Xenopus homologs of the Drosophila Polycomb gene and the vertebrate bmi-1 gene have been isolated. XPolycomb and Xbmi-1 genes are expressed in overlapping patterns in the central nervous system of Xenopus embryos. However, XPolycomb is also expressed in the somites, whereas Xbmi-1 is not. The XPolycomb and Xbmi-1 proteins are able to interact with each other via conserved sequence motifs. These data suggest that vertebrate Polycomb group proteins also form multimeric complexes (Reijnen, 1995).

The bmi-1 proto-oncogene can be activated by Moloney murine leukemia proviral insertions in transgenic mice. It encodes a highly conserved nuclear protein of 324 amino acids which belongs to a family of proteins containing a putative new zinc-finger. Another closely related member of this family is the mouse protein Mel-18. Distinct domains of the mouse Bmi-1 protein, including the putative zinc-finger motif, are highly conserved within the much larger PSC protein. The conserved domains between mouse BMI and PSC are also conserved within Suppressor-2 of Zeste ( van Lohuizen, 1991).

The mammalian mel-18/bmi-1 gene products share an amino acid sequence and a secondary structure, including a RING-finger motif, with the Drosophila Polycomb group (Pc-G) gene products PSC and Su(z)2, implying that they represent a gene family with related functions. As Drosophila Pc-G gene products are thought to function as transcriptional repressors by modifying chromatin structure, Mel-18/Bmi-1 might be expected to have similar activities. Mel-18 acts as a transcriptional repressor via its target DNA sequence, 5'-GACTNGACT-3'. Interestingly, this binding sequence is found within regulatory or non-coding regions of various genes, including the c-myc, bcl-2 and Hox genes, suggesting diverse functions of mel-18 as the mammalian homolog of the Pc-G gene. Mel-18 has tumor suppressor activity, in contrast to bmi-1, which has been defined as a proto-oncogene (Kanno, 1995).

Murine mel-18 and closely related bmi-1 are homologous to Drosophilaposterior sex combs and suppressor two of zeste. Mel-18 protein mediates a transcriptional repression via direct binding to specific DNA sequences. To gain further insight into the function of Mel-18, it was inactivated by homologous recombination. Mice lacking mel-18 survive to birth and die around 4 weeks after birth after exhibiting strong growth retardation. Similar to the Drosophila posterior sex combs mutant, posterior transformations of the axial skeleton are reproducibly observed in mel-18 mutants. The homeotic transformations are correlated with ectopic expression of Homeobox cluster genes along the anteroposterior axis in the developing paraxial mesoderm. Surprisingly, mel-18-deficient phenotypes are reminiscent of bmi-1 mutants. These results indicate that the vertebrate Polycomb group genes mel-18 and bmi-1, like Drosophila Polycomb group gene products, might play a crucial role in maintaining the silent state of Homeobox gene expression during paraxial mesoderm development (Akasaka, 1996).

The oncogene bmi-1, which was originally found to be involved in B- and T-cell lymphoma formation encodes a protein with a domain of homology to the Drosophila protein Posterior sex combs (Psc) and its relative Suppressor 2 of Zeste (Su[z]2). The possibility that bmi-1 may play a similar role in vertebrates has been suggested by the finding that mice lacking the bmi-1 gene show posterior transformations of the axial skeleton. Transgenic mice overexpressing Bmi-1 protein show the opposite phenotype, namely a dose-dependent anterior transformation of vertebral identity. The anterior expression boundary of the Hoxc-5 gene is shifted in the posterior direction, indicating that Bmi-1 is involved in the repression of Hox genes. It is proposed that Bmi-1 is a member of a vertebrate Polycomb complex that regulates segmental identity by repressing Hox genes throughout development (Alkema, 1995).

Mice lacking the Bmi-1 gene reveal posterior transformations along the axial skeleton, and transgenic mice overexpressing Bmi-1 display anterior transformations. Whereas the expression boundaries of certain Hox genes are more anterior in the paraxial mesoderm in the absence of Bmi-1 than in wild type embryos, such a shift in expression boundaries is not observed in the neurectoderm. Nevertheless, mice overexpressing the Bmi-1 gene have a second spinal ganglion which degenerates like the first ganglion does in wild types. Bmi-1 is involved in the repression of a subset of Hox genes from different clusters from at least day 9.5 onwards (van der Lugt, 1996).

