176
Xenopus
et al. 2009). Chromatin state maps of zebraf sh, mouse, and
human embryos would follow suit (Vastenhouw et al. 2010;
Lindeman et al. 2011; Liu et al. 2016; Xia et al. 2019). The
functional genomic elements associated with particular chromatin states in Xenopus are very similar to those in other
vertebrate species, despite substantial differences in GC content and CpG density between genomes. However, the differences in overall nucleotide composition do affect the relative
frequencies of promoter elements. For example, there is a
more frequent use of the TATA box relative to the GC-rich
Sp1 motif in Xenopus compared to human promoters (van
Heeringen et al. 2011). Nonetheless, similar to human promoters, frog promoters frequently feature a relatively high
density of unmethylated CpG dinucleotides (Bogdanovic et
al. 2011; Long et al. 2013; Hontelez et al. 2015). Also similar
to the promoters in human and other genomes, active promoters in frogs feature accessible chromatin and are enriched for
methylated histone H3 lysine 4 (H3K4me1, H3K4me3) and
acetylated H3K9 (H3K9ac; Tables 11.1, 11.2) (Akkers et al.
2009; van Heeringen et al. 2014; Hontelez et al. 2015; Bright
et al. 2021). Compared with promoters, enhancers typically
show an intermediate level of chromatin accessibility when
active and are marked by H3K4me1 (but not H3K4me3)
and H3K27ac. The H3K27 acetylation mark is deposited by
the p300 (Ep300) coactivator. H3K4 methylation is considered permissive; active regulatory elements usually exhibit
H3K4 methylation, but some inactive or “poised” elements
have these modifcations as well. The H3K27me3 mark
(mutually exclusive with H3K27ac) is involved in repression of transcription by the Polycomb Repression Complex
2 (PRC2). It is often found in broad domains of unmethylated CpG islands at developmentally and spatially regulated
genes. Trimethylation of H3K36 is a proxy for elongating
RNA polymerase II, whereas methylated H3K9 and H4K20
(H3K9me2, H3K9me3, H4K20me3) are modif cations
associated with heterochromatin and repetitive elements
(Hontelez et al. 2015; van Kruijsbergen et al. 2017). The
epigenome maps of both X. tropicalis and X. laevis allow
for studies of genome evolution (Elurbe et al. 2017) and provide a rich resource for both whole genome and gene-centric
analyses of embryonic development.
11.3. CHROMATIN ACCESSIBILITY: OPENING
THE DOOR FOR GENE EXPRESSION
11.3.1. HISTONE VARIANTS AND LINKER
HISTONES MODULATE NUCLEOSOME
DYNAMICS AND ACCESSIBILITY
With the exception of sperm, histones are abundant nuclear
proteins in all cells. In Xenopus oocytes and early embryos,
these proteins are even more abundant. Normally cells contain a mass of core histones that is approximately equal to the
mass of their genomic DNA, but oocytes and early embryos
contain excess histone protein, stored in the cytoplasm. The
abundance of stored histone proteins and histone-encoding
mRNAs in Xenopus embryos made them attractive targets
for early studies of the histones in relation to their incorporation in embryonic chromatin and the mid-blastula transition
(Woodland and Adamson 1977; Woodland 1980; Koster et al.
1988). In agreement with these early studies, quantitative proteomics has established that oocytes and early embryos contain over 1012 molecules of core histone proteins ( Smits et al.
2014). This is more than enough to package newly replicated
DNA into chromatin until well after the onset of embryonic
transcription during blastula stages.
Most of the histone protein is incorporated in chromatin by DNA replication-coupled chromatin assembly
mechanisms. In somatic cells, this happens through S phasespecif c expression of canonical histone genes, tightly coordinated with the need for chromatin assembly during DNA
replication. However, additional non-canonical histone
genes exist. They encode variant histones that serve distinct
functions and that can be incorporated in chromatin independent of replication-coupled chromatin assembly (Martire
and Banaszynski 2020). Variant histones are subject to
dynamic exchange (deposition and eviction), which is mediated by histone chaperones. For example, in the histone H3
family of proteins, histones H3.1/3.2 are incorporated during S phase by the CAF-1 complex. By contrast, H3.3, which
differs at just fve and four amino acids with H3.1 and H3.2,
respectively, is selectively deposited at gene-regulatory
regions by the HIRA histone chaperone complex in a DNA
synthesis-independent manner. This causes nucleosomes to
be more dynamic at regulatory regions. Depletion of histone
H3.3 by morpholinos leads to problems with blastopore closure in Xenopus embryos (Szenker et al. 2012; Sitbon et al.
