Reprogramming of Chromatin Accessibility in Somatic ... - Cell Press

Report

Reprogramming of Chromatin Accessibility in Somatic Cell Nuclear Transfer Is DNA Replication Independent Graphical Abstract

Authors reprogrammed 1cell

SCNT donor cumulus cell

12 hours with or without Aphidicolin

Mohamed Nadhir Djekidel, Azusa Inoue, Shogo Matoba, ..., Falong Lu, Lan Jiang, Yi Zhang

Correspondence [email protected]

No. of DHS 14924

closed

open 179

open resistant

closed

open 524

closed

closed 262

resemble paternal pronucelus

open

Highlights d

Genome-wide DHS maps for donor cumulus cells, one-cell SCNT, and IVF embryos

d

Rapid and DNA replication-independent DHS reprogramming in SCNT embryos

d

Transcription factor network switch accompanies DHS reprogramming in SCNT embryos

d

Loss of donor cell DHSs correlates with suppression of somatic transcription program

Djekidel et al., 2018, Cell Reports 23, 1939–1947 May 15, 2018 ª 2018 The Authors. https://doi.org/10.1016/j.celrep.2018.04.036

In Brief Djekidel et al. used low-input DNase-seq to map the chromatin accessibility dynamics of donor cells and SCNT onecell embryos. They revealed a drastic and fast global DHS reprogramming of donor cells in a DNA replication-independent manner.

somatic memory

successful reprogramiming

Replication-independent reprogramming of DHSs

Data and Software Availability GSE110851

Cell Reports

Report Reprogramming of Chromatin Accessibility in Somatic Cell Nuclear Transfer Is DNA Replication Independent Mohamed Nadhir Djekidel,1,2,3,9 Azusa Inoue,1,2,3,6,9 Shogo Matoba,1,2,3,7 Tsukasa Suzuki,1,2,3 Chunxia Zhang,1,2,3 Falong Lu,1,2,3,8 Lan Jiang,1,2,3 and Yi Zhang1,2,3,4,5,10,* 1Howard

Hughes Medical Institute, Boston Children’s Hospital, Boston, MA 02115, USA in Cellular and Molecular Medicine, Boston Children’s Hospital, Boston, MA 02115, USA 3Division of Hematology/Oncology, Department of Pediatrics, Boston Children’s Hospital, Boston, MA 02115, USA 4Department of Genetics, Harvard Medical School, Boston, MA 02115, USA 5Harvard Stem Cell Institute, WAB-149G, 200 Longwood Avenue, Boston, MA 02115, USA 6Present address: RIKEN Center for Integrative Medical Sciences, Yokohama, Kanagawa 230-0045, Japan 7Present address: RIKEN Bioresource Center, Tsukuba, Ibaraki 305-0074, Japan 8Present address: State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China 9These authors contributed equally 10Lead Contact *Correspondence: [email protected] https://doi.org/10.1016/j.celrep.2018.04.036 2Program

SUMMARY

Mammalian oocytes have the ability to reset the transcriptional program of differentiated somatic cells into that of totipotent embryos through somatic cell nuclear transfer (SCNT). However, the mechanisms underlying SCNT-mediated reprogramming are largely unknown. To understand the mechanisms governing chromatin reprogramming during SCNT, we profiled DNase I hypersensitive sites (DHSs) in donor cumulus cells and one-cell stage SCNT embryos. To our surprise, the chromatin accessibility landscape of the donor cells is drastically changed to recapitulate that of the in vitro fertilization (IVF)derived zygotes within 12 hr. Interestingly, this DHS reprogramming takes place even in the presence of a DNA replication inhibitor, suggesting that SCNTmediated DHS reprogramming is independent of DNA replication. Thus, this study not only reveals the rapid and drastic nature of the changes in chromatin accessibility through SCNT but also establishes a DNA replication-independent model for studying cellular reprogramming. INTRODUCTION Among the currently available systems for cell fate reprogramming, somatic cell nuclear transfer (SCNT) is the only one capable of reprogramming terminally differentiated cells to a totipotent state (Jullien et al., 2011; Mitalipov and Wolf, 2009). SCNT therefore provides an excellent model for understanding how cell memory can be fully reprogrammed to generate totipotent cells, and thus can provide important clues on how to improve other reprogramming systems. However, despite more than 50 years

