The recombinant BMP-2 loaded silk fibroin microspheres improved the bone phenotype of mild osteogenesis imperfecta mice

Mice models

In this investigation, two genotypes of OI-modeled mice with mild disease phenotypes were utilized. The wild-type (WT) C57BL/6 mice were purchased from the Laboratory Animal Center of the Academy of Military Medical Science (Beijing, China). B6C3Fe a/a- Col1a2^oim/J mice (#001815) bought from Jackson Laboratory (Bar Harbor, ME, USA) were backcrossed with WT C57BL/6 mice for at least 11 generations. The heterozygotes were marked as Col1a2^oim/+mice. Another genotype of OI mice was Col1a1^+/−365 which mimics OI type I (Liu et al., 2019). Col1a1^+/−365 and Col1a2^oim/+mice were bred and identified following previously described procedures (Liu et al., 2019; Carleton et al., 2008)). All mice on a C57BL/6 background were bred under specific pathogen-free (SPF) conditions and the protocols of the study were approved by the Animal Care and Use Committee of Tianjin Medical University (TMUaMEC 2017012). This research employed only male mice.

rBMP-2-loaded SF microspheres preparation and morphologically characterization

The lyophilized SF microspheres were bought from Suzhou Simatech Biotechnology Co., Ltd (China). A suspension of 2 mg/mL was prepared by dissolving the microspheres in a PBS solution using ultrasound. 0.5 µg of recombinant BMP-2 (rBMP-2; Peprotech, Cranbury, NJ, USA) was added into one mL of microsphere solution to produce a 0.5 µg rBMP-2/mL fibroin mixture, which was subsequently incubated for 30 min at 4 °C. Subsequently, the rBMP-2/SF microspheres were acquired through an overnight lyophilization process. The composite microspheres were sterilized by radiation. Both blank SF and rBMP-2/SF microspheres were dissolved in PBS to prepare a 0.5 mg/mL solution in the following experiments.

The surface morphology of unloaded and loaded microspheres dissolved in PBS buffer was characterized by scanning electron microscopy (SEM). The solution of two kinds of spheres was centrifuged (1500 rpm/min, 5 min at 4 °C) and the supernatants were discarded. Then precipitates were resuspended in one mL sterile water separately. The SEM analysis was performed on gold-coated air-dried samples using a conventional SE2 detector (accelerating voltage of 2.00 kV) by GeminiSEM 300 microscope (Carl Zeiss Microscopy S.L., Oberkochen, Germany). The particle size of spheres was measured by ImageJ software.

Adipose-derived mesenchymal stem cells (ADSCs) isolation, culture, and identification

ADSCs were used for testing cytotoxicity and the osteogenic activity of rBMP-2 released from the loaded spheres. ADSCs were isolated from 4-week-old WT and OI-modeled mice, following the methodology outlined in our previous publication (Liu et al., 2021). Briefly, the epididymis and groin adipose tissue were obtained and subjected to digestion using 0.2% type I collagenase (Sigma-Aldrich, USA). Subsequently, the cells underwent filtration and were cultured in αMEM medium (Gibco, Billings, MT, USA), which was supplemented with 15% fetal bovine serum (FBS; Gibco) and 1% penicillin and streptomycin (PS; Gibco). Cells were passaged once the confluence reached 80%–90%. The three types of ADSCs were designated as ADSCs^WT, ADSCs^+/−365, and ADSCs^oim/+, respectively. Passage 3 (P3) generation of ADSCs after identification was used in this investigation.

P3 generation of ADSCs was identified via the detection of surface markers by flow cytometry (FCM) and differentiation ability assay. The cells were subjected to incubation with distinct antibodies, namely phycoerythrin (PE) conjugated-CD29 antibody, PE conjugated-CD44 antibody, PE conjugated-Sca1 antibody, fluorescein isothiocyanate (FITC) conjugated-CD45 antibody, and PerCP conjugated-CD90 antibody, all of which were obtained from eBioscience, USA. The incubation was carried out for 30 min at 4 °C. The detection of positive cells was performed using FCM. To analyze the data, FlowJo (Version 7.6.1) was employed. To evaluate the differentiation potential, ADSCs were cultivated for 14 days in the osteogenic, adipogenic, and chondrogenic induction medium (Cyagen, Santa Clara, CA, USA). Alkaline phosphatase (ALP; Solarbio, Beijing, China), Oil Red O (Solarbio), and Alcian blue (Solarbio) staining were used to determine the results.

