The large SpCAF-1 subunit includes four intrinsically disordered regions (IDR)

a Names for the large, the medium and the small subunits of CAF-1 in H. sapiens, S. cerevisiae and S. pombe. b Upper panel: The magenta line shows the predicted disorder of Pcf1 (spot disorder software) and the black line the Cα Local Distance Difference Test (pLDDT) calculated for Pcf1 residues by the AlphaFold2 model of the full SpCAF-1 complex. Lower panel: Cα chemical shift index calculated for the 101 assigned residues. This Cα chemical shift index is consistent with their disordered nature. The four IDR regions are highlighted with pink semi-transparent vertical bars. The predicted domains of Pcf1 are labeled. c General strategy for the production of SpCAF-1. The lower panel shows the purification SEC profile and the SDS-PAGE purity of the sample. d 1H-15N SOFAST-HMQC spectrum of the FL SpCAF-1 complex composed of uniformly labeled 15N-Pcf1 and unlabeled Pcf2 and Pcf3 (SpCAF-1(15N-Pcf1)). The assigned signals are labeled. e AF2 model of the SpCAF-1 complex. The four IDR segments are shown with a dashed line. The relative orientation of the four modules is arbitrary.

The acidic ED domain binds histones alone and in the full SpCAF-1 complex

a SEC profile and the SDS-PAGE purity of SpCAF1-H3-H4 histones at 150 mM NaCl and 1 M NaCl. b Mapping of the interaction between SpCAF-1(15N-Pcf1) and SpH3-H4 histones, using the intensities ratio (I/I0), where I and I0 are the intensity of the signals 1H-15N SOFAST-HMQC spectra before and after addition of histones, respectively. c Sequence Logo of the ED domain generated with a large data set of Pcf1 homologues. The position of the two conserved residues Y340 and W348, mutated in ED* are indicated with stars and conserved Pcf1 L359 and F380 residues with five and four branch stars respectively. d Mapping of the interaction between Pcf1_ED or Pcf1_ED* with SpH3−H4 histones using the intensities ratio (I/I0) as in b. Histones were added alone or previously complexed with histones chaperones. e Cartoon representation of the complex between human histones H3−H4 (light blue and cyan), Asf1 (light grey) and Mcm2 (dark grey) (PDB: 5BNX). f AlphaFold2 model of Pcf1 (353-385) (as red cartoon), corresponding to the segment of the ED domain indicated in red, in complex with histones H3−H4 (light blue and cyan surface) superimposed with Mcm2 and Asf1 as in panel e. The two insets represent zoomed views of the sidechains of the conserved Pcf1 L359 and F380 residues (red sticks) binding into H4 and H3 pockets, respectively. The same four and five branch stars are used to label these positions in the logo panel c.

Pcf1_KER is the main DNA binding domain of SpCAF-1:

a EMSA with SpCAF-1 and 40 dsDNA (1µM) revealed with SYBR SAFE staining. b Microscale thermophoresis (MST) fitted curves of SpCAF-1 WT and mutants with 40bp dsDNA. c Upper panel: Modelled structure of the Pcf1_KER-PIP domain (56-185) rainbow coloured according to the pLDDT of each residue. Red corresponds to pLDDT values of 1 and dark blue of 0. Middle panel same model represented with its electrostatic surface. Lower panel: zoom of the C-terminus of the KER domain and the PIP motif. The five mutated residues are labeled and highlighted with spheres. d EMSA of Pcf1_KER binding with a 40bp dsDNA (1µM) revealed with SYBR SAFE staining e MST fitted curves of Pcf1_KER constructs and mutants with 40bp dsDNA. f Overlay of the calculated model of the WHD domain obtained with the CS-rosetta software (light orange) using NMR assignments of the domain (Figure S3n), with AlphaFold2 (gold) and the structure of Cac1 WHD from budding yeast (PDB 5jbm, in grey)(Liu et al. 2016)(Grey). g EMSA revealed with SYBR SAFE staining of Pcf1_WHD domain with a 40 dsDNA (1µM).

Experimental affinities of different SpCAF-1 constructs with a 40bp dsDNA measured by Microscale thermophoresis (MST).

Interactions parameter with SpPCNA measured by isothermal microcalorimetry (ITC): *The stoichiometry (N) is calculate as a molar ratio of monomeric PCNA.

