Enhanced hemostasis and bone regeneration achieved by thrombin–simvastatin-polydopamine loaded spray-dried β-TCP/ gelatin/ soy protein isolate biocomposites in critical rat calvarial defects

Abstract

Effective bone tissue engineering strategies require scaffolds with tailored properties to address specific bone abnormalities and fractures. Herein, porous biocomposite scaffolds were fabricated via repeated freeze-casting and three-layer coating. The prepared system consisted of spray-dried β-TCP granules and bio-macromolecule gelatin-based soya protein isolate (S) combined with a layer of the biomimetic adherent poly-dopamine (PDA) for simvastatin (SIM) coating via dopamine self-polymerization through the dip-coating process. Thrombin (Th) was then loaded to make the final hemostatic hybrid bone graft (TGS/SIM/Th). The PDA adhered layer improved the mechanical qualities as expected (13.93 MPa). Additionally, MC3T3-E1 cell migration (>90%) and osteogenic differentiation (∗∗∗p < 0.001) were superior in vitro. The sustained release of SIM (∼78%) can help form new capillary networks, and the burst release of Th (∼97.2%) supports the creation of stable blood coagulation. The hemostatic time of the TGS/SIM/Th significantly decreased (∗∗∗∗p < 0.0001) in the critical-sized rat calvarial defect model and tail amputation model, which would effectively enhance the bone repair. This study shows that the fabricated biocomposite could represent a new approach for treating bone defects while absorbing a large amount of blood from onsite bleeding.

Graphical abstract

1. Introduction

Bone is an essential, active organ that provides the body with structural support and physical protection [1,2]. The majority of bone abnormalities heal on their own, although the healing process can be impaired by inadequate regeneration under pathological conditions, including bone tumors, osteoporosis, and severe trauma [3,4]. A potential approach for enhancing bone transplant materials is bone tissue engineering, which combines biological and material engineering. The current gold standard for treating bone abnormalities is autografting [5]. However, this has various drawbacks, such as reduced bone mass at the donor site and bone resorption. Another major therapeutic approach is allografting, however, it is constrained by immunological rejection and infection [6,7]. As a result, developing a substitute graft material to repair bone defects is essential. Such materials must be suitable for manufacturing processes that produce high surface area porous structures and have the appropriate mechanical qualities [8]. The use of a scaffold that functions as an artificial extracellular matrix (ECM) aims to promote tissue regeneration and bone restoration in the damaged region [9,10]. The scaffold should promote cell adhesion, facilitating cell division and growth.

Beta tricalcium phosphate (β-TCP) is one of the biomaterials that has attracted interest as a potential replacement for bone grafts because its calcium/phosphate ratio is remarkably similar to that of natural bone tissue [11,12]. However, β-TCP alone is insufficient as it lacks certain essential biological and mechanical properties needed for coordinated healing. In recent years, a variety of polymers and organic/inorganic biomaterials have been used as scaffolds for bone tissue engineering to meet the requirements that cannot be fulfilled by single-component scaffolds [13]. For this reason, bioactive polymers, peptides, or drugs are frequently mixed with β-TCP for robust bone regeneration. Gelatin mixed with β-TCP is a well-known material for matching the needs of bone tissue engineering scaffolds since it can mimic the characteristics of the native organic phase of bone tissue [[14], [15], [16], [17]]. Additionally, gelatin contains the arginine-glycine-asparagine (RGD) sequence, which is crucial for cell attachment, differentiation, and proliferation [[18], [19], [20]].

Surface modification of biomaterials is crucial in bone tissue engineering because it confers certain characteristics that control local cellular activities [21]. Glutaraldehyde was used here as a cross-linking agent, and results found that it enhances the adhesion of soya protein isolate (S) [22]. Being the purest form of soybean proteins, S is often regarded as the preferred environmentally friendly substitute in biomedical applications because of its low immunogenic potential, long-term storage stability as a natural component of the ECM, and biocompatibility [23]. These ECM-mimicking compounds, particularly isoflavones, are among the most potent phytoestrogens during bone remodeling due to their ability to promote improved bone formation and mechanical performance [24].

Simvastatin (SIM) is the most commonly used medication to treat hypercholesterolemia in clinical practice, and it has been shown to have favorable anabolic effects on bone [25,26]. SIM can be administered topically or systemically. However, because of the liver's high first-pass metabolism, systemically administered SIM has low bioavailability at the sites of bone injury [27]. The optimal drug delivery is associated with the loading technique and materials used in the delivery system [28]. For instance, scaffolds coated with SIM accelerate bone healing by releasing SIM at the location of bone defects. A polydopamine (PDA) layer often serves as a carrier supporting biomaterials and osteogenic medications. The ortho-dihydroxyphenyl functional group of dopamine allows it to form covalent or noncovalent bonds with various biomaterials, including metal, inorganic, and organic materials [29,30]. Dopamine polymerization was observed on substrates and in solution [31]. Additionally, PDA coating adds amine groups to the biomaterial's surface, enabling SIM binding [29,32].

Bone healing is an intricate and well-coordinated physiological process that involves the regulation of osteogenic cell migration from the bone marrow compartment to the site of injury [33]. Rapid hemostasis is an essential consideration in determining the efficacy of bone grafts because early bleeding control stabilizes the scaffold and enables fibrin formation [34]. Thrombin (Th) has been reported to actively promote and regulate the inflammatory microenvironment. Therefore, conjugation of Th into bone graft systems reduces the hemostatic time and triggers several osteogenic events biologically, such as early clot formation, extracellular matrix deposition, and osteoblast proliferation [35]. Yang et al. created a drug delivery method by combining SIM acid with a β-TCP scaffold through a PDA coating. Their findings indicated that this composite system enhanced ALP expression while exhibiting good biocompatibility. However, no further in vivo studies of the system were performed [36].

Hydrogels, polymeric scaffolds, and ceramics based on calcium phosphate are only a few of the biomaterial systems that have been investigated recently in periodontal and craniofacial regeneration. Recently, Liang, Zhaoquan et al. have developed a mussel-inspired GelMA-based alendronate-loaded hydrogel that showed excellent bone regeneration ability [37]. Although hydrogels offer good drug delivery and biocompatibility, they often have low mechanical strength and are not a suitable choice for critical bone defects. In another recent study, Chang, Dongbiao et al. reported that Janus collagen membrane with PLGA microsphere-based polymeric scaffold can promote complex bone regeneration through immunomodulation [38]. Mesoporous bioactive glass hybrid nanoparticles developed by Chen, Yongjun et al. can promote bone healing via synergistic angiogenesis and immunomodulation [39]. Wu, Hao et al. developed a porous polycaprolactone/β-tricalcium phosphate composite, which also promotes bone regeneration through immunomodulation [40]. Recent research suggests that, in addition to osteogenesis, rapid hemostasis is a critical yet often overlooked factor in bone defect management. Most existing systems focus primarily on either osteogenesis or hemostasis, lacking a unified approach that addresses both aspects simultaneously.

To address this limitation, we here describe the creation of a novel hemostatic bone graft using spray-dried β-TCP and gelatin crosslinked with glutaraldehyde in combination with natural S, surface-modified by PDA, loaded with SIM, and conjugated with Th using freeze-drying and dip-coating methods. The outer layer of the scaffold was fabricated using rapidly released SIM and Th, while the core structure was composed of biodegradable β-TCP and gelatin. The TGS/SIM/Th scaffold is designed to integrate immunomodulation, drug delivery, hemostasis, and bone regeneration within a single system. Morphological and physical characteristics were examined using a scanning electron microscope (SEM), Thermogravimetric analysis (TGA), and Fourier transform infrared spectroscopy (FTIR). The cumulative sustained drug release of SIM and the burst release of Th were quantified. Pre-osteoblast MC3T3-E1 cells were differentiated in vitro, hemostatic time was calculated, and scaffolds were implanted in critical-sized bone lesions in rat skulls to examine the in vivo consequences. Lastly, the effects of bone regeneration were identified by several osteogenic markers in in vivo samples.

A schematic diagram of the fabricated graft is presented in Scheme 1.

Scheme 1.

Schematic illustration of A. Fabrication process of the TGS/SIM/Th biocomposite scaffold. B. In vivo application, hemostatic function, and C. assumed bone remodeling process after implantation.

2. Experimental

2.1. Materials

Soya flour type I powder (Protein, ∼52% (85+% dispersible and 1% fat, Sigma, USA), Gelatin from porcine skin type A powder (Gel strength 300, Sigma, USA), β-TCP powder, and PMMA were supplied from Inobone Co., Ltd. (Cheonan, South Korea), Simvastatin powder (≥97% (HPLC), Sigma, USA), Dopamine hydrochloride powder ((≥98% (TLC), Sigma-Aldrich, USA), Thrombin (1KU, Sigma, USA). Bio-Rad Laboratories and Sigma, USA, were the suppliers of Triton X-100. The American Type Culture Collection (ATCC) provided pre-osteoblast cells MC3T3-E1(CRL-2593, USA), Eagle's Minimum Essential Medium (α-MEM; CORNING), fetal bovine serum (FBS; Thermo Fisher, USA), Penicillin (100 U/mL) Streptomycin (100 U/mL) (PS; Thermo Fisher, USA), Gibco™ trypsin-EDTA (Thermo Fisher, USA), Osteogenic differentiation medium (iXCells Biotechnologies, USA, Gibco™3-[4, 5-dimethylthiazol-2-yl]-2,5 diphenyltetrazolium bromide (MTT; Thermo Fisher, USA), Dimethyl sulfoxide 99.5 % (DMSO; Daejung Chemicals & Metals Co. Ltd., Korea), Bovine Serum Albumin (BSA; Sigma, USA), Acetone 99.9% (Daejung Chemicals & Metals Co. Ltd., Korea), Ethyl alcohol anhydrous 99.5 % (Daejung Chemicals & Metals Co. Ltd., Korea), Glutaraldehyde 50% aqueous solution (Daejung Chemicals & Metals Co. Ltd., Korea), Glycine 99.5+% powder (Alfa aesar, Johnson Matthey Company), Tris hydroxymethyl aminomethane (TRIS; MW 121.14, Bioshop, Canada Inc., Burlington), Fibrinogen from Bovine plasma powder (Sigma-Aldrich), Phosphate buffer saline tablets (PBS; Sigma, USA, Simulated body fluid (SBF), 4% paraformaldehyde (PFA) and Deionized (DI) water. LIVE/DEAD™ Viability/Cytotoxicity Kit (Thermo Fisher Scientific, USA). For the in vivo study, male Sprague-Dawley (SD) rats aged 12 weeks, weighing 250–300 g each, were purchased from DBL (Chungcheongbuk, Korea).

