
==== Front
Proc Natl Acad Sci U S A
Proc Natl Acad Sci U S A
PNAS
Proceedings of the National Academy of Sciences of the United States of America
0027-8424
1091-6490
National Academy of Sciences

38285941
202315596
10.1073/pnas.2315596121
research-articleResearch ArticlegeneticsGenetics419
Biological Sciences
Genetics
Specialized replication of heterochromatin domains ensures self-templated chromatin assembly and epigenetic inheritance
Nathanailidou Patroula a https://orcid.org/0000-0003-3532-0939

Dhakshnamoorthy Jothy a
Xiao Hua a
Zofall Martin a https://orcid.org/0000-0001-6331-0127

Holla Sahana a https://orcid.org/0000-0001-5186-0180

O’Neill Maura b
Andresson Thorkell b
Wheeler David a
Grewal Shiv I. S. grewals@mail.nih.gov
a 1
aLaboratory of Biochemistry and Molecular Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892
bCancer Research Technology Program, Frederick National Laboratory for Cancer Research, Frederick, MD 21701
1To whom correspondence may be addressed. Email: grewals@mail.nih.gov.
Edited by Douglas Koshland, University of California, Berkeley, CA; received September 7, 2023; accepted December 21, 2023

29 1 2024
6 2 2024
29 7 2024
121 6 e231559612107 9 2023
21 12 2023
Copyright © 2024 the Author(s). Published by PNAS.
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This article is distributed under Creative Commons Attribution-NonCommercial-NoDerivatives License 4.0 (CC BY-NC-ND).

Significance

Heterochromatin domains defined by specific histone methylation patterns are a major feature of mammalian genomes. Epigenetic pathways ensure stable heterochromatin inheritance through cell division to prevent the expression of lineage-inappropriate genes, and defects in this process can lead to many diseases including cancer. Previous studies have shown that heterochromatin maintenance requires a critical threshold of methylated histones, but how modified parental histones are retained during DNA replication is unclear. We find that heterochromatin domains are specialized replication zones enriched in replisome components, which together with other factors help retain parental histones. This mechanism is further supported by the boundary elements surrounding the heterochromatin domains. DNA replication is thus coordinated with histone preservation to facilitate epigenetic inheritance of heterochromatin.

Heterochromatin, defined by histone H3 lysine 9 methylation (H3K9me), spreads across large domains and can be epigenetically inherited in a self-propagating manner. Heterochromatin propagation depends upon a read–write mechanism, where the Clr4/Suv39h methyltransferase binds to preexisting trimethylated H3K9 (H3K9me3) and further deposits H3K9me. How the parental methylated histone template is preserved during DNA replication is not well understood. Here, we demonstrate using Schizosaccharomyces pombe that heterochromatic regions are specialized replication domains demarcated by their surrounding boundary elements. DNA replication throughout these domains is distinguished by an abundance of replisome components and is coordinated by Swi6/HP1. Although mutations in the replicative helicase subunit Mcm2 that affect histone binding impede the maintenance of a heterochromatin domain at an artificially targeted ectopic site, they have only a modest impact on heterochromatin propagation via the read–write mechanism at an endogenous site. Instead, our findings suggest a crucial role for the replication factor Mcl1 in retaining parental histones and promoting heterochromatin propagation via a mechanism involving the histone chaperone FACT. Engagement of FACT with heterochromatin requires boundary elements, which position the heterochromatic domain at the nuclear peripheral subdomain enriched for heterochromatin factors. Our findings highlight the importance of replisome components and boundary elements in creating a specialized environment for the retention of parental methylated histones, which facilitates epigenetic inheritance of heterochromatin.

epigenetic
heterochromatin
histone methylation
gene silencing
HHS | NIH | National Cancer Institute (NCI) 100000054 ZIA BC 011208 Patroula NathanailidouJothy DhakshnamoorthyHua XiaoMartin ZofallSahana HollaDavid WheelerShiv I. S. Grewal
==== Body
pmcGene expression patterns in eukaryotic genomes are tightly regulated to direct differentiation of cell types. To achieve this, the genome is organized into two broad chromatin categories: euchromatin, which contains actively expressed genes, and heterochromatin, which coats silenced regions (1). Euchromatin and heterochromatin domains contain specific posttranslational modifications of histones. In euchromatin, histones are acetylated and methylated on histone 3 lysine 4, while in heterochromatin, histones are generally hypoacetylated and methylated on histone 3 lysine 9 (2–4). Remarkably, heterochromatin domains can be epigenetically inherited, thereby preventing the misexpression of lineage-inappropriate genes and maintaining silencing of repetitive DNA elements (1). However, how heterochromatic structures are stably inherited during cell division is not well understood.

The fission yeast Schizosaccharomyces pombe is an excellent model system for studying epigenetic inheritance of heterochromatin (1). Heterochromatin coats extended chromosomal domains associated with pericentromeric repeats, subtelomeric regions, and the well-characterized silent mating-type (mat) region (5). These domains contain dg and dh repeat elements that serve as heterochromatin nucleation centers. Cotranscriptional processing of repeat transcripts by RNAi machinery including Argonaute (Ago1), Dicer (Dcr1), and RNA-dependent RNA polymerase (Rdp1), is linked to the recruitment of Clr4 (6–10), the sole S. pombe homolog of the mammalian SUV39H1 and SUV39H2 methyltransferase enzymes (11, 12). As a subunit of the ClrC complex (13), Clr4 methylates H3K9 (“write”) to nucleate heterochromatin (11), but also binds H3K9me3 via its chromodomain (“read”) (10). Read–write activity allows Clr4/Suv39h to bind preexisting H3K9me3 and further deposit H3K9me, promoting the spreading of heterochromatin (10, 14). At the silent mat interval, H3K9me is initially targeted to the cenH element sharing homology to dg/dh repeats and a nearby site (7, 15, 16), and then spreads across a 20-kb domain surrounded by inverted repeat (IR) boundary elements (2). H3K9me allows chromatin association of HP1 proteins (11, 17–19). Besides binding to H3K9me, HP1 proteins engage in low-affinity multivalent interactions (20, 21) to form high-concentration territories, providing an increased interface to concentrate various effector proteins (1), including the histone deacetylase (HDAC) Clr3, a component of the Snf2/HDAC repressor complex (SHREC) (22, 23).

Remarkably, heterochromatin can be faithfully passed to daughter cells during cell division. Heterochromatin inheritance occurs in cis via a self-templating mechanism (7, 24), and H3K9me3 plays a crucial role as an epigenetic carrier (10, 25). A critical density of H3K9me3 is required to maintain a high local concentration of Clr4/Suv39h on chromatin to support read–write activity (25). Various strategies maintain H3K9me3 levels at different genomic loci. Pericentromeric regions depend on RNAi, targeting multiple copies of dg/dh repeats, to maintain H3K9me3, while the silent mat region containing a single copy of dg/dh element utilizes a distinctive mechanism (1). At mat, high levels of the Clr3 HDAC recruited by HP1-dependent and -independent mechanisms safeguard H3K9me3 by preventing nucleosome turnover (26, 27). Robust chromatin association of Clr3, or the absence of the anti-silencing factor Epe1, results in the stabilization of methylated nucleosomes and allows epigenetic inheritance of heterochromatin via H3K9me3 at endogenous and ectopic loci (7, 27–29). Additionally, the RNA processing Rix1 containing complex RIXC, which binds Swi6/HP1 with high affinity and tethers mat to the nuclear periphery, facilitates heterochromatin propagation (30).