The SpCAF-1-KER* mutant is affected for PCNA binding.

a EMSA showing interactions of purified SpCAF-1 (at the indicated concentrations), with or without recombinant SpPCNA (3µM) in the presence and absence of 40bp dsDNA (1µM), revealed with Coomassie blue (upper panel) and with SYBR SAFE staining (lower panel). b Quantification of bound SpPCNA in the EMSA shown in panel a and in Figure S4d for SpCAF-1 and mutants. Values are indicated in % compared to the free PCNA reference (PCNA alone in line 1 in panel a) after addition of SpCAF-1 (WT or mutant) at the indicated concentration and in the presence (filled bars) or absence (dashed bars) of 40bp dsDNA (1µM). c Quantification of bound DNA for EMSA shown in panel a and in Figure S4d for SpCAF-1 and mutants. Bound DNA in % is compared to the free DNA reference (line 5 in panel a) after addition of SpCAF-1 (WT or mutant) at the indicated concentration and in the presence (filled bars) or absence (dashed bars) of SpPCNA (3µM). All experiments were done in duplicates. Mean values are indicated and error bars shows their standard deviation.

Rescue using Y340 and W348 in the ED domain, the intact KER DNA binding domain and the C-terminal WHD of Pcf1 in SpCAF-1 mediated nucleosome assembly.

Supercoiling analysis after 45 (left) and 120 (right) minutes incubation to monitor chromatin assembly in control mock and xenopus p150-depleted HSE. Total DNA visualized by EtBr staining (top) and synthesized DNA visualized by biotin detection (bottom) are shown. The xenopus p150-depleted HSE is either mock complemented (-) or complemented using SpCAF-1 complex composed of wild type Pcf1(WT) or mutants Pcf1-PIP*, Pcf1-ED*, Pcf1-KER*-, or Pcf1-ΔWHD- as indicated. T: pBS plasmid incubated without extract run in parallel serves as a migration control to locate supercoiled DNA. The position of relaxed (II) and supercoiled (I) DNA are indicated.

Association of CAF-1 with histone modulates PCNA interaction in vivo and foci formation.

a Anti-FLAG Pulldown to address PCNA−CAF-1 interaction in vivo in indicated strains. b Quantification of bound PCNA from (a). c Simultaneous acquisition of Pcf2-GFP in WT (red arrow) and pcf1 mutated strains (blue arrow). Strains were grown separately and equally mixed before to process them for cell imaging. WT pcf1 cells expressed the fusion Cut11-mCherry, a component of the nuclear pore complex, leading to a red labelling of the nuclear periphery. Under same illumination and acquisition conditions, Pcf2-GFP foci are detected in WT and pcf1-ΔWHD cells, but not in pcf1-PIP* and pcf1-KER* cells. In contrast, Pcf2-GFP foci are brighter and more abundant in pcf1-ED* cells. d Quantification of cells showing Pcf2-GFP foci, according to cell morphology in indicated strains. Mono-nucleated cells mark G2-phase cells and bi-nucleated cells with septum mark S-phase cells. Values are means of at least 3 independent experiments ± standard error of the mean (sem). At least 1000 nuclei were analysed per strain. P values are indicated with stars and were calculated using the student test.

The WHD domain of SpCAF-1 specifies CAF-1 function.

a Top: Schematic representation of the silencing assay used. Otr: outer repeats, imr; inner repeats: cnt1; central core of the centromere 1. Bottom: Serial fivefold dilution on indicated strains on indicated media. b Example of Rad52-GFP foci in WT cells. Blue arrows indicate Rad52 foci-positive cells. c Quantification of Rad52-GFP foci in indicated strains. Values are means of at least 3 independent experiments ± sem. P values are indicated as stars and were calculated with the student test. At least 1000 nuclei were analyzed per strain. d. Co-lethality assay. Tetrad dissections of cells deleted for hip1 (hip1Δ) crossed with cells deleted for pcf1 (pcf1Δ) (grey) or harbouring pcf1-PIP* (magenta), pcf1-ED* (red), pcf1-KER* (purple) or pcf1-ΔWHD* (orange). Spores with double mutations are surrounded and were deducted from the analysis of viable spores from each tetrad (see materiel and method). At least 18 tetrads were analysed per cross.