-
HIGHLIGHTED ARTICLE| INVESTIGATION
Cohesin Impedes Heterochromatin Assembly in FissionYeast Cells
Lacking Pds5
H. Diego Folco, Andrea McCue,1 Vanivilasini Balachandran, and
Shiv I. S. Grewal2
Laboratory of Biochemistry and Molecular Biology, National
Cancer Institute, National Institutes of Health, Bethesda,Maryland
20892
ABSTRACT The fission yeast Schizosaccharomyces pombe is a
powerful genetic model system for uncovering fundamental
principlesof heterochromatin assembly and epigenetic inheritance of
chromatin states. Heterochromatin defined by histone H3 lysine 9
meth-ylation and HP1 proteins coats large chromosomal domains at
centromeres, telomeres, and the mating-type (mat) locus.
Althoughgenetic and biochemical studies have provided valuable
insights into heterochromatin assembly, many key mechanistic
details remainunclear. Here, we use a sensitized reporter system at
themat locus to screen for factors affecting heterochromatic
silencing. In additionto known components of heterochromatin
assembly pathways, our screen identified eight new factors
including the cohesin-associatedprotein Pds5. We find that Pds5
enriched throughout heterochromatin domains is required for proper
maintenance of heterochromatin.This function of Pds5 requires its
associated Eso1 acetyltransferase, which is implicated in the
acetylation of cohesin. Indeed, introducingan acetylation-mimicking
mutation in a cohesin subunit suppresses defects in heterochromatin
assembly in pds5D and eso1D cells. Ourresults show that in cells
lacking Pds5, cohesin interferes with heterochromatin assembly.
Supporting this, eliminating cohesin from themat locus in the pds5D
mutant restores both heterochromatin assembly and gene silencing.
These analyses highlight an unexpectedrequirement for Pds5 in
ensuring proper coordination between cohesin and heterochromatin
factors to effectively maintain genesilencing.
KEYWORDS heterochromatin; fission yeast; mat locus; Pds5;
cohesin acetylation
THE eukaryotic genome is regulated by epigenetic modifi-cations
that create structurally distinct “open” euchroma-tin and “closed”
heterochromatin domains. Heterochromatinis defined by
hypoacetylation of histones and methylation ofhistone H3 at lysine
9 (H3K9me), and is implicated in diversefunctions including
transcriptional and post-transcriptionalsilencing, as well as the
maintenance of genome stability(Grewal and Jia 2007; Wang et al.
2016; Freitag 2017;Allshire andMadhani 2018). Understanding the
mechanismsof heterochromatin assembly is vital to elucidate the
causesof human diseases linked to defects in this process.
The fission yeast Schizosaccharomyces pombe is an idealmodel
genetic organism for studying heterochromatin assem-bly pathways.
The S. pombe genome contains facultative het-erochromatin islands
as well as constitutive heterochromatindomains coating centromeres,
telomeres, and the silentmating-type (mat) region (Cam et al.
2005). At centromeres, arrays ofrepetitive dg and dh elements
embedded within pericentro-meric regions serve as heterochromatin
nucleation centers,whereas at the mat locus a cenH element bearing
homologyto dg and dh repeats serves to nucleate
heterochromatin(Grewal and Jia 2007). The hallmark heterochromatin
mod-ification H3K9me is added by the sole methyltransferase
Clr4(SUV39H in mammals) (Nakayama et al. 2001b), which ex-ists in
the multisubunit H3K9 methyltransferase (CLRC) pro-tein complex
(Hong et al. 2005; Horn et al. 2005; Jia et al.2005). Once added,
H3K9me spreads across extended chro-mosomal domains surrounded by
boundary DNA elements(Hall et al. 2002). Spreading requires a
unique feature ofClr4 to both “read” and “write” H3K9me (Nakayama
et al.2001b; Zhang et al. 2008; Al-Sady et al. 2013).
Clr4-mediatedH3K9me deposition also provides binding sites for
the
Copyright © 2019 by the Genetics Society of Americadoi:
https://doi.org/10.1534/genetics.119.302256Manuscript received
April 26, 2019; accepted for publication June 24, 2019;
publishedEarly Online July 5, 2019.Supplemental material available
at Figshare: https://doi.org/10.25386/genetics.8307248.1Present
address: Department of Molecular Genetics, The Ohio State
University,Columbus, OH 43210.
2Corresponding author: Laboratory of Biochemistry and Molecular
Biology, NationalCancer Institute, National Institutes of Health,
Bethesda, MD 20892. E-mail: [email protected]
Genetics, Vol. 213, 127–141 September 2019 127
https://doi.org/10.1534/genetics.119.302256https://doi.org/10.25386/genetics.8307248https://doi.org/10.25386/genetics.8307248mailto:[email protected]:[email protected]
-
chromodomains of HP1 (Heterochromatin Protein 1) familyproteins
such as Chp2 and Swi6 (Thon and Verhein-Hansen2000; Bannister et
al. 2001; Nakayama et al. 2001b; Sadaieet al. 2004), which in turn
create a platform for the recruit-ment of effectors such as SHREC
[Snf2-histone deacetylase(HDAC) repressor complex] containing the
Clr1, Clr2, Clr3,and Mit1 proteins (Thon and Klar 1992; Ekwall and
Ruusala1994; Thon et al. 1994; Sugiyama et al. 2007; Job et
al.2016). Interestingly, Swi6 also recruits the
cohesin-loadingcomplex that is critical for preferential cohesin
enrichment atheterochromatic loci (Bernard et al. 2001; Nonaka et
al.2002; Fischer et al. 2009).
Paradoxically, heterochromatin nucleation is dependentupon
transcription by RNA polymerase II. Highly conservedRNA-processing
factors, including RNA interference (RNAi)-dependent and
-independent mechanisms, target coding andlong noncodingRNAs to
assemble facultative and constitutiveheterochromatin (Reyes-Turcu
and Grewal 2012; Allshireand Madhani 2018). Transcripts produced
from repeat ele-ments at constitutive heterochromatin domains are
pro-cessed by RNAi machinery into small interfering RNAs(siRNAs).
The siRNAs bind to Argonaute (Ago1) and providespecificity for
targeting the RNAi-induced transcriptionalsilencing complex, which
facilitates recruitment of CLRC(Verdel et al. 2004; Zhang et al.
2008; Bayne et al. 2010).RNAi-independent mechanisms involving
RNA-processingand RNA polymerase II termination factors operate in
paral-lel to RNAi to mediate the assembly of constitutive
hetero-chromatin domains (Reyes-Turcu et al. 2011; Marina et
al.2013; Chalamcharla et al. 2015; Tucker et al. 2016;
Touat-Todeschini et al. 2017).
Early studies of the mat locus in fission yeast were
partic-ularly valuable in uncovering mechanisms directing the
nu-cleation, spreading, and propagation of heterochromatinthrough
mitosis and meiosis (Hall et al. 2002; Klar 2007).The mating-type
region is comprised of three cassettes:mat1,mat2, andmat3 (Figure
1A). In wild-type (WT) homo-thallic cells (h90), the mating type is
determined by the allelepresent in the activemat1 cassette, which
is eithermat1P in P(plus) or mat1M in M (minus) cells (Klar 2007).
Cells switchmating type by copying genetic information from
silentmat2ormat3 donor cassettes that harbor the P andM
information,respectively. Silencing at donor mating-type cassettes
isenforced by two mechanisms. First, heterochromatin nucle-ated at
the cenH element, which is located betweenmat2 andmat3, spreads
across a �20-kb region surrounded by bound-ary elements (referred
to as IR-L and IR-R) to repress geneexpression across the entire
domain (Grewal and Klar 1997;Noma et al. 2001; Hall et al. 2002).
Second, the cis-actingelements REII and REIII, which are located
adjacent to mat2andmat3 cassettes, respectively (Thon et al. 1994,
1999), actlocally to recruit HDACs and silence donor cassettes
indepen-dently of heterochromatin (Grewal et al. 1998; Cam et
al.2008). The redundancy in silencing mechanisms underscoresthe
fundamental importance of silencingmat2 andmat3 loci.Indeed,
impairment of both silencing pathways results in
coexpression of M and P mating-type alleles in haploid
cells,leading to an aberrant sporulation phenotype called
“haploidmeiosis” (Thon et al. 1994).
Despite significant progress, it is not fully understood
howheterochromatic structures are propagated in the context
ofothernuclearprocesses. Togain insight into thesemechanisms,we
performed a systematic screen for additional factors re-quired for
heterochromatic silencing at the silent mat region.Our screen
identified several factors with previously unde-scribed roles in
heterochromatin regulation. Among the geneswe identified was Pds5,
a conserved protein known to asso-ciate with cohesin. Our results
suggest that defective hetero-chromatin assembly in cells lacking
Pds5 is functionallyconnected to its role in promoting the
acetylation of cohesinby the Eso1 acetyltransferase. Our discovery
reveals an unex-pected and intriguing role for Pds5 and Eso1 in
preventingdeleterious effects of cohesin on heterochromatin
assembly.
Materials and Methods
Yeast strains and deletion library
Standard procedures were used for fission yeast cell cultureand
genetic manipulations (Sabatinos and Forsburg 2010).S. pombe
strains used in this study are listed in SupplementalMaterial,
Table S1. Gene deletions and epitope tagging wereperformed by
homologous recombination (Bähler et al.1998). The following mutant
alleles and reporters were pre-viously described: KD::ade6+ (Ayoub
et al. 1999), KD::ura4+
(Grewal and Klar 1996), Kint2::ura4+ (Grewal and Klar1997),
REIID mat2P::ura4+ (Thon et al. 1994), psc3-4T(Nonaka et al. 2002),
psm3KK/QQ (Kagami et al. 2011),rad21-K1 (Tatebayashi et al. 1998),
and rpl42(sP56Q )(Roguev et al. 2007). The haploid deletion library
version4.0 (Bioneer, Daejeon, Korea), which contains 3400
strains,was used to systematically screen for factors involved in
het-erochromatin formation. We obtained data for 3042 strains,which
represent �90% of all nonessential genes in fissionyeast. The
tester strain SPDF734 was created as follows.The his2 gene was
deleted with a nourseothricin resistancecassette (NAT) in a
heterothallic strain (mat1M-smt0) har-boring the REII deletion and
a ura4+ marker insertion(mat2P::ura4+). The mat locus can be
followed with theNAT marker given its genetic linkage with his2 (,
0.5 cM).To prevent diploids from forming in the first round of
screen-ing, the strain also included the rpl42(sP56Q) allele,
whichconfers resistance to cycloheximide.
