Although green fluorescent protein (GFP) labeling is widely accepted as a tracking method, much remains uncertain regarding the retention of injected GFP-labeled cells implanted in ischemic organs. In this study, we evaluate the effectiveness of GFP for identifying and tracking implanted bone marrow- mesenchymal stem cells (BM-MSCs) and the effect of GFP on the paracrine actions of these cells. MSCs isolated from rat femur marrow were transduced with a recombinant adenovirus carrying GFP. After transplantation of the GFP-labeled BM-MSCs into the infarct zone of rat hearts, the survival, distribution, and migration of the labeled cells were analyzed at 3, 7, 14, and 28 days. To evaluate the effect of GFP on the paracrine actions of BM-MSCs, Western blot analysis was performed to detect the expression of vascular endothelial growth factor (VEGF), b fibroblast growth factor (b FGF), tissue inhibitor of metalloproteinase-1 (TIMP-1) and matrix metalloproteinases-2 (MMP-2). GFP was successfully expressed by BM-MSCs in vitro. At 14 days after cell transplantation the GFP-positive cells could not be detected via confocal microscopy. By using a GFP antibody, distinct GFP-positive cells could be seen and quantitative analysis showed that the expression volume of GFP was 6.42 ± 0.92 mm(3) after 3 days, 1.24 ± 0.76 mm(3) after 7 days, 0.33 ± 0.03 mm(3) after 14 days, and 0.09 ± 0.05 mm(3) after 28 days. GFP labeling did not adversely affect the paracrine actions of BM-MSCs. GFP labeling could be used to track MSC distribution and their fate for at least 28 days after delivery to rat hearts with myocardial infarction, and this stem cell tracking strategy did not adversely affect the paracrine actions of BM-MSCs.