Both H4K20 mono-methylation and H3K56 acetylation mark transcription-dependent histone turnover in fission yeast

Biochemical and Biophysical Research Communications 476 (2016) 515e521

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

Both H4K20 mono-methylation and H3K56 acetylation mark transcription-dependent histone turnover in fission yeast Hanna Yang a, Chang Seob Kwon b, Yoonjung Choi a, **, Daeyoup Lee a, * a b

Department of Biological Sciences, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu, Daejeon 34141, South Korea Department of Chemistry and Biology, Korea Science Academy of KAIST, Busan, 614-822, South Korea

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 May 2016 Accepted 28 May 2016 Available online 3 June 2016

Nucleosome dynamics facilitated by histone turnover is required for transcription as well as DNA replication and repair. Histone turnover is often associated with various histone modifications such as H3K56 acetylation (H3K56Ac), H3K36 methylation (H3K36me), and H4K20 methylation (H4K20me). In order to correlate histone modifications and transcription-dependent histone turnover, we performed genome wide analyses for euchromatic regions in G2/M-arrested fission yeast. The results show that transcription-dependent histone turnover at 50 promoter and 30 termination regions is directly correlated with the occurrence of H3K56Ac and H4K20 mono-methylation (H4K20me1) in actively transcribed genes. Furthermore, the increase of H3K56Ac and H4K20me1 and antisense RNA production was observed in the absence of the histone H3K36 methyltransferase Set2 and histone deacetylase complex (HDAC) that are involved in the suppression of histone turnover within the coding regions. These results together indicate that H4K20me1 as well as H3K56Ac are bona fide marks for transcription-dependent histone turnover in fission yeast. © 2016 Elsevier Inc. All rights reserved.

Keywords: Histone turnover Histone turnover marker H3K56Ac H4K20me1 Schizosaccharomyces pombe

1. Introduction The dynamic chromatin environment, by modulating the access of regulatory factors to nucleosomal DNA, is crucial for many biological processes such as transcription and DNA replication [1,2]. The timely opening and closing of chromatin can be regulated by histone turnover, which is carried out through replacement of prior existing histones with newly synthesized histones or histone variants. Various factors have been reported to facilitate histone turnover (also referred to as histone exchange); these include histone modifying enzymes, ATP-dependent remodelers, and chaperones [3]. Histone turnover occurs predominantly through two

Abbreviations: H3K56Ac, H3 lysine 56 acetylation; H4K20me, H4 lysine 20 methylation; Set2, histone H3 lysine 36 (K36) methyltransferase; HDAC, histone deacetylase complex; H3-Flag, flag tagged H3; S. pombe, Schizosaccharomyces pombe; S. cerevisiae, Saccharomyces cerevisiae. * Corresponding author. Department of Biological Sciences, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu, Daejeon 34141, South Korea. ** Corresponding author. Department of Biological Sciences, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu, Daejeon 34141, South Korea. E-mail addresses: [email protected] (Y. Choi), [email protected] (D. Lee). http://dx.doi.org/10.1016/j.bbrc.2016.05.155 0006-291X/© 2016 Elsevier Inc. All rights reserved.

distinct pathways differing in their dependence on DNA replication. Histone deposition is an important process to restore chromatin state following DNA replication and is accomplished by the DNA replication-related chaperone, Chromatin Assembly Factor 1 (CAF1) [4,5]. DNA replication-independent histone turnover is largely associated with transcriptional activity. Transcribing RNA polymerase II (Pol II) disassembles nucleosomes from the promoter. Histone chaperones such as Spt6 and FACT complex then reassemble the nucleosome after Pol II passage, thereby maintaining nucleosome integrity for correct transcription in subsequent rounds [3,6e8]. Histone acetylation also contributes to nucleosome dynamics by weakening the histone-DNA interaction. In particular, histone H3K56Ac functions in both DNA replication-dependent and eindependent histone turnover marking newly incorporated H3 [9,10]. Currently, H3K56Ac enriched at promoters of active genes is widely used as a maker of transcription-dependent histone turnover [11]. Methylation of the newly deposited H3 is also associated with histone turnover. This modification seems to be more stable than other histone modifications, which allows methylated histones to be accumulated on the stable nucleosome [12]. Indeed, in the fission yeast Schizosaccharomyces pombe, heterochromatin with low histone turnover rate is marked by H3K9 methylation. H3K36 methylation is also enriched in coding regions by preventing histone exchange, thereby slowing histone turnover rate compared to

516

H. Yang et al. / Biochemical and Biophysical Research Communications 476 (2016) 515e521

promoter regions [3,13]. Recently, methylation of H4K20 has been implicated in nucleosome age, and old nucleosomes bear H4K20 dimethylation and tri-methylation (H4K20me2/3) in fission yeast. In contrast to H4K20me2/3, generally enriched in less transcribed genes, H4K20me1 in mouse and human is found in the promoter of active genes reflecting distinct functions according to different methylation levels [12,14e16]. Using the genome-wide analyses, we show here that histone turnover occurs primarily at TSS and TES of actively transcribed genes and the active histone turnover is highly correlated with the presence of both H3K56Ac and H4K20me1 in fission yeast.

