2018), suggesting proximity to the chromosome
end is sufficient to drive assembly of a facultative heterochromatin domain. In fact, telomere
repeats (TTAGGG) themselves can induce
H3K27me2/3 placement, as ectopically placing
8 or 17 telomere repeats drives formation of
novel facultative heterochromatin domains
(Jamieson et al. 2018). However, positionindependent domains of facultative heterochromatin exist as well; these domains are typically internal on the chromosome and are
maintained upon chromosome translocation.
No underlying signal for H3K27me2/3 has
been identified in position-independent
domains, as ectopic placement of underlying
DNA from one such domain failed to induce
H3K27me2/3 (Jamieson et al. 2018).
An interesting dynamic exists between
H3K27me2/3-marked facultative heterochromatin and constitutive heterochromatin
marked by H3K9me3: the placement of
H3K9me3 at some genomic loci actually
represses the establishment of facultative heterochromatin (Fig. 1.2c). Deletion or mutation
of components of DCDC or HP1 results in relocalization of the H3K27me2/3 mark to former
sites of constitutive heterochromatin (Basenko
et al. 2015; Jamieson et al. 2016). The reverse
relationship does not hold true: loss of PRC2
function does not cause H3K9me3 movement
(Basenko et al. 2015; Jamieson et al. 2016).
The inhibition of PRC2 components actually
rescues two phenotypes of constitutive heterochromatin mutants: defective growth rates and
sensitivity to DNA-damaging agents (Basenko
et al. 2015; Jamieson et al. 2016). Interestingly,
the movement of facultative heterochromatin to
formerly
constitutive
heterochromatin
domains is also observed in a Dhpo strain, in
which the H3K9me3 binding HP1 is removed,
but H3K9me3 is still catalyzed on histone tails.
Mass spectrometry of Dhpo histones showed
tails with substantial levels of both H3K9me3
and the dimethylated posttranslational modification of H3K27 (Jamieson et al. 2016). Thus,
the loss of the canonical H3K9me3-binding
protein HP1 must signal, directly or indirectly,
for PRC2 recruitment and specific dimethylation on lysine 27.
It is currently unknown how H3K27me2/3marked heterochromatin repression is established and maintained, given the lack of PRC1
homologs in the Neurospora genome—clearly
H3K27me2/3 must be bound by and interpreted
as a repressive mark. All four members of the
Neurospora PRC2Polycomb Repressive Complex 2 (PRC2) complex are required for wildtype levels of H3K27me2/3, but deletion of npf/
cac-3 causes loss of only the most subtelomeric
H3K27me2/3, whereas facultative heterochromatin domains located internally on the chromosome are maintained (Jamieson et al. 2013;
Klocko et al. 2016). Other fungal species encode
a PRC2 complex that itself, through the EED
protein, allosterically binds the H3K27me2/3
mark for positive induction of catalytic activity
(Jiao and Liu 2015), thereby providing a convenient mechanism for propagating the dimethylation or trimethylation to nearby histones.
Moreover, one of the most common mechanisms to bind a methylated histone tail is
mediated by chromodomains, such as that
found in HP1 for H3K9me3.
Within the chromodomain, three aromatic residues—
most often tyrosines and tryptophans—form an aromatic cage enclosing the methylated, and now nonpolar, lysine side chain; further amino acid determinants
of the chromodomain interact with the surrounding
residues within the N-terminal histone tail to provide
specificity to that specific histone residue (Nielsen et al.
2002).
Currently,
it
is
unclear
whether
chromodomain-containing proteins have a
role in binding H3K27me2/3. Further, the first
H3K27me2/3 reader protein, Early Bolting in
Short Day (EBS), which contains a bromoadjacent homology (BAH) domain that also
forms an aromatic cage for binding
H3K27me2/3, was recently characterized in
Arabidopsis (Yang et al. 2018), thus providing
a new research direction with a different
domain. Neurospora appears to have a weak
homolog of EBS, but it is unclear what role, if
any, this homolog may play in H3K27me2/3
recognition. Lastly, many eukaryotic model
systems rely on noncoding RNAs for the establishment of H3K27me2/3, including on the
1 Chromatin Structure and Function in Neurospora crassa
17
end is sufficient to drive assembly of a facultative heterochromatin domain. In fact, telomere
repeats (TTAGGG) themselves can induce
H3K27me2/3 placement, as ectopically placing
8 or 17 telomere repeats drives formation of
novel facultative heterochromatin domains
(Jamieson et al. 2018). However, positionindependent domains of facultative heterochromatin exist as well; these domains are typically internal on the chromosome and are
maintained upon chromosome translocation.
No underlying signal for H3K27me2/3 has
been identified in position-independent
domains, as ectopic placement of underlying
DNA from one such domain failed to induce
H3K27me2/3 (Jamieson et al. 2018).
An interesting dynamic exists between
H3K27me2/3-marked facultative heterochromatin and constitutive heterochromatin
marked by H3K9me3: the placement of
H3K9me3 at some genomic loci actually
represses the establishment of facultative heterochromatin (Fig. 1.2c). Deletion or mutation
of components of DCDC or HP1 results in relocalization of the H3K27me2/3 mark to former
sites of constitutive heterochromatin (Basenko
et al. 2015; Jamieson et al. 2016). The reverse
relationship does not hold true: loss of PRC2
function does not cause H3K9me3 movement
(Basenko et al. 2015; Jamieson et al. 2016).
The inhibition of PRC2 components actually
rescues two phenotypes of constitutive heterochromatin mutants: defective growth rates and
sensitivity to DNA-damaging agents (Basenko
et al. 2015; Jamieson et al. 2016). Interestingly,
the movement of facultative heterochromatin to
formerly
constitutive
heterochromatin
domains is also observed in a Dhpo strain, in
which the H3K9me3 binding HP1 is removed,
but H3K9me3 is still catalyzed on histone tails.
Mass spectrometry of Dhpo histones showed
tails with substantial levels of both H3K9me3
and the dimethylated posttranslational modification of H3K27 (Jamieson et al. 2016). Thus,
the loss of the canonical H3K9me3-binding
protein HP1 must signal, directly or indirectly,
for PRC2 recruitment and specific dimethylation on lysine 27.
It is currently unknown how H3K27me2/3marked heterochromatin repression is established and maintained, given the lack of PRC1
homologs in the Neurospora genome—clearly
H3K27me2/3 must be bound by and interpreted
as a repressive mark. All four members of the
Neurospora PRC2Polycomb Repressive Complex 2 (PRC2) complex are required for wildtype levels of H3K27me2/3, but deletion of npf/
cac-3 causes loss of only the most subtelomeric
H3K27me2/3, whereas facultative heterochromatin domains located internally on the chromosome are maintained (Jamieson et al. 2013;
Klocko et al. 2016). Other fungal species encode
a PRC2 complex that itself, through the EED
protein, allosterically binds the H3K27me2/3
mark for positive induction of catalytic activity
(Jiao and Liu 2015), thereby providing a convenient mechanism for propagating the dimethylation or trimethylation to nearby histones.
Moreover, one of the most common mechanisms to bind a methylated histone tail is
mediated by chromodomains, such as that
found in HP1 for H3K9me3.
Within the chromodomain, three aromatic residues—
most often tyrosines and tryptophans—form an aromatic cage enclosing the methylated, and now nonpolar, lysine side chain; further amino acid determinants
of the chromodomain interact with the surrounding
residues within the N-terminal histone tail to provide
specificity to that specific histone residue (Nielsen et al.
2002).
Currently,
it
is
unclear
whether
chromodomain-containing proteins have a
role in binding H3K27me2/3. Further, the first
H3K27me2/3 reader protein, Early Bolting in
Short Day (EBS), which contains a bromoadjacent homology (BAH) domain that also
forms an aromatic cage for binding
H3K27me2/3, was recently characterized in
Arabidopsis (Yang et al. 2018), thus providing
a new research direction with a different
domain. Neurospora appears to have a weak
homolog of EBS, but it is unclear what role, if
any, this homolog may play in H3K27me2/3
recognition. Lastly, many eukaryotic model
systems rely on noncoding RNAs for the establishment of H3K27me2/3, including on the
1 Chromatin Structure and Function in Neurospora crassa
17