Accèder directement au contenu

Olivier Hyrien

Eukaryotic Chromosome Replication

Our long-term goal is a quantitative understanding of whole-genome replication in multiple eukaryotes.

Research context and recent results

DNA replication is tightly regulated to accurately duplicate the genome despite endogenous and exogenous obstacles. Replication origins are "licensed" in G1 for use in the next S-phase by loading the replicative helicase in an inactive form around DNA. During S phase, protein kinases and accessory factors cooperate to activate the helicase and assemble bidirectional replisomes. Only a fraction of the helicase complexes are activated during S phase. Redundant complexes provide backup, potential origins that can rescue replication downstream of stalled forks and boost S phase completion, but are more often dislodged by freely passing forks during normal replication (a process we call origin passivation). Helicase complexes are activated at different times through S phase, which generates cell-type specific replication timing (RT) profiles.

Despite decades of investigation, the nature of mammalian replication origins is still unclear (1). To elucidate this, we mapped replication fork directionality (RFD) genome-wide in human cells by purification and strand-oriented sequencing of Okazaki fragments (OK-seq) (2). We demonstrated a fundamental mathematical link between RFD and RT profiles that was corroborated by experimental observation. RFD profiles revealed the location and extent of thousands of replication initiation and termination zones as well as unidirectionally replicating and randomly replicating regions, and genome compartments whose replication mode is constant or plastic between cell lines. We notably identified oncogene-repressed initiation zones in a model for chronic myeloid leukemia progression (3). We propose that replication first initiates at efficient master initiation zones then propagates by activation of more dispersed origins. Further characterization of dispersive initiation will require high-throughput single-molecule analyses (see below).

We have confirmed master initiation zones mapped by OK-seq with a novel origin mapping technique called EdUseq-HU (4). This study also highlighted an interesting connection between preferential initiation sites and preferential sites of DNA breakage during S phase.

To understand how loaded helicases in G1 are chosen for activation in S phase, we compared the genomic binding sites of helicases and their loader with the RFD, RT and RNA-seq profiles of the same cells (8). Interestingly, helicases are more abundant in early- than late-replicating domains yet are excluded from the body of active genes. Indeed, many initiation zones abruptly end at active gene borders. However, no comparable drop in helicase density is observed across initiation zone borders that do not flank an active gene. There, helicase density is not sufficient to predict initiation zones and RFD profile. We propose that epigenetic marks contribute to select the appropriate helicase complexes for activation, either by opening chromatin or by facilitating the recruitment of limiting helicase activation factors.

Using DNA combing, a widespread but low-throughput technique for visualizing replication of stretched DNA molecules, we have obtained evidence for dispersed replication initiaton between master initiation zones at a few exemplary loci (10). In parallel, we have developped a high-throughput methodology to simultaneously visualize in nanochannel arrays the entire length, replication tracts, and restriction map of tens of thousands of stretched DNA molecules replicated in Xenopus egg extracts (6), which shows promises for mammalian studies.

More recently, we developed FORK-seq, a method to map replication genome-wide at the single molecule level by nanopore sequencing of pulse-labeled replication intermediates with nucleoside analogs such as BrdU (7). Using artificial intelligence to precisely quantify BrdU incorporation along single DNA molecules, we detected and oriented 60,000 replication forks in the yeast genome. The results confirmed the known replication initiation and termination sites in this organism, but also revealed a large fraction of previously undetected, dispersive initiation and termination events. We believe FORK-seq has a huge potential for future DNA replication studies in many experimental systems.

Cell-population and single-molecule data contain key information to mathematically model the DNA replication process. We previously demontrated that the time-dependent rate of origin activation has a universal bell-shape in yeasts, insects, frogs and mammals. Recently, we reproduced this experimental shape by modeling genome replication as a probabilistic activation of helicases by limiting factors incorporated into replication forks and recycled upon merging of convergent forks (5). We found that the universal shape emerges from a competition between origin activation and passivation times, which predicts a novel mathematical relationship between maximum initiation rate, fork speed and helicase density that we could precisely verify in all tested eukaryotes.