In Drosophila and mouse, Polycomb group genes are involved in the maintenance of homeotic gene expression patterns throughout development. Skeletal phenotypes are found in mouse compound mutants for two Polycomb group genes bmi1 and M33. Mice deficient for both bmi1 and M33 present stronger homeotic transformations of the axial skeleton as compared to each single Polycomb group mutant, indicating strong dosage interactions between those two genes. These skeletal transformations are accompanied with an enhanced shift of the anterior limit of expression of several Hox genes in the somitic mesoderm. These results demonstrate that in mice the Polycomb group genes act in synergy to control the nested expression pattern of some Hox genes in somitic mesodermal tissues during development (Bel, 1998).

When Pc-G mutant mice are compared, loss of each Pc-G gene shows a unique subset of affected Hox genes. For example, in M33 mutant mice, only anterior shifts for Hoxa3 can be detected and, in some cases, for Hoxc8. mel18 -/-, bmi1 -/- and rae28 -/- mice present a more extensive overlap in affected Hox genes, encompassing one prevertebrae anterior shifts of Hoxa5 and Hoxc8. However, Hoxc6 and Hoxc5 are uniquely affected in bmi1 -/- mice, while Hoxa7 and Hoxd4 are only affected in mel18 -/- mice and Hoxb5 is unaffected in all those mutant mice. In M33 bmi1 double mutant mice, the anterior limit of expression of at least two Hox genes, Hoxc9 and Hoxc8, is significantly more severely shifted as compared to both single mutants, demonstrating the additive effect of Pc-G products in maintaining the boundaries of selected Hox genes. One striking observation is that deleting three gene doses of these Pc-G genes does not increase the derepressive effect of Hoxc8 or Hoxc9 expression; full deficiency of M33 and bmi1 is required to induce extensive ectopic expression of these two genes in mesodermal tissues. Nevertheless, according to the differential expression of Hoxc8 and Hoxc9 in both single mutants, it seems that Hoxc8 is more sensitive to M33 regulation since a one-segment anterior shift is observed in M33 -/- bmi1 +/- (prevertebrae 10), whereas that shift is not visible in bmi1 -/- M33 +/- (prevertebrae 11); reciprocally, Hoxc9 is more sensitive to bmi1 regulation. In contrast, some other Hox genes like Hoxb1, Hoxd4 and Hoxd11 are not affected, either in single or in double mutant mice. This suggests that several murine multimeric Pc-G protein complexes of different composition might exist that differ in their affinities for specific Hox genes. Alternatively, since in mammals Pc-G genes exist as highly related gene pairs (such as mel18/bmi1, Enx1/Enx2, M33/MPc2, hPc1/hPc2 and Hph1/rae28/Hph2), a potential redundancy likely exists. This suggests that the homologous gene complements part of the function and thus maintains the boundaries of expression for a subset of Hox genes. Analysis of double mutant mice for a related pair such as mel18 and bmi1 should clarify the degree of redundancy. Differential effects of Pc-G mutations on Hox gene expression in different tissues like the somitic mesoderm and the neural tube, also suggest that different Pc-G complexes may regulate a specific subset of Hox genes in a tissue-dependent manner (Bel, 1998).

Bmi1-interacting proteins are constituents of a multimeric mammalian Polycomb complex. Bmi, and the closely related Mel18 gene share sequence homology to Drosophila PSC and SU(Z)2. M33 is a murine homolog of Drosophila Polycomb protein. A Bmi binding protein, Mph1 has similarity to Drosophila Polyhomeotic. Although the overall identity of the Mph1 protein to Polyhomeotic is only 31%, Mph1 and PH share significant similarity at the C-terminal region. A C-terminal homology domain shares weak similarity to members of the Ets family of transcription factors. This domain is most similar to the Drosophila protein Lethal(3)malignant brain tumor, Bmi1 and Mph1, as well as the Mel18 and M33 proteins are constituents of a multimeric protein complex in mouse embryos and human cells. A central domain of Bmi1 interacts with the carboxyl terminus of Mph1, a conserved alpha-helical domain in the Mph1 protein is required for its homodimerization. Transgenic mice overexpressing various mutant Bmi1 proteins demonstrate that the central domain of Bmi1 is required for the induction of anterior transformations of the axial skeleton. Bmi1, M33 and Mph1 show an overlapping speckled distribution in interphase nuclei. Localization of Bmi1 to subnuclear domains depends on an intact RING finger (Alkema, 1997).

