Polycomb

Protein Interactions

The proximal Polyhomeotic gene product has 193 amino-terminal amino acids that are absent from distal Ph; in addition, it makes use of internal initiation to give an alternate product that is shorter by 244 amino acids. A notable feature of this unique proximal domain is the presence of a PxxPxxPxxP motif (amino acids 156 to 165) with proline spacing the same as that of the polyproline type II helix recognized by the SH3 domain. Ph also has many glutamine repeats and a serine/threonine-rich region. Near the carboxyl terminus are two blocks of sequence (amino acids 1297 to 1388 and 1511 to 1576) that are shared with the mammalian Ph homologs. The first sequence, named H1, consists of 28 highly conserved amino acids followed by an unusual C4 zinc finger with intercysteine spacing Cx2C...Cx3C. The second sequence has been variously referred to as H2 or SEP (for the mouse homolog); as SPM (for the PcG protein Sex combs on midleg as well as for the human Ph homologs HPH1 and HPH2), and as SAM (for a variety of yeast signal transduction proteins). This domain can mediate homotypic and heterotypic self-association between Ph and Scm proteins in vitro. In view of this result, the domain is referred to as a self-association motif (SAM), but the specific subset of SAMs with greatest similarity to Ph and Scm is called SPM. The only internal region of sequence dissimilarity between the proximal and distal Ph are the 52 amino acids immediately preceding the SPM domain (Kyba, 1998).

Polyhomeotic and Posterior sexcombs are shown to coprecipitate with Polycomb from nuclear extracts. The domains required for the association of Psc with Ph and Pc were analyzed by using the yeast two-hybrid system and an in vitro protein-binding assay. Psc and Ph interact through regions of sequence conservation with mammalian homologs, i.e., the H1 domain of Ph (amino acids 1297 to 1418) and the helix-turn-helix-containing region of Psc (amino acids 336 to 473). Psc contacts Pc primarily at this region of Psc, and secondarily at the ring finger (amino acids 250 to 335). The Pc chromobox is not required for this interaction. There is also a discussion of the implication of these results for the nature of the complexes formed by Polycomb group proteins (Kyba, 1998).

The Polycomb group (PcG) genes are required for maintenance of homeotic gene repression during development. Mutations in these genes can be suppressed by mutations in genes of the SWI/SNF family. A complex, termed PRC1 (Polycomb repressive complex 1), has been purified that contains the products of the PcG genes Polycomb, Posterior sex combs, polyhomeotic, Sex combs on midleg, and several other proteins. Preincubation of PRC1 with nucleosomal arrays blocks the ability of these arrays to be remodeled by SWI/SNF. Addition of PRC1 to arrays at the same time as SWI/SNF does not block remodeling. Thus, PRC1 and SWI/SNF might compete with each other for the nucleosomal template. Several different types of repressive complexes, including deacetylases, interact with histone tails. In contrast, PRC1 is active on nucleosomal arrays formed with tailless histones (Shao, 1999).

It is apparent from the composition of PRC1 that there must be other PcG complexes in addition to PRC1. PRC1 purified via either tagged PH or PSC contains Pc, Psc, Ph-p, Ph-d, and Scm, as well as several other proteins. PRC1 does not contain Pcl and E(z). Previous studies using immunoprecipitation, in vitro binding, and/or yeast two-hybrid analysis have shown that Pc, Psc, and Ph interact with each other, and that Scm interacts with Ph. E(z) and Esc have been shown to interact with each other by similar approaches, and E(z) separates from PRC1 during chromatography. Similarly, mammalian homologs of PcG can also be separated into roughly two complexes, one containing homologs to Pc, Psc, and Ph, and the other containing homologs to E(z) and Esc. Another argument that E(z) and Esc form a separate complex with a distinct function is based on the observation that homologs to these genes are found in the C. elegans genome, whereas homologs to Pc, Ph, or Psc are not. The activities of PRC1 suggest that it may be directly involved in creating the repressed state, and that it may require other complexes for targeting. Through screens for homeotic derepression, 14 PcG genes have been well characterized genetically. It is possible that a subset of these genes are required for direct repression, while other PcG proteins function in targeting, regulation of repression activity, or maintenance of the repressed state through mitosis. How PcG proteins are recruited to their targets is still unknown, but several proteins have been suggested as candidates for this function, such as Esc and E(z), and sequence-specific DNA-binding proteins Pho, Trithorax-like, Hunchback (Hb), and the Hb interacting protein dMi-2. PRC1 does not contain E(z) and Trithorax-like. Using antibodies against a region of human YY1 that is conserved in Pho, it was found that Pho is unlikely to be in PRC1; an antibody made specifically against Pho is needed to verify this result. Due to the lack of antibodies, whether PRC1 contains Esc, Hb, or dMi-2 was not tested (Shao, 1999 and references).

Carboxy-terminal truncation of PC protein do not affect chromosomal binding. However, mutations affecting only the chromo domain abolish chromosomal binding (Messmer, 1992).

Functionally relevant domains of the PC protein have been identified by sequencing different Pc alleles. Additionally, it was found that a particular mutant protein cannot bind to four particular target loci, but otherwise does not change the remaining overall binding pattern. In contrast to the dotted subnuclear localization of the wild-type protein, the nuclear distribution of mutant proteins becomes homogeneous. Surprisingly, in Pc mutants the Polyhomeotic protein, another member of the Pc-G, is also redistributed in the nucleus. (Franke, 1995).

Studies of deletion mutants of trithorax and Polycomb demonstrate that Trithorax-dependent activation requires the central zinc-binding domain, while Polycomb-dependent repression requires the intact chromodomain. In addition, trithorax-dependent activity can be abrogated by increasing the amount of Polycomb, suggesting a competitive interaction between the products of trithorax and Polycomb (Chang, 1995).

The HP1 chromo domain, like the Polycomb chromo domain, evinces chromosome binding activity, but only to distinct chromosomal sites. A chimeric HP1-Polycomb protein was constructed, consisting of the chromo domain of Polycomb in the context of HP1. It binds to both heterochromatin and Polycomb binding sites in polytene chromosomes. In flies expressing chimeric protein, endogenous HP1 is mislocalized to Polycomb binding sites, and endogenous polycomb is misdirected to the heterochromatic chromocenter, suggesting that both proteins are recruited to their distinct chromosomal binding sites through protein-protein contacts. Chimeric HP1-Polycomb protein promotes heterochromatin-mediated gene silencing, supporting the view that the chromo domain homology reflects a common mechanistic basis for homeotic and heterochromatic silencing (Platero, 1995).

The ability of a chimeric HP1-Polycomb (PC) protein to bind both to heterochromatin and to euchromatic sites of PC protein binding was exploited to detect stable protein-protein interactions in vivo. Endogenous PC protein is recruited to ectopic heterochromatic binding sites by the chimeric protein. Posterior sex combs (PSC) protein also is recruited to heterochromatin by the chimeric protein, demonstrating that PSC protein participates in direct protein-protein interaction with PC protein or PC-associated proteins. In flies carrying temperature-sensitive alleles of Enhancer of zeste[E(z)] the general decondensation of polytene chromosomes that occurs at the restrictive temperature is associated with loss of binding of endogenous PC and chimeric HP1-Polycomb protein to euchromatin, but binding of HP1 and chimeric HP1-Polycomb protein to the heterochromatin is maintained. The E(z) mutation also results in the loss of chimera-dependent binding to heterochromatin by endogenous PC and PSC proteins at the restrictive temperature, suggesting that interaction of these proteins is mediated by E(Z) protein. A myc-tagged full-length Suppressor 2 of zeste [SU(Z)2] protein interacts poorly or not at all with ectopic Pc-G complexes, but a truncated SU(Z)2 protein is strongly recruited to all sites of chimeric protein binding. Trithorax protein is not recruited to the heterochromatin by the chimeric HP1-Polycomb protein, suggesting either that this protein does not interact directly with Pc-G complexes or that such interactions are regulated. Ectopic binding of chimeric chromosomal proteins provides a useful tool for distinguishing specific protein-protein interactions from specific protein-DNA interactions important for complex assembly in vivo (Platero, 1996).

The distribution of the PC protein has been mapped in the homeotic bithorax complex (BX-C) of Drosophila tissue culture cells. The PC protein quantitatively covers large regulatory regions of repressed BX-C genes. Conversely, the Abdominal-B gene is active in these cells; the region is devoid of any bound PC protein (Orlando, 1993). This is the first biochemical evidence found that the PC protein covers large chromosomal domains encompassing inactive genes of the BX-C.

Two members of the Pc-G, Polycomb and Polyhomeotic, are constituents of a soluble multimeric protein complex. Size fractionation indicates that a large portion of the two proteins are found in a distinct complex of molecular weight 2-5 x 10(6) Da. During embryogenesis the two proteins show the same spatial distribution. In addition, using double-immunofluorescence labeling, Polycomb and polyhomeotic show exactly the same binding patterns on polytene chromosomes of larval salivary glands. It is proposed that some Pc-G proteins act in multimeric complexes to compact the chromatin of stably repressed genes like the homeotic regulators (Franke, 1992).

Antibodies raised against the product of Polycomblike were used to show that the Polycomblike protein is found in all nuclei during embryonic development. Antibody staining also reveals that Polycomblike protein is found on larval salivary gland polytene chromosomes at about 100 specific loci, the same loci at which are found Polycomb and Polyhomeotic proteins. These data add further support for a model in which Polycomb group proteins form multimeric protein complexes at specific chromosomal loci to repress transcription at those loci (Lonie, 1994).

Polycomb directly competes with a transcription factor for binding at the chromatin level. The functional role of the FAB fragment, a PRE (Pc-G Response element) of the bithorax complex has been tested. Transgenic fly lines were constructed containing a PRE adjacent to a reporter gene inducible by the yeast GAL4 trans-activator. Polycomb and GAL4 have mutually exclusive binding patterns, supporting the notion that Pc-G-induced chromatin structures can prevent activators from binding to their target sequences. The antagonistic function can be overcome with high doses of GAL4, even in the absence of DNA replication (Zink, 1995).

The products of Psc and Su(z)2 were immunohistochemically detected at 80-90 sites on polytene chromosomes. The chromosomal binding sites of these two proteins were compared with those of Zeste protein and two other Polycomb group proteins, Polycomb and Polyhomeotic. The five proteins co-localize at a large number of sites, suggesting that they frequently act together on target genes (Rastelli, 1993).

Trithorax proteins are bound to 63 specific sites on the polytene chromosomes of the larval salivary gland. Trithorax binding is detected at the sites of its known targets, the Bithorax and Antennapedia complexes, despite the transcriptionally repressed state of these loci in the salivary gland. A temperature-sensitive trithorax mutation greatly reduces the number of binding sites. Simultaneous localization of Trithorax and Polycomb indicates that many of their chromosomal binding sites coincide. One Trithorax binding site is located within a portion of the large 5' regulatory region of the Ubx gene, to an interval which also contains binding sites for Polycomb group proteins. Co-localization of TRX with Polycomb suggests that interactions between activators and repressors may be a significant feature of their mode of action (Chinwalla, 1995).

Recruitment of components of Polycomb Group chromatin complexes in Drosophila

Polycomb Group complexes assemble at polycomb response elements (PREs) in vivo and silence genes in the surrounding chromatin. To study the recruitment of silencing complexes, various Polycomb Group (PcG) proteins have been targeted by fusing them to the LexA DNA binding domain. When LexA-Pc, -Psc, -Ph or -Su(z)2 are targeted to a reporter gene, they recruit functional PcG-silencing complexes that recapitulate the silencing behavior of a PRE: silencing is sensitive to the state of activity of the target chromatin. When the target is transcriptionally active, silencing is not established but when the target is not active at syncytial blastoderm, it becomes silenced. The repressed state persists through embryonic development but cannot be maintained in larval imaginal discs even when the LexA-PcG fusion is constitutively expressed, suggesting a discontinuity in the mechanism of repression. These proteins also interact with other PC-containing complexes in embryonic nuclear extracts. In contrast LexA-Pho is neither able to silence nor to interact with Pc-containing complexes. Analysis of pho mutant embryos and of PRE constructs whose Pho-binding sites are mutated suggests that, while Pho is important for silencing in imaginal discs, it is not necessary for embryonic PcG silencing (Poux, 2001).

These results show that several PcG proteins, targeted by fusion to a DNA-binding domain, can recruit a repressive PcG complex. Several new conclusions can be drawn from these experiments. (1) The recruitment of the silencing complex cannot occur before blastoderm. The alpha1-tubulin-LexA-Pc construct reveals that even when the protein is deposited in the egg during oogenesis, as well as being zygotically expressed, it does not block the initiation of transcription from the Ubx promoter. These results suggest that the LexA-PcG protein cannot establish repression at this early stage, just as the endogenous PcG proteins known to be present in the normal pre-blastoderm embryo do not prevent the initiation of Ubx transcription. Thus, PcG silencing directed by a PRE or by LexA-Pc appears only to set in after blastoderm. One possible explanation is that the assembly of a functional PcG complex at the PRE or at the LexA-binding sites is a multistep process that requires time and is not accomplished until after blastoderm. Another interesting possibility is that the state of the chromatin in nuclei whose very rapid nuclear divisions are just beginning to slow down, cannot yet support the establishment of PcG silencing, for example, because the nucleosomes still bear the deposition-associated histone acetylation pattern. A similar argument might explain why centric heterochromatin is not detectable until blastoderm (Poux, 2001).

(2) The repression established by the LexA-PcG protein is not unconditional but it is sensitive to the state of activity of the target. Like a genuine PRE complex, the LexA-PcG protein establishes silencing only in cells in which the reporter gene is inactive, thus discriminating between active and inactive chromatin targets. The later-acting Ubx H1 enhancer can still function but only in the progeny of cells that were active at early times. The fact that, like the endogenous PRE, the action of the LexA-PcG protein distinguishes between active and silent chromatin, indicates that the discrimination occurs after the binding of the first PcG protein. The sensitive step could be the assembly of a sufficient nucleus of PcG proteins or still later, the involvement of other factors that effect the silencing. It has been shown that Pc-containing complex purified from embryonic nuclear extracts can prevent chromatin remodelling in vitro if it is bound to chromatin before the addition of purified SWI/SNF complex but not if it is added simultaneously or afterwards. If the activation of the Ubx-lacZ reporter gene by the enhancers involves the recruitment of the Drosophila SWI/SNF complex, this observation could help to understand how the LexA binding sites function as a genuine synthetic PRE possessing at least one aspect of the cellular memory displayed by endogenous PREs (Poux, 2001).

(3) Nevertheless, the LexA-binding sites do not constitute a fully functional PRE. Silencing by LexA-PcG proteins is much less effective during larval development. It is possible that the activity of the Ubx H1 enhancer in imaginal discs is more difficult to repress, e.g. because of very high activator concentration. More likely, once the H1 enhancer is active, it is much more difficult to repress by LexA-Pc induced at later times. It cannot be explained, however, why neither daily heat shocks, nor constitutive expression of LexA-Pc can maintain a continuity between embryonic silencing and larval silencing. Repeated heat shocks do eventually reduce the level of expression of the reporter in imaginal discs but the memory of the early domains of repression is lost. The apparent discontinuity in PcG silencing between the embryo and the larva, suggests that there might be a real mechanistic difference between the two states. There are also differences in the maintenance properties of embryos and larvae. One possible explanation is that additional proteins, recruited at a true PRE but not by the LexA fusion proteins, might be necessary for continuous repression at postembryonic developmental stages (Poux, 2001).

PcG proteins Psc, Su(z)2 and Ph are as effective as Pc in recruiting a silencing complex, implying that any one of the 'core' PcG proteins can reconstitute a silencing complex. In contrast, LexA-Gaga and -Pho do not silence the reporter gene. That LexA-Gaga does not silence is hardly surprising. Many promoters, including the hsp70 promoter, the alpha1-tubulin promoter or the Ubx promoter itself, contain Gaga-binding sites but are not thereby targets for PcG silencing in vivo. The possibility that the LexA-Gaga fusion protein is defective in some respect cannot be excluded, although it is able to bind to the normal endogenous sites on polytene chromosomes and participate in PcG complexes. Most likely, however, Gaga/Trithorax-like factor by itself cannot recruit PcG complexes. It is supposed that, like many other nuclear factors, Gaga can have either a stimulating or a repressing activity, depending on the binding of other proteins. In the chromatin context of the PRE, however, Gaga interacts with PcG proteins to form a stable complex (Poux, 2001).

