A DNA binding winged helix domain in CAF-1 functions with PCNA to stabilize CAF-1 at replication forks

Abstract

Chromatin assembly factor 1 (CAF-1) is a histone H3-H4 chaperone that deposits newly synthesized histone (H3-H4)2 tetramers during replication-coupled nucleosome assembly. However, how CAF-1 functions in this process is not yet well understood. Here, we report the crystal structure of C terminus of Cac1 (Cac1C), a subunit of yeast CAF-1, and the function of this domain in stabilizing CAF-1 at replication forks. We show that Cac1C forms a winged helix domain (WHD) and binds DNA in a sequence-independent manner. Mutations in Cac1C that abolish DNA binding result in defects in transcriptional silencing and increased sensitivity to DNA damaging agents, and these defects are exacerbated when combined with Cac1 mutations deficient in PCNA binding. Similar phenotypes are observed for corresponding mutations in mouse CAF-1. These results reveal a mechanism conserved in eukaryotic cells whereby the ability of CAF-1 to bind DNA is important for its association with the DNA replication forks and subsequent nucleosome assembly.

Figure 1.

Structure of Cac1C. (A) A overview of Cac1C. Key residues predicted to be involved in DNA binding are shown as stick representations (magentas). Secondary structure elements of Cac1C are labeled. (B and C) Structural comparison of Cac1C (cyan) with RECQ1 (magenta PDB 2WWY, panel B) and RFX1 (green PDB 1DP7, panel C). All α helices and w1 are labeled. The root mean square deviations (RMSD) of these two structures with Cac1C are 2.60 Å and 2.15 Å, respectively. (D) The electrostatic potential surface of Cac1C shows a positively charged region composed by eight basic residues. The positively charged, negatively charged and neutral regions are shown in blue, red and white, respectively.

Figure 2.

The DNA-binding ability of Cac1C assessed by EMSA. (A) The DNA-binding ability of Cac1C was reduced by mutations at charged residues locating on the positively charged surface. A total of 5 μM of 58 bp FAM labeled DNA was incubated with 20 μM WT Cac1C or different Cac1C mutant proteins and analyzed in 4% native PAGE. The gels were scanned with Typhoon Troi + (GE Healthcare). (B) Quantification of DNA binding of WT and each mutant in panel A. The integrated density value of each band was quantified by ImageJ and normalized against WT proteins. Error bars indicate the average and SD of three independent experiments calculated by Student's t-test. *P < 0.05 compared to WT. (C) Cac1C showed similar binding affinity to AT-rich and GC-rich DNA. (D) Mutations N565A and Y570A had no apparent effect on the DNA-binding ability of Cac1C. The results of panel C and D came from analyzing of data shown in Supplementary Figure S5.

Figure 3.

Cac1C contributes to the interaction between CAF-1 and DNA, and DNA-binding mutations have no apparent effect on the ability of CAF-1 to bind its partners. (A) Cac1C was involved in the interaction between CAF-1 and DNA. Equal amounts of streptavidin sepharose beads labeled with 58 bp DNA were used to pull down WT or mutated CAF-1 complex (Cac1, GST-Cac2 and Cac3). Pull-downed proteins were detected by immunoblotting with antibodies against GST. Compared with WT, two CAF-1 mutants, K564E/K568E and R582E/K583E, showed reduced DNA-binding abilities. GST-tagged Cac2 alone was used as a negative control. (B) Cac1C does not contribute to the interaction between CAF-1 and H3–H4. GST-Cac1C, WT and mutated GST-CAF-1 complexes (Cac1, GST-Cac2, Cac3) were used to pull down histone H3–H4 at 300 mM NaCl. Pulled-down proteins were detected by Coomassie Blue staining. Note that Cac1 and GST-Cac2 were at the similar position in the SDS-PAGE gel (Supplementary Figure S7). (C) Cac1C does not interact with PCNA in vitro. GST-Cac1 PIP domain, GST-Cac1 PIP mutant (F233A/F234A, GST-mPIP) and GST-Cac1C were used to pull down recombinant his-tagged PCNA. PCNA was detected by immunoblotting with antibodies against his. GST tagged proteins were detected by Coomassie Blue staining. GST was use as negative control. (D) CAF-1 mutants defective in DNA binding do not affect the CAF-1-PCNA interaction in vitro. The experiments were performed as described above except that GST-Cac2 (negative control) and WT or mutated GST-CAF-1 complex were used in the experiments.

Figure 4.

The Cac1C-DNA interaction and the Cac1-PCNA interaction function in parallel to mediate Cac1's role in HMR silencing and response to DNA damage. (A and B) Cells harboring Cac1 DNA binding mutant exhibit defects in transcriptional silencing at the HMR locus (A) as well as increased sensitivity toward DNA damaging agent CPT (B). Cells with indicated genotypes were analyzed for GFP expression using FACS. The W303 cells (less than 1%) and cells with sir3Δ (over 95%) were used as standards for GFP expression. Error bars indicate SEM analysis of at least three independent experiments. (B) The sensitivity to DNA damage was analyzed through a 10-fold serial dilution of cells of indicated genotype onto regular growth media (SCM-His) or media with CPT. (A and C) Cells expressing Cac1 mutants harboring mutations at both DNA-binding domain and PCNA binding motif showed an synthetic defects in HMR silencing (A) and response to CPT (C). The experiments were repeated at least three times and Student's t-test was used to calculate P-value. *P < 0.05 compared to WT, ^#P < 0.05 compared to empty vector. The expression of the WT and mutated Cac1 is shown in Supplementary Figure S10.

Figure 5.

Mutations at mouse p150 DNA-binding domain resulted in mis-localization at replication foci. (A) Representative images of WT GFP-p150, PCNA and DAPI staining in early, middle and late S phase and non-S phase cells. GFP-tagged WT p150 was expressed in mouse NIH3T3 cells using lentivirus-based system. GFP and PCNA were stained with monoclonal antibodies. DAPI staining was used to mark foci of pericentric heterochromatin. PCNA does not bind to chromatin in non-S phase cells and the soluble PCNA will be washed away after pre-extraction and therefore little PCNA IF signals (black) were detected in non-S phase cells. (B) Representative images of mutant GFP-p150 showing partial- and full-localization defects. In cells exhibiting partial-localization defect, GFP-p150 co-localized with both PCNA and heterochromatin foci (DAPI), while all the GFP-p150 mis-targeted to heterochromatin foci in cells exhibiting full-localization defect cells during S phase. PCNA staining was used to mark S phase cells. (C) Analysis of the percentage of cells showing partial- and full-localization defects. Error bars indicate standard error of mean (SEM) of at least three experiments calculated by Student's t-test. *P < 0.05 partial defect compared to WT, ^#P < 0.05 full defect compared to WT. The expression of the WT and mutated p150 is shown in Supplementary Figure S16.

Figure 6.

A model for the WHD–DNA interaction in stabilizing CAF-1 at replication forks. (A) CAF-1 is recruited to replication forks through its interaction with PCNA. (B) After recruited, CAF-1 binds to DNA in close proximity through its WHD, thereby stably associating with replication forks.