Contents
Replication timing regulation of eukaryotic replicons:Rif1 as a global regulator of replication timing
DNA複製タイミング制御の分子機構
Abstract
Origins of DNA replication on eukaryotic genomes have been observed to fire during S phase in a coordinated manner. In higher eukaryotes, they are clustered into sets of neighboring replication origins that fire at specific times within the S phase. Studies in yeast indicate that origin firing is affected by a number of factors, including checkpoint regulators and chromatin modifiers. However, it is unclear what the mechanisms orchestrating this coordinated process are. Recent studies have identified factors that regulate the timing of origin activation. Among them is Rif1, which plays crucial roles in the regulation of the origin firing program in fission yeast as well as in higher eukaryotes. In mammalian cells, Rif1 appears to regulate the structures of replication timing domains, each of which may replicate at defined timings during S phase, through its ability to organize the chromatin loop structures. Replication timing regulation would be associated with local chromatin structures as well as with more global chromatin architecture, and thus may be linked to regulation of other chromosome transactions including recombination, repair or transcription. This review summarizes recent progress in the effort to elucidate the mechanism of replication timing regulation of eukaryotic replicons.
Origins of DNA replication
Marking and activation-DNA replication is initiated at defined loci known as replication origins. In prokaryotic replicons, replication is initiated from a single locus in most cases, and the sequence specificity of origin activation is very high; generally one base substitution within origin leads to loss of initiation [1]. By contrast, replication initiates at multiple loci on eukaryotic genomes [2]. Although initiation occurs within specific loci on each genome at specific times during S phase, the sequence specificity can be significantly relaxed compared to the bacterial systems. It appears that cells prepare many potential origins for possible uses during S phase, but only a subset are utilized during the normal course of S phase. Other origins may be used at later stages of S phase or may not be used at all (dormant origins).
It is well established now that preparation for DNA replication starts as early as late M or early G1 with assembly of pre-RCs (pre-replicative complexes) at the selected locations on chromosomes. This step, also called origin licensing, proceeds through the step-wise assembly of Orc, Cdc6 and Cdt1-Mcm, resulting in the loading of Mcm (helicase loading) onto the chromatin. The selected pre-RCs are activated by the actions of Cdc7 kinase and Cdk when cells enter S phase [2,3]. Once in S phase, origin licensing is strictly inhibited by layers of mechanisms that prevent rereplication [4]. These mechanisms are largely conserved from yeasts to human. Regulation of origin firing during S phase-Once S phase is initiated, origins are fired in a coordinated and regulated manner, until the entire genome is replicated. There are origins every 50-150 kb; about 300 in budding yeast, about 1100 in fission yeast, and more than 20,000 in human (Figure 1). These origins are fired in specific orders. Yeasts (budding and fission yeasts) have served as excellent model organisms for the study of regulation of origin firing due to its small genome size and ease of genetic manipulation. Thus, the precise locations of all the origins and the order by which they are fired have been known (Figure 1). Mechanisms that may regulate the timing of origin firing in yeasts have been the focus of many studies and are discussed below. In metazoans, firing of origins appears to be regulated on a domain basis [5,6], that is, clusters of nearby origins present in the same domain may be spatially and temporally coregulated [7,8]. Recent genomics studies demonstrated the presence of cell-type specific “replication domains” that define the segments of the chromosomes containing the coregulated origins (ranging in size from several hundred kilobases to megabase) [9]. It has been elusive how these replication domains are generated on chromosomes and how they are regulated in different cell types [9,10]. Rif1, originally identified in budding yeast as a Rap1-interacting factor involved in telomere length regulation [11], has recently come into the spotlight because of the unexpected discovery of its participation in origin regulation.
In this article, we will first summarize various factors and conditions that regulate the origin firing/ replication timing program in various eukaryotes (see Table 1 for a list; see Supplemental Table for factor nomenclature in various species.) We will then discuss the newly identified Rif1, which may be a global regulator of replication timing domains in metazoans.
Regulation of the origin firing/ replication timing program
Origin firing timing and replication timing-Origin firing timing and replication timing are not identical. One can determine the firing timing of a given origin in a given cell, which can be at a certain time during S phase. Replication timing at a certain genome position could be similar to the firing time of the nearest origin, but could be quite different if the nearest origins are far away. Firing efficiency of an origin can be very low, and a given origin may be fired in only a small subset of the cells during a given S phase. Thus, the average timing of DNA replication (usually replicated passively from origins further away) can be quite different from the time of the origin firing. However, firing occurs more or less simultaneously in the clusters of the neighboring origins. Thus, even though firing efficiency of each origin is low, the segments harboring early-firing origins are most likely to be replicated early and those harboring late-firing origins late. Therefore, in this article, origin firing program and replication timing program are generally used interchangeably unless specified otherwise. However, special note should be made on the timing transition region (TTR) that separates two distinct timing zones. TTR, identified more notably on the chromosomes of higher eukaryotes, is generally composed of a long stretch of chromosome with suppressed origins and is replicated unidirectionally [12,13]. The replication timing of TTR can obviously be different from the firing timing of the even nearest origin.
Possible mechanisms-For simplicity, let us assume there are two classes of replication origins; early- and late-firing. There are two ways to distinguish these two classes of origins. The first is to mark the early-firing origins. The chromatin structures in which the origins are embedded could dictate early-firing. Alternatively, covalent modification of the pre-RC components or other factor(s) that selectively associate with specific origins prior to the initiation could mark early-firing origins (Figure 2A). This “marking” step would be required or at least highly facilitating for firing at early-S. The late origins may need to be marked later in S phase for initiation. Alternatively, firing of early origins may somehow cause changes of chromatin structures of the late origin segments, leading to activation of late origins. Alternatively, the initiation factors, which are limiting, are available for late origins only after the early segments are replicated. The second is to assume that all the origins are in the state that are ready to fire at the onset of the S phase (the default state). Late origins are somehow actively prevented from firing by being sequestered from interacting with replication initiation factors (Figure 2B). At later S phase, this constraint is released and late origins are fired. Combinations of these two are also certainly possible. As explained below, data supporting both mechanisms have recently been reported.
Temporal and spatial consideration-Before discussing the factors involved in the regulation, we would like to discuss the cell cycle and spatial regulation of replication timing. Using the combination of nuclei isolated from various cell cycle stages of CHO cells and extracts from Xenopus eggs that are capable of replicating the added nuclei, it has been shown that the replication timing program is established in nuclei isolated from cells that traversed a discrete point during the early G1 phase [14,15]. At this point, termed the TDP (Timing Decision Point), major chromatin repositioning, where the chromatin is relocated to its respective sub-nuclear positions, takes place in nuclei.
