Biologic Functionalization of a Biphasic Bone Substitute With Enamel Matrix Derivative Modulates Early Foreign Body Reaction in a Subcutaneous Mouse Model

Abstract

Background/Aim

Biologic enhancement of bone grafts with enamel matrix derivative (EMD) may improve regenerative outcomes by modulating the host tissue response. This study evaluated the in vivo tissue reaction to a biphasic bone substitute (Maxresorb^®), used alone or in combination with EMD, in a subcutaneous mouse model.

Materials and Methods

Forty BALB/c mice received subcutaneous implants of Maxresorb^®, EMD, Maxresorb^® + EMD, or underwent sham surgery. Tissue samples were harvested at 15- and 30-days post-implantation. Histological and immunohistochemical analyses were performed to assess inflammation, neovascularization, fibrosis, and the presence of macrophages and biomaterial-associated multinucleated giant cells (BMGCs). Statistical analysis was conducted using ANOVA (p≤0.05).

Results

The Maxresorb^® + EMD group showed higher inflammatory scores, increased numbers of macrophages and BMGCs, and enhanced neovascularization compared with the other groups, particularly at 30 days. Despite this elevated cellular activity, all test conditions were classified as non-irritant or slightly irritant according to ISO 10993-6.

Conclusion

The combination of EMD and Maxresorb^® enhanced early tissue responses without adverse effects. These findings support the biocompatibility of this combination and suggest potential regenerative benefits through biologically driven material functionalization.

Introduction

Periodontal therapy aims to restore the anatomical structure and functional integrity of the periodontium by regenerating lost supporting tissues, including alveolar bone, periodontal ligament, and cementum (1). The periodontium is a complex, highly specialized tissue system, and its predictable regeneration remains one of the major challenges in dental medicine. Although numerous surgical and non-surgical techniques have been developed and refined over recent decades - ranging from guided tissue regeneration (GTR) to the application of growth factors and biomimetic materials - most clinical outcomes still result predominantly in periodontal repair rather than true regeneration (1-4). Repair is characterized by the formation of a long junctional epithelium and scar-like connective tissue, whereas regeneration requires the re-establishment of all three essential periodontal components in their original architecture and function.

Among the best-studied biologics in periodontal regenerative therapy is enamel matrix derivative (EMD; Straumann^® Emdogain^®, Basel, Switzerland), a protein-based preparation derived from the enamel of developing porcine teeth (3,5). Since its introduction into clinical practice in the late 1990s, EMD has been extensively investigated for its capacity to recapitulate key events of tooth development and wound healing. Human histological studies have demonstrated that EMD can stimulate the formation of new acellular extrinsic fiber cementum with inserting Sharpey’s fibers, promote regeneration of the periodontal ligament, and induce new alveolar bone formation (3,5,6). Mechanistically, EMD exerts its regenerative influence by modulating cell behavior - enhancing attachment, proliferation, migration, and survival of periodontal ligament fibroblasts and cementoblasts - while orchestrating a network of growth factors, cytokines, and extracellular matrix molecules that are crucial for tissue repair and bone remodeling (7-8). In addition, EMD has been shown to attenuate inflammatory cytokine production and create a more favorable microenvironment for healing (9-11).

In challenging clinical scenarios, such as defects with unfavorable morphology, large volumetric tissue loss, or reduced intrinsic stability, the regenerative outcome often depends on the use of bone grafts to provide mechanical support and maintain space for tissue ingrowth. The viscous nature of EMD alone may be insufficient to prevent flap collapse or stabilize the blood clot, particularly in wide three-wall defects or furcation involvements (3,12). In these situations, biphasic bone substitutes (BBSs) - typically composed of hydroxyapatite (HA) and β-tricalcium phosphate (β-TCP) - are frequently employed due to their osteoconductivity, biocompatibility, and gradual resorption profile (13,14). The HA component provides long-term structural stability, whereas β-TCP is more readily resorbed, allowing progressive replacement by newly formed bone. However, these materials are intrinsically bioinert; they lack active signaling properties and rely predominantly on passive osteoconduction rather than active stimulation of regenerative pathways (15,16). Consequently, while they may support bone ingrowth, they do not inherently trigger the complex cascade of cellular events required for complete periodontal regeneration.

