The Fkh1 Forkhead associated domain promotes ORC binding to a subset of DNA replication origins in budding yeast

Abstract

The pioneer event in eukaryotic DNA replication is binding of chromosomal DNA by the origin recognitioncomplex (ORC). The ORC-DNA complex directs the formation of origins, the specific chromosomal regions where DNA synthesis initiates. In all eukaryotes, incompletely understood features of chromatin promote ORC-DNA binding. Here, we uncover a role for the Fkh1 (Forkhead homolog) protein and its forkhead associated (FHA) domain in promoting ORC-origin binding and origin activity at a subset of origins in Saccharomyces cerevisiae. Several of the FHA-dependent origins examined required a distinct Fkh1 binding site located 5' of and proximal to their ORC sites (5'-FKH-T site). Genetic and molecular experiments provided evidence that the Fkh1-FHA domain promoted origin activity directly through Fkh1 binding to this 5' FKH-T site. Nucleotide substitutions within two relevant origins that enhanced their ORC-DNA affinity bypassed the requirement for their 5' FKH-T sites and for the Fkh1-FHA domain. Significantly, assessment of ORC-origin binding by ChIPSeq provided evidence that this mechanism was relevant at ∼25% of yeast origins. Thus, the FHA domain of the conserved cell-cycle transcription factor Fkh1 enhanced origin selection in yeast at the level of ORC-origin binding.

Figure 1.

Model to explain the ORC-binding differences between two distinct yeast origin cohorts (24). Origins in the positive-DNA cohort rely on direct interactions between ORC (orange crescent) and the the origin’s ORC binding site to achieve normal levels of ORC binding, while origins in the positive-chromatin cohort require additional interactions with an origin-adjacent protein(s) to achieve normal levels of ORC binding to their intrinsically weak ORC sites. The weaker ORC–DNA interactions at positive-chromatin origins are symbolized with open ORC. In this simple model, a sequence-specific DNA binding protein (green cylinder) binds near the origin and promotes ORC binding (arrow). The arrow is not meant to imply a specific mechanism by which the green factor promotes ORC binding.

Figure 2.

The positive-chromatin origin ARS1529.5 required a 5′ proximal FKH binding site for optimal function. (A) The indicated ARS1529.5 chromosomal fragments (left) were cloned into a plasmid and the corresponding plasmid loss rates (PLRs; mean ± SE) were determined (right). Asterisks here and in all subsequent figures indicate P-values for the relevant comparisons (***P < 0.001, **P < 0.01 and *P < 0.05). P-values are derived from two-tailed Student’s t-tests comparing constructs to ARS1529.5-standard. The sequences that were used to substitute for native ARS1529.5 sequences to create the indicated Δ clones are derived from KANMX and are denoted in red, or red shading when they overlapped OriDB regions (B) DNA sequences between positions –155 and –86 of the ARS1529.5 5′Δ + 50Δ ARS construct were substituted with the 8 bp expanded Sal1 restriction site, gGTCGACg, as indicated. Mutations that significantly raised PLRs (P-values < 0.05) relative to ARS1529.5-standard are boxed in gray shading. (C) PLRs of Sal1 linker-containing ARS1529.5 mutants in (B). Horizontal dashed lines represent the PLR mean ± se of the wild-type ARS1529.5 50Δ fragment in A. (D) (Top) An FKH consensus motif (23). Sequence of the wild-type region of ARS1529.5 containing the FKH site (top), and the SalI linker substitution in ars1529.5-FKHΔ::Sal1. Basal ARS activity is defined as whether colonies were recovered when yeast were transformed with the indicated ARSs. The indicated yeast cells were transformed with the ARS1-like version, OriDB version, or OriDB version containing the FKHΔ::Sal1 substitution (mutOriDB) of ARS1529.5. ‘+’ indicates that colonies were recovered. ‘−’ indicates that no colonies were recovered.

Figure 3.

