gms | German Medical Science

Objective: Given a limited capacity for self-repair of the central nervous system, cell replacement strategies have been demonstrated to be a promising therapeutic approach for dysfunction caused by death or impairment of neural cell types. Correct identification and localization of implanted cells is crucial when evaluating the success of transplantation therapy. Various methods of cell labelling and trafficking employed in previously published studies, are contrasted to our results and practical experiences.

Methods: Undifferentiated pluripotent enhanced-GFP-transfected murine embryonic stem (ES) cells were implanted into the ipsi/contralateral cortex of Sprague-Dawley rats 72 hours after traumatic brain injury (TBI). Animals were sacrificed at day 5 or week 7 postimplantation and evaluated by conventional, fluorescence and confocal microscopy.

Results: Five days after implantation, the identification of implanted cell graft by intrinsic GFP-autofluorescence failed. Anti-GFP labelling coupled to fluorescent and conventional antibodies was needed to visualize the implanted cells. Moreover, GFP-signal of the cells differentiating along neural axis seemed to be decreasing. 7 weeks post implantation, cell assemblies with non-specific autofluorescence were found at site of transplantation, additionally showing some positive reaction in anti-GFP-staining. This turned out to be a false positive phenomenon, for these cells were later identified as macrophages, mimicking the implanted stem cells. GFP-positive cells were eventually correctly localized, using advanced histological techniques.

Conclusions: We demonstrate a whole variety of previously undiscussed pitfalls and fallacies interfering with correct identification of implanted stem cells. The phagocytosis of implanted cells, observed in this study, also raises concerns about the methodological safety of in-vivo trafficking of nanoparticle-labelled stem cells.