Replication landscape of the human genome.

Petryk N, Kahli M, d'Aubenton-Carafa Y, Jaszczyszyn Y, Shen Y, Silvain M, Thermes C, Chen CL, Hyrien O. · 2016

Read Replication landscape of the human genome. by Petryk N, Kahli M, d'Aubenton-Carafa Y, Jaszczyszyn Y, Shen Y, Silvain M, Thermes C, Chen CL, Hyrien O. (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4729899.txt.

Introduction

Nataliya Petryk, Malik Kahli, Yves d'Aubenton-Carafa, and six colleagues mapped where human DNA replication starts and stops, in Nature Communications in 2016. Each cell cycle, copying must begin at many origins and end cleanly between them, yet origin maps from earlier methods disagreed wildly. The team purified Okazaki fragments, the short pieces made on one strand during copying, from two cell types and sequenced them deeply. Strand counts revealed which direction forks traveled everywhere in the genome.

Copying starts at random points inside broad zones, often beside active genes and open chromatin, and ends scattered between zones. Forks mostly travel the same way genes are read, avoiding head-on collisions. Start zones frequently touch chromosome folding domains and carry marks of gene enhancers. The map gives the clearest genome-wide picture yet of a process every cell performs, and shows how chromatin and transcription choreograph it.

Catalog record

CREATORPetryk N, Kahli M, d'Aubenton-Carafa Y, Jaszczyszyn Y, Shen Y, Silvain M, Thermes C, Chen CL, Hyrien O.
YEAR2016
COLLECTIONPAPERS
FORMATTXT
FILE SIZE65 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4729899

