A role for CENP-A/Cse4 phosphorylation on serine 33 in deposition at the centromere

A role for CENP-A/Cse4 phosphorylation on serine 33 in deposition at the centromere Abstract Centromeres are the sites of assembly of the kinetochore, which connect the chromatids to the microtubules for sister chromatid segregation during cell division. Centromeres are characterized by the presence of the histone H3 variant CENP-A (termed Cse4 in Saccharomyces cerevisiae). Here, we investigated the function of serine 33 phosphorylation of Cse4 (Cse4-S33ph) in S. cerevisiae, which lies within the essential N-terminal domain (END) of the extended Cse4 N-terminus. Significantly, we identified histone H4-K5, 8, 12R to cause a temperature-sensitive growth defect with mutations in Cse4-S33 and sensitivity to nocodazole and hydroxyurea. Furthermore, the absence of Cse4-S33ph reduced the levels of Cse4 at centromeric sequences, suggesting that Cse4 deposition is defective in the absence of S33 phosphorylation. We furthermore identified synthetic genetic interactions with histone H2A-E57A and H2A-L66A, which both cause a reduced interaction with the histone chaperone FACT and reduced H2A/H2B levels in chromatin, again supporting the notion that a combined defect of H2A/H2B and Cse4 deposition causes centromeric defects. Altogether, our data highlight the importance of correct histone deposition in building a functional centromeric nucleosome and suggests a role for Cse4-S33ph in this process. Cse4, Hat1, Ser33, centromere, kinetochore INTRODUCTION Centromeres are defined regions of the chromosomes where the kinetochores are assembled and provide a physical attachment between centromeric chromatin and the microtubules for chromosome segregation in the metaphase of the cell cycle (reviewed in Pesenti, Weir and Musacchio 2016). The centromeres are distinguished from other chromosomal regions by the presence of a variant version of histone H3, CENP-A (termed Cse4 in Saccharomyces cerevisiae), which determines the site of recruitment of kinetochore proteins to the chromosome (Earnshaw and Rothfield 1985; Palmer et al.1987; Stoler et al.1995; Meluh et al.1998). In higher eukaryotes, CENP-A-carrying nucleosomes are interspersed with nucleosomes containing canonical H3 (McKinley and Cheeseman 2016). In contrast, S. cerevisiae has a point centromere consisting of what is most likely a single centromeric nucleosome that contains Cse4, although there has been controversy surrounding the precise nature of this particle (Black and Cleveland 2011). One difference between Cse4 and its CENP-A homologs is that it contains a large, 130-amino acid N-terminal domain (Stoler et al.1995). Part of this domain is essential for centromere function, because deletion of the END region (essential N-terminal domain, amino acids 28 - 60) is lethal (Chen et al.2000). This domain recruits inner kinetochore complexes to the centromere (Ortiz et al.1999) and possibly serves to compensate for the fact that S. cerevisiae has only one rather than multiple CENP-A/Cse4 nucleosomes at the centromere, as is the case in higher eukaryotes. CENP-A/H4 and Cse4/H4 are deposited at the centromeric sequences by histone chaperones that are specialized for the purpose, namely HJURP (Dunleavy et al.2009; Foltz et al.2009; Shuaib et al.2010) in higher eukaryotes and Scm3 in yeast (Camahort et al.2007; Stoler et al.2007). They distinguish the centromeric H3 variant from canonical H3 via the so-called CENP-A targeting domain (CATD), which lies in the core region of CENP-A/Cse4 (Black et al.2004) and makes direct contacts to HJURP/Scm3 (Cho and Harrison 2011; Zhou et al.2011). Interestingly, posttranslational modifications (PTMs) on CENP-A and H4 regulate their interaction with HJURP and thus deposition in chromatin. CENP-A is phosphorylated by the kinase Cdk1 on serine 68 (S68), which lies close to the CATD. This phosphorylation reduces CENP-A binding to HJURP and thus prevents premature loading of CENP-A at the centromeres (Yu et al.2015; Fachinetti et al.2017; Wang et al.2017). Furthermore, H4 K5 and K12 are acetylated in the context of CENP-A/H4 by the RbAp46/48-Hat1 complex before they are incorporated into centromeric chromatin, and CENP-A deposition is reduced in cells deficient for RbAp48 (Shang et al.2016). A conserved role for H4 acetylation in CENP-A deposition is further underscored by the finding that the CENP-A homolog in Drosophila (Cid) is present in a complex with Hat1 and the chromatin assembly factor Caf1, and Hat1 is required for proper Cid loading into centromeric chromatin (Boltengagen et al.2016). It has been proposed that the absence of H4 acetylation reduces the interaction of the CENP-A/H4 dimer with the histone chaperone HJURP, which thus restricts CENP-A deposition at the centromere. CENP-A-containing nucleosomes show specific structural properties compared to canonical nucleosomes that are necessary to mediate the binding of inner kinetochore proteins, for instance CENP-C/Mif2 (Carroll, Milks and Straight 2010; Falk et al.2016), which then recruits other kinetochore sub-complexes. The most chromatin-proximal complexes in S. cerevisiae are the CBF3 complex and the CTF19 complex, the yeast equivalent of the constitutive centromere-associated network (CCAN) in higher eukaryotes. These complexes interact with the KMN (Knl1 complex, Mis12 complex and Ndc80 complex) network, within which the NDC80 complex provides a direct contact with the microtubule (reviewed in Cheeseman 2014). Proper kinetochore—microtubule attachment is monitored by the chromosomal passenger complex (CPC), which consists of the Ipl1/Aurora B kinase, Sli5/INCENP, Bir1/Survivin and Nbl1/Borealin (Biggins and Murray 2001; Tanaka et al.2002; Sandall et al.2006; Ruchaud, Carmena and Earnshaw 2007; Nakajima et al.2009). The recruitment of the CPC to the kinetochore in part is regulated by an interaction between Bir1 and shugoshin (Sgo1) (Kawashima et al.2007). Thus, a complex hierarchy of multi-protein complexes is assembled at centromeric sequences to link chromatin to the microtubule and to pull sister chromatids apart. Phosphorylation events play a major role in regulating interactions among kinetochore proteins as well as the interaction between the microtubule and the kinetochore (reviewed in Funabiki and Wynne 2013). Aurora B phosphorylates multiple targets in the KMN network and thus regulates interactions to the microtubule in response to changes in tension and attachment state (Welburn et al.2010). Furthermore, cyclin-dependent kinase (CDK) phosphorylates CENP-T, which regulates its interaction with the NDC80 complex as well as the KMN network (Gascoigne et al.2011; Malvezzi et al.2013; Huis In’t Veld et al.2016). Other phosphorylation events control the activation of checkpoints, most prominently the spindle assembly checkpoint (SAC). As an example, the kinase Mps1 phosphorylates the Knl1 protein, which regulates the recruitment of the Bub1-Bub3 complex (Yamagishi et al.2012). Mps1 subsequently phosphorylates sites within Bub1, which promotes binding of Mad1 to the kinetochore to initiate the SAC (Krenn et al.2014; London and Biggins 2014; Moyle et al.2014). Next to the phosphorylation of kinetochore components, other PTMs on CENP-A, including methylation, acetylation and ubiquitination play an important role in centromere function (Fukagawa 2017). Similarly to CENP-A, S. cerevisiae Cse4 carries multiple PTMs. Ubiquitination in the C-terminal region of Cse4 by the E3 ubiquitin ligase Psh1 prevents its inappropriate incorporation at non-centromeric regions (Hewawasam et al.2010; Ranjitkar et al.2010). Furthermore, we identified methylation on arginine 37 (R37) of Cse4, which lies within the END domain, and the absence of methylation causes selective defects in combination with mutations in components of the CTF19 complex, indicating that this PTM regulates interactions of the Cse4 N-terminus with kinetochore components (Samel et al.2012). Cse4 furthermore is acetylated on K49 and phosphorylated on serines 20, 33, 40 and 105 (Boeckmann et al.2013). Simultaneous mutation of all four phosphorylation sites causes centromeric defects, but the individual contribution of each site has not been dissected. Here, we sought to determine how Cse4-S33 phosphorylation (Cse4-S33ph) regulates centromere function. Screens were performed of cse4-S33 mutations with libraries of systematic mutations in the canonical histones H2A, H2B, H3 and H4. Significantly, this revealed selected mutations in H4-K5, 8, 12R, H2A-E57A and H2A-L66A to cause synthetic phenotypes, whereas no defects were uncovered in a comprehensive analysis of mutations in genes encoding kinetochore components. Interestingly, both the H4 and the H2A residues have been implicated in the deposition of the respective histones in chromatin, suggesting an equivalent role for Cse4-S33ph in its deposition in chromatin. Accordingly, we found reduced levels of unphosphorylated Cse4 at centromeric regions. Altogether, our analysis shows an important role for PTMs on H4 and Cse4 as well as of H2A deposition in building a functional centromeric nucleosome. MATERIAL AND METHODS Yeast strains and plasmids The S. cerevisiae strains and plasmids used in this study are listed in Tables S1 and S2 (Supporting Information), respectively. Yeast was grown and manipulated according to standard procedures (Sherman 1991). Yeast was grown on full medium (YPD) and selective minimal plates (YM). Genomic integration of cse4 alleles was performed by cloning the allele on a URA3-marked integrative vector and introducing it into the strain by integrative transformation followed by loop-out on 5-fluoroorotic acid (5- FOA)-containing medium. For genetic crosses, cse4 was marked with natMX or kanMX by the integration of the respective resistance cassettes downstream of the open reading frame. Deletions of chromosomal genes were performed using the integration of knockout cassettes (Longtine et al.1998). Growth curves of yeast cultures were measured using a microplate reader (SynergyH1, BioTek). 200 μl cultures in a 96-well plate were inoculated to an optical density at 600 nm (OD600) of 0.1 in YPD. OD600 was measured in 10 min intervals with double orbital shaking. Mass spectrometric analysis of Cse4 Purification and analysis of 3xHA-tagged Cse4 from yeast cells was performed essentially as described (Samel et al.2012). Briefly, partially purified histones from cells expressing 3xHA-Cse4 were separated on 10% SDS-PAGE gels, the Cse4 band excised and digested in-gel with trypsin. The peptide mixture were analyzed by online nano-flow liquid chromatography tandem mass spectrometry using an EASY-nLC™ system (ProxeonBiosystems, Odense, Denmark) connected to the LTQ-OrbitrapVelos (Thermo Electron, Bremen, Germany) through a nano-electrospray ion source (see Materials and Methods for details, Supporting Information). Synthetic lethal screens with libraries of mutations in histone genes A library of mutations in H3 and H4 ((Dai et al.2008), obtained from ThermoFisher, catalog number YSC5106) was screened for synthetic growth defects with cse4-S33A (AEY6071) or -S33E (AEY6070) and, as a control, with a CSE4 strain (AEY6067) using a Rotor HDA spotting robot (Singer Instruments). Diploids were selected, sporulated, and haploids of the desired genotype were selected in three consecutive rounds of selection. The first round selected for lyp1Δ can1Δ::MFA1pr-his5 hht1-hhf2Δ::NatMX haploids, the second round additionally applied selection for the cse4::kanMX alleles, and the third round selected for hht2–hhf2::[HHTS-HHFS]*-URA3 in addition to the previously selected markers. Selection from the last and the penultimate selection were compared for growth differences. The screening procedure was repeated three independent times for each cse4 strain. Mutations with a potential effect were re-arrayed and retested. For the search of mutations in H2A and H2B with a synthetic growth defect with cse4-S33 alleles, the strains AEY5984 (CSE4), AEY5988 (cse4-S33E) and AEY6105 (cse4-S33A) were crossed to the mutant collection (Nakanishi et al.2009). After diploid selection and sporulation, haploids were selected in a first round for lyp1Δ, pHIS3-HTA1*-flag-HTB1**hta1-htb1Δ::LEU2 and hta2-htb2Δ::URAMX, in a second round the same selection was applied as in the first round, and the third round added a selection for the cse4 allele. Candidate strains for a synthetic growth defect were chosen, re-arrayed and retested as above. Chromatin immunoprecipitation Chromatin immunoprecipitation (ChIP) and quantitative real-time PCR were performed essentially as previously described (Weber, Irlbacher and Ehrenhofer-Murray 2008) using strains AEY6242, AEY6244 and AEY6280. Cells were grown at 23°C to an OD600 of approx. 0.5 and shifted to 34°C for 3 h prior to formaldehyde crosslinking, cell harvest and chromatin isolation. Primer sequences are available upon request. RESULTS Phosphorylation on serine 33 of Cse4 has a functionally distinct role from Cse4-R37 methylation The centromeric histone H3 variant Cse4 from S. cerevisiae has previously been reported to be phosphorylated on serine 33 (Boeckmann et al.2013), a modification site that we also identified in four independent purifications of Cse4 from yeast cells (Figs S1 and S2, Supporting Information). However, the precise function of this modification has not been investigated. Interestingly, this site lies within the END domain (AA 28–60), close to the methylated residue R37 (Samel et al.2012). Our quantification showed that S33 phosphorylation is less abundant (approx. 2%) than mono-methylation of R37 (approx. 