Sex combs on midleg
Lethal(3)malignant brain tumor: a Sex combs on midleg homolog When mutated, the lethal(3)malignant brain tumor [l(3)mbt] gene causes malignant growth of the adult optic neuroblasts and ganglion mother cells in the larval brain and imaginal disc overgrowth. Via overlapping deficiencies a genomic region of approximately 6.0 kb has been identified, containing l(3)mbt+ gene sequences. The l(3)mbt+ gene encodes seven transcripts of 5.8 kb, 5.65 kb, 5.35 kb, 5.25 kb, 5.0 kb, 4.4 kb and 1.8 kb. The putative MBT163 protein, encompassing 1477 amino acids, is proline-rich and contains a novel zinc finger. In situ hybridizations of whole mount embryos and larval tissues reveal l(3)mbt+ RNA ubiquitously present in stage 1 embryos and throughout embryonic development in most tissues. In third instar larvae l(3)mbt+ RNA is detected in the adult optic anlagen and the imaginal discs, the tissues directly affected by l(3)mbt mutations, but also in tissues, showing normal development in the mutant, such as the gut, the goblet cells and the hematopoietic organs (Wismar, 1995).
The lethal(3)malignant brain tumor [D-l(3)mbt] gene is considered to be one of the tumor suppressor genes of Drosophila, and its recessive mutations are associated with malignant transformation of the neuroblasts in the larval brain. The structure of D-l(3)mbt protein is similar to Drosophila Sex combs on midleg (Scm) protein which is a member of Polycomb group (PcG) proteins. The first human homolog of the D-l(3)mbt gene, designated h-l(3)mbt, has been isolated. Radiation hybrid mapping and fluorescence in situ hybridization (FISH) analysis has localized the h-l(3)mbt gene to chromosome 20q12. The h-l(3)mbt transcript is expressed in most of the human adult normal tissues and cultured cell lines. However, some cancer cells markedly reduce h-l(3)mbt protein expression. Immunocytochemical study reveals that the h-l(3)mbt protein shows a speckled and scattered distribution in interphase nuclei and completely associates with condensed chromosomes in mitotic cells. This subcellular localization has been shown to be different from that of Bmi1 protein, which is a component of PcG complex. Furthermore, overexpression of h-l(3)mbt protein by using a Cre-mediated gene activation system leads to failures of proper chromosome segregation and cytokinesis, which result in formation of multinuclei in U251MG cells. These observations suggest that h-l(3)mbt protein has functions distinct from those of PcG proteins and may play a role in proper progression of cell division (Koga, 1999).
Mammalian Sex combs on midleg homologs: Rae28, SCML1 and SCML2 The rae28 gene, a mouse homolog of the Drosophila polyhomeotic gene, is involved in the maintenance of the transcriptional repression states of Hox genes. A glutathione S transferase-RAE28 (GST-RAE28) fusion protein has been synthesized and sequence-specific DNA binding activity in the RAE28 protein was examined by using the selected and amplified binding site method. After five rounds of enrichment, the eluted DNAs were amplified, cloned and sequenced. The sequences of individual oligonucleotides included the following consensus sequences; 5'-ACCA-3', 5'-ACCCA-3', 5'-CTATCA-3' and 5'-TGCC-3'. The oligonucleotides including these consensus sequences have significant affinity with the GST-RAE28 fusion protein. The RAE28 protein forms multimeric protein complexes with other members of mouse Pc-G proteins in the nucleus. These findings strongly suggest that the RAE28 protein constitutes a sequence-specific DNA binding domain in multimeric Pc-G protein complexes (Nomura, 1998).
The Polycomb group loci in Drosophila encode chromatin proteins required for repression of homeotic loci in embryonic development. Mouse Polycomb group homologs, RAE28, BMI1 and M33, have overlapping but not identical expression patterns during embryogenesis and in adult tissues. These three proteins coimmunoprecipitate from embryonic nuclear extracts. Gel filtration analysis of embryonic extracts indicates that RAE28, BMI1 and M33 exist in large multimeric complexes. M33 and RAE28 coimmunoprecipitate and copurify as members of large complexes from F9 cells, which express BMI1 at very low levels, suggesting that different Polycomb group complexes can form in different cells. RAE28, BMI1 and M33 interact homotypically, and both RAE28 and M33 interact with BMI1, but not with each other. The domains required for interaction have been localized. Together, these studies indicate that murine Polycomb group proteins are developmentally regulated and function as members of multiple, heterogeneous complexes (Hashimoto, 1998).