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Replication landscape of the human genome By Nataliya Petryk, Malik Kahli, Yves d'Aubenton-Carafa, Yan Jaszczyszyn, Yimin Shen, Maud Silvain, Claude Thermes, Chun-Long Chen, Olivier Hyrien Published in: Nature Communications DOI: 10.1038/ncomms10208 ABSTRACT Abstract Despite intense investigation, human replication origins and termini remain elusive. Existing data have shown strong discrepancies. Here we sequenced highly purified Okazaki fragments from two cell types and, for the first time, quantitated replication fork directionality and delineated initiation and termination zones genome-wide. Replication initiates stochastically, primarily within non-transcribed, broad (up to 150 kb) zones that often abut transcribed genes, and terminates dispersively between them. Replication fork progression is significantly co-oriented with the transcription. Initiation and termination zones are frequently contiguous, sometimes separated by regions of unidirectional replication. Initiation zones are enriched in open chromatin and enhancer marks, even when not flanked by genes, and often border ‘topologically associating domains' (TADs). Initiation zones are enriched in origin recognition complex (ORC)-binding sites and better align to origins previously mapped using bubble-trap than λ-exonuclease. This novel panorama of replication reveals how chromatin and transcription modulate the initiation process to create cell-type-specific replication programs. Genome-wide identification of mammalian replication origins1,2,3 has been attempted by various methods including trapping of replication-bubbles4,5, purification of RNA-primed single DNA strands6,7,8,9,10,11, immunoprecipitation of BrdU-labelled DNA12 or chromatin immunoprecipitation of origin recognition complex (ORC)13, followed by microarray hybridization or high-throughput sequencing. The number and position of origins thus detected present discrepancies1,3 (from 12,000 (ref. 4) to 250,000 (ref. 9) putative origins detected genome-wide). In addition, these methods cannot precisely quantify origin efficiency nor map termination. An orthogonal method to map replication is based on the analysis of nucleotide compositional asymmetries of the two DNA strands1,14,15. In most bacterial genomes, the GC and TA skews SGC=(G−C)/(G+C) and STA=(T−A)/(T+A) are strictly correlated with replication direction, suggesting that mutational asymmetry of the leading and lagging strands led to skew accumulation over evolutionary times14. Thus, detection of an abrupt sign inversion of the skew is a standard tool to predict origin and termini in bacteria16. Upward skew jumps similar to bacterial origins have been detected at 1,546 sites in the human genome17,18. Between jumps, the skew decreased in a linear manner, suggesting a progressive inversion of the average direction of replication (in germline cells) across N-shaped domains of skew, perhaps due to the random termination between the two origins19. When compared with somatic cell replication timing profiles20,21,22, skew N-domains indeed coincided with U-shaped domains of replication timing, with borders (skew jumps) replicating early and centres replicating later15,21,23,24,25. However, detectable N/U-domains together cover only one-third to one-half of the human genome and it is unclear if replication of the remainder follows similar rules. Furthermore, N/U-domains are much longer (0.5–3 Mb) than typical inter-origin distances (100–200 kb) suggesting that additional intradomain origins may fire in a sequential pattern from N/U-domain borders to centres15,25,26. The low resolution of skew and replication timing profiles precluded precise location of initiation and termination events along N/U-domains. A novel method that permits a high-resolution, quantitative analysis of replication fork initiation, progression and termination has been recently developed in Saccharomyces cerevisiae27,28. The principle is to isolate and sequence Okazaki fragments to reveal the proportions of rightward- (R) and leftward- (L) moving forks throughout the genome. Zones of predominant initiation (termination) can then be detected as upward (downward) slopes in replication fork directionality (RFD=R−L) profiles (Fig. 1a–c). Since the very low amount of Okazaki fragments present in wild-type cells represents a challenge, these studies used conditional lethal mutants that massively increase Okazaki fragment abundance27,28. The results were consistent with preexisting origin identifications and allowed to estimate origin efficiency. Furthermore, the results demonstrated that a large fraction of the genome can be replicated in either orientation and that termination occurs between origins at positions dictated by their relative firing times. [Figure: Principle of RFD mapping by OK-Seq.(a) Schematic leftward (blue) and rightward (red) forks with Okazaki fragments. RFD is computed as the difference between the proportions of rightward- and leftward-moving forks in 1 kb windows. (b) Expected RFD profiles for a fixed origin or a broad initiation zone. (c) Expected RFD profiles for a fixed terminus or a broad termination zone. The bubbles in a,b and X-shaped structures in c indicate the location of potential initiation and termination sites. The red and blue lines inside these structures indicate the Okazaki fragments synthesized during the progression of each fork. These Okazaki fragment populations are randomly sampled when extracted from asynchronously growing cell populations.] Here we present a new methodology to isolate and sequence Okazaki fragments (OK-Seq) that does not require overproducer mutant cells. This allowed us to determine RFD genome-wide in asynchronous cultures of two human cell lines, without any genetic modification. We infer the location of replication initiation and termination events at high resolution and paint a novel, comprehensive landscape of human genome replication in these two cell types. We validate the replication model of N/U-domains. We identify in the two cell lines 9,386 and 5,684 (4,150 shared) broad initiation zones that typically support a single but randomly located initiation event. These zones are mostly non-transcribed, often but not always circumscribed by active genes, generally enriched in open chromatin and enhancer epigenetic marks, and often located at TAD borders. Initiation zones associated and not associated with active genes fire early and late in S phase, respectively. Replication terminates very dispersively between initiation zones, either within transcribed, early-replicating gene bodies or within large, late-replicating, silent DNA regions. A significant though incomplete co-orientation of replication with transcription is observed, due to preferential (but not exclusive) initiation of DNA replication upstream of active genes. The number of initiation zones detected in this work accounts for only a fraction of total initiation events. We, therefore, suggest that the human genome undergoes a superposition of efficient initiation at circumscribed ‘master' zones identified here and more dispersed, less efficient initiation elsewhere. RESULTS Genome-wide profiling of RFD Our methodology is based on the replicative incorporation of the thymidine analogue 5-ethynyl-2′-deoxyuridine (EdU) followed by purification and sequencing of labelled Okazaki fragments (Fig. 2a and Methods section). Okazaki fragments purified from two cell types, HeLa (adenocarcinoma) and GM06990 (lymphoblastoid) revealed a totally new, full panorama of replication fork progression along the genome with excellent reproducibility (Fig. 2b,c, Supplementary Fig. 1). RFD profiles displayed series of alternating quasi-linear ascending, descending and flat segments (AS, DS, FS) of varying size and slope. RFD often reached values >0.9 or <−0.9, indicating nearly complete purity of Okazaki fragments. A small-scale (<300 kb) saw-tooth pattern was mostly observed in early-replicating regions (for example, 118–123 Mb, Fig. 2b,c). By contrast, most late-replicating regions displayed megabase-sized DS and/or FS of low absolute values of RFD (|RFD|) (for example, 115–118 Mb, Fig. 2b,c). [Figure: Replication directionality revealed by OK-Seq.(a) Okazaki fragment purification scheme (see text and Methods section). (b,c) Exemplary HeLa (b) and GM06990 (c) profiles. Top, Okazaki fragments mapping to Crick (C, red) or Watson (W, blue) strands. Middle, RFD computed in 1 kb adjacent windows. Bottom, replication timing profile (S50; see Methods section); red arrows, early peaks (S50<0.5).] OK-Seq validates replication model of N/U domains Previously, we detected in these cells, independently of skew analysis, 900–1,500 megabase-scaled domains characterized by parabolic (U-shaped) replication timing profiles

Read the complete text