Background

The anterior cruciate ligament (ACL) is one of the most important stable structures of the knee joint, and the meniscus and cartilage are easily injured after ACL rupture, predisposing the knee to early degenerative changes [1]. Similar to that of the articular cartilage and meniscus, the healing potential of the ACL is extremely poor [2], and reconstruction surgery is often required. Anterior cruciate ligament reconstruction (ACLR) can restore knee stability and prevent further damage to the injured knee [3, 4]. ACLR has been widely used in clinical practice, and the incidence of ACLR has significantly increased over time [5]. Although tendon grafts, including hamstring and patellar tendon grafts, have been used in ACLR for decades with satisfactory long-term clinical outcomes [6, 7], there are still some cases that require revision surgery, which is related to poor tendon–bone healing after reconstruction [8]. The success of ACLR largely depends on the biological healing between the graft and bone tunnel, and ideal ACLR using tendon grafts requires biological healing of the bone-to-tendon interface between the host bone and transplanted tendon. However, problems such as the relatively long healing time between tendon and bone and insufficient early biomechanical strength have not been fundamentally resolved. Therefore, promoting tendon–bone healing after ACL reconstruction is still an urgent problem to be solved.

To enhance the bone graft healing process, biological augmentation techniques, such as transplantation of bone marrow stem cells (BMSCs) [9], platelet-rich plasma (PRP) injection [10], and growth factor and gene transfer [11], have been applied in ACLR in recent years. It was reported that radiolabeled BMSCs were evenly distributed at the tendon–bone interface during the healing process when they were implanted into the tendon–bone interface after ACL reconstruction [12], suggesting that BMSCs might play an important role in promoting tendon-to-bone tunnel healing. In view of the role of BMSCs in tendon–bone healing and their self-renewal and multi-differentiation potential, stem cell transplantation is considered to be a promising method to promote tendon–bone healing [13], and clinical application has been attempted [14]. Bone formation plays an important role in early tendon–bone healing [15], and the biomechanical strength of the tendon–bone connection is related to the bone mass and mineralization of regenerative tissue at the tendon–bone interface [16], implying that new bone formation in the bone tunnel is essential for early healing between bone and tendon. Previous studies using BMSCs to promote tendon–bone healing showed that transplantation of BMSCs could promote direct tendon–bone healing to some extent [9, 17, 18], but new bone formation was insufficient at the tendon–bone interface.

Dedifferentiation refers to the transformation of cells from a given differentiated state to a less differentiated or stem cell-like state and leads to reacquisition of pluripotency; dedifferentiation is a cellular process associated with reentry into the cell cycle, trans/redifferentiation, or tissue regeneration [19]. Dedifferentiated cells possess certain characteristics of primitive stem cells, and the efficiency of redifferentiation is significantly improved. Dedifferentiation also occurs during the process of directed differentiation of stem cells [20]. Studies focused on the differentiation and dedifferentiation of BMSCs found that BMSCs could be reprogrammed in vitro via neuronal differentiation and dedifferentiation, with enhanced therapeutic efficacy [21]. During the process of osteogenic differentiation of BMSCs, BMSCs could be dedifferentiated after removing the induction conditions, and the resulting cells were termed dedifferentiated osteogenic BMSCs (De-BMSCs). De-BMSCs have improved osteogenic potential in vitro and exhibit great superiority in ectopic bone formation in vivo [21, 22]. These studies, which focused on De-BMSCs, provided new insights for stem cell transplantation. Because of the essential role of bone formation at the tendon–bone interface in early tendon–bone healing, De-BMSCs implantation at the tendon–bone interface could theoretically differentiate into more osteoblasts and osteocytes during the tendon–bone healing procedure, increase bone growth in the bone tunnel and promote early tendon–bone healing.

Although the osteogenic effect of De-BMSCs is significantly enhanced, its molecular mechanism is still unclear. Nanog is a newly reported transcription factor that is expressed in primordial germ cells and embryonic stem cells and plays a key role in maintaining the self-proliferation and undifferentiated state of stem cells [23]. It selectively inhibits or promotes gene expression by binding to the regulatory regions of target genes. The expression of Nanog in De-BMSCs was significantly increased [20]. Osterix is an osteoblast-specific transcription factor belonging to the SP/KLF family [24]; it is also the key transcription factor for osteogenic differentiation and can induce the expression of many mature osteoblast marker genes [25] and play a vital role in the maintenance of bone formation [1.