Scientific (cat.5143). Prepared libraries were sequenced on a HiSeq2500 with singleend method (50-bp reads) and raw reads were aligned to S. pombe genome (ASM294v2) using NovoAlign (Novocraft technologies, Malaysia). In case of H3K56Ac and H3 in S. cerevisiae, sequenced reads were aligned to mapped to the SacCer3 using Bowtie2 [21]. Significantly enriched peaks were identified using Homer [22], and mapped reads were quantified using DeepTools [23] and bwtool [24].

2. Materials and methods

Cells were arrested in G2/M by growing for 3 h at 37  C. Total RNA was prepared from approximately 1  107 cells by ‘hot phenol’ method [25]. mRNA was isolated from 5 ug of total RNA using NEBNext® Poly (A) mRNA Magnetic Isolation Module (NEB #E7490). mRNA-seq libraries were made using NEXTflexTM Rapid RNA-Seq kit (cat. 5138) from isolated mRNA. mRNA library was sequenced using a HiSeq2500 with single-end method (50-bp reads) and raw reads were aligned to Schizosaccharomyces pombe genome (ASM294v2) using STAR aligner [26]. The number of mapped reads was further analyzed using HOMER and DESeq [27]. Transcriptomes in G2-arrested cells of S. pombe and asynchronized cells of Saccharomyces cerevisiae were respectively classified as high (More than 75th percentile), low (less than 25th percentile), and intermediate (25th to 75th percentile) levels of expression. Genes corresponding to the lower 5% and upper 5% were omitted.

2.1. Yeast strains and plasmids All yeast strains used in this research are listed in Supplementary Table S1. Deletion strains were constructed using standard PCR-based method [17]. Strains used in histone turnover experiments were transformed by electroporation method to bear the pINV1-H3.2-Flag plasmid from Dr. Grewal. 2.2. Histone turnover experiment Cells were cultured in 240 ml PMG-LEU þ 8% glucose for overnight at 25  C. When cell numbers reached ~2e4  106 cells/ml, cells were pelleted and resuspended in pre-warmed (37  C) 440 ml PMG-LEU þ 8% glucose. Cells were synchronized to G2/M boundary by growing for 3 h at 37  C. Finally, media was changed to 220 ml PMG-LEU þ 4% sucrose to induce expression of H3.2-Flag under the inv1 promoter. Cell were cultured at 37  C for 2 h. 2.3. Fast chromatin immunoprecipitation (ChIP) and quantitative real time PCR Fast ChIP was conducted as previously described [18]. To analyze the DNA obtained from fast ChIP method, quantitative PCR (qPCR) analysis was performed. Each sample was prepared and performed in triplicate. ChIP DNA and cDNA were analyzed by real-time PCR using BioFACT™ Real-Time PCR Kit (BioFACT, cat. DQ116) and CFX96 real-time System (Bio-Rad). 2.4. Antibodies For chromatin immunoprecipitation (ChIP) and western blot analysis, a-H3K56Ac (Abcam ab76307), a-H4K20me1 (Abcam ab9051), a-H4K20me3 (Abcam ab9053) and a-FLAG M2 monoclonal (sigma-aldrich, F1804) were used. a-H3, rabbit polyclonal antibody was made in-house as previously described [19]. 2.5. Western blotting Protein preparation was performed as previously described [20]. The boiled samples were analyzed by SDS-PAGE (15% acrylamide gels) and detected using each antibodies. 2.6. Genome-wide ChIP-seq and data processing ChIP DNA for ChIP-seq library preparation was obtained using fast ChIP method with minor modifications. The chromatin bound to beads was eluted using SigmaPrep spin column (Sigma) with elution buffer instead of Chelex 100. Qiagen PCR purification kit was used to obtain de-crosslinked DNA. ChIP-seq libraries for genome-wide sequencing were prepared using 10 ng of input DNA or 1e10 ng of eluted ChIP DNA using NEXTflex ChIP-Seq kit, Bioo

