diabetes. 3.4. Osteogenic potential of the composite hydrogel under diabetic condition in vitro and in vivo BMSCs are essential for bone regeneration. To assess the impact of B-TNs@CMS on osteogenic differentiation and mineralization, we treated BMSCs with B-TNs@CMS and conducted ALP and ARS staining, followed by statistical analysis ([145]Fig. 4H–K). Under DM conditions, both differentiation and mineralization were significantly impaired. While CMS treatment partially restored mineralization, a considerable deficit remained. However, B-TNs@CMS fully restored mineralization capacity. qPCR analysis ([146]Fig. 4L) showed that osteogenic gene expression (RUNX2, OPN, OCN, and ALP) was markedly suppressed in the DM group, whereas B-TNs@CMS significantly upregulated these genes. WB analysis ([147]Fig. 4M) further confirmed that B-TNs@CMS restored osteogenic protein expression and BMSC functionality under DM conditions. To validate these findings in vivo, we induced DM in rats using STZ and a high-fat, high-sugar diet, followed by femoral defect creation ([148]Fig. 5A). Rats were divided into four groups: Control, DM, CMS, and B-TNs@CMS. The B-TNs@CMS group received hydrogel implantation with daily US stimulation (0.2 W/cm^2, 10 min). Micro-CT imaging at 4 and 8 weeks ([149]Fig. 5B–S21) revealed severe impairment of bone regeneration in DM rats, with minimal healing. However, B-TNs@CMS significantly improved bone repair. Micro-CT analysis showed that BV/TV ([150]Fig. 5C) in the DM group was only ∼6 % of the control at 4 weeks, increasing to 17 % at 8 weeks. In the CMS group, values reached 48 % and 50 %, respectively. Notably, the B-TNs@CMS group achieved 93 % and 97 % of the control at 4 and 8 weeks, demonstrating substantial improvement. Similar trends were observed in Tb.N, Tb.Sp, Tb.Th, and AGV ([151]Fig. 5D–S22-S24). These results confirm that DM severely impairs bone regeneration, while B-TNs@CMS effectively enhances bone repair in DM environments. Fig. 5. [152]Fig. 5 [153]Open in a new tab B-TNs@CMS influences macrophages and osteogenic effects in vivo experiments. (A) Schematic diagram of model construction in DM rats. (B–D) Micro-CT scan results and quantitative analysis of bone regeneration in DM rats at 4 and 8 weeks. (E–G) HE and Masson staining results and quantitative analysis at 4 and 8 weeks, scale bars: 100 μm. (H–J) Fluorescence staining results and quantitative analysis of cell surface markers CD86 and CD206, scale bars: 100 μm. (K, L) Fluorescence staining results and quantitative analysis of osteogenic protein RUNX2 in bone tissue, scale bars: 100 μm. Data presented as mean ± SD, n = 3 per group. ∗p < 0.05, ∗∗p < 0.01,by one-way ANOVA with Tukey's post-hoc test. HE staining revealed inflammatory cell aggregation at the bone defect site in the DM group, with significantly reduced new bone formation (red-stained area) at 4 and 8 weeks, measuring approximately 16.9 % and 29.7 %, respectively. In the CMS group, these values increased to 24.6 % and 37.5 %. The B-TNs@CMS group not only alleviated inflammation and improved the osteogenic immune microenvironment but also enhanced new bone formation to 39.8 % and 56.0 %, closely approaching control levels (46.8 % and 59.1 %). Masson's trichrome staining highlighted collagen fibers (blue) indicative of new bone formation. At 4 and 8 weeks, collagen deposition was 49.5 % and 59.1 % in the control group, while the DM, CMS, and B-TNs@CMS groups exhibited 17.6 % and 29.7 %, 27.2 % and 43.2 %, and 41.6 % and 56.0 %, respectively. These results confirm that B-TNs@CMS effectively promotes in vivo bone regeneration, suggesting that enhancing the osteogenic immune microenvironment combined with piezoelectric stimulation improves osteogenic activity in a DM setting ([154]Fig. 5E). Immunofluorescence staining further corroborated these results. CD86 (red) and CD206 (green)) staining, markers for M1 and M2 macrophages, respectively, showed significant M1 macrophage accumulation in the DM group. B-TNs@CMS treatment reduced M1 macrophages while increased M2 macrophage expression, indicating an improved osteogenic immune microenvironment. Additionally, Runx2 (green) staining demonstrated that DM suppressed osteogenesis, whereas B-TNs@CMS restored osteogenic protein expression to near-control levels. These results confirm that B-TNs@CMS enhances BMSC osteogenic mineralization in a DM environment both in vitro and in vivo. HE staining of major organs (liver, kidney, and