Loss of HP1 causes depletion of H3K27me3 from facultative heterochromatin and gain of H3K27me2 at constitutive heterochromatin

Methylated lysine 27 on histone H3 (H3K27me) marks repressed “facultative heterochromatin,” including developmentally regulated genes in plants and animals. The mechanisms responsible for localization of H3K27me are largely unknown, perhaps in part because of the complexity of epigenetic regulatory networks. We used a relatively simple model organism bearing both facultative and constitutive heterochromatin, Neurospora crassa, to explore possible interactions between elements of heterochromatin. In higher eukaryotes, reductions of H3K9me3 and DNA methylation in constitutive heterochromatin have been variously reported to cause redistribution of H3K27me3. In Neurospora, we found that elimination of any member of the DCDC H3K9 methylation complex caused massive changes in the distribution of H3K27me; regions of facultative heterochromatin lost H3K27me3, while regions that are normally marked by H3K9me3 became methylated at H3K27. Elimination of DNA methylation had no obvious effect on the distribution of H3K27me. Elimination of HP1, which “reads” H3K9me3, also caused major changes in the distribution of H3K27me, indicating that HP1 is important for normal localization of facultative heterochromatin. Because loss of HP1 caused redistribution of H3K27me2/3, but not H3K9me3, these normally nonoverlapping marks became superimposed. Indeed, mass spectrometry revealed substantial cohabitation of H3K9me3 and H3K27me2 on H3 molecules from an hpo strain. Loss of H3K27me machinery (e.g., the methyltransferase SET-7) did not impact constitutive heterochromatin but partially rescued the slow growth of the DCDC mutants, suggesting that the poor growth of these mutants is partly attributable to ectopic H3K27me. Altogether, our findings with Neurospora clarify interactions of facultative and constitutive heterochromatin in eukaryotes.It has become increasingly clear that covalent modifications of chromatin, such as methylation of specific histone residues and methylation of DNA, can have profound effects on genome functions. In animals, even partial disruption of DNA methylation leads to developmental defects and disease states (Robertson 2005). Similarly, methylation of histone H3 lysine 27 (H3K27me) by the Polycomb Repressive Complex 2 (PRC2) is critical for normal development in flies, plants, and other systems (Schwartz and Pirrotta 2007), and recent work implicates perturbation of H3K27me in a high fraction of pediatric gliomas (Schwartzentruber et al. 2012; Sturm et al. 2012; Wu et al. 2012; Chan et al. 2013; Lewis et al. 2013). It is of obvious interest to understand the normal regulation of epigenetic features such as methylation of DNA and H3K27, which normally mark constitutive and facultative heterochromatin, respectively. Unfortunately, despite numerous studies in a variety of systems, little is understood about how these chromatin modifications are controlled.Epigenetic marks frequently influence one another, confounding analyses. For example, Schmitges et al. (2011) demonstrated that the amino terminus of histone H3 is recognized by the Nurf55-Suz12 submodule of PRC2 and that this binding is blocked by marks of active chromatin, namely, K4me3 and K36me2/3. Although the mechanism of such “crosstalk” is sometimes reasonably obvious, more often it is not, as illustrated by observations of H3K27me3 redistribution in response to defects in constitutive heterochromatin. More than a decade ago, Peters et al. (2003) noticed that mouse cells defective in both of the SUV39H methyltransferases, which are responsible for the trimethylation of histone H3 lysine 9 (H3K9me3) characteristic of pericentric heterochromatin, show redistribution of H3K27me3; both cytological and molecular analyses suggested that this Polycomb mark “relocated” to the neighborhood abandoned by H3K9me3. DNA methylation typically colocalizes with H3K9me3 but not with H3K27me3 (Rose and Klose 2014). Because DNA methylation can depend on H3K9me, and vice versa (Tariq and Paszkowski 2004), it was of interest to determine whether loss of DNA methylation would also result in redistribution of H3K27me. In early studies with mouse embryonic stem cells, reduced DNA methylation resulting from mutation of either the maintenance methyltransferase gene dnmt1 or the de novo DNA methyltransferase genes dmnt3a and dnmt3b did not result in an obvious change in the distribution of H3K27me (Martens et al. 2005). However, subsequent studies with Arabidopsis (Mathieu et al. 2005; Deleris et al. 2012), mouse embryonic fibroblasts (Lindroth et al. 2008; Reddington et al. 2013), embryonic stem cells (Hagarman et al. 2013), and neural stem cells (Wu et al. 2010) revealed that loss of DNA methylation, caused by disruption of DNA methyltransferase genes or treatment with the demethylating agent 5-azacytidine, provided the most potent trigger of H3K27me3 redistribution.Considering that DNA methylation has been reported to stimulate H3K9 methylation, in both plants (Tariq and Paszkowski 2004) and animals (Jin et al. 2011), and that both of these epigenetic marks are tied to additional nuclear processes, interpretation of these fascinating results is problematic. We took advantage of a relatively simple system to explore possible relationships between marks of constitutive and facultative heterochromatin. Specifically, we used the filamentous fungus Neurospora crassa, which unlike many other simple model eukaryotes (e.g., budding and fission yeasts, Drosophila and Caenorhabditis elegans) has both DNA methylation and H3K27me (Aramayo and Selker 2013; Jamieson et al. 2013). The pathway for formation of constitutive heterochromatin in Neurospora is relatively well understood and essentially unidirectional, as illustrated in Figure 1A. Constitutive heterochromatin, which is primarily in centromere regions, is characterized by AT-rich (GC-poor) DNA resulting from the action of the genome defense system RIP (repeat-induced point mutation) operating on transposable elements (Selker 1990; Aramayo and Selker 2013). DIM-5, in the DIM-5/DIM-7/DIM-9/DDB1/CUL4 complex (DCDC) (Fig. 1B), methylates H3K9 associated with RIP''d DNA (Lewis et al. 2010a,b). Heterochromatin Protein 1 (HP1) specifically binds the resulting H3K9me3 (Freitag et al. 2004) and recruits the DNA methyltransferase DIM-2 (Honda and Selker 2008). Consequently, the genomic distribution of 5mC, HP1, H3K9me3, AT-rich DNA, and repeated sequences correlate almost perfectly (Fig. 1A). Importantly, mutation of dim-2 does not affect the distributions of H3K9me3 and HP1, unlike the situation in plants (Tariq and Paszkowski 2004) and animals (Espada et al. 2004; Gilbert et al. 2007). Similarly, mutation of the gene encoding HP1 (hpo) has almost no effect on the distribution of H3K9me3 (Lewis et al. 2009). Although mutations in genes encoding DCDC proteins render the organism slow-growing and sensitive to certain drugs (Lewis et al. 2010a), neither the components of the DNA methylation/constitutive heterochromatin machinery nor the components of the H3K27me2/3/facultative heterochromatin machinery are essential for viability of the organism, allowing us to test knockouts of genes for components of these processes for possible epigenetic interactions.Open in a separate windowFigure 1.Heterochromatin formation in Neurospora crassa. (A) An ∼500-kb region, including the centromere of LG III, is shown to illustrate the pathway leading to constitutive heterochromatin in Neurospora. During the sexual phase of the life cycle, the genome defense system RIP (repeat-induced point mutation) recognizes repeated DNA (e.g., transposons and other repeated DNA shown as black rectangles) and litters them with C:G-to-T:A mutations (Selker 1990). In vegetative cells, lysine 9 of histone H3 (H3K9) associated with G:C-poor DNA is methylated by DIM-5, generating H3K9me3 (orange track), which is bound by the HP1-DIM-2 complex (yellow track) and catalyzes DNA methylation (green track). Perturbation of any step in the pathway eliminates the downstream steps without significantly influencing earlier steps. (B) Key proteins required to form constitutive (left) or facultative (right) heterochromatin. Methylation of H3K9 by DIM-5 depends on all five members of the DCDC (DIM-5/-7/-9, CUL4/DDB1 complex) (Lewis et al. 2010a). HP1 directly binds to H3K9me3 (cluster of three red hexagons labeled “Me”) and recruits the DNA methyltransferase DIM-2 (Honda and Selker 2008), resulting in a genome-wide correlation between H3K9me3 and DNA methylation (orange hexagons labeled “Me”) (Lewis et al. 2009). HP1 is also involved in the HCHC deacetylase silencing complex (Honda et al. 2012) and the DMM (Honda et al. 2010) complex, which limits spreading of constitutive heterochromatin. Methylation of H3K27 (cluster of three blue hexagons labeled “Me”) is carried out by the PRC2 complex consisting of SET-7, EED, SU(Z)12, and NPF (Jamieson et al. 2013).     

Two new insulator proteins,Pita and ZIPIC,target CP190 to chromatin

Insulators are multiprotein–DNA complexes that regulate the nuclear architecture. The Drosophila CP190 protein is a cofactor for the DNA-binding insulator proteins Su(Hw), CTCF, and BEAF-32. The fact that CP190 has been found at genomic sites devoid of either of the known insulator factors has until now been unexplained. We have identified two DNA-binding zinc-finger proteins, Pita, and a new factor named ZIPIC, that interact with CP190 in vivo and in vitro at specific interaction domains. Genomic binding sites for these proteins are clustered with CP190 as well as with CTCF and BEAF-32. Model binding sites for Pita or ZIPIC demonstrate a partial enhancer-blocking activity and protect gene expression from PRE-mediated silencing. The function of the CTCF-bound MCP insulator sequence requires binding of Pita. These results identify two new insulator proteins and emphasize the unifying function of CP190, which can be recruited by many DNA-binding insulator proteins.Insulators in the Drosophila and vertebrate genomes have been identified based on their ability to disrupt the communication between an enhancer and a promoter when inserted between them (Raab and Kamakaka 2010; Ghirlando et al. 2012; Herold et al. 2012; Matzat and Lei 2013; Chetverina et al. 2014; Kyrchanova and Georgiev 2014). The growing amount of data show that insulator proteins fulfil an architectural function in mediating inter- and intrachromosomal interactions and in contacting regulatory elements such as promoters or enhancers (Maksimenko and Georgiev 2014).The best studied Drosophila insulator proteins, dCTCF (homolog of vertebrate insulator protein CTCF) and Su(Hw) are DNA-binding zinc-finger proteins (Herold et al. 2012; Matzat and Lei 2013; Kyrchanova and Georgiev 2014). Binding sites for dCTCF have been identified in the insulators that separate functional regulatory domains of the bithorax complex and in many promoter regions (Moon et al. 2005; Holohan et al. 2007; Mohan et al. 2007; Nègre et al. 2010, 2011; Ni et al. 2012). The Su(Hw) protein more frequently associates with intergenic sites (Adryan et al. 2007; Bushey et al. 2009; Nègre et al. 2010, 2011; Soshnev et al. 2012, 2013). As shown in a transgenic assay, dCTCF and Su(Hw) binding sites can support specific distant interactions (Kyrchanova et al. 2008a,b), which suggests a key role for these proteins in organizing chromatin architecture.The Su(Hw), dCTCF, and BEAF-32 proteins interact with Centrosomal Protein 190 kD, named CP190 (Pai et al. 2004; Gerasimova et al. 2007; Mohan et al. 2007; Bartkuhn et al. 2009; Oliver et al. 2010; Liang et al. 2014). CP190 (1096 amino acids) contains an N-terminal BTB/POZ domain, an aspartic-acid-rich D-region, four C2H2 zinc-finger motifs, and a C-terminal E-rich domain (Oliver et al. 2010; Ahanger et al. 2013). The BTB domain of CP190 forms stable homodimers that may be involved in protein–protein interactions (Oliver et al. 2010; Bonchuk et al. 2011). In addition to these motifs, CP190 also contains a centrosomal targeting domain (M) responsible for its localization to centrosomes during mitosis (Butcher et al. 2004). It has been shown that CP190 is recruited to chromatin via its interaction with the Su(Hw) and dCTCF proteins (Pai et al. 2004; Mohan et al. 2007). Inactivation of CP190 affects the activity of the dCTCF-dependent insulator Fab-8 from the bithorax complex (Gerasimova et al. 2007; Mohan et al. 2007; Moshkovich et al. 2011) and the gypsy insulator, which contains 12 binding sites for the Su(Hw) protein (Pai et al. 2004). Binding of Su(Hw) and CP190 at gypsy-like sites is mutually dependent, indicating a stabilizing role of CP190 in these cases (Schwartz et al. 2012).Recent genome-wide ChIP-chip studies provide evidence for an extensive overlap of the CP190 distribution pattern with dCTCF, BEAF-32, and Su(Hw) insulator proteins and the promoters of active genes (Bartkuhn et al. 2009; Bushey et al. 2009; Nègre et al. 2010, 2011; Schwartz et al. 2012; Soshnev et al. 2012). Very recently, it has been demonstrated that CP190 bridges DNA-bound insulator factors with promoters (Liang et al. 2014). These data support the model that CP190 has a global role in the function of insulator proteins. However, there are a number of sites in the Drosophila genome where CP190 does not colocalize with any known insulator DNA binding protein (IBP), suggesting that there may be some other proteins that recruit CP190 to chromatin (Schwartz et al. 2012).To identify new factors that associate with CP190, we purified the FLAG-tagged CP190 protein from S2 cells and identified two zinc-finger proteins, CG7928 and Pita, which were shown to interact with CP190 in vivo and in vitro. Genome-wide identification of binding sites for Pita and CG7928 in S2 cells revealed their extensive colocalization with CP190, providing evidence for direct interactions between these proteins, which was supported by binding and in vivo functional assays. Based on these results we termed CG7928 the “zinc-finger protein interacting with CP190” (ZIPIC).     

Spreading of X chromosome inactivation via a hierarchy of defined Polycomb stations

X chromosome inactivation (XCI) achieves dosage balance in mammals by repressing one of two X chromosomes in females. During XCI, the long noncoding Xist RNA and Polycomb proteins spread along the inactive X (Xi) to initiate chromosome-wide silencing. Although inactivation is known to commence at the X-inactivation center (Xic), how it propagates remains unknown. Here, we examine allele-specific binding of Polycomb repressive complex 2 (PRC2) and chromatin composition during XCI and generate a chromosome-wide profile of Xi and Xa (active X) at nucleosome-resolution. Initially, Polycomb proteins are localized to ∼150 strong sites along the X and concentrated predominantly within bivalent domains coinciding with CpG islands (“canonical sites”). As XCI proceeds, ∼4000 noncanonical sites are recruited, most of which are intergenic, nonbivalent, and lack CpG islands. Polycomb sites are depleted of LINE repeats but enriched for SINEs and simple repeats. Noncanonical sites cluster around the ∼150 strong sites, and their H3K27me3 levels reflect a graded concentration originating from strong sites. This suggests that PRC2 and H3K27 methylation spread along a gradient unique to XCI. We propose that XCI is governed by a hierarchy of defined Polycomb stations that spread H3K27 methylation in cis.X chromosome inactivation (XCI) provides an excellent model by which to study Polycomb regulation and the role of long noncoding RNAs (lncRNAs) in inducing facultative heterochromatin (Lyon 1999; Wutz and Gribnau 2007; Payer and Lee 2008; Lee 2011). XCI is controlled by the X-inactivation center (Xic), an X-linked region that controls the counting of X chromosomes, the mutually exclusive choice of Xa and Xi, and the recruitment and propagation of silencing complexes. The 17-kb Xist RNA initiates the silencing step as it accumulates on the X (Brockdorff et al. 1992; Brown et al. 1992; Clemson et al. 1996). Although recent studies have shown that Xist RNA directly recruits Polycomb repressive complex 2 (PRC2) to the Xi (Zhao et al. 2008) and that loading of the Xist-PRC2 complex occurs first at a YY1-bound nucleation center located within the Xic (Jeon and Lee 2011), how the silencing complexes spread throughout the X after this obligatory nucleation step remains a major unsolved problem.Because autosomes with ectopic Xic sequences are subject to long-range silencing (Wutz and Gribnau 2007; Payer and Lee 2008), it is thought that spreading elements cannot be unique to the X. One hypothesis suggests that repetitive elements of the LINE1 class facilitate spreading (Lyon 2000). However, this hypothesis has been difficult to test, as linking repeats to locus-specific function has been complicated by their repetitive nature. Some studies have provided correlative evidence (Bailey et al. 2000; Wang et al. 2006; Chow et al. 2010), whereas others find that species lacking active LINE1s nonetheless possess XCI (Cantrell et al. 2009). Other classes of repeats may be more enriched on the X (Chow et al. 