Evaluation of Acellular Bovine Cancellous Bone Matrix Seeded with Foetal Rabbit Osteoblasts for Segmental Bone Defects in Rabbits

Bone grafting is the appropriate option to facilitate the repair in cases of segmental bone defects that may result due to disease, trauma or through tumour resection. Currently various therapeutic techniques are available for repair of large bone flaw viz autologus, allogenic and Xenogenic grafts and other synthetic materials like bioceramics and metals (Costantino and Friedman 1994). However, shortcomings like minimal availability of graft, site demise, inadequate strength, disease spread from donor to host and graft rejection would reduce their application (Trentz et al., 2003). Drawback in the use of autologous and allograft tissues can be prevailed over by application of xenografts (Bauermeister et al., 1961).
       
Xenogeneic extracellular matrix (ECM) scaffolds are derived after successful removal of antigenic determinant present in the tissues minimizing immune reactions by allograft and xenograft materials (Badylak and Gilbert, 2008). Decellularization protocols remove only the cellular matters without altering the structural composition of ECM (Dhandayuthapani et al., 2011). Rapid freeze- thawing technique is one of the physical methods of decellularization found to be a more efficient and reliable way for the generation of acellular bovine bone scaffold (Pathak et al., 2012; Tamilmahan, 2013). Composite grafts are better than acellular grafts as they would contain bone forming cells and growth factors secreting cells. These bioscaffolds are formed by incorporating bone producing osteogenic cells and multilineage stem cells and other growth factors to the acellular matrices (Trentz et al., 2003). Some in vitro studies demonstrated the evidences of osteoblasts adhesion, cell viability, osteocalcin synthesis and osteoblastic differentiation within xenogenic decellularized, processed bovine cancellous bone matrices (Rashmi et al., 2017). This study assessed the effect of the composite bone grafts in vivo for the repair of segmental bone defects in rabbits.

Experimental design
The experiment was carried out during the year 2014-2015 at Division of Surgery, Indian Veterinary and Research Institute, Izatnagar. This experiment was conducted after obtaining prior permission from the Institute Animal Ethics Committee (IAEC) for the in vivo study. Thirty six adult New Zealand white rabbits of either sex were allotted to three groups of 12 each. A standard 20 mm defects was created in left fore limb radius bone in all the animals. In group A, defect was filled with auto graft and served as positive control. Group B received acellular bovine cancellous bone scaffold (Fig 1a). In group C (Composite graft), foetal osteoblasts seeded bovine cancellous bone matrices (Fig 1b) were implanted. At days 60 and 90, six specimens per group were harvested and subjected to histological and radiological evaluations.
 

 
Isolation of osteoblast cells
 
The osteoblasts cells were isolated and characterised based on the method standardized in our stem cell lab. The acellular bovine bone graft was prepared by freeze and thaw methods as per the decellularization techniques (Tamilmahan, 2013).  The acellular bovine cancellous bone scaffolds were cut into small pieces and were sterilized with 70% ethanol for 1 hr before seeding. After washing with sterile 1X PBS for 1 hr, finally the bone pieces were sterilized by ‘6 hr in UV’ treatment. Cultured third passaged rabbit foetal osteoblast cells were seeded on acellular bovine matrices (Rashmi et al., 2017). 
 
Surgical procedures
 
The animals were sedated with xylazine @ 6 mg/kg and ketamine @ 60 mg/kg administered through intramuscular route. The rabbits were positioned left lateral recumbence and incision was made on the medial side of the left limb and muscles were separated bluntly. A segmental defect was created at the length of 20 mm on the central diaphysis of the radius bone using bone cutting saw (Fig 1c). The segmental defect was replaced by the same size of the scaffold (Fig 1d). Once the graft was placed in position the muscles and skin were closed by routine manner. Postoperatively limbs were stabilized with thin splint for fourteen days. 
 
Survey radiography
 
Post operatively, implanted graft position was measured by radiography on days 30, 60 and 90. The radiographs were observed for the position of implanted scaffold, new osseous tissue formation and amount of callus and healing of the bone defect and adopted the grading system of Lane and Sandhu (1987) and Heiple (1987) (Table 1).
 
Gross observations
 
The gross healing was assessed as described earlier (Udehiya et al., 2013) on days 60 and 90 after completion of trail period.  
 
Histological observation
 
The test bones were immersed in Goodling and Stewart’s decalcification fluid for 40-72 hrs after euthanizing animals on days 60 and 90 (Wallington, 1972). Then the tissues slices were stained with Haematoxycilin and Eosin (Luna, 1968). Progress of bone union was measured in all the groups using modified procedure explained by Lane and Sandhu (1987) and Heiple (1987) histological grading system (Table 1).
 

 
Statistical analysis
 
All the statistical data were estimated using Statistical Program for Social Sciences (SPSS) and graph pad prism 6 Software. For comparative data One- way ANOVA test was done. Kruskal-Wallis test was used for the analysis of nonparametric parameters (Siegel and Castellan, 1988).

Postoperatively all the animals were near normal and non weight bearing initially, but later started bear weight on the operated limb.    
 
Radiographic observations
 
At 30^th day radiographic density was significantly high in group A (2 ±0) trailed by group C (1±0.51) and group B (1±0). The score for radiographic density improved with time in all the three groups (Table 2). In autograft, the gap was almost filled with callus but complete union was lacking as compared to groups B and group C (Fig 2). Similar observations were made by the study of Gomes (2011), in radiographs after 21 days of surgery with no bone callus bridging. Remodelling was not found in initial stage of fracture repair in any of the groups by day 30, because of action of osteoblastic cells (Lieberman et al., 2002). At 60^th day, higher radiographic density in groups A and C could again be attributed to enhanced osteogenesis occurring in these groups as compared to group B. These results were matching with gross healing, where in, the group A and C scaffolds were minimally distinguished with moderate amount of integration with host bone, whereas in group B, scaffolds were considerably perceptible (Fig 3). This finding correlated with the study of Caporali (2006), where in bone callus was mostly noted near to the ulnar margin at 60 days. In group A, complete remodelling was seen, whereas, animals of groups B and C showed less remodelling. Scaffold started to become more radiolucent, which is a sign of scaffold resorption and remodelling. In an ideal osteoconduction process grafted bone slowly gets resorbed and replaced by new bone (Yildirim et al., 2001). At the end of observation period (90^th day), radiographic density value was noted highest in group A (3±0) followed by group C (2±0.52) and group B (2±0). However, no significant difference was seen between groups C and B at all the intervals. Grossly at 90 days, seeded graft was superior to acellular graft however no significant difference was found (Table 4). Irregular callus was seen in most of the animals of group A, whereas in groups B and C soft callus (Fig 3). The radiopacity gradually decreased at 90^th day interval as compared to 60^th day in groups B and C (Fig 2). This may be due to gradual loss of graft material (Caporali et al., 2006). The radiographic density at gap and intercortical junction had increased and periosteal reaction subsided overall. Among all the groups, groups A and C showed the highest reduction in bone defect as compared to group B. This can be attributed to provision of a prompt supply of osteogenic progenitor cells at the grafted area resulting in osteogenesis (Burwell, 1985). In group A, complete remodelling was found and medullary canal formation was also seen. But remodelling was not completed in groups B and C. The incorporation of implanted bone scaffold may occur early or in later part of healing or may be prolonged without synthesis of new ostogenic tissue (Choi et al., 1996).
 

 

 

 

  
Overall, Periosteal reaction was more intense in group A as compared to other groups, but it was observed less at 90^th day. A periosteal reaction occurs prominently in early part of bone healing and vanishes progressively (Table 3).
 

 
Synthesis of new bone formation occurs in reaction to any injury or stimuli to the periosteum (Ved and Haller, 2002). 
 
Histological observations
 
Histological evaluation and Median ±SD histological score of all the group of test animals on days 60 and 90 were given in Fig 4 and Table 5. Score of all parameters were added for total histological score. The total histological score at day 60 was significantly elevated in group A (9±0) trailed by groups C (5.5±0.81) and B (1.5±1.26).
 

 

       
The specimens of group A showed bony union, group C showed osteochondral union, while fibrous union was mainly evident in group B (Fig 4 A). Presence of pores in the scaffold, the newly formed bony tissue was easily identified from the host tissue in groups C and B. These results were similar to the study of Wang (2010) on MSCs loaded tricalcium phosphate scaffold after 8 weeks of in vivo implantation. Integration of graft into the host bone were maximum in groups C and A as compared to group B after 60 days, due to faster degradation of scaffold in group A than in group B. In group A, woven bone is seen due to progression of new bone growth. Active bone marrow formation and remodelling of graft was seen in group A. However, at day 60 in seeded graft, it started to appear and in acellular group bone marrow formation was still under progress in defect area (Fig 4 B 60 days). In group C, intense new bone formation was found in the edge of defect area as compared to group B, where the osteogenesis was noticed in the middle of defect. This results correlated with the work of Kaveh (2010), in which they found that bone marrow seeded corticocancellous graft showed osteogenesis at the centre of graft, as compared to non seeded graft. Another study also suggested that bone synthesis in the middle region of implant was more pronounced in osteoblast containing processed bovine cancellous matrix in comparison to acellular (Kneser et al., 2006). At day 90, there was significant difference between groups B and C. Complete organization of shaft was found in group A, but in seeded group union was found bony, followed by osteochondral to bony in group B (Fig 4 90C). Early calcification of grafted bone due to transformation of osteoblasts into osteocytes and liberation of Ca²⁺ ion in the matrix was bringing about formation of bony union in control group and group C (Breitbart et al., 2010). Neovascularisation was seen near the side of the graft and not at the edge of the graft, due to integration of graft to the host bone (Hur et al., 2012). At host-graft junction intense osteoblastic activity and disintegrated scaffold were found in group C (Fig 4 C 90). Overall score for all histological parameters was highest for group A with reorganizing cortical bone formation and active red marrow formation with bony union. Seeded group was found better than acellular group in all parameters. This study suggested that composite grafts provide a better healing by new bone synthesis, less immunogenic and augmented integration to the host in comparison to acellular cancellous graft. Development of cell based bone grafting opens up new avenue for successful clinical use of composite graft for bone regenerations.