2020). One H3.3-specifc residue, S31, is critical for rescue
of this phenotype. This serine is phosphorylated, and a phospho-mimetic S31D mutation not only rescues H3.3 depletion
but also increases H3.3 K27 acetylation. This modif cation
is strongly associated with gene activation. This suggests
that the gastrulation phenotype is caused by a requirement
for effcient H3K27 acetylation that depends on histone
exchange and S31 phosphorylation (Sitbon et al. 2020).
H2A.Z (H2az1/H2az2) is one of many variants of histone H2A that is predominantly found at gene-regulatory
regions, where it facilitates access of transcription factors
and other chromatin-associated proteins to DNA (Martire
and Banaszynski 2020). Mutations in Srcap, a chromatin
remodeler mediating the incorporation of H2A.Z into chromatin, cause Floating-Harbor syndrome (FHS). There are
two H2A.Z variants, which differ by just three amino acids.
Knockdown of H2A.Z.2 (H2az2) mimics the FHS craniofacial phenotype in Xenopus (Greenberg et al. 2019). Also, for
this molecule, a single amino acid difference with H2A.Z.1
is critical for H2A.Z.2 to rescue the phenotype. In contrast
to the amino acid substitutions in H3.3, this amino acid difference in H2A.Z.2 may affect the stability of nucleosomes.
The altered H2A.Z-containing nucleosomes appear critical
for a specifc category of AT-rich enhancers that regulate
neural crest gene expression (Greenberg et al. 2019). H2A.Z
specifcally recruits the Pwwp2a protein to DNA (Pünzeler
et al. 2017). Knockdown of this protein in Xenopus also
Xenopus
et al. 2009). Chromatin state maps of zebraf sh, mouse, and
human embryos would follow suit (Vastenhouw et al. 2010;
Lindeman et al. 2011; Liu et al. 2016; Xia et al. 2019). The
functional genomic elements associated with particular chromatin states in Xenopus are very similar to those in other
vertebrate species, despite substantial differences in GC content and CpG density between genomes. However, the differences in overall nucleotide composition do affect the relative
frequencies of promoter elements. For example, there is a
more frequent use of the TATA box relative to the GC-rich
Sp1 motif in Xenopus compared to human promoters (van
Heeringen et al. 2011). Nonetheless, similar to human promoters, frog promoters frequently feature a relatively high
density of unmethylated CpG dinucleotides (Bogdanovic et
al. 2011; Long et al. 2013; Hontelez et al. 2015). Also similar
to the promoters in human and other genomes, active promoters in frogs feature accessible chromatin and are enriched for
methylated histone H3 lysine 4 (H3K4me1, H3K4me3) and
acetylated H3K9 (H3K9ac; Tables 11.1, 11.2) (Akkers et al.
2009; van Heeringen et al. 2014; Hontelez et al. 2015; Bright
et al. 2021). Compared with promoters, enhancers typically
show an intermediate level of chromatin accessibility when
active and are marked by H3K4me1 (but not H3K4me3)
and H3K27ac. The H3K27 acetylation mark is deposited by
the p300 (Ep300) coactivator. H3K4 methylation is considered permissive; active regulatory elements usually exhibit
H3K4 methylation, but some inactive or “poised” elements
have these modifcations as well. The H3K27me3 mark
(mutually exclusive with H3K27ac) is involved in repression of transcription by the Polycomb Repression Complex
2 (PRC2). It is often found in broad domains of unmethylated CpG islands at developmentally and spatially regulated
genes. Trimethylation of H3K36 is a proxy for elongating
RNA polymerase II, whereas methylated H3K9 and H4K20
(H3K9me2, H3K9me3, H4K20me3) are modif cations
associated with heterochromatin and repetitive elements
(Hontelez et al. 2015; van Kruijsbergen et al. 2017). The
epigenome maps of both X. tropicalis and X. laevis allow
for studies of genome evolution (Elurbe et al. 2017) and provide a rich resource for both whole genome and gene-centric
analyses of embryonic development.
11.3. CHROMATIN ACCESSIBILITY: OPENING
THE DOOR FOR GENE EXPRESSION
11.3.1. HISTONE VARIANTS AND LINKER
HISTONES MODULATE NUCLEOSOME
DYNAMICS AND ACCESSIBILITY
With the exception of sperm, histones are abundant nuclear
proteins in all cells. In Xenopus oocytes and early embryos,
these proteins are even more abundant. Normally cells contain a mass of core histones that is approximately equal to the
mass of their genomic DNA, but oocytes and early embryos
contain excess histone protein, stored in the cytoplasm. The
abundance of stored histone proteins and histone-encoding
mRNAs in Xenopus embryos made them attractive targets
for early studies of the histones in relation to their incorporation in embryonic chromatin and the mid-blastula transition
(Woodland and Adamson 1977; Woodland 1980; Koster et al.
1988). In agreement with these early studies, quantitative proteomics has established that oocytes and early embryos contain over 1012 molecules of core histone proteins ( Smits et al.
2014). This is more than enough to package newly replicated
DNA into chromatin until well after the onset of embryonic
transcription during blastula stages.
Most of the histone protein is incorporated in chromatin by DNA replication-coupled chromatin assembly
mechanisms. In somatic cells, this happens through S phasespecif c expression of canonical histone genes, tightly coordinated with the need for chromatin assembly during DNA
replication. However, additional non-canonical histone
genes exist. They encode variant histones that serve distinct
functions and that can be incorporated in chromatin independent of replication-coupled chromatin assembly (Martire
and Banaszynski 2020). Variant histones are subject to
dynamic exchange (deposition and eviction), which is mediated by histone chaperones. For example, in the histone H3
family of proteins, histones H3.1/3.2 are incorporated during S phase by the CAF-1 complex. By contrast, H3.3, which
differs at just fve and four amino acids with H3.1 and H3.2,
respectively, is selectively deposited at gene-regulatory
regions by the HIRA histone chaperone complex in a DNA
synthesis-independent manner. This causes nucleosomes to
be more dynamic at regulatory regions. Depletion of histone
H3.3 by morpholinos leads to problems with blastopore closure in Xenopus embryos (Szenker et al. 2012; Sitbon et al.
2020). One H3.3-specifc residue, S31, is critical for rescue
of this phenotype. This serine is phosphorylated, and a phospho-mimetic S31D mutation not only rescues H3.3 depletion
but also increases H3.3 K27 acetylation. This modif cation
is strongly associated with gene activation. This suggests
that the gastrulation phenotype is caused by a requirement
for effcient H3K27 acetylation that depends on histone
exchange and S31 phosphorylation (Sitbon et al. 2020).
H2A.Z (H2az1/H2az2) is one of many variants of histone H2A that is predominantly found at gene-regulatory
regions, where it facilitates access of transcription factors
and other chromatin-associated proteins to DNA (Martire
and Banaszynski 2020). Mutations in Srcap, a chromatin
remodeler mediating the incorporation of H2A.Z into chromatin, cause Floating-Harbor syndrome (FHS). There are
two H2A.Z variants, which differ by just three amino acids.
Knockdown of H2A.Z.2 (H2az2) mimics the FHS craniofacial phenotype in Xenopus (Greenberg et al. 2019). Also, for
this molecule, a single amino acid difference with H2A.Z.1
is critical for H2A.Z.2 to rescue the phenotype. In contrast
to the amino acid substitutions in H3.3, this amino acid difference in H2A.Z.2 may affect the stability of nucleosomes.
The altered H2A.Z-containing nucleosomes appear critical
for a specifc category of AT-rich enhancers that regulate
neural crest gene expression (Greenberg et al. 2019). H2A.Z
specifcally recruits the Pwwp2a protein to DNA (Pünzeler
et al. 2017). Knockdown of this protein in Xenopus also