In mammals, it has been well known that chromatin in the nuclear interior is replicated in early S, whereas the chromatin at nuclear periphery is preferentially replicated at late S. The late-replicating inactive X chromosome allele is associated with nuclear periphery, while the early-replicating active chromosome is located in the interior of the nuclei [16]. Thus, spatial arrangement of chromosomes may play an important role in origin firing program. In budding yeast, forced cell cycle-specific dissociation of telomeres from the origins demonstrated that the decision for late activation is established between mitosis and START (corresponding to the restriction point in mammalian cells) in the subsequent G1 phase. It is also interesting to note that, once established, late origin activation can be enforced even if the telomere is released from the target origin [17,18]. Furthermore, late origins associate with the nuclear envelope during G1 phase, whereas early origins are randomly localized within the nucleus throughout the cell cycle [18]. However, the artificial tethering of an early replicating origin at nuclear membrane in budding yeast did not cause late replication of this origin [19], suggesting that the association with nuclear periphery alone is not sufficient for enforcing late replication on an early-firing origin.
Factors involved in regulation of origin firing program
Checkpoint signaling-Treatment of cells with hydroxyurea (HU) or methyl-methane sulfate (MMS) delays or blocks the initiation from the late-firing origins [20,21]. However, they are precociously activated in checkpoint mutants, rad53 or mec1 [22-27], indicating that the origin firing program is regulated by the checkpoint pathway. Tof1-Csm3-Mrc1 are conserved factors that play important roles in mediating the stalled fork signal to downstream, checkpoint machinery [28]. In mrc1∆, late or dormant origins are activated in the presence of HU or in early S phase both in budding and fission yeasts. Further analyses of various checkpoint mutants for their ability to suppress late origin firing showed that the RFCctf18 clamp loader (a complex consisting of RFC, Ctf18, Ctf8 and Dcc1) plays an important role in replication stress-induced repression of late origin firing [29]. This complex is essential for mediating replication-stress induced checkpoint signaling that results in Rad53 kinase activation, and thus plays important roles in late origin suppression. Suppression of dormant or late-firing origins by the checkpoint pathway was also reported in human cells [30].
Transcription and transcription factors-Although not very obvious in yeast [31], the correlation between active transcription and early replication has been shown in other organisms including Drosophila and mammals [32-35]. DNA combing analyses of replicated molecules at the human IgH locus showed that replication origin activation correlated with changes in the chromatin structure and transcriptional activity at different stages of B cell differentiation [36]. It appears that open chromatin or a potential for transcription activity, rather than active transcription per se, is related to replication in early S phase.
Recently, Forkhead transcription factors, Fkh1 and Fkh2, were shown to regulate the origin firing time of many replication origins in budding yeast by binding near a subset of early-firing origins and facilitating the association of Cdc45 with these origins in G1 phase, possibly by promoting origin clustering. Fkh1/2 binding sites are enriched near Fkh-activated origins and depleted near Fkh-repressed origins. Fkh1/2 interacts with ORC and potentially tether together Fkh-activated origins in trans, which may facilitate the early firing of these origins by increasing the effective concentration of replication factors [37]. These functions of Fkh1/2 are independent on their transcription activity. Although it is not known whether similar mechanisms operate in other species, transcription factors have been implicated in origin activation/ selection in mammals [38,39]. In replication initiation at oriP, the origin of Epstein Barr Virus, EBNA1 transcription factor was shown to facilitate the recruitment of ORC [40].
Chromatin structures-Histone modifiers play crucial roles in regulation of origin firing program. For example, Sir3, which interacts with the histone deacetylase Sir2, was shown to delay or suppress firing of subtelomeric origins in budding yeast [41]. Sir2 suppresses pre-RC formation through a specific sequence present near a subset of origins by generating unfavorable chromatin structures for pre-RC assembly [42]. In human cells, HBO1, a histone acetylase, that binds to Cdt1 [43], was shown to promote pre-RC assembly through acetylation of histone H4, which is inhibited by Geminin known to play a crucial role in prevention of rereplication [44]. It was also reported that the artificial tethering of HAT1, the Drosophila homologue of HBO1, to an origin presumably leads to efficient pre-RC formation, which could in turn cause early-firing at this origin in S phase [45] (see below). The histone H3 deacetylase Rpd3, a component of a gene-specific transcriptional repressor, is arguably the most important histone modifier with regard to origin firing. It exerts a profound effect on genome-wide origin firing profiles in a manner independent of checkpoint signaling [46-48]. Correlations between regional transcriptional activation and origin deregulation suggested that hyper-acetylation near origins caused by the lack of Rpd3 facilitates both transcription and the origin activation process potentially through induction of open chromatin structures. It should be noted that chromatin structures can affect efficiency of both formation and activation of pre-RC. Although the efficient pre-RC formation can lead to early-firing events, its mechanism is elusive. Therefore, we will mostly focus on replication timing regulation that acts at the stage of the pre-RC activation (firing step).
Replication timing of chromosomes with special chromatin structures may be regulated by distinct mechanisms. The fission yeast centromere segments are replicated early in spite of their heterochromatin character (Figure 1). This is made possible by Swi6/HP1 which recruits the kinase Hsk1 (the Cdc7 homologue) through interacting with the Dfp1/Him1 subunit (Dbf4 homologue) of the Hsk1 kinase complex [49]. Telomere/ subtelomere segments are replicated very late in S phase in spite of the presence of clusters of pre-RCs [50] (Figure 1). Transfer of an early-firing origin on the arm region to the vicinity of the subtelomere region renders it late-firing. Special chromatin structures in the subtelomere regions may be responsible for the suppression of initiation on these regions, since disruption of telomere binding factors, known to be required for generation of telomere-specific chromatin structures, deregulates replication and causes theses regions to replicate early [51-53]. It was also reported that telomere shortening by the yKu mutation stimulates the firing in the subtelomere segment up to ~40 kb from the end [54]. These results indicate chromatin structures are major regulatory elements for origin firing.
On the other hand, effect of replication timing on chromatin structures have been argued [35]. The process of DNA replication can induce changes of chromatin status by altering the histone modification patterns or spatial arrangement of chromatin in nuclei. The detailed genome-wide studies on replication timing and chromatin features in human cells indicated the presence of the correlation of chromatin modification features and replication timing in TTR [55], a long stretches of chromosome that is unidirectionally replicated. This correlation is very similar to what is found in the other constant replication timing segments, arguing in favor for the idea that replication timing affects the chromatin structures.
Other factors-Mrc1 is a checkpoint adaptor protein and late origins are fired in the presence of HU in mrc1∆ due to loss of checkpoint. It was reported that weak early-firing origins are stimulated in the fission yeast mrc1 mutant proficient in checkpoint responses, suggesting that Mrc1 regulates early origin firing also in a checkpoint-independent manner as well, although its precise mechanism is unknown. Mrc1 selectively binds to early-firing origins at the pre-firing stage at the onset of S-phase, leading to a proposal that Mrc1 may mark the early-firing origins [56].
A simpler model was also proposed. In fission yeast, early recruitment of Orc to chromatin during the preceding M phase facilitates pre-RC formation in G1 as well as early firing of these origins in S phase presumably through increased recruitment of initiation factors [57]. Extension of M phase resulted in more uniform levels of initiation at “early” and “late” origins, leading the authors to conclude that differential recruitment of ORC during M phase determines the timing and efficiency of origin firing. However, a possibility remains that the altered chromatin structures caused during the extended M phase are responsible for the change of origin activation.
