A rapidly fabricated bone marrow enrichment material with a controllable pore size promotes osteogenesis close to autogenous bone graft in clinic

Graphical abstract

A: Bone marrow is a natural repository for a large number of effective osteogenic associated cells and cytokines, but the traditional bone marrow enrichment method (directly immersed or injection) is difficult to achieve the ideal effective concentration of the scaffold. B-C: Bone marrow enrichment machine with the mechanism of selective cell enrichment technology can efficiently enrich the cells or cytokines and the MIX scaffold owns a better enrichment efficiency and a better osteogenesis ability in different bone defect conditions.

Highlights

• •A rapidly fabricated DBM/BS scaffold (MIX) achieves controllable pore size for bone marrow enrichment.
• •MIX (1:1 ratio) enhances MSC osteogenesis and angiogenesis via optimized pore structure.
• •Scaffold enriched with autologous marrow achieves goat critical-size defect healing comparable to autograft.
• •Clinical trial validates MIX’s efficacy in lumbar fusion, matching gold-standard outcomes.
• •Policy-friendly solution bridges scalable fabrication with translational orthopedic applications.

Abstract

Introduction

The shortage of highly osteogenic grafts significantly impacts clinical bone repair and reconstruction in clinic. Selective cell retention technology increases the number of bone marrow (BM)-derived osteogenic cells on scaffold materials. A suitable pore size is key for improving the efficiency of BM cell enrichment of scaffold materials. Many materials with a controllable pore size have been developed, but due to ethics policies, only a few have been approved for clinical use.

Objectives

In this study, cortical bone shavings with a small pore size were mixed in different ratios with decalcified bone matrix, which has a relatively larger pore size, to form a new BM enrichment scaffold material with an adjustable pore size (MIX materials).

Methods

We evaluated the in vitro and in vivo enrichment efficiency and osteogenic ability of materials with different pore sizes.

Results

MIX material (1:1) had suitable pore size that was beneficial for the effective enrichment of BM-derived cells and cytokines and promoted the proliferation and osteogenic differentiation of mesenchymal stem cells in vitro. A subcutaneous heterotopic transplantation experiment in a nude mouse model confirmed that MIX material enhanced osteoinductive potential and angiogenesis in vivo. Similarly, a large animal model using goats with a critical bone defect (2.5 cm) showed that the osteogenic ability of enriched MIX was superior to that of nonenriched MIX after 3 months. A subsequent self-controlled prospective clinical study demonstrated that the osteogenic ability of enriched MIX material was close to that of an autograft (the clinical gold standard) in lumbar intervertebral fusion surgery.

Conclusion

MIX scaffold material is an effective choice for the clinical treatment of bone repair and bone reconstruction.

Introduction

With the development of society, bone defects caused by trauma, infection, tumor, and other reasons are very common. In most cases, large bone defects are difficult to heal, and clinical treatment has always been problematic [1,2]. The most commonly used tool for clinical treatment is an autogenous bone graft, which is considered the gold standard. However, there are risks, such as pain and infection in the bone harvesting area, particularly when a relatively large amount of autogenous bone is required, as is the case for spinal fusion and other operations. In such situations, a long incision is made to harvest part of the iliac bone for grafting, further traumatizing the patient [3]. Currently, allogeneic bone, synthetic bone, and other bone graft materials are used to avoid harvesting autogenous bone. However, these materials lack cells and factors involved in osteogenesis, posing a risk of absorption in vivo, leading to bone nonunion. Although approaches such as the Masquelet induced membrane technique or Ilizarov distraction osteogenesis technique for large bone defects have some therapeutic effect, they have yet to fully meet the needs of clinical treatment [4]. Thus, the development of bone graft replacement materials with osteogenic effects similar to those of autogenous bone has long been a focus of research by orthopedic surgeons.

Bone marrow (BM) is an important natural repository of bioactive osteogenic components. Because it contains a large number of stem cells and cytokines crucial for osteogenesis, it is often used for improving the osteogenic activity of scaffold materials and treating bone fracture nonunion. The traditional method of BM enrichment involves immersing the graft in BM solution or injecting BM into the scaffold to achieve a higher local concentration of cells and cytokines. Muschler et al. [5] proposed selective cell retention (SCR) technology to overcome the difficulties encountered in traditional BM enrichment. The underlying principle of SCR technology is the pore size effect of physical interception. When liquid BM passes through the scaffold material, the amount of BM-derived progenitor cells, cytokines, and growth factors with proliferation and differentiation potential retained on the scaffold can be altered by changing the pore size of the scaffold. Additionally, the application of negative suction pressure in the container housing the scaffold material during the enrichment process forces the BM to filter through the scaffold pores, increasing direct contact between the BM and scaffold materials, such as allogenic or artificial bone [6]. Repeated cycles of enrichment increase the amount of cells and factors retained on the scaffold material, thereby improving the biological activity of the bone graft materials, promoting bone reconstruction, and enhancing the biological activity of traditional scaffold materials. Hernigou et al. [7,8] confirmed that SCR resulted in a relatively high concentration of cells, including mesenchymal stem cells (MSCs), karyocytes, endothelial progenitor cells (EPCs), and monocytes, and increased local concentrations of cytokines, such as bone morphogenetic protein, platelet-derived growth factor, and vascular endothelial growth factor (VEGF), on scaffold materials. Furthermore, the overall enrichment time is short (approximately 10 min), facilitating rapid fabrication. SCR technology was employed to prepare BM-enriched materials for use in conjunction with decompression and debridement to treat three patients with femoral head necrosis. All patients showed significantly improved radiographic union scores within 2 years after surgery [9]. Yousef et al. [10] used SCR technology and collagen-modified scaffold for spinal fusion in 21 patients, achieving good long-term clinical results (average follow-up: 7 years). Similarly, Yang et al. [11] used SCR technology in 18 cases of adolescent idiopathic scoliosis, achieving satisfactory clinical results during follow-up. Some studies have compared the results of treatment using traditional bone graft materials with approaches using SCR technology. Shen et al. [12] found that the time to new bone formation was shorter in the SCR group compared with the traditional graft healing group, proving the satisfactory clinical effect of SCR technology.

