Nano-kinetic implant surface mediates osteo-antibacterial synergy through electromagnetic-mechanical coupling

Yangmengfan Chen; Hao Du; Heng Sun; Xingcan Huang; Duan Wang; Xuming Chen; Jiehao Chen; Xuanjun Zhang; Liqiang Zhou; Zongke Zhou

doi:10.1038/s41467-026-78462-4

. 2026 Oct 19;17:10692. doi: 10.1038/s41467-026-78462-4

Nano-kinetic implant surface mediates osteo-antibacterial synergy through electromagnetic-mechanical coupling

Yangmengfan Chen ¹, Hao Du ¹, Heng Sun ^2,³, Xingcan Huang ⁴, Duan Wang ¹, Xuming Chen ¹, Jiehao Chen ⁵, Xuanjun Zhang ^2,³, Liqiang Zhou ^2,^3,^4,^✉, Zongke Zhou ^1,^✉

PMCID: PMC13680427 PMID: 42897207

Abstract

Periprosthetic joint infection (PJI) is a devastating complication of total joint arthroplasty, mainly caused by bacterial biofilm formation and antibiotic resistance. Traditional static surfaces rely on contact killing-based bactericidal properties and cannot reliably clear mature biofilms. In this study, we fabricate a remote-activated dynamic biomimetic system, namely a movable magnetic topographic surface (MMTS). Here we show that under the exposure of external electromagnetic field (EMF), the anchored Janus nano-heterodimers oscillate periodically to mechanically disrupt bacterial membranes and eradicates biofilms of Staphylococcus aureus and Escherichia coli. This mechanical force can also activates Piezo1 mechanosensitive ion channel in bone marrow mesenchymal stem cells (BMSCs), inducing calcium oscillations, upregulating gene expression of Runx2 and BMP2, and enhancing extracellular matrix mineralization. In vivo, MMTS combined with EMF exposure eliminates local infection and promoted osseointegration. This remote-activated dynamic biomimetic strategy offers a promising multifunctional platform for PJI treatment, and improved implant osteointegration in clinical orthopedics.

Subject terms: Biomedical engineering, Biomedical materials, Tissue engineering and regenerative medicine

Periprosthetic joint infection (PJI) is a complication of total joint arthroplasty mainly caused by bacterial biofilm formation and antibiotic resistance. Here, authors develop a remote-activated dynamic biomimetic system for mechanical disruption of bacterial membranes and promotion of implant osteointegration for treatment of PJI.

Introduction

Total joint arthroplasty (TJA) is a well-established intervention for end-stage joint disease, as it effectively alleviates pain and restores physical function in patients with osteoarthritis. It has been reported that approximately 800,000 TJAs are performed annually in the UK and the United States, and this number is projected to surpass 4 million by 2030¹. Nevertheless, the growing volume of TJA surgery has led to a rising rate of postoperative complications, with periprosthetic joint infection (PJI) being the most prominent concern. PJI occurs in approximately 1% of total hip arthroplasties (THA) and 1–2% of total knee arthroplasties (TKA), imposing substantial clinical and economic burdens². Patients with PJI suffer from higher morbidity and mortality rate¹, while revision surgeries for PJI also place a heavy financial burden on healthcare systems. It was reported that treating a single PJI case costs an average of €95,000, which is five times the cost of a primary arthroplasty, owing to repeated surgical interventions, prolonged rehabilitation, and long-term administration of antibiotics and analgesics³. The increase in revision surgery poses a major threat to global healthcare systems, highlighting an urgent need to develop more effective strategies for PJI prevention and treatment.

Cefazolin (CEZ) is a guideline-recommended antibiotic against methicillin-susceptible Staphylococcus aureus⁴. Owing to its favorable safety, it has been widely applied in clinical practice⁵. For TJA, CEZ is the standard of care for preoperative antibiotic prophylaxis⁶. However, bacterial biofilm formation on implant surfaces constitutes a major obstacle in treating PJI. Biofilms are organized bacterial communities embedded within a protective extracellular polymeric matrix. This matrix functions as a barrier, protecting bacteria from antimicrobial agents including CEZ and host immune defenses, thereby leading to chronic infections⁷. Another challenge is the growing prevalence of antibiotic-resistant pathogens. For example, in the United States, up to 46.7% of S. aureus strains are methicillin-resistant (MRSA), and up to 23% of Enterococci spp. are vancomycin-resistant¹. It is projected that drug-resistant pathogens will cause 10 million deaths worldwide each year by 2050⁸; there is an urgent need for innovative antimicrobial strategies, particularly non-antibiotic therapeutics⁹. Current biomaterial-based strategies often rely on antibiotic coatings or metallic antibacterial agents, which face limitations such as potential cytotoxicity, burst release profiles, and the risk of inducing further bacterial resistance¹⁰. For instance, antibacterial coatings on titanium (Ti) implants often suffer from poor stability, rapid decomposition¹¹, and might lead to implant failure over time¹². Collectively, these limitations have driven extensive research into non-antibiotic and physical antimicrobial modalities¹³.

In this context, biomimetic design has emerged as a promising research direction¹⁴. In particular, bionic approaches that mimic natural antibacterial surface topographies or inherent physiological defense systems offer viable strategies for antimicrobial therapy without leading to bacterial resistance¹⁵. A conventional example is the micro-nanostructure of insect wings, which exhibits contact-killing effects by physically rupturing bacterial cells without inducing resistance¹⁶. However, our previous studies revealed that many static biomimetic designs only focus on structural imitation, ignoring adaptability to the dynamic physiological environment, leading to unsatisfactory bactericidal performance^17–19. To overcome this limitation, dynamic bionic strategies that simulate responsive biological behaviors have been developed²⁰. A recent breakthrough in this field is the fabrication of magnetic urchin-like capsule robots. Driven by an external magnetic field, these robots form microswarms that mechanically disrupt bacterial biofilms, showing great potential for the treatment of implant-associated infections²¹.

The utilization of precisely regulated physical stimuli represents a particularly promising direction of research. Mechanical forces are essential regulators of bone remodeling²², and they can also directly impair bacterial viability and disrupt biofilm²³. An ideal therapeutic strategy for PJI should not only eradicate infection without inducing bacterial resistance, but also restore the local physiological environment to facilitate tissue repair and functional recovery²⁴. In this study, we developed a dynamic bionic system. By integrating magnetic components, the system mimics both the nanostructure of insect wings and their flapping motions upon the application of an electromagnetic field (EMF) (Fig. 1). This platform is composed of Fe₃O₄-Ag Janus nano-heterodimers (JNHDs) tethered to a nano-hydroxyapatite (nHA)-modified titanium (Ti) substrate through a flexible linker, forming a movable magnetic topographic surface (MMTS). Upon exposure to an EMF with designated frequencies and intensities, the JNHDs generate periodic mechanical vibration. This design serves two major purposes: (1) mimicking insect wing flapping kinematics (30–50 Hz)²⁵, to achieve contact-killing through remotely modulated mechanical vibration; (2) mimicking exercise-induced mechanical stimulation (10–128 Hz)²⁶ to accelerate bone regeneration and osseointegration. We further investigate the underlying molecular mechanisms by which mechanical activation of Piezo1 channels on primary cilia converts extracellular mechanical stimuli into biochemical signals through calcium oscillations, thereby enhancing osteogenesis and mineral deposition.

Fig. 1 — a The synthesis process of Fe₃O₄-Ag Janus nano-heterodimers (JNHDs) and movable magnetic topographic surface (MMTS). b Under external EMF exposure, MMTS undergoes reversible structural changes mimicking insect wing flapping. During the activated state, magnetic actuation exposes Ag NPs, enabling contact-killing antibacterial activity. When suppressed, Ag NPs are concealed, reducing cytotoxicity and favoring tissue integration. Alternating EMF drives periodic mechanical oscillation of JNHDs on MMTS, physically disrupting bacterial integrity. Furthermore, these mechanical oscillations are transduced via Piezo1-mediated calcium signaling in BMSCs, promoting osteogenic differentiation.

In this study, a non-antibiotic strategy that synergistically combines remote-activated mechano-bactericidal activity with mechano-stimulated bone regeneration (Fig. 1) was compared with the first-line clinical antibiotic CEZ in the orthopedic department²⁷. This advanced electromagnetic-mechanical coupling offers a promising therapeutic approach to eliminate infection and restore function.

Results

Preliminary exploratory analysis of S. aureus and bone harvested from human hip prosthesis suggest the key role of mechanical stimulation

To provide preliminary exploratory insight into bacterial and host adaptations in the periprosthetic microenvironment, we first conducted transcriptomic and metabolomic sequencing of S. aureus isolated from the femoral component of patients undergoing revision total hip arthroplasty for PJI (Fig. 2A–E). The diagnosis of PJI was confirmed by the radiographic identification of bone destruction, focal osteosclerotic lesions, and inflammation (Supplementary Figs. 1–3). Analysis of Clusters of Orthologous Groups (COG) revealed differences in energy production, cell wall/membrane biogenesis, and structural organization between patient-derived S. aureus and laboratory reference strains (Fig. 2F). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis also indicated alterations in drug resistance, membrane transport, and metabolism of essential nutrients (Fig. 2G). To further evaluate the alterations in metabolism of the periprosthetic S. aureus derived from patients, metabolomic sequencing was conducted. A Venn plot revealed 1022 metabolites that were common to both groups, while a considerable number of metabolites were uniquely identified in each group (Fig. 2H). A volcano plot confirmed significant differences in the abundance of metabolites between the two groups (Fig. 2I). Subsequent compound classification showed that these metabolic differences primarily involved monosaccharides, oligosaccharides, phospholipids, and fatty acids, which are key components associated with the bacterial wall and membrane biosynthesis (Fig. 2J). The pathway classification also highlighted changes in amino sugar and nucleotide sugar metabolism, glycerophospholipid metabolism, and ABC transporters. Taken together, our multi-omics data (Supplementary Fig. 4) suggested that patient-derived S. aureus undergoes remodeling of the cell wall and membrane synthesis pathways. These adaptations are likely a response to the unique microenvironment shaped by mechanical stimulation from patient rehabilitation and daily activities, which promotes functional, metabolic, and genetic alterations in the periprosthetic bacteria.

Bone tissue was also collected from the cementless acetabular cup of the PJI patient (Fig. 2L–O). Hematoxylin and eosin (H&E) staining (Fig. 2P) and Masson’s trichrome staining (Fig. 2Q) revealed new bone formation at the interface of bone and acetabular cup, despite the presence of focal fat deposition. However, no bacterial colonies were observed upon Gram staining, probably due to effective pre-surgical infection control (Fig. 2R). Furthermore, immunofluorescent staining for BMP2 (Fig. 2S) and COL1A1 (Fig. 2T) demonstrated active bone formation at the implant–bone interface. Notably, Piezo1 was also widely distributed along this interface, suggesting its potential involvement in the process of osteointegration (Fig. 2U). Taking the obtained findings together, the transcriptomic and metabolomic analyses revealed alterations in the synthesis and metabolism of the membrane and cell wall of S. aureus around the implant. Together with the observed bone formation and localization of Piezo1, these preliminary exploratory findings underscore the critical role of mechanical stimulation in modulating host-pathogen interactions and suggest its potential therapeutic application in the management of PJI.

Characterization and principle of the dynamic biomimetic MMTS

Inspired by biomimetic principles, we examined the micro/nanostructural features of the wing of Pomponia linearis (Fig. 3A) as well as those of other winged insects (Supplementary Fig. 5). To emulate the surface topography of the P. linearis wing, we first synthesized Fe₃O₄-Ag JNHDs, as shown by transmission electron microscopy imaging (Fig. 3B, C). Elemental mapping confirmed the spatial distribution of Fe, Ag, and O within these JNHDs (Fig. 3D–G), while energy-dispersive X-ray spectroscopy (EDS) revealed characteristic peaks consistent with their expected composition (Fig. 3H). Particle size analysis indicated average diameters of 18.54 nm for Fe₃O₄ nanoparticles (NPs) and 5.18 nm for Ag NPs (Fig. 3I). The paramagnetic behavior of (a) Ag NPs, (b) Fe₃O₄ NPs, and (c) JNHDs was assessed under a static magnetic field (SMF). Notably, both Fe₃O₄ NPs and JNHDs could be oriented and enriched, transitioning from a (1) dispersed state to an aggregated state, accumulating at (2) the top and (3) the side of the droplet (Fig. 3J and Supplementary Movie 1).

To endow titanium (Ti) orthopedic implants with dynamic biomimetic properties, JNHDs were either directly deposited onto nHA-Ti surfaces to create a static magnetic topographic surface (UMTS), or elastically tethered via a flexible linker (MAL-PEG-NHS) to form a MMTS. Scanning electron microscopy (SEM) images illustrated the nanostructural features of the Ti substrate, UMTS, and MMTS (Fig. 3K). EDS spectra of the Ti substrate showed only Ti peaks (Fig. 3L), whereas UMTS exhibited additional Fe and Ag signals (Fig. 3M). Importantly, the presence of a Si peak in MMTS was noted (Fig. 3N). Elemental mapping further verified the distribution of O, Ca, Si, Ti, Fe, and Ag across all specimens (Fig. 3O), confirming successful functionalization. Atomic force microscopy (AFM) revealed that both UMTS and MMTS surfaces exhibited greater roughness and higher spatial frequency than the pure Ti substrate (Fig. 3P, Q), as supported by representative AFM images (Fig. 3R). Fourier-transform infrared (FT-IR) spectroscopy identified vibrational bands associated with C–N, Si–O–Si, and C=O bonds, corroborating the successful conjugation of the flexible linker to the functionalized JNHD and nHA-Ti substrate (Fig. 3S). The incorporation of the polymer linker and SiO₂ coating significantly enhanced the hydrophilicity of UMTS (Fig. 3T). Critically, to evaluate the stability afforded by the flexible linker, we measured the release kinetics of Fe (Fig. 3U) and Ag ions (Fig. 3V). The results demonstrated that the flexible linker substantially decelerated ion release. Although exposure to an EMF induced mechanical instability in the JNHDs, MMTS exhibited superior sustained-release properties compared with UMTS, attributable to the covalent bonding provided by the flexible linker (Fig. 3W).

Dynamic biomimetic MMTS under EMF exposure enhances cell viability and adhesion

Live/dead staining was performed to evaluate the biocompatibility of each group. The results showed that UMTS led to substantial cell death, especially upon EMF exposure. Whereas the cytotoxicity of MMTS was reduced, and CEZ was similar to Ti, which exhibited no obvious cytotoxicity (Fig. 4A). CCK-8 assay showed slight reductions in cell viability in UMTS, MMTS, and CEZ groups compared with Ti (Fig. 4B). This reduced cytotoxicity was correlated with the increased intracellular reactive oxygen species (ROS) levels (Fig. 4C). Specifically, UMTS can induce excessive ROS accumulation, but it was significantly suppressed in MMTS due to controlled JNHD burst release (Supplementary Fig. 6).

Fig. 4 — A Live/dead staining of BMSCs. B Cell viability analysis assessed by CCK-8 assay; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. C Quantitative analysis of ROS generation in BMSCs; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. D Quantification of BMSCs spreading area; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. E Representative SEM images of BMSCs on each specimen. F Cytoskeleton (phalloidin, green) and nuclei (DAPI, blue) in BMSCs on the surface of each specimen. The immunofluorescent staining of G Itg-β3 and H Itg-α5 in BMSCs. I Nuclear co-localization analysis of Runx2 (Green) and DAPI (Blue). CEZ group: clinical control, without EMF exposure. a.u. denotes arbitrary units. Data were compared by one-way ANOVA followed by Tukey’s multiple comparison test. Source data are provided as a Source data file.

For BMSCs, the cell spreading suggests a high capacity for activity and differentiation. Quantitative analysis indicated that UMTS, with or without EMF exposure, significantly reduced cell spreading. In contrast, no significant difference was observed between Ti, MMTS, and CEZ groups. This improvement of MMTS can be attributed to the bendable linker that mitigates the burst release of JNHDs from MMTS (Fig. 4D). As the morphology of BMSCs showed, BMSCs on UMTS and UMTS + EMF appeared slightly crumpled compared with those on Ti, likely due to the cytotoxicity of Ag NPs. However, the crumpled morphology was relieved on the surface of MMTS and MMTS + EMF, while CEZ had almost no effect on the morphology of BMSCs (Fig. 4E, F). Due to the importance of cell adhesion to BMSC function, the integrin levels were also evaluated in BMSCs with immunofluorescent staining. Both UMTS and MMTS surfaces markedly elevated Itg-β3 levels, which were further enhanced upon EMF exposure, while Itg-β3 levels remained at basal levels in the CEZ group (Fig. 4G). Similarly, the level of Itg-α5 also showed increased levels in both UMTS + EMF and MMTS + EMF, but showed no change in the CEZ group (Fig. 4H). As a key transcription factor, immunofluorescence staining of Runx2 revealed increased fluorescence intensity in UMTS + EMF and MMTS + EMF (Supplementary Fig. 7). The nuclear translocation was also enhanced in both UMTS + EMF and MMTS + EMF groups, while CEZ had no effect on the level or nuclear translocation of Runx2 in BMSCs (Fig. 4I).

Periodic mechanical stimulation induced by dynamic biomimetic MMTS under EMF exposure enhances osteogenesis of BMSCs

Periodic mechanical stimulation plays a crucial role in promoting the osteogenic differentiation of BMSCs and preventing bone loss. In this study, such stimulation was applied to the dynamic biomimetic MMTS surface under an EMF. To investigate the molecular response of BMSCs to this stimulation, RT-qPCR was performed to assess the expression of Itg-α5, Itg-β3, OPN, Runx2, and BMP2. The results revealed that both UMTS and MMTS upregulated the expression of these integrin and osteogenic genes in the BMSCs. Moreover, EMF exposure further significantly enhanced their expression in both UMTS + EMF and MMTS + EMF groups (Fig. 5A). Western blot analysis (Fig. 5B) and quantification (Fig. 5C) corroborated these findings, demonstrating markedly elevated protein levels of ITG-α5, ITG-β3, OPN, Runx2, and BMP2 in the MMTS + EMF group. The osteogenic differentiation capacity of BMSCs was further evaluated using an ALP activity assay. The results showed that both UMTS and MMTS slightly increased ALP activity, while EMF exposure substantially enhanced it (Fig. 5D). Matrix mineralization, assessed by Alizarin Red S staining after 28 days of its induction, showed that EMF strongly promoted new bone formation on both UMTS and MMTS (Fig. 5E). Similarly, ALP staining confirmed the strong pro-osteogenic effect of MMTS upon EMF exposure (Fig. 5F). In addition, the influences of CEZ, which is a first-line antibiotic, on osteogenesis were also evaluated with Alizarin Red S staining (Supplementary Fig. 8A) and ALP staining (Supplementary Fig. 8B), the results demonstrated that CEZ can not promote osteogenesis, rather, it even showed a slight inhibitory effects.

Fig. 5 — A RT-qPCR analysis of integrins (*Itg-α5, Itg-β3*), and osteogenesis-associated genes (OPN, Runx2, BMP2) of BMSCs on each specimen; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. B Representative western blot bands, and C Quantitative data of protein levels of Itg-α5, Itg-β3, OPN, Runx2, and BMP2; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. D ALP activity of BMSCs; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. BMSCs were seeded and cultured on a silicon dioxide coverslip-based control, UMTS, and MMTS for E Alizarin Red S staining after 28 days, and F ALP staining after 14 days of culture. G Schematic illustration of the influences of mechanical stimulation on bone homeostasis. Created in BioRender. Chen, Y. (2026) https://BioRender.com/gohw1kx. Data were compared by one-way ANOVA followed by Tukey’s multiple comparison test. Source data are provided as a Source data file.

The enhanced osteogenic effects conferred by MMTS and EMF are attributed to the periodic mechanical stimulation generated on the MMTS surface upon the application of an EMF. This mechanism is biomimetically designed to simulate the natural mechanical stimulation to which bone is subjected during exercise. Specifically, upon exercise, mechanical force is transduced into the bone, thus activating the mechanical sensor Piezo1, leading to calcium influx that in turn promotes new bone formation (Fig. 5G). In summary, both UMTS and MMTS promoted the osteogenic differentiation of BMSCs, and EMF-induced periodic mechanical stimulation further amplified this osteogenic response.

Activation of Piezo1 and calcium ion oscillations induced by dynamic biomimetic MMTS upon EMF exposure

Piezo1 is a mechanosensitive ion channel present on most mammalian cells, capable of sensing mechanical stress and transducing it into biochemical signals to regulate cellular functions. Initially, we assessed the protein level of Piezo1 across different stages of BMSC differentiation. Western blot analysis revealed that the level of Piezo1 on days 7 and 14 was higher than on day 3, while it decreased after day 28. This indicated that the level of Piezo1 is crucial in the middle stage of osteogenic differentiation and determines the fate of BMSCs (Fig. 6A, B). Mechanical activation of Piezo1 triggered calcium influx, leading to elevated intracellular Ca²⁺ levels and resulting in calcium oscillations (Fig. 6C). To investigate the functional role of Piezo1 in BMSCs, we employed its pharmacological activator Yoda1. Treatment with Yoda1 induced calcium influx and oscillations in a concentration-dependent manner (Fig. 6D). Conversely, pretreatment with the Piezo1 antagonist Dooku1 effectively blocked Yoda1-induced calcium responses (Fig. 6E), confirming that Piezo1 is functionally expressed on BMSCs and essential for mechanotransductive signaling (Supplementary Fig. 9). We next evaluated the osteogenic ability of Piezo1. Alizarin Red S staining (Fig. 6F) and ALP staining (Fig. 6G) demonstrated that Yoda1 enhanced matrix mineralization and ALP activity, whereas Dooku1 suppressed these effects, underscoring the critical role of Piezo1-mediated calcium influx in osteogenesis.

Fig. 6 — A Representative western blot images of Piezo1 in BMSCs. B Quantification of Piezo1 protein levels; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. C Schematic illustrating Piezo1-mediated mechanotransduction. Created in BioRender. Chen, Y. (2026) https://BioRender.com/nwwp37d. D Real-time intracellular calcium dynamics in BMSCs treated with 0, 5, 10, or 20 μM Yoda1. E Calcium responses in BMSCs pre-treated with Dooku1 (0-20 μM) followed by 10 μM Yoda1. F Alizarin Red S and G ALP staining of BMSCs treated with Yoda1 and/or Dooku1. H Calcium dynamics in BMSCs cultured on Ti, UMTS, or MMTS with/without EMF. I Calcium responses in Dooku1-pre-treated BMSCs under the same conditions. J Fluorescent staining of primary cilia (green, anti-acetylated tubulin, indicated with white arrows), and nuclei (blue, DAPI). K Alizarin red staining and L ALP staining. M Schematic illustrating the importance of primary cilia in BMSCs in perceiving mechanical stimulation. a.u. denotes arbitrary units. Created in BioRender. Chen, Y. (2026) https://BioRender.com/tb5ct1c. Data were compared by one-way ANOVA followed by Tukey’s multiple comparison test. Source data are provided as a Source data file.

To determine whether the dynamic biomimetic MMTS under EMF exposure could generate periodic mechanical stimulation sufficient to activate Piezo1, we monitored real-time calcium dynamics in BMSCs. No calcium oscillations were observed in cells cultured on Ti, UMTS, or MMTS in the absence of EMF. However, EMF exposure induced pronounced calcium oscillations and increased fluorescence intensity in both UMTS and MMTS groups (Fig. 6H). These effects were abolished by pretreatment with Dooku1 (Fig. 6I), confirming the involvement of Piezo1 in this mechanoresponsive process (Supplementary Fig. 10). As primary cilia serve as critical mechanosensors in mammalian cells, immunofluorescence staining was performed to evaluate their presence and structural integrity in BMSCs. The results confirmed that primary cilia were present in BMSCs, and their formation was effectively inhibited upon transfection with si-IFT20 (Fig. 6J). Meanwhile, knockdown of IFT20 disrupted ciliary integrity, leading to suppressed matrix deposition (Fig. 6K) and reduced ALP activity (Fig. 6L). These findings indicate that the presence and structural maintenance of primary cilia in BMSCs are essential for Piezo1-mediated perception of extracellular mechanical signals and subsequent osteogenic differentiation (Fig. 6M). In summary, Piezo1 is functionally expressed in BMSCs and plays a pivotal role in osteogenic differentiation. The dynamic biomimetic MMTS, when activated by EMF, generates periodic mechanical stimuli that activate Piezo1 and induce calcium oscillations, thereby promoting osteogenesis.

Dynamic biomimetic MMTS with EMF exposure effectively kills bacteria and disrupts biofilms via mechanical destruction of bacterial structures

The dynamic biomimetic MMTS was fabricated to generate periodic mechanical stimulation not only to promote osteogenesis but also to eradicate bacteria and disrupt biofilms. This dual functionality is achieved through the electromagnetically driven capture of bacteria and subsequent mechanical disruption of their structural integrity (Fig. 7A).

Fig. 7 — A Schematic illustrating the antibacterial mechanism of MMTS: EMF-induced capture and mechanical destruction of bacteria. Created in BioRender. Chen, Y. (2026) https://BioRender.com/irvwm4o. B Live/dead staining of *S. aureus* and *E. coli*. C SEM images of bacterial morphology on different substrates. OD 600 measurements of planktonic D *S. aureus* and E *E. coli*; n = 4 biological replications, data were presented as mean values ± SD; error bars = SD. F Colony-adhesion assay. G Disk diffusion assay showing inhibition zones. *S. aureus* and *E. coli*. were seeded and cultured on a silicon dioxide coverslip-based control, UMTS, and MMTS for H Crystal violet-stained biofilms. Quantitative analysis of biofilm content for I *S. aureus* and J *E. coli*; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. CEZ group: clinical control, without EMF exposure. Data were compared by one-way ANOVA followed by Tukey’s multiple comparison test. Source data are provided as a Source data file.

The bactericidal efficacy against adherent bacteria was evaluated using live/dead staining. Both UMTS and MMTS exhibited slight antibacterial activity against S. aureus and E. coli, which was then significantly enhanced upon EMF exposure, due to the induced mechanical stimulation (Fig. 7B). The CEZ groups showed a similar bactericidal ability compared with the mechanical force-induced bactericidal ability (Supplementary Fig. 23). SEM imaging further revealed that, while UMTS and MMTS alone caused minor structural damage to bacterial cells, EMF activation resulted in more severe disruption of bacterial walls and membranes, leading to the leakage of intracellular contents. Similarly, CEZ treatment severely broke the integrity of the bacterial walls of both S. aureus and E. coli (Fig. 7C).

The quantitative results of bactericidal ability against planktonic bacteria (OD = 600) demonstrated significant reductions in bacterial concentration for both S. aureus (Fig. 7D) and E. coli (Fig. 7E) after treatment with UMTS or MMTS. With EMF exposure, their bactericidal ability has been greatly enhanced, even to a level similar to that of the first-line antibiotic CEZ. Colony-forming unit (CFU) assays showed that UMTS and MMTS alone can only slightly inhibit bacterial proliferation, but EMF exposure can dramatically enhance their bactericidal performance against S. aureus (Fig. 7F) and E. coli (Supplementary Fig. 11A). Similarly, the inhibition zones around UMTS and MMTS specimens indicated effective antibacterial diffusion. EMF exposure further enlarged these zones, likely due to the enhanced release of JNHDs against S. aureus (Fig. 7G). However, its bactericidal ability against E. coli is relatively weak (Supplementary Fig. 11B).

Biofilm formation is the major challenge of PJI. As the crystal violet staining showed, UMTS and MMTS alone partially inhibited biofilm formation, while EMF activation completely eradicated the biofilms, attributable to their mechanical disruption via the dynamic surface. Interestingly, CEZ showed a greater anti-biofilm effect against S. aureus than against E. coli (Fig. 7H). Quantitative analysis verified the significantly improved anti-biofilm efficacy of EMF-induced mechanical stress against both S. aureus (Fig. 7I) and E. coli (Fig. 7J).

Dynamic biomimetic MMTS eliminates infection and promotes osteointegration upon EMF exposure in vivo

To assess the impact of dynamic biomimetic MMTS on infection clearance and bone regeneration, an infected Sprague–Dawley (SD) rat tibial osteomyelitis model was constructed. Cross-sectional μCT imaging revealed that, in the Ti group ( ± EMF exposure), the implant was surrounded by necrotic bone and a dead cavity (red arrows). In contrast, no necrotic bone or cavity was observed around the UMTS, MMTS, and CEZ implants. Moreover, robust new bone formation and stable osteointegration (green arrows) were only shown around both UMTS and MMTS upon EMF exposure (Fig. 8A). Three-dimensional reconstructed images further confirmed extensive peri-implant–bone formation in the UMTS and MMTS groups following EMF exposure (Fig. 8B). Coronal sections of the tibiae also showed substantial new bone formation around UMTS and MMTS, particularly upon EMF stimulation (Fig. 8C).

Fig. 8 — A Cross-sectional μCT imaging of tibial bone with different implants. B Three-dimensional reconstructed μCT images used to show the peri-implant–bone formation. C The three-dimensional reconstructed μCT image showing the implant (blue) and newly formed bone (red) in the tibial cavity. Quantitative analysis of the lengths of the D BV/TV, E Tb.Th, F Tb.N, and G Tb.Sp, of the tibial bone around the implant; n = 3 biological replications, data were presented as mean values ± SD; error bars = SD. H H&E staining. I Masson’s trichrome staining. J Gram staining used to show the bacterial colonies (black arrow). K Giemsa staining used to show the bacterial colonies (black arrow). CEZ group: clinical control, without EMF exposure. Data were compared by one-way ANOVA followed by Tukey’s multiple comparison test. Source data are provided as a Source data file.

Quantitative μCT analysis verified the synergistic effect of dynamic biomimetic MMTS and EMF exposure on enhancing new bone formation. The bone volume/tissue volume (BV/TV) ratio was significantly higher in the UMTS and MMTS groups, and EMF exposure further enhance BV/TV, even more than CEZ group (Fig. 8D). Although trabecular thickness (Tb.Th) did not differ significantly among the groups, the MMTS + EMF group showed a notable increasing trend (Fig. 8E). Trabecular number (Tb.N) was significantly elevated in both UMTS and MMTS groups under EMF exposure, more than CEZ group (Fig. 8F). Conversely, trabecular separation (Tb.Sp) was markedly reduced in the MMTS group, although EMF exposure have not further reduce it, the Tb.Sp of MMTS + EMF can still keep a similar level to the CEZ group (Fig. 8G). Taken together, these results indicate that MMTS + EMF exposure significantly enhances new bone formation, even more than conventional CEZ treatment, because of the periodic mechanical stimulation induced by the application of an EMF on the MMTS surface.

Histological staining was also performed to evaluate cellular and tissue responses in the vicinity of the implant. H&E staining revealed no notable differences among the groups, suggesting that neither surface modification nor EMF exposure provoked a local immune reaction (Fig. 8H). Masson’s trichrome staining indicated comparable collagen deposition around the UMTS and MMTS implants, and EMF exposure substantially increased the intensity of collagen staining, implying that both implant types can be activated by EMF to promote collagen synthesis, even better than the CEZ group (Fig. 8I). In addition, both Gram staining (Fig. 8J) and Giemsa staining (Fig. 8K) showed abundant bacterial colonies (black arrows) around the Ti implants, whereas few bacterial presence was detected around UMTS + EMF, MMTS + EMF, or CEZ. Moreover, the long-term relapse of infection was also evaluated after a 2-week treatment cessation (without EMF exposure or CEZ administration). Both Gram staining (Supplementary Fig. 27A) and Giemsa staining (Supplementary Fig. 27B) showed no bacterial colonies around UMTS + EMF, MMTS + EMF, indicating complete elimination of infection.

In summary, using an infected tibial osteomyelitis model, we demonstrated that MMTS + EMF exposure synergistically eliminates peri-implant infection and facilitates robust osteointegration, which is even better than using first-line antibiotic CEZ. Therefore, this electromagnetic-mechanical coupling strategy is promising for PJI treatment.

In vivo evaluation of fluorescence distribution and systemic biocompatibility of dynamic biomimetic MMTS upon EMF exposure

To further evaluate the osteogenic microenvironment induced by MMTS, immunofluorescence staining was performed to assess the expression and spatial distribution of key proteins involved in bone regeneration. The results showed that type I alpha 1 collagen (COL1A1), a major component of bone matrix, was expressed at baseline levels around the Ti implant. In contrast, both UMTS and MMTS groups, especially upon EMF exposure, exhibited strong COL1A1 fluorescence at the bone–implant interface, indicating enhanced deposition of extracellular matrix (Fig. 9A). As the quantitative result showed, UMTS + EMF and MMTS + EMF showed a significant increased COL1A1 intensity than CEZ, indicating that mechanical stimulation can is important to further enhance the osteointegration (Fig. 9B). Bone morphogenetic protein 2 (BMP2), a critical inducer of osteogenic differentiation in the TGF-β superfamily, was minimally detected around the Ti implant. However, its expression was markedly elevated at the interface of UMTS and MMTS implants, particularly upon EMF stimulation, suggesting enhanced osteoinductive activity (Fig. 9C). However, as a popular antibiotic, CEZ had almost no effect in promoting BMP2 secretion around the bone–implant surface (Fig. 9D). Vascular endothelial growth factor A (VEGFA), which plays essential roles in angiogenesis and early bone healing, was primarily localized in vessel-like structures near the interface. The intensity and distribution of VEGFA signals were significantly stronger and more extensive in UMTS and MMTS groups with EMF exposure, indicating the promotion of vascularization during new bone formation (Fig. 9E). Interestingly, the quantitative analysis demonstrated that CEZ significantly increased VEGFA levels around implant (Fig. 9F). As HIF-1α is one of the most important factor that promote angiogenesis, the immunofluorescent staining HIF-1α was also performed. The result showed that UMTS + EMF, MMTS + EMF, and CEZ showed comparable HIF-1α intensity and distribution around the implant across all groups (Supplementary Fig. 12). For the mechanosensitive ion channel Piezo1, its distribution was broader around UMTS and MMTS implants, and the intensity was further enhanced with EMF exposure, implying possible mechanotransduction activation (Fig. 9G). While, as the quantitative result showed, CEZ has no obvious influence on Piezo1 distribution and intensity compared with Ti (Fig. 9H).

Finally, systemic biocompatibility was assessed via H&E staining for the major organs (heart, liver, spleen, lung, and kidney) after 4 weeks (Fig. 9E) and 6 weeks (Fig. 9F) of implantation and treatment. No pathological abnormalities, metaplasia, or necrosis were observed in any group, confirming the biosafety of the implants and EMF treatment.

In conclusion, the results of these in vivo evaluations demonstrate that dynamic biomimetic MMTS combined with EMF exposure significantly promotes key osteogenic and angiogenic factors and exhibits favorable systemic biosafety. These synergistic antibacterial and mechanostimulatory activities showed greater strengths even more than traditional antibiotic CEZ, thus providing a promising clinic approach for PJI.

Discussion

Despite the widespread use of antibiotics in reducing infection-related morbidity, bacterial infections remain a severe threat to global public health. A major concern lies in the remarkable ability of bacteria to evolve sophisticated drug-resistance mechanisms. First, planktonic bacteria secrete enzymes that degrade or structurally modify antibiotics, blocking their binding to molecular targets and diminishing therapeutic effects²⁸. Second, mutations in genes encoding penicillin-binding proteins, DNA helicase, or enzymatic alteration of drug target sites further reduce antibiotic efficacy²⁹. Third, bacteria also restrict intracellular drug accumulation by altering membrane permeability, downregulating drug influx channels or upregulating efflux pumps, constituting another critical resistance mechanism³⁰. Most importantly, the formation of biofilms further complicates antibiotic resistance. As structured communities encased in extracellular polymeric substances, biofilms protect resident bacteria by hindering antibiotic penetration, driving metabolic heterogeneity, and facilitating horizontal gene transfer. In the context of anti-biofilm, the “race to the surface” concept is well known, as it highlights the competition between host cells and bacteria to colonize implant surfaces. This race is between osseointegration and bacterial colonization of the implant. In 2019, CEZ was recommended by the American Academy of Orthopaedic Surgeons guidelines as the gold standard prophylactic antibiotics³¹. Therefore, the CEZ group was included as a clinical reference, given its extensive usage³² and high bactericidal ability³³ in orthopedic surgery. In this study, we demonstrated that nano-kinetic surface of MMTS can be activated with EMF exposure, to induce a electromagnetic-mechanical coupling and disrupt the bacterial wall and membrane (Supplementary Fig. 16), without the generation of ROS (Supplementary Fig. 17). It not only produce a comparable bactericidal ability as CEZ against S. aureus and E. coli, but also exhibited similar bactericidal ability of vancomycin against MRSA (Supplementary Fig. 21). More importantly, this electromagnetic-mechanical coupling can activate Piezo1 and Ca²⁺ influx to strongly promote new bone formation and osteointegration, which cannot be achieved by the antibiotic like CEZ.

Preventing bacterial colonization is critical for rapid osseointegration. Once bacterial colonization adherent to the implant surface first, they substantially impair host cell adhesion and tissue repair, thereby accelerating biofilm formation and hindering implant osseointegration³⁴. Accordingly, an effective anti-biofilm strategy needs to fulfill the following requirements: (1) prevention of initial bacterial adhesion while maintaining high viability of host cells on implant surfaces in the early post-implantation stage; (2) delivery of antimicrobial agents specifically to infected sites to improve local therapeutic efficacy; (3) improve penetration of antimicrobial agents across the biofilm’s extracellular polymeric substances, so that bactericidal concentrations can reach embedded bacteria; and (4) exert broad-spectrum activity against polymicrobial communities to address diverse and coexisting pathogenic populations²¹.

To meet these requirements, surface engineering has undergone rapid development³⁵. Specifically, it has been shown that micro- and nanostructural modifications of implants can inhibit bacterial adhesion and kill bacteria via either contact-killing or localized antimicrobial release, providing important protection³⁶. It has been reported that mechanical stress within the bacterial wall disrupts the tripartite efflux complex CusCBA in E. coli. This disassembly increases bacterial susceptibility to metal toxicity, demonstrating that mechanical forces can directly compromise bacterial survival³⁷. To endow implant surfaces with bactericidal capacity and optimize their biological performance, biomimetic modification approaches have attracted substantial research attention in recent years. Integrating nanotechnology with bioinspired design principles has been proven to improve targeting accuracy, therapeutic efficacy, and biocompatibility of implant materials³⁸. Particularly, biomimetic antibacterial materials represent an emerging class of nature-derived biomaterials. By virtue of anti-adhesive and bactericidal mechanisms, these materials exhibit high efficacy in eliminating planktonic bacteria and biofilms and can effectively treat implant-associated infections via anti-adhesion and bactericidal mechanisms³⁹. These advantages have generated growing interest in the use of bionic strategies for treating PJI due to their great potential for effectively preventing and eradicating infection. To date, many studies have explored various antibacterial approaches⁴⁰. For instance, Dai et al. developed cobalt-doped peony flower-like Bi₂O₃ (CBO). The CBO sample with a cobalt doping concentration of 0.4% exhibited a narrowed band gap, which significantly facilitated electron-hole separation under light irradiation. This structural optimization enhanced the generation of superoxide anions, hydroxyl radicals (•OH), and hydrogen peroxide (H₂O₂), thereby enhancing the laser-mediated bactericidal effect against E. coli⁴¹. Similarly, Wang et al. designed a biomimetic nanoadjuvant to combat biofilm-related infections. This nanoplatform is based on hollow mesoporous organosilica nanoparticles coated with macrophage membranes, which enables precise targeted delivery to infection sites. Loaded with immunomodulators (CpG-ODN or miR-21-5p), the platform provides sequential immunotherapy for infection treatment. In the early stage of infection, CpG-ODN induces the formation of neutrophil extracellular traps and promotes M1 macrophage polarization via the TLR9/MyD88/TRAF6 pathway. During the tissue repair phase, miR-21-5p induces M2 macrophage polarization through the PTEN/PI3K/Akt pathway. Combined with sonodynamic therapy that generate ROS, this integrated system effectively disrupts biofilms, eliminates bacteria, and enhances tissue integration, offering a multifunctional and comprehensive antibacterial therapeutic strategy⁴². Additionally, Zhang et al. developed a biodegradable Janus sonozyme (MON(CuMn)-Q) for the treatment of infected bone defects. This sonozyme exerts sonodynamic antibacterial activity through dual active sites (MnN and CuO) to generate antibacterial ROS. After degradation, the released Cu²⁺, Mn²⁺, and quercetin provide sustained antimicrobial effects. In a rat MRSA-infected cranial defect model, this sonozyme rapidly eliminated bacteria, alleviated local inflammation, and promoted bone regeneration, demonstrating effective and continuous antibacterial functionality alongside osteogenic promotion⁴³. Moreover, mussel-inspired hydrogels have also been reported to exhibit strong tissue adhesion and anti-infective capabilities⁴⁴. While conventional static biomimetic surfaces successfully replicate morphological features, they fundamentally overlook the dynamic responses that living organisms or implants exhibit in response to changing natural conditions⁴⁵. For instance, in our previous study, only limited intrinsic bactericidal ability was exerted by the insect wing-inspired TiO₂ nanotube via its morphology. Instead, its antibacterial effect was primarily dependent on the controlled release of encapsulated bactericidal agents¹⁸. This suggests that more potent bioinspired biomaterials can be achieved via dynamic rather than static mimicry. Advanced biomimicry is not limited to the replication of biological morphology, more importantly, it focuses on emulating the dynamic functions and adaptive responses to environmental stimuli. Based on the flapping kinematics of insect and human exercise-induced stimulation, we have performed a screening of EMF parameters; the results showed that EMF frequencies (0–100 Hz) and intensities (0–5 mT) have no obvious effects on cell viability, ALP activity, intracellular ROS level (Supplementary Fig. 24), and infection (Supplementary Fig. 25). These results further clarified it was the mechanical stimulation that induce significant biologiccal effects on cells.

Two critical biosafety concerns associated with the dynamic mimicry have been carefully considered. Although JHNDs can be effectively aligned in situ with an external field (Supplementary Movie 2, Supplementary Fig. 19), the mechanical vibration generated on the surface of MMTS was strictly controlled (Supplementary Fig. 20), within a physiologically safe mechanical window that mimics human exercise-induced stimulation. The systemic evaluation showed that JHNDs have a stable characterization after long-term daily EMF exposure (Supplementary Fig. 26). The flexible linker further served as a mechanical buffer to avoid excessive shear force or burst release to host cells. Notably, in vitro and in vivo biocompatibility results confirmed no obvious cytotoxicity or mechanical impairment, indicating negligible harmful shear stress to host cells and major organs (Supplementary Fig. 22). To evaluate the potential damage caused by EMF-induced magnetic hyperthermia, real-time thermal imaging was used; the results showed that magnetic hyperthermia does not occur in our system and poses no adverse effects (Supplementary Fig. 18).

Dynamic control biomaterials, such as magnetic ligand nano-assemblies on Fe₃O₄, enable the remote regulation of ligand presentation. This tunable property further modulates stem cell adhesion and differentiation through mechanotransductive pathways⁴⁶. It has also been reported that actuated anisotropic magnetic nanochains can effectively disperse biofilms via mechanical force, showing the effective removal of >90% of adherent bacteria under a magnetic field, without harming host epithelial cells⁴⁷. Notably, most conventional antimicrobial strategies primarily focus on preventing infection, while largely neglecting their potential effects on osseointegration has often been overlooked^11,48. However, it has been reported that mechanical stimulation plays a vital role in bone regeneration and osteointegration. For example, low-magnitude high-frequency vibration (LMHFV) at 35 Hz and 0.3 g was reported to accelerate fracture healing and rescue impaired healing in osteoporotic models by enhancing callus formation⁴⁹. In addition, in rat implant models, LMHFV (0.2–0.5 g; 30–45 Hz) significantly increased bone–implant contact, bone volume ratio, and mineral density⁵⁰. During such mechanical stimulation, the mechanosensitive ion channel Piezo1 is a critical mediator of mechanical signaling in bone cells. It operates on a force-from-lipid principle, being activated upon changes in membrane tension⁵¹. For instance, exercise-induced osteogenic gene expression was found to be increased in Piezo1^fl/fl mice but not in Piezo1-deficient mice⁵², suggesting its key role in bone mechanotransduction and its potential as a target for treating disuse osteoporosis⁵³. Yoda1, the agonist of Piezo1, can stabilize the channel’s open state, lowering its activation threshold⁵⁴. Yoda1 has thus been incorporated into biomaterials to promote osteogenesis and angiogenesis via the Piezo1/RhoA/ROCK1/YAP pathway⁵⁵. Meanwhile, Guo et al. developed a composite scaffold (PGS-P@MGel) that delivers dynamic mechanical stimulation under external fields, activating Piezo1 and enhancing osteogenesis via β-catenin and YAP nuclear translocation⁵⁶. Mechanical stimulation also upregulates specific integrins, which cooperate with Piezo1 to promote the nuclear entry of YAP and Runx2 expression, driving osteogenic differentiation⁵⁷. In this study, we carefully evaluated the influences of MMTS and EMF exposure on the activation of Piezo1 in BMSCs. The findings demonstrated that calcium influx and calcium oscillations were induced to facilitate osteogenesis. However, it has been reported that Piezo1 deletion does not fully abolish the skeletal mechanoresponse, indicating contributions from other sensors including integrins, focal adhesions, or primary cilia⁵⁸. Moreover, beyond mechanical and electromagnetic stimulation, triboelectric stimulation via wearable nanogenerators (30 µA peak current) was shown to promote the osteogenic differentiation of aged BMSCs through Piezo1 activation⁵⁹. Piezo1 may thus function not just as a mechanosensor; however, its potential roles require further investigation to elucidate its full mechanistic implications.

Immunomodulatory biomaterials that trigger controllable immune responses have attracted extensive research attention, as they can enhance tissue regeneration and indirectly prevent bacterial infection⁶⁰. For instance, breathing-mimetic mechanical motions have been reported to suppress viral replication by activating innate immune responses via TRPV4 and RAGE signaling⁶¹. In addition, anisotropic hydrogels were shown to promote integrin β1 expression and polarization of macrophages toward the pro-regenerative M2 phenotype, accelerating wound healing⁶². However, it has remained unclear whether the dynamic biomimetic MMTS with EMF exposure can affect the osteoimmune microenvironment, such as macrophage polarization¹⁹ and NETosis⁶³, to promote bone regeneration and eliminate infections. Another limitation of this study is that human multi-omics and histological are a preliminary exploratory finding, further studies with larger patient cohorts are needed.

Despite the rapid advances in this field, the fabrication of high-performance bioinspired biomaterials for preventing PJI remains a major challenge. On one hand, implant materials are required to possess biocompatibility, mechanical robustness, and corrosion resistance⁶⁴. On the other hand, beyond basic material performance, the mechanical cues play a pivotal role in regulating bone remodeling and host-pathogen interactions⁶⁵. This study integrated advanced antimicrobial and mechanobiology-based strategies into a dynamic biomimetic implant surface, which showed greater advances than the clinical gold standard CEZ. Therefore, this work holds great promise for developing implants that not only resist infection but also promote functional tissue regeneration.

Methods

Ethical statement

All animal procedures were conducted in accordance with the institutional guidelines of Sichuan University and approved by the University’s Animal Ethics Committee (Approval No. 20250224026). All personally identifiable information was anonymized at the point of sample collection. The study protocol involving human specimens was approved by the Ethics Committee of Sichuan University (Approval No. 2022422) and conducted in accordance with the ethical standards outlined in the 2008 Declaration of Helsinki. We confirm that written informed consent was obtained from the patient for the use of their medical data and biological tissues, and that the patient participated in this study without any compensation.

Synthesis process of specimen

Fe₃O₄-Ag JNHDs were synthesized based on an established method⁶⁶. In detail, 500 μL Fe₃O₄ NPs were dispersed in 20 mL trisodium citrate solution (25% w/v in ddH₂O), and stirred at 60 °C. Subsequently, 250 μL AgNO₃ solution (10 mM) was added to the mixture and stirred for 2 h at 60 °C. To harvest JNHDs, the mixture was exposed to UV irradiation for 30 min. For the SiO₂ coating of JNHDs, 250 μg JNHDs were dispersed in 500 μL of ethanol with 25 μL NH₃·H₂O and 75 μL ddH₂O. Then, the suspension was vigorously stirred for 10 min. Subsequently, the suspension was supplemented with 0.2 μL tetraethyl orthosilicate (TEOS) and was stirred for 2 h. Lastly, SiO₂@JNHDs were washed three times with ethanol. SiO₂@JNHDs were dispersed in 500 μL ethanol with 10 μL (3-aminopropyl) triethoxysilane (APTES), then it was stirred for 24 h. Afterward, the amino-functionalized SiO₂@JNHDs were washed three times with ddH₂O. To synthesize the transformable topological surface, the amino-functionalized SiO₂@JNHDs was stirred in 500 μL ddH₂O with 5 mg of long bendable linker [maleimide-poly(ethylene glycol)-succinimidyl carboxymethyl ester] for 5 min, then added with 2 μl of N,N-diisopropylethylamine (DIPEA) and stirred for 24 h. The PEGylated SiO₂@JNHDs were then washed three times with ddH₂O. For the synthesis of a non-transformable topological surface, the amino-functionalized SiO₂@JNHDs were not PEGylated. Similarly, for the SiO₂ coating of Ti substrate, the pure-Ti substrates (purity: 99.9%) were placed in a 24-well plate and immersed in 500 μL hydroxyapatite nanoparticle (nHA) suspension (1 mg/mL). The specimen was thereafter dried for 24 h at room temperature. Subsequently, the specimen was dispersed in 500 μL of ethanol with 25 μL NH₃·H₂O and 75 μL ddH₂O. Then, the plate was shaken for 15 min. Afterward, 0.2 μL TEOS was added to each well, and the plate was shaken for a further 2 h. Lastly, SiO₂-nHA@Ti were washed three times with ethanol. To thiolated the specimen, each specimen was then immersed in a mixture of 475 μL ethanol and 25 μL (3-mercaptopropyl)trimethoxysilane (MPTMS) for 24 h. Afterward, the specimen was gently washed with ddH₂O. The thiolated specimen was treated with 500 μl PEGylated SiO₂@JNHDs (1:10 diluted in ddH₂O) for 16 h to create the transformable conjugation and fabricated MMTS. For UMTS synthesis, JNHDs were directly deposited on the surface of nHA-Ti. To compare with regular clinic treatment, Ti substrate was immersed in 50 μg/ml CEZ (Cat. No. HY-B1892, MedChemExpress, China) for 24 h, after air dry it was then used as positive control. To facilitate high-resolution imaging of specific staining assays, the conventional Ti substrate was substituted with a silicon dioxide round coverslip (Cat. No. WHB-12-CS, WHB Scientific, China). All subsequent fabrication steps for UMTS and MMTS coatings remained identical to those applied on Ti substrates, ensuring consistent surface modification and functionalization across different material bases (Supplementary Fig. 13). Requests for experimental materials should be directed to the corresponding authors.

Characterization of specimens

The morphology and elemental distribution of JNHDs were analyzed using transmission electron microscopy (TEM; JEOL JEM-F200, Japan). Particle size distribution was quantified from TEM images using ImageJ software. Chemical composition and elemental mapping were conducted via EDS. The surface morphology and elemental distribution of Ti, UMTS, and MMTS samples were examined using SEM (ZEISS Sigma 300, Germany). Additionally, SEM was employed to observe the nanostructure of insect wings as well as bacterial adhesion on implant surfaces. Surface topography and roughness parameters were assessed using an atomic force microscope (AFM; Bruker Dimension Icon, Germany). Chemical bond characteristics of Ti, UMTS, and MMTS were evaluated by Fourier-transform infrared spectroscopy (Thermo Fisher Scientific Nicolet iS20, USA). The release profiles of Ag and Fe ions from each specimen were determined using inductively coupled plasma optical emission spectrometry/mass spectrometry (ICP-OES/MS; Agilent 7800, USA). Pre-dissected insect wings were obtained from a commercial supplier, Beijing Jiaying Art Insectarium (Beijing, China).

Magnetic hysteresis loop measurement

Magnetic hysteresis loops of Ti, UMTS, and MMTS were recorded at room temperature using a vibrating sample magnetometer (VSM, LakeShore 7404/8604, USA). The maximum applied magnetic field was set to ±2 T. Hysteresis data were collected and analyzed to characterize the magnetic properties of the samples.

Magnetothermal effect assessment

The magnetothermal performance of Ti, UMTS, MMTS, and JNHDs was evaluated using a thermal imaging camera. Each sample was placed in an alternating EMF coil. Thermal images were captured before and after EMF exposure. The temperature at the sample surface (marked ROI) was recorded, and the temperature change was calculated. All measurements were repeated in triplicate.

Magnetic-field-induced motion of JNHDs

The directional movement of JNHD nanoparticles under an external MF was observed by microscopy. JNHDs were placed in a custom MF. Bright-field images and cross-sectional fluorescence scans (X–Z plane) were captured. For the static control, images were recorded without the applied field. Under the aligned MF, JNHDs’ movement in both horizontal and vertical directions was visualized. Multiple fields of view were captured to assess the consistency of directional alignment.

Human specimens

Bone tissue samples were collected from a patient undergoing one-stage revision surgery for severe chronic periprosthetic hip infection. The diagnosis of periprosthetic joint infection was confirmed by senior orthopedic surgeons, radiologists, and supporting laboratory results. Following removal of all infected prosthetic components, the femoral component was gently rinsed with sterile physiological saline. Both the saline solution and the explanted components were promptly transferred to the laboratory for further processing. To isolate bacteria on the prosthetic surfaces, the saline was centrifuged, and the pellet was cultured in Luria-Bertani (LB) medium (Cat. No. L1010, Solarbio, China). After 48 h of incubation, bacterial cultures were collected for subsequent analysis. Concurrently, the acetabular cup and any adherent bone tissue were subjected to fixation and decalcification for further histological or morphological examination.

Cell isolation and culture

Bone marrow mesenchymal stromal cells (BMSCs) were isolated from the tibiae and femurs of 2-week-old C57 mice via whole bone marrow flushing and adherent culture. Briefly, mice were humanely euthanized by cervical dislocation under deep anesthesia. Intact tibiae and femurs were aseptically dissected, and all adjacent soft tissues were carefully removed. The epiphyses were trimmed, and the marrow cavities were flushed repeatedly using a sterile 25-gauge needle and syringe filled with α-Minimum Essential Medium (α-MEM; Gibco, Cat. No. 12571063) supplemented with 10% fetal bovine serum (FBS; Gibco, Cat. No. 16010159) and 1% penicillin–streptomycin (Beyotime Biotechnology, Cat. No. C0222).

The resulting cell suspension was collected and centrifuged at 500 × g for 5 min. The pellet was resuspended in fresh complete medium and seeded into culture flasks. Cells were maintained at 37 °C in a humidified atmosphere with 5% CO₂. After 24 h, non-adherent cells were removed by medium replacement. The medium was refreshed every 2 days thereafter. Upon reaching 80–90% confluence, adherent cells were passaged using 0.25% trypsin-EDTA (Gibco, Cat. No. 25200056) at a split ratio of 1:3. BMSCs between passages 3 and 5 were used for all experiments.

Cell viability assay

Cell viability was evaluated using the Cell Counting Kit-8 (Cat. No. C0038, CCK-8; Beyotime Biotechnology, China), following the manufacturer’s protocol. BMSCs were seeded onto the specimens and cultured with or without daily EMF exposure. To prevent interference between experimental groups, the culture plates for EMF-treated cells were separated from non-exposed conditions during stimulation. After 3 days of culture, the cells were gently rinsed three times with phosphate-buffered saline (PBS). CCK-8 working solution was then added to each well, and the plates were incubated at 37 °C for 1 h. The absorbance was measured at 450 nm using a microplate reader (PerkinElmer, USA).

Live/Dead staining

Cytotoxicity and bactericidal ability were assessed using a Live/Dead Cell Staining Kit (Cat. No. L3224, Thermo Fisher Scientific, China). BMSCs and adherent bacteria were seeded onto each specimen and cultured with or without daily EMF exposure. After 24 h, the cells were gently washed three times with PBS and incubated with a solution containing Calcein-AM and propidium iodide (PI) for 1 h at 37 °C in the dark. Fluorescent images were acquired using a fluorescence microscope (CLSM; Nikon, Japan). Bacterial viability was evaluated under identical staining conditions.

Cytoskeleton staining

Cellular morphology was evaluated through cytoskeleton staining. Briefly, BMSCs were gently washed three times with PBS and fixed with 4% paraformaldehyde for 10 min at room temperature. The cells were then permeabilized using 0.2% Triton X-100 for 10 min, followed by refixation with 2% paraformaldehyde for an additional 10 min. After washing, samples were incubated with 0.5 µg/ml Phalloidin (Cat. No. Y249127-10μg, Beyotime Biotechnology, China) and DAPI (5 μg/mL, Cat. No. C1002, Beyotime Biotechnology, China) for 1 h at 37 °C in the dark. Imaging was performed using a fluorescent microscope (CLSM; Nikon, Japan). Cell spreading area was quantified with ImageJ software.

DCFH-DA assay

Intracellular ROS levels in BMSCs and adherent bacteria were detected using a fluorometric assay kit (Cat. No. E-BC-K138-F, Elabscience, China). Following treatment, cells were washed three times with PBS and incubated with 10 μM 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) for 30 min at 37 °C in the dark. After incubation, unincorporated dye was removed by washing three times with PBS. Fluorescence images were captured using a fluorescence scanning microscope (CLSM; Nikon, Japan). Quantitative analysis of fluorescence intensity was performed using ImageJ software.

SYTOX Green staining

SYTOX Green staining was performed to evaluate the disruption of bacterial cell membranes and cell walls. Bacterial suspensions were incubated with Ti, UMTS, MMTS specimens with or without EMF exposure and NAC (Cat. No. S0077, Beyotime Biotechnology, China) for 24 h at 37 °C. After incubation, the samples were gently rinsed twice with PBS. Then, the specimens were stained with SYTOX Green staining buffer (Cat. No. C1186-100ml, Beyotime Biotechnology, China) in the dark at room temperature for 60 min. Following three washes with PBS, the fluorescent images were immediately captured using a fluorescent microscope (CLSM; Nikon, Japan). The fluorescence intensity, indicative of membrane and wall damage allowing dye penetration, was quantitatively analyzed using ImageJ software.

Immunofluorescence staining

BMSCs were washed three times with PBS and fixed with 4% paraformaldehyde for 10 min at room temperature. Cells were then permeabilized using 0.2% Triton X-100 for 10 min, followed by refixation with 2% paraformaldehyde for an additional 10 min. Non-specific binding sites were blocked with 5% bovine serum albumin (BSA; Boster, China, Cat. No. AR1006) under gentle shaking for 1 h at room temperature. Subsequently, cells were incubated overnight at 4 °C with the following primary antibodies: Itg-β3 (Cat. No. 66952-1-Ig, 1:1000, mouse monoclonal; Proteintech), Itg-α5 (Cat. No. 27096-1-AP, 1:500, rabbit polyclonal; Proteintech), and Runx2 (Cat. No. 20700-1-AP, 1:500, rabbit polyclonal; Proteintech). After washing with PBS, samples were incubated for 2 h in the dark with CoraLite 488-conjugated goat anti-rabbit IgG (Cat. No. SA00013-2, 1:100; Proteintech, China) or goat anti-mouse IgG (Cat. No. SA00013-1, 1:100; Proteintech, China). Nuclei were stained with DAPI (5 μg/mL, Cat No. C1002, Beyotime Biotechnology, China) for 30 min in the dark. Images were acquired using a fluorescence microscope, and Runx2 co-localization was analyzed using ImageJ software.

Immunofluorescent staining of primary cilia in BMSCs

BMSCs were first transfected with or without si-IFT20, then undergo a serum-starved for at least 48 h to induce primary cilia formation. For immunofluorescence staining, cells were fixed with 4% paraformaldehyde for 10 min at room temperature, washed three times with PBS, and permeabilized using 0.2% Triton X-100 for 10 min. A second fixation step was performed with 2% paraformaldehyde for 10 min. Non-specific binding sites were blocked with 5% BSA for 1 h at room temperature. BMSCs were then incubated overnight at 4 °C with a primary antibody against acetylated tubulin (1:200l; Cat No. 66200-1-Ig, Proteintech, China). After PBS washes, samples were incubated for 2 h in the dark with a CoraLite 488-conjugated goat anti-mouse secondary antibody (1:100; Cat No. SA00013-1, Proteintech, China). Nuclei were stained with DAPI (5 μg/mL; Cat No. C1002, Beyotime Biotechnology, China) for 30 min protected from light. Images were acquired using a fluorescence microscope.

RNA isolation and RT-qPCR

Total RNA was extracted from BMSCs using the Eastep® Super Total RNA Extraction Kit (Cat. No. LS1040, Promega Biotech, China) following the manufacturer’s instructions. RNA concentration and purity were determined using a NanoDrop™ spectrophotometer (Cat. No. 840-317400, Thermo Fisher Scientific, China). Complementary DNA (cDNA) was synthesized from 1 µg of total RNA using the First Strand cDNA Synthesis Kit (No. D7190S, Beyotime Biotechnology, China, Cat.). Quantitative real-time PCR was performed with the SYBR Green One-Step RT-qPCR Kit (Cat. No. D7268S, Beyotime Biotechnology, China) according to the provided protocol. All reactions were run in triplicate. The relative expression of target genes was calculated using the 2^−ΔΔ^CT method, with GAPDH as the endogenous control. Primer sequences are listed in Supplementary Table 1. The primer sequences used for amplification were synthesized by Tsingke Biotech, Beijing, China.

Primary cilia knockdown

IFT20 mRNA in BMSCs was knocked down using IFT20 siRNA (Forward Primer: GAAUAUGAAGCUUUGUGUA, Reverse Primer: UACACAAAGCUUCAUAUUC) with Lipofectamine 2000 (Cat. No. 11668019, Invitrogen, USA). Cells were seeded in 6-well plates at 2.5 × 10⁵ cells/well in complete medium without antibiotics and transfected at 60–70% confluence. For each well, 50 nM siRNA was diluted in 250 µL Opti-MEM (Cat. No. 11058021, Gibco, USA), mixed with 5 µL Lipofectamine 2000 in 250 µL Opti-MEM, and incubated for 20 min. The complex was added dropwise to wells containing fresh antibiotic-free medium. After 6 h, the medium was replaced with standard complete medium. Cells were harvested 48 h after transfection for subsequent research.

Transcriptome sequencing and analysis

Total RNA was extracted using TRIzol Reagent (Cat. No. R0016, Beyotime Biotechnology, China). RNA quality was assessed using an Agilent 5300 Bioanalyzer and a NanoDrop ND-2000 spectrophotometer. Only high-quality RNA samples (OD260/280 = 1.8-2.2, OD260/230 ≥ 2.0, RQN ≥ 6.5, 28S:18S ≥ 1.0, total RNA > 1 μg) were used for subsequent library preparation. Sequencing libraries were constructed using either the Illumina® Stranded mRNA Prep Ligation Kit (1 μg input) or the SMART-Seq V4 Ultra Low Input RNA Kit (10 ng input), strictly following the manufacturer’s protocols. Library quantification was performed using a Qubit 4.0 fluorometer, and paired-end sequencing (150 bp) was carried out on an Illumina NovaSeq X Plus platform. Raw sequencing reads were quality-filtered and adapter-trimmed using fastp. Clean reads were aligned to the reference genome with HISAT2. Transcript assembly and expression quantification were conducted using StringTie and RSEM, respectively. Differential gene expression analysis was performed with DESeq2 and DEGseq, with significantly differentially expressed genes (DEGs) defined as those with |log₂FC| ≥ 1 and FDR ≤ 0.05 (DESeq2) or FDR ≤ 0.001 (DEGseq). For bacterial metabolomic analysis, microbial metabolism was rapidly quenched, and metabolites were extracted accordingly. Metabolic profiling was performed using LC-MS and GC-MS platforms. Raw data were processed to align peaks, identify metabolites by matching against standard databases, and were subjected to multivariate statistical analysis for biological interpretation. Functional enrichment analyses of DEGs, including Gene Ontology and KEGG pathways, were conducted using the DAVID bioinformatics database.

Western blot

Total protein was extracted from BMSCs using ice-cold RIPA lysis buffer (Cat No.PC101, Epizyme Biotech, Shanghai, China) supplemented with protease and phosphatase inhibitor cocktails (Cat No.GRF103, Epizyme Biotech, Shanghai, China). Lysates were centrifuged at 3000 × g for 10 min at 4 °C to remove cellular debris. Protein concentration was determined using a BCA protein assay kit (Cat No.ZJ102, Epizyme Biotech, Shanghai, China). Equal amounts of protein (25 µg per lane) were separated by SDS-PAGE on 10% or 15% acrylamide-bisacrylamide gels at 120 V for 150 min and then transferred to 0.22 µm PVDF membranes at 400 mA for 30 min using Ice-Bath Free Western Superquick Transfer Buffer (Cat No.PS201S, Epizyme Biotech, Shanghai, China).

Membranes were blocked with 5% BSA in TBST for 2 h at room temperature with gentle shaking, followed by incubation overnight at 4 °C with the following primary antibodies: Itg-β3 (1:2000, mouse, Cat No. 66952-1-Ig, Proteintech, China), Itg-α5 (1:2000, rabbit, Cat No. 27096-1-AP, Proteintech, China), Runx2 (1:2000, rabbit, Cat No. 20700-1-AP, Proteintech, China), Bmp2 (1:1000, mouse, Cat No.66383-1-Ig, Proteintech, China), OPN (1:1000, rabbit, Cat No.22952-1-AP, Proteintech, China), Piezo1 (1:1000, rabbit, Cat No.82625-4-RR, Proteintech, China), and GAPDH (1:5000, rabbit, Cat No.10494-1-AP, Proteintech, China). After washing, membranes were incubated with corresponding HRP-conjugated secondary antibodies (1:10,000 in TBST) for 2 h at room temperature. Protein bands were visualized using a chemiluminescence detection kit (Cat No.SQ202, Epizyme Biotech, Shanghai, China) and imaged with a CCD camera. Signal intensities were quantified using ImageJ software.

Fabrication and calibration of EMF exposure system

EMF exposure system (Supplementary Fig. 14) was custom-designed and consisted of four main components: a one-dimensional Helmholtz coil for in vivo animal experiments (custom model based on Cat. No. DXHC15-10, Dexin Mag, Xiamen, China), a solenoid for in vitro cell experiments (custom design, Dexin Mag, Xiamen, China), a power amplifier (Cat. No. DXACP-0332, Dexin Mag, Xiamen, China), and a multi-waveform generator (Cat. No. 8300S, Dexin Mag, Xiamen, China). The magnetic field intensity was emulated (Supplementary Fig. 15) and calibrated before each experiment using a Gaussmeter (Cat. No. DX-102F, Dexin Mag, Xiamen, China).

Intracellular Ca²⁺ staining

BMSCs were seeded on each specimen for 24 h in a Petri dish, and then loaded with 2 µM Fluo-4 AM (Cat No. S1060, Beyotime Biotechnology, China) at 37 °C for 2 h. Subsequently, BMSCs were gently washed three times with PBS and incubated in Hanks Balanced Salt Solution (with Ca²⁺, Cat No. C0219, Beyotime Biotechnology, China). BMSCs were then undergo real time video recording with a fluorescent microscope (CLSM; Nikon, Japan). Piezo1 activator Yoda1 (Cat No. HY-18723, MCE, China) and Yoda1 antagonist Dooku1 (Cat No. HY-126010, MCE, China) were used to pharmacologically modulate the activity of Piezo1. Yoda1 were used at 10 µM for subsequent experiments, as dose-response screening showed that while 20 µM yielded slightly higher signals, the marginal gain did not justify the increased risk of off-target cytotoxicity at the higher concentration. The quantitative analysis was conducted with ImageJ software.

ALP staining and Alizarin red staining

Alkaline phosphatase (ALP) staining and Alizarin red staining were performed to evaluate early osteogenic differentiation and late-stage matrix mineralization, respectively. For ALP staining, BMSCs were induced under osteogenic conditions for 14 days, with or without pharmacological treatment. Cells were washed three times with PBS and stained using the BCIP/NBT Alkaline Phosphatase Color Development Kit (Cat No.C3206, Beyotime Biotechnology, China) according to the manufacturer’s instructions. For Alizarin red staining, BMSCs were osteogenically induced for 28 days, with or without pharmacological treatment, then fixed with ice-cold 99% ethanol at −20 °C for 1 h. After fixation, cells were gently washed three times with tap water and incubated with Alizarin red solution (Cat No.C0138, Beyotime Biotechnology, China) for 30 min at room temperature. Unbound dye was removed by washing three times with tap water. Stained samples were imaged using a confocal laser scanning microscope (CLSM; Nikon, Japan).

Antibacterial assay

Antibacterial activity was evaluated against both Gram-positive Staphylococcus aureus and Gram-negative Escherichia coli. Bacterial suspensions collected from specimen surfaces or culture suspension (with or without EMF exposure) were inoculated into LB culture medium (Cat No. L1010, Solarbio, China) and incubated at 37 °C for 24 h with shaking. Subsequently, the suspensions were appropriately diluted, spread onto LB agar plates (Cat No. L1015, Solarbio, China), and incubated statically at 37 °C for an additional 24 h. The number of bacterial colonies and macroscopic morphology of the agar plates were documented to assess antibacterial activity.

Crystal violet staining

Bacterial suspensions collected from specimen surfaces with or without EMF exposure were cultured in LB medium (Cat No. L1010, Solarbio, China) at 37 °C for 24 h under shaking conditions. Subsequently, 0.5 mL of each suspension was mixed with an equal volume of fresh LB medium and transferred to a 24-well plate for static incubation at 37 °C for 48 h to allow biofilm formation. After careful removal of the supernatant, the adherent biofilms were stained with crystal violet solution (Cat No. C0121-100ml, Beyotime Biotechnology, China) for 10 min at room temperature. The wells were gently rinsed with ddH₂O to remove unbound dye. Images of the stained biofilms were acquired using an optical microscope. For quantitative analysis, the bound dye was dissolved in 95% ethanol, and the absorbance was measured at 570 nm to evaluate anti-biofilm activity.

Animal experiment

Male Sprague–Dawley rats (4 weeks old) were obtained from Dashuo Laboratory Animal Co., Ltd. (China). A tibial osteomyelitis model was established in these animals. A total of 42 male Sprague–Dawley rats (4 weeks old) were used in this study. Rats were randomly assigned to seven experimental groups (n = 6 per group) using a computer random number generator: Ti, UMTS, MMTS, Ti+EMF, UMTS + EMF, MMTS + EMF, and CEZ. All rats completed the experiment without exclusion or replacement.

Following anesthesia, the surgical site was shaved and disinfected. An incision was made along the knee joint to expose the proximal tibial metaphysis. A bone channel was drilled from the tibial plateau into the marrow cavity using an electric drill. Pre-inoculated specimens were then implanted into the tibia. Muscle, fascia, and skin were sutured layer by layer. Postoperative health status and activity of the rats were monitored daily. Animals in the EMF exposure groups were subjected to daily EMF (50 Hz, 5 mT) treatment for 30 min. The CEZ control group did not receive EMF exposure. The clinical dose for adults is 0.5–1 g, i.v., t.i.d, accordingly, an intraperitoneal dose of 50 mg/kg, q.d., was administered to the SD rats of the CEZ group in this study.

Micro-CT evaluation

Bone tissue samples from the tibia were analyzed using a high-resolution μ-CT system (NMC-100, PINGSENG Healthcare, Shanghai, China). Scanning parameters were set at 80 kV and 0.08 mA. Consistent regions of interest (ROIs) were selected and scanned under identical conditions for all samples. Coronal CT images and three-dimensional reconstructions were generated, and quantitative analyses were performed using Imaris software (version 10.2, Oxford Instruments, USA).

Histological analysis

Tissue specimens were decalcified in 10% ethylenediaminetetraacetic acid (EDTA) for 30 days, dehydrated in a graded ethanol series, and embedded in paraffin. Sections of 10 μm thickness were prepared using a rotary microtome. Consecutive sections were stained with H&E for general histoarchitecture evaluation, Masson’s trichrome for collagen deposition, Gram staining, and Giemsa staining for bacterial visualization. For immunofluorescence analysis, sections were subjected to antigen retrieval, blocked with 5% bovine serum albumin, and incubated overnight at 4 °C with primary antibodies against COL2A1, BMP2, VEGFA, Piezo1, and HIF1A, following established protocols⁶⁷. After washing, appropriate fluorophore-conjugated secondary antibodies were applied, and nuclei were counterstained with DAPI. All stained sections were imaged under a fluorescence microscope under consistent exposure settings.

Statistics and reproducibility

All experiments were performed with at least three independent replications. Error bars represent the standard deviation of the mean. Statistical significance was determined using one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test. Differences were considered statistically significant at p < 0.05. Statistical analyses were performed with the GraphPad Prism software V.8.0.1 (El Camino Real, USA).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

Supplementary Information^{(16MB, docx)}

41467_2026_78462_MOESM2_ESM.docx^{(13.6KB, docx)}

Description of Additional Supplementary Files

Supplementary movie 1^{(6.7MB, mp4)}

Supplementary movie 2^{(14MB, mp4)}

Reporting Summary^{(99.4KB, pdf)}

Transparent Peer Review file^{(9.6MB, pdf)}

Source data

Source data^{(6.8MB, xlsx)}

Acknowledgements

The authors thank the public instrument platform of West China School of Medicine of Sichuan University for the technical support. The authors thank the laboratory team of the Research Institute of Orthopedics of West China Hospital of Sichuan University for providing experimental platforms and solving technical challenges.

Author contributions

Conceptualization: H.D. and Z.Z.; methodology: L.Z., H.D., and Y.C.; validation: L.Z., Y.C., and X.H.; formal analyses: Y.C., L.Z., and H.S.; investigation: L.Z., Y.C., and X.H.; resources: D.W., H.D., H.S., X.Z., and Z.Z.; data curation: L.Z., H.D., and Z.Z.; original draft preparation: Y.C.; review and editing: all authors; visualization: J.C., H.S., and Y.C.; supervision: L.Z. and Z.Z.; project administration: L.Z. and Z.Z.; funding acquisition: L.Z., Y.C., X.Z., and Z.Z. All authors have read and agreed to the published version of the manuscript.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

This work was supported by the National Key Research and Development Program of China (2022YFC2503100, 2022YFC2503104, and 2022YFC2503103, Z.Z.), Shenzhen Medical Research Fund (A2503015, L.Z.), the 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University (ZYGD23033, Z.Z.), the National Natural Science Foundation of China (82402473, Y.C., and 82302369, L.Z.), the Science and Technology Development Fund, Macau SAR (0032/2023/ITP1, L.Z.), the Research Grant of University of Macau (MYRG-2024-00162-FHS, MYRG-GRG2025-00095-FHS, X.Z.), Guangdong Basic and Applied Basic Research Foundation (2025A1515012894, X.Z.), the Natural Science Foundation of Sichuan (2024YFFK0208, Y.C.), the Postdoctor Research Fund of West China Hospital, Sichuan University (2024HXBH002, Y.C.), the China Postdoctoral Science Foundation (2024M762246, Y.C.), and Sichuan University Postdoctoral Interdisciplinary Innovation Fund (Y.C.).

Data availability

The authors declare that all data supporting the findings of this study are available within the article and the Supplementary Information. Supplementary Table 1 was provided in the Supplementary Information file. The metabolomics data generated in this study have been deposited in the MetaboLights database under accession MTBLS15301. The RNA-sequencing data generated in this study have been deposited in the Sequence Read Archive (SRA) under accession PRJNA1529328 [https://www.ncbi.nlm.nih.gov/bioproject/1529328]. Any additional requests for information can be directed to, and will be fulfilled by, the corresponding authors. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Liqiang Zhou, Email: liqiangzhou@um.edu.mo.

Zongke Zhou, Email: zhouzongke@scu.edu.cn.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-78462-4.

References

1.Kapadia, B. H. et al. Periprosthetic joint infection. Lancet387, 386–394 10.1016/S0140-6736(14)61798-0 (2016). [DOI] [PubMed] [Google Scholar]
2.Kurtz, S. M. et al. Prosthetic joint infection risk after TKA in the Medicare population. Clin. Orthop. Relat. Res.468, 52–56 10.1007/s11999-009-1013-5 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Kapadia, B. H. et al. The economic impact of periprosthetic infections following total knee arthroplasty at a specialized tertiary-care center. J Arthroplasty29, 929–932 10.1016/j.arth.2013.09.017 (2014). [DOI] [PubMed] [Google Scholar]
4.Tong, S. Y. C., Fowler, V. G. Jr., Skalla, L. & Holland, T. L. Management of Staphylococcus aureus Bacteremia: a review. JAMA334, 798–808 10.1001/jama.2025.4288 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Burdet, C. et al. Cloxacillin versus cefazolin for meticillin-susceptible Staphylococcus aureus bacteraemia (CloCeBa): a prospective, open-label, multicentre, non-inferiority, randomised clinical trial. Lancet406, 2349–2359 10.1016/s0140-6736(25)01624-1 (2025). [DOI] [PubMed] [Google Scholar]
6.Bains, S. S. et al. Addition of vancomycin to cefazolin is often unnecessary for preoperative antibiotic prophylaxis during total joint arthroplasties. Arthroplasty6, 20 10.1186/s42836-023-00222-2 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Flemming, H. C. & Wingender, J. The biofilm matrix. Nat. Rev. Microbiol.8, 623–633 10.1038/nrmicro2415 (2010). [DOI] [PubMed] [Google Scholar]
8.Hu, Y., Shao, J., Dong, H., Yang, D. & Dong, X. Bacterial programmed cell death induced by nanotherapeutic strategies. ACS Mater. Lett.6, 4209–4229 10.1021/acsmaterialslett.4c01165 (2024). [DOI] [Google Scholar]
9.Patel, R. Periprosthetic Joint Infection. N. Engl. J. Med.388, 251–262 10.1056/NEJMra2203477 (2023). [DOI] [PubMed] [Google Scholar]
10.Xi, Y. et al. Biomimetic bioactive multifunctional poly(citrate-siloxane)-based nanofibrous scaffolds enable efficient multidrug-resistant bacterial treatment/non-invasive tracking in vitro/in vivo. Chem. Eng. J.383, 123078 10.1016/j.cej.2019.123078 (2020). [DOI] [Google Scholar]
11.Wu, S. et al. Long-lasting renewable antibacterial porous polymeric coatings enable titanium biomaterials to prevent and treat peri-implant infection. Nat Commun12, 3303 10.1038/s41467-021-23069-0 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Xu, Y. et al. Fabrication and functional regulation of biomimetic interfaces and their antifouling and antibacterial applications: a review. Small20, e2308091 10.1002/smll.202308091 (2024). [DOI] [PubMed] [Google Scholar]
13.Nowotnick, A. G. et al. Antimicrobial biomaterials based on physical and physicochemical action. Adv. Healthc. Mater.13, e2402001 10.1002/adhm.202402001 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Zhang, C., McAdams, D. A. 2nd & Grunlan, J. C. Nano/Micro-manufacturing of bioinspired materials: a review of methods to mimic natural structures. Adv. Mater.28, 6292–6321 10.1002/adma.201505555 (2016). [DOI] [PubMed] [Google Scholar]
15.Mouterde, T. et al. Antifogging abilities of model nanotextures. Nat. Mater.16, 658–663 10.1038/nmat4868 (2017). [DOI] [PubMed] [Google Scholar]
16.Baulin, V. A., Linklater, D. P., Juodkazis, S. & Ivanova, E. P. Exploring broad-spectrum antimicrobial nanotopographies: implications for bactericidal, antifungal, and virucidal surface design. ACS Nano19, 12606–12625 10.1021/acsnano.4c15671 (2025). [DOI] [PubMed] [Google Scholar]
17.Guan, M. et al. Long-lasting bactericidal activity through selective physical puncture and controlled ions release of polydopamine and silver nanoparticles-loaded TiO₂ nanorods in vitro and in vivo. Int. J. Nanomed.14, 2903–2914 10.2147/ijn.S202625 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Gao, C. et al. Poly(dopamine) and Ag nanoparticle-loaded TiO₂ nanotubes with optimized antibacterial and ROS-scavenging bioactivities. Nanomedicine14, 803–818 10.2217/nnm-2018-0131 (2019). [DOI] [PubMed] [Google Scholar]
19.Chen, Y. et al. Improved immunoregulation of ultra-low-dose silver nanoparticle-loaded TiO₂ nanotubes via M2 macrophage polarization by regulating GLUT1 and autophagy. Int. J. Nanomed.15, 2011–2026 10.2147/ijn.S242919 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Wu, Y., Liu, P., Mehrjou, B. & Chu, P. K. Interdisciplinary-inspired smart antibacterial materials and their biomedical applications. Adv. Mater.36, e2305940 10.1002/adma.202305940 (2024). [DOI] [PubMed] [Google Scholar]
21.Sun, M. et al. Magnetic microswarm and fluoroscopy-guided platform for biofilm eradication in biliary stents. Adv. Mater.34, 2201888 10.1002/adma.202201888 (2022). [DOI] [PubMed] [Google Scholar]
22.Johnson-Love, O., Salmeron-Sanchez, M., Reid, S., Childs, P. G. & Dalby, M. J. Vibration-based cell engineering. Nat. Rev. Bioeng.3, 408–429 10.1038/s44222-025-00273-x (2025). [DOI] [Google Scholar]
23.Ozcivici, E. et al. Mechanical signals as anabolic agents in bone. Nat. Rev. Rheumatol.6, 50–59 10.1038/nrrheum.2009.239 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Kalelkar, P. P., Riddick, M. & Garcia, A. J. Biomaterial-based delivery of antimicrobial therapies for the treatment of bacterial infections. Nat. Rev. Mater.7, 39–54 10.1038/s41578-021-00362-4 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Iwamoto, H. The tymbal muscle of cicada has flight muscle-type sarcomeric architecture and protein expression. Zool. Lett.3, 15 10.1186/s40851-017-0077-4 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Liu, Y. et al. Biomimetic electroactive materials and devices for regenerative engineering. Nat. Rev. Electr. Eng.2, 188–204 10.1038/s44287-025-00145-x (2025). [DOI] [Google Scholar]
27.Jolissaint, J. E., Mallett, K. E., Thomson, A. L., Carli, A. V. & Austin, M. S. Cefazolin and the R1 side chain: Why patients with a cephalosporin allergy can be safely given cefazolin while undergoing joint arthroplasty. J. Bone Jt. Surg. Am. 10.2106/jbjs.25.01349 (2026). [DOI] [PubMed]
28.Dance, A. Five ways science is tackling the antibiotic resistance crisis. Nature632, 494–496 10.1038/d41586-024-02601-4 (2024). [DOI] [PubMed] [Google Scholar]
29.Zhang, J., Dong, H., Liu, B. & Yang, D. Biomimetic materials for antibacterial applications. Small21, e2408543 10.1002/smll.202408543 (2025). [DOI] [PubMed] [Google Scholar]
30.Super, M. et al. Biomaterial vaccines capturing pathogen-associated molecular patterns protect against bacterial infections and septic shock. Nat. Biomed. Eng.6, 8–18 10.1038/s41551-021-00756-3 (2022). [DOI] [PubMed] [Google Scholar]
31.Carter, B. J. et al. Cefazolin is associated with a lower periprosthetic joint infection risk than alternative antibiotic regimens following total hip arthroplasty. J. Arthroplasty40, 2671–2678 10.1016/j.arth.2025.04.079 (2025). [DOI] [PubMed] [Google Scholar]
32.Bergstein, V. E., Taylor, W. L. T., Weinblatt, A. I. & Long, W. J. Hidden costs of first choice alternatives: a financial model of thromboprophylaxis and prosthetic joint infection prophylaxis in total knee arthroplasty. J. Orthop.63, 87–92 10.1016/j.jor.2024.10.053 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Rumalla, K. C. et al. Noncefazolin antibiotic prophylaxis is associated with increased periprosthetic joint infection risk following total knee arthroplasty. J. Arthroplasty41, 601–607 10.1016/j.arth.2025.06.057 (2026). [DOI] [PubMed] [Google Scholar]
34.Arciola, C. R., Campoccia, D. & Montanaro, L. Implant infections: adhesion, biofilm formation and immune evasion. Nat. Rev. Microbiol.16, 397–409 10.1038/s41579-018-0019-y (2018). [DOI] [PubMed] [Google Scholar]
35.Lin, Y. et al. Systemic T-cell and local macrophage interactions mediate granule size-dependent biological hydroxyapatite foreign body reaction. BMEMat3, e12133 10.1002/bmm2.12133 (2025). [DOI] [Google Scholar]
36.Zhang, Y. et al. A smart coating with integrated physical antimicrobial and strain-mapping functionalities for orthopedic implants. Sci. Adv.9, eadg7397 10.1126/sciadv.adg7397 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Genova, L. A. et al. Mechanical stress compromises multicomponent efflux complexes in bacteria. Proc. Natl. Acad. Sci. USA116, 25462–25467 10.1073/pnas.1909562116 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Bian, S. et al. Antibacterial hydrogel: the sniper of chronic wounds. BMEMat3, e12135 10.1002/bmm2.12135 (2025). [DOI] [Google Scholar]
39.Vishwakarma, A., Dang, F., Ferrell, A., Barton, H. A. & Joy, A. Peptidomimetic polyurethanes inhibit bacterial biofilm formation and disrupt surface established biofilms. J. Am. Chem. Soc.143, 9440–9449 10.1021/jacs.1c02324 (2021). [DOI] [PubMed] [Google Scholar]
40.Chopra, D., Guo, T., Jayasree, A., Gulati, K. & Ivanovski, S. Bioinspired, bioactive, and bactericidal: anodized nanotextured dental implants. Adv. Funct. Mater.34, 2314031 10.1002/adfm.202314031 (2024). [DOI] [Google Scholar]
41.Ma, H., Yang, X., Tang, X., Cao, X. & Dai, R. Self-assembled Co-doped β-Bi2O3 flower-like structure for enhanced photocatalytic antibacterial effect under visible light. Appl. Surf. Sci.572, 151348 10.1016/j.apsusc.2021.151348 (2022). [DOI] [Google Scholar]
42.Wang, L. et al. Biomimetic targeting nanoadjuvants for sonodynamic and chronological multi-immunotherapy against holistic biofilm-related infections. Adv. Mater.36, e2308110 10.1002/adma.202308110 (2024). [DOI] [PubMed] [Google Scholar]
43.Ou, Z. et al. Biodegradable Janus sonozyme with continuous reactive oxygen species regulation for treating infected critical-sized bone defects. Nat. Commun.15, 10525 10.1038/s41467-024-54894-8 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Fan, Y. et al. Mussel-mimetic polysaccharide-based injectable hydrogels for biomedical applications. BMEMat e12089 10.1002/bmm2.12089 (2024). [DOI]
45.Tang, W., Fischer, N. G., Kong, X., Sang, T. & Ye, Z. Hybrid coatings on dental and orthopedic titanium implants: current advances and challenges. BMEMat e12105 10.1002/bmm2.12105 (2024). [DOI]
46.Lee, S. et al. Magnetic control and real-time monitoring of stem cell differentiation by the ligand nanoassembly. Small17, e2102892 10.1002/smll.202102892 (2021). [DOI] [PubMed] [Google Scholar]
47.Savli, M. et al. Magnetomechanical detachment of bacterial biofilms using anisotropic magnetic iron oxide nanochains. ACS Appl. Bio. Mater.8, 8059–8071 10.1021/acsabm.5c01029 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Tang, Y. et al. Photoelectrons sequentially regulate antibacterial activity and osseointegration of titanium implants. Adv. Mater.36, e2307756 10.1002/adma.202307756 (2024). [DOI] [PubMed] [Google Scholar]
49.Shi, H. F., Cheung, W. H., Qin, L., Leung, A. H. & Leung, K. S. Low-magnitude high-frequency vibration treatment augments fracture healing in ovariectomy-induced osteoporotic bone. Bone46, 1299–1305 10.1016/j.bone.2009.11.028 (2010). [DOI] [PubMed] [Google Scholar]
50.Jing, D. et al. Low-1 level mechanical vibration improves bone microstructure, tissue mechanical properties and porous titanium implant osseointegration by promoting anabolic response in type 1 diabetic rabbits. Bone106, 11–21 10.1016/j.bone.2017.10.001 (2018). [DOI] [PubMed] [Google Scholar]
51.Unnithan, A. R. et al. Remotely actuated magnetic nanocarpets for bone tissue engineering: non-invasive modulation of mechanosensitive ion channels for enhanced osteogenesis. Adv. Funct. Mater.32, 2201311 10.1002/adfm.202201311 (2022). [DOI] [Google Scholar]
52.Sun, W. et al. The mechanosensitive Piezo1 channel is required for bone formation. eLife8, e47454 10.7554/eLife.47454 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Hu, Y. et al. The critical role of the Piezo1/β-catenin/ATF4 axis on the stemness of Gli1+ BMSCs during simulated microgravity-induced bone loss. Adv. Sci.10, 2303375 10.1002/advs.202303375 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Guan, H. et al. Magnetic aggregation-induced bone-targeting nanocarrier with effects of Piezo1 activation and osteogenic-angiogenic coupling for osteoporotic bone repair. Adv. Mater.36, e2312081 10.1002/adma.202312081 (2024). [DOI] [PubMed] [Google Scholar]
55.Yang, J. et al. Accelerated bone reconstruction by the Yoda1 bilayer membrane via promotion of osteointegration and angiogenesis. Adv. Healthc. Mater.12, e2203105 10.1002/adhm.202203105 (2023). [DOI] [PubMed] [Google Scholar]
56.Guo, X. et al. Magnetically guided mechanoactive mineralization scaffolds for enhanced bone regeneration. Adv. Funct. Mater.35, 2503903 10.1002/adfm.202503903 (2025). [DOI] [Google Scholar]
57.Gao, Q., Wang, M., Hou, X., Li, M. & Li, L. Substrate stiffness modulates osteogenic and adipogenic differentiation of osteosarcoma through PIEZO1 mediated signaling pathway. Cell. Signal.127, 111601 10.1016/j.cellsig.2025.111601 (2025). [DOI] [PubMed] [Google Scholar]
58.Li, X. et al. Stimulation of Piezo1 by mechanical signals promotes bone anabolism. Elife8 10.7554/eLife.49631 (2019). [DOI] [PMC free article] [PubMed]
59.Wang, B. et al. Bone repairment via mechanosensation of Piezo1 using wearable pulsed triboelectric nanogenerator. Small18, 2201056 10.1002/smll.202201056 (2022). [DOI] [PubMed] [Google Scholar]
60.Chu, G. et al. Mechanochemically reprogrammed interface orchestrates neutrophil bactericidal activity and apoptosis for preventing implant-associated infection. Adv. Mater.36, e2311855 10.1002/adma.202311855 (2024). [DOI] [PubMed] [Google Scholar]
61.Bai, H. et al. Mechanical control of innate immune responses against viral infection revealed in a human lung alveolus chip. Nat. Commun.13, 1928 10.1038/s41467-022-29562-4 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Tian, M. et al. Anisotropic nanocomposite hydrogel dressings with thermo-responsive contraction for actively stimulated tissue regeneration. Chem. Eng. J.514, 163202 10.1016/j.cej.2025.163202 (2025). [DOI] [Google Scholar]
63.Guo, G. et al. Neutrophil function conversion driven by immune switchpoint regulator against diabetes-related biofilm infections. Adv. Mater.36, e2310320 10.1002/adma.202310320 (2024). [DOI] [PubMed] [Google Scholar]
64.Zhu, Z. et al. Tissue/Organ adaptable bioelectronic silk-based implants. Adv. Mater.36, e2405892 10.1002/adma.202405892 (2024). [DOI] [PubMed] [Google Scholar]
65.Zhang, H. et al. A bioinspired virus-like mechano-bactericidal nanomotor for ocular multidrug-resistant bacterial infection treatment. Adv. Mater.37, e2408221 10.1002/adma.202408221 (2025). [DOI] [PubMed] [Google Scholar]
66.Zeng, J. et al. Direct synthesis of water-dispersible magnetic/plasmonic heteronanostructures for multimodality biomedical imaging. Nano Lett.19, 3011–3018 10.1021/acs.nanolett.9b00171 (2019). [DOI] [PubMed] [Google Scholar]
67.Chen, Y. et al. Multifunctionally disordered TiO nanoneedles prevent periprosthetic infection and enhance osteointegration by killing bacteria and modulating the osteoimmune microenvironment. Theranostics14, 6016–6035 10.7150/thno.98219 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information^{(16MB, docx)}