16%, Fig. S1, Supporting Information). To investigate the function of Cse4-S33ph, we created mutant versions of Cse4 in which S33 was replaced by alanine (cse4-S33A) or glutamate (cse4-S33E) to imitate the unphosphorylated or constantly phosphorylated form of Cse4, respectively. Strains carrying these Cse4 versions alone were viable and showed no obvious growth defects (Fig. 1), indicating that this phosphorylation is not essential for Cse4 function in cell viability. Figure 1. View largeDownload slide Mutation of the phosphorylation site S33 of Cse4 caused synthetic growth defects with H4-K5, 8, 12R. Serial dilutions of strains carrying H4 alleles with lysine 5 and 8 or 5 and 12 mutated to arginine and CSE4 (top), cse4-S33E (middle) or -S33A (bottom) were spotted on full medium and grown at the indicated temperatures for 3 days. Figure 1. View largeDownload slide Mutation of the phosphorylation site S33 of Cse4 caused synthetic growth defects with H4-K5, 8, 12R. Serial dilutions of strains carrying H4 alleles with lysine 5 and 8 or 5 and 12 mutated to arginine and CSE4 (top), cse4-S33E (middle) or -S33A (bottom) were spotted on full medium and grown at the indicated temperatures for 3 days. Since S33 is located in the proximity of the methylated R37 residue of Cse4, we asked whether S33 mutations caused similar phenotypes as abrogation of Cse4-R37 methylation (cse4-R37A). Unlike cse4-R37A, however, neither cse4-S33A nor -S33E caused lethality or synthetic growth defects with mutations in the CDEI-binding protein Cbf1 (Cai and Davis 1990) or in Ctf19 complex components (Table S3, Supporting Information). There also were no additional growth defects in combination with mutations in CBF3, NDC80, Knl1 or Ctf3 complex components, and no synthetic growth defects were found when combining cse4-S33 mutations with the mutation of Cse4-R37 (Table S3, Supporting Information). This suggests that Cse4-S33ph has a function at the centromere that is distinct from that of R37 methylation, although the two sites are in close proximity within the END domain of Cse4. Absence of Cse4-S33 phosphorylation causes a synthetic defect in combination with the mutation K5, 8, 12R in histone H4 To obtain insight into the function of Cse4-S33ph, we sought to identify mutations in other genes that cause a synthetic growth defect with cse4-S33 mutations. No reproducible candidates were recovered from a screen with a library of gene deletions (Tong et al.2001) (data not shown), indicating that cse4-S33 mutations have no synthetic interactions with non-essential genes. We next asked whether synthetic interactions with mutations in histone genes could be identified. For this purpose, a genetic screen was conducted in which cse4-S33A or -S33E was combined with a library of 486 mutations in histone H3 and H4. In this collection, each histone amino acid is systematically replaced by alanine; natural alanines are replaced by serine; and the collection contains several other unique mutations, for instance in histone residues that are known to carry PTMs and where the mutation mimics the (un-) modified state (see Materials and Methods, Dai et al.2008). Interestingly, the screen revealed a defect for combinations of cse4-S33 mutations with H4-K8A and -K12Q, which both represent sites of lysine acetylation in H4 (Davie et al.1981; Nelson 1982). No mutations in H3 were recovered, which is notable in the light of the fact that the centromeric nucleosome may also contain some H3 (Lochmann and Ivanov 2012). To further evaluate these effects in a well-defined genetic background, plasmid shuffle experiments were conducted in strains carrying the different cse4 alleles as well as deletions of the HHT1–HHF1 and HHT2–HHF2 gene copies, which were kept viable by a URA3-marked HHT1–HHF1 plasmid. Plasmids were introduced with mutations in histone H4 acetylation sites that we had at our disposal. After the elimination of the wild-type HHT1–HHF1 plasmid by counter-selection on 5-FOA, the growth characteristics of the strains were tested. Importantly, the combination of cse4-S33A or -S33E with H4-K5, K8, K12R caused a pronounced growth defect at elevated temperatures that was not observed with the cse4-S33 or H4 mutations alone (Fig. 1). The defect was more pronounced for cse4-S33A, and it was not observed for the individual mutations H4-K5R, 8R or H4-K5R, 12R, showing that simultaneous mutation of all three residues was necessary to cause the synthetic growth defect. To further quantitate the growth differences, growth curves were generated. A mild growth defect for H4-K5, 8, 12R was observed in the CSE4 strain. Significantly, both cse4-S33E and -S33A showed a pronounced reduction in growth with H4-K5, 8, 12R, but not with wt H3 and H4 (Fig. 2), further supporting the notion that the absence of Cse4 S33 phosphorylation causes a defect in combination with mutations of the lysine residues K5, K8 and K12 of H4, which are sites that are subject to lysine acetylation. Figure 2. View largeDownload slide H4-K5, 8, 12R with cse4-S33E or -S33A showed slow growth in liquid medium. Growth curves of the H4-K5, 8, 12R strains were measured at 36°C using a microtiter dish plate reader. Arbitrary units of the optical density at 600 nm (OD600) are shown relative to cultivation time. Mean values ± SD (n = 3) are shown. *P < 0.05; **P < 0.01: ***P < 0.001 (two-sided t-test). Figure 2. View largeDownload slide H4-K5, 8, 12R with cse4-S33E or -S33A showed slow growth in liquid medium. Growth curves of the H4-K5, 8, 12R strains were measured at 36°C using a microtiter dish plate reader. Arbitrary units of the optical density at 600 nm (OD600) are shown relative to cultivation time. Mean values ± SD (n = 3) are shown. *P < 0.05; **P < 0.01: ***P < 0.001 (two-sided t-test). We also tested the effect of H4-K12Q using plasmid shuffle, but even though this allele had shown defects in the high-throughput screen of the histone mutant collection, we observed no phenotypes in the plasmid shuffle strain, which may be due to strain background differences. Also, the mutations of H4-K5A, -K12A or mutations in K16 (K16A, K16Q), which also can be acetylated, did not display a defect with cse4-S33 mutations (not shown). Thus, the synthetic growth defect of cse4-S33 mutations showed selectivity for K5, K8 and K12 and was not shared with the acetylation site K16. The discovery of a synthetic interaction of the absence of Cse4-S33ph with acetylation sites in H4 was interesting, because H4 acetylation has previously been linked to a defect in CENP-A deposition (Boltengagen et al.2016; Shang et al.2016), and likewise suggested a role for S33ph in Cse4 deposition at the centromere. cse4-S33 mutations cause a mitotic defect with H4-K5, 8, 12R We next asked whether the growth defect of cse4-S33A and -S33E with the H4 mutations reflected a defect in mitosis. Therefore, sensitivity of the strains to the microtubule-inhibiting drug nocodazole and the S-phase inhibitor hydroxy-urea (HU) was tested. In the CSE4 background, H4-K5,8,12R alone showed a mild sensitivity towards nocodazole and HU. Importantly, the sensitivity was strongly enhanced in the cse4-S33A and -S33E strains, which themselves were insensitive to the compounds (Fig. 3). This showed that the inability to phosphorylate Cse4 S33 by itself did not disturb mitotic function or replication. However, the combination of Cse4-S33 mutation and H4-K5, 8, 12R caused strong defects in the two processes, indicating that centromeric function was defective in these mutants. Figure 3. View largeDownload slide Absence of Cse4 phosphorylation caused a defect in mitosis and replication in cells carrying H4-K5, 8, 12R. Serial dilutions of the strains as in Fig. 1 were spotted on YPD containing 3 μM nocodazole (noc) or 100 mM hydroxyurea (HU) and grown at 30°C for 3 days. Figure 3. View largeDownload slide Absence of Cse4 phosphorylation caused a defect in mitosis and replication in cells carrying H4-K5, 8, 12R. Serial dilutions of the strains as in Fig. 1 were spotted on YPD containing 3 μM nocodazole (noc) or 100 mM hydroxyurea (HU) and grown at 30°C for 3 days. hat1Δ and esa1-L327S mutations does not cause a synthetic defect with cse4-S33 mutations Since the allele H4-K5,8,12R mutates three acetylated lysines of H4 and imitates the deacetylated state, we asked whether mutations in the histone acetyltransferases responsible for their acetylation likewise caused a synthetic growth defect with cse4-S33A or -S33E. Hat1 is the catalytic subunit of the Hat1-Hat2 complex and acetylates H4-K5 and K12 (Kleff et al.1995; Parthun, Widom and Gottschling 1996). Furthermore, the HAT Esa1 (e ssential S as2-related a cetyltransferase) as a component of the NuA4 HAT complex (Allard et al.1999) acetylates H4-K5 and K8 as well as K12 (to a lesser degree) and H2A (Suka et al.2001; Torres-Machorro et al.2015). However, neither hat1Δ, esa1-L327S (ESA1 is essential) nor hat1Δ esa1-L327S double mutation caused additional growth defects with cse4-S33A or -S33E (not shown). This was surprising and suggested that the defect of the cse4-S33 mutations with H4-K5, 8, 12R was not caused by the absence of acetylation at these residues. Alternatively, a yet unknown acetyltransferase may be responsible for the acetylation that is relevant in this context, since there is residual acetylation at the lysine residues in both mutants (Suka et al.2001). We also considered the possibility that a modification at these lysines other than acetylation may be relevant for the synthetic phenotype with cse4-S33 mutations. The methyltransferase Set5 methylates lysines 5, 8 and 12 of H4 (Green et al.2012), and we therefore tested for synthetic growth defects of set5Δ with cse4-S33 mutations. However, no defects were observed (data not shown), suggesting that the defects of H4-K5,8,12R were not caused by the absence of Set5-dependent lysine methylation in the H4 N-terminus. cse4-S33A caused reduced Cse4 levels at the centromere in H4-K5, 8, 12R Since the absence of H4 acetylation by Hat1 has previously been linked to reduced deposition of CENP-A (Boltengagen et al.2016; Shang et al.2016), we asked whether the H4-K5, 8, 12R mutation caused a reduction in the amount of mutant Cse4 protein present at the centromere, which might explain why the cells have a mitotic defect. Interestingly, ChIP showed mildly reduced levels of Cse4-S33A, though not in Cse4-S33E, at the centromere in H4-K5,8,12R strains (Fig. 4). This reduction in Cse4 level was not present in strains with wt H4 (data not shown). This indicated that the absence of S33 phosphorylation caused a mild defect in the deposition of Cse4 at the centromere when the H4 N-terminus was mutated. The observation that the defect is only seen in cse4-S33A agrees with the fact that cse4-S33A has a stronger growth defect than cse4-S33E with the histone mutation (Figs 1 and 2). Figure 4. View largeDownload slide The absence of Cse4-S33 phosphorylation reduced levels of Cse4 at the centromere in H4-K5,8,12R strains. ChIP analysis of 3xHA-tagged Cse4 (α-HA) was performed. Cells were grown at 34°C for 3 h prior to ChIP. Values give the enrichment at CEN4 relative to the control region POL1. Error bars give SD of three independent experiments. *P < 0.05. Figure 4. View largeDownload slide The absence of Cse4-S33 phosphorylation reduced levels of Cse4 at the centromere in H4-K5,8,12R strains. ChIP analysis of 3xHA-tagged Cse4 (α-HA) was performed. Cells were grown at 34°C for 3 h prior to ChIP. Values give the enrichment at CEN4 relative to the control region POL1. Error bars give SD of three independent experiments. *P < 0.05. Synthetic defects of cse4-S33 with the mutations H2A-E57A and -L66A We next asked whether mutations in the canonical histones H2A and H2B can be identified that cause defects with cse4-S33A or -S33E. To test this, a genetic screen was conducted to combine the cse4-S33 alleles with a library of 225 histone alleles in which the amino acids of histones H2A and H2B were systematically replaced by alanine (Nakanishi et al.2009). This revealed synthetic growth defects for H2A-E57A and -L66A with the cse4-S33 alleles. As above, we evaluated the defects in a defined strain background using plasmid shuffle. In the presence of wt CSE4, both H2A-E57A and -L66A caused a growth defect on their own as compared to strains with wt H2A (Fig. 5). Significantly, this defect was exacerbated by cse4-S33E. Interestingly, cse4-S33A did not cause a defect with H2A-L66A, and it partially suppressed the defect of H2A-E57A. Since cse4-S33E mimics the constantly phosphorylated form of Cse4, this suggests that the defect of H2A-E57A is exacerbated by Cse4-S33ph and partially suppressed by its absence (in cse4-S33A). Figure 5. View largeDownload slide Mimicking the constantly phosphorylated state of Cse4-S33 (cse4-S33E) caused synthetic growth defects with the mutations E57A and L66A of histone H2A. Strains with the indicated genotypes were serially diluted, spotted on YPD and grown at the respective temperatures for 3 days. Figure 5. View largeDownload slide Mimicking the constantly phosphorylated state of Cse4-S33 (cse4-S33E) caused synthetic growth defects with the mutations E57A and L66A of histone H2A. Strains with the indicated genotypes were serially diluted, spotted on YPD and grown at the respective temperatures for 3 days. The identification of these H2A residues as being relevant in the context of Cse4-S33 phosphorylation is interesting in light of the fact that both H2A mutations cause a reduced interaction with the H2A/H2B histone chaperone FACT (facilitates chromatin transcription) and a defect of H2A/H2B deposition (Hodges, Gloss and Wyrick 2017). This supports the interpretation that Cse4–33 phosphorylation regulates its deposition in chromatin, and that the combined effect of reduced H2A and Cse4 at the centromere impair centromere function (see Discussion). DISCUSSION Post-translational modifications on histones and histone variants are crucial for regulating their role within chromatin. Here, we characterized the function of phosphorylation of serine 33 on the centromeric H3 variant Cse4 from S. cerevisiae. Importantly, we