Malek Djabali

A Vertebrate Polycomb Response Element Governs Segmentation of the Posterior Hindbrain

Chromatin remodeling by Polycomb group (PcG) and trithorax group (trxG) proteins regulates gene expression in all metazoans. Two major complexes, Polycomb repressive complexes 1 and 2 (PRC1 and PRC2), are thought to mediate PcG-dependent repression in flies and mammals. In Drosophila, PcG/trxG protein complexes are recruited by PcG/trxG response elements (PREs). However, it has been unclear how PcG/trxG are recruited in vertebrates. Here we have identified a vertebrate PRE, PRE-kr, that regulates expression of the mouse MafB/Kreisler gene. PRE-kr recruits PcG proteins in flies and mouse F9 cells and represses gene expression in a PcG/trxG-dependent manner. PRC1 and 2 bind to a minimal PRE-kr region, which can recruit stable PRC1 binding but only weak PRC2 binding when introduced ectopically, suggesting that PRC1 and 2 have different binding requirements. Thus, we provide evidence that similar to invertebrates, PREs act as entry sites for PcG/trxG chromatin remodeling in vertebrates.

Polycomb Mediated Epigenetic Silencing and Replication Timing at the INK4a/ARF Locus during Senescence

Background The INK4/ARF locus encodes three tumor suppressor genes (p15Ink4b, Arf and p16Ink4a) and is frequently inactivated in a large number of human cancers. Mechanisms regulating INK4/ARF expression are not fully characterized. Principal Findings Here we show that in young proliferating embryonic fibroblasts (MEFs) the Polycomb Repressive Complex 2 (PRC2) member EZH2 together with PRC1 members BMI1 and M33 are strongly expressed and localized at the INK4/ARF regulatory domain (RD) identified as a DNA replication origin. When cells enter senescence the binding to RD of both PRC1 and PRC2 complexes is lost leading to a decreased level of histone H3K27 trimethylation (H3K27me3). This loss is accompanied with an increased expression of the histone demethylase Jmjd3 and with the recruitment of the MLL1 protein, and correlates with the expression of the Ink4a/Arf genes. Moreover, we show that the Polycomb protein BMI1 interacts with CDC6, an essential regulator of DNA replication in eukaryotic cells. Finally, we demonstrate that Polycomb proteins and associated epigenetic marks are crucial for the control of the replication timing of the INK4a/ARF locus during senescence. Conclusions We identified the replication licencing factor CDC6 as a new partner of the Polycomb group member BMI1. Our results suggest that in young cells Polycomb proteins are recruited to the INK4/ARF locus through CDC6 and the resulting silent locus is replicated during late S-phase. Upon senescence, Jmjd3 is overexpressed and the MLL1 protein is recruited to the locus provoking the dissociation of Polycomb from the INK4/ARF locus, its transcriptional activation and its replication during early S-phase. Together, these results provide a unified model that integrates replication, transcription and epigenetics at the INK4/ARF locus.

Activation of the beta globin locus by transcription factors and chromatin modifiers

Locus control regions (LCRs) alleviate chromatin‐mediated transcriptional repression. Incomplete LCRs partially lose this property when integrated in transcriptionally restrictive genomic regions such as centromeres. This frequently results in position effect variegation (PEV), i.e. the suppression of expression in a proportion of the cells. Here we show that this PEV is influenced by the heterochromatic protein SUV39H1 and by the Polycomb group proteins M33 and BMI‐1. A concentration variation of these proteins modulates the proportion of cells expressing human globins in a locus‐dependent manner. Similarly, the transcription factors Sp1 or erythroid Krüppel‐like factor (EKLF) also influence PEV, characterized by a change in the number of expressing cells and the chromatin structure of the locus. However, in contrast to results obtained in a euchromatic locus, EKLF influences the expression of the γ‐ more than the β‐globin genes, suggesting that the relief of silencing is caused by the binding of EKLF to the LCR and that genes at an LCR proximal position are more likely to be in an open chromatin state than genes at a distal position.

Vertebrate orthologues of the Drosophila region-specific patterning gene teashirt

In Drosophila the teashirt gene, coding for a zinc finger protein, is active in specific body parts for patterning. For example, Teashirt is required in the trunk (thorax and abdomen) tagmata of the embryo, parts of the intestine and the proximal parts of appendages. Here we report the isolation of vertebrate cDNAs related to teashirt. As in Drosophila, human and murine proteins possess three widely spaced zinc finger motifs. Additionally, we describe the expression patterns of the two murine genes. Both genes show regionalized patterns of expression, in the trunk, in the developing limbs and the gut.

The gene encoding L1, a neural adhesion molecule of the immunoglobulin family, is located on the X chromosome in mouse and man.

The murine and human genes for the L1 neural adhesion molecule were shown to lie on conserved regions of the X chromosome to which genes responsible for several neuromuscular diseases have been mapped and which are adjacent to the fragile site (FRAXA) associated with mental retardation. By pulsed-field gel mapping we have demonstrated physical linkage between the L1 gene and other genes located in Xq28: L1 lies between the eye pigment RCP, GCP locus and the glucose-6-phosphate dehydrogenase (G6PD) gene. This location is compatible with the implication of the L1 molecule in one of the X-linked neuromuscular diseases mapped to this region.