41467_2026_78462_MOESM2_ESM.docx^{(13.6KB, docx)}

Description of Additional Supplementary Files

Supplementary movie 1^{(6.7MB, mp4)}

Supplementary movie 2^{(14MB, mp4)}

Reporting Summary^{(99.4KB, pdf)}

Transparent Peer Review file^{(9.6MB, pdf)}

Source data^{(6.8MB, xlsx)}

Data Availability Statement

[CR1] 1.Kapadia, B. H. et al. Periprosthetic joint infection. Lancet387, 386–394 10.1016/S0140-6736(14)61798-0 (2016). [DOI] [PubMed] [Google Scholar]

[CR2] 2.Kurtz, S. M. et al. Prosthetic joint infection risk after TKA in the Medicare population. Clin. Orthop. Relat. Res.468, 52–56 10.1007/s11999-009-1013-5 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR3] 3.Kapadia, B. H. et al. The economic impact of periprosthetic infections following total knee arthroplasty at a specialized tertiary-care center. J Arthroplasty29, 929–932 10.1016/j.arth.2013.09.017 (2014). [DOI] [PubMed] [Google Scholar]

[CR4] 4.Tong, S. Y. C., Fowler, V. G. Jr., Skalla, L. & Holland, T. L. Management of Staphylococcus aureus Bacteremia: a review. JAMA334, 798–808 10.1001/jama.2025.4288 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Burdet, C. et al. Cloxacillin versus cefazolin for meticillin-susceptible Staphylococcus aureus bacteraemia (CloCeBa): a prospective, open-label, multicentre, non-inferiority, randomised clinical trial. Lancet406, 2349–2359 10.1016/s0140-6736(25)01624-1 (2025). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Bains, S. S. et al. Addition of vancomycin to cefazolin is often unnecessary for preoperative antibiotic prophylaxis during total joint arthroplasties. Arthroplasty6, 20 10.1186/s42836-023-00222-2 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR7] 7.Flemming, H. C. & Wingender, J. The biofilm matrix. Nat. Rev. Microbiol.8, 623–633 10.1038/nrmicro2415 (2010). [DOI] [PubMed] [Google Scholar]

[CR8] 8.Hu, Y., Shao, J., Dong, H., Yang, D. & Dong, X. Bacterial programmed cell death induced by nanotherapeutic strategies. ACS Mater. Lett.6, 4209–4229 10.1021/acsmaterialslett.4c01165 (2024). [DOI] [Google Scholar]

[CR9] 9.Patel, R. Periprosthetic Joint Infection. N. Engl. J. Med.388, 251–262 10.1056/NEJMra2203477 (2023). [DOI] [PubMed] [Google Scholar]

[CR10] 10.Xi, Y. et al. Biomimetic bioactive multifunctional poly(citrate-siloxane)-based nanofibrous scaffolds enable efficient multidrug-resistant bacterial treatment/non-invasive tracking in vitro/in vivo. Chem. Eng. J.383, 123078 10.1016/j.cej.2019.123078 (2020). [DOI] [Google Scholar]

[CR11] 11.Wu, S. et al. Long-lasting renewable antibacterial porous polymeric coatings enable titanium biomaterials to prevent and treat peri-implant infection. Nat Commun12, 3303 10.1038/s41467-021-23069-0 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Xu, Y. et al. Fabrication and functional regulation of biomimetic interfaces and their antifouling and antibacterial applications: a review. Small20, e2308091 10.1002/smll.202308091 (2024). [DOI] [PubMed] [Google Scholar]

[CR13] 13.Nowotnick, A. G. et al. Antimicrobial biomaterials based on physical and physicochemical action. Adv. Healthc. Mater.13, e2402001 10.1002/adhm.202402001 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Zhang, C., McAdams, D. A. 2nd & Grunlan, J. C. Nano/Micro-manufacturing of bioinspired materials: a review of methods to mimic natural structures. Adv. Mater.28, 6292–6321 10.1002/adma.201505555 (2016). [DOI] [PubMed] [Google Scholar]

[CR15] 15.Mouterde, T. et al. Antifogging abilities of model nanotextures. Nat. Mater.16, 658–663 10.1038/nmat4868 (2017). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Baulin, V. A., Linklater, D. P., Juodkazis, S. & Ivanova, E. P. Exploring broad-spectrum antimicrobial nanotopographies: implications for bactericidal, antifungal, and virucidal surface design. ACS Nano19, 12606–12625 10.1021/acsnano.4c15671 (2025). [DOI] [PubMed] [Google Scholar]

[CR17] 17.Guan, M. et al. Long-lasting bactericidal activity through selective physical puncture and controlled ions release of polydopamine and silver nanoparticles-loaded TiO₂ nanorods in vitro and in vivo. Int. J. Nanomed.14, 2903–2914 10.2147/ijn.S202625 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Gao, C. et al. Poly(dopamine) and Ag nanoparticle-loaded TiO₂ nanotubes with optimized antibacterial and ROS-scavenging bioactivities. Nanomedicine14, 803–818 10.2217/nnm-2018-0131 (2019). [DOI] [PubMed] [Google Scholar]

[CR19] 19.Chen, Y. et al. Improved immunoregulation of ultra-low-dose silver nanoparticle-loaded TiO₂ nanotubes via M2 macrophage polarization by regulating GLUT1 and autophagy. Int. J. Nanomed.15, 2011–2026 10.2147/ijn.S242919 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Wu, Y., Liu, P., Mehrjou, B. & Chu, P. K. Interdisciplinary-inspired smart antibacterial materials and their biomedical applications. Adv. Mater.36, e2305940 10.1002/adma.202305940 (2024). [DOI] [PubMed] [Google Scholar]

[CR21] 21.Sun, M. et al. Magnetic microswarm and fluoroscopy-guided platform for biofilm eradication in biliary stents. Adv. Mater.34, 2201888 10.1002/adma.202201888 (2022). [DOI] [PubMed] [Google Scholar]

[CR22] 22.Johnson-Love, O., Salmeron-Sanchez, M., Reid, S., Childs, P. G. & Dalby, M. J. Vibration-based cell engineering. Nat. Rev. Bioeng.3, 408–429 10.1038/s44222-025-00273-x (2025). [DOI] [Google Scholar]

[CR23] 23.Ozcivici, E. et al. Mechanical signals as anabolic agents in bone. Nat. Rev. Rheumatol.6, 50–59 10.1038/nrrheum.2009.239 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Kalelkar, P. P., Riddick, M. & Garcia, A. J. Biomaterial-based delivery of antimicrobial therapies for the treatment of bacterial infections. Nat. Rev. Mater.7, 39–54 10.1038/s41578-021-00362-4 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Iwamoto, H. The tymbal muscle of cicada has flight muscle-type sarcomeric architecture and protein expression. Zool. Lett.3, 15 10.1186/s40851-017-0077-4 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Liu, Y. et al. Biomimetic electroactive materials and devices for regenerative engineering. Nat. Rev. Electr. Eng.2, 188–204 10.1038/s44287-025-00145-x (2025). [DOI] [Google Scholar]