2.7. mRNA-seq and data processing

3. Results 3.1. DNA replication-independent histone turnover occurs primarily at 50 promoter and 30 termination regions in a transcription dependent manner To investigate histone turnover, we used yeast cells ectopically expressing C-terminal Flag-tagged H3.2 (hhtþ2) under the control of sucrose-inducible promoter (inv1þ), which can be rapidly and fully activated in sucrose-based medium [28,29]. Cells carrying a temperature sensitive cdc25þ allele were cultured at the permissive temperature (25  C) and then arrested in G2/M for 3 h at the restrictive temperature (37  C). H3-Flag was induced under G2/M arrest conditions to minimize the effect of replication-dependent histone deposition (Supplementary Fig. S1). We performed chromatin immunoprecipitation (ChIP) against Flag-tagged H3 induced in a sucrose medium, and followed by high-throughput sequencing to generate a genome-wide map showing deposition of newlysynthesized H3 in a replication-independent manner. Histone turnover was measured as enrichment of Flag signals, normalized to endogenous H3 in G2-arrested wild-type cells. We observed widespread exchange of histones in euchromatic regions (Fig. 1A see the positive values for each promoter region). Consistent with previous studies, newly synthesized histone H3 was primarily incorporated into the 50 promoter and 30 termination regions of transcribed genes, while Flag-H3 incorporation over coding regions showed relatively lower turnover rates (Fig. 1B) [10,30]. In particular, incorporation of newly synthesized histones at 50 promoters was correlated with higher transcription rate (Fig. 1B. Compare red line and grey line), indicating that histone turnover independent of DNA replication is involved in transcriptional activity. To further examine the patterns of histone dynamics in the whole-genome, we clustered them using k-means clustering according to the similarity of histone turnover signals (Fig. 1C). Average profiles of cluster 1 and 2 showed similar pattern to that of canonical average profiles with peaks near the 50 and 3’-end of

H. Yang et al. / Biochemical and Biophysical Research Communications 476 (2016) 515e521

Chromosome 2

A 2,260 kb

2,270 kb

2,280 kb

2 0 -2 SPBC15C4.05.1

taf10

SPBC15C4.06c.1

B

dbp7

sfc6

SPBC21H7.03c.1

SPBC16H5.14c.1

opi10

C high intermediate low

omh2 SPBC16H5.08c.1

1.0 0 -1.0

TSS

TES

0.5kb

TES 0.5kb

Gene length >2kb 1kb~2kb 500bp~1kb <500bp

1

-1 TSS

TES

0.5

0

cluster3

-0.5

cluster1

0

1.5

1.0

cluster2

-1

-0.5

prp43

2.0

0

D log2(H3-Flag/H3)

cluster1 cluster2 cluster3

-0.5 TSS

-0.5

Histone turnover

SPBC16H5.12c.1 skb1

SPBC16H5.13.1

log2(H3-Flag/H3)

log2(H3-Flag/H3)

Histone turnover

Expression 1

his5

Histone exchange

log2(H3-Flag/H3)

2,250 kb

Histone turnover

517

-1.0

-1.5

0.5kb

Fig. 1. DNA replication-independent histone turnover occurs highly at 50 /30 -end of genes in a transcription dependent manner. (A) A browser view of histone turnover signals along with a 30 kb region of chromosome 2. Histone turnover values were measured by ChIP-seq. Results of Flag-tagged H3 induced in a sucrose medium and endogenous H3 (Flag versus H3) are shown. (B) Average profile of histone turnover across the genes. TSS and TES indicate transcription start site and transcription termination site, respectively. The plots were stratified according to the expression levels defined in the Materials and methods section. (C) Heat map showing the distribution of histone turnover in the protein-coding genes. Profiles were divided into three clusters using the basic K-means algorithm (k ¼ 3). Top, clustered average profiles. Heat map colors represent histone turnover values of (A). (D) Average profile of histone turnover plotted on the basis of the indicated ORF length.

genes, whereas cluster 3 showed increased incorporation of newly synthesized histone H3 in the coding regions as well as promoter. About 65% of the genes in cluster 3 had RNA coding regions shorter than 1 kb (Supplementary Fig. S2A). Indeed, when we compared the level of histone turnover according to the gene length, genes smaller than 500 bp showed higher turnover rate in the coding regions while histone turnover at promoter regions was not affected by gene length (Fig. 1D). To further investigate the properties of long genes in cluster 3, we checked the dependency of histone turnover on transcription rate since histone turnover in coding regions has been reported to be dependent on transcription level [11]. Interestingly, we could not observe a correlation between histone turnover in the coding regions and transcription level in this cluster (Supplementary Fig. S2), suggesting the existence of another driving force for nucleosome dynamics in the absence of transcription and replication. Evidence for other functions for histone turnover, such as incorporation of mammalian H3.3 for epigenetic memory of an active gene state, has accumulated [31]. We thus examined histone turnover marks that are known to be associated with histone exchange such as H3K56Ac and H4K20me in G2/M phase to more precisely measure replication-independent histone turnover. H3K56Ac is a representative histone modification involved in