The Pho protein binds to DNA sequences contained in the bxd PRE and has been proposed as a major recruiter of PcG complexes. Mutating the Pho binding sites in the bxd PRE causes loss of silencing in imaginal discs. In the experiments carried out in this study, however, LexA-Pho is unable to silence the reporter gene. Since LexA-Pho efficiently rescues pho mutants in vivo, it is concluded that LexA-Pho is fully functional and that Pho cannot, by itself, recruit PcG complexes able to silence the reporter gene either in the embryo or in imaginal discs (Poux, 2001).

The ability of LexA fusion proteins to assemble PcG complexes that can be targeted to LexA-binding sites allow for the exploration in vitro of the composition of the complexes. It is important to note that what is seen in the immunoprecipitation experiments is the bulk of PcG complexes that contain the LexA-Pc protein. It is not possible to distinguish between complexes recruited by the LexA-Pc protein at the LexA-binding site and complexes formed at other PREs, to which the LexA-Pc protein has been recruited. These experiments therefore do not necessarily reflect the nature of the complex formed at the reporter gene, which is limited to what components can be directly or indirectly recruited by the LexA fusion protein. Nevertheless, the in vitro binding experiments reveal the ability of the LexA fusion protein to participate in complexes containing other PcG proteins (Poux, 2001).

The binding experiments show that the four 'core' proteins Pc, Psc, Su(z)2 and Ph are recruited to PcG complexes and are themselves able to recruit repressive complexes. It cannot be shown that all four proteins are simultaneously associated in one complex but it is very likely that at least Pc, Psc and Ph can participate in the same complex. Similar experiments have also shown that Gaga factor is associated with at least some PcG complexes, although it is not itself able to recruit them to the reporter gene. In contrast, Pho neither recruits any of the core PcG proteins nor participates in the majority of PcG complexes present in embryonic extracts. Conceivably, when the LexA-binding domain is bound to DNA, it might hinder the access of other PcG proteins to the Pho moiety. It is thought that this is unlikely because even wild-type Pho does not seem to be associated with PcG proteins. A PRE fragment containing three Pho binding sites but no GAGA sites does not bind PC-containing complexes in vitro, although it interacts readily with endogenous Pho or LexA-Pho in the same embryonic extracts, indicating that also endogenous Pho is not stably associated with PcG complexes. Similarly, Pho is not detected in a purified PRC1 complex (Poux, 2001).

Long-range repression by multiple Polycomb Group (PcG) proteins targeted by fusion to a defined DNA-binding domain in Drosophila

A tethering assay was developed to study the effects of Polycomb group (PcG) proteins on gene expression in vivo. This system employed the Su(Hw) DNA-binding domain (ZnF) to direct PcG proteins to transposons that carried the white and yellow reporter genes. These reporters constituted naive sensors of PcG effects, since bona fide PcG response elements (PREs) were absent from the constructs. To assess the effects of different genomic environments, reporter transposons integrated at nearly 40 chromosomal sites were analyzed. Three PcG fusion proteins, ZnF-PC, ZnF-SCM, and ZnF-ESC, were studied, since biochemical analyses place these PcG proteins in distinct complexes. Tethered ZnF-PcG proteins repress white and yellow expression at the majority of sites tested, with each fusion protein displaying a characteristic degree of silencing. Repression by ZnF-PC is stronger than ZnF-SCM, which is stronger than ZnF-ESC, as judged by the percentage of insertion lines affected and the magnitude of the conferred repression. ZnF-PcG repression is more effective at centric and telomeric reporter insertion sites, as compared to euchromatic sites. ZnF-PcG proteins tethered as far as 3.0 kb away from the target promoter produce silencing, indicating that these effects are long range. Repression by ZnF-SCM requires a protein interaction domain, the SPM domain, which suggests that this domain is not primarily used to direct SCM to chromosomal loci. This targeting system is useful for studying protein domains and mechanisms involved in PcG repression in vivo (Roseman, 2001).

Biochemical studies indicate that PcG repression involves multiple, distinct PcG complexes. Thus, an underlying assumption of the assay system that was used is that gene silencing by the tethered ZnF-PcG protein involves assembly with endogenous PcG proteins at the reporter site. This hypothesis leads to the prediction that repression by a tethered ZnF-PcG protein should be compromised by loss of function for an endogenous PcG partner. It was difficult to test this prediction for the comprehensive set of endogenous PcG proteins because the basic assay system involved generating a very complex genotype. Nevertheless, the requirement was tested for endogenous PH protein, which is encoded by an X-linked gene and for which a hemizygous viable allele is available (Roseman, 2001).

A reporter integration site was identified that normally lacks PH binding, as scored on polytene chromosomes: it was repressed by all three ZnF-PcG proteins. Genetic tests show that reduction in PH dosage relieves tether-based repression by PC and SCM at this site. These results can be reconciled with the known PC-PH and SCM-PH molecular interactions. Surprisingly, ZnF-ESC repression is sensitive to PH dosage. This result was not expected since ESC-PH interactions have not been reported and there is evidence that ESC and PH are in separate complexes in embryos (Roseman, 2001).

A Drosophila Polycomb group complex includes Zeste and dTAFII proteins

A goal of modern biology is to identify the physical interactions that define 'functional modules' of proteins that govern biological processes. One essential regulatory process is the maintenance of master regulatory genes, such as homeotic genes, in an appropriate 'on' or 'off' state for the lifetime of an organism. The Polycomb group (PcG) of genes maintain a repressed transcriptional state, and PcG proteins form large multiprotein complexes, but these complexes have not been described owing to inherent difficulties in purification. A major PcG complex, PRC1, has been purified to 20%-50% homogeneity from Drosophila embryos. Thirty proteins have been identified in these preparations, then the preparation has been further fractionated and Western analyses have been used to validate unanticipated connections. The known PcG proteins Polycomb, Posterior sex combs, Polyhomeotic and dRING1 exist in robust association with the sequence-specific DNA-binding factor Zeste and with numerous TBP (TATA-binding-protein)-associated factors that are components of general transcription factor TFIID (dTAFIIs). Thus, in fly embryos, there is a direct physical connection between proteins that bind to specific regulatory sequences, PcG proteins, and proteins of the general transcription machinery (Saurin, 2001).

The inheritance of established expression patterns of certain genes during multiple cell divisions is essential for the correct development of an animal. In Drosophila, the expression patterns of the homeotic genes that govern body segment identity are established early in embryogenesis by the products of the gap and pair-rule genes, but are maintained throughout the rest of development by proteins of the PcG and trithorax group (trxG). The trxG maintains the transcriptionally active state of the homeotic genes, whereas the PcG prevents ectopic expression by maintaining a repressive state. The PcG genes encode components of multiple complexes. One of these complexes, Polycomb repressive complex 1 (PRC1), contains the Polycomb (PC), Polyhomeotic (PH) and Posterior sex combs (PSC) proteins (Saurin, 2001).

To better understand the mechanisms of this cellular memory system, an epitope-tag strategy was used to purify PRC1 over 3,000-fold from Drosophila embryos. This complex has been extensively washed in 1 M salt and has a high specific activity in functional analyses; however, contaminating proteins remain associated. Extensive efforts to fractionate this complex to homogeneity in reasonable quantity were blocked by unacceptably low yields on a wide variety of subsequent purification steps. The advent of genome-wide sequence analysis provided an alternative route to identify the components of PRC1. Using mass spectrometry and the recently completed Drosophila genome, almost all of the proteins were identified in the highly fractionated material derived from the M2-affinity column. The sensitivity of Western analysis was used to validate of the association of proteins during subsequent chromatography steps (Saurin, 2001).

The presence of the previously identified PcG proteins PH, PC, and PSC was confirmed by mass spectrometry. In addition, a Drosophila homolog of the mammalian RING1 protein (dRING1) was identified; dRING1 has been found to colocalize with PC on polytene chromosome preparations and its mammalian counterparts have previously been shown to associate with mammalian PcG proteins. Thus, the PcG complement of PRC1 is made up from PH, PSC, PC, dRING1 and sub-stoichiometric amounts of Sex Combs on Midleg (SCM) (Saurin, 2001).

Of the remaining proteins identified by mass spectrometry, the presence of several dTAFII proteins and Zeste is particularly striking. Zeste is a sequence-specific DNA-binding factor, with binding sites in the promoter and regulatory regions of some homeotic genes. The dTAFII proteins were initially identified in the general transcription factor TFIID, a central component for transcriptional initiation, but are also found in histone acetyltransferase complexes. The dTAFII proteins identified by mass spectrometry and Zeste all appear approximately stoichiometric with the PcG complement in the M2 fraction, and maintain a quantitative association with PcG proteins on gel filtration and heparin agarose (Saurin, 2001).

PRC1 fractionates as an extremely large complex, and contains several other proteins in addition to the PcG, dTAFII and Zeste proteins described above. The sequence information provides speculative information on the identity of these proteins, but further work is needed to validate each of these associations. The constitutively expressed Heat shock cognate 3 and 4 (HSC3 and HSC4) proteins were found. The requirement for HSCs in PcG action during development has been demonstrated genetically in flies, where a mutant allele of HSC4 enhances the homeotic phenotype of PC-heterozygous flies. Proteins were found that have been linked to histone deacetylase complexes, including HDAC (RPD3), dMi-2, dSin3A, p55 and SMRTER, a functional homolog of the human SMRT/N-CoR corepressors. While dMi-2 has been linked genetically to PcG repression, and HDAC and p55 have been found present in the Esc/E(z) PcG complex, further studies are clearly needed to examine their association with PRC1. These proteins are present in low stoichiometry: cofractionation of these proteins with PcG on subsequent steps could not be accurately assessed owing to lack of signal, and PRC1 has low deacetylase activity when acetylated core histones and histone peptides are used as substrate (Saurin, 2001).

The most surprising connection revealed in this study is that between PcG proteins and several dTAFIIs. TAFII proteins have previously been found in TFIID and in histone acetyltransferase complexes, and in both contexts have been linked to transcriptional activation. This study suggests that they may also function in PRC1-mediated PcG repression. PcG complexes are targeted to specific genes by sequences called Polycomb response elements (PREs), but are also known to associate at promoters. The presence of dTAFII proteins in PRC1 provides a direct physical connection between PcG proteins and components of the general transcription machinery that bind at promoters. In addition, several of these dTAFIIs have similarities with core histone proteins (dTAFIIs 62, 42 and 30beta) and have been biochemically and structurally demonstrated to associate with each other in a histone octamer-like substructure. Although quite speculative, the structural similarities between the dTAFII42/62 heterotetramer with the histone H3/H4 heterotetramer might indicate a direct role in interacting with nucleosomes and/or DNA to help maintain a stable association of PcG proteins across the numerous rapid cell divisions of the embryo (Saurin, 2001).

The presence of Zeste in PRC1 may serve to assist in the targeting of PcG proteins to repressed loci. Indeed, Zeste can be found localized with PcG proteins at some PcG-repressed loci and recent data demonstrate that Zeste is directly involved in the maintenance of the repressed state of some of these loci. Zeste binds to both PRE and promoter sequences, and thus may serve to bridge the connection of the PcG proteins to these elements. Zeste has also been shown to interact directly with the BRM complex of the trxG. Zeste thus appears be involved in both PcG function and trxG function, consistent with previous genetic studies implying a role in activation and repression (Saurin, 2001).

Reconstitution of a functional core Polycomb repressive complex

The opposing actions of polycomb (PcG) and trithorax group (trxG) gene products maintain essential gene expression patterns during Drosophila development. PcG proteins are thought to establish repressive chromatin structures, but the mechanisms by which this occurs are not known. Polycomb repressive complex 1 (PRC1) contains PcG proteins Polyhomeotic (Ph), Polycomb (Pc), Posterior sexcombs (Psc) and Sex combs extra/Ring and inhibits chromatin remodeling by trxG-related SWI/SNF complexes. A functional core of PRC1 has been defined by reconstituting a stable complex using four recombinant PcG proteins. One subunit, Psc, can also inhibit chromatin remodeling on its own. These PcG proteins create a chromatin structure that has normal nucleosome organization and is accessible to nucleases but excludes hSWI/SNF (Francis, 2001).

To assemble an active recombinant complex of PcG proteins that could be purified in large amounts and used for detailed functional analyses, Sf9 cells were coinfected with Flag-Ph, Pc, Psc, and Ring. Flag-Ph and associated proteins were purified by affinity chromatography. Pc, Psc, and Ring were purified quantitatively with Flag-Ph. If Sf9 cells are coinfected with Psc, Ph, and Pc, with either Psc or Ph carrying the Flag epitope, the three proteins copurify but the eluted fractions contain a significant excess of the epitope-tagged subunit; this suggests Ring is important for stable complex formation. Purification of Flag-Ph from cells expressing Flag-Ph, Ring, and Pc did not result in purification of a complex, but mainly of Flag-Ph. Thus, Pc, Ph, Psc, and Ring form the most stable complex, which will be referred to as PCC (PRC1 core complex) for simplicity. PCC fractions invariably contain a fifth, 70 kDa protein; this protein may be an HSC70 homolog from Sf9 cells since it is immunoreactive with antibodies to HSP70. HSC70 may interact specifically with PcG proteins since HSC proteins copurify with PcG proteins in PRC1 and mutations in HSC4 enhance Pc phenotypes; alternatively, it may associate with the complex simply because the PcG proteins are overexpressed (Francis, 2001).

To determine whether PCC shares the ability of PRC1 to block chromatin remodeling by hSWI/SNF, PCC was tested in the plasmid supercoiling assay originally used to define PRC1 activity. When incubated with a nucleosomal plasmid array and topoisomerase I, hSWI/SNF induces an ATP-dependent decrease in negative supercoiling, which can be identified on agarose gels as a change in the distribution of plasmid topoisomers. PRC1 blocks the ability of hSWI/SNF to alter plasmid topology when preincubated with the template, but not when added simultaneously with hSWI/SNF. Preparations of PCC inhibit remodeling in this assay in a preincubation-dependent manner that mimics the activity of PRC1 (Francis, 2001).

Each component of PCC was purified to determine whether any individual protein shared functional characteristics with PCC or PRC1. One subunit, Psc, also inhibits chromatin remodeling in the plasmid supercoiling assay in a manner that is strictly dependent on preincubation. Preparations of Pc, Ph, and Ring were less active than Psc in this assay. It is concluded that PCC and at least one of its subunits share the ability of PRC1 to inhibit chromatin remodeling (Francis, 2001).

To compare the relative efficiency of PRC1, PCC, and Psc, a quantifiable assay for inhibition was developed based on restriction enzyme accessibility. Assembly of DNA into chromatin blocks the ability of restriction enzymes to digest DNA at nucleosomal sites, but this decrease in accessibility can be counteracted by ATP-dependent chromatin remodeling factors such as hSWI/SNF. Nucleosomal arrays were assembled by the use of templates consisting of two sets of five 5S nucleosome positioning sequences that flank DNA sufficient to assemble two nucleosomes, one of which is predicted to overlap a unique HhaI site. The extent of digestion of the HhaI site in a chromatin template exposed to hSWI/SNF was used to quantify inhibition of chromatin remodeling by PcG proteins (Francis, 2001).

PRC1 blocks the ATP-dependent stimulation of restriction enzyme digestion by hSWI/SNF when preincubated with the template at a concentration of approximately 1 nM. Under the same conditions, approximately 2-4 nM PCC also inhibits chromatin remodeling, suggesting the core complex is up to 50% as efficient as PRC1. Psc inhibits chromatin remodeling at higher concentrations than PCC. Thus, both Psc and PCC inhibit chromatin remodeling and are nearly as efficient as PRC1 (Francis, 2001).

PcG preparations could inhibit remodeling through interactions with the nucleosomal array, by directly inhibiting hSWI/SNF function, or by a combination of these mechanisms. PRC1 does not efficiently inhibit remodeling of mononucleosomes. This suggestes that the complex might require a nucleosomal array substrate. Because large amounts of concentrated PCC and Psc could be purified, tests were performed to see whether these proteins inhibit remodeling of mononucleosomes over a wide range of protein concentrations in excess of nucleosomes. PCC and Psc were titrated into a restriction enzyme accessibility assay for hSWI/SNF remodeling of mononucleosomes assembled on a 155 bp DNA fragment. PCC and Psc have no effect on remodeling of mononucleosomes at concentrations greater than those required to completely inhibit remodeling of arrays, and Psc inhibits remodeling only when present at greater than 20-fold excess over template. These results suggest that PCC or Psc inhibit chromatin remodeling by interacting with the substrate and that these proteins have specific substrate requirements not present in mononucleosomes, such as linker DNA or multiple contiguous nucleosomes. These experiments also confirm that concentrations of PCC and Psc that inhibit remodeling of nucleosomal arrays do not directly inhibit hSWI/SNF activity (Francis, 2001).

