Investigation of the medium-term effect of osteoprotegerin/bone morphogenetic protein 2 combining with collagen sponges on tendon-bone healing in a rabbit

Abstract

Background

Osteoprotegerin (OPG) and bone morphogenetic protein-2 (BMP-2) could be administered sequentially to promote tendon-bone healing. There remain several unresolved issues in our previously published study: a) the release kinetics of OPG/BMP-2 from the OPG/BMP-2/collagen sponge (CS) combination in vitro remained unclear; b) the medium-term effect of the OPG/BMP-2/CS combination was not analyzed. Hence, we design this study to address the issues mentioned above.

Methods

30 rabbits undergoing anterior cruciate ligament reconstruction (ACLR) with an Achilles tendon autograft randomly received one of the 3 delivery at the femoral and tibial tunnels: OPG/BMP-2, OPG/BMP-2/CS combination, and nothing (blank control). At 8 and 24 weeks post-surgery, the biomechanical tests and histologic analysis were used to evaluate the tendon-bone healing.

Results

In mechanical tests, the OPG/BMP-2/CS group showed a higher final failure load and stiffness than the other groups at 8 and 24 weeks. Additionally, the maximum stretching distance showed a decreasing trend. The mechanical failure pattern of samples shifted from a tunnel pull-away to a graft midsubstance rupture after OPG/BMP-2/CS-treated. From histological analysis, the OPG/BMP-2/CS treatment increased the amount of collagen fibers (collagen I and II) and promoted fibrocartilage attachment.

Conclusion

CS as a carrier promotes the medium-term effect of OPG and BMP-2 on tendon-bone healing at the tendon-bone interface in a rabbit ACLR model. OPG, BMP-2 and CS were already applied in several clinical practice, but a further study of clinic use of OPG/BMP-2/CS is still needed.

Keywords

anterior cruciate ligament biomechanical collagen sponges tendon-bone healing

Introduction

Anterior cruciate ligament (ACL) tears account for 50% of injuries to the knee.¹ Consequently, an ACL rupture results in instability of the knee joint² and inferior clinical outcomes.³ Notably, arthroscopic anterior cruciate ligament reconstruction (ACLR) has become a primary method for treating of ACL rupture. The key healing process following ACLR relies primarily on the integration of tendon graft with bone and the type of fixation.⁴ Although ACLR is an effective surgery, with 75%–90% of patients reporting good or excellent outcomes, slow and incomplete healing of the tendon-bone interface can result in knee joint instability, secondary cartilage damage, meniscal injury, and worsening of osteoarthritis.^5,6 The success of reconstruction depends to a large extent on the biological healing of the tendon-bone interface, which is a weak point and a key site to withstand exercise intensity.⁷ Therefore, strengthening the healing between tendon graft and bone tunnel will reduce the time required to return to normal daily life.⁸

Tendon-bone healing is accelerated, tendon graft failure is reduced, and early aggressive rehabilitation is possible with interventions.⁹ Studies have demonstrated that tendon-bone healing can be improved by exogenous bone-growth factors like osteoprotegerin (OPG)^10,11 and bone morphogenetic protein-2 (BMP-2).^12,13 Notably, OPG has been found to inhibit osteoclast maturation and bone resorption in the clinical setting.¹⁴ Our previous study also supported that OPG improved osteointegration in tendon-bone healing.¹¹ Additionally, the osteoinductive cytokine BMP-2 induces the formation of bones and cartilage.¹⁵ Therefore, incorporating BMP-2 binding peptides into the materials used for ACLR enhances bone formation in the femoral tunnels, thereby improving graft quality.¹⁶

A collagen sponge (CS) is a biomaterial that can deliver intrinsic signals within the structure to enhance tissue formation and be degraded by the body when new tissue is formed.¹⁷ Its three-dimensional structure acts as a growth factor delivery carrier, providing extended retention, improving osteogenesis, and allowing extracellular matrix synthesis and slow release of growth factors.^17,18 In our previous work, CS carriers containing OPG/BMP-2 contribute to more consistent, dense interface tissues in the tendon-bone interface.¹⁹ In addition, OPG/BMP-2/CS combination facilitates the formation of the direct enthesis and promotes the tendon-bone healing following ACLR. However, some issues remain open: a) the release rate of OPG/BMP-2 from the OPG/BMP-2/CS combination in vitro remained unclear; b) the medium-term effects of the OPG/BMP-2/CS combination were not analyzed. Therefore, this study used a rabbit ACLR model to evaluate the medium-term effect of OPG/BMP-2/CS intervention on tendon-bone healing by biomechanical and histopathological analysis.

Materials and methods

Ethical statement

The study protocol was approved by the ethics review board of Guilin Medical University (Approval No. GY2017052) and was conducted in strict accordance with the rules of the Animal Care and Use Committee of Zhejiang province.

Release kinetics of OPG/BMP-2/CS combination in vitro

The OPG/BMP-2/CS combination was prepared following our previously published protocol.¹⁹ Briefly, OPG (1.0 mg) and BMP-2 (1.0 μg) powder were dissolved in 10 mL aseptic distilled water to prepare the 100 μg/mL OPG/BMP-2 solutions; the CS (10 mm × 5 mm, Figure 1(a)) were immersed into OPG/BMP-2 solution, and then OPG/BMP-2/CS combination was prepared. A release kinetics experiment was undertaken to determine the release of OPG or BMP-2 from OPG/BMP-2/CS combination. Each sample was placed into 12-well plates containing 2 mL of Phosphate-Buffered Saline (PBS). Subsequently, PBS (containing released OPG and BMP-2) was collected for ELISA analysis after 12 h, 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 7 days, 9 days, 11 days, and 13 days. All ELISA kits were obtained from BIOSS (Beijing, China). The temporal release of OPG or BMP-2 was expressed as a percentage of the total amount of adsorbed OPG or BMP-2.

Figure 1.

In vitro release kinetics curves with respect to irradiated light treatment time. (a) The plots of CS. (b) Schematic diagram of OPG/BMP-2/CS wrapping tendon graft. (c) Cumulative release of OPG and BMP-2 from OPG/BMP-2/CS combination. Data are shown as the mean ± SD (n = 3). OPG, osteoprotegerin; BMP-2, bone morphogenetic protein-2; CS, collagen sponges.

Establishment of the ACLR model

A total of 30 New Zealand white rabbits (male; aged 4 ∼ 5 months; weight 2.0 ∼ 2.4 kg) were purchased from the Experimental Animal Center of Guilin Medical University and used for bilateral ACLR model building, according to our previously reported procedure.¹⁹ Briefly, rabbits were anesthetized by 2.5% sodium pentobarbital solution (30 mg/kg; Propbs, Beijing, China) injection at the ear vein and lidocaine (2%, 2 mL; Jumpcan, Jiangsu, China) paracentesis at the bilateral knee. The whole Achilles tendon of the limb was obtained and used as the autograft for the bilateral ACLR (a macro photo of the ACLR, see reference 19). Subsequently, these animals were grouped as follows: blank control, OPG/BMP-2, OPG/BMP-2/CS combination. The blank control group rabbits received Achilles tendons implantation in the bone tunnel and no other intervention; OPG/BMP-2 group, approximately 0.1 mL of OPG/BMP-2 were injected into the femur, tibial bone tunnel, and graft tendon cavity, respectively; OPG/BMP-2/CS group, both ends of tendon graft (circumference of 4–5 mm) was wrapped with OPG/BMP-2/CS combination (1 cm in length) in one layer and then sutured with braided suture (Figure 1(b)). The press-fit technique was applied to affix the CS on the tendon. The patella was restored, and the incision was sutured. Postoperatively, rabbits were injected with penicillin (400,000 U/d; NCPC, Hebei, China) for 5 days. The animals had free access to food and water and were monitored daily for complications or abnormal behaviors during the healing period. After 8 and 24 weeks of surgery, five rabbits were sacrificed by administering an overdose of sodium pentobarbital (150 mg/kg, IV).

Biomechanical testing

The harvested bilateral knee-joint specimens (total = 30) were subjected to biomechanical test within 2 h of the specimen preparation. Two anchor surfaces were resected 4 cm above the femoral condyle and below the tibial plateau. Two tunnels at 1 mm on both ends of knees were obtained and fixed by wire. The femur-tendon-tibia specimens (only remaining ACL, the rest of the tissue removed) were obtained. Using specially designed clamps, the proximal part of the femur and the distal part of the tibia were loaded on a computer-controlled universal testing machine (Sans Material Test Instrument Company, Shenzhen, China). The displacement rate was 25 mm/min, and the data (the ultimate failure load, the maximum stretching distance, and stiffness) was recorded.

Hematoxylin-eosin staining

A total of 30 knee-joint specimens were used for hematoxylin and eosin (H&E), Safranin/fast green staining, and immunohistochemistry (IHC) analysis. The specimens were fixed in a 10% neutral-buffered formalin solution for 48 h. After the specimens were decalcified with 20% formic acid over 2 weeks, they were cut longitudinally in a sagittal plane passing through the bone tunnels. After embedding in paraffin blocks, 5-micrometer-thick sections were cut. Pathological morphological changes were evaluated by two pathologists independently. They were stained with H&E, then examined under an optical microscope (Carl Zeiss, Germany).

Safranin/fast green staining

Safranin/fast green staining followed the protocol of Servicebio Technology CO., LTD (Wuhan, China). In brief, the paraffin-embedded slices were placed in the fast green dye liquid for 1–5 min, and the excess dye solution was washed away until the cartilage was colorless. For staining with safranin, the slices were stained in Safranin dye liquid for 1–5 s and then quickly dehydrated with three cylinders of anhydrous ethanol. The slides were immersed in xylene to become transparent for 5 min, sealing with neutral resin. The cartilage stained red, and bone formation stained green.

Immunohistochemistry

The paraffin-embedded sections were dewaxed and dehydrated, followed by antigen retrieval. After serum sealing, the sections were incubated overnight at 4°C with anti-collagen I (Col I), and anti-collagen II (Col II) (diluted 1:200; Servicebio, Wuhan, China). After the primary antibody was removed, the HRP-labeled secondary antibody (diluted 1:100, Servicebio) was added, then the cells were visualized with diaminobenzidine (DAB) and the nucleus counterstained with hematoxylin solution. The nucleus of the hematoxylin stain is blue, and the positive expression of DAB is brownish yellow.

Statistical analysis

Data were analyzed using SPSS 26.0 (IBM, Chicago, IL, USA). The homogeneity test of variance was performed among the groups, suggesting homogeneity of variance (p > 0.05). Measurement data were displayed as the mean ± standard deviation (SD). One-way ANOVA and the post-hoc multiple comparison Tukey tests were used for comparisons among multiple groups. p < 0.05 were considered significant.

Results

Release kinetics of OPG or BMP-2

The pore structures of CS were distinct and facilitated the adsorption of OPG and BMP-2 (Figure 1(a)). From the OPG/BMP-2/CS combination, OPG and BMP-2 were released in two phases: a rapid phase at the beginning (4 days) and a subsequent slower phase (Figure 1(c)). The OPG or BMP-2 showed a slowly release profile with nearly 10% per day within the first 4 days. During the subsequent slower phase (days 5–13), the release rate of OPG or BMP-2 gradually decreased, nearly 80% depleted by the end of 13 days.

Biomechanical properties of OPG/BMP-2/CS on ACLR in rabbits

Biomechanical properties are the ultimate index for evaluating tendon-bone healing.²⁰ Femoral-tendon-tibial specimens were analyzed (Figure 2(a)), and the results of biomechanical testing are shown. Most tendon autografts had pulled away from femoral or tibial bone tunnels among all groups at week 8 (Table 1). At 24 weeks post-operation, 4 of the 5 OPG/BMP-2/CS-treated joints failed by rupture at the midsubstance of the tendon, whereas the tendon mainly pulled away from femoral or tibial bone tunnels in other groups.

Figure 2.

Summary of biomechanical data among the control, OPG/BMP-2, and OPG/BMP-2/CS groups at weeks 8 and 24 after the operation (n = 5). (a) The femur and tibia were fixed at both ends of the fixture to detect biomechanical tests. (b) Comparisons of the ultimate failure load. The addition of OPG/BMP-2/CS resulted in higher ultimate loads to failure at weeks 8 and 24 (all p < 0.01). (c) Comparisons of the maximum stretching distance. No statistical differences were noted among groups at week 8 (p > 0.05). The maximum stretching distance in the OPG/BMP-2/CS group was significantly lower than that in the other groups at week 24 (p < 0.01). (d) Comparisons of stiffness Data. Stiffness in the OPG/BMP-2/CS group was significantly higher than in the other groups at weeks 8 and 24 (all p < 0.01). Data are shown as the mean ± SD (n = 5). *p < 0.05; **p < 0.01; ***p < 0.001; ns, no significance. OPG, osteoprotegerin; BMP-2, bone morphogenetic protein-2; CS, collagen sponges.

Table 1.

The site of ligament pulled-out or ruptured in each group at different time points. Tendon autograft extraction occurs mainly on the tibia side.

Groups	n	Pulled out from femoral bone tunnels		Tendon midsubstance ruptured		Pulled out from tibial tunnels
Groups	n	8 weeks	24 weeks	8 weeks	24 weeks	8 weeks	24 weeks
Control	5	1	2	0	0	4	3
OPG/BMP-2	5	2	1	0	1	3	3
OPG/BMP-2/CS	5	1	0	0	4	4	1

In addition, the ultimate failure load increased with time in all groups (Figure 2(b)). The OPG/BMP-2/CS group generally had a higher failure load than that of other groups at all time points (all p < 0.05). At week 24, the ultimate failure load of the OPG/BMP-2/CS group (120.06 ± 15.9) was significantly higher than that in the control group (59.93 ± 9.56, p < 0.001) and the OPG/BMP-2 group (93.55 ± 8.4, p < 0.01). Additionally, there were significant increases in the OPG/BMP-2/CS group versus the other groups at week 8. Furthermore, the maximum stretching distance showed no significant difference among groups at week 8 (all p > 0.05) (Figure 2(c)). At 24 weeks post operation, the stretching distance in the OPG/BMP-2/CS-treated group (2.32 ± 0.26) was lower than that in the control group (3.2 ± 0.31, p < 0.01) and the OPG/BMP-2 group (2.84 ± 0.3, p = 0.038). Moreover, the stiffness in the OPG/BMP-2/CS group was significantly higher than that in other groups (week 8, p < 0.05; week 24, p < 0.001) (Figure 2(d)).

Histologic analysis of OPG/BMP-2/CS on the tendon-bone interface

Successful ACLR requires a firm tendon-bone interface, and bone formation is critical for tendon-bone healing.²¹ Histological analysis was shown in Figure 3. Eight weeks after the operation, the conjunction of tendon-bone in the control group was not tight and mainly consisted of uncalcified fibrocartilage (UFC). Interestingly, in the OPG/BMP-2/CS group, tight conjunction was formed, and UFC layers were observed. Notably, the performance difference across the three groups was evident at week 24 (Figure 3(a)). Tendon-bone interface in both the OPG/BMP-2 and OPG/BMP-2/CS groups formed a clear structural layer, which consisted of tendon graft, noncalcified fibrocartilage, calcified fibrocartilage (CFC), and bone tissue. However, the 4-layer structures of the tendon-bone interface in the OPG/BMP-2/CS group were more closely packed and regular than that of the OPG/BMP-2 group. Moreover, the CFC layer of the tendon-bone interface was not observed in the control group. The degree of tendon-bone healing is correlated with the fibrocartilage layer (UFC and CFC) present on the tendon-bone interface.²² Safranin/fast green staining was used to detect the proteoglycan expression level at the tendon-bone interface and directly indicate the cartilage infiltration. The proteoglycan expression in the OPG/BMP-2/CS group was higher than in the other groups at weeks 8 and 24 (Figure 3(b)).

Figure 3.

H&E staining and Safranin/fast green staining of the tendon-bone interface at weeks 8 and 24 (n = 5). (a) Representative H&E staining image. The histologic maturation at the tendon-bone interface occurred earlier and was more improved in the OPG/BMP-2/CS-treated group than in the other groups. Scale bar: 200 μm. (b) Representative images of Safranin O/fast green staining. OPG/BMP-2/CS promoted proteoglycan expression at the tendon-bone interface at weeks 8 and 24. Scale bar: 200 μm. H&E, Hematoxylin and Eosin; B, bone; UFC, uncalcified fibrocartilage; CFC, calcified fbrocartilage; T, tendon graft; IF, tendon-bone interface. OPG, osteoprotegerin; BMP-2, bone morphogenetic protein-2; CS, collagen sponges.

Immunohistochemical analysis of Col I and Col II expression

As noted in previous investigations, there is mainly Col I in the fibrous connective tissue layer and mainly Col II in the UFC and CFC layers.²² We observed Col I and Col II expression at the tendon-bone interface of all groups at weeks 8 and 24 after ACLR (Figure 4(a) and (c)). Furthermore, Col I and Col II expressions in the OPG/BMP-2/CS group were significantly higher than in the other groups (all p < 0.05) (Figure 4(b) and (d)).

Figure 4.

Expression of type I and type II collagen were followed by immunohistochemical staining (n = 5). (a) OPG/BMP-2/CS promoted the Col I expression of the tendon-bone interface at weeks 8 and 24. Scar bar: 100 μm. (b) Quantitative analysis of Figure 4(a). (c) OPG/BMP-2/CS promoted the Col II expression of the tendon-bone interface at weeks 8 and 24. Scar bar: 100 μm. (d) Quantitative analysis of Figure 4(c). Data are shown as the mean ± SD. *p < 0.05; **p < 0.01; ns, no significance. B, bone; T, tendon graft; IF, tendon-bone interface. OPG, osteoprotegerin; BMP-2, bone morphogenetic protein-2; CS, collagen sponges.

Discussion

Growth factors improves tendon-bone healing after surgery, and combination use of growth factors has a stronger healing effect than one growth factor alone. However, exogenous growth factors are susceptible to degradation in vivo. An effective delivery system that ensures slow release and protects them against degradation will necessitate successful healing. CS is an emerging biocompatible material with a broad range of medical applications. Here, we reported CS as a vehicle to deliver OPG and BMP-2 in a rabbit ACLR model and showed that CS carrying OPG and BMP-2 improved the tendon-bone healing and biomechanical properties in rabbits under ACLR. These findings support that ACLR using an OPG/BMP-2/CS compound reinforcement device gives satisfactory mid-term results (24 weeks).

Previously, we reported that the OPG and BMP-2 levels of synovial fluid of the knee gradually decreased with prolonged time.¹⁹ Importantly, the OPG and BMP-2 were detected at week 12. Here, we investigated the release profile of OPG and BMP-2 from the OPG/BMP-2/CS compound in vitro, in which the results demonstrated that the OPG/BMP-2/CS compound was successfully prepared.

The rabbit tendon-bone interface has four distinct continuous structures: ligament, uncalcified fibrocartilage, calcified fibrocartilage, and bone.²³ Healing occurs with the maturation of vascular and fibrous tissues as the matrix collagen fibers and the fibrous interface become indistinct.²⁴ Adequate mechanical loading is required for the longtime survival of the tendon graft, and poor biomechanical integrity between tendon graft and bone is a significant contributor to procedure failures.²⁴ Additionally, an extra-articular model has demonstrated the association between biomechanical strength and tendon-bone healing.¹⁸ Our study’s mechanical tests and histological observations suggest that the OPG/BMP-2/CS group had the best tendon-bone healing. The OPG/BMP-2/CS group had a firm interface, the greatest amounts of collagen fibers and fibrocartilage, the highest ultimate failure load and stiffness, and the shortest maximum stretching distance. Notably, Col I is recognized to be the most common type of collagen in the tendon, whereas fibrocartilage mainly consists of Col II. Huang and colleagues reported that bone marrow mesenchymal stem cell-derived exosomes promoted the Col I and Col II expression at the rotator cuff tendon-bone interface.²² Likewise, we have observed similar results in our study, in which IHC revealed that Col I and Col II expressions in the OPG/BMP-2 group were higher than in the control group. However, this difference was not statistically significant.

Healing is strengthened over time. In support of this, all measurements of our study showed a time-dependent effect. From histological observations, the amount of collagen fibers and fibrocartilage increased with time. In mechanical tests, the ultimate failure load and stiffness increased, and the maximum stretching distance decreased with time after surgery. The tendon graft midsubstance ruptures after OPG/BMP-2/CS-treated occurred mainly at week 24, suggesting that OPG/BMP-2/CS develops a stronger tendon-bone interface. The mechanical failure patterns shifted from a tunnel pull-away at early weeks to a tendon midsubstance rupture at late weeks post-surgery. The tendon graft pulled away mainly occurred in the tibial tunnel (Table 1), similar to Anderson’s study.²⁵ This result likely occurred in the tibial tunnel because of a relative paucity of cancellous bone in the tibial metaphysis in rabbits, with a correspondingly less vigorous healing response.²⁵ Additionally, the tendon graft gradually integrated with the bone, consistent with previous autologous tendon ACLR rabbit models.^11,26 No statistically significant differences between groups for the maximum stretching distance were found at week 8, possibly caused by the improvement of tendon plasticity and toughness requiring a longer time. Lim et al.²⁷ also stated that ultimate failure load was positively correlated with stiffness and negatively correlated with the maximum stretching distance, which was also observed in our study. In short, the tendon-bone healing after OPG/BMP-2/CS-treated may also have benefited from the stronger biomechanics at week 24.

This study has some notable limitations. First, we did not find an optimum ratio of OPG and BMP-2. Further studies are needed to investigate this aspect in more depth. Second, the histologic findings presented were preliminary, and more accurate evaluations such as micro-CT and molecular chemistry would better serve the study. Third, healing is a consequence of the complex interaction of multiple pathways. for example, the OPG/RANK/RANKL system is believed to be the pathway through which OPG promotes tendon-bone healing.^28–30 However, a more sophisticated understanding of these processes requires further study.

Conclusions

Collagen sponges can be used as an alternative carrier to further promote the effect of growth factors on tendon-bone healing. Importantly, by 24 weeks post-operation, the medium-term effect of the OPG/BMP-2/CS treatment was satisfactory.

Footnotes

Acknowledgements

We are grateful for the experimental support of the public laboratory platform at Southeast University and Guilin Medical University. The authors thank AiMi Academic Services () for English language editing and review services.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by a grant from the National Natural Science Foundation of China [grant number: 82072427] and the Natural Science Foundation of Yongkang City [grant number: 201721].

ORCID iD

Qing Wang

References

Maestro

Herruzo

Varillas-Delgado

, et al. Subjective assessment reported by patients shows differences between single-bundle and double-bundle anterior cruciate ligament reconstruction, systematic review and meta-analysis. Sci Rep 2021; 11: 15385. DOI: 10.1038/s41598-021-94868-0.

Volpi

. Editorial commentary: timing of anterior cruciate ligament reconstruction is just as important as a correct surgical procedure. Arthroscopy : The Journal of Arthroscopic and Related Surgery: Official Publication of the Arthroscopy Association of North America and the International Arthroscopy Association 2021; 37: 1221–1222.

Berthold

Willinger

LeVasseur

, et al. High rate of initially overlooked kaplan fiber complex injuries in patients with isolated anterior cruciate ligament injury. Am J Sports Med 2021; 49: 2117–2124.

Huang

Tsai

, et al. Three-Dimensional printed porous titanium screw with bioactive surface modification for bone-tendon healing: a rabbit animal model. Int J Mol Sci 2020; 21: 3628.

Frobell

Roos

, et al. A randomized trial of treatment for acute anterior cruciate ligament tears. N Engl J Med 2010; 363: 331–342.

Wang

, et al. Biological and biomechanical evaluation of autologous tendon combined with ligament advanced reinforcement system artificial ligament in a rabbit model of anterior cruciate ligament reconstruction. Orthop Surg 2018; 10: 144–151. DOI: 10.1111/os.12370.

Muller

Bowman

Bedi

. ACL graft healing and biologics. Clin Sports Med 2013; 32: 93–109.

Chai

Wan

Jiang

, et al. Use of estradiol promotes graft-bone healing in rabbit model of anterior cruciate ligament reconstruction with a polyethylene terephthalate ligament. Artif Organs 2017; 41: 1153–1161.

Kosaka

Nakase

Hayashi

, et al. Adipose-derived regenerative cells promote tendon-bone healing in a rabbit model. Arthroscopy: The Journal of Arthroscopic and Related Surgery: Official Publication of the Arthroscopy Association of North America and the International Arthroscopy Association 2016; 32: 851–859.

10.

Dynybil

Kawamura

Kim

, et al. [The effect of osteoprotegerin on tendon-bone healing after reconstruction of the anterior cruciate ligament: a histomorphological and radiographical study in the rabbit]. Z Orthop Ihre Grenzgeb 2006; 144: 179–186.

11.

Zou

Song

Wei

. Effects of tendon-bone healing of anterior cruciate ligament reconstruction by osteoprotegerin combined with deproteinized bovine bone. Muscles Ligaments Tendons J 2017; 7: 256–262. DOI: 10.11138/mltj/2017.7.2.256.

12.

Germann

Snedeker

Stalder

, et al. Incorporating BMP-2 and skeletal muscle to a semitendinosus autograft in an oversized tunnel yields robust bone tunnel ossification in rabbits: Toward single-step revision of failed anterior cruciate ligament reconstruction. Knee 2018; 25: 765–773.

13.

Han

Jin

, et al. Sustained BMP-2 release and platelet rich fibrin synergistically promote tendon-bone healing after anterior cruciate ligament reconstruction in rat. Eur Rev Med Pharmacol Sci 2019; 23: 8705–8712.

14.

Steff

Lobert

, et al. Rhizomelia and Impaired Linear Growth in a Girl with Juvenile Paget Disease: The Natural History of the Condition. Hormone Research in Paediatrics 2021: 1–8.

15.

Shoji

Uchida

Satio

, et al. Acceleration of bone union by in situ-formed hydrogel containing bone morphogenetic protein-2 in a mouse refractory fracture model. J Orthop Surg Res 2020; 15: 426. DOI: 10.1186/s13018-020-01953-7.

16.

Crispim

Lee

, et al. Bioactive Tape With BMP-2 Binding Peptides Captures Endogenous Growth Factors and Accelerates Healing After Anterior Cruciate Ligament Reconstruction. Am J Sports Med 2018; 46: 2905–2914.

17.

Munhoz

MAS

Hirata

Plepis

AMG

, et al. Use of collagen/chitosan sponges mineralized with hydroxyapatite for the repair of cranial defects in rats. Injury 2018; 49: 2154–2160. DOI: 10.1016/j.injury.2018.09.018.

18.

Neybecker

Henrionnet

Pape

, et al. In vitro and in vivo potentialities for cartilage repair from human advanced knee osteoarthritis synovial fluid-derived mesenchymal stem cells. Stem Cell Res Ther 2018; 9: 329.

19.

Wei

Wang

Yan

, et al. Osteoprotegerin/bone morphogenetic protein 2 combining with collagen sponges on tendon-bone healing in rabbits. J Bone Miner Metab 2020; 38: 432–441. DOI: 10.1007/s00774-019-01078-w.

20.

Shi

Kang

Wang

, et al. Exosomes derived from bone marrow stromal cells (BMSCs) enhance tendon-bone healing by regulating macrophage polarization. Med Sci Monit 2020; 26: e923328. DOI: 10.12659/MSM.923328.

21.

Tie

Cai

Qin

, et al. Nanog/NFATc1/Osterix signaling pathway-mediated promotion of bone formation at the tendon-bone interface after ACL reconstruction with De-BMSCs transplantation. Stem Cell Res Ther 2021; 12: 576. DOI: 10.1186/s13287-021-02643-9.

22.

Huang

Wang

, et al. Bone marrow mesenchymal stem cell-derived exosomes promote rotator cuff tendon-bone healing by promoting angiogenesis and regulating M1 macrophages in rats. Stem Cell Res Ther 2020; 11: 496. DOI: 10.1186/s13287-020-02005-x.

23.

Liu

Yang

Zhang

, et al. Biomimetic tendon extracellular matrix composite gradient scaffold enhances ligament-to-bone junction reconstruction. Acta Biomater 2017; 56: 129–140.

24.

Chen

Liu

Tsai

, et al. Photoencapsulation of bone morphogenetic protein-2 and periosteal progenitor cells improve tendon graft healing in a bone tunnel. Am J Sports Med 2008; 36: 461–473.

25.

Anderson

Seneviratne

Izawa

, et al. Augmentation of tendon healing in an intraarticular bone tunnel with use of a bone growth factor. Am J Sports Med 2001; 29: 689–698.

26.

Chen

Ouyang

Xiao

, et al. Co-injection of human adipose stromal cells and rhBMP-2/fibrin gel enhances tendon graft osteointegration in a rabbit anterior cruciate ligament-reconstruction model. Am J Transl Res 2018; 10: 535–544.

27.

Lim

Hui

, et al. Enhancement of tendon graft osteointegration using mesenchymal stem cells in a rabbit model of anterior cruciate ligament reconstruction. Arthroscopy 2004; 20: 899–910. DOI: 10.1016/j.arthro.2004.06.035.

28.

Rodeo

Kawamura

, et al. The effect of osteoclastic activity on tendon-to-bone healing: an experimental study in rabbits. J Bone Joint Surg Am 2007; 89: 2250–2259.

29.

Honma

Ikebuchi

Kariya

, et al. Regulatory mechanisms of RANKL presentation to osteoclast precursors. Curr Osteoporos Rep 2014; 12: 115–120.

30.

Zhang

Peng

, et al. The role of RANKL/RANK/OPG system in the canine model of hip periprosthetic infection osteolysis. Int J Artif Organs 2017; 39: 619–624.