Two human proteins, HPH1 and HPH2 coimmunoprecipitate and cofractionate with each other and BMI1, a homolog of Drosophila Posterior sex combs. They also colocalize with BMI1 in interphase nuclei of human cultured cells. HPH1 and HPH2 have little sequence homology with each other, except in two highly conserved domains, designated homology domains I and II. They share these homology domains I and II with the Drosophila PcG protein Polyhomeotic. Homology domain II is C terminal in the two human proteins and Polyhomeotic. Homology domain I is located about 230 amino acids upstream of Homology domain II in all three proteins. HPH1, HPH2 and BMI1 show distinct, although overlapping expression patterns in different tissues and cell lines. The highest level of expression of HPH1 is found in thymus, testis and ovary. Low expression levels are detected in spleen, prostate, small intestine, colon and peripheral blood leukocytes. In contrast, HPH2, like BMI1, is expressed ubiquitously. Homology domain II of HPH1 interacts with both homology domains I and II of HPH2. In contrast, homology domain I of HPH1 interacts with homology domain II of HPH2, but not with homology domain I of HPH2. Furthermore, BMI1 does not interact with the individual homology domains. Instead, both intact homology domains I and II need to be present for interactions with BMI1. HPH1 and HPH2 colocalize with BMI1 in large domains in the nuclei of cultured cells. The occurrence of two proteins with homology to PH may also point towards the interesting possibility of the existence of different mammalian multimeric PcG complexes. This idea is supported by the differential expression of HPH1 and of HPH2 and BMI1. This idea is reinforced by the observation that PSC shares many, but not all binding sites with the PcG proteins PC, PH and Polycomblike. This provides flexibility, and it increases the possibilities of regulating a wide range of target genes with only a limited number of components. In conclusion, homology domains I and II are involved in protein-protein interactions and the results indicate that HPH1 and HPH2 are able to heterodimerize (Gunster, 1997).

The bmi-1 oncogene cooperates with c-myc in transgenic mice, resulting in accelerated lymphoma development. Altering the expression of Bmi-1 affects normal embryogenesis. Chimeric LexA-Bmi-1 protein has previously been shown to repress transcription. How Bmi-1 functions in embryogenesis and whether this relates to the ability of Bmi-1 to mediate cellular transformation is unknown. Bmi-1 is able to transform rodent fibroblasts in vitro, providing a system that has allowed its molecular properties to be correlated with its ability to transform cells. Functional domains of Bmi-1 involved in transcriptional suppression were mapped by using the GAL4 chimeric transcriptional regulator system. Deletion analysis shows that the centrally located helix-turn-helix-turn-helix-turn (HTHTHT) motif is necessary for transcriptional suppression whereas the N-terminal RING finger domain is not required. Nuclear localization requires KRMK (residues 230 to 233); the absence of nuclear entry ablates transformation. The subnuclear localization of wild-type Bmi-1 to the rim of the nucleus requires the RING finger domain and correlates with its ability to transform. Studies with Bmi-1 deletion mutants suggest that the ability of Bmi-1 to mediate cellular transformation correlates with its unique subnuclear localization but not its transcriptional suppression activity (Cohen, 1996).

Control of cell identity during development is specified in large part by the unique expression patterns of multiple homeobox-containing (Hox) genes in specific segments of an embryo. Trithorax and Polycomb-group (Trx-G and Pc-G) proteins in Drosophila maintain Hox expression or repression, respectively. Mixed lineage leukemia (MLL) is frequently involved in chromosomal translocations associated with acute leukemia and is the one established mammalian homolog of Trx. Bmi-1 was first identified as a collaborator in c-myc-induced murine lymphomagenesis and is homologous to the Drosophila Pc-G member Posterior sex combs. The axial-skeletal transformations and altered Hox expression patterns of Mll-deficient mice and Bmi-1-deficient mice are normalized when both Mll and Bmi-1 are deleted, demonstrating their antagonistic role in determining segmental identity. Embryonic fibroblasts from Mll-deficient mice, compared with Bmi-1-deficient mice, demonstrate reciprocal regulation of Hox genes as well as an integrated Hoxc8-lacZ reporter construct. Reexpression of MLL is able to overcome repression, rescuing expression of Hoxc8-lacZ in Mll-deficient cells. Consistent with this, MLL and BMI-I display discrete subnuclear colocalization. Although Drosophila Pc-G and Trx-G members have been shown to maintain a previously established transcriptional pattern, MLL can also dynamically regulate a target Hox gene (Hanson, 1999).