Table 1 Primers used for qPCR
Full size table

Western blotting

To obtain total protein, the cells were harvested and dissolved in RIPA (Beyotime, Nan**g, China) buffer. The protein concentrations were determined with a BCA protein assay kit. Equal amounts of protein lysates (40 mg/lane) were loaded and resolved by 10% sodium-dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) (Sigma-Aldrich) and then transferred to nitrocellulose filters and probed with rabbit anti-Nanog, anti-NFATc1, anti-Osterix, anti-OPN, anti-OCN, and anti-COL1A1 antibodies at 4 °C overnight. After incubation with a horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology), the blots were developed by enhanced chemiluminescence following the manufacturer’s protocol and were visualized by exposure to a Fusion FX system (Vilber Lourmat, Marne-la-Vallee, France). The protein amounts in electrophoresis gels were analyzed with Quantity One 4.6 analysis software (Bio-Rad Laboratories, Hercules, CA, USA). All solutions in this procedure contained a mixture of protease and phosphatase inhibitors.

Immunoprecipitation

Immunoprecipitation was performed following the method described in a previous study [33]. Cell lysates were prepared in NP40 lysis buffer (50 mM Tris–HCl pH 7.4, 150 mM NaCl, 0.5% NP40, all from Sigma) or high-salt lysis buffer (20 mM HEPES pH 7.4, 10% glycerol, 0.35 M NaCl, 1 mM MgCl2, 0.5% Triton X-100, and 1 mM DTT, all from Sigma-Aldrich) with proteinase inhibitors. The supernatant was then incubated with protein G beads (GE Healthcare) and the NFATc1 antibody (Abcam, Cambridge, MA, USA) at 4 °C for 4 h. Beads conjugated with the lysates and antibodies were collected by centrifugation and washed three times with lysis buffer. The final volume of wash buffer was aspirated, and SDS loading buffer was added to the beads. The prepared proteins were resolved using 10% SDS–PAGE and then transferred to nitrocellulose membranes. Finally, the membranes were incubated with antibodies against Osterix (Abcam, Cambridge, MA, USA) for 12 h. Chemiluminescence was detected using the abovementioned ECL system.

Statistical analysis

SPSS 17 (SPSS Science, Chicago, IL, USA) was used for data analysis. Quantitative data were expressed as the mean ± SD. One-way ANOVA followed by Dunnett’s post hoc test and Student’s t tests were used to analyze differences in the quantitative results. Statistical significance was defined as p < 0.05.

Results

Characterization of BMSCs and De-BMSCs

The general process of osteogenic differentiation and dedifferentiation is schematically illustrated (Fig. 1a). To determine whether BMSCs and De-BMSCs have stem cell characteristics, we used flow cytometry to analyze cell surface markers. The profiling showed that De-BMSCs expressed the stemness markers CD29 and CD44 but were negative for the hematopoietic markers CD34 and CD45 (Fig. 1b). The cell surface antigen profiles of De-BMSCs were similar to those of BMSCs. To determine whether De-BMSCs have the potential to differentiate into multiple lineages, osteogenic, chondrogenic, and adipogenic differentiation was induced. The results demonstrated that De-BMSCs could differentiate into osteoblasts, chondrocytes, and adipocytes, which was verified by positive staining for alkaline phosphatase (ALP), safranin O, and oil red O, respectively (Fig. 1c). The above results indicated that De-BMSCs retained stem cell properties. CCK-8 assay was used to detect the ability of cell proliferation in BMSCs and De-BMSCs, the result showed that De-BMSCs proliferated rapidly compared to BMSCs at the same time point (Fig. 1d).

Optimal dedifferentiation culture time to promote the osteogenic differentiation potential of De-BMSCs

It was reported that De-BMSCs had improved osteogenic potential in vitro and in vivo, and the time for dedifferentiation culture was 7 to 14 days [22, 34]. To explore the optimal dedifferentiation culture time to promote the osteogenic differentiation ability of De-BMSCs, we cultured Os-BMSCs in medium with no osteogenic factors for 3, 7, and 10 days after osteogenic induction and then transferred the De-BMSCs to osteogenic medium again for 2 weeks. The osteogenic differentiation ability of the cells was tested. Alizarin red S staining revealed that there were significantly more positive calcium nodules formed in the De-BMSCs that were dedifferentiated and cultured for 3 days (Fig. 2a). ALP staining showed similar results (Fig. 2a). We further examined the mRNA and protein expression of osteoblast marker genes after osteogenic differentiation of De-BMSCs cultured for different dedifferentiation times. The results demonstrated that the mRNA expression of COL1A1, OCN, and OPN in De-BMSCs subjected to dedifferentiation culture for 3 days was significantly increased (Fig. 2b), and the same trend was also found for protein expression (Fig. 2c). The above results indicated that the optimal dedifferentiation culture time to promote the osteogenic differentiation potential of De-BMSCs was 3 days.

