Integrating BMSC-inlaid GelMA hydrogels and melatonin-loaded PLGA microspheres into silk fibroin/nanohydroxyapatite scaffolds to repair osteoporotic bone defects

Abstract

Background

Bone defects are commonly accompanied by disuse-induced osteoporosis, which in turn exacerbates the difficulty of bone regeneration. Biomimetic bone scaffolds represent one of the most promising strategies for bone defect repair.

Methods

In this study, we constructed poly(lactic-co-glycolic acid) (PLGA) microspheres with sustained release of melatonin (M-PLGA), an anti-osteoporosis hormone that simultaneously promotes osteogenesis and inhibits osteoclastogenesis. The microspheres were then integrated into a silk fibroin (SF)/nanohydroxyapatite (nHA) scaffold. To improve the cell inoculation rate and survival rate, a methacrylated gelatine (GelMA) hydrogel was laden with bone marrow mesenchymal stem cells (BMSCs, G-B). The gel acted as the cell carrier and was then inlaid into the M-PLGA/SF/nHA scaffold via negative pressure inhalation. The biomimetic scaffold was then photocrosslinked to yield the M-PLGA/SF/nHA/B-G scaffold. The microstructure, mechanical properties, pore diameters and sustained release of melatonin from the scaffold were evaluated. In vitro, the cytocompatibilities of the scaffolds were assessed using a CCK-8 assay, live/dead staining and phalloidin staining, and the osteogenesis-inducing abilities of the scaffolds were assessed by alkaline phosphatase, Alizarin Red S, and von Kossa staining, as well as immunohistochemical staining for collagen type 1. In vivo, a rat femoral distal bone defect was constructed and then immobilized by tail suspension to imitate disuse osteoporotic bone defects. Once the model was established, different scaffolds were implanted within the bone defect regions. Six weeks after implantation with different scaffolds, microcomputed tomography and histological staining were conducted to evaluate new bone formation within the bone defects in vivo.

Results

In vitro, live/dead staining revealed that almost all the cells inoculated into the three types of scaffolds remained viable. Specifically, on day seven, the number of living cells in the M-PLGA/SF/nHA/G-B group was 621.7 ± 8.1% greater than that in the SF/nHA group. Phalloidin staining revealed that all the scaffolds were suitable for BMSC attachment, among which the M-PLGA/SF/nHA/G-B scaffolds were most conducive to BMSC distribution. Alizarin Red S staining and Von Kossa staining revealed more mineralized nodules in the M-PLGA/SF/nHA/G-B scaffolds following osteogenic induction, and ALP staining and collagen type 1 immunohistochemistry analysis confirmed that the M-PLGA/SF/nHA/G-B scaffolds were more conducive to osteogenic differentiation. In vivo, analysis by micro-CT and histological staining showed more new bone formation in the osteoporotic bone defect in the M-PLGA/SF/nHA/G-B group than in the SF/nHA or M-PLGA/SF/nHA group six weeks after implantation.

Conclusions

The novel multifunctional composite scaffold holds great promise as a bone repair biomaterial for future clinical translation.

Introduction

Bone defects, which are frequently caused by trauma, tumour resection, infection, or congenital disorders, self-repair is usually difficult and consequently pose significant clinical and economic burdens [1]. Current clinical treatments include autografts, allografts, and artificial bone substitutes [2]. However, these treatments often necessitate prolonged limb immobilization and extended bed rest, which can lead to disuse osteoporosis (DOP) and further impair bone healing. Osteoporosis, in turn, increases the risk of fractures [3] and makes the healing process of fractures with bone defects more difficult [4, 5]. Treatment for osteoporosis represents a difficult therapeutic challenge in clinical [6, 7]. DOP is difficult to reverse, and pharmacological treatment remains the primary treatment approach for this condition [8]. Although autografts are considered the gold standard because of their excellent osteogenic properties, their application is limited by donor site scarcity and the need for additional surgery [9, 10]. Allografts carry risks of immune rejection and infection, whereas artificial grafts often lack adequate osteoconductive and osteoinductive properties [11]. Overall, these limitations hinder effective bone defect treatment in clinical practice, highlighting the need for innovative strategies to enhance bone regeneration and functional recovery.

Bone tissue engineering (BTE) is a promising strategy for bone defect repair and regeneration and relies on the interplay of four key elements: scaffolds, seed cells, growth factors, and a culture environment [12]. Among various biomaterial combinations, the silk fibroin (SF)/hydroxyapatite (HAp) scaffold has shown great promise for BTE and has been successfully fabricated using various methods [13]. SF, the primary component of natural silk, has been clinically applied in the field of biomaterials because of its excellent biocompatibility and low immunogenicity [14]. HAp is the primary inorganic component of bone tissue. Nanoscale hydroxyapatite (nHA) closely mimics the microstructure of natural bone and significantly increases the mechanical strength and fracture toughness of scaffolds [15]. Through methods such as electrospinning, 3D printing or lyophilization, SF and nHA can be combined to further prepare various forms of materials, such as porous scaffolds, membranes and hydrogels, which can be applied to repair bone defects in different locations with different requirements [16]. Li et al. prepared a silk fibroin/hydroxyapatite/tricalcium phosphate composite scaffold using 3D printing technology to simulate bony microenvironment structure and successfully repaired rabbit tibia bone defects [17]. Through a freeze-drying process, a bionic SF/nHA/chitosan scaffold that incorporated IL-37, which is antibacterial, and pamidronate, which suppresses bone resorption, was developed, and this scaffold could effectively treat infectious bone defects [18].

Building on the properties of scaffold materials, bioactive molecules such as melatonin have been introduced to increase osteogenesis. Melatonin is an amphiphilic neuroendocrine hormone secreted by the pineal gland that is known for its low toxicity, rapid metabolism, and wide range of physiological activities [19]. It promotes osteoblast differentiation and inhibits osteoclast formation. Jarrar et al. fabricated a chitosan/hydroxyapatite scaffold embedded with melatonin and BMP-2. Melatonin has dual functions in promoting bone formation in cooperation with BMP-2 and simultaneously alleviates osteoclast differentiation enhanced by BMP-2 [20]. Furthermore, sustained release of melatonin from a GelMA-dopamine composite hydrogel could effectively inhibit oxidative stress-induced osteoblast apoptosis and promote osteogenesis around a prosthesis, thereby attenuating implant loosening in osteoporotic patients [21]. Thus, melatonin exhibits significant antiosteoporotic and osteogenic effects.

However, its clinical application is hindered by low bioavailability and a short half-life [19]. Additionally, systemic administration requires high doses of melatonin, which may increase the risk of adverse effects. Poly(lactic-co-glycolic acid) (PLGA) microspheres have been widely recognized as effective sustained-release carriers for medical drugs and cytokines [22, 23]. They have also been employed as injectable delivery systems to promote the repair of bone defects. Zhang et al. successfully encapsulated melatonin within PLGA microspheres, enabling its sustained and controlled release, which significantly promoted the osteogenic differentiation of human mesenchymal stem cells (hMSCs) in vitro [24]. Chen et al. formulated melatonin into PLGA nanoparticles and then blended them with a sodium alginate hydrogel, which was then loaded into a polycaprolactone/β-tricalcium phosphate scaffold. Composite scaffolds can promote both osteogenesis and angiogenesis under diabetic conditions [25].

