Transcriptome Profiling Reveals Degree of Variability in Induced Pluripotent Stem Cell Lines: Impact for Human Disease Modeling.

Induced pluripotent stem cell (iPSC) technology has become an important tool for disease modeling. Insufficient data on the variability among iPSC lines derived from a single somatic parental cell line have in practice led to generation and analysis of several, usually three, iPSC sister lines from each parental cell line. We established iPSC lines from a human fibroblast line (HDF-K1) and used transcriptome sequencing to investigate the variation among three sister lines (iPSC-K1A, B, and C). For comparison, we analyzed the transcriptome of an iPSC line (iPSC-K5B) derived from a different fibroblast line (HDF-K5), a human embryonic stem cell (ESC) line (ESC-HS181), as well as the two parental fibroblast lines. All iPSC lines fulfilled stringent criteria for pluripotency. In an unbiased cluster analysis, all stem cell lines (four iPSCs and one ESC) clustered together as opposed to the parental fibroblasts. The transcriptome profiles of the three iPSC sister lines were indistinguishable from each other, and functional pathway analysis did not reveal any significant hits. In contrast, the expression profiles of the ESC line and the iPSC-K5B line were distinct from that of the sister lines iPSC-K1A, B, and C. Differentiation to embryoid bodies and subsequent analysis of germ layer markers in the five stem cell clones confirmed that the distribution of their expression profiles was retained. Taken together, our observations stress the importance of using iPSCs of different parental origin rather than several sister iPSC lines to distinguish disease-associated mechanisms from genetic background effects in disease modeling.