replication-dependent and -independent histone deposition [10,32], but a causal relationship between H3K56Ac and histone turnover has not yet been determined in fission yeast. In contrast to H3K56Ac, which marks active histone turnover, H4K20me2/3 are present in old nucleosomes and H4K20me3 is negatively correlated with histone turnover [11,12,33]. However, the genomic location of H4K20me1 according to transcription rate is still controversial. Moreover, the relationship between H4K20me1 and histone turnover is poorly understood. Therefore, we decided to revisit these histone turnover marks under G2/M-arrested conditions. 3.2. Histone turnover is positively correlated with H3K56Ac and H4K20me1 We performed ChIP-seq experiments against H3K56Ac, H4K20me1 and H4K20me3 in G2/M-arrested cells and extracted values from the region spanning 300 bp upstream of the transcription start site (TSS) showing high turnover rate. These data were normalized to the level of H3 and hierarchically clustered by the Pearson correlation (Fig. 2A). Consistent with previous studies, H3K56Ac showed a strong correlation with histone turnover rate (Pearson correlation ¼ 0.7) independent of DNA replication, whereas H4K20me3 was inversely correlated (Pearson

518

H. Yang et al. / Biochemical and Biophysical Research Communications 476 (2016) 515e521

A

B -0.45

-0.31

0.7

1

Histone turnover

−0.45

0.4

1

0.7

Pearson Correlation

H4K20me1/H3

−0.52

1

0.4

0.31

1.0 0.5 0.0 -0.5

H4K20me3/H3

1

H3K56Ac/H3

Histone turnover log2(H3-Flag/H3)

H3K56Ac log2(H3K56Ac/H3)

H4K20me1

−0.52 −0.45 −0.45

log2(H4K20me1/H3)

2 0 -2 4 0 -4 3 0 -3

3 3 3 er c/H 1/H 3/H ov 6A turn 0me 0me 5 K 2 2 e 3 K K H ton H4 H4 His

C

sec73

H3K56Ac

-0.5 -1.5

TSS

TES

log2(H3K56Ac/H3)

0.5

-0.5

npr2

S. cerevisiae high intermediate low

1.5

0 -1

-0.5

TSS

TES

0.5kb

E 0.5 -0.5 -1.5

-0.5

F

TSS

TES

12 10 8 6 4 2

e ton r His ove n r tu

H3

K5

c 6A

1 me

0

K2

H4

0 -1

-0.5

5’end

14

0

K2

H4

3 me

high intermediate low

1

0.5kb

Relative enrichment normalized to H3 (log2 value)

Relative enrichment normalized to H3 (log2 value)

H4K20me1/H3 H3K56Ac/H3

1.5

H4K20me1

log2 enrichment

D

high intermediate low

1

0.5kb

log2(H4K20me1/H3)

H3K56Ac

log2(H3K56Ac/H3)

S. pombe

ini1

TSS

TES

0.5kb

3’end

14 12 10 8 6 4 2

e ton r His ove n r tu

e3

e1

c

6A

K5

H3

0m

K2

H4

0m

K2

H4

Fig. 2. Histone turnover is positively correlated with H3K56Ac and H4K20me1. (A) Pearson correlation heat map matrix of ChIP signals. Average log2 ratios were measured across the region spanning over 300 bp upstream of TSS. Histone turnover values were calculated, as in Fig. 1(A). All ChIP experiments were performed at G2/M arrested condition. (B) ChIP data of histone turnover, H3K56Ac, and H4K20me1 at the sec73 and npr2 loci. (C) Average profile of H3K56Ac in S. pombe (left panel) and S. cerevisiae (right panel), as in Fig. 1(B). (D) Average profile of H3K56Ac and H4K20me1 across the genes. (E) Average profile of H4K20me1, as in Fig. 1(B). (F) Boxplots showing the distribution of each ChIP signal corresponding to the 50 end (left panel) and 30 end (right panel) of genes. 50 end: spanning over 300 bp upstream of TSS. 30 end: spanning over 300 bp downstream of TES.

correlation ¼ minus 0.45). However, surprisingly, we found that H4K20m1 in G2/M phase is positively correlated with histone turnover ((Pearson correlation ¼ 0.4), which implicates dual aspects of H4K20 methylation as a mark for histone turnover. These correlation was explored on a genome-wide scale using Integrative Genomic Viewer (IGV) (Fig. 2B, Supplementary Fig. S3) [34]. To further characterize histone turnover marks including H3K56Ac and H4K20me1 in S. pombe, we first compared the pattern of H3K56 acetylation in fission versus budding yeast. Similar to H3K56Ac in S. cerevisiae [10], H3K56Ac in S. pombe was also enriched around the transcription start site (TSS) of transcribed genes in a transcription-dependent manner suggesting the conserved role of H3K56Ac in transcription-dependent histone turnover (Fig. 2C). It is also noteworthy that H3K56Ac in S. pombe is incorporated into 30 end as well as 50 end of transcribed genes, indicating the occurrence of dynamic histone turnover in the nucleosome-depleted region (NDR).

We next asked whether H4K20me1 can reflect the pattern of H3K56Ac and histone turnover that are associated with transcriptional activity. Average profile of H4K20me1 showed overlapping colocalization with H3K56Ac in the chromatin context with high histone turnover (Fig. 2D). In addition, the genomic distribution of H4K20me1 also depended on transcription rate, indicating that H4K20me1 has a direct relationship between H3K56Ac and histone turnover (Fig. 2E). It thus appears that histone turnover in wild-type is high at 5’/ 3’-end NDR regions of transcribed genes and low over the coding regions, and histone turnover is consistent with the levels of H3K56Ac and H4K20me1. Meanwhile, H4K20me3 has been shown to be enriched in the coding regions of lowly transcribed genes [14]. When we compared histone turnover and the histone modifications, rapid histone turnover at 5’/3’-end of genes was clearly consistent with H3K56Ac and H4K20me1 signals (Fig. 2F), while slow histone turnover at coding regions was correlated with the

H. Yang et al. / Biochemical and Biophysical Research Communications 476 (2016) 515e521

log2(H3K56Ac/H3)

0.5

B

alp13Δ/wt set2Δ/wt

log2(H4K20me1/H3)

A

0

-0.5

-1.0 TES

0.4 0 -0.4 -0.8

D

TSS

TES

0.4

0.5kb

high intermediate low

0.2 0 -0.2 -0.4 -0.6

-0.5

TSS

TES

0.5kb

high intermediate low

0.4 0.2 0 -0.2 -0.4

-0.5

TSS

TES

0.5kb

F log2(H4K20me1/H3) (set2Δ/wt)

log2(H3K56Ac/H3) (alp13Δ/wt)

-0.25

-0.5

high intermediate low

E log2(H4K20me1/H3) (alp13Δ/wt)

0

0.5kb

log2(H3K56Ac/H3) (set2Δ/wt)

TSS

C

-0.6

alp13Δ/wt set2Δ/wt

0.25

-0.50 -0.5

-1.2

519

high intermediate low

0.1 0 -0.1 -0.2

-0.5

TSS

TES

0.5kb

-0.5

TSS

TES

0.5kb

Fig. 3. Increased levels of H3K56Ac and H4K20me1 are highly associated with histone turnover in Set2 and HDAC mutants. Average profiles were plotted as a ratio of mutant versus the wild-type. Average plot in whole genes of H3K56Ac (A), H4K20me1 (B). The data in (A) was separated according to expression levels in alp13D (C) and set2D (D). The data in (B) was also separated according to expression levels in alp13D (E) and set2D (F).

increased levels of H4K20me3 (Compare Fig. 2F and Supplementary Fig. S4). Taken together, we clarified the conserved properties of histone turnover and -associated makers in fission yeast and, importantly, we found a new function of H4K20me1 as a histone turnover mark.

3.3. H3K56Ac and H4K20me1 mark increased histone turnover in coding regions in Set2 and HDAC mutants To validate the role of H4K20me1, we checked whether H4K20me1 would reflect the change of histone turnover rate in the absence of histone methyltransferase Set2. In budding yeast, H3K36 methylation by Set2 suppresses histone turnover in the coding region by recruiting Rpd3S histone deacetylase complex (HDAC). The hyperacetylated nucleosomes of coding regions in the set2D cells and HDAC mutants become highly dynamic displaying elevated levels of H3K56Ac, which eventually leads to aberrant transcription within the coding regions [35,36]. In S. pombe, Set2 and the Rpd3S homolog, Alp13, are also required to prevent antisense transcription from cryptic promoters within the coding regions [37]. We thus examined the levels of H3K56Ac in strains lacking Set2 and Alp13. As expected, H3K56Ac was increased in the coding regions while there is no significant change in global H3K56Ac level (Fig. 3A, Supplementary Fig. S5A). Interestingly, H4K20me1 was also increased within the coding regions, similar to the pattern of

