Wet-adhesive and antibacterial PAH-TPP coacervates with self-mineralizing capability for cranial flap fixation

Abstract

Repositioning and securing the cranial bone flap after craniotomy remain significant challenges in neurosurgery. Traditional fixation methods often suffer from weak mechanical strength, bioinertness, limited osteogenic capacity, and a lack of antibacterial properties, complicating clinical outcomes. Recent medical adhesives offer superior fixation but face significant limitations in cranial bone applications. In this study, we explored the application of PAH (Poly (allylamine) hydrochloride)-TPP (Tripolyphosphate) coacervate (PT) as a bone adhesive. The PT coacervate demonstrated excellent anti-swelling (anti-swelling ratio less than 1 %), self-healing, and injectable properties, as well as exceptional shape adaptability and cytocompatibility. Adhesion tests revealed its outstanding adhesion (99.06 ± 11.76 kPa for lap shear and 121.42 ± 16.73 kPa for end to end), long-term durability, and tunable adhesion strength. Furthermore, the coacervate demonstrated broad-spectrum antibacterial activity against both Gram-positive and Gram-negative bacteria (antibacterial rate more than 90 %), with mechanistic studies revealing promising strategies to address localized and systemic drug-resistant infections. Additionally, the coacervate's self-mineralizing properties significantly enhanced its osteogenic performance. In vivo studies confirmed its effective fixation, robust antibacterial activity, and improved osteogenesis, underscoring its suitability for cranial bone flap repositioning and fixation after craniotomy. In summary, this coacervate adhesive offers a promising therapeutic solution for addressing the challenges of cranial flap fixation in neurosurgery.

Graphical abstract

The coacervates, formed by the straightforward combination of oppositely charged PAH and TPP through non-covalent interactions, exhibit excellent adhesive properties and potent antibacterial activity. By leveraging insights into their antibacterial mechanisms, these coacervates hold great promise as a novel strategy for combating drug-resistant infections. Moreover, their self-mineralization capability significantly enhances osteogenic performance, enabling effective fixation, antibacterial action, and osteogenic effects in vivo.

Highlights

• •The coacervate exhibited exceptional wet adhesion, long-term durability, and adjustable adhesion strength, presenting a promising therapeutic solution for overcoming challenges in cranial flap fixation during neurosurgery.
• •It demonstrated broad-spectrum antibacterial activity against both Gram-positive and Gram-negative bacteria.
• •The coacervate's self-mineralizing properties significantly enhanced its osteogenic performance.

1. Introduction

A multitude of neurological conditions, including comminuted skull fractures, intracranial neoplasms, and cerebrovascular disorders, are treated via craniotomy [[1], [2], [3], [4]]. After surgical intervention, cranial bone flaps must be redeposited to their original locations and firmly attached to the skull [[5], [6], [7]], which is crucial for reconstructing the skull integrity and protecting the cranial cavity, as well as for aesthetic purposes [8,9]. An ideal system for securing the bone flap should ensure safety, reliability, and straightforward application while minimizing immune response and preventing interference with neuroimaging [10]. In contemporary clinical practice, sutures, polyether ether ketone (PEEK)-based fixation devices, and titanium fixation systems (plates, screws, and clamp-like devices) are the main means for bone flap fixation. Suturing a bone flap is often insufficient to provide adequate securement, predisposing the flap displacement [[11], [12], [13]]. PEEK-based fixation devices exhibit superior mechanical properties and X-ray permeability, however, they are mostly bioinert, which restricts their widespread application [14,15]. Titanium fixation devices demonstrate exceptional corrosion resistance and offer reliable stabilization to avert bone flap displacement [8,16]. Nonetheless, these devices may compromise postoperative CT and MRI scans by inducing imaging artifacts, which complicates the monitoring of disease progression; moreover, titanium-based materials may affect osseointegration, leading to aseptic loosening in clinical scenarios [[17], [18], [19]].

Nowadays, adhesives are widely applied in various medical scenarios and are expected to replace invasive fixation tools such as sutures and staples [20]. However, there are still no reports on the use of adhesives in clinical cranial bone fixation, although, given their straightforward use and strong fixation capabilities, adhesives can offer a novel and promising strategy for cranial bone fixation. Ideally, cranial bone adhesives should possess excellent cytocompatibility, adhesive strength, osteogenic capability, and antibacterial properties. Current adhesives include synthetic adhesives and nature-derived adhesives [21]. Cyanoacrylate adhesives are synthetic adhesives used in clinical applications. However, cyanoacrylate adhesives rapidly polymerize in water and form a hard and brittle film on the tissue, leading to weak wet adhesion and tissue damage, limiting their further applications [[22], [23], [24]]. Synthetic adhesives prepared using thiol-ene [25] or thiol-yne [26] coupling chemistry, Michael addition reactions [27], Schiff base reaction [28], and phenol-amination chemistry [29] have been recently reported to achieve strong adhesion. However, by-products generated during the synthetic process and extreme reaction conditions may cause cytotoxicity, restricting their use [30,31]. Natural-derived adhesives, including fibrin glue [32], alginate- [33], silk fibroin- [34], gelatin- [35], and bio-inspired adhesives, such as catechol-containing adhesives [36], exhibit excellent cytocompatibility. However, their adhesion strength is significantly lower compared to synthetic adhesives [37]. Therefore, neither synthetic nor natural-derived adhesives are fully suitable for cranial bone fixation, urging further improvements in the field.

Recently, coacervates have gained widespread attention and are expected to become the next generation of bone adhesives [21]. Coacervates are formed through liquid-liquid phase separation (LLPS) by simply combining oppositely charged macromolecules through non-covalent interactions, such as electrostatic interactions, hydrophobic interactions, and hydrogen bonding [21,38]. Therefore, little by-product formation accompanies coacervate synthesis. These interactions also endow coacervates with excellent cytocompatibility, wet adhesion, and anti-swelling properties, allowing also their tunable performance (e.g., by altering pH or temperature) [39,40]. Coacervates fabricated from polyethyleneimine (PEI)/polyacrylic acid (PAA) [41] and hyaluronic acid (HA)/chitosan [42] exhibited excellent cytocompatibility and wet adhesion, but they lacked osteogenic properties, which has limited their applications as bone adhesives. Bone is mainly composed of organic matrix and inorganic minerals, and the bones' fracture repair process involves the deposition of minerals in the bone matrix [43], making mineralization critical for bone repair. Coacervates with mineralization properties can enhance osteogenic performance. Mineralized adhesives are typically prepared via dispersing minerals into a precursor solution of adhesives, but this step is still challenging due to uneven mineral distribution [44]. Moreover, the mismatch in mechanical properties at the interface of a rigid mineral and a soft polymer matrix often makes the direct addition of minerals critical, because it can deteriorate the adhesive's properties [45]. Thus, self-mineralizing adhesives may solve the interfacial mismatch problem by in situ chelating calcium ions from the environment, thereby maintaining the adhesive's performance [46].

Antibacterial properties are also important for cranial bone adhesives since anti-infection therapy is crucial for a successful cranial bone fixation following craniotomy. The incidence rate of post-neurosurgical site infections is as high as 10 % [47], which can severely impede the cranial repair process, delay healing, or even cause a non-union of the skull [[48], [49], [50]]. Antibiotics are the primary means for antibacterial treatment of bone adhesives. However, the misuse of antibiotics can lead to increased bacterial resistance, significantly impeding bone healing [51,52]. This underscores the critical need for advancing bone adhesives with inherent antibacterial properties, an area that still requires substantial development.

Multifunctional adhesives with antibacterial and self-mineralizing properties and sufficient adhesion strength have not been reported so far. Previous studies have developed PAH (Poly (allylamine) hydrochloride)-TPP (Tripolyphosphate) coacervate (PT) and explored its potential as a wet and self-healing adhesive, demonstrating its promising application in medical, household, and industrial field [[53], [54], [55]]. Based on these findings, this study further investigates the additional properties and application of PT coacervate as a novel bone adhesive. The PT coacervate was synthesized from PAH and TPP. PAH is expected to provide excellent antibacterial activity [56], meanwhile TPP possesses the superior ability to chelate calcium ions and promote osteogenesis [57], which may endow the coacervate with triple functionalities of anti-infection, self-mineralization, and adhesion for cranial flap fixation (Scheme 1). Primarily, the physicochemical properties of the PT coacervate were rigorously tested, followed by an evaluation of its adhesive capacity. Subsequent cytocompatibility assessments and antibacterial activity analyses were performed. Then, we explored the antibacterial mechanism of the PT coacervate. The self-mineralizing properties of the coacervate were confirmed in vivo and in vitro. Ultimately, a rat model of cranial infected comminuted fracture was employed to evaluate the material's potential for strong adhesion, efficient antibacterial effect and ultimate bone repair.

Scheme 1.

Schematic diagram of the preparation of PT coacervate and treatment principle. The PT coacervate can fix the cranial bone flap after cranial surgery, killing bacteria by disrupting their cell membranes, and promoting osteogenesis by self-mineralization.

2. Materials and methods

2.1. Materials

PAH (Mw 18,000–20000) was acquired from Admas-beta Chemical Reagent Co. TPP (sodium salts, Mw 367.86) was obtained from Aladdin. Co. Staphylococcus aureus (S. aureus), Escherichia coli (E. coli) and Methicillin-resistant Staphylococcus aureus (MRSA) were purchased from Ningbo Testobio Co. Cyanoacrylate was supplied from Jitian Biotech Co.

2.2. Preparation of the PT coacervate

Solution 1 was synthesized by dissolving 1.23 g of TPP into 10 mL of deionized (DI) water, and Solution 2 was synthesized by dissolving 1.56 g of PAH into 10 mL of DI water. The PT coacervate was synthesized through the combination of Solution 1 with Solution 2 according to the previous report [53]. pH levels of Solution 1 and Solution 2 were adjusted to an equivalent value (1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0) using diluted HCl and NaOH, and measured via pH meter (PB-10, Sartorius). Under the condition of identical pH levels, Solution 1 was introduced into Solution 2 while being vigorously agitated at room temperature. Then, the coacervate was isolated from the solvent via centrifugation. Ultimately, the PT coacervate was immersed in phosphate-buffered saline (PBS) for 12 h to remove the excess PAH and TPP.

2.3. In vitro swelling and degradation tests of PT coacervates

Prior to weighing, the water from the PT coacervate surface was removed using filter paper, with the initial mass recorded as W₀. Subsequently, the samples were individually immersed in an equivalent volume of PBS solution and incubated at 37 °C. At the specified time points (0, 0.5, 1, 3, 5, 7, 14, 28 days), the moisture from the PT coacervates was removed using filter paper, and the samples were weighed, with the obtained values recorded as W_t. The formulas for calculating the swelling ratio and degradation rate are as follows:

Weight remaining ratio ( %) = W_t/W₀ × 100 % (1)

2.4. Rheological study

The investigation of the rheological properties of the coacervates was conducted employing a rheometer (Anton Paar GmbH, Austria). For each test, a 25-mm diameter circular parallel plate geometry was employed at 25 °C. A stress sweep was performed over a range from 0.1 to 100 % at a frequency of 1 Hz to ascertain the loss modulus (G″) and storage modulus (G′) for the various coacervate formulations.

2.5. Fourier transform infrared spectroscopy

Freeze-dried PT7 coacervate samples were ground to fine powders. An infrared spectrometer (Bruker VERTEX 70+HYPERION 2000) was utilized to capture the fourier transform infrared spectroscopy (FTIR) spectra.

2.6. Self-gelling performance

The PT7 coacervate was promptly submerged in liquid nitrogen for approximately 15 min, followed by freeze-drying to eliminate moisture content. Ultimately, the dried PT7 coacervate was pulverized into a powder utilizing a mortar and pestle. Rhodamine B dye was physically mixed into DI water for better visualization of self-gelling. Subsequently, Rhodamine B solution was added to the PT7 coacervate powder to investigate its self-gelling properties.

2.7. Self-healing performance

Rhodamine B dye was mixed into the PT7 coacervate for measuring the self-healing property of coacervate. Then, two coacervate (weighing 0.2 g) with different color were connected by compression using tweezers for 10 s, and followed tensile force (1 N) was applied from both sides after 10 s to observe the self-healing performance of the coacervate.

2.8. Injectability study

The PT7 coacervate was loaded into a 2.5 mL syringe and injected into a glass bottle at a rate of 0.1 mL/s. The injection process was observed using a camera with a magnification of 4 × to evaluate its injectability.

2.9. Shape-adaptive properties

The coacervate was loaded into a custom-made pentagonal mold at a temperature of 37 °C. Under the action of gravity, the coacervate could conform to the mold shape. The process was observed using a camera with a magnification of 4 × to evaluate shape-adaptive properties.

2.10. Adhesion tests

For the lap-shear adhesion strength test, 0.2 g of the coacervates, from which surface moisture had been removed, were applied to the adherend (bovine bone), with an implementation area of 20 mm × 10 mm. Then, another adherend was affixed using a lap shear manner. After the system was placed at room temperature for a duration of 15 min, the shear strength tests were performed using a universal testing machine (Shanghai Hengyi, China). The tests were conducted at a consistent rate of 5 mm/min. The adhesive strength was determined by dividing the peak force by the area of the adhesive interface. For the end-to-end adhesion strength test of bone tissue, 0.1 g of the coacervates, from which surface moisture had been removed, were applied to bovine bone (20 mm × 7 mm). Then, another bone block was bonded in a butt joint manner. The end-to-end strength tests were performed after 15 min of implementation at a constant tensile speed of 5 mm/min. The adhesive strength was determined by dividing the peak force exerted by the area of the adhesive interface. Then, we test the end-to-end adhesive strength of bone tissue in different environments. The glued bone tissue samples were put under different environments, including dry conditions, SBF (simulated body fluid), and acidic SBF (pH = 5.5) for one day, then we test the adhesion strength. Subsequently, we conducted tensile mechanical tests to evaluate the long-term adhesive strength by end-to-end strength tests (1, 3, 7, 14, 28 days). Last, the samples were detached and then reassembled to assess the tensile force and they were subjected to a 180° rotation to further evaluate their tensile strength.

2.11. In vitro antibacterial properties

To prepare an LB (Luria Bertani) medium, the following steps were taken: 10 g tryptone, 5 g yeast extract, and 10 g of NaCl were each dissolved in 950 mL of DI water. The mixture was shaken until complete dissolution of the components. The pH was then adjusted to a range of 7.35–7.45, DI water was added to bring the total volume up to 1 L. The LB medium was subsequently sterilized by autoclaving. For bacterial growth assessment, S. aureus was cultured in the prepared LB medium at 37 °C with shaking. The bacterial growth was assessed by measuring the optical density at a wavelength of 600 nm. A S. aureus solution with a concentration of 1 × 10⁸ CFU/mL was prepared, diluted to a range of 1 × 10⁴ CFU/mL, and then co-cultured with the coacervates at 37 °C for 24 h. Subsequently, 10 μL bacterial suspension was spread onto a solid LB medium, and the resulting colonies were enumerated after incubation at 37 °C for 24 h. The antibacterial properties of coacervates against MRSA and E. coli were evaluated using the same method. In a 24-well plate, 1 mL of S. aureus at a concentration of 10⁶ CFU/mL was introduced and incubated with coacervates for 24 h at 37 °C under a 5 % CO₂ atmosphere. The bacterial suspension from each well was then transferred to a new 24-well plate and incubated for an additional 72 h. Biofilms were fixed with 4 % paraformaldehyde (PFA) for 20 min and stained with a 1 % crystal violet solution for 10 min. Afterward, 300 μL of 95 % ethanol was added to each well, and the plates were agitated to decolorize the stain for 10 min before transferring the samples to a 96-well plate. The absorbance at 595 nm was measured using a microplate reader (BioTek Instruments, USA). 1 mL of S. aureus at a concentration of 1 × 10⁶ CFU/mL were added to each well of a 24-well plate and co-cultured with coacervates to incubate for 24 h under the same incubation conditions. After incubation, the S. aureus samples were subjected to staining using an acridine orange/ethidium bromide (AO/EB) staining kit for a duration of 20 min and examined under an inverted fluorescence microscope. Lastly, 900 μL of the LB medium and 100 μL of S. aureus at a concentration of 1 × 10⁶ CFU/mL were added to each well of a 24-well plate and co-cultured with the coacervates for 24 h at 37 °C in 5 % CO₂. Subsequently, the bacteria were fixed using a 4 % PFA solution for 20 min, followed by dehydration through a graded ethanol series, and then sputter-coated to augment electrical conductivity. The morphological examination of the bacteria was performed using scanning electron microscopy (SEM).5

2.12. Zeta potential property

The samples were divided into 3 groups: PT7 group; S. aureus group; PT7+S. aureus group. 1 mL DI water was added to the PT7 powder in the PT7 group, 1 mL of S. aureus (10⁶ CFU/mL) in the S. aureus group, and 1 mL of S. aureus (10⁶ CFU/mL) was co-cultured with PT7 in the PT7+S. aureus group; these groups were cultured in a shaker at 37 °C for 24 h. Zeta potential property was recorded by a ZEN 3690 Zetasizer Nano ZS.

2.13. RNA sequencing and analysis for S. aureus

S. aureus and S. aureus co-cultured with the PT7 coacervate was collected, and subsequently submitted to Majorbio Co., Ltd. (Shanghai, China) for analysis. For the functional categorization of genes, Gene Ontology (GO) (http://www.geneontology.org) was employed. Differential gene expression analysis was performed using DESeq2. Genes that met the criteria of a p-value less than 0.05 and an absolute log2 fold change (|log2FC|) greater than 1 were identified as differentially expressed, indicating a statistically significant change in gene expression between the groups being compared.

2.14. Cytocompatibility assessment

Primary bone mesenchymal stem cells (BMSCs) were extracted from the femurs of male rats aged six weeks. These cells were incubated in α-MEM medium (Gibco, USA) supplemented with 1 % streptomycin-penicillin and 10 % fetal bovine serum (FBS) (Gibco, USA). These initial cells were then propagated to become second-generation BMSCs for subsequent experimental procedures. The rat BMSCs were seeded in the lower chamber of a 6-well transwell plate at a cell density of 1 × 10⁵ cells per well. Subsequently, the coacervates were positioned in the upper compartment after 24 h. A control group was set up with the upper chamber being free of coacervates. Ultimately, the cytocompatibility of the coacervates was assessed at specified time points using the CCK-8 assay (Beyotime, China) and live/dead staining (Solarbio, China).

2.15. Cytoskeleton staining

BMSCs were initially co-cultured with the coacervates at a concentration of 1 × 10⁵ cells per well for a duration of 24 h. After a 24-hour incubation period, the cultured cells were washed three times with PBS and subsequently fixed with a 4 % paraformaldehyde (PFA) solution (Biosharp, China). Following fixation, the cells were treated with a mixture containing 0.3 % Triton X-100 and 3 % bovine serum albumin (BSA). Finally, the samples were labeled with RFP-conjugated phalloidin (Invitrogen, USA), counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (Solarbio, China), and examined under a fluorescence microscope.

2.16. Hemolysis

The whole blood of rats was collected and preserved in sodium heparin-containing tubes. The samples were diluted with PBS, subjected to centrifugation at 3000 rpm for 8 min, and then resuspended in DI water and PBS to create an erythrocyte solution. 1.5 mL erythrocyte solution in DI water, 1.5 mL erythrocyte solution in PBS, and 1.5 mL of the erythrocyte solution co-cultured with PT coacervates in PBS were incubated at 37 °C. Following a 4-h incubation period at 37 °C, the mixtures were subsequently subjected to centrifugation at 3000 rpm for a duration of 5 min. Following this, the absorbance of the supernatants was quantified at a wavelength of 540 nm utilizing a spectrophotometer. A control group treated with DI water served as the positive control, indicating significant hemolysis, while erythrocytes treated with PBS acted as the negative control, indicating no hemolysis. The hemolysis rate was calculated using the formula:

[(OD₅₄₀^sample − OD₅₄₀^negative) / (OD₅₄₀^positive − OD₅₄₀^negative)] × 100 % (2)

2.17. Subcutaneous implantation model of nude mice

The protocols were authorized by the Institutional Animal Care and Use Committee of the Soochow University (SUDA20240911A22) by following Chinese national standards for the care and use of experimental animals. To study the in vivo mineralization of the materials, PT7 coacervates (0.02 g) and cyanoacrylates (0.02 g) were subcutaneously implanted into 8-week-old nude mice. The PT7 coacervates and cyanoacrylates were retrieved at 14 and 28 days.

2.18. SEM

PT7 coacervates, mineralized PT7 coacervates, and cyanoacrylates were carefully cleaned to remove any surface contaminants and were sectioned to reveal their cross-sectional profiles before lyophilization using a freeze-dryer (Alpha 1–2 LDplus, Christ, Germany). After freeze-drying, the samples were fixed onto a titanium sheet using conductive carbon tape. Then, to enhance the electrical conductivity of the samples during SEM observation, a thin layer of gold was sputtered onto the sample surface using a sputter coater. The internal microstructure of the PT7 coacervates was examined using a SEM (Hitachi Regulus 8100, Japan), and elemental distributions were analyzed through X-ray spectroscopy (EDS).

2.19. Alizarin red S (ARS) staining

The mineralization of PT7 coacervates and cyanoacrylates was evaluated by ARS staining (Solarbio, China). The samples underwent washing with DI water three times. Subsequently, the samples were immersed in Optimal Cutting Temperature (OCT) compound for an extended period of 24 h to facilitate embedding, followed by sectioning to achieve slices with a thickness of 20 μm. The ARS staining protocol, as provided by the manufacturer, was meticulously adhered to for the execution of the staining process.

2.20. X-ray powder diffraction (XRD)

Dried PT7 coacervate samples were pulverized into a powdery form and subjected to analysis using an X-ray diffractometer (Bruker D8 Advance, Germany), which was equipped with a copper X-ray source operating at a current of 40 mA and a voltage of 40 kV. The diffraction data were acquired over a 2θ range from 10° to 80° using CuKα radiation. The exposure time for each measurement was set to 0.1 s, with a step increment of 0.02°.

2.21. In vitro osteogenesis properties of mineralized PT7 coacervates

BMSCs were seeded into a six-well culture plate at a cell density of 1 × 10⁵ cells per well. After 24 h, the growth medium was replaced with an osteogenic differentiation medium supplemented with 10 mM β-glycerophosphate, 10 nM dexamethasone, and L-ascorbic acid 2-phosphate at a concentration of 50 μg/mL (Sigma Aldrich, USA). BMSCs were then co-cultured with PT7 coacervates, as well as with mineralized PT7 coacervates. After 7 days of culture, alkaline phosphatase (ALP) staining was conducted utilizing an ALP staining kit (Beyotime, China). Subsequently, ARS staining was carried out after 14 days.

2.22. Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) analysis

BMSCs were seeded into a six-well culture plate at a cell density of 1 × 10⁵ cells per well. After 24 h, the growth medium was replaced with an osteogenic differentiation medium supplemented with 10 mM β-glycerophosphate, 10 nM dexamethasone, and L-ascorbic acid 2-phosphate at a concentration of 50 μg/mL (Sigma Aldrich, USA). BMSCs were then co-cultured with PT7 coacervates, as well as with mineralized PT7 coacervates for 7 days. Subsequently, total RNA was isolated from BMSCs using Trizol reagent (Invitrogen, Carlsbad, USA). RNA was reverse-transcribed to cDNA by a cDNA synthesis kit after DNase I treatment, followed by real-time quantitative reverse transcription-polymerase chain reaction (RT-qPCR) using a Miscript SYBR Green PCR kit (Qiagen, Inc.). The primer sequences utilized in this study are detailed in Table 1.

Table 1.

QPCR primer sequence.

Gene Name	Primers	Type
Alpl	TATGTCTGGAACCGCACTGAAC CACTAGCAAGAAGAAGCCTTTGG	Forward (5′-3′) Reversed (5′-3′)
Runx2	ATCCAGCCACCTTCACTTACACC GGGACCATTGGGAACTGATAGG	Forward (5′-3′) Reversed (5′-3′)
Gapdh	GACATGCCGCCTGGAGAAAC AGCCCAGGATGCCCTTTAGT	Forward (5′-3′) Reversed (5′-3′)

2.23. In vitro mineralizing property

The mineralization solution was prepared according to the method described by Zhu et al. [58]. A 10 × SBF solution was supplemented with 100 μg/mL polyaspartic acid and 10 mM sodium carbonate. Subsequently, the PT7 coacervates was placed into the solution. The volume of solution was calculated through the equation: V_s = S_a/10. S_a was the apparent surface area of the specimen. The solution was replaced every day. After 3 and 7 days, the samples were removed from the fluid and washed with DI water, respectively. Samples were lyophilized via freeze-dryer (Alpha 1–2 LDplus, Christ, Germany). The internal microstructure of the samples was examined using a SEM (Hitachi Regulus 8100, Japan), and elemental distributions were analyzed through X-ray spectroscopy (EDS).

2.24. Rat cranial infected comminuted fracture construction

The protocols were authorized by the Institutional Animal Care and Use Committee of the Soochow University (SUDA20240911A22) by following Chinese national standards for the care and use of experimental animals. All Sprague-Dawley (SD) rats were categorized into four distinct groups: defect group, control (ctrl) group, cyanoacrylates group, and PT7 group. The rats involved in these experiments were male, weighed approximately 300 g, and were 8 weeks old. Anesthesia was induced by administering 1.5 % pentobarbital at a dosage of 30 mg/kg via intraperitoneal injection. Once the rats were anesthetized and the surgical site was sterilized, calvarial defects measuring 5 mm in diameter were created using a micro bone drill. After the skull was shattered, the bone fragments were not implanted back in the defect group, while bone fragments were placed back to the original position without fixation in the control group; the cyanoacrylate adhesives were used to bond the bone fragments, and then, the bone fragments were implanted back at the fracture site in the cyanoacrylate group, and the PT7 coacervates were used to bond the bone fragments, which were then implanted back at the fracture site. Next, S. aureus was added to the fracture area (10 μL, 10⁴ CFU per rat) to simulate infected bone fracture. Following the closure of the pericranium and skin with sutures, the surgical site underwent a secondary sterilization process.

2.25. Micro-CT analysis

After surgery, at the 2-week and 4-week time points, all nude mice were sacrificed via lethal injection of pentobarbital, after which the PT7 coacervates and cyanoacrylates were extracted. Similarly, at the 4-week and 8-week time points, all rats were sacrificed using a lethal dose of pentobarbital, and their skulls were retrieved. Both the collected materials and skulls were immersed in a 10 % formalin solution for fixation prior to undergoing high-resolution Micro-CT imaging (SkyScan 1276, USA). The micro-CT analysis was conducted with the following settings: 100 kV, 100 mA, 16 slices, with a 0.5 mm aluminum filter for the harvested materials and a 1 mm aluminum filter for the skulls. The SkyScan NRecon software was then used for the reconstruction of the scanned images, which were subsequently analyzed using the SkyScan CTAn software.

2.26. Nanoindentation analysis

Nanoindentation was employed for the quantitative analyses of nanoscale elastic and plastic responses using Anton Paar NHT2 (Austria) with a diamond Berkovich pyramidal indenter under trapezoidal loading. Calibration was performed using standard-fused quartz and aluminum. The elastic modulus and hardness of the samples were deduced from the unloading phase of load-displacement curves, based on Oliver and Pharr's method.

2.27. Histological and immunofluorescent staining

The harvested skull samples were initially fixed in a 10 % formalin solution for 48 h, followed by decalcification. Subsequently, the skull specimens underwent graded dehydration, after which they were embedded in paraffin. Histological sections were prepared with a thickness of 6 μm. The sections were then stained with hematoxylin and eosin (H&E) and Masson's trichrome (MT) stains, and subsequently observed under an inverted microscope. For immunofluorescence staining, the sections were permeabilized for 20 min and blocked for 30 min to prevent nonspecific binding. The primary antibody was applied to the sections and incubated at 4 °C overnight to allow for specific antigen-antibody interactions. On the following day, the sections were incubated with the corresponding fluorescently labeled secondary antibody for 120 min and DAPI for 3 min. Finally, the sections were analyzed using a fluorescence microscope (Zeiss, Germany) to assess the immunofluorescence signals.

2.28. Statistical analysis

Statistical analyses were conducted using OriginPro 2024b software from OriginLab Corporation, Northampton, USA. Data are presented as the mean ± standard deviation, and inter-group differences were assessed via one-way analysis of variance (ANOVA) when the data followed a normal distribution and exhibited homogeneity of variance, with Tukey's post-hoc test for multiple comparisons. If these requirements were not satisfied, the non-parametric ANOVA was used instead. A p-value below 0.05 was deemed to indicate statistical significance; whereas “ns" denotes a lack of statistical significance.

3. Results and discussion

The coacervate in this study was synthesized from two oppositely charged polyelectrolyte solutions (PAH and TPP). The charge density of the polyelectrolytes is closely related to the pH of their respective solutions, thus pH may influence the synthesis of the coacervate [39]. To investigate the effect of pH on coacervate formation, the pH values of PAH and TPP solutions were adjusted to a range of 1.0–10.0, and the solutions were mixed in equal volumes. Upon mixing, transparent PAH and TPP solutions rapidly transformed into a white emulsion and were observed under light microspore as drops with the liquid. Following centrifugation, the synthesized coacervate was separated at the bottom of the tube (Fig. 1A). However, when the pH of precursor solution was 9.0 and 10.0, no coacervate separation was observed at the bottom of the test tube (Fig. S1), indicating that coacervate was not formed when the precursor solution pH was greater than or equal to 9.0. Additionally, the formed coacervate exhibited macroscopically visible and unique adhesive properties (Fig. S2). Given the near-neutral physiological pH of the human body, PT coacervates with pH values of 6.0, 7.0, and 8.0 were selected for further study. The yields of the coacervate at pH values of 6.0, 7.0, and 8.0 were further investigated, and the results showed that the yields of PT6, PT7, and PT8 coacervates were 71.7 %, 71.1 %, and 39.0 %, respectively. PT6 and PT7 coacervates exhibited higher yields. The coacervate formation process was relatively simple and required no specialized equipment. Furthermore, the raw materials used in the synthesis were commercially available and cost-effective, further supporting the potential for large-scale production in clinic. However, challenges in maintaining consistent quality during large-scale synthesis need to be addressed. The swelling and degradation experiments confirmed that the PT coacervate exhibited a swelling ratio of less than 1 % and gradually degraded by the third day (Fig. 1B). The consistent predominance of G″ over G′ determined in amplitude sweep rheological measurements affirms the fluid rheological characteristics of all prepared coacervates, which facilitates their better conformability to diverse shapes and enhances their injectability (Fig. 1C). Considering PT7 coacervate's excellent anti-swelling performance, rheological fluid properties, and the highest G′ value, we selected PT7 for further evaluation.

Fig. 1.

Characterization of PT coacervates. (A) Photographs show the mixing of PAH and TPP to form PT coacervates. (B) Swelling and degradation test of PT coacervates (n = 3). (C) Amplitude sweep rheological measurements of PT coacervates. (D) Self-healing properties. (E) Self-gelling properties. (F) Injectability. (G) Shape adaptability. (H) SEM and EDS elemental distributions of the PT7 coacervate.

The FTIR spectra of PAH, TPP, and the PT7 coacervate are depicted in Fig. S3. In the PT7 coacervate, the amine group (-NH) peak of PAH at 1600 cm⁻¹ shifts to 1530 cm⁻¹ [41], while the phosphate group (-P-O) peak of TPP at 883 cm⁻¹ shifts to 873 cm⁻¹ [59]. The characteristic shifts of carboxylic and amine groups within the PT7 coacervate suggest that the PT7 coacervate is formed via physical interactions, such as hydrogen bonding and electrostatic interactions, rather than covalent bonding. Besides, PT7 coacervate exhibits robust self-healing properties (Fig. 1D). When applied to fracture sites, bone adhesives are often subjected to mechanical stresses from the surrounding bone tissue, which can lead to a reduced lifetime of the adhesive. Adhesives with self-healing properties can better accommodate the mechanical strains around the bone [60]. To explore the powder self-gelling properties of the PT7 coacervate, the PT7 powder was obtained by freeze-drying and subsequent grinding of the PT7 coacervate. The rapid formation of the PT7 coacervate upon the addition of water to the PT7 powder demonstrates the self-gelling property of the coacervate powder (Fig. 1E). The reversible solid-gel phase transition meets the long-term stable storage requirements of the coacervate, while enabling immediate reconstruction during surgery to adapt to complex fracture morphologies and enhancing clinical operability. The excellent injectability and shape adaptability of PT7 coacervate were beneficial for accommodating various irregular fracture ends and facilitating minimally invasive procedures in clinical practice (Fig. 1F and G). SEM and EDS substantiate the formation of coacervates (Fig. 1H).

Lap shear and end-to-end tests were performed on bovine bones to evaluate the adhesive capacity of coacervates. The results demonstrate the best adhesive performance at pH = 7.0, with a bonding strength of 99.06 ± 11.76 kPa for lap shear (Fig. 2A) and 121.42 ± 16.73 kPa for end to end (Fig. 2B). This excellent performance is likely attributed to the elevated linear charge density of PAH under near-neutral pH, maximizing the ionic cross-linking between PAH and TPP. This results in the formation of a stable and robust network structure, thereby conferring superior adhesive properties. We also evaluated the adhesion strength under various conditions: dry, neutral wet, and acidic wet environments, employing an end-to-end connection method. As shown in Fig. 2C, the PT7 coacervate exhibits robust adhesive strength in wet and acidic environments. This property is especially critical under physiological conditions and during infections [61]. Most conventional adhesives suffer from significantly reduced adhesion in such conditions. In contrast, the stable adhesion in different environment of PT7 coacervate could enhance surgical outcomes. In addition, we assessed the long-term adhesive performance of the PT7 coacervate, as shown in Fig. 2D. A gradual decline in the adhesive strength occurs over time; however, robust adhesive strength (70.99 ± 3.43 kPa) is retained even after 4 weeks. Furthermore, we evaluated the adhesive strength of the PT7 coacervate to a range of commonly used medical implants, confirming its robust adhesion to diverse materials and indicating its potential to offer new fixation strategies for surgery (Fig. S4). A solid fixation of bone fractures often necessitates multiple modifications. Therefore, secondary adhesion and adjustable adhesion are crucial for bone fragment bonding [62]. We illustrate the secondary adhesion by dismantling and re-bonding the bovine bone blocks (Fig. 2E). The subsequent evaluation of the adhesive strength in re-bonded bone blocks indicates that the re-bonded pieces exhibit adhesive strength comparable to that of the initial bonds (Fig. 2F). Furthermore, adjustable adhesiveness is demonstrated by rotating the connected bone blocks for 180° (Fig. 2G). After rotation, the adhesion strength of the connected bone pieces shows no significant changes compared to unrotated bone pieces (Fig. 2F). Thus, unlike rigidly fixed medical devices, the adjustable adhesiveness of the PT7 coacervate allows surgeons an opportunity to make adjustments during the postoperative assembly process.

Fig. 2.

Adhesion performance of PT coacervates tested on bovine bone blocks. (A) Lap-shear adhesive strength test of PT coacervates (n = 4). (B) End-to-end adhesive strength tests of PT coacervates (n = 4). (C) Adhesive strength test under dry, wet (pH = 7.0), and acidic (pH = 5.5) conditions (n = 4). (D) Long-term adhesive strength tests of the PT7 coacervate (n = 4). (E) Demonstration of secondary adhesion through rejoining bone pieces with the residual PT7 coacervate. (F) Secondary and adjustable adhesion strength of PT7 coacervates (n = 4). (G) Demonstration of the adjustable adhesion via rotating one of the bone pieces for 180° after binding them with the PT7 coacervate. (∗ indicates p < 0.05; ns means no significance.)

S. aureus, MRSA, and E. coli were co-cultured with the coacervates and incubated on solid LB agar plates for 24 h. The coacervates exhibit excellent antibacterial activity against S. aureus, E. coli, and MRSA (Fig. 3A–D). Potent broad-spectrum antibacterial properties are displayed by PT6, PT7, and PT8 coacervates, with PT7 and PT8 coacervates demonstrating over 90 % efficacy relative to the control group. In skeletal infections, S. aureus emerges as the most prevalent pathogen [63]. To further investigate the clinical antibacterial potential of coacervates, we conducted subsequent experiments using S. aureus.

Fig. 3.

Antibacterial properties of the PT coacervates. The CFUs (A) and antibacterial rates of S. aureus (B), E. coli (C), and MRSA (D) were assessed on a solid LB agarose medium after incubation with PT coacervates (n = 3). (E) Quantitative analysis of biofilm inhibition tests (n = 3). (F) Photographs of the corresponding biofilm inhibition tests. (G) AO/EB staining of S. aureus. (H) SEM images of S. aureus. (∗ indicates p < 0.05, ns means no significance.)

A significant inhibitory effect of PT6, PT7, and PT8 on biofilm formation is shown in Fig. 3E and F. A notable efficiency in killing S. aureus is shown by PT6, PT7, and PT8 coacervates (Fig. 3G). The SEM assessment of the morphology of S. aureus co-cultured with the coacervates (Fig. 3H) indicates that S. aureus bacteria in the control group are densely packed and intact, with preserved bacterial membranes. In contrast, S. aureus bacteria in the PT6, PT7, and PT8 groups are shrunken, suggesting that the PT coacervates exhibit remarkable antibacterial properties, which is crucial for implanted materials to prevent local bacterial infections. Compared to adhesive without antibacterial properties, the PT coacervate can enhance the success rate of surgeries that have failed due to infection.

Nevertheless, although these coacervates exhibit strong antibacterial properties (Fig. 3), the specific antibacterial mechanism remains unclear. Investigating this mechanism could expand their use in various infection-related diseases. Given the superior physicochemical properties and remarkable antibacterial efficacy of the PT7 coacervate, PT7 coacervate was selected for an in-depth investigation of antibacterial mechanism. As shown in Fig. 3H, the PT7 coacervate has damaged the S. aureus cell membrane. Zeta potential analysis showed the positive charge of PT7 (Fig. 4A), which could increase the interaction with negatively charged S. aureus, thereby damaging the cell membrane. Further investigation into the antibacterial mechanism was carried out through RNA sequencing of the viable bacteria, aiming to discern the variations in gene expression patterns of S. aureus under two distinct conditions: untreated and treated with PT7 coacervate. A volcano map was generated to depict the differential expression between the control group and the PT7 group. Among these differentially expressed genes, 539 are upregulated and 381 are down-regulated in the PT7 group compared to the control group (Fig. 4B). Then, we further analyzed the differential genes by gene ontology (GO) enrichment. In the GO enrichment analysis, transmembrane-related molecular function, transmembrane-related biological processes, and membrane-related cellular components are significantly enriched (Fig. 4C and D), indicating that the combination of the PT7 coacervate's positive charge and the negative charge of the bacterial cell membrane may destabilize and disrupt the bacterial membrane, thereby inhibiting the transmembrane transport of substances and potentially accelerating bacterial demise. Furthermore, by leveraging the comprehensive antibiotic resistance database (CARD) [64], we conducted a heatmap analysis to assess the alterations in the antibiotic resistance gene expression of S. aureus (Fig. 4E). The treatment of the PT7 coacervate significantly downregulates the expression of resistance genes. This finding may offer a novel strategy to combat the growing issue of bacterial antibiotic resistance. Additionally, utilizing the virulence factors of pathogenic bacteria (VFPB) database [65], we analyzed the expression of virulence genes and observed a significant downregulation following the PT7 coacervate treatment (Fig. 4F), which reveals that the PT7 coacervate can reduce the pathogenicity of S. aureus.

Fig. 4.

Antibacterial mechanism of the PT7 coacervate. (A) Zeta potential in the PT7 group, S. aureus group, and S. aureus/PT7 group (n = 3). (B) The volcano map for the differential gene expression analysis of S. aureus in the control group vs the PT7 group. (C) GO enrichment analysis in the PT7 group compared with the control group. (D) GO enrichment analysis and their division into three major categories: biological process (BP), cellular component (CC), and molecular function (MF). (E) The heat map of antibiotic resistance genes in the PT7 group compared with the control group according to CARD. (F) The heat map of virulence genes in the PT7 group compared with the control group according to the VFPB database.

Cytocompatibility is a critical factor for the clinical application of biomaterials in human tissues and organs [66]. We assessed the cytocompatibility of PT coacervates in vitro by co-culturing with BMSCs using a transwell system (Fig. 5A). The proliferation rate of BMSCs was determined by CCK-8 assay (Fig. 5B), showing excellent cytocompatibility of PT6 and PT7 groups, whereas the PT8 group exhibits a significant inhibition of BMSC proliferation. The live-dead cell staining assessment confirms the good cytocompatibility of PT6 and PT7 coacervates; however, the PT8 coacervate exhibits cytotoxic behavior (Fig. 5C). Additionally, cell morphology was elucidated by cytoskeletal staining (Fig. 5D). BMSCs in the PT6 and PT7 groups exhibit an elongated morphology, similar to the control group, while the cells in the PT8 group show a crumpled state. These findings suggest that the PT6 and PT7 coacervates were biocompatible and suitable for surgical use. To further verify the biosafety of PT coacervates, hemolysis experiments were conducted. Fig. 5E shows a post-centrifugation image, with clearly separated supernatants for all samples except the positive control. Fig. 5F displays the spectrophotometric results for the supernatants from each group, revealing that all PT coacervates have a lower hemolysis ratio (less than 3 %) than the positive control DI water group, indicating that these coacervates were compatible with blood and can be used in contact with blood [67].

Fig. 5.

In vitro cytocompatibility of PT coacervates. (A) Schematic illustration of the conducted in vitro cytocompatibility test. (B) BMSCs proliferation (1, 3, 5 days) (n = 3). (C) Live/dead staining of BMSCs. (D) Cytoskeleton staining. (E) An optical image of the hemolysis test of coacervates. (F) Hemolysis ratio of each group (n = 3). (∗ indicates p < 0.05, ns means no significance.)

According to the adhesive performance, antibacterial properties and cytocompatibility, the PT7 coacervate was chosen for further investigation. Common clinically applied cyanoacrylate, as a representative of synthetic adhesives [37], was subcutaneously co-implanted in nude mice with the PT7 coacervate to elucidate their self-mineralization potential (Fig. 6A). Fig. 6B shows no changes in the cyanoacrylates' appearance 2 and 4 weeks after implantation. In contrast, the PT7 coacervate exhibits a marked improvement in the mineral content.

Fig. 6.

Self-mineralization properties of adhesives. (A) Schematic illustration describing the subcutaneous implantation model of nude mice. (B) The appearance of adhesives 0, 2, and 4 weeks after implantation. (C) EDS of adhesives 0, 2, and 4 weeks after implantation. (D) Representative Micro-CT reconstructions of adhesives 0, 2, and 4 weeks after implantation. (E) ARS staining of adhesives sections 0, 2, and 4 weeks after implantation. (F) ALP staining after 7 days. (G) ARS staining after 14 days.

The self-mineralization effects of both cyanoacrylate and the PT7 coacervate were further substantiated via SEM and EDS elemental analyses. These results suggested minimal mineralization in cyanoacrylates and substantial mineralization in PT7 coacervates 2 and 4 weeks after implantation (Fig. 6C, Fig. S5). The Micro-CT reconstructions show PT7's superior mineralization effects (Fig. 6D, Fig. S6). Alizarin red staining shows a small number of calcium nodules in cyanoacrylate, in contrast to the PT7 coacervate, where a substantial number of calcium nodules appear with increasing implantation time (Fig. 6E). XRD analysis indicates that the mineralized PT7 is non-crystalline, pointing to the formation of amorphous calcium phosphate (ACP) (Fig. S7), a pivotal mineral constituent in the human body, holding significant potential for bone repair. As an intermediary product in bone mineralization, ACP serves as a reservoir of calcium and phosphate ions, facilitating the proliferation and differentiation of osteoblasts, thereby hastening new bone formation [68,69]. To further explore the differences in osteogenic potential between the mineralized coacervates (PT7-M-2W: PT7 after mineralization for 2 W, PT7-M-4W: PT7 after mineralization for 4 W) and the non-mineralized PT7 coacervate, ALP and ARS staining were employed. ALP staining reveals that the self-mineralized PT7 coacervate notably augments ALP expression compared to both the control and PT7 groups. With the extension of mineralization time, the ability to promote osteogenic differentiation of BMSC is further improved (Fig. 6F). ARS staining demonstrates that BMSCs cultured with the self-mineralizing PT7 coacervate yield an extensive area of calcium nodules, signifying the capacity of the self-mineralized PT7 coacervate to foster BMSC differentiation into osteoblasts, which can significantly promote osteogenesis (Fig. 6G). The expression of osteogenic differentiation genes in BMSCs were shown in Fig. S8. The expression of Alpl and Runx2 in BMSCs of mineralized PT7 coacervates were higher than the control group and non-mineralized PT7 coacervate, which reveals that the self-mineralized properties of PT7 coacervate could enhance osteogenic differentiation of BMSCs. The in vitro self-mineralization effects of PT7 coacervates were further explored. The results indicate that PT7 coacervates also exhibited remarkable in vitro self-mineralization capacity (Fig. S9).