Additionally, BMSCs could migrate to injury sites and differentiate into osteoblasts induced by osteoinductive factors in external environment or exosomes secreted by itself, ultimately promoting bone repair [26, 27]. However, the poor homing proportion and low survival rate of the stem cells limited the implantation effectiveness [28]. The delivery and functionality of BMSCs rely on suitable biomaterial carriers. Gelatine methacryloyl (GelMA), synthesized from gelatine and methacrylic anhydride, is a thermally stable, cross-linked hydrogel formed through photoinitiator activation or ultraviolet (UV) irradiation [29]. Studies have shown that GelMA can facilitate bone defect repair, support extracellular matrix deposition enriched in type II collagen, and effectively promote angiogenesis [30]. He et al. encapsulated strontium into GelMA hydrogels, which exhibited significant osteogenic and antiresorptive effects in the treatment of osteoporotic bone defects [31]. Compared with cells directly inoculated onto the scaffold, survival and cell seeding efficiency were improved for cells encapsulated in the GelMA hydrogel and then combined with the osteochondral scaffold [32, 33].

Considering these significant findings, we aimed to integrate multiple functional components into a composite scaffold to further enhance bone regeneration. The purpose of this study was to develop a biomimetic SF/nHA scaffold incorporating M-PLGA and GelMA-BMSCs (G-B) and to evaluate its ability to promote BMSC proliferation and osteogenic differentiation, as well as its efficacy in repairing critical-sized bone defects due to DOP in vivo.

Materials and methods

Preparation and characterization of M-PLGA

An explanation of abbreviations is summarized in Table 1. M-PLGA microspheres were prepared via the emulsion solvent evaporation method [24, 34]. Briefly, 1000 mg of PLGA (50:50, Mw = 94,000 Da; Jinan Daigang Biomaterial Co., China; cat. no. 141245-359) was dissolved in 6 mL of dichloromethane (99.9%, 3rd Branch of Tianjin Chemical Reagent Co. Ltd., China). Melatonin (100 mg; Solarbio, China; cat. no. M8600) was dissolved in 0.1 mL of methyl alcohol (99.9%; Tianda Tianjin Chemical Reagent Factory, China). The two solutions were homogenized, added dropwise to 300 mL of 0.25% (w/v) polyvinyl acetate (PVA) solution, and stirred at room temperature for solvent evaporation. The microspheres were subsequently washed, frozen at −80 °C, and lyophilized. The morphology was observed by scanning electron microscopy (SEM, S-4300, Japan). The particle size and distribution were analysed using Measure software (Nano Measure 1.2, China) and Origin software (Origin 7.5, USA), respectively.

Table 1.

Abbreviations and explanation of the different scaffolds

Name	Explanation and description
M-PLGA	PLGA microspheres that encapsulate melatonin
G-B	GelMA hydrogel loaded with BMSCs
SF/nHA	Silk fibroin/nanohydroxyapatite composite scaffold
M-PLGA/SF/nHA	SF/nHA scaffold with melatonin encapsulated within PLGA microspheres
M-PLGA/SF/nHA/G-B	GelMA hydrogel loaded with BMSCs integrated with M-PLGA/SF/nHA scaffold

The encapsulation efficiency and loading efficiency of melatonin

PLGA microspheres (500 mg) were dispersed in 1 mL of 0.04 M NaOH solution, and the melatonin concentration was measured by high-performance liquid chromatography (HPLC) at 282 nm. The encapsulation efficiency (EE) and loading efficiency (LE) of melatonin were calculated using the following equations.

Release test of M-PLGA

Melatonin release from M-PLGA microspheres, M-PLGA/SF/nHA, and M-PLGA/SF/nHA/G (GelMA without BMSCs) scaffolds was assessed by HPLC. The samples were incubated in PBS (50 mL, pH = 7.4) at 37 °C. At set intervals, 0.5 mL of medium was collected and replaced with fresh PBS. The results were calculated on the basis of a standard calibration curve.

Synthesis of M-PLGA/SF/nHA scaffolds

On the basis of previous reports and with some modifications [15], Bombyx mori silk fibres (Simatech, China) were degummed in 0.02 M Na₂CO₃, dissolved in 9.3 M LiBr at 60 °C for 4 h, and dialyzed (a dialysis bag; MWCO 3500 Da; Solarbio, China; cat. no. YA1078) in ultrapure water for 48 h. The solution was then concentrated to 15% using polyethylene glycol (PEG) (Biosharp, China). To stabilize the nHA (300 mg/mL; Macklin, China; cat. no. H811001) in aqueous solution, carboxylated cellulose nanofibrils (cCNFs; Naxian Tech, China) were added at a 4:1 (v/v) ratio. The nHA/cCNF suspension was mixed with 10% (w/v) SF solution (1:9 v/v) under stirring for 1 h and then combined with M-PLGA microspheres (5 mg/mL) to form the M-PLGA/SF/nHA scaffold (1.5 mm diameter, 3 mm height), which was freeze-dried and sterilized by irradiation.

BMSC culture and identification

BMSCs were purchased from Fuyuan Bio Co. (China; cat. no. FY200001). Upon receipt of the BMSCs, their osteogenic, chondrogenic and adipogenic differentiation potentials were assessed to verify their stemness. BMSCs were cultured in α-MEM supplemented with 10% FBS and 100 U/mL penicillin/streptomycin in an incubator at 37 °C and 5% CO₂. In accordance with the methods in the literature [34], for adipogenic and osteogenic identification, BMSCs at passage three at a density of 2 × 10⁴ cells/cm² were seeded into 6-well culture plates. After the cells reached 70–80% confluence, the medium was replaced with adipogenic or osteogenic medium (Fuyuan Bio Co., China; cat. no. FY200007 or FY200006, respectively) and then renewed every three days. After 7 or 21 days, the adipogenic capacity of the BMSCs was explored using an Oil Red O staining kit (Solarbio, China; Solarbio, China; cat. no. G1262), and osteogenic activity was detected with an Alizarin Red S staining kit (Solarbio, China; cat. no. G3284) according to the manufacturer’s instructions. For chondrogenic differentiation, 1.5 × 10⁶ BMSCs resuspended in a 15 ml centrifuge tube were centrifuged at 800 × g for 5 min, after which the BMSC pellets were obtained and cultured in chondrogenic medium (Fuyuan Bio Co., China; cat. no. FY200008). After 4 weeks, the pellet was incubated with Alcian blue solution (Solarbio, China; cat. no. G1563) for 20 min. The stained samples were subsequently observed under a microscope (Nikon, Japan). Following identification, the BMSCs were mixed with the GelMA hydrogel for subsequent experiments.

Synthesis of M-PLGA/SF/nHA/G-B scaffolds

A GelMA solution (5% w/v) (Engineering for Life, China; cat. no. EFL-GM-30) with 0.25% (w/v) lithium phenyl (2,4,6-trimethylbenzoyl) phosphinate photoinitiator (LAP, Engineering for Life China; cat. no.85073-4) was filtered through a 0.22 μm membrane and mixed with BMSCs (1 × 10⁷ cells/mL) to form a G-B suspension. The trilineage differentiation of the BMSCs (Fuyuan Bio Co., Ltd., China) was confirmed by Alizarin Red S staining, Oil Red O staining, and Alcian blue (all from Solarbio, China) after induction with the corresponding media. Sterilized cylindrical scaffolds (6 mm diameter, 3 mm height) were placed into custom polyethylene moulds (6 mm diameter, 6 mm height). The G-B suspension fully infiltrated the scaffolds under negative pressure and was crosslinked with 405 nm blue light for 40 s (Fig. 1A).

Fig. 1.

Characterization of M-PLGA and composite scaffolds. A Schematic illustration of the fabrication of the M-PLGA/SF/nHA/G-B scaffold for osseous tissue repair. B–C Morphology under TEM and particle size distribution of M-PLGA. D Cumulative release profile of melatonin in vitro from M-PLGA, M-PLGA/SF/nHA and M-PLGA/SF/nHA/G scaffolds. E–F Appearance and morphology of the SF/nHA scaffold as determined by TEM. G Morphology of the M-PLGA/SF/nHA scaffold under TEM. H-I TEM image and morphology of the M-PLGA/SF/nHA/G-B scaffold. J Micro-CT image of cells within the M-PLGA/SF/nHA/G-B scaffold. K GelMA hydrogel before and after ultraviolet light exposure. L‒M Compression strength and pore diameter of the three types of scaffolds. The data are presented as the means ± SDs. One-way ANOVA with the Bonferroni multiple comparison test was employed to assess the differences among multiple groups. The statistically significant difference was at p < 0.05. Scale bar in Fig. 1B, G and I: 10 μm. Scale bar in Fig. 1F: 100 μm

Physical properties of the scaffolds

After freeze-drying, the scaffold microstructures were examined by SEM, and the pore diameters were measured using Nano Measure 1.2 software. M-PLGA/SF/n-HA/G-B scaffolds were scanned using micro-CT (Siemens, Germany). The mechanical properties of the SF/nHA, M-PLGA/SF/nHA, and M-PLGA/SF/nHA/G-B scaffolds were evaluated using an Electro-Force 3230 testing system (BOSE, USA) under uniaxial compression at a loading rate of 0.5 mm/min until failure.

The cytocompatibility of the scaffolds

The cytocompatibilities of the SF/nHA, M-PLGA/SF/nHA, and M-PLGA/SF/nHA/G-B scaffolds were evaluated using a CCK-8 assay (Dojindo Laboratories, Japan; cat. no. CK04). Equal numbers of BMSCs were seeded onto the scaffolds for 1, 3, 7, and 14 days. After being washed with PBS, the cells were incubated with 10% CCK-8 for 1.5 h at 37 °C, and 100 µL of the supernatant was transferred to a 96-well plate for absorbance measurement at 450 nm using a microplate reader (BioTek, USA). BMSCs cultured on the three scaffolds were subjected to live/dead staining (Solarbio, China; cat. no. CA1633) and phalloidin staining (Solarbio, China; cat. no. CA1610) after 1 and 7 days of incubation. Cell morphology and viability were observed using confocal laser scanning microscopy (CLSM, LSM 800, Germany).

Osteogenic differentiation of BMSCs in the scaffolds

SF/nHA, M-PLGA/SF/nHA, and M-PLGA/SF/nHA/G-B scaffolds seeded with equal numbers of BMSCs were cultured in osteoinductive medium, with the medium changed every 48 h. After 7 days, the scaffolds were collected for histological sectioning. Alkaline phosphatase (ALP) activity was assessed using an ALP assay kit (Beyotime Biotechnology, China; cat. no. C3206). Collagen type 1 (COL1) expression was assessed by immunohistochemical staining using a primary antibody against COL1 (1:1000; Proteintech, USA; Cat No. 14695-1-AP), followed by a biotin-labelled secondary antibody, after which SABC (Boster, China; Cat No. SA1046 according to the manufacturer’s instructions) and haematoxylin (Solarbio, China; cat. no. G1080) were used for nuclear counterstaining. After 14 days of incubation, von Kossa staining (Solarbio, China; cat. no. G3282) was used to assess mineralized matrix deposition, and on Day 21, Alizarin Red S staining (Solarbio, China; cat. no. G3284) was performed to evaluate calcium deposition according to the manufacturer’s directions. Stained samples were observed under a microscope (Nikon, Japan).

Gene expression of the BMSCs in the scaffolds

Total RNA was extracted from the BMSCs on Day 7 using TRIzol reagent (Invitrogen, USA), and cDNA was synthesized from 5 µg of RNA using the ReverTra Ace qPCR RT Kit (Toyobo, Japan). Quantitative real-time PCR (RT‒qPCR) was performed with SYBR^® Green Real-Time PCR Master Mix (Toyobo Co., Japan) on a LightCycler^® 480 II system (Roche, Switzerland) to evaluate the expression of COL1A, Runt-related transcription factor 2 (RUNX2) and osteopontin (OPN). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the reference gene, and gene expression was analysed using the ΔΔCt method (Table 2). The sequences of primers used were as follows:

Table 2.

The sense and antisense primer sequences of target genes for quantitative real-time PCR

Gene	Forward primers (5’-3’)	Reverse primers (5’-3’)
GAPDH	TGTGTCCGTCGTGGATCTG	TTGCTGTTGAAGTCGCAGGA
RUNX2	TCTTAGAACAAATTCTGCCCTTT	TGCTTTGGTCTTGAAATCACA
COL1A	GCTGATGATGCCAATGTGGTT	CCAGTCAGAGTGGCACATCTTG
OPN	TCCAAAGCCAGCCTGGAAC	TGACCTCAGAAGATGAACTC

Animal model

This study was approved by the ethical committee of Tianjin Hospital (No. 2022 105). Male SD rats (240–280 g) were purchased from the Experimental Animal Centre of Tianjin Hospital. Bone defect models (1.5 mm diameter, 2 mm depth) were established in the bilateral femurs. DOP was induced by tail suspension at a 30° head-down tilt for 6 weeks using specialized clips (Yamashita Giken, Japan) to prevent hindlimb contact with the cage floor [35]. During the suspension period, the rats retained forelimb support and free 360° movement within the cage. Nonsuspended rats served as the control group. After the bone defect model was established, the rats were randomly divided into three groups and treated with the SF/nHA, M-PLGA/SF/nHA or M-PLGA/SF/nHA/G-B scaffolds. Six weeks post-implantation, the rats were euthanized via intraperitoneal injection of butorphanol (5.0 mg/kg), medetomidine (0.3 mg/kg), or midazolam (4.0 mg/kg). Bilateral femurs were collected for further analysis.

