Purpose: To investigate the effect of different ultrasound intensities on enhanced green fluorescence protein (EGFP) gene expression in tissue between bone defects by ultrasound-mediated microbubble destruction (UMMD) and to evaluate the tissue damage after transfer. Method: Thirty rabbits were randomly divided into six groups: A1, A2, A3, A4, A5 and A6 (six in each group of A1-A4, and three in each group of A5-A6). Bone defect models were generated on the right ulnas. On the 10th day after model generation, suspension of EGFP plasmids and microbubbles (PM, 0.3 ml/kg) was locally injected to groups A1-A4, while plasmid-saline (PS, 0.3 ml/kg) was locally injected to groups A5 and A6. Then, ultrasound was given to the 6 groups at 0.5 W/cm(2), 1.0 W/cm(2), 1.5 W/cm(2), 2.0 W/cm(2), 1.0 W/cm(2), and 0 W/cm(2), respectively, with all other parameters remaining constant. Two rabbits were sacrificed for sample harvesting on the 3rd day, 7th day, and 14th day after gene transfer in groups A1-A4, and all rabbits were sacrificed on the 7th day after gene transfer in groups A5 and A6. The EGFP expression was quantified by average optical density under a fluorescence microscope. Tissue damage was observed under an optical microscope and an electron microscope. Results: On the 3rd day after transfection, groups A1 and A2 both had higher AOD (P<0.05) when compared to the others. However, there was no significant difference between these two groups (P>0.05). On the 7th day after transfection, AOD was significantly higher in groups A1-A4 than that in A5 and A6 (P. 0.05), among which, groups A1 and A2 were the groups with the highest AOD (P<0.05). On the 14th day after transfection, there was no significant difference in AOD measurement among all groups (P>0.05). Tissue damage was detected under an optical microscope and an electron microscope in all groups at all observation points. A correlation between the severity of tissue damage and the intensity of experimental ultrasound was observed. Conclusions: EGFP gene can be efficiently transfected without obvious toxicity in bone defects of rabbits at 0.5 W/cm(2), when other parameters were set as follows: frequency: 1 MHz, duty ratio: 20%, and exposure duration: 1 min.