2.2. Fabrication

2.2.1. Soya protein isolate (S) separation from soya flour (SF)

S was extracted from SF by employing a modified version of previously reported standard protocols [41]. Specifically, H. Wang et al. used a 10:1 water: SF ratio, whereas an 8:1 ratio was used for a basic wash. The centrifugation rate was 12,000 rpm at a temperature of 15 °C for 20 min to separate the soluble and insoluble portions. The supernatant was refrigerated overnight at 4 °C after adjusting the pH to 4.5, which will facilitate the formation of S curds and whey. The curds were neutralized at pH 7, and the freeze-dried S powder was crushed.

2.2.2. Spray-dried beta TCP preparation

Here, 75% w/v β-TCP powder and 25% w/v PMMA were mixed. The slurry was sprayed into a spray-dryer. To burn out the PMMA, the calcination temperature was set to 600 °C for 3 h in a tube furnace. The sintering was conducted at 1100 °C for 6 h in a box furnace. The spray-dried β−TCP (T) powder was approximately ≈ 8 μm in size.

2.2.3. TGS/SIM/Th scaffold preparation

Gelatin (7% w/v) was dissolved in DI water at 60 °C. Spray-dried β-TCP (20% w/v) was added and mixed thoroughly using a high-speed homogenizer. The slurry was then frozen at −80 °C for 24 h and freeze-dried at −70 °C for 48 h to remove the remaining DI water. This is the base structure. The glutaraldehyde solution was crosslinked with the base structure. Crosslinked samples were three times washed with a 0.1 M glycine-0.02 M HCl buffer (pH = 3) to remove excess aldehyde, and finally coated with 2% isolated S, defined as (TGS) at pH 9 at 50 °C. 0.2g of dopamine was dissolved in 100 mL of 10 mM TRIS buffer at pH 8.5. TGS samples were immersed in the solution. After some time, the colorless solution turned black, which indicated PDA formation. SIM (1 μM), 0.1% BSA, and 2% DMSO were mixed in DI water, and PDA-coated samples were transferred into this solution. Lastly, (1KU) Th solution was added to make the final biocomposite TGS/SIM/Th.

2.3. Sodium Dodecyl Sulphate-Polyacrylamide gel Electrophoresis (SDS-PAGE)

A total of 0.2g of SF and S was suspended in 5 mL of DI water, and diluted 1:5 using DI water. SDS-PAGE was performed in a discontinuous buffer system based on the Laemmli method [42]. Both a 12 % separating gel and a 5% stacking gel were used in the experiment. The molecular weight range of the protein markers was 25 kDa to 250 kDa. Until the tracking dye reached the bottom of the gel, electrophoresis was conducted at 70 V in the stacking gel and 100 V in the separating gel. The gels were stained with Coomassie Brilliant Blue G-250.

2.4. SEM & EDS

SEM, JEOL, JSM-6701F, Japan imaging was used to study the cross-sectioned lyophilized microstructures. Samples were initially loaded into the SEM sample stage and coated using a platinum sputter coater (Cressington Scientific Instruments, United Kingdom). To comprehend the shape of the scaffolds, an accelerated voltage of 10 KV was applied. To find out which elements were present in the samples, EDS mapping was used. ImageJ Fiji (NIH, USA) was applied to investigate the distribution of pore sizes.

2.5. Fourier Transform Infrared Spectroscopy (FTIR)

Fourier transform infrared spectroscopy was applied to analyze the composites and raw components (Nicolet Ia10, Thermo-Fisher Scientific, USA). Using the OMNIC 7.3 spectra software, the samples were evaluated with a resolution of around 8 cm⁻¹ at wavenumbers between 400 and 4000 cm⁻¹.

2.6. Compressive strength analysis

The compression strength of the individual cylindrical-shaped scaffolds was examined in the axial direction using a universal testing machine (RB 302 Microload R&B, Korea). Testing was performed at 1 mm/min speed with Heilo-X software.

2.7. Contact angle

Wettability was evaluated using the sessile drop method (Drop-shape analyzer, DSA 100, KRUSS, Germany). Five microliters of distilled water were dropped onto the scaffold surface, which was securely fastened to a level base, using a stainless-steel needle. Every 20 s, a CCD video camera recorded angular deflections. The contact angle was then determined by averaging three data points obtained at different points.

2.8. Swelling behavior

PBS absorption behavior was determined using the conventional weighing method. Briefly, each dry sample was weighed (W_dry), placed on separate petri plates, and then immersed in 5 mL of PBS (W_wet). Subsequently, the samples were kept in an incubator at 37 °C. After 1, 3, 5, 7, 9, and 14 days, the scaffolds wet weights were measured again. The swelling rate (%) was calculated using the following equation:

$$\text{Swelling}\text{rate}=\left({\mathrm{W}}_{\text{wet}–}{\mathrm{W}}_{\text{dry}}\right)/{\mathrm{W}}_{\text{dry}}\times 100\%$$

2.9. Degradation

Degradation studies were conducted at 37 °C in PBS. W₀ represents the dry samples initial weight. Each sample was pre-wetted before degradation to allow PBS to penetrate the whole scaffold. Samples were gently cleaned three times with DI water at a prearranged time and then freeze-dried. The acquired solid mass was annotated as W_t, and the scaffold's rate of degradation was computed using the formula below:

$$\text{Residual}\text{mass}\text{ratio}=\left({\mathrm{W}}_{0-}{\mathrm{W}}_{\mathrm{t}}\right)/{\mathrm{W}}_0\times 100\%$$

2.10. pH sensitivity evaluation

The scaffolds were immersed in 5 mL of PBS to evaluate the change in pH. Initially, the pH of the PBS was 7.4. Using a pH meter (Orion Star A211, Thermo Fisher Scientific), it was measured after 1, 3, 5, 7, 9, and 14 days. During the immersion period, no additional PBS was added.

2.11. In vitro drug release profiles

The in vitro release behavior of simvastatin (SIM) scaffolds was evaluated. The scaffolds were placed in 24-well cell culture plates, and 1 mL of PBS was added to each well and incubated in a humidified CO₂ incubator, at 37 °C and 5% CO₂ (ASTEC, Japan) under gentle shaking (50 rpm). The supernatant was withdrawn at predetermined time points and replaced with fresh PBS to maintain constant volume and conditions. The extract solution's UV absorbance was measured at 240 nm using a UV Spectrophotometer (Biodrop-1005363, Denville Scientific Inc., USA). A standard calibration curve with known SIM concentrations was used to quantify the amount of SIM loaded and released from the scaffolds. All experiments were performed in triplicate (n = 3).

Thrombin (Th) release was determined by following the Th fluorometric activity assay kit (ab197006).

2.12. Degradation and apatite formation

The solution-mediated degradation properties of the scaffolds were evaluated in SBF. The preparation of SBF was performed according to the protocol of Kokubo et al. [43,44]. The chemical components are listed in Table 1, which are similar to the composition of human blood plasma. The pH was adjusted to 7.3 ± 0.5. Samples were placed in a clean petri dish and completely immersed in 5 mL of SBF solution. Next, the dishes were sealed and placed in an incubator that maintained a constant temperature of 37 °C. Every 3 days, the release medium (SBF) was replaced. The predetermined intervals allowed the assessment of weight loss, pH change, and apatite formation. The morphology and microstructures were examined using SEM (JSM-6701F, JEOL, Japan). Apatite coverage percentage was calculated from the obtained SEM images using ImageJ Fiji. The ratio of the apatite-covered area to the overall scaffold surface area was used to determine the percentage surface coverage (%).

Table 1.

Chemical composition of SBF solution.

Order	Reagent	Amount
1.	NaCl	7.996 g
2.	NaHCO3	0.350 g
3.	KCl	0.224 g
4.	K2HPO4. 3H2O	0.228 g
5.	MgCl2. 6H2O	0.305 g
6.	12 M HCl	3.33 mL
7.	CaCl2	0.278 g
8.	Na2SO4	0.071 g
9.	(CH2OH)3CNH2	6.057 g

2.13. Cell culture

MC3T3-E1 pre-osteoblast cells were cultivated in a growth medium with α-MEM that containing 10% FBS, and 1% penicillin-streptomycin (PS). The medium was changed every 48 h, and the cells were maintained in an incubator at 37 °C, 5% humidity, and 5% CO₂.

Osteogenic differentiation medium was used to cultivate MC3T3-E1 cells to facilitate osteogenic differentiation. These cells were kept in an incubator with 5% CO2 at 37 °C (Thermo Fisher Scientific, USA).

2.14. Preparation of the sample media

A total amount of 15 mL falcon tube containing α−MEM medium was filled with 30 mg of each sample. Following a 24 h incubation period at 37 °C, the tube was centrifuged for 5 min at 5000 rpm. Finally, the extracted media from different sample tubes were separated from the sediments at the bottom.

2.15. In vitro biocompatibility study

2.15.1. MTT assay and cell proliferation

The MTT colorimetric test was conducted to evaluate cytotoxicity following the standard protocol [ISO10993-5] [45]. Approximately 2 × 10⁴ MC3T3-E1 cells in 1 mL were seeded in each well of 24-well cell culture plates. The culture medium was changed every 48 h. A hemocytometer was used to count the cells. Following aspiration of the cell media at 1, 3, and 7 days, MTT solution (5 mg/mL) was added to each composition at a 1:10 ratio. After incubation for 4 h, the MTT was removed. Each well plate was then filled with 400 μL of DMSO, and the formazan crystals were left to dissolve for half an hour. Each sample was thereafter moved to a 96-well plate. Lastly, absorbance was measured at 570 nm.