found that mutations that abrogate Cse4-S33ph caused synthetic growth defects when combined with mutations in histone H4-K5, 8, 12R, which was accompanied by decreased levels of Cse4 at the centromere. Furthermore, we observed synthetic growth defects with H2A-E57A and -L66A, which cause reduced H2A deposition in chromatin. Altogether, our data highlight the important role of histone deposition in building a functional centromeric nucleosome and indicates that Cse4-S33ph regulates its deposition in chromatin. Our findings of centromeric defects and a reduced deposition of S33-mutated Cse4 at the centromere when the H4 tail was mutated reflects findings in chicken and human cells that H4 and K12 acetylation by the Hat1 complex are required for CENP-A/H4 deposition (Shang et al.2016), but also show some differences to the observations in higher eukaryotes. Whereas the absence of H4 acetylation alone was sufficient to reduce CENP-A incorporation at the centromere in higher eukaryotes (Boltengagen et al.2016; Shang et al.2016), we only observed a defect in conjunction with the absence of Cse4-S33ph in yeast. Thus, Cse4 deposition apparently is less sensitive to H4 acetylation defects in yeast than CENP-A in higher eukaryotes. Furthermore, we did not observe a defect of Cse4-S33 mutations together with the absence of Hat1, and the defect was only seen when not only the Hat1 targets H4 K5 and K12 were mutated, but also K8. Thus, as above, this indicates that Cse4 is less sensitive to defects in H4 acetylation than CENP-A from higher eukaryotes for deposition at centromeric sequences. Notably, while the HAT Esa1, the catalytic entity of the NuA4 HAT complex (Allard et al.1999), acetylates all three respective lysine residues in H4 (Suka et al.2001), the mutation of Esa1 did not cause a defect with the absence of Cse4-S33ph. Therefore, a yet unknown HAT or combination of HATs is responsible for the H4 acetylation in the context of Cse4/H4 deposition. A role for H4 acetylation sites in Cse4 deposition furthermore raised the question whether the respective H4 residues are acetylated on the centromeric, Cse4-containing nucleosome. However, our query of existing genome-wide data for these modifications in S. cerevisiae (Weiner et al.2015) revealed that H4 K5Ac, K8Ac and K12Ac are undetectable at the centromeric nucleosome while being present on the flanking nucleosomes (Fig. S3, Supporting Information), which is in contrast to findings in human and chicken cells, where K5Ac and K12Ac are present on centromeric sequences (Shang et al.2016). Possibly, the modifications are removed in yeast shortly after deposition by histone deacetylases. An alternative scenario for the role of H4 acetylation in Cse4 function is that H4 K5, 8, 12R causes defects in gene expression, for instance in the expression of kinetochore components, and that this indirectly affects centromere function. Our data would imply that there is a gene (or group of genes) that is unaffected in single H4 lysine mutations and only affected in the K5, 8,12R triple mutant. However, a study of transcriptional effects of combinations of H4 lysine mutations showed that the cumulative effect of H4 K5, K8 and K12 mutation has a broad additional effect on gene expression that is mild at any single gene, and gene ontology analysis does not show an effect on chromatin or kinetochore components (Weiner et al.2015). It therefore seems unlikely that the H4 mutations show synthetic defects with Cse4-S33 due to misregulation of a centromere factor. The H4 residues highlighted here have not been described in centromere function in yeast before. Several studies have investigated effects of histone mutations in centromere function, though none report effects for N-terminal H4 mutations (Hyland et al.2005; Matsubara et al.2007; Sakamoto et al.2009; Kawashima et al.2011; Yu et al.2011). An early study isolated the first temperature-sensitive H4 allele, hhf1-20 (hhf1-T82I, A89V), which is defective in chromosome segregation (Megee et al.1990). More recently, one of the histone mutation libraries used here (Dai et al.2008) was specifically screened for centromeric defects and sensitivity to low levels of the Ilp1 kinase, and several sites in the H3 and H4 core regions were identified in this context to be important for chromosome segregation and kinetochore bi-orientation (Ng et al.2013). Some of the residues lie at the entry/exit site of DNA on the nucleosome, and it was proposed that they affect the structural properties of the centromeric nucleosome. Furthermore, H4 R36 was found to be important for the interaction of Cse4 with the E3 ubiquitin ligase Psh1 and Psh1-mediated degradation of Cse4 (Deyter et al.2017). We furthermore show here a genetic interaction between Cse4-S33 and the mutations E57A and L66A in histone H2A. These mutations lie in the α2 helix of the histone fold of H2A (Luger et al.1997) and are part of the ‘acidic patch’ on the nucleosome (Fig. S4, Supporting Information), which is important for contacts between nucleosomes, as it interacts with the H4 tail of the neighbouring nucleosome (Luger et al.1997; Kalashnikova et al.2013). Interestingly, H2A-E57A abrogates the interaction of H2A with the Spt16 subunit of the histone chaperone FACT (Hodges, Gloss and Wyrick 2017). Accordingly, gene-specific decreases in histone occupancy have been observed in both H2A-E57A and -L66A. Both residues also cause severe decreases in H2B-K123 ubiquitination as well as other PTMs (Cucinotta et al.2015). Furthermore, H2A-E57 and -L66 are part of the thiabendazole/benomyl-sensitive region II (TBS-II) of H2A. Mutations in this region cause defects in the establishment of chromosome bi-orientation, and they impair the localization of shugoshin (Sgo1) to the kinetochore (Kawashima et al.2011). Based on our conclusions above of a reduced deposition of mutant Cse4-S33 forms, we hypothesize that the two H2A mutations cause a defect in H2A/H2B occupancy at the centromeric nucleosome, which is further exacerbated by a reduced deposition of Cse4 in the cse4-S33E mutant as a result of its impaired interaction with the Scm3 chaperone. Interestingly, the defect of the H2A mutations is only enhanced by cse4-S33E, but not -S33A, suggesting that the imitation of the constantly phosphorylated state of Cse4 causes a stronger defect. How does S33 phosphorylation affect Cse4 function? The discovery of a synthetic defect with Cse4/H4 deposition suggests that the S33 phosphorylation likewise affects Cse4 deposition. In analogy to H4 acetylation, we postulate that the mutation of S33 reduces the interaction of Cse4/H4 with the histone chaperone Scm3 (the yeast homolog of HJURP). Our data indicate that the mutation of Cse4-S33 on its own causes a mild defect in Cse4 deposition that alone is insufficient to result in a growth defect, but which is enhanced when the interactions of both H4 and Cse4 to Scm3 are disturbed. This view parallels studies of CENP-A that demonstrate an important role for Cdk1-dependent phosphorylation serine 68 of the CATD of CENP-A in its interaction with HJURP and its deposition in chromatin (Yu et al.2015; Fachinetti et al.2017; Wang et al.2017). The respective structure of Scm3/Cse4/H4 reveals a similar involvement of the Cse4 CATD in the interaction with Scm3 (Cho and Harrison 2011; Zhou et al.2011). Our data indicate that, in addition to the CATD domain of Cse4, the Cse4 N-terminus, including S33 phosphorylation, mediates interaction with Scm3 for Cse4/H4 deposition. It therefore will be of interest to determine the molecular details of the interactions of the full-length Cse4 and Scm3 proteins in order to dissect the function of S33 phosphorylation. The fact that we identify circumstances here in which Cse4 levels at the centromere are reduced furthermore sheds light on the question of the composition of the centromeric nucleosome and the distribution of Cse4 at the centromere. It has variably been argued that the nucleosome contains one or two Cse4 molecules, depending on the phase of the cell cycle (Shivaraju et al.2012; Wisniewski et al.2014; Dhatchinamoorthy et al.2017). Also, microscopic studies show a broader radial distribution of Cse4 at the kinetochore compared to the Ndc80 complex, and this distribution is altered in selected mutants (Haase et al.2013). Our observations, together with the latter study, show that Cse4 levels can be reduced at the centromere, indicating that more than one Cse4 molecule is localized there for most of the cell cycle. Our analysis of Cse4-S33 phenotypes allows important insights into the role of this residue in centromere function. An earlier study identified S33 and additionally S22, S40 and S105 of Cse4 to be phosphorylated (Boeckmann et al.2013). An anti-phospho-Cse4 antibody was raised and revealed increased Cse4 phosphorylation upon nocodazole treatment, though the antibody did not distinguish between phosphorylation sites. Interestingly, all sites except S33 were phosphorylated in vitro by Ipl1, and overexpression of a phosphomimetic version of CSE4 including all phosphorylated sites (cse4-4SD) partially suppressed the temperature sensitivity of the ipl1-2 allele. Also, the cse4-4SD allele caused synthetic growth defects with mutations in OKP1 and AME1. Given our observations on S33, it seems likely that these effects reflect functions of the phosphorylation sites S22, S40 and S105, but not of S33. Which kinase is responsible for Cse4-S33ph? The earlier study excluded Ipl1/Aurora B as the S33 kinase (Boeckmann et al.2013). Obvious candidates are CDKs, in analogy to CENP-A phosphorylation by Cdk1 (Yu et al.2015; Fachinetti et al.2017), which is interesting in the light of the fact that S33 is predicted to be phosphorylated by Cdc2/Cdc28 (NetPhos 3.1) (Blom, Gammeltoft and Brunak 1999). Another possibility is the kinase Mps1, which monitors kinetochore—microtubule attachments and phosphorylates multiple targets including kinetochore components, checkpoint proteins and components of the spindle pole body in a sequential phosphorylation cascade to elicit a checkpoint response (Ji et al.2017). In summary, our genetic analyses of the Cse4-S33ph site in combination with mutations in the canonical histones have pinpointed selected residues in the H4 N-terminus and the H2A acidic patch as sites that cause synthetic growth defects. Interestingly, both regions are involved in the interaction of the histones with their respective chaperones Scm3 and FACT for deposition into chromatin. We therefore suggest that Cse4-S33ph likewise regulates its interaction with the Scm3 chaperone and that a reduced interaction reduces its deposition at centromeric sequences, which disturbs proper centromere function and chromosome segregation. SUPPLEMENTARY DATA Supplementary data are available at FEMSYR online. Acknowledgements We thank Ali Shilatifard for reagents (collection of histone mutations) and Josta Hamann and Silke Steinborn for technical support. FUNDING This work was supported by the Deutsche Forschungsgemeinschaft (EH237/10-1 and 12-1) and Humboldt-Universität zu Berlin. Conflict of interest. None declare. REFERENCES Allard S, Utley RT, Savard J et al.   NuA4, an essential transcription adaptor/histone H4 acetyltransferase complex containing Esa1p and the ATM-related cofactor Tra1p. EMBO J  1999; 18: 5108– 19. Google Scholar CrossRef Search ADS PubMed  Biggins S, Murray AW. The budding yeast protein kinase Ipl1/Aurora allows the absence of tension to activate the spindle checkpoint. Genes Dev  2001; 15: 3118– 29. Google Scholar CrossRef Search ADS PubMed  Black BE, Cleveland DW. Epigenetic centromere propagation and the nature of CENP-a nucleosomes. Cell  2011; 144: 471– 9. Google Scholar CrossRef Search ADS PubMed  Black BE, Foltz DR, Chakravarthy S et al.   Structural determinants for generating centromeric chromatin. Nature  2004; 430: 578– 82. Google Scholar CrossRef Search ADS PubMed  Blom N, Gammeltoft S, Brunak S. Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol  1999; 294: 1351– 62. Google Scholar CrossRef Search ADS PubMed  Boeckmann L, Takahashi Y, Au WC et al.   Phosphorylation of centromeric histone H3 variant regulates chromosome segregation in Saccharomyces cerevisiae. Mol Biol Cell  2013; 24: 2034– 44. Google Scholar CrossRef Search ADS PubMed  Boltengagen M, Huang A, Boltengagen A et al.   A novel role for the histone acetyltransferase Hat1 in the CENP-A/CID assembly pathway in Drosophila melanogaster. Nucleic Acids Res  2016; 44: 2145– 59. Google Scholar CrossRef Search ADS PubMed  Cai M, Davis RW. Yeast centromere binding protein CBF1, of the helix-loop-helix protein family, is required for chromosome stability and methionine prototrophy. Cell  1990; 61: 437– 46. Google Scholar CrossRef Search ADS PubMed  Camahort R, Li B, Florens L et al.   Scm3 is essential to recruit the histone h3 variant cse4 to centromeres and to maintain a functional kinetochore. Mol Cell  2007; 26: 853– 65. Google Scholar CrossRef Search ADS PubMed  Carroll CW, Milks KJ, Straight AF. Dual recognition of CENP-A nucleosomes is required for centromere assembly. J Cell Biol  2010; 189: 1143– 55. Google Scholar CrossRef Search ADS PubMed  Cheeseman IM. The kinetochore. Cold Spring Harb Perspect Biol  2014; 6: a015826. Google Scholar CrossRef Search ADS PubMed  Chen Y, Baker RE, Keith KC et al.   The N terminus of the centromere H3-like protein Cse4p performs an essential function distinct from that of the histone fold domain. Mol Cell Biol  2000; 20: 7037– 48. Google Scholar CrossRef Search ADS PubMed  Cho US, Harrison SC. Recognition of the centromere-specific histone Cse4 by the chaperone Scm3. P Natl Acad Sci USA  2011; 108: 9367– 71. Google Scholar CrossRef Search ADS   Cucinotta CE, Young AN, Klucevsek KM et al.   