Faithful propagation of DNA methylation patterns during DNA replication is critical for maintaining cellular phenotypes of individual differentiated cells. Although it is well established that Uhrf1 (ubiquitin-like with PHD and ring finger domains 1; also known as Np95 and ICBP90) specifically binds to hemi-methylated DNA through its SRA (SET and RING finger associated) domain and has an essential role in maintenance of DNA methylation by recruiting Dnmt1 to hemi-methylated DNA sites, the mechanism by which Uhrf1 coordinates the maintenance of DNA methylation and DNA replication is largely unknown. Here we show that Uhrf1-dependent histone H3 ubiquitylation has a prerequisite role in the maintenance DNA methylation. Using Xenopus egg extracts, we successfully reproduce maintenance DNA methylation in vitro. Dnmt1 depletion results in a marked accumulation of Uhrf1-dependent ubiquitylation of histone H3 at lysine 23. Dnmt1 preferentially associates with ubiquitylated H3 in vitro though a region previously identified as a replication foci targeting sequence. The RING finger mutant of Uhrf1 fails to recruit Dnmt1 to DNA replication sites and maintain DNA methylation in mammalian cultured cells. Our findings represent the first evidence, to our knowledge, of the mechanistic link between DNA methylation and DNA replication through histone H3 ubiquitylation.

We generated knockout mice for MCM8 and MCM9 and show that deficiency for these genes impairs homologous recombination (HR)-mediated DNA repair during gametogenesis and somatic cells cycles. MCM8(-/-) mice are sterile because spermatocytes are blocked in meiotic prophase I, and females have only arrested primary follicles and frequently develop ovarian tumors. MCM9(-/-) females also are sterile as ovaries are completely devoid of oocytes. In contrast, MCM9(-/-) testes produce spermatozoa, albeit in much reduced quantity. Mcm8(-/-) and Mcm9(-/-) embryonic fibroblasts show growth defects and chromosomal damage and cannot overcome a transient inhibition of replication fork progression. In these cells, chromatin recruitment of HR factors like Rad51 and RPA is impaired and HR strongly reduced. We further demonstrate that MCM8 and MCM9 form a complex and that they coregulate their stability. Our work uncovers essential functions of MCM8 and MCM9 in HR-mediated DSB repair during gametogenesis, replication fork maintenance, and DNA repair.

DNA methylation is a chemical modification that defines cell type and lineage through the control of gene expression and genome stability. Disruption of DNA methylation control mechanisms causes a variety of diseases, including cancer. Cancer cells are characterized by aberrant DNA methylation (i.e., genome-wide hypomethylation and site-specific hypermethylation), mainly targeting CpG islands in gene expression regulatory elements. In particular, the early findings that a variety of tumor suppressor genes (TSGs) are targets of DNA hypermethylation in cancer led to the proposal of a model in which aberrant DNA methylation promotes cellular oncogenesis through TSGs silencing. However, recent genome-wide analyses have revealed that this classical model needs to be reconsidered. In this review, we will discuss the molecular mechanisms of DNA methylation abnormalities in cancer as well as their therapeutic potential. DNA methylationDNA methylation is a chemical modification that plays an important role in the regulation of epigenetic gene expression [1-3], genomic imprinting [4-6], X chromosome inactivation [7,8], transposon (see Glossary) silencing [9], and genome stability [10,11]. DNA methylation is mainly catalyzed by three enzymes, DNMT1, DNMT3A, and DNMT3B (Figure 1). De novo DNA methylation is mainly catalyzed by the DNA methyltransferases (DNMTs) DNMT3A and 3B (Box 1), while established DNA methylation patterns are maintained by the daughter DNA through a maintenance DNA methylation mechanism during cell proliferation (Box 2). Both de novo DNA methylation and maintenance DNA methylation are important for normal development. DNMT1 inactivation and DNMT3A/3B double knockout (KO) mouse embryos show significant growth inhibition and are lethal before mid-gestation [12,13]. In contrast, DNA methylation is not necessarily required in embryonic stem (ES) cells; even when CpG methylation is completely lost by combined KO of three DNMTs Dnmt1, Dnmt3a, and Dnmt3b, there is a minimal change in phenotype in undifferentiated ES cells [14]. Although DNA methylation is a stable modification, there are also several pathways of demethylation and these pathways play an important role in the regulation of DNA methylation in various biological contexts (Box 3). In normal cells, most CpG sequences in the genome are methylated, but CpG islands and the nearby CpG island shores (the region within 2 kb of the islands) are exceptionally hypomethylated [15,16]. Many of these hypomethylated regions of DNA function as elements that regulate gene expression, such as promoters and enhancers. In addition, systematic analysis of unmethylated DNA and methylated CpG as ligands has revealed that DNA methylation promotes the binding of many transcription factors [17]. Recently, broad unmethylated regions were also reported as DNA methylation canyons [18] or DNA methylation valleys [19] that are associated with either the active histone mark H3K4me3 or the inhibitory mark H3K27me3.Accumulating evidence demonstrate that cancer cells show largely di...

The proper location and timing of Dnmt1 activation are essential for DNA methylation maintenance. We demonstrate here that Dnmt1 utilizes two-mono-ubiquitylated histone H3 as a unique ubiquitin mark for its recruitment to and activation at DNA methylation sites. The crystal structure of the replication foci targeting sequence (RFTS) of Dnmt1 in complex with H3-K18Ub/23Ub reveals striking differences to the known ubiquitin-recognition structures. The two ubiquitins are simultaneously bound to the RFTS with a combination of canonical hydrophobic and atypical hydrophilic interactions. The C-lobe of RFTS, together with the K23Ub surface, also recognizes the N-terminal tail of H3. The binding of H3-K18Ub/23Ub results in spatial rearrangement of two lobes in the RFTS, suggesting the opening of its active site. Actually, incubation of Dnmt1 with H3-K18Ub/23Ub increases its catalytic activity in vitro. Our results therefore shed light on the essential role of a unique ubiquitin-binding module in DNA methylation maintenance.