The diagnostic criteria for osteoporosis are mainly based on bone mineral density (BMD) tests [36, 37]. The BMD of the distal femur was measured by dual-energy X-ray absorptiometry (DXA, Lunar-Prodigy; GE). Following anaesthesia with chloral hydrate (300 mg/kg, ip), the rats were fixed in the supine position, with hind limb extorsion and hip, knee and ankle joint flexion at 90°. We used DXA (Lunar-Prodigy; GE) to analyse the BMD of the metaphyseal part of the distal femur, which was scanned in “small animal” mode with a high resolution of 0.1 mm×0.1 mm. The accompanying software automatically analysed the data and output the BMD (g/cm²).

Microcomputed tomography evaluation

Bilateral femurs were analysed using micro-CT with 3D reconstruction performed automatically. Microstructure was assessed using the bone volume-to-total volume ratio (BV/TV), bone surface area-to-bone volume ratio (BS/BV), and trabecular thickness (TB.TH) as key evaluation parameters.

Histological analysis

The harvested specimens were decalcified in 15% ethylenediaminetetraacetic acid (EDTA) (Solarbio, China; cat. no. E8030), dehydrated through graded ethanol (70–100%), embedded in paraffin, and sectioned at a thickness of 5 μm. The sections were stained with haematoxylin and eosin (HE; Solarbio, China; cat. no. G1120), toluidine blue (Solarbio, China; cat. no. G3668), and safranin O (all from Solarbio, China; cat. no. G1371) according to the manufacturer’s instructions and then observed under a microscope (Nikon, Japan).

Statistical analysis

Statistical analysis was performed on independent data, including the release profile of melatonin, pore diameter of the scaffolds, compressive strength, CCK-8 OD values, osteogenic gene expression, distal femur BMD, and micro-CT parameters, and the results are presented as the means ± SDs. GraphPad Prism software (GraphPad Software Inc., USA) was used for statistical analyses and graph generation. One-way ANOVA with Bonferroni post hoc correction was employed for multiple group comparisons, with significance at P < 0.05. Significant differences between groups are indicated as follows: * < 0.05, ** < 0.01, and *** < 0.001 vs. the SF/nHA group; # < 0.05, ## < 0.01, and ### < 0.001 vs. the G-PLGA/SF/nHA group; and & < 0.05, && < 0.01, and &&& < 0.001 vs. the G-PLGA/SF/nHA/G-B group.

Results

Microsphere and scaffold characterization and detection of melatonin release

SEM analysis revealed that the M-PLGA microspheres were uniform and spherical (Fig. 1B), with an average diameter of 9.8 ± 3.6 μm (Fig. 1C). The melatonin LE was 0.025.3 ± 0.0062 mg/mg PLGA, and the EE was 28.7 ± 3.5%. In vitro release profiles were similar for the microspheres and composite scaffolds (Fig. 1D), with approximately 38.7% release from the microspheres, 33.6% from the M-PLGA/SF/nHA scaffold, and 33.3% from the M-PLGA/SF/nHA/G scaffold in the first 2 days. The release gradually increased, reaching 82.0%, 74.5%, and 73.5%, respectively, after 30 days. SEM images revealed that the SF/nHA scaffold had a well-interconnected porous structure, with M-PLGA microspheres evenly distributed in the M-PLGA/SF/nHA scaffold (Fig. 1F–G). The GelMA concentration used for crosslinking was 5% w/v and the LAP concentration was 0.25% (w/v) in the reaction mixture, which was then filtered through a 0.22 μm membrane and mixed with BMSCs (1 × 10⁷ cells/mL) to form a G-B suspension. The G-B suspension fully infiltrated the scaffolds under negative pressure and was crosslinked with 405 nm blue light for 40 s. In the M-PLGA/SF/n-HA/G-B scaffolds, the BMSCs and microspheres were mostly encapsulated by the hydrogel, which was rarely observed directly (Fig. 1I). 3D micro-CT reconstruction revealed a uniform cell distribution within the scaffolds (green dots; Fig. 1J). The average pore diameters were 101.0 μm, 99.2 μm and 94.2 μm for SF/nHA, M-PLGA/SF/nHA and M-PLGA/SF/nHA/G-B, respectively (Fig. 1L). The compressive strength did not significantly differ among the three scaffolds (Fig. 1M).

Assessment of the biocompatibility of the scaffolds

The BMSCs demonstrated clear evidence of osteogenic, adipogenic, and chondrogenic differentiation (Fig. 2A). Compared with the SF/nHA and M-PLGA/SF/nHA scaffolds, the M-PLGA/SF/nHA/G-B scaffold significantly increased cell proliferation (Fig. 2B).

Fig. 2.

Effects of different scaffolds on the cytocompatibility of BMSCs in vitro. A The ability of the cultured BMSCs to undergo multilineage differentiation, including osteogenic, adipogenic and chondrogenic differentiation, was excellent. B The proliferation of the BMSCs in the scaffolds was determined by a CCK-8 assay. C The viability of the BMSCs was assessed by live (green staining)/dead (red staining) cell staining on Day 1 and 7. The data are presented as the means ± SDs. One-way ANOVA with the Bonferroni multiple comparison test was employed to assess the differences among multiple groups. The statistically significant difference was considered to be at p < 0.05. Significant differences between groups are indicated as follows: * < 0.05, ** < 0.01, and *** < 0.001 vs. the SF/nHA group; # < 0.05, ## < 0.01, and ### < 0.001 vs. the G-PLGA/SF/nHA group

Live/dead staining was performed to assess the number of viable BMSCs in the scaffolds (Fig. 2C). After 1 day of culture, live/dead staining revealed that all the cells on the three types of scaffolds were alive (green fluorescence), and dead cells (red fluorescence; red staining indicated nonspecific staining of the wall of the scaffold) were not observed. The numbers of living cells were evaluated by analysing the relative area of green fluorescence with ImageJ software (National Institutes of Health, USA). Compared with that in the SF/nHA group, the numbers of viable cells in the M-PLGA/SF/nHA and M-PLGA/SF/nHA/G-B groups was comparable to that in the SF/nHA group (102.5 ± 1.6% and 104.4 ± 2.0%, respectively, with no significant difference (p ≥ 0.05)). Moreover, after 7 days of culture, only a very small proportion of cells on all the scaffolds were dead. Compared with the number of living cells in the SF/nHA group, significantly greater viability (p < 0.001) was detected in the M-PLGA/SF/nHA group (287.2 ± 7.2%) and M-PLGA/SF/nHA/G-B group (621.7 ± 8.1%). In addition, the number of living cells in the M-PLGA/SF/nHA/G-B group was dramatically greater than that in the M-PLGA/SF/nHA group (p < 0.001).

Phalloidin staining confirmed that the BMSCs in all the scaffolds had well-organized cytoskeletons and elongated morphologies, suggesting that all the scaffolds facilitated cell adhesion and were not cytotoxic. Among the three types of scaffolds, the M-PLGA/SF/nHA/G-B scaffold resulted in greater cell spreading and distribution after 7 days (Fig. 3). These results indicated that the M-PLGA/SF/nHA scaffold, especially with G-B, provided an optimal microenvironment for BMSC survival and proliferation.