after the first successful cloning by SCNT (Gurdon, 1962), the molecular mechanisms underlying SCNT-mediated reprogramming are almost completely unknown. Reprogramming requires change to the chromatin, epigenetic, and transcriptional landscapes of somatic cells. Many studies have been performed to characterize these changes during the induced pluripotent stem cell (iPSC) reprogramming process. These studies used different assays including RNA sequencing (RNA-seq), chromatin immunoprecipitation sequencing (ChIP-seq), assays for transposase-accessible chromatin using sequencing (ATAC-seq), Hi-C, and proteomics analyses (Hussein et al., 2014; Knaupp et al., 2017; Koche et al., 2011; Krijger et al., 2016; Li et al., 2017; Sridharan et al., 2013; Stadhouders et al., 2018). The studies revealed the dynamic nature of the chromatin, epigenetics, and transcriptome during the iPSC generation process and identified important factors and molecular events that facilitate or impede the reprogramming process. While such multi-dimensional analyses have been applied to the iPSC reprogramming system, only transcriptome analyses have been performed for SCNT reprogramming (Chung et al., 2015; Hormanseder et al., 2017; Inoue et al., 2015; Matoba et al., 2014). Although these studies revealed that a donor cell transcriptional program is largely reprogrammed to an embryonic program by the time of zygotic genome activation (ZGA), with the exception of reprogramming-resistant regions (Chung et al., 2015; Matoba et al., 2014), its molecular basis is still unknown and further study of the chromatin landscape changes during the reprogramming process is necessary. Chromatin accessibility is a good indicator of transcriptional regulatory elements and can serve as a predictor of gene transcription activity. It can be identified genome-wide by DNase I sequencing or ATAC-seq (Boyle et al., 2008; Buenrostro et al., 2013). Recent refinements to these techniques have allowed the profiling of the open chromatin landscape using limited number of cells by low-input DNase I sequencing (liDNase-seq) or at

Cell Reports 23, 1939–1947, May 15, 2018 ª 2018 The Authors. 1939 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

A

C Cumulus cell

Cumulus

SCNT

IVF (Pat)

G2 phase Activation 12 h

OO n=3,092

SCNT

MII oocyte G2 phase IVF 12 h Paternal pronucleus (Pat)

B

PCA clustering of DHS profiles 100

OC n=14,924

Cumulus1 Cumulus2

PC2 (3.13%)

RPKM

15 10 5 0

0

Pat3

−100

rOC n=524

Pat1 SCNT1

rCO n=262 CO n=179

Pat2

−200

SCNT2

−1600

−1200

−800

5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb 5kb

−400

PC1 (94.6%)

D

IVF(Pat)