In both budding and fission yeasts, late-firing or dormant origins can be precociously activated by overexpression of some of the replication factors including Cdc7-Dbf4, Sld3 and Cdc45, which are limiting in numbers and appear to be preferentially utilized by “early” origins [58,59]. It is not known whether this is the case in higher eukaryotes.
In summary, whereas a number of factors may regulate the origin firing program, local chromatin structures are likely to be one of the most crucial elements in its determination. In yeasts, preferential recruitment to the early-firing origins of the limiting factors could determine the origin firing pattern. This could be facilitated not only by preferred chromatin structures but also by origin tethering mediated by a factor like Fkh or by origin marking with some early origin-specific binding factor such as Mrc1. Late origins may be available for firing at early-S phase but are fired only after early origins are fired and limiting factors become available. On the other hand, another model, which is not exclusive with above, hypothesizes that the late origins are actively suppressed by some unknown mechanisms, permitting the firing of only the origins present in the early-replicating segments. In the following sections, we will discuss a novel factor that might participate in this latter mechanism.
Rif1 as a novel regulator of genome-wide replication timing
Fission yeast cells-In order to understand the molecular basis for the replication origin firing program, it would be important to identify the crucial factors that may determine the genome-wide pattern of early- and late-firing of origins. In budding yeast, the Cdc7 kinase is a critical regulator of origin firing and may determine which potential origins are activated. In the absence of Cdc7, firing efficiency is too low to support the completion of S phase under normal growth conditions. Similarly, fission yeast hsk1∆ (encoding the fission yeast homologue of Cdc7) exhibit a delayed S phase and eventually die [60,61]. However, a variety of genetic or physiological conditions (mrc1∆, cds1∆, and high temperature etc.) allow hsk1∆ cells to survive and form colonies ([62]; see Box 1). Under these conditions, the origin-firing program is modified and firing occurs in the absence of Hsk1 kinase. Systematic screening for mutations that permit growth of hsk1∆ cells lead to the identification of Rif1 (Rap1 interacting factor). rif1∆ efficiently restores growth of hsk1∆, and this bypass appears to be distinct from that of checkpoint mutants (Box 1)[63].
Rif1 was originally discovered as a telomere binding factor in budding yeast, and is involved in telomere length regulation in both budding and fission yeasts (telomere elongation in the absence of Rif1). It binds to subtelomere regions and suppresses their replication. In rif1∆ of fission yeast, not only the subtelomere regions but also many late or dormant origins on the arms of the chromosome are vigorously fired in the presence of HU (Figure 1). Some origins that are not fired in checkpoint mutants are fired in rif1∆. Indeed, twice as many origins are activated in rif1∆ than in checkpoint mutants, including 90% of those activated in the latter. More interestingly, more than 100 early-firing origins are downregulated or their firing timing is delayed in rif1∆ cells, suggesting that Rif1 negatively regulates late-firing origin and positively regulates early-firing origins. Thus, it appears that the combination of these opposing effects results in replication timing that occurs most frequently in mid-S phase in rif1∆. A possibility remains, however, that extensive deregulation of late-firing/ dormant origins in rif1∆ causes titration of limiting initiation factors, resulting in reduced firing of origins normally fired early. Together, these results indicate that Rif1 is required for establishing genome-wide origin firing program in fission yeast [63].
Rif1 binding sites identified by ChIP-chip analyses do not generally overlap with the known origins in yeast. However, Rif1 tends to bind at sites closer to the upregulated origins than to the downregulated origins. Analyses of sequences of the binding segments near the upregulated origins revealed the presence of a conserved sequence, although its functional significance is not clear. Rif1 deletion does not affect the pre-RC formation but the loading of Cdc45 onto the pre-RC is either stimulated or decreased at the upregulated or downregulated origins, respectively [63].
The telomere binding protein, Taz1 (a homologue of human TRF1/2), is also involved in the origin activation program in fission yeast [53]. Deregulation of origin firing was observed at telomeres as well as some of the non-telomeric late origins in the arm of the chromosomes in taz1∆. Sequences similar to Taz1 binding sites were identified near the suppressed origins and these sequences were shown to be required for suppression. Taz1 may enforce late origin firing through binding to these telomere-like repeats present near the affected origins [53]. Unlike rif1∆, suppression of early-firing origins was not observed in taz1∆. All the Taz1-regulated origins are also regulated by Rif1, suggesting that suppression of late origins by Taz1 involves the function of Rif1.
Mammalian cells- In mammalian cells, clusters of origins are activated at distinct times [64], and they constitute chromosome, so called “replication (timing) domains” that are replicated at specific timings within the S phase. Replication domains were initially observed as adjacent chromosome segments that incorporated [3H]-thymidine asynchronously during the S phase. Recent genome-wide studies revealed the presence of distinct replication timing domains that presumably contain sets of temporally coregulated origins and their distribution pattern on the entire chromosomes has been clarified in many cell types [65](Figure 1).
It has long been known that replication occurs at specific locations within nuclei [66]. These foci, detectable by incorporated nucleotide analogues, adopt characteristic patterns at distinct stages of S phase. Most notably, foci are detected at nuclear and nucleoli periphery at the mid-S phase stage, whereas foci are scattered uniformly over nuclei during early-S phase [64]. These foci may correspond to each replication domain replicating at a distinct time within S phase. Recent studies demonstrate that replication timing profiles change during development [9]. During the differentiation of ES cells into neural stem cells, changes in replication timing occur in ~ 20% of the genome, often resulting in a shift of the boundary or consolidation of replication domains into a large coordinatedly regulated regions [9]. Similarly, the genome-wide patterns of replication domains change significantly in different cell types [10].
Mammalian Rif1 does not play a role in telomere maintenance but it is required for establishment of replication timing domains in mammalian cells [67,68]. The most striking effect of Rif1 depletion in mammalian cells is the loss of the mid-S phase specific replication foci pattern. In Rif1-depleted cells, an early-S phase like foci pattern prevails throughout S phase, and late-S phase pattern, characterized by replication foci at the heterochromatin segments, appears at the end of S phase. Replication timing profiles also undergo dramatic changes. Genome-wide analyses in Rif1 knockout mouse MEF cells indicated that both early to late and late to early changes in replication timing occurred in over 40% of the replication segments, resulting in fragmentation of replication timing domains [68]. Knockdown of Rif1 in human cancer cells led to marked alteration of replication domain profiles in the 42Mb segment of the human chromosome 5. These changes are complex, including late to early and early to mid conversion and consolidation of domains (fusion of smaller domains)[67]. Generally, the distribution of replication timing occurred on average in mid S-phase in the absence of Rif1, suggesting that the replication timing regulation is generally lost, consistent with the analyses of rif1∆ fission yeast cells. During early-S phase, Cdc7-mediated phosphorylation and chromatin loading of replication factors are enhanced in Rif1-depleted cells, most likely pointing to the fact that more pre-RCs gain access to replication initiation factors and are fired, albeit during a limited time window. Rif1 is localized in nuclease-insoluble structures within nuclei and maintains chromatin loops and facilitates nucleosome formation. Rif1 localization overlaps with mid-S replication foci, suggesting that Rif1 generates nuclear structures specifically required for establishing mid-S replication domains (see Figure 3). Chromatin loop sizes increase in Rif1-depleted cells, indicating that Rif1 is required for correct chromatin loop formation. Chromatin loop sizes may be related to the sizes of the replication timing domains, since chromatin loops can generate chromatin domains that contain synchronously firing origins and are insulated from the neighboring segment. Rif1 is highly expressed in undifferentiated ES cells, which may be related to the smaller replication domain sizes observed in these cells. In differentiated cells, Rif1 levels are much lower, while replication domain sizes are increased [9].