2005). Matrix-associated proteins, such as HNRNPU (also known as SAF-A), have also been proposed to facilitate spreading (Helbig and Fackelmayer 2003; Hasegawa et al. 2010; Pullirsch et al. 2010), but a direct link has also not been demonstrated.In general, the identification of spreading elements has been thwarted by the lack of high-throughput approaches that distinguish Xi and Xa at sufficient resolution. Epigenomic studies have primarily focused on male cells (Bernstein et al. 2006; Boyer et al. 2006; Barski et al. 2007; Mikkelsen et al. 2007; Ku et al. 2008), though one recent ChIP-seq analysis with partial allele-specific coverage used female mouse embryonic stem (ES) cells but without addressing PRC2 binding. The reported 1.2-fold enrichment of H3K27me3 on Xi (Marks et al. 2009) is unexpectedly low and at odds with intense cytological H3K27me3 immunostaining (Plath et al. 2003; Silva et al. 2003)—likely caused by low-density polymorphisms between Xi and Xa. As a result, the quest for an Xi chromatin state map and spreading elements has remained unrealized.In principle, silencing complexes could initially load at the Xic and spread serially from nucleosome to nucleosome. Alternatively, they could spread outwardly via “way stations” located at defined sites along the X that would anchor and relay silencing complexes (Gartler and Riggs 1983). To test these models, we herein devise an allele-specific ChIP-seq strategy that enables the generation of chromosome-wide developmental profiles at unprecedented allelic resolution. We report a high-density Xi chromatin state map and identification of discrete Polycomb stations.     

Abundant and dynamically expressed miRNAs,piRNAs, and other small RNAs in the vertebrate Xenopus tropicalis

Small regulatory RNAs have recently emerged as key regulators of eukaryotic gene expression. Here we used high-throughput sequencing to determine small RNA populations in the germline and soma of the African clawed frog Xenopus tropicalis. We identified a number of miRNAs that were expressed in the female germline. miRNA expression profiling revealed that miR-202-5p is an oocyte-enriched miRNA. We identified two novel miRNAs that were expressed in the soma. In addition, we sequenced large numbers of Piwi-associated RNAs (piRNAs) and other endogenous small RNAs, likely representing endogenous siRNAs (endo-siRNAs). Of these, only piRNAs were restricted to the germline, suggesting that endo-siRNAs are an abundant class of small RNAs in the vertebrate soma. In the germline, both endogenous small RNAs and piRNAs mapped to many high copy number loci. Furthermore, endogenous small RNAs mapped to the same specific subsets of repetitive elements in both the soma and the germline, suggesting that these RNAs might act to silence repetitive elements in both compartments. Data presented here suggest a conserved role for miRNAs in the vertebrate germline. Furthermore, this study provides a basis for the functional analysis of small regulatory RNAs in an important vertebrate model system.Short RNAs have recently emerged as abundant regulators of gene expression in many eukaryotes, including plants, animals, and fungi (Sharp 2009). The lin-4 and let-7 miRNAs were the first type of endogenous short regulatory RNAs to be identified in eukaryotes (Lee et al. 1993; Reinhart et al. 2000); since then many functional small RNAs have been identified in organisms as diverse as roundworms, flies, fish, frogs, mammals, flowering plants, mosses, anemones, sponges, and even viruses, using genetics, molecular cloning, and predictions from bioinformatics (Lagos-Quintana et al. 2001; Lau et al. 2001; Lee and Ambros 2001; Llave et al. 2002; Reinhart et al. 2002; Lim et al. 2003; Pfeffer et al. 2004; Arazi et al. 2005; Axtell and Bartel 2005; Watanabe et al. 2005; Grimson et al. 2008). In cells, miRNAs are tightly bound by proteins of the Ago clade of the Argonaute superfamily of RNA-binding proteins (Cerutti et al. 2000). miRNAs are thought to inhibit efficient translation of target mRNAs or to control mRNA decay.Another class of small RNAs, 21–24 nucleotides (nt) endogenous siRNAs, was first discovered in plants in response to viral infection (Hamilton and Baulcombe 1999; Llave et al. 2002). These RNAs are thought to represent endogenous instances of short interfering RNAs (siRNAs), the mediators of RNAi (Fire et al. 1998; Tuschl et al. 1999; Zamore et al. 2000). More recently, endo-siRNAs have also been identified in Caenorhabditis elegans (Ambros and Lee 2004), Drosophila (Czech et al. 2008; Ghildiyal et al. 2008; Kawamura et al. 2008; Okamura et al. 2008), and mouse oocytes (Tam et al. 2008; Watanabe et al. 2008). endo-siRNAs are enriched in the germline of animals and map to various genomic loci including repetitive elements, pseudogenes, palindromes, and regions where both strands are transcribed. Like miRNAs, endo-siRNAs interact with Argonaute proteins. endo-siRNAs likely have roles in silencing of transposable elements or pseudogenes (Okamura et al. 2008).A third class of 25–30 nt RNAs has been identified in Drosophila, zebrafish, mice, rats, anemones, and sponges and has been named Piwi-associated RNAs or piRNAs (Grimson et al. 2008; Klattenhoff and Theurkauf 2008). By definition piRNAs interact with proteins of the Piwi clade of the Argonaute superfamily. piRNA populations are complex; there are hundreds of thousands of unique piRNAs in mammals. Piwi and piRNAs are required for transposon silencing: for example, in Drosophila the piRNAs of the flamenco locus control the gypsy retrotransposon (Desset et al. 2003; Brennecke et al. 2007). The piRNAs of C. elegans are unique in that they are 21 nt short RNAs with distinct genomic organization and biogenesis, but a conserved role in transposon silencing (Ruby et al. 2006; Batista et al. 2008; Das et al. 2008; Wang and Reinke 2008).Previously, small RNAs have also been grouped together based on their genomic location as repeat-associated small RNAs (rasiRNAs) in plants, fungi, Drosophila, and zebrafish (Llave et al. 2002; Reinhart et al. 2002; Aravin et al. 2003; Chen et al. 2005b). These can now be reclassified as endo-siRNAs or piRNAs based on their size, biogenesis, and associated Argonaute superfamily proteins (Okamura et al. 2008; Malone and Hannon 2009). Although endo-siRNA and piRNA pathways are distinct, in animals, endo-siRNAs and piRNAs are 2′O-methylated at the 3′ end (Horwich et al. 2007; Tam et al. 2008; Watanabe et al. 2008). While the function of this modification remains unclear in animals, in plants, 2′O-methylation stabilizes miRNAs and endo-siRNAs (Yang et al. 2006).Xenopus laevis has been used widely as a model system for the study of oocyte development and maturation, including the regulation of gene expression at the level of translation and RNA localization. Xenopus oogenesis is subdivided into six stages (I–VI) based on features such as diameter, pigmentation color, and the amount of yolk in the cytoplasm. Stage VI oocytes are arrested in first meiotic prophase, and can be matured into eggs, arrested in MII metaphase, by progesterone. While previous work in Xenopus has identified a number of miRNAs through cloning and comparative genomic approaches, little is know about small RNAs population during Xenopus oogenesis. Microarrays, Northern blotting, and in situ hybridization have been used to determine miRNA expression during embryogenesis and adult frog (Watanabe et al. 2005; Hikosaka et al. 2007; Michalak and Malone 2008; Tang and Maxwell 2008; Walker and Harland 2008). However, recent advances in sequencing technology have allowed the more complete assessment of small RNA species in animals, plants, and fungi.Here we applied Illumina sequencing (formerly known as Solexa sequencing) to determine the expression of small RNAs in the Xenopus tropicalis germline and somatic tissues. This work represents the first example of small RNA high-throughput sequencing in an amphibian. Using this approach we identify abundant populations of miRNAs, piRNAs, and other small RNAs in the germline and soma of X. tropicalis. We hope that these data might set the stage for the biochemical analysis of small RNA pathways in a powerful model system, the Xenopus oocyte.