SCNT Cumulus

OO 350

1

Wipi1

0

350

OC

Mesdc2

0

400

Ccdc163

0

400

rOC

Zyx

0

350

Fstl3

0

350

rCO

Mfsd4b4

0

150

Scn8a

0

175

CO

Hivep3

0

250

Atxn7

0

150

Nav2

0

2 1 2 1 2 5 kb

5 kb

5 kb

5 kb

5 kb

5 kb

2 kb

5 kb

2 kb

5 kb

Figure 1. Rapid Reprogramming of Donor Cell Chromatin Accessibility in SCNT (A) Schematic illustration of the experimental design for studying the chromatin accessibility dynamics in SCNT and that of the paternal pronucleus from in vitrofertilized (IVF) zygotes. The collected samples for liDNase-seq analysis are shown inside dotted boxes. (B) PCA of the genome-wide DHS profile of the cumulus, IVF(Pat), and one-cell SCNT samples. Each dot represents one sample. (C) Heatmap showing the DHSs clustered according to their combinatorial enrichment in cumulus, one-cell SCNT, and IVF(Pat) samples. Each row represents a locus (DHS center ± 5 kb), and the red gradient color indicates the liDNase-seq signal intensity. OO, open in cumulus cells, SCNT, and IVF(Pan); OC, open in cumulus cells and closed in SCNT and IVF(Pat); rOC, open in cumulus cells and closed in IVF(Pat), but failed to be closed in SCNT; rCO, closed in cumulus cells and open in IVF(Pat), but failed to be opened in SCNT; CO, closed in cumulus cells and open in IVF(Pat) and SCNT. (D) Genome browser views showing the normalized coverage (per million mapped reads) at representative loci of the different DHS categories. See also Figure S1.

the single-cell level by ATAC-seq (Buenrostro et al., 2015; Jin et al., 2015; Lu et al., 2016), thereby facilitating the study of chromatin accessibility in mouse preimplantation embryos (Inoue et al., 2017; Lu et al., 2016; Wu et al., 2016). In this work, we used liDNase-seq to study chromatin accessibility changes during SCNT reprogramming, which revealed the quick and DNA replication-independent nature of the reprogramming process.

1940 Cell Reports 23, 1939–1947, May 15, 2018

RESULTS AND DISCUSSION Fast DNase I Hypersensitive Site Reprogramming upon SCNT To understand how the chromatin accessibility of somatic donor cells is reprogrammed to that of the totipotent one-cell embryo, we attempted to generate the DNase I hypersensitive site (DHS)

A

B Genomic distribution of the different DHS categories

OO OC rOC rCO CO

Pathway enrichment of genes in proximity of OO DHS

0

25

50

75

0

Cell Cycle, Mitotic DNA Replication Mitotic M-M/G1 phases Gene Expression Synthesis of DNA Formation and Maturation of mRNA Transcript S Phase Mitotic G1-G1/S phases Processing of Capped Intron-Containing Pre-mRNA Mitotic G2-G2/M phases

Promoter Exon Intron Intergenic

100

Percentage (%)

2

4

6

8 10 12 14 16 18 19.56 16.23 13.50 13.38 11.83 10.41 9.61 9.48 9.27 9.15

-log 10 (Binom ial p-value)

C Pathway enrichment of genes in proximity of OC DHS 0.1

Cumulus

0 −0.1 −0.2

SCNT 2-Cell

H3K9me3 (Chip − Input) RPKM

H3K9me3 enrichment in DHS groups

0.1 0 −0.1 −0.2 −2.5Mb

DHS Center

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

DHS category OO OC rOC rCO CO

Integrins in angiogenesis HIF-1-alpha transcription factor network Hypoxic and oxygen homeostasis regulation of HIF-1-alpha Osteopontin-mediated events IL6-mediated signaling events Alpha4 beta1 integrin signaling events Alpha6Beta4Integrin megakaryocyte development and platelet production Ceramide signaling pathway Retinoic acid receptors-mediated signaling Cell junction organization

28.24 27.16 23.74 20.57 16.95 16.53 14.91 13.83 13.51 10.56 10.25

-log 10 (Binom ial p-value)

2.5Mb

D %HDS with motif

OO OC

5

25

50

10

40

70

rOC

-log10(p-value)