During the mitotic phase, Rif1 dissociates from chromatin and re-associates with it at late M/ early G1 in a manner that is resistant to nuclease treatment. Thus, Rif1 must generate mid-S replication domain structures by early G1. The TDP was reported to occur during early G1 concomitant with chromatin repositioning [14,15]. It is an intriguing possibility that Rif1 may play a role at the TDP in establishing replication timing domains.
A model for regulation of replication timing domains in mammalian cells
We propose that the association of potential origins with replication factories triggers initiation and that this process is dynamic and stochastic to some extent (Figure 3). In early-S phase replicating domains, replication origins are not constrained in G1 phase, and upon entering S phase, multiple origins may be tethered at a replication factory and simultaneously fired for initiation. The choice of which origins are brought to the factories may be made stochastically, although origins in “open chromatin” structures may have higher probability of being recruited and firing. In mid-S phase replicating domains, Rif1, which is bound to nuclear insoluble structures (e.g. nuclear lamina), tethers the chromatin and forms a loop (“Rif1-loop”) containing many potential origins. Association of these origins with replication factories (similarly present at the nuclear insoluble structures) may be inhibited by Rif1. At mid-S phase, this inhibition is released and the mid-S phase origins are now able to interact with the replication factories to initiate replication. The choice of origins to be activated could be stochastic but might be influenced by various local epigenetic factors, as in the case for early-firing origins. In the absence of Rif1, the Rif1-loops are not generated and the mid-S phase replication domains are released from constraint, and thus the origins, normally prevented from being activated, are brought to replication factories and fired at early S phase. Under these conditions, more origins would compete for limiting replication factors and some of the origins normally firing early may be competed out by more “active” origins, resulting in delayed or reduced firing, although more positive roles of Rif1 in generating early-S replication domains cannot be ruled out. Several unsolved issues include (i) what tethers replisomes to the replication factories in early S phase? (ii) how does the Rif1-loop prevent the origins from firing in early S phase? (iii) How is this inhibition released to permit the firing of origins in mid-S phase? (iv) How does Rif1 (close to 2500 amino acids in human Rif1 without any notable motifs except for Heat repeats at the N-terminus) promote chromatin loop formation? (v) Does localization of Rif1 at nuclear periphery play any roles in preferred localization of late replicating chromatin close to nuclear membrane? and (v) are there similar mechanisms by which the firing of late-firing origins (associated with heterocrhomatin) are inhibited until late S phase?
Biological significance of replication timing
The regulation of the replication timing program is likely to be evolutionarily conserved because (i) factors regulating origin firing/ replication timing are conserved, and (ii) in animal cells, replication domain structures are conserved between different species in spite of significant divergence of primary sequences and rearrangement of genomic segments [65,69]. However, the apparent normal growth of rif1∆ fission yeast despite severely misregulated origin firing patterns suggests that cells can tolerate aberrant origin activation. Similarly, Rif1-depleted cancer cells do not show apparent defects in S phase progression despite the genome-wide alteration of replication timing program. These observations are in line with the speculation that the order and spatial location of origin firing in nuclei are quite flexible and adapt to changes in the environment.
However, the biological significance of replication timing regulation in eukaryotes should not be dismissed altogether. Regulated timing may function to avoid too many replication forks at one time that would cause a shortage of nucleotide pools, resulting in stalled replication forks and eventually in DNA damage. Precise replication timing may facilitate the coordination with transcription timing, which may be required for proper progression of the cell cycle. Early replication may facilitate more protein synthesis of the genes present on the early-replicating segments due to doubled copy numbers and this increased expression may be needed for progression through the cell cycle. As discussed above, replication timing could affect the chromatin structures, which may in turn affect various chromatin activities including transcription, recombination and repair. Special mid-S phase replication timing domain may play roles for cell-type specific chromatin packaging or efficient condensation of chromosomes. Indeed, the Rif1 knockout MEF cells grow poorly with extended S phase [68]. Finally, it was reported that the mutation frequency varies during S phase. (see Box 2) It is higher in late S phase than in early S phase, although the significance of differential mutation rates regulated by replication timing is not clear.
Concluding remarks
Precise and complete replication of the entire genome is crucial for cell growth and survival. Many different measures are taken to ensure that replication occurs once and only once. After entering S phase, long chromosomes need to be replicated in an ordered manner. Regulation of the timing of replication and the nuclear localization of chromosome are both important for efficient and coordinated replication of the entire chromosome.
In contrast to the prokaryotic genomes which typically carry only a single, very efficient origin, eukaryotic chromosomes carry many potential origins that are in many cases much less efficient. In addition, any given origin is utilized only in a fraction of the cells and cell cycles [7,8]. These features may provide eukaryotic cells with plasticity and adaptability to respond to various environmental or epigenetic changes [9,62,70]. Although the firing at each origin may be stochastic, the order of origin firing and timing of replication in a population of cells are predetermined and this feature is evolutionally conserved. Chromatin structures may play a major role in determination of the activity of each origin. Generally, “open chromatin” configurations characterized by “transcription-permissive” histone marks promote early replication.
We propose the presence of two layers of regulation for origin firing timing. The first is on the level of replication domains of several hundred kb to megabase. The domains that are insulated by Rif1 may be suppressive for initiation at early S phase and defines mid-S replication domains. There may be early-firing origin domains defined by factors such as Fkh1/2. Another would be on the level of individual replicons or smaller clusters of replicons, the firing of which may be affected by local chromatin structures. This would determine which origin would be used for firing in each individual cell in each cell cycle, and the selection process could bear stochastic nature (Figure 3).
The mid-S replication “factories” are very unique in their nuclear distribution and it is noteworthy that the factories appear to be generated as early as at late M/ early G1. The pre-RC, which is generated also at a similar timing during the cell cycle, is not affected by the absence of Rif1 both in fission yeast and in human cells, and thus, origin licensing and so-called “mid-S licensing” occur independently at a similar timing during the cell cycle, and may constitute the two major steps for regulated and coordinated replication of the genome during S phase.