How H3K9me3 histones are transmitted to sister chromatids during DNA replication, which involves the unraveling of parental nucleosomes, requires further exploration. DNA replication initiates at replication origins distributed across the genome (31). The origin recognition complex (ORC) bound to origins, along with the licensing factors Cdt1 and Cdc6/Cdc18, loads the minichromosome maintenance (MCM) complex to assemble the prereplicative complex (pre-RC) in the G1 phase of the cell cycle (32). Subsequently, origins are activated by the conserved cyclin-dependent kinase (CDK) and Dfp4-dependent kinase (DDK), which are required for the recruitment of Cdc45 and the GINS (Go-Ichi-Ni-San) complex in S phase. This process leads to activation of the MCM helicase, unwinding of parental DNA, and assembly of DNA polymerases for synthesis of the leading and lagging strands (33). Interestingly, replisome components are believed to direct the transfer of parental histones onto daughter DNA strands during DNA replication (34, 35). How these factors coordinate the replication of silenced domains with their role in transmission of parental histones, required for self-propagation of heterochromatin, remains to be explored.

Here, we investigated the requirement for replication factors in heterochromatin propagation in S. pombe. We find that heterochromatic domains, such as those at the silent mat and pericentromeric regions, are specialized replication zones delimited by boundary elements. Replisome components are preferentially enriched across these zones and Swi6/HP1, a factor previously implicated in the epigenetic inheritance of heterochromatin (7, 36), coordinates replication activation across extended domains. Our analyses identified a crucial role for the replication factor Mcl1 in heterochromatin propagation via a mechanism involving the histone chaperone FACT. Boundary elements facilitate loading of FACT at mat by positioning it at the nuclear periphery. Together with our analysis of the histone chaperone function of the MCM complex subunit, Mcm2, these findings underscore the critical requirement of replication proteins in retaining parental methylated histones to transmit epigenetic memory for heterochromatin propagation.

Results

Genetic Screen for Factors Affecting Heterochromatin Propagation.

Previous studies of the archetypical mating-type heterochromatin domain comprising the mat2P and mat3M loci provided important insights into the establishment and maintenance of heterochromatic structures. Genetic screens using silencing reporter systems identified many conserved heterochromatin factors (30, 37–40). We screened for additional factors that affect heterochromatin propagation by exploiting the fact that silencing at mat is enforced by two distinct mechanisms. In addition to heterochromatin-mediated silencing, REII and REIII elements recruit HDACs to locally silence mat2P and mat3M loci, respectively (27). When the REII element is deleted, mat2P silencing becomes solely dependent on the spreading of heterochromatin, creating a sensitized region. In nonswitching mat1M (mat1M-smt0) cells lacking REII, defects in heterochromatin propagation cause mat2P expression and aberrant sporulation (haploid meiosis) caused by coexpression of both M and P mating-type information in haploid cells (37). Cells undergoing haploid meiosis stain a dark color upon exposure to iodine vapor, whereas wild-type (WT) colonies stain yellow. A ura4+ reporter inserted close to mat2P (mat2P::ura4+) provides an additional readout for heterochromatic silencing.

Following UV mutagenesis of the reporter strain carrying REIIΔ and mat2P::ura4+, colonies were screened for haploid meiosis and mat2P::ura4+ expression. Dark-stained colonies that grew on medium lacking uracil were backcrossed to a nonmutagenized parental strain (Fig. 1A). Whole genome sequencing of mutant and WT segregants revealed mutations in known heterochromatin factors, such as the ClrC components Clr4 and Raf1 (13), and the SHREC components Clr3 and Mit1 (22) (SI Appendix, Fig. S1 A and B). Additionally, mutations in the chromatin remodeler Fft3, implicated in heterochromatin propagation (40, 41), factors linked to DNA replication, including the replisome component Mcl1 (an ortholog of S. cerevisiae Ctf4 and mammalian AND1) and the Mediator of Replication Checkpoint Mrc1, were identified (Fig. 1A and SI Appendix, Fig. S1 A and B) (42–46).

Fig. 1. The replisome component Mcl1 affects heterochromatin assembly. (A) Schematic representation of the sensitized mating-type locus (REIIΔ mat2P::ura4+) reporter used for the genetic screen (Top). Step-by-step illustration of the UV mutagenesis screen (Bottom). The mcl1-2 mutant contains a single nucleotide insertion in the mcl1 gene. (B) Ten-fold serial dilutions of the indicated strains were plated on nonselective (N/S) medium, medium lacking uracil (−Ura), or FOA-containing medium (+FOA) to assess derepression of the ura4+ reporter. Haploid meiosis was assessed by iodine staining. (C) H3K9me3 fold enrichment determined by ChIP-qPCR analysis at cenH and mat2P loci (mean ± SEM; n = 3). The location of the oligos used is indicated in (A). (D) Strains expressing untagged Mcl1 or Mcl1-GFP were subjected to immuno-affinity purification and analyzed by SDS-PAGE. (E) The total peptide coverage (%) of the identified Mcl1-interacting proteins was determined by mass spectrometry. (F) Co-IP of Mcm2-FLAG or the FACT subunit Pob3-FLAG with Mcl1-GFP. See also SI Appendix, Fig. S1.

Although Mrc1 and Mcl1 were found in previous screens (30, 39, 47–49), their exact function was unknown. The mutants showed a variable increase in haploid meiosis and mat2P::ura4+ derepression, as indicated by growth on medium lacking uracil (−Ura) and counterselective medium containing 5-Fluoroorotic Acid (FOA) (SI Appendix, Fig. S1A). The mrc1R558stop mutant, which lacks the conserved Mrc1-like domain, and mrc1Δ showed only mild silencing defects (SI Appendix, Fig. S1 A and C) (47). By contrast, mcl1-2 cells bearing a single nucleotide insertion at position 618 showed a more severe silencing defect, and we therefore focused on the role of Mcl1 in heterochromatin assembly.

The slow-growth phenotype of mcl1-2 limited our analyses. Fortuitously, cells expressing Mcl1 tagged at the carboxy terminus with the HA or FLAG epitope (referred to as mcl1-3 and mcl1-4, respectively) showed a severe silencing defect without affecting growth (Fig. 1B). By contrast, Mcl1 tagged with GFP (Mcl1-GFP) did not affect silencing or H3K9me3 levels, indicating that it remained functional (SI Appendix, Fig. S1D). Chromatin Immunoprecipitation and quantitative PCR (ChIP-qPCR) analyses of H3K9me3 in mcl1-3 and mcl1-4 showed a reduction near the mat2P locus and no major effect at the cenH nucleation site (Fig. 1C). Similar changes were observed in the slow-growing mcl1-2 (Fig. 1C). Thus, targeting of heterochromatin to the nucleation site occurs independently of Mcl1, which instead affects the spreading and/or maintenance of heterochromatin.

Mcl1 Associates with the Replisome and the Histone Chaperone FACT.

To explore the role of Mcl1 in heterochromatin assembly, we performed mass spectrometry analysis of immuno-affinity purified Mcl1-GFP (Fig. 1 D and E and SI Appendix, Fig. S1E). Several replisome factors, including the MCM and GINS holo-complexes, as well as Cdc45, DNA polymerase alpha (Polα), replication factor C (RFC), and RPA components were identified (Fig. 1E and SI Appendix, Fig. S1E). Thus, Mcl1 is a component of the active replisome in S. pombe, as in other systems (43, 44, 46). Components of the SCF (Skp1-Cul1-F box) E3 ubiquitin ligase and proteasome components potentially involved in controlling Mcl1 protein levels were also present (SI Appendix, Fig. S1E) (50). Intriguingly, the Spt16 and Pob3 subunits of FACT (Fig. 1E), which is implicated in the propagation of heterochromatin domains (30, 51–54), also copurified. Coimmunoprecipitation (co-IP) experiments confirmed the association of Mcl1 with Mcm2 and FACT (Fig. 1F).

Heterochromatin Domains Are Specialized Replication Zones.

