Background: Magnetic resonance imaging (MRI) is a promising tool for monitoring stem cell- based therapy. Conventionally, cells loaded with ironoxide nanoparticles appear hypointense on MR images. However, the contrast generated by ironoxide labeled cells is neither specific due to ambiguous background nor quantitative. A strategy to overcome these drawbacks is F-19 MRI of cells labeled with perfluorocarbons. We show here for the first time that human neural stem cells (NSCs), a promising candidate for clinical translation of stem cell- based therapy of the brain, can be labeled with F-19 as well as detected and quantified in vitro and after brain implantation. Methodology/Principal Findings: Human NSCs were labeled with perfluoropolyether (PFPE). Labeling efficacy was assessed with F-19 MR spectroscopy, influence of the label on cell phenotypes studied by immunocytochemistry. For in vitro MRI, NSCs were suspended in gelatin at varying densities. For in vivo experiments, labeled NSCs were implanted into the striatum of mice. A decrease of cell viability was observed directly after incubation with PFPE, which re-normalized after 7 days in culture of the replated cells. No label-related changes in the numbers of Ki67, nestin, GFAP, or bIII-tubulin+ cells were detected, both in vitro and on histological sections. We found that 1,000 NSCs were needed to accumulate in one image voxel to generate significant signal-to-noise ratio in vitro. A detection limit of similar to 10,000 cells was found in vivo. The location and density of human cells (hunu+) on histological sections correlated well with observations in the F-19 MR images. Conclusion/Significance: Our results show that NSCs can be efficiently labeled with (19)Fwith little effects on viability or proliferation and differentiation capacity. We show for the first time that F-19 MRI can be utilized for tracking human NSCs in brain implantation studies, which ultimately aim for restoring loss of function after acute and neurodegenerative disorders.