Graft materials that can be enriched using SCR technology include decalcified bone matrix (DBM), irradiated allograft bone, and tricalcium phosphate. Among them, the most commonly used is DBM, which has a natural mesh structure with a large pore size (400–600 µm) and is conducive to MSC proliferation and new bone growth [13,14]. In contrast, a pore size of 200–300 µm is conducive to blood vessel regeneration in areas with bone defects [15,16]. The size of osteogenic and angiogenic cells in BM is approximately 20 µm [17]. Thus, the relatively large pore size of DBM may not be conducive to SCR. However, the optimal pore size of bone-enriched scaffold materials remains undetermined.

This study is based on previous research on BM enrichment technology and investigates the effects of applying SCR technology to allogeneic bone materials with different controllable pore sizes on enrichment and osteogenesis (Fig. 1). First, cortical bone shavings (BS) with small pore sizes were prepared using bone shaving technology and mixed in different proportions with DBM materials with a larger pore size to obtain a new BM enrichment material (MIX) with adjustable scaffold pore sizes. Through a series of material characterization and evaluation experiments and experiments determining BM enrichment efficiency, we concluded that enrichment plateaued after five enrichment cycles, and the MIX material with a 1:1 ratio of DBM to BS had the highest enrichment efficiency. We also compared the osteogenic ability of several materials in vitro, demonstrating that the MIX BM enrichment material with a relatively moderate pore size was the most conducive to the enrichment of osteogenically active components in BM and MSC proliferation. Furthermore, we evaluated the in vivo osteogenic efficiency of MIX materials using a nude mouse model of subcutaneous heterotopic transplantation and compared the osteogenic efficiency of enriched MIX material and nonenriched MIX material using a critical bone defect model in goats. Finally, we performed a prospective clinical study comparing the effects of MIX-enriched materials and autogenous bone in lumbar intervertebral fusion surgery. The results demonstrated that MIX-enriched materials had robust osteogenic efficacy close to that of autogenous bone. To summarize, we used a new strategy to prepare mixed scaffold materials with a controllable pore size for the clinical treatment of bone defects that significantly improved osteogenesis due to enrichment with BM-derived osteogenic cells and factors. During this process, we used bone shaving technology to prepare small bone chips from allogeneic cortical bone which were used to fill in and around the DBM to construct a mixed scaffold material with a controllable pore size.

Fig. 1.

A: Bone marrow is a natural repository for a large number of osteogenic-associated cells and cytokines, but the traditional bone marrow enrichment method (direct immersion or injection) makes it difficult to achieve the ideal effective concentration of the scaffold. B, C: Bone marrow enrichment device. This machine uses selective cell enrichment technology to efficiently enrich the cells and cytokines on the MIX scaffold. The enrichment efficiency and osteogenesis ability is superior to that of traditional bone marrow enrichment for treating bone defects.

Materials and methods

Material preparation and characterization

Material preparation

Preparation of DBM material: After deep freezing goat hind leg bone for 3 months, the attached tissues were removed, and the bone was washed with purified water. The femoral head was cut with a bone saw and used to prepare pieces of cancellous bone (5 × 5 × 5 mm). The bone pieces were repeatedly soaked and washed with purified water until the rinsing solution was clear and bloodless. Then they were soaked in a beaker filled with purified water, and the beaker was placed into the water bath of an ultrasonic cleaner for 10 min. The soaking and ultrasonic cleaning processes were repeated three times. Following degreasing, deproteinization, and surface decalcification, the pieces were vacuum packed and subjected to irradiation sterilization (25 kGy).

Preparation of BS material: Cortical bone was prepared from goat hind leg bone using the same degreasing and deproteinization steps as the cancellous bone. After drying, a milling cutter was used to obtain cortical BS (Fig. S1A), which were collected and screened between 1 mm and 2 mm meshes, followed by soaking in decalcified solution. Then the BS were subjected to ultrasonic cleaning with purified water as described above until reaching a pH of 7.0. They also underwent irradiation sterilization in the same manner used for preparing the cancellous bone.

Preparation of MIX material: DBM bone (3 g) with a decalcified surface was mixed with BS at different volume ratios (2:1, 1:1, and 1:2) to adjust the pore size. Each sample was placed in a 20-mL syringe inside a 50-mL centrifuge tube and centrifuged at a low speed. Irradiation sterilization was also performed in the same manner used for preparing the cancellous bone.

Preparation of the enriched MIX materials used in the prospective clinical study: DBM (porosity: 50 %, pore size 350 µm) and BS were provided by Beijing Kejian Biotechnology Co., Ltd. During the surgical procedure, DBM and BS were rapidly and thoroughly combined in a precise 1:1 ratio directly in the operating field. This critical mixing step was performed intraoperatively to achieve quick, on-demand fabrication of the graft material required for implantation. The resulting homogeneous mixture was promptly processed using a specialized portable enrichment device (National Machinery Standards: 20223101740, Fuwosi, Chongqing, China). This integrated intraoperative workflow ensured efficient graft preparation and optimization in real-time, facilitating immediate clinical application during the same surgical session and minimizing preparation time outside the operating theater.

Micro-computed tomography (micro-CT) and scanning electron microscopy (SEM)

Micro-CT (scanning voltage, 60 kV; current, 165 μA; and resolution, 5 μm/pixel) was performed on a Micro-CT SkyScan 1272 system (Bruker, Belgium) to characterize the materials and quantify the bone growth on the implanted porous scaffolds. A bone defect area (length: 2.5 mm) was selected for quantitative analysis to determine the bone mineral density (g/cm³), trabecular bone volume fraction (BV/TV, %), trabecular number (mm⁻¹), and trabecular separation (Tb.Sp, mm).

The ultrastructure of the materials was assessed by SEM (SU8020, Hitachi, Japan) after immersion for 24h. Before SEM, the sample was attached to conductive carbon tape, and the sample table was placed in the sample bin of the scanning electron microscope for vacuum pumping. The scanning voltage was 3.0 or 5.0 kV, and the magnification range was 100–500×, depending on the specific sample. The characterization results of SEM were supported by Beijing Zhongkebaice Technology Service Co., Ltd.