A cell suspension (2 × 10⁴ cells) was seeded onto 24-well plate cell culture dishes at the start of the cell proliferation study. In a petri dish, the same sample extracts were applied to each group of samples. After that, the plates were kept in a humidified incubator with 5% CO2 at 37 °C for 1, 3, and 7 days. After the incubation period, 4% PFA was utilized for 10 min for fixation. To permeabilize the fixed cells, 0.5% Triton X-100 was administered for 10 min following three PBS washes. Then, 2.5% BSA was applied for 2 h to nonspecifically block the cells. Following application of FITC (25 μg/mL) for 3 h, the attached cells' actin cytoskeleton was bound and tagged. For 10 min, HOECHST-33342 (1 μg/mL) was used to stain the nuclei. After thorough washing and air drying, one drop of mounting solution was added to each dish. A fluorescence microscope FV10i-W, Olympus, Tokyo, Japan) was then used to examine the materials, and photographs were taken. ImageJ was used to examine the cell count and F-actin-positive area (%) in relation to the overall image area.

2.15.2. Live and dead assay

Following the manufacturer's instructions, a live and dead assay was conducted using a LIVE/DEAD™ viability Kit. Approximately 1 × 10⁵ MC3T3-E1 cells were first seeded and cultured for 24 h. For the staining solution, 10 mL of PBS, 20 μL of ethidium homodimer-1 (EthD1), and 5 μL of calcein AM were used. After adding 200 μL of the test solution to each sample, it was incubated for 45 min at 20 to 25 °C. A fluorescence FV10i-W microscope (Olympus) was used for visualization. Total live and dead cells in a sample were counted and used to calculate the cell viability using the following equation:

$$\text{Cell}\text{viability}\%=\left(\text{Total}\text{cells}-\text{Dead}\text{cells}\right)/\text{Total}\text{cells}\times 100\%$$

2.15.3. Cell migration assay

To create the conditional media, each sample was immersed for 24 h in the α-MEM containing 10% FBS and 1% PS. To measure cell migration, a 200 μL sterile pipette tip was used to create a linear wound midline across the bottom of the dish on a confluent monolayer of MC3T3-E1 cells. After that, cell migration was recorded at 0 and 24 h by taking magnified photos using an optical microscope (Olympus). Migration rates were calculated from the images using the t-scratch software.

2.16. In vitro hemocompatibility

2.16.1. Hemolysis assay

To perform the hemocompatibility test, 10 mL of fresh blood was extracted from an SD rat and placed in a tube with an anticoagulant acid citrate dextrose (ACD) solution in a 9:1 ratio. To separate the blood components, the tube was centrifuged for 10 min at 25 °C at 3000 rpm. The RBC sediment was obtained by centrifuging the cells separated from the blood plasma and then rinsing them twice in PBS. For each sample, a 2% erythrocyte solution (0.2 mL) was made by mixing the blood sediment with 10 mL of PBS extract. Next, the mixture underwent another centrifugation. Triton X-100 (10 mL of 0.1% Triton X-100 + 0.2 mL of 2% erythrocyte suspension) and PBS (10 mL of PBS + 0.2 mL of 2% erythrocyte suspension) were used as positive and negative controls, respectively. Finally, a microplate reader (Infinite F50, Tecan, Männedorf, Switzerland) was used to measure the absorbance at 540 nm. The hemolysis ratio (HR %) was computed using the following equation:

$$\text{HR}\%=\left(\text{AS}-\text{AN}\right)/\left(\text{AP}-\text{AN}\right)\times 100\%$$

Here, AP stands for the absorbance of the positive control, AN for the absorbance of the negative control, and AS for the absorbance of the sample supernatant.

2.16.2. Blood clotting index

The total blood clotting capacity was evaluated using the blood clotting index (BCI). Typically, 100 μL of fresh rat blood and 5 mg of precisely weighted sample material were combined in a centrifuge tube with 10% 0.1 M calcium chloride solution. Each test tube was filled with 3 mL of filtered water after 30 min of incubation. The absorbance was then measured at 540 nm after 200 μL of each sample's supernatant was transferred to a 96-well plate. The blood clotting index was calculated as follows:

$$\text{BCI}\%=\text{AS}/\text{AC}\times 100\%$$

AS stands for sample absorbance and AC for control absorbance (100 μL of fresh rat blood mixed with 3 mL of water and 10% 0.1 M calcium chloride solution).

2.16.3. Platelet adhesion and activation

To initiate platelet adhesion and activation, the samples were exposed to PRP. In brief, PRP was extracted from whole blood by centrifugation for 5 min at 2500 rpm. Each sample was incubated for 1 h at 37 °C after the addition of 200 μL of PRP. Then, the samples were fixed for 2 h using a 2.5% glutaraldehyde solution, and any unattached PRP was removed using three PBS washes. An ethanol series was used to dehydrate the samples, gradually preparing them for SEM inspection. After that, they were left to air dry at ambient temperature for a whole day under a fume hood. The samples were then gently coated with platinum for the SEM observation and examined at an accelerating voltage of 10 KV. ImageJ software was used to quantify platelet adhesion from SEM images. For each sample, at least three randomly chosen fields were examined. Platelet density was computed as the number of platelets per unit area (10³ μm²) using the following equation after the number of adherent platelets was counted within a certain area:

$$\text{Density}\text{of}\text{platelets}=\left(\mathrm{N}/\text{Area}\right)\times {10}^3$$

N stands for the number of platelets, and Area is the measured area (μm²).

2.16.4. Fibrin polymerization

A turbidity-based assay was used to measure fibrin polymerization in order to assess the scaffolds' capacity to produce clots. In short, a 2 mg/mL fibrinogen solution was made in PBS (pH 7.4). Fibrin production was started using a calcium chloride (CaCl₂) solution (2.5 mM). Scaffold samples were produced as discs (∼5 mg each) and were placed one at a time into each well of a 96-well plate. Each well was then filled with 100 μL of fibrinogen solution and 20 μL of CaCl₂ solution to initiate polymerization. A fibrinogen-only system without CaCl₂ was used as the negative control, and thrombin (1 U/mL) was added to the fibrinogen solution in the absence of scaffolds for the positive control. A microplate reader with a wavelength of 405 nm was used to instantly monitor the polymerization process. At 37 °C, absorbance was measured every 30 s for 30 min. The creation of fibrin networks as a result of light scattering is correlated with an increase in absorbance [[46], [47], [48]].

2.17. In vitro osteogenic differentiation ability

2.17.1. Alkaline phosphatase (ALP) activity

Approximately 1 × 10⁵ cells were seeded onto a 6-well plate and incubated for 7 and 14 days. To visualize the ALP activity of the cells, the well plates were incubated with primary antibody anti-ALP (1:100, Abcam, USA) overnight and secondary antibody Alexa Fluor-488 conjugated Goat anti-rabbit IgG (H + L, Invitrogen, USA) for 2 h. Nuclei were counterstained with HOECHST-33342. The scaffolds were observed under a fluorescence microscope (FV10i-W, Olympus).

2.17.2. Alizarin Red Staining (ARS)

Following co-culturing of the cells with the sample extracts for 14 and 21 days, the cells were fixed for 30 min in 4% PFA and rinsed with PBS. The treated samples were then left at room temperature for 30 min for staining using the ARS solution. Next, the stained samples were rinsed three times with DI water to remove any remaining ARS dye. A high-resolution digital microscope was used to image the stained samples. A solution of 10% cetylpyridinium chloride (Sigma-Aldrich) was made with DI water. To determine the amount of calcium deposition, the stained samples were shaken while being treated with this solution for 1 h at room temperature. Finally, a BioTek SYN-ERGY/H1 multiplate reader was used to measure the supernatant at an absorbance of 570 nm.

2.17.3. Real-time PCR

Total RNA was isolated from cultured cells using RiboEX reagent (GENE ALL, Seoul, Korea), and cDNA was reverse-transcribed using Maxime RT PreMix (Oligo dT Primer) Kit (iNtRON Biotechnology, Gyeonggi-do, Korea), in accordance with the manufacturer's instructions. Gene expression of cell-specific markers (RUNX2, COL1, and OPN) in each group was examined by real-time PCR using PowerUp™SYBR™Green Master Mix (Thermo Fisher Scientific, CA, USA). The specific primers used in this experiment are listed in Table 2. The Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the housekeeping gene.

Table 2.

Primer sequence for Real-Time PCR.

Genes	Strand (5' → 3′)
′	Sense	Antisense
Runx2	AGAAGAGCCAGGCAGGTGCTT	TTCGTGGGTTGGAGAAGCG
OPN	GATGATGATGACGATGGAGAC	GACTGTAGGGACGATTGGA
COL1	CGAGACCCTTCTCACTCCTG	GCATCCTTGGTTAGGGTCAA
GAPDH	TCTCCTGCGACTTCAACA	CTGTAGCCGTATTCATTGTC

2.18. In vivo study

2.18.1. SD Rat calvarial defect model

At first, the ethylene oxide (EO) gas sterilizer procedure was used to sanitize the scaffolds using the HS-3241 EO machine. A total of 18 animals were randomly divided into three groups: control (n = 6), TGS (n = 6), and TGS/SIM/Th (n = 6) in this experiment to ensure that initial animal weights and physiological conditions were evenly distributed across groups. As previously described, inhalation of isoflurane was applied for anesthesia for all surgical procedures. The SD rat skull was thoroughly cleaned with 70% ethanol and povidone-iodine solution, and the hair was completely shaved. A circular defect of 5 mm diameter and 2 mm depth was created using a trephine drill under constant saline irrigation to prevent thermal damage. First, each scaffold was applied to the wound site to check the hemostatic time using preweighed gauges to absorb the excess blood. To record the time, a stopwatch was used. Then the samples were implanted, and the wound site was sutured. Painkillers (Maritrol, Jeilpharm, Korea) and antibiotics (Baytril, Bayer, Korea) were provided. After the anticipated duration of 4-8 weeks, SD rats were euthanized by administering an overdose of isoflurane. Upon reopening the skull, the implanted samples were extracted and preserved in 10% formalin.

2.18.2. SD rat tail amputation model

SD Rats aged 12 weeks were selected for the in vivo rat tail model study. First, each rat received enough oxygen to be sedated by isoflurane (Terrell, TX, USA). To properly sterilize the tails, 70% ethanol was used. The tail was cut at a precise, pre-measured distance (around 3 cm from the tip) using a surgical blade to make sure a uniform wound across all animals. The scaffolds were applied on the wound site with a minimal force (using standardized forceps pressure) just sufficient to maintain contact, rather than vigorous manual compression. In each rat group, n = 6 in this experiment. A timer was used to record how long each administered sample bled overall. To control the bleeding, a preweighed surgical gauge was placed over the wound and used to absorb blood until it stopped flowing.

