Abstract: Objective To compare different ways to trace bone marrow mesenchymal stem cells (BMSCs) after being transplanted in cerebral ischemia-reperfusion injury. Methods Male SD rats of SPF grade were randomly divided into sham group, model group (ischemia-reperfusion，IR), BrdU tracing group, PKH26 tracing group and GFP tracing group. Fo⁃ cal cerebral ischemia-reperfusion model was established by blocking middle cerebral artery. 24 hours after cerebral isch⁃ emia-reperfusion injury, 10 μL BMSCs that were labeled respectively by BrdU, PKH26, GFP were added respectively into BrdU, PKH26 and GFP tracing group while equal volum of normal saline was added into sham group and model group. Mod⁃ el and transplanting cells efficacy was determined by neural behavioral score, TTC staining and brain water content; Neurons were counted using tar violet staining; The number of transplant cells in the transplanting site was assessed by fluorescence microscopy. Results Before transplanting, there was no significant difference among BrdU, PKH26 and GFP group in cell labeled efficacy. By contrast, neural behavioral score, brain infarct volume and brain tissue water content were significantly lower in all three tracing groups than that in model group 4 weeks after transplantation while neuron counts were markedly higher. There was no significant difference of above parameters among the three tracing groups. However, the number of traced transplanting cells in damaging area in GFP group is significantly higher than that in BrdU group and PKH26 group. Conclusion In cerebral ischemia-reperfusion injury, the tracing effect of GFP last longer, therefore it is significantly more effective than BrdU and PKH26.

Key words: reperfusion injury, mesenchymal stem cell transplantation, green fluorescent proteins, fluorescent antibody technique, cerebral ischemia-reperfusion, bone marrow mesenchymal stem cells, trace