Noninvasive Method For In Vivo Tracking Of The Regeneration Of Tissue Engineering Articular Cartilage

Articular cartilage defects are common and caused by many reasons. However,treatments including conservative treatment, joint debridement, autologous or allogeneicbone cartilage transplantation and artificial joint replacement have obvious limitations suchas incomplete repair. Tissue-engineering is considered as a new and effective therapyproposal of articular cartilage defects in future. But it is a key problem to know about thecharacteristics of seed cells in repairing process because of the lack of methods regardingthe efficient and noninvasive technique to monitor the in vivo behavior of delivered cells inhost tissue. Recently, magnetic resonance imaging (MRI) in vivo tracking of magneticallylabeled cells technology provides a new train of thought. Superparamagnetic iron oxideparticles (SPIO), as a negative image contrast agent can significantly reduce T2WI and T2*WI signal, thus the inplanted cells and original tissue cells can be distinguished from eachother easily. However, as to labeling of MSCs with SPIO, the uptake of general SPIO bycells is very low, which results in insuffieientiron load in single cell. Also, the quick cellproliferation will lead to rapid dilution of iron load in single cell. And the current SPIOproducts can’t meet the requirements for long-term dynamic MR tracking.PurposeThe aim of this study is to coat iron oxide nanoparticles with polyethylenimine（PEI）to form a new type of superparamagnetic particles, to investigate the influence of magneticlabeling on the biological properties of the seeded cells, to in vivo track of SPIO-labeledBMSCs to evaluate the migration and prognosis in animal cartilage defect model, and todetermine the feasibility of MRI tracking magnetically labeled seeded cells for articularcartilage tissue engineering noninvasively and repeatedly. Seeded cells labelled with greenfluorescent protein (GFP) were used as the control. So that we developed an efficient and noninvasive imaging technique for clinical application of tissue engineered articularcartilage products.Method1. Preparation of new controllable degradable superparamagnetic nanoparticlesSPIO nanoparticles were prepared according to synthetic method described in aprevious publication by Sun et al. These nanocrystals were dried under argon andredispersed in chloroform together with alkylated PEI. Then, physicochemical propertieswere detected by using scanning electron microscope (SEM), dynamic light scattering(DLS), the magnetic induction intensity detection and relaxation rate assay.2. Feasibility study of SPIO labeling of seeding cells for articular cartilage tissueengineeringPorcine BMSCs were isolated by gradient centrifugation method and then expanded.Third passage of BMSCs were coincubated with PEI/SPIO with different concentration of4μg/ml、6μg/ml、8μg/ml、10μg/ml and12μg/ml for24h. Labeled cells were compared withunlabeled cells by Prussian blue staining, transmission electron microscope test, atomicabsorption spectrometer detection, Trypan blue staining, MTT test and multipledifferentiation induction of osteogenesis, chondrogenesis and adipogenesis to evaluate theeffects of SPIO on cell labeling efficiency, cell viability, proliferation and differentiationability. Different concentration of BMSCs (5×10~6、1×10~6、5×10~5、1×10~5、5×10~4、1×10~4、5×10~3) labeled with different dose of PEI/SPIO(4μg/ml、6μg/ml、8μg/ml、10μg/ml、12μg/ml)underwent3.0T MRI with SET2*WI weighted sequence.3. In vivo MR Imaging Tracking of SPIO Labeled Bone Marrow Mesenchymal StemCells for Cartilage Defect Repair in MinipigSPIO (8μg/ml) and GFP double labeled BMSCs were mixed into1ml of type IIcollegen hydrogel and inplanted into the cartilage cartilage defect of knee joints in minipigmodel. Eighteen Guizhou minipigs were randomly divided to three groups. In experimentalgroup (grou A), BMSCs labeled with SPIO and GFP that were mixed into type II collegenhydrogel were transplanted into articular cartilage defects. In nective control group (groupB), BMSCs labeled with GFP that were mixed into type II collegen hydrogel weretransplanted. In blank control group (group C), no treatment was performed to cartilagedefects of minipig. At4,8,12and24weeks postsurgery, the minipigs underwent3.0-T MR imaging with SET2*WI weighted sequence. And the results were compared with histologyat24weeks.Result1. A new type of SPIO was successfully synthesized. Particle size was arround79±28nm by scanning electron microscopic and DLS detection. The saturationmagnetization rate was48emu/g, while the T2relaxation rate was323Fe mM/s. Theparticle possessed a positive surface charge of40mv solution with a final concentration of1.3275mg/ml.2. Prussian blue staining showed numerous blue-stained particles were in thecytoplasm of the labeled cells, transmission electron microscopy showed SPIO particlesexisted in cytoplasm. Atomic absorption spectrophotometer testing showed thatintracellular iron increased when the concentration of SPIO increased. Data of intracellulariron at8μg/ml reached its peak. The intracellular iron droped when the SPIOconcentration continued to increase. In4,6and8μg/ml group, Trypan blue staining andMTT test showed that no statistically significantly difference was found in cell viability,proliferation between labeled and unlabeled cells (P＞0.05). Osteogenic, chondrogenic andadipogenic inductions demonstrated that BMSC had the ability to differentiate intoosteoblasts chondroblasts and adiptoblasts. In10,12μg/ml group, there were obviouslystatistically significantly differences in the viability, proliferation between labeled andunlabeled cells (P <0.05), and most of cells deceased in the procedure of osteogenic,chondrogenic and adipogenic inductions. Results of3.0T MR scan indicated that signal of5×10~6、1×10~6、5×10~5、1×10~5、5×10~4labeled cells were significantly lower than that ofunlabeled cells at same concentration (P <0.05); percent of SI loss (△SI) increased whenthe cell concentration increased, with significant difference (P <0.05).3. Four week postsurgery, we found that the signal in the defect of group A apparentlydecreased, there was no similar change in the surrounding synovial tissue or joint space. Ingroup B, we found that there was no signal change in the defect region while there was anincreased signal change in the defect of group C. At8w postoperative, MRI showed thathypointense signal of group A was still obvious with no change signal in the surroundingsynovial tissue and joint space. After12weeks, hypointense signal of group A weakenedalthough there was still a significant difference between that of group B. At24weeks postsurgery, cartilage defect signal of group A got weaker, signal change range was alsosignificantly reduced. B and C control group had no obviously signal changes. On differentphases of cartilage defect repair, SI values of SET2*WI MRI image sequences has a verysignificant difference between the4weeks,8weeks,12weeks and24weeks groups. Grossand histological analysis found that postoperative24weeks hyaline-like cartilage repairregenerated with a smooth surface and integration to the surrounding cartilage closely ingroup A and B. In the defect region of group A, we found the Prussian blue positive cellsand expression of GFP while in group B no Prussian blue positive cell was found. In groupC fibrous tissue regenerated with rough surface and derangement of extracellular matrixConclusion1. The newty SPIO which has good physical and chemical properties (saturationmagnetization rate and T2relaxation rate) was successfully synthesized.2. PEI/SPIO of8μg/mL can effectively label porcine BMSCs, without influencing thecell viability, proliferation and the multiple differentiation ability．The magnetically labeledcells in the MR produced characteristic low signal changes. Tracking SPIO-labeled cells invitro by clinical3.0T MR SET2*WI sequences were viable. The minimum detectionthreshold was for5×10~4/cm~3.3. Obvious change of signal was found in the defect region after surgery onSET2-weighted MR image and persisted for at least24weeks. In vivo MRI tracking ofPEI/SPIO labeled BMSCs mixed with type II collagen hydrogel transplanted to cartilagedefect in minipig model is feasible and efficient. This would show the distribution anddiffusion of the labeled cells, thereby elucidating the regenerative mechanisms andproviding opportunities to improve current repair strategies.