Determination of calcium content in the different materials

The calcium content in the material was determined by inductively coupled plasma mass spectrometry (ICP-MS). Briefly, the sample was placed in a PTFE digestion tube containing approximately 5 mL HNO₃, 1 mL HF, 1 mL H₂O₂, and 1 mL HCl. The mixture was heated in an oven at 180 °C oven for 8 h and then transferred to an ICP optical emission spectrometer (ICP-OES730, Agilent, Japan) to measure the calcium content.

Determination of hydrophilicity (via contact angle measurement) and porosity

The contact angle of the scaffold material was measured with a Digidrop contact angle meter (GBX Scientific Ltd., France) at ambient temperature after a drop of water. The measurement was repeated three times, and the average value was used to evaluate the hydrophilicity of the material. The pore volume and pore distribution of the materials with corresponding pore sizes were determined using a Micromeritics 9620 mercury intrusion porosimeter (Mike Company, USA) by measuring the amount of mercury entering the pores under different external pressures.

Enrichment efficiency evaluation of the different materials

The enrichment ability of MIX was evaluated to determine the optimal mixing ratio of DBM and BS. Briefly samples of goat iliac BM (∼25 mL) were used to enrich scaffold materials using a enrichment device to simulate the clinical process. Briefly, the MIX material (5 mL) was placed between the upper and lower filter screens of the cone cup in the machine . The BM sample was injected into the cone cup, and six cycles were performed (approximately 3–5 min). After each cycle, 1 mL of BM was removed to determine the cell count and concentration using an XN-1000 Automated Hematology Analyzer (Sysmex, Japan). The concentration enrichment ratio was calculated as follows: cell shelf concentration [i.e., number of cells on shelves / scaffold volume]) / (total cell concentration) × 100 %. The enrichment ratio was calculated using the following formula: (total cells in BM before enrichment − total cells in BM after enrichment) / (total cells in BM before enrichment) × 100 % [18].

Cell activity of the different materials

Trypan Blue staining is used to distinguish live from dead cells. Prepare a single-cell suspension and mix an equal volume of cell sample (10 µL) with 0.4 % Trypan Blue solution (10 µL). Gently mix and immediately load a small volume (10–15 µL) onto a hemocytometer. After Trypan Blue staining, use a Corning cell counter (6794, USA) to automatically determine live or dead cells (green indicates live cells, red indicates dead cells).

In vitro evaluation of osteogenesis induced by gradient pore size enrichment materials

Cell proliferation experiment

Human mesenchymal stem cells (hMSCs) inherited from a previous laboratory were inoculated into the wells of a 24-well plate (5 × 10⁴ cells/well) containing DBM, BS, or MIX scaffolds. A blank control group without scaffold materials was also prepared. The cells were cultured in an incubator at 37 °C under 5 % CO₂. On days 1, 3, and 7, the proliferation ability of hMSCs on the scaffold was analyzed using a CCK8 kit (MCE, China) per the manufacturer’s instructions The absorbance was measured at 450 nm using a multifunctional enzyme-labeled instrument (BioTek, USA) to determine the level of hMSC proliferation [19].

Alizarin red staining analysis and quantitative analysis of alkaline phosphatase (ALP)

According to previous experimental methods [20], 5 × 10⁴ hMSCs/well were inoculated on DBM, BS, or MIX scaffolds and cultured in 24-well plates. After 7 days of induction culture in a specialized culture medium for osteogenic induction (Saiye, China), the supernatant and materials were removed. The cells were fixed with 4 % paraformaldehyde for 30 min, dyed with alizarin red (Beyotime Biotechnology, Shanghai, China), and gently washed three times with phosphate-buffered saline. They were photographed under microscopy and quantitatively assessed using ImageJ software (USA). For the quantitative detection of ALP, after 7 days of coculture under the above conditions, the ALP content in the culture medium was directly analyzed using an Alkaline Phosphatase Assay Kit (Beyotime Biotechnology).

Reverse transcription quantitative polymerase chain reaction (RT-qPCR) assay

RT-qPCR was used to determine the levels of mRNA expressed by osteogenic and angiogenic genes. Total RNA was isolated from hMSCs using an RNA-Quick Purification Kit (Jianshi, Beijing, China) and transcribed into complementary DNA (cDNA) using RT Master Mix for qPCR (Takara, Beijing, China). The cDNA was used as the template for qPCR using SYBR Green qPCR Master Mix (ABclonal, Wuhan, China). The relative expression of the target genes was normalized to that of GAPDH (Beyotime Biotechnology) using the 2^−ΔΔCt method. Table S1 lists the related primer sequences.

In vivo evaluation of osteogenesis induced by gradient pore size enrichment materials in vivo

Construction of a subcutaneous heterotopic transplantation model in nude mice

DBM, BS, and MIX materials (1.0 × 1.0 × 0.3 cm) were prepared in the same manner as described for the previous experiments. Nude mice (6 weeks old) were selected and anesthetized using 3 %–4 % isoflurane gas in an anesthesia induction chamber. The skin between the thigh root and the spine along the back of the nude mouse was cut, keeping the incision direction consistent with the long axis of the body. The subcutaneous tissue and muscle were separated, the materials were placed, and the wound was sutured layer by layer. The mice gradually recovered from the anesthesia at room temperature and were returned to the specific pathogen-free animal house where their vital signs were observed daily. They were euthanized after 4 weeks, and samples were collected for further investigations, such as hematoxylin and eosin (HE), Masson, and immunofluorescence (IMF) staining and micro-CT analysis. In vivo biocompatibility was evaluated by HE staining of heart, liver, spleen, lung, and kidney tissues.