These experiments demonstrate that the stable PcG complex PCC and the single subunit Psc share the ability of PRC1 to inhibit remodeling. The definition of this minimal system and the ability to obtain large quantities of highly purified PCC and Psc allows a detailed characterization of the ability of PcG proteins to inhibit remodeling. The ability of PCC and Psc to bind DNA and to interact with nucleosomal arrays to prevent remodeling was examined (Francis, 2001).

Although one mammalian Psc homolog, MEL-18, has been demonstrated to bind DNA, this has not been shown to be the case with the PcG proteins present in PCC. The ability of PCC and Psc to inhibit remodeling on arrays but not on mononucleosomes has suggested that they might recognize free DNA. PCC and Psc were tested for DNA binding activity using filter binding with a 155 bp probe; both PCC and Psc bind DNA with high affinity. The measured KDs of PCC and Psc for DNA were similar. Filter binding assays were used to determine what fraction of molecules in each of the protein preparations were active for DNA binding. A high fraction of the molecules in PCC and Psc preparations were active for DNA binding (PCC 30%-60%; Psc 20%-50%), assuming binding as monomers (Francis, 2001).

It was of interest to determine whether PcG proteins generally block access of all enzymes to chromatin or, more specifically, prevent chromatin remodeling. Considerable attention has been given to the ability of remodeling factors to increase accessibility of DNA sites, such as is measured in the restriction enzyme accessibility assay. However, nucleosome movement stimulated by remodelers can also cause previously exposed sites to become occluded. A SacI site in the 5S array template is predicted to reside between nucleosomes; consistent with this prediction, the SacI site is accessible in 40%-60% of the nucleosomal arrays in the absence of hSWI/SNF. If the template is remodeled with hSWI/SNF, the accessibility of this site is dramatically reduced. The SacI site therefore allows the dissociation of effects of PCC and Psc on chromatin remodeling from direct effects on the restriction enzyme or on template accessibility, since in this case inhibition of remodeling would be expected to increase SacI digestion after incubation with hSWI/SNF. When the template was remodeled by SWI/SNF for 30 min and subsequently digested with SacI for 5 min, a clear decrease in restriction enzyme accessibility was observed, consistent with repositioning of nucleosomes over this site on some templates. This hSWI/SNF-dependent decrease in accessibility is blocked by preincubation of the template with PCC or Psc so that digestion levels were similar to those in the absence of hSWI/SNF (40%-60% digestion). It is concluded that Psc and PCC block ATP-dependent remodeling without blocking access of restriction enzymes to the template (Francis, 2001).

To examine whether inhibition of remodeling by hSWI/SNF is accompanied by changes in nucleosome organization, the effect of PRC1, PCC, and Psc on MNase digestion of nucleosomal arrays was examined. MNase preferentially digests the linker DNA between nucleosomes. On arrays of nucleosomes that are positioned by 5S sequences, digestion with MNase produces a distinct ladder, which can be visualized by probing Southern blots of the digested array with the 5S sequence. Incubation of the array with hSWI/SNF and ATP causes a smearing of the banding pattern, reflecting randomization of nucleosome positioning. PRC1, PCC, and Psc block disruption of the MNase ladder by hSWI/SNF. Significantly, none of these PcG proteins or complexes alter the MNase digestion pattern of the array, implying that MNase is still able to access the array and that nucleosome organization is not grossly altered by the presence of these PcG proteins. These results suggest that complexing these PcG proteins with the template prevents hSWI/SNF-mediated nucleosome movement but not nuclease access to the template (Francis, 2001).

In these MNase experiments, it was possible that PCC or Psc altered nucleosome positions but nonetheless maintained a regular organization of nucleosomes on the template. To determine whether nucleosome positions were altered, indirect end labeling was carried out after digestion with MNase or DNaseI and nucleosome position was examined in the presence or absence of PCC or Psc. Neither Psc nor PCC altered nucleosome position as assayed by either MNase or DNaseI digestion, both of which yield nucleosomal ladders due to the tight positioning of the nucleosomes on the 5S array. Similar results were obtained in MNase experiments with PRC1. Interestingly, however, arrays incubated with PCC or Psc were approximately 5- to 10-fold less sensitive to digestion by MNase, but about 10-fold more sensitive to digestion by DNase I. Although as yet there is no explanation for this observation, it is consistent with PCC or Psc altering the structure of the chromatin template and perhaps the orientation of the linker DNA (Francis, 2001).

Psc and PCC could prevent remodeling by preventing hSWI/SNF from binding to the template, by interfering with the remodeling of chromatin carried out by bound hSWI/SNF, or by a combination of both mechanisms. To determine whether binding of Psc or PCC to the template interferes with the binding of hSWI/SNF, in vitro chromatin immunoprecipitations (ChIPs) were carried out using antisera to BRG1, the major ATPase subunit and remodeling protein of hSWI/SNF. The amount of DNA precipitated by anti-BRG1 was compared on templates incubated with or without Psc or PCC and these results were correlated with inhibition of remodeling in the same reactions. When arrays were preincubated with concentrations of Psc or PCC that inhibit chromatin remodeling, the amount of BRG1 bound to the template was decreased by at least 10-fold. Titrations of Psc and PCC indicate that exclusion of BRG1 from the template correlates with inhibition of remodeling. Thus, these experiments suggest that Psc or PCC can exclude hSWI/SNF from a nucleosomal array, and are consistent with exclusion accounting, at least in part, for inhibition of remodeling (Francis, 2001).

Psc is active as an isolated polypeptide, raising the possibility that this protein is an important component of PCC activity and could function in the absence of the other PCC subunits. Indeed, recent evidence suggests Psc is an essential component of the silencing mechanism and may function independently of the other PCC subunits in certain circumstances. (1) Loss of Psc and its homolog suppressor of zeste 2 [Su(Z)2] or Ph, but not of other PcG proteins, has been shown to result in rapid upregulation of homeotic genes. Repression can be restored by resupply of Psc [and Su(Z)2] provided only a few cell generations has passed. These results are consistent with an essential role for Psc in the actual silencing mechanism. (2) Immunocytochemistry in Drosophila embryos demonstrates that the majority of Ph, Pc, and Psc dissociate from chromosomes during mitosis; Psc reassociates with the chromosomes before Pc and Ph, suggesting Psc may function independently of these subunits in vivo during reestablishment of repression (Francis, 2001).

In vivo, PcG proteins are targeted to appropriate genes by PREs, but PCC does not require PRE sequences to prevent chromatin remodeling. Preliminary results with templates that include PRE sequences from the Ubx gene suggest that PREs may not target PCC and Psc in vitro. This raises the possibility that the basic mechanisms by which PcG proteins influence chromatin can be mechanistically separated from those which target repression to appropriate genes. It seems likely that PcG proteins interact, perhaps directly, with DNA binding factors that target them to PREs. The characterization of PCC activity provides a system in which targeting of PcG proteins can be analyzed in vitro (Francis, 2001).

Essential role of Drosophila Hdac1 in homeotic gene silencing

Deacetylation of the N-terminal tails of core histones plays a crucial role in gene silencing. Rpd3 and Hda1 represent two major types of genes encoding trichostatin A-sensitive histone deacetylases. Drosophila Rpd3, referred to here by its alternative name HDAC1, interacts cooperatively with Polycomb group repressors in silencing the homeotic genes that are essential for axial patterning of body segments. The biochemical copurification and cytological colocalization of HDAC1 and Polycomb group repressors strongly suggest that HDAC1 is a component of the silencing complex for chromatin modification on specific regulatory regions of homeotic genes (Chang, 2001).

In Drosophila, five potential genes encoding TSA-sensitive HDACs have been identified. Of these five, Rpd3 (Hdac1) and Hdac3 are Rpd3 types; Hdac2 and Hdac4 are Hda1 types, and CG10899 appears to diverge significantly from both types. Because the expressions of homeotic genes are oppositely controlled by Pc-G and trx-G proteins in a dosage-sensitive manner, it is possible to assess the roles of these HDACs in homeotic gene regulation by examining the genetic interactions between Pc-G or trx-G mutations and HDAC mutations. In a preliminary study, deficiencies that delete four of five potential HDAC genes (a deficiency for Hdac3 is not available currently) have been examined for genetic interactions with Pc. Only Df(3R)10H, which deletes Hdac1, shows a significant genetic interaction with a Pc mutation, resulting in a more than 2-fold increase in ectopic sex comb teeth on the second and third legs of male adults. In addition, this deficiency substantially reduces the frequency of mesothoracic transformation typically found in trx-G mutants. These results are consistent with a negative role for Hdac1 in homeotic gene regulation (Chang, 2001).

Df(3R)10H deletes not only Hdac1, but several other genes as well. To show that deletion of the Hdac1 gene is responsible for the genetic interactions with Pc, genetic interactions between Pc-G mutants and Hdac1^P-UTR, a semilethal mutant with a P element inserted at +47 of Hdac1 were examined. Hdac1^P-UTR also shows dosage-sensitive genetic interactions with Pc and Psc mutations, indicating that Hdac1 is important in regulating the function of homeotic genes. In contrast to the results with Pc and Psc, no genetic interactions were observed between Hdac1^P-UTR and extra sex combs (esc) or Enhancer of zeste [E(z)] mutations. The differences in the genetic interactions with the Pc-G mutations might reflect the presence of two physically distinct complexes formed by these proteins, because the PC and PSC proteins copurify in one Pc-G complex and the ESC and E(Z) proteins copurify in a different Pc-G complex. It is interesting to note that Df(3R)10H and Hdac1^P-UTR by themselves did not cause leg transformation. Thus, the homeotic effects of Hdac1 appear to manifest themselves only in combination with Pc-G mutations as noted for several other Pc-G, including Enhancer of Polycomb, Suppressor 2 of zeste, and Mi-2 (Chang, 2001).

Further support for the role of Hdac1 in homeotic gene regulation was obtained by analyzing several newly characterized Hdac1 mutations. As observed for Hdac1^P-UTR, two missense mutations (Hdac1³⁰³ and Hdac1³¹³) and one small deletion (Hdac1^def8) enhance the Pc mutant phenotype. Surprisingly, one missense mutation, Hdac1³²⁶, suppresses the Pc phenotype significantly. This unexpected suppression probably results from a stronger effect on ectopic expression of posterior homeotic genes, causing repression of more anterior ones (Chang, 2001).

To demonstrate that the effect of Hdac1 mutations is exerted at the level of expression of homeotic genes, the expressions were examined of Sex combs reduced (Scr) and Ultrabithorax (Ubx) proteins in wild-type and Pc mutant imaginal discs. Scr proteins normally are expressed at high levels in the first leg discs, but are not expressed in the second and third leg discs. In Pc⁴ mutant heterozygotes, however, Scr proteins also can be detected at low levels in second and third leg discs. Consistent with the increase in ectopic sex comb teeth, dramatic increases in the levels of Scr proteins are observed in the second and third leg discs from Pc⁴ mutant heterozygotes that were also heterozygous for any of the Hdac1 alleles except Hdac1³²⁶. In addition, Ubx proteins are marginally detectable only in the peripodial membranes of imaginal wing discs of wild-type or Pc⁴ mutant heterozygous larvae. In larvae heterozygous for both Pc⁴ and an Hdac1 mutation, high levels of Ubx proteins are observed in the medial sections of the wing discs proper. In contrast to the lack of ectopic Scr expression in Pc⁴ heterozygotes carrying the Hdac1³²⁶ allele, a much stronger effect on ectopic Ubx expression is observed; Ubx protein levels in both first and second leg discs are increased substantially. It is highly likely that the expanded Ubx expression reduces Scr expression, resulting in suppressed Pc phenotype (i.e., reduced numbers of ectopic sex comb teeth) in Pc⁴/Hdac1³²⁶ trans-heterozygotes. These results strongly suggest that Hdac1 acts cooperatively with Pc to repress homeotic genes during larval and pupal development (Chang, 2001).

Experiments also were performed to explore the role of Hdac1 in regulating the embryonic expressions of two homeotic genes, Abd-B and Ubx. Abd-B proteins normally are expressed in a graded fashion in the posterior part of ventral nerve cord, starting from parasegment 10 (PS10). Although this pattern is not altered in homozygous Hdac1³⁰³ mutants, significant levels of Abd-B proteins are observed in more anterior parasegments of homozygous Psc^e24 mutants. Much higher levels of ectopic Abd-B proteins are found in Psc^e24 Hdac1³⁰³ double mutants, indicating a synergistic effect of Hdac1 and Psc on Abd-B repression. Consistent (but less striking) effects also are observed on Ubx protein levels. The anterior boundary of the Ubx expression domain is PS5, with the exception of a small cluster of cells in the middle of PS4 that also express Ubx proteins. Although homozygous Psc^e24 mutants only show sporadic low levels of Ubx expression in more anterior parasegments, Psc^e24 Hdac1³⁰³ double mutants show significantly higher levels of ectopic Ubx expression in more cells. In PS5, more cells with higher levels of Ubx proteins are observed in the double mutants than in either of the single mutants. In contrast, Ubx expression is reduced substantially in the abdominal parasegments of the double mutants compared with that in the single mutants, presumably reflecting Ubx repression by more extensive ectopic expression of Abd-B and possibly ABD-A. These data indicate that Hdac1 is essential for homeotic gene silencing in embryos (Chang, 2001).

The genetic interactions between Hdac1 and Pc-G mutations suggest that the encoded proteins might be physically associated. This idea was tested by examining whether HDAC1 and PC proteins can be copurified from cultured Drosophila cells. A permanent S2 cell line was established that expresses a PC protein with a FLAG-epitope tag at the C-terminal end under the control of a metallothionein promoter. Nuclear extracts prepared from induced cells were passed over a FLAG antibody column. After the addition of FLAG peptide, tagged PC and its associated proteins were eluted. Using ³H-labeled, acetylated core histones as substrates to assay these fractions, it was found that HDAC activity elutes with the same profile as PC. In addition, this activity is sensitive to the HDAC-specific inhibitor TSA (Chang, 2001).

To determine the identity of the HDAC associated with PC, immunoblotting was performed with an affinity-purified antibody against the C-terminal part of HDAC1. HDAC1 was detected in the eluted fraction. In addition, substantial amounts of PSC and PH also were copurified, consistent with previous findings that they are components of large PC protein complexes. Much lower amounts of another Pc-G protein, Sex combs on mid-leg (Scm), were detected in these preparations. Thus, these results indicate that HDAC1 is associated with the PC protein complexes in cultured cells (Chang, 2001).

To examine whether HDAC1 is associated with PC complexes in embryos, HDAC1 proteins were immunoprecipitated from embryonic nuclear extracts. PC was detected when an HDAC1 antibody was used for the immunoprecipitation. These results further support the idea that the associations between PC and HDAC1 proteins are physiologically relevant (Chang, 2001).

Given the genetic and biochemical interactions between Hdac1 and Pc, it might be anticipated that a fraction of HDAC1 would colocalize with Pc-G complexes on polytene chromosomes. Approximately 100 common binding sites have been identified for several Pc-G proteins. At least 70% of these sites (identified by staining with PSC mAbs) also stain with the HDAC1 antibody, including the Antennapedia complex at 84AB and the bithorax complex at 89E. These results suggest that HDAC1 proteins act together with a substantial fraction of the Pc-G complex. However, the relative intensities of the signals for PSC and HDAC1 at these sites do not always correlate, suggesting a regulatory, rather than a constitutive function. Furthermore, HDAC1 is much more widely distributed along the chromosomes than is PSC, consistent with its role in global gene regulation and/or chromatin structure (Chang, 2001).

The colocalizations of HDAC1 and PSC were examined further on polytene chromosomes from a transgenic line that carries a Ubx upstream cis-regulatory region (i.e., bxd-14) inserted at 62A. This insert contains a functional PRE and creates a new Pc-G-binding site. Staining with both PSC and HDAC1 antibodies reveals that a new PSC site coincides with a new HDAC1-binding site. This new binding site is beside an HDAC1 site present in the wild-type chromosome, creating a broader signal of HDAC1 at this site. These results strongly suggest that HDAC1 and Pc-G proteins are recruited to this ectopic PRE (Chang, 2001).