Polycomb group (PcG) genes were initially described in Drosophila melanogaster as regulators of the homeobox gene. Four mammalian homologs, mel-18, bmi-1, M33 and rae-28, have been analyzed in this study. They not only regulate mammalian homeotic genes by analogy with their Drosophila counterparts, but also have some influence on the growth and differentiation of B lymphocytes. These four mammalian PcG genes are rapidly induced after antigen-receptor cross-linking in B cells. Thus it is proposed that mammalian PcG genes can be categorized as a new type of immediate early gene (Hasegawa, 1998).
Using exon trapping, a new human gene in Xp22 has been identified encoding a 3-kb mRNA. Expression of this RNA is detectable in a range of tissues but is most pronounced in skeletal muscle and heart. The gene, designated 'sex comb on midleg-like-1' (SCML1), maps 14 kb centromeric of marker DXS418, between DXS418 and DXS7994, and is transcribed from telomere to centromere. SCML1 spans 18 kb of genomic DNA, consists of six exons, and has a 624-bp open reading frame. The predicted 27-kDa SCML1 protein contains two domains that each have a high homology to two Drosophila transcriptional repressors of the polycomb group (PcG) genes and their homologues in mouse and human. PcG genes are known to be involved in the regulation of homeotic genes, and the mammalian homologs of the PcG genes repress the expression of Hox genes. SCML1 appears to be a new human member of this gene group and may play an important role in the control of embryonal development (van de Vosse, 1998).
A novel gene with homologies to the Drosophila Sex combs on midleg (Scm) gene from the short arm of the X chromosome has been identified. Scm is a member of
the Polycomb group (PcG) genes, which encode transcriptional repressors essential for appropriate development in the fly and in mammals. The newly identified
transcript named SCML2 (sex comb on midleg like-2, HGMW-approved symbol) is ubiquitously expressed and encodes a protein of 700 amino acids. SCML2
maps very close to the recently identified SCML1, revealing the presence of a new gene cluster in Xp22. The homology and map location identify SCML2 as a
candidate gene for Xp22-linked developmental disorders, including the oral-facial-digital type I (OFDI) syndrome. A study of the SCML1-SCML2 cluster in
primates indicates that the two genes are localized to the same region in Old World monkeys, New World monkeys, and prosimians, suggesting that the duplication
event leading to the formation of the SCML cluster on Xp22 occurred before primate divergence (Montini, 1999).
Expressed sequence tag (EST) databases were searched to recover putative
mammalian Sex combs on midleg homologs. A cDNA recovered that encodes two mbt repeats and the SPM domain that
characterizes Scm, but that lacks the cysteine clusters and the serine/threonine-rich region has been found at the amino terminus of Scm. Accordingly, the gene has been named Sex
comb on midleg homolog 1 (Scmh1). Like their Drosophila counterparts, Scmh1 and the mammalian polyhomeotic homolog RAE28/mph1 interact in vitro via their
SPM domains. The expression of Scmh1 and rae28/mph1 was examined using Northern analysis of embryos and adult tissues, and in situ hybridization to embryos.
The expression of Scmh1 and rae28/mph1 is well correlated in most tissues of embryos. However, in adults, Scmh1 expression is detected in most tissues,
whereas mph1/rae28 expression is restricted to the gonads. Scmh1 is strongly induced by retinoic acid in F9 and P19 embryonal carcinoma cells. Scmh1 maps to
4D1-D2.1 in mice. These data suggest that Scmh1 will have an important role in regulation of homeotic genes in embryogenesis and that the interaction with
RAE28/mph1 is important in vivo (Tomotsune, 1999).
cDNAs encoding two isoforms of a human
homolog of Drosophila Sex comb on midleg (Scm) have been isolated and named Sex comb on midleg homolog-1 (SCMH1). Overall, SCMH1 has 94% identity to its mouse counterpart
Scmh1, and 41% identity to Scm; it contains two 1(3)mbt domains and the SPM domain; these are both characteristic of Scm. SCMH1 is widely expressed in adult
tissues, and maps to chromosome 1p34 (Berger, 1999).