Cell viability assay

The cell viability assay using 3-(4, 5-Dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) method (Ghasemi et al., 2023) was used to determine the cytotoxicity of SF microspheres. ADSCs were seeded in 96-well plates at a concentration of 7000 cells/well and cultured in a 37 °C, 5% CO₂ incubator for 24 h. The supernatant was then replaced with 200 µL of medium supplemented with 0, 0.5, 1, 2.5, 5, 25, 50, and 100 µg of SF microsphere. A total of 20 µL of MTT reagent (Solarbio) was added to each well after a 24-hour incubation period, and the cells were then continuously incubated for 4 h. The medium was then taken out, and 150 µL of DMSO (Solarbio) was used per well to dissolve the crystals. The optical density (OD) value at 490 nm was determined using the Microplate Reader (Bio-Rad, Hercules, CA, USA).

In vitro release study

To study the release behavior, a mixture was prepared by immersing 0.3 mg rBMP-2/SF microspheres containing 0.075 µg rBMP-2, in three mL of PBS. The suspension was subjected to agitation while being placed in a water bath maintained at a temperature of 37 °C. A total of 100 µL of supernatant was collected at 1, 3, 6, 12, 24, 48, and 72 h, and then every three days for the next 30 days. The samples were centrifuged (13,000 rpm, 10 min), filtered by 0.22 µm membranes, and stored at −20 °C until the concentration was measured. After each sampling, an equivalent volume of PBS was added. The enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN, USA) was utilized to detect the concentration of rBMP-2 in all samples following the manufacturer’s instructions. All operational steps are completed under sterile conditions to avoid contamination.

The biological activity of rBMP-2/SF microspheres by real-time quantitative polymerase chain reaction (qPCR)

To identify the osteoinductive activity of factors released from rBMP-2/SF microspheres, P3 ADSCs from OI-modeled mice were co-cultured with 20 µL of blank SF or rBMP-2/SF microspheres for 7 days. The medium was replaced on the third day and supplemented with an equivalent dosage of microspheres. Defective and wild-type ADSCs under normal culture served as control cells. ADSCs^WT under normal culture conditions acted as the control. The extraction of total RNA from ADSCs was performed utilizing Trizol reagent (Invitrogen, USA) following the manufacturer’s instructions. Reverse Transcription of mRNA to cDNA was performed using HiScript III RT SuperMix for qPCR (Vazyme, Nanjing, China). Real-time qPCR was performed using AceQ qPCR SYBR Green Master Mix (Vazyme). The procedures were established using the protocols reported in a prior report (Liu et al., 2019). The mRNA expression levels of several osteogenesis-related genes were normalized to Glyceraldehyde-3-phosphate dehydrogenase (Gapdh). The primers given in Table 1 were produced by Sangon Biotech Co., Ltd (Shanghai, China).

Animal studies

A total of 6 WT and 48 OI modeled mice, each 4 weeks old, were divided into the following groups: (1) WT + PBS group (n = 6) ; (2) Col1a1^+/−365/ Col1a2^oim/+ + PBS group (n = 6); (3) Col1a1^+/−365/ Col1a2^oim/++ SF group (for each genotype n = 6); (4) Col1a1^+/−365/ Col1a2^oim/++ rBMP-2 group (for each genotype n = 6); (5) Col1a1^+/−365/ Col1a2^oim/++ rBMP-2/SF group (for each genotype n = 6). Each mouse in the study received a single daily injection of PBS, rBMP-2 solution (0.5 µg/mL in PBS), SF microsphere solution, rBMP-2/SF microsphere solution, or rBMP-2 solution into the left femur. The right side femur received no medical treatment. The course of treatment lasted for two weeks. All mice were euthanized two weeks following the last injection, and the femurs were collected for later analysis.