Genetic screen
All manipulations were performed with a multiblot replicatorin a
96-well plate format (Figure S1A) (V&P Scientific, SanDiego,
CA). The entire deletion library, consisting of 36plates, was
crossed to SPDF734 in SPAS (sporulation agarwith supplements)
medium for$ 3 days at 26�. Spores weregerminated in YES (yeast
extract plus supplements) plus cy-cloheximide (100 mg/ml), ClonNat
(100 mg/ml), and G418
128 H. D. Folco et al.
-
(100 mg/ml) for $ 3 days at 32�. The multiple selectionstrategy
ensured that haploid progeny harbored each deletedgene in the
sensitizedmat locus background. Next, cells werereplicated onto
PMG5S (pombe minimal glutamate plus fivesupplements) plus
cycloheximide (200 mg/ml), ClonNat(200 mg/ml), and G418 (300 mg/ml)
and incubated for2 days at 26�, and then exposed to iodine vapor
for the initialidentification of candidates. Candidates were then
examinedfor haploid meiosis using a standard bright-field
microscope.False positives displaying azygotic (from diploids) or
zygoticasci (from mating) were eliminated. The verified
positivecandidates (Table S2) were tested for haploid meiosis
(i.e.,iodine staining) and expression of mat2P::ura4+ in plate
di-lution assays employing a combined system that scores fourtraits
as follows: (a) PMG5S: iodine staining as 1 = black,0.5 =
brown/variegated, and 0 = yellow; (b) PMG – URA:iodine staining
[same as for (a)]; (c) PMG + 5-flourooroticacid (FOA): growth as 1=
none, 0.5 =medium, and 0= full;and (d) PMG – URA: growth as 1 =
full, 0.5 = medium, and
0 = none (Table S3). Only candidates with a score$ 2 werecarried
forward for further analyses. Strains carrying dele-tions in the
genes of interest were genotyped by PCR, andsubjected to tetrad
dissection and random spore analyses toconfirm the phenotypes.
Finally, cells from the shortlistedcandidates were spotted on PMG5S
for 3 days at 30�. Thepercentage of cells undergoing haploid
meiosis was quanti-fied using a bright-field microscope and
candidates thatexhibited , 2% haploid meiosis were discarded.
Oligonucleotides, RT-PCR, and quantitative PCR
Oligonucleotides used for chromatin
immunoprecipitation(ChIP)-quantitative PCR (qPCR) and RT-qPCR are
listed inTable S4. RNAwas isolated using the MasterPure Yeast
RNAPurification Kit (Epicentre) according to the
manufacturer’sprotocol and random primers were used for cDNA
amplifica-tion. qPCR reactions were performed in a Quant Studio
3Real-Time PCR system (Thermo Fisher Scientific) usingSYBR Green
Supermix (Bio-Rad, Hercules, CA).
Figure 1 Genetic screen for new factors in-volved in
heterochromatin formation. (A)Schematic representation of the
mating-type(mat) region in fission yeast. (B) Depiction ofthe
genetic screen used to identify new het-erochromatin factors. The
screen exploitsthe haploid meiosis phenomenon to identifyfactors
required to maintain silencing at themat locus. A tester strain
containing the REIIdeletion and mat2P::ura4+ reporter wascrossed to
the fission yeast gene deletionlibrary in a 96-well format. Haploid
mutantcolonies carrying a silencing reporter werethen screened for
derepression of the matlocus, which leads to the expression of
bothmating-type factors and haploid meiosis, byiodine staining. See
details in the Materialsand Methods section. (C) Results from
thescreen indicating the number of candidategenes remaining after
each validation step.(D) Comparison of the iodine staining
resultsobtained for strains carrying deletions of thenewly
identified factors. The percentage ofcells undergoing haploid
meiosis is indicatedbelow. N. 200. (E) Analysis ofmat2P::ura4+
expression in strains carrying deletions of thenewly identified
factors. Ten-fold serial dilu-tions of the indicated strains were
spottedon YEA rich media, with or without the ad-dition of FOA and
grown at 32�. WT andclr4D strains serve as growth controls.
FOA,5-flouroorotic acid; KAN, kanamycin; NAT,nourseothricin; WT,
wild-type; YEA, yeast ex-tract plus adenine.
Effect of Pds5 on Heterochromatin 129
-
Chromatin immunoprecipitation
ChIP and ChIP-chip experiments were performed as
previouslydescribed (Cam et al. 2005; Zofall et al. 2016). Anti-HA
(12CA5,Roche or 16B12; BioLegend), anti-GFP (ab290; Abcam),
anti-H3K9me2 (ab1220 and ab115159; Abcam), and anti-H3K9me3(ab8898)
antibodies were used for immunoprecipitation. DNAisolated from
immunoprecipitated chromatin or fromwhole-cellextracts was labeled
with Cy5/Cy3 for microarray-based ChIP-chip analyses using a custom
4 3 44K oligonucleotide array(Agilent). Cy5/Cy3 ratios were further
processed by a seven-probe sliding window filter to reduce
noise.
Data availability
Strains are available upon request. The authors affirm that
alldata necessary for confirming the conclusions of the article
arepresentwithin the article, figures, and tables.Microarray
dataare available at the National Center for Biotechnology
In-formation Gene Expression Omnibus repository under theaccession
number GSE130233. Supplemental material avail-able at Figshare:
https://doi.org/10.25386/genetics.8307248
Results
Sensitized genetic screen for heterochromaticsilencing
factors
Genetic screens have proven valuable in identifying
factorsinvolved in heterochromatin assembly (Thon and Klar
1992;Ekwall and Ruusala 1994; Thon et al. 1994; Bayne et al.
2014;Zofall et al. 2016; Taneja et al. 2017; Jahn et al. 2018).
Touncover additional factors, we used a sensitized reporter sys-tem
to screen the S. pombe gene deletion library for factorsaffecting
silencing at the silent mat region. We employed anonswitchable
(mat1-Msmt0) tester strain carrying a deletionof the local
silencerREII, which locally recruits HDACs (Grewalet al. 1998; Cam
et al. 2008) near themat2P cassette (Figure 1,A and B) (Thon et al.
1994). In this strain, mat2P silencing isentirely dependent upon
heterochromatin formation acrossthe silent mat region. Defects in
heterochromatin lead to lossof mat2P silencing, which triggers
haploid meiosis resultingfrom coexpression ofM and P information in
otherwise haploidcells. Colonies formed bymutants defective inmat2P
silencingstain dark brown/black in the presence of iodine vapor due
tothe starch-like compound produced by cells undergoing hap-loid
meiosis. In contrast, colonies formed byWT cells that lackthis
compound stain yellow. A ura4+marker adjacent tomat2P(mat2P::ura4+)
provided an orthogonal readout for hetero-chromatic silencing, as
evidenced by growth on counterselec-tive medium containing FOA.
To study the effects of individual gene deletions on
het-erochromatic silencing, we crossed the tester strain to theS.
pombe deletion library (Figure 1B and Figure S1A). Sincethe mat
reporter in our tester strain is marked with a tightlylinked NAT
marker, and gene deletions in the library wereconstructed using a
kanamycin (KAN) marker, we used selec-tion on NAT and G418 plates
to obtain the desired segregants.
In addition, the recessive cycloheximide resistance gene
(theP56Q allele of the ribosomal protein gene rpl42+)
allowedpositive selection for haploid cells of interest while
eliminatingdiploids bearing the dominant cycloheximide-sensitive
gene(Roguev et al. 2007). Of the deletion mutants, . 100
testedpositive for iodine staining (Figure 1C). The haploid
meiosisphenotype of these candidates was further examined by
mi-croscopy. Fifty-two candidates that scored positive were
se-lected for further analyses (Table S2). We then scored thedegree
of haploidmeiosis and themat2P::ura4+ silencing phe-notype, based
on the intensity of iodine staining and growth ofmutants on medium
either lacking uracil or containing FOA.A maximum score of 4 was
assigned to mutants that showedthe highest derepression in both
assays (Table S3 andMaterials and Methods). Candidates with scores,
2 were dis-carded. This scoring strategy resulted in a final list
of 26 genesthat are required for silencing at the mat region (Table
1).
Identification of new heterochromatic silencing factors
Our screen revealed 18 previously reported factors
alongwitheight factors with no described role in heterochromatic
si-lencing (Figure 1C and Table 1). As expected, the majorityof
strains carrying deletions in known heterochromatin as-sembly
factors exhibited the highest scores for haploid mei-osis and
derepression of mat2P::ura4+ (Table S3). Theseincluded components
of CLRC (Horn et al. 2005), HP1 familyproteins Swi6 and Chp2 (Thon
and Verhein-Hansen 2000;Sadaie et al. 2004), components of SHREC
(Sugiyama et al.2007; Job et al. 2016), and chromatin remodelers
and his-tone chaperones such as Fft3, Hip1, and Ani1 (Blackwell et
al.2004; Strålfors et al. 2011; Yamane et al. 2011; Taneja et
al.2017; Jahn et al. 2018; Tan et al. 2018). The new
genesidentified encoded unnamed products, ribosomal proteins,Yap18,
the proteasome regulatory subunit Rpn10 (Wilkinsonet al. 2000), and
the cohesin-associated protein Pds5(Hartman et al. 2000; Panizza et
al. 2000; Sumara et al.2000; Tanaka et al. 2001; Wang et al. 2002).
We namedthe unannotated SPCP1E11.10 and SPAC17C9.15c genesdhm1 and
dhm2 (deleterious haploid meiosis), respectively.dhm1 encodes
an�23-kDaprotein containing ankyrin repeats,whereas dhm2 encodes an
S. pombe-specific small protein(�11 kDa) that lacks annotated
domains. Both of these pro-teins exhibit a nuclear localization
(Matsuyama et al. 2006).
We independently generateddeletionsof eachof thenewlyidentified
genes. The deletion mutants showed haploid mei-osis that closely
correlatedwithderepressionofmat2P::ura4+
(Figure 1, D and E). In particular, cells carrying a deletion
ofpds5 displayed a high level of haploid meiosis and alleviationof
mat2P::ura4+ reporter gene silencing. Since Pds5 is asso-ciated
with the cohesin protein complex (Tanaka et al. 2001;Wang et al.
2002; Schmidt et al. 2009), which is enriched atheterochromatic
loci (Bernard et al. 2001; Nonaka et al.2002; Fischer et al. 2009),
we focused our further effortson understanding the possible
function(s) of this conservedprotein in heterochromatin
assembly.
130 H. D. Folco et al.
https://doi.org/10.25386/genetics.8307248
-
Loss of Pds5 causes variegated expression and
defectiveheterochromatin assembly
Interestingly, in contrast to the uniform staining patterns of
WT(yellow) and clr4D cells (black), pds5D cells formed coloniesthat
were differentially stained to yield a mixture of yellow andblack
patches (Figure 2A). A similar phenotype was also dis-played by
several other mutants identified in our screen (FigureS1B and Table
S2). This variegated staining pattern is a char-acteristic of
mutants that are known to be defective in themain-tenance of
heterochromatin and that show a reduction, but notloss, of H3K9me
levels (Taneja et al. 2017). Indeed, ChIP anal-yses of H3K9 di- and
trimethylation (H3K9me2/3) showed areduction in heterochromatic
H3K9 marks at or near mat2P inpds5D (Figure 2B). The decrease
correlated with higher expres-sion levels of mat2-Pc and the ura4+
reporter compared to WT(Figure 2C). Thus, the loss of silencing and
triggering of haploidmeiosis that are observed in cells lacking
Pds5 are functionallyconnected to a failure in the proper assembly
of heterochroma-tin at the mat locus.
Pds5 is broadly required for heterochromatin assembly
We next wondered if Pds5 affects H3K9me at other
hetero-chromatin domains. Genome-wide analyses of WT and pds5D
cells revealed that H3K9me2 was notably decreased at
cen-tromeres and telomeres in pds5D (Figure 3, A and B). Thisresult
is consistent with ChIP-qPCR analysis showing a re-duction, but not
a complete loss, of both H3K9me2 andH3K9me3 at centromeric repeat
elements (Figure 3C). To-gether, our results suggest that the
function of Pds5 in het-erochromatin assembly is not limited to
themat locus, and infact that Pds5 plays a broader role in
heterochromatin forma-tion at other loci.
Pds5 is required for heterochromatin maintenance
Ourpreviousstudieshaveshownthatheterochromatincanself-propagate
via an epigenetic mechanism, wherein parentalH3K9 methylated
nucleosomes recruit Clr4 to stably maintainthe heterochromatic
state in cis both during mitosis and mei-osis (Grewal and Klar
1996; Nakayama et al. 2000; Hall et al.2002; Zhang et al. 2008).
The crucial requirement for theepigenetic propagation of
heterochromatin is unveiled in theabsence of de novo nucleation
mechanisms. Indeed, the reten-tion of parental methylated
nucleosomes, and the ability ofClr4 to read and write H3K9me, are
essential for clonal prop-agation of heterochromatin in cells
lacking the cenH nucleationsite at the mat locus (Zhang et al.
2008; Aygün et al. 2013).
Table 1 List of genes identified in this study
Gene ID Name Description Synonyms Human ortholog
Chromatin remodelers & chaperonesSPBC1347.02 ani1 CENP-A
N-terminus domain isomerase Ani1 fkbp39SPAC25A8.01c fft3 SMARCAD1
family ATP-dependent DNA helicase Fft3 snf2SR SMARCAD1SPBC31F10.13c
hip1 Histone H3.3 H4 chaperone, hira family Hip1 hir1
HIRASPBC609.05 pob3 Histone H2A-H2B chaperone, FACT complex subunit
Pob3 SSRP1SPBC15D4.03 slm9 Histone H3.3 H4 chaperone, hira family
Slm9 HIRA
Heterochromatin proteinsSPBC16C6.10 chp2 Heterochromatin (HP1)
family chromodomain protein Chp2 CBX1, CBX3, CBX5SPAC664.01c swi6
Heterochromatin (HP1) family chromodomain protein Swi6 SPAC824.10c
CBX1, CBX3, CBX5
Histone deacetylasesSPBC2D10.17 clr1 SHREC complex intermodule
linker subunit Clr1SPAC1B3.17 clr2 Chromatin silencing protein
Clr2SPBC800.03 clr3 Histone deacetylase (class II) Clr3 HDAC6,
HDAC10SPBP35G2.10 mit1 SHREC complex ATP-dependent DNA helicase
subunit Mit1 CHD3
H3K9 methyltransferase complexSPBC428.08c clr4 Histone H3
methyltransferase Clr4 SUV39H1, SUV39H2SPCC613.12c raf1 CLRC
ubiquitin ligase complex WD repeat subunit Raf1/Dos1 dos1, cmc1,
clr8SPCC970.07c raf2 CLRC ubiquitin ligase complex subunit Raf2
dos2, cmc2, clr7SPCC11E10.08 rik1 CLRC ubiquitin ligase complex WD
repeat protein Rik1 DDB1
OthersSPBC428.07 meu6 Pleckstrin homology domain protein
Meu6SPAC694.06c mrc1 Claspin, Mrc1 CLSPNSPAC110.02 pds5 Mitotic and
meiotic cohesin loader subunit Pds5 PDS5A, PDS5BSPCC338.16 pof3
F-box protein Pof3 STIP1SPAC637.10c rpn10 19S proteasome regulatory
subunit Rpn10 pus1 PSMD4SPBC19F8.03c yap18 ENTH/VHS domain protein
(predicted) SNAP91, PICALM
Ribosomal proteinsSPCC663.04 rpl39 60S ribosomal protein L39
RPL39SPBC11C11.09c rpl502 60S ribosomal protein L5 rpl5-2, rpl5b
RPL5SPAC959.08 rpl2102 60S ribosomal protein L21 (predicted)
rpl21-2, rpl21 RPL21
UnannotatedSPCP1E11.10 dhm1 Ankyrin repeat protein, unknown
biological roleSPAC17C9.15c dhm2 Schizosaccharomyces-specific
protein
Newly identified factors required for heterochromatic silencing
are highlighted in bold. CLRC, multisubunit H3K9 methyltransferase
protein complex; ENTH, epsin N-terminalhomology; FACT, facilitates
transcription activation; ID, identifier; SHREC, Snf2-histone
deacetylase repressor complex; VHS, Vps27p, Hrs and STAM;WD,
tryptophan-aspartic acid.
Effect of Pds5 on Heterochromatin 131
-
The variegated expression pattern of pds5D was charac-teristic
of mutants defective in epigenetic inheritance ofheterochromatin
(Figure 2A) (Taneja et al. 2017). To de-termine if Pds5 affects
epigenetic stability, we testedwhether the heterochromatic “OFF”
state in cells carryinga replacement of cenHwith ade6+ (KD::ade6+)
(Ayoub et al.1999) could be maintained in the absence of Pds5.
Since theepigenetically maintained OFF state is stably
propagatedthrough meiosis (Grewal and Klar 1996; Ayoub et al.1999),
we introduced pds5D by a genetic cross. Tetrad anal-yses of meiotic
segregants showed that the KD::ade6+ OFFstate was maintained in WT
progeny, as expected (Figure 4A).In contrast, maintenance of
KD::ade6+ was severely af-fected in pds5D segregants, as indicated
by growth on me-dium lacking adenine (Figure 4A). Loss of
KD::ade6+
silencing in the pds5D background was confirmed by
replicaplating onto low-adenine medium plates. All KD::ade6+
pds5D segregants formed white colonies indicating loss ofade6+
silencing (Figure 4A). Defective silencing correlatedwith a
reduction in H3K9me3 levels at the mat region (Fig-ure 4B). A
similar loss of silencing occurred in pds5D cellscarrying a
KD::ura4+ reporter (Figure S2A), which also cor-related with
reduced levels of H3K9me3 (Figure S2B). Im-portantly, cells
carrying a Kint2::ura4+ reporter, in whichthe cenH nucleation
region is unaltered, did not show de-fects in silencing (Figure
S2A). These results indicate thatPds5 is required for the
maintenance of heterochromatin atthe mat locus.
Pds5 and RNAi play distinct roles inheterochromatic
silencing
RNAi machinery plays an important role in the nucleation
ofheterochromatin and gene silencing (Hall et al. 2002; Volpeet al.
2002; Verdel et al. 2004). To understand the relation-ship between
Pds5 and the RNAi machinery, we deleted Pds5in a strain lacking the
RNAi factor Ago1 and looked for effectson silencing of the
Kint2::ura4+ reporter inserted at cenH.Compared to the single ago1D
or pds5D deletionmutants, theago1D pds5D double mutant showed
severe loss-of-silencingof Kint2::ura4+ (Figure S2C). Moreover,
double mutantsshowed a cumulative increase in transcripts
originating fromcenH (Figure S2D). A similar result was obtained
when weanalyzed the expression of centromeric repeats in single
anddouble mutants. Whereas loss of Pds5 alone causedonly minor
changes in the expression of dg centromeric re-peats, the
combination of pds5D with ago1D resulted in acumulative increase in
dg transcripts that was comparableto that of clr4D cells (Figure
S3). These results are consistentwith Pds5 and RNAi performing
distinct roles in promotingsilencing at centromeres and the mat
locus.
Localization of Pds5 across heterochromatin domainsrequires
Swi6
We next asked whether Pds5 acts directly to facilitate
hetero-chromatic silencing. Pds5associateswith the cohesin
complex(Tanaka et al. 2001; Wang et al. 2002; Schmidt et al.
2009)and localizes to centromeres in a manner dependent on
Figure 2 Pds5 is required for heterochroma-tin formation at the
mat locus. (A) Iodinestaining of single colonies. The
indicatedstrains were grown in YEA, replica plated ontoPMG5S, and
grown at 32� for 3 days prior tostaining. (B) H3K9me2 and H3K9me3
ChIP-qPCR analysis of indicated loci in WT andpds5D cells. The
euchromatic locus fbp1serves as a control. The percentage of
inputthat was immunoprecipitated is shown (% In-put). The position
of the oligonucleotides isindicated in the schematic above.
Errorbars denote SEM; N = 3 independent strains.* P , 0.05, ** P ,
0.01, and **** P ,0.0001 (Student’s t-test). (C) RT-qPCR analysisof
mat2-Pc and ura4+ genes in the indicatedstrains is shown as the
relative fold enrichmentover act1. Error bars denote SEM; N $ 3
in-dependent experiments. * P, 0.05 (Student’st-test). ChIP,
chromatin immunoprecipitation;PMG5S, pombe minimal glutamate plus
fivesupplements; qPCR, quantitative PCR; WT,wild-type; YEA, yeast
extract plus adenine.
132 H. D. Folco et al.
-
cohesin (Tanaka et al. 2001). However, cohesin localizationat
heterochromatin locations remains unaffected in the ab-sence of
Pds5 (Yamagishi et al. 2010). Consistent with theselocus-specific
studies, our analyses revealed that Pds5 is lo-calized in the
nucleus and is enriched throughout heterochro-matin domains,
including at centromeres, telomeres, and thesilent mat region
(Figure 5, A–D). Also, we noted a remark-able genome-wide
colocalization of Pds5 with the cohesinsubunits Rad21 and Psc3
(Mizuguchi et al. 2014; Folcoet al. 2017) (Figure S4, A–C). In
addition to their identicaldistribution profiles within
heterochromatin domains, Pds5and cohesin subunits colocalized at
the 39-ends of convergentgenes (Figure 5E and Figure S4C).
We previously reported that Swi6 recruits the cohesin-loading
complex to heterochromatic regions (Fischer et al.2009). Cells
lacking Clr4, which is required for Swi6 locali-zation, therefore
fail to load cohesin to heterochromatin do-mains (Bernard et al.
2001; Nonaka et al. 2002; Schmidt et al.2009; Folco et al. 2017).
We wondered whether the hetero-chromatin machinery is also required
for Pds5 localization.ChIP-chip analyses revealed that loss of Swi6
or Clr4 causedsevere defects in Pds5 localization at centromeres,
and at the
silent mat region (Figure 5, A–D). However, the localizationof
Pds5 to euchromatic locations was unaffected in
hetero-chromatin-deficient cells (Figure 5E), consistent with
previ-ous studies showing that heterochromatin factors
aredispensable for cohesin localization at euchromatic
regions(Schmidt et al. 2009; Mizuguchi et al. 2014; Folco et
al.2017). The finding that Pds5 mimics cohesin
localizationsuggested a close connection between these proteins,
whichcould reveal the cause of silencing defects in pds5D
cells.
Pds5-associated Eso1 is required forheterochromatic
silencing
Wenext tested factors associated with Pds5 for a possible rolein
heterochromatic silencing. Previous studies have shownthat Pds5
interacts with conserved factors Hrk1, Wpl1, andEso1 to achieve
proper sister chromatid cohesion and chro-mosome segregation (Goto
et al. 2017). Hrk1, a haspin-re-lated kinase, phosphorylates
histone H3 Thr3 to recruitthe chromosomal passenger complex (CPC)
(Yamagishiet al. 2010). Wpl1 is an anticohesion factor that
promotescohesion loss with or without cohesin release from
DNA(Kueng et al. 2006; Feytout et al. 2011; Birot et al. 2017).
Figure 3 Pds5 is required for genome-wideheterochromatin
formation. (A) Distributionof H3K9me2 along ChrI in WT and
pds5Dcells as determined by ChIP-chip. (B) ChIP-chip analysis of
H3K9me2 enrichment atcentromere 1 and tel1L is shown. The
foldenrichment of H3K9me2 (y-axis) is plotted atthe indicated
chromosome position (x-axis).(C) ChIP-qPCR analysis of H3K9me2
andH3K9me3 enrichment at dg repeats, andthe control fbp1 locus, in
WT and pds5D.The percentage of input that was immuno-precipitated
is shown (% Input). Error barsdenote SEM; N = 3 independent strains
pergenotype. ChIP, chromatin immunoprecipi-tation; ChrI, chromosome
I; qPCR, quantita-tive PCR; WT, wild-type.
Effect of Pds5 on Heterochromatin 133
-
On the other hand, Eso1 is an acetyltransferase that acety-lates
two conserved lysine residues (K105 and K106) withinthe globular
head domain of the cohesin subunit Psm3 topromote proper cohesion
establishment (Tanaka et al.2000; Skibbens 2009).
To uncover clues to the underlying mechanism, we askedwhether
the loss of any of the Pds5-associated factors causessilencing
defects similar to those in pds5D. Deletion of thenonessential gene
hrk1 (Yamagishi et al. 2010) did not affectmat2P::ura4+ repression
and failed to trigger noticeable lev-els of haploid meiosis (Figure
6A). Thus, Pds5-mediated re-cruitment of the CPC has no apparent
role in heterochromaticsilencing. Next, we focused on Wpl1 and
Eso1. Because of itsessential function in promoting sister
chromatid cohesion,cells lacking Eso1 are not viable (Tanaka et al.
2000). How-ever, due to their antagonistic roles in cohesion
establish-ment, the lethality of eso1D can be suppressed by
deletion
of wpl1 (Feytout et al. 2011; Kagami et al. 2011). Whereaswpl1D
did not show defects in heterochromatic silencing,the eso1D wpl1D
double mutant showed derepression ofmat2P::ura4+ and haploid
meiosis similar to pds5D cells(Figure 6A). Moreover, when we
deleted pds5 in cellslacking Eso1 and/or Wpl1, the levels of
haploid meiosisdisplayed by double or triple mutants were
comparable tothat of single-mutant pds5D (Figure S5). Taken
together,these results indicate that the cohesin
acetyltransferaseEso1 is the Pds5-associated factor that is
critical for hetero-chromatic silencing.
Acetylation-mimicking cohesin mutations mitigateheterochromatin
defects in pds5D cells
We considered that the lack of acetylation of cohesin maybe the
major cause of heterochromatin defects observed inpds5D cells. To
directly test this, we asked if a mutation in the
Figure 4 Pds5 is a heterochromatin mainte-nance factor. (A)
Examples of haploidgrowth phenotypes obtained from tetraddissection
analysis of KD::ade6+ “OFF” 3pds5D crosses. Dissection plates were
replicaplated onto PMG – ADE and YE (low ADEplates). (B) ChIP-qPCR
analysis of H3K9me3enrichment at indicated loci in WT andpds5D
cells. Data are shown as relative foldenrichment compared to the
control fbp1locus. The position of the oligonucleotidesis indicated
in the schematic in (A). Error barsdenote SEM; N = 3 independent
strainsper genotype. * P , 0.05, ** P , 0.01,and *** P , 0.001
(Student’s t-test). ADE,adenine; ChIP, chromatin
immunoprecipitation;PMG, pombe minimal glutamate; qPCR,
quan-titative PCR; WT, wild-type.
134 H. D. Folco et al.
-
cohesin subunit Psm3 that mimics acetylation by Eso1 canrestore
heterochromatin assembly in eso1D or pds5D cells.We combined the
psm3K105QK106Q (psm3KK/QQ) mutantallele with eso1D, eso1D wpl1D, or
pds5D. As expected,psm3KK/QQ suppressed the lethality caused by
eso1D (Figure6B) (Feytout et al. 2011; Kagami et al. 2011).
Notably, we
found that eso1D psm3KK/QQ showed neither haploid meio-sis nor
derepression of mat2P::ura4+, indicating that thepsm3KK/QQ
acetylation-mimicking mutation can preventheterochromatin defects
caused by loss of Eso1 (Figure6B). Further underscoring a
potentially important require-ment for Psm3 acetylation, the
psm3KK/QQ allele rescued
Figure 5 Pds5 localizes throughout heterochromatin do-mains. (A)
Microscopy analysis of Pds5-GFP localization inWT, swi6D, and clr4D
cells. On the bottom, fluorescenceand brightfield images are
merged. (B–E) ChIP-chip analysisof Pds5 distribution. Pds5-GFP
localization along ChrII (B),centromere 1 (C), the mat locus (D),
and a euchromaticchromosomal arm region (E) is shown for the
indicatedstrains. The fold enrichment of Pds5 (y-axis) is plotted
atthe indicated chromosome position (x-axis). Green barsrepresent
open reading frames according to the 2007 -S. pombe genome
assembly. ChIP, chromatin immunopre-cipitation; ChrI/II, chromosome
I/II; WT, wild-type.
Effect of Pds5 on Heterochromatin 135
-
defective heterochromatic silencing in eso1D wpl1D andpds5D
(Figure 6B). In the case of pds5D psm3KK/QQ, cellsgrew on
counterselective FOA medium, indicating mainte-nance ofmat2P::ura4+
silencing, but still exhibited some de-gree of haploid meiosis
(Figure 6, B and C).
We next explored whether restoration of silencing bypsm3KK/QQ is
linked to heterochromatin formation. Remark-ably, pds5D cells
carrying the psm3KK/QQ allele showed amarked increase in H3K9me2/3
levels at the silent mat re-gion (Figure 6D). Together, these
results suggest that defec-tive heterochromatic silencing in pds5D
cells is functionallyconnected to the lack of cohesin acetylation
by the Eso1acetyltransferase.
Cohesin removal prevents heterochromatin defects inpds5D
cells
As defective cohesin acetylation is linked to the
phenotypesobserved in pds5D, we wondered if cohesin itself is
requiredfor heterochromatic silencing. We tested this possibility
byassaying the effects of psc3-4T and rad21-K1 cohesin mutants
on silencing at the mat locus. Cohesin mutants displayed
nei-ther haploidmeiosis nor derepression ofmat2P::ura4+ (Figure7A).
This finding is consistent with previous work suggestingthat
cohesin is dispensable for heterochromatic silencing(Nonaka et al.
2002; Yamagishi et al. 2008). However, a pos-sibility remained that
residual levels of cohesin in the partialloss-of-function psc3-4T
and rad21-K1mutants might be suffi-cient to support heterochromatic
silencing. To rule this out, wecreated a strain lacking cohesin
specifically at heterochromaticlocations (Figure 7B). We took
advantage of a previous obser-vation that lethality caused by loss
of mitotic cohesin subunitsPsc3 and Rad21 can be bypassed by
overexpression of meioticcohesin subunits Rec11 and Rec8 (Kitajima
et al. 2003). Im-portantly, Rec8 and Rec11 specifically localize to
chromosomalarms but not to heterochromatic regions. Indeed, our
ChIPanalyses showed that Rec8 is enriched at chromosomal
armregions, but not at the silentmat region and at
pericentromericregions (Figure 7C). As expected, cells lacking
cohesin at cen-tromeres were sensitive to the spindle poison
thiabendazole(Figure 7B).Whenwe used this strain to assay silencing
at the
Figure 6 Lack of cohesin acetylation impairsheterochromatic
silencing. (A and B) Ten-fold serial dilutions of the indicated
strainswere spotted on YEA, with or withoutFOA, and PMG5S minimal
media, and grownfor 3 days at 30 and 32�, respectively.PMG5S plates
were stained with iodine va-por. (C) Iodine staining of single
coloniesgenerated from the indicated strains. Cellswere grown on
YEA at 30�, replica platedonto PMG5S minimal media, and grown at32�
for 3 days prior to staining with iodinevapor. (D) H3K9me2 and
H3K9me3 ChIP-qPCR analyses of the indicated loci areshown as the
percentage of immunoprecipi-tated input (% Input) in pds5D and
pds5Dpsm3KK/QQ. The position of the oligonucle-otides is shown in
the schematic above. Errorbars denote SEM; N = 3 independent
strains.* P, 0.05 (Student’s t-test). ChIP,
chromatinimmunoprecipitation; PMG5S, pombe mini-mal glutamate plus
five supplements; qPCR,quantitative PCR; WT, wild-type; YEA,
yeastextract plus adenine.
136 H. D. Folco et al.
-
mat locus, we found that depletion of cohesin had no
impact.Indeed, unlike pds5D, cells lacking cohesin at the mat
locusshowed no haploid meiosis (Figure 7D).
Our results suggested that cohesin itself is dispensable
forheterochromatic silencing. However, the fact that a
mutationmimicking cohesin acetylation can restore silencing in
pds5Dcells (Figure 6, B–D) indicated that the unacetylated form
ofcohesin may impede heterochromatin formation. One predic-tion of
this reasoning is that defective heterochromatin as-sembly in pds5D
cells could be prevented by removingcohesin at heterochromatic
loci. To test this, we deletedpds5 in cells lacking cohesin at the
silentmat region. Remark-ably, the absence of cohesin suppressed
the heterochromatic
silencing defect in pds5D (Figure 7, D and E). Moreover,
thisspecific loss of cohesin from heterochromatic loci in
pds5Dcells resulted in a marked increase in H3K9me3 levels at
thesilent mat region as compared to the pds5D mutant (Figure7F).
These results support the conclusion that in cells lackingPds5 and
its associated Eso1, the unacetylated form of cohe-sin interferes
with the proper assembly of heterochromaticstructures.
Discussion
Previous studies using the genetically tractable S. pombemodel
system have led to valuable insights into conserved
Figure 7 Loss of cohesin suppresses haploidmeiosis in pds5D. (A)
Left: ten-fold serial di-lutions of the indicated strains were
spottedon YEA rich media, with or without additionof FOA, and grown
at 30�. Right: iodinestaining of single colonies generatedfrom the
indicated strains. Cells weregrown on YEA at 26�, replica plated
ontoPMG5S minimal media, and grown at 32�for 4 days prior to
staining with iodine vapor.(B) Top: schematic of mitotic and
meioticcohesin complexes (Kitajima et al. 2003).Bottom: ten-fold
serial dilutions of the indi-cated strains were spotted on YEA rich
me-dia, with or without the spindle poison TBZ,and grown at 32�.
(C) ChIP-qPCR analysis ofRec8-HA enrichment at individual loci in
theindicated strains. Data are shown as relativefold enrichment
compared to the fbp1 con-trol locus. Error bars denote SD; N = 2
inde-pendent experiments. Mean values markedwith different letters
(a or b) indicate resultsthat are significantly different from
eachother, as established by one-way ANOVAand Holm–Sidak test for
multiple compari-sons, respectively (P, 0.05). (D) Iodine stain-ing
of single colonies generated from theindicated strains. Cells were
grown on YEAat 32�, replica plated onto EMM minimalmedia, and grown
at 30� for 3 days priorto staining with iodine vapor. (E)
RT-qPCRanalysis of mat2-Pc in the indicated strainsis shown as the
relative fold enrichmentcompared to the control leu1 locus.
Errorbars denote SD; N = 2 independent experi-ments. One-way ANOVA
followed by Holm–Sidak test (P , 0.01). (F) ChIP-qPCR analysisof
H3K9me3 enrichment at the indicated lociin pds5D cells expressing
either mitotic or mei-otic cohesin. The percentage of input thatwas
immunoprecipitated is shown (% Input).Error bars denote SEM; N = 9
independentexperiments. ** P , 0.01 and *** P ,0.001 (Student’s
t-test). ChIP, chromatin im-munoprecipitation; FOA, 5-flouroorotic
acid;PMG5S, pombe minimal glutamate plus fivesupplements; qPCR,
quantitative PCR; TBZ,thiabendazole; WT, wild-type; YEA, yeast
ex-tract plus adenine.
Effect of Pds5 on Heterochromatin 137
-
heterochromatin assembly pathways (Grewal and Jia 2007;Allshire
and Madhani 2018). Indeed, . 50 proteins havebeen shown to
contribute to the assembly of heterochromatindomains. In
particular, unbiased genetic screens have identi-fied a variety of
factors that impact nucleation, spreading,and/or epigenetic
inheritance of heterochromatin. However,significant gaps remain in
our understanding of the underly-ing mechanisms.
In this study, we used a highly sensitive reporter system
toperform a systematic screen for factors that affect
heterochro-matic silencing at the silent mat locus. In addition to
identi-fying previously known heterochromatin assembly factors,such
as components of the CLRC (Hong et al. 2005; Hornet al. 2005; Jia
et al. 2005), HP1 proteins (Thon andVerhein-Hansen 2000; Sadaie et
al. 2004), SHREC (Thonand Klar 1992; Ekwall and Ruusala 1994; Thon
et al. 1994;Sugiyama et al. 2007), FACT (facilitates chromatin
transcrip-tion) (Lejeune et al. 2007), and Fft3 (Strålfors et al.
2011;Taneja et al. 2017), we identified eight additional
factors.Among the newly identified factors were 60S ribosomal
pro-teins (Rpl2102, Rpl39, and Rpl502) and the 19S
proteasomeregulatory subunit Rpn10. While it remains unclear how
ri-bosomal proteins impact heterochromatic silencing, it is
pos-sible that loss of Rpn10 increases levels of
antisilencingfactors that affect the spreading and epigenetic
inheritanceof heterochromatin. Indeed, levels of Epe1, which
negativelyaffects heterochromatin stability (Ayoub et al. 2003),
are reg-ulated by the protein degradation machinery (Braun et
al.2011). Moreover, a mutation in the Rpt4 subunit of the
19Sproteasome was reported to affect heterochromatin spread-ing
(Seo et al. 2017). However, the major finding from thiswork is that
the cohesin-associated factor Pds5 is required forthe stable
maintenance of heterochromatin. Our analysessuggest that Pds5
prevents the cohesin protein complex,which binds preferentially
across heterochromatin domains(Bernard et al. 2001; Nonaka et al.
2002), from interferingwith heterochromatin assembly.
Pds5 localizes throughout heterochromatin domains andits loss
causes defective silencing that correlates with reducedH3K9me
levels. Evidence suggests thatPds5 is required for thestable
propagation of heterochromatin. Indeed, deletion ofpds5 in cells
lacking Ago1, which is required for RNAi-mediatednucleation of
heterochromatin, causes a cumulative de-crease in H3K9me levels.
How might Pds5 contribute toheterochromatic silencing? Considering
that Pds5 forms acomplex with cohesin that has been shown to play
an impor-tant role in chromosome architecture (Mizuguchi et al.
2014;Kim et al. 2016), it is possible that pds5D cells are
defective incohesin-dependent higher-order chromatin organization
atheterochromatic loci. However, several lines of evidence sug-gest
that, with the exception of a rad21 mutant that displaysreduced
heterochromatin formation at subtelomeric regions(Dheur et al.
2011), cohesin itself is dispensable for hetero-chromatic
silencing. First, heterochromatin is not affected incells carrying
thermosensitive mutations in cohesin subunits(Nonaka et al. 2002)
(this study). Second, we find that cells
specifically devoid of cohesin at themat locus are proficient
inheterochromatin assembly and gene silencing. Third, it hasbeen
shown that artificial recruitment of cohesin to hetero-chromatic
loci in cells lacking Swi6 is not sufficient totrigger gene
silencing (Yamagishi et al. 2008). Thus, whilePds5–cohesin may
contribute to higher-order organization atheterochromatic loci, it
is unlikely that defects in such organi-zation are responsible for
the silencing phenotype displayed bypds5D cells.
As described above, Pds5 interacts with multiple factors
tocoordinate diverse chromosomal events. In addition to recruit-ing
Hrk1 kinase and the CPC, which are involved in sisterchromatid
biorientation during cell division, Pds5 associateswith Eso1
acetyltransferase and Wpl1, which are linked tocohesin dynamics
(Vaur et al. 2012; Goto et al. 2017). Amongthese, we find that Eso1
is themost critical factor for the role ofPds5 in heterochromatin
maintenance, although silencing atthe edge of a heterochromatin
domain at the mat locus hasbeen shown to be affected in cells
lacking Wpl1 (Jahn et al.2018). Given our results showing that an
acetylation-mimick-ing cohesinmutant can suppress silencing defects
displayed byeso1D and pds5D, it is likely that the lack of cohesin
acetylationcreates a major impediment to heterochromatin
assembly.Acetylated cohesin might limit the effects of
antisilencing fac-tors such as Epe1 (Ayoub et al. 2003), which is
also recruitedby Swi6 (Zofall and Grewal 2006). Another possibility
is thatunacetylated cohesin may directly interfere with
heterochro-matin assembly. This second possibility is supported by
ourfinding that eliminating cohesin from heterochromatic loca-tions
can restore the proper maintenance of heterochromatinin cells
lacking Pds5.
Cohesin acetylation is a central determinant of replicationfork
processivity in mammalian systems (Terret et al. 2009;Sherwood et
al. 2010), and loss of Pds5 can hinder replica-tion progression
(Carvajal-Maldonado et al. 2019). More-over, depletion of Rad21 in
a Pds5-deficient backgroundrescues the replication defect
(Carvajal-Maldonado et al.2019). Considering that factors required
for DNA replicationalso impact heterochromatin maintenance
(Nakayama et al.2001a; Jahn et al. 2018), it is possible that
defects in hetero-chromatin formation caused by loss of Pds5 and
its associatedEso1 might be linked to impaired replication. Changes
inreplication might affect the preservation of parental
modifiedhistones required for Clr4 loading, which in turn
epigeneti-cally maintains heterochromatin domains through its
abilityto both read and write H3K9me nucleosomes (Zhang et al.2008;
Aygün et al. 2013). Regardless of the mechanism, ourfindings
highlight an important role for the conserved Pds5protein in
coordinating interplay between cohesin involvedin sister chromatid
cohesion and factors contributing to het-erochromatin
maintenance.
Acknowledgments
We thank Takeshi Sakuno (Osaka University, Japan) and
theNational BioResource Project-Yeast (Japan) for strains,
Jemima
138 H. D. Folco et al.
-
Barrowman for editing the manuscript, Martin Zofall andSahana
Holla for technical assistance, and members of theGrewal laboratory
for discussions. This work was supportedby the Intramural Research
Program of the NationalInstitutes of Health, National Cancer
Institute.
Author contributions: H.D.F. and S.I.S.G. conceived theproject,
and designed experiments. H.D.F. performed mostof the experiments
including the genome-wide screen. A.M.constructed strains, and
performed RT-PCR and ChIP-chipexperiments. V.B. constructed
strains. All authors contrib-uted to data interpretation. H.D.F.
and S.I.S.G. wrote themanuscript with input from all authors.
Literature Cited
Allshire, R. C., and H. D. Madhani, 2018 Ten principles of
hetero-chromatin formation and function. Nat. Rev. Mol. Cell Biol.
19:229–244. https://doi.org/10.1038/nrm.2017.119
Al-Sady, B., H. D. Madhani, and G. J. Narlikar, 2013 Division
oflabor between the chromodomains of HP1 and Suv39 methylaseenables
coordination of heterochromatin spread. Mol. Cell 51:80–91.
https://doi.org/10.1016/j.molcel.2013.06.013
Aygün, O., S. Mehta, and S. I. Grewal, 2013
HDAC-mediatedsuppression of histone turnover promotes epigenetic
stabilityof heterochromatin. Nat. Struct. Mol. Biol. 20:
547–554.https://doi.org/10.1038/nsmb.2565
Ayoub, N., I. Goldshmidt, and A. Cohen, 1999 Position effect
var-iegation at the mating-type locus of fission yeast: a
cis-actingelement inhibits covariegated expression of genes in the
silentand expressed domains. Genetics 152: 495–508.
Ayoub, N., K. Noma, S. Isaac, T. Kahan, S. I. Grewal et al.,
2003 Anovel jmjC domain protein modulates heterochromatization
infission yeast. Mol. Cell. Biol. 23: 4356–4370.
https://doi.org/10.1128/MCB.23.12.4356-4370.2003
Bähler, J., J. Q. Wu, M. S. Longtine, N. G. Shah, A. McKenzie,
III et al.,1998 Heterologous modules for efficient and versatile
PCR-basedgene targeting in Schizosaccharomyces pombe. Yeast 14:
943–951.https://doi.org/10.1002/(SICI)1097-0061(199807)14:10,943::AID-YEA292.3.0.CO;2-Y
Bannister, A. J., P. Zegerman, J. F. Partridge, E. A. Miska, J.
O.Thomas et al., 2001 Selective recognition of methylated lysine9
on histone H3 by the HP1 chromo domain. Nature 410: 120–124.
https://doi.org/10.1038/35065138
Bayne, E. H., S. A. White, A. Kagansky, D. A. Bijos, L.
Sanchez-Pulido et al., 2010 Stc1: a critical link between RNAi
andchromatin modification required for heterochromatin
integrity.Cell 140: 666–677.
https://doi.org/10.1016/j.cell.2010.01.038
Bayne, E. H., D. A. Bijos, S. A. White, F. de Lima Alves, J.
Rappsilberet al., 2014 A systematic genetic screen identifies new
factorsinfluencing centromeric heterochromatin integrity in
fissionyeast. Genome Biol. 15: 481.
https://doi.org/10.1186/s13059-014-0481-4
Bernard, P., J. F. Maure, J. F. Partridge, S. Genier, J. P.
Javerzat et al.,2001 Requirement of heterochromatin for cohesion at
centromeres.Science 294: 2539–2542.
https://doi.org/10.1126/science.1064027
Birot, A., K. Eguienta, S. Vazquez, S. Claverol, M. Bonneu et
al.,2017 A second Wpl1 anti-cohesion pathway requires
dephos-phorylation of fission yeast kleisin Rad21 by PP4. EMBO J.
36:1364–1378. https://doi.org/10.15252/embj.201696050
Blackwell, C., K. A. Martin, A. Greenall, A. Pidoux, R. C.
Allshireet al., 2004 The Schizosaccharomyces pombe HIRA-like
proteinHip1 is required for the periodic expression of histone
genes and
contributes to the function of complex centromeres. Mol.
Cell.Biol. 24: 4309–4320.
https://doi.org/10.1128/MCB.24.10.4309-4320.2004
Braun, S., J. F. Garcia, M. Rowley, M. Rougemaille, S. Shankaret
al., 2011 The Cul4-Ddb1(Cdt)(2) ubiquitin ligase inhibitsinvasion
of a boundary-associated antisilencing factor intoheterochromatin.
Cell 144: 41–54. https://doi.org/10.1016/j.cell.2010.11.051
Cam, H. P., T. Sugiyama, E. S. Chen, X. Chen, P. C. FitzGerald
et al.,2005 Comprehensive analysis of heterochromatin- and
RNAi-mediated epigenetic control of the fission yeast genome.
Nat.Genet. 37: 809–819. https://doi.org/10.1038/ng1602
Cam, H. P., K. Noma, H. Ebina, H. L. Levin, and S. I.
Grewal,2008 Host genome surveillance for retrotransposons by
trans-poson-derived proteins. Nature 451: 431–436.
https://doi.org/10.1038/nature06499
Carvajal-Maldonado, D., A. K. Byrum, J. Jackson, S. Wessel, D.
Le-maçon et al., 2019 Perturbing cohesin dynamics drives
MRE11nuclease-dependent replication fork slowing. Nucleic Acids
Res.47: 1294–1310. https://doi.org/10.1093/nar/gky519
Chalamcharla, V. R., H. D. Folco, J. Dhakshnamoorthy, and S.
I.Grewal, 2015 Conserved factor Dhp1/Rat1/Xrn2 triggers pre-mature
transcription termination and nucleates heterochroma-tin to promote
gene silencing. Proc. Natl. Acad. Sci. USA 112:15548–15555.
Dheur, S., S. J. Saupe, S. Genier, S. Vazquez, and J. P.
Javerzat,2011 Role for cohesin in the formation of a
heterochromaticdomain at fission yeast subtelomeres. Mol. Cell.
Biol. 31: 1088–1097. https://doi.org/10.1128/MCB.01290-10
Ekwall, K., and T. Ruusala, 1994 Mutations in rik1, clr2, clr3
andclr4 genes asymmetrically derepress the silent mating-type
lociin fission yeast. Genetics 136: 53–64.
Feytout, A., S. Vaur, S. Genier, S. Vazquez, and J. P.
Javerzat,2011 Psm3 acetylation on conserved lysine residues is
dispens-able for viability in fission yeast but contributes to
Eso1-mediatedsister chromatid cohesion by antagonizing Wpl1. Mol.
Cell. Biol.31: 1771–1786. https://doi.org/10.1128/MCB.01284-10
Fischer, T., B. Cui, J. Dhakshnamoorthy, M. Zhou, C. Rubin et
al.,2009 Diverse roles of HP1 proteins in heterochromatin assem-bly
and functions in fission yeast. Proc. Natl. Acad. Sci. USA
106:8998–9003. https://doi.org/10.1073/pnas.0813063106
Folco, H. D., V. R. Chalamcharla, T. Sugiyama, G.
Thillainadesan,M. Zofall et al., 2017 Untimely expression of
gametogenicgenes in vegetative cells causes uniparental disomy.
Nature543: 126–130. https://doi.org/10.1038/nature21372
Freitag, M., 2017 Histone methylation by SET domain proteins
infungi. Annu. Rev. Microbiol. 71: 413–439.
https://doi.org/10.1146/annurev-micro-102215-095757
Goto, Y., Y. Yamagishi, M. Shintomi-Kawamura, M. Abe, Y. Tannoet
al., 2017 Pds5 regulates sister-chromatid cohesion and chro-mosome
bi-orientation through a conserved protein interactionmodule. Curr.
Biol. 27: 1005–1012. https://doi.org/10.1016/j.cub.2017.02.066
Grewal, S. I., and S. Jia, 2007 Heterochromatin revisited.
Nat.Rev. Genet. 8: 35–46. https://doi.org/10.1038/nrg2008
Grewal, S. I., and A. J. Klar, 1996 Chromosomal inheritance
ofepigenetic states in fission yeast during mitosis and meiosis.
Cell86: 95–101. https://doi.org/10.1016/S0092-8674(00)80080-X
Grewal, S. I., and A. J. Klar, 1997 A recombinationally
repressedregion between mat2 and mat3 loci shares homology to
centro-meric repeats and regulates directionality of
mating-typeswitching in fission yeast. Genetics 146: 1221–1238.
Grewal, S. I., M. J. Bonaduce, and A. J. Klar, 1998 Histone
deace-tylase homologs regulate epigenetic inheritance of
transcrip-tional silencing and chromosome segregation in fission
yeast.Genetics 150: 563–576.
Effect of Pds5 on Heterochromatin 139
https://doi.org/10.1038/nrm.2017.119https://doi.org/10.1016/j.molcel.2013.06.013https://doi.org/10.1038/nsmb.2565https://doi.org/10.1128/MCB.23.12.4356-4370.2003https://doi.org/10.1128/MCB.23.12.4356-4370.2003https://doi.org/10.1002/(SICI)1097-0061(199807)14:103.0.CO;2-Yhttps://doi.org/10.1002/(SICI)1097-0061(199807)14:103.0.CO;2-Yhttps://doi.org/10.1002/(SICI)1097-0061(199807)14:103.0.CO;2-Yhttps://doi.org/10.1002/(SICI)1097-0061(199807)14:103.0.CO;2-Yhttps://doi.org/10.1038/35065138https://doi.org/10.1016/j.cell.2010.01.038https://doi.org/10.1186/s13059-014-0481-4https://doi.org/10.1186/s13059-014-0481-4https://doi.org/10.1126/science.1064027https://doi.org/10.15252/embj.201696050https://doi.org/10.1128/MCB.24.10.4309-4320.2004https://doi.org/10.1128/MCB.24.10.4309-4320.2004https://doi.org/10.1016/j.cell.2010.11.051https://doi.org/10.1016/j.cell.2010.11.051https://doi.org/10.1038/ng1602https://doi.org/10.1038/nature06499https://doi.org/10.1038/nature06499https://doi.org/10.1093/nar/gky519https://doi.org/10.1128/MCB.01290-10https://doi.org/10.1128/MCB.01284-10https://doi.org/10.1073/pnas.0813063106https://doi.org/10.1038/nature21372https://doi.org/10.1146/annurev-micro-102215-095757https://doi.org/10.1146/annurev-micro-102215-095757https://doi.org/10.1016/j.cub.2017.02.066https://doi.org/10.1016/j.cub.2017.02.066https://doi.org/10.1038/nrg2008https://doi.org/10.1016/S0092-8674(00)80080-X
-
Hall, I. M., G. D. Shankaranarayana, K. Noma, N. Ayoub, A.
Cohenet al., 2002 Establishment and maintenance of a
heterochroma-tin domain. Science 297: 2232–2237.
https://doi.org/10.1126/science.1076466
Hartman, T., K. Stead, D. Koshland, and V. Guacci, 2000 Pds5p is
anessential chromosomal protein required for both sister
chromatidcohesion and condensation in Saccharomyces cerevisiae. J.
Cell Biol.151: 613–626. https://doi.org/10.1083/jcb.151.3.613
Hong, E. J., J. Villen, E. L. Gerace, S. P. Gygi, and D.
Moazed,2005 A cullin E3 ubiquitin ligase complex associates
withRik1 and the Clr4 histone H3–K9 methyltransferase and is
re-quired for RNAi-mediated heterochromatin formation. RNABiol. 2:
106–111. https://doi.org/10.4161/rna.2.3.2131
Horn, P. J., J. N. Bastie, and C. L. Peterson, 2005 A
Rik1-associated,cullin-dependent E3 ubiquitin ligase is essential
for heterochro-matin formation. Genes Dev. 19: 1705–1714.
https://doi.org/10.1101/gad.1328005
Jahn, L. J., B. Mason, P. Brogger, T. Toteva, D. K. Nielsen et
al.,2018 Dependency of heterochromatin domains on
replicationfactors. G3 (Bethesda) 8: 477–489.
https://doi.org/10.1534/g3.117.300341
Jia, S., R. Kobayashi, and S. I. Grewal, 2005 Ubiquitin ligase
com-ponent Cul4 associates with Clr4 histone methyltransferase
toassemble heterochromatin. Nat. Cell Biol. 7:
1007–1013.https://doi.org/10.1038/ncb1300
Job, G., C. Brugger, T. Xu, B. R. Lowe, Y. Pfister et al., 2016
SHRECsilences heterochromatin via distinct remodeling and
deacetyla-tion modules. Mol. Cell 62: 207–221.
https://doi.org/10.1016/j.molcel.2016.03.016
Kagami, A., T. Sakuno, Y. Yamagishi, T. Ishiguro, T. Tsukaharaet
al., 2011 Acetylation regulates monopolar attachment atmultiple
levels during meiosis I in fission yeast. EMBO Rep.12: 1189–1195.
https://doi.org/10.1038/embor.2011.188
Kim, K. D., H. Tanizawa, O. Iwasaki, and K. Noma, 2016
Transcriptionfactors mediate condensin recruitment and global
chromosomal or-ganization in fission yeast. Nat. Genet. 48:
1242–1252. https://doi.org/10.1038/ng.3647
Kitajima, T. S., S. Yokobayashi, M. Yamamoto, and Y.
Watanabe,2003 Distinct cohesin complexes organize meiotic
chromosomedomains. Science 300: 1152–1155.
https://doi.org/10.1126/science.1083634
Klar, A. J., 2007 Lessons learned from studies of fission yeast
mat-ing-type switching and silencing. Annu. Rev. Genet. 41:
213–236.https://doi.org/10.1146/annurev.genet.39.073103.094316
Kueng, S., B. Hegemann, B. H. Peters, J. J. Lipp, A. Schleiffer
et al.,2006 Wapl controls the dynamic association of cohesin
withchromatin. Cell 127: 955–967.
https://doi.org/10.1016/j.cell.2006.09.040
Lejeune, E., M. Bortfeld, S. A. White, A. L. Pidoux, K. Ekwall
et al.,2007 The chromatin-remodeling factor FACT contributesto
centromeric heterochromatin independently of RNAi. Curr.Biol. 17:
1219–1224. https://doi.org/10.1016/j.cub.2007.06.028
Marina, D. B., S. Shankar, P. Natarajan, K. J. Finn, and H. D.
Mad-hani, 2013 A conserved ncRNA-binding protein recruits
silenc-ing factors to heterochromatin through an
RNAi-independentmechanism. Genes Dev. 27: 1851–1856.
https://doi.org/10.1101/gad.226019.113
Matsuyama, A., R. Arai, Y. Yashiroda, A. Shirai, A. Kamata et
al.,2006 ORFeome cloning and global analysis of protein
locali-zation in the fission yeast Schizosaccharomyces pombe.
Nat.Biotechnol. 24: 841–847 (erratum: Nat. Biotechnol. 24:
1033).https://doi.org/10.1038/nbt1222
Mizuguchi, T., G. Fudenberg, S. Mehta, J. M. Belton, N. Tanejaet
al., 2014 Cohesin-dependent globules and heterochromatinshape 3D
genome architecture in S. pombe. Nature 516: 432–435.
https://doi.org/10.1038/nature13833
Nakayama, J., A. J. Klar, and S. I. Grewal, 2000 A
chromodomainprotein, Swi6, performs imprinting functions in fission
yeast dur-ing mitosis and meiosis. Cell 101: 307–317.
https://doi.org/10.1016/S0092-8674(00)80840-5
Nakayama, J., R. C. Allshire, A. J. Klar, and S. I. Grewal,2001a
A role for DNA polymerase alpha in epigenetic controlof
transcriptional silencing in fission yeast. EMBO J. 20: 2857–2866.
https://doi.org/10.1093/emboj/20.11.2857
Nakayama, J., J. C. Rice, B. D. Strahl, C. D. Allis, and S. I.
Grewal,2001b Role of histone H3 lysine 9 methylation in
epigeneticcontrol of heterochromatin assembly. Science 292:
110–113.https://doi.org/10.1126/science.1060118
Noma, K., C. D. Allis, and S. I. Grewal, 2001 Transitions in
distincthistone H3 methylation patterns at the heterochromatin
domainboundaries. Science 293: 1150–1155.
https://doi.org/10.1126/science.1064150
Nonaka, N., T. Kitajima, S. Yokobayashi, G. Xiao, M. Yamamotoet
al., 2002 Recruitment of cohesin to heterochromatic regionsby
Swi6/HP1 in fission yeast. Nat. Cell Biol. 4: 89–93.
https://doi.org/10.1038/ncb739
Panizza, S., T. Tanaka, A. Hochwagen, F. Eisenhaber, and K.
Nasmyth,2000 Pds5 cooperates with cohesin in maintaining sister
chroma-tid cohesion. Curr. Biol. 10: 1557–1564.
https://doi.org/10.1016/S0960-9822(00)00854-X
Reyes-Turcu, F. E., and S. I. Grewal, 2012 Different means,
sameend-heterochromatin formation by RNAi and RNAi-independentRNA
processing factors in fission yeast. Curr. Opin. Genet. Dev.22:
156–163. https://doi.org/10.1016/j.gde.2011.12.004
Reyes-Turcu, F. E., K. Zhang, M. Zofall, E. Chen, and S. I.
Grewal,2011 Defects in RNA quality control factors reveal
RNAi-inde-pendent nucleation of heterochromatin. Nat. Struct. Mol.
Biol.18: 1132–1138. https://doi.org/10.1038/nsmb.2122
Roguev, A., M. Wiren, J. S. Weissman, and N. J. Krogan,2007
High-throughput genetic interaction mapping in the fis-sion yeast
Schizosaccharomyces pombe. Nat. Methods 4: 861–866.
https://doi.org/10.1038/nmeth1098
Sabatinos, S. A., and S. L. Forsburg, 2010 Molecular genetics
ofSchizosaccharomyces pombe. Methods Enzymol. 470:
759–795.https://doi.org/10.1016/S0076-6879(10)70032-X
Sadaie, M., T. Iida, T. Urano, and J. Nakayama, 2004 A
chromo-domain protein, Chp1, is required for the establishment of
het-erochromatin in fission yeast. EMBO J. 23: 3825–3835.
https://doi.org/10.1038/sj.emboj.7600401
Schmidt, C. K., N. Brookes, and F. Uhlmann, 2009 Conserved
fea-tures of cohesin binding along fission yeast chromosomes.
Ge-nome Biol. 10: R52. https://doi.org/10.1186/gb-2009-10-5-r52
Seo, H. D., Y. Choi, M. Kim, K. Kang, T. Urano et al., 2017
The19S proteasome is directly involved in the regulation of
hetero-chromatin spreading in fission yeast. J. Biol. Chem. 292:
17144–17155. https://doi.org/10.1074/jbc.M117.790824
Sherwood, R., T. S. Takahashi, and P. V. Jallepalli, 2010
Sisteracts: coordinating DNA replication and cohesion
establishment.Genes Dev. 24: 2723–2731.
https://doi.org/10.1101/gad.1976710
Skibbens, R. V., 2009 Establishment of sister chromatid
cohesion. Curr.Biol. 19: R1126–R1132.
https://doi.org/10.1016/j.cub.2009.10.067
Strålfors, A., J. Walfridsson, H. Bhuiyan, and K. Ekwall, 2011
TheFUN30 chromatin remodeler, Fft3, protects centromeric and
sub-telomeric domains from euchromatin formation. PLoS Genet.
7:e1001334. https://doi.org/10.1371/journal.pgen.1001334
Sugiyama, T., H. P. Cam, R. Sugiyama, K. Noma, M. Zofall et
al.,2007 SHREC, an effector complex for heterochromatic
tran-scriptional silencing. Cell 128: 491–504.
https://doi.org/10.1016/j.cell.2006.12.035
Sumara, I., E. Vorlaufer, C. Gieffers, B. H. Peters, and J. M.
Peters,2000 Characterization of vertebrate cohesin complexes
andtheir regulation in prophase. J. Cell Biol. 151:
749–762.https://doi.org/10.1083/jcb.151.4.749
140 H. D. Folco et al.
https://doi.org/10.1126/science.1076466https://doi.org/10.1126/science.1076466https://doi.org/10.1083/jcb.151.3.613https://doi.org/10.4161/rna.2.3.2131https://doi.org/10.1101/gad.1328005https://doi.org/10.1101/gad.1328005https://doi.org/10.1534/g3.117.300341https://doi.org/10.1534/g3.117.300341https://doi.org/10.1038/ncb1300https://doi.org/10.1016/j.molcel.2016.03.016https://doi.org/10.1016/j.molcel.2016.03.016https://doi.org/10.1038/embor.2011.188https://doi.org/10.1038/ng.3647https://doi.org/10.1038/ng.3647https://doi.org/10.1126/science.1083634https://doi.org/10.1126/science.1083634https://doi.org/10.1146/annurev.genet.39.073103.094316https://doi.org/10.1016/j.cell.2006.09.040https://doi.org/10.1016/j.cell.2006.09.040https://doi.org/10.1101/gad.226019.113https://doi.org/10.1101/gad.226019.113https://doi.org/10.1038/nbt1222https://doi.org/10.1038/nature13833https://doi.org/10.1016/S0092-8674(00)80840-5https://doi.org/10.1016/S0092-8674(00)80840-5https://doi.org/10.1093/emboj/20.11.2857https://doi.org/10.1126/science.1060118https://doi.org/10.1126/science.1064150https://doi.org/10.1126/science.1064150https://doi.org/10.1038/ncb739https://doi.org/10.1038/ncb739https://doi.org/10.1016/S0960-9822(00)00854-Xhttps://doi.org/10.1016/S0960-9822(00)00854-Xhttps://doi.org/10.1016/j.gde.2011.12.004https://doi.org/10.1038/nsmb.2122https://doi.org/10.1038/nmeth1098https://doi.org/10.1016/S0076-6879(10)70032-Xhttps://doi.org/10.1038/sj.emboj.7600401https://doi.org/10.1038/sj.emboj.7600401https://doi.org/10.1186/gb-2009-10-5-r52https://doi.org/10.1074/jbc.M117.790824https://doi.org/10.1101/gad.1976710https://doi.org/10.1016/j.cub.2009.10.067https://doi.org/10.1371/journal.pgen.1001334https://doi.org/10.1016/j.cell.2006.12.035https://doi.org/10.1016/j.cell.2006.12.035https://doi.org/10.1083/jcb.151.4.749
-
Tan, H. L., K. K. Lim, Q. Yang, J. S. Fan, A. M. M. Sayed et
al.,2018 Prolyl isomerization of the CENP-A N-terminus
regulatescentromeric integrity in fission yeast. Nucleic Acids Res.
46:1167–1179. https://doi.org/10.1093/nar/gkx1180
Tanaka, K., T. Yonekawa, Y. Kawasaki, M. Kai, K. Furuya et
al.,2000 Fission yeast Eso1p is required for establishing
sisterchromatid cohesion during S phase. Mol. Cell. Biol. 20:
3459–3469. https://doi.org/10.1128/MCB.20.10.3459-3469.2000
Tanaka, K., Z. Hao, M. Kai, and H. Okayama, 2001
Establishmentand maintenance of sister chromatid cohesion in
fission yeast bya unique mechanism. EMBO J. 20: 5779–5790.
https://doi.org/10.1093/emboj/20.20.5779
Taneja, N., M. Zofall, V. Balachandran, G. Thillainadesan,
T.Sugiyama et al., 2017 SNF2 family protein Fft3
suppressesnucleosome turnover to promote epigenetic inheritance
andproper replication. Mol. Cell 66: 50–62.e6.
https://doi.org/10.1016/j.molcel.2017.02.006
Tatebayashi, K., J. Kato, and H. Ikeda, 1998 Isolation of
aSchizosaccharomyces pombe rad21ts mutant that is aberrant
inchromosome segregation, microtubule function, DNA repair
andsensitive to hydroxyurea: possible involvement of Rad21
inubiquitin-mediated proteolysis. Genetics 148: 49–57.
Terret, M. E., R. Sherwood, S. Rahman, J. Qin, and P. V.
Jallepalli,2009 Cohesin acetylation speeds the replication fork.
Nature462: 231–234. https://doi.org/10.1038/nature08550
Thon, G., and A. J. Klar, 1992 The clr1 locus regulates the
expres-sion of the cryptic mating-type loci of fission yeast.
Genetics131: 287–296.
Thon, G., and J. Verhein-Hansen, 2000 Four chromo-domain
pro-teins of Schizosaccharomyces pombe differentially repress
transcrip-tion at various chromosomal locations. Genetics 155:
551–568.
Thon, G., A. Cohen, and A. J. Klar, 1994 Three additional
linkagegroups that repress transcription and meiotic recombination
inthe mating-type region of Schizosaccharomyces pombe. Genetics138:
29–38.
Thon, G., K. P. Bjerling, and I. S. Nielsen, 1999 Localization
andproperties of a silencing element near the mat3-M
mating-typecassette of Schizosaccharomyces pombe. Genetics 151:
945–963.
Touat-Todeschini, L., Y. Shichino, M. Dangin, N. Thierry-Mieg,
B.Gilquin et al., 2017 Selective termination of lncRNA
transcrip-tion promotes heterochromatin silencing and cell
differentia-tion. EMBO J. 36: 2626–2641.
https://doi.org/10.15252/embj.201796571
Tucker, J. F., C. Ohle, G. Schermann, K. Bendrin, W. Zhang et
al.,2016 A novel epigenetic silencing pathway involving thehighly
conserved 59-39 exoribonuclease Dhp1/Rat1/Xrn2
inSchizosaccharomyces pombe. PLoS Genet. 12: e1005873.
https://doi.org/10.1371/journal.pgen.1005873
Vaur, S., A. Feytout, S. Vazquez, and J. P. Javerzat, 2012
Pds5promotes cohesin acetylation and stable cohesin-chromosome
interaction. EMBO Rep. 13: 645–652.
https://doi.org/10.1038/embor.2012.72
Verdel, A., S. Jia, S. Gerber, T. Sugiyama, S. Gygi et al., 2004
RNAi-mediated targeting of heterochromatin by the RITS complex.
Sci-ence 303: 672–676. https://doi.org/10.1126/science.1093686
Volpe, T. A., C. Kidner, I. M. Hall, G. Teng, S. I. Grewal et
al.,2002 Regulation of heterochromatic silencing and histoneH3
lysine-9 methylation by RNAi. Science 297:
1833–1837.https://doi.org/10.1126/science.1074973
Wang, J., S. T. Jia, and S. Jia, 2016 New insights into the
regu-lation of heterochromatin. Trends Genet. 32: 284–294.
https://doi.org/10.1016/j.tig.2016.02.005
Wang, S. W., R. L. Read, and C. J. Norbury, 2002 Fission
yeastPds5 is required for accurate chromosome segregation and
forsurvival after DNA damage or metaphase arrest. J. Cell Sci.
115:587–598.
Wilkinson, C. R., K. Ferrell, M. Penney, M. Wallace, W. Dubiel
et al.,2000 Analysis of a gene encoding Rpn10 of the fission
yeastproteasome reveals that the polyubiquitin-binding site of
thissubunit is essential when Rpn12/Mts3 activity is compromised.J.
Biol. Chem. 275: 15182–15192.
https://doi.org/10.1074/jbc.275.20.15182
Yamagishi, Y., T. Sakuno, M. Shimura, and Y. Watanabe,2008
Heterochromatin links to centromeric protection by re-cruiting
shugoshin. Nature 455: 251–255 [corrigenda: Nature563: E21 (2018)].
https://doi.org/10.1038/nature07217
Yamagishi, Y., T. Honda, Y. Tanno, and Y. Watanabe, 2010
Twohistone marks establish the inner centromere and
chromosomebi-orientation. Science 330: 239–243.
https://doi.org/10.1126/science.1194498
Yamane, K., T. Mizuguchi, B. Cui, M. Zofall, K. Noma et al.,2011
Asf1/HIRA facilitate global histone deacetylation andassociate with
HP1 to promote nucleosome occupancy at het-erochromatic loci. Mol.
Cell 41: 56–66. https://doi.org/10.1016/j.molcel.2010.12.009
Zhang, K., K. Mosch, W. Fischle, and S. I. Grewal, 2008 Roles
ofthe Clr4 methyltransferase complex in nucleation, spreadingand
maintenance of heterochromatin. Nat. Struct. Mol. Biol.15: 381–388.
https://doi.org/10.1038/nsmb.1406
Zofall, M., and S. I. Grewal, 2006 Swi6/HP1 recruits a JmjC
do-main protein to facilitate transcription of heterochromatic
re-peats. Mol. Cell 22: 681–692.
https://doi.org/10.1016/j.molcel.2006.05.010
Zofall, M., D. R. Smith, T. Mizuguchi, J. Dhakshnamoorthy, and
S.I. S. Grewal, 2016 Taz1-Shelterin promotes facultative
hetero-chromatin assembly at chromosome-internal sites
containinglate replication origins. Mol. Cell 62: 862–874.
https://doi.org/10.1016/j.molcel.2016.04.034
Communicating editor: M. Freitag
Effect of Pds5 on Heterochromatin 141
https://doi.org/10.1093/nar/gkx1180https://doi.org/10.1128/MCB.20.10.3459-3469.2000https://doi.org/10.1093/emboj/20.20.5779https://doi.org/10.1093/emboj/20.20.5779https://doi.org/10.1016/j.molcel.2017.02.006https://doi.org/10.1016/j.molcel.2017.02.006https://doi.org/10.1038/nature08550https://doi.org/10.15252/embj.201796571https://doi.org/10.15252/embj.201796571https://doi.org/10.1371/journal.pgen.1005873https://doi.org/10.1371/journal.pgen.1005873https://doi.org/10.1038/embor.2012.72https://doi.org/10.1038/embor.2012.72https://doi.org/10.1126/science.1093686https://doi.org/10.1126/science.1074973https://doi.org/10.1016/j.tig.2016.02.005https://doi.org/10.1016/j.tig.2016.02.005https://doi.org/10.1074/jbc.275.20.15182https://doi.org/10.1074/jbc.275.20.15182https://doi.org/10.1038/nature07217https://doi.org/10.1126/science.1194498https://doi.org/10.1126/science.1194498https://doi.org/10.1016/j.molcel.2010.12.009https://doi.org/10.1016/j.molcel.2010.12.009https://doi.org/10.1038/nsmb.1406https://doi.org/10.1016/j.molcel.2006.05.010https://doi.org/10.1016/j.molcel.2006.05.010https://doi.org/10.1016/j.molcel.2016.04.034https://doi.org/10.1016/j.molcel.2016.04.034