The murine Polycomb-Group (PcG) proteins Bmi1 (B cell-specific Mo-MLV integration site 1) and eed (embryonic ectoderm development) govern axial patterning during embryonic development by segment-specific repression of Hox gene expression. The two proteins engage in distinct multimeric complexes that are thought to use a common molecular mechanism to render the regulatory regions of Hox and other downstream target genes inaccessible to transcriptional activators. Beyond axial patterning, Bmi1 is also involved in hemopoiesis because a loss-of-function allele causes a profound decrease in bone marrow progenitor cells. Here, evidence is presented that is consistent with an antagonistic function of eed and Bmi1 in hemopoietic cell proliferation. Heterozygosity for an eed null allele causes marked myelo- and lympho-proliferative defects, indicating that eed is involved in the negative regulation of the pool size of lymphoid and myeloid progenitor cells. This antiproliferative function of eed does not appear to be mediated by Hox genes or the tumor suppressor locus p16INK4a/p19ARF because expression of these genes was not altered in eed mutants. Intercross experiments between eed and Bmi1 mutant mice reveal that Bmi1 is epistatic to eed in the control of primitive bone marrow cell proliferation. However, the genetic interaction between the two genes is cell-type specific, because the presence of one or two mutant alleles of eed trans-complements the Bmi1-deficiency in pre-B bone marrow cells. Thus, these studies suggest that hemopoietic cell proliferation is regulated by the relative contribution of repressive (Eed-containing) and enhancing (Bmi1-containing) PcG gene complexes. Biochemical studies have indicated that the PcG proteins Bmi1, Mel18, M33, and Mph1/Rae28 are constituents of a multimeric protein complex A, which localizes to discrete nuclear foci in U-2 OS osteosarcoma cells. Importantly, Eed neither interacts physically with Bmi1 nor engages in this protein complex. Instead, Eed forms a complex B with the PcG proteins Enx1/EzH2 and Enx2/EzH1, which lack signs of a discrete subnuclear distribution and are found rather uniformly throughout the nucleoplasm of U-2 OS osteosarcoma cells (Lessard, 1999).

The bmi-1 and myc oncogenes collaborate strongly in murine lymphomagenesis, but previously, the basis for this collaboration has not been understood. The ink4a-ARF tumor suppressor locus is a critical downstream target of the Polycomb-group transcriptional repressor Bmi-1. Part of Myc's ability to induce apoptosis depends on induction of p19arf. Down-regulation of ink4a-ARF by Bmi-1 underlies its ability to cooperate with Myc in tumorigenesis. Heterozygosity for bmi-1 inhibits lymphomagenesis in Eľ-myc mice by enhancing c-Myc-induced apoptosis. Increased apoptosis is observed in bmi-1 -/- lymphoid organs. This apoptosis can be rescued by deletion of ink4a-ARF or overexpression of bcl2. Furthermore, Bmi-1 collaborates with Myc in enhancing proliferation and transformation of primary embryo fibroblasts (MEFs) in an ink4a-ARF dependent manner, by prohibiting Myc-mediated induction of p19arf and apoptosis. Strong collaboration is observed between the Eľ-myc transgene and heterozygosity for ink4a-ARF. This heterozygosity is accompanied by loss of the wild-type ink4a-ARF allele and formation of highly aggressive B-cell lymphomas. Together, these results reinforce the critical role of Bmi-1 as a dose-dependent regulator of ink4a-ARF, which in its turn acts to prevent tumorigenesis upon activation of oncogenes such as c-myc (Jacobs, 1999).

Polycomb group genes were identified as a conserved group of genes whose products are required in multimeric complexes to maintain spatially restricted expression of Hox cluster genes. Unlike in Drosophila, in mammals Polycomb group (PcG) genes are represented as highly related gene pairs, indicative of duplication during metazoan evolution. Mel18 and Bmi1 are mammalian homologs of Drosophila Posterior sex combs. Mice deficient for Mel18 or Bmi1 exhibit similar posterior transformations of the axial skeleton and display severe immune deficiency, suggesting that their gene products act on overlapping pathways/target genes. However unique phenotypes are also observed upon loss of either Mel18 or Bmi1. Using embryos doubly deficient for Mel18 and Bmi1 it has been shown that Mel18 and Bmi1 act in synergy and in a dose-dependent and cell type-specific manner to repress Hox cluster genes and mediate cell survival of embryos during development. In addition, Mel18 and Bmi1, although essential for maintenance of the appropriate expression domains of Hox cluster genes, are not required for the initial establishment of Hox gene expression. Furthermore, there is an unexpected requirement for Mel18 and Bmi1 gene products to maintain stable expression of Hox cluster genes in regions caudal to the prospective anterior expression boundaries during subsequent development (Akasaka, 2001).