To overcome these limitations, combinatorial regenerative strategies have been introduced that aim to synergistically integrate the scaffold function of synthetic bone grafts with the bioactivity of agents such as EMD. Preclinical and clinical data suggest that such combinations can enhance early wound stability, accelerate angiogenesis, and promote favorable soft-tissue integration (17-19). In vitro findings further indicate that EMD can bind to calcium phosphate surfaces and modulate host cell behavior at the graft interface (20). Nevertheless, despite promising in vitro evidence, there is still limited in vivo data on how EMD modifies the inflammatory and immunological tissue response when used together with synthetic bone substitutes (21). In particular, its potential role in modulating macrophage polarization, regulating the formation of biomaterial-associated multinucleated giant cells (BMGCs), and influencing the kinetics of neovascularization remains insufficiently clarified.

Therefore, the aim of the present in vivo study was to evaluate the host tissue response to the BBS (Maxresorb^®, botiss biomaterials GmbH, Zossen, Germany) used either alone or in combination with the EMD Emdogain^® (Straumann^® Emdogain^®, Basel, Switzerland). For this purpose, a well-established subcutaneous implantation model in BALB/c mice was employed, enabling controlled analysis of biomaterial-tissue interactions in the absence of confounding periodontal variables such as bacterial biofilm. The investigation focused on both early and later healing phases, using standardized histological and histopathological assessments to evaluate biocompatibility according to the scoring system of DIN EN ISO 10993-6, including (a) the degree and nature of the inflammatory response, (b) the extent and quality of neovascularization and (c) the presence of BMGCs at the biomaterial interface.

Using this approach, the study aimed to provide mechanistic insights into how the combination of EMD with a synthetic bone scaffold may influence not only structural aspects of tissue regeneration but also the immunomodulatory processes underlying healing - potentially paving the way for more predictable and biologically driven periodontal regenerative therapies.

Materials and Methods

In vivo trial, im- and explantation procedure. The in vivo experiment was performed at the Faculty of Medicine, University of Niš, Serbia, after approval by the Local Ethical Committee (Faculty of Medicine, University of Niš, Serbia) and in accordance with the regulations of the Veterinary Directorate of the Ministry of Agriculture, Forestry and Water Management of the Republic of Serbia (Decision No. 323-07-00278/2017-05/3, dated 13 July 2017). Throughout the study, all mice were housed under standard laboratory conditions with free access to water, artificial lighting, and a diet of regular mice pellets. Standardized preoperative and postoperative care protocols were strictly followed.

For the subcutaneous implantation procedure, forty female BALB/c mice (8-10 weeks old; obtained from Military Medical Academy, Belgrade, Serbia) were randomly allocated to four experimental groups, with ten animals per group (group 1: BBS+EMD, group 2: EMD alone, group 3: BBS alone, group 4: sham operation) and five animals assigned to each observation period (n=5 per time point at 15 and 30 days). The implantation was performed according to the method described by Barbeck et al. (22-26). Following induction of general anesthesia by intraperitoneal injection of ketamine combined with xylazine, according to general anesthesia protocol for mice, the dorsal region was shaved and disinfected. A subcutaneous pocket was created through a skin incision extending into the connective tissue, followed by blunt dissection in the rostral region. The biomaterials were then placed into the prepared pocket (except in the sham group), and the incision was closed with sutures. At each predetermined time point, the animals were humanely euthanized by anesthetic overdose. The implantation sites, including surrounding tissues, were excised and immediately fixed in 10% neutral-buffered formalin.

Histological preparation and histopathological analysis. For the histological workup, each tissue explant was first divided into two segments of identical dimensions and dehydrated in a graded series of increasing alcohol concentrations. Following xylene exposure, the samples were embedded in paraffin, and sections with a thickness of 3-5 μm were prepared using a rotary microtome (SLEE, Mainz, Germany). Three sections from each tissue explant were subjected to histochemical staining with haematoxylin and eosin (H&E), Movat pentachrome, and Alcian blue (all from Morphisto GmbH, Offenbach, Germany).

Histopathological analyses of tissue-biomaterial interactions within the implantation beds and the surrounding tissue were performed using an Axio.Scope.A1 microscope (Zeiss, Oberkochen, Germany), as described previously (22-26). These analyses focused on evaluating parameters associated with early and late tissue responses to the implants, including fibrosis, hemorrhage, necrosis, vascularization, and the presence of neutrophils, lymphocytes, plasma cells, macrophages, and BMGCs. Microphotographs were acquired with an Axiocam 305 color camera connected to a computer system running ZEN Core software (Zeiss).