A novel cDNA clone encoding a protein structurally related to the transcriptional repressor Polycomb group (PcG) proteins, which regulate homeotic genes and others, was isolated from mouse and rat brain. The coding protein contains the SPM domain and mbt repeats, both of which are characteristic of the PcG proteins, and show significant similarity in amino acid sequence to the Drosophila Sex comb on midleg (Scm) protein. Since this novel protein contains the mbt repeats in four tandem copies, this murine gene was designated as Sfmbt for Scm-related gene containing four mbt domains. Cloning and characterization of the mouse Sfmbt gene revealed that the coding sequence comprises 20 exons, dispersed along approximately 40kb, and maps to the proximal part of chromosome 14. Northern blot analysis showed that the Sfmbt mRNAs are expressed most abundantly in the adult testis, and less intensively in all other tissues examined (Usui, 2000).
The product of the Scmh1 gene, a mammalian homolog of Drosophila
Sex comb on midleg, is a constituent of the mammalian Polycomb repressive
complexes 1 (Prc1). Scmh1 has been identified as an indispensable component of
the Prc1. During progression through pachytene, Scmh1 was shown to be excluded
from the XY body at late pachytene, together with other Prc1 components such
as Phc1 and Phc2 (Polyhomeotic homologs), Rnf110 (Pcgf2), Bmi1 [Drosophila homologs Psc and Su(z)2] and Cbx2 (Polycomb homolog). The role of
Scmh1 in mediating the survival of late pachytene spermatocytes has been identified. Apoptotic
elimination of Scmh1-/- spermatocytes is accompanied by
the preceding failure of several specific chromatin modifications at the XY
body, whereas synapsis of homologous autosomes is not affected. It is
therefore suggested that Scmh1 is involved in regulating the sequential
changes in chromatin modifications at the XY chromatin domain of the pachytene
spermatocytes. Restoration of defects in Scmh1-/-
spermatocytes by Phc2 mutation indicates that Scmh1 exerts its
molecular functions via its interaction with Prc1. Therefore, for the first
time, it is possible to indicate a functional involvement of Prc1 during the
meiotic prophase of male germ cells and a regulatory role of Scmh1 for Prc1,
which involves sex chromosomes (Takada, 2007).
Based on the present observations, it is postulated that Scmh1 could primarily
promote the exclusion of Prc1 components from the XY body in the pachytene
spermatocytes because Scmh1 itself is a functional component of Prc1. By
contrast, failure to maintain exclusion of trimethylated H3-K27 and to undergo
H3-K9 methylation at the XY body in Scmh1-/- spermatocytes
may occur secondarily to the failure to exclude Prc1 from the XY body. At many
loci, epistatic engagement of Prc1 by Prc2 has been shown to be essential for
the mediation of transcriptional repression.
Preceding exclusion of trimethylated H3-K27, which represents Prc2 actions,
for Prc1 exclusion from the XY body, is consistent with epistatic roles of
Prc2 for Prc1 at the XY body. Therefore, Scmh1 may affect H3-K27
trimethylation at the XY body through the Prc1-Prc2 engagement. It is
noteworthy that H3-K27 trimethylation has been shown to be regulated by Prc1
at the XY body. This may imply that Prc1-Prc2 engagement is a reciprocal
rather than epistatic process at the XY body. This possibility should be
addressed by using conditional mutants for Prc2 components. A functional correlation between Prc1 exclusion and H3-K9 methylations at the XY body is also
hypothesized because the indispensable H3-K9 methyltransferase
complex, composed of G9a and GLP, is constitutively associated with E2F6
complexes, which share at least Rnf2 and Ring1 components with Prc1. Moreover,
several components of respective complexes are structurally related to each
other. Intriguingly, although Prc1 components, apart from Rnf2, have been shown to be
excluded from the XY body at late pachytene stage, components of E2F6 complexes including Rnf2, RYBP, HP1gamma and G9a are retained. The most attractive scenario would be that exclusion of Prc1 is a prerequisite for the functional manifestation of E2F6 complexes to mediate the hypermethylation of H3-K9 at the XY body. It is thus proposed that Scmh1-mediated exclusion of Prc1 from the XY body might be a prerequisite for maintaining appropriate chromatin structure to undergo subsequent sequential chromatin remodeling of the XY chromatin in pachytene spermatocytes (Takada, 2007).