Compound Mel18/Bmi1-deficient mice reveal a strong exacerbation of the phenotypes of Bmi1 or Mel18 single mutant mice. Foremost, this is evident from the significant shortened life span of the doubly deficient embryos, which die around 9.5 dpc, displaying severe growth retardation and abnormalities in somite alignment, neural tube, paraxial mesoderm and notochord morphology. This is in sharp contrast to the single mutant mice, which are born without these severe malformations. Furthermore, Mel18+/-;Bmi1-/- and Mel18 -/-;Bmi1+/- newborn mice show a more pronounced and complete posterior transformations of the axial and appendicular skeleton than those seen in mice that lack either Bmi1 or Mel18, or even in M33-/-;Bmi1-/- doubly deficient mice. The severe morphological abnormalities and posterior transformations are accompanied with pronounced anterior shifts in Hox gene expression boundaries. These findings are in line with the ability of the highly related Mel18 and Bmi1 proteins to regulate largely overlapping sets of Hox target genes. In addition to their effects on Hox gene expression, Bmi1 and Mel18 are both required to regulate other target genes. This is well illustrated by the dramatically enhanced apoptosis in doubly deficient embryos. Thus, both genetic and biochemical evidence favor the conclusion that Bmi1 and Mel18 act in the same PcG complex, which most probably represents the mammalian ortholog of the recently described Drosophila PRC1 complex (Shao, 1999). This is further supported by the observation of characteristic scapulae defects in Mel18+/-;Bmi1-/- and Mel18-/-;Bmi1+/- mice. These defects are not present in single Mel18- or Bmi1-deficient mice, but are found in mice deficient for another mammalian 'PRC1' complex member, M33/Cbx2. Together, these results clearly indicate a strict dose-dependent requirement for both Mel18 and Bmi1 in regulating proper repression of Hox and other target genes (Akasaka, 2001).

An emerging concept in the field of cancer biology is that a rare population of 'tumor stem cells' exists among the heterogeneous group of cells that constitute a tumor. This concept, best described with human leukemia, indicates that stem cell function (whether normal or neoplastic) might be defined by a common set of critical genes. The Polycomb group gene Bmi-1 was shown to have a key role in regulating the proliferative activity of normal stem and progenitor cells. Most importantly, evidence has been provided that the proliferative potential of leukemic stem and progenitor cells lacking Bmi-1 is compromised because they eventually undergo proliferation arrest and show signs of differentiation and apoptosis, leading to transplant failure of the leukemia. Complementation studies showed that Bmi-1 completely rescues these proliferative defects. These studies therefore indicate that Bmi-1 has an essential role in regulating the proliferative activity of both normal and leukemic stem cells (Lessard, 2003).

In summary, the findings reported here reinforce the notion of a structure in leukemic hierarchy where 'stemness' would be conferred by the continual expression of Bmi-1. Importantly, it was possible under conditions of stimulation with high concentrations of growth factors in vitro to generate nonleukemogenic Bmi-1^-/- clones with an impaired expression of p19^ARF and p16^INK4a, both of which are known functional targets of ^Bmi-1. Retroviral introduction of ^Bmi-1 into these clones deficient in p19^ARF and p16^INK4a readily rescues their tumorigenic properties, suggesting that Bmi-1 has one or more additional functions in L-HSCs, in addition to repression of these CKIs. The expression of currently known regulators of early haematopoiesis (such as tal-1/SCL and ^Hoxb4) was not altered by the status of Bmi-1 in leukemic and nonleukemic clones (Lessard, 2003).

The apparent difficulties of bypassing the requirement for Bmi-1 in leukemic stem/progenitor cells in vivo suggests that adroit molecular targeting of ^Bmi-1 in leukemic stem/progenitor cells might have potent and specific therapeutic effects. Interestingly, BMI-1 expression has been recently reported in several cases of human non-small-cell lung cancer, breast cancer cell lines and immortalized mammary epithelial cells (MECs). It will therefore be interesting to determine whether the findings reported here might also extend to other types of 'cancer stem cells' (Lessard, 2003).