The infected comminuted skull fracture rat model was established to evaluate cranial bone repair and the antibacterial efficiency of the PT7 coacervate in vivo. Four groups were tested in this study: the defect group with cranial bone fragments that were not re-implanted to the fracture site; the ctrl group with cranial fragments that were re-implanted to the fracture site; the cyanoacrylate group with fragments fixed by cyanoacrylate re-implanted to the fracture site; the PT7 group with fragments fixed by PT7 re-implanted to the fracture site (Fig. 7A and B). After 4 and 8 weeks postoperatively, the healing of cranial fractures was evaluated using Micro-CT. The defect, control, and cyanoacrylate groups demonstrate a minimal amount of new bone formation under infected conditions. Bone fragments untreated with adhesive cannot be stabilized in situ, while the bone fragments treated with cyanoacrylate and PT7 adhesives remain substantially stabilized. However, the inferior biocompatibility and antibacterial properties of cyanoacrylate result in a less effective bone repair compared to PT7 (Fig. 7C). Additionally, the mechanical strength of both normal bone and fracture sites was dynamically evaluated after 4 and 8 weeks by nanoindentation (Fig. 7D and E). The normal skulls from SD rats were used as the control (ctrl) group. After 4 weeks, the skull fracture site exhibits significantly lower Young's modulus and hardness than the normal skulls, while they are similar to the normal skulls after 8 weeks. The depth-load curves reveal the difference in hardness and Young's modulus, i.e., harder samples show shallower penetration and softer samples exhibit deeper penetration (Fig. 7F). The findings indicated the excellent properties of the PT7 coacervates for bone repair, facilitating the restoration of the normal skull's mechanical properties at the fracture site within 8-week period.

Fig. 7.

PT7 coacervate-guided skull fixation and healing (A) Schematic illustration describing the rat cranial infected comminuted fracture construction. (B) The appearance of the cranial flap adhesion. (C) Representative Micro-CT reconstructions of different groups after 4 and 8 weeks. (D) Hardness analysis of different groups obtained by nanoindentation (n = 5). (E) Young's modulus analysis of different groups obtained by nanoindentation (n = 5). (F) Depth-load curves of different groups obtained by nanoindentation. (∗ indicates p < 0.05, ns means no significance.)

H&E staining and Masson's trichrome staining confirmed that both cyanoacrylate and PT7 groups show effective fixation of the comminuted bones. A small amount of new bone is formed in the defect, ctrl, and cyanoacrylate groups 4 and 8 weeks postoperatively. In contrast, the PT7 group exhibits significantly improved new bone formation at both time points (Fig. 8A and B). The immunofluorescence staining of S. aureus confirms bacterial presence in the defect, ctrl, and cyanoacrylate groups, whereas no bacteria in the PT7 group (Fig. 8C–S10), suggesting that the PT7 coacervate not only provides stable fixation for cranial bone segments and promotes cranial repair but also effectively eliminates bacterial infections.

Fig. 8.

H&E staining (A) and Masson staining (B) of different groups. (C) S. aureus immunofluorescence staining of the histological sections (green fluorescence represents S. aureus, blue fluorescence represents cell nuclei).

4. Conclusion

In this study, we demonstrate the efficacy of PAH-TPP coacervate as a bone adhesive with superior antibacterial and adhesive properties, for securing cranial bone flaps following craniotomy, while simultaneously promoting subsequent bone repair. The PT7 coacervate, synthesized by coacervation of PAH solution (pH = 7) and TPP solution (pH = 7), demonstrated straightforward preparation and exceptional performance characteristics, including shape adaptability, injectability, and self-gelling capability. In vitro adhesion assays confirmed the coacervate's robust, tunable adhesive and re-adhesive properties, maintaining stable adhesive strength under diverse environmental conditions. Additionally, the coacervate exhibited excellent cytocompatibility and broad-spectrum antibacterial activity against both Gram-positive and Gram-negative bacteria. Mechanistic investigations into its antibacterial effects revealed potential strategies for combating localized and systemic drug-resistant bacterial infections. Notably, the coacervate's self-mineralization capacity supports effective bone repair. In vivo experiments further validated the coacervate's potent antibacterial performance, efficacy in bone fragment fixation, and facilitation of bone regeneration. These findings underscore its promise as a multifunctional biomaterial for cranial surgeries, addressing critical needs in bone fixation, repair, and infection prevention.

CRediT authorship contribution statement

Weicheng Chen: Writing – review & editing, Writing – original draft, Methodology, Investigation, Data curation, Conceptualization. Jiaxu Shi: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Dachuan Liu: Writing – original draft, Validation, Investigation. Kai Lu: Validation, Investigation, Data curation. Jianlong Fu: Validation, Investigation. Jingxi Xu: Writing – review & editing, Validation. Huan Wang: Validation, Investigation. Zhiliang Guo: Validation, Investigation. Li Dong: Validation, Investigation. Di Li: Investigation. Xin Li: Investigation. Miodrag J. Lukic: Writing – review & editing, Visualization. Wei Xia: Writing – review & editing, Methodology. Song Chen: Writing – review & editing, Supervision, Resources, Project administration, Investigation, Funding acquisition, Conceptualization. Bin Li: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization.

Ethics approval and consent to participate

All animal studies were performed with approval from the Institutional Animal Care and Use Committee of the Soochow University (SUDA20240911A22), and all experiments were performed after approval by a local ethics committee at the Institutional Animal Care and Use Committee of the Soochow University.

Declaration of competing interest

Wei Xia is an editorial board member for Bioactive Materials and was not involved in the editorial review or the decision to publish this article. The authors declare that they have no known competing financial interests or personal relationships that could appear to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article.