These results do not imply that there is a direct physical interaction between HDAC1 and any Pc-G proteins that have been characterized to date. It is possible that an adapter-like molecule might be involved. In several organisms, direct interactions or cytological colocalization between HDAC1 and SIN3A proteins have been demonstrated. However, there is currently no evidence that SIN3A is required for homeotic gene repression. The observations that HDAC1 is not detected at about 30% of PSC sites and that HDAC1 intensity does not always correlate with that of remaining PSC sites suggests that HDAC1 is not constantly associated with the PC complexes. This is consistent with a catalytic rather than a structural function. Different levels of HDAC1 staining might reflect varying degrees of repression at these sites. Because more acetyl groups need to be removed when a gene becomes repressed from an active state, it is also possible that higher levels of HDAC1 are required for the initiation of a repressed state than for the maintenance of a repressed state. Thus, the significance of the relative levels of HDAC1 should be interpreted with caution (Chang, 2001).

Drosophila Enhancer of Zeste/Esc complexes have a histone H3 methyltransferase activity that marks chromosomal polycomb sites

The chromodomain of HP1 has been shown to bind specifically both dimethyl and trimethyl lysine 9 N-terminal peptide of histone H3. The structure of the chromodomain and the sequence conservation of critical residues in Pc suggest that the Pc chromodomain might also have such a specific affinity. Immobilized H3 N-terminal peptide unmethylated or trimethylated at lysine 9 was used to see if these peptides were able to bind recombinant Pc protein. The Pc protein has some affinity for the trimethylated peptide but binds to the same extent to the unmethylated peptide. The interaction was examined of GST-Pc to full-length histone H3 either methylated with purified HIMalpha or acetylated with GCN5. Compared to the acetylated H3, methylated H3 binds five times better to GST-Pc. The binding is nearly unaffected by the K9A mutation but decreases more than 2-fold with the K27A mutant H3. These results indicate that methylation by HIMalpha significantly increases the affinity of H3 for Pc but most of this affinity is due to K27 methylation (Czermin, 2002).

Native and recombinant polycomb group complexes establish a selective block to template accessibility to repress transcription in vitro

Polycomb group (PcG) proteins are responsible for stable repression of homeotic gene expression during Drosophila melanogaster development. They are thought to stabilize chromatin structure to prevent transcription, though how they do this is unknown. An in vitro system has been established in which the PcG complex PRC1 and a recombinant PRC1 core complex (PCC; a stable complex of the core PcG subunits of the PRC1 complex—Polycomb, Polyhomeotic, Posterior Sex Combs, and dRING1) are able to repress transcription by both RNA polymerase II and by T7 RNA polymerase. Assembly of the template into nucleosomes enhances repression by PRC1 and PCC. The subunit Psc is able to inhibit transcription on its own. PRC1- and PCC-repressed templates remain accessible to Gal4-VP16 binding, and incubation of the template with HeLa nuclear extract before the addition of PCC eliminates PCC repression. These results suggest that PcG proteins do not merely prohibit all transcription machinery from binding the template but instead likely inhibit specific steps in the transcription reaction (King, 2002).

These results are consistent with the hypothesis that an interaction with the template is primarily responsible for PcG repression, rather than an interaction with the general transcription machinery. PRC1 represses transcription of both RNA Pol II and T7 RNA polymerase, and repression of both of these polymerases is quantitatively similar, suggesting that repression requires neither a specific interaction with either polymerase nor one with eukaryotic transcription factors. In addition, the enhancement of repression observed for Pol II transcription on a chromatin template in comparison to naked DNA is also quite similar to the enhancement observed with T7 transcription. Since T7 transcription does not involve activators or any of the general transcription factors required for Pol II transcription, it seems unlikely that PRC1 represses transcription predominantly by interacting with or sequestering these factors. The simplest way to explain how PRC1 and PCC behave in this system is that they interact with the template to block steps required for transcription (King, 2002).

Though this study indicates that PcG complexes are capable of repressing transcription in vitro through an interaction between the PcG complexes and the template, it is still unknown whether interactions with the general machinery also play a role in PcG repression. Indeed, the finding that PRC1 contains a number of TAF proteins, which are also a part of the TFIID complex, raises the possibility of interactions between PRC1 and general factors at the promoter. However, PCC, which contains only PcG proteins, is sufficient for repression in this system, implying that the TAF subunits of PRC1 are not essential for the repression observed here. The TAF subunits may be involved in targeting PcG activity, since TAF proteins are capable of interacting with promoter sequences and with transcription factors (King, 2002).

The Psc subunit of PCC has substantial activity as a single protein, suggesting that Psc may be central to the mechanism of transcriptional repression by PcG complexes. This is in accord with previous results that show that Psc on its own is sufficient to inhibit chromatin remodeling by Swi/Snf in vitro and raises the possibility that a template interaction mediated mainly by Psc is responsible for both of these in vitro activities. The Ph subunit also has a small amount of activity on its own, which is consistent with in vivo studies that have also indicated that Psc and Ph may be central to the mechanism of repression. When Psc and its homolog Su(z)2, or Ph, are removed from imaginal disc clones by recombination, homeotic gene expression is disrupted more quickly and to a greater extent than it is when other PcG genes including Pc are removed. The other subunits of PCC that do not appear to inhibit transcription might function primarily in other aspects of repression that are not limiting in this assay, such as the targeting of complexes to specific genes (King, 2002).

Structural basis for specific binding of Polycomb chromodomain to histone H3 methylated at Lys 27

The chromodomain of Drosophila Polycomb protein is essential for maintaining the silencing state of homeotic genes during development. Recent studies suggest that Polycomb mediates the assembly of repressive higher-order chromatin structures in conjunction with the methylation of Lys 27 of histone H3 by a Polycomb group repressor complex. A similar mechanism in heterochromatin assembly is mediated by HP1, a chromodomain protein that binds to histone H3 methylated at Lys 9. To understand the molecular mechanism of the methyl-Lys 27 histone code recognition, a 1.4-Å-resolution structure was determined of the chromodomain of Polycomb in complex with a histone H3 peptide trimethylated at Lys 27. The structure reveals a conserved mode of methyl-lysine binding and identifies Polycomb-specific interactions with histone H3. The structure also reveals a Pc dimer in the crystal lattice that is mediated by residues specifically conserved in the Polycomb family of chromodomains. The dimerization of Drosophila Pc can effectively account for the histone-binding specificity and provides new mechanistic insights into the function of Polycomb. It is proposed that self-association is functionally important for Polycomb (Min, 2003).

The crystal structure presented in this study shows that the chromodomains of Pc and HP1 have a common overall structure, and they also bind methyl-lysine-containing substrates similarly. The structure identifies that Arg 67 and Asp 65 of Pc are important for Polycomb-specific interactions with histone H3. Curiously, these residues interact with the main chain of the histone peptide. This is also true in the HP1 structure, where most of the chromodomain-histone interactions are through main-chain atoms. An important question concerning the recognition of the methylated histone tail by Polycomb and HP1 is what determines their binding specificities. The difference in histone-binding affinity of Polycomb and HP1 is not sufficient to account for their substrate specificities. Crystallographic analyses have identified a Pc chromodomain dimer in the crystal lattice that can account for the binding specificity of Pc, compelling reasons in support of the potential physiological significance of the observed dimeric interactions are outlined (Min, 2003).

Although dimerization of Pc was never pointed out explicitly before, self-association of the Pc chromodomain was noted in several previous studies. Interestingly, an in vivo domain-swap experiment replacing the chromodomain of HP1 with that of Pc has shown that the chimeric protein binds to both heterochromatin and Pc binding sites in polytene chromosomes. Furthermore, some endogenous Pc is misdirected to the heterochromatic center, whereas some endogenous HP1 is mislocalized to the Pc binding sites. These observations not only confirmed earlier observations that the chromodomain of Pc and the C-terminal chromo shadow domain of HP1 possess intrinsic nuclear localization and chromosomal binding properties, they also indicate that the dPC chromodomain directs the mislocalization of endogenous dPC through protein-protein contacts (Min, 2003).

The dimerization model predicts that the binding specificity of the Pc chromodomain arises from histone-histone interactions resulting from the close proximity of the two histone-tail-binding sites in the Pc chromodomain dimer. The binding of two methyl-Lys9 histone H3 tails by the Pc chromodomain dimer is disallowed because of potential steric clashes. It is possible that the Pc chromodomain dimer may bind only one histone H3 tail in some instances. In this case, the available structural information cannot exclude the binding of a methyl-Lys9 histone H3 tail to the Pc chromodomain dimer. The observation that the Pc chromodomain binds specifically to methyl-27 histone H3 peptides in vitro clearly supports the binding of two histone H3 tails to the Pc dimer. The structure also provides insights into the function of Pc proteins in the assembly of repressive higher-order chromatin structure. Because the two histone H3-binding sites are closely juxtaposed, the two histone tails are unlikely to come from the same nucleosome. In the nucleosome core particle structure, Lys 27 is ordered in one of the H3 tails, whereas the ordered residues start from Pro 38 in the other H3 molecule. The observed distance between the Calpha atoms of Pro 38 and Lys 27 of the same H3 molecule is 26 Å. The two histone-H3 Pro 38 residues are 73 Å apart (linear distance) in the nucleosome core particle structure, which coverts to ~80 Å along the arc of wrapped DNA. The two Lys 27 Calpha atoms must be within 27 Å to occupy the two binding sites in the Pc chromodomain dimer. Assuming that the two H3 tails N-terminal to Pro 38 can be maximally stretched and free to adopt any conformation, it has not been possible to model the simultaneous binding of two histone tails from the same nucleosome to the Pc chromodomain dimer without steric clashes. Thus, it is believed that the histone tails binding to the Pc chromodomain dimer must come from two separate nucleosomes. The in vivo binding of two methyl-Lys 27 histone-H3 tails, from spatially adjacent nucleosomes, will effectively lock the nucleosomes into a more compact configuration. This compaction will lead to a repressive chromatin state associated with the silencing of homeotic genes. A similar function for HP1 in the assembly of heterochromatin has been proposed, although dimerization of HP1 via the C-terminal chromo shadow domain makes the binding of two histone tails from the same nucleosome, as well as separate ones, possible (Min, 2003).

On the histone H3 tail, Lys 9 and Lys 27 are both methylation sites associated with epigenetic repression, and reside within a highly related sequence motif ARKS, a sequence common to chromodomain proteins Polycomb (Pc) and heterochromatin protein 1 (HP1). Pc and HP1 are highly discriminatory for binding to these sites in vivo and in vitro. In Drosophila S2 cells, and on polytene chromosomes, methyl-Lys 27 and Pc are both excluded from areas that are enriched in methyl-Lys 9 and HP1. Swapping of the chromodomain regions of Pc and HP1 is sufficient for switching the nuclear localization patterns of these factors, indicating a role for their chromodomains in both target site binding and discrimination. To better understand the molecular basis for the selection of methyl-lysine binding sites, the 1.8 Å structure of the Pc chromodomain was solved in complex with a H3 peptide bearing trimethyl-Lys 27, and it was compared with the structure of the HP1 chromodomain in complex with a H3 peptide bearing trimethyl-Lys 9. The Pc chromodomain distinguishes its methylation target on the H3 tail via an extended recognition groove that binds five additional residues preceding the ARKS motif (Fischle, 2003).

These studies show a clear preference of the Pc chromodomain for the H3 Lys 27 methyl mark. How does this activity of Pc contribute to PcG function? In the case of the formation of heterochromatin and the initially genetically defined pathway of suppression of variegation, it has been suggested that methylation of H3 on Lys 9 by Suv3-9 generates a docking site for the HP1 (also known as Suv2-5) chromodomain. Further recruitment of Suv3-9 by the chromo shadow domain of HP1 has been postulated to lead to a perpetuation and spreading of a heterochromatic domain until blocked by yet unknown mechanisms. Similarly, Esc-E(z)-dependent methylation of Lys 27 (and possibly Lys 9) and consecutive recruitment of Pc and Pc-containing complexes might contribute to the stability of the PcG complex, particularly in the early stages of assembly at a PRE by permitting complex formation to spread to neighboring sequences. This interpretation of the specific binding of Pc to methyl-Lys 27 is in agreement with studies demonstrating loss of chromosome binding for several components of PRC1 upon inactivation of E(z) and is consistent with several other in vivo results that imply synergy between these complexes. However, it is unclear at present if dynamic perpetuated spreading of a Lys 27 mark indeed exists and is dependent on Pc recruitment and Esc-E(z) enzymatic activity. Other possible functions for the binding of Pc to the Lys 27 methyl mark include a more static maintenance effect that could contribute to epigenetic memory. In this model, the recruitment of Esc-E(z) would be independent of and precede any involvement of Pc binding. Alternatively, although complex recruitment could be constitutive, the decision to repress or not could depend on an epigenetic switch mediated by Lys 27 methylation and its interaction with local Pc/PRC1 (Fischle, 2003 and references therein).

However, alternative routes and mechanisms of Pc and PcG recruitment to local sites of chromatin (in addition to recognition of Lys 27 methylation) might also exist. For example, studies involving the localization of a chimeric HP1 protein containing the chromodomain of Pc on polytene chromosomes imply critical interactions of the Pc chromodomain with other PcG components that are recruited to PcG sites. Therefore, the particular localization patterns observed for the wild-type and chimeric proteins in swapping experiments might be the result of additive effects, including other targeting mechanisms besides methyl-lysine binding. Nevertheless, it is intriguing to note that the subnuclear localization patterns for wild-type and chimeric Pc and HP1 proteins are coincident with the localization of specifically recognized methyl-lysine marks on the histone H3 tail. It is unclear to what extent additional regions of the proteins C-terminal to the chromodomains could contribute to subnuclear localization and function. For example, different studies have implicated the C-terminal chromo shadow domain and hinge regions of HP1 in addition to the chromodomain in the specific subnuclear targeting of this factor. However, a C-terminal truncation of Pc does not affect its specific chromosomal localization. Additional studies will have to address the exact contribution of the chromodomains and their recognition and binding of particular methyl-lysine marks to the specific functions of the Pc and HP1 proteins (Fischle, 2003 and references therein).

The Drosophila RYBP gene functions as a Polycomb-dependent transcriptional repressor

The Polycomb and trithorax groups of genes control the maintenance of homeotic gene expression in a variety of organisms. A putative participant in the regulation of this process is the murine RYBP (Ring and YY1 Binding Protein). Sequence comparison between different species has identified the homologous gene in Drosophila, the dRYBP gene. Whether dRYBP participates in the mechanisms of silencing of homeotic genes expression was investigated. dRYBP expression, examined by RNA in situ hybridisation, was found ubiquitously throughout development. Moreover, a polyclonal anti-dRYBP antibody was generated that recognises the dRYBP protein. dRYBP protein is nuclear and expressed maternally and ubiquitously throughout development. To study the transcriptional activity of dRYBP, a fusion protein was generated containing the entire dRYBP protein and the GAL4 DNA binding domain. This fusion protein functions, in vivo, as a transcriptional repressor throughout development. Importantly, this repression is dependent on the function of the Polycomb group genes. Furthermore, using the GAL4/UAS system, dRYBP was over-expressed in the haltere and the wing imaginal discs. In the haltere discs, high levels of dRYBP repress the expression of the homeotic Ultrabithorax gene. This repression is Polycomb dependent. In the wing discs, dRYBP over-expression produces a variety of phenotypes suggesting the overall miss-regulation of the many putative genes affected by high levels of dRYBP. Taking together, these results indicate that dRYBP is able to interact with PcG proteins to repress transcription suggesting that the dRYBP gene might belong to the Polycomb group of genes in Drosophila (Bejarano, 2005).

The mouse homologous gene, RYBP, was identified in a two-hybrid screen for murine Ring1 interacting proteins. RYBP family members include the human YEAF1 homologous gene and the murine and human YAF2 gene coding for structurally related proteins. Although very similar in sequences, they seem to have different functions as transcriptional regulators of the hGABP gene, i.e. YAF2 positively regulates the transcriptional activity of hGABP but YEAF1 negatively regulates this activity (Sawa, 2002; Bejarano, 2005).

dRYBP is expressed maternally and throughout development in all the nuclei of the embryo and the imaginal discs cells. The murine RYBP gene is also expressed ubiquitiously in the mouse embryo (Garcia, 1999). The ubiquitous and nuclear pattern of dRYBP expression coincides with the pattern of expression of the Polycomb group proteins so far described (Bejarano, 2005).

When dRYBP is tethered to DNA sequences, it is able to repress the transcriptional state of minigene reporter constructs. Moreover, GALDB-dRYBP transcriptional repression function requires the products of at least the Pc, Sce and pho genes, suggesting that GALDB-dRYBP represses transcription by interacting with PcG protein complexes. The Pho protein (homologous to mouse YY1) is able to bind DNA in a sequence specific manner and it has been proposed to recruit the PcG complexes to DNA. However, the results show that the transcriptional repression function of GALDB-dRYBP cannot be achieved in the absence of Pho protein. Although silencing in these experimental conditions could formally result solely from the interaction of dRYBP with Pho, the need of Pho to execute the transcriptional repression may also suggest that in the process of maintenance of homeotic gene expression, the Pho protein serve other functions than the recruitment of PcG complexes to DNA (Bejarano, 2005).

Additional evidence for the transcriptional repressor function of dRYBP comes from the experiments of over-expression of dRYBP using the GAL4/UAS system. UbxGAL4/UASdRYBP halteres show partial transformation towards wing which is correlated with the repression of UBX expression in the haltere imaginal discs due to high levels of dRYBP. The partial transformation of the haltere towards wing is not fully understood. It is speculated that the over-expression of dRYBP may also affect genes involved in proliferation that act downstream the Ubx gene. The repressive effect is Polycomb dependent, suggesting that dRYBP transcriptional repression function needs the interaction with Polycomb proteins. Moreover, although no changes have been detected in the levels of engrailed expression, some of the phenotypes observed in enGAL4/UASdRYBP flies are indicative of engrailed repression, revealing again the repressor effect of dRYBP over expression (Bejarano, 2005).

A model has been proposed in which RYBP protein, through its interaction with DNA-binding proteins like YY1, function as a ‘bridge’ to ensure interactions of DNA and non-DNA binding proteins in multimeric protein complexes. It is not yet known if dRYBP serves a similar bridging function in Drosophila. The YY1 protein (homologous to Drosophila Pho) is able to bind DNA in a sequence specific manner and directly interacts with dRYBP. It is speculated that dRYBP, serves a similar bridging function, bridging between DNA binding proteins like Pho and the multimeric PcG complexes. Further work and mutations in the dRYBP gene will be necessary to define whether dRYBP serves this putative bridge function (Bejarano, 2005).

dRYBP over-expression in the wing produces homeotic and non-homeotic phenotypes indicative of miss regulation of a variety of genes. High levels of dRYBP in the wing (i.e. sdGAL4/UASdRYBP flies) produces, among others, transformation towards haltere with the corresponding expression of the Ubx protein in the wing cells, i.e. outside its normal domain of expression. This effect could seem opposite to the repressor effect observed when dRYBP is tethered to DNA (GALDBdRYBP) or when dRYBP is over expressed under the control of the UbxGAL4 line. However, interference with the assembling/recruting of the PcG and trxG complexes either because of sequestration of PcG/trxG proteins, perturbation of the PcG/trxG balance or disruption of the cross regulatory interactions between PcG proteins could perhaps explain the observed expression of UBX protein in the wing disc due to over-expression of dRYBP. Alternatively, over abundance of dRYBP or dRYBP containing complexes might lead to a unique target gene repretoire that lead to the effects observed. Finally, the cross regulatory interactions between the genes patterning the wing, that are perhaps being miss regulated by the high levels of dRYBP could also explain the range of phenotypes observed in the wing due to over expression of dRYBP (Bejarano, 2005).

In conclusion, these results show that dRYBP protein is nuclear, maternal and ubiquotiously expressed throughout development. The results also show that dRYBP functions, in a Polycomb dependent manner, as a transcriptional repressor, suggesting that dRYBP is able to interact with the PcG proteins to repress transcription and therefore might belong to the Polycomb group of genes of Drosophila. Finally, the study of the multiple phenotypes produced by high levels of dRYBP in the wing might be indicative of the involvement of dRYBP on the regulation of many genes as also described for the PcG genes in Drosophila (Bejarano, 2005).

Architecture of a polycomb nucleoprotein complex; Chromatin immunoprecipitation and nuclease mapping demonstrate that PREs are nucleosome depleted

Polycomb group (PcG) epigenetic silencing proteins act through cis-acting DNA sequences, named Polycomb response elements (PREs). Within PREs, Pleiohomeotic (Pho) binding sites and juxtaposed Pc binding elements (PBEs) function as an integrated DNA platform for the synergistic binding of Pho and the multisubunit Polycomb core complex (PCC). This study analyzed the architecture of the Pho/PCC/PRE nucleoprotein complex. DNase I footprinting revealed extensive contacts between Pho/PCC and the PRE. Scanning force microscopy (SFM) in combination with DNA topological assays suggested that Pho/PCC wraps the PRE DNA around its surface in a constrained negative supercoil. These features are difficult to reconcile with the simultaneous presence of nucleosomes at the PRE. Indeed, chromatin immunoprecipitations (ChIPs) and nuclease mapping demonstrated that PREs are nucleosome depleted in vivo. The implications of these findings for models explaining PRE function are discussed (Mohd-Sarip, 2006).

How specialized DNA elements such as PREs can bring a linked gene under epigenetic control remains poorly understood. An important breakthrough was the identification of Pho as a sequence-specific PcG protein. Subsequent research firmly established that Pho forms a critical component of the 'PRE code.' Another building block of PREs, the PBE is located directly downstream of Pho binding sites. Pho and PCC interact only weakly in solution, but docking onto Pho/PBE modules drives the assemblage of a stable Pho/PCC/PRE silenceosome. The mechanistical properties of silenceosome and enhanceosome formation are strikingly similar. Both involve synergistic interactions between a stereo-specific arrangement of binding sites and a reciprocal network of protein-protein interactions. This study investigated the architecture of a PcG silenceosome (Mohd-Sarip, 2006).

The results revealed that Pho/PCC contacts the bxd PRE over 400 bp and wraps the PRE DNA around its surface in a constrained negative supercoil. It has been found that Pho/PCC binding to the PRE can overcome chromatinization. Moreover, the resulting DNase I digestion pattern of the Pho/PCC/PRE complex suggested the absence of nucleosomes. It is estimated that a Pho/PCC oligomer can wrap more DNA around its surface than a nucleosome. The extensive contacts between Pho/PCC and PRE DNA together with the left-handed wrapping are likely to affect histone-DNA interactions (Mohd-Sarip, 2006).

The 400 bp bxd PRE core is nucleosome poor in vivo, as revealed by nuclease mapping and quantitative ChIP assays. Likewise, the PREs from the Abd-B cis-regulatory domains were found to be nuclease hypersensitive in chromatin digests. Other researchers have also suggested that the core PRE region is largely devoid of nucleosomes. While this work was in progress, ChIP analysis by others independently established that PREs are depleted for histones. The iab-7 PRE sequences required for the pairing-sensitive silencing of mini-white in transgene assays closely coincide with the nuclease-hypersensitive region bound by Pho/PCC. Finally, there appears to be a good correlation between PRE activity and the extent of nuclease hypersensitivity (Mohd-Sarip, 2006).

These findings dovetail nicely with the results of recent genome-scale determination of nucleosome positioning in yeast. These studies suggested that RNA polymerase II promoters comprise a nucleosome-free region flanked by positioned nucleosomes, bearing a stereotyped pattern of histone modifications. It is proposed that, like promoters and enhancers, PREs are in a nucleosome-depleted conformation in vivo (Mohd-Sarip, 2006).

It is suggested that PcG-directed gene silencing is a multistep process, initiated by silenceosome formation on the PRE. The next step requires the establishment of a silenced state onto PRE-linked genes. PCC-histone interactions, modulated by covalent histone modifications, are likely to be the main driving force of sequence-independent spreading over a target gene. Thus, it is imagined that histone modifications would generally follow, rather than precede, Polycomb nucleocomplex formation on PREs. Collectively, the available evidence enforces the notion that a cooperative network of contextual protein-DNA and protein-protein interactions nucleates silenceosome formation. This work presents a view of the architecture of a Pho/PCC/PRE nucleoprotein complex and provides a framework for models explaining PRE function (Mohd-Sarip, 2006).

Polycomb complexes and the propagation of the methylation mark at the Drosophila Ubx gene

Polycomb group proteins are transcriptional repressors that control many developmental genes. The Polycomb group protein Enhancer of Zeste has been shown in vitro to methylate specifically lysine 27 and lysine 9 of histone H3 but the role of this modification in Polycomb silencing is unknown. This study shows that H3 trimethylated at lysine 27 is found on the entire Ubx gene silenced by Polycomb. However, Enhancer of Zeste and other Polycomb group proteins stay primarily localized at their response elements, which appear to be the least methylated parts of the silenced gene. These results suggest that, contrary to the prevailing view, the Polycomb group proteins and methyltransferase complexes are recruited to the Polycomb response elements independently of histone methylation and then loop over to scan the entire region, methylating all accessible nucleosomes. It is proposed that the Polycomb chromodomain is required for the looping mechanism that spreads methylation over a broad domain, which in turn is required for the stability of the Polycomb group protein complex. Both the spread of methylation from the Polycomb response elements, and the silencing effect can be blocked by the gypsy insulator (Kahn, 2006).

The experiments described in this study show clearly that all three PcG proteins tested, Pc, Psc, and E(z), are preferentially located at the PREs. This specificity is clearest and most sharply delineated in the case of Psc and E(z). In the case of PC, the peaks centered at the PREs are much broader, including secondary peaks, and although the binding detected at other Ubx regions decreases to low values, it never reaches the level seen at control sites such as the white gene that possess no PRE. The second basic conclusion from these experiments is that, in contrast to the localization of PcG proteins, the H3 me3K27 profile forms a broad domain that includes the entire Ubx transcription unit and upstream regulatory region. The third important observation is that, contrary to previously published accounts, the PREs themselves appear to contain the lowest levels of me3K27 of the entire domain. This surprising result will be considered first. The lack of apparent methylation at the PREs does not depend on the antibody used or on the level of cross-linking. Comparable results were obtained with two different anti-me3K27 antibodies and with anti-me3K9. Furthermore, the fact that a similar result was obtained with antibody against total histone H3 or histone H2B suggests that nucleosomes are underrepresented at the bxd PRE core. The result is not because of lack of accessibility to the histone or to the epitope: GAGA factor bound to the PRE appears as easily accessible as GAGA factor bound to the Ubx promoter. Salt extraction of the PcG complexes before cross-linking does not qualitatively change the me3K27 binding profile (Kahn, 2006).

In sum, careful quantitative analysis of ChIP indicates that while PcG proteins are principally localized at the PRE, the histone H3 methylation they produce is distributed over the entire Ubx gene. It is evident from this and from the undermethylation of the PRE core that that K27 methylation does not, by itself, recruit PcG complexes. This does not preclude an important role for methylation in PcG binding and silencing but suggests that the relationship between the two requires a more dynamic model (Kahn, 2006).

PREs have been shown to recruit PcG complexes and to produce new binding foci detectable in polytene chromosomes. It is not surprising therefore to find the three PcG proteins tested are associated with the two Ubx PREs. A much smaller peak in microarray profiles for all three proteins can be discerned in the vicinity of the Ubx 3'-exon but its significance is unknown. More surprising was the striking difference between the distributions of Psc and E(z) and that of Pc. E(z) and Psc belong to two different complexes that do not co-precipitate (except in the very early embryo) but are both recruited to the PRE. Pc and Psc are core components of the PRC1-type of PcG complex yet, while Psc is detected almost exclusively at the PREs, Pc has a much broader distribution peaking at the PREs but tailing over considerable distances along the Ubx gene and regulatory regions. The simplest interpretation of this is that a second type of complex containing Pc but not Psc is recruited by a different mechanism to the rest of the Ubx sequences. Alternatively, the same complex, containing both proteins is involved in both cases but the nature of the chromatin contact is different, such that in one case both proteins are well cross-linked to the chromatin but in the second case only Pc is efficiently cross-linked (Kahn, 2006).

Just as striking is the fact that, although the E(z) complex is responsible for the H3 K27 methylation spread over the entire Ubx gene, the E(z) protein is found localized at the PREs. It is concluded that the E(z) complex methylates the Ubx domain by a hit-and-run type of mechanism. Because the methylation is stable, the E(z) complex needs only visit each nucleosome once on the average every cell cycle. It is noted that E(z)-dependent histone H3 K27 dimethylation is highly abundant and widely distributed in the genome but E(z) complexes are not associated with it. Where then does the E(z) that methylates PcG target genes come from? While more complicated scenarios may be imagined, the simplest one involves the E(z) complex bound at the PRE (Kahn, 2006).

It is supposed that the PcG complexes are recruited to PREs by DNA-binding proteins independently of histone methylation. To methylate the entire Ubx domain, the E(z)/ESC histone MTase complex might then detach from the PRE and slide along the chromatin from one nucleosome to the next to survey the entire domain. However, it more likely that both the Pc and the E(z) complexes assembled at the PRE remain associated with the PRE sequences, where they are detected, but that the whole PRE assembly loops over to scan the entire region, methylating all accessible nucleosomes. Such looping models were originally proposed to be mediated by sites of weak PcG complex formation. In a modern version of this type of model, the looping activity would be mediated by the distinct affinity of the Pc protein for histone H3, which is greatly increased by K27 methylation. These affinities would mediate transient interactions of the complexes bound at the PRE with the surrounding chromatin and allow continuous scanning and methylation of unmethylated or hemimethylated nucleosomes (Kahn, 2006).

In such a model, ChIP experiments would always detect a strong PcG presence at the PRE but PcG interactions with the rest of the repressed gene would be distributed over a region, which is very large in the case of the Ubx gene, smaller in the case of the YGPhsW transposon, hence the signal detected at any one site would be weaker in proportion to the extent of the methylated domain. In addition, the contacts between the PcG complex and the rest of the silenced gene would be much more transient than contacts with the PRE. Together, these considerations would explain why ChIP assay gives such low values for PcG proteins over the rest of the methylated domain (Kahn, 2006).

The looping mechanism proposed for the PRE-bound complex strongly resembles that suggested for the interaction between the Locus Control Region and β-globin genes or for enhancer-promoter interactions. Like these interactions, the silencing of a promoter by the PRE is blocked by insulator elements. In transposon constructs, the insertion of a gypsy Su(Hw) insulator between PRE and promoter blocks the spread of methylation. At present, the mechanism of insulator action is not clear and how the block to methylation is achieved is unknown. It is possible that the insulator element produces topological constraints that prevent the PRE-bound complexes from looping beyond the insulator. This would be consistent with the observation that a significant level of Pc presence becomes detectable over the yellow gene when the insulator block is lifted (Kahn, 2006).

Although the data argue against a principal role of histone methylation in the recruitment of Polycomb proteins to their response elements, it seems to be important for both transcriptional repression and stable association of PcG proteins with chromatin. Loss of catalytic E(z) function eventually results in derepression of HOX genes and dissociation of PcG proteins from polytene chromosomes. It is speculated that once the me3K27 domain is established, modified nucleosomes will pave the way for looping interactions of the PRE-bound PcG proteins with the parts of the silenced gene including promoter or enhancer regions. Silencing might then result from hit-and-run interactions with either or both, possibly even resulting in methylation of the associated factors. Alternatively or in addition, trimethylation of K27 and possibly K9 may directly interfere with the signaling cascade of consecutive histone modifications that guide the multistep process of transcription initiation and elongation. Since histone methylation is thought to persist through cell division its immediate presence at the very beginning of the subsequent interphase might win the time necessary for the full assembly of PcG complexes on the PREs before competing transcription has taken over (Kahn, 2006).

Polycomb proteins remain bound to chromatin and DNA during DNA replication in vitro

The transcriptional status of a gene can be maintained through multiple rounds of cell division during development. This epigenetic effect is believed to reflect heritable changes in chromatin folding and histone modifications or variants at target genes, but little is known about how these chromatin features are inherited through cell division. A particular challenge for maintaining transcription states is DNA replication, which disrupts or dilutes chromatin-associated proteins and histone modifications. PRC1-class Polycomb group protein complexes, consisting of four core PcG subunits, polyhomeotic (Ph), posterior sex combs (PSC), dRING, and Polycomb (Pc), are essential for development and are thought to heritably silence transcription by altering chromatin folding and histone modifications. It is not known whether these complexes and their effects are maintained during DNA replication or subsequently re-established. When PRC1-class Polycomb complex-bound chromatin or DNA is replicated in vitro, Polycomb complexes remain bound to replicated templates. Retention of Polycomb proteins through DNA replication may contribute to maintenance of transcriptional silencing through cell division (Francis, 2009).

The data suggest that PCC is not released into solution during passage of the DNA replication fork. Furthermore, nucleosomes facilitate PCC binding to and retention on templates, but are not essential for either. The finding that PCC can be maintained on either chromatin or naked DNA is interesting in light of the finding that PREs are sites of rapid histone turnover and can be depleted of nucleosomes (Francis, 2009).