Results

Histopathological analysis. The histopathological analysis revealed that the combination of the synthetic bone substitute and EMD induced a moderate inflammatory tissue reaction at 15 and 30 days post-implantation (Figure 1A and B). In the surrounding connective tissue, the infiltrate consisted predominantly of macrophages, lymphocytes, and granulocytes, while BMGCs were observed at the surfaces of both biomaterials. The implanted EMD appeared as small spherical structures in close proximity to the bone substitute granules. A moderate vascularization of the implantation beds was also evident. In addition, a very slight fibrosis associated with individual bone substitute granules was detected.

Figure 1.

Tissue reactions to the mixture of the biphasic bone substitute (BBS) with enamel matrix derivative (EMD) (asterisks or E) on day 15 (A) and day 30 (B) post-implantation, to EMD alone on day 15 (C) and day 30 (D) post-implantation, and to the BBS alone on day 15 (E) and day 30 (F) postimplantation within the subcutaneous connective tissue (CT). White arrows: macrophages; red arrows: blood vessels; green arrows: granulocytes; blue arrowheads: biomaterial-associated multinucleated giant cells at the surfaces of the BBS; purple arrowheads: biomaterial-associated multinucleated giant cells at the surfaces of the EMD (H&E staining; original magnification 400×; scale bars: 20 μm).

The analysis further showed that EMD alone likewise induced a moderate inflammatory tissue reaction at both time points (Figure 1C and D). Again, macrophages, lymphocytes, and granulocytes, together with a moderate vascularization, were present at days 15 and 30. At the surface of the EMD, which appeared as a bulk material, only macrophages were observed. No signs of fibrosis were detectable.

Histopathological evaluation of the synthetic bone substitute alone demonstrated an inflammatory tissue reaction surrounding the granules at days 15 and 30 post-implantation (Figure 1E and F). Macrophages, lymphocytes, and moderate numbers of granulocytes, along with few vessels, were present within the surrounding connective tissue, and moderate numbers of BMGCs were found at the granule surfaces. Slight signs of fibrosis related to the material granules were also observed.

In the sham-operated group, histopathological analysis revealed a moderate inflammatory reaction, mainly involving macrophages, lymphocytes, and granulocytes at both study time points, without any evidence of fibrosis (data not shown).

Histopathological scoring. At 15 days post-implantation, evaluation showed that the BBS+EMD group exhibited the highest overall degree of inflammation, whereas the EMD and BBS groups each demonstrated moderate but distinct inflammatory responses (Table I). The sham operation group displayed the lowest inflammatory scores. Moderate numbers of polymorphonuclear cells were consistently observed in all three material groups, while lower counts were noted in the sham group. Similarly low lymphocyte levels were detected in the BBS+EMD and EMD groups; in contrast, only approximately half these counts were recorded in the BS and sham groups (Table I).

Table I. Results of the scoring evaluation at 15 days post-implantation.

Overall inflammation; Inflammation associated with membrane/region of membrane; Scoring Matrix: 0: Absent; 1: Minimal; 2: Mild; 3: Moderate; 4: Marked. Polymorphonuclear Cells; Lymphocytes; Plasma Cells; Macrophages; Giant Cells Scoring Matrix: 0: 0; 1: Rare, 1-5 per high powered (400x) field (phf) (giant cells: 1-2/phf); 2: 6-10/phf (giant cells: 3-5/phf); 3: Heavy infiltrate; 4: Packed. BBS: Biphasic bone substitute; EMD: Emdogain^®; SO: sham operation.

Plasma cells were absent in all study groups except the sham operation group, where only minimal numbers were observed. Moderate macrophage infiltration was present in all material groups, with slightly lower levels in the control group. Multinucleated giant cells were detected exclusively in the BBS+EMD and BBS groups, with slightly higher counts in the BBS+EMD group. Neovascularization was comparable across all study groups. Uniformly low fibrosis scores were recorded in all groups, except for the EMD group (Table I). No fatty infiltration or necrosis was observed in any group at this early time-point.

At 30 days post-implantation, the BBS+EMD group still exhibited the highest overall degree of inflammation, followed by the EMD and BBS groups, which both showed mild inflammatory reactions, while the sham operation group again displayed the lowest inflammatory scores (Table II). Polymorphonuclear cells were most prominent in the BBS+EMD group and remained present at lower but comparable levels in the EMD and BBS groups, with only rare cells observed in the sham group. Lymphocyte infiltration was mild in the BBS+EMD and BBS groups, lower in the EMD group, and minimal in the sham group.