The nucleosome acidic patch regulates the H2B K123 monoubiquitylation cascade and transcription elongation in saccharomyces cerevisiae. PLoS Genet  2015; 11: e1005420. Google Scholar CrossRef Search ADS PubMed  Dai J, Hyland EM, Yuan DS et al.   Probing nucleosome function: a highly versatile library of synthetic histone H3 and H4 mutants. Cell  2008; 134: 1066– 78. Google Scholar CrossRef Search ADS PubMed  Davie JR, Saunders CA, Walsh JM et al.   Histone modifications in the yeast S. Cerevisiae. Nucleic Acids Res  1981; 9: 3205– 16. Google Scholar CrossRef Search ADS PubMed  Deyter GM, Hildebrand EM, Barber AD et al.   Histone H4 facilitates the proteolysis of the budding yeast CENP-ACse4 centromeric histone variant. Genetics  2017; 205: 113– 24. Google Scholar CrossRef Search ADS PubMed  Dhatchinamoorthy K, Shivaraju M, Lange JJ et al.   Structural plasticity of the living kinetochore. J Cell Biol  2017; 216: 3551– 70. Google Scholar CrossRef Search ADS PubMed  Dunleavy EM, Roche D, Tagami H et al.   HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell  2009; 137: 485– 97. Google Scholar CrossRef Search ADS PubMed  Earnshaw WC, Rothfield N. Identification of a family of human centromere proteins using autoimmune sera from patients with scleroderma. Chromosoma  1985; 91: 313– 21. Google Scholar CrossRef Search ADS PubMed  Fachinetti D, Logsdon GA, Abdullah A et al.   CENP-A modifications on Ser68 and Lys124 are dispensable for establishment, maintenance, and long-term function of human centromeres. Dev Cell  2017; 40: 104– 13. Google Scholar CrossRef Search ADS PubMed  Falk SJ, Lee J, Sekulic N et al.   CENP-C directs a structural transition of CENP-a nucleosomes mainly through sliding of DNA gyres. Nat Struct Mol Biol  2016; 23: 204– 8. Google Scholar CrossRef Search ADS PubMed  Foltz DR, Jansen LE, Bailey AO et al.   Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURP. Cell  2009; 137: 472– 84. Google Scholar CrossRef Search ADS PubMed  Fukagawa T. Critical histone post-translational modifications for centromere function and propagation. Cell Cycle  2017; 16: 1259– 65. Google Scholar CrossRef Search ADS PubMed  Funabiki H, Wynne DJ. Making an effective switch at the kinetochore by phosphorylation and dephosphorylation. Chromosoma  2013; 122: 135– 58. Google Scholar CrossRef Search ADS PubMed  Gascoigne KE, Takeuchi K, Suzuki A et al.   Induced ectopic kinetochore assembly bypasses the requirement for CENP-A nucleosomes. Cell  2011; 145: 410– 22. Google Scholar CrossRef Search ADS PubMed  Green EM, Mas G, Young NL et al.   Methylation of H4 lysines 5, 8 and 12 by yeast Set5 calibrates chromatin stress responses. Nat Struct Mol Biol  2012; 19: 361– 3. Google Scholar CrossRef Search ADS PubMed  Haase J, Mishra PK, Stephens A et al.   A 3D map of the yeast kinetochore reveals the presence of core and accessory centromere-specific histone. Curr Biol  2013; 23: 1939– 44. Google Scholar CrossRef Search ADS PubMed  Hewawasam G, Shivaraju M, Mattingly M et al.   Psh1 is an E3 ubiquitin ligase that targets the centromeric histone variant Cse4. Mol Cell  2010; 40: 444– 54. Google Scholar CrossRef Search ADS PubMed  Hodges AJ, Gloss LM, Wyrick JJ. Residues in the nucleosome acidic patch regulate histone occupancy and are important for FACT binding in saccharomyces cerevisiae. Genetics  2017; 206: 1339– 48. Google Scholar CrossRef Search ADS PubMed  Huis In’t Veld PJ, Jeganathan S, Petrovic A et al.   Molecular basis of outer kinetochore assembly on CENP-T. eLife  2016; 5: e21007. Google Scholar PubMed  Hyland EM, Cosgrove MS, Molina H et al.   Insights into the role of histone H3 and histone H4 core modifiable residues in Saccharomyces cerevisiae. Mol Cell Biol  2005; 25: 10060– 70. Google Scholar CrossRef Search ADS PubMed  Ji Z, Gao H, Jia L et al.   A sequential multi-target Mps1 phosphorylation cascade promotes spindle checkpoint signaling. eLife  2017; 6: e22513. Google Scholar CrossRef Search ADS PubMed  Kalashnikova AA, Porter-Goff ME, Muthurajan UM et al.   The role of the nucleosome acidic patch in modulating higher order chromatin structure. J R Soc, Interface  2013; 10: 20121022. Google Scholar CrossRef Search ADS   Kawashima S, Nakabayashi Y, Matsubara K et al.   Global analysis of core histones reveals nucleosomal surfaces required for chromosome bi-orientation. EMBO J  2011; 30: 3353– 67. Google Scholar CrossRef Search ADS PubMed  Kawashima SA, Tsukahara T, Langegger M et al.   Shugoshin enables tension-generating attachment of kinetochores by loading Aurora to centromeres. Genes Dev  2007; 21: 420– 35. Google Scholar CrossRef Search ADS PubMed  Kleff S, Andrulis ED, Anderson CW et al.   Identification of a gene encoding a yeast histone H4 acetyltransferase. J Biol Chem  1995; 270: 24674– 7. Google Scholar CrossRef Search ADS PubMed  Krenn V, Overlack K, Primorac I et al.   KI motifs of human Knl1 enhance assembly of comprehensive spindle checkpoint complexes around MELT repeats. Curr Biol  2014; 24: 29– 39. Google Scholar CrossRef Search ADS PubMed  Lochmann B, Ivanov D. Histone H3 localizes to the centromeric DNA in budding yeast. PLoS Genet  2012; 8: e1002739. Google Scholar CrossRef Search ADS PubMed  London N, Biggins S. Mad1 kinetochore recruitment by Mps1-mediated phosphorylation of Bub1 signals the spindle checkpoint. Genes Dev  2014; 28: 140– 52. Google Scholar CrossRef Search ADS PubMed  Longtine MS, McKenzie A3rd, Demarini DJ et al.   Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast  1998; 14: 953– 61. Google Scholar CrossRef Search ADS PubMed  Luger K, Mader AW, Richmond RK et al.   Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature  1997; 389: 251– 60. Google Scholar CrossRef Search ADS PubMed  Malvezzi F, Litos G, Schleiffer A et al.   A structural basis for kinetochore recruitment of the Ndc80 complex via two distinct centromere receptors. EMBO J  2013; 32: 409– 23. Google Scholar CrossRef Search ADS PubMed  Matsubara K, Sano N, Umehara T et al.   Global analysis of functional surfaces of core histones with comprehensive point mutants. Genes Cells  2007; 12: 13– 33. Google Scholar CrossRef Search ADS PubMed  McKinley KL, Cheeseman IM. The molecular basis for centromere identity and function. Nat Rev Mol Cell Biol  2016; 17: 16– 29. Google Scholar CrossRef Search ADS PubMed  Megee PC, Morgan BA, Mittman BA et al.   Genetic analysis of histone H4: essential role of lysines subject to reversible acetylation. Science  1990; 247: 841– 5. Google Scholar CrossRef Search ADS PubMed  Meluh PB, Yang P, Glowczewski L et al.   Cse4p is a component of the core centromere of Saccharomyces cerevisiae. Cell  1998; 94: 607– 13. Google Scholar CrossRef Search ADS PubMed  Moyle MW, Kim T, Hattersley N et al.   A Bub1-Mad1 interaction targets the Mad1-Mad2 complex to unattached kinetochores to initiate the spindle checkpoint. J Cell Biol  2014; 204: 647– 57. Google Scholar CrossRef Search ADS PubMed  Nakajima Y, Tyers RG, Wong CC et al.   Nbl1p: a Borealin/Dasra/CSC-1-like protein essential for Aurora/Ipl1 complex function and integrity in Saccharomyces cerevisiae. Mol Biol Cell  2009; 20: 1772– 84. Google Scholar CrossRef Search ADS PubMed  Nakanishi S, Lee JS, Gardner KE et al.   Histone H2BK123 monoubiquitination is the critical determinant for H3K4 and H3K79 trimethylation by COMPASS and Dot1. J Cell Biol  2009; 186: 371– 7. Google Scholar CrossRef Search ADS PubMed  Nelson DA. Histone acetylation in baker's yeast. Maintenance of the hyperacetylated configuration in log phase protoplasts. J Biol Chem  1982; 257: 1565– 8. Google Scholar PubMed  Ng TM, Lenstra TL, Duggan N et al.   Kinetochore function and chromosome segregation rely on critical residues in histones H3 and H4 in budding yeast. Genetics  2013; 195: 795– 807. Google Scholar CrossRef Search ADS PubMed  Ortiz J, Stemmann O, Rank S et al.   A putative protein complex consisting of Ctf19, Mcm21, and Okp1 represents a missing link in the budding yeast kinetochore. Genes Dev  1999; 13: 1140– 55. Google Scholar CrossRef Search ADS PubMed  Palmer DK, O’Day K, Wener MH et al.   A 17-kD centromere protein (CENP-A) copurifies with nucleosome core particles and with histones. J Cell Biol  1987; 104: 805– 15. Google Scholar CrossRef Search ADS PubMed  Parthun MR, Widom J, Gottschling DE. The major cytoplasmic histone acetyltransferase in yeast - links to chromatin replication and histone metabolism. Cell  1996; 87: 85– 94. Google Scholar CrossRef Search ADS PubMed  Pesenti ME, Weir JR, Musacchio A. Progress in the structural and functional characterization of kinetochores. Curr Opin Struct Biol  2016; 37: 152– 63. Google Scholar CrossRef Search ADS PubMed  Ranjitkar P, Press MO, Yi X et al.   An E3 ubiquitin ligase prevents ectopic localization of the centromeric histone H3 variant via the centromere targeting domain. Mol Cell  2010; 40: 455– 64. Google Scholar CrossRef Search ADS PubMed  Ruchaud S, Carmena M, Earnshaw WC. Chromosomal passengers: conducting cell division. Nat Rev Mol Cell Biol  2007; 8: 798– 812. Google Scholar CrossRef Search ADS PubMed  Sakamoto M, Noguchi S, Kawashima S et al.   Global analysis of mutual interaction surfaces of nucleosomes with comprehensive point mutants. Genes Cells  2009; 14: 1271– 330. Google Scholar CrossRef Search ADS PubMed  Samel A, Cuomo A, Bonaldi T et al.   Methylation of CenH3 arginine 37 regulates kinetochore integrity and chromosome segregation. Proc Natl Acad Sci U S A  2012; 109: 9029– 34. Google Scholar CrossRef Search ADS PubMed  Sandall S, Severin F, McLeod IX et al.   A Bir1-Sli15 complex connects centromeres to microtubules and is required to sense kinetochore tension. Cell  2006; 127: 1179– 91. Google Scholar CrossRef Search ADS PubMed  Shang WH, Hori T, Westhorpe FG et al.   Acetylation of histone H4 lysine 5 and 12 is required for CENP-A deposition into centromeres. Nature Commun  2016; 7: 13465. Google Scholar CrossRef Search ADS   Sherman F. Getting started with yeast. Methods Enzymol  1991; 194: 3– 21. Google Scholar CrossRef Search ADS PubMed  Shivaraju M, Unruh JR, Slaughter BD et al.   Cell-cycle-coupled structural oscillation of centromeric nucleosomes in yeast. Cell  2012; 150: 304– 16. Google Scholar CrossRef Search ADS PubMed  Shuaib M, Ouararhni K, Dimitrov S et al.   HJURP binds CENP-A via a highly conserved N-terminal domain and mediates its deposition at centromeres. Proc Natl Acad Sci USA  2010; 107: 1349– 54. Google Scholar CrossRef Search ADS PubMed  Stoler S, Keith KC, Curnick KE et al.   A mutation in CSE4, an essential gene encoding a novel chromatin-associated protein in yeast, causes chromosome nondisjunction and cell cycle arrest at mitosis. Genes Dev  1995; 9: 573– 86. Google Scholar CrossRef Search ADS PubMed  Stoler S, Rogers K, Weitze S et al.   Scm3, an essential Saccharomyces cerevisiae centromere protein required for G2/M progression and Cse4 localization. Proc Natl Acad Sci USA  2007; 104: 10571– 6. Google Scholar CrossRef Search ADS PubMed  Suka N, Suka Y, Carmen AA et al.   Highly specific antibodies determine histone acetylation site usage in yeast heterochromatin and euchromatin. Mol Cell  2001; 8: 473– 9. Google Scholar CrossRef Search ADS PubMed  Tanaka TU, Rachidi N, Janke C et al.   Evidence that the Ipl1-Sli15 (Aurora kinase-INCENP) complex promotes chromosome bi-orientation by altering kinetochore-spindle pole connections. Cell  2002; 108: 317– 29. Google Scholar CrossRef Search ADS PubMed  Tong AH, Evangelista M, Parsons AB et al.   Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science  2001; 294: 2364– 8. Google Scholar CrossRef Search ADS PubMed  Torres-Machorro AL, Clark LG, Chang CS et al.   The Set3 complex antagonizes the MYST Acetyltransferase Esa1 in the DNA damage response. Mol Cell Biol  2015; 35: 3714– 25. Google Scholar CrossRef Search ADS PubMed  Wang K, Yu Z, Liu Y et al.   Ser68 phosphorylation ensures accurate cell-cycle-dependent CENP-A deposition at centromeres. Develop Cell  2017; 40: 5– 6. Google Scholar CrossRef Search ADS   Weber JM, Irlbacher H, Ehrenhofer-Murray AE. Control of replication initiation by the Sum1/Rfm1/Hst1 histone deacetylase. BMC Mol Biol  2008; 9: 100. Google Scholar CrossRef Search ADS PubMed  Weiner A, Hsieh TH, Appleboim A et al.   High-resolution chromatin dynamics during a yeast stress response. Mol Cell  2015; 58: 371– 86. Google Scholar CrossRef Search ADS PubMed  Welburn JP, Vleugel M, Liu D et al.   Aurora B phosphorylates spatially distinct targets to differentially regulate the kinetochore-microtubule interface. Mol Cell  2010; 38: 383– 92. Google Scholar CrossRef Search ADS PubMed  Wisniewski J, Hajj B, Chen J et al.   Imaging the fate of histone Cse4 reveals de novo replacement in S phase and subsequent stable residence at centromeres. eLife  2014; 3: e02203. Google Scholar CrossRef Search ADS PubMed  Yamagishi Y, Yang CH, Tanno Y et al.   MPS1/Mph1 phosphorylates the kinetochore protein KNL1/Spc7 to recruit SAC components. Nat Cell Biol  2012; 14: 746– 52. Google Scholar CrossRef Search ADS PubMed  Yu Y, Srinivasan M, Nakanishi S et al.   A conserved patch near the C terminus of histone H4 is required for genome stability in budding yeast. Mol Cell Biol  2011; 31: 2311– 25. Google Scholar CrossRef Search ADS PubMed  Yu Z, Zhou X, Wang W et al.   Dynamic phosphorylation of CENP-A at Ser68 orchestrates its cell-cycle-dependent deposition at centromeres. Develop Cell  2015; 32: 68– 81. Google Scholar CrossRef Search ADS   Zhou Z, Feng H, Zhou BR et al.   Structural basis for recognition of centromere histone variant CenH3 by the chaperone Scm3. Nature  2011; 472: 234– 7. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Yeast Research Oxford University Press

A role for CENP-A/Cse4 phosphorylation on serine 33 in deposition at the centromere

Loading next page...
 
/lp/ou_press/a-role-for-cenp-a-cse4-phosphorylation-on-serine-33-in-deposition-at-biIPXZS44h
Publisher
Blackwell
Copyright
© FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1567-1356
eISSN
1567-1364
D.O.I.