One model for the transfer of PCC during DNA replication is that the complex remains in direct contact with DNA during passage of the DNA replication fork. Contacts between PcG proteins and nucleosomes or DNA could be disrupted in front of the replication fork, but replaced by contacts with nucleosomes or DNA behind the replication fork. This mechanism has been proposed for transfers of histone-DNA contacts during replication and transcription in vitro. PCC can likely contact multiple nucleosomes or a long stretch of DNA, which may allow the complex to remain on chromatin when some template contacts are disrupted. A second model is that PCC interacts with the replication machinery, either directly or through intermediary factors. These interactions could retain PCC near DNA during replication, even if direct DNA contacts are disrupted, allowing rapid rebinding of PCC to newly replicated chromatin. Consistent with this idea, several chromatin-modifying proteins can interact with components of the DNA replication machinery (Francis, 2009).

The inhibition of DNA and chromatin replication by PCC in vitro raises the question of how PcG-bound regions are replicated if PRC1-class complexes are indeed continuously bound. If PCC inhibits replication initiation but not elongation, as the results suggest, then PRC1-class complexes would limit replication only if they were bound near replication origins (Francis, 2009).

Intriguingly, targeting of Pc to a replication origin in Drosophila that mediates developmental chorion gene amplification in follicle cells decreased gene amplification (Aggarwal, 2004) and PcG-silenced regions of polytene chromosomes (such as Hox gene clusters) are underreplicated, although this involves additional genes such as Suppressor of DNA Underreplication (Marchetti, 2003; Moshkin, 2001; Francis, 2009 and references therein).

Reduction of PcG protein levels leads to reactivation of their target genes, suggesting that these genes are continuously susceptible to transcriptional activation. It may therefore be important that PRC1-class complexes, which can directly repress transcription, maintain constant association with genes marked for silencing (Francis, 2009).

It was surprising to find that H3K27me3 is not essential for maintaining PRC1-class complexes through DNA replication in vitro. It is possible that retention of parental PRC1-class complexes and recruitment of new complexes are mechanistically distinct because no evidence was found for recruitment of new PCC during replication, and in vivo data suggest that PSC is present on newly replicated chromatin but that additional PSC is recruited after replication. This may be similar to histone proteins in that it is thought that parental histones are transferred randomly to the two daughter strands, followed by deposition of new histones by replication-coupled assembly complexes. In vivo data raise the possibility that recruitment of new PRC1 is not directly coupled to DNA replication; perhaps it involves H3K27me3 (Francis, 2009).

In these experiments, PCC interacts with chromatin through mass action, but in vivo, PRC1-class complexes are specifically targeted to PREs. It is hypothesized that the stable association of PCC with chromatin observed in this study reflects how the complex could behave once it is recruited to a PRE, but it will be important to test this mechanism in a system where PCC is targeted (Francis, 2009).

In conclusion, the ability of parental PCC to be transferred to daughter chromatin may help explain how PcG-mediated repression established by transiently acting factors can be propagated through cell generations. These data also suggest that maintenance of chromatin regulatory proteins through DNA replication might be an important mechanism of epigenetic inheritance (Francis, 2009).

Accessibility of the Drosophila genome discriminates PcG repression, H4K16 acetylation and replication timing

Histone modifications are thought to regulate gene expression in part modulating DNA accessibility. This study measured genome-wide DNA accessibility in Drosophila melanogaster by combining M.SssI methylation footprinting with methylated DNA immunoprecipitation. Methylase accessibility demarcates differential distribution of active and repressive histone modifications as well as sites of transcription and replication initiation. DNA accessibility is increased at active promoters and chromosomal regions that are hyperacetylated at H4K16, particularly at the male X chromosome, suggesting that transcriptional dosage compensation is facilitated by permissive chromatin structure. Conversely, inactive chromosomal domains decorated with H3K27me3 are least accessible, supporting a model for Polycomb-mediated chromatin compaction. In addition, higher accessibility was detect at chromosomal regions that replicate early and at sites of replication initiation. Together, these findings indicate that differential histone-modification patterns and the organization of replication have distinct and measurable effects on the exposure of the DNA template (Bell, 2010).

Covalent modifications of histones, including acetylation and methylation, are important determinants of chromatin structure. These modifications are thought either to serve as platforms to recruit proteins to chromatin or to directly alter its physical properties, yet little is known about the exact molecular mechanisms involved. Chromatin structure is often separated into 'active chromatin' versus 'heterochromatin'. However, this bimodal classification does not take into account the degree of DNA accessibility or the multiple variables of chromatin in vivo. Active modifications such as histone H3 Lys4 methylation (H3K4me) and histone H4 hyperacetylation localize to promoters and the 5' ends of active genes, which also show low nucleosomal density and frequently harbor sites of DNase I hypersensitivity. Direct links have been established between histone H4 acetylation and higher-order chromatin folding. In vitro, acetylation of H4 at Lys16 (H4K16ac) can inhibit the formation of compact 30-nm fibers and impair the activity of ATP-dependent chromatin remodeling. As H4K16ac is critical for male dosage compensation in Drosophila, the effects observed in vitro may directly contribute to increase the accessibility of factors that promote transcriptional upregulation in vivo In contrast, heterochromatin has the property to remain condensed during interphase and is characterized by histone hypoacetylation and methylation at Lys9 of histone H3 (H3K9me2). H3K9 methylation is enriched at pericentric heterochromatin|, where it specifically interacts with heterochromatin protein 1 (HP1). HP1 binding to methylated Lys9 is proposed to serve as a scaffold to promote the formation of compact chromatin. Histone H3 lysine 27 methylation (H3K27me3), which is set by the Polycomb system, has also been implicated in the formation of repressive chromatin domains, as it inversely correlates with gene activity. Unlike the localized patterns of active marks, H3K27me3 spreads over larger regions harboring many target genes. The exact mechanism of repression remains unclear but may involve the recruitment of components of the Polycomb repressive complex 1 (PRC1). PRC1 core components have been shown to promote the compaction of nucleosomal arrays in vitro and to mediate long-range interactions in vivo, indicating that a change in chromatin organization could account for its repressive activity. Thus, it is generally assumed that repressive histone modifications form compact structures that reduce the accessibility for DNA-binding proteins such as transcription factors and the transcriptional machinery, whereas permissive marks would favor activator binding by exposing the DNA template. This study provide a genome-wide view of DNA accessibility in Drosophila using a modified DNA methylase accessibility assay (MeDIP footprint). Relating accessibility to a comprehensive analysis of histone modifications, different levels of chromatin structure were identified associated with distinct modification patterns. Gene-rich regions are embedded in generally accessible chromatin. Notably, DNA is highly exposed at active promoters, yet no appreciable differences were detected in accessibility between transcribed and silent intragenic regions, suggesting similar degrees of chromatin compaction outside of cis-regulatory regions. However, it was found that early-replicating regions, especially zones of replication initiation, are generally more accessible than late-replicating regions. This suggests that an open chromatin conformation that exposes DNA is not only a feature of active promoters but also of active origins of replication. In contrast, large chromosomal regions covered with H3K27me3 are least accessible for M.SssI methylation. This result supports a model of Polycomb-dependent chromatin compaction as a mode of repression of target genes. In contrast, regions marked with H3K9 methylation do not show reduced methylation footprinting, arguing for a mode of repression that, at the level of DNA accessibility, is distinct from that of the Polycomb pathway (Bell, 2010).

This genome-wide study provides a comprehensive view of DNA accessibility relative to a set of other chromosomal measures. The results suggest complex levels of chromatin organization delineated by distinct patterns of post-translational histone modifications and DNA replication (Bell, 2010).

Concerted epigenetic signatures inheritance at PcG targets through replication

Polycomb group of proteins (PcG), by controlling gene silencing transcriptional programs through cell cycle, lock cell identity and memory. Recent chromatin genome-wide studies indicate that PcG targets sites are bivalent domains with overlapping repressive H3K27me3 and active H3K4me3 marked domains. During S phase, the stability of epigenetic signatures is challenged by the replication fork passage. Hence, specific mechanisms of epigenetic inheritance might be provided to preserve epigenome structures. Recently, a critical time window before replication was identified, during which high levels of PcG binding and histone marks on BX-C PRE target sites set the stage for subsequent dilution of epigenomic components, allowing proper transmission of epigenetic signatures to the next generation. This study has extended this analysis to promoter elements, showing the same mechanism of inheritance. Furthermore, to gain insight into the inheritance of PREs bivalent marks, dynamics were analyzed of H3K4me3 deposition, a mark that correlates with transcriptionally active chromatin. Likewise, an early S-phase enrichment of H3K4me3 mark was found preceding the replication-dependent dilution. This evidence suggests that all epigenetic marks are inherited simultaneously to ensure their correct propagation through replication and to protect the 'bivalency' of PREs (Lanzuolo, 2012).

In Drosophila, genome-wide ChIP-seq and transcriptional analysis, in parallel with the detection of transcription start sites (TSS), revealed new features of Polycomb distribution along the Drosophila genome. One particularly clear feature is that Polycomb often targets TSSs with a stalled RNAPolII. These sites are also enriched in H3K4me1/me2, and these specific signatures at TSSs might serve transcriptional pausing of key developmental genes. Although the H3K27me3 and H3K4me3 marks do not generally coexist in Drosophila, these transcriptionally paused promoters could be functionally considered as the fly analogs of the 'bivalent domains' in mammals, which represent poised states for lineage-specific activation of developmental genes (Enderle, 2011). This was further confirmed in S2 cells, where repressed PcG targets show occupancy of repressive marks in combination with active marks, general transcription factors and RNA PolII. It was previously shown that the inheritance of distinct repressive marks at BX-C PREs during replication share a common timeframe (Lanzuolo, 2012). In particular, it was found that the H3K9me3 repressive mark, present on PREs, shows a trend similar to H3K27me3, being enriched in early S phase and subsequently diluted in late S phase, when PRE replication takes place. This result is in agreement with the evidence that H3K9me3 histone mark is controlled by PcG proteins and suggests that repressive epigenetic signatures are simultaneously inherited during replication. This study analyzed the S-phase dynamic of H3K4me3 deposition, a mark that correlates with transcriptionally active chromatin. As expected, the presence of H3K4me3 mark at PREs was confirmed, although at lower levels compared with H3K27me3. Surprisingly, quantification of H3K4me3 enrichment in different S-phase fractions revealed that the active mark deposition follows a similar tendency compared with H3K27me3 and H3K9me3 marks. This dynamic is specific for PREs, because H3K4me3 levels on the transcriptionally active and early replicating Gapdh promoter show a different trend, being diluted from G₁ to early S phase. Altogether, these results indicate that all three epigenetic marks responsible for the bivalent transcriptional potential of PREs are inherited at the same time to preserve their epigenetic state (Lanzuolo, 2012).

These findings determine the early S phase as the critical time point for the Polycomb cell memory system integrating recent observations. It is suggested that PcG complex's binding and enrichment for all repressive and active histone marks that determine the 'epigenetic bivalency' of PcG bound elements are uncoupled from and precede PcG targets replication, when epigenetic signatures are redistributed on daughter strands. This time-dependent dynamics would allow the local conservation of the epigenetic structures through DNA replication and is necessary for the inheritance of the epigenome (Lanzuolo, 2012).

PcG-mediated higher-order chromatin structures modulate replication programs at the Drosophila BX-C

Polycomb group proteins (PcG) exert conserved epigenetic functions that convey maintenance of repressed transcriptional states, via post-translational histone modifications and high order structure formation. During S-phase, in order to preserve cell identity, in addition to DNA information, PcG-chromatin-mediated epigenetic signatures need to be duplicated requiring a tight coordination between PcG proteins and replication programs. However, the interconnection between replication timing control and PcG functions remains unknown. Using Drosophila embryonic cell lines, this study found that, while presence of specific PcG complexes and underlying transcription state are not the sole determinants of cellular replication timing, PcG-mediated higher-order structures appear to dictate the timing of replication and maintenance of the silenced state. Using published datasets it was shown that PRC1, PRC2, and PhoRC complexes differently correlate with replication timing of their targets. In the fully repressed BX-C, loss of function experiments revealed a synergistic role for PcG proteins in the maintenance of replication programs through the mediation of higher-order structures. Accordingly, replication timing analysis performed on two Drosophila cell lines differing for BX-C gene expression states, PcG distribution, and chromatin domain conformation revealed a cell-type-specific replication program that mirrors lineage-specific BX-C higher-order structures. This work suggests that PcG complexes, by regulating higher-order chromatin structure at their target sites, contribute to the definition and the maintenance of genomic structural domains where genes showing the same epigenetic state replicate at the same time (Lo Sardo, 2013).

The epigenome in its overall complexity, including covalent modifications of DNA and histones, higher-order chromatin structures and nuclear positioning, influences transcription and replication programs of the cell. It is well known that timing of DNA replication is correlated with relative transcription state, in particular transcriptionally active genes tend to replicate early and inactive genes tend to replicate late. However, in recent years, genome-wide analyses revealed several exceptions to this rule. These and other evidence suggested that the transcriptional potential of chromatin, expressed as histone modifications and transcription factors binding (rather than the process of transcription per se) is most closely related to replication timing. A recent work in Drosophila has shown that the selection and the timing of firing of replication origins are associated with distinct sets of chromatin marks and DNA binding proteins (Eaton, 2011). This reinforces previous works showing that mutation, overexpression, depletion or tethering of chromatin modifying proteins to specific loci in yeast, Drosophila and vertebrates determines changes in replication timing locally or/and at a global level. In mammals, it has been suggested that higher-order chromatin structures more than basal epigenome modifications better correlate with replication timing profiles. Although several proteins have been reported to control higher-order chromatin structure formation, their role in replicon structure and replication timing regulation remains to be elucidated. Among these, cohesins have been shown to co-localize with ORC binding sites and to influence replication origin choice and density through the regulation of specific chromatin loops. Previously, it has been reported that PcG proteins are key regulators of higher-order chromatin structures and that condensins complex components and Topoisomerase II take part in PRE and BX-C silencing functio. Moreover, depletion of the mammalian PC homologue M33 determines a switch of the INK4a/ARF locus replication timing, suggesting a role for PcG proteins in the regulation of replication programs at their targets (Lo Sardo, 2013).

However, the interplay between PcG-mediated silencing, higher-order structures and control of replication timing in Drosophila has not been elucidated. This issue has been addressed on a genome-wide level; H3K27me3 enriched and PRC2 bound sequences were found to replicate later than their unbound counterparts. Surprisingly, the same is not true for PRC1 or PhoRC target sites, where the binding of PcG proteins does not significantly correlate with genome-wide replication timing distributions, highlighting a difference between PRC1, PhoRC and PRC2 complexes at a genome-wide scale. Notably, replication timing is more correlated to PRC1 binding at transcribed TSSs than at silenced TSSs (Lo Sardo, 2013).

To investigate the possible contribution of PRC1 and other PcG complexes at repressed genes, the functional interplay between PcG-dependent epigenetic signatures and maintenance of replication programs were analyzed at one of the major PcG targets: the Drosophila BX-C. After depletion of single PcG proteins in S2 cells, reactivation was found of BX-C genes and their related PREs. Interestingly, depletion of PHO protein causes only a mild effect on homeotic genes transcription, although this protein has been reported to be required for recruitment of PRC1 and 2. This suggests that multiple additional mechanisms of recruitment, such as ncRNAs or other protein partners, may act simultaneously at PcG target loci, as described particularly in mammalian cells. Interestingly, also 3C analysis in single PcG depleted cells reveals a different response to E(z) depletion with respect to PC and PHO depletions. In particular, in PC and PHO depleted cells no change or the small reduction of some PRE-PRE and PRE-promoter BX-C interactions were seen, while in E(z) depleted cells even an increase in specific crosslinking frequencies for some interactions was observed. Moreover, both 3C and replication timing analysis in single PcG depleted cells show that transcription per se cannot dramatically perturb the BX-C higher-order structures neither change the timing of replication. This result is in agreement with recent findings in mammalian cells showing that spatial chromatin organization and replication timing are not a direct consequence of transcription (Lo Sardo, 2013).

Conversely, simultaneous depletion of components of the three major PcG complexes (PhoRC, PRC2 and PRC1) determines major changes in BX-C transcription as well as in higher-order structure and an anticipation in replication timing, suggesting that PcG proteins act synergistically on three-dimensional structures and on replication program maintenance. In line with these findings, in recovered cells, BX-C topological structure and PRE replication timing are indistinguishable from controls, suggesting that the observed variations are not sufficient to determine a stable epigenetic switch. In this context the more stable contacts might hamper an irreversible disruption of the three-dimensional BX-C structure (Lo Sardo, 2013).

These findings were further confirmed by the comparison of two different cell lines: S2 and S3 that differ for their embryonic origin. Previous studies have shown that in S3 cells, active transcription of AbdB is associated with different topological conformation of the locus, where AbdB gene and its regulative PREs are topologically separated from the other repressed and clustered epigenetic elements of the locus. This study found that distinct chromatin structures in S2 and S3 are associated with different replication timings, thus confirming that these epigenetic parameters vary in parallel (Lo Sardo, 2013).

This analysis, in line with recent observations, indicates that the genome may be organized into distinct structural and functional domains in which DNA regions that stay together replicate together as a stable unit for many cell generations irrespective of single gene transcription state. It was shown that major adjustments of chromatin higher-order structure and replication program are necessary for a correct differentiation and are required for reprogramming of cell identity. The high stability of higher-order chromatin structures and replication programs can explain one of the underlying molecular basis counteracting cellular reprogramming and representing an epigenetic barrier and PcG complexes may play an important role in the maintenance of this barrier. The data show that correct levels of PcG components can fully restore silencing, higher-order structures and late replication timing at derepressed BX-C gene loci. Of course, additional functions may be involved in the maintenance of these epigenetic parameters either at the BX-C and in the rest of the genome. For example, other factors involved in the regulation of higher-order chromatin structure, including the insulator CTCF protein, condensin complex subunits and Topoisomerase II, were shown to have a role in PcG-mediated gene silencing function. Interestingly, Topoisomerase II has been shown to be required for a global resetting of replicon organization in the context of somatic cell reprogramming. Hence, a deeper understanding of the functional interplay between epigenetic mechanisms modulating the stability of higher-order chromatin structure and replication program will be crucial to unravel the molecular basis (Lo Sardo, 2013).

Interactions among Polycomb domains are guided by chromosome architecture

Polycomb group (PcG) proteins bind and regulate hundreds of genes. Previous evidence has suggested that long-range chromatin interactions may contribute to the regulation of PcG target genes. This study adapted the Chromosome Conformation Capture on Chip (4C) assay to systematically map chromosomal interactions in Drosophila melanogaster larval brain tissue. The results demonstrate that PcG target genes interact extensively with each other in nuclear space. These interactions are highly specific for PcG target genes, because non-target genes with either low or high expression show distinct interactions. Notably, interactions are mostly limited to genes on the same chromosome arm, and it was demonstrated that a topological rather than a sequence-based mechanism is responsible for this constraint. These results demonstrate that many interactions among PcG target genes exist and that these interactions are guided by overall chromosome architecture (Tolhuis, 2011).

This study successfully adapted the 4C method to systematically map long-range chromatin contacts with limited material from a single fly tissue. With this method, interactions were detect between the ANT-C and BX-C in central brain. This observation is in good agreement with earlier microscopic reports, underscoring the strength of the method. Importantly, it was further demonstrated that not only the two Homeotic gene clusters, but also many other PcG target genes interact, suggesting that long-range chromatin contacts between PcDs are common in central brain tissue. The control fragments (wntD, CG5107, Crc, and RpII140), which do not reside in PcDs, have interactions that are distinct from PcDs, emphasizing the specificity of the findings (Tolhuis, 2011).

PcG targets show a strong preference for interaction with other PcG targets, suggesting that PcG proteins help to establish these interactions. This is in line with earlier in vivo and in vitro results that indicated that PcG proteins can keep certain DNA sequences together. However, interactions among PcG target genes are constrained by overall chromosome architecture, because the data demonstrate that loci need to be on the same chromosome arm for efficient interaction (Tolhuis, 2011).

Discrete interaction domains (DIDs) range in size from 6 to 600 kb, with an average of ~170 kb. Thus, highly local strong enrichments are found as well as moderate enrichments over larger regions, which may reflect different types of long-range interactions. It is emphasized that interactions as detected by 4C technically represent events of molecular proximity of DNA sequences, and not necessarily physical binding. Based on the current data it is therefore not possible to identify within the DIDs sequence elements that may mediate direct contact with other DIDs. It is conceivable that contacts between PcDs may occur at any position within the PcDs; if PcG protein complexes have an intrinsic propensity to aggregate, as has been observed in vitro, then large PcDs may have a higher chance of interacting with each other due to their larger ‘sticky’ surface area (Tolhuis, 2011).

The 4C method is a cell population based assay that only detects the most frequent interactions in the population. Previous 4C studies in mammalian cells suggested an extensive network of long-range interactions. The current data also suggests an extensive network among interacting PcDs. However, microscopic studies in mammalian cells revealed that specific long-range interactions occur only in a small proportion of the cells. Likewise, only a proportion of D. melanogaster cells show contacts between the two Homeotic clusters. Therefore, 4C data have to be carefully interpreted, because the identified interactions are in part stochastic and do not all occur simultaneously. As a consequence, it is not known how many PcDs interact in a single cell, but the most common interactions are known in the population of larval brain cells (Tolhuis, 2011).

Interphase chromosomes in most eukaryotes occupy distinct 'territories' inside the nucleus, with only a limited degree of intermingling. Although some studies have reported interactions between some loci that are on two different chromosomes (interchromosomal), unbiased 4C mapping in mouse tissues has indicated that interactions within the same chromosome (intrachromosomal) occur much more frequently than between different chromosomes. In addition, a recent genome-wide map of chromatin interactions in human cells showed that intrachromosomal interactions occur with higher frequency than interchromosomal contacts. The current data are in agreement with these observations, and show that most interactions are even limited to single chromosome arms, at least in D. melanogaster larval brain. Since the experiments indicate that a topological mechanism prevents interactions between the two arms of a chromosome, it is proposed that each arm (rather than the chromosome as a whole) forms a distinct territory. This is consistent with early microscopy studies of chromosome architecture in D. melanogaster, which suggested that chromosome arms are units of spatial organization (Tolhuis, 2011).

What topological mechanism may limit contacts between the two chromosome arms? About ~16 Mbp of pericentric heterochromatin are located in between the two euchromatic arms of chromosome 3. This heterochromatin region could act as a long spacer and prevent efficient interactions between DNA fragments that are located on either chromosome arm. However, the data show that interactions within one arm can span even longer distances, such as between Ptx1 and grn (~22.7 Mbp). Another explanation may be that the pericentric regions of all chromosomes assemble into a nuclear compartment, called chromocenter. This large structure could physically obstruct interactions between chromosome arms (Tolhuis, 2011).

Previous reports have demonstrated that certain PcG-bound PREs can pair in trans (i.e. when they are located on different chromosomes). First, a Fab7 PRE-element integrated on the X chromosome (Fab-X) was often found in close spatial proximity to the endogenous Fab-7 in the BX-C (chromosome 3R), although this phenomenon appears to be tissue-specific and dependent on the transgene integration site. Second, a microscopy assay based on Lac repressor/operator recognition showed that the Mcp PRE-element is able to pair with copies of that same element inserted at remote sites in the genome either in cis (i.e. when they are located on the same chromosome) or in trans. The 4C experiments also identified cases of trans interactions between endogenous loci, although they occur with low frequency (approximately 5% occurs in trans) (Tolhuis, 2011).

Although rare, such interactions between loci on different chromosome arms are of interest, because they indicate that the topological constraints imposed by chromosome architecture can in principle be overcome. In mammalian cells, there is evidence that the relative position of a gene locus within its chromosome territory (CT) influences its ability to form either cis- or trans-interactions. Peripheral regions of mammalian CTs intermingle their chromatin, which may allow for interactions between chromosomes. Indeed, more trans contacts are identified by 4C using a bait that often resides in the CT periphery compared to a bait located in the interior of a CT. Likewise, activation of the HoxB gene cluster during differentiation coincides with relocation away from its CT interior, and the active HoxB1 gene more frequently contacts sequences on other chromosomes compared to the inactive gene. In line with this, a varying degree of trans interactions are observed among eight bait sequences, suggesting distinct capacities to contact chromatin on other chromosomes. The trh gene has the strongest capacity to contact other chromosome arms. Interestingly, trh is located within 500 Kbp of the telomere of chromosome 3L, and 5 out of 6 contacts in trans occur within 500 Kbp of other telomeres. Thus, interactions between chromosome arms may be possible if loci are favorably positioned on the edge of chromosome (arm) territories, which could be the case for telomeric sequences in larval brain cells (Tolhuis, 2011).

The experiments with strain In(3LR)sep showed dramatic changes in interactions, such as loss of contacts between the Homeotic gene clusters and gain of contacts with other PcDs. Despite these changed interactions, no convincing evidence was found for global gene expression alterations on the In(3LR)sep chromosome. This raises the question: how relevant are long-range chromatin contacts between PcG-target genes for regulation of expression?

The lack of detectable expression changes may indicate that long-range interactions have only quantitatively subtle effect on the regulation of gene expression. Nevertheless, such subtle effects on gene expression could be very important for long-term viability and species survival. In(3LR)sep animals do suffer from an overall reduced viability during several stages of development, which may indicate a generally reduced fitness, possibly due to a dimly altered regulation of gene expression (Tolhuis, 2011).

Alternatively, PcG gene regulation may not be affected in strain In(3LR)sep, because Abd-B and Antp, although they no longer interact with each other, still prefer to interact with other PcDs, suggesting that it is not relevant which PcDs interact. In such a model, the complement of all interactions contributes to PcG-mediated gene silencing in a population of cells (Tolhuis, 2011).

Finally, it is interesting to note that over ~100 million years of evolution of the Drosophila genus, exchange of genes between chromosome arms has been rare despite extensive rearrangements within each arm. Chromosome arm territories ensure that genes within a single arm are relatively close compared to genes on other arms, which may have resulted in an increased chance of rearrangements within one arm. Alternatively, the importance of long-range interactions among sets of genes, which are topologically limited to the same arm, may have contributed to the selective pressure that has led to this remarkable conservation of the gene complement of each chromosome arm (Tolhuis, 2011).

Polycomb purification by in vivo biotinylation tagging reveals cohesin and Trithorax group proteins as interaction partners

The maintenance of specific gene expression patterns during cellular proliferation is crucial for the identity of every cell type and the development of tissues in multicellular organisms. Such a cellular memory function is conveyed by the complex interplay of the Polycomb and Trithorax groups of proteins (PcG/TrxG). These proteins exert their function at the level of chromatin by establishing and maintaining repressed (PcG) and active (TrxG) chromatin domains. Past studies indicated that a core PcG protein complex is potentially associated with cell type or even cell stage-specific sets of accessory proteins. In order to better understand the dynamic aspects underlying PcG composition and function, an inducible version of the biotinylation tagging approach was established to purify Polycomb and associated factors from Drosophila embryos. This system enabled fast and efficient isolation of Polycomb containing complexes under near physiological conditions, thereby preserving substoichiometric interactions. Novel interacting proteins were identified by highly sensitive mass spectrometric analysis. Many TrxG related proteins were found, suggesting a previously unrecognized extent of molecular interaction of the two counteracting chromatin regulatory protein groups. Furthermore, this analysis revealed an association of PcG protein complexes with the cohesin complex and showed that Polycomb-dependent silencing of a transgenic reporter depends on cohesin function (Strübbe, 2011).

Combinations of one-step capture with streptavidin, low stringency washes, specific elution, and detection of peptides using a highly sensitive LTQ-FT-ICR mass spectrometer enabled the identification of even labile and transient interactions. It has been well recognized that PcG and TrxG proteins exert their counteracting activities at the level of chromatin by employing various biochemical activities directed against histones, like methylation, acetylation, and chromatin remodeling. Indeed, this study reveals a substantial number of Pc-interacting proteins implicated in TrxG action. The genes encoding for Rdx, Ebi, CG1845, Rad21, and Fs(1)h have been shown genetically to belong to the TrxG suppressing PcG mutant phenotypes and activating HOX gene expression, for example. Additionally, Pp1-87B has been found to interact with Trx or its homologue MLL. These data indicate that Pc and specific members of the TrxG may physically cooperate to maintain the on/off state of genes (Strübbe, 2011).

So far, the DNA-binding proteins Zeste, Gaf, Pho, Dsp1, Sp1/Klf family members, Psq, and Grh have been connected to PcG function on the basis of genetic interactions, biochemical copurification, functional assays, and/or colocalization on PREs. This study found direct biochemical interactions of Grh and Pho with Pc. Moreover, a Pc-interacting protein called Fs(1)h was identified that might, as well, contribute to recruitment of PRCs to chromatin. Fs(1)h interacts strongly with Ubx, trx, and ash1 mutations and leads to homeotic phenotypes when overexpressed. Fs(1)h is essential for development and conserved in mammals. Whether Pc is recruited by Fs(1)h or opposes its function in gene activation needs to be established. Beside the aforementioned DNA-binding proteins, Enok is a Pc interactor with a putative DNA-binding domain. Enok forms part of the MYST domain family of histone acetyl transferases (HATs), and mutants with defects in the HAT domain show retarded development and pupal lethality (Strübbe, 2011).

Enok's HAT domain is conserved in the vertebrate Moz/Morf proteins. They typically form complexes comprising one protein per BRPF-, ING-, and EAF family member. In Drosophila, a Moz/Morf like complex may consist of Enok, CG1845 (homologue of Brpf1-3), and Eaf6 as all these proteins copurified with Pc and were detected with high confidence. Moz and Brpf1 are TrxG proteins required for HOX gene expression in vertebrates. Although MYST-domain-containing HATs have generally been associated with transcriptional activation, there are also examples with a link to HOX gene repression in Drosophila (Strübbe, 2011).

This work uncovered a connection between Pc and the cohesin complex. Cohesin has been described in detail for its roles in mitosis and meiosis, embracing sister chromatids in mitotic cells. Interestingly, mutations in Ph-p, Psc, and Pc have been reported to result in chromosome missegregation phenotypes in embryos. Besides its traditional role in sister chromatid cohesion, cohesin has also been implicated in both activation and repression of transcription. Furthermore, mutations in the Rad21 subunit of the cohesin complex strongly enhance TrxG and suppress PcG loss of function phenotypes. Pc and cohesins are not colocalized on salivary gland chromatin, and removal of cohesin does not affect Pc binding. It cannot be ruled out that Pc is needed for recruitment of cohesin, however. For example, chromatin binding of cohesin in S. pombe depends on formation of heterochromatin, requiring another chromo domain protein, HP1 (Strübbe, 2011).

A hallmark of PcG repression in flies is pairing-sensitive silencing (PSS), depending on pairing of homologous chromosomes in interphase chromatin. It is known that multiple copies of a transgenic PRE interact with each other if inserted on the same or even on different chromosomes. Because cohesin plays a role in pairing of homologous chromosomes in meiosis and has been suggested to facilitate long-range DNA interactions, it may also facilitate PRE pairing. The transgenic reporter for PSS used in this study only showed PRE-dependent silencing upon PRE pairing. The observation that cohesin mutant alleles reduce PSS supports a model in which cohesins contribute to PRE pairing in interphase chromatin. The identification of Pc-interacting proteins was made possible by employing the in vivo biotinylation system combined with highly sensitive mass spectrometric analysis, thereby preserving near physiological conditions for protein purification. The identification of substoichiometric levels of interacting proteins shows that in vivo biotinylation was effective in capturing even weakly or underrepresented associated proteins. Inducible biotinylation tagging is currently limited to the use of Gal4 drivers that trigger biotinylation well above the background levels. Generation of libraries of different UAS-BirA transgenic lines with less leaky expression and flies carrying BirA under direct control of tissue-specific promoters will further improve and expand this tool, making it a versatile system for proteomic and genomic studies in specialized cell types. As a major advantage over tissue-specific expression of tagged bait proteins, biotin tagging allows to express the bait protein under control of endogenous promoter sequences, whereas the induction of the BirA ligase can be independently induced via the Gal4/UAS system avoiding bait protein misexpression artifacts. This work opens the perspective for tissue-specific applications, potentially enabling a systems analysis on how protein networks can control subsets of genes in specialized cells (Strübbe, 2011).