It is also suggested that sequential changes in chromatin modifications of the
sex chromosomes in the pachytene spermatocytes might be monitored by some
meiotic checkpoint mechanisms. This is supported by the temporal concurrence
of Prc1 exclusion from the XY body and apoptotic depletion of meiotic
spermatocytes, their coincidental restorations by Phc2 mutation, and
normal oogenesis and fertility in Scmh1-/- females. In addition, defects in the XY body formations have been shown to correlate with apoptotic depletion of meiotic spermatocytes by studies using H2A.X and Brca1 mutants, although developmental arrests occurred by early pachytene stage. However, this link has not been substantially demonstrated (Takada, 2007).
Although Scmh1 has been shown to act together with Prc1, the role of Scmh1
for Prc1 might be modified in a tissue- or locusspecific manner because
spermatogenic defects by Scmh1 mutation are restored by Phc2
mutation, whereas premature senescence of MEFs is enhanced mutually by both mutations. This is supported by an immunofluorescence study revealing the co-localization of Scmh1 with other class 2 PcG proteins in subnuclear speckles in U2OS cells, whereas in female
trophoblastic stem (TS) cells it is excluded from the inactive X chromosome
domain, which is intensely decorated by Rnf2, Phc2 and Rnf110. It may be possible to postulate some additional factors that modify the molecular functions or subnuclear localization of Scmh1. Indeed, most of the soluble pool of SCM in Drosophila embryos is not stably associated with Prc1, although SCM is capable of assembling with the Polyhomeotic protein by their respective SPM domains in the Polycomb core
complex. As the SPM domain is shared, not only by
polyhomeotic homologs, but also by multiple paralogs of the
Drosophila Scm gene, namely Scml1, Scml2, Sfmbt, l(3)mbt3
and others in mammals, these structurally related gene products may
potentially interact with Scmh1 and modulate its functions. Conservations of
crucial amino acid residues required for the mutual interaction of SPM domains
and multiple mbt repeats in these proteins may further suggest functional
overlap with Scmh1. It is notable that phenotypic expressions of
Scmh1 mutation are quite variable during spermatogenesis and axial
development even after more than five times backcrossing to a C57Bl/6 background. This incomplete penetrance might involve multiple paralogs of the canonical Scm proteins, which may act in compensatory manner for Scmh1 mutation, as revealed between Rnf110 and Bmi1 or Phc1 and Phc2 (Takada, 2007).
The Polycomb-group (PcG) repressive complex-1 (PRC1) forms microscopically visible clusters in nuclei; however, the impact of this cluster formation on transcriptional regulation and the underlying mechanisms that regulate this process remain obscure. This study reports that the sterile alpha motif (SAM) domain of a PRC1 core component Phc2 plays an essential role for PRC1 clustering through head-to-tail macromolecular polymerization, which is associated with stable target binding of PRC1/PRC2 and robust gene silencing activity. A role is proposed for SAM domain polymerization in this repression by two distinct mechanisms: first, through capturing and/or retaining PRC1 at the PcG targets, and second, by strengthening the interactions between PRC1 and PRC2 to stabilize transcriptional repression. These findings reveal a regulatory mechanism mediated by SAM domain polymerization for PcG-mediated repression of developmental loci that enables a robust yet reversible gene repression program during development (Isono, 2013).