Table II. Results of the scoring evaluation at 30 days post-implantation.

Overall inflammation; Inflammation associated with membrane/region of membrane; Scoring Matrix: 0: Absent; 1: Minimal; 2: Mild; 3: Moderate; 4: Marked. Polymorphonuclear Cells; Lymphocytes; Plasma Cells; Macrophages; Giant Cells Scoring Matrix: 0: 0; 1: Rare, 1-5 per high powered (400x) field (phf) (giant cells: 1-2/phf); 2: 6-10/phf (giant cells: 3-5/phf); 3: Heavy infiltrate; 4: Packed. BBS: Biphasic bone substitute; EMD: Emdogain®; SO: sham operation.

Plasma cells were absent in all groups, except for the sham operation group, where they were detected only in minimal numbers. Macrophage infiltration remained a dominant feature in all material groups, reaching the highest scores in the BS+EMD and BS groups and being less pronounced in the EMD and sham groups. Multinucleated giant cells were again restricted to the BBS+EMD and BBS groups, with higher scores in the BBS+EMD group, whereas no giant cells were found in the EMD or sham groups. Neovascularization was most pronounced in the BBS+EMD group, followed by the BBS group, and was less developed in the sham and particularly in the EMD group. Fibrosis scores remained low overall, with only minimal values recorded in the BBS+EMD and sham groups and absence in the EMD and BBS groups. As on day 15, no fatty infiltration or necrosis was observed in any study group at this later time -point.

At 15 days post-implantation, both test materials were classified as non-irritant (Table III). Test 1 (Maxresorb^® + Emdogain^®) showed a treatment irritancy score of 13.33 with an overall irritancy score of 1.83, whereas Test 2 (Emdogain^® alone) exhibited a lower treatment irritancy score of 9.33 and an overall irritancy score of 0.00, indicating no relevant irritant effect. The control (Maxresorb^® alone) and sham-operated (SO) groups showed treatment and overall irritancy scores of 11.50 and 6.00, respectively, but no formal irritant status was assigned.

Table III. Irritancy scores and irritancy status.

Test 1: Maxresorb^® + Emdogain^®; Test 2: Emdogain^®; Control 1: Maxresorb^®

At 30 days, Test 1 showed an increased treatment irritancy score of 17.00 and an overall irritancy score of 4.50, leading to its classification as slight irritant (Table III). In contrast, Test 2 maintained a low treatment irritancy score of 6.67 with an overall irritancy score of 0.00 and was still rated as non-irritant. The control group presented with a treatment and overall irritancy score of 12.50, while the SO group showed further reduced values (4.83), yet again without a specific irritant status. Overall, these data indicate that Emdogain^® alone is non-irritant at both time points, whereas the combination with the bone substitute develops a slight irritant potential by day 30.

Discussion

This in vivo study aimed to evaluate the tissue response to Maxresorb^® when applied alone or in combination with Emdogain^® in a subcutaneous mouse model. The analysis focused on histopathological parameters such as inflammation, vascularization, fibrosis, and the presence of BMGCs, along with immunohistochemical characterization of macrophage activity. The findings offer valuable insights into how biologic functionalization of bone substitutes may modulate the host’s early inflammatory and healing response (27-30).

The addition of EMD to the bone substitute resulted in higher inflammatory scores and greater presence of BMGCs compared to either Maxresorb^® alone or Emdogain^® alone, especially at 30 days. While elevated inflammation and foreign body response might typically be considered undesirable, this observation should be interpreted in light of EMD’s known capacity to modulate immune responses and stimulate angiogenesis and tissue turnover during early phases of healing. Several studies have demonstrated that EMD can increase monocyte/macrophage recruitment and promote tissue remodeling via upregulation of growth factors such as TGF-β and VEGF (3,31,32).

Interestingly, while macrophage infiltration was present in all experimental groups, the combination group (Maxresorb^® + EMD) consistently demonstrated a higher proportion of both macrophages and multinucleated giant cells. This may reflect a more active biodegradation or phagocytic response to the BBS when combined with EMD, which could facilitate further bone remodeling. It has been proposed that the foreign body reaction, including the presence of BMGC, is not inherently detrimental in biomaterial integration but may represent a dynamic phase in the transition from inflammation to healing. In line with this, Barbeck et al. have shown that the presence of BMGCs can correlate with controlled degradation and integration of calcium phosphate-based materials (25).

In terms of vascularization, the EMD-containing group exhibited higher neovascularization scores at 30 days compared to controls. This aligns with existing evidence that EMD promotes angiogenesis, likely via up-regulation of VEGF and stimulation of endothelial cell proliferation (33). Enhanced vascularization is a critical component of biomaterial integration and long-term regeneration, supporting the rationale for biologic augmentation of bone grafts (34).However, the Emdogain-only group demonstrated a relatively mild inflammatory profile and absence of BMGCs, consistent with EMD’s anti-inflammatory properties and lack of foreign body characteristics when applied alone. Previous subcutaneous studies have confirmed that EMD does not elicit a chronic immune response and may even promote resolution of inflammation through M2 macrophage polarization (35-38).

The overall irritancy score at both time points classified the combination group as “non-irritant” at day 15 and “slight irritant” at day 30. These values fall within acceptable thresholds for biocompatibility per DIN EN ISO 10993-6 standards and support the material’s safe use in regenerative applications. The slightly increased irritancy score at the later time point in the EMD + Maxresorb^® group likely reflects the ongoing cellular activity and remodeling rather than pathological inflammation (36).

Study limitations. First, the use of a subcutaneous mouse model represents an ectopic environment that does not fully replicate the complex anatomical and microbiological conditions of periodontal defects, limiting the direct transferability of the results to the clinical situation. Second, the observation period was restricted to two early time points (15 and 30 days), so neither long-term tissue reactions nor the complete course of material degradation and tissue remodeling could be assessed. Third, the evaluation relied mainly on semi-quantitative histopathological scoring and irritancy grading, which, although standardized, cannot provide detailed information on the underlying molecular mechanisms or on the quality of newly formed tissue. Finally, this experimental setting did not include any functional or structural readouts of periodontal regeneration (e.g., new cementum, periodontal ligament, or alveolar bone formation), any conclusions regarding potential benefits for periodontal regenerative therapy remain hypothetical and must be confirmed in orthotopic defect models and clinical studies (39,40).

Taken together, the results suggest that the biologic functionalization of a BBS with EMD leads to a more dynamic early tissue response, characterized by enhanced inflammatory infiltration, neovascularization, and presence of giant cells. This may reflect an accelerated or amplified regenerative environment, although long-term studies would be necessary to determine whether these early responses translate into improved tissue regeneration and material integration. From a clinical perspective, these findings may have implications for periodontal regeneration, where a finely balanced early inflammatory reaction, sufficient angiogenesis, and controlled biomaterial turnover are crucial for successful tissue restoration. Emdogain^® has been widely used as a biologic adjunct in periodontal defects, primarily to enhance periodontal ligament and cementum regeneration, while BBSs such as Maxresorb^® mainly serve as space-maintaining scaffolds within intrabony defects. The more pronounced, yet still acceptable, foreign body-type response and the increased vascularization observed in the Maxresorb^® + EMD group in this ectopic model may therefore reflect a microenvironment that supports graft remodeling and vascular ingrowth, both of which are desirable features in regenerative periodontal procedures. However, whether this more dynamic tissue reaction ultimately translates into superior clinical outcomes - such as enhanced defect fill, new attachment formation, and stable long-term results - needs to be verified in orthotopic periodontal defect models and controlled clinical studies.

Conclusion

In this subcutaneous mouse model, the BBS Maxresorb^® showed good local tolerability both alone and in combination with Emdogain^®. Functionalization with Emdogain^® induced a more pronounced but still acceptable early foreign body-type response, with higher inflammatory, macrophage/BMGC, and neovascularization scores, whereas Emdogain^® alone remained non-irritant and did not induce BMGC formation. These findings support the concept that biologically functionalized bone substitutes can modulate early tissue responses in a potentially beneficial way, but confirmation of any regenerative advantage will require studies in orthotopic periodontal defect models and clinical settings.

Conflicts of Interest

All 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.

Authors’ Contributions

Conceptualization: T.K., S.N. and M.B.; methodology: M.B., S.S. and S.N.; resources: S.S., S.N., O.J. and M.B.; data curation: T.K. and M.B.; writing - original draft preparation: T.K. and M.B.; writing - review and editing: T.K., S.S., S.N., O.J. and M.B.; visualization: M.B.; funding acquisition: S.N., O.J. and M.B.. All Authors have read and agreed to the published version of the manuscript.

Artificial Intelligence (AI) Disclosure

No artificial intelligence (AI) tools, including large language models or machine-learning software, were used in the preparation, analysis, or presentation of this manuscript.