Construction of a critical femur bone defect model in goats

Clean-grade male black goats (n = 5; 8–12 months old, 32–35 kg) were provided by the Experimental Animal Center of the Army Military Medical University (Chongqing, China). Each goat was operated on its left and right hind limbs for the implantation of enriched MIX materials(n=5) and unenriched MIX materials(n=5), respectively. Goat feeding was stopped 12h before the operation. The goats were anesthetized via intramuscular injection of luminacin (0.15 mL/kg). The right side of the goat was laid flat on the experimental platform to prepare the skin of the right hind limb using routine disinfection and drying. A longitudinal incision (approximately 7 cm) was made in the in skin of the right lateral femur to expose the middle portion of the femur. A small incision was made at the femoral trochanteric fossa to insert a guide pin, and the hole was reamed with a reaming drill. Using the guide needle, an intramedullary nail was inserted into the femoral medullary cavity, and the guide needle was pulled out. Two holes each were drilled at the distal and the proximal ends respectively, and four interlocking screws were tightened in turn. After examination, the middle femur was removed using a wire saw to create a 2.5-cm defect model. After washing the wound, the sterilized materials were implanted into the bone defect. The same procedure was performed for the left hind limb, with the left side of the goat laid flat on the experimental platform. Before, during, and 3 days after the operation, penicillin (80,000 U) was injected intramuscularly twice a day. Three months after the operation. the goats were anesthetized and underwent a computed tomography (CT) scan before being euthanized, and the intramedullary nails were removed.

HE and Masson staining as well as finite element reconstruction analysis were performed. Immunohistochemical (IHC) staining was carried out on formalin-fixed, paraffin-embedded sections (4–5 µm) after deparaffinization in xylene and rehydration through a graded series of ethanol. Citrate buffer (pH 6.0) was used for heat-mediated antigen retrieval, followed by endogenous peroxidase quenching with 3 % H₂O₂ for 15 min at room temperature. The sections were blocked with 10 % normal serum in Tris-buffered saline/0.1% Tween-20 (TBST) for 1h, followed by overnight incubation at 4 °C with type 1 collagen (COL-1; 1:100; Cat. No. A1352, ABclonal) and VEGF (1:50; Cat. No. A17877 ABclonal) as the primary antibody, diluted in blocking buffer (1:100). After washing with TBST, the sections were incubated with an appropriate HRP-conjugated secondary antibody for 1h at room temperature. DAB substrate was added, and the resultant staining was monitored microscopically monitored, followed by counterstaining with hematoxylin. Finally, the sections were dehydrated, cleared in xylene, mounted, and coverslipped.

Histological and IMF staining

The nude mouse subcutaneous heterotopic osteogenesis samples and goat large critical bone defect samples were fixed with 4 % paraformaldehyde for 1 week. The nude mouse samples were then decalcified with 10 % EthyleneDiamineTetraaceticAcid(EDTA) for 4 weeks, and the goat large bone samples were decalcified with 10 % HCl and EDTA because of the thick cortical bone. Images were taken and observed using a SLIDEVIEW VS200 whole slide scanning system (Olympus, Japan).

For IMF staining of subcutaneous heterotopic osteogenesis in the nude mouse model, formalin-fixed, paraffin-embedded sections were fixed and dehydrated as described in section 2.3.3. Following blocking, the samples were incubated with primary antibody followed by incubation with an appropriate secondary antibody using a conventional method. The primary antibodies used were anti-COL-1 (1:100, Cat. No. ab270993, Abcam), anti-osteopontin (OPN; 1: 100; Cat. No. AF808, R&D Systems), anti-VEGF receptor 1 (1:200; Cat. No. ab32152, Abcam), and anti-CD31 (1:100; Cat. No. AF3628, R&D Systems). The secondary antibody was purchased from Abcam(Donkey Anti-Goat IgG H&L Alexa Fluor® 647 for ab150131 and Goat Anti-Rabbit IgG H&L Alexa Fluor® 488 for ab150077). Images were taken using a laser confocal microscope (LSM 880, ZEISS).

Comparison of osteogenic efficacy between enriched MIX materials and autogenous bone in clinical practice

After assessing subcutaneous heterotopic osteogenesis in the nude mouse model and critical bone defect repair in the goat model, we decided to further evaluate the clinical repair effect and osteogenesis ability of autogenous bone compared with enriched MIX in different lumbar fusion segments of the same patient by conducting a self-controlled prospective clinical study.

Inclusion and exclusion criteria

The research protocol was approved by the Ethics Committee of the First Affiliated Hospital of PLA Army Medical University (approval No. (B)KY2025034).

The inclusion criteria were as follows: (1) male or female volunteers aged 18–80 years who had received conservative treatment for at least 1 month for lumbar degenerative diseases, such as lumbar disk herniation, lumbar spinal stenosis, and lumbar spondylolisthesis (grade I–II), during which the symptoms did not abate, with indications for two-level lumbar fusion; (2) voluntary acceptance of different bone grafting materials randomly selected and placed in different intervertebral spaces and related testing; (3) good understanding of and compliance with the experimental procedures and good communication with the researchers; and (4) signed informed consent.

The exclusion criteria were as follows: (1) active systemic infection; (2) fever or abnormal complete blood count results; (3) mental illness; (4) any conditions that may potentially affect the implantation effect of the fusion cage, such as malignant tumor, congenital malformation, fracture of the surgical site, and other basic diseases; (5) severe osteoporosis with osteoporotic fracture; (6) history of severe allergy or rejection of medical materials; (7) no indication for fusion surgery or only requiring single-segment fusion surgery; (8) insufficient tissue coverage of the surgical site; (9) poor compliance and late follow-ups; (10) participation in other clinical trials; and (11) other circumstances considered by the researcher as unsuitable for enrolment.

Surgical methods

All patients underwent surgery performed by the same surgeon, who followed the standard procedure for two-level unilateral decompression and transforaminal lumbar interbody fusion. The patients underwent general anesthesia and were placed in the prone position, and routine disinfection and draping were performed. A posterior midline incision was made, and the skin, subcutaneous tissue, and lumbar fascia were incised layer by layer. The approach was taken along the paraspinal muscles (between the longest muscle and multifidus muscle, approximately 2 cm lateral to the spinous process) to expose the bilateral laminae, superior and inferior articular processes, and transverse processes. The entry point for pedicle screw placement was determined at the intersection of the vertical line from the outer edge of the superior articular process and the bisector of the transverse process. Pedicle screws of the appropriate length and diameter were inserted. BM blood flowing out of the screw placement channel was collected for later use. Decompression was performed through the symptomatic intervertebral foramen by removing the lamina and part of the superior articular process of the inferior vertebra of the affected segment, resecting the ligamentum flavum, and fully exposing the intervertebral foramen. The dura mater and nerve roots were explored, and the posterior annulus fibrosus was incised using a sharp knife. The nucleus pulposus-like tissue was removed using a nucleus pulposus clamp, and the endplates were prepared. Throughout the entire process, BM blood and peripheral blood flowing from the removed lamina, articular processes, and prepared intervertebral space were collected for later use. After reexploring and confirming adequate nerve root decompression, the intervertebral space was irrigated, and an appropriately sized interbody cage was selected after trialing. The other affected segment was treated similarly. BM blood mixed with peripheral blood (total: 60 mL) was combined with MIX materials and underwent enrichment as described in section 2.1.1. Autogenous bone from the spinal canal and articular processes cut from the previous decompression were processed into granules. Finally, the autogenous bone and the enriched allogenic bone material were implanted into the corresponding segments, and the incision was sutured in turn.