Rif1 has been shown to interact with transcription factors, and was also implicated in DNA damage checkpoint pathways [71-78. The chromatin architecture generated by Rif1 may regulate not only DNA replication, but also other chromosome transactions including transcription. Indeed, the transcriptional profile changes significantly in Rif1-depleted cells in fission yeast and human cancer cells (although significant changes were not observed in MEF cells). Thus, it would be interesting to examine the roles of Rif1 in transcription, recombination, repair and other related processes. The roles of mammalian Rif1 in DNA damage responses have been investigated in detail and is well-established. It was recently reported that Rif1 is required for non-homologous end joining and inhibits DSB end resection [73-76].
Rif1 is essential for embryonic development (Note: in one genetic background, male Rif1 knockout mice were viable for reasons still unclear [75]), and this may be due to aberrant transcription profiles in knockout mice. The importance of Rif1 in early development is further underscored by its high expression level in undifferentiated embryonic stem cells [77; its level quickly decreases upon induction of differentiation [78]. This high level of Rif1 protein is required to maintain the undifferentiated state [77,78]. Thus, the chromatin architecture defined by Rif1 may be established early in development and continues to influence various biological events throughout the life of the organism. Further elucidation of its roles would help clarify the biological significance of replication timing regulation and explain the connection between the Rif1-generated chromatin loops and origin firing.
Box 1
Cdc7 kinase and its bypass mutations
Cdc7, originally identified in budding yeast in the Hartwell collection, encodes a serine-threonine kinase that seems to be essential for initiation of S phase, and was later shown to form a complex with Dbf4, identified in an independent screening as another mutant defective in S phase initiation. The complex formation activates Cdc7 kinase. Cdc7 phosphorylates Mcm in the pre-RC on the chromatin and this phosphorylation stimulates interaction of Cdc45with the pre-RC [79,80], facilitating the generation of an active helicase complex (CMG [Cdc45-Mcm-GINS] helicase) [81]. The haploid budding yeast cdc7(ts) mutant cells arrest with 1C DNA content (representing the cell cycle stage before DNA replication) at the non-permissive temperature and can be released into the cell cycle upon return to the permissive temperature. In fission yeast, hsk1(ts)（encoding the Cdc7 homologue）, at the non-permissive temperature, temporarily arrests with 1C DNA, later starts to synthesize DNA, arrest in S phase and eventually undergoes cell death [61]. In embryonic stem cells, Cdc7 knockout results in S phase arrest, G2-M checkpoint induction, and eventually in p53-dependent cell death [82]. Thus, Cdc7 is generally essential for DNA replication and for growth under normal conditions.
The first bypass mutation of cdc7∆, bob1, was isolated in budding yeast and was identified as mcm5P83L. This mutation appears to bypass the requirement for Cdc7 through modifying the structure of the Mcm complex so it can initiate replication in the absence of Cdc7-mediated phosphorylation [83]. More recently, deletion within the N-terminal segments of Mcm4 were found to bypass cdc7 [84]. In fission yeast, checkpoint mutants, mrc1∆ and cds1∆, were found to bypass hsk1∆. In mrc1∆ and cds1∆, late or dormant origins were activated early presumably due to loss of inhibitory checkpoint signals preventing replication [56,62]. Thus, increased initiation potential in checkpoint mutants appears to reduce the requirement of Cdc7 for initiation. Notably, growth at a higher temperature (37°C) also restores the growth of hsk1∆ cells, which do not grow at 30°C or lower. Firing of some dormant origins was observed in the wild-type cells at 37°C, although the precise mechanisms of this deregulation of origin firing at a higher temperature are not known [62]. Random screening of bypass mutants of hsk1∆ yielded rif1∆, which restores the growth of hsk1∆ most efficiently [63]. The efficiency of bypass appears to be correlated with the extent of late/ dormant origin firing in the presence of HU. Indeed, about two times more origins are activated in rif1∆ than in checkpoint mutants.
Box 2
DNA replication timing and diseases
Altered replication timing was reported in various diseases such as ICF immunodeficiency syndrome, Gilles de la Tourette syndrome and DiGeorge and Velocardio facial syndromes [85-87]. In some cases, transcriptional silencing and late replication were strongly correlated.
Genome-wide analyses of DNA replication timing in cancer cells from patients indicated the alteration of DNA replication timing pattern specific to cancer [88]. The changes occur along the developmentally regulated boundary of replication timing domains, supporting the concept that replication domains represent units of chromosome structure and function. In cancer cells, misregulation occurs at the level of these units [88].
Replication timing profile data are useful to detect karyotypic abnormalities and copy-number variation. Aneuploidies or translocation breakpoints could be detected as lower signal or unnaturally sharp transitions in replication. Cancer SCNAs (Somatic copy-number alterations) arise preferentially in genomic regions that have both the same replication timing and share long-range interactions in the nucleus [89].
All these data indicate that the genome rearrangements responsible for genome duplications, deletions or translocations tend to occur at the boundaries of replication domains or chromosome interaction units. These results suggest the potential of replication timing profiles as a diagnostic biomarker.
It has been suggested that there is an association between mutation rate and DNA replication timing. Initial observation suggested that cancer-related genes are frequently present at the TTR (boundary of domains with different replication timing) [90]. Similarly, SNPs are more enriched in the TTR and late replicating regions [90]. Analyses using ENCODE comprising 1% of the human genome revealed that mutation rate, as reflected in evolutionary divergence [91] and human nucleotide diversity (SNPs), is markedly increased in later-replicating segments of the human genome [92]. Analyses of somatic mutations in > 400 cancer genomes also showed that the frequency of somatic SNPs increases with replication time during the S phase. The late-replicating regions are enriched for recurrently mutated genes but are depleted of cancer driver genes. A higher mutation rate in the late replicating chromosome segments was also reported in yeast [93].
The mechanisms for increased mutation rate in the late replicating regions are not clear. The depletion of nucleotide pools in late S phase may cause increased incidences of fork arrest. Alternatively, replication through heterochromatic regions may pause at obstacles for fork progression. The stalled replication fork could generate unstable single-stranded DNA segments that could be the cause for mutagenesis.