Stable inheritance of DNA methylation is critical for maintaining differentiated phenotypes in multicellular organisms. We have recently identified dual mono-ubiquitylation of histone H3 (H3Ub2) by UHRF1 as an essential mechanism to recruit DNMT1 to chromatin. Here, we show that PCNA-associated factor 15 (PAF15) undergoes UHRF1-dependent dual monoubiquitylation (PAF15Ub2) on chromatin in a DNA replication-coupled manner. This event will, in turn, recruit DNMT1. During early S-phase, UHRF1 preferentially ubiquitylates PAF15, whereas H3Ub2 predominates during late S-phase. H3Ub2 is enhanced under PAF15 compromised conditions, suggesting that H3Ub2 serves as a backup for PAF15Ub2. In mouse ES cells, loss of PAF15Ub2 results in DNA hypomethylation at early replicating domains. Together, our results suggest that there are two distinct mechanisms underlying replication timing-dependent recruitment of DNMT1 through PAF15Ub2 and H3Ub2, both of which are prerequisite for high fidelity DNA methylation inheritance.

Origins of DNA replication are licensed by recruiting MCM2-7 to assemble the prereplicative complex (pre-RC). How MCM2-7 is inactivated or removed from chromatin at the end of S phase is still unclear. Here, we show that MCM-BP can disassemble the MCM2-7 complex and might function as an unloader of MCM2-7 from chromatin. In Xenopus egg extracts, MCM-BP exists in a stable complex with MCM7, but is not associated with the MCM2-7 hexameric complex. MCM-BP accumulates in nuclei in late S phase, well after the loading of MCM2-7 onto chromatin. MCM-BP immunodepletion in Xenopus egg extracts inhibits replication-dependent MCM dissociation without affecting pre-RC formation and DNA replication. When excess MCM-BP is incubated with Xenopus egg extracts or immunopurified MCM2-7, it binds to MCM proteins and promotes disassembly of the MCM2-7 complex. Recombinant MCM-BP also releases MCM2-7 from isolated late-S-phase chromatin, but this activity is abolished when DNA replication is blocked. MCM-BP silencing in human cells also delays MCM dissociation in late S phase. We propose that MCM-BP plays a key role in the mechanism by which pre-RC is cleared from replicated DNA in vertebrate cells.

on the specific proteolysis of the cyclin B subunit, whereas the Cdc2 subunit remains present at nearly constant levels throughout the cell cycle. It is unknown how Cdc2 escapes degradation when cyclin B is destroyed. In Xenopus egg extracts that reproduce the exit from M phase in vitro, we have found that dissociation of the cyclin B-Cdc2 complex occurred under conditions where cyclin B was tethered to the 26S proteasome but not yet degraded. The dephosphorylation of Thr 161 on Cdc2 was unlikely to be necessary for the dissociation of the two subunits. However, the dissociation was dependent on the presence of a functional destruction box in cyclin B. Cyclin B ubiquitination was also, by itself, not sufficient for separation of Cdc2 and cyclin B. The 26S proteasome, but not the 20S proteasome, was capable of dissociating the two subunits. These results indicate that the cyclin B and Cdc2 subunits are separated by the proteasome through a mechanism that precedes proteolysis of cyclin B and is independent of proteolysis. As a result, cyclin B levels decrease on exit from M phase but Cdc2 levels remain constant. The cyclin B-Cdc2 kinase is a universal regulator of M phase (for review, see Nurse 1990; Nigg 1995). Its activation induces entry into M phase and its inactivation is necessary for exit from M phase. The activity of cyclin B-Cdc2 kinase is regulated primarily by the formation of a complex between the catalytic Cdc2 subunit and the cyclin B regulatory subunit. This complex is stabilized by phosphorylation of Cdc2 on Thr 161, and is kept inactive by inhibitory phosphorylation of Cdc2 on Thr 14 and Tyr 15 (for review, see Nigg 1995). The amount of Cdc2 remains relatively constant throughout the cell cycle, whereas cyclin B accumulates during interphase, reaching a peak at metaphase, and is suddenly destroyed at the exit from M phase (for review, see King et al. 1996; Townsley and Ruderman 1998). The differences in the stability of Cdc2 and cyclin B at the end of M phase is seen even though the two proteins are tightly associated in a complex prior to cyclin B degradation. Although it has not been previously demonstrated, it is probable that the cyclin B-Cdc2 complex must dissociate prior to the degradation of cyclin B subunit so that Cdc2 might escape degradation. However, when the dissociation of cyclin B and Cdc2 takes place and the actual steps involved in the regulation of this dissociation are not known.The degradation of the mitotic B-type cyclins is performed by ubiquitin-proteasome-mediated proteolysis. The N-terminal region of cyclin B contains a conserved motif called the destruction box, which serves as a signal for ubiquitination and is necessary for cell cycle-regulated proteolysis (Glotzer et al. 1991). The formation of ubiquitin conjugates requires the concerted activity of a series of enzymes that first activate ubiquitin (E1) and then recognize and transfer ubiquitin (E2 and E3) to proteins destined for turnover (for review, see Hershko and Ciechanover 1998). Cyclin B is polyubiquitinated by...

No abstract