10.1093/femsyr/fox094
Publisher site
See Article on Publisher Site

Abstract

Abstract Centromeres are the sites of assembly of the kinetochore, which connect the chromatids to the microtubules for sister chromatid segregation during cell division. Centromeres are characterized by the presence of the histone H3 variant CENP-A (termed Cse4 in Saccharomyces cerevisiae). Here, we investigated the function of serine 33 phosphorylation of Cse4 (Cse4-S33ph) in S. cerevisiae, which lies within the essential N-terminal domain (END) of the extended Cse4 N-terminus. Significantly, we identified histone H4-K5, 8, 12R to cause a temperature-sensitive growth defect with mutations in Cse4-S33 and sensitivity to nocodazole and hydroxyurea. Furthermore, the absence of Cse4-S33ph reduced the levels of Cse4 at centromeric sequences, suggesting that Cse4 deposition is defective in the absence of S33 phosphorylation. We furthermore identified synthetic genetic interactions with histone H2A-E57A and H2A-L66A, which both cause a reduced interaction with the histone chaperone FACT and reduced H2A/H2B levels in chromatin, again supporting the notion that a combined defect of H2A/H2B and Cse4 deposition causes centromeric defects. Altogether, our data highlight the importance of correct histone deposition in building a functional centromeric nucleosome and suggests a role for Cse4-S33ph in this process. Cse4, Hat1, Ser33, centromere, kinetochore INTRODUCTION Centromeres are defined regions of the chromosomes where the kinetochores are assembled and provide a physical attachment between centromeric chromatin and the microtubules for chromosome segregation in the metaphase of the cell cycle (reviewed in Pesenti, Weir and Musacchio 2016). The centromeres are distinguished from other chromosomal regions by the presence of a variant version of histone H3, CENP-A (termed Cse4 in Saccharomyces cerevisiae), which determines the site of recruitment of kinetochore proteins to the chromosome (Earnshaw and Rothfield 1985; Palmer et al.1987; Stoler et al.1995; Meluh et al.1998). In higher eukaryotes, CENP-A-carrying nucleosomes are interspersed with nucleosomes containing canonical H3 (McKinley and Cheeseman 2016). In contrast, S. cerevisiae has a point centromere consisting of what is most likely a single centromeric nucleosome that contains Cse4, although there has been controversy surrounding the precise nature of this particle (Black and Cleveland 2011). One difference between Cse4 and its CENP-A homologs is that it contains a large, 130-amino acid N-terminal domain (Stoler et al.1995). Part of this domain is essential for centromere function, because deletion of the END region (essential N-terminal domain, amino acids 28 - 60) is lethal (Chen et al.2000). This domain recruits inner kinetochore complexes to the centromere (Ortiz et al.1999) and possibly serves to compensate for the fact that S. cerevisiae has only one rather than multiple CENP-A/Cse4 nucleosomes at the centromere, as is the case in higher eukaryotes. CENP-A/H4 and Cse4/H4 are deposited at the centromeric sequences by histone chaperones that are specialized for the purpose, namely HJURP (Dunleavy et al.2009; Foltz et al.2009; Shuaib et al.2010) in higher eukaryotes and Scm3 in yeast (Camahort et al.2007; Stoler et al.2007). They distinguish the centromeric H3 variant from canonical H3 via the so-called CENP-A targeting domain (CATD), which lies in the core region of CENP-A/Cse4 (Black et al.2004) and makes direct contacts to HJURP/Scm3 (Cho and Harrison 2011; Zhou et al.2011). Interestingly, posttranslational modifications (PTMs) on CENP-A and H4 regulate their interaction with HJURP and thus deposition in chromatin. CENP-A is phosphorylated by the kinase Cdk1 on serine 68 (S68), which lies close to the CATD. This phosphorylation reduces CENP-A binding to HJURP and thus prevents premature loading of CENP-A at the centromeres (Yu et al.2015; Fachinetti et al.2017; Wang et al.2017). Furthermore, H4 K5 and K12 are acetylated in the context of CENP-A/H4 by the RbAp46/48-Hat1 complex before they are incorporated into centromeric chromatin, and CENP-A deposition is reduced in cells deficient for RbAp48 (Shang et al.2016). A conserved role for H4 acetylation in CENP-A deposition is further underscored by the finding that the CENP-A homolog in Drosophila (Cid) is present in a complex with Hat1 and the chromatin assembly factor Caf1, and Hat1 is required for proper Cid loading into centromeric chromatin (Boltengagen et al.2016). It has been proposed that the absence of H4 acetylation reduces the interaction of the CENP-A/H4 dimer with the histone chaperone HJURP, which thus restricts CENP-A deposition at the centromere. CENP-A-containing nucleosomes show specific structural properties compared to canonical nucleosomes that are necessary to mediate the binding of inner kinetochore proteins, for instance CENP-C/Mif2 (Carroll, Milks and Straight 2010; Falk et al.2016), which then recruits other kinetochore sub-complexes. The most chromatin-proximal complexes in S. cerevisiae are the CBF3 complex and the CTF19 complex, the yeast equivalent of the constitutive centromere-associated network (CCAN) in higher eukaryotes. These complexes interact with the KMN (Knl1 complex, Mis12 complex and Ndc80 complex) network, within which the NDC80 complex provides a direct contact with the microtubule (reviewed in Cheeseman 2014). Proper kinetochore—microtubule attachment is monitored by the chromosomal passenger complex (CPC), which consists of the Ipl1/Aurora B kinase, Sli5/INCENP, Bir1/Survivin and Nbl1/Borealin (Biggins and Murray 2001; Tanaka et al.2002; Sandall et al.2006; Ruchaud, Carmena and Earnshaw 2007; Nakajima et al.2009). The recruitment of the CPC to the kinetochore in part is regulated by an interaction between Bir1 and shugoshin (Sgo1) (Kawashima et al.2007). Thus, a complex hierarchy of multi-protein complexes is assembled at centromeric sequences to link chromatin to the microtubule and to pull sister chromatids apart. Phosphorylation events play a major role in regulating interactions among kinetochore proteins as well as the interaction between the microtubule and the kinetochore (reviewed in Funabiki and Wynne 2013). Aurora B phosphorylates multiple targets in the KMN network and thus regulates interactions to the microtubule in response to changes in tension and attachment state (Welburn et al.2010). Furthermore, cyclin-dependent kinase (CDK) phosphorylates CENP-T, which regulates its interaction with the NDC80 complex as well as the KMN network (Gascoigne et al.2011; Malvezzi et al.2013; Huis In’t Veld et al.2016). Other phosphorylation events control the activation of checkpoints, most prominently the spindle assembly checkpoint (SAC). As an example, the kinase Mps1 phosphorylates the Knl1 protein, which regulates the recruitment of the Bub1-Bub3 complex (Yamagishi et al.2012). Mps1 subsequently phosphorylates sites within Bub1, which promotes binding of Mad1 to the kinetochore to initiate the SAC (Krenn et al.2014; London and Biggins 2014; Moyle et al.2014). Next to the phosphorylation of kinetochore components, other PTMs on CENP-A, including methylation, acetylation and ubiquitination play an important role in centromere function (Fukagawa 2017). Similarly to CENP-A, S. cerevisiae Cse4 carries multiple PTMs. Ubiquitination in the C-terminal region of Cse4 by the E3 ubiquitin ligase Psh1 prevents its inappropriate incorporation at non-centromeric regions (Hewawasam et al.2010; Ranjitkar et al.2010). Furthermore, we identified methylation on arginine 37 (R37) of Cse4, which lies within the END domain, and the absence of methylation causes selective defects in combination with mutations in components of the CTF19 complex, indicating that this PTM regulates interactions of the Cse4 N-terminus with kinetochore components (Samel et al.2012). Cse4 furthermore is acetylated on K49 and phosphorylated on serines 20, 33, 40 and 105 (Boeckmann et al.2013). Simultaneous mutation of all four phosphorylation sites causes centromeric defects, but the individual contribution of each site has not been dissected. Here, we sought to determine how Cse4-S33 phosphorylation (Cse4-S33ph) regulates centromere function. Screens were performed of cse4-S33 mutations with libraries of systematic mutations in the canonical histones H2A, H2B, H3 and H4. Significantly, this revealed selected mutations in H4-K5, 8, 12R, H2A-E57A and H2A-L66A to cause synthetic phenotypes, whereas no defects were uncovered in a comprehensive analysis of mutations in genes encoding kinetochore components. Interestingly, both the H4 and the H2A residues have been implicated in the deposition of the respective histones in chromatin, suggesting an equivalent role for Cse4-S33ph in its deposition in chromatin. Accordingly, we found reduced levels of unphosphorylated Cse4 at centromeric regions. Altogether, our analysis shows an important role for PTMs on H4 and Cse4 as well as of H2A deposition in building a functional centromeric nucleosome. MATERIAL AND METHODS Yeast strains and plasmids The S. cerevisiae strains and plasmids used in this study are listed in Tables S1 and S2 (Supporting Information), respectively. Yeast was grown and manipulated according to standard procedures (Sherman 1991). Yeast was grown on full medium (YPD) and selective minimal plates (YM). Genomic integration of cse4 alleles was performed by cloning the allele on a URA3-marked integrative vector and introducing it into the strain by integrative transformation followed by loop-out on 5-fluoroorotic acid (5- FOA)-containing medium. For genetic crosses, cse4 was marked with natMX or kanMX by the integration of the respective resistance cassettes downstream of the open reading frame. Deletions of chromosomal genes were performed using the integration of knockout cassettes (Longtine et al.1998). Growth curves of yeast cultures were measured using a microplate reader (SynergyH1, BioTek). 200 μl cultures in a 96-well plate were inoculated to an optical density at 600 nm (OD600) of 0.1 in YPD. OD600 was measured in 10 min intervals with double orbital shaking. Mass spectrometric analysis of Cse4 Purification and analysis of 3xHA-tagged Cse4 from yeast cells was performed essentially as described (Samel et al.2012). Briefly, partially purified histones from cells expressing 3xHA-Cse4 were separated on 10% SDS-PAGE gels, the Cse4 band excised and digested in-gel with trypsin. The peptide mixture were analyzed by online nano-flow liquid chromatography tandem mass spectrometry using an EASY-nLC™ system (ProxeonBiosystems, Odense, Denmark) connected to the LTQ-OrbitrapVelos (Thermo Electron, Bremen, Germany) through a nano-electrospray ion source (see Materials and Methods for details, Supporting Information). Synthetic lethal screens with libraries of mutations in histone genes A library of mutations in H3 and H4 ((Dai et al.2008), obtained from ThermoFisher, catalog number YSC5106) was screened for synthetic growth defects with cse4-S33A (AEY6071) or -S33E (AEY6070) and, as a control, with a CSE4 strain (AEY6067) using a Rotor HDA spotting robot (Singer Instruments). Diploids were selected, sporulated, and haploids of the desired genotype were selected in three consecutive rounds of selection. The first round selected for lyp1Δ can1Δ::MFA1pr-his5 hht1-hhf2Δ::NatMX haploids, the second round additionally applied selection for the cse4::kanMX alleles, and the third round selected for hht2–hhf2::[HHTS-HHFS]*-URA3 in addition to the previously selected markers. Selection from the last and the penultimate selection were compared for growth differences. The screening procedure was repeated three independent times for each cse4 strain. Mutations with a potential effect were re-arrayed and retested. For the search of mutations in H2A and H2B with a synthetic growth defect with cse4-S33 alleles, the strains AEY5984 (CSE4), AEY5988 (cse4-S33E) and AEY6105 (cse4-S33A) were crossed to the mutant collection (Nakanishi et al.2009). After diploid selection and sporulation, haploids were selected in a first round for lyp1Δ, pHIS3-HTA1*-flag-HTB1**hta1-htb1Δ::LEU2 and hta2-htb2Δ::URAMX, in a second round the same selection was applied as in the first round, and the third round added a selection for the cse4 allele. Candidate strains for a synthetic growth defect were chosen, re-arrayed and retested as above. Chromatin immunoprecipitation Chromatin immunoprecipitation (ChIP) and quantitative real-time PCR were performed essentially as previously described (Weber, Irlbacher and Ehrenhofer-Murray 2008) using strains AEY6242, AEY6244 and AEY6280. Cells were grown at 23°C to an OD600 of approx. 0.5 and shifted to 34°C for 3 h prior to formaldehyde crosslinking, cell harvest and chromatin isolation. Primer sequences are available upon request. RESULTS Phosphorylation on serine 33 of Cse4 has a functionally distinct role from Cse4-R37 methylation The centromeric histone H3 variant Cse4 from S. cerevisiae has previously been reported to be phosphorylated on serine 33 (Boeckmann et al.2013), a modification site that we also identified in four independent purifications of Cse4 from yeast cells (Figs S1 and S2, Supporting Information). However, the precise function of this modification has not been investigated. Interestingly, this site lies within the END domain (AA 28–60), close to the methylated residue R37 (Samel et al.2012). Our quantification showed that S33 phosphorylation is less abundant (approx. 2%) than mono-methylation of R37 (approx. 