2.19. Micro-CT analysis

The new bone volume in the implants was measured using micro-CT. The Skyscan 1172 micro-CT scanner was used to build reconstructed 3D images, via an 11-megapixel camera, software version 1.5, and a pixel size of 8.81 μm. Each data scan was processed by batch reconstruction using the Nrecon software version 1.6.9.8. During conventional data reconstruction, post-alignment compensation and related ring artifact correction were applied. Utilizing Data Viewer software version 1.5.1, the recovered data were rotated and reoriented. The following parameters were analyzed using CTAn version 1.18: percent bone volume/tissue volume (BV/TV %), trabecular separation Tb. Sp (mm), trabecular number Tb. N (mm⁻¹), trabecular thickness Tb. Th (mm). All samples were then viewed three-dimensionally using CTvoc version 3.0 software.

2.20. Histological analysis

Samples were fixed in 10% formaldehyde, cleaned, and then subsequently decalcified using 10% EDTA solution with regular solution replacement and complete decalcification. The decalcified samples were dehydrated with sequential ethanol treatment, embedded in paraffin blocks, and sectioned using a microtome (Thermo Fisher Scientific, USA) at a thickness of 5 μm, followed by staining with H & E (ab245880, Abcam) according to the standard protocol. A light microscope was employed to examine the tissue slides to determine the histological structure. For analysis of the images, eXCope software was used.

2.21. Immunofluorescence analysis

The expression of the osteogenesis-related genes such as OPN, COL1, CD34, BMP-2, OCN, CD68, and CD206 was determined by immunofluorescence staining. In short, the sectioned samples were permeabilized by applying 0.5% Triton X-100. After three washes with PBS, 10% goat serum was used to block the samples for 50 min at room temperature. Primary antibodies such as anti-OPN (1:50, Novus, USA), anti-COL1 (1:100, Abcam, USA), anti-CD34 (1:100, Abcam, USA), anti-BMP-2 (1:500, Santa Cruz, USA), anti-OCN (10 μg/ml, Abcam, USA), anti-CD68 (1:100, Bio-Rad, USA), and anti-CD206 (1:800, Cell Signaling) were then added to the tissue sections, and left overnight at 4 °C. The slices were given three washes with PBS, followed by the application of the proper secondary antibodies (1:100) and 2 h at room temperature. The nuclei were stained with HOECHST. A fluorescence microscope was used for visualization and imaging. The recorded images were quantified using ImageJ software.

2.22. Statistical analysis

GraphPad Prism Software v8.0 (GraphPad, San Diego, CA, USA) was used to perform the statistical analysis. One-way ANOVA (Tukey's test) was used for multiple comparisons for a single variable, whereas two-way ANOVA was used for multiple comparisons for two variables. The data are presented as mean ± standard deviation (SD). The criteria for statistical significance were as follows: ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001, and not significant (n.s.).

3. Results and discussion

3.1. Fabrication, morphological and physicochemical characterization of TGS/SIM/Th hemostatic bone grafts

This study started by creating the porous base structure using spray-dried 20% β-TCP (w/v) and gelatin (7%) by the freeze-drying method. The core structure was crosslinked using a 0.2% (w/v) glutaraldehyde solution. S was separated from soya flour (SF) using the pH solubilization technique (Fig. S1). Proteins were first solubilized to remove insoluble fiber during the S preparation process, then precipitated to remove soluble sugars while maintaining pH. Glutaraldehyde modified the inner core surface and strengthened the adhesion of the S. Dopamine was dissolved in TRIS buffer solution at a high pH to form a deep black colored PDA layer from a colorless solution. For SIM coating, the samples were transferred into the SIM solution. Finally, the Th solution was dropped carefully to make the final hemostatic bone graft composition TGS/SIM/Th (Fig. S2). Using SDS-PAGE, the protein profiles of SF and S were examined. The resulting pattern showed bands corresponding to the main constituents of S, which are represented by glycinin (11S) and α’, α, and β subunits of β-conglycinin (7S). Fig. S3 displays the usual protein bands that correspond to the acidic and basic subunits of glycinin and β-conglycinin [49]. The three bands were linked to the trimer, β-conglycinin, and had molecular weights of roughly 75 kDa, 65 kDa, and 54 kDa. The acidic and basic polypeptides of glycinin were identified by the other two bands, which had molecular weights of roughly 33 kDa and 18 kDa, respectively. Many research groups previously reported similar molecular weights of the glycinin and β-conglycinin subunits [[50], [51], [52], [53], [54]].

The honeycomb-like interconnected porous gelatin structure with a uniform distribution of spray-dried TCP granules was visible in the cross-sectional SEM image (Fig. 1A). Even after coatings of various compositions, the fundamental structure remains unchanged, but the shape of the TCP granules is altered from porous to almost entirely enclosed and even. EDS mapping analysis showed the presence of oxygen (O), phosphorus (P), and calcium (Ca), as shown in Fig. 1B and C. The assumed SIM coating interaction reaction based on the PDA adhesion (Fig. 1D).

Fig. 1.

A. Cross-sectional SEM image. (1-3) after coating the surface morphology of TGS, TGS/SIM, and TGS/SIM/Th (yellow arrow represents gelatin structure, green arrow represents spray-dried β-TCP granule), B-C. Element identification by EDS mapping analysis, EDS image of the surface of TGS/SIM/Th, D. assumed PDA reaction with SIM, E. FTIR spectra of different composites, F. Mechanical properties of the scaffolds. We used Tukey's post-hoc test for pairwise comparisons following the One-way ANOVA for Fig. 1F. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the TGS group.

Fourier transform infrared (FTIR) spectra displayed characteristic absorption bands for the raw materials spray-dried β-TCP, gelatin, S, and dopamine (Fig. S4). Spray-dried β-TCP FTIR is presented in Fig. S5. The distinctive absorption peaks of PMMA, including the C=O stretching band (∼1730 cm⁻¹) and C–H stretching bands (∼3000–2850 cm⁻¹), are absent from the sintered sample SIM FTIR spectra. The complete thermal breakdown and elimination of the PMMA during sintering at 1100 °C is confirmed by the lack of these peaks [55,56]. The presence of the characteristic phosphate (PO₄³⁻) vibrational bands is visible particularly at 967 cm⁻¹, which is attributed to the ν₁ symmetric stretching vibration, whereas the peak at 543 cm⁻¹ is associated with the v₄ bending mode (O–P–O). The σ₃ asymmetric stretching vibrations of PO₄³⁻ groups are responsible for the peaks seen at 1001 cm⁻¹. The absence of any other peaks that matched secondary calcium phosphate phases, such as HA or α-TCP, confirmed the effective synthesis and phase purity of β-TCP during sintering [57,58].

Fig. S6 displays the thermogravimetric analysis (TGA) of both raw and spray-dried β-TCP. Over the temperature range from ambient temperature to 1000 °C, both samples showed very little weight loss (<0.3%), demonstrating the exceptional thermal durability of β-TCP. The elimination of physically adsorbed moisture is responsible for the slight weight loss seen below 100 °C. At higher temperatures, no discernible phase shift or breakdown was found, suggesting that β-TCP retains its structural integrity under thermal conditions. Interestingly, compared to raw β-TCP, the spray-dried β-TCP showed slightly less weight loss and more residual mass, indicating lower moisture content and better material purity following processing. These findings verify that the thermal stability of β-TCP is not negatively impacted by the spray-drying procedure.

The scaffolds TGS, TGS/SIM, and TGS/SIM/Th are shown in Fig. 1E. Spray-dried β-TCP structure was confirmed by the presence of the phosphate vibrational bands representing ν₄ bending (O–P–O), ν₁ symmetric stretching (P–O), and C–O stretching (CO₃²⁻) near 563 cm⁻¹, 974 cm⁻¹, and 1437 cm⁻¹. The prominent C=O, C–H, O–H peaks in the FTIR spectrum were observed approximately at 1633 cm⁻¹, 2925 cm⁻¹, and 3312 cm⁻¹ in TGS/SIM/Th, respectively. The C=O stretching vibrations of peptide bonds are the main cause of the Amide I band. Both the gelatin matrix and the Th integrated into the scaffold are responsible for this absorption since they are both protein-based molecules with a large number of amide bonds [59]. Successful drug-protein interaction inside the composite matrix is confirmed by the increased intensity and location of this peak, which indicates conformational changes most likely caused by hydrogen bonding or molecular interactions between SIM and Th. Furthermore, a weaker band that was seen at approximately 1541 cm⁻¹, which corresponded to the Amide II band (N–H bending and C–N stretching), supports the protein component's structural changes during composite production. The distinctive C–H stretching vibrations around 2925 cm⁻¹ proved the presence of the aliphatic chains from SIM. Because SIM contains hydroxyl groups and gelatin, and Th contains amide groups, a wide absorption band at around 3307 cm⁻¹ and 3312 cm⁻¹ in TGS/SIM and TGS/SIM/Th, respectively, corresponds to overlapping N–H and O–H stretching vibrations. This band's broadening and intensification suggest possible stabilization of the composite matrix and the creation of hydrogen bonds, most likely between the drug and protein components [60].

The design and manufacture of scaffolds intended for use in bone tissue engineering depend heavily on the characterization of mechanical properties, particularly compressive strength. Bone healing materials must have strong mechanical qualities in order to withstand physiological pressures without losing their structural integrity [61]. Herein, the compressive strength of the scaffolds was assessed using a uniaxial compression test. Through the analysis of Fig. 1F, TGS exhibited a compression strength of 10.86 MPa. However, compared with that of TGS/SIM (13.53 MPa), the compression value of TGS/SIM/Th (13.93 MPa) was slightly higher due to the crosslinking, which confirms the stability and reliability of the PDA coating of the scaffold [62]. The strength is enhanced in TGS/SIM/Th because of the addition of Th and SIM layers, as these coatings improve particle cohesion and decrease intergranular gaps, acting as a binding matrix and increasing compressive strength. Additionally, it reduces structural failures by dispersing mechanical stress, which can be beneficial in preventing sudden fracture during in vivo degradation, enabling the potential use in bone regeneration.