[CR27] 27.Jolissaint, J. E., Mallett, K. E., Thomson, A. L., Carli, A. V. & Austin, M. S. Cefazolin and the R1 side chain: Why patients with a cephalosporin allergy can be safely given cefazolin while undergoing joint arthroplasty. J. Bone Jt. Surg. Am. 10.2106/jbjs.25.01349 (2026). [DOI] [PubMed]

[CR28] 28.Dance, A. Five ways science is tackling the antibiotic resistance crisis. Nature632, 494–496 10.1038/d41586-024-02601-4 (2024). [DOI] [PubMed] [Google Scholar]

[CR29] 29.Zhang, J., Dong, H., Liu, B. & Yang, D. Biomimetic materials for antibacterial applications. Small21, e2408543 10.1002/smll.202408543 (2025). [DOI] [PubMed] [Google Scholar]

[CR30] 30.Super, M. et al. Biomaterial vaccines capturing pathogen-associated molecular patterns protect against bacterial infections and septic shock. Nat. Biomed. Eng.6, 8–18 10.1038/s41551-021-00756-3 (2022). [DOI] [PubMed] [Google Scholar]

[CR31] 31.Carter, B. J. et al. Cefazolin is associated with a lower periprosthetic joint infection risk than alternative antibiotic regimens following total hip arthroplasty. J. Arthroplasty40, 2671–2678 10.1016/j.arth.2025.04.079 (2025). [DOI] [PubMed] [Google Scholar]

[CR32] 32.Bergstein, V. E., Taylor, W. L. T., Weinblatt, A. I. & Long, W. J. Hidden costs of first choice alternatives: a financial model of thromboprophylaxis and prosthetic joint infection prophylaxis in total knee arthroplasty. J. Orthop.63, 87–92 10.1016/j.jor.2024.10.053 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Rumalla, K. C. et al. Noncefazolin antibiotic prophylaxis is associated with increased periprosthetic joint infection risk following total knee arthroplasty. J. Arthroplasty41, 601–607 10.1016/j.arth.2025.06.057 (2026). [DOI] [PubMed] [Google Scholar]

[CR34] 34.Arciola, C. R., Campoccia, D. & Montanaro, L. Implant infections: adhesion, biofilm formation and immune evasion. Nat. Rev. Microbiol.16, 397–409 10.1038/s41579-018-0019-y (2018). [DOI] [PubMed] [Google Scholar]

[CR35] 35.Lin, Y. et al. Systemic T-cell and local macrophage interactions mediate granule size-dependent biological hydroxyapatite foreign body reaction. BMEMat3, e12133 10.1002/bmm2.12133 (2025). [DOI] [Google Scholar]

[CR36] 36.Zhang, Y. et al. A smart coating with integrated physical antimicrobial and strain-mapping functionalities for orthopedic implants. Sci. Adv.9, eadg7397 10.1126/sciadv.adg7397 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Genova, L. A. et al. Mechanical stress compromises multicomponent efflux complexes in bacteria. Proc. Natl. Acad. Sci. USA116, 25462–25467 10.1073/pnas.1909562116 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Bian, S. et al. Antibacterial hydrogel: the sniper of chronic wounds. BMEMat3, e12135 10.1002/bmm2.12135 (2025). [DOI] [Google Scholar]

[CR39] 39.Vishwakarma, A., Dang, F., Ferrell, A., Barton, H. A. & Joy, A. Peptidomimetic polyurethanes inhibit bacterial biofilm formation and disrupt surface established biofilms. J. Am. Chem. Soc.143, 9440–9449 10.1021/jacs.1c02324 (2021). [DOI] [PubMed] [Google Scholar]

[CR40] 40.Chopra, D., Guo, T., Jayasree, A., Gulati, K. & Ivanovski, S. Bioinspired, bioactive, and bactericidal: anodized nanotextured dental implants. Adv. Funct. Mater.34, 2314031 10.1002/adfm.202314031 (2024). [DOI] [Google Scholar]

[CR41] 41.Ma, H., Yang, X., Tang, X., Cao, X. & Dai, R. Self-assembled Co-doped β-Bi2O3 flower-like structure for enhanced photocatalytic antibacterial effect under visible light. Appl. Surf. Sci.572, 151348 10.1016/j.apsusc.2021.151348 (2022). [DOI] [Google Scholar]

[CR42] 42.Wang, L. et al. Biomimetic targeting nanoadjuvants for sonodynamic and chronological multi-immunotherapy against holistic biofilm-related infections. Adv. Mater.36, e2308110 10.1002/adma.202308110 (2024). [DOI] [PubMed] [Google Scholar]

[CR43] 43.Ou, Z. et al. Biodegradable Janus sonozyme with continuous reactive oxygen species regulation for treating infected critical-sized bone defects. Nat. Commun.15, 10525 10.1038/s41467-024-54894-8 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Fan, Y. et al. Mussel-mimetic polysaccharide-based injectable hydrogels for biomedical applications. BMEMat e12089 10.1002/bmm2.12089 (2024). [DOI]

[CR45] 45.Tang, W., Fischer, N. G., Kong, X., Sang, T. & Ye, Z. Hybrid coatings on dental and orthopedic titanium implants: current advances and challenges. BMEMat e12105 10.1002/bmm2.12105 (2024). [DOI]

[CR46] 46.Lee, S. et al. Magnetic control and real-time monitoring of stem cell differentiation by the ligand nanoassembly. Small17, e2102892 10.1002/smll.202102892 (2021). [DOI] [PubMed] [Google Scholar]

[CR47] 47.Savli, M. et al. Magnetomechanical detachment of bacterial biofilms using anisotropic magnetic iron oxide nanochains. ACS Appl. Bio. Mater.8, 8059–8071 10.1021/acsabm.5c01029 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Tang, Y. et al. Photoelectrons sequentially regulate antibacterial activity and osseointegration of titanium implants. Adv. Mater.36, e2307756 10.1002/adma.202307756 (2024). [DOI] [PubMed] [Google Scholar]

[CR49] 49.Shi, H. F., Cheung, W. H., Qin, L., Leung, A. H. & Leung, K. S. Low-magnitude high-frequency vibration treatment augments fracture healing in ovariectomy-induced osteoporotic bone. Bone46, 1299–1305 10.1016/j.bone.2009.11.028 (2010). [DOI] [PubMed] [Google Scholar]

[CR50] 50.Jing, D. et al. Low-1 level mechanical vibration improves bone microstructure, tissue mechanical properties and porous titanium implant osseointegration by promoting anabolic response in type 1 diabetic rabbits. Bone106, 11–21 10.1016/j.bone.2017.10.001 (2018). [DOI] [PubMed] [Google Scholar]

[CR51] 51.Unnithan, A. R. et al. Remotely actuated magnetic nanocarpets for bone tissue engineering: non-invasive modulation of mechanosensitive ion channels for enhanced osteogenesis. Adv. Funct. Mater.32, 2201311 10.1002/adfm.202201311 (2022). [DOI] [Google Scholar]

[CR52] 52.Sun, W. et al. The mechanosensitive Piezo1 channel is required for bone formation. eLife8, e47454 10.7554/eLife.47454 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Hu, Y. et al. The critical role of the Piezo1/β-catenin/ATF4 axis on the stemness of Gli1+ BMSCs during simulated microgravity-induced bone loss. Adv. Sci.10, 2303375 10.1002/advs.202303375 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Guan, H. et al. Magnetic aggregation-induced bone-targeting nanocarrier with effects of Piezo1 activation and osteogenic-angiogenic coupling for osteoporotic bone repair. Adv. Mater.36, e2312081 10.1002/adma.202312081 (2024). [DOI] [PubMed] [Google Scholar]

[CR55] 55.Yang, J. et al. Accelerated bone reconstruction by the Yoda1 bilayer membrane via promotion of osteointegration and angiogenesis. Adv. Healthc. Mater.12, e2203105 10.1002/adhm.202203105 (2023). [DOI] [PubMed] [Google Scholar]

[CR56] 56.Guo, X. et al. Magnetically guided mechanoactive mineralization scaffolds for enhanced bone regeneration. Adv. Funct. Mater.35, 2503903 10.1002/adfm.202503903 (2025). [DOI] [Google Scholar]

[CR57] 57.Gao, Q., Wang, M., Hou, X., Li, M. & Li, L. Substrate stiffness modulates osteogenic and adipogenic differentiation of osteosarcoma through PIEZO1 mediated signaling pathway. Cell. Signal.127, 111601 10.1016/j.cellsig.2025.111601 (2025). [DOI] [PubMed] [Google Scholar]

[CR58] 58.Li, X. et al. Stimulation of Piezo1 by mechanical signals promotes bone anabolism. Elife8 10.7554/eLife.49631 (2019). [DOI] [PMC free article] [PubMed]

[CR59] 59.Wang, B. et al. Bone repairment via mechanosensation of Piezo1 using wearable pulsed triboelectric nanogenerator. Small18, 2201056 10.1002/smll.202201056 (2022). [DOI] [PubMed] [Google Scholar]

[CR60] 60.Chu, G. et al. Mechanochemically reprogrammed interface orchestrates neutrophil bactericidal activity and apoptosis for preventing implant-associated infection. Adv. Mater.36, e2311855 10.1002/adma.202311855 (2024). [DOI] [PubMed] [Google Scholar]

[CR61] 61.Bai, H. et al. Mechanical control of innate immune responses against viral infection revealed in a human lung alveolus chip. Nat. Commun.13, 1928 10.1038/s41467-022-29562-4 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR62] 62.Tian, M. et al. Anisotropic nanocomposite hydrogel dressings with thermo-responsive contraction for actively stimulated tissue regeneration. Chem. Eng. J.514, 163202 10.1016/j.cej.2025.163202 (2025). [DOI] [Google Scholar]

[CR63] 63.Guo, G. et al. Neutrophil function conversion driven by immune switchpoint regulator against diabetes-related biofilm infections. Adv. Mater.36, e2310320 10.1002/adma.202310320 (2024). [DOI] [PubMed] [Google Scholar]

[CR64] 64.Zhu, Z. et al. Tissue/Organ adaptable bioelectronic silk-based implants. Adv. Mater.36, e2405892 10.1002/adma.202405892 (2024). [DOI] [PubMed] [Google Scholar]

[CR65] 65.Zhang, H. et al. A bioinspired virus-like mechano-bactericidal nanomotor for ocular multidrug-resistant bacterial infection treatment. Adv. Mater.37, e2408221 10.1002/adma.202408221 (2025). [DOI] [PubMed] [Google Scholar]

[CR66] 66.Zeng, J. et al. Direct synthesis of water-dispersible magnetic/plasmonic heteronanostructures for multimodality biomedical imaging. Nano Lett.19, 3011–3018 10.1021/acs.nanolett.9b00171 (2019). [DOI] [PubMed] [Google Scholar]

[CR67] 67.Chen, Y. et al. Multifunctionally disordered TiO nanoneedles prevent periprosthetic infection and enhance osteointegration by killing bacteria and modulating the osteoimmune microenvironment. Theranostics14, 6016–6035 10.7150/thno.98219 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Nano-kinetic implant surface mediates osteo-antibacterial synergy through electromagnetic-mechanical coupling

Yangmengfan Chen

Hao Du

Heng Sun

Xingcan Huang

Duan Wang

Xuming Chen

Jiehao Chen

Xuanjun Zhang

Liqiang Zhou

Zongke Zhou

Abstract

Introduction

Fig. 1. Schematic diagram illustrating the synthesis and working principle of dynamic biomimetic MMTS with EMF exposure.

Results

Preliminary exploratory analysis of S. aureus and bone harvested from human hip prosthesis suggest the key role of mechanical stimulation

Fig. 2. Multi-omics profiling of patient-derived S. aureus and histological analysis of periprosthetic bone tissue.

Characterization and principle of the dynamic biomimetic MMTS

Fig. 3. Characterization of the dynamic biomimetic MMTS.

Dynamic biomimetic MMTS under EMF exposure enhances cell viability and adhesion

Fig. 4. Cell viability, adhesion, and Runx2 levels in BMSCs cultured on dynamic biomimetic MMTS.

Periodic mechanical stimulation induced by dynamic biomimetic MMTS under EMF exposure enhances osteogenesis of BMSCs

Fig. 5. Osteogenic differentiation ability of BMSCs on Dynamic Biomimetic MMTS under EMF Exposure.

Activation of Piezo1 and calcium ion oscillations induced by dynamic biomimetic MMTS upon EMF exposure

Fig. 6. Regulation of Piezo1 expression and function in BMSCs by dynamic biomimetic MMTS under EMF exposure.

Dynamic biomimetic MMTS with EMF exposure effectively kills bacteria and disrupts biofilms via mechanical destruction of bacterial structures

Fig. 7. Antibacterial and anti-biofilm performance of dynamic biomimetic MMTS under EMF exposure.

Dynamic biomimetic MMTS eliminates infection and promotes osteointegration upon EMF exposure in vivo

Fig. 8. In vivo evaluation of dynamic biomimetic MMTS efficacy in eliminating infection and promoting osteointegration.

In vivo evaluation of fluorescence distribution and systemic biocompatibility of dynamic biomimetic MMTS upon EMF exposure

Fig. 9. In vivo evaluation of functional protein distribution and biocompatibility of Dynamic Biomimetic MMTS.

Discussion

Methods

Ethical statement

Synthesis process of specimen

Characterization of specimens

Magnetic hysteresis loop measurement

Magnetothermal effect assessment

Magnetic-field-induced motion of JNHDs

Human specimens

Cell isolation and culture

Cell viability assay

Live/Dead staining

Cytoskeleton staining

DCFH-DA assay

SYTOX Green staining

Immunofluorescence staining

Immunofluorescent staining of primary cilia in BMSCs

RNA isolation and RT-qPCR

Primary cilia knockdown

Transcriptome sequencing and analysis

Western blot

Fabrication and calibration of EMF exposure system

Intracellular Ca2+ staining

ALP staining and Alizarin red staining

Antibacterial assay

Crystal violet staining

Animal experiment

Micro-CT evaluation

Histological analysis

Statistics and reproducibility

Reporting summary

Supplementary information

Source data

Acknowledgements

Author contributions

Peer review

Peer review information

Funding

Data availability

Competing interests

Footnotes

Contributor Information

Supplementary information

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

Intracellular Ca²⁺ staining