Follow-up and evaluation indicators

Follow-up to monitor postoperative complications was conducted for 12 months via outpatient services, telephone, and mail. Pain was assessed using the visual analog scale (VAS) and the level of funtion related to lower back pain was measured using the Oswestry disability index (ODI) at 1, 3, 6, and 12 months after surgery. To evaluate segmental fusion, CT three-dimensional (3D) reconstruction of the lumbar spine was performed at 3 and 6 months after surgery. Patient characteristics (age and gender) perioperative data (intraoperative blood loss, fusion segment, and surgery time), hospitalization time, imaging findings, and the clinical efficacy of the two groups were compared. Bone formation was compared between the enriched bone segment and the autogenous bone segment using CT DICOM data, and finite element reconstruction was carried out using Mimics software. Bone formation in the intervertebral space and the Hounsfield (HU) value of the autogenous bone segment and the enriched bone segment were compared.

Statistical analysis

All experiments in this study were repeated in triplicate, and the results were expressed as the mean and standard deviation. Pairwise comparison between samples were performed using t-tests, and comparisons among groups used one-way analysis of variance. P-values < 0.05 were considered statistically significant.

Results

Characterization of materials

DBM, BS, and MIX (1:1) were subjected to micro-CT (Fig. 2A). The DBM scaffold had a larger pore size than the BS, and the pore size of MIX was moderate. The porosity distribution (Fig. 2B) of BS was mainly concentrated in the smaller-sized pores (∼0.5 × 10⁵ nm), which was close to that of MIX (1:2), while that of DBM was close to MIX (2:1). The porosity distribution of MIX materials was mainly around 0.5 × 10⁵–2 × 10⁵ nm, which was smaller than the DBM and larger than the BS pore size. The porosity of MIX materials differed according to the ratio of DBM to BS, which may have influenced the concentrations of cells and cytokines achieved following BM enrichment and bone growth during the later stage of bone regeneration. Determination of hydrophobicity using contact angle measurement (Fig. 2C) revealed that the MIX group had a relatively smaller contact angle compared with the other two materials. ICP-MS (Fig. 2D) showed that there was no statistically significant difference in the calcium content of the three materials. Quantitative analysis using the micro-CT results (Fig. 2Ea–c) showed that the direct BV/TV of DBM was higher than that of the other two groups. The bone surface-to-bone volume ratio (BS/BV) was the highest and the Tb.Sp was the lowest in the BS group. And the results of Micro CT evaluation of the MIX (1:2) and MIX(2:1) was presented in Fig. S1(B-C).

Fig. 2.

A: General view, micro-computed tomography 3D reconstruction, and scanning electron microscopy images of decalcified bone matrix (DBM), bone shavings (BS), and the new BM enrichment material (MIX), indicating that the pore size of the MIX material was moderate. B, C: Porosity of DBM, BS, and MIX shows that the MIX material with a 1:1 ratio of DBM to BS has optimal porosity. Contact angle measurement shows that all three materials demonstrate hydrophobicity, and the hydrophilic angle of MIX material is the lowest. D, E: The results of surface calcium content determination show no significant differences between the three materials. Quantitative analysis using the micro-CT results for DBM, BS, and MIX showing the properties of the materials. Yellow square: represents DBM, red triangle: BS.

Enriched efficiency of the materials

According to the enrichment efficiency results, enrichment plateaued after five cycles. The optimal enrichment ratio of DBM to BS was 1:1 (Fig. S2). SEM showed that compared with the other two materials, the MIX material retained more cells during the enrichment process (Fig. 3A). This was confirmed by the concentration enrichment ratio and enrichment ratio results (Fig. 3B, C) calculated from the results of routine blood analysis of the BM before and after enrichment. The performance of five enrichment cycles for MIX (1:1) resulted in the following enrichment ratios: leukocytes 46.85 % ± 3.25 %; erythrocytes, 41.81 % ± 2.02 %; neutrophilic granulocytes, 46.93 % ± 2.51 %; lymphocytes, 48.86 % ± 6.66 %; and monocytes, 47.38 % ± 3.21 % (Fig. S1). The SEM results further confirmed that the enrichment ability of the BS group was close to that of the DBM group, and that of the MIX group was the best. Furthermore, the live/dead staining results (Fig. 3A) and the quantitative analysis results (Fig. 3D) suggested that all scaffolds maintained cytocompatibility and that enrichment did not adversely affect cell viability.

Fig. 3.

A: Scanning electron microscopy results of decalcified bone matrix (DBM), bone shavings (BS), and the new BM enrichment material (MIX) after enrichment. The MIX material was the most effective for cell enrichment. Trypan Blue staining results before and after BM enrichment shows that the enrichment process did not affect cell activity in any of the materials (green fluorescence: live cells, red fluorescence: dead cells) B, C: Analysis of the concentration enrichment ratio and enrichment ratio of DBM, BS, and MIX. D: Quantitative analysis of living and dead cells before and after the enrichment operation of DBM, BS, and MIX.

The MIX materials promote osteogenesis in vitro

Analysis of cell proliferation using CCK8 showed that all three groups of materials promoted MSC proliferation. Furthermore, the proliferation effect of the MIX group was better than the other two groups in the early stage (day 1). However, there was no significant difference between the MIX and the BS group in the later stages (day 3 and 7). This may be related to the moderate pore size of MIX and the BS, which are more conducive to MSC proliferation. Alizarin red staining analysis of osteogenic differentiation induced in vitro for 7 days revealed that there were more calcified nodules and a higher relative ALP content in the MIX scaffold group (Fig. 4A, B). qRT-PCR targeting differentiation-related and osteogenic genes, including COL-1, OCN, OPN, and RUNX2, showed upregulated expression of all targeted genes, further confirming the results of ours in vitro osteogenesis experiments. The upregulation of RUNX2, OPN, COL-1, and OCN expression (Fig. 4Ca-d) was responsible for the better osteogenic ability of MIX material.

Fig. 4.

A: Representative images of alizarin red staining of decalcified bone matrix (DBM), bone shavings (BS), and new BM enrichment material (MIX). MIX material significantly enhances osteogenesis in vitro. B: (a) Results of quantitative analysis of the Alizarin red staining and ALP content. The average values are shown. (b) Cell proliferation activity results of the DBM, BS, MIX groups indicate that the MIX material effectively promotes cell proliferation. C: PCR verification that DBM, BS, and MIX material promote changes in the expression of osteogenesis-related genes in (a) OPN, (b) Runx2, (c)COL-1, (d)OCN.

The materials promote osteogenesis in vivo

Subcutaneous ectopic osteogenesis in nude mouse

Animal experiments were first performed using a subcutaneous ectopic osteogenesis model in nude mice in which DBM, BS, and MIX were implanted under the skin. At the end of the experiment, the implanted materials were removed. The general appearance of the samples was assessed, and they were stained using HE and Masson stain The osteogenic effect of the material in the MIX group was better than that in the other two groups (Fig. 5A, B). This was supported by the micro-CT findings (Fig. 5C), which showed that the MIX material had a better osteogenic effect in vivo than the other two materials after 4 weeks, as evidenced by the BV and BV/TV results. The BS/BV was not significantly different between the MIX and BS groups. And the Micro-CT analysis comparison of enriched and none-enriched group was presented in Fig. S4. All materials were biocompatible, and no biotoxicity in vivo was observed (Fig. S3).

Fig. 5.

A: General photograph of the decalcified bone matrix (DBM), bone shavings (BS), and new BM enrichment material (MIX) removed from the nude mouse models of subcutaneous osteogenesis. B: HE and Masson staining results of DBM, BS, and MIX materials removed from the nude mouse models of subcutaneous osteogenesis. C: Micro-CT results and quantitative analysis of DBM, BS, and MIX materials removed from nude mouse models of subcutaneous osteogenesis. The MIX material has a better osteogenesis ability and the enriched group has a better osteogenesis ability than the nonenriched groups.

The moderate pore size of the MIX material provided an environment conducive to cell proliferation. IMF staining (Fig. 6) showed higher levels of OPN and COL-1 in the MIX group than in the DBM and BS groups. Furthermore, comparison of the results before and after enrichment showed that that the enrichment process promoted effective osteogenesis and angiogenesis, with the enriched components remaining on the scaffold materials. Thus, the enrichment efficiency was highest in the MIX group.

Fig. 6.

A: Immunofluorescence staining of CD31, VEGF, OPN, and COL-1 in decalcified bone matrix (DBM), bone shavings (BS), and new BM enrichment material (MIX) removed from the nude mouse models of subcutaneous osteogenesis. B: Quantitative analysis of the immunofluorescence staining results show that the MIX material has relatively more markers related to osteogenesis and angiogenesis, and the expression in the enriched MIX group is better than that of the nonenriched MIX group.

Materials promote the repair of large critical bone defects in goats

We further verified the osteogenic effect of the MIX materials in the repair of large critical bone defects using a goat model. This animal was chosen because it has a similar bone structure to humans and undergoes comparable remodeling, transformation, and bone formation. All goats recovered from anesthesia at 2–6h after surgery. There were no signs of postoperative infection, internal fixation failure, or foreign body reaction. All goats were unable to bear weight immediately after the operation, but they were able to walk with partial weight bearing after 1 week, walk upright with weight bearing after 1 month (although there was claudication), and walk with a normal gait after 3 months.

Three months after the operation, the animals were anesthetized and a CT scan was performed. The imaging results showed a gradual increase in the density of new bone at the defect site, finally approaching that of normal bone. The enriched group had more callus and new bone formation compared with the control group (Fig. 7A). HE and Masson staining further confirmed the superior osteogenesis ability of the enriched materials (Fig. 7B). The results of the finite element reconstruction analysis, IHC staining (Fig. S5), and quantitative analysis also indicated that the enriched MIX material had better osteogenic efficiency, Although the BS/BV results of the finite element reconstruction analysis in the cortical bone have no significant differences, but the enriched group still has a higher BS/BV result (Fig. 7C-D).

Fig. 7.

A: General picture and computed tomography scan image of a representative critical bone defect model goat treated with enriched new scaffold material (MIX) or nonenriched MIX. B: Representative image of HE and Masson staining results. The enriched group shows better osteogenesis then the nonenriched group. C: Finite element reconstruction analysis indicates that the enriched MIX material has better osteogenic efficiency than the unenriched MIX material. D: The relevant quantitative results of finite element analysis data indicates that the enriched MIX has stronger osteogenic ability.

Evaluation of the osteogenic efficiency of MIX material in intervertebral fusion surgery

DBM and BS materials are both commonly used to clinically repair bone. We conducted a prospective clinical study to compare the osteogenic ability of the enriched MIX materials with that of autogenous bone materials (considered the gold standard) for spinal fusion. After applying the inclusion and exclusion criteria, eight patients were enrolled in our study. Table S2 presents the basic patient information. The X-rays and the 3D images of CT scanning at 3 and 6 months and the finite element reconstruction indicated that the bone graft material in the intervertebral space had become bone and gradually fused (Fig. 8A). Finite element quantitative analysis of the CT scans showed that there was no statistically significant difference between the HU values inside and outside the lumbar fusion cage at 3 and 6 months after the operation, indicating that the quality of the enriched bone material and the autogenous bone material was similar during early osteogenesis. This was further supported by the absence of a statistically significant difference in the total bone mass between the two time points (Fig. 8B). The VASs and ODI scores showed that the surgery had improved the patients’ quality of life (Fig. S6).