Bmi-1 is required for the post-natal maintenance of stem cells in multiple tissues including the central nervous system (CNS) and peripheral nervous system (PNS). Deletion of Ink4a or Arf from Bmi-1^-/- mice partially rescues stem cell self-renewal and stem cell frequency in the CNS and PNS, as well as forebrain proliferation and gut neurogenesis. Arf deficiency, but not Ink4a deficiency, partially rescues cerebellum development, demonstrating regional differences in the sensitivity of progenitors to p16^Ink4a and p19^Arf. Deletion of both Ink4a and Arf does not affect the growth or survival of Bmi-1^-/- mice or completely rescue neural development. Bmi-1 thus prevents the premature senescence of neural stem cells by repressing Ink4a and Arf, but additional pathways must also function downstream of Bmi-1 (Molofsky, 2005).

The Polycomb group (PcG) gene Bmi1 promotes cell proliferation and stem cell self-renewal by repressing the Ink4a/Arf locus. A genetic approach was used to investigate whether Ink4a or Arf is more critical for relaying Bmi1 function in lymphoid cells, neural progenitors, and neural stem cells. Arf is a general target of Bmi1, however particularly in neural stem cells, derepression of Ink4a contributes to Bmi1^-/- phenotypes. Additionally, haploinsufficient effects have been demonstrated for the Ink4a/Arf locus downstream of Bmi1 in vivo. This suggests differential, cell type-specific roles for Ink4a versus Arf in PcG-mediated (stem) cell cycle control (Bruggeman, 2005).

Best known as epigenetic repressors of developmental Hox gene transcription, Polycomb complexes alter chromatin structure by means of post-translational modification of histone tails. Depending on the cellular context, Polycomb complexes of diverse composition and function exhibit cooperative interaction or hierarchical interdependency at target loci. The present study interrogated the genetic, biochemical and molecular interaction of BMI1 [Drosophila homologs Psc and Su(z)2] and EED (Drosophila homolog; Esc), pivotal constituents of heterologous Polycomb complexes, in the regulation of vertebral identity during mouse development. Despite a significant overlap in dosage-sensitive homeotic phenotypes and co-repression of a similar set of Hox genes, genetic analysis implicated eed and Bmi1 in parallel pathways, which converge at the level of Hox gene regulation. Whereas EED and BMI1 formed separate biochemical entities with EzH2 and Ring1B, respectively, in mid-gestation embryos, YY1 engaged in both Polycomb complexes. Strikingly, methylated lysine 27 of histone H3 (H3-K27), a mediator of Polycomb complex recruitment to target genes, stably associated with the EED complex during the maintenance phase of Hox gene repression. Juxtaposed EED and BMI1 complexes, along with YY1 and methylated H3-K27, were detected in upstream regulatory regions of Hoxc8 and Hoxa5. The combined data suggest a model wherein epigenetic and genetic elements cooperatively recruit and retain juxtaposed Polycomb complexes in mammalian Hox gene clusters toward co-regulation of vertebral identity (Kim, 2006).

At least two PcG complexes with diverse composition and function in chromatin remodeling have been identified in mammals. The Polycomb repressive complex 1 (PRC1) involves the paralogous PcG proteins BMI1/MEL18, M33/PC2, RAE28, and RING1A. Evidence for PRC1-mediated chromatin modification derived from ubiquitylation at lysine 119 of histone H2A (H2A-K119). A second PcG complex, PRC2, encompasses EED, the histone methyltransferase EZH2, the zinc finger protein SUZ12, the histone-binding proteins RBAP46/RBAP48, and the histone deacetylase HDAC1. Several EED isoforms, generated by alternate translation start site usage of eed mRNA, differentially engage in PRC2-related complexes (PRC2/3/4), targeting the histone methyltransferase activity of EZH2 to H3-K27 or H1-K26. PcG complexes bind to cis-acting Polycomb response elements (PREs), which encompass several hundred base pairs and are necessary and sufficient for PcG-mediated repression of target genes. Whereas the function of several PREs has been delineated in Drosophila, similar elements await characterization in mammals (Kim, 2006 and references therein).