Fig. 3.

The adhesion of BMSCs on different scaffolds assessed by phalloidin staining on Day 1 and 7 to assess the cytocompatibility; blue = DAPI, red = phalloidin. Scale bar: 100 μm

Evaluation of the osteogenic induction properties of the scaffolds

The osteogenic potential of the scaffolds was evaluated by ALP activity and Alizarin Red S staining. ALP is an important biomarker of bone ossification [38]. After 7 days in osteogenic medium, compared with the SF/nHA group, both the M-PLGA/SF/nHA group and the M-PLGA/SF/nHA/G-B group showed greater ALP activity and more extensive areas of staining. By Day 21, Alizarin Red S staining revealed greater calcium deposition in the M-PLGA/SF/nHA and M-PLGA/SF/nHA/G-B groups, suggesting enhanced late-stage osteogenesis (Fig. 4A). COL1 immunohistochemistry on Day 7 revealed intense brown staining of the pericellular area and extracellular matrix, with spindle-shaped BMSCs aligned along the scaffold, especially in the M-PLGA/SF/nHA/G-B group. Von Kossa staining on Day 14 revealed prominent black deposits, particularly in the M-PLGA/SF/nHA/G-B group, indicating calcium salt accumulation and mineralization (Fig. 4A).

Fig. 4.

Effects of the scaffolds on osteogenic differentiation of the BMSCs in vitro. A ALP staining and Alizarin Red S staining of BMSCs after 14 days and 21 days of culture in osteogenic medium in the presence of the scaffolds. Von Kossa staining and COL1 immunohistochemical staining of the scaffolds seeded with BMSCs after 21 and 7 days of culture in osteogenic medium. B Osteogenesis-related gene expression in the BMSCs cultured in the scaffolds was detected by RT‒PCR. C Analysis of BMD in the distal femur of rats. Scale bar: 100 μm. The data are presented as the means ± SDs. One-way ANOVA with the Bonferroni multiple comparison test was employed to assess the differences among multiple groups. The statistically significant difference was at p < 0.05. Significant differences between groups are indicated as follows: * < 0.05, ** < 0.01, and *** < 0.001 vs. the SF/nHA group; # < 0.05, ## < 0.01, and ### < 0.001 vs. the G-PLGA/SF/nHA group. & < 0.05, && < 0.01, &&& < 0.001 vs. the G-PLGA/SF/nHA/G-B group

Osteogenic gene expression of BMSCs cultured in the scaffolds

To further assess osteogenic differentiation, RT‒qPCR was performed after 7 days of induction. The expression of osteogenic markers (COL1A, RUNX2, and OPN) was significantly greater in BMSCs cultured in the M-PLGA/SF/nHA/G-B scaffold than in BMSCs from other groups (Fig. 4B), demonstrating increased mineralization and osteogenic potential. The primer sequences are shown in Table 1.

Effects of the scaffolds in an animal osteoporotic bone defect model

At six weeks post-operation, the BMD measurements confirmed osteoporotic levels in all the bone defect models. However, the BMD was significantly greater in the M-PLGA/SF/nHA/G-B and M-PLGA/SF/nHA groups than in the SF/nHA group (Fig. 4C). Micro-CT imaging revealed increased new bone formation in both the M-PLGA/SF/nHA/G-B and M-PLGA/SF/nHA groups, with the M-PLGA/SF/nHA/G-B group exhibiting the strongest regenerative effect (Fig. 5C). Quantitative analysis revealed significantly higher BV/TV and TB.TH values and lower BS/BV ratios in the M-PLGA/SF/nHA/G-B group, indicating superior bone repair (Fig. 5D).

Fig. 5.

Effects of the scaffolds on bone defect repair in vivo. A Bone defect site implanted with scaffolds. B–C General appearance and 3D reconstructed micro-CT images of the bone defect region repaired by the scaffolds for 6 weeks. D Quantitative analyses of BV/TV, TB. TH and BS/BV to evaluate regenerative new bone within bone defect sites treated with the scaffolds. The data are presented as the means ± SDs. One-way ANOVA with the Bonferroni multiple comparison test was employed to assess the differences among multiple groups. The statistically significant difference was at p < 0.05. Significant differences between groups are indicated as follows: * < 0.05, ** < 0.01, and *** < 0.001 vs. the SF/nHA group; # < 0.05, ## < 0.01, and ### < 0.001 vs. the G-PLGA/SF/nHA group. & < 0.05, && < 0.01, &&& < 0.001 vs. the G-PLGA/SF/nHA/G-B group

HE staining revealed an eosinophilic matrix with osteocyte-like cells in the lacunae, indicating mature bone formation. In SF/nHA group, there existed a mass of undegraded material (RM, Remnant Material) and some connective tissue (CT), while, in the M-PLGA/SF/nHA groups, more connective tissue was observed and remnant material decreased. Compared with the SF/nHA and M-PLGA/SF/nHA group, the M-PLGA/SF/nHA/G-B group exhibited marked new bone (NB, new bone)regeneration and increased collagen fibre deposition at defect sites. The toluidine blue staining could distinguish between newly formed bone and original-mineralized bone (OB), with the newly formed bone showing a darker staining. In safranin O-fast green staining, bone tissue was stained green while cartilage tissue stained red. Toluidine blue and safranin O-fast green staining further confirmed enhanced osteoid matrix and new bone formation in the M-PLGA/SF/nHA/G-B group, demonstrating its superior regenerative capacity over the other groups (Fig. 6).

Fig. 6.

Effects of the scaffolds on bone defect repair in vivo assessed by histological analysis: HE, toluidine blue and safranin O-fast green staining of the bone defect region 6 weeks after implantation surgery. The images of the three types of staining in each group (SF/nHA, M-PLGA/SF/nHA, M-PLGA/SF/nHA/G-B or control) were obtained from consecutive slices of the same bone tissue sample paraffin block. However, the samples used for staining varied among the four groups. Scale bar: 500 μm. OB: Original Bone; NB: New Bone; CT: Connective Tissue; RM: Remnant Material

Discussion

Although bone has a natural capacity for regeneration, this regeneration is often inadequate for repairing large or critical-sized defects, which typically require surgical intervention or bone grafting [39]. Posttreatment immobilization can lead to DOP, further complicating bone defect repair. With advances in bone pathophysiology and tissue engineering, BTE scaffolds have emerged as promising solutions. An ideal BTE scaffold should exhibit excellent biocompatibility, biodegradability, osteoconductivity, osteoinductivity, and mechanical strength, along with the ability to support angiogenesis and modulate the bone microenvironment [40]. These features are essential for providing initial mechanical support at the defect site while enhancing stem cell and osteoblast attachment, proliferation and differentiation, as well as promoting bone matrix formation and mineralization through a well-designed scaffold.