Acknowledgments
We would like to thank Dave Gilbert for critical reading of the manuscript and discussion, Yutaka Kanoh for the original drawing used in Figure 1, and Seiji Matsumoto, Yutaka Kanoh, Michie Shimmoto and all other members of our laboratory for collaboration and helpful discussion. We also thank Claire Renard-Guillet and Katsuhiko Shirahige for informatics analyses of microarray data in fission yeast. Finally, we would like to express our deepest gratitude to the editor and reviewers for comments and extensive corrections of the manuscript. The research in our laboratory was supported by grants-in-aid for scientific research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
References
1. Katayama, T., et al. (2010) Regulation of the replication cycle: conserved and diverse regulatory systems for DnaA and oriC. Nat. Rev. Microbiol. 8, 163-170
2. Masai, H., et al. (2010) Eukaryotic chromosome DNA replication: where, when, and how? Annu. Rev. Biochem. 79, 89-130
3. Labib, K. (2010) How do Cdc7 and cyclin-dependent kinases trigger the initiation of chromosome replication in eukaryotic cells? Genes Deve. 24, 1208-1219
4. Arias, E.E., and Walter, J.C. (2007) Strength in numbers: preventing rereplication via multiple mechanisms in eukaryotic cells. Genes Deve. 21, 497-518
5. Farkash-Amar, S., et al. (2008) Global organization of replication time zones of the mouse genome. Genome Res.18, 1562-1570
6. Karnani, N., et al. (2009) Genomic study of replication initiation in human chromosomes reveals the influence of transcription regulation and chromatin structure on origin selection. Mol. Biol. Cell 21, 393-404
7. Cayrou, C., et al. (2010) Genome-scale analysis of metazoan replication origins reveals their organization in specific but flexible sites defined by conserved features. Genome Res. 21, 1438-1449
8. Besnard, E., et al. (2012) Unraveling cell type-specific and reprogrammable human replication origin signatures associated with G-quadruplex consensus motifs. Nat. Struct. Mol. Biol.19, 837-844
9. Hiratani, I., et al. (2008) Global reorganization of replication domains during embryonic stem cell differentiation. PLoS Biol. 6, e245
10. Hansen, R.S., et al. (2010) Sequencing newly replicated DNA reveals widespread plasticity in human replication timing. Proc. Natl. Acad. Sci. U. S. A. 107, 139-144
11. Hardy, C.F., et al. (1992) A RAP1-interacting protein involved in transcriptional silencing and telomere length regulation. Genes Deve. 6, 801-814
12. Guan, Z., et al. (2009) Decreased replication origin activity in temporal transition regions. J. Cell Biol. 187, 623-635
13. Farkash-Amar, S., and Simon, I. (2009) Genome-wide analysis of the replication program in mammals. Chromosome Res. 18, 115-125
14. Dimitrova, D.S., and Gilbert, D.M. (1999) The spatial position and replication timing of chromosomal domains are both established in early G1 phase. Mol. Cell 4, 983-993
15. Li, F., et al. (2001) The replication timing program of the Chinese hamster beta-globin locus is established coincident with its repositioning near peripheral heterochromatin in early G1 phase. J. Cell Biol. 154, 283-292
16. Wu, R., et al. (2005) Differential subnuclear localization and replication timing of histone H3 lysine 9 methylation states. Mol. Biol. Cell 16, 2872-2881
17. Raghuraman, M.K., et al. (1997) Cell cycle-dependent establishment of a late replication program. Science 276, 806-809
18. Heun, P., et al. (2001) The positioning and dynamics of origins of replication in the budding yeast nucleus. J. Cell Biol. 152, 385-400
19. Ebrahimi, H., et al. (2010) Early initiation of a replication origin tethered at the nuclear periphery. J. Cell Sci. 123, 1015-1019
20. Alvino, G.M., et al. (2007) Replication in hydroxyurea: it’s a matter of time. Mol. Cell. Biol. 27, 6396-6406
21. Poli, J., et al. (2012) dNTP pools determine fork progression and origin usage under replication stress. EMBO J. 31, 883-894
22. Shirahige, K., et al. (1998) Regulation of DNA-replication origins during cell-cycle progression. Nature 395, 618-621
23. Santocanale, C., and Diffley, J.F. (1998) A Mec1- and Rad53-dependent checkpoint controls late-firing origins of DNA replication. Nature 395, 615-618
24. Santocanale, C., et al. (1999) Activation of dormant origins of DNA replication in budding yeast. Genes Deve. 13, 2360-2364
25. Feng, W., et al. (2006) Genomic mapping of single-stranded DNA in hydroxyurea-challenged yeasts identifies origins of replication. Nat. Cell Biol. 8, 148-155
26. Raveendranathan, M., et al. (2006) Genome-wide replication profiles of S-phase checkpoint mutants reveal fragile sites in yeast. EMBO J. 25, 3627-3639
27. Bermejo, R., et al. (2011) The replication checkpoint protects fork stability by releasing transcribed genes from nuclear pores. Cell 146, 233-246
28. Katou, Y., et al. (2003) S-phase checkpoint proteins Tof1 and Mrc1 form a stable replication-pausing complex. Nature 424, 1078-1083
29. Crabbe, L., et al. (2010) Analysis of replication profiles reveals key role of RFC-Ctf18 in yeast replication stress response. Nat. Struct. Mol. Biol. 17, 1391-1397
30. Karnani, N., and Dutta, A. (2011) The effect of the intra-S-phase checkpoint on origins of replication in human cells. Genes Deve. 25, 621-633
31. Wyrick, J.J., et al. (2001) Genome-wide distribution of ORC and MCM proteins in S. cerevisiae: high-resolution mapping of replication origins. Science 294, 2357-2360
32. Schubeler, D., et al. (2002) Genome-wide DNA replication profile for Drosophila melanogaster: a link between transcription and replication timing. Nat. Genet. 32, 438-442
33. Goren, A., and Cedar, H. (2003) Replicating by the clock. Nat. Rev. Mol. Cell Biol. 4, 25-32
34. Cadoret, J.C., et al. (2008) Genome-wide studies highlight indirect links between human replication origins and gene regulation. Proc. Natl. Acad. Sci. U. S. A. 105, 15837-15842
35. Hiratani, I., et al. (2009) Replication timing and transcriptional control: beyond cause and effect–part II. Curr. Opin. Genet. Dev. 19, 142-149
36. Norio, P., et al. (2005) Progressive activation of DNA replication initiation in large domains of the immunoglobulin heavy chain locus during B cell development. Mol. Cell 20, 575-587
37. Knott, S.R., et al. (2012) Forkhead transcription factors establish origin timing and long-range clustering in S. cerevisiae. Cell 148, 99-111
38. Oda, M., et al. (2012) Regulation of DNA replication timing on human chromosome by a cell-type specific DNA binding protein SATB1. PloS One 7, e42375
39. Minami, H., et al. (2006) Binding of AlF-C, an Orc1-binding transcriptional regulator, enhances replicator activity of the rat aldolase B origin. Mol. Cell. Biol. 26, 8770-8780
40. Moriyama, K., et al. (2012) Epstein-Barr nuclear antigen 1 (EBNA1)-dependent recruitment of origin recognition complex (Orc) on oriP of Epstein-Barr virus with purified proteins: stimulation by Cdc6 through its direct interaction with EBNA1. J. Biol. Chem. 287, 23977-23994
41. Stevenson, J.B., and Gottschling, D.E. (1999) Telomeric chromatin modulates replication timing near chromosome ends. Genes Deve. 13, 146-151
42. Crampton, A., et al. (2008) An ARS element inhibits DNA replication through a SIR2-dependent mechanism. Mol. Cell 30, 156-166
43. Miotto, B., and Struhl, K. (2008) HBO1 histone acetylase is a coactivator of the replication licensing factor Cdt1. Genes Deve. 22, 2633-2638
44. Miotto, B., and Struhl, K. (2010) HBO1 histone acetylase activity is essential for DNA replication licensing and inhibited by Geminin. Mol. Cell 37, 57-66
45. Aggarwal, B.D., and Calvi, B.R. (2004) Chromatin regulates origin activity in Drosophila follicle cells. Nature 430, 372-376
46. Vogelauer, M., et al. (2002) Histone acetylation regulates the time of replication origin firing. Mol. Cell 10, 1223-1233
47. Aparicio, J.G., et al. (2004) The Rpd3-Sin3 histone deacetylase regulates replication timing and enables intra-S origin control in Saccharomyces cerevisiae. Mol. Cell. Biol. 24, 4769-4780
48. Knott, S.R., et al. (2009) Genome-wide replication profiles indicate an expansive role for Rpd3L in regulating replication initiation timing or efficiency, and reveal genomic loci of Rpd3 function in Saccharomyces cerevisiae. Genes Deve. 23, 1077-1090
49. Hayashi, M.T., et al. (2009) The heterochromatin protein Swi6/HP1 activates replication origins at the pericentromeric region and silent mating-type locus. Nat. Cell Biol. 11, 357-362
50. Hayashi, M., et al. (2007) Genome-wide localization of pre-RC sites and identification of replication origins in fission yeast. EMBO J. 26, 1327-1339
51. Cosgrove, A.J., et al. (2002) Ku complex controls the replication time of DNA in telomere regions. Genes Deve. 16, 2485-2490
52. Lian, H.Y., et al. (2011) The effect of Ku on telomere replication time is mediated by telomere length but is independent of histone tail acetylation. Mol. Biol. Cell 22, 1753-1765
53. Tazumi, A., et al. (2012) Telomere-binding protein Taz1 controls global replication timing through its localization near late replication origins in fission yeast. Genes Deve. 26, 2050-2062
54. Bianchi, A., and Shore, D. (2007) Early replication of short telomeres in budding yeast. Cell 128, 1051-1062
55. Farkash-Amar, S., et al. (2012) Systematic determination of replication activity type highlights interconnections between replication, chromatin structure and nuclear localization. PloS One 7, e48986
56. Hayano, M., et al. (2011) Mrc1 marks early-firing origins and coordinates timing and efficiency of initiation in fission yeast. Mol. Cell. Biol. 31, 2380-2391
57. Wu, P.Y., and Nurse, P. (2009) Establishing the program of origin firing during S phase in fission Yeast. Cell 136, 852-864
58. Tanaka, S., et al. (2011) Origin association of Sld3, Sld7, and Cdc45 proteins is a key step for determination of origin-firing timing. Curr. Biol. 21, 2055-2063
59. Mantiero, D., et al. (2011) Limiting replication initiation factors execute the temporal programme of origin firing in budding yeast. EMBO J. 30, 4805-4814
60. Masai, H., et al. (1995) hsk1+, a Schizosaccharomyces pombe gene related to Saccharomyces cerevisiae CDC7, is required for chromosomal replication. EMBO J. 14, 3094-3104
61. Takeda, T., et al. (2001) Regulation of initiation of S phase, replication checkpoint signaling, and maintenance of mitotic chromosome structures during S phase by Hsk1 kinase in the fission yeast. Mol. Biol. Cell 12, 1257-1274
62. Matsumoto, S., et al. (2011) Multiple pathways can bypass the essential role of fission yeast Hsk1 kinase in DNA replication initiation. J. Cell Biol. 195, 387-401
63. Hayano, M., et al. (2012) Rif1 is a global regulator of timing of replication origin firing in fission yeast. Genes Deve. 26, 137-150
64. Dimitrova, D.S., and Berezney, R. (2002) The spatio-temporal organization of DNA replication sites is identical in primary, immortalized and transformed mammalian cells. J. Cell Sci. 115, 4037-4051
65. Ryba, T., et al. (2010) Evolutionarily conserved replication timing profiles predict long-range chromatin interactions and distinguish closely related cell types. Genome Res. 20, 761-770
66. Nakamura, H., et al. (1986) Structural organizations of replicon domains during DNA synthetic phase in the mammalian nucleus. Exp. Cell Res. 165, 291-297
67. Yamazaki, S., et al. (2012) Rif1 regulates the replication timing domains on the human genome. EMBO J. 31, 3667-3677
68. Cornacchia, D., et al. (2012) Mouse Rif1 is a key regulator of the replication-timing programme in mammalian cells. EMBO J. 31, 3678-3690
69. Yaffe, E., et al. (2010) Comparative analysis of DNA replication timing reveals conserved large-scale chromosomal architecture. PLoS Genet. 6, e1001011
70. Goren, A., et al. (2008) DNA replication timing of the human beta-globin domain is controlled by histone modification at the origin. Genes Dev. 22, 1319-1324
71. Buonomo, S.B., et al. (2009) Mammalian Rif1 contributes to replication stress survival and homology-directed repair. J. Cell Biol. 187, 385-398
72. Xu, D., et al. (2010) Rif1 provides a new DNA-binding interface for the Bloom syndrome complex to maintain normal replication. EMBO J. 29, 3140-3155
73. Zimmermann, M., et al. (2013) 53BP1 regulates DSB repair using Rif1 to control 5′ end resection. Science 339, 700-704
74. Di Virgilio, M., et al. (2013) Rif1 prevents resection of DNA breaks and promotes immunoglobulin class switching. Science 339, 711-715
75. Chapman, J.R., et al. (2013) RIF1 Is Essential for 53BP1-Dependent Nonhomologous End Joining and Suppression of DNA Double-Strand Break Resection. Mol. Cell. In press.
76. Escribano-Diaz, C., et al. (2013) A Cell Cycle-Dependent Regulatory Circuit Composed of 53BP1-RIF1 and BRCA1-CtIP Controls DNA Repair Pathway Choice. Mol. Cell. In press.
