Biology Reference
In-Depth Information
undergone by female and male germ cells. Interestingly, oocyte- and sperm-
specific methylation of distinct genomic regions according to their CpG
content and localization with respect to genes is consistent with the different
nature of maternal and paternal ICRs. This divergence may reflect the dif-
ferent DNA methylation-guiding pathways in the two parental germlines.
The acquisition of methylation patterns in gametes depends on the
de novo
methyltransferase Dnmt3A and its cofactor Dnmt3L in both sexes (
Fig. 9.1
A;
Bourc'his & Bestor, 2004; Bourc'his, Xu, Lin, Bollman, & Bestor, 2001;
Kaneda et al., 2004
). It is also believed that active H3K4me2/3 marks could
repel DNA methyltransferases at defined genomic sequences in the two
parental germlines. The strongest evidence for this comes from
in vitro
bio-
chemical studies, which revealed that DNMT3A and DNMT3L bind lysine
4 of histone H3 directly, but that these interactions are antagonized when the
lysine residue is methylated (
Ooi et al., 2007; Otani et al., 2009
). Moreover,
the histone H3K4 demethylase KDM1B (also known as LSD2 or AOF1) is
required for the establishment of DNA methylation for at least four maternal
ICRs in the oocyte (
Ciccone et al., 2009
). In agreement with this, a general
depletion of DNA methylation occurs at loci with H3K4me3 enrichment in
mouse oocytes (
Smallwood et al., 2011
). Conversely, enrichment of
H3K4me3 marks is observed at maternal ICRs during male germ-cell devel-
opment, potentially underlying their exclusion from the paternal
de novo
methylation program (
Henckel, Chebli, Kota, Arnaud, & Feil, 2012
).
Anadditional component that shapes theoocyte and spermmethylomes lies
in transcription-dependent mechanisms. Pioneering genetic studies demon-
strated that transcription traversing the imprinted
Gnas
maternal ICRs is essen-
tial for oocyte methylation establishment at this locus (
Chotalia et al., 2009
).
This requirement for transcription appears tobe a genome-wide phenomenon,
asRNA-seq studies in oocytes linkDNAmethylation in gene units with active
transcription of these units (
Kobayashi et al., 2012
). Interestingly, alternative
oocyte-specific promoters are commonly located upstream of canonical,
somatically used promoters. These downstream promoters frequently become
methylated in the oocyte and not in the sperm(
Smallwoodetal.,2011
), such as
genes encoding members of the DNMT family,
Dnmt1
,
Dnmt3b
,and
Dnmt3L
(
Guenatri, Duffie, Iranzo, Fauque, &Bourc'his, 2013; Smallwood et al., 2011;
Smith et al., 2012
). The mechanism for transcription-associated DNA meth-
ylation in the oocyte has yet to be elucidated. A similar mechanism to gene-
body methylation in somatic cells could be at play. Gene-body methylation
has been proposed to occur through the recruitment of DNA
methyltransferases along the path of the elongating transcript (
Ball et al.,
