[Repairing porcine knee joint osteochondral defects at non-weight bearing area by autologous BMSC].

OBJECTIVE

To test the possibility of using bone marrow stromal cells (BMSC) and biodegradable polymers to repair articular osteochondral defects at non-weight bearing area of porcine knee joints.

METHODS

Bone marrows were harvested from 18 hybrid pigs. BMSC were cultured and in vitro expanded and induced with dexamethasone (group A) or with dexamethasone and transforming growth factor-beta1 (TGF-beta1) (group B) respectively. Immunohistochemistry and RT-PCR were used to evaluate chondrogenic differentiation of induced cells. Part of BMSC of 2 animals were retrovirally-labeled with green fluorescent protein (GFP). After induction and label, cells were seeded on a construct of polyglycolic acid (PGA) and polylactic acid (PLA) and co-cultured for 1 week before implantation. Total 4 osteochondral defects (8 mm in diameter, 5 mm in depth) in each animal were created at the non-weight bearing areas of knee joints on both sides. The defects were repaired with dexamethasone induced BMSC-PGA/PLA construct in group A, with dexamethasone and TGF-beta1 induced BMSC-PGA/PLA construct in group B, with PGA/PLA construct alone (group C) or left untreated (group D) as controls. Animals were sacrificed at 3 months (n = 6) or 6 months (n = 10) post-repair. Gross observation, histology, glycosaminoglycan (GAG) quantification and biomechanical test were applied to analyze the results. The two animals with GFP-labeled cells were sacrificed at 7 months post-repair to observe with confocal microscope the distribution of GFP-labeled cells in repaired tissue.

RESULTS

Stronger expression of type II collagen and aggrecan were observed in BMSCs induced with both dexamethasone and TGF-beta1. At both time points, Gross observation and histology showed that the defects in most of group A were repaired by engineered fibrocartilage and cancellous bone with an irregular surface, minority defects were repaired by engineered hyaline cartilage and cancellous bone. However, in most of group B, the defects were completely repaired by engineered hyaline cartilage and cancellous bone. No repair or only fibrous tissue were observed in groups C and D. Besides, the compressive moduli of repaired cartilage in groups A and B reached 30.37% and 43.82% of normal amount at 3 months and 62.69% and 80.27% at 6 months respectively, which was further supported by the high levels of GAG contents in engineered cartilage of group A (78.03% of normal contents) and group B (no statistical difference from normal contents). More importantly, confocal microscope revealed the presence of GFP-labeled cells in engineered cartilage lacuna and repaired underlying cancellous bone.

CONCLUSION

The results demonstrated that implanted BMSC can differentiate into either chondrocytes or osteoblasts at different local environments and repair a complex articular defect with both engineered cartilage and bone. TGF-beta1 and dexamethasone in vitro induction can promote chondrogenic differentiation of BMSC and thus improve the results of repairing articular defects.