proteins, known to be important in development, repress many developmental regulators in mammalian embryonic stem (ES) cells until these genes are ready to be turned on during differentiation, according to two
in this week?s Nature
. A third
study, published in Cell
, shows that a characteristic pattern of histone methylation
in ES cells also represses genes involved in cell differentiation. Understanding epigenetic differences between stem cells and differentiated cells may help researchers exert more control over ES cell differentiation, the authors suggest.
The repressive mechanisms
discovered in ES cells differ from those seen in somatic cells, because repression must be reversible for cells to remain pluripotent, said Vincenzo Pirrotta
of Rutgers University in Piscataway, NJ, who was not involved in the studies. ?Somehow these genes, although repressed, are poised to be expressed again,? Pirrotta told The Scientist
. ?Given the right signal, they can take off.?
Polycomb group (PcG) proteins, which silence transcription by modifying chromatin
structure, repress homeotic
genes in Drosophila
until they need to be turned on. PcG proteins have been implicated
in preserving pluripotency in ES cells, but few of their target genes in mammals are known, Rudolf Jaenisch
of Whitehead Institute for Biomedical Research in Cambridge, Mass., a co-author on all three papers, told The Scientist
Led by Laurie Boyer, also of Whitehead Institute, researchers identified genes in mouse ES cells occupied by the two major PcG protein complexes: Polycomb Repressive Complex 1 (PRC1) and Polycomb Repressive Complex 2 (PRC2). Most of these genes were also enriched for an epigenetic mark of repressed chromatin: tri-methylation of lysine 27 on histone H3, a modification that seems
to be necessary for Polycomb gene silencing.
Most of the genes repressed by Polycomb complexes and lysine 27 methylation were transcription factors involved in developmental processes like organogenesis, pattern specification, cell-fate determination, and differentiation. Transcripts of the genes targeted by Polycomb were present at very low levels in ES cells but became expressed in differentiated cells. For example, neural-specific genes that were repressed by PRCs and lysine 27 methylation in ES cells lost these marks and become expressed in neural precursors. Genes not needed in neural precursors, however, remained repressed.
In a Cell
paper that included many of the same researchers, Ton Ihn Lee of Whitehead and colleagues reported similar findings in human ES cells. These researchers mapped sites of the human genome occupied by a key subunit of PRC2. They found that, as in mice, PRC2 and histone H3 methylation of lysine 27 were associated with repression of a wide variety of developmental regulators, including the majority of homeodomain genes and large subsets of gene families involved in axial patterning, tissue development, cell-fate specification, and lineage differentiation.
In another Cell
paper, researchers led by Bradley Bernstein
of Harvard University found a previously undescribed epigenetic pattern that represses some of the same developmental genes. In this study, the researchers mapped histone methylation across highly conserved regions of the mouse genome in ES cells. They focused on two marks: the lysine 27 methylation that represses gene expression and lysine 4 methylation of the same histone, which activates gene expression. These modifications were thought to occupy exclusive genomic regions, Jaenisch said, but Bernstein and colleagues found large stretches of lysine 27 methylation with smaller regions of lysine 4 methylation embedded inside them. Because of the opposing actions of the two types of modifications, the researchers named these regions ?bivalent domains.?
These bivalent domains overlapped with many repressed developmental transcription factor genes -- similar to what?s seen with lysine 27 methylation only. But by having an activating signal there as well, ?these genes in this particular chromatin configuration are poised for activation when differentiation is induced,? Jaenisch said. The bivalent-domain pattern disappeared in differentiated cells, with genes retaining lysine 4 methylation if they become active or lysine 27 methylation if they remain repressed, but not both.
?The idea that there?s poised chromatin in these developmentally important areas of the genome is pretty cool,? said Michael Atchison
of the University of Pennsylvania, who was not involved in the studies. This bivalent pattern could underlie stem cells? unique ability to differentiate into any adult cell type, Atchison added.
All three papers also found an association between marks of repressed chromatin and three transcription factors -- OCT4, SOX2, and NANOG -- that are essential for ES cells to remain pluripotent. These transcription factors often activate gene expression, Jaenisch said, but they are also associated with repressed genes when a Polycomb complex is present, too. It?s not clear exactly what this means, Pirrotta told The Scientist
, ?but it?s suggesting that these factors are in some way contributing to the recruitment of the Polycomb complexes.? It?s possible that these transcription factors may keep developmental genes ready for activation when a signal comes along that turns off Polycomb repression, Atchison said.
The identity of such differentiation signals and the mechanisms they use to tell cells to differentiate are still largely unknown, said Pirrotta. ?How exactly that works will have to be worked out.?
Melissa Lee Phillips
Links within this article
B. Maher, ?Polycomb getting its due at Keystone,? The Scientist
, January 23, 2006.
L.A. Boyer et al., ?Polycomb complexes repress developmental regulators in murine embryonic stem cells,? Nature
, published online April 20, 2006.
T.I. Lee et al., ?Control of developmental regulators by Polycomb in human embryonic stem cells,? Cell
, April 21, 2006.
B.A. Maher, ?Histone methylation is making its mark,? The Scientist
, January 27, 2003.
L.A. Pray, ?Epigenetics: Genome, meet your environment,? The Scientist
, July 5, 2004.
B.E. Bernstein et al., ?A Bivalent chromatin structure marks key developmental genes in embryonic stem cells,? Cell
, April 21, 2006.
B.A. Cunningham, ?Unraveling chromatin's secrets,? The Scientist
, January 7, 2002.
V. Orlando, R. Paro, ?Chromatin multiprotein complexes involved in the maintenance of transcription patterns,? Current Opinion in Genetics and Development
, April 1995.
D. O'Carroll et al., ?The polycomb-group gene Ezh2 is required for early mouse development,? Molecular and Cellular Biology
, July 2001.
R. Cao et al., ?Role of histone H3 lysine 27 methylation in Polycomb-group silencing,? Science
, November 1, 2002.