Possessing two X chromosomes can significantly reduce DNA methylation in mouse embryonic stem cells, potentially explaining why such cell lines have proven unstable in culture, scientists report online in
These findings suggest an embryo's sex could have subtle but significant consequences on "diseases where epigenetics plays a difference," Brockdorff said. Uncovering the human homologs of the mouse genes responsible for this reduced methylation could illuminate differences between the species in X inactivation, he added.
In mammals, methylation typically occurs on cytosine residues of CpG dinucleotides. In mice, phosphorimager analysis showed roughly 70 percent of CpG dinucleotides were methylated in two different XY embryonic stem cell lines and in XY and XX somatic cells. However, in two mouse XX embryonic stem cell lines, less than 35 percent of CpG sites were methylated, according to the new report.
CpG sites are found in high concentrations in repetitive sequences. Southern blot analysis of XX embryonic stem cells showed reduced methylation levels in major satellite repeat sequences that were fully methylated in XY embryonic stem cells, embryonic stem cells possessing just an X chromosome, and differentiated XX cells where one X chromosome is normally inactivated. This suggests this reduced methylation is due to the presence of two active X chromosomes and not the absence of a Y chromosome, the authors note.
Focusing on imprinted genes, which are methylated differently depending on which parent they came from, Brockdorff and his colleagues found the
Frequently, when scientists try establishing XX embryonic stem cell lines, an X chromosome is lost. During serial passaging of six embryonic stem cell lines derived from unfertilized eggs, Southern blot analysis suggested losing an X chromosome restored methylation of
Western blot analysis of Dnmt3a and Dnmt3b, enzymes linked to de novo methylation, revealed significantly lower Dnmt3a levels in XX than in XY embryonic stem cells. The researchers speculate the X chromosome encodes a repressor of de novo methyltransferases such as Dnmt3a and Dnmt3b, and that cells with two active X chromosomes overexpress this gene. "Finding out what this gene is could shed light on regulation through DNA methylation," Brockdorff said.
Transfecting XX embryonic stem cells with either Dnmt3a or Dnmt3b expression constructs mostly restored global methylation levels, although not at KvDMR1 and
While more mouse XX embryonic stem cell lines should be studied to confirm these findings, "this may explain why there have been difficulty in isolating embryonic stem cell lines from different species," Mahendra Rao at the National Institute on Aging in Bethesda, Md., who did not participate in this study, told
It is "intriguing" that obtaining human XX embryonic stem cell lines has not been that difficult. This possibly suggests X-inactivation happens early in human cells or is regulated differently than from mice, Rao noted. "It's proven very hard to study X chromosome inactivation," so these findings are helpful in that light, Rao said.
The only other model system to look at reduced global methylation "is cancer," Rao said. "What they have provided here is another handle onto global methylation, which will be very useful in understanding this fundamental biological regulation process." Future research can see if X chromosome-related trisomy such as XXY also reduces methylation, he added.