The degree of contact angle measurements of TGS, TGS/SIM, and TGS/SIM/Th are displayed in Fig. 2A. The findings indicate that, compared with the TGS-coated surface and the TGS/SIM dual-coated surface, the TGS/SIM/Th three-layered coated surface showed slightly enhanced hydrophilicity. The wetting angles of the biocomposites TGS and TGS/SIM were 76.2° and 74.5°, respectively. The TGS/SIM/Th coating has a comparatively hydrophilic surface, as evidenced by the measured contact angle of 72.6°. However, all three biocomposites showed angles below 90°; thus, the composite surfaces had high water affinity, which is advantageous for early biological processes, including cell adhesion and protein adsorption.

Fig. 2.

Physical characterization. A. Contact angle, B. Swelling rate of the scaffolds, C. Degradation behavior, D. Evaluation of pH change in PBS, E. Cumulative release of SIM, F. Thrombin release rate, G. SEM observation of the surface of TGS/SIM/Th sample in SBF after days 7 and 28, H-I. EDS mapping analysis after 28 days in TGS/SIM/Th, J. Schematic representation of the assumed degradation process, K. Apatite coverage % in SBF, L. Residual mass %, M. pH change. Statistical significance was calculated using Two-way ANOVA, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the TGS group at each respective time point.

PBS uptake behavior as presented in Fig. 2B, followed by increasing weights after 1, 3, and 5 days. On day 5, the swelling rate was the highest in TGS, TGS/SIM, and TGS/SIM/Th. The increased swelling for TGS/SIM/Th was due to the large number of hydrophilic groups, such as the amine and hydroxyl functional groups. Meanwhile, from day 7 to day 14, the weights of the groups were reduced. To improve the tissue regeneration process, the swelling capacity is a key factor and is responsible for the polymeric matrix's expansion, which results in an increase in pore size and facilitates both cell attachment and internalization. Additionally, the wide pores help nutrients and waste to spread throughout the structure. Conversely, persistent swelling may result in compressive stress and a loss of mechanical integrity in the surrounding tissue [63].

Following a 14-day immersion in PBS, residual mass (%) was monitored to assess the scaffold SIM degradation behavior (Fig. 2C). Over time, residual mass gradually decreased in all groups, indicating progressive degradation. With values of 81.5% and 70.0% after 1 and 2 weeks, respectively, TGS displayed the lowest residual mass across the groups, suggesting quicker degradation. TGS/SIM and TGS/SIM/Th, on the other hand, showed greater residual mass retention. In particular, at 1 and 2 weeks, TGS/SIM exhibited 88.8% and 82.3% residual mass, respectively, whereas TGS/SIM/Th maintained 89.9% and 85.1%. These findings imply that adding SIM improves the scaffold's structural stability and matrix integrity, while adding Th is mostly surface-associated and has minimal impact on the overall degrading behavior.

Fig. 2D illustrates different pH trends of the hemostatic bone system over 28 days of immersion in PBS. Eventually, the pH of the incubation medium becomes alkaline. After 7 days, the pH of TGS plateaued approximately at 8.04, while the addition of SIM and SIM/Th slightly reduced the alkalinity, to pH of 7.90 and 7.82, respectively. Over time, the scaffolds eventually achieved equilibrium, and the rate of release of OH⁻ ions decreased. Owing to the stabilization of OH⁻ ion release from the scaffolds and the buffering ability of the medium, the pH of all samples progressively dropped by day 14 in TGS (7.85), TGS/SIM (7.77), and TGS/SIM/Th (7.71). The pH range of 7.2-7.4 in TGS/SIM/Th is favorable for the growth of tissue engineering cells.

Fig. 2E presents the curve of the cumulative release rate of SIM measured by UV spectroscopy. The graph shows that there was no large initial burst release from the scaffold. The system followed a sustained release profile, and the SIM release behavior tended toward reaching equilibrium at 28 days. Approximately 78% of the total concentration was released after day 28.

In this study, Th was loaded in the TGS/SIM/Th system. Fig. 2F shows that after 60 min of incubation, approximately 97.2% of Th was released, demonstrating the burst release behavior characteristic of surface-adsorbed proteins. This suggests a rapid and effective hemostatic reaction at the wound site.

Osteoblast proliferation, differentiation, and the process of bone formation were all positively impacted by the local increase in ionic concentration brought about by the bio-composites. In vitro bioactivity and potential for bone regeneration are shown through their interaction with simulated body fluid (SBF). SEM images, as presented in Fig. 2G, show the degradation behavior on days 7 and 28 after immersion in SBF solution (pH 7.4). The agglomerated broccoli-shaped and spike-shaped white particles on the surface of TGS/SIM/Th are an indication of apatite formation. The biocomposite pore walls developed rapid mineralization, which resulted in particles being deposited on the surface pores and an expansion of the bio-ceramic grains. Owing to β-TCP dissolution and Ca²⁺ chelation by PDA, immersion in SBF produced a surface apatite layer, indicating the scaffold's bioactive potential and its capacity to support bone-like mineral formation. TGS and TGS/SIM SEM images are shown in Fig. S7. EDS elemental analysis showed the presence of O, P, and Ca and supported the apatite formation, as shown in Fig. 2H and I. A schematic representation of SBF ions on the scaffold is shown in Fig. 2J.

Fig. 2K illustrates the apatite coverage % of the composites. After day 7, the average coverage was 15.49%, 32.22%, and 43.60%, which increased to 23.56%, 44.07%, and 75.91% in TGS, TGS/SIM, and TGS/SIM/Th, respectively. The use of β-TCP exhibits good osteoconductivity, high mechanical rigidity, biocompatibility, and biodegradability [64]. Following stimulation of bone formation, the resorbable properties of β-TCP during bone regeneration allow it to be completely replaced by fresh bone tissue [65].

Fig. 2L shows the degradation behavior after 28 days of incubation. TGS exhibited the lowest residual mass (69.69%), whereas TGS/SIM and TGS/SIM/Th scaffolds showed enhanced structural stability, 75.96% and 79.78%, respectively, which suggests that surface modification successfully decreased matrix dissolution. The pH variation over 28 days is shown in Fig. 2M. A gradual increase in pH was observed after day 7, reaching 8.21, 8.10, and 7.99 in TGS, TGS/SIM, and TGS/SIM/Th, respectively. This continuous rise in pH may indicate ongoing ionic interactions between the scaffold and the surrounding medium. After 28 days, the pH of all of the scaffolds stabilized and shifted to the neutral range (7.7). Materials containing β-TCP release calcium (Ca²⁺) and phosphate (PO₄³⁻) ions, which promote the formation of apatite [66]. During the first phase, the interaction of phosphate ions with the surrounding media causes the pH to slightly rise. By controlling ion release and having a slight buffering effect, PDA performs a supplementary role in preventing drastic pH changes. Because of the low concentration, SIM contributes very little to pH change and is not anticipated to have a major impact on the overall ionic balance. Similarly, Th has little effect on bulk pH behavior since it is mostly found at the surface. Compared to the surface-modified groups, TGS exhibits a comparatively higher pH increase due to its increased degradation rate, which permits faster ion release. A dynamic balance between ion release and reprecipitation activities, such as apatite production, is responsible for the pH gradually stabilizing toward near-neutral levels over time. Bone regeneration and biocompatibility are supported by this regulated pH environment [67].

3.2. In vitro cell viability, cell proliferation, and hemocompatibility of hemostatic bone grafts

One fundamental biological test for biomaterials is the evaluation of in vitro cytocompatibility. The initial indicator of the biocompatibility of a material extraction is cell growth, followed by cell adhesion to the interface between biomaterials. In this study, the samples were seeded with preosteoblast cells (MC3T3-E1) to evaluate their cytocompatibility. Confocal microscopy imaging was employed to monitor cellular proliferation, displaying the expressions of Hoechst (blue) and F-actin (green) in the seeded cells after 1, 3, and 7 days of incubation, as illustrated in Fig. 3A. The findings demonstrated that cells were joined on all scaffolds during the culture periods and that the number of growing cells increased over time. Initially, there were no noticeable differences in cell viability between the samples. However, on day 7, the TGS/SIM and TGS/SIM/Th samples showed the highest cell proliferation and increased propagation compared with the control group and TGS. Previous research showed that statins increase MC3T3-E1 cell proliferation at lower dosages while causing cytotoxicity at higher ones [68]. Because cholesterol is essential for maintaining cell membranes, concentrated statins have a lethal impact as they significantly lower cholesterol levels [69]. Additionally, TCP and gelatin promote superior cellular migration and adhesion [63]. Implementing the MTT assay, cellular proliferation and viability were assessed for 7 days. After days 1 and 3, the cell viability of the samples did not differ noticeably, as shown in Fig. 3B. On day 7, PDA and SIM improved cell viability in TGS/SIM; however, when SIM and Th were combined in TGS/SIM/Th, the overall impact resulted in a significant increase in biocompatibility, suggesting that these substrates are nontoxic [70]. Based on MTT data, the released SIM fell within the range needed to promote cell viability and proliferation without impairing cell behavior. TGS did not demonstrate any toxicity to the cells; however, it demonstrated less cell proliferation than TGS/SIM and TGS/SIM/Th. The composite coating led to increased cell attachment [71] and characteristic elongated-spindle and osteoblastic-like morphology, demonstrating the pro-osteogenic potential of SIM on cell proliferation [72,73]. The f-actin area percentage of the composites in Fig. 3C. Cells covered almost 75% of the surface area in TGS/SIM/Th, the highest percentage of any group. Nevertheless, the percentages in control, TGS, and TGS/SIM are 55%, 57%, and 71%, respectively.

Fig. 3.

In vitro biocompatibility of TGS, TGS/SIM, and TGS/SIM/Th biocomposites with MC3T3-E1 cells compared with the control. A. Fluorescence images of cell proliferation after 1, 3, and 7 days. B. OD value of the MTT assay. C. Percentage of F-actin area. D. Live (green) and dead (red) staining assay images after 24 h. E. Rates of cell viability for different scaffolds after 24 h. F. Approximate number of live and dead cells. G. Representative images of the cell migration assay performed by the scratch test. H. Quantitative analysis of the rates of wound area closure. I. Schematic illustration of the blood coagulation mechanism and multiple platelet shapes. J. Representative SEM images of in vitro platelet adhesion on the samples (scale bar = 1 μm). K. Total number of platelets adhering to the biocomposites (per 10³ μm²). L. Hemolysis ratio. M. Blood clotting index. Statistical significance was calculated using Two-way ANOVA for Fig. 3B & H, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the control group at each respective time point. We used Tukey's post-hoc test for pairwise comparisons following the One-way ANOVA for Fig. 3C–F & 3K-3L.

