You are here

| Pluripotent Stem Cells

Methylation Dictates Fate? Cell fate potential of human pluripotent stem cells is encoded by histone modifications

From Cell Stem Cell
By Stuart P. Atkinson

Multiple studies over the last 5 years have shown us that human embryonic stem cells (hESCs) growing under self-renewing conditions in vitro are surprisingly heterogeneous with individual cells displaying dynamic phenotypes (Hayashi et al and Stewart et al) while others studies have demonstrated cell-to-cell variance in the levels of pluripotency-associated transcription factors (Chambers et al, Hayashi et al and Toyooka et al). This heterogeneity is most easily observed during directed differentiation of hESCs which generally generates a spectrum of differentiated cell types which impacts on the usefulness of such strategies for therapeutic use. The study by Hayashi et al begins to unravel the links between heterogeneous ESC states in mouse and specific chromatin modifications and now results from Hong et al from the lab of Mick Bhatia at the McMaster University, Hamilton, Canada, presented in Cell Stem Cell, reveal the identification of cells in hESC cultures that are lineage biased, through the use of specific surface markers, and link this lineage bias to specific chromatin modification patterns.

Initial studies demonstrated that markers for mesoderm progenitors (c-KIT) and neural precursors (A2B5) (Carpenter et al) were present in hESC cultures under self-renewing conditions. c-KIT+ cells were found throughout colonies, whereas A2B5+ cells were found at the periphery of colonies only associated with the fibroblastic cells produced by hESCs in culture. Similar results were observed in multiple different hESC lines and in feeder-free and irradiated MEF co-culture growth systems suggesting that this is a general phenomenon. Flow cytometric analysis of hESC cultures using these markers revealed a composition of 35% c-KIT+, 10% A2B5+, 50% c-KIT-/A2B5-cells and 5% of cells c-KIT+/A2B5+; with this population strictly localized to the periphery of hESC colonies, perhaps representative of a dynamic transient state of primitive to lineage-primed cells. Upon studying pluripotency associated cell markers, it was found that they occupied specific fractions. OCT4 and SSEA3 negative cells were enriched in the c-KIT- fraction, OCT4 and SSEA3 positive cells in the A2B5- fraction, and SSEA3- cells enriched in the A2B5+ fraction, altogether suggesting an appreciable level of heterogeneity in hESCs cultures.

Any differentiation bias of these distinct populations was next studied by analysing the self-renewal potential of purified single cells in a quantitative hESC-colony initiating cell assay. This demonstrated that A2B5- cells were enriched for clonogenic capabilities compared to A2B5+ cells, while c-KIT+ and c-KIT- cells showed similar clonogenic capability. This was underlined by mRNA analysis of OCT4 and NANOG which showed that while levels remain the same in both c-KIT positive and negative populations, A2B5- cells exhibited 25- to 35-fold higher OCT4 and NANOG transcript levels than A2B5+ cells. Further mRNA analysis of mesodermal (BRA and MIXL1) and neural-associated transcripts (PAX6 and NF-68) showed a 2-fold increase in mesodermal mRNA levels in c-KIT+ cells versus c-KIT- cells and higher levels of neuronal genes in the A2B5+ cells as compared with A2B5- cells. The differentiation bias of these fractions was then investigated through embryoid body (EB)-mediated differentiation under specific conditions. EB generation with c-KIT+ and c-KIT- cells under hematopoietic-inducing conditions showed that c-KIT+ EBs gave higher frequencies of hemogenic precursors, primitive hematopoietic progenitors, and mature hematopoietic cells than the c-KIT- EBs indicating a mesodermal differentiation bias. Similar studies with A2B5+ and A2B5- cells demonstrated that EBs generated from A2B5+ cells could go on to produce extensive neurite-like outgrowths when plated back onto a fibronectin growth substrate, while purified A2B5- cells showed low viability and did not allow for EB formation. For a better comparison of the A2B5 populations, EBs consisting of similar amounts of A2B5+ and A2B5- cells derived from hESC lines expressing green fluorescent protein or red fluorescent protein were formed and showed that under conditions conducive to neural differentiation A2B5+ cells were present in higher numbers and expressed higher levels of neural lineage markers (astrocyte marker GFAP and neuron marker MAP2) than A2B5- cells. This shows that these two markers, found in cells in hESC cultures, are generally indicative of lineage-specific differentiation towards hematopoietic (c-KIT) and neuronal (A2B5) fates.