rCO CO

E

200

300

Sf 1 p5 3 Ze b1 Fo xo 3 Kl f5 G fi1 b

f9 Kl

El f Sm 1 ad 3

x3

5

a1 Fo xk 1 D ux bl So x5 AP −1

Fr

Pb

fx R

N

Sp

1

FY

100

F de novo motifs enriched in OC DHS Motif GA

G C A

Best match TF TCC A CAG

TGA TCA TTGTTATGGTTA GATTTGGTCAGC TCAGAATTCAGT TG A

CC T G C

A

C

T A

G C

A

G C C T A A

G C

A

G C

C T

A

G T C

G C

A

G

G C

A

G T

G C

A

A

G C T A C T

T

G T C

A

G T C

C

G T C

A

C T

G T C

A

G C

A

G T

A G T C

G C

A

C T

C T

A

G

A

G T

A

C

C T

G C

A

G C

A

G T

A

G T C

C T

A

C T

T G AGT

A

C

G C

A

G C

A

G T C

A

T G

A

G T

A

G C

A

de novo motifs enriched in CO DHS Motif

p−value % target % background 46.67

C TG C T A AA

1e−150

Foxk1

1e−46

Duxbl

1e−44

0.24

0.01

Sox5

1e−39

0.22

0.01

GT G

G

0.25

Best match TF C

ATTGGCT GG T A C CCTCCC G A G TCGGAGATTGGT CC

9.44

Fra1(bZIP)

0.01

A

C

C

TC G T A T

A

G T

A

C T

T

A

G T C

C C

C T

A

T

G

C G

A

A

C

A

C T

A

G

T A

G

G T

C

TC T

A

G T C

G

G C

A

TCA

G T

G C

A

A

1e−75

66.48

9.59

Klf5

1e−17

17.32

2.34

Gfi1b(Zf)

1e−12

5.59

0.15

TA

C T

A

A

C T

A

G C

A

20

Nfya

FPKM

CO enriched TF expression

Fra1

FPKM

OC enriched TF expression 2.0 1.5 1.0 0.5 0.0 8 6 4 2 0

p−value % target % background

NFY(CCAAT)

A

T G C

G

GAA T

T AGG

A

A

G C

10 0 Cumulus

Oocyte

1-Cell SCNT 1-Cell IVF

Sox5

Cumulus

Oocyte

1-Cell SCNT 1-Cell IVF

(legend on next page)

Cell Reports 23, 1939–1947, May 15, 2018 1941

map of SCNT one-cell embryos. To this end, we collected mouse cumulus cells to serve as somatic donor cells and performed SCNT. Twelve hours post-activation (hpa), pseudopronuclei were isolated from SCNT one-cell embryos for liDNase-seq (Figure 1A) with biological duplicates for both the donor cells and one-cell SCNT embryos (Figures S1A and S1B). Since sperm chromatin is reprogrammed under physiological conditions upon fertilization (Inoue et al., 2017), we used the DHS map of paternal pronuclei (Pat) of 12 hr post-fertilization (hpf) zygotes as a control (Figure 1A). Using stringent criteria for peak calling and reproducibility (irreproducibility discovery rate [IDR] < 0.05, mean reads per kilobase million [RPKM] > 2, RPKM in all replicate > 1, sex chromosomes were excluded), we identified 23,353, 3,005, and 3,610 DHSs in donor cumulus cells, SCNT one-cell embryo, and Pat, respectively (Table S1). Principalcomponent analysis (PCA) indicates that the overall DHS landscape of SCNT embryos is similar to that of Pat (Figure 1B), suggesting that SCNT-mediated reprogramming of chromatin accessibility is largely complete by 12 hr after activation. To closely examine the DHS dynamics, we classified the detected DHSs into five groups using the following criteria (3-fold difference between DHS-negative and -positive categories, mean RPKM > 2 and RPKM in all replicates > 1 in DHS-positive categories). Groups 1 to 5, respectively, represent DHSs detected in all samples (open to open [OO], n = 3,092), those only detected in cumulus cells (open to closed [OC], n = 14,924), those detected in both cumulus and SCNT samples (resistant to open to closed [rOC], n = 524), those detected in both SCNT and in vitro fertilization (IVF) (Pat) (closed to open [CO], n = 179), and those specifically detected in IVF(Pat) (resistant to closed to open [rCO], n = 262) (Figures 1C, 1D, and S1C). The great majority of DHSs were in the OC category (78.6%; Figure 1C), suggesting that SCNT-mediated DHS reprogramming is accompanied with a global loss of DHSs. The DHS categories show an interesting genomic distribution, among the non-reprogrammed DHSs, the majority of OO (87.8%) and rOC (66.4%) categories are located in promoters (transcription start site [TSS] ± 1 kb) (Figure 2A), while only 11.5% of the rCO category are located near promoter regions. In contrast, the majority of reprogrammed DHSs, the OC (74.2%) and CO (87.7%) categories, are located outside the promoter region (Figure 2A). These results suggest that distal regulatory elements are prone to be reprogrammed upon SCNT, while failure to close accessible somatic promoters or to open distal regulatory regions required for differentiation program may be the major reprogramming barriers. This observation is similar to recent ATAC-seq analysis of chromatin reprogram-

ming, which demonstrated that accessible chromatin at promoter regions is relatively stable during transcription factor (TF)-mediated iPSC generation (Knaupp et al., 2017; Li et al., 2017). Biological pathway and gene ontology (GO) enrichment analysis of the genes associated with OO and OC DHSs revealed that the OO-associated genes are enriched in ubiquitous cellular functions, such as cell cycle and DNA replication, while OCassociated genes are enriched in specific somatic cell programs such as angiogenesis and HIF1-alpha network (Figures 2B, S2A, and S2B). No specific GO or pathway enrichment in the other DHS categories was found (data not shown). These data suggest that SCNT reprogramming involves a specific loss of somatic cell memory while maintaining ubiquitous cellular functions. It has been shown that H3K9me3 in donor cells functions as an epigenetic barrier preventing SCNT-mediated reprogramming (Chung et al., 2015; Matoba et al., 2014). Thus, we hypothesized that reprogramming-resistant regions, particularly the rCO category, should be enriched for H3K9me3 in donor cells and SCNT embryos. Analysis of a public H3K9me3 ChIP-seq dataset in cumulus cells and SCNT two-cell embryos (Liu et al., 2016) indeed revealed that H3K9me3 is specifically enriched in rCO sites in cumulus cells and SCNT two-cell embryos (Figure 2C). This result provides another piece of evidence supporting the role of H3K9me3 in preventing SCNT-mediated reprogramming. TF Network Switch Likely Accompanies DHS Reprogramming Data presented in Figure 2C indicate that OC loci are devoid of H3K9me3 despite lack of chromatin accessibility in SCNT embryos. This observation suggests that global loss of DHSs at the OC loci is not due to gain of H3K9me3 but by the displacement of TFs from the transplanted somatic cell chromatin. To assess this possibility, we performed TF motif enrichment analysis for each DHS category (p value < 0.001, fragments per kilobase of transcript per million mapped reads [FPKM] > 1 in cumulus, SCNT, or IVF, and at least 2-fold enrichment compared to background) (Figure 2D). Interestingly, we found that 46.7% of the OC DHSs contain the Fra1 binding motif (Figure 2E), which does not show enrichment in any other DHS category (Figures 2D and 2E). Fra1 is known to regulate the specific transcriptional program in cumulus cells (Sharma and Richards, 2000) and is very lowly expressed in oocytes (Xue et al., 2013) or one-cell SCNT and IVF embryos (Matoba et al., 2014) (Figure 2E). Similar expression pattern was also observed for Sox5, a TF also with binding motif enriched in the OC loci (Figure 2E). Thus, lack of Fra1 and Sox5 binding to chromatin in SCNT embryos at least partly explains the dramatic loss of DHS in the OC category.

Figure 2. Transcription Factor Network Switch Likely Accompanies DHS Reprogramming (A) Bar graph showing the genomic distribution of the different DHS categories. DHS peaks within TSS ± 1 kb are considered promoter DHS, and those not located in promoters, exons, or introns are labeled as intergenic. (B) Bar plot showing the top ten enriched pathways of OO- or OC-associated genes using GREAT enrichment (McLean et al., 2010). (C) Averaged H3K9me3 ChIP-seq signal profile normalized to that of the input in cumulus and two-cell stage SCNT embryos in a DHS ± 2.5-Mb window. (D) Dot plot showing the enriched de novo TFs motifs (x axis) in the different DHS categories (y axis). Circle size represents the proportion of DHS having the de novo TF motif, and the gradient red color indicates the p value enrichment. The names of TFs with motif enriched in the different DHS categories are indicated. (E and F) TFs with binding motif enriched in the OC (E) and CO (F) DHSs and their relative expression levels in donor cells, oocytes, one-cell SCNT, and IVF embryos. See also Figure S2.

1942 Cell Reports 23, 1939–1947, May 15, 2018

A

liDNase-Seq

-5kb 5kb

-5kb 5kb

OC

ATAC-Seq

n= 12,858

OC

ICM

8-Cell

4-Cell

2-Cell

Morula

8-Cell

4-Cell

2-Cell

IVF(Pat)

Cumulus

SCNT

n= 2,066

B

liDNase-Seq

ATAC-Seq

-5kb 5kb

-5kb 5kb

ICM

8-Cell

4-Cell

2-Cell

Morula

8-Cell

4-Cell

IVF (Pat)

Cumulus

2-Cell

OO SCNT

OO

C OO

log2(FPKM + 0.1)

Cumulus

1-Cell

OC

2-Cell

4-Cell

8-Cell

ICM

< 1.4e−188

< 4.5e−184

< 7.3e−163

< 5.2e−160

20 < 9.8e−59

< 1.3e−144

15 10 5 0

D

expression of CO-associated genes (TSS±10kb)

15 log2(FPKM + 0.1)

0.039

10 5 0 −5 Cumulus

1-Cell

2-Cell

4-Cell

8-Cell

ICM

(legend on next page)

Cell Reports 23, 1939–1947, May 15, 2018 1943

These observations suggest that loss of donor cell-specific TF occupancy may contribute to the global loss of DHSs in SCNT embryos. In addition to the large number of OC category, which is likely responsible for the loss of donor cell chromatin identity, the other reprogrammed CO category could be important for acquiring totipotency. Analysis of the CO loci identified NFY as the most enriched motif (Figures 2D and 2F). Since Nfya, a subunit of the NFY complex, has been shown to be involved in ZGA (Lu et al., 2016), it is possible that the NFY complex also contributes to SCNT-mediated DHS reprogramming. Indeed, Nfya is expressed at a relatively lower level in cumulus cells while higher in one-cell SCNT and IVF embryos (Figure 2F). Another interesting finding from the motif enrichment analysis comes from the rOC category. Although NFY motif was enriched in rOC category as in the CO category, we found that the AP-1(Jun) and Elf1 motifs were exclusively enriched in this category (Figures 2D and S2C). Interestingly, it has been recently shown that chromatin binding of AP-1 in differentiated cells functions as a reprogramming barrier preventing iPSC generation (Li et al., 2017). This suggests that failure of specific somatic cell TFs to dissociate from chromatin can also be a barrier in SCNT reprogramming. Taken together, our data support the notion that the donor cell-specific TF network is quickly lost and a new zygotic TF network is established upon SCNT within 12 hr. Loss of Donor Cell DHSs Is Associated with Downregulation of Somatic Cell Transcription Program To understand the consequences of DHS reprogramming, we asked whether the OC loci maintain inaccessible states or become accessible later during preimplantation development. To this end, we analyzed publicly available liDNase-seq and ATAC-seq datasets at each preimplantation embryo development stages (Lu et al., 2016; Wu et al., 2016). Interestingly, both datasets consistently revealed that most of the OC loci maintain their inaccessibility throughout preimplantation development (Figure 3A). However, a small number of OC loci regain accessibility at the eight cell-to-morula stages in the liDNaseseq dataset (morula/cumulus > 3, RPKM > 1, and mean RPKM > 2 in morula) and at the inner cell mass (ICM) stage in the ATAC-seq dataset (Figure 3A). This chromatin accessibility dynamics is strikingly different from the OO loci, which remain accessible throughout preimplantation development (Figure 3B). These data suggest that, once lost upon SCNT, the majority of DHSs do not reappear during preimplantation development. To gain insight into the functional significance of the loss of DHSs, we examined whether OC-associated genes are transcriptionally turned off in preimplantation embryos. To this end,

