You are here

| Pluripotent Stem Cells

Changing Methylation Accompanies Changing Culture

"Epigenetic stability, adaptability, and reversibility in human embryonic stem cells”

Human embryonic stem cells (hESCs) are known to adapt during extended periods of ­­in vitro culture, a process potentially detrimental to later differentiation ability. However, the mode of initiation of transcriptional changes and the potential for reversibility of adaptation mediated by culture conditions is relatively unknown. DNA methylation is a potential mechanism behind transcriptional changes during culture adaptation through the reinforcement of environmental cues (Bird 2002 and Riggs 1989), but the recently observed plasticity of DNA methylation (Gong and Zhu 2011 and Riggs and Xiong 2004) suggests that it may also allow for the reversal of observed transcriptional changes. Researchers from the laboratory of Arthur D. Riggs have now analysed hESCs transcriptional and DNA methylation status during passaging and transfer between different growth conditions, finding that DNA methylation patterns appear to be remarkably stable but that a few regions of differential methylation are significant and often irreversible (Tompkins et al).

The model studied herein involved the initial growth of HES-2 hESCs on murine embryonic fibroblasts (MEF)/DMEM F12 with knockout serum replacement (MEF/DF12), transfer to one of two widely used and commercially available defined systems (Matrigel/mTesR1 [Mat/mTR] or CELLstart/STEMPRO [CS/SP] adhesive matrix and medium combinations), and after several passages return of the hESCs to the original MEF/DF12 conditions. hESCs under all conditions exhibited a normal karyotype, high pluripotency marker expression and minimal expression of differentiation markers. Transcriptional analyses found that 386 and 627 transcripts were differentially expressed in Mat/mTR and CS/SP adapted cells respectively, compared with initial MEF/DF12 cultures, with the vast majority of these changes being respective culture-specific as only 37 up-regulated and 54 down-regulated transcripts were shared between Mat/mTR and CS/SP cultures. As could be expected, upregulation of factors which facilitate cellular adhesion unique to Matrigel and CELLstart matrices were observed, alongside large expression variation in factors controlling TGF-β and Wnt signaling, consistent with both pathways being targeted by mTesR1 and STEMPRO media (Akopian et al).

Interestingly, upon reversal of culture conditions back to MEF/DF12, the original hESC transcriptional profile did not appear and hESCs grown under Mat/mTR and CS/SP conditions were distinct, suggesting a lack of reversibility. The number of differentially expressed genes between Mat/mTR and CS/SP conditions compared to initial adaptation was higher and several genes upregulated in the Mat/mTR and CS/SP conditions continued to be expressed for up to 7-10 passages after reversion; these generally being transcripts involved in cell adhesion. Additional transcriptional changes involved genes associated with cellular stress responses, including p53 signaling, cellular homeostasis, cell adhesion, and metabolism, perhaps indicating the difficulty of hESC condition reversion.

Next, DNA methylation associated with these changes was analysed and findings initially suggested that DNA methylation patterns across samples were similar. Despite the large amount of transcriptional changes observed after adaption of MEF/DF12 grown cells to Mat/mTR and CS/SP, only 20 and 14 differentially methylated regions (DMRs) were identified. These sites, which were often found at, or near, regulatory sequences for protein coding genes, were generally sites of hypermethylation and were highly specific and stable over additional culture time. Overall, this suggests that the cellular response to a change in culture environment involves site-specific and stable DNA hypermethylation. Excitingly, upon reversion of the Mat/mTR and CS/SP grown cells back to MEF/DF12 conditions, nearly all DMRs returned to their initial status, cumulatively representing a 90% reversion of culture-induced DNA methylation changes. Also noted were several additional sites of DNA hypomethylation, most clustered on chromosome X p11.4, a region which has been shown to be prone to escape from X-chromosome inactivation (XCI) (Esposito et al 1997 and Teichroab et al 2011). However, some DMRs were seen to be irreversible, and these few sites of residual DNA methylation may persist as a culture-induced “memory” of cellular manipulation. One shared site between Mat/mTR and CS/SP was in the PHF17 (JADE-1) gene promoter, which is a candidate tumor suppressor in renal tumorigenesis through a Wnt-signaling connection (Chitalia et al 2008), while others were discovered at the monocyte-differentiation-associated miRNAs mir-503 and mir- 424 in CS/SP cultures (Forrest et al 2010).

DMRs were also noted at several adhesion genes (MXRA5, PCDHB18, CDH22, and, PLEKHA2), as expected from the transcriptional analyses and several DMRs identified during adaptation, from and back to MEF/DF12, were shared between Mat/mTR and CS/SP-cultured hESCs. DMRs were only observed at a few developmentally-associated genes, alongside the miRNAs mentioned, examples being HOXB13, NOTCH4, and ALXI. Analysis of the remaining DMRs found them to be associated with genes involved in G-protein-coupled signaling (MLNR, RAB14, RAB5C, ARRB2, and GPSM3).

Analysis of sites of methylation and relevance to expression was next assessed. As previously observed, DNA methylation at promoter sequences was associated with lower gene expression; however there was no clear link between culture-induced intragenic DMRs and gene expression or repression. Strong correlation between DNA methylation and gene repression was observed only for the promoter-proximal region [−500 to +610 bp of transcription start sites (TSS)], while DNA methylation at upstream promoter distal regions (−501 to −2,440 bp) correlated more strongly with increased gene expression. Indeed, in the promoter-distal region, loss of DNA methylation at DMRs was well correlated with gene silencing, such as for the GPSM3 gene. GPSM3 promoter-distal DNA methylation is associated with GPSM3 expression in Mat/mTR and CS/SP adapted hESCs and subsequent reversion of conditions back to MEF/DF12 led to the loss of DNA methylation and a decrease in gene expression.

Overall, this work suggests that transcriptional changes occur during culture adaptation are associated with changes in DNA methylation, and that these methylation events are ‘plastic’ and generally revert to their original status after reversion of hESCs to initial culture conditions after adaption. Few DMRs exhibited irreversible DNA methylation, but some of these were associated with pluripotency-associated miRNA expression, the influence of which on subsequent differentiation is unknown.  However, hundreds of genes transcriptional status changed in reverse-adapted hESCs, adopting unique transcriptional profiles from the initial MEF/DF12 conditions, suggesting that DNA methylation is not implicitly linked to the expression status of these genes. Indeed, transcriptional analysis suggested a stress response in hESCs which may initiate a positive-feedback which then reinforces a subset of genes expression through other means, such as changes in  histone modifications whose dynamics are far more plastic than DNA methylation and may play a more substantial role in culture adaptation. It also raises concerns that long term stresses on hESCs over prolonged culture and changes to growth conditions may affect their use for cell-transplantation therapy, indicating that properly controlled conditions may allow their successful implementation.



Akopian V et al.
International Stem Cell Initiative Consortium (2010) Comparison of defined culture systems for feeder cell free propagation of human embryonic stem cells.
In Vitro Cell Dev Biol Anim 46:247–258.

Bird A (2002)
DNA methylation patterns and epigenetic memory.
Genes Dev 16: 6–21.

Chitalia VC et al. (2008)
Jade-1 inhibits Wnt signalling by ubiquitylating beta-catenin and mediates Wnt pathway inhibition by pVHL.
Nat Cell Biol 10:1208–1216.

Esposito T et al. (1997)
Escape from X inactivation of two new genes associated with DXS6974E and DXS7020E.
Genomics 43:183–190.

Forrest AR et al. (2010)
Induction of microRNAs, mir-155, mir-222, mir-424 and mir-503, promotes monocytic differentiation through combinatorial regulation.
Leukemia 24:460–466.

Gong Z, Zhu JK (2011)
Active DNA demethylation by oxidation and repair.
Cell Res 21: 1649–1651.

Riggs AD (1989)
DNA methylation and cell memory.
Cell Biophys 15:1–13.

Riggs AD, Xiong Z (2004)
Methylation and epigenetic fidelity.
Proc Natl Acad Sci USA 101:4–5.

Teichroeb JH et al (2011)
Suppression of the imprinted gene NNAT and X-chromosome gene activation in isogenic human iPS cells.
PLoS ONE 6:e23436.

Tompkins JD, et al (2012)
Epigenetic stability, adaptability, and reversibility in human embryonic stem cells.
Proc Natl Acad Sci U S A. 109:12544-9.


Article originally appeared from PNAS.

STEM CELLS correspondent Stuart P Atkinson reports on those studies appearing in current journals that are destined to make an impact on stem cell research and clinical studies.