Fig. 8.

A: X-ray, computed tomography (CT), and 3D finite element reconstruction diagram showing the promotion of lumbar fusion at 3 and 6 months after the implantation of MIX bone repair materials and autogenous bone materials in different intervertebral spaces. B: CT 3D finite element analysis was used to quantitatively analyze the internal and external Hounsfield values and bone mass of the fusion cage at 3 and 6 months after implantation. The enriched MIX material has the same clinical effect as the autogenous bone graft.

Discussion

Bone transplantation is a fundamental technique in orthopedics, and the choice of bone graft material is key [18,19]. The main bone graft materials currently in use are autogenous bone, allogenic bone, or artificial bone materials. Although autogenous bone transplantation has an excellent osteogenic effect, its clinical application is limited by the injury site, risk of infection, and pain in the bone harvest area [20]. Additionally, the existing regenerative treatment approaches, such as pedicled vascularized bone grafts, the Masquelet induced membrane technique, and the Ilizarov distraction osteogenesis technique, have specific indications and limitations [4]. The gold standard is autogenous bone. However, due to the increasing clinical demand for excellent osteogenic bone materials, the development of artificial bone materials that can be substituted for autogenous bone is critical. The excellent osteogenic ability of tissue-engineered bone has resulted in it becoming a potentially important material for clinical grafts [[21], [22], [23]]. Basic experimental studies have improved our understanding of bone tissue engineering [19,24]. Although attention has been focused on clinical research and application of tissue-engineered bone, due to ethical restrictions and high costs, only a few tissue-engineered bones have been approved for clinical research. The basic research on bone tissue engineering technology provides strong theoretical support for treating bone defects. However, there are many limitations that interfere with the clinical application of tissue-engineered bone, particularly the approval policy for new scaffold materials [25]. Furthermore, the expansion and inoculation of seed cells in vitro and the maturation of tissue-engineered bone is a lengthy and costly process, which is disadvantageous for routine use [[26], [27], [28], [29], [30], [31]]. The excellent osteogenic ability of tissue-engineered bone is due to the proteins secreted by the seed cells, usually obtained from BM [[26], [27], [28], [29], [30], [31]]. Therefore, BM-enriched bone repair technology based on efficient BM use was developed. Its core principle is to concentrate BM-derived cells and cytokines with osteogenic effects into scaffold materials for the rapid construction of bone grafts with high osteogenic activity.

Compared with traditional BM injection and soaking, bone marrow enrichment technology uses several times more bone marrow than scaffold materials, and the majority of cells in BM can be enriched on the scaffold material, significantly increasing the number and concentration of BM-derived cells and cytokines and enhancing growth on the scaffold material. Numerous studies have confirmed that enrichment efficiency depends on the physical and chemical properties of the materials [[32], [33], [34]]. Hou et al. [18] modified a DBM scaffold with nanoscale self-assembling peptide RADA16-I, which reduced the scaffold pore size from 396.24 to 20.22 µm, greatly improving the physical interception efficiency (pore size effect) of SCR technology and improved the efficiency of BM enrichment. Moreover, the presence of RADA16-I results in a cell microenvironment more suited to osteogenesis, promoting interaction between the DBM and cells, regulating cell function, and further promoting osteogenesis of the DBM scaffold. Luo et al. [35] modified DBM with lysine-cyclic RGD peptide, which enhanced enrichment using SCR technology. This provided a microenvironment conducive to osteogenic induction, resulting in the retention of a higher concentration and quantity of MSCs. Xing et al. [36] modified DBM scaffold using a layer-by-layer self-assembly technique to construct a new composite material. This material was based on the principle of combining physical interception and chemical adsorption, which significantly improved the efficiency and specificity of SCR. It also provided a microenvironment conducive to bone tissue growth, substantially promoting the migration, proliferation, and osteogenic differentiation of BM MSCs, thereby accelerating the bone repair process. Tang et al. [34] designed a collagen-binding domain containing the core functional amino acid sequence of laminin α4. This was used to modify DBM to enhance scaffold function so that the scaffold contained bioactive ligands capable of binding to laminin receptors on MSCs. Combined with SCR technology, this functionalized scaffold provides a microenvironment with better cell adhesion, further promoting bone formation and vascularization. However, due to the restrictions of administrative approval, the above-mentioned basic research materials cannot be immediately applied in clinical practice, and there is also no relevant conclusion about which pore size is the most conducive to enrichment.

It is crucial to determine the pore size most conducive to the proliferation of enriched osteogenic cells. In this study, BS with a small pore size was prepared using original bone shaving technology and mixed in different proportions with DBM scaffold material, which has with a relatively larger pore size, to create MIX, a new BM enrichment material with controllable pore size. DBM and BS are allogenic bone repair and transplantation materials that have received administrative approval and are widely used in clinical practice. Investigation of the pore size distribution of the controllable pore size enrichment material (Fig. 2A) revealed that the pore size of MIX was substantially smaller than that of DBM but larger than that of BS (Fig. 2B). Thus, the cortical BS were interspersed and stacked together within DBM pores or between the DBM, resulting in relatively close contact. Regarding the process of enrichment, the optimal number of cycles has yet to be determined and may be influenced by differences between instruments and enrichment materials [35,36]. To explore the enrichment efficiency of controllable pore size BM enrichment materials, including the optimal number of cycles and MIX material ratio, we carried out an experiment to investigate enrichment efficiency (Fig. S2). The MIX scaffold demonstrated superior cell retention capacity when the material ratio was 1:1, and the enrichment of all the materials plateaued after five cycles. The more moderate pore size of the MIX group enhanced the effective interception of cells in the enrichment process more than the DBM group, which had a larger pore size. Furthermore, the MIX pore size did not affect BM circulation during the enrichment process. However, the smaller pore size of the BS significantly slowed down the enrichment process. After determining the optimal material proportion and enrichment times for the MIX group, the osteogenic capacity of the materials was further studied in vitro. DBM is a biomaterial with low stiffness and good biocompatibility that provides a suitable microenvironment for MSCs, promoting their proliferation and differentiation. It reduces the release of inflammatory mediators by promoting the polarization of M2 macrophages, thereby mitigating the inhibitory effect on MSC proliferation [37]. Our vitro experiments demonstrated that MSCs become strongly attached to the MIX scaffold, which is beneficial to cell proliferation and differentiation [38]. All three materials contain calcium ions, which promote MSC proliferation through various mechanisms, including regulating cell signal transduction, influencing the cell cycle process, regulating the expression of related genes, and influencing the extracellular matrix to promote MSC proliferation. More MSCs also led to more calcium nodules and relatively higher ALP content. The PCR results were consistent with those of the in vitro osteogenesis experiments, showing increased expression of OPN, OCN, RUNX2, COL-1, and other genes.

