ESCs differences can result in functional differences
ESCs and iPS cells sharesimilarities in morphology, self-renew capacity, differentiation potential, andage-affected cellular systems such as telomeres and mitochondria. However,numbers studies comparing ESCs and iPS cells at the epigenetic,transcriptional, proteomic and metabolic levels have demonstrated the moleculardifferences between ESCs and iPS cells (Bock et al., 2011; Chin et al., 2009, 2010;Doi et al.
, 2009; Ghosh et al., 2010; Kim et al., 2010b; Lister et al., 2011;Loewer et al., 2010; Marchetto et al., 2009; Polo et al., 2010b).
At transcriptional level, human iPS cells and ESCs can bedistinguished by their differential expression of protein-coding RNAs whichmainly attributes to the residual expression of somatic genes but dissipateupon extended passaging (Chin et al., 2009, 2010). In addition, ten large intergenic non-coding RNAs (lincRNAs) aredifferentially expressed between human iPS cells and ESCs. Some of theselincRNAs participate in the reprogramming process that their overexpression enhances and thedownregulation inhibits reprogramming (Loewer et al., 2010). Thesetranscriptional differences can result in functional differences between iPScells and ESCs.
For example, therepression of a small group of non-coding RNAs encoded in the Dlk1–Dio3gene cluster could affect the functionality of mouse iPS cells. Althoughsome iPS cells lacking of expression at this locus are capable of generatingchimaeras, fail the tetraploid complementation assay (the gold standard forexamining mouse pluripotency) to show animals are entirely derived from thesecells (Stadtfeld et al., 2010). The chromatin state of iPSCs and ESCs has beenextensively examined to date and consistent differences are observed. Among histone modifications, genome-widestudies showed the expression patterns for H3K4me3 and H3K27me3 areindistinguishable between bothmouse and human iPS cells and their respective ESCs counterparts (Maherali et al.
, 2007; Mikkelsen et al.,2008). However, the expressionpatterns of H3K9me3 within promoter regions are different (Hawkins et al., 2010) and its overexpression were found among the genes which aredifferentially expressed between human iPS cells and ESCs. Notably, many of the transcriptional and chromatin differencesdescribed are observed in the early-passage iPS cells and disappeared at alater passage, suggestive of a residual ‘epigenetic memory’ persisting inthe early-passage iPS cells, reflecting the cell oforigin (Chin et al., 2009; Ghosh et al., 2010;Marchetto et al.
, 2009). Specifically, Kim et.al and Polo et al. provide functionalevidence showing that an epigenetic memory of cell of origin persist in mouseiPS cells which is linked to the residual DNAmethylation within lineage-specific genes. This persisting DNA methylationpattern influences the functionality and differentiation potential of derivediPS cells (Kim et al., 2010; Polo et al., 2010).
For example, iPS cells derived from blood cells are more easily to bedifferentiated to their original blood cell lineage than fibroblast-derivediPS. This may due to the DNA hypermethylation of blood cell markers infibroblast-derived iPS cells that potentially prevent their upregulation underthe induction of blood lineage differentiation. Reversely, treatments ofnon-blood cell-derived iPS cells with DNA methylation inhibitors enable a moreefficient differentiation towards blood lineage. Notably, this residualepigenetic memory in mouse iPS cells could be erased upon extended passaging. Furthermore, this epigenetic memory has been uncovered inhuman iPS cells. Single-cell based whole-genome DNA methylation mapping studiesreveal that the somatic DNA methylome was only partially erased duringreprogramming and an epigenetic memory of the somatic DNA methylation patternpersists in human iPS cells. In addition, some iPS cells fail to establishESC-like methylation pattern which is associated with transcriptional andfunctional differences found between late-passage iPS cells and ESCs (Lister et al., 2011).
Nevertheless, genetic abnormalities have been seen insome iPS cells. One study suggests that these abnormalities might due to theoncogenic stress induced by reprogramming factors. They observed a higher levelof phosphorylated histone H2AX, one of the earliest indicators of DNAdouble-strand breaks, in the cells induced with OSKM or OSK. They demonstratedthat the homologous recombination pathway is essential for repair DNAdouble-strand breaks to maintain genomic integrity during reprogramming process(González et al.
, 2013). Nevertheless, more evidence are required to settle whether thesedefects are as a result of reprogramming process or due to the genetic andepigenetic differences existing within the individual parental fibroblasts (Abyzov et al., 2012; Cheng et al., 2012).