No severe wound infection, inability to walk or death occurred in the four groups after regular observation. After 10 days, the weight gain of the mice was measured (Table 1). There was no significant difference in body weight increment among the four groups.
Micro - CT imaging
Micro-CT images (Fig.1E-H)showed that the site, size and depth of the bone defect were basically the same in the TBD and TBSTD groups, and the cortical distances on both sides of the fracture end of the bone defect were approximately the same. the fracture sites were the same in the TF and TFSTD groups, however, the TFSTD group showed a significant widening of the gap at the fracture end compared with the TF group, while the latter did not.
Micro-CT quantitative analysis showed (Fig.2) that in the bone defect group, BV was significantly less in the TBD group than in the TBSTD group (P < 0.05). In the oblique row fracture group, the BV in the TF group was significantly lower than that in the TFSTD group (P < 0.05). Moreover, the BV/TV ratio in the TBSTD group was significantly higher than that in the TBD group (P < 0.05), and the BV/TV ratio in the TFSTD group was significantly greater than that in the TF group (P < 0.05). These indicated that in the same bone injury model, the BV and the BV/TV ratio significantly increased in soft tissue defect group compared to preserved soft tissue group. Furthermore, in the different bone injury models, the bone callus volume and the BV/TV ratio did not show significant difference between the TBD and TF groups, while the TFSTD group showed significantly higher bone callus volume than the TBSTD group (P < 0.05), and the BV/TV ratio in the TFSTD group was significantly higher than that in the TBSTD group (P < 0.05).
Three-dimensional images showed there was no sign of significant separation, misalignment and angulation of the fracture ends in the TF group, however, all samples in the TFSTD group exhibited significant separation and different degrees of misalignment and angulation of the fracture ends (Fig.1J, K).
Histopathology
The results of Safraine O-Fast Green stained indicated that the distribution of hard callus and cartilage callus was consistent with the Mirco-CT images (Fig. 2A-H). Histologically, cartilage callus formation was observed on the 10th day after injury.
Histologically (Fig.2A), in the TBD, it could be observed that there was a little bone callus, which consisted mainly of hard callus stained as light blue-stained and a few orange-red chondrocytes distributed ipsilateral side to the tibial bone defect, some osteoid calluses were also visible in the bone marrow cavity. However, in the TBSTD (Fig.2B), callus was showed markedly irregularly larger and increased, mainly hard callus and chondrocytes, distributed ipsilateral and contralateral side to the tibial bone defect. In the TF (Fig.2C), the callus consisted mainly of hard callus distributed on the inner and outer tibia, a very small amount of orange chondrocytes presented. In the TFSTD (Fig.2D), the callus was significantly larger and more numerous, the gap between the fracture ends was significantly widened and separated, and the distribution direction of the callus was more extensive, light blue hard callus and orange chondrocytes were distributed in all directions of the tibia. Micro-CT showed that the distribution of callus with different densities was consistent with the histological results (Fig. 2E-H).
We found differences in the distribution direction of callus in different models (Table 2). The callus formed in the TBD group (10/10) was only distributed on the same side of tibial bone defect. Most of the TBSTD group (8/10) had callus mainly distributed in the ipsilateral and contralateral side of the tibia defect, and only a few samples (2/10) had thin layer callus distributed in the posterior of the tibia.
In the oblique fracture model, only one sample (1/10) of the TF group had a small amount of callus in the posterior of tibia, and most (9/10) of the oblique fracture model had callus distributed in the ipsilateral and contralateral side. Interestingly, in the TFSTD Group (10/10), callus was distributed in ipsilateral side, contralateral side, and posterior. At the same time, there were differences in the gap widening of the broken end of bone injury in the four groups (Table 3). In TFSTD group, the gap at the broken end was larger than the other three groups at 10 days after the fracture.
In addition, histomorphometric analysis revealed significant differences in the CAr between the soft tissue defect group and the soft tissue preservation group after operation (Fig.3A). Besides,tissue distribution within the callus showed substantial differences among 4 groups (Fig.3 B-D). The CAr,BAr,CgAr,FbAr of in the TBSTD were significantly more than that in the TBD (P<0.05, Fig. 3A,C,D). Accordingly, the CgAr/CAr and FbAr/CAr in the TBSTD were significantly higher than in the TBD(P<0.01, Fig.3E), but the BAr/CAr was significantly lower than that in the TBD (P< 0.05, Fig. 3B). Compared with the TF, the CgAr and FbAr in the TFSTD were significantly increased(P<0.01, Fig.3A,C), the BAr was significantly decreased (P <0.05, Fig.3B). Therefore, the CgAr/ CAr and the FbAr/ CAr in the TFSTD was significantly higher than that in the TF, while the BAr/ CAr was significantly lower than the TF(P<0.01, Fig.3E).
We found that the CAr and BAr of the TF were slightly higher than those of the TBD, but there were no significant differences (P>0.05, Fig.3A,3B), and there were also no significant differences in the CgAr, FbAr, BAr/CAr, CgAr/CAr and FbAr/ CAr (P > 0.05, Fig.3C,3D). However, the CAr, CgAr, and FbAr in the TFSTD were significantly higher than those in the TBSTD (P < 0.001, Fig.3A, 3C, 3D), and there was no significant difference in the BAr between the two groups with soft tissue defects. Moreover, the CgAr/CAr and FbAr/ CAr in the TBSTD were significantly lower than those in the TFSTD (P < 0.001, Fig.3E), while the BAr/CAr in the TBSTD was significantly higher than that in the TFSTD (P < 0.001, Fig.3E). This showed that the soft tissue reservation in bone defect and oblique fracture model for the formation of the CAr, BAr, CgAr and FbAr had no significant statistical differences. While in the oblique fracture model after soft tissue defect, the proportion of cartilage callus and fibrous tissue, the CgAr and the FbAr were significantly higher than those in the drilled bone defect model with soft tissue defect, the amount of hard callus was not significantly different between the two groups, and the proportion of hard callus was significantly reduced.
HE staining(Fig.4)confirmed no significant inflammatory cell in the TBD, but a few inflammatory cells in the TBSTD and the TF. Quantitative analysis (Fig.4E) showed that there was no significant difference in the TBD compared with the TBSTD, and there was also no statistically significant difference between the TF and the TFSTD.
HE staining (Fig. 4) showed that the TBD had only a small amount of inflammatory cells, the TBSTD and the TF had a low degree of inflammatory cells, and the TFSTD had a slightly higher degree of inflammatory cells. Quantitative analysis showed (Fig. 4E) that the pairwise comparison between experimental groups in the number of inflammatory cells was no difference (P > 0.05).
ELISA
The levels of pro-inflammatory cytokine IL-1β were measured by ELISA kit in different bone injury models (Fig.5). The results showed that on day 10 after injury, the level of IL-1β was slightly higher in the experimental group than in the control group, but the pairwise comparison between experimental groups was no difference (P > 0.05).
Immunohistochemistry
Analysis of CYR61
Positive expression of CYR61 in different tissues was observed on day 10 post-operation. It was detected in proliferating chondrocytes, fibroblasts, osteoblasts, and immature osteocytes, as well as in the osteoid matrix, but not in hypertrophic chondrocytes or in mature cortical osteocytes (Fig.6A-D). The negative control group was shown in Figure 6E. The percentage of positive area for CYR61 in the 4 groups revealed that (Fig.8A) it was significantly high in the TBSTD and the TFSTD, while the TBSTD being larger than the TBD and the TFSTD larger than the TF. In addition, there was no significant difference in the expression of CYR61 between the TBD and the TF (P > 0.05), while it was significantly higher in the TFSTD than that in the TBSTD (P < 0.001).
Analysis of VEGFR2
The results showed VEFR2 was detected in osteoblasts and their precursors, fibroblast cells, endothelial cells, proliferative chondrocytes and hypertrophic chondrocytes, not in the tibia cortical bone (Fig.7A-D). The negative control group was shown in Figure 6E. Comparative analysis of the percentage of VEGFR2 in each group (Fig.8B) showed that it was smaller in the TBD than that in the TBSTD, and in the TF smaller than that in the TFSTD. However, there was no significant difference in VEGFR2 expression between the TBD and the TF (P > 0.05), nor between the TBSTD and the TFSTD (P > 0.05).