An antibody raised against residues 123-140 of the EED amino terminus precipitated three distinct isoforms of approximately 50 and 75 kDA from E12.5 trunk, representing three of the four EED isoforms previously reported in 293 cells. In addition to EZH2 and YY1, dimethylated H3-K27 co-immunoprecipitated with EED. Immunoprecipitation identified three BMI1 isoforms of approximately 39-41 kDA. BMI1 was found in a complex with RING1B, but not dimethylated H3-K27. Similar to the EED complex, the BMI1 complex also contained YY1. It should be emphasized that all (co-)immunoprecipitating bands were detected by at least two antibodies against different epitopes. Strikingly, while dimethylated H3-K27 engaged in the EED complex, trimethylated H3-K27 did not appear to associate with either the EED or the BMI1 complex. Importantly, reciprocal co-immunoprecipitation detected EED and BMI1 in separate protein complexes (Kim, 2006).

Ectopic expression in mutant embryos revealed Hoxc8 and Hoxa5 as downstream targets of EED and BMI1 function. ChIP detected EED and BMI1 binding immediately upstream of the Hoxc8 transcribed region near putative promoter elements. The binding sites could not be separated, indicating close proximity of the complexes. EED and BMI1 binding also clustered within a small fragment 1.5 kb upstream of the Hoxc8 transcription start site, suggesting long-range juxtaposition of heterologous PcG complexes. Similar to EED and BMI1, YY1 localized to both regions. In support of YY1 binding to Hox regulatory regions, inspection of the mouse genome sequence revealed clusters of putative YY1 binding sites in both regions a and b, including TGTCCATTAG and CCCCCATTCC (region a), as well as ACACCATGGC, TTTCCATTAG and TCCCCATAAA (region b). CCAT represents the core of the YY1 consensus binding site, while flanking sequences exhibited significant tolerance for multiple nucleotides. EED, BMI1 and YY1 also co-localized approximately 1.5 kb upstream of the transcription start site of Hoxa5. In addition to PcG binding, ChIP detected trimethylated H3-K27 throughout the regulatory regions of Hoxc8 and Hoxa5. Furthermore, dimethylated H3-K27 localized to region b of Hoxc8 (Kim, 2006).

Spatial regulation of EED and BMI1 binding to Hox regulatory regions was evident from ChIP analysis of dissected anterior and posterior regions of E12.5 trunk. In agreement with transcriptional silencing of Hoxc8 and Hoxa5, EED and BMI1 binding was detected upstream of these loci in anterior regions of the trunk. By contrast, EED and BMI1 binding was absent from posterior regions of the trunk, where Hoxc8 and Hoxa5 are transcribed. These findings implicate PcG complexes in Hox gene repression in anterior regions of the AP axis (Kim, 2006).

The combined interpretation of the co-immunoprecipitation and ChiP results indicates that trimethylated H3-K27 did not form a complex with EED or BMI1, despite co-localization of the three proteins in Hox regulatory regions. By contrast, co-immunoprecipitation demonstrated physical association of the EED complex with dimethylated H3-K27. In aggregate, the results support a model in which EED- and BMI1-containing chromatin remodeling complexes exist as separate, but juxtaposed, biochemical entities at Hox target loci (Kim, 2006).

Genetic studies have demonstrated that Bmi1 promotes cell proliferation and stem cell self-renewal with a correlative decrease of p16^INK4a expression. Polycomb genes EZH2 and BMI1 repress p16 expression in human and mouse primary cells, but not in cells deficient for pRB protein function. The p16 locus is H3K27-methylated and bound by BMI1, RING2, and SUZ12. Inactivation of pRB family proteins abolishes H3K27 methylation and disrupts BMI1, RING2, and SUZ12 binding to the p16 locus. These results suggest a model in which pRB proteins recruit PRC2 to trimethylate p16, priming the BMI1-containing PRC1L ubiquitin ligase complex to silence p16 (Kotake, 2007).