The live/dead staining, as shown in Fig. 3D, presents the cell survival rates of the control, TGS, TGS/SIM, and TGS/SIM/Th groups within 24 h. This assay relies on the enhanced permeability of dead and dying cells and the esterase activity of living cells. Specifically, by allowing living cells to liberate green fluorescent calcein from nonfluorescent cell-permeant calcein-AM, the test detects living cells based on their esterase activity. The substance ethidium homodimer, which enters cells with broken membranes and exhibits a substantial increase in red fluorescence when bound to nucleic acids, is used to identify dead and dying cells [74]. Additionally, the cell viability ratio was >86 % after 24 h in all scaffolds in Fig. 3E, suggesting that these composites are biocompatible [75]. The results of the live/dead staining revealed a gradual increase in the number of living cells among all scaffold types, suggesting that there were comparatively few dead cells present and that the cell development was favorable over time. Increased cell viability indicates that the composites can promote postoperative wound repair by enhancing cell adhesion and proliferation and functioning as a tissue-engineered scaffold. Fig. 3F presents the numbers of living and dead cells. The results showed that, the maximum number of live cells was seen for TGS/SIM/Th (1897 ± 7). Meanwhile, the other groups, TGS/SIM (1836 ± 42) and TGS (1657 ± 50), showed more living cells than the control (1643 ± 36). In contrast, the number of dead cells was much higher in the control (156 ± 44) than in TGS (113 ± 20), TGS/SIM (110 ± 23), and TGS/SIM/Th (85 ± 17).

Cell migration is an integral part of cell biology that controls morphogenesis and inflammation. A pipette tip was used to create a tiny scratch to create a cell-free zone for the evaluation of the cell migration. The entire scratch was imaged at 0 h for all samples and then periodically examined using optical microscopy. Next, the images were analyzed to determine the extent to which the scratch area had closed over time. (Fig. 3G). The greatest number of migrating cells was visible in TGS/SIM/Th compared with the control. Bone scaffolds should promote cell migration, which improves wound healing, besides their hemostatic properties. The size of the open wounds, as shown in Fig. 3H, demonstrated that the wound area significantly shrank after a day. This is because, after polarization, cells close to the wound edge migrated into the wound space. The scratch closed faster in treated wells than in controls, indicating that the samples promoted MC3T3-E1 migration. Overall, bio-implantable polymers facilitate natural biological processes, such as tissue regeneration, cell adhesion, migration, and proliferation, which are all essential for effective healing.

In vitro, hemostatic qualities refer to the capacity of a material or system to promote blood coagulation and prevent excessive bleeding. Fig. 3I briefly represents the common blood coagulation pathway that activates a complex series of coagulation factors, ensuring efficient clot formation while preventing unwanted thrombosis, and illustrates different platelet shapes. Here, the hemostatic properties were evaluated by analyzing SEM images of platelet adhesion. SEM images demonstrate that the surface of the TGS/SIM/Th activated the greatest number of platelets when compared with TGS and TGS/SIM in Fig. 3J. Inactivated platelets are usually spherical in shape, and upon activation, they form pseudopodia [76]. The number of platelets adhering to the surface is shown in Fig. 3K. Overall, both SIM and Th-loaded grafts, TGS/SIM (∗p < 0.05) and TGS/SIM/Th (∗∗∗∗p < 0.0001), had substantially more platelets adhered to their surfaces than was the case for TG/S. The total number of attached platelets per 10³ μm² on TGS, TGS/SIM, and TG/SIM/Th was 77 ± 6.1, 99 ± 8, and 138 ± 7.5, respectively. Platelet adhesion initiates the coagulation cascade, creating a stable blood clot. Rapid hemostasis is facilitated by regulated platelet adhesion in biomaterials used in wound care applications.

Fig. 3L describes that the hemolysis rates (HRs) of TGS, TGS/SIM, and TGS/SIM/Th were 3.62%, 3.43%, and 2.72%, respectively. The hemolysis rate indicates how much hemostatic substances damage red blood cells (RBCs) and is the key indicator of how well the substance works with blood. A high HR means the material promotes red blood cell rupture, releasing hemoglobin into the plasma, which can lead to toxicity, inflammation, or thrombosis. According to ISO 10993-4, a hemolysis ratio of <5% is considered to reflect blood-compatibility [77].

Whole blood clotting measurement was performed to assess the blood coagulation effect of TGS/SIM/Th compared with those of TGS/SIM and TGS in order to further explore the hemostatic properties of the scaffolds. As shown in Fig. 3M, the BCI indexes of TGS/SIM/Th and TGS/SIM were 3.70% and 4.76%, respectively, whereas that for TGS was 10.1%. Anticoagulant materials retain higher absorbance, while procoagulant materials often show lower absorbance values due to clot formation and decreased availability of soluble reaction products. Therefore, the reduced absorbance found in this investigation suggests that the biomaterials have procoagulant action.

Fig. S8 shows the fibrin polymerization behavior of the scaffolds. The assay's validity was confirmed by the positive control, PC = (fibrinogen + thrombin + CaCl₂), which demonstrated quick and visible clot formation, whereas the negative control, NC = (fibrinogen + PBS), showed no clotting. TGS exhibited weak and loosely formed clots within the scaffold groups, suggesting a limited capacity for intrinsic hemostasis. Due to surface alteration, the TGS/SIM scaffold showed better clot formation, indicating increased interaction with fibrinogen. Notably, similar to the positive control, the TGS/SIM/Th scaffold showed rapid, thick, well-formed fibrin clots. This suggests that the thrombin integrated into the scaffold maintains its bioactivity and efficiently catalyzes the production of fibrin. The scaffold + PBS groups, on the other hand, showed no clot formation, indicating that the effects were solely caused by fibrin polymerization rather than scaffold-induced interference. These results show that adding thrombin considerably improves the scaffold's hemostatic efficacy by encouraging the quick development of fibrin networks.

3.3. In vitro osteogenic capability

We examined the impact of composite scaffolds on osteogenic differentiation in vitro. Through alkaline phosphatase (ALP) antibody staining, calcium deposition, and the expression of genes and proteins linked to osteogenesis, the potential of MC3T3-E1 cells was investigated in the sample extracts. Following 7 and 14 days of co-culture, fluorescence images showed the ALP activity, as illustrated in Fig. 4A. The images indicated that TGS/SIM and TGS/SIM/Th had the highest osteogenic activity, whereas the positive control exhibited higher activity only from TGS. ALP expression increases during the early phases of osteogenic differentiation and plays a crucial role in matrix mineralization.

Fig. 4.

Evaluation of the in vitro osteogenic differentiation potential of different scaffolds on MC3T3-E1 cells. A. Representative fluorescence images of alkaline phosphatase antibody staining at days 7 and 14 (scale bar = 100 μm), B. Relative rates of ALP area coverage after 1 and 2 weeks, C. Representative microscopic images of alizarin red staining after 14 and 21 days of cell culture (scale bar = 100 μm), D. Absorbance value of alizarin red activity, E-G. Quantitative analysis of RUNX2, COL1, and OPN relative mRNA expression with the conditioned medium on days 7 and 14. Statistical significance was calculated using Two-way ANOVA, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the Control (Fig. 4B–D) and Non osteogenic (Fig. 4E–G) group at each respective time point.

Fig. 4B displays the relative area rates of ALP activity for 7 and 14 days, which were determined by using ImageJ to analyze the images. While there was no statistically significant difference (ns) between TGS and Control, TGS/SIM showed a slight increase, and TGS/SIM further increased. These findings suggest that adding SIM greatly promotes osteogenic differentiation, resulting in the TGS/SIM/Th group having the highest ALP activity. According to previous in vitro research, statins SIM have been reported to stimulate osteoblast development and mineralization, as evidenced by higher osteogenic markers. Since SIM increased BMP-2 expression in osteoblast-like MC3T3-E1 cells, it is further hypothesized that an up-regulation of BMP-2 may contribute to these effects [78]. The increased ALP activity seen in the SIM-containing groups in this study implies that SIM stimulates early-stage osteogenic differentiation.

Alizarin red staining (ARS) is frequently used to analyze calcium deposition on biomaterials in vitro and serves as a late osteogenic marker. The MC3T3-E1 cells grown in the sample extracted osteogenic medium were exposed to ARS after co-culturing for 14 and 21 days to observe calcium deposition. The stained samples were captured using an optical microscope, as shown in Fig. 4C. Interestingly, when compared to the positive control group, the samples showed no visible alterations after 14 days. However, TGS/SIM and TGS/SIM/Th produced more ARS-calcium complex (shown by red dots) than the control after 21 days of incubation. Greater calcium deposition is reflected in more intense staining, suggesting improved support for the creation of bone tissue. Fig. 4D exhibits that no significant optical density on day 14, but there were higher OD values in TGS/SIM (2.074 ± 0.046) and TGS/SIM/Th (2.089 ± 0.068) than in TGS (1.881 ± 0.146) and the control (1.763 ± 0.142) on day 21. Overall findings indicated that the spray-dried gelatin-based SIM and Th-loaded bio-functionalized TGS/SIM/Th scaffold had a greater ability to promote the development and nucleation of minerals, resulting in osteoinductive capability in MC3T3-E1 cells.