H3K56Ac in both mutants (Fig. 3B), suggesting that newly deposited nucleosomes contain mono-methylated H4K20. Deletion of set2þ and alp13þ did not affect the global H4K20me1 levels, but slight reduction of H4K20me3 levels was observed in alp13D (Supplementary Fig. S5B). It is known that the Set2-Rpd3 S pathway is required for suppression of cryptic transcription on lowly transcribed genes [38]. Consistent with this, the increase of H3K56Ac in set2 and alp13 mutants was higher at infrequently transcribed genes (Fig. 3C, D). H4K20me1 also reflected the result of K56 acetylation exhibiting more significant increase at Set2/Rpd3S-dependent genes (Fig. 3E, F). Collectively, these data suggest that H4K20me1 in set2 and alp13 mutants can mark transcription-dependent histone turnover associated with cryptic transcription.

3.4. H4K20me1 is associated with transcription-dependent histone turnover To address the role of K20me1 as a transcription-dependent histone turnover mark, we performed mRNA-seq in either set2D or alp13D cells, then observed aberrant antisense RNA (asRNA) production in these mutants. If H4K20me1 acts as a positive mark of transcription-dependent histone turnover, the increase of aberrant transcription by RNA polymerase II (RNAPII) in set2D and alp13D cells should be correlated with the increase of H4K20me1 levels. We identified genes harboring both K56Ac peaks and

520

H. Yang et al. / Biochemical and Biophysical Research Communications 476 (2016) 515e521

Fig. 4. H4K20me1 marks transcription-dependent histone turnover. (A) Venn diagram of the genes exhibiting increased signals of H3K56Ac and H4K20me1 in alp13D compared to wild-type (log2 [alp13D/wt] > 0.5). (B) Venn diagram shows the overlaps between the overlapped genes (n ¼ 1849) in (A) and genes that produce antisense RNAs in alp13D. (C) Boxplots showing distribution of asRNAs, H3K56Ac, and H4K20me1 levels in alp13D and set2D, compared with those in wild-type. ChIP signals of H3K56Ac, H4K2me1 were normalized to H3 in the coding regions. Asterisk (**) means p-value < 0.0001.

K20me1 peaks that were increased in alp13D compared to wildtype (Fig. 4A). Indeed, genes with asRNAs in alp13D were significantly overlapping with genes marked by these two modifications (Fig. 4B). To further examine asRNAs and histone turnover marks, we selected 1400 genes that produce antisense RNAs in both mutants, and compared the levels of H4K20me1and H3K56Ac within coding regions. Consistent with the above results, production of asRNAs was accompanied with the increase of the histone turnover marks (Fig. 4C). The increase of histone turnover and H4K20me1and H3K56Ac in a specific genomic site was also confirmed by ChIP-qPCR analyses in the two mutants (Fig. 4C, Supplementary Fig. S6). Taken together, we conclude that transcription-dependent histone turnover can be marked by H4K20me1 as well as H3K56Ac. 4. Discussion In this study, we generated a genome-wide map of DNA replication-independent histone turnover using a sucroseinducible system in fission yeast. Consistent with previous studies, histone exchange occurred highly at 5’/3’-end NDR regions of transcribed genes, and was dependent on the transcription rate. H3K56Ac was also correlated with the pattern of histone turnover. So far, there is no direct evidence for the role of H3K56Ac as a histone turnover mark in fission yeast, but our data clearly show that this modification is highly associated with histone turnover in fission yeast. Unexpectedly, we found that H4K20me1 is also correlated with

high histone turnover rate in wild-type cells. Since we could observe this correlation in G2/M-arrested cells, it would be interesting to check whether H4K20me1 can be more suitable for uniquely representing replication-independent histone turnover. Indeed, it is difficult to distinguish the relative contribution of H3K56Ac to replication-dependent and eindependent histone turnover because the correlation between H3K56Ac and histone exchange is observed even in unsynchronized cells (Data not shown). H4K20me1 marks histone turnover independent of replication and shows an increased signal according to transcription rate. We speculate that histones incorporated at promoters during transcription are transiently monomethylated, and then further methylated into H4K20me2/3 on a stable nucleosome. The role of H4K20me1 as a transcription-dependent histone turnover mark was confirmed by correlation with H3K56 acetylation levels in set2 and HDAC mutants. These mutants resulted in the increase of antisense RNAs produced from the coding regions, which is consistent with the increased levels of H3K56Ac and H4K20me1 as well as higher histone turnover in the regions. Most of all, our results uncover H4K20me1 as a novel mark of transcriptiondependent histone turnover in fission yeast. Acknowledgements This work was supported by grants from the Stem Cell Research Program (2011e0019509 and 2012M 3A9B 4027953), the KAIST Future Systems Healthcare Project, and the Intelligent Synthetic Biology Center of Global Frontier Project funded by the Ministry of