Micro-computed tomography (Micro-CT)

The micro-CT system (Scanco µCT-40; Bruttisellen, Zurich, Switzerland) was utilized following the instructions provided, to identify the femoral microstructure post-treatment by rBMP-2/SF microspheres. The attached software (CTAn and DataViewer) performed a three-dimensional analysis and calculation on scanned slice data. The region of interest (ROI) was selected from the proximal-distal growth plate, where the epiphyseal cap disappeared and continued for 200 slices to the end of the femur. The trabecular bone parameters assessed in this study included bone volume / total volume (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), and trabecular spacing (Tb.Sp). Cortical bone parameters at mid-diaphysis were assessed through the measurement of cortical thickness (ct.Th) and cross-sectional area (CSA).

Histological and immunohistochemical (IHC) staining

The femur samples were first fixed for three days in a 4% formaldehyde buffer, then decalcified for 21 days in a 10% EDTA buffer. The samples were then sectioned into tissue slices with a thickness of 5 µm and immersed in paraffin. Hematoxylin and eosin (H&E) staining (Solarbio) was used following the established methodology to evaluate the histological morphology.

IHC staining of Bone γ-carboxyglutamate protein (Bglap) was carried out for in vivo quantification of osteoblastogenesis The sections were dewaxed, hydrated, and subjected to antigen retrieval by microwave heating method at 96 °C for 30 min. The sections were then subjected to treatment with 0.1% Triton X-100 for 30 min, followed by blocking with 5% BSA for another 30 min at room temperature after cooling. The samples were treated overnight at 4 °C with rabbit-derived anti- Bglap (1:200; Bioss, China). HRP-conjugated goat anti-rabbit IgG (1:500; Bioss, Beijing, China) was employed as secondary antibodies the next day. Positive signals were detected through the utilization of the universal DAB color development kit (Beyotime, Jiangsu, China). The integral optical density (IOD) of Bglap positive signals across all sections was determined by Image-Pro Plus software (Media Cybernetics, Rockville, MD, USA). All the results were obtained from a minimum of six randomly selected fields for each sample.

Statistical analysis

The statistical analysis of the data was performed using SPSS 17.0 software (SPSS Inc., Chicago, IL, USA), and the results were presented as mean ± SD. An ordinary one-way ANOVA of variance was used when more than two groups were compared. P < 0.05 was considered statistically significant.

The rBMP-2/SF microspheres sustainably released rBMP-2 over 15 days

The surface morphology of unloaded and loaded SF microspheres was observed through Scanning electron microscopy as depicted in Fig. 1A. The sphere solutions have been stored at −20 °C for a long term which might affect the morphology of microspheres. The SEM images showed that most of the unloaded and loaded spheres were spherical with rough surfaces (Fig. 1A). Some spheres manifested damaged surface morphology (Fig. 1A), which might be related to their degradation following storage in PBS. The particle size of blank and loaded spheres was 26.16 ± 10 µm and 26.42 ± 9.28 µm respectively (Fig. 1B). Overall, there was no significant difference in surface morphology and size between the two kinds of microspheres. The MTT assay was utilized to determine the cytotoxicity of SF microspheres. The results indicated that the introduction of SF particles ranging from 0.5 µg to 100 µg did not have any significant impact on the cell viability, as depicted in Fig. 1C. This observation indicates that the SF particles exhibit favorable biocompatibility. The results of the release experiment suggested that the SF microspheres loaded with rBMP-2 exhibited sustained release of the rBMP-2 for a minimum duration of 15 days in vitro (Fig. 1D). The cumulative release curve shown in Fig. 1D indicated that the release pattern of rBMP-2 exhibited an initial rapid release within the first three days (22.15 ± 2.88% of the loaded factor), followed by a transition to a slower and more consistent release, that persisted until the 15th day. The cumulatively released rBMP-2 did not change a lot post 15th day. After 15 days, 28.72 ± 0.96% of loaded rBMP-2 was released.