16%, Fig. S1, Supporting Information). To investigate the function of Cse4-S33ph, we created mutant versions of Cse4 in which S33 was replaced by alanine (cse4-S33A) or glutamate (cse4-S33E) to imitate the unphosphorylated or constantly phosphorylated form of Cse4, respectively. Strains carrying these Cse4 versions alone were viable and showed no obvious growth defects (Fig. 1), indicating that this phosphorylation is not essential for Cse4 function in cell viability. Figure 1. View largeDownload slide Mutation of the phosphorylation site S33 of Cse4 caused synthetic growth defects with H4-K5, 8, 12R. Serial dilutions of strains carrying H4 alleles with lysine 5 and 8 or 5 and 12 mutated to arginine and CSE4 (top), cse4-S33E (middle) or -S33A (bottom) were spotted on full medium and grown at the indicated temperatures for 3 days. Figure 1. View largeDownload slide Mutation of the phosphorylation site S33 of Cse4 caused synthetic growth defects with H4-K5, 8, 12R. Serial dilutions of strains carrying H4 alleles with lysine 5 and 8 or 5 and 12 mutated to arginine and CSE4 (top), cse4-S33E (middle) or -S33A (bottom) were spotted on full medium and grown at the indicated temperatures for 3 days. Since S33 is located in the proximity of the methylated R37 residue of Cse4, we asked whether S33 mutations caused similar phenotypes as abrogation of Cse4-R37 methylation (cse4-R37A). Unlike cse4-R37A, however, neither cse4-S33A nor -S33E caused lethality or synthetic growth defects with mutations in the CDEI-binding protein Cbf1 (Cai and Davis 1990) or in Ctf19 complex components (Table S3, Supporting Information). There also were no additional growth defects in combination with mutations in CBF3, NDC80, Knl1 or Ctf3 complex components, and no synthetic growth defects were found when combining cse4-S33 mutations with the mutation of Cse4-R37 (Table S3, Supporting Information). This suggests that Cse4-S33ph has a function at the centromere that is distinct from that of R37 methylation, although the two sites are in close proximity within the END domain of Cse4. Absence of Cse4-S33 phosphorylation causes a synthetic defect in combination with the mutation K5, 8, 12R in histone H4 To obtain insight into the function of Cse4-S33ph, we sought to identify mutations in other genes that cause a synthetic growth defect with cse4-S33 mutations. No reproducible candidates were recovered from a screen with a library of gene deletions (Tong et al.2001) (data not shown), indicating that cse4-S33 mutations have no synthetic interactions with non-essential genes. We next asked whether synthetic interactions with mutations in histone genes could be identified. For this purpose, a genetic screen was conducted in which cse4-S33A or -S33E was combined with a library of 486 mutations in histone H3 and H4. In this collection, each histone amino acid is systematically replaced by alanine; natural alanines are replaced by serine; and the collection contains several other unique mutations, for instance in histone residues that are known to carry PTMs and where the mutation mimics the (un-) modified state (see Materials and Methods, Dai et al.2008). Interestingly, the screen revealed a defect for combinations of cse4-S33 mutations with H4-K8A and -K12Q, which both represent sites of lysine acetylation in H4 (Davie et al.1981; Nelson 1982). No mutations in H3 were recovered, which is notable in the light of the fact that the centromeric nucleosome may also contain some H3 (Lochmann and Ivanov 2012). To further evaluate these effects in a well-defined genetic background, plasmid shuffle experiments were conducted in strains carrying the different cse4 alleles as well as deletions of the HHT1–HHF1 and HHT2–HHF2 gene copies, which were kept viable by a URA3-marked HHT1–HHF1 plasmid. Plasmids were introduced with mutations in histone H4 acetylation sites that we had at our disposal. After the elimination of the wild-type HHT1–HHF1 plasmid by counter-selection on 5-FOA, the growth characteristics of the strains were tested. Importantly, the combination of cse4-S33A or -S33E with H4-K5, K8, K12R caused a pronounced growth defect at elevated temperatures that was not observed with the cse4-S33 or H4 mutations alone (Fig. 1). The defect was more pronounced for cse4-S33A, and it was not observed for the individual mutations H4-K5R, 8R or H4-K5R, 12R, showing that simultaneous mutation of all three residues was necessary to cause the synthetic growth defect. To further quantitate the growth differences, growth curves were generated. A mild growth defect for H4-K5, 8, 12R was observed in the CSE4 strain. Significantly, both cse4-S33E and -S33A showed a pronounced reduction in growth with H4-K5, 8, 12R, but not with wt H3 and H4 (Fig. 2), further supporting the notion that the absence of Cse4 S33 phosphorylation causes a defect in combination with mutations of the lysine residues K5, K8 and K12 of H4, which are sites that are subject to lysine acetylation. Figure 2. View largeDownload slide H4-K5, 8, 12R with cse4-S33E or -S33A showed slow growth in liquid medium. Growth curves of the H4-K5, 8, 12R strains were measured at 36°C using a microtiter dish plate reader. Arbitrary units of the optical density at 600 nm (OD600) are shown relative to cultivation time. Mean values ± SD (n = 3) are shown. *P < 0.05; **P < 0.01: ***P < 0.001 (two-sided t-test). Figure 2. View largeDownload slide H4-K5, 8, 12R with cse4-S33E or -S33A showed slow growth in liquid medium. Growth curves of the H4-K5, 8, 12R strains were measured at 36°C using a microtiter dish plate reader. Arbitrary units of the optical density at 600 nm (OD600) are shown relative to cultivation time. Mean values ± SD (n = 3) are shown. *P < 0.05; **P < 0.01: ***P < 0.001 (two-sided t-test). We also tested the effect of H4-K12Q using plasmid shuffle, but even though this allele had shown defects in the high-throughput screen of the histone mutant collection, we observed no phenotypes in the plasmid shuffle strain, which may be due to strain background differences. Also, the mutations of H4-K5A, -K12A or mutations in K16 (K16A, K16Q), which also can be acetylated, did not display a defect with cse4-S33 mutations (not shown). Thus, the synthetic growth defect of cse4-S33 mutations showed selectivity for K5, K8 and K12 and was not shared with the acetylation site K16. The discovery of a synthetic interaction of the absence of Cse4-S33ph with acetylation sites in H4 was interesting, because H4 acetylation has previously been linked to a defect in CENP-A deposition (Boltengagen et al.2016; Shang et al.2016), and likewise suggested a role for S33ph in Cse4 deposition at the centromere. cse4-S33 mutations cause a mitotic defect with H4-K5, 8, 12R We next asked whether the growth defect of cse4-S33A and -S33E with the H4 mutations reflected a defect in mitosis. Therefore, sensitivity of the strains to the microtubule-inhibiting drug nocodazole and the S-phase inhibitor hydroxy-urea (HU) was tested. In the CSE4 background, H4-K5,8,12R alone showed a mild sensitivity towards nocodazole and HU. Importantly, the sensitivity was strongly enhanced in the cse4-S33A and -S33E strains, which themselves were insensitive to the compounds (Fig. 3). This showed that the inability to phosphorylate Cse4 S33 by itself did not disturb mitotic function or replication. However, the combination of Cse4-S33 mutation and H4-K5, 8, 12R caused strong defects in the two processes, indicating that centromeric function was defective in these mutants. Figure 3. View largeDownload slide Absence of Cse4 phosphorylation caused a defect in mitosis and replication in cells carrying H4-K5, 8, 12R. Serial dilutions of the strains as in Fig. 1 were spotted on YPD containing 3 μM nocodazole (noc) or 100 mM hydroxyurea (HU) and grown at 30°C for 3 days. Figure 3. View largeDownload slide Absence of Cse4 phosphorylation caused a defect in mitosis and replication in cells carrying H4-K5, 8, 12R. Serial dilutions of the strains as in Fig. 1 were spotted on YPD containing 3 μM nocodazole (noc) or 100 mM hydroxyurea (HU) and grown at 30°C for 3 days. hat1Δ and esa1-L327S mutations does not cause a synthetic defect with cse4-S33 mutations Since the allele H4-K5,8,12R mutates three acetylated lysines of H4 and imitates the deacetylated state, we asked whether mutations in the histone acetyltransferases responsible for their acetylation likewise caused a synthetic growth defect with cse4-S33A or -S33E. Hat1 is the catalytic subunit of the Hat1-Hat2 complex and acetylates H4-K5 and K12 (Kleff et al.1995; Parthun, Widom and Gottschling 1996). Furthermore, the HAT Esa1 (e ssential S as2-related a cetyltransferase) as a component of the NuA4 HAT complex (Allard et al.1999) acetylates H4-K5 and K8 as well as K12 (to a lesser degree) and H2A (Suka et al.2001; Torres-Machorro et al.2015). However, neither hat1Δ, esa1-L327S (ESA1 is essential) nor hat1Δ esa1-L327S double mutation caused additional growth defects with cse4-S33A or -S33E (not shown). This was surprising and suggested that the defect of the cse4-S33 mutations with H4-K5, 8, 12R was not caused by the absence of acetylation at these residues. Alternatively, a yet unknown acetyltransferase may be responsible for the acetylation that is relevant in this context, since there is residual acetylation at the lysine residues in both mutants (Suka et al.2001). We also considered the possibility that a modification at these lysines other than acetylation may be relevant for the synthetic phenotype with cse4-S33 mutations. The methyltransferase Set5 methylates lysines 5, 8 and 12 of H4 (Green et al.2012), and we therefore tested for synthetic growth defects of set5Δ with cse4-S33 mutations. However, no defects were observed (data not shown), suggesting that the defects of H4-K5,8,12R were not caused by the absence of Set5-dependent lysine methylation in the H4 N-terminus. cse4-S33A caused reduced Cse4 levels at the centromere in H4-K5, 8, 12R Since the absence of H4 acetylation by Hat1 has previously been linked to reduced deposition of CENP-A (Boltengagen et al.2016; Shang et al.2016), we asked whether the H4-K5, 8, 12R mutation caused a reduction in the amount of mutant Cse4 protein present at the centromere, which might explain why the cells have a mitotic defect. Interestingly, ChIP showed mildly reduced levels of Cse4-S33A, though not in Cse4-S33E, at the centromere in H4-K5,8,12R strains (Fig. 4). This reduction in Cse4 level was not present in strains with wt H4 (data not shown). This indicated that the absence of S33 phosphorylation caused a mild defect in the deposition of Cse4 at the centromere when the H4 N-terminus was mutated. The observation that the defect is only seen in cse4-S33A agrees with the fact that cse4-S33A has a stronger growth defect than cse4-S33E with the histone mutation (Figs 1 and 2). Figure 4. View largeDownload slide The absence of Cse4-S33 phosphorylation reduced levels of Cse4 at the centromere in H4-K5,8,12R strains. ChIP analysis of 3xHA-tagged Cse4 (α-HA) was performed. Cells were grown at 34°C for 3 h prior to ChIP. Values give the enrichment at CEN4 relative to the control region POL1. Error bars give SD of three independent experiments. *P < 0.05. Figure 4. View largeDownload slide The absence of Cse4-S33 phosphorylation reduced levels of Cse4 at the centromere in H4-K5,8,12R strains. ChIP analysis of 3xHA-tagged Cse4 (α-HA) was performed. Cells were grown at 34°C for 3 h prior to ChIP. Values give the enrichment at CEN4 relative to the control region POL1. Error bars give SD of three independent experiments. *P < 0.05. Synthetic defects of cse4-S33 with the mutations H2A-E57A and -L66A We next asked whether mutations in the canonical histones H2A and H2B can be identified that cause defects with cse4-S33A or -S33E. To test this, a genetic screen was conducted to combine the cse4-S33 alleles with a library of 225 histone alleles in which the amino acids of histones H2A and H2B were systematically replaced by alanine (Nakanishi et al.2009). This revealed synthetic growth defects for H2A-E57A and -L66A with the cse4-S33 alleles. As above, we evaluated the defects in a defined strain background using plasmid shuffle. In the presence of wt CSE4, both H2A-E57A and -L66A caused a growth defect on their own as compared to strains with wt H2A (Fig. 5). Significantly, this defect was exacerbated by cse4-S33E. Interestingly, cse4-S33A did not cause a defect with H2A-L66A, and it partially suppressed the defect of H2A-E57A. Since cse4-S33E mimics the constantly phosphorylated form of Cse4, this suggests that the defect of H2A-E57A is exacerbated by Cse4-S33ph and partially suppressed by its absence (in cse4-S33A). Figure 5. View largeDownload slide Mimicking the constantly phosphorylated state of Cse4-S33 (cse4-S33E) caused synthetic growth defects with the mutations E57A and L66A of histone H2A. Strains with the indicated genotypes were serially diluted, spotted on YPD and grown at the respective temperatures for 3 days. Figure 5. View largeDownload slide Mimicking the constantly phosphorylated state of Cse4-S33 (cse4-S33E) caused synthetic growth defects with the mutations E57A and L66A of histone H2A. Strains with the indicated genotypes were serially diluted, spotted on YPD and grown at the respective temperatures for 3 days. The identification of these H2A residues as being relevant in the context of Cse4-S33 phosphorylation is interesting in light of the fact that both H2A mutations cause a reduced interaction with the H2A/H2B histone chaperone FACT (facilitates chromatin transcription) and a defect of H2A/H2B deposition (Hodges, Gloss and Wyrick 2017). This supports the interpretation that Cse4–33 phosphorylation regulates its deposition in chromatin, and that the combined effect of reduced H2A and Cse4 at the centromere impair centromere function (see Discussion). DISCUSSION Post-translational modifications on histones and histone variants are crucial for regulating their role within chromatin. Here, we characterized the function of phosphorylation of serine 33 on the centromeric H3 variant Cse4 from S. cerevisiae. Importantly, we found that mutations that abrogate Cse4-S33ph caused synthetic growth defects when combined with mutations in histone H4-K5, 8, 12R, which was accompanied by decreased levels of Cse4 at the centromere. Furthermore, we observed synthetic growth defects with H2A-E57A and -L66A, which cause reduced H2A deposition in chromatin. Altogether, our data highlight the important role of histone deposition in building a functional centromeric nucleosome and indicates that Cse4-S33ph regulates its deposition in chromatin. Our findings of centromeric defects and a reduced deposition of S33-mutated Cse4 at the centromere when the H4 tail was mutated reflects findings in chicken and human cells that H4 and K12 acetylation by the Hat1 complex are required for CENP-A/H4 deposition (Shang et al.2016), but also show some differences to the observations in higher eukaryotes. Whereas the absence of H4 acetylation alone was sufficient to reduce CENP-A incorporation at the centromere in higher eukaryotes (Boltengagen et al.2016; Shang et al.2016), we only observed a defect in conjunction with the absence of Cse4-S33ph in yeast. Thus, Cse4 deposition apparently is less sensitive to H4 acetylation defects in yeast than CENP-A in higher eukaryotes. Furthermore, we did not observe a defect of Cse4-S33 mutations together with the absence of Hat1, and the defect was only seen when not only the Hat1 targets H4 K5 and K12 were mutated, but also K8. Thus, as above, this indicates that Cse4 is less sensitive to defects in H4 acetylation than CENP-A from higher eukaryotes for deposition at centromeric sequences. Notably, while the HAT Esa1, the catalytic entity of the NuA4 HAT complex (Allard et al.1999), acetylates all three respective lysine residues in H4 (Suka et al.2001), the mutation of Esa1 did not cause a defect with the absence of Cse4-S33ph. Therefore, a yet unknown HAT or combination of HATs is responsible for the H4 acetylation in the context of Cse4/H4 deposition. A role for H4 acetylation sites in Cse4 deposition furthermore raised the question whether the respective H4 residues are acetylated on the centromeric, Cse4-containing nucleosome. However, our query of existing genome-wide data for these modifications in S. cerevisiae (Weiner et al.2015) revealed that H4 K5Ac, K8Ac and K12Ac are undetectable at the centromeric nucleosome while being present on the flanking nucleosomes (Fig. S3, Supporting Information), which is in contrast to findings in human and chicken cells, where K5Ac and K12Ac are present on centromeric sequences (Shang et al.2016). Possibly, the modifications are removed in yeast shortly after deposition by histone deacetylases. An alternative scenario for the role of H4 acetylation in Cse4 function is that H4 K5, 8, 12R causes defects in gene expression, for instance in the expression of kinetochore components, and that this indirectly affects centromere function. Our data would imply that there is a gene (or group of genes) that is unaffected in single H4 lysine mutations and only affected in the K5, 8,12R triple mutant. However, a study of transcriptional effects of combinations of H4 lysine mutations showed that the cumulative effect of H4 K5, K8 and K12 mutation has a broad additional effect on gene expression that is mild at any single gene, and gene ontology analysis does not show an effect on chromatin or kinetochore components (Weiner et al.2015). It therefore seems unlikely that the H4 mutations show synthetic defects with Cse4-S33 due to misregulation of a centromere factor. The H4 residues highlighted here have not been described in centromere function in yeast before. Several studies have investigated effects of histone mutations in centromere function, though none report effects for N-terminal H4 mutations (Hyland et al.2005; Matsubara et al.2007; Sakamoto et al.2009; Kawashima et al.2011; Yu et al.2011). An early study isolated the first temperature-sensitive H4 allele, hhf1-20 (hhf1-T82I, A89V), which is defective in chromosome segregation (Megee et al.1990). More recently, one of the histone mutation libraries used here (Dai et al.2008) was specifically screened for centromeric defects and sensitivity to low levels of the Ilp1 kinase, and several sites in the H3 and H4 core regions were identified in this context to be important for chromosome segregation and kinetochore bi-orientation (Ng et al.2013). Some of the residues lie at the entry/exit site of DNA on the nucleosome, and it was proposed that they affect the structural properties of the centromeric nucleosome. Furthermore, H4 R36 was found to be important for the interaction of Cse4 with the E3 ubiquitin ligase Psh1 and Psh1-mediated degradation of Cse4 (Deyter et al.2017). We furthermore show here a genetic interaction between Cse4-S33 and the mutations E57A and L66A in histone H2A. These mutations lie in the α2 helix of the histone fold of H2A (Luger et al.1997) and are part of the ‘acidic patch’ on the nucleosome (Fig. S4, Supporting Information), which is important for contacts between nucleosomes, as it interacts with the H4 tail of the neighbouring nucleosome (Luger et al.1997; Kalashnikova et al.2013). Interestingly, H2A-E57A abrogates the interaction of H2A with the Spt16 subunit of the histone chaperone FACT (Hodges, Gloss and Wyrick 2017). Accordingly, gene-specific decreases in histone occupancy have been observed in both H2A-E57A and -L66A. Both residues also cause severe decreases in H2B-K123 ubiquitination as well as other PTMs (Cucinotta et al.2015). Furthermore, H2A-E57 and -L66 are part of the thiabendazole/benomyl-sensitive region II (TBS-II) of H2A. Mutations in this region cause defects in the establishment of chromosome bi-orientation, and they impair the localization of shugoshin (Sgo1) to the kinetochore (Kawashima et al.2011). Based on our conclusions above of a reduced deposition of mutant Cse4-S33 forms, we hypothesize that the two H2A mutations cause a defect in H2A/H2B occupancy at the centromeric nucleosome, which is further exacerbated by a reduced deposition of Cse4 in the cse4-S33E mutant as a result of its impaired interaction with the Scm3 chaperone. Interestingly, the defect of the H2A mutations is only enhanced by cse4-S33E, but not -S33A, suggesting that the imitation of the constantly phosphorylated state of Cse4 causes a stronger defect. How does S33 phosphorylation affect Cse4 function? The discovery of a synthetic defect with Cse4/H4 deposition suggests that the S33 phosphorylation likewise affects Cse4 deposition. In analogy to H4 acetylation, we postulate that the mutation of S33 reduces the interaction of Cse4/H4 with the histone chaperone Scm3 (the yeast homolog of HJURP). Our data indicate that the mutation of Cse4-S33 on its own causes a mild defect in Cse4 deposition that alone is insufficient to result in a growth defect, but which is enhanced when the interactions of both H4 and Cse4 to Scm3 are disturbed. This view parallels studies of CENP-A that demonstrate an important role for Cdk1-dependent phosphorylation serine 68 of the CATD of CENP-A in its interaction with HJURP and its deposition in chromatin (Yu et al.2015; Fachinetti et al.2017; Wang et al.2017). The respective structure of Scm3/Cse4/H4 reveals a similar involvement of the Cse4 CATD in the interaction with Scm3 (Cho and Harrison 2011; Zhou et al.2011). Our data indicate that, in addition to the CATD domain of Cse4, the Cse4 N-terminus, including S33 phosphorylation, mediates interaction with Scm3 for Cse4/H4 deposition. It therefore will be of interest to determine the molecular details of the interactions of the full-length Cse4 and Scm3 proteins in order to dissect the function of S33 phosphorylation. The fact that we identify circumstances here in which Cse4 levels at the centromere are reduced furthermore sheds light on the question of the composition of the centromeric nucleosome and the distribution of Cse4 at the centromere. It has variably been argued that the nucleosome contains one or two Cse4 molecules, depending on the phase of the cell cycle (Shivaraju et al.2012; Wisniewski et al.2014; Dhatchinamoorthy et al.2017). Also, microscopic studies show a broader radial distribution of Cse4 at the kinetochore compared to the Ndc80 complex, and this distribution is altered in selected mutants (Haase et al.2013). Our observations, together with the latter study, show that Cse4 levels can be reduced at the centromere, indicating that more than one Cse4 molecule is localized there for most of the cell cycle. Our analysis of Cse4-S33 phenotypes allows important insights into the role of this residue in centromere function. An earlier study identified S33 and additionally S22, S40 and S105 of Cse4 to be phosphorylated (Boeckmann et al.2013). An anti-phospho-Cse4 antibody was raised and revealed increased Cse4 phosphorylation upon nocodazole treatment, though the antibody did not distinguish between phosphorylation sites. Interestingly, all sites except S33 were phosphorylated in vitro by Ipl1, and overexpression of a phosphomimetic version of CSE4 including all phosphorylated sites (cse4-4SD) partially suppressed the temperature sensitivity of the ipl1-2 allele. Also, the cse4-4SD allele caused synthetic growth defects with mutations in OKP1 and AME1. Given our observations on S33, it seems likely that these effects reflect functions of the phosphorylation sites S22, S40 and S105, but not of S33. Which kinase is responsible for Cse4-S33ph? The earlier study excluded Ipl1/Aurora B as the S33 kinase (Boeckmann et al.2013). Obvious candidates are CDKs, in analogy to CENP-A phosphorylation by Cdk1 (Yu et al.2015; Fachinetti et al.2017), which is interesting in the light of the fact that S33 is predicted to be phosphorylated by Cdc2/Cdc28 (NetPhos 3.1) (Blom, Gammeltoft and Brunak 1999). Another possibility is the kinase Mps1, which monitors kinetochore—microtubule attachments and phosphorylates multiple targets including kinetochore components, checkpoint proteins and components of the spindle pole body in a sequential phosphorylation cascade to elicit a checkpoint response (Ji et al.2017). In summary, our genetic analyses of the Cse4-S33ph site in combination with mutations in the canonical histones have pinpointed selected residues in the H4 N-terminus and the H2A acidic patch as sites that cause synthetic growth defects. Interestingly, both regions are involved in the interaction of the histones with their respective chaperones Scm3 and FACT for deposition into chromatin. We therefore suggest that Cse4-S33ph likewise regulates its interaction with the Scm3 chaperone and that a reduced interaction reduces its deposition at centromeric sequences, which disturbs proper centromere function and chromosome segregation. SUPPLEMENTARY DATA Supplementary data are available at FEMSYR online. Acknowledgements We thank Ali Shilatifard for reagents (collection of histone mutations) and Josta Hamann and Silke Steinborn for technical support. FUNDING This work was supported by the Deutsche Forschungsgemeinschaft (EH237/10-1 and 12-1) and Humboldt-Universität zu Berlin. Conflict of interest. None declare. REFERENCES Allard S, Utley RT, Savard J et al.   NuA4, an essential transcription adaptor/histone H4 acetyltransferase complex containing Esa1p and the ATM-related cofactor Tra1p. EMBO J  1999; 18: 5108– 19. Google Scholar CrossRef Search ADS PubMed  Biggins S, Murray AW. The budding yeast protein kinase Ipl1/Aurora allows the absence of tension to activate the spindle checkpoint. Genes Dev  2001; 15: 3118– 29. Google Scholar CrossRef Search ADS PubMed  Black BE, Cleveland DW. Epigenetic centromere propagation and the nature of CENP-a nucleosomes. Cell  2011; 144: 471– 9. Google Scholar CrossRef Search ADS PubMed  Black BE, Foltz DR, Chakravarthy S et al.   Structural determinants for generating centromeric chromatin. Nature  2004; 430: 578– 82. Google Scholar CrossRef Search ADS PubMed  Blom N, Gammeltoft S, Brunak S. Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol  1999; 294: 1351– 62. Google Scholar CrossRef Search ADS PubMed  Boeckmann L, Takahashi Y, Au WC et al.   Phosphorylation of centromeric histone H3 variant regulates chromosome segregation in Saccharomyces cerevisiae. Mol Biol Cell  2013; 24: 2034– 44. Google Scholar CrossRef Search ADS PubMed  Boltengagen M, Huang A, Boltengagen A et al.   A novel role for the histone acetyltransferase Hat1 in the CENP-A/CID assembly pathway in Drosophila melanogaster. Nucleic Acids Res  2016; 44: 2145– 59. Google Scholar CrossRef Search ADS PubMed  Cai M, Davis RW. Yeast centromere binding protein CBF1, of the helix-loop-helix protein family, is required for chromosome stability and methionine prototrophy. Cell  1990; 61: 437– 46. Google Scholar CrossRef Search ADS PubMed  Camahort R, Li B, Florens L et al.   Scm3 is essential to recruit the histone h3 variant cse4 to centromeres and to maintain a functional kinetochore. Mol Cell  2007; 26: 853– 65. Google Scholar CrossRef Search ADS PubMed  Carroll CW, Milks KJ, Straight AF. Dual recognition of CENP-A nucleosomes is required for centromere assembly. J Cell Biol  2010; 189: 1143– 55. Google Scholar CrossRef Search ADS PubMed  Cheeseman IM. The kinetochore. Cold Spring Harb Perspect Biol  2014; 6: a015826. Google Scholar CrossRef Search ADS PubMed  Chen Y, Baker RE, Keith KC et al.   The N terminus of the centromere H3-like protein Cse4p performs an essential function distinct from that of the histone fold domain. Mol Cell Biol  2000; 20: 7037– 48. Google Scholar CrossRef Search ADS PubMed  Cho US, Harrison SC. Recognition of the centromere-specific histone Cse4 by the chaperone Scm3. P Natl Acad Sci USA  2011; 108: 9367– 71. Google Scholar CrossRef Search ADS   Cucinotta CE, Young AN, Klucevsek KM et al.   