This report shows that microscopically visible subnuclear PRC1 domains represent SAM polymerization-dependent PRC1 clustering at PcG target genes and that such clustering appears to be closely related to their gene silencing activity in mammalian cells. Further, PRC1 clustering itself is stable, but the participating PRC1 components within such foci are continuously interchanged with the PRC1 reservoir outside the clusters. It is therefore proposed that the SAM-mediated interaction likely contributes to capture and/or retainment of PRC1 at the target loci by counteracting the constitutive exchange of PRC1 components to and from the clusters. The data further point to a role of SAM domain-mediated PRC1 clustering to form chromatin configurations that are fit for stable binding of PRC1 and PRC2 and exclusion of RNA polymerase II. Importantly, a previous structural study revealed a capacity for the Ph-SAM domain to form head-to-tail helices with 6-fold screw symmetry, which could be necessary for the outward positioning for the rest of the Ph protein from the polymer axis (Kim, 2002). The same structural capacity could be possessed by SAM-domains of Phc2, Phc1, and Phc3 as well, as predicted by sequence homology. It is speculated that multivalency and periodicity of PRC1 conferred by SAM-mediated polymerization play a key role to reinforce the PRC1/nucleosome interaction suitable for gene silencing. The periodic nature of the polymerized PRC1 by SAM-mediated interactions may contribute to nucleosome alignment at regular intervals by the chromodomain/H3K27me3 interaction. This may in turn facilitate the organization of an optimum nucleosome density that is preferable for binding and biochemical activity of PRC2 (Yuan, 2012). Through these multiple and diverse mechanisms, SAM-mediated PRC1 clustering likely strengthens the PRC1/PRC2 interaction to yield robustly repressed chromatin landscapes, which efficiently blocks the access of RNA polymerase II and preventing chromatin remodeling. It is also noteworthy that only 12% of Ring1B+H3K27me3+ genes exhibited robust and significant de-repression in Phc2L307R/L307R mouse embryonic
fibroblasts (MEFs), despite the dominant negative function of Phc2L307R. This observation implies that although Phc2-SAM polymerization could be a key mechanism for stable PRC1 binding and/or to silence PcG target genes, this process is complemented and buffered by other silencing mechanisms embedded in the PRC1 circuitry such as the H2A monoubiquitination (Endoh, 2012). Phc1-SAM and Phc3-SAM, which are closely related to Phc2-SAM, are also speculated to aid in PRC1 cluster formation and gene silencing, because Phc1 is shown to colocalize with Phc2 at PRC1 clusters and synergistically regulate the repression with Phc2 (Isono, 2013).
Although SAM polymerization could be a critical process for silencing of PcG target genes, its contribution to the long-range interactions of separate PRC1-binding sites is still controversial. Although this study showed that defective Phc2-SAM polymerization affects condensation of the Hoxb cluster, the data did not exclude the possibility that this condensation defect could be due to altered local chromatin. It is, however, fascinating to speculate that the multivalency of polymerized PRC1 might yield a large amount of hypothetically free/exposed chromodomain motifs that could be used to bind to a second array of target chromatin. This model is consistent with a previous observation (Lavigne, 2004) that PCC1-bound chromatin could recruit and repress a second nucleosomal array. Interestingly, that study found that Phc1 plays a critical role in this recruitment process. Based on those findings, it is proposed that SAM-mediated PRC1 polymerization could also be used for propagation of silencing by facilitating the binding of PRC1 to a second chromatin array and that this process might play an essential role in PRC1- (and PRC2-) mediated transcriptional silencing, especially for genes that are located within multiple gene clusters such as the Hox loci (Isono, 2013).
The stable existence and/or maintenance of PRC1 clusters despite the dynamic and continuous exchange of PRC1 components suggests that Phc2-SAM polymerization should be accompanied by, and balanced with, its depolymerization at these clusters. This depolymerizing effect may potentially contribute to keeping PcG repression reversible in response to developmental cues. Indeed, activation of Hox cluster genes by developmental inputs is accompanied by decondensation of the cluster in ESCs and developing tissues. Consistent with this model, Phc2 itself has been closely linked to developmental signals. Phc2 interacts with MAPK activated kinases at subnuclear PcG foci and is likely involved in mediating MAPK signals that maintain hematopoietic stem cells (Schwermann, 2009). Collectively, it is proposed that Phc2-SAM polymerization is involved in conferring robustness yet reversibility to PRC1-mediated repression of developmental genes that enables successful and robust implementation of developmental programs at PcG target loci (Isono, 2013).
Polycomb group (PcG) proteins are transcriptional repressors of genes involved in development and differentiation, and also maintain repression of key genes involved in the cell cycle, indirectly regulating cell proliferation. The human SCML2 gene, a mammalian homologue of the Drosophila PcG protein SCM, encodes two protein isoforms: SCML2A that is bound to chromatin and SCML2B that is predominantly nucleoplasmic. SCML2B was purified and found to form a stable complex with CDK/CYCLIN/p21 and p27, enhancing the inhibitory effect of p21/p27. SCML2B participates in the G1/S checkpoint by stabilizing p21 and favoring its interaction with CDK2/CYCE, resulting in decreased kinase activity and inhibited progression through G1. In turn, CDK/CYCLIN complexes phosphorylate SCML2, and the interaction of SCML2B with CDK2 is regulated through the cell cycle. These findings highlight a direct crosstalk between the Polycomb system of cellular memory and the cell-cycle machinery in mammals (Leconam 2013).
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Sex combs on midleg:
Biological Overview
| Regulation
| Developmental Biology
| Effects of Mutation
| References
Society for Developmental Biology's Web server.