spleen) from B-TNs@CMS-treated rats subjected to US intervention showed no significant pathological abnormalities ([155]Fig. S25), demonstrating the biosafety of the treatment. We explored the role of ROS in regulating BMSC proliferation and differentiation. BMSCs were cultured in a DM environment with or without B-TNs@CMS treatment, followed by transcriptome analysis. Volcano plots ([156]Fig. 6A and B) highlighted differentially expressed genes among the control, DM, and B-TNs@CMS groups. GO analysis ([157]Fig. 6C and D) revealed that the DM environment impaired bone development and morphogenesis, whereas B-TNs@CMS treatment enriched osteogenesis-related functions. KEGG pathway analysis ([158]Fig. 6E and F) identified Wnt signaling alterations between the control and DM groups and between the DM and B-TNs@CMS groups. GSEA analysis ([159]Fig. 6G and H) further confirmed that the DM environment suppressed Wnt pathway-related genes, which were restored by B-TNs@CMS intervention. This trend was further confirmed in qPCR experiments ([160]Fig. S26). Fig. 6. [161]Fig. 6 [162]Open in a new tab Transcriptomic analysis of BMSCs under different intervention. (A, B) Volcano plots of differential proteins. (C, D) GO functional enrichment analysis. (E, F) KEGG pathway enrichment analysis. (G–J) GSEA analysis of Wnt pathway and gene heatmap results. The Wnt signaling pathway is critical for cell proliferation and differentiation. Moderate ROS levels activate Dishevelled (DVL), leading to dissociation of the Axin-CK1-GSK3β complex and stabilization of β-catenin, thereby triggering Wnt/β-catenin signaling. Transcriptome analysis showed that B-TNs@CMS treatment at 0.2 W/cm^2 ultrasound generated ROS at levels sufficient to activate Wnt signaling, promoting BMSC proliferation and differentiation [[163]47]. However, excessive ROS accumulation in a DM environment disrupted homeostasis, inhibiting the Wnt pathway and impairing osteogenesis. These findings suggest that controlling ROS levels can regulate cellular behavior and influence cell fate, offering a potential therapeutic approach for improving osteogenesis in DM conditions. 4. Conclusion This study introduces a composite piezoelectric hydrogel with immune-regulatory and targeted stimulation properties for diabetic bone regeneration. By modulating macrophage polarization, the hydrogel promotes the transition from pro-inflammatory M1 to anti-inflammatory M2 macrophages in a DM environment, reducing local inflammation and creating a conducive microenvironment for BMSC proliferation and differentiation. Furthermore, US-responsive B-TNs stimulation generates controlled intracellular ROS, triggering BMSC proliferation and differentiation to facilitate bone regeneration under DM conditions. Mechanistically, CMS modulates macrophage polarization via the PI3K-Akt-mTORC1-LXR pathway, inhibiting excessive M1 macrophage production while increasing M2 macrophage prevalence, thereby supporting cell regeneration. Simultaneously, US-induced ROS activate the Wnt/β-catenin pathway in BMSCs, further enhancing osteogenesis. This engineered hydrogel system combines BMSC-targeting with osteo-immunomodulatory functions, offering an effective therapeutic strategy for diabetic bone regeneration. Experimental results confirm that B-TNs@CMS restores immune-osteogenic homeostasis, significantly improving bone regeneration efficiency. These findings establish B-TNs@CMS as a promising biomaterial for diabetic bone repair. CRediT authorship contribution statement Jintao Dong: Validation, Investigation, Writing – original draft, Methodology, Conceptualization. Wengang Dong: Validation, Investigation, Writing – original draft, Methodology. Huijie Ran: Validation, Conceptualization, Methodology. Dongxu Wu: Validation, Methodology. Xinli Wang: Validation, Methodology. Hongli Chen: Validation, Methodology. Jiahao Cao: Validation, Methodology. Xu Wang: Validation, Methodology. Xinsen Lin: Methodology. Wei Lei: Supervision, Funding acquisition, Writing – review & editing, Project administration. Tianji Wang: Funding acquisition, Writing – review & editing. Yafei Feng: Writing – review & editing, Methodology, Funding acquisition, Validation, Investigation. Ethics approval and consent to participate The Fourth Military Medical University Animal Research Ethics Committee's guiding principles (Protocol Number: 20220770) were followed in all experiments. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements