Bone Generation Following Repeated Administration of Recombinant Bone Morphogenetic Protein 2

Abstract

Background:

The delivery of recombinant human bone morphogenetic protein 2 (rhBMP2) by using various carriers has been used to successfully induce bone formation in many animal models. However, the effect of multiple administration of rhBMP2 on bone formation and BMP2 antibody production has not been determined. Our aim was to examine the bone formation activity of rhBMP2 and serum levels of anti-BMP2 antibodies following the repeated administration of rhBMP2 in mice.

Methods:

Absorbable collagen sponges or polyphosphazene hydrogels containing rhBMP2 were subcutaneously implanted or injected into one side on the back of six-week-old C57BL/6 mice. Three or 4 weeks later, the same amount of rhBMP2 was administered again with the same carrier into the subcutaneous regions on the other side of the back or into calvarial defects. The effects of a single administration of rhBMP2 on the osteoinductive ability in the ectopic model were compared with those of repeated administrations. In vivo ectopic or orthotopic bone formation was evaluated using microradiography and histological analyses. Serum concentrations of anti-rhBMP2 antibodies were measured by ELISAs.

Results:

Re-administration of the same amount of rhBMP2 into the subcutaneous area showed a comparable production of ectopic bone as after the first administration. The bone forming ability of repeated rhBMP2 administrations was equal to that of single rhBMP2 administration. The administration of rhBMP2 into calvarial defects, following the first subcutaneous administration of rhBMP2 on the back, completely recovered the defect area with newly regenerated bone within 3 weeks. Repeated administration of rhBMP2 at 4-week intervals did not significantly alter the serum levels of anti-BMP2 antibodies and did not induce any inflammatory response. The serum obtained from rhBMP2-exposed mice had no effect on the ability of rhBMP2 to induce osteogenic gene expressions in MC3T3-E1.

Conclusion:

We suggest that the osteoinductive ability of rhBMP2 is not compromised by repeated administrations. Thus, rhBMP2 can be repeatedly used for bone regeneration at various sites within a short duration.

Electronic supplementary material

The online version of this article (10.1007/s13770-020-00290-4) contains supplementary material, which is available to authorized users.

Introduction

Bone morphogenetic protein 2 (BMP2) is a potent osteoinductive cytokine that plays a critical role in bone regeneration and repair [1–3]. BMP2 activates endochondral ossification by inducing the differentiation of mesenchymal stem cells (MSCs) into chondrocytes and osteoblasts, and induces bone regeneration in various animal models as well as in several clinical applications [4]. Especially, the clinical use of BMPs is actively considered in orthopedic and oral/maxillofacial fields when dealing with bone fracture, spinal fusion, alveolar cleft defects, and craniofacial bone defects. Recombinant human BMP (rhBMP)-containing products have been approved by the FDA for the treatment of several spinal disc diseases and open tibial fractures [5]. Owing to the developments in recombinant DNA technology, rhBMPs derived from Escherichia coli and mammalian cells have been obtained, and adequate production of rhBMPs for clinical utilization is possible [3, 6–8].

Although the in vitro osteogenic effects of BMPs are observed at very low doses, for clinical use, large amounts of rhBMPs are required for bone formation. As with all therapeutic proteins, treatments involving high concentrations of rhBMP2 have the potential to elicit immune responses with adverse effects [9–11]. Several clinical trials involving rhBMPs have reported the results of immunogenicity testing, most of which have demonstrated low antibody formation rates [12]. However, most of these trials were performed for orthopedic applications, and only single treatment outcomes were demonstrated. Under current clinical circumstances, the application of rhBMP2 in dentistry is expected to be repeated and frequent, particularly in the placement of dental implants. However, the clinical significance of exogenously re-administered rhBMP2 for bone regeneration and immune response remains unclear. Furthermore, it is unknown whether the bone formation ability of rhBMP2 is maintained even when it is re-administered into the craniofacial region within a few weeks of administration into other regions.

Our aim was to examine whether repeated administration of rhBMP2 alters the serum levels of anti-BMP2 antibodies and affects the bone formation ability of re-administered rhBMP2 in the cranial and back regions of mice. We found that repeated administration of rhBMP2 can induce orthotopic and heterotopic bone generation without increasing inflammatory responses in mice. We suggest that repeated rhBMP2 administration retains its bone formation activity in various clinical applications such as in orthopedics, craniofacial, and dentistry.

Materials and methods

Materials

rhBMP2 was purchased from Cowellmedi Co., Ltd. (Busan, Korea) and dissolved in PBS (final concentration 1 μg/mL) according to manufacturer’s instructions. Absorbable collagen wound dressing sponge (CollaDermTM) was obtained from Bioland (Ochang, Korea) and cut into pieces of 0.5 cm × 0.5 cm × 0.3 cm for experimental use. A thermosensitive absorbable polyphosphazene hydrogel (15%) was fabricated and prepared for the delivery of rhBMP2 as described in a previous study [13].

Animal preparations

All animal studies were reviewed and approved by the Animal Ethics Committee of Chonnam National University (CNU IACUC-YB-2017-73). Six-week-old male C57BL/6 mice were purchased from Damool Science (Daejeon, Korea). Mice were randomly divided into six experimental groups as described in Table 1 and Figure S1 (n = 3–5 per group). Animals were anesthetized by intraperitoneal injection of a mixture of Zoletil (30 mg/kg; Virbac Lab, Carros Cedex, France) and Rompun (10 mg/kg; Bayer Korea Ltd, Seoul, Korea). For the subcutaneous delivery of rhBMP2 into the back, a sagittal incision of 1 cm was made on the back, and the mice were implanted with either rhBMP2-soaked collagen sponges or control collagen sponges. After 4 weeks, bone formation was monitored using dual energy X-ray absorptiometry (DEXA) scanning system (InAlyzer, MEDIKORS Inc., Seoul, Korea) and a second implantation of rhBMP2/collagen sponge was then performed. In another two groups, each mouse was shaved and injected with 200 μL of polyphosphazene hydrogel containing 7 μg rhBMP2 or control hydrogel at 4-week intervals, as described previously [13]. To examine the effect of repeated administration of rhBMP2 on calvarial bone regeneration, a sagittal incision of 1 cm was made on the back and rhBMP2-soaked collagen sponges were implanted. After rhBMP2-induced ectopic bone formation was confirmed by two-dimensional (2D) soft X-ray radiographs (Hitex Ltd., Osaka, Japan), calvarial bone defects were generated and an equal amount of rhBMP2 was administered as follows. Briefly, a sagittal incision was made on the scalp, and the calvarium was exposed. A critical-sized bone defect was generated using a trephine bur with an inner diameter of 5 mm (Fine Science Tools, Foster City, CA, USA) under low speed drilling and cool saline irrigation conditions. The defects were filled with rhBMP2-soaked collagen sponges. In the control group, PBS-soaked collagen sponges were implanted at the first subcutaneous site and the second calvarium defect site. Six weeks after the first surgery, the animals were euthanized by CO₂ asphyxiation, then tissue and blood specimens were harvested from each group of mice. The specimens (bones and blood) collected were used for further studies, as described below. Table 1 summarizes the experimental groups, based on regions and order of rhBMP2 delivery.

Table 1.

Experimental groups for rhBMP2 delivery with absorbable collagen sponge carriers

Experimental model	Group	Carrier	1st Administration (Subcut site 1)	2nd Administration (Subcut site 2)
Ectopic model	I	C-sponge	PBS	PBS
	II	C-sponge	rhBMP2 (7 μg)	PBS
	III	C-sponge	PBS	rhBMP2 (7 μg)
	IV	C-sponge	rhBMP2 (7 μg)	rhBMP2 (7 μg)
Orthotopic model	V	C-sponge	PBS	PBS
	VI	C-sponge	rhBMP2 (7 μg)	rhBMP2 (7 μg)

C-sponge Collagen-sponge, Subcut site subcutaneous implant site

2D X-ray and micro-computed tomography (micro-CT) scanning

To identify bone generation following the first administration of rhBMP2, microradiographic analysis was performed in living mice using a DEXA scanning (Fig. 1A) or 2D soft X-ray apparatus (Hitex Ltd., Japan) and diagnostic X-ray film (X-OMAT V, Kodak, Rochester, NY, USA) (Fig. 2A and Figure S2). The operating conditions for 2D soft X-ray imaged were as follows: tube voltage, 35 kVp, tube current, 400 μA; and exposure time, 45 s. For 3-dimensional (3D) quantitative analysis, each specimen was scanned using a micro-CT apparatus (Skyscan 1172; Bruker, Kontich, Belgium) in the cone-beam acquisition mode. The X-ray source was set at 50 kV and 200 μA with a 0.5-mm aluminum filter at 17.09 μm resolution. The exposure time was 1.2 s. Four hundred and forty-nine projections were acquired over an angular range of 180° (angular step; 0.4°). The image slices were reconstructed using the NRecon program (version 1.6.2.0, Bruker, Belgium), and bone volume and bone mineral density (BMD) were measured using the CT-Analyzer program (version 1.10.0.5, Bruker). 3D-surface rendering images were obtained using Mimics software 14.0 (Materialise NV, Leuven, Belgium).

Fig. 1.

Repeated administration of rhBMP2 induces subcutaneous ectopic bone generation in mice with efficacy equivalent to that of a single administration. rhBMP2 (7 μg) with absorbable collagen sponges was administered once or twice at a 4-week interval into the subcutaneous spaces in the back of mice, as mentioned in Table 1. A Radiographical findings. X-ray images or micro-CT (2D and 3D micro-CT) images were obtained 8 weeks after the first administration. B Total volume of newly formed ectopic bone and bone mineral density (BMD) at the primary or secondary administration regions were measured by using NRecon and CT-analyzer software programs. C Histological analysis. The specimens used in radiographical analyses were prepared for hematoxylin and eosin staining and micro photographed. There were no significant changes in the volumes and structures of newly formed bones between the primary and secondary treatment with rhBMP2. ns, not significant (n = 4); NB, new bone; CS, collagen sponge

Fig. 2.

Repeated administration of rhBMP2 effectively regenerates cranial bone in critical-sized calvarial defects of mice. A Microradiographic (soft X-ray) and micro-CT findings. rhBMP2 (7 μg) with collagen sponges were primarily implanted into the subcutaneous space on the back of mice, and after a 3-week interval, equal amounts of rhBMP2 with collagen sponges were surgically implanted into calvarial defects in the same mice (Group VI). In the control group, PBS-soaked collagen sponges were implanted at the first subcutaneous site and second calvarium defect site (Group V). Radiographical images of the mice or isolated tissues 3 weeks after secondary calvarial administration. B Volumetric analysis. Volumes of newly formed bones at the primary (left) or secondary (right) administration regions were measured using NRecon and CT-Analyzer software programs. C Histological analysis. All specimens from radiographical analyses were prepared for hematoxylin and eosin staining and microphotographed. Repeated use of rhBMP2 in critical-sized calvarial defects effectively regenerated cranial bone, even after primary administration of rhBMP2 induced ectopic bone in a subcutaneous region of the back. **p < 0.01 compared to the control group (n = 5)

Histological analysis

All specimens were fixed using 10% neutral buffered formalin (Sigma-Aldrich, Inc., St. Louis, MO, USA) for 3 days, decalcified in a rapid decalcifying solution (Calci-Clear Rapid, National Diagnostics, Atlanta, GA, USA) for 10 days, and then embedded in paraffin and cut into 5-μm thick serial slices. The sections were deparaffinized in xylene at room temperature for 20 min, and then rehydrated using a graded series of alcohols. The sections were then stained using hematoxylin and eosin (H&E). The H&E-stained sections from each group were then examined under a light microscope (Leica, Wetzlar, Germany) to evaluate new bone formation.

Measurement of serum anti-rhBMP2 antibody concentration and serum inflammatory cytokines levels

The changes in blood anti-BMP2 antibody levels following rhBMP2 administration were evaluated using enzyme-linked immunosorbent assays (ELISAs), as previously mentioned [14]. Briefly, blood samples were collected by cardiac puncture at the time of euthanization. Serum samples were pooled from four mice of each group and employed for ELISAs. rhBMP2 was diluted to a concentration of 1 μg/mL using coating buffer (eBioscience, Vienna, Austria). Flat-bottomed 96-well ELISA plates (Corning Laboratories, Corning, NY, USA) were coated with diluted rhBMP2 and incubated at 4 °C for 12 h. After washing with distilled water, plates were incubated with a blocking buffer (0.05% Tween 20, 1 mM EDTA, and 0.5% BSA in phosphate-buffered saline) for 1 h, and serum samples (collected from Groups I, IV, V and VI) were added to each well and incubated for 2 h at room temperature. Then, horseradish peroxidase (HRP)-conjugated anti-mouse IgG antibody (Southern Biotechnology, Birmingham, AL, USA; 1:2000) was added and this mixture was incubated for 2 h, and the reaction was developed using 3,3′,5,5′-tetramethyl-benzidine (TMB; BD Bioscience, Bedford, MA, USA). The absorbance of each well was read on a microplate reader (Molecular Devices, Menlo Park, CA, USA) at 450 nm. To determine the levels of IL-1β and TNF-α, sera collected from Group I, II, III and IV were tested using ELISA MAX Deluxe kits (BioLegend, San Diego, CA, USA) in accordance with the manufacturer’s instructions. Absorbance of each well was read on a microplate reader at 450 nm.

Cell culture

MC3T3-E1 preosteoblasts were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were cultured in α-minimal essential medium (α-MEM; Gibco/Thermo Fisher Scientific, Waltham, MA, USA), supplemented with 10% fetal bovine serum (FBS; Gibco/Thermo Fisher Scientific), 100 U/mL penicillin, and 100 µg/mL streptomycin (Invitrogen/Thermo Fisher Scientific), in a humidified 5% CO₂ atmosphere at 37 °C. Cells were seeded in 6-well culture plates at 2 × 10⁵ cells/well. After 24 h, cells were treated with rhBMP2 (150 ng/mL) for 2 days in the absence or presence of serum, which was collected from Groups I, IV, V and VI as mentioned above.

RNA isolation and real-time PCR analysis

Total RNA was isolated from the cultured cells using TRIzol reagent (Invitrogen/Thermo Fisher Scientific), according to the manufacturer’s instructions. To amplify the transcripts of osteoblast-specific genes, 1 μg of total RNA was used to synthesize cDNA using random primers and M-MLV reverse transcriptase (Promega, Madison, WI, USA). Real-time PCR (RT-PCR) was performed with StepOnePlus Real-Time PCR System (ABI, Abilene, TX, USA) using Power SYBR Green PCR Master Mix (ABI, Valencia, CA, USA) according to the manufacturer’s protocol. The expression of all mRNAs was normalized to that of endogenous β-actin. Relative target gene expression was determined using a comparative CT method. The primer sequences for alkaline phosphatase (ALP), bone sialoprotein (BSP), osterix (OSX), and β-actin are as follows; ALP, (F) 5′-ATCTTTGGTCTGGCTCCCAT-3′ and (R) 5′-TTTCCCGTTCACCGTCCAC-3′; BSP, (F) 5′-AAGCAGCACCGTTGAGTATG-3′ and (R) 5′-CCTTGTAGTAGCTGTATTCA-3′; OSX, (F) 5′-AGCGACCACTTGAGCAAACA-3′ and (R) 5′-GCGGCTGATTGGCTTCTTCT-3′; β-actin, (F) 5′- ACCCACACTGTGCCCATCTA-3′ and (R) 5′- GCCATCTCCTGCTCGAAGTC-3′.

Statistical analysis

All data are presented as the mean ± SEM. Statistical analysis was performed using the Student’s t test using SPSS version 18 (SPSS Inc., Chicago, IL, USA). p < 0.05 was considered statistically significant. In vitro experiments were repeated three times independently.

Results

Ectopic bone generation following repeated administration of rhBMP2

To examine whether the osteoinductive ability of a second rhBMP2 administration is affected by the bone-forming response induced by prior rhBMP2 administration, 7 μg of rhBMP2 was subcutaneously administered using collagen sponge on one side on the back of mice, and after 4 weeks the same amount of rhBMP2 was re-administered on the opposite side on the back of mice (Table 1 and Figure S1). A single administration of rhBMP2 on one backside of mice was performed to compare with the osteoinductive ability of repeated administrations (Table 1 and Figure S1). Four weeks after the second administration, ectopic bone generation was determined in living animals using 2D X-ray analysis. A rounded radio-opaque shadow was observed at the sites of the first as well as of the second administration of rhBMP2, but not at the PBS administration sites (upper panels of Fig. 1A). For further analyses, all the tissues, including radio-opaque regions, were harvested from the mice and analyzed using micro-CT. Consistent with the results of 2D X-ray analysis, rounded or spherical radio-opaque shadows were observed in rhBMP2-treated groups. The margin of the newly formed tissue was continuously radio-opaque, similar to the cortical bone, whereas the inside looked like a trabecular bone in all the rhBMP2-administrated sites (lower panels of Fig. 1A). No radio-opaque image was observed when the control carrier alone was implanted into the mice (Fig. 1A).

Volumetric micro-CT analysis of rhBMP2-induced bone formation showed that the volume of the newly formed bone was 4.52 ± 2.97 mm³ at the first implantation site and 7.35 ± 3.36 mm³ at the second implantation site in the repeated administrated group (Fig. 1B). In single administration groups, the volume of the newly formed bone was 5.20 ± 2.22 mm³ at the first site and 6.31 ± 1.67 mm³ at the second site (Fig. 1B). There was no significant difference between all groups (p > 0.05, n = 3). In addition, there was no difference in the bone mineral density (BMD) of the bones formed at the first site or second site by repeated administration as well as by single administration of rhBMP2 (Fig. 1B). Histological analysis showed bone-like mineralization tissues and bone marrow structure in all sites of rhBMP2 administration; these results show that the inner structure—composed of trabecular bones and blood cells—as well as the outer structure—densely calcified tissue—were induced upon rhBMP2 administration (Fig. 1C). There were no qualitative differences in the newly formed tissues between the first and the second site of rhBMP2 administration, regardless of single or repeated administration conditions. These results suggest that the bone forming ability of rhBMP2 in repeated administration conditions is equal to that of single administration of rhBMP2; they also suggest that the intact bone-forming ability of rhBMP2 is not compromised by repeated administration at 4-week intervals.

Generation of cranial bone following repeated administration of rhBMP2

We also wanted to check whether rhBMP2 can be repeatedly used for cranial bone generation in different regions. Hence, 7 μg rhBMP2 with collagen sponge was first implanted into the back of mice, 3 weeks after which, the same amount of rhBMP2 was administered into the critical-sized calvarial defects (Table 1). Three weeks after the second administration, 2D radiographic and micro-CT analyses were performed separately. A rounded radio-opaque image was observed at the site of the first implantation with rhBMP2 (Fig. 2A: 1st Subcut site of Group VI). In addition, a new radio-opaque tissue was found at the cranial defective region where the second administration of rhBMP2 was performed (Fig. 2A; 2nd Calvarium site of Group VI). 2D micro-CT images showed that the outer side of the newly formed tissues—which were continuously radio-opaque—resembled the cortical bone, while the inside of the new bone—which was partially radio-opaque—looked like the trabecular bone (middle panels of Fig. 2A). 3D images showed that the first administration of rhBMP2 resulted in the generation of a spherical radio-opaque tissue in the back, and the second administration of rhBMP2 into the cranial defects completely removed the defects, forming bone-like tissues (lower panels of Fig. 2A).

Volumetric analysis indicated that the volume of the new bone at the first subcutaneous implantation site of rhBMP2/collagen sponge was 12.05 ± 2.13 mm³ (Fig. 2B). No new bone formation was observed in the first subcutaneous implantation site of control sponge group (Figs. 2A, B). The second administration of rhBMP2/collagen sponge into the cranial defect area produced 28.5 ± 4.87 mm³ of new bone; the collagen sponge alone also produced 1.79 ± 1.63 mm³ of new bone (Fig. 2B). Histological analysis revealed that the newly formed tissues possessed cortical bone structures in the periphery and those of trabecular bone in the central region (Fig. 2C: 1st Subcut/rhBMP2 and 2nd Calvarium/rhBMP2). These results indicated that repeated administration of rhBMP2 can successfully induce bone formation under experimental conditions.

Effect of repeated administration of rhBMP2 on immunological responses

The immunogenicity of recombinant BMP2 can be caused by the immune reaction itself, while the lesser protein efficacy can be attributed to the formation of neutralizing antibodies against rhBMP2 [9]. Therefore, we examined the changes in the serum anti-rhBMP2 antibody concentration following two administrations of rhBMP2. In the subcutaneous administration model, there were no significant differences in the serum anti-BMP2 antibody concentrations (experimental vs. control; 303.00 ± 14.63 ng/mL vs. 296.5 ± 23.60 ng/mL, p = 0.822, n = 4) (Fig. 3A). Likewise, there were no significant differences between the two groups in a subsequent administration model of a calvarial defect after subcutaneous administration (experimental vs. control; 351.50 ± 10.78 ng/mL vs. 281.00 ± 33.46 ng/mL, p = 0.092, n = 4) (Fig. 3B). These results suggest that anti-rhBMP2 antibodies are not produced in vivo in response to the repeated administration of rhBMP2 at least for the time periods examined by us.

Fig. 3.

The immunological responses to repeated administration of rhBMP2 in mice. A and B Anti-rhBMP2 antibody levels in the serum from Groups I and IV (A) and Groups V and VI (B). Blood samples were collected by puncturing the heart 8 weeks (A) or 6 weeks (B) after the first administration. Serum anti-BMP2 antibody levels were measured by ELISA. Repeated use of rhBMP2 did not significantly alter the serum levels of anti-rhBMP2 antibody. ns, not significant (n = 4). C Histological analysis of the connective tissues surrounding carriers or newly formed bones in subcutaneous implant sites. Eight weeks after the first administration, hematoxylin and eosin staining was performed by harvesting the tissue of the subcutaneous implant sites. D Serum TNFα levels were determined by ELISA as described in the Sect. 2. Administration of rhBMP2 did not affect the serum TNFα levels. ns, not significant compared to the control group (n = 4)

To check whether an inflammatory reaction is induced in response to the repeated administration of rhBMP2, histological analysis was performed by harvesting the tissue of the subcutaneous implant sites of all groups in the ectopic model at the experimental endpoint. Some inflammatory cells were observed in the connective tissues, not only around the collagen sponges containing PBS but also around the bone formed by rhBMP2, thus indicating that mild inflammation occurred in all subcutaneous implant sites (Fig. 3C). However, there was no significant difference between the four experimental groups (Fig. 3C). In addition, we measured the serum level of pro-inflammatory cytokines, TNFα and IL1β, which were reported to be elevated by rhBMP2 [9, 15]. TNFα levels were very low and there was no difference between all groups (Fig. 3D); IL1β was not detected because it might have been under the lower limit of detection (data not shown). Taken together, these results suggest that repeated administration of rhBMP2 does not significantly induce inflammation, although the initial inflammatory response after administrations had not be confirmed.

Effects of the sera on rhBMP2-induced gene expression

To further confirm that no anti-BMP2 antibodies were present in serum after the repeated administration of rhBMP2, we examined the neutralizing activity of serum on rhBMP2-induced osteogenic gene expression. The presence of anti-rhBMP2 neutralizing antibodies in the serum was determined by evaluating whether the serum affects the biological activity of rhBMP2 newly added to the in vitro experiment. MC3T3-E1 cells were cultured with rhBMP2 in the presence of sera, which were collected from PBS (control) or rhBMP2-administrated mice, as performed in Fig. 3A. The expression levels of osteoblast-specific genes, including alkaline phosphatase (ALP), bone sialoprotein (BSP), and osterix (OSX), was assessed by RT-PCR. In the presence of control serum, rhBMP2 increased the mRNA levels of osteoblast-specific genes. In the presence of serum from rhBMP2-treated mice, rhBMP2 also induced the expression of osteoblast-specific genes (Fig. 4). There were no significant differences between the two groups (control serum and BMP2-exposed serum) with respect to rhBMP2-induced gene expression. These results indicate that the repeated administration of rhBMP2 did not generate anti-BMP2 neutralizing antibodies.

Fig. 4.

Effects of exposure to serum from rhBMP2-treated mice on osteoblast-specific gene expression by rhBMP2 in MC3T3-E1 cells. Cells were cultured with rhBMP2 (150 ng/mL) for 2 days in the presence of sera collected as described in Fig. 3A. Total mRNA was extracted and RT-PCR was performed with primers specific for alkaline phosphatase (ALP), bone sialoprotein (BSP), and osterix (OSX). ns, not significant; control serum, PBS/PBS administrated serum; rhBMP2 serum, rhBMP2/rhBMP2 administrated serum

Discussion

Recombinant BMP2 has excellent bone formation ability and has been studied extensively as a biosimilar for bone repair in the dental or orthopedic fields. However, there are concerns about its clinical usage via repeated administration for bone regeneration at multiple sites. These concerns are based on the fact that administration of exogenous proteins can lead to antibody formation and consequently inflammatory responses. Previous studies have mentioned that the repeated use of rhBMP for spinal fusion did not produce the risk of wound healing problems and allergic reactions [16, 17]. In the present study, we found that the repeated administration of rhBMP2 into subcutaneous spaces in the back or in calvarial defects did not significantly affect the bone formation ability of rhBMP2 as well as the serum anti-BMP2 antibody concentration and inflammatory responses in mice. It is noteworthy that our study is the first known effort to identify the effects of repeated use of rhBMP2 on bone formation in orthotopic cranial and ectopic subcutaneous regions, and to directly identify the changes in serum anti-BMP2 antibody levels.

In vivo bone formation ability of bioactive agents and biomaterials can be diverse, depending on experimental models, such as ectopic (subcutaneous, intramuscular, and kidney capsule) and orthotopic (calvarial and long bones) transplantations [18, 19]. The subcutaneous ectopic model has unique advantages over orthotopic (bone) ones, including a relative lack of bone cytokine stimulation and cell-to-cell interaction with endogenous (host) bone-forming cells. Therefore, ectopic models can be properly evaluated for the sole osteoinductive activity of administered bioactive molecules. Meanwhile, orthotopic (bone) models provide better physiological and clinical environments for evaluating bone formation effects of substances [18, 19]. In addition to the experimental animal models, the carrier type is also important for evaluating bone regeneration induced by bioactive compounds. Absorbable collagen sponges, which possess an adequately porous structure and biodegradable properties, have been used for bone regeneration with topical administration of bioactive substances, such as rhBMP2 [20, 21]. For convenient usage with no surgery, injectable and biodegradable polymers have been developed as rhBMP2 carriers, which exist as a liquid at room temperature and gels at body temperature [7, 13]. Recently, we have developed a sustained rhBMP2-releasing polyphosphazene hydrogel system and demonstrated an optimal bone regeneration effect.

In this study, we examined the effectiveness of repeated rhBMP2 administration in a subcutaneous ectopic model and an orthotopic calvarial defect model with absorbable collagen sponges and thermosensitive polyphosphazene hydrogels. When rhBMP2 was administered along with collagen sponges or polyphosphazane hydrogel carriers into subcutaneous spaces on the backs of mice, ectopic bone generation was observed equally in both groups (Fig. 1 and Figure S2). This result indicated that rhBMP2 possesses its own osteoinductive activity regardless of the type of carrier. When the same amount of rhBMP2 was administered again into the other side of the back in carrier-matched individuals, equal amounts and quality of ectopic bones were generated again (Fig. 1 and Figure S2). In our orthotopic models, repeated administration of rhBMP2 into calvarial defects following subcutaneous administration also regenerated new bones. The quantity of regenerated bone in the defect region was more than that in the subcutaneous region (Fig. 2). The enhanced bone formation ability of repeated rhBMP2 administration in the calvarial bone might depend on the orthotopic microenvironment, which hosted more osteogenic precursor cells and cytokine signals than the ectopic subcutaneous region. Histological analysis additionally showed that the regenerated bone is well fused to the nascent calvarial bone. Our results are clinically significant because we show that rhBMP2 can be repeatedly used without loss of effectiveness in the cranial regions or others although, it has only been administered in the extremities or trunk regions so far.

In vivo administration of exogenous proteins can induce immunogenicity, which results in an inflammatory adverse reactions and the production of antibodies against the proteins, thus reducing their biological effectiveness. We confirmed that significant formation of neither anti-BMP2 antibodies nor neutralizing antibodies was detected in our experimental conditions, which suggests that repeated administration of rhBMP2 does not show reduced efficacy in a living organism at least within the indicated time periods. In addition, no visible signs of swelling or surface inflammation were observed during the experiment after administration of rhBMP2. At the endpoint of the experiment, histological analysis showed that mild inflammation occurred in each subcutaneous implant of all groups, but there was no significant difference between the four experimental groups. In addition, there was no change in the serum levels of pro-inflammatory cytokines by repeated rhBMP2 administrations. Taken together, we believe that repeated administration of rhBMP2 alone does not induce inflammation reaction, although admittedly there is still a need to investigate the initial inflammatory response after administration. The antibody titer might increase rapidly on consistent exposure to the antigen at thrice the concentration. Under clinical circumstances, rhBMP2 administration might be required two or more times at different intervals and at different sites to avoid subsequent bone damage. Therefore, we need to further examine the changes in antibody levels and inflammatory responses under various experimental conditions, for example by altering the interval between rhBMP2 treatments.

Overall, our results suggest that rhBMP2 can be repeatedly used for bone regeneration in orthotopic cranial regions, as well as in other ectopic regions, without typical immunological interruptions, at least with two administrations in a 4-week interval. For clinical applications, further extensive studies on the effects of repeated rhBMP2 treatment under various conditions, such as administration doses, intervals, and frequency, are required.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical statement

All animal studies were reviewed and approved by the Animal Ethics Committee of Chonnam National University (CNU IACUC-YB-2017-73).