High-throughput, single molecule analysis of DNA replication intermediates in Xenopus, yeast and human cells will allow us to quantitate replication origin usage genome-wide and to fully evaluate cell-to-cell heterogeneity in origin usage. We will further clarify the nature of human DNA replication origins and whether adjacent origins are activated in an independent or coordinated manner. These studies will also reveal the genome-wide distributions of single replication fork speeds, which has so far not been reported in any organism. Genetic and epigenetic analyses, integrative modelling of multiple datasets and artificial intelligence will allow us to extract the rules that govern the choice of replication initiation sites and the speed of fork progression and their links to chromatin structure, gene activity and the three-dimensional folding of the genome. The study of these fundamental mechanisms is important for a full understanding of the impact of replication perturbations on genome stability and their dire consequences in developmental diseases and cancer.

Selected publications

1. Hyrien O (2015) Peaks cloaked in the mist : The landscape of mammalian replication origins. J. Cell Biol, 208, 147-160.

2. Petryk N, Kahli M, d’Aubenton-Carafa Y, Jaszczyszyn Y, Shen Y, Sylvain M, Thermes C, Chen CL, Hyrien O (2016) Replication landscape of the human genome. Nature Comm., 7, 10208. doi : 10.1038/ncomms10208.

3. Wu X*, Kabalane H*, Kahli M, Petryk N, Laperrousaz B, Jaszczyszyn Y, Drillon G, Nicolini FE, Perot G, Robert A, Fund C, Chibon F, Xia R, Wiels J, Argoul F, Maguer-Satta V, Arneodo A, Audit B, and Hyrien O (2018) Developmental and cancer-associated plasticity of DNA replication preferentially targets GC-poor, lowly expressed and late-replicating regions. Nucleic Acids Res., Nov 2 ; 46(19):10157-10172. doi : 10.1093/nar/gky797. * co-premiers auteurs.

4. Tubbs A, Sridharan S, van Wietmarschen N, Maman Y, Callen E, Stanlie A, Wu W, Wu X, Day A, Wong N,Yin M, Canela A, Fu H, Redon C, Pruitt SC, Jaszczynszyn Y, Aladjem MI, Aplan PD, Hyrien O, Nussenzweig A (2018) Dual roles of poly(dA:dT) tracts in replication initiation and fork collapse. Cell, 174 (5) 1127-1142. doi : 10.1016/j.cell.2018.07.011.

5. Arbona JM, Goldar A, Hyrien O, Arneodo A, Audit B (2018) The eukaryotic bell-shaped temporal rate of DNA replication origin firing emanates from a balance between origin activation and passivation. Elife. 2018 Jun 1 ;7. pii : e35192. doi : 10.7554/eLife.35192.

6. De Carli F*, Menezes N*, Berrabah W, Barbe V, Genovesio A, Hyrien O (2018) High-throughput optical mapping of replicating DNA. Small Methods, 2018, Sep 11 ; 2(9):1800146, doi : 10.1002/smtd.201800146. *contribution égale.

7. Hennion M, Arbona JM, Lacroix L, Cruaud C, Theulot B, Le Tallec B, Proux F, Wu X, Novikova E, Engelen S, Lemainque A, Audit B, Hyrien O (2020) FORK-seq : replication landscape of the Saccharomyces cerevisiae genome by nanopore sequencing. Genome Biol., 2020 May 26 ;21(1):125. doi : 10.1186/s13059-020-02013-3.

8. Kirstein N, Buschle A, Wu X, Krebs S, Blum H, Hammerschmidt W, Vorberg N, Kremmer E, Lacroix L, Hyrien O*, Audit B*, Schepers A* (2021) Human ORC/MCM density is low in active genes and correlates with replication time but does not delimit initiation zones. Elife, 2021 Mar 8 ;10:e62161. doi : 10.7554/eLife.62161.

9. Ciardo D, Haccard O, Narassimprakash H, Arbona JM, Hyrien O, Audit B, Marheineke K, Goldar A. Organization of DNA replication origin firing in Xenopus egg extracts : the role of intra-S checkpoint. BioRxiv, 2020.06.22.164673 ; doi:10.1101/2020.06.22.164673

10. Blin M., Lacroix L., Petryk N, Jaszczyszyn Y, Chen CL, Hyrien O, LeTallec B (2021) DNA molecular combing-based replication fork directionality profiling. Nucleic Acids Res., en révision.

Principle of RFD mapping by OK-Seq
Principle of RFD mapping by OK-Seq
Replication directionality revealed by OK-seq
Replication directionality revealed by OK-seq
Two corroborating fluorescent methods for replication bubble (...)
Two corroborating fluorescent methods for replication bubble detection
Simultaneous visualization of DNA (blue), replicated tracks (red) and (...)
Simultaneous visualization of DNA (blue), replicated tracks (red) and restriction map (green)