Analysis of reprogramming-associated alterations using an isogenic human stem cell system

Abstract

Human induced pluripotent stem cells (iPSCs) could provide a valuable tool for the production of specific somatic cell types for disease modeling or cell replacement. Considering the molecular complexity of the reprogramming process and the requirement of major epigenetic rearrangements for the generation of induced pluripotent stem cells, it is critical whether and to what extent reprogramming-associated alterations can confound readout parameters in disease modeling or influence clinical safety of human iPSC-derived cell populations. Several comparative analyses of iPSCs and human embryonic stem cells (ESCs) already tried to unravel this question. In the majority of studies, however, variances due to distinct genetic backgrounds challenged assertion of their equivalence, indicating the need for genetically matched cells in order to discover even minor differences. Moreover, most approaches focused on pluripotent populations and only little is known about the effects on iPSC-derived somatic cell types. Since some reprogramming-associated alterations may not be detectable in the pluripotent state but, nevertheless, could become relevant in somatic cell populations used in research and therapy, the present study aimed at addressing to what extent somatic stem cells, passed through reprogramming and subsequent differentiation into their original fate, maintain their native transcriptional and methylation signatures. To that end, highly standardized and well-characterized human ESC-derived neural stem cells (ESC-NSCs) were reprogrammed into iPSCs, which were subsequently re-differentiated into NSCs (iPSC-NSCs). Global transcription and DNA methylation profiling of this isogenic system revealed a remarkably similar transcriptome and methylome of both NSC populations with only minor differences. Among these yet identified alterations, there was a disproportionately large fraction of X-chromosomal genes and methylation sites, which regionally coincided with each other. While this data point to a extensive overall reinstallation of transcriptomic and methylation signatures upon sojourn through pluripotency, they also indicate that X-chromosomal genes may escape this reinstallation process and thus corrupt downstream applications such as iPSC-based disease modeling and regenerative approaches. Therefore, these results strongly recommend comprehensive and tight epigenetic monitoring of iPSCs and iPSC-derived populations in their biomedical application as well as considerate experimental design and choice of readout parameters in disease modeling and drug discovery approaches.