RT-PCR was used to analyze the mRNAs responsible for particular protein expression quantitatively. Following 7 and 14 days of culture, the levels of RUNX2, COL1, and OPN mRNA expression were investigated. All of the scaffold control, positive control, TGS, TGS/SIM, and TGS/SIM/Th samples showed trends of increased expression for both RUNX2 and COL1 markers at 7 and 14 days, respectively, as shown in Fig. 4E and F. OPN showed significant expression when compared with the control group at both 7 days and 14 days, as shown in Fig. 4G. TGS/SIM and TGS/SIM/Th scaffold groups showed higher levels of all three mRNA expressions than control groups. RUNX2 and COL1 play essential roles in osteogenic differentiation and bone formation and are considered to be early-stage bone expression markers. RUNX2, a potent transcription activator specific to osteoblasts, can increase the expression of genes specific to osteoblasts [79]. OPN is a noncollagenous protein that promotes collagen attachment to the bone's inorganic fraction. By controlling the development of hydroxyapatite crystals inside the bone matrix, OPN is known to play a role in the adhesion of cells to the extracellular matrix [21]. Soy derivatives can be considered as alternative agents for treating metabolic bone diseases such as osteoporosis because these bioactive peptides effectively improve bone regeneration performance and upregulate the expression of osteoblast differentiation markers (e.g., Runx2). Another explanation is that MC3T3-E1 cell differentiation is facilitated by the continuous release of calcium ions from the TGS/SIM/Th scaffolds, which increases mRNA translation by activating calcium ion-sensing receptor signaling [24,80]. The osteogenesis induced by a biomaterial is one of the most important factors in applications involving bone regeneration. The best bone implants must be biocompatible and have outstanding osteogenic activity.

3.4. In vivo hemostatic properties, micro-CT imaging, and histological analysis

The in vivo hemostatic performance was assessed by measuring how long it took for the bleeding to stop in the critical-sized SD rat calvarial defect model and tail amputation model (Fig. 5A). The bleeding times in the SD rat calvarial defect model for TGS, TGS/SIM, and TGS/SIM/Th were 69.33s, 52.66s, and 37s, respectively, while for the blank control, the time was 140.3s, as shown in Fig. 5B. The amount of blood loss was calculated by using a previously weighted gauge. Fig. 5C illustrates blood loss in the SD rat calvarial defect model for TGS (1.61g), TGS/SIM (1.11g), and TGS/SIM/Th (0.95g, ∗∗∗∗p < 0.0001), when compared with the control value of 2.25g. Moreover, Fig. 5D presents the hemostatic time for TGS (103.66s, ∗∗∗∗p < 0.0001), TGS/SIM (92s, ∗∗∗∗p < 0.0001), and TGS/SIM/Th (73s, ∗∗∗∗p < 0.0001), which were significantly different compared with the control time of 213s in the SD rat tail amputation model. Furthermore, the blood loss decreased for TGS (1.36g), TGS/SIM (1.35g), and TG/SIM/Th (1.16g, ∗∗∗∗p < 0.0001) relative to that in the control 2.79g, as shown in Fig. 5E. For any hemostatic agent, it must encourage rapid blood clotting in the early stages of injury. The main mechanism was the initial hemostatic phase, which set off further blood coagulation cascades that transformed fibrinogen into insoluble fibrin. A blood clot was formed by the union of platelets and fibrin threads.

Fig. 5.

In vivo hemostatic property of the scaffold A. Digital images of bleeding and hemostasis in the SD rat calvarial defect model and tail amputation models after applying hemostatic bone materials. B. In vivo blood clotting time, C. Blood loss in the SD rat calvarial bone defect. D. Hemostatic time and E. quantitative analysis of blood loss from the bleeding SD rat tail with different prepared scaffolds. F. Representative photos of extracted samples from the SD rat calvarial defect model. G-H. Quantitative analysis of microstructural parameters of regenerated bone tissues (BV/TV, Tb. Sp). I-J. 3D reconstructed micro-CT images of sectioned tissue samples after 1 and 2 months of implantation. K. H & E staining of cross-sectioned low (1.2x) and high magnification (40.0x) optical microscopy images, Scale bars = 100 & 10 μm, HB = host bone, NB = new bone, FT = fibrous tissue, S = remaining scaffold. We used Tukey's post-hoc test for pairwise comparisons following the One-way ANOVA for Fig. 5B–E. Statistical significance was calculated using Two-way ANOVA for Fig. 5G and H, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the control group at each respective time point.

For 4 and 8 weeks, in vivo tests were conducted using the SD rat calvarial model to further investigate the biodegradation and efficiency of inducing new bone formation of TGS/SIM/Th scaffolds. The critical-sized SD rat calvarial defect model is frequently used to study the bone-healing potential of biomaterials, particularly for defects that require treatment to heal. Fig. 5F presents photographs of the defect site and scaffold extraction of the hemostatic bone grafts during in vivo operation.

Histomorphometric analysis of the systems is shown in Fig. 5G and H. Fig. 5G displays the findings of the bone volume fraction percentage for the control, TGS, and TGS/SIM/Th. The percentages of BV/TV for TGS/SIM/Th were 26.03% (∗∗∗∗p < 0.0001) and 51.74% (∗∗∗∗p < 0.0001), while for TGS they were 13.37% and 31.96% respectively, and both appeared to be higher than that of the control, both 1 and 2 months after implantation. A high BV/TV ratio is suggestive of active remodeling or the development of early-phase healing. The lower values of trabecular separation (Tb. Sp) at 8 weeks postimplantation confirmed the effectiveness of new bone tissue formation in the TGS (∗∗∗p < 0.001) and TGS/SIM/Th (∗∗∗p < 0.001) hybrid groups, as shown in Fig. 5H. Lower trabecular spacing suggests early bridging inside the gap.

Interestingly, compared to the control group, structural parameters such as Tb. N (mm⁻¹) increased considerably in the TGS (∗∗p < 0.01) and TGS/SIM/Th (∗∗∗p < 0.001). Additionally, there was an increase in Tb. Th (mm) in TGS (∗p < 0.05), and TGS/SIM/Th (∗∗p < 0.01) than control (Fig. S9). In the early stages of bone regeneration, freshly formed, immature trabeculae are usually thinner, and this indicates an active remodeling phase. The most noticeable effect was seen in the TGS/SIM/Th group, suggesting that Th-mediated fibrin production speeds up early osteogenesis and creates a more dynamic remodeling environment. These results demonstrate how Th and SIM might work together to promote quick but structurally dynamic bone repair.

Reconstructed coronal and sagittal three-dimensional (3-D) micro-CT images of the defect zones at 4 and 8 weeks post-implantation are presented in Fig. 5I and J. In this figure, white, red, green, and black represent native bone, new bone tissue, implanted scaffolds, and background, respectively. Multiple criteria based on contrast differences in micro-CT images were used to identify new bone development inside each scaffold and at neighboring gaps between the cylindrical scaffolds. The red region (new bone) in the images demonstrated that, at 1 and 2 months after implantation, TGS/SIM/Th began to produce new bone, along with degradation of the implanted scaffolds. In contrast, in the control, there was no significant bone formation. The gradual TCP breakdown at the TGS/SIM/Th scaffold's surface likely creates continuous ionic interactions that promote mineral deposition without compromising the scaffold's structural integrity. A dynamic balance between ion releases occurs because of the apatite production, which is aided by the release of calcium (Ca²⁺) and phosphate (PO4³⁻) ions from materials containing β-TCP. With the linked porosity and advantageous calcium-phosphate surface chemistry, β-TCP offers an osteoconductive framework that promotes cell adhesion, vascular infiltration, and guided bone tissue ingrowth. Instead of accelerating scaffold degradation, SIM activated Bone morphogenetic protein (BMP2), which is responsible for the majority of bone's osteoinductive capacity [11,81,82]. Th-mediated fibrin network significantly functions as a transient extracellular matrix at the defect site and may also encourage cell adhesion. Furthermore, early-stage biological responses, including hemostasis and osteogenic signaling, may be further enhanced by the presence of Th and SIM, which would work in alignment to increase bone regeneration. The increased trabecular quantity at early time points and increasing thickening shown in micro-CT analysis are consistent with this combined impact, which speeds up both the development of immature trabeculae and their later maturity. These results demonstrate how osteoinduction and enhanced biological integration work together to promote superior bone regeneration in the TGS/SIM/Th group.

The porous structure of scaffolds is essential for the ingrowth of bone tissue. Post-implantation histological evaluation was conducted using Hematoxylin and Eosin (H&E) staining to confirm the newly formed bone and to analyze the tissue response in the defect area. Fig. 5K shows the cross-sectioned H & E images of the interface and inner view of the control, TGS, and TGS/SIM/Th scaffolds. Some scaffold fragments were discovered inside the scaffolds, and they began to partially resorb at 1 month, whereas at 2 months, the resorption rate improved alongside the quantity of scaffold pieces. After 1 month in TGS/SIM/Th, bone formation started at the edge, and after 2 months, it extended to the center of the defect zone, and the remaining portions of the scaffold were filled with vascularized soft connective tissue. SIM and Th-coated TGS/SIM, and TGS/SIM/Th demonstrated good osteointegration since the sample did not exhibit interface loosening or inflammation. Tiny clusters of osteoid were also seen in addition to peripheral extension within the pores of the scaffold. However, in TGS, bone formation was not as significant as the majority of the osteoid was still immature. Although some bone growth was visible on the edges, the sample had a large number of inflammatory and fibroblast-like cells. A small amount of woven bone was peripherally expanded in the blank control at 8 weeks. The margins of the recently deposited bone masses had active osteoblasts attached to them. The thickness of the extended new bone from the scaffolds' periphery inside the defect was greater than that of the control after 2 months. An excessive number of inflammatory cells encircled by fibrous tissue prevented the proper development of bones in the control group. According to the H&E images, the TGS/SIM group did not exhibit any unusual effects and promoted bone regeneration provided in Fig. S10. Newly generated tissue, mostly from the defect borders, fills a portion of the defect site. There are still gaps and a less dense matrix in the core area, indicating incomplete bridging. Early-stage tissue infiltration, which indicates active but not yet continuous bone development, is highlighted by the marked area. At one month, compared to TGS/SIM, TGS/SIM/Th showed more woven bone growth with greater cellularity and less fibrous tissue, suggesting early matrix deposition and rapid osteoblast development. By two months, the TGS/SIM/Th group had low remaining fibrous tissue and well-organized lamellar bone, while the TGS/SIM group had more immature bone architecture. Since the TGS/SIM group demonstrated bone repair effects in our initial study. We did not further evaluate the SIM group due to limitations in animal availability and ethical standards aimed at reducing animal use. Here, the TGS group was compared using immunofluorescence (IF) staining and micro-CT.