The nucleosome acidic patch regulates the H2B K123 monoubiquitylation cascade and transcription elongation in saccharomyces cerevisiae. PLoS Genet  2015; 11: e1005420. Google Scholar CrossRef Search ADS PubMed  Dai J, Hyland EM, Yuan DS et al.   Probing nucleosome function: a highly versatile library of synthetic histone H3 and H4 mutants. Cell  2008; 134: 1066– 78. Google Scholar CrossRef Search ADS PubMed  Davie JR, Saunders CA, Walsh JM et al.   Histone modifications in the yeast S. Cerevisiae. Nucleic Acids Res  1981; 9: 3205– 16. Google Scholar CrossRef Search ADS PubMed  Deyter GM, Hildebrand EM, Barber AD et al.   Histone H4 facilitates the proteolysis of the budding yeast CENP-ACse4 centromeric histone variant. Genetics  2017; 205: 113– 24. Google Scholar CrossRef Search ADS PubMed  Dhatchinamoorthy K, Shivaraju M, Lange JJ et al.   Structural plasticity of the living kinetochore. J Cell Biol  2017; 216: 3551– 70. Google Scholar CrossRef Search ADS PubMed  Dunleavy EM, Roche D, Tagami H et al.   HJURP is a cell-cycle-dependent maintenance and deposition factor of CENP-A at centromeres. Cell  2009; 137: 485– 97. Google Scholar CrossRef Search ADS PubMed  Earnshaw WC, Rothfield N. Identification of a family of human centromere proteins using autoimmune sera from patients with scleroderma. Chromosoma  1985; 91: 313– 21. Google Scholar CrossRef Search ADS PubMed  Fachinetti D, Logsdon GA, Abdullah A et al.   CENP-A modifications on Ser68 and Lys124 are dispensable for establishment, maintenance, and long-term function of human centromeres. Dev Cell  2017; 40: 104– 13. Google Scholar CrossRef Search ADS PubMed  Falk SJ, Lee J, Sekulic N et al.   CENP-C directs a structural transition of CENP-a nucleosomes mainly through sliding of DNA gyres. Nat Struct Mol Biol  2016; 23: 204– 8. Google Scholar CrossRef Search ADS PubMed  Foltz DR, Jansen LE, Bailey AO et al.   Centromere-specific assembly of CENP-a nucleosomes is mediated by HJURP. Cell  2009; 137: 472– 84. Google Scholar CrossRef Search ADS PubMed  Fukagawa T. Critical histone post-translational modifications for centromere function and propagation. Cell Cycle  2017; 16: 1259– 65. Google Scholar CrossRef Search ADS PubMed  Funabiki H, Wynne DJ. Making an effective switch at the kinetochore by phosphorylation and dephosphorylation. Chromosoma  2013; 122: 135– 58. Google Scholar CrossRef Search ADS PubMed  Gascoigne KE, Takeuchi K, Suzuki A et al.   Induced ectopic kinetochore assembly bypasses the requirement for CENP-A nucleosomes. Cell  2011; 145: 410– 22. Google Scholar CrossRef Search ADS PubMed  Green EM, Mas G, Young NL et al.   Methylation of H4 lysines 5, 8 and 12 by yeast Set5 calibrates chromatin stress responses. Nat Struct Mol Biol  2012; 19: 361– 3. Google Scholar CrossRef Search ADS PubMed  Haase J, Mishra PK, Stephens A et al.   A 3D map of the yeast kinetochore reveals the presence of core and accessory centromere-specific histone. Curr Biol  2013; 23: 1939– 44. Google Scholar CrossRef Search ADS PubMed  Hewawasam G, Shivaraju M, Mattingly M et al.   Psh1 is an E3 ubiquitin ligase that targets the centromeric histone variant Cse4. Mol Cell  2010; 40: 444– 54. Google Scholar CrossRef Search ADS PubMed  Hodges AJ, Gloss LM, Wyrick JJ. Residues in the nucleosome acidic patch regulate histone occupancy and are important for FACT binding in saccharomyces cerevisiae. Genetics  2017; 206: 1339– 48. Google Scholar CrossRef Search ADS PubMed  Huis In’t Veld PJ, Jeganathan S, Petrovic A et al.   Molecular basis of outer kinetochore assembly on CENP-T. eLife  2016; 5: e21007. Google Scholar PubMed  Hyland EM, Cosgrove MS, Molina H et al.   Insights into the role of histone H3 and histone H4 core modifiable residues in Saccharomyces cerevisiae. Mol Cell Biol  2005; 25: 10060– 70. Google Scholar CrossRef Search ADS PubMed  Ji Z, Gao H, Jia L et al.   A sequential multi-target Mps1 phosphorylation cascade promotes spindle checkpoint signaling. eLife  2017; 6: e22513. Google Scholar CrossRef Search ADS PubMed  Kalashnikova AA, Porter-Goff ME, Muthurajan UM et al.   The role of the nucleosome acidic patch in modulating higher order chromatin structure. J R Soc, Interface  2013; 10: 20121022. Google Scholar CrossRef Search ADS   Kawashima S, Nakabayashi Y, Matsubara K et al.   Global analysis of core histones reveals nucleosomal surfaces required for chromosome bi-orientation. EMBO J  2011; 30: 3353– 67. Google Scholar CrossRef Search ADS PubMed  Kawashima SA, Tsukahara T, Langegger M et al.   Shugoshin enables tension-generating attachment of kinetochores by loading Aurora to centromeres. Genes Dev  2007; 21: 420– 35. Google Scholar CrossRef Search ADS PubMed  Kleff S, Andrulis ED, Anderson CW et al.   Identification of a gene encoding a yeast histone H4 acetyltransferase. J Biol Chem  1995; 270: 24674– 7. Google Scholar CrossRef Search ADS PubMed  Krenn V, Overlack K, Primorac I et al.   KI motifs of human Knl1 enhance assembly of comprehensive spindle checkpoint complexes around MELT repeats. Curr Biol  2014; 24: 29– 39. Google Scholar CrossRef Search ADS PubMed  Lochmann B, Ivanov D. Histone H3 localizes to the centromeric DNA in budding yeast. PLoS Genet  2012; 8: e1002739. Google Scholar CrossRef Search ADS PubMed  London N, Biggins S. Mad1 kinetochore recruitment by Mps1-mediated phosphorylation of Bub1 signals the spindle checkpoint. Genes Dev  2014; 28: 140– 52. Google Scholar CrossRef Search ADS PubMed  Longtine MS, McKenzie A3rd, Demarini DJ et al.   Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast  1998; 14: 953– 61. Google Scholar CrossRef Search ADS PubMed  Luger K, Mader AW, Richmond RK et al.   Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature  1997; 389: 251– 60. Google Scholar CrossRef Search ADS PubMed  Malvezzi F, Litos G, Schleiffer A et al.   A structural basis for kinetochore recruitment of the Ndc80 complex via two distinct centromere receptors. EMBO J  2013; 32: 409– 23. Google Scholar CrossRef Search ADS PubMed  Matsubara K, Sano N, Umehara T et al.   Global analysis of functional surfaces of core histones with comprehensive point mutants. Genes Cells  2007; 12: 13– 33. Google Scholar CrossRef Search ADS PubMed  McKinley KL, Cheeseman IM. The molecular basis for centromere identity and function. Nat Rev Mol Cell Biol  2016; 17: 16– 29. Google Scholar CrossRef Search ADS PubMed  Megee PC, Morgan BA, Mittman BA et al.   Genetic analysis of histone H4: essential role of lysines subject to reversible acetylation. Science  1990; 247: 841– 5. Google Scholar CrossRef Search ADS PubMed  Meluh PB, Yang P, Glowczewski L et al.   Cse4p is a component of the core centromere of Saccharomyces cerevisiae. Cell  1998; 94: 607– 13. Google Scholar CrossRef Search ADS PubMed  Moyle MW, Kim T, Hattersley N et al.   A Bub1-Mad1 interaction targets the Mad1-Mad2 complex to unattached kinetochores to initiate the spindle checkpoint. J Cell Biol  2014; 204: 647– 57. Google Scholar CrossRef Search ADS PubMed  Nakajima Y, Tyers RG, Wong CC et al.   Nbl1p: a Borealin/Dasra/CSC-1-like protein essential for Aurora/Ipl1 complex function and integrity in Saccharomyces cerevisiae. Mol Biol Cell  2009; 20: 1772– 84. Google Scholar CrossRef Search ADS PubMed  Nakanishi S, Lee JS, Gardner KE et al.   Histone H2BK123 monoubiquitination is the critical determinant for H3K4 and H3K79 trimethylation by COMPASS and Dot1. J Cell Biol  2009; 186: 371– 7. Google Scholar CrossRef Search ADS PubMed  Nelson DA. Histone acetylation in baker's yeast. Maintenance of the hyperacetylated configuration in log phase protoplasts. J Biol Chem  1982; 257: 1565– 8. Google Scholar PubMed  Ng TM, Lenstra TL, Duggan N et al.   Kinetochore function and chromosome segregation rely on critical residues in histones H3 and H4 in budding yeast. Genetics  2013; 195: 795– 807. Google Scholar CrossRef Search ADS PubMed  Ortiz J, Stemmann O, Rank S et al.   A putative protein complex consisting of Ctf19, Mcm21, and Okp1 represents a missing link in the budding yeast kinetochore. Genes Dev  1999; 13: 1140– 55. Google Scholar CrossRef Search ADS PubMed  Palmer DK, O’Day K, Wener MH et al.   A 17-kD centromere protein (CENP-A) copurifies with nucleosome core particles and with histones. J Cell Biol  1987; 104: 805– 15. Google Scholar CrossRef Search ADS PubMed  Parthun MR, Widom J, Gottschling DE. The major cytoplasmic histone acetyltransferase in yeast - links to chromatin replication and histone metabolism. Cell  1996; 87: 85– 94. Google Scholar CrossRef Search ADS PubMed  Pesenti ME, Weir JR, Musacchio A. Progress in the structural and functional characterization of kinetochores. Curr Opin Struct Biol  2016; 37: 152– 63. Google Scholar CrossRef Search ADS PubMed  Ranjitkar P, Press MO, Yi X et al.   An E3 ubiquitin ligase prevents ectopic localization of the centromeric histone H3 variant via the centromere targeting domain. Mol Cell  2010; 40: 455– 64. Google Scholar CrossRef Search ADS PubMed  Ruchaud S, Carmena M, Earnshaw WC. Chromosomal passengers: conducting cell division. Nat Rev Mol Cell Biol  2007; 8: 798– 812. Google Scholar CrossRef Search ADS PubMed  Sakamoto M, Noguchi S, Kawashima S et al.   Global analysis of mutual interaction surfaces of nucleosomes with comprehensive point mutants. Genes Cells  2009; 14: 1271– 330. Google Scholar CrossRef Search ADS PubMed  Samel A, Cuomo A, Bonaldi T et al.   Methylation of CenH3 arginine 37 regulates kinetochore integrity and chromosome segregation. Proc Natl Acad Sci U S A  2012; 109: 9029– 34. Google Scholar CrossRef Search ADS PubMed  Sandall S, Severin F, McLeod IX et al.   A Bir1-Sli15 complex connects centromeres to microtubules and is required to sense kinetochore tension. Cell  2006; 127: 1179– 91. Google Scholar CrossRef Search ADS PubMed  Shang WH, Hori T, Westhorpe FG et al.   Acetylation of histone H4 lysine 5 and 12 is required for CENP-A deposition into centromeres. Nature Commun  2016; 7: 13465. Google Scholar CrossRef Search ADS   Sherman F. Getting started with yeast. Methods Enzymol  1991; 194: 3– 21. Google Scholar CrossRef Search ADS PubMed  Shivaraju M, Unruh JR, Slaughter BD et al.   Cell-cycle-coupled structural oscillation of centromeric nucleosomes in yeast. Cell  2012; 150: 304– 16. Google Scholar CrossRef Search ADS PubMed  Shuaib M, Ouararhni K, Dimitrov S et al.   HJURP binds CENP-A via a highly conserved N-terminal domain and mediates its deposition at centromeres. Proc Natl Acad Sci USA  2010; 107: 1349– 54. Google Scholar CrossRef Search ADS PubMed  Stoler S, Keith KC, Curnick KE et al.   A mutation in CSE4, an essential gene encoding a novel chromatin-associated protein in yeast, causes chromosome nondisjunction and cell cycle arrest at mitosis. Genes Dev  1995; 9: 573– 86. Google Scholar CrossRef Search ADS PubMed  Stoler S, Rogers K, Weitze S et al.   Scm3, an essential Saccharomyces cerevisiae centromere protein required for G2/M progression and Cse4 localization. Proc Natl Acad Sci USA  2007; 104: 10571– 6. Google Scholar CrossRef Search ADS PubMed  Suka N, Suka Y, Carmen AA et al.   Highly specific antibodies determine histone acetylation site usage in yeast heterochromatin and euchromatin. Mol Cell  2001; 8: 473– 9. Google Scholar CrossRef Search ADS PubMed  Tanaka TU, Rachidi N, Janke C et al.   Evidence that the Ipl1-Sli15 (Aurora kinase-INCENP) complex promotes chromosome bi-orientation by altering kinetochore-spindle pole connections. Cell  2002; 108: 317– 29. Google Scholar CrossRef Search ADS PubMed  Tong AH, Evangelista M, Parsons AB et al.   Systematic genetic analysis with ordered arrays of yeast deletion mutants. Science  2001; 294: 2364– 8. Google Scholar CrossRef Search ADS PubMed  Torres-Machorro AL, Clark LG, Chang CS et al.   The Set3 complex antagonizes the MYST Acetyltransferase Esa1 in the DNA damage response. Mol Cell Biol  2015; 35: 3714– 25. Google Scholar CrossRef Search ADS PubMed  Wang K, Yu Z, Liu Y et al.   Ser68 phosphorylation ensures accurate cell-cycle-dependent CENP-A deposition at centromeres. Develop Cell  2017; 40: 5– 6. Google Scholar CrossRef Search ADS   Weber JM, Irlbacher H, Ehrenhofer-Murray AE. Control of replication initiation by the Sum1/Rfm1/Hst1 histone deacetylase. BMC Mol Biol  2008; 9: 100. Google Scholar CrossRef Search ADS PubMed  Weiner A, Hsieh TH, Appleboim A et al.   High-resolution chromatin dynamics during a yeast stress response. Mol Cell  2015; 58: 371– 86. Google Scholar CrossRef Search ADS PubMed  Welburn JP, Vleugel M, Liu D et al.   Aurora B phosphorylates spatially distinct targets to differentially regulate the kinetochore-microtubule interface. Mol Cell  2010; 38: 383– 92. Google Scholar CrossRef Search ADS PubMed  Wisniewski J, Hajj B, Chen J et al.   Imaging the fate of histone Cse4 reveals de novo replacement in S phase and subsequent stable residence at centromeres. eLife  2014; 3: e02203. Google Scholar CrossRef Search ADS PubMed  Yamagishi Y, Yang CH, Tanno Y et al.   MPS1/Mph1 phosphorylates the kinetochore protein KNL1/Spc7 to recruit SAC components. Nat Cell Biol  2012; 14: 746– 52. Google Scholar CrossRef Search ADS PubMed  Yu Y, Srinivasan M, Nakanishi S et al.   A conserved patch near the C terminus of histone H4 is required for genome stability in budding yeast. Mol Cell Biol  2011; 31: 2311– 25. Google Scholar CrossRef Search ADS PubMed  Yu Z, Zhou X, Wang W et al.   Dynamic phosphorylation of CENP-A at Ser68 orchestrates its cell-cycle-dependent deposition at centromeres. Develop Cell  2015; 32: 68– 81. Google Scholar CrossRef Search ADS   Zhou Z, Feng H, Zhou BR et al.   Structural basis for recognition of centromere histone variant CenH3 by the chaperone Scm3. Nature  2011; 472: 234– 7. Google Scholar CrossRef Search ADS PubMed  © FEMS 2017. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Yeast ResearchOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial