GINS, a novel multiprotein complex required for chromosomal DNA replication in budding yeast

Abstract

Eukaryotic chromosomal DNA replication requires a two-step assembly of replication proteins on origins; formation of the prereplicative complex (pre-RC) in late M and G1 phases of the cell cycle, and assembly of other replication proteins in S phase to load DNA polymerases to initiate DNA synthesis. In budding yeast, assembly of Dpb11 and the Sld3-Cdc45 complex on the pre-RC at origins is required for loading DNA polymerases. Here we describe a novel replication complex, GINS (Go, Ichi, Nii, and San; five, one, two, and three in Japanese), in budding yeast, consisting of Sld5, Psf1 (partner of Sld five 1), Psf2, and Psf3 proteins, all of which are highly conserved in eukaryotic cells. Since the conditional mutations of Sld5 and Psf1 confer defect of DNA replication under nonpermissive conditions, GINS is suggested to function for chromosomal DNA replication. Consistently, in S phase, GINS associates first with replication origins and then with neighboring sequences. Without GINS, neither Dpb11 nor Cdc45 associates properly with chromatin DNA. Conversely, without Dpb11 or Sld3, GINS does not associate with origins. Moreover, genetic and two-hybrid interactions suggest that GINS interacts with Sld3 and Dpb11. Therefore, Dpb11, Sld3, Cdc45, and GINS assemble in a mutually dependent manner on replication origins to initiate DNA synthesis.

Figure 1

Sld5, Psf1, Psf2, and Psf3 are conserved among eukaryotes. Amino acid sequence alignment of S. cerevisiae Sld5 (Sc; A), Psf1 (Sc; B), Psf2 (Sc; C), and Psf3 (Sc; D) and their homologs from S. pombe (Sp), Caenorhabditis elegans (Ce), Drosophila melanogaster (Dm), Homo sapiens (Hs). Amino acids conserved in all aligned sequences are shown with dark shadow, and those conserved in at least two sequences but not all are shown with light shadow. sld5 mutation sites and psf1-1 mutation sites are shown by asterisks with the amino acid changes in A and B. sld5-2, sld5-12, and sld5-13 mutations were identified at positions 448 (A to G), 199 (T to C), and 878 (T to C) in the nucleotide sequence (nucleotide 1 is A of the first ATG of the ORF). These mutations change the amino acids of Sld5 from conserved lysine, tryptophan, and leucine to glutamic acid, arginine, and proline, respectively. The sld5-8 allele has two alterations occurring at nucleotides 61 (T to C) and 196 (T to C), which both result in the replacement of serine with proline (A). However, we do not know whether two-site mutations are required for temperature sensitivity. The psf1-1 mutation site was identified at position 250 (A to G) in the nucleotide sequence, which changes an arginine to glycine (B).

Figure 2

The genetic and two-hybrid interactions between Dpb2, Dpb11, Sld2, Sld3, Cdc45, Psf1, Psf2, and Psf3. (A) Suppression of thermosensitive growth of sld5-12 and psf1-1. YYK38 (sld5-12) cells and YYT1 (psf1-1) cells carrying different plasmids were streaked onto YPDA plates and incubated at the indicated temperatures. The genes in the left half of the diagram were cloned into YCplac22 (the low-copy vector), and the genes in the right half were cloned into YEp195 (the high-copy vector). (B) Two-hybrid interactions between Psf1 and Dpb2, Dpb11, Sld3, or Sld5. “Vector” and “Psf1” in pBTM116 denote plasmids that express LexABD and LexABD–Psf1, respectively. V, DPB2, DPB11, SLD3, and SLD5 in pACT2 denote plasmids that express Gal4AD and Gal4AD fusion with Dpb2, Dpb11, Sld3, and Sld5, respectively. Transformants of L40 each carrying a pair of plasmids were assayed for β-galactosidase activity by colony color with X-Gal. (C) Summary of genetic and two-hybrid interactions among Dpb2, Dpb11, Sld2, Sld3, Cdc45, Psf1, Psf2, and Psf3. The blue line shows multicopy suppression; the red line shows a positive signal in the two-hybrid assay.

Figure 3

Sld5, Psf1, Psf2, and Psf3 form a complex. (A) Protein extracts were prepared from cells of the indicated genotypes. +, an epitope-tagged gene; −, a wild-type allele. Anti-Flag immunoprecipitations were performed. The immunoprecipitated proteins were separated by 4%–20% SDS-PAGE and subsequently silver-stained (left and center panels). In the right panel, immunoprecipitates from 6Flag–Psf1 or 6Flag–Psf1 GST–Sld5 cell extracts were probed with anti-Sld5 or anti-GST antibodies. (B) A gel filtration chromatography of 6Flag–Psf1 immunoprecipitates using Superdex 200 was performed as described in Materials and Methods. The proteins in each column fraction were identified by 5%–20% SDS-PAGE followed by silver staining. The first lane contains protein molecular-weight standards (M), and the second lane contains the input sample. The Superdex 200 column was calibrated with tyroglobin (669 kD), ferritin (440 kD), aldolase (158 kD), BSA (63 kD), and RNase A (13 kD). (C) YYK46 (6Flag–Psf1, WT) and YYK50 (6Flag–Psf1 in sld5-12) cells were cultured at 23°C to 1 × 10⁷ cells/mL. Half of each culture was shifted to 30°C while the other half was maintained at 23°C, and incubations were continued until cells reached 2 × 10⁷ cells/mL (left panel). YEp195 (V), YEp195–SLD5 (SLD5), and YEp195–PSF2 (PSF2) plasmids were introduced into YYK46 (6Flag–Psf1; WT) and YYK50 (6Flag–Psf1 in sld5-12) cells. Cells carrying each plasmid were cultured at 23°C (center and right panels). Protein extracts were prepared from each culture, and anti-Flag immunoprecipitations were performed. The amount of GINS was estimated by staining with silver (left and center panels). In the right panel, whole cell extracts (WCE) and immunoprecipitates (IP) were immunoblotted with anti-Sld5 or anti-Flag antibodies.