Cell-based bone tissue engineering holds great promise for defect repair, but it requires scaffolds with specific properties and the sustained viability of implanted cells during healing [41]. SF offers excellent biocompatibility and low immunogenicity but lacks sufficient osteoinductivity for effective bone regeneration [42]. The incorporation of nHA, which mimics the composition of natural bone, enhances both the mechanical strength and osteogenic potential of the scaffold [43]. A well-designed pore structure further supports cell organization, interactions, and extracellular matrix formation. Previous studies have shown that pore sizes of 100–300 μm facilitate cell infiltration, neovascularization, and bone tissue regeneration, whereas 2–50 nm mesopores promote osteoblast proliferation and differentiation [44, 45]. The SF/nHA, M-PLGA/SF/nHA and M-PLGA/SF/nHA/G-B scaffolds exhibited pore sizes of 101.0 μm, 99.2 μm, and 94.2 μm, respectively, all of which were within the optimal range for cell growth. Furthermore, compressive strength testing revealed no significant differences among the groups, indicating that the scaffold modifications maintained mechanical integrity and structural stability (Fig. 1L–M).

Biopolymeric-based microspheres have been widely served as the controlled and sustained transmitting carriers for the therapeutic agents [46]. This study demonstrated sustained melatonin release from M-PLGA microspheres over 30 days, showing that preencapsulating melatonin in PLGA before it was integrated into the SF/nHA scaffold enabled prolonged and controlled drug delivery. The PLGA microparticles were uniformly distributed in the M-PLGA/SF/nHA scaffold and M-PLGA/SF/nHA/G-B scaffold, as observed by SEM and micro-CT, respectively, and the incorporation of the microspheres did not affect the porous structures of the composite scaffold (Fig. 1G and J). Zhao et al. combined naringin-loaded PLGA microspheres with SF/nHA scaffolds. The release of naringin from the PLGA microspheres or the composite scaffold was similar to the burst release observed within the initial 24 h, but the release gradually slowed in the following 36 days, indicating a similar controlled-release trend [34]. Zhang et al. encapsulated curcumin in PLGA microparticles, which were then mixed with GelMA hydrogel to repair infected skin wounds. The release profile of curcumin from the PLGA microparticles was nearly identical to that of the PLGA microparticle-GelMA composite scaffold, with burst release after the first 24 h of incubation and sustained release in the following 20 days [47]. Our results were similar, i.e., the release profiles and cumulative release of the PLGA microspheres, M-PLGA/SF/nHA scaffold and M-PLGA/SF/nHA/G-B scaffold percentage did not significantly differ at the detection time points (Fig. 1D). Therefore, the SF/nHA scaffold and GelMA hydrogel likely do not significantly affect the sustained-release profiles of melatonin from PLGA microspheres. As PLGA degrades, the loaded factors are gradually released. The degradation rate of PLGA depends on the monomer ratio. We selected a copolymer with a 50:50 monomer ratio because the resultant microspheres rapidly degrade (in approximately 2 months) [24]. The burst release during the first few hours was attributed, to a certain extent, to the detachment of the adsorbed factors to the microsphere walls, after which the adsorbates continuously diffused from PLGA in a sustained manner for the following days [48]. However, we observed that the cumulative release percentage of the two scaffolds was relatively low compared with that of the PLGA microparticles (Fig. 1D). This difference may have occurred because in the M-PLGA/SF/nHA scaffold, most of the microspheres were partially embedded in the inner walls of the composite scaffold, thus limiting the sustained release rate. However, the almost unaffected interconnected porous structures of the M-PLGA/SF/nHA scaffold could facilitate the diffusion released factor to the surrounding area. As hydrogels possess retentive properties [25], the M-PLGA/SF/nHA/G-B scaffold GelMA demonstrated a slightly slower release rates than that of the pure PLGA microparticles (Fig. 1D). However, SEM revealed that the GelMA hydrogel structure was a honeycomb, with pores on the surface interconnected with those in the interior (Fig. 1I). The PLGA microspheres were homogeneously distributed in the GelMA hydrogel (Fig. 1J), and such a structure promoted the dispersion of factors.

Melatonin, which is known primarily for its ability to regulate sleep–wake cycles, also has bone-promoting, antioxidant, and anti-inflammatory effects [49]. Current research suggests that melatonin suppresses osteoclast formation and bone resorption by modulating the NF-κB signalling pathway and the RANKL/OPG balance [50]. In our study, live/dead and phalloidin staining revealed that the BMSC viability, spreading, and surface coverage were greater on the M-PLGA/SF/nHA scaffold than on the SF/nHA, indicating that M-PLGA increases cell survival and proliferation. Moreover, ALP, Alizarin Red S, Von Kossa, and COL1 staining, along with RT‒qPCR analysis, consistently demonstrated that the SF/nHA scaffold containing M-PLGA promoted the osteogenic differentiation and maturation of BMSCs into osteoblasts (Fig. 2). Stem cells play a vital role in tissue regeneration [51, 52], but traditional seeding onto scaffolds often results in poor adhesion and significant cell loss [41]. Enhancing cell–scaffold interactions is crucial for improving therapeutic outcomes. Hydrogels, which mimic the extracellular matrix with excellent hydrophilicity, are well suited for encapsulating and delivering stem cells, thereby increasing cell retention and viability within scaffolds [53]. Compared with gelatine, GelMA offers advantages such as slower degradation and increased mechanical strength [54]. In this study, a BMSC-laden GelMA hydrogel was integrated with the M-PLGA/SF/nHA scaffold to promote bone regeneration. Cytocompatibility assays revealed that compared with the M-PLGA/SF/nHA, the M-PLGA/SF/nHA/G-B scaffold better supported BMSC growth. Osteogenic induction experiments further demonstrated that the M-PLGA/SF/nHA/G-B scaffold significantly increased BMSC differentiation and maturation into osteoblasts. The M-PLGA/SF/nHA/G-B scaffold provided both mechanical support and osteoinductive signals, notably upregulating key osteogenic genes, including COL1A, RUNX2, and OPN (Fig. 3).

Ultraviolet (UV) photopolymerization not only affects the cross-linking of GelMA hydrogels but also influences cell survival. The shorter the UV light exposure, the higher the cellular survival rate obtained. An excessively long crosslinking time could lead to free radical accumulation, triggering oxidative stress and thereby reducing the cell survival. However, an irradiation time that is too short leads to low mechanical strength [29]. High-concentration GelMA (such as 10%) could provide more stable mechanical support, but this composition may cause uneven cross-linking because of the light scattering effect. The GelMA concentration also needs to be optimized to ensure deep cross-linking of the material. Experimental results revealed that after exposure to 405 nm UV light for 30 s, 5% GelMA is conducive to BMSC survival because of its larger pore size and greater porosity [55]. Li et al. encapsulated BMSCs into a GelMA hydrogel and then exposed them to UV irradiation for 40 s, which also resulted in good cell activity [30]. Our studies adopted 5% GelMA with 405 nm UV light exposure for 40 s, and the obtained composite scaffold not only had good mechanical properties but also good cell compatibility (Figs. 1M and 2C).