H. Yang et al. / Biochemical and Biophysical Research Communications 476 (2016) 515e521

Science, ICT and Future Planning (2011-0031955). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2016.05.155. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2016.05.155. References [1] K. Luger, M.L. Dechassa, D.J. Tremethick, New insights into nucleosome and chromatin structure: an ordered state or a disordered affair? Nat. Rev. Mol. Cell Biol. 13 (2012) 436e447. [2] J.L. Workman, Nucleosome displacement in transcription, Genes Dev. 20 (2006) 2009e2017. [3] S. Venkatesh, J.L. Workman, Histone exchange, chromatin structure and the regulation of transcription, Nat. Rev. Mol. Cell Biol. 16 (2015) 178e189. [4] H. Tagami, D. Ray-Gallet, G. Almouzni, Y. Nakatani, Histone H3. 1 and H3. 3 complexes mediate nucleosome assembly pathways dependent or independent of DNA synthesis, Cell 116 (2004) 51e61. [5] A. Verreault, P.D. Kaufman, R. Kobayashi, B. Stillman, Nucleosome assembly by a complex of CAF-1 and acetylated histones H3/H4, Cell 87 (1996) 95e104. [6] C.D. Kaplan, L. Laprade, F. Winston, Transcription elongation factors repress transcription initiation from cryptic sites, Science 301 (2003) 1096e1099. [7] R. Pavri, B. Zhu, G. Li, P. Trojer, S. Mandal, A. Shilatifard, D. Reinberg, Histone H2B monoubiquitination functions cooperatively with FACT to regulate elongation by RNA polymerase II, Cell 125 (2006) 703e717. [8] S. Carvalho, A.C. Raposo, F.B. Martins, A.R. Grosso, S.C. Sridhara, J. Rino, M. Carmo-Fonseca, S.F. de Almeida, Histone methyltransferase SETD2 coordinates FACT recruitment with nucleosome dynamics during transcription, Nucleic Acids Res. 41 (2013) 2881e2893. [9] J. Han, H. Zhou, B. Horazdovsky, K. Zhang, R.-M. Xu, Z. Zhang, Rtt109 acetylates histone H3 lysine 56 and functions in DNA replication, Science 315 (2007) 653e655. [10] A. Rufiange, P.-E. Jacques, W. Bhat, F. Robert, A. Nourani, Genome-wide replication-independent histone H3 exchange occurs predominantly at promoters and implicates H3 K56 acetylation and Asf1, Mol. Cell 27 (2007) 393e405. [11] J. Schneider, P. Bajwa, F.C. Johnson, S.R. Bhaumik, A. Shilatifard, Rtt109 Is Required for Proper H3K56 Acetylation A CHROMATIN MARK ASSOCIATED WITH THE ELONGATING RNA POLYMERASE II, J. Biol. Chem. 281 (2006) 37270e37274. [12] A.N. Scharf, T.K. Barth, A. Imhof, Establishment of histone modifications after chromatin assembly, Nucleic Acids Res. 37 (2009) 5032e5040. [13] O. Aygün, S. Mehta, S.I. Grewal, HDAC-mediated suppression of histone turnover promotes epigenetic stability of heterochromatin, Nat. Struct. Mol. Biol. 20 (2013) 547e554. [14] J.P. Svensson, M. Shukla, V. Menendez-Benito, U. Norman-Axelsson, P. Audergon, I. Sinha, J.C. Tanny, R.C. Allshire, K. Ekwall, A nucleosome turnover map reveals that the stability of histone H4 Lys20 methylation depends on histone recycling in transcribed chromatin, Genome Res. 25 (2015) 872e883. [15] H. Talasz, H.H. Lindner, B. Sarg, W. Helliger, Histone H4-lysine 20 monomethylation is increased in promoter and coding regions of active genes and correlates with hyperacetylation, J. Biol. Chem. 280 (2005) 38814e38822. [16] C.R. Vakoc, M.M. Sachdeva, H. Wang, G.A. Blobel, Profile of histone lysine methylation across transcribed mammalian chromatin, Mol. Cell. Biol. 26

