GINS motion reveals replication fork progression is remarkably uniform throughout the yeast genome

Abstract

Previous studies have led to a picture wherein the replication of DNA progresses at variable rates over different parts of the budding yeast genome. These prior experiments, focused on production of nascent DNA, have been interpreted to imply that the dynamics of replication fork progression are strongly affected by local chromatin structure/architecture, and by interaction with machineries controlling transcription, repair and epigenetic maintenance. Here, we adopted a complementary approach for assaying replication dynamics using whole genome time-resolved chromatin immunoprecipitation combined with microarray analysis of the GINS complex, an integral member of the replication fork. Surprisingly, our data show that this complex progresses at highly uniform rates regardless of genomic location, revealing that replication fork dynamics in yeast is simpler and more uniform than previously envisaged. In addition, we show how the synergistic use of experiment and modeling leads to novel biological insights. In particular, a parsimonious model allowed us to accurately simulate fork movement throughout the genome and also revealed a subtle phenomenon, which we interpret as arising from low-frequency fork arrest.

Figure 1

Time-course ChIP-chip analysis of GINS complex binding to chromosome XVI through the cell cycle. S. cerevisiae cells (W303 strain, genomically tagged on Psf2—a component of the GINS complex) were synchronized in YPD media with α-factor. Starting from G₁, ChIP-chip was performed at 11 time points through the cell cycle (see Materials and methods). The GINS complex was immunoisolated at each time point. Associated DNA was amplified and hybridized to yeast whole genome microarrays. Relative occupancies of the GINS complex for sites across chromosome XVI are shown for each time point. Locations of the initial binding sites are noted with dashes. Cell cycle state was determined by monitoring budding indices as well as DNA content by flow cytometry.

Figure 2

Time-course ChIP-chip analysis of GINS complex binding to chromosome XV through S phase. (A) At the beginning of S phase (20 min time points) the GINS complex has not yet associated with any part of the chromosome. As S phase progresses, replication initiates at the indicated origins (25–30 min time points) whereupon the GINS complex moves away bi-directionally with the replication forks. Replication of a particular chromosomal region is finished when adjacent forks collide, giving rise to broad peaks in the inter-origin spaces. The locations of previously defined ARSs (excluding dubious origins as defined by Nieduszynski et al, 2007) are shown as red columns in the bottom panel. (B) Schematic of fork progression rate calculations.

Figure 3

S-phase time-course data (experiment and simulations) for several genomic regions of interest. (A) ChrVIII coordinates 240–370 kb. A category 1 origin flanked by two category 2 origins. (B) Seven category 1 origins spaced unevenly between 320 and 600 kb of ChrIV. (C) ChrXII coordinates 510–600 kb. The gap in between the two noted origins is 97 kb, among the longest inter-origin distances observed in the yeast genome. (D) The right-hand telomeric region of ChrXIII. The telomeric region is shaded. (E) The centromeric region of ChrIX (shaded gray). (F) A category 3 origin (shaded gray) located at ∼695 kb (ARS1625) in ChrXVI.

Figure 4

Varying efficiencies applied a group of origins located in the right arm of ChrXV. The six category 1 origins on either end are given 100% efficiencies, whereas the efficiency applied to the cluster of seven category 2 origins in the middle of this region is indicated above each panel.

Figure 5

Persistent features associated with GINS progression through highly transcribed genes. ChIP-chip data for the GINS with the entire ChrVII are shown for the 75 min time point. The baseline has been raised to emphasize peaks. Persistent features can be observed at all of the tRNA genes (t), snoRNA genes and snRNA genes (s), rRNA genes (r), as well as six out of the other nine most highly expressed genes on this chromosome (indicated by the gene name).

Figure 6

The GINS complex undergoes long-term arrest at tRNA genes. (Center) ChIP-chip data for the association of the GINS complex with a region of chromosome VI (125–220 kb) containing eight tRNA genes. Data are shown in 15 min increments through the cell cycle. During late S phase sharp spikes are observed, which coincide with the position of the tRNA genes. We observe similar long-lived spikes that coincide with 267 of the 275 tRNA genes in the yeast genome. Two alternative models for explaining these spikes are presented: (left) in each cell, the replication forks are assumed to pause for 6 s at each tRNA gene; (right) the replication forks are assumed to have a 0.2% chance of long-term (t=∞) arrest at each tRNA gene. Additional model parameter conditions are shown in the Supplementary information.