To assess whether Mcl1 plays a direct role in heterochromatin assembly, we blocked cells carrying the cdc25-22 temperature-sensitive allele at the G2/M phase boundary, then released them synchronously into S phase, indicated by the presence of cell septa (Fig. 2A). The distributions of Mcl1 and the Mcm2 replisome component were determined alongside the incorporation of 5-bromo-2-deoxyuridine (BrdU) by ChIP-sequencing (ChIP-seq). Mcl1 localized to heterochromatic loci in S phase, like Mcm2 (Fig. 2B). Moreover, its localization was observed concurrently with BrdU incorporation, suggesting that Mcl1 is involved in DNA replication of these domains. Significantly, the prominent enrichment pattern of Mcl1, Mcm2, and BrdU incorporation across heterochromatin domains at the silent mat region and pericentromeric repeats was delimited by their surrounding boundary elements (Fig. 2 B and C). This distribution contrasted with euchromatin regions, where Mcm2 appeared as distinct peaks corresponding to replication origins around which BrdU was incorporated (SI Appendix, Fig. S2A).

Fig. 2. Replisome components coat heterochromatin domains. (A) Illustration of the cell cycle synchronization process using the cdc25-22 temperature-sensitive mutant. Cells were arrested at the G2/M boundary and then synchronously released into the cell cycle. The septation index was used to follow cell cycle progression. (B and C) Fold enrichments of Mcl1 and Mcm2 at mat (B) and centromere 2 (C) during G2 and S phases determined by ChIP-seq. S phase entry was confirmed by the incorporation of the thymidine analog BrdU. (D and E) Fold enrichments of Mcm2 and BrdU in WT and swi6Δ cells at mat (D) and centromere 2 (E) during the S phase determined by ChIP-seq. ChIP-seq analysis of Orc1 in WT cells upon cdc10-v50 block at G1 is also presented. See also SI Appendix, Figs. S2 and S3.

Swi6/HP1 controls early replication at specific sites within the mat and centromeric loci by targeting DDK (55). To determine whether Swi6/HP1 coordinates replication throughout these heterochromatin domains, we mapped Mcm2 and BrdU in cells lacking Swi6/HP1 during S phase (Fig. 2 D and E and SI Appendix, Figs. S2 A–D and S3). Mcm2 mapping was performed 60 min after release from the cdc25-22 block. A similar septation index for WT and swi6Δ cells indicated normal progression of swi6Δ cells through S phase (SI Appendix, Fig. S2B). For BrdU, ChIP analysis was performed 60, 70, and 80 min after release in the presence of HU, which blocks cells in early S phase. Indeed, BrdU was absent from the late replicating subtelomeric regions at the time points tested (55) (SI Appendix, Fig. S2C). Compared to WT, swi6Δ cells showed a considerable decrease in BrdU incorporation across mat, which persisted through the time course (Fig. 2D and SI Appendix, Fig. S3). Mcm2 was still detected, although with more defined peaks, likely reflecting licensed, but not activated, replication origins across the domain (Fig. 2D). The Mcm2 peaks in swi6Δ coincided with sites of preferential enrichment of the ORC subunit Orc1 in cells arrested in G1 phase (Fig. 2D). BrdU incorporation was also affected at centromeres (Fig. 2E). Mcm2 was still detectable in this case as well (Fig. 2E). At subtelomeric regions, despite the absence of BrdU incorporation, which is consistent with their late replication timing (55), ORC and Mcm2 were again heavily enriched, confirming this is as a common feature among heterochromatin domains regardless of their replication timing (SI Appendix, Fig. S2C). Of note, swi6Δ had no apparent effect on replication in other parts of the genome, indicating a specific role for Swi6/HP1 in the replication of heterochromatin domains (SI Appendix, Figs. S2A and S2D).

Collectively, our results suggest that heterochromatic regions are specialized replication zones in which Swi6/HP1 coordinates replication of the entire domain.

Boundary Elements Are Not Necessary for Replication in Heterochromatin Propagation.

The boundary elements IR-L and IR-R surrounding the silent mat region are suggested to have replication origin activity, and their deletion affects heterochromatin propagation (56–59). We therefore wondered whether IRs affect heterochromatin propagation by impacting the replication of this domain.

We used CRISPR to simultaneously delete sequences in both IR-R and IR-L and then looked for effects on heterochromatin propagation using the sensitized REIIΔ mat2P::ura4+ reporter. We targeted regions containing B-boxes, which are critical for their function (56, 57). IR-Δ1 removed a T-rich region and an adjacent DNAase I hypersensitive site, IR-Δ2 deleted B-boxes 3 to 5 and a DNAase I hypersensitive site, and IR-Δ3 eliminated B-boxes 1 and 2 (Fig. 3A). While IR-Δ1 and IR-Δ3 had minor effects, IR-Δ2 showed a variegated pattern of derepression, indicated by dark iodine staining from elevated levels of haploid meiosis and growth analysis on −Ura and FOA medium (Fig. 3B). Consistently, defects in H3K9me3 spreading from the nucleation site to surrounding sequences were observed in IR-Δ2, but not IR-Δ1 or ΙR-Δ3 (Fig. 3C).

Fig. 3. Boundary sequences required for heterochromatin propagation are dispensable for replisome recruitment. (A) Schematic showing deletion constructs IR-Δ1, IR-Δ2, and IR-Δ3. (B) Ten-fold serial dilutions were plated on the indicated media to assess derepression of ura4+. (C) H3K9me3 fold enrichments determined by ChIP-chip. (D) Mcm2 and BrdU fold enrichments at mat determined by ChIP-seq in cells synchronized in the S phase. (E) Fold enrichment of Rix1 determined by ChIP-seq. (F) Representative images showing localization of mat in three distinct nuclear zones (Left). IR-R>lacO/LacI-GFP marks mat. Nup40-mCherry marks the nuclear envelope. The percentage of cells in each zone is plotted for WT (n = 183), IR-Δ2 (n = 181), swi6Δ (n = 192), and IR-Δ2 swi6Δ (n = 167). n represents the combined number of cells from two independent experiments. Pairwise comparisons between the WT and each mutant were performed using two-way ANOVA (mean ± SEM). *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. (G) Ten-fold serial dilutions were plated to assess derepression of ura4+ in WT or IR-Δ2 cells carrying an empty vector (pREP1) or a plasmid containing swi6+. Representative immunofluorescence images against Swi6 are shown (Bottom). See also SI Appendix, Fig. S4.

We then tested whether silencing defects are related to defects in replication across the silent mat region and found that IR-Δ2 showed neither major changes in Mcm2 localization nor significant changes in BrdU incorporation across the domain (Fig. 3D). Therefore, the heterochromatin propagation function of the boundary elements is not linked to dramatic changes in the replication program of the mat region, which most likely utilizes origins within the domain to replicate.

IR boundaries recruit the RNA processing complex RIXC, which also binds Swi6/HP1 with high affinity, and tethers the mat region to a nuclear peripheral subdomain enriched for silencing proteins (30). This facilitates the loading of FACT, which prevents the turnover of histones to maintain heterochromatin (30, 53). Unlike IR-Δ1 and IR-Δ3, IR-Δ2 showed severe disruption of Rix1 localization at IRs (Fig. 3E and SI Appendix, Fig. S4A), although Rix1 was still detected internally at the silent mat domain where Swi6/HP1 loads RIXC (Fig. 3E) (30). IR-Δ2 cells presented delocalization of mat, marked with lacO/lacI-GFP, from the periphery during S phase (Fig. 3F). Loss of Swi6/HP1 that removes RIXC from the silent mat interval further affected its peripheral localization (Fig. 3F). Swi6 overexpression, which causes a pan-nuclear localization, rescued the silencing defects in IR-Δ2 cells resulting from delocalization of mat from the periphery (Fig. 3G). Thus, boundaries are important for the peripheral localization of the mat domain during S phase.

