The accurate duplication and transmission of genetic information is critical for cell growth and proliferation, and this is ensured in part by the multi-layered regulation of DNA synthesis. One of the key steps in this process is the selection and activation of the sites of replication initiation, or origins, across the genome. Interestingly, origin usage changes during development and in different pathologies, suggesting an integral interplay between the establishment of replication initiation along the chromosomes and cellular function. The present review discusses how the spatiotemporal organization of replication origin activation may play crucial roles in the control of biological events.
Genome duplication is an essential biological process that is tightly regulated and intricately co-ordinated with other cellular functions, including transcription, proliferation and differentiation. DNA synthesis in eukaryotes initiates at discrete sites called origins of replication that are distributed across the genome. Origins are used at characteristic times during S-phase and with particular frequencies, or efficiencies, in a population of cells, giving rise to the overall programme of DNA replication. The selection of these origins is highly complex: for instance, different cell types can exhibit distinct patterns of activation along the chromosomes (Figure 1A), and this flexibility in origin usage allows cells to adapt to developmental and environmental changes [1,2]. However, the functional impact of using specific replication programmes remains largely unexplored. In fact, the output of replication may not be limited to genome duplication, and the organization of DNA synthesis may have specific effects on distinct biological processes. In the present review, we approach this simple yet fundamental question by discussing the effects of changing the replication programme on cellular physiology.
Flexibility in replication origin activation
How many origins does it take to faithfully duplicate a genome?
Before replication initiation, a large number of sites across the genome are licensed for DNA synthesis through the assembly of pre-RCs (pre-replicative complexes). However, not all of these potential origins actually fire in each cell during S-phase, and many remain dormant and are eventually passively replicated. Hence origin activation has a stochastic component: at the single-cell level, different subsets of sites are used in different cells in a population as well as from one cell cycle to the next (Figure 1B). Despite this plasticity, overall replication programmes at the population level, which take into account the average timing of firing and efficiency of each origin in a genome, are highly characteristic for each cell type . This indicates that there is also a deterministic element in origin selection. It is possible that this control of the programme of replication contributes to the maintenance of origin density, which has been demonstrated to be important for genome integrity : both decreases and increases in the number of active origins can result in deleterious effects (see below). Is there then an ‘optimal’ number of origins for a cell to use for proper genome duplication?
A shortage of active origins: genome instability and developmental disorders
Initiating DNA synthesis with a reduced number of origins (Figure 2A) generates long inter-origin regions that increase the risk of replication fork collapse , potentially contributing to entry into mitosis with incompletely duplicated DNA . In budding yeast, inhibition of pre-RC assembly through overexpression of the G1 cyclin Cln2 or deletion of the G1 regulator and CDK (cyclin-dependent kinase) inhibitor Sic1 brings about a decrease in the number of initiation events [6,7], resulting in GCRs (gross chromosomal rearrangements). Strikingly, in the case of Cln2 overexpression, this phenotype is suppressed by the introduction of additional origins . Consistent with these results, the genomic instability linked to cyclin E deregulation found in many cancers may also be due to changes in origin activity . Indeed, disruption of the loading of MCM (minichromosome maintenance) helicase subunits in human cells by overexpression of cyclin E leads to a reduction in the number of origins used in early S-phase .
Changes in the programme of DNA replication
Interestingly, decreased origin firing is associated not only with defects in genome maintenance, but also with pleiotropic developmental phenotypes. In Drosophila, hypomorphic mutations in the ORC3 subunit of the origin recognition complex (latheo mutants) result in a reduction in volume of the mushroom body, a region of the brain involved in olfactory learning . It will be interesting to determine whether this phenotype is linked directly to changes in the pattern of replication initiation. A more clear case is that of MGS (Meier–Gorlin syndrome), which has been associated with mutations in pre-replicative complex components, including multiple ORC subunits, Cdc6 and Cdt1 [11–13]. These mutations impair pre-RC formation and result in a reduced capacity to activate origins, generating a hypocellular dwarfism phenotype that affects particular tissues . In both latheo mutants and MGS, the ultimate effect of decreased levels of pre-RC formation appears to vary depending on the cell type. This suggests that cell-specific characteristics, such as the rates of growth and division, may impose different constraints on the ‘optimal’ number of origins to use during S-phase [14,15]. Thus alterations in replication initiation can have consequences in specific developmental conditions while remaining innocuous in other situations.
It therefore appears that cells may have a ‘minimal’ number of initiation events to use for proper genome duplication, which could change depending on the developmental state. Reducing this number below cell-type-specific thresholds may lead to developmental defects, genome instability and potentially cancer.
An excess of active origins: more is not necessarily better
As is the case with activating insufficient numbers of origins during S-phase, replicating with an excess of origins (Figure 2A) may also be deleterious. For instance, the higher origin activity generated genome-wide by simultaneous overexpression of the pre-IC (pre-initiation complex) components Sld3, Sld7 and Cdc45 in budding yeast has been suggested to contribute to severe growth defects and cell death . In fact, increased origin firing may lead to higher levels of ssDNA and to a greater chance of collisions between replication and transcriptional machineries, which may result in damage to the genome [17,18]. Similarly, induction of origin firing in fission yeast through elevating the levels of Hsk1/Cdc7, the DDK (Dbf4-dependent kinase) critical for pre-IC assembly, is accompanied by greater genome instability, as determined by plasmid and minichromosome loss . In human cells, the misregulation of pre-IC assembly may also have negative effects, as hinted by the increased expression of Cdc7 found in many cancer cell lines . Indeed, an excess of initiation events has been implicated in the early steps of cell transformation: overexpression of the c-Myc oncogene gives rise to a higher density of early-replicating origins, which increases replication fork collapse and associated DNA damage [21,22]. This non-transcriptional effect of c-Myc is independent of its function in cell growth and metabolism. Taken together, these observations from diverse models show that having higher numbers of active origins does not provide a particular advantage for the cell and may, on the contrary, lead to severe defects in genome maintenance.
It is thus tempting to conclude that a copy of the genome is not the only input that replication provides for proper cell growth and proliferation. Cells may replicate with different numbers of active origins to provide particular functions in the context of specific physiological states or external conditions (see below for further discussion). The modulation of the number of replication initiation events during each S-phase is achieved via a conserved mechanism in which potential origins compete for limiting replication factors [23,24]. Through this regulation, cells may set the limits for origin activation, ensuring genome integrity as well as normal development and differentiation.
Specific changes in the pattern of origin usage
Up to this point, we have focused on the deleterious effects of reducing or increasing the overall number of initiation events during S-phase, but alterations in the pattern of replication that differentially affect distinct groups of origins (Figure 2A) can also have an impact on cellular function. An interesting example is provided by the Rif1 telomere-binding protein, which is often mutated in breast cancer cell lines [25,26]. In fission yeast mutants of rif1, a subset of late-firing and dormant origins is activated earlier, whereas some early-firing and efficient origins are delayed or not used . A similar phenotype is observed with the human and mouse homologues of Rif1, which function to establish replication timing domains [28,29]. Thus one appealing possibility is that deregulation of Rif1 may play a role in cellular transformation in part through changes in the replication pattern. Along with the observation that cancer cells generally show altered replication timing domains , this suggests that specific changes in the programme of origin usage may be detrimental for cellular physiology.
However, modifications of origin selection can play an integral part in complex biological processes. In metazoa, the genome is organized in domains that replicate at characteristic times throughout S-phase, and changes in this timing of replication have been observed during development and differentiation . For instance, the genome-wide replication patterns of mouse and human ES (embryonic stem) cells change upon differentiation, and specific replication profiles are correlated with distinct cell types [31,32]. Although this may simply be a consequence of modulating various signalling pathways, a more intriguing possibility is that the replication programme contributes directly to critical physiological transitions . One example that supports this idea can be found in the Drosophila ovary: during oogenesis, follicle cells utilize a modified replication pattern in which only six of the origins in the genome are fired in successive rounds through a process of re-replication . This results in the amplification of a cluster of chorion genes, promoting the production of high levels of chorion proteins, which are required for eggshell formation and egg development . In this system, a regulated modification of origin usage is therefore crucial for cell function.
The results of these studies suggest that programmed alterations in origin usage play a key role in cellular processes. Sufficient flexibility is required in order to respond to developmental signals and to promote distinct functions, but, at the same time, limiting plasticity may be important for preventing uncontrolled deregulation of essential pathways. If specific replication programmes have evolved to support different processes, then this may place additional constraints on the plasticity of origin usage.
Modulating origin selection in response to external conditions
As discussed above, alterations in origin selection may be programmed events that contribute to key physiological events. Could the flexibility in origin usage also be utilized as part of an immediate and targeted response to changes in external conditions? One example of this type of modulation occurs when cells are exposed to replication stress such as nucleotide depletion (Figure 2B). In this situation, the number of stalled forks and the abundance of stretches with ssDNA must be limited. One strategy to achieve this is (i) to finish duplicating regions that have already started DNA synthesis through activation of additional origins, and (ii) to prevent new initiation events in areas that have not yet begun replicating. On one hand, cells take advantage of an excess of assembled pre-RCs across the genome, locally activating back-up dormant origins [36,37]. On the other hand, a number of late-firing origins are inhibited by the S-phase checkpoint in areas devoid of prior initiation events [38–40]. These complementary aspects of the regulation of origin activation are a crucial part of the response to challenges to DNA synthesis.
Recent studies in which origin licensing is restricted have begun to reveal how the inability to alter the replication programme in response to stress conditions affects genome maintenance. In human cells and in Caenorhabditis elegans, decreasing the levels of MCMs causes hypersensitivity to replication stress due to the absence of firing of back-up origins. This results in lethality even at low, normally non-toxic, levels of the replication inhibitor hydroxyurea [36,37]. In contrast, a reduction in MCM levels has no observable effect on a normal S-phase, showing that dormant origins are critical specifically in response to challenging conditions. But what happens in genomic regions in which potential additional initiation sites are rare? This is likely to be the case at CFSs (common fragile sites), which are prone to forming double-strand breaks upon exposure to replication stress . The fragility of these sites has been linked to an absence of dormant origins  as well as to a lack of initiation events . As the instability of CFSs is considered to be an early step in oncogenesis , there may be additional mechanisms to limit their negative effects. Indeed, the fragility of a CFS is cell-type-specific and likely to be dependent on the particular replication programme used in those conditions . All together, these findings highlight the importance of the flexibility and buffering capacity of the replication programme.
The regulation of origin usage in conditions of replication stress integrates the different ideas that we have discussed in the present review: controlled alterations in the replication programme give rise to a modified ‘optimal’ number of active origins in particular regions of the genome, contributing to its faithful duplication and maintenance.
In the present review, we have discussed the biological relevance and significance of the programme of origin firing. Genome integrity and proper cell function require the maintenance of a minimal yet limited number of active origins, and these thresholds are altered depending on the external conditions or the physiological state of the cell. In addition, although uncontrolled deregulation of the patterns of replication initiation have negative consequences, they can be important for development and differentiation. Finally, the flexibility of the replication programme allows cells to induce immediate responses to external stimuli, such as challenges to DNA synthesis, promoting genome stability. All this points toward an integral role for specific replication programmes in cellular processes.
Paradoxically, whereas origin activation can be a source of damage that may lead to disease if it is improperly regulated, it is also considered to be a contributing force for evolution. Early-firing origins in yeast have been co-localized with breakpoints that generate gene amplifications and chromosome rearrangements [45,46], suggesting an intriguing role for origin selection in genome architecture. The plasticity of the programme of replication may therefore allow for a delicate but dynamic balance between genome maintenance and genome evolution.
The 7th International Fission Yeast Meeting: Pombe 2013: An Independent Meeting/EMBO Conference held at University College London, London, U.K., 24–29 June 2013. Organized and Edited by Jürg Bähler (University College London, U.K.) and Jacqueline Hayles (Cancer Research UK London Research Institute, U.K.).
We thank Damien Coudreuse for a critical reading of the paper and insightful discussions before submission. We also thank members of the Genome Duplication and Maintenance laboratory for helpful suggestions. We apologize to any authors whose work was not cited owing to space restrictions.
This work was supported by the Association for International Cancer Research [grant number 11-0685], the Fondation pour la Recherche Médicale and the Action Thématiques et Incitatives sur Program (ATIP)/Avenir programme of the Centre National de la Recherche Scientifique (CNRS)–Institut National de la Santé et de la Recherche Médicale (Inserm).