gms | German Medical Science

Objective: Current experimental therapies for spinal cord injury (SCI) include focal application of neurotrophic factors, biocompatible materials, and cellular transplantation. Recently, we engineered neural-tube-like neuroepithelial cells (NEC) from mouse embryonic stem cells (ESC). NEC are capable of forming an ependymal tube in vitro that contains an arranged network of precursors, neurons and their glial support. Regeneration after SCI in lower vertebrates is starting from such an ependymal tube. The aim of this study was to investigate the fate of these pre-conditioned and differentiated cells in a SCI model.

Methods: Sox1-GFP knock-in ES cells were generated in E14Tg2a.IV cells (129P2/OlaHsd mouse strain). GFP+ cells emerged in tubular conformations, were seeded in matrigel and revealed a neural phenotype that recapitulated spinal cord development and cellular organization including a lumen and an apical/basal orientation. Pure GFP+ tubes or neurospheres derived from primary murine cerebellar neural stem cells (NSC) were transplanted into the lesioned rat spinal cord. The lesion consisted of a hemimyelonectomy at the level of Th9-10 which was filled with the cellular graft. Five experimental groups were formed: 1) NEC tubes in matrigel, 2) NEC tubes only, 3) primary neurospheres in matrigel or 4) without matrigel, and 5) matrigel only.

Results: In 34 rats with the NEC tubes in matrigel, a tumor formation was seen. An initial slight recovery from the acute neurological deficit did not improve. The animals deteriorated after postinjury day (PID) 14. From PID 21 on, macroscopic tumor growth was seen. Histological staining revealed an inhomogeneous tumor with different tissue types classified as an immature teratoma. In animals with purified, matrigel-free NEC tubes, the tumors occurred in 2 out of 10 rats after 3 months. Reculturing of explanted tumors showed that the cells retained their ability to from NEC tubes. The animals transplanted with cell-free matrigel or with NSC did not show tumor formation.

Conclusions: ES cell derived implants need careful monitoring for tumor growth. In this case a differentiating step and selection for a differentiating marker was thought to minimize ES cell contaminations, however, tumorigenicity prevailed. Cotransplantation of matrigel potentiates the tumorigenicity. Changes in the protocol for purification and fall-back mechanisms are necessary for ES cell based therapies.