Epigenetic regulation of Nanog expression by Ezh2 in pluripotent stem cells

Nanog levels in pluripotent stem cells are heterogeneous and this is thought to reflect two different and interchangeable cell states, respectively poised to self-renew (Nanog-high subpopulation) or to differentiate (Nanog-low subpopulation). However, little is known about the mechanisms responsible for this pattern of Nanog expression. Here, we have examined the impact of the histone methyltransferase Ezh2 on pluripotent stem cells and on Nanog expression. Interestingly, induced pluripotent stem (iPS) cells lacking Ezh2 presented higher levels of Nanog due to a relative expansion of the Nanog-high subpopulation, and this was associated to severe defects in differentiation. Moreover, we found that the Nanog promoter in embryonic stem (ES) cells and iPS cells coexists in two alternative univalent chromatin configurations, either H3K4me3 or H3K27me3, the latter being dependent on the presence of functional Ezh2. Finally, the levels of expression of Ezh2, as well as the amount of H3K27me3 present at the Nanog promoter, were higher in the Nanog-low subpopulation of ES/iPS cells. Together, these data indicate that Ezh2 directly regulates the epigenetic status of the Nanog promoter affecting the balance of Nanog expression in pluripotent stem cells and, therefore, the equilibrium between self-renewal and differentiation.

Laser microdissection of the fragile X region: Identification of cosmid clones and of conserved sequences in this region

Laser microdissection has been used to dissect material from the X-chromosome region involved in fragile-X-linked mental retardation. After dissection, single chromosome slices corresponding to this fragile site were subjected to DNA amplification using either a vector ligation method (to provide known anchor sequences) or primer oligonucleotides corresponding to the ubiquitous Alu sequences. Amplified material was then cloned or, alternately, used to screen a gridded cosmid library. Eight cosmid clones identified in this way were regionally mapped using a panel of hybrid cell lines and shown to originate from a narrow interval centered on the fragile X site. Two clones are included in the approximately 6-cM interval defined by probes RNI (DXS369, 5 cM proximal) and VK21 (DXS 296, 1-2 cM distal) and which includes the fragile site, and at least one clone contains sequences conserved across species suggestive of a gene. This method combines the focused approach of microdissection and the convenience of obtaining cosmid (rather than small-insert) clones; it may be useful for studies of other defined chromosomal regions.

Disruption of E2F signaling suppresses the INK4a-induced proliferative defect in M33-deficient mice

Polycomb group (Pc-G) proteins associate to form large complexes that repress Hox genes, thereby imposing Hox gene expression pattern required for development. However, Pc-G proteins have a Hox-independent function in controlling cell proliferation. Here we show that embryonic fibroblasts derived from M33-deficient mice are impaired in the progression into the S phase of the cell cycle, as shown by a reduced rate of incorporation of bromodeoxyuridine. These cells have a senescent phenotype, associated to an abnormal accumulation of the cyclin-dependent kinase inhibitor p16INK4a protein. We demonstrate that this defect is bypassed in mutant embryonic fibroblasts expressing a transdominant negative form of the cell cycle controlling transcription factor E2F (E2F-DB). In addition, we show that the polycomb protein M33 controls critical expansion of B- and T-lymphocyte precursors. Together, our results emphasize M33-Polycomb protein function in cell cycle control.

cDNA cloning, expression and chromosomal localization of the murine AF-4 gene involved in human leukemia