77. Loh, Y.H., et al. (2006) The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nat. Genet. 38, 431-440
78. Wang, J., et al. (2006) A protein interaction network for pluripotency of embryonic stem cells. Nature 444, 364-368
79. Masai, H., et al. (2006) Phosphorylation of MCM4 by Cdc7 kinase facilitates its interaction with Cdc45 on the chromatin. J. Biol. Chem. 281, 39249-39261
80. Sheu, Y.J., and Stillman, B. (2006) Cdc7-Dbf4 phosphorylates MCM proteins via a docking site-mediated mechanism to promote S phase progression. Mol. Cell 24, 101-113
81. Moyer, S.E., et al. (2006) Isolation of the Cdc45/Mcm2-7/GINS (CMG) complex, a candidate for the eukaryotic DNA replication fork helicase. Proc. Natl. Acad. Sci. U. S. A. 103, 10236-10241
82. Kim, J.M., et al. (2002) Inactivation of Cdc7 kinase in mouse ES cells results in S-phase arrest and p53-dependent cell death. EMBO J. 21, 2168-2179
83. Hoang, M.L., et al. (2007) Structural changes in Mcm5 protein bypass Cdc7-Dbf4 function and reduce replication origin efficiency in Saccharomyces cerevisiae. Mol. Cell. Biol. 27, 7594-7602
84. Sheu, Y.J., and Stillman, B. (2010) The Dbf4-Cdc7 kinase promotes S phase by alleviating an inhibitory activity in Mcm4. Nature 463, 113-117
85. State, M.W., et al. (2003) Epigenetic abnormalities associated with a chromosome 18(q21-q22) inversion and a Gilles de la Tourette syndrome phenotype. Proc. Natl. Acad. Sci. U. S. A.100, 4684-4689
86. D’Antoni, S., et al. (2004) Altered replication timing of the HIRA/Tuple1 locus in the DiGeorge and Velocardiofacial syndromes. Gene 333, 111-119
87. Yehezkel, S., et al. (2008) Hypomethylation of subtelomeric regions in ICF syndrome is associated with abnormally short telomeres and enhanced transcription from telomeric regions. Hum. Mol. Genet. 17, 2776-2789
88. Ryba, T., et al. (2012) Abnormal developmental control of replication-timing domains in pediatric acute lymphoblastic leukemia. Genome res. 22, 1833-1844
89. De, S., and Michor, F. (2011) DNA replication timing and long-range DNA interactions predict mutational landscapes of cancer genomes. Nat. Biotechnol. 29, 1103-1108
90. Watanabe, Y., et al. (2002) Chromosome-wide assessment of replication timing for human chromosomes 11q and 21q: disease-related genes in timing-switch regions. Hum. Mol. Genet. 11, 13-21
91. Stamatoyannopoulos, J.A., et al. (2009) Human mutation rate associated with DNA replication timing. Nat. Genet. 41, 393-395
92. Woo, Y.H., and Li, W.H. (2012) DNA replication timing and selection shape the landscape of nucleotide variation in cancer genomes. Nat. Commun. 3, 1004
93. Lang, G.I., and Murray, A.W. (2011) Mutation rates across budding yeast chromosome VI are correlated with replication timing. Genome Biol. Evol. 3, 799-811
94. Kitamura, E., et al. (2006) Live-cell imaging reveals replication of individual replicons in eukaryotic replication factories. Cell 125, 1297-1308
95. Koren, A., et al. (2010) MRC1-dependent scaling of the budding yeast DNA replication timing program. Genome Res. 20, 781-790
96. Pasero, P., et al. (2002) Single-molecule analysis reveals clustering and epigenetic regulation of replication origins at the yeast rDNA locus. Genes Deve. 16, 2479-2484
97. Zappulla, D.C., et al. (2002) Control of replication timing by a transcriptional silencer. Curr. Biol. 12, 869-875
98. Donaldson, A.D., et al. (1998) CLB5-dependent activation of late replication origins in S. cerevisiae. Mol. Cell 2, 173-182
99. McCune, H.J., et al. (2008) The temporal program of chromosome replication: genomewide replication in clb5{Delta} Saccharomyces cerevisiae. Genetics 180, 1833-1847
100. Heichinger, C., et al. (2006) Genome-wide characterization of fission yeast DNA replication origins. EMBO J. 25, 5171-5179
101. Schwaiger, M., et al. (2010) Heterochromatin protein 1 (HP1) modulates replication timing of the Drosophila genome. Genome Res. 20, 771-780
102. Katsuno, Y., et al. (2009) Cyclin A-Cdk1 regulates the origin firing program in mammalian cells. Proc. Natl. Acad. Sci. U. S. A. 106, 3184-3189
103. Jorgensen, H.F., et al. (2007) The impact of chromatin modifiers on the timing of locus replication in mouse embryonic stem cells. Genome Biol. 8, R169
104. Yokochi, T., et al. (2009) G9a selectively represses a class of late-replicating genes at the nuclear periphery. Proc. Natl. Acad. Sci. U. S. A. 106, 19363-19368
Legends to Figures
Figure 1 Replication programs of budding yeast, fission yeast, and human chromosomes
A. The locations of replication origins on chromosome VI (0.27 Mb) of budding yeast. Red, early-firing origins; green, mid-firing origins; blue, late-firing origins [93]. B. The locations of replication origins on the chromosome II (4.5 Mb) of fission yeast. Red vertical lines, early-firing origins (origins firing in the presence of HU); blue vertical lines, late-firing or dormant origins [50]. C. Replication timing profile of the human chromosome 20 (62 Mb) in K562 cells is shown. Early-, mid- and late-replicating domains are deduced from data in [10] and are shown by red, green and blue horizontal bars, respectively.
Figure 2 Possible mechanisms for determination of origin firing timing,
A. Early-firing origins are molecularly marked. Marking can be chromatin structures favorable for firing, covalent modification of pre-RC components, or association of other factor(s) that facilitate recruitment of initiation proteins. B. A hypothetical Factor X inhibits firing of late origins by preventing the recruitment of initiation factors. As a result, only the early-firing origins can get access to initiation factors during early S phase.
Figure 3 A model for regulation of replication timing domains in higher eukaryotes
Replication occurs at factories where two replisomes are held together and replicating DNA strands are passed through as bidirectional DNA synthesis proceeds, generating the loop consisting of the replicated daughter molecules (replication loops; shown in gray) [94]. In the early-replicating domains (upper), chromosomes, whose conformations are not constrained during G1, can associate with replication factories where the clusters of early origins are simultaneously replicated. In the mid-replicating domains (lower), Rif1 generates specific chromatin loop structures (which we call “Rif1-loops” to distinguish from replication loops) in G1, and origins present in the Rif1-loops are sequestered and kept inactive until mid-S phase. Rif1 associates with nuclear insoluble structures, which could hold together multiple Rif1-loops. At mid-S phase, the origins in the Rif1-loop are activated through association with the axis of the Rif1-loop, generating an active replication factory. Multiple origins could be simultaneously activated within the Rif1-loop, generating multiple smaller replicating loops. Again, the selection of origins to be activated could be dynamic and stochastic, and thus, the sizes and numbers of replication loops generated from one Rif1-loop may vary from one cell to another and from one cell cycle to next. How the origins in the Rif1-loops are kept from activation in early S phase and how they get activated after mid-S phase are unknown. We also do not know if any factors are responsible for replication loop formation during early S phase and if any factors sequester the late-replicating domains, which are not described here but show distinct spatial distribution. In the absence of Rif1, the early S phase domains are intact but the Rif1-loops are disrupted, releasing mid-S phase origins from sequestration. Thus, the majority of the chromosomes (except for the late replicating heterochromatin segments) are replicated in the early-S phase pattern throughout the S phase except for very late S phase. The replication loops in the main part of the figure are shown by single lines, even though they are made up with two daughter molecules, and both replicated and unreplicated DNAs are shown in black.
