Osteochondral Angiogenesis In Rat Mandibular Condyles With Osteoarthritis-like Changes

Osteoarthritis (OA) is a group of joint diseases characterized by the progressivedegradation of articular cartilage, osteophyte formation, and thickening of thecalcified cartilage layer. The pathogenesis of OA has not yet been clear. Recentstudies have showed that osteochondral angiogenesis may play an important rolein the development of OA. The TMJ is unique compared with other load-bearingjoints in some aspects, for example, the surface of mandibular condyle iscomposed of fibrocartilage instead of hyaline cartilage. Some authors evenhypothesize the molecular events that govern responses to injury in the TMJ aresignificantly different from those of other joints. But unfortunately, so far nostudies on the osteochondral angiogenesis in TMJ-OA were reported. In thepresent study, we used our recently reported rat model with obvious OA-likechanges in the mandibular condyle to investigate angiogenesis at theosteochondral junction of TMJ. We further evaluated, at the protein and mRNAexpression levels, the pro-angiogenic and anti-angiogenic factors in the mandibular condyle of these rats. In addition, recombinant adeno-associatedvirus vector was used to explore the effect of vascular endothelial growth factoron the hypoxia-inducible factor1expression of chondrocytes.Experiment one: Long term observation of the condylar cartilage of ratsinduced by experimentally created disordered occlusionObjective: To investigate the long term effects of occlusal disorder on the TMJcondoylar cartilage. Materials and methods: Twenty-four femaleSprague–Dawley rats at8weeks of age were randomly divided into experimentalgroups (Exp) and sham-operated control groups. An elastic rubber band wasinserted between the first and second molars of the left side of the maxillarydentition and of the right side of the mandibular dentition. Four weeks after thebeginning of the experiment, the occlusion was further disturbed, using the samemethod to move the left maxillary and right mandibular third molars distally. Thesham-control rats were all subjected to the same procedure but without disturbingthe occlusion. Experimental animals, together with their age-matched controls,were sacrificed at the end of the20th and24th week after the beginning of theexperiment, named accordingly as the20-and24-week subgroup. HE and Alcianblue (AB)&periodic acid-Schiff (PAS) staining were used for histologicalassessment. The thickness of the condylar cartilage was measured at quarterpoints of the centre and posterior thirds using a computer assisted imageanalysing system. Results: In Exp groups, two first and two third molars weremoved away from their original positions and no longer intercuspated with theiropposite molars, which led to the experimentally created disordered occlusion.Histological observation showed that in control groups the condylar cartilagesurface was smooth and chondrocytes were homogenously distributed throughoutthe cartilage. In contrast, the Exp groups showed features of OA-like changes, e.g. loss of cartilage surface integrity, disarrangement of cellular disposition, theabnormal thickening of the fibrous layer, cell-free area, and osteophyte-likeformation. The thickness of the whole layer of condylar cartilage significantlyincreased in the20-week Exp group compared to that in the age-matched control(P <0.05), but not in the24-week Exp group. The thicknessesof the calcifiedcartilage layer and fibrous layer in both Exp groups also increased significantlyover their controls (all P <0.05). Conclusions: Experimentally createddisordered occlusion had a long lasting damaging effect on the condylar cartilageof rats. And the confirmation of the tidemark established a solid foundation forthe further experiments.Experiment two: Evaluation of osteochondral vascularityObjective: To detect the osteochondral angiogenesis in the mandibular condyleof rats induced by the experimentally created disordered occlusion. Materialsand methods: After the establishment of the rat model, the rats were sacrificedat the end of the20th and24th week after the beginning of the experiment.Osteochondral vascularity was determined by respectively counting the numberof vascular channels terminating in the calcified cartilage and non-calcifiedcartilage along the centre and posterior thirds of the condyle, and theirrelationships with OA-like changes in the condylar cartilage were also evaluated.CD34immunofluorescence staining was used to further assess the osteochondralvascularity. Results：HE staining showed in the control groups vascular channelswere clearly confined within the calcified cartilage layer, below the tidemark. Inthe Exp groups, vascular channels containing erythrocytes could be observedcrossing the tidemark and breaching into the non-calcified cartilage, and thechondrocytes around these channels were disorderly arranged or even appearedto be necrotic. CD34immunofluorescence showed that in the Exp groups the blood vessels in the vascular channels could reach the deep calcified cartilage,and even breakthrough the tidemark into the non-calcified cartilage. The numberof vascular channels terminating in the calcified cartilage was greater in both Expgroups over their age-matched controls (both P <0.05). Also, the number ofvascular channels terminating in the non-calcified cartilage was significantlygreater in both Exp groups compared with control groups (both P <0.05).Conclusions: From different angles, osteochondral angiogenesis was identifiedfor the first time on the present TMJ-OA rat model, and this phenomenon mayhave a direct link with the condylar cartilage degradation.Experiment three: Evaluation of pro-and anti-angiogenic factors at theprotein and gene levelsObjective: To study the mechanisms of osteochondral angiogenesis in themandibular condyle with OA-like changes. Materials and methods: The ratswere sacrificed at the end of the20th and24th week after the beginning of theexperiment. The right TMJ tissue blocks in each group were dissected forparaffin sections making, and the left condylar head for RNA extraction. UsingReal-time PCR and immunohistochemistry to detect the gene and protein levelsof vascular endothelial growth factor (VEGF), connective tissue growth factor(CTGF), matrix metalloproteinase9(MMP9), and Chondromodulin-I (CHM-I).Results: In the Exp groups, the hypertrophic chondrocytes adjacent to theosteochondral interface showed increased expression for VEGF, CTGF andMMP9, and immunopositive cells were also found in the vascular channels at theosteochondral junction. The immunopositive staining for CHM-I was stillapparent, but the expression was more evident in the superior hypertrophicchondrocytes, and not in the inferior hypertrophic layer adjacent to the bloodvessels at the osteochondral junction. Compared to their age-matched controls, the protein levels of VEGF and CTGF were higher in20-week experimentalgroup, and the protein and mRNA levels of CTGF, MMP-9, and CHM-Iincreased in the24-week experimental group (all P <0.05). Conclusions:Abnormal mechanical forces caused by the disordered occlusion, may stimulatethe chondrocytes to secret angiogenic factors including VEGF, CTGF andMMP9. With disruption of the tidemark, angiogenic factors could reach theosteochondral junction by mass transport and diffusion from the cartilage matrix.And the subchondral bone may also contribute the angiogenic stimuli throughexpression of angiogenic factors by osteoblasts. The accumulation of VEGF,CTGF and MMP9at osteochondral junction established an angiogenicmicro-environment which stimulated endothelial cell migration and proliferation,and finally led to the osteochondral vascularization. Although the increase inCHM-I may moderate pro-angiogenic factors effects in the superior cartilage, thedeficiency of deep hypertrophic chondrocytes to express CHM-I may permitvascular invasion into condylar cartilage.Experiment four: The induction and identification of ATDC5cell line andthe effects of recombinant adeno-associated virus vector rAAV-VEGF onthe experisson of hypoxia-inducible factor1Objective: To induce differentiation of ATDC5cell into chondrocyte-like cells,and to detect the effects of VEGF on the cell proliferation and hypoxia-induciblefactor1(HIF-1) expression. Materials and methods:Insulin-transferring-selenium was used to induce ATDC5cells to differentiateinto chondrocyte-like cells. Differentiation into chondrocyte-like cells wasconfirmed by the expression of type II collagen, as well as the production ofglycosaminoglycan visualized by Alcian Blue staining. The rAAV vectorencoding VEGF gene was transfected into the chondrocyte-like cells and the methyl thiazolyl tetrazolium method was used to evaluate celll proliferation. Theexperissions of VEGF and HIF-1were further detected by theimmunofluorescence. Results:1week after the beginning of the induction, thecells, at condensation stage, had begun to secret glycosaminoglycan;2weeksafter the beginning of the induction, the cells expressed type II collagen and theglycosaminoglycan production was further increased;3weeks after thebeginning, cartilage nodules were observed. rAAV-VEGF could be successfullytransfected into the chondrocyte-like cells. VEGF could enhance the cellproliferation and HIF-1expression in the chondrocyte-like cells. Conclusions:ATDC5cells could be induced differentiation into chondrocyte-like cells. VEGFcould increase the HIF-1expression in the chondrocyte-like cells, suggesting thatthere may be a positive, mutually reinforcing relationship between HIF-1andVEGF in the development of OA.