SIM by itself encourages osteoblast development, but adding Th seems to improve early cellular recruitment and microenvironmental stability via forming fibrin. Early-stage interconnected pore-like structures support the scaffold's function in directing tissue ingrowth, and their eventual replacement by mineralized matrix validates successful scaffold-assisted bone regeneration in TGS/SIM/Th system.

The relationship between scaffold degradation and bone regeneration is complex and does not necessarily require rapid material resorption at early stages. In the present study, the TGS/SIM/Th scaffold exhibited a relatively slow degradation profile (as shown in Fig. 2C and L) during the initial phase, maintaining structural integrity while supporting cellular infiltration and tissue stabilization. Despite this, significant osteogenic activity and new bone formation were observed, which can be attributed to the surface bioactivity of the scaffold and controlled release of therapeutic agents, rather than bulk degradation alone.

3.5. Immunofluorescence analysis of implanted bone defects

Bone tissue samples were immunohistochemically stained to detect markers of structural protein collagen type 1 (COL1) and bone growth (osteopontin, bone morphogenic protein, and osteocalcin) at the defect. OPN and COL1 production are essential for the repair of bones, especially during the mineralization and remodeling stages. Fig. 6A–C shows fluorescence images of the control, TGS, and TGS/SIM/Th samples at 1-2 months. Fig. 6D–F demonstrates the expression levels of the aforementioned osteogenic markers in the control, TGS, and TG/SIM/Th groups. Among these groups, TGS/SIM/Th exhibited higher expression of OPN, COL1, and CD34. Significantly weaker signals were detected in the control, while the highest expression levels were found in the TGS/SIM/Th hybrid composite. The well-known procoagulant serine protease Th catalyzes the transformation of fibrinogen into fibrin, which starts and maintains the formation of clots. At the defect site, this Th-mediated fibrin network notably acts as a temporary extracellular matrix, which can aid in cell attachment [83], support the local microenvironment and promote the regenerative responses [[84], [85], [86]], and may be linked to the increased marker expression seen in this investigation. Therefore, the sustained release of SIM and the procoagulant effects of Th are possible reasons for the elevated marker expression.

Fig. 6.

Evaluation of in vivo defect repair after sample implantation through immunofluorescence staining. Fluorescence images of A. OPN (red), B. COL1 (red), and C. CD34 (red). D-F Quantitative analysis of positively stained coverage areas using ImageJ software (scale bar = 100 μm). Statistical significance was calculated using Two-way ANOVA, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the Control group at each respective time point.

OPN increases osteoclast attachment and woven bone resorption. During the early stage of bone remodeling, the COL1 protein offers structural support for mineral deposition. The amount of collagen expression was higher in TGS and TGS/SIM/Th than in the control. Collagen deposition plays a crucial role in bone remodeling and has a favorable link with bone healing because collagen is the primary organic component of the bone matrix [87]. The peripheral region increased and contributed to the display of sparse bone-like structures in the central view of the composites. The gradual breakdown of the core structure of spray-dried β-TCP and gelatin, as well as the SIM and Th-coated layers, favored bone growth and enhanced osteogenesis in the defect zone. According to our current understanding of bone healing, osteoclasts stimulate osteoblasts and are key players in bone repair. In the absence of osteoclastic action, osteogenesis may result in unwanted bone formation. Furthermore, because angiogenesis and osteogenesis function in combination, angiogenesis plays a critical role in bone repair [88].

Fig. 7A and B illustrates fluorescence images of BMP-2 and OCN markers. TGS/SIM/Th (∗∗∗∗p < 0.0001) followed the increased rate after the implantation period in both Fig. 7C and D. The expression of BMP-2 stimulates osteoinductive factors that promote osteoblast differentiation, whereas OCN contributes to stabilizing the bone matrix by binding to hydroxyapatite [89]. The interconnected microarchitecture of the TGS/SIM/Th scaffold promotes cell proliferation and new vessel formation at sites of sample degradation, which contributes to increased bone formation. Additionally, the breakdown of β-TCP releases calcium and phosphate, which may support the biological activity of osteoblasts and accelerate bone development.

Fig. 7.

Representative fluorescence images of A. BMP-2 (red), and B. OCN (red). C-D. Quantitative analysis of positively stained coverage areas using ImageJ software (scale bar = 100 μm). Statistical significance was calculated using Two-way ANOVA, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the Control group at each respective time point.

Fig. 8A and B shows the results for CD68 (M1) and CD206 (M2) markers, respectively. Both TGS (∗∗p < 0.01, ∗∗∗∗p < 0.0001) and TGS/SIM/Th (∗∗p < 0.01, ∗∗∗∗p < 0.0001) showed significant expression compared to control, both in Fig. 8C and D, respectively, after 1 and 2 months. Fig. 8A shows the expression of CD68 (commonly associated with early inflammatory response M1) in control, TGS, and TGS/SIM/Th groups after 1 and 2 months of implantation. It is clearly observed that CD68 (green) expression in the control group is very high compared to the other groups after 1 month. However, after 2 months, CD68 expression had declined in the control group, but the TGS and TGS/SIM/Th groups showed a more pronounced reduction, indicating better inflammation control. Notably, more CD206 (commonly associated with pro-regenerative response, M2) positive cells (green) were observed in TGS and TGS/SIM/Th groups in both 1 and 2 months timepoints (Fig. 8B). After 2 months, the TGS/SIM/Th exhibited a strong expression of CD206-positive cells. This indicates enhanced M2 polarization and tissue remodeling in the TGS/SIM/Th groups. Fig. 8C and D demonstrates the quantified relative coverage of the CD68 and CD206, respectively, at 1 and 2 months. The relative coverage of CD68 was significantly lower in both TGS (p < 0.01 at 1 and 2 months) and TGS/SIM/Th (∗∗∗∗p < 0.0001 at 1 month, ∗∗p < 0.01 at 2 months) groups. Contrastingly, the relative coverage of CD206 was significantly higher in both TGS (∗∗∗p < 0.001 at 1 and 2 months) and TGS/SIM/Th (∗∗∗∗p < 0.0001 at 1 and 2 months) groups. Overall, the presence of gelatin and soy protein in both the TGS and TGS/SIM/Th groups appears to facilitate the transition from M1 to M2 macrophage phenotypes. Notably, the SIM-loaded group exhibited significantly higher M2 expression, consistent with previous reports [90]. By secreting a variety of chemicals, they are primarily responsible for controlling the inflammatory response and preserving immunological balance. Furthermore, macrophages play a crucial role in attracting bone mesenchymal stem cells (BMSCs) to the fracture site, where they support development. The interaction between MSCs and macrophages is important for the complex processes of inflammation and osteogenesis, which are necessary for bone healing. This cooperation is vital because macrophages not only use signaling to initiate bone regeneration, but also moderate the surrounding environment to ensure continuous healing. Studies have shown that different macrophage phenotypes have a significant impact on the healing process. In particular, M2 macrophages release pro-regenerative and anti-inflammatory cytokines (IL-10, BMP-2, and VEGF) that affect the rate and quality of tissue regeneration through several coordinated processes, including promoting osteogenic differentiation, promoting cell migration and angiogenesis, and reducing inflammation [91,92], whereas M1 macrophages facilitate pro-inflammatory responses [93]. It is thus essential to maintain an appropriate balance between the different macrophage types in order to achieve effective bone regeneration [94,95]. These results suggest that proinflammatory macrophages polarized toward a pro-regenerative phenotype and infiltrated the graft wall.

Fig. 8.

Representative fluorescence images of A. CD68 (green), and B. CD206 (green). C-D. Quantitative analysis of positively stained coverage areas using ImageJ software (scale bar = 100 μm). E. Schematic illustration of the assumed healing process of the SD rat calvarial defect after TGS/SIM/Th implantation. Statistical significance was calculated using Two-way ANOVA, considering Material Composition (Groups) and Immersion Time as the two independent variables, followed by Tukey's post-hoc test for multiple comparisons. Significant differences are indicated as ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 relative to the Control group at each respective time point.

4. Conclusion

In the current study, we fabricated a multifunctional layered biocomposite scaffold combining spray-dried β-TCP granules, gelatin, and soy protein isolate matrix, polydopamine-mediated drug delivery, and an outer layer based on thrombin (TGS/SIM/Th) for hemostasis and bone regeneration. The scaffold structure provides an optimal microenvironment, so its biodegradability successfully achieved the sustained release of SIM for 28 days. SIM facilitated MC3T3-E1 cell proliferation, accelerated cell migration, and promoted the in vitro expression of osteogenic genes and proteins such as RUNX2, COL1, and OPN. In vivo implantation of TGS/SIM/Th scaffold into the rat calvarial bone defect model greatly promoted the development of new bone, as supported by the micro-CT and histological investigation. The system showed high osteoblast activity, examined using early (OPN, COL1) and late osteogenic markers (BMP-2, OCN), and enhanced angiogenesis, detected by using CD34. Taking the obtained findings together, this study introduced unique drug protein co-loaded biomaterials as hemostatic bone grafts with excellent mechanical strength, cell attachment, and blood coagulation properties. In addition to its encouraging in vitro and in vivo outcomes, the scaffold shows great promise for clinical translation in applications like craniofacial reconstruction, post-extraction socket preservation, and periodontal bone defect repair, where improved bone regeneration and quick bleeding control are vital. Overall, this study offers a viable alternative for complex bone defect care by bridging the gap between hemostatic function and bone tissue engineering with a flexible and clinically applicable biomaterial platform. Nonetheless, this research still has some limitations, which should be mentioned here. For example, osteogenesis may be affected by all of these surface modifications, for instance, if they impact the retention and release profiles of intrinsic bioactive chemicals. Future research will be required to investigate the relevant mechanisms and delivery systems. We believe that a more comprehensive investigation will help to overcome current obstacles and create new opportunities for using composite scaffolds for bone defects.

CRediT authorship contribution statement

Nusrat Jahan: Conceptualization, Data curation, Investigation, Methodology, Writing – original draft. Prayas Chakma Shanto: Data curation, Investigation. Myeongki Park: Data curation, Investigation. Md Abdullah Al Fahad: Data curation, Investigation. Hai-Doo Kim: Methodology. Byong-Taek Lee: Conceptualization, Formal analysis, Funding acquisition, Project administration, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

The authors do not have permission to share data.