GelMA has been extensively studied in tissue engineering for bone and cartilage regeneration, and it can be purchased commercially. The literature has reported that lower concentrations of GelMA are more conducive to promoting cell activity than high concentrations of GelMA [56]. Therefore, our study also adopted a 5% GelMA hydrogel for cell encapsulation. Degradation analysis of 5% GelMA revealed that following soaking in deionized water, the mass loss rate was 14.4% on Day 7, 68.1% on Day 14 and 91.0% on Day 28, which was advantageous for cell growth and factor diffusion. The swelling ratio of the 5% GelMA hydrogel was approximately 176.8% following immersion in deionized water for 24 h, which was much greater than that of the 10–15% hydrogels [57]. Moreover, the water retention capacity of the 5% GelMA hydrogel was much stronger than that of the 10–15% hydrogels [58]. Thus, the 5% GelMA hydrogel is more hydrophilic, which facilitates cell survival and nutrient transport. A rheological test revealed that the storage modulus (G′) of 5% GelMA at 1 Hz was 574.0 Pa, the loss modulus (G′) was 20.4 Pa, and the G′′/G′ ratio was 0.04, suggesting that 5% GelMA had good viscoelastic-solid capabilities and could provide stabilized mechanical support [57]. Therefore, the GelMA hydrogels could provide a stable microenvironment for BMSC proliferation and differentiation. Silk fibroin/hydroxyapatite composite scaffolds have been proven to be highly favourable materials for bone tissue engineering [16]. Appropriate biodegradability and biostability are critical for biomimetic scaffolds. The degradation rate should match the degree of new tissue formation within the scaffolds. When composite scaffolds composed of 10% (wt%) silk fibroin and hydroxyapatite were cultured in lysozyme degradation solution for 7 days, the mass loss percentage was approximately 5%. When incubated in simulated body fluid, the weight loss percentage was still less than 10% [59]. Tungtasana et al. reported that when silk fibroin/hydroxyapatite scaffolds were incubated in collagenase solution for 28 days, approximately half the mass remained at the end of the experiment [60]. Zhao et al. fabricated silk fibroin/hydroxyapatite scaffolds with a silk fibroin concentration of 10% w/v, and naringin-inlaid PLGA microspheres adhered to this scaffold. After being seeded with BMSCs, the scaffolds were incubated in osteogenic differentiation medium for 21 days, and the porous structure of the scaffold remained mostly intact and demonstrated excellent osteogenic induction properties [34]. Wang et al. adopted directional temperature field freezing technology to fabricate a silk fibroin/nanohydroxyapatite scaffold that contained BMSC-laden GelMA hydrogels. After being incubated in protease solution for 15 days, the degradation rates of SF/nHA and GelMA-SF/nHA were 20.4–36.6%, respectively, whereas in PBS more than 90% and 85% of the mass remained, respectively [41]. Previous studies had revealed that the degradation process of PLGA should last approximately 2 months [24].

Therefore, our M-PLGA/SF/nHA/G-B scaffolds provide a stable environment for cell growth and have an appropriate degradation rate for BMSC ingrowth and factor penetration. The use of a silk fibroin/nanohydroxyapatite scaffold improved the rapid degradation of the GelMA hydrogel to a certain extent.

Following osteogenic induction for 7 days, the mRNA expression of RUNX2 in the BMSCs seeded in the M-PLGA/SF/nHA/G-B scaffold dramatically increased (Fig. 3B). Runx-2 not only is a crucial marker of osteogenic differentiation but also participates in osteoclastogenesis, promoting bone formation and suppressing bone resorption [61]. Thus, the increase in RUNX2 expression reflected the inhibition of bone resorption and decreased osteoclast activity. Melatonin has been reported to increase MSC proliferation, promote the osteogenic differentiation of MSCs, and increase matrix calcium deposits [61], which may explain the increases in COL1A, RUNX2, and OPN gene expression in our study. In addition to promoting bone formation, osteoporosis therapies involve the inhibition of bone resorption and osteoclastogenesis [62]. Melatonin also promotes RAW264.7 apoptosis and inhibits osteoclast activity [63]. Moreover, melatonin can antagonize unloading-induced disuse osteoporosis [64]. Jarrar et al. constructed a chitosan/hydroxyapatite scaffold that sustained the release of melatonin and BMP-2, and the incorporated melatonin successfully inhibited osteoclastogenesis induced by BMP-2, enhancing bone regeneration and inhibiting bone resorption [20]. Thus, the M-PLGA/SF/nHA/G-B scaffold in our study may also theoretically inhibit osteoclast differentiation and bone resorption.

As patients with bone defects often suffer from osteoporosis, which hinders the healing of fractures, timely diagnosis and intervention are necessary [65, 66]. BMD is generally adopted as the proposed detection indicator of osteoporosis and Dual-energy X-ray is widely recognized as the gold standard for measuring bone density [67]. The distal femur of a rat is often chosen as the region of interest for BMD measurement, and the BMD of the distal femur in a rat usually ranges from approximately 0.15 g/cm² to more than 0.20 g/cm² under normal breeding conditions [68, 69]. Differences in BMD measurements are related to factors such as the age, weight and sex of the rats, as well as the type of detection instrument used. In our in vivo studies, we measured BMD in the distal femur to assess the degree of osteoporosis at six weeks post-operation. In the control group (bred under normal conditions), the measured BMD was approximately 0.1333 ± 0.0096 g/cm², whereas the BMD was 0.0701 ± 0.0055 in the SF/nHA group, 0.0903 ± 0.0028 in the M-PLGA/SF/nHA group, and 0.0915 ± 0.0039 in the M-PLGA/SF/nHA/G-B group; these values were significantly lower than those in the control group. This finding indicated that tail suspension led to significant disuse osteoporosis. However, the BMDs in the M-PLGA/SF/nHA/G-B and M-PLGA/SF/nHA groups were dramatically greater than that in the SF/nHA group, suggesting that our composite scaffold supplemented with melatonin-PLGA microspheres and GelMA-BMSCs promoted bone regeneration, increasing the BMD.

In vivo, the bone repair capacity of the M-PLGA/SF/nHA/G-B scaffold was evaluated in the femoral defects of rats with DOP. The results of the histological staining (HE, toluidine blue, and safranin O-fast green) were consistent with the micro-CT results, revealing substantial new bone formation in the M-PLGA/SF/nHA/G-B group compared with the other groups. These findings confirmed that the scaffold was biocompatible and promoted bone growth in terms of both bone quantity and quality (Fig. 4). Similarly, Wang et al. reported that the BMSCs@GelMA-SF/nHAp composite scaffold significantly enhanced skull defect repair in rats by promoting vascularization and osteogenesis, further supporting the benefits of such multifunctional scaffolds [19]. In this study, our biomimetic scaffold demonstrated tremendous potential in treating disuse osteoporosis bone defects.

Regenerative medicine and tissue engineering approach have demonstrated to be one of the most promising approach for tissue regeneration [70, 71]. Tissue engineering involves three major components: seed cells, biological signalling factors and scaffold materials. An eligible bone tissue substitute requires the following: (1) the scaffold materials are readily available, and the structure, function and microenvironment provided should be similar to those of autologous bone; (2) the seed cells have an abundant source, good differentiation ability and robust activity; and (3) the release of signalling molecules should be controlled to effectively maintain the biological activity of seed cells for long periods [72]. Zhao et al. produced PLGA microspheres loaded with naringin, which were then combined with a silk fibroin/hydroxyapatite scaffold. The scaffolds effectively released naringin in a controlled manner, and in vitro and in vivo assays revealed good cell compatibility and osteogenic potential [34]. However, the composite scaffold did not retain seeded cells when it was implanted in a bone defect model. In vitro, BMSCs were loaded by the dropwise addition of a cell suspension onto the scaffold surface and allowed to slowly infiltrate. The cells easily detached, were not evenly distributed throughout the scaffold, and needed a certain amount of time to adhere to the porous structure inside the scaffold; therefore, this method was not efficient for seeding scaffolds with sells [73]. Nevertheless, tissue repair necessitates a sufficient quantity of viable cells [74]. The structure of GelMA is highly hydrated and porous structure, which supports a microenvironment that simulates the cell extracellular matrix. Moreover, GelMA can be easily and controllably cured by photocrosslinking [33]. In our studies, GelMA hydrogel was used as the cell carrier. Compared with directly inoculating cells onto the scaffold, this method resulted in greater cell inoculation and survival rates [74]. The M-PLGA/M-PLGA/SF/nHA scaffold in our study was similar to the composite scaffold fabricated by Zhao et al. For the same culture conditions and number of initially inoculated cells, the cell density, cell viability, number of mineralized nodules, and expression of osteogenic genes were all greater for the M-PLGA/SF/nHA/G-B scaffold than for M-PLGA/SF/nHA scaffold (Fig. 3).

Wang et al. chose a GelMA hydrogel as the BMSC carrier, which was then laden into a silk fibroin/nanohydroxyapatite scaffold; this scaffold provided a stable structure for cell diffusion and delivery and effectively repaired cranial bone defects [41]. However, the biomimetic scaffold lacked signalling factors that guided stem cells to differentiate in a specific lineage, i.e., cell differentiation was not controlled. Compared with traditional physical adsorption methods, PLGA microspheres have a high loading efficiency for cytokines and avoid burst release, attenuating the toxicity and side effects of high doses and maintaining continuous and effective drug concentrations in the microenvironment [24]. In our studies, the PLGA microspheres were loaded with melatonin, which was effectively released in a controlled manner following integration with the silk fibroin/nanohydroxyapatite/GelMA scaffold complex (Fig. 1D). Melatonin not only promotes osteoblast differentiation and inhibits osteoclastogenesis but also inhibits the differentiation of BMCSs into adipocytes [75]. The incorporation of GelMA hydrogel and melatonin in our research not only improved the ability of the composite scaffold to maintain BMSC viability but also potently guided the directed differentiation of stem cells. GelMA hydrogel loaded with melatonin was shown to attenuate, oxidative stress-induced apoptosis in MC3T3-E1 cells, an osteoblast precursor cell line, and improved bone quality around implanted prosthesis in osteoporotic rats [22].

Bone tissue consists of multiple components including organic and inorganic substances, which means that it is difficult to construct an ideal biomimetic bone substitute using a single material. Numerous studies have integrated various different natural or synthetic materials to construct biomimetic bone materials [13, 15, 24, 30]. Due to the limited availability of donors and ethical concerns, extracting materials from other species and then making improvements is one of the inevitable research directions. Just as mentioned above, each component of the composite scaffold in our study had been confirmed to be easily obtained, relatively inexpensive and have excellent biocompatibility in various studies. However, further challenges need to be addressed before this can be applied in clinical settings: (1) Material Source and batch variance: for any materials derived from biological sources (such as silk fibroin and gelatin), sources differences (such as bovine or marine organisms) and various preparation processes may affect the physical and chemical properties, purity, and batch-to-batch consistency of the materials. This requires strict raw material screening, standardized production processes, and comprehensive quality inspections for control; (2) Clinical approval pathway: to introduce this composite scaffold into clinical practice, it is necessary to follow strict clinical approval processes of medical devices or biological products, which usually requires providing detailed data on safety (such as cytotoxicity, biocompatibility, and degradability for long-term application in vivo) and effectiveness (such as the ability to regenerate tissues); (3) Immunogenicity: materials from diverse biological origins sources (such as bovine or marine collagen) may trigger immune rejection reactions. Solutions may include further chemical modification (such as cross-linking) to the materials to enhance their stability, or the selection of low immunogenicity alternative materials; (4) Ethical considerations: as for materials derived from animal, it is required to ensure compliance with ethical standards and consider the patient’s acceptance of using materials originated from animals. Using materials that are certified to meet international standards may help enhance the traceability and ethical compliance of the materials. For example, PLGA has been approved by the US Food and Drug Administration (FDA) for using in drug delivery systems, biodegradable materials, and tissue engineering fields.

Conclusions

In summary, we successfully developed an innovative bone tissue engineering scaffold by combining M-PLGA and G-B with an SF/nHA scaffold. In vitro experiments demonstrated that the M-PLGA/SF/nHA/G-B scaffold provided sustained melatonin release and effectively supported BMSC adhesion, proliferation, and osteogenic differentiation. In vivo experiments using a femoral defect model with disuse osteoporosis further confirmed that the multifunctional composite scaffold significantly increased new bone formation and facilitated the tissue-specific repair of bone defects. These findings suggest that the biomimetic M-PLGA/SF/nHA/G-B scaffold represents a promising and practical strategy for treating bone defects associated with DOP in clinical applications.

Abbreviations

PLGA

poly(lactic-co-glycolic acid)

M-PLGA

PLGA microspheres for the sustained-release of melatonin

SF

Silk fibroin

nHA

Nanohydroxyapatite

GelMA

Methacrylated gelatine

BMSCs

Bone marrow mesenchymal stem cells

DOP

Disuse osteoporosis

BTE

Bone tissue engineering

B-G

GelMA-BMSCs

PVA

Polyvinyl acetate

SEM

Scanning electron microscopy

HPLC

High-performance liquid chromatography

EE

The encapsulation efficiency

LE

Loading efficiency

cCNFs

Carboxylated cellulose nanofibrils

PEG

Polyethylene glycol

LAP

Lithium phenyl (2,4,6-trimethylbenzoyl) phosphinate photoinitiator

Micro-CT

Microcomputed tomography

ALP

Alkaline phosphatase

COL1

Collagen Type 1

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

RUNX2

Runt-related transcription factor 2

OPN

Osteopontin

BMD

Bone mineral density

DXA

Dual-energy X-ray absorptiometry

HE

Haematoxylin and eosin

Author contributions

JL and ZW designed the research strategy and conceived the research. JL and BM fabricated the tissue engineering scaffold and conducted the in vitro assays. JL and JL performed the in vivo assays, performed statistical analysis and wrote the manuscript.ZW revised the paper. All the authors have reviewed and approved the final manuscript.

Funding

Financial support for this work was sponsored by the Tianjin Health Research Project (Grant No. TJWJ2022QN056).

Data availability

The datasets supporting the findings of the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.