The Fkh1-FHA domain functioned through the FKH site in ARS1529.5 and contributed to the ARS function of the majority of positive-chromatin origins. (A) Diagram of the Fkh1 protein and relevant protein domains. Residue 80 of Fkh1 is a conserved arginine essential for the threonine-phosphopeptide-binding activity of FHA domains (44–46)(38). (B) PLRs measured for the indicated ARS1529.5 plasmids in FKH1, fkh1R80A, or fkh1Δ cells. P-values are derived from two-tailed Student’s t-tests comparing constructs to ARS1529.5-standard. (C) ARS activities (inverse of plasmid loss rates) were determined for the indicated ARSs, cloned as standard fragments (see Figure 2A), in fkh1R80A and FKH1 cells. The data are expressed as ratios of ARS activity measured in fkh1R80A cells to that measured in FKH1 cells. The ARSs are ordered from most sensitive (FHA-dependent) to least affected (FHA-independent). (D) Enrichment of FHA-dependent ARSs (fkh1R80A-sensitive) within the positive-chromatin and positive–DNA origin groups. P-values were derived from a hypogeometric functon comparing the distribution of FHA-dependent or -independent origins in each cohort to their distribution in the entire collection of 32 origins. Significance is denoted as in Figure 2. (E) Chromosomal S-phase copy numbers as measured by ddPCR for six FHA-dependent and four FHA-independent ARSs in C. For each locus, 6 (for wild-type) or 12 (for fkh1R80A mutant strain) independent S-phase cell samples were assessed and the mean ± 95% confidence interval determined (see Supplementary Figure S1). Asterisks indicate P-values from the relevant tests in panels B and D (***P < 0.001, **P < 0.01 and *P < 0.05).

Figure 4.

FHA-dependent ARSs were enriched for T-rich oriented FKH motifs positioned 5′ of their ORC binding sites. (A) FHA-dependent and -independent origin fragments were examined for FKH site matches in either the T-rich (FKH-T) or the A-rich (FKH-A) orientation relative to the T-rich ORC site as depicted. Origin fragments were defined as chromosomal regions with approximately 300 nucleotides 5′ of the ORC site start of ‘0’ (–300 to –1, 5′ region) and 3′ of the end of the ORC site ‘+32’ (+33 to +300, 3′ region). (B) Positions of FKH-T and FKH-A motifs in FHA-dependent (left) and FHA-independent (right) origins. Vertical dotted lines through each graph demarcate the T-rich ORC site boundaries (0 to +32) used to align each origin. (C) Fraction of origins in the indicated cohorts (x-axis) with at least one of the indicated FKH motifs as defined in A (see also Supplementary Figure S3). Asterisks indicate P-values from the Hypergeometric Distribution function comparing a cohort’s fraction with that measured for all origins (dashed line) (***P < 0.001, **P < 0.01 and *P < 0.05). (D) The 5′ FKH-T sites were replaced with an expanded Sal1 restriction site, gGTCGACg, as with ARS1529.5 in Figure 2. The positions of the substituted sites are indicated with a red X. E) Plasmid loss rates for the indicated ARSs were determined in FKH1 or fkh1R80A cells.

Figure 5.

High-affinity ORC sites bypassed requirements for the 5′ FKH-T site and the FHA domain. (A) Top: Web Logo for ORC consensus site derived from 393 chromosomal origins. Bottom: ORC binding sites for three indicated origins, and below that, the nucleotide substitutions, indicated in red, used to generate high-affinity mutant ORC binding sites. (B) The Kds measured by gel shift assays with purified ORC and origin DNA fragments were normalized by dividing by the Kd measured for the tight-binding ORC binding site from ARS317 (see Supplementary Figure S5). The normalized Kds for wildtype ARS1529.5 and ARS1114 are indicated in parentheses below the text of those origins (24). (C) Plasmid loss rates for wildtype and the indicated mutant versions of ARS1529.5 in either FKH1 or fkh1R80A cells. (D) Plasmid loss rates for wild-type and the indicated mutant versions of ARS1114 in FKH1 cells.

Figure 6.

The Fkh1-FHA domain promoted ORC-origin interactions at a substantial fraction of yeast chromosomal origins. (A) ORC signals of ARS1529.5 and ars1529.5-FKHΔ in ORC ChIP-qPCR. (B) ORC signal measured at ARS1529.5 by ORC ChIPSeq in FKH1 and fkh1R80A cells. (C) ORC signals as in B at four additional FHA-dependent positive-chromatin origins identified in Figure 3 C and with 5′ FKH matches validated in ARS assays in Figure 4D and E. (D) Distributions of ratios of fkh1R80A/FKH1 ORC binding areas spanning nucleotides –100 to +100. Distributions of ratios are plotted as box-and-whisker plots overlayed on violin plots. (E) Enrichment analysis of distinct groups of yeast origins defined by their ratios of fkh1R80A/FKH1 ORC binding areas as defined in D. Enrichments (indicated with a ‘+’) or depletion (indicated with a ‘−’) of categories of fkh1R80A/FKH1 scaled ORC ratios were challenged against distributions of those same categories within all confirmed origins with the Hypergeometric Distribution function. ++/−−, P value < 0.01; +/−, P value < 0.05.