Optimization of in vitro cell labeling methods for human umbilical cord-derived mesenchymal stem cells.

BACKGROUND AND OBJECTIVES: Human umbilical cord-derived mesenchymal stem cells (hUCMSCs) are a novel source of seed cells for cell therapy and tissue engineering. However, in vitro labeling methods for hUCMSCs need to be optimized for better detection of transplanted cells.

AIM OF THE STUDY: To identify the most stable and efficient method for labeling hUCMSCs in vitro.

MATERIALS AND METHODS: hUCMSCs were isolated using a modified enzymatic digestion procedure and cultured. hUCMSCs of passage three (P3) were then labeled with BrdU, PKH26, or lentivirus-GFP and passaged further. Cells from the first labeled passage (LP1), the fourth labeled passage (LP4) and later passages were observed using a fluorescence microscope. The differentiation potential of LP4 cells was assessed by induction with adipogenic and osteogenic medium. Flow cytometry was used to measure the percentage of labeled cells and the percentage of apoptotic or dead cells. The labeling efficiencies of the three hUCMSC-labeling methods were compared in vitro.

RESULTS: BrdU, PKH26, and lentivirus-GFP all labeled LP1 cells with high intensity and clarity. However, the BrdU labeling of the LP4 cells was vague and not localized to the cell nuclei; LP9 cells were not detected under a fluorescence microscope. There was also a significant decrease in the fluorescence intensity of PKH26-labeled LP4 cells, and LP11 cells were not detected under a fluorescence microscope. However, the fluorescence of LP4 cells labeled with lentivirus-GFP remained strong, and cells labeled with lentivirus-GFP were detected up to LP14 under a fluorescence microscope. Statistical analyses indicated that percentages of LP1 cells labeled with PKH26 and lentivirus-GFP were significantly higher than that of cells labeled with BrdU (p < 0.05), and that the LP4 cells were more efficiently labeled with lentivirus-GFP than with PKH26 or BrdU (p < 0.05). BrdU-, PKH26-, and lentivirus-GFP labeled LP4 cells were all differentiated to adipocytes or osteoblasts with adipogenic and osteogenic medium. No statistical significance (p > 0.05) was observed between the death rates of labeled and unlabeled cells.

CONCLUSIONS: Lentivirus-GFP is a valid method for long-term in vitro labeling, and it may be used as a long-term hUCMSC tracker following transplantation in vivo.