Identification of ADSCs

The results depicted in Fig. 2A indicate that the P3 ADSCs exhibited a notable expression of CD44, CD29, CD90, and Sca1 while demonstrating minimal expression of CD45. These findings are in accordance with the typical features of mesenchymal stem cells (MSCs). The induction differentiation assay further confirmed their osteogenic, adipogenic, and chondrogenic differentiation capacities (Fig. 2B).

The rBMP-2/SF microspheres promoted the osteogenic differentiation of ADSCs derived from OI-modeled mice

The bioactivity of rBMP-2 released from rBMP-2/SF microspheres was determined by analyzing the osteogenesis-related gene expression of mutant ADSCs cultured with rBMP-2/SF microspheres for 7 days. qPCR was employed to detect the mRNA expression levels of Alp, Col1a1, Runx2, and Bglap in stem cells. The results confirmed that the mutant ADSCs exhibited a reduced expression of the aforementioned genes in comparison to the normal stem cells (Fig. 3). Compared to untreated mutant cells, the transcription of these genes was significantly upregulated in defective stem cells following coculturing with rBMP-2/SF microparticles instead of blank SF spheres (Fig. 3). These results demonstrated that rBMP-2 released from rBMP-2/SF microspheres exhibited effective osteoinductive activity.

Infusion of rBMP-2/SF microspheres improved femoral microstructure of OI-modeled mice

The structural parameters of murine femurs were analyzed using Micro-CT. Figure 4 depicts the reconstructed 3D images of each group of mice. The Col1a1^+/365 mice with a null Col1a1 allele closely mimic the OI type I (Liu et al., 2019). Another OI mouse model, Col1a2^oim/+, with a mutation(c.3978del) in the Col1a2 gene, resembles OI type IV (Saban et al., 1996). In accordance with earlier findings (Liu et al., 2021), it was observed that both PBS-treated OI-modeled mice exhibited significant loss of trabecular and cortical bone in comparison to WT mice treated with PBS (Fig. 5). The injection of blank SF spheres did not affect any skeletal parameters (Fig. 5). Compared to PBS-treated mice of the same genotype, a few parameters improved significantly in OI-modeled mice that received rBMP-2 factor alone (Figs. 5A, 5C, 5E–5H and 5K). Both OI-modeled mice treated with rBMP-2/SF microspheres demonstrated substantially enhanced trabecular and cortical performance compared to mice of the same genotype treated with PBS only (Fig. 5).

Infusion of rBMP-2SF/ microspheres induced bone formation of OI-modeled mice

H&E staining showed the microanatomy of the bone in each mouse. Compared to PBS-treated WT mice, both Col1a1^+/−365 and Col1a2^oim/+ mice appeared to have fewer and smaller trabeculae, particularly at the lower edge of the growth plate (Figs. 6A and 6B). Quantitative results of bone area (B.Ar) and bone perimeter (B.Pm) showed no significant changes in the histological performance of blank SF microspheres or rBMP-2 treated-OI mice when compared to the same genotype treated with PBS (Figs. 6C–6F). Mice treated with rBMP-2/SF microspheres had significantly larger B.Ar and B.Pm than the same genotype treated with PBS, SF particles, or rBMP-2 (Figs. 6C–6F). These results indicated that rBMP-2/SF spheres effectively promoted bone formation in the two genotypes of OI mice.

Infusion of rBMP-2/SF microspheres promoted Bglap expression of OI-modeled mice

Bglap, also known as osteocalcin, is primarily produced by osteoblasts and regulates bone calcium metabolism (Komori, 2020). The Bglap expression can reflect the extent of bone formation. Immunohistochemical staining of Bglap demonstrated that OI mice had a much lower level of Bglap in the femurs than PBS-treated WT mice (Fig. 7). Blank SF particles or rBMP-2 treatment did not increase Bglap expression in OI mice (Fig. 7). Infusion of rBMP-2/SF microspheres significantly increased Bglap expression in comparison to the same genotype treated with PBS, SF particles, or rBMP-2 (Figs. 7C and 7D). These results indicated that rBMP-2/SF microspheres could promote bone metabolism and bone formation in both OI mice.