521

(2006) 9185e9195. [17] A.B. Steever, A. Wach, P. PHILIPPSEN, J.R. Pringle, Heterologous modules for efficient and versatile PCR-based gene targeting in Schizosaccharomyces pombe, Yeast 14 (1998) 943e951. [18] J.D. Nelson, O. Denisenko, K. Bomsztyk, Protocol for the fast chromatin immunoprecipitation (ChIP) method, Nat. Protoc. Electron. Edition 1 (2006) 179. [19] S. Oh, K. Jeong, H. Kim, C.S. Kwon, D. Lee, A lysine-rich region in Dot1p is crucial for direct interaction with H2B ubiquitylation and high level methylation of H3K79, Biochem. Biophys. Res. Commun. 399 (2010) 512e517. [20] R.M. Ricke, A.-K. Bielinsky, Easy detection of chromatin binding proteins by the histone association assay, Biol. Proced. Online 7 (2005) 60e69. [21] B. Langmead, S.L. Salzberg, Fast gapped-read alignment with Bowtie 2, Nat. Methods 9 (2012) 357e359. [22] S. Heinz, C. Benner, N. Spann, E. Bertolino, Y.C. Lin, P. Laslo, J.X. Cheng, C. Murre, H. Singh, C.K. Glass, Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities, Mol. Cell 38 (2010) 576e589. [23] F. Ramírez, F. Dündar, S. Diehl, B.A. Grüning, T. Manke, deepTools: a flexible platform for exploring deep-sequencing data, Nucleic Acids Res. 42 (2014) W187eW191. [24] A. Pohl, M. Beato, bwtool: a tool for bigWig files, Bioinformatics 30 (2014) 1618e1619. [25] M.E. Schmitt, T.A. Brown, B.L. Trumpower, A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae, Nucleic Acids Res. 18 (1990) 3091. [26] A. Dobin, C.A. Davis, F. Schlesinger, J. Drenkow, C. Zaleski, S. Jha, P. Batut, M. Chaisson, T.R. Gingeras, STAR: ultrafast universal RNA-seq aligner, Bioinformatics 29 (2013) 15e21. [27] S. Anders, W. Huber, Differential expression analysis for sequence count data, Genome Biol. 11 (2010) R106. [28] J.S. Iacovoni, P. Russell, F. Gaits, A new inducible protein expression system in fission yeast based on the glucose-repressed inv1 promoter, Gene 232 (1999) 53e58. [29] E.S. Choi, J.A. Shin, H.S. Kim, Y.K. Jang, Dynamic regulation of replication independent deposition of histone H3 in fission yeast, Nucleic Acids Res. 33 (2005) 7102e7110. [30] D.C. Kraushaar, W. Jin, A. Maunakea, B. Abraham, M. Ha, K. Zhao, Genomewide incorporation dynamics reveal distinct categories of turnover for the histone variant H3. 3, Genome Biol. 14 (2013) R121. [31] R.K. Ng, J. Gurdon, Epigenetic memory of an active gene state depends on histone H3. 3 incorporation into chromatin in the absence of transcription, Nat. Cell Biol. 10 (2008) 102e109. [32] Q. Li, H. Zhou, H. Wurtele, B. Davies, B. Horazdovsky, A. Verreault, Z. Zhang, Acetylation of histone H3 lysine 56 regulates replication-coupled nucleosome assembly, Cell 134 (2008) 244e255. [33] J.J. Pesavento, H. Yang, N.L. Kelleher, C.A. Mizzen, Certain and progressive methylation of histone H4 at lysine 20 during the cell cycle, Mol. Cell. Biol. 28 (2008) 468e486.  ttir, W. Winckler, M. Guttman, E.S. Lander, [34] J.T. Robinson, H. Thorvaldsdo G. Getz, J.P. Mesirov, Integrative genomics viewer, Nat. Biotechnol. 29 (2011) 24e26. [35] S. Venkatesh, M. Smolle, H. Li, M.M. Gogol, M. Saint, S. Kumar, K. Natarajan, J.L. Workman, Set2 methylation of histone H3 lysine 36 suppresses histone exchange on transcribed genes, Nature 489 (2012) 452e455. [36] C.K. Govind, H. Qiu, D.S. Ginsburg, C. Ruan, K. Hofmeyer, C. Hu, V. Swaminathan, J.L. Workman, B. Li, A.G. Hinnebusch, Phosphorylated Pol II CTD recruits multiple HDACs, including Rpd3C (S), for methylation-dependent deacetylation of ORF nucleosomes, Mol. Cell 39 (2010) 234e246. [37] E. Nicolas, T. Yamada, H.P. Cam, P.C. FitzGerald, R. Kobayashi, S.I. Grewal, Distinct roles of HDAC complexes in promoter silencing, antisense suppression and DNA damage protection, Nat. Struct. Mol. Biol. 14 (2007) 372e380. [38] B. Li, M. Gogol, M. Carey, S.G. Pattenden, C. Seidel, J.L. Workman, Infrequently transcribed long genes depend on the Set2/Rpd3S pathway for accurate transcription, Genes Dev. 21 (2007) 1422e1430.