Figure 4

sld5-12 and psf1-1 cells are defective in DNA replication. (A) FACS analysis of cells released from G1-phase arrest. Wild-type (WT), sld5-12, and psf1-1 cells were synchronized with α-factor at 23°C and released from α-factor at 36°C. At the indicated times, aliquots were treated with propidium iodide and the DNA content was measured by FACScan. 1C and 2C indicate DNA contents of G1 and G2/M cells. (B) Viability and cell morphology of sld5-12 and psf1-1 mutant cells. Portions of the same samples incubated at 36°C and described in A were used to determine cell number and cell morphology. The cells were spread onto YPD plates to measure viability. ▪, viable cells; ▵, cells without bud; □, cells with small bud; ○, cells with large bud. (C) N/N 2-D gel analysis of the chromosomal ARS1 locus in wild-type (WT), sld5-12, and psf1-1 cells. Cells were grown and harvested at 23°C or shifted to 37°C for 3 h prior to harvest. DNA was digested with NcoI and probed with ARS1-containing fragment. The diagrams besides photographs show the bubble to Y fork and Y fork arcs.

Figure 5

Association of Psf1 with ARS regions. (A) Genomic intervals near or at ARSs amplified by PCR primers. (B) Flag–Psf1 specifically associated with ARS1 in vivo. YYK9 (tag-) or YYT2 (6Flag–Psf1) cells were grown in YPAR at 23°C. Immunoprecipitation (IP) was performed with anti-Flag antibody. PCR was performed on chromatin fragments isolated after IP or on those from the whole cell extracts (WCE). (C) Flag–Psf1 specifically associated with functional ARS1. YYT2 (6Flag–Psf1, ARS1-A; WT) or YYK51 (6Flag–Psf1, ARS1-A/860T → G; m) cells were arrested in G1 phase by α-factor at 23°C for 3 h and released in YPAR containing 0.15 M HU at 23°C. Cells were withdrawn after 30 min. 6Flag–Psf1 was immunoprecipitated from each extract with an anti-Flag antibody. PCR was performed on immunoprecipitates derived from the same number of cells. (D) DNA content of synchronized cells used for the ChIP assay (E) was measured by FACS analysis. The percentage of budded cells is also shown. (E) Association of Flag–Psf1 with the ARS1, ARS305, or ARS501 regions. YYT2 cells were arrested in G1 phase with α-factor and released in YPAD medium at 16°C. Cells were withdrawn from the culture every 15 min and fixed with formaldehyde. Cell lysates were sonicated and used for immunoprecipitation. PCR was performed either on immunoprecipitates (IP) derived from the same number of cells at each time point or on the 0-min chromatin fraction from whole cells extracts (WCE). (F) Association of Psf1 with the origin depends on Sld3. YYT2 (6Flag–Psf1; WT) and YYT3 (6Flag–Psf1; sld3-5) cells were arrested in G1 phase by α-factor at 23°C for 3 h and released in YPAR containing 0.15 M HU at 33°C. Cells were withdrawn from the culture every 30 min. 6Flag–Psf1 was immunoprecipitated from each extract with an anti-Flag antibody. PCR was performed on immunoprecipitates derived from the same number of cells at each time point.

Figure 6

Cdc45 association with chromatin is reduced in the psf1-1 mutant cells. YYK20 (WT) and YYT4 (psf1-1) cells expressing Cdc45-3HA were synchronized in G1 phase by α-factor and released at 37°C. The cells were collected at the α-factor block or 75 min after release. Chromatin-binding assay was performed as described (Kamimura et al. 2001). The proteins present in the different fractions of the chromatin purification were examined by immunoblotting of SDS-PAGE: W, whole cell extract; S, supernatant; P, pellet fraction. Extracts were incubated on ice either without (−) or with (+) DNase I. The bottom panel shows the DNA content of the samples used in the top panel.

Figure 7

Associations of Psf1 and Dpb11 with origins are mutually dependent. (A) Association of Psf1 with origins requires Dpb11. YYT2 (6Flag–Psf1; WT) and YYK42 (6Flag–Psf1; dpb11-26) cells were arrested in G1 phase by α-factor at 23°C for 3 h and released in YPAR containing 0.15 M HU at 33°C. Cells were withdrawn from the culture every 30 min. 6Flag–Psf1 was immunoprecipitated from each extract with an anti-Flag antibody. (B) Association of Dpb11 with origins requires Psf1. YHM011 (Dpb11-9myc; WT) and YYK43 (Dpb11-9myc; psf1-1) cells were arrested in G1 phase with α-factor at 23°C for 3 h and released in YPAR containing 0.15 M HU at 34°C. Cells were withdrawn from the culture every 30 min. PCR was performed on immunoprecipitates derived from the same number of cells at each time point.