Elimination of undifferentiated human embryonic stem cells in vitro

Abstract

Stem cells have been a focus of growing scientific interest both as a tool for studying early development and as a potential source of transplantable cells for regenerative medicine. Since their derivation in 1998, human embryonic stem (ES) cells have received much of this attention principally because of their ability to proliferate seemingly indefinitely in vitro and to differentiate along multiple lineages, essentially giving rise to every cell in the body (Reubinoff et ai, 2000; Thomson et al., 1998). However, before human ES-derived transplantation becomes a reality for patients, problems associated with prolonged and functional engraftment, histocompatability, homogeneous cell populations, and the risk of tumours arising from the accidental transplantation of undifferentiated human ES cells, must first be addressed.

Despite the risk of malignancy associated with undifferentiated human ES cells, their indefinite growth in culture has provided an opportunity to manipulate their fate. This thesis has investigated the possibility of using the exogenous cell surface markers; the Galal-3Galpl-4GlcNAc-R (a-gal) epitope and the murine major histocompatibility complex (MHC) class I molecule H2-Kk, under the transcription control of the human telomerase reverse transcriptase (hTERT) promoter to selectively identify undifferentiated human ES cells for selective elimination. Provided within is evidence to show the successful genetic manipulation of undifferentiated human ES cells with al,3galactosyltransferase (al,3Gal) gene, providing comparable cell surface expression of the a-gal epitope with endogenous a-gal expression on the ovine foetal fibroblast cell line (PDFF). However, the use of the H2-Kk gene as a potential lysis epitope was not so successful. Despite successful integration and transcription of the H2-Kk gene, cell surface expression of the H2-Kk epitope was not achieved, and proof of protein translation could not be found.

Expression of the a-gal epitope on undifferentiated human ES cells elicits a severe, yet highly specific cytolytic response; on average 95% of the transgenic human ES cells were lysed compared with just 8-12% of wild type non-expressing H9 cells, when exposed to human serum containing active-complement. In addition, when transgenic human ES cells were differentiated the a-gal epitope was down regulated, in the same manner as established markers of undifferentiated human ES cells (TRA1-81 and SSEA-4). Following differentiation the transgenic a-gal expressing cell line (M2) survived exposure to active serum-complement.

This novel system for selective ablation could potentially provide natural immune protection, through the presence of circulating antibodies to a-gal that would protect graft-recipients against the presence of, or de-differentiation of, human ES-cell derivatives following engraftment.