The role of the histone H2A ubiquitinase Sce in Polycomb repression

Polycomb group (PcG) proteins exist in multiprotein complexes that modify chromatin to repress transcription. Drosophila PcG proteins Sex combs extra (Sce; dRing) and Posterior sex combs (Psc) are core subunits of PRC1-type complexes. The Sce:Psc module acts as an E3 ligase for monoubiquitylation of histone H2A, an activity thought to be crucial for repression by PRC1-type complexes. This study created an Sce knockout allele and showed that depletion of Sce results in loss of H2A monoubiquitylation in developing Drosophila. Genome-wide profiling identified a set of target genes co-bound by Sce and all other PRC1 subunits. Analyses in mutants lacking individual PRC1 subunits reveals that these target genes comprise two distinct classes. Class I genes are misexpressed in mutants lacking any of the PRC1 subunits. Class II genes are only misexpressed in animals lacking the Psc-Su(z)2 and Polyhomeotic (Ph) subunits but remain stably repressed in the absence of the Sce and Polycomb (Pc) subunits. Repression of class II target genes therefore does not require Sce and H2A monoubiquitylation but might rely on the ability of Psc-Su(z)2 and Ph to inhibit nucleosome remodeling or to compact chromatin. Similarly, Sce does not provide tumor suppressor activity in larval tissues under conditions in which Psc-Su(z)2, Ph and Pc show such activity. Sce and H2A monoubiquitylation are therefore only crucial for repression of a subset of genes and processes regulated by PRC1-type complexes. Sce synergizes with the Polycomb repressive deubiquitinase (PR-DUB) complex to repress transcription at class I genes, suggesting that H2A monoubiquitylation must be appropriately balanced for their transcriptional repression (Gutiérrez, 2012).

This study analyzed how PRC1 regulates target genes in Drosophila to investigate how the distinct chromatin-modifying activities of this complex repress transcription in vivo. Because H2A monoubiquitylation is thought to be central to the repression mechanism of PRC1-type complexes, focus was placed on the role of Sce. The following main conclusions can be drawn from the work reported in this study. First, in the absence of Sce, bulk levels of H2A-K118ub1 are drastically reduced but the levels of the PRC1 subunits Psc and Ph are undiminished. Sce is therefore the major E3 ligase for H2A monoubiquitylation in developing Drosophila but is not required for the stability of other PRC1 subunits. Second, PRC1-bound genes fall into two classes. Class I target genes are misexpressed if any of the PRC1 subunits is removed. Class II target genes are misexpressed in the absence of Ph or Psc-Su(z)2 but remain stably repressed in the absence of Sce or Pc. At class II target genes, Ph and the Psc-Su(z)2 proteins work together to repress transcription by a mechanism that does not require Sce and Pc and is therefore independent of H2A monoubiquitylation. Third, removal of the Ph, Psc-Su(z)2 or Pc proteins results in imaginal disc tumors that are characterized by unrestricted cell proliferation. However, removal of Sce does not cause this phenotype, suggesting that this tumor suppressor activity by the PcG system does not require H2A monoubiquitylation. Finally, these analyses reveal that PRC1 subunits are essential for repressing the elB, noc, dac and pros genes outside of their normal expression domains in developing Drosophila. This expands the inventory of developmental regulator genes in Drosophila for which PcG repression has been demonstrated in a functional assay (Gutiérrez, 2012).

In the Sce^33M2 allele Arg65 is mutated to Cys, but this mutant Sce protein is undetectable and therefore does not appear to be stable in vivo. The crystal structure of the Ring1B-Bmi1 complex provides a molecular explanation for this observation: the Arg70 residue in Ring1B that corresponds to Arg65 in Sce is thought to be critical for interaction with Bmi1. A likely scenario therefore is that the Sce^Arg65Cys protein in Drosophila is unstable and is degraded because it is unable to associate with Psc or its paralog Su(z)2. Interestingly, removal of Sce protein has no detectable effect on the levels of the Psc and Ph proteins. Psc is therefore stable in the absence of its binding partner Sce. This is in contrast to the situation in mice in which Ring1B mutant ES cells show a drastic reduction in the levels of the Ring1B partner protein Bmi1 and its paralog Mel18 (Pcgf62) and also a reduction in the levels of Mph2 (Phc2) and Mpc2 (Cbx4) (Leeb, 2007). The interdependence between PRC1 subunits for protein stability is therefore different in mammals and Drosophila (Gutiérrez, 2012).

Reconstitution of the Drosophila PRC1 core complex in a baculovirus expression system suggests that Sce is important for complex stability. At present, it is not know whether the Psc, Ph and Pc proteins still form a complex in vivo in the absence of Sce. It is currently unknown whether Psc, Ph and Pc are still bound to all PRC1 target genes in the absence of Sce. However, the finding that class II genes remain repressed in the absence of Sce, even though their repression depends on Psc-Su(z)2 and Ph, argues against a crucial role of Sce in the targeting of these other PRC1 subunits to these genes. Interestingly, the repression of all class II target genes analyzed in this study always requires both the Ph and the Psc-Su(z)2 proteins. A possible explanation for this observation is that Ph and Psc-Su(2) still form a PRC1 subcomplex in the absence of Sce and that this complex is fully functional to repress class II target genes. Alternatively, it is possible that Ph and Psc-Su(z)2 repress class II target genes as components of as yet uncharacterized complexes that are distinct from PRC1 and Drosophila dRing-associated factors (dRAF) complex (Gutiérrez, 2012).

In vitro, Psc and Su(z)2 proteins compact nucleosome templates, inhibit nucleosome remodeling by SWI/SNF complexes and repress transcription on chromatin templates. The observation that repression of class II target genes requires Psc-Su(z)2 and Ph but not Pc and Sce supports the idea that the chromatin-modifying activities of Psc-Su(z)2 identified in vitro are the main mechanism by which PRC1 represses these genes. Previous structure/function analyses in Drosophila showed that the same domains of the Psc protein responsible for chromatin compaction and remodeling inhibition in vitro are crucial for HOX gene repression in vivo. Chromatin modification by Psc and Su(z)2 is therefore also crucial for repression of class I target genes. Regulation of the class I target gene en further illustrates this point. In some tissues (e.g. in the dorsal hinge region of the wing imaginal disc) repression of en requires all PRC1 core subunits, but in other tissues (e.g. in the pouch of the wing imaginal disc) en remains repressed in the absence of Sce and Pc, and only Psc-Su(z)2 and Ph seem to be crucial to keep the gene inactive. At present, the molecular mechanism of Ph is not well understood. In vitro, Ph protein has the capacity to inhibit chromatin remodeling and transcription but it does so less effectively than Psc. At the target genes analyzed in this study, Ph is required for transcriptional repression wherever Psc-Su(z)2 is required, suggesting that Ph and Psc-Su(z)2 act in concert in this repression. Nevertheless, it is possible that repression of other PRC1 target genes requires a different subset of PRC1 subunits, or that, as in the case of en, the subunit requirement for repression changes depending on the cell type (Gutiérrez, 2012).

In mammals, Ring1B and Ring1A are responsible for the bulk of H2A-K119 monoubiquitylation. Similarly, Sce generates the bulk of H2A-K118 monoubiquitylation in Drosophila, both in tissue culture cells (Lagarou, 2008) and in the developing organism (this study). The requirement for Sce at class I target genes is consistent with the idea that H2A monoubiquitylation of their chromatin is part of the repression mechanism. Repression of a subset of class I genes, namely the HOX genes, also requires the H2A deubiquitinase PR-DUB (Gaytán de Ayala Alonso, 2007; Scheuermann, 2010). Moreover, PR-DUB and Sce strongly synergize to repress HOX genes. Specifically, the phenotype of Sce PR-DUB double mutants suggests that H2A monoubiquitylation becomes ineffective for HOX gene repression if PR-DUB is absent. However, embryos that lack PR-DUB alone show a 10-fold increase in the bulk levels of H2A-K118ub1 and it is estimated that ~10% of all H2A molecules become monoubiquitylated in these animals. How could this conundrum be explained? One possibility is that H2A monoubiquitylation and deubiquitylation at HOX gene chromatin need to be regulated in a precisely balanced manner. However, an alternative explanation considers H2A-K118ub1 levels in the context of ubiquitin homeostasis. In particular, the high H2A-K118ub1 levels in PR-DUB mutants suggest that Sce generates widespread H2A monoubiquitylation at most Sce-bound genes and possibly also elsewhere in the genome, but that in wild-type animals PR-DUB continuously deubiquitylates H2A-K118ub1 at these locations and thereby recycles ubiquitin. The observation that PR-DUB is widely co-bound with Sce, not only at HOX but also at many other class I and class II target genes, is consistent with this idea. It is tempting to speculate that the widespread H2A monoubiquitylation in PR-DUB mutants sequesters a substantial fraction of the pool of free ubiquitin. It is therefore possible that removal of PR-DUB effectively depletes the pool of free ubiquitin in the nucleus to an extent that H2A monoubiquitylation at HOX target genes becomes inefficient and, consequently, their repression can no longer be maintained. According to this model, the crucial function of PR-DUB would not be the deubiquitylation of H2A-K118ub1 at HOX genes but rather at class II target genes and elsewhere in the genome where Sce 'wastefully' monoubiquitylates H2A (Gutiérrez, 2012).

Enhancer of Split Gene Complex: architecture and coordinate regulation by Notch, Cohesin and Polycomb group proteins

The cohesin protein complex functionally interacts with Polycomb group (PcG) silencing proteins to control expression of several key developmental genes, such as the Drosophila Enhancer of split gene complex [E(spl)-C]. The E(spl)-C contains twelve genes that inhibit neural development. In a cell line derived from central nervous system, cohesin and the PRC1 PcG protein complex bind and repress E(spl)-C transcription, but the repression mechanisms are unknown. The genes in the E(spl)-C are directly activated by the Notch receptor. This study shows that depletion of cohesin or PRC1 increases binding of the Notch intracellular fragment (NICD) to genes in the E(spl)-C, correlating with increased transcription. The increased transcription likely reflects both direct effects of cohesin and PRC1 on RNA polymerase activity at the E(spl)-C, and increased expression of Notch ligands. By chromosome conformation capture this study found that the E(spl) C is organized into a self-interactive architectural domain that is co-extensive with the region that binds cohesin and PcG complexes. The self-interactive architecture is formed independently of cohesin or PcG proteins. It is posited that the E(spl)-C architecture dictates where cohesin and PcG complexes bind and act when they are recruited by as yet unidentified factors, thereby controlling the E(spl)-C as a coordinated domain (Schaaf, 2013).

These studies investigated the regulation of the E(spl)-C complex by cohesin, PRC1, and the Jarid2-Aebp2-containing PRC2 and promotes H3K27 trimethylation on H2Aub nucleosomes. Jarid2, Aebp2 and H2Aub thus constitute components of a positive feedback loop establishing H3K27me3 chromatin domains (Kalb, 2014).

Nucleosomes constitute the building blocks of eukaryotic chromosomes. They consist of a core of histone proteins around which DNA is wrapped in two helical turns. The post-translational modification of histones is a key step for the regulation of diverse processes that occur on nucleosomal DNA. Specific histone modifications often decorate arrays of nucleosomes that comprise many kilobases of DNA, but how such extended stretches of chromatin become modified is not well understood. A paradigm for a long-range chromatin-modification mechanism is transcriptional repression by Polycomb protein complexes. The Polycomb system generates two distinct histone modifications: methylation of K27 in histone H3 and monoubiquitination of K119 in histone H2A in vertebrates and of the corresponding K118 in Drosophila H2A. Polycomb repressive complex 2 (PRC2) catalyzes mono-, di- and trimethylation at H3 K27. At inactive Polycomb-target genes, H3 K27 trimethyl marks typically decorate nucleosomes across the entire upstream, promoter and coding region and are essential for repression of these genes. The H3K27me3 modification is recognized by Polycomb, a subunit of the canonical Polycomb repressive complex 1 (PRC1), and is thought to promote PRC1 interaction with chromatin across the entire length of repressed genes. PRC1 has been proposed to repress transcription through chromatin compaction and also through its ubiquitin-ligase activity for H2Amonoubiquitination. To gain insight into the function of H2Aub, this study set out to identify interactors of this modification (Kalb, 2014).

Arrays of four nucleosomes (referred to as oligonucleosomes) were reconstituted with recombinant Drosophila or Xenopus histones and monoubiquitinated H2A in these templates, using appropriate recombinant enzymes. Drosophila monoubiquitinated H2AK118 (H2AK118ub) oligonucleosomes and the corresponding unmodified oligonucleosome control template were used for affinity purification of H2AK118ub-binding proteins from Drosophila embryo nuclear extracts. In parallel, Xenopus monoubiquitinated H2A K119 (H2AK119ub) and unmodified control oligonucleosomes were used to identify vertebrate H2AK119ub interactors in nuclear extracts from mouse embryonic stem cells. In both experiments, quantitative MS analyses identified PRC2 subunits as being among the most highly enriched H2Aub interactors. Jarid2 and Aebp2 were the PRC2 subunits showing highest enrichment in both cases (Kalb, 2014).

The identification of PRC2 as an H2Aub interactor in both flies and vertebrates prompted an analysis of PRC2 histone methyltransferase (HMTase) activity on H2Aub nucleosomes. Recombinant human PRC2 containing EED, EZH2, SUZ12 and RBBP4 (referred to as PRC2) and assemblies of the same complex that in addition contained AEBP2 (AEBP2-PRC2), JARID2 (JARID2-PRC2) or both JARID2 and AEBP2 (JARID2-AEBP2-PRC2) were reconstituted. For substrates, Xenopus mononucleosomes were used that were either unmodified or monoubiquitinated at H2A K119, and in all cases western blot analyses were used with antibodies against either monomethylated H3 K27 (H3K27me1) or H3K27me3 to monitor PRC2 activity. A time-course experiment was performed to compare the activity of PRC2 and JARID2-AEBP2-PRC2 on H2A and H2Aub nucleosomes. It was found that, consistently with earlier reports, the catalytic activity of PRC2 alone is largely unchanged on H2Aub nucleosome templates. As expected, inclusion of JARID2 and AEBP2 in PRC2 resulted in stronger activity for H3 K27 methylation on unmodified nucleosome templates. However, a much stronger increase was used in H3K27me3 formation when JARID2-AEBP2-PRC2 was used for HMTase reactions on H2Aub nucleosomes. It was estimated that JARID2-AEBP2-PRC2 trimethylates H3K27 in H2Aub nucleosomes with an efficiency 25-fold higher than that of PRC2. To assess the contributions of JARID2 and AEBP2 to this stimulation of HMTase activity, the catalytic activity was compared of all four forms of PRC2 on H2A and H2Aub nucleosome substrates. JARID2-PRC2 showed higher H3K27 methyltransferase activity than did PRC2 on unmodified nucleosomes, as previously reported, but this was not further increased on H2Aub nucleosomes. In contrast, AEBP2-PRC2 methylated H3K27 in H2Aub nucleosomes with considerably higher efficiency than in unmodified nucleosomes. This suggests that AEBP2 is critical for the specific activation of PRC2 by H2Aub, whereas JARID2 has a more general function in boosting PRC2 HMTase activity, independently of the H2A modification state (Kalb, 2014).

The work reported in this study reveals that Jarid2-Aebp2-containing PRC2 binds to H2Aub nucleosomes and demonstrates that H3K27 trimethylation by this complex is strongly enhanced on H2Aub nucleosomes. This establishes H2Aub, Aebp2 and Jarid2 as components of a positive feedback loop in which H2Aub promotes PRC2 binding and H3K27 trimethylation, and H3K27me3 in turn promotes binding of the canonical PRC1 via the chromodomain of Polycomb. It is currently not clear whether canonical PRC1 indeed has E3 ligase activity for H2Amonoubiquitination or whether this modification is generated only by forms of PRC1 lacking Polycomb. Intriguingly, in embryonic stem cells, the PRC1-type complexes PRC1.1 and PRC1.6 were also identified as H2Aub interactors, results suggesting an additional feedback loop for H2A ubiquitination in vertebrates. The positive feedback loop for H3K27me3 formation by H2Aub uncovered in this study provides a rationale for how extended domains of Polycomb-repressed chromatin could be generated in both Drosophila and vertebrates. These findings could explain why H3K27me3 levels at Polycomb-target genes are reduced in mouse embryonic stem cells in which H2AK119ub has been depleted. However, it was previously found that bulk H3K27me3 levels were undiminished in late-stage Drosophila larvae in which bulk H2Aub levels had been depleted, thus suggesting that maintenance of H3K27me3-containing chromatin domains does not strictly depend on H2Aub. The H2Aub-mediated feedback loop may thus primarily be required for the initial formation of H3K27me3 chromatin domains when Polycomb repression is first established during the early stages of embryogenesis (Kalb, 2014).

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.