The AF-4/FEL (Gu et al. 1992; Morrissey et al. 1993) gene was first identified in the human by its involvement with Mll in the translocation t(4;11)(q21;q23) in acute mixed-lineage leukemia (Mirro et al. 1986; Cimino et al. 1992; Gu et al. 1992). The Mll gene is very frequently rearranged with at least 30 different loci associated with acute lymphoblastic or myeloid leukemias (Waring and Cleary, 1997). The most frequent reciprocal translocation found in infant leukemias is the t(4;11)(q21;q23) and has a poor prognosis. The leukemic cells express the stem cell markers CD34, HLA-DR, and a proB cell marker CD19 as well as the myelomonocytic marker CD15 (Raimondi et al. 1989; Mirro et al. 1986). These characteristics of the blast cells suggest that the t(4;11) occurs and transforms a multipotential progenitor cell. The mechanism by which the fusion protein is able to induce neoplasia is not fully understood, and two models are proposed. The first one implies a gain of function of the MLL/AF4 chimeric protein in which the DNA binding domain is fused to a transactivation domain of AF-4 altering the regulation of target genes (Lavau et al. 1997). Some proposed that the truncation of Mll is sufficient to induce neoplasia (Schichman et al. 1994). Recently, the creation of a fusion protein between MLL and AF9 by homologous recombination in ES cells (Corral et al. 1996) and between MLL and ENL by retroviral gene transfer into cell populations enriched in hematopoietic stem cells (Lavau et al. 1997) has demonstrated that the expression of such a fusion protein is sufficient to induce the development of tumors. Chimeric mice carrying the fusion gene developed tumors, which were restricted to acute myeloid leukemias despite the large expression domain of the Mll gene, implying that the Mll-AF9 fusion protein is present in myeloid and in lymphoid committed cells (Corral et al. 1996). This suggests that the partner genes may influence the tumor phenotype. The t(4;11), generating an Mll-AF-4 fusion product, is predominantly associated in acute lymphoblastic leukemia (Pui et al. 1986; Djabali et al. 1992; Domer et al. 1993), suggesting that the AF-4 gene is important and has a role in defining the leukemic lineage. For more insight into the function of the most frequent Mll-associated gene in leukemia, we have undertaken to further study the murine AF-4 gene in the wild-type mouse. By screening adult mouse brain and macrophage cDNA library, we isolated three cDNA clones spanning all the murine AF4 transcript (from ATG to TAA). The coding sequence of the murine AF4 cDNA is 3962 bp. Since the mouse transcript appears on Northern blot to be around 10 kb, we conclude that the rest of the molecule is corresponding to the 3 8 non-coding region, as is the case for the human mRNA (Morrissey et al. 1993). The murine AF-4 sequence contains a predicted open reading frame of 1217 amino acids, which presents 64% homology to its human counterpart of 1210 aa. However, five different domains present a stronger sequence conservation (Fig. 1A). The first is an N-terminal domain of about 66 AA (AA 6–72; 91%). The second is a 63-AA domain (AA 258–321) with 89% homology. The third domain encompasses the putative transactivation domain of the human protein (Prasad et al. 1995; Ma and Staudt 1996) and shows 86% homology (AA 405–510; Fig. 1B). The fourth domain is from AA 753 to 780. Finally, the last conserved domain is from AA 968 to 1126. Computer-assisted comparison with the mouse AF-4 protein sequence revealed that the five conserved domains are shared with two other human proteins: FMR2 and LAF-4 (Fig. 1B). FMR2 is associated with a CpG island adjacent to FRAXE, a folatesensitive site in Xq28 (Gecz J. 1996; Gu Y. 1996). The FMR2 protein is a nuclear protein involved in a non-specific mental retardation (Gecz et al. 1996; Gu et al. 1996). LAF-4 has been characterized, as a potential transcription factor with DNA-binding capacity and transactivation potential. LAF-4 possesses two potent transcriptional activation domains, one of which is conserved both in human and murine AF-4 protein (Fig. 1A). It is interesting to note that LAF-4 is probably involved in lymphocyte differentiation (Prasad et al. 1995) as it is strongly expressed in B and T cell lineages (Ma and Staudt 1996). Homologies between mAF4, AF-4, LAF-4, and FRM2, presented here, highlight the conserved domains in this family of potential transcriptional regulators. This should help in future functional analysis of this gene family. The precise chromosomal localization of the mAF-4 gene locus was assessed by fluorescence in situ hybridization of a biotinlabeled mAF-4 probe to mouse metaphase chromosome (Bonhomme and Gue ́net J.-L. 1989; Pinkel et al. 1986; Matsuda et al. 1992; Lemieux et al. 1992). In total, 50 metaphase cells were analyzed, and 95% of them showed specific fluorescent spots on the E region of murine Chromosomes 5 (Fig. 2). This region is syntenic with the 4q21 human region. This strengthens the conclusion that this is the map position for mAF4 in mouse. The expression of mAF-4 in different mouse adult tissues (6week-old mice) was assessed by Northern analysis probed with an mAF-4 cDNA fragment corresponding to a part of the coding region (clone Ia). A single transcript of ∼10 kb was detected in all the adult tissues examined. The thymus, the lymph nodes, and the kidney expressed higher levels. Weak signals were, however, detected in the spleen, bone marrow, heart, muscle, lung, and liver. The transcript was very weakly detected in the testis and the brain of adult mice (Fig. 3A). These results indicate that mAF4 is mostly accumulated in organs where T lymphopoiesis occurs but also in B cells compartment, though at a lower level. Because of the strong expression detected in the adult thymuses, mAF-4 mRNA accumulation was also studied in the developing thymus starting from day 15.5 dpc of development. As seen in Fig. 3Ba,b, mAF-4 expression is detected at this age as a single transcript; but as development proceeds and starting from day 16.5 dpc, a second transcript of higher molecular weight is Correspondence to: M. Djabali Mammalian Genome 9, 1065–1068 (1998).

Characterisation of set-1, a conserved PR/SET domain gene in Caenorhabditis elegans

The SET domain is a highly conserved domain shared between proteins of the antagonistic trithorax and Polycomb groups. It has been shown to play an important role in the assembly of either transcriptional activating or repressing protein complexes, and possesses a histone methyl-transferase activity. We report here the characterisation of the Caenorhabditis elegans gene, set-1, encoding a conserved SET-domain protein. We have analysed the developmental expression pattern of set-1 and show that maximal expression is observed early in development when set-1 is ubiquitously expressed. Its expression is more and more restricted as development progress. Gene inactivation by RNA interference shows that set-1 is an essential gene. Functional analysis of set-1 may contribute to the understanding of the molecular role of the SET domain.