We asked whether delocalization of mat from the periphery in IR-Δ2 cells is linked to a decrease in FACT loading. During S phase in IR-Δ2 cells, the Pob3 subunit of FACT was significantly reduced, while the anti-silencing factor Epe1 was increased at the silent mat domain (SI Appendix, Fig. S4 B and C). Loss of FACT and elevated levels of Epe1 can lead to increased histone turnover (26, 30, 53), ultimately affecting the H3K9me3 density required to support Clr4/Suv39h read–write activity (25). Cells carrying IR-Δ2 alone or in combination with a deletion of the HDAC clr3 showed increased histone H3 turnover compared to WT (SI Appendix, Fig. S4 D–F). Additionally, the heterochromatin defects in IR-Δ2 could be rescued by expressing the chimeric Clr4Chp1CD protein (25), in which the Clr4 chromodomain (CD) is replaced by the Chp1 CD, which restores a high local concentration of Clr4 through the high-affinity binding of Chp1 CD to H3K9me2 (SI Appendix, Fig. S4G).

These results show that the function of boundaries in heterochromatin propagation is closely associated with RIXC association and its downstream effects, such as the loading of FACT, which is required to maintain methylated histones and support read–write.

Mcm2 Affects Heterochromatic Silencing through Histone Binding.

Mcm2 is believed to transfer parental histones to daughter strands during DNA replication in other species (60–63). To determine whether Mcm2 also contributes to heterochromatic silencing in S. pombe, we used an error-prone PCR technique to generate mcm2 mutants affecting mat silencing using the REIIΔ mat2P::ura4+ reporter (Fig. 4A). We isolated 3 mutants: mcm2-1, mcm2-2, and mcm2-6. Interestingly, the mutations in mcm2-1 (D87G and D112G) and mcm2-2 (D79V and D112G) correspond to conserved residues that are required for the H3-H4 histone binding activity of the human MCM2 (62) (Fig. 4B and SI Appendix, Fig. S5A), while the third mutant (mcm2-6) contains a mutation (S218F) near the MCM box (SI Appendix, Fig. S5B). The mcm2-6 mutant showed sensitivity to hydroxyurea (HU) (SI Appendix, Fig. S5C), indicating defective replication, and was not further analyzed. The other two mutants showed defective silencing of the mat2P::ura4+ reporter and haploid meiosis (Fig. 4C), as well as detectable levels of the mat2P transcript (Fig. 4D).

Fig. 4. Mutations in the histone binding domain (HBD) of Mcm2 cause silencing defects. (A) The random mutagenesis screen used to isolate mcm2 mutants defective in heterochromatic silencing. (B) The mcm2-1 and mcm2-2 mutants isolated from the screen. The blue solid box represents the HBD of Mcm2. The identified amino acid substitutions are indicated by red lines. (C) Ten-fold serial dilutions were spotted on nonselective, −Ura or +FOA medium to assess derepression of ura4+. (D) RT-qPCR analysis of mat2P expression (mean ± SD; n = 2). (E and F) In vitro histone binding assay using cells expressing the WT or mutated HBD of Mcm2. The ability of WT or mutant HBDs to bind H3/H4 histones was assessed by affinity chromatography (E). The fraction of H3/H4 histones pulled down is plotted as the percentage of the HBD input signal in each lane (F). (G) Co-IP analysis of histone H3 with Mcm2 in WT and mcm2-1 cells (Top) or mcm2-2 cells (Bottom). See also SI Appendix, Fig. S5.

To determine whether WT and mutant Mcm2 bind to histones H3 and H4, we purified the histone binding domains (HBD). Both H3 and H4 histones copurified with the WT HBD, but H3 did not copurify with either the mcm2-1 or mcm2-2 mutant HBDs (Fig. 4 E and F). Furthermore, the mcm2-1 HBD was severely compromised in its ability to bind histone H4 (Fig. 4 E and F). Notably, the addition of the WT Mcm2 HBD during the purification of the H3 and H4 histones promoted their solubilization, highlighting its role as a histone chaperone (SI Appendix, Fig. S5D). In vivo co-IP experiments confirmed defective binding of H3 in the mcm2-1 and mcm2-2 mutants (Fig. 4G). These analyses demonstrate that Mcm2 binds histone H3 and H4 via its HBD and that mutations that affect this interaction cause silencing defects.

mcl1 and mcm2 Mutants Show Progressive Loss of Heterochromatic Silencing.

Mutations in replisome components might cause progressive loss of heterochromatic silencing as cells undergo successive replication cycles. To assess this, we employed a fluorescent reporter system consisting of a YFP gene inserted within the cenH nucleation element (Kint2::YFP) and a mCherry gene inserted near mat3M (mat3M::mCherry) (56) (Fig. 5A). This system allows simultaneous monitoring of silencing at cenH and a site distal to the nucleation center within the mat region at the single cell level, and quantitative fluorescence imaging of the cassettes determines derepression at each position.

Fig. 5. Effects of mutations in Mcl1 and Mcm2 on retention of parental histones and heterochromatin propagation. (A) Schematic of the reporter cassettes used to assay silencing by time-lapse microscopy. YFP is embedded in the cenH (Kint2::YFP), while mCherry is embedded close to the distal mat3M gene (mat3M::mCherry). (B) Representative images from the time-lapse analysis of WT, mcm2-1, and mcl1-4 for generations 0 and 6. (C) The numbers of mcm2-1 and mcl1-4 cells expressing mCherry per generation are plotted. The numbers of cells from independent fields followed using time-lapse microscopy are presented for each case (mcm2-1 n = 4; mcl1-4 n = 3). (D) Mean intensities of mCherry positive mcm2-1 and mcl1-4 cells per generation. Cumulative numbers are plotted. mCherry and YFP expression were absent in WT cells in all fields tested. (E) Schematic representation of the histone retention assay (Top). H3-FLAG enrichment at mat2P relative to mitochondrial tRNA was quantified by qPCR at the indicated time points after H3-FLAG suppression for mcm2-1 and mcl1-4 (mean ± SEM; n = 2). (F) Fold enrichment of H3K9me3 at the silent mat interval was determined by ChIP-seq. See also SI Appendix, Figs. S6–S8.

We monitored the silencing of Kint2::YFP and mat3M::mCherry reporter genes in WT, mcm2-1, and mcl1-4 cells that were allowed to divide through 6 generations. Similar doubling times for both WT and mutant cells ruled out any impact from different generation times between the strains (SI Appendix, Fig. S6A). In WT cells both cassettes remained repressed through six cell divisions. However, mcm2-1 and mcl1-4 showed derepression of mat3M::mCherry beginning at generation 2 to 3, and the number of mCherry positive cells increased with cell divisions (Fig. 5 B and C). Interestingly, mcl1-4 caused derepression of mat3M::mCherry in a wider cell population compared to mcm2-1 (Fig. 5C and SI Appendix, Fig. S6B). Quantification showed that successive cell divisions led to cells with higher mCherry intensities, reflecting increasingly severe silencing defects (Fig. 5D).

Heterochromatin Propagation at an Ectopic Site Requires Mcl1 and Mcm2.

We next tested whether Mcl1 and Mcm2 are required for maintenance of heterochromatin at an ectopic site. We inducibly expressed Clr4/Suv39h fused to a TetRoff DNA-binding domain (TetR-Clr4) in cells harboring an ade6+ reporter located downstream of six tetracycline operators (6xtetO-ade6+) (28, 30). The strains also expressed Clr3 HDAC fused to two chromodomains (Clr3-CDx2) (27). TetR-Clr4 recruitment to tetO causes accumulation of H3K9me3 and silencing of the ade6+ reporter. Clr3-CDx2 binds to methylated nucleosomes with high affinity, leading to histone deacetylation and stabilization of nucleosomes by preventing their turnover, supporting stable propagation of heterochromatin (27). Indeed, upon release of TetR-Clr4 in cells expressing Clr3-CDx2, the read–write activity of endogenous Clr4/Suv39h maintains heterochromatin for multiple generations (27) (SI Appendix, Fig. S7). Tethering of TetR-Clr4 established heterochromatic silencing of ade6+ in mcm2-1 and mcl1-4 mutant cells, as indicated by the red coloration of colonies, and H3K9me3 levels were comparable to WT (SI Appendix, Fig. S7). However, following the release of TetR-Clr4, mcm2-1 and mcl1-4 cells failed to maintain heterochromatin, as indicated by white colony color and loss of the H3K9me3 mark (SI Appendix, Fig. S7). Thus, both Mcl1 and Mcm2 are required for propagation of heterochromatin at an ectopic site.

Mcl1 Is More Critical Than Mcm2 for the Retention of Parental Histones.

Defects in heterochromatin propagation in mcl1 and mcm2 mutant cells led us to explore their involvement in the retention of parental histones in cycling cells. To test whether the “dilution” of parental histones coupled to DNA replication occurs more readily in mcl1 or mcm2 mutant cells, we employed an assay that directly measures the retention of histones at heterochromatic loci in cycling cells. Flag-tagged histone H3 (H3-FLAG) under the control of the inducible promoter inv1 allows the rapid switch between H3-FLAG expression or repression simply by using sucrose or glucose as a carbon source for growth, respectively (64). Cells grown overnight in sucrose-containing medium to express FLAG-tagged H3 were shifted to glucose-containing medium to turn off H3-FLAG expression (SI Appendix, Fig. S8A). Time course analysis of H3-FLAG dilution at the silent mat region showed more rapid loss in mcm2-1 and mcl1-3 compared to WT, and mcl1 mutant cells lost H3-FLAG considerably faster than the mcm2-1 mutant cells, defective in binding to histones (Fig. 5E). Thus, although both mutants affect the retention of parental histones, Mcl1 plays a more critical role in the process.

Mcl1, but not Mcm2, Is Vital for Endogenous Heterochromatin Propagation.

We next assessed propagation of an endogenous heterochromatin domain in mcl1 and mcm2 mutants. Heterochromatin at the silent mat region, initially established by an RNAi-dependent pathway, can propagate for multiple generations in the absence of RNAi via the Clr4/Suv39h read–write mechanism (7, 10). This process requires factors such as Clr3 HDAC and FACT, which maintain high H3K9me3 density (26, 30, 53). When Clr3 and FACT are mutated, heterochromatin can be nucleated at cenH via RNAi, but it cannot spread due to failure to preserve H3K9me3.

ChIP-seq analysis of H3K9me3 distribution was performed in mcl1-4 and mcm2-1 single mutants, and in double mutants that also contained ago1Δ, which disrupts heterochromatin nucleation at cenH (7). In mcl1-4, H3K9me3 was detected at the cenH nucleation site but failed to spread (Fig. 5F), similar to mutants defective in heterochromatin propagation. Moreover, unlike ago1Δ, which can maintain heterochromatin at mat via the read–write mechanism (7, 10), mcl1-4 ago1Δ showed complete loss of H3K9me3 throughout mat (Fig. 5F). Cells expressing mcm2-1 defective in binding to histone H3 showed only a moderate reduction in H3K9me3 across mat (Fig. 5F), suggesting that heterochromatin propagation can occur, albeit slightly less efficiently. Further supporting this argument, a substantial level of H3K9me3 was retained throughout the silent mat interval in mcm2-1 ago1Δ cells (Fig. 5F). The retention of H3K9me3 required the read–write activity of Clr4/Suv39h. A mutation in Clr4 chromodomain (Clr4W31G), which disrupts its binding to methylated H3K9 (10), abolished H3K9me3 in mcm2-1 ago1Δ cells (Fig. 5F). Based on these results, the read–write mechanism can still maintain H3K9me3 across an endogenous heterochromatin domain in cells with defects in the histone chaperone function of Mcm2. On the other hand, Mcl1 is vital for this process.

We also investigated the role of Mcl1 and Mcm2 at pericentromeric repeats, where factors such as Clr3 HDAC, critical for read–write heterochromatin propagation, are not enriched (27). Pericentromeric regions contain multiple copies of dg/dh repeats and rely on persistent targeting of Clr4/Suv39h via RNAi to maintain H3K9me3 (9). Neither mcl1-4 nor mcm2-1 showed major changes in H3K9me3 at centromeres, except when combined with ago1Δ (SI Appendix, Fig. S8B).

Mcl1 Promotes FACT Association with the Replisome for Heterochromatin Propagation.

The above results prompted us to investigate how Mcl1 contributes to the retention of parental histones. A critical clue was provided by our finding that Mcl1 copurifies with both subunits of FACT (Fig. 1 E and F). Furthermore, FACT associates with the replisome in S. cerevisiae, although how this happens remains to be fully investigated (60, 65). We asked whether heterochromatin maintenance defects in mcl1-4 and mcm2-1 mutants are related to defective association of FACT with the replisome. FACT associates with subunits of MCM (Mcm2) and GINS (Psf3) when Mcl1 is present, but these interactions are lost in cells lacking Mcl1 (Fig. 6A), suggesting that Mcl1 is required for the association between FACT and the replisome. Importantly, FACT was able to associate with the replisome, albeit to a lesser extent, in cells carrying the mutation in the histone binding domain of Mcm2 (Fig. 6A).

Fig. 6. Mcl1 affects heterochromatin propagation by controlling the association of FACT with the replisome. (A) Co-IP analyses of FACT (Pob3) and MCM (Mcm2) or GINS (Psf3) in the presence or absence of Mcl1 (Left and Middle). Co-IP analysis of FACT and WT or mutated Mcm2 (mcm2-1) (Right). (B and C) Fold enrichment of H3K9me3 at mat was determined by ChIP-seq. See also SI Appendix, Fig. S9.

To explore the functional relationship between replisome components and FACT in heterochromatin propagation, we performed ChIP-seq analysis of H3K9me3 distribution in single and double mutants. Compared to WT cells, the mcl1-4 and pob3Δ single mutants showed loss of H3K9me3 at regions distal to the cenH nucleation center within the silent mat interval. Interestingly, the mcl1-4 pob3Δ double mutant did not show an additive effect on H3K9me3 levels (Fig. 6B). On the other hand, mcm2-1 pob3Δ showed a cumulative loss of H3K9me3, including at the cenH nucleation site (Fig. 6C), suggesting that Mcm2 and FACT might have additional roles beyond their shared function. Moreover, the single and double mutants showed no major changes in H3K9me3 at pericentromeric repeats where RNAi machinery targets Clr4/Suv39h to maintain H3K9me3 (SI Appendix, Fig. S9 A and B).

Collectively, our work reveals that replisome components such as Mcl1 and Mcm2 are preferentially enriched across specialized replication zones at heterochromatic regions and help retain parental methylated histones to support heterochromatin propagation via a mechanism involving the histone chaperone FACT.

Discussion

Stable heterochromatin propagation preserves cell type-specific gene expression patterns and silences repetitive DNA elements. We showed previously that heterochromatin is inherited via the read–write mechanism in a self-templating manner, whereby H3K9me3 locally concentrates Clr4/Suv39h on chromatin to further deposit H3K9me and promote heterochromatin propagation (10, 25). Understanding how methylated histones are transferred to daughter DNA strands during replication is of fundamental importance. Here, we describe the involvement of replication machinery in the retention of parental histones, which transmit epigenetic memory to support heterochromatin propagation (Fig. 7).

Fig. 7. Model showing specialized replication of the silent mat region and its role in heterochromatin propagation. (Top) Anchoring of mat at the nuclear periphery via RIXC and the boundary elements spatially constrains mat at a peripheral subdomain enriched for the heterochromatin protein Swi6/HP1. During S phase, Swi6/HP1 promotes the efficient activation of origins, leading to the enrichment of replisome components, which together with FACT help retain parental histones during DNA replication. The ability of Swi6/HP1 to engage in low-affinity interactions (straight dotted lines) and to form dimers (curved dotted lines) creates a high-concentration microenvironment that might help retain FACT and released parental histones within the vicinity of the replication fork to augment histone recycling during replication. (Bottom) Replication-coupled histone transfer in WT, mcl1 and mcm2 mutant cells. In WT cells, FACT and replisome components including Mcm2 facilitate the transfer of parental histones to daughter strands. Full maturation of chromatin behind the replication fork is achieved through the “read–write” activity of Clr4/Suv39h (Top). Loss of Mcl1 disrupts replisome and the transfer of parental histones by FACT, leading to defects in heterochromatin propagation (Middle). FACT can still associate with the replisome when the HBD of Mcm2 is mutated. This allows the transfer of sufficient parental histones to support the read–write activity of Clr4/Suv39h, which compensates for the loss of modified histones during DNA replication and allows epigenetic inheritance of heterochromatin (bottom).

We find that in S. pombe, major heterochromatin domains are replicated as specialized zones. These regions are heavily coated with replication factors, such as ORC and MCM, distinguishing them from the rest of the genome. Moreover, we show that Swi6/HP1 coordinates replication throughout heterochromatin domains such as the silent mat region and pericentromeric repeats, in addition to its previously described role in early replication by recruiting DDK (55). Cells lacking Swi6/HP1 exhibit defects in replication initiation specifically within heterochromatin. In swi6Δ cells, Mcm2 protein is observed in multiple peaks across heterochromatin domains, corresponding to ORC localization during G1, likely representing inactive but licensed replication origins. These observations, along with evidence showing that features predictive of S. pombe origins are enriched at the mat locus, centromeres, and subtelomeric regions (66), further support the presence of clustered origins within heterochromatin domains. The clustering of replication origins, typically found in efficiently replicated parts of the genome (67), together with the activating role of Swi6/HP1, may help overcome challenges in replicating these repressive domains. This specialized replication environment leads to a high local concentration of replisome components, which overcomes yet another challenge-the need to retain parental histones to maintain heterochromatin.

Disassembly of nucleosomes and rapid reassembly at the back of the replication fork preserves chromatin structure (68–70) and requires the coordinated action of factors including replication proteins (34, 35, 60, 61, 63, 71). Replication proteins localized across heterochromatin facilitate the retention of parental histones and likely other factors. As in other systems, S. pombe Mcm2 possesses histone chaperone activity, and we find that the mcm2 HBD mutant cells, that are defective in binding histones, fail to propagate an artificially assembled heterochromatin domain at an ectopic site. However, the same mutant presents only modest defects in the retention of parental histones and the self-propagation of endogenous heterochromatin. Other replication factors localized in high abundance at endogenous heterochromatic loci may act in an overlapping manner with Mcm2 to mediate histone transfer. Supporting this, mutations in replication components, such as Mrc1, Polα, and the fork protection complex, impair heterochromatin propagation (39, 61, 71, this study). Importantly, besides chaperoning histones, replication factors engage other factors involved in retention of parental histones. In this regard, we show that Mcl1 is required for connecting the replisome to the histone chaperone FACT, implicated in heterochromatin propagation (30, 53). By contrast, mutation of the Mcm2 HBD compromises but does not completely disrupt FACT association with the replisome. This differential effect correlates with the heterochromatin propagation defects, such that mutations in Mcl1 severely affect the retention of parental histones in cycling cells and result in defective heterochromatin propagation at both ectopic and endogenous loci.

The ortholog of Mcl1, Ctf4, exists as a homotrimeric complex and connects multiple factors to the replication fork, such as the Cdc45-MCM-GINS (CMG) to polymerases, like Polα (43, 44, 46). Mcl1/Ctf4 might provide a platform for directly engaging FACT with the replisome or may simply maintain the structural integrity of the replisome to ensure effective cooperation with FACT. In either case, FACT likely captures histones released during replication and redeposits them onto daughter strands to preserve H3K9me3 (51, 54, 72, 73). This transfer is possibly guided by the histone chaperone function of replication factors and may require other factors like the SMARCAD1 homolog Fft3 that also associates with replication machinery (40). Importantly, both Fft3 and FACT facilitate the retention of parental histones, and their loss causes similar defects in the epigenetic propagation of heterochromatin domains (30, 40, 53).

We find that boundary elements, specifically those surrounding the silent mat domain, are crucial for loading FACT to maintain heterochromatin (Fig. 7), while not essential for DNA replication per se. This boundary function in loading FACT and heterochromatin propagation is tightly linked to the recruitment of RIXC. Among other functions, RIXC positions the mat domain at the nuclear periphery (30), which helps form high-concentration Swi6/HP1 territories. This process involving nonstoichiometric multivalent HP1 interactions is thought to create a microenvironment for efficient recruitment of effectors (1). To this end, DDK targeted by Swi6/HP1 might allow effective coupling of replisome abundance with FACT engagement for the retention of parental histones. We envision that high-concentration Swi6/HP1 territories facilitate the local retention of FACT and its associated parental histones and their transfer to daughter strands, promoting the self-propagation of heterochromatin (Fig. 7).

Specialized replication for heterochromatin propagation is likely conserved in other eukaryotes. In S. cerevisiae, silencer elements, which possess replication origin activity, are bound by ORC, which ensures silencing of mat loci by directly recruiting Sir2 (74–76). Similarly, in S. pombe, ORC copurifies with Swi6/HP1 (77), suggesting a more direct role for high levels of pre-RC for the propagation of heterochromatin domains. Regardless, in addition to directly recruiting silencing factors, ORC might help establish a high local concentration of replisome components required for retaining parental histones during replication in yeast and higher eukaryotes. In human cells, the ORC-associated protein ORCA stabilizes ORC binding on chromatin (78). Interestingly, ORCA binds to H3K9me3 that coats heterochromatin domains (79). ORCA may stabilize ORC binding to promote efficient replication of heterochromatin regions. This may enrich the replisome components, ultimately facilitating efficient DNA and chromatin replication. All of the above observations support that targeting replication factors in high concentrations across heterochromatin domains has evolved as a general strategy across species, aiding the stable inheritance of heterochromatin domains.

Materials and Methods

Strains are listed in SI Appendix, Table S1. Standard techniques were used to culture S. pombe. All experiments were performed in yeast extract–rich medium supplemented with adenine (YEA) at 30 °C unless indicated otherwise. For BrdU chromatin immunoprecipitation experiments, Edinburgh’s Minimal Medium (EMM) was used. Strains were constructed by genetic crosses, followed by tetrad dissection or random spore analysis, and by CRISPR (SI Appendix). Standard methods for cell cycle synchronization were used (SI Appendix). The Mcl1-GFP purification and mass spectrometry were performed following standard procedures (30) (SI Appendix), and the Mcl1 interactors are listed in SI Appendix, Table S2. The silencing assay at the ectopic site was performed as previously described (27) (SI Appendix). Histone turnover and histone retention assays were previously described (40) (SI Appendix). A description of random and targeted genetic screens, co-IP analysis, ChIP-chip, ChIP-seq, qPCR, RT-qPCR, FACS, and western blot procedures can be found in SI Appendix. Primers used in this study are listed in SI Appendix, Table S3.

Supplementary Material

Appendix 01 (PDF)

Click here for additional data file.

We thank members of the Grewal lab for stimulating discussions, Amar Klar and Genevieve Thon for providing strains, and Jemima Barrowman for editing the manuscript. We also thank Diego Folco for comments. This study used the Helix Systems and Biowulf Linux cluster at the NIH. This work was supported by the Intramural Research Program of the NIH, National Cancer Institute.

Author contributions

P.N., H.X., M.Z., S.H., D.W., and S.I.S.G. designed research; P.N., J.D., H.X., M.Z., S.H., M.O., and T.A. performed research; P.N., J.D., H.X., M.Z., and S.H. contributed new reagents/analytic tools; P.N., J.D., H.X., M.Z., S.H., M.O., T.A., D.W., and S.I.S.G. analyzed data; and P.N. and S.I.S.G. wrote the paper.

Competing interests

The authors declare no competing interest.

Data, Materials, and Software Availability

ChIP-seq and microarray data are deposited in the Gene Expression Omnibus (GEO) under accession numbers GSE242353 (80), GSE242429 (81), and GSE242430 (82). All other data are included in the manuscript and/or SI Appendix.

Supporting Information

This article is a PNAS Direct Submission.
==== Refs
1 S. I. S. Grewal, The molecular basis of heterochromatin assembly and epigenetic inheritance. Mol. Cell 83 , 1767–1785 (2023).37207657
2 K. Noma, C. D. Allis, S. I. Grewal, Transitions in distinct histone H3 methylation patterns at the heterochromatin domain boundaries. Science 293 , 1150–1155 (2001).11498594
3 M. D. Litt, M. Simpson, M. Gaszner, C. D. Allis, G. Felsenfeld, Correlation between histone lysine methylation and developmental changes at the chicken beta-globin locus. Science 293 , 2453–2455 (2001).11498546
4 T. Kouzarides, Chromatin modifications and their function. Cell 128 , 693–705 (2007).17320507
5 H. P. Cam , Comprehensive analysis of heterochromatin-and RNAi-mediated epigenetic control of the fission yeast genome. Nat. Genet. 37 , 809–819 (2005).15976807
6 E. H. Bayne , Stc1: A critical link between RNAi and chromatin modification required for heterochromatin integrity. Cell 140 , 666–677 (2010).20211136
7 I. M. Hall , Establishment and maintenance of a heterochromatin domain. Science 297 , 2232–2237 (2002).12215653
8 Y. Shimada, F. Mohn, M. Buhler, The RNA-induced transcriptional silencing complex targets chromatin exclusively via interacting with nascent transcripts. Genes Dev. 30 , 2571–2580 (2016).27941123
9 T. A. Volpe , Regulation of heterochromatic silencing and histone H3 lysine-9 methylation by RNAi. Science 297 , 1833–1837 (2002).12193640
10 K. Zhang, K. Mosch, W. Fischle, S. I. Grewal, Roles of the Clr4 methyltransferase complex in nucleation, spreading and maintenance of heterochromatin. Nat. Struct. Mol. Biol. 15 , 381–388 (2008).18345014
11 J. Nakayama, J. C. Rice, B. D. Strahl, C. D. Allis, S. I. Grewal, Role of histone H3 lysine 9 methylation in epigenetic control of heterochromatin assembly. Science 292 , 110–113 (2001).11283354
12 S. Rea , Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406 , 593–599 (2000).10949293
13 P. J. Horn, J. N. Bastie, C. L. Peterson, A Rik1-associated, cullin-dependent E3 ubiquitin ligase is essential for heterochromatin formation. Genes Dev. 19 , 1705–1714 (2005).16024659
14 B. Al-Sady, H. D. Madhani, G. J. Narlikar, Division of labor between the chromodomains of HP1 and Suv39 methylase enables coordination of heterochromatin spread. Mol. Cell 51 , 80–91 (2013).23849629
15 S. Jia, K. Noma, S. I. Grewal, RNAi-independent heterochromatin nucleation by the stress-activated ATF/CREB family proteins. Science 304 , 1971–1976 (2004).15218150
16 J. F. Nickels , The transcription factor Atf1 lowers the transition barrier for nucleosome-mediated establishment of heterochromatin. Cell Rep. 39 , 110828 (2022).35584672
17 A. J. Bannister , Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410 , 120–124 (2001).11242054
18 M. Lachner, D. O’Carroll, S. Rea, K. Mechtler, T. Jenuwein, Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410 , 116–120 (2001).11242053
19 M. Sadaie, T. Iida, T. Urano, J. Nakayama, A chromodomain protein, Chp1, is required for the establishment of heterochromatin in fission yeast. EMBO J. 23 , 3825–3835 (2004).15372076
20 A. G. Larson , Liquid droplet formation by HP1alpha suggests a role for phase separation in heterochromatin. Nature 547 , 236–240 (2017).28636604
21 A. R. Strom , Phase separation drives heterochromatin domain formation. Nature 547 , 241–245 (2017).28636597
22 T. Sugiyama , SHREC, an effector complex for heterochromatic transcriptional silencing. Cell 128 , 491–504 (2007).17289569
23 G. Job , SHREC silences heterochromatin via distinct remodeling and deacetylation modules. Mol. Cell 62 , 207–221 (2016).27105116
24 S. I. Grewal, A. J. Klar, Chromosomal inheritance of epigenetic states in fission yeast during mitosis and meiosis. Cell 86 , 95–101 (1996).8689692
25 A. R. Cutter DiPiazza , Spreading and epigenetic inheritance of heterochromatin require a critical density of histone H3 lysine 9 tri-methylation. Proc. Natl. Acad. Sci. U.S.A. 118 , e2100699118 (2021).34035174
26 O. Aygun, S. Mehta, S. I. Grewal, HDAC-mediated suppression of histone turnover promotes epigenetic stability of heterochromatin. Nat. Struct. Mol. Biol. 20 , 547–554 (2013).23604080
27 M. Zofall, R. Sandhu, S. Holla, D. Wheeler, S. I. S. Grewal, Histone deacetylation primes self-propagation of heterochromatin domains to promote epigenetic inheritance. Nat. Struct. Mol. Biol. 29 , 898–909 (2022).36064597
28 P. N. Audergon , Epigenetics. Restricted epigenetic inheritance of H3K9 methylation. Science 348 , 132–135 (2015).25838386
29 K. Ragunathan, G. Jih, D. Moazed, Epigenetics., Epigenetic inheritance uncoupled from sequence-specific recruitment. Science 348 , 1258699 (2015).25831549
30 S. Holla , Positioning heterochromatin at the nuclear periphery suppresses histone turnover to promote epigenetic inheritance. Cell 180 , 150–164.e115 (2020).31883795
31 S. P. Bell, A. Dutta, DNA replication in eukaryotic cells. Annu. Rev. Biochem. 71 , 333–374 (2002).12045100
32 H. Nishitani, Z. Lygerou, DNA replication licensing. Front. Biosci. 9 , 2115–2132 (2004).15353274
33 A. Costa, J. F. X. Diffley, The initiation of eukaryotic DNA replication. Annu. Rev. Biochem. 91 , 107–131 (2022).35320688
34 A. Serra-Cardona, Z. Zhang, Replication-coupled nucleosome assembly in the passage of epigenetic information and cell identity. Trends Biochem. Sci. 43 , 136–148 (2018).29292063
35 K. R. Stewart-Morgan, N. Petryk, A. Groth, Chromatin replication and epigenetic cell memory. Nat. Cell Biol. 22 , 361–371 (2020).32231312
36 J. Nakayama, A. J. Klar, S. I. Grewal, A chromodomain protein, Swi6, performs imprinting functions in fission yeast during mitosis and meiosis. Cell 101 , 307–317 (2000).10847685
37 G. Thon, A. Cohen, A. J. Klar, Three additional linkage groups that repress transcription and meiotic recombination in the mating-type region of Schizosaccharomyces pombe. Genetics 138 , 29–38 (1994).8001791
38 K. Ekwall, T. Ruusala, Mutations in rik1, clr2, clr3 and clr4 genes asymmetrically derepress the silent mating-type loci in fission yeast. Genetics 136 , 53–64 (1994).8138176
39 L. J. Jahn , Dependency of heterochromatin domains on replication factors. G3 (Bethesda) 8 , 477–489 (2018).29187422
40 N. Taneja , SNF2 family protein Fft3 suppresses nucleosome turnover to promote epigenetic inheritance and proper replication. Mol. Cell 66 , 50–62.e56 (2017).28318821
41 A. Stralfors, J. Walfridsson, H. Bhuiyan, K. Ekwall, The FUN30 chromatin remodeler, Fft3, protects centromeric and subtelomeric domains from euchromatin formation. PLoS Genet. 7 , e1001334 (2011).21437270
42 A. A. Alcasabas , Mrc1 transduces signals of DNA replication stress to activate Rad53. Nat. Cell Biol. 3 , 958–965 (2001).11715016
43 A. Gambus , A key role for Ctf4 in coupling the MCM2-7 helicase to DNA polymerase alpha within the eukaryotic replisome. EMBO J. 28 , 2992–3004 (2009).19661920