The molecular basis of this bias was investigated by studying the epigenetic landscape of gene-specific regulatory regions using chromatin immunoprecipitation (ChIP). Using specific antibodies for the permissive histone modification tri-methylated lysine 4 histone H3 (TriK4) and the repressive histone modification tri-methylated lysine 27 histone H3 (TriK27), epigenetic maps were generated for the regulatory regions of both pluripotency and differentiation associated genes. Unfractionated hESCs showed high levels of TriK4 at the OCT4 and NANOG promoter which correlates to the expression of these genes, while genes associated with the mesodermal (BRA, MIXL1, MEOX1, EOMES and TBX6) and neuronal (PAX6, NF-68, MASH1, NESTIN, SOX1) lineage showed a bivalent pattern, that is, the presence of both permissive and repressive histone marks, which is understood to allow for the rapid expression or silencing of gene expression required upon lineage specific differentiation. Further in depth analysis of the specific fractions used a sequential ChIP technique which allows for the demonstration of two specific histone modifications on the same genomic region. Low levels of TriK27 were observed in c-KIT+, c-KIT- and A2B5- fractions at the OCT4 locus which correlated to OCT4 gene expression in hESCs, however low levels of TriK4 were found at the OCT4 locus in A2B5+ cells. An enrichment of TriK4 was observed at mesoderm-associated genes in c-KIT+ hESCs, while TriK27 was found to be enriched at the same genes in c-KIT- cells. Meanwhile, neuronal associated genes in the c-KIT fractions were enriched with marks indicative of gene repression, while neuronal-associated genes in the A2B5 fractionated cells showed a loss of TriK27 in A2B5+ cells, indicating neural gene expression. As around 50% of cultured hESC cells are devoid of c-KIT and A2B5 expression, chromatin studies were also undertaken in these cells which, interestingly, showed enrichment of TriK4 at OCT4 and NANOG, but instead of showing bivalent structure at differentiation associated genes, TriK27 was strongly enriched, suggesting that these cells are epigenetically “locked” into a pluripotent status.

Taken together, these studies suggest that any lineage bias of hESC subfractions marked by c-KIT and A2B5 is encoded by specific chromatin patterns found at lineage-associated and pluripotency genes. However, how these modifications are established are unknown but the identification of transcription factors, co-factors, histone modifying enzymes and the signalling cascades which control them can now begin, using systems such as is demonstrated in this paper. The identification of such markers also may make directed differentiation studies more efficient by eradicating unwanted cell types, making hESC-derived cells for therapeutic use a more attractive proposition.



Hayashi K, Lopes SM, Tang F, Surani MA.
Dynamic equilibrium and heterogeneity of mouse pluripotent stem cells with distinct functional and epigenetic states.
Cell Stem Cell. 2008 Oct 9;3(4):391-401.

Stewart MH, Bossé M, Chadwick K, Menendez P, Bendall SC, Bhatia M.
Clonal isolation of hESCs reveals heterogeneity within the pluripotent stem cell compartment.
Nat Methods. 2006 Oct;3(10):807-15.

Chambers I, Silva J, Colby D, Nichols J, Nijmeijer B, Robertson M, Vrana J, Jones K, Grotewold L, Smith A.
Nanog safeguards pluripotency and mediates germline development.
Nature. 2007 Dec 20;450(7173):1230-4.

Toyooka Y, Shimosato D, Murakami K, Takahashi K, Niwa H.
Identification and characterization of subpopulations in undifferentiated ES cell culture.
Development. 2008 Mar;135(5):909-18.

Cell fate potential of human pluripotent stem cells is encoded by histone modifications.
Hong SH, Rampalli S, Lee JB, McNicol J, Collins T, Draper JS, Bhatia M.
Cell Stem Cell. 2011 Jul 8;9(1):24-36.

Enrichment of neurons and neural precursors from human embryonic stem cells.
Carpenter MK, Inokuma MS, Denham J, Mujtaba T, Chiu CP, Rao MS.
Exp Neurol. 2001 Dec;172(2):383-97.