The mammalian pRB family proteins, pRB, p107, and p130 (also known as pocket proteins), play a key role in controlling the G1-to-S transition of the cell cycle and maintaining differentiated cells in a reversible quiescent or permanent senescent arrest state. The pocket proteins are hypophosphorylated in cells exiting mitosis as well as in quiescent cells, where they bind to and negatively regulate the function of the E2F family transcription factors. In cells entering the cell cycle, extracellular mitogens first induce the expression of D-type cyclins, which bind to and activate CDK4 and CDK6, leading to the phosphorylation of pRB family proteins, causing functional inactivation by E2F dissociation, thereby promoting a G1-to-S transition. Inhibition of CDK4 and CDK6 by the INK4 family of CDK inhibitors (p16^INK4a, p15^INK4b, p18^INK4c, and p19^INK4d) retains pRB family proteins in their hypophosphorylated, growth-suppressive states and prevents G1-to-S progression. Disruption of the INK4-RB pathway, consisting of INK4-cyclinDs-CDK4/6-RB-E2Fs, deregulates G1-to-S control and represents a common event in the development of most, if not all, types of cancer (Kotake, 2007).

Among the major challenges toward a better understanding of G1 control by the INK4-pRB pathway is how different INK4 genes are regulated, thereby linking G1 control to different cellular pathways. INK4 proteins are relatively stable, and the primary regulation of INK4 is through transcriptional control. The expression of each of the INK4 genes is distinctly different during development, in different adult tissues, and in response to different cellular conditions. There have been only a few reports wherein a transcriptional regulator has been demonstrated to bind to an INK4 promoter by either gel shift or chromatin immunoprecipitation (ChIP) assay. Identification of factors that directly bind to INK4 promoters holds the key to linking different cellular pathways to G1 control by the INK4-pRB pathway, but these links remain disproportionately poorly understood in comparison with knowledge of the function of the INK4-pRB pathway (Kotake, 2007).

To elucidate the molecular mechanisms regulating p16 expression, whether p16 gene expression is directly regulated by BMI1, an oncogene that encodes a transcriptional repressor of the Polycomb group (PcG) of proteins, was directly tested. Deletion of Bmi1 retards cell proliferation, causes premature senescence in mouse embryonic fibroblasts (MEFs), and reduces the number of hematopoietic stem cells, with an associated up-regulation of p16 (and to a lesser extent of p19^Arf). Codeletion of p16 (or p16-Arf) partially rescues the proliferative defects of Bmi1-null cells, providing genetic evidence supporting a functional interaction between the Bmi1 and p16 genes. However, whether BMI1 directly binds to and regulates the transcription of the p16 gene has not been demonstrated. A notable feature of p16 is its high level of expression in virally transformed cells and its inverse correlation with pRB function, suggesting a negative regulation of p16 gene expression by pRB. Therefore whether pRB and BMI1 collaboratively repress p16 expression was also examined (Kotake, 2007).

These results provide the first biochemical evidence supporting a direct regulation of p16 transcription by the PRC2 histone methyltransferase complex and the BMI1-RING2-containing PRC1 histone ubiquitin ligase complex. Both H3K27 methylation at and BMI1/RING2 binding to the p16 locus require the function of the pRB family proteins, linking for the first time H3K27 methylation and the function of BMI1 with the pRB proteins. The detailed biochemical mechanism by which pRB family proteins collaborate with BMI1 to repress p16 transcription is yet to be determined. In repeated attempts, binding of pRB to the p16 locus could not be detected. The simplest model suggested by these results is that the pRB family proteins are either involved in regulating the enzymatic activity or the recruitment of PRC2 to the p16 locus. H3K27 methylation by PRC2 would then facilitate recruitment of the BMI1-containing PRC1L complex to ubiquitinate H2A, leading to p16 silencing (Kotake, 2007).

The results also suggest a regulatory loop between p16 and the pocket proteins, with p16 acting as an upstream activator of the pocket proteins and the pocket proteins repressing p16 transcription as negative feedback. INK4 proteins are intrinsically stable and, once synthesized, stably bind to and inhibit the activity of CDK4/6 by both interfering with ATP binding and by reducing the cyclin-CDK4/6 surface. Without a mechanism for repressing INK4 expression, mitogen-induced cyclin D synthesis would not be able to compete off INK4 from CDK4/6, and displaced, monomeric cyclin D proteins would be rapidly degraded, leaving a constitutive activation of RB function and locking cells in a permanent G1-arrested state. Repression of p16 expression by pRB family proteins thus also constitutes a feedback loop to set up a balance between INK4-mediated inhibition and cyclin D-mediated activation of G1 progression. This function of p16, however, must be repressed in stem cells, which undergo continuous proliferation and self-renewal in vivo. It is speculated that one mechanism to achieve this is through expression of BMI1 in the stem cell compartment (Kotake, 2007).