44 A. C. Simon , A Ctf4 trimer couples the CMG helicase to DNA polymerase alpha in the eukaryotic replisome. Nature 510 , 293–297 (2014).24805245
45 K. Tanaka, P. Russell, Mrc1 channels the DNA replication arrest signal to checkpoint kinase Cds1. Nat. Cell Biol. 3 , 966–972 (2001).11715017
46 F. Villa , Ctf4 Is a hub in the eukaryotic replisome that links multiple CIP-box proteins to the CMG helicase. Mol. Cell 63 , 385–396 (2016).27397685
47 H. D. Folco, A. McCue, V. Balachandran, S. I. S. Grewal, Cohesin impedes heterochromatin assembly in fission yeast cells lacking Pds5. Genetics 213 , 127–141 (2019).31278118
48 T. Natsume , A DNA polymerase alpha accessory protein, Mcl1, is required for propagation of centromere structures in fission yeast. PLoS ONE 3 , e2221 (2008).18493607
49 G. Shipkovenska, A. Durango, M. Kalocsay, S. P. Gygi, D. Moazed, A conserved RNA degradation complex required for spreading and epigenetic inheritance of heterochromatin. eLife 9 , e54341 (2020).32491985
50 Y. M. Mamnun, S. Katayama, T. Toda, Fission yeast Mcl1 interacts with SCF(Pof3) and is required for centromere formation. Biochem. Biophys. Res. Commun. 350 , 125–130 (2006).16997270
51 T. Formosa, The role of FACT in making and breaking nucleosomes. Biochim. Biophys. Acta. 1819 , 247–255 (2013).24459727
52 E. Lejeune , The chromatin-remodeling factor FACT contributes to centromeric heterochromatin independently of RNAi. Curr. Biol. 17 , 1219–1224 (2007).17614284
53 M. Murawska , The histone chaperone FACT facilitates heterochromatin spreading by regulating histone turnover and H3K9 methylation states. Cell Rep. 37 , 109944 (2021).34731638
54 J. Yang , The histone chaperone FACT contributes to DNA replication-coupled nucleosome assembly. Cell Rep. 14 , 1128–1141 (2016).26804921
55 M. T. Hayashi, T. S. Takahashi, T. Nakagawa, J. Nakayama, H. Masukata, The heterochromatin protein Swi6/HP1 activates replication origins at the pericentromeric region and silent mating-type locus. Nat. Cell Biol. 11 , 357–362 (2009).19182789
56 S. J. Charlton, M. M. Jorgensen, G. Thon, Integrity of a heterochromatic domain ensured by its boundary elements. Proc. Natl. Acad. Sci. U.S. A. 117 , 21504–21511 (2020).32817556
57 K. Noma, H. P. Cam, R. J. Maraia, S. I. Grewal, A role for TFIIIC transcription factor complex in genome organization. Cell 125 , 859–872 (2006).16751097
58 G. Thon, P. Bjerling, C. M. Bunner, J. Verhein-Hansen, Expression-state boundaries in the mating-type region of fission yeast. Genetics 161 , 611–622 (2002).12072458
59 T. Toteva , Establishment of expression-state boundaries by Rif1 and Taz1 in fission yeast. Proc. Natl. Acad. Sci. U.S.A. 114 , 1093–1098 (2017).28096402
60 M. Foltman , Eukaryotic replisome components cooperate to process histones during chromosome replication. Cell Rep. 3 , 892–904 (2013).23499444
61 H. Gan , The Mcm2-Ctf4-Polalpha axis facilitates parental histone H3–H4 transfer to lagging strands. Mol. Cell 72 , 140–151.e143 (2018).30244834
62 H. Huang , A unique binding mode enables MCM2 to chaperone histones H3–H4 at replication forks. Nat. Struct. Mol. Biol. 22 , 618–626 (2015).26167883
63 N. Petryk , MCM2 promotes symmetric inheritance of modified histones during DNA replication. Science 361 , 1389–1392 (2018).30115746
64 J. S. Iacovoni, P. Russell, F. Gaits, A new inducible protein expression system in fission yeast based on the glucose-repressed inv1 promoter. Gene 232 , 53–58 (1999).10333521
65 A. Gambus , GINS maintains association of Cdc45 with MCM in replisome progression complexes at eukaryotic DNA replication forks. Nat. Cell Biol. 8 , 358–366 (2006).16531994
66 M. Segurado, A. de Luis, F. Antequera, Genome-wide distribution of DNA replication origins at A+T-rich islands in Schizosaccharomyces pombe EMBO Rep. 4 , 1048–1053 (2003).14566325
67 A. Kaykov, P. Nurse, The spatial and temporal organization of origin firing during the S-phase of fission yeast. Genome Res. 25 , 391–401 (2015).25650245
68 T. M. Escobar , Active and repressed chromatin domains exhibit distinct nucleosome segregation during DNA replication. Cell 179 , 953–963.e911 (2019).31675501
69 G. Schlissel, J. Rine, The nucleosome core particle remembers its position through DNA replication and RNA transcription. Proc. Natl. Acad. Sci. U.S.A. 116 , 20605–20611 (2019).31511420
70 D. J. Smith, I. Whitehouse, Intrinsic coupling of lagging-strand synthesis to chromatin assembly. Nature 483 , 434–438 (2012).22419157
71 J. Nakayama, R. C. Allshire, A. J. Klar, S. I. Grewal, A role for DNA polymerase alpha in epigenetic control of transcriptional silencing in fission yeast. EMBO J. 20 , 2857–2866 (2001).11387218
72 H. Ehara, T. Kujirai, M. Shirouzu, H. Kurumizaka, S. I. Sekine, Structural basis of nucleosome disassembly and reassembly by RNAPII elongation complex with FACT. Science 377 , eabp9466 (2022).35981082
73 Y. Liu , FACT caught in the act of manipulating the nucleosome. Nature 577 , 426–431 (2020).31775157
74 S. P. Bell, J. Mitchell, J. Leber, R. Kobayashi, B. Stillman, The multidomain structure of Orc1p reveals similarity to regulators of DNA replication and transcriptional silencing. Cell 83 , 563–568 (1995).7585959
75 C. A. Fox, A. E. Ehrenhofer-Murray, S. Loo, J. Rine, The origin recognition complex, SIR1, and the S phase requirement for silencing. Science 276 , 1547–1551 (1997).9171055
76 P. Laurenson, J. Rine, Silencers, silencing, and heritable transcriptional states. Microbiol. Rev. 56 , 543–560 (1992).1480108
77 T. Fischer , Diverse roles of HP1 proteins in heterochromatin assembly and functions in fission yeast. Proc. Natl. Acad. Sci. U.S.A. 106 , 8998–9003 (2009).19443688
78 Z. Shen , A WD-repeat protein stabilizes ORC binding to chromatin. Mol. Cell 40 , 99–111 (2010).20932478
79 K. M. Chan, Z. Zhang, Leucine-rich repeat and WD repeat-containing protein 1 is recruited to pericentric heterochromatin by trimethylated lysine 9 of histone H3 and maintains heterochromatin silencing. J. Biol. Chem. 287 , 15024–15033 (2012).22427655
80 P. Nathanailidou , Specialized replication of heterochromatin domains ensures self-templated chromatin assembly and epigenetic inheritance. Gene Expression Omnibus (GEO). https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE242353. Deposited 6 September 2023.
81 P. Nathanailidou , Specialized replication of heterochromatin domains ensures self-templated chromatin assembly and epigenetic inheritance. Gene Expression Omnibus (GEO). https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE242429. Deposited 6 September 2023.
82 P. Nathanailidou , Specialized replication of heterochromatin domains ensures self-templated chromatin assembly and epigenetic inheritance. Gene Expression Omnibus (GEO). https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE242430. Deposited 6 September 2023.