Fig. 2
figure 2

The osteogenic differentiation ability of De-BMSCs cultured for different dedifferentiation times. a Staining with alizarin red S (×100) and staining for alkaline phosphatase (×40) to detect the osteogenic differentiation of De-BMSCs. b mRNA expression of osteoblast marker genes, including COL1A1, OCN, and OPN. c Protein expression of osteoblast marker genes, including COL1A1, OCN, and OPN. Data represent the mean ± SD (n = 3). *p < 0.05, **p < 0.01. BMSCs, bone marrow mesenchymal stem cells; De-BMSCs dedifferentiated osteogenic bone marrow mesenchymal stem cells, COL1A1 α1 chain of type I collagen, OCN osteocalcin, OPN osteopontin

Full size image

De-BMSC transplantation could promote tendon–bone healing after ACLR

Tendon–bone healing is an important factor for successful ACLR, and bone formation plays an important role in the process of tendon–bone healing [15]. In view of the enhanced osteogenic differentiation ability of De-BMSCs, we transplanted De-BMSCs that were dedifferentiated and cultured for 3 days at the tendon–bone interface after ACLR. At week 12, the rabbits in each group were sacrificed to evaluate tendon–bone healing by histologic staining, micro-CT examination, and biomechanical testing. HE staining showed that the interface zone was organized and that perpendicular collagen fibers formed crossing the junction between tendon and bone in the control group; a mature zone of cartilage was observed gradually blending into the adjacent bone and tendon, and tight fibrous tissue and a small amount of osteogenic bone tissue could be seen in the BMSC group; more newly formed surrounding bone was observed in the De-BMSC group, and there was an obvious migration connection between the newly formed bone tissue and bone tunnel (Fig. 3a). Micro-CT analysis showed more new bone formation around the tendon graft in the De-BMSC group than in the other two groups (Fig. 3d), the BV/TV value of the De-BMSC group was significantly higher than those of the control group and BMSCs group (Fig. 3e), and the bone tunnel areas of the De-BMSCs group were smaller than those of the control group (Fig. 3f). To further confirm the formation of bone tissue at the tendon–bone interface, RNA was extracted from the regenerated tissues at the tendon–bone interface of all groups. The mRNA expression of COL1A1, OCN, and OPN in the De-BMSC group was significantly increased (Fig. 3h, j, l). Immunohistochemical staining of sections with COL1A1, OCN, and OPN antibodies showed that the tendon–bone interface stained positive for these proteins. Staining for Col1A1, OCN, and OPN was stronger in the De-BMSC group than in the other two groups (Fig. 3g, i, k), which was consistent with the mRNA expression results. To clarify the effect of bone formation on tendon–bone healing after ACLR, we examined the biomechanics of the graft. Failure mode was divided into ruptures at the midsubstance or pullout from the bone tunnel. The failure pattern of the grafts was different on biomechanical test in the three groups, the majority of grafts were ruptured at the midsubstance (rupture/pullout: 4/1 grafts in Control group, 3/2 grafts in BMSCs group, and 5/0 grafts in De-BMSCs group). The maximum failure load values in the control, BMSCs, and De-BMscs groups were 42.01 ± 6.95 N, 49.24 ± 3.22 N, and 65.03 ± 7.96 N, and the stiffness values were 12.39 ± 1.56 N/mm, 16.26 ± 2.95 N/mm, 25.47 ± 2.58 N/mm. The maximum failure load and stiffness in the De-BMSC group were higher than those in the other two groups (Fig. 3b, c). These results implied that the transplantation of De-BMSCs could increase bone formation at the tendon–bone interface, thereby promoting tendon–bone healing after ACLR.

Fig. 3
figure 3

Tendon–bone healing 12 weeks after anterior cruciate ligament reconstruction (ACLR). a Hematoxylin and eosin staining of the tendon–bone interface. b Maximum failure load. c Stiffness. d Microcomputed tomography evaluation of the tendon–bone interface. e Bone volume/total volume (BV/TV). f Bone tunnel area. g Immunohistochemical staining for COL1A1 and quantification of the MOD at the tendon–bone interface. h mRNA expression of COL1A1 at the tendon–bone interface. i Immunohistochemical staining for OCN and quantification of the MOD at the tendon–bone interface. j mRNA expression of OCN at the tendon–bone interface. k Immunohistochemical staining for OPN and quantification of the MOD at the tendon–bone interface. l mRNA expression of OPN at the tendon–bone interface. Data represent the mean ± SD (n = 5). *p < 0.05, **p < 0.01. Scale bar = 50 μm. BMSCs bone marrow mesenchymal stem cells, De-BMSCs differentiation bone marrow mesenchymal stem cells, COL1A1 α1 chain of type I collagen, OCN osteocalcin, OPN osteopontin, MOD mean of density, B bone, IF interface T tendon

Full size image

Nanog/NFATc1/Osterix expression increased during the osteogenic differentiation of De-BMSCs

Nanog plays an essential role in maintaining the self-proliferation and undifferentiated state of stem cells [23]. Osterix is the key transcription factor for osteogenic differentiation and can induce the expression of many mature osteoblast marker genes [25]. Nuclear factor of activated T-cells c1 (NFATc1) also plays an important role in bone formation. It combines with Osterix to form a complex and regulate osteoblast differentiation [28]. Therefore, we investigated the expression of Nanog, NFATc1, and Osterix in regenerated tissues at the tendon–bone interface. The results showed that the mRNA and protein expression levels of Nanog (Fig. 4a, b), NFATc1 (Fig. 4c, d) and Osterix (Fig. 4e, f) in the De-BMSC group were significantly higher than those in the other two groups at 12 weeks after ACLR. During the osteogenic differentiation of De-BMSCs, the mRNA and protein expression levels of Nanog (Fig. 4g, h), NFATc1 (Fig. 4i, j) and Osterix (Fig. 4k, l) were also obviously increased. This result suggested that the Nanog/NFATc1/Osterix signaling pathway may play an important role in the enhanced osteogenic differentiation ability of De-BMSCs.

Fig. 4
figure 4

Expression of Nanog/NFATc1/Osterix signaling pathway components at the tendon–bone interface at 12 weeks after anterior cruciate ligament reconstruction (ACLR). a Immunohistochemical staining for Nanog and quantification of the MOD at the tendon–bone interface. b mRNA expression of Nanog at the tendon–bone interface. c Immunohistochemical staining for NFATc1 and quantification of the MOD at the tendon–bone interface. d mRNA expression of NFATc1 at the tendon–bone interface. e Immunohistochemical staining for Osterix and quantification of the MOD at the tendon–bone interface. f mRNA expression of Osterix at the tendon–bone interface. g Protein expression of Nanog. h mRNA expression of Nanog. i Protein expression of NFATc1. j mRNA expression of NFATc1. k Protein expression of Osterix. l mRNA expression of Osterix. Data represent the mean ± SD (n = 5 in vivo, n = 3 in vitro). Scale bar = 50 μm. *p < 0.05, **p < 0.01. BMSCs bone marrow mesenchymal stem cells, De-BMSCs dedifferentiated osteogenic bone marrow mesenchymal stem cells, NFATc1 nuclear factor of activated T-cells c1, B bone, IF interface, T tendon

Full size image

The Nanog/NFATc1/Osterix signaling pathway mediated the enhanced osteogenic differentiation ability of De-BMSCs

To investigate the molecular mechanism of the enhanced osteogenic differentiation ability of De-BMSCs, we performed SiRNA-mediated stable knockdown of NFATc1 expression in cultured De-BMSCs. NFATc1 expression was significantly suppressed at both the mRNA and protein levels (Fig. 5a, b). Alizarin red S staining revealed more calcium nodules in the De-BMSC group than in the BMSC group. When si-NFATc1 treatment was performed, the number of stained calcium nodules decreased (Fig. 5c). ALP staining showed a similar result (Fig. 5c). To further verify the essential role of NFATc1 in the observed osteogenic effect, we investigated the expression of osteogenic marker genes. The results demonstrated that the mRNA and protein expression levels of COL1A1 (Fig. 5d, e), OCN (Fig. 5f, g), and OPN (Fig. 5h, i) were increased in De-BMSCs compared with BMSCs during osteogenic differentiation, and the expression of the above genes was significantly decreased when si-NFATc1 was used and lower than that in BMSCs. The above results indicated that NFATc1 plays an important role in the enhanced osteogenic differentiation ability of De-BMSCs. During the differentiation of stem cells, Nanog upregulates the expression of NFATc1, and osteoblast differentiation is regulated by the complex formed by NFATc1 and Osterix. Therefore, we examined the expression of Nanog (Fig. 5j, k), NFATc1 (Fig. 5l, m), and Osterix (Fig. 5n, o). The results showed that the mRNA and protein expression levels of Nanog, NFATc1, and Osterix in De-BMSCs were increased compared with those in BMSCs during osteogenic differentiation. After si-NFATc1 administration, the expression of NFATc1 and Osterix decreased significantly and was lower than that in BMSCs, while no significant change was observed in the expression of Nanog (Fig. 5j, k). This suggested that NFATc1 might mediate the enhanced osteogenic differentiation ability of De-BMSCs through the interaction with Osterix. To clarify this mechanism, we used Co-IP to determine whether NFATc1 interacted with Osterix. The results demonstrated that during the osteogenic differentiation of De-BMSCs, the binding of NFATc1 and Osterix was significantly increased compared with that observed during the differentiation of BMSCs, and the interaction of NFATc1 and Osterix was significantly reduced after knocking down NFATc1 (Fig. 5p). The above results indicated that increased Nanog expression upregulated the expression of NFATc1 during the osteogenic differentiation of De-BMSCs, which enhanced the interaction of NFATc1 and Osterix and subsequently increased the expression of Osterix, thereby significantly enhancing the osteogenic differentiation ability.

Fig. 5
figure 5

Mechanism underlying the enhanced osteogenic differentiation ability of De-BMSCs. a and b Effect of Si-NFATc1 in De-BMSCs. c Staining with alizarin red S and staining for alkaline phosphatase to detect the osteogenic differentiation of BMSCs and De-BMSCs. d mRNA expression of COL1A1. e Protein expression of COL1A1. f mRNA expression of OCN. g Protein expression of OCN. h mRNA expression of OPN. i Protein expression of OPN. j mRNA expression of Nanog. k Protein expression of Nanog. l mRNA expression of NFATc1. m Protein expression of NFATc1. n mRNA expression of Osterix. o Protein expression of Osterix. p Effects of Si-NFATc1 on the interaction of NFATc1 and Osterix. Data represent the mean ± SD (n = 3). *p < 0.05, **p < 0.01. BMSCs, bone marrow mesenchymal stem cells; De-BMSCs dedifferentiated osteogenic bone marrow mesenchymal stem cells, COL1A1 α1 chain of type I collagen, OCN osteocalcin, OPN osteopontin, NFATc1 nuclear factor of activated T-cells c1

Full size image

Discussion

In the present study, we found that De-BMSCs could retain the characteristics of stem cells and regain the potential for multilineage differentiation. The capacity for osteogenic differentiation was improved in De-BMSCs, as indicated by increased osteoblast marker gene expression. We transplanted De-BMSCs to the tendon–bone interface after ACLR, and the results showed that more newly formed bone was observed at the tendon–bone interface than observed after transplantation of BMSCs; this change improved the biomechanical strength. It was further found that the Nanog/NFATc1/Osterix signaling pathway mediated the enhanced potential for osteogenic differentiation of De-BMSCs.

Tendon-to-bone insertion healing after ACLR is divided into direct healing and indirect healing [35]. During the process of tendon–bone healing, BMSCs infiltrate and are recruited to the interface, and a variety of cytokines are released and promote the proliferation and differentiation of stem cells to achieve tendon–bone healing [36]. Considering the role of stem cells in tendon–bone healing, many types of stem cells, such as adipose-derived mesenchymal stem cells [37, 38], BMSCs [18, 39], induced pluripotent stem cells (iPSCs) [40], umbilical cord stem cells [41], tendon-derived stem cells [42] and CD34+ ACL-derived stem cells [43], have been used to promote tendon–bone healing after ACLR. Among these stem cells, BMSCs are recognized as having the best proliferation and differentiation potential [44]. Therefore, there are many studies on the application of BMSCs to promote tendon–bone healing, and the results showed that transplantation of BMSCs could promote tendon–bone healing after ACL reconstruction, mainly indicated by increased cartilage formation at the tendon–bone interface [18, 39]. The strength of the tendon-to-bone attachment correlated with new bone formation around the tendon [45]. Although the application of BMSCs can promote healing of the tendon–bone interface, insufficient new bone formation may affect the early biomechanical strength after reconstruction.

Studies have shown that the osteogenic differentiation ability of De-BMSCs in vitro and the bone formation of De-BMSCs in vivo were increased [23, 34], but research on the application of De-BMSCs in tendon–bone healing after ACLR has not been reported. To our knowledge, this is the first report of the use of De-BMSCs to enhance bone formation at the tendon–bone interface. Micro-CT scans showed that there was more new bone generation around the graft in the De-BMSCs group than in the BMSCs group. The results of biomechanical testing showed that the maximum load and stiffness of the De-BMSC group were both higher than those of the BMSC group. These results implied that transplantation of De-BMSCs could increase new bone formation at the tendon–bone interface and enhance biomechanical strength after ACLR. Lim et al. [38] and Soon et al. [18] used BMSCs to promote tendon–bone healing after ACLR with autograft and allograft tendons, respectively. The biomechanical strength of the BMSC group in our study was similar to that reported in the abovementioned studies, and the biomechanical strength of De-BMSCs group was higher than that of BMSCs group in this study. Therefore, transplantation of De-BMSCs could increase bone formation at the tendon–bone interface after ACLR and enhance the biomechanical strength of the reconstructed ligament, which was helpful for early postoperative rehabilitation.

Although stem cells have been widely used in various regenerative medicine studies, the low cell survival rate and differentiation efficiency in vivo after transplantation has significantly reduced the effectiveness of stem cell therapy. Studies have shown that the proliferation and differentiation efficiency of De-BMSCs in vitro and in vitro is higher than that of BMSCs [23, 34], and the mechanism is mainly due to increased Nanog expression. In the present study, we also found that the expression of Nanog in De-BMSCs was increased, and its expression at the tendon–bone interface also increased after De-BMSC transplantation. Nanog is the key transcription factor controlling MSCs identity and fate conversion and plays a vital role in maintaining the self-proliferation and undifferentiated state of stem cells [24]. Nanog also plays a major role in the increased capacity for osteogenic differentiation in De-BMSCs [23], but the molecular mechanism is still unclear. Osterix is a zinc finger-containing osteoblast-specific transcription factor that can induce the expression of many mature osteoblast genes. It has been confirmed to be involved in osteoblast differentiation, maturation, and activity [46] and is also necessary for the maturation and function of osteocytes postnatally [47]. Osteogenic differentiation is a key step in osteogenesis, so Osterix also plays an important role in the process of osteogenic differentiation of stem cells. Our study showed that the expression of Osterix increased during the osteogenic differentiation of BMSCs and De-BMSCs. NFATc1 is an important transcription factor for osteoclast formation [48], and it can induce osteoclast differentiation and promote bone resorption. However, some studies have shown that the NFAT signaling pathway also plays a role in the differentiation of osteoblasts, the consequences of NFAT signaling in osteoblastic cells are controversial, and both stimulatory and inhibitory effects on osteoblastic differentiation have been reported [49]. The results of our study showed that the expression of NFATc1 was higher during the osteogenic differentiation of De-BMSCs than during that of BMSCs. The cooperation of NFATc1 and Osterix plays an important role in the formation of new bone [28]. In the present study, we found that NFATc1 interacted with Osterix during the osteogenic differentiation of BMSCs; the interaction between NFATc1 and Osterix was significantly enhanced in the osteogenic differentiation of De-BMSCs, and the expression of osteoblast marker genes was obviously increased. When siRNA was used to knock down the expression of NFATc1, the interaction between NFATc1 and Osterix was reduced, and the expression of osteoblast marker genes was decreased accordingly. During the process of osteogenic differentiation of stem cells, Nanog could promote osteogenic differentiation by enhancing the interaction between NFATc1 and Osterix, and the increased NFATc1 expression induced by Nanog might play an important role [29]. Our results demonstrated that the expression of Nanog was increased in De-BMSCs compared with BMSCs and that the expression of NFATc1 was also elevated during the osteogenic differentiation of De-BMSCs. Therefore, during the process of osteogenic differentiation of De-BMSCs, the increased Nanog expression enhanced the binding of NFATc1 and Osterix by upregulating NFATc1 and promoted the expression of Osterix at the same time; the expression of downstream osteoblast marker genes was increased accordingly, thereby enhancing the osteogenic differentiation ability of dedifferentiated BMSCs.

Conclusions

In summary, our study found that the osteogenic differentiation ability of De-BMSCs was significantly enhanced. The mechanism was that the Nanog/NFATc1/Osterix signaling pathway increased the expression of osteoblast marker genes. Transplantation of De-BMSCs after ACLR could increase bone formation at the tendon–bone interface, thereby increasing the biomechanical strength of the reconstructed ligament. This study provides a new method for promoting tendon–bone healing after ACLR and offers new insight into the clinical application of stem cells.