We used a subcutaneous heterotopic osteogenesis model to evaluate the osteogenic properties of materials containing BM-derived cells [39]. The enrichment process enhanced the retention of a large number of osteogenic-associated cells on the scaffold materials, and the micro-CT results showed that the osteogenic effect of the enriched group was better than that of the nonenriched group. The MIX material (1:1) had the best osteogenic effect, which may be related to the fact that the MIX scaffold retained more cells beneficial for osteogenesis and angiogenesis due to its appropriate pore size and porosity, which provided a suitable microenvironment for osteogenesis and angiogenesis. HE and Masson staining showed that more new bones and blood vessels were present in the enriched group than in the nonenriched group. Moreover, the IMF staining results indicated that VEGF, CD31, and other angiogenesis-related markers in the MIX group were improved. It has been confirmed that angiogenesis is crucial for the repair of bone defects because it provides sufficient nutrition for the regeneration of new bones. Thus, there is a coupling effect between osteogenesis and angiogenesis in vivo. Furthermore, a large number of studies have also confirmed that Hif-1α and VEGF are important regulators in the coupling process of angiogenesis and osteogenesis [40]. Under the coculture system with hMSCs provided by the selective cell enrichment scaffold, the expression of HIF-1α and VEGF, two key genes that regulate bone formation, was also upregulated in EPCs. However, there was no significant difference between the results of the enriched and nonenriched groups (Fig. S4). This may be because the model of heterotopic osteogenesis was insufficient to reflect the difference between enriched materials and nonenriched materials using micro-CT. Thus, animal models with a higher evidence level should be used to verify the efficiency of enrichment.

The repair of large segmental bone defects, particularly critical segmental bone defects, remains one of the most challenging problems in orthopedics [41]. The critical size defect (CSD) refers to the smallest bone defect that cannot be repaired using its own regenerative capacity or defects where the bone regeneration is <10 % during its growth cycle, which easily leads to a poor wound environment and biomechanical instability of the defect site. If affected by factors such as metabolic diseases, abnormal hormone secretion, poor surgical techniques, and malnutrition, CSDs become incapable of healing on their own and are more prone to bone nonunion [42]. To further simulate and evaluate the application of MIX, we designed a preclinical animal experiment of critical bone defects in goat femurs. The CT and histological staining results indicated that the MIX material with enrichment showed better osteogenic ability than the MIX material without enrichment.

One of the reasons why we choose DBM and BS as the materials was because their clinical use has been ethically approved, meaning that they can be directly used in clinical bone grafting fusion surgery. However, the mixing of these two materials would alter the physical properties of the scaffold, and this could not be accurately simulated using 3D printing materials. After passing ethical review, we performed a prospective self-controlled study. The follow-up results showed that the patients had a good prognosis and they were able to gradually return to normal life 3 months after the operation, and there were no related complications, such as infection or broken screws.

The finite element analysis revealed that there was no statistical difference in the total amount of osteogenesis between the MIX material and autogenous bone at 3 and 6 months after the operation, indicating similar osteogenic ability in the early stage. Thus, MIX material can be an effective substitute for autogenous bone in intervertebral fusion surgery, providing a therapeutic option. However, the enrichment process used in our experiments is nonselective. If the selectivity of the enrichment process could be enhanced to target only osteogenic components, it may be more conducive to the osteogenesis of selective cell-enriched materials, and this is a direction for future research.

Conclusion

We demonstrated the preparation and osteogenic effect of MIX-enriched materials with controllable pore size. We prepared BS and mixed them with different ratios of DBM to construct MIX material with an adjustable pore size. We postulated that this new scaffold material would be beneficial to the enrichment of BM-derived osteogenically active components, and this was supported byte results of ours in vitro and in vivo osteogenic experiments. Finally, in a self-controlled prospective clinical study of lumbar interbody fusion surgery, it was confirmed that there were no statistically significant differences between BM-enriched MIX material and autogenous bone in early osteogenesis. Although the sample size was small, this study demonstrates that selective BM cell-enriched MIX scaffold is a clinically viable alternative to autogenous bone and has excellent osteogenic ability. Our findings are expected to arouse broad interest among researchers and orthopedists, providing a new therapeutic choice for treating bone defects and bone reconstruction.

Compliance with ethics requirements

Ethical approval and consent to participate

This study was conducted in accordance with the ethical standards of the The First Affiliated Hospital of Army Military Medical University((B)KY2025034) and the 1964 Helsinki Declaration (and its later amendments).

For animal studies: All procedures followed the ARRIVE guidelines and were approved by the Army medical university(AMUWEC20232848). Efforts were made to minimize animal suffering.

CRediT authorship contribution statement

Qiandong Yang: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Bo Yu: Methodology, Funding acquisition, Data curation. Chunyu Liu: Methodology, Funding acquisition, Data curation. Jiangling Zhou: Software. Yang Zhang: Methodology, Investigation, Formal analysis, Data curation. Ming Yang: Methodology, Investigation, Formal analysis, Data curation. Sihao He: Software. Juan Cai: Methodology, Conceptualization. Qijie Dai: Validation, Conceptualization. Zhenzhen Tang: Visualization, Validation. Jianzhong Xu: Writing – review & editing, Supervision, Resources, Funding acquisition, Conceptualization. Zhongrong Zhang: Writing – review & editing, Supervision, Resources, Funding acquisition, Conceptualization. Tianyong Hou: Writing – review & editing, Supervision, Resources, Funding acquisition, Conceptualization.

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 are the Supplementary data to this article: