Cells are generally labeled during in vivo implantation studies enabling the cells to be traced. The relationship between the labeling efficiency and cellular proliferation after transplantation is critical for the interpretation of data obtained by detection of the signals on tissue sections. Here, we compare cellular labeling methods of rat marrow stromal cells that were labeled with 5-bromo-2-deoxyuridine (BrdU), 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI) and fluorescence in situ hybridization (FISH). Our data show that (i) BrdU uniformly labeled the nuclei, (ii) DiI-labeled cells had many dots or stained clear and uniform when a longer exposure time was used during detection and (iii) FISH labeled the cells with dots along the edges of the nuclei. The labeling efficiency was 94.1+/-8.6%, 97.6+/-3.4% and 90.5+/-3.0%, in BrdU, DiI- and FISH-labeled cells, respectively. After sub-culturing of labeled cells, the percentage of BrdU-positive cells was found to be 71.9+/-18.0% and 18.4+/-6.9%, after the first and second passages, respectively. The percentage of DiI-labeled cells detected depended on the exposure time: a long exposure time (>10 s) resulted in identification of 95.1+/-4.0% and 94.5+/-3.9% DiI-positive cells after the first and second sub-cultures, respectively. The percentage of FISH-positive cells was found to be 87.0+/-3.0% and 89.1+/-9.7%. The BrdU labeling signal quickly decreased over time. Thus, BrdU should only be used to temporarily label dividing cells. In contrast, our data indicate that DiI and FISH labeling may be used to steadily trace cells during in vivo experiments. To our knowledge, this is the first time that the effects of different labeling methods over time have been examined during a cell transplantation study.