Post-licensing Specification of Eukaryotic Replication Origins by Facilitated Mcm2-7 Sliding along DNA

Abstract

Eukaryotic genomes are replicated from many origin sites that are licensed by the loading of the replicative DNA helicase, Mcm2-7. How eukaryotic origin positions are specified remains elusive. Here we show that, contrary to the bacterial paradigm, eukaryotic replication origins are not irrevocably defined by selection of the helicase loading site, but can shift in position after helicase loading. Using purified proteins we show that DNA translocases, including RNA polymerase, can push budding yeast Mcm2-7 double hexamers along DNA. Displaced Mcm2-7 double hexamers support DNA replication initiation distal to the loading site in vitro. Similarly, in yeast cells that are defective for transcription termination, collisions with RNA polymerase induce a redistribution of Mcm2-7 complexes along the chromosomes, resulting in a corresponding shift in DNA replication initiation sites. These results reveal a eukaryotic origin specification mechanism that departs from the classical replicon model, helping eukaryotic cells to negotiate transcription-replication conflict.

Figure 1. Isolation of functional pre-RCs free in solution

(A) Reaction scheme. (B) Gel-filtration analysis of reconstituted pre-RC assembly reaction using linearized pARS305 (9.8 kb) as a template. Proteins were analyzed by SDS-PAGE and silver stain, DNA was analyzed by agarose gel-electrophoresis and ethidium-bromide staining. Pre-RCs were assembled in the presence (top two panels) or absence (bottom two panels) of ORC. (C) Representative electron micrograph (left) and 2D class average (right) of Mcm2-7 complexes eluting in fractions 8-10 in presence of ORC in B. (D) In vitro DNA replication assay using purified Mcm2-7 DHs isolated on relaxed pARS305 (9.8 kb); reaction scheme on left. Input protein was analyzed by SDS-PAGE and silver stain, input DNA by agarose gel-electrophoresis and ethidium bromide stain. Replication products were visualized by autoradiography after native agarose gel-electrophoresis. RI's = replication intermediates.

Figure 2. T7 RNAP can push Mcm2-7 DHs off the ends of linear DNA

(A) Reaction scheme. (B) Pre-RCs were tested for displacement by T7 RNAP on linear or circular relaxed forms of pARS305 (9.8 kb) indicated on the right. Protein-DNA complexes were analyzed by gel-filtration; relative Mcm2-7 content in each fraction was determined after SDS-PAGE and silver stain (Figure S2). P_T7, T7 promoter.

Figure 3. Mcm2-7 DHs remain functional after displacement by T7 RNAP

(A) Native agarose gel analysis of in vitro DNA replication products obtained with pre-RCs reconstituted on circular closed pARS1 (5.8 kb) treated with T7 RNAP in the absence or presence of a T7 promoter or Vaccinia Topo I as indicated. Averages and standard deviations of ³²P-dCTP incorporation from three independent experiments are shown in the bar diagram. Reaction scheme on the left. (B) Mcm2-7 DH location analysis by co-immunoprecipitation of Mcm2-7 complexes and DNA fragments generated from pre-RCs assembled on circular closed pARS1 according to reaction scheme on left. Restriction sites used are indicated by red triangles in the plasmid map. P_T7, T7 promoter; T_T7, T7 terminator. DNA was analyzed by ethidium-bromide stain. (C) In vitro replication analysis using Mcm2-7/DNA fragment complexes obtained according to reaction scheme on left. Replication products were analyzed by native agarose gel-electrophoresis and autoradiography (right panel) and ³²P incorporation into respective fragments was quantitated by phosphorimaging (middle graph). (D) In vitro replication analysis of Mcm2-7/DNA complexes obtained according to reaction scheme on left. The pARS1 template included T7 terminator sequences (see map in b); restriction was carried out with SpeI / NdeI. Replication products were analyzed by native agarose gel-electrophoresis and autoradiography and ³²P incorporation into respective fragments was quantitated by phosphorimaging.

Figure 4. ChIP-seq analysis of ORC and Mcm2-7 in rat1-1 cells

(A) ORC and Mcm2-7 enrichment around ARS1414, ARS447, and ARS305 at 24°C and 37°C is plotted over chromosomal position. Annotated ORFs on the Watson (yellow) and Crick (blue) strands are indicated. (B) Heat maps of ORC and Mcm2-7 enrichment at 210 origin sites where Mcm2-7 and ORC peaks coincide. Positive signal within 6 kb window is indicated in white. ORC maps are centered on ORC peak positions at 24°C; Mcm2-7 maps are centered on Mcm2-7 peak positions at 24°C. Mcm2-7 peaks obtained at 37°C were ordered into eight clusters (I-VIII). ORC (24°C and 37°C) and Mcm2-7 (24°C) maps were ordered according to Mcm2-7 (37°C). Panel on the right indicates average TDM at respective origin sites (see Figure S6B). (C) TDMs at the 210 ORC/Mcm2-7 peaks plotted as a function of Mcm2-7 peak position at 37°C vs. 24°C. red: Mcm2-7 shift to the right; blue: Mcm2-7 shift to the left; black: Mcm2-7 peak position unchanged. (D) Data from (C) summarized as box and whisker plots; significance of difference between origins at which Mcm2-7 at 37°C is displaced to the right and origins at which Mcm2-7 at 37°C is displaced to the left is indicated.

Figure 5. Okazaki fragment distribution in rat1-1 cells

(A) Okazaki fragments mapping to Watson and Crick strands on S. cerevisiae chromosome XV in RAT1 or rat1-1 cells at 37°C. Select ARS IDs are indicated on top. (B) Distribution of origin efficiencies in rat1-1 and RAT1 cells. N = number of origins (OEM ≥ 0.1). (C) Distribution of origin positions in rat1-1 and RAT1 cells relative to those reported previously for RAT1 cells at 30°C (McGuffee et al., 2013).

Figure 6. Initiation site shift at ARS1415 and ARS1303 in rat1-1 cells at 37°C

Okazaki fragments mapping around ARS1415 and ARS1303 in RAT1 (blue / orange) or rat1-1 (red / blue) cells at 37°C. Dashed vertical lines denote positions of ARS1415 and ARS1303, respectively; dashed vertical arrows denote main new initiation sites in rat1-1 cells. Transcript maps (Xu et al., 2009) and positions of annotated open reading frames (ORFs) are indicated below.

Figure 7. Initiation sites shift in the direction of transcription

(A) Average Okazaki fragment densities at origins located in regions of prevailing Watson strand transcription (left panels; N = 58), or prevailing Crick strand transcription (right panels; N = 58) in Rat1 (top panels) or rat1-1 (bottom panels) cells at 37°C. Red triangles indicate origin position. Dotted vertical lines indicate previously reported positions for respective origins determined at 30°C (McGuffee et al., 2013). Δ indicates the shift in origin position. (B) Model for origin relocation during defective transcription termination in rat1-1 cells.