we analyzed RNA-seq datasets of preimplantation embryos (Wu et al., 2016) and cumulus cells (Matoba et al., 2014). Comparative analyses between OC- and OO-associated genes (DHS within TSS ± 10 kb) (Tables S2 and S3) revealed that the gene expression level of OC-associated genes (median, 1 FPKM) is significantly lower than that of OO-associated genes (median, 4 FPKM) in all stages (Figure 3C). These data indicate that the OC-associated genes are expressed at a higher level in cumulus cells than in preimplantation embryos. A similar result was obtained when analyzing the genes harboring promoter DHSs (TSS ± 1 kb) (Figure S3A), excluding the possibility that these results could account for the difference in genomic distribution between OC and OO DHSs (Figure 2A). This observation was further confirmed by comparing the expression levels of the OC-associated genes between cumulus and preimplantation embryos (Figure S3B). Taken together, these results suggest that SCNT-mediated loss of DHS may be required for turning off the somatic cell transcriptional program. We also analyzed the expression of CO-associated genes. Interestingly, this group of genes exhibits an expression pattern that peaks at the two-cell stage consistent with ZGA (Figure 3D). Indeed, 12 out of the 52 CO-proximal genes (23.08%) are activated during the one-cell to two-cell transition (FC > 3, mean FPKM in two cell > 2) (Figure S3C). DHS Reprogramming in SCNT Is Independent of DNA Replication SCNT embryos complete the first round of DNA replication within 12 hpa. Previous studies indicated that DNA replication is required for somatic cell reprogramming in a heterokaryonmediated reprogramming setting (Tsubouchi et al., 2013). Given that TF-mediated iPSC generation generally takes 1 week, DNA replication must be required. To determine whether DNA replication is also required for SCNT-mediated DHS reprogramming, we treated SCNT embryos with a potent DNA replication inhibitor, aphidicolin, from 4 to 12 hpa (Figure 4A). We first confirmed that aphidicolin treatment effectively blocked DNA replication in SCNT embryos using a bromodeoxyuridine (BrdU) incorporation assay (Figure 4B). We then collected pseudopronuclei from aphidicolin-treated SCNT embryos at 12 hpa and performed liDNase-seq with biological duplicates (Figure S4A). Hierarchical clustering and PCA revealed that the DHS landscape of aphidicolin-treated SCNT embryos is much more similar to those of the non-treated SCNT and IVF(Pat) than to that of the cumulus cells (Figures 4C and S4B). A closer examination of the reprogrammed DHSs (CO and OC, Figure 1C) showed a limited effect of aphidicolin treatment on DHS reprogramming compared to that of the change from cumulus to SCNT (Figures 4D, S4C,

Figure 3. Loss of Donor Cell DHSs Is Associated with Downregulation of Somatic Cell Transcription Program (A and B) Heatmap showing the dynamics of OC (A) and OO (B) DHSs during preimplantation development analyzed by liDNase-seq (left) and ATAC-seq (right). Each row represents a locus (DHS center ± 5 kb), and the red and blue gradient colors represent the liDNase-seq and ATAC-seq signal intensity, respectively. OC DHSs were separated into two groups based on whether they reappear at the morula stage (FC[morula/cumulus] > 3, RPKM > 1, and mean RPKM > 2 in morula embryos). Peaks were ordered based on the signal intensity in cumulus cells. (C) Averaged gene expression levels of the OO- or OC-associated genes (TSS ± 10 kb) in cumulus cells and preimplantation embryos. The p values were calculated using the Wilcoxon rank sum test. (D) Averaged gene expression levels of the CO-associated genes (TSS ± 10 kb) in cumulus cells and preimplantation embryos. The p values were calculated using the paired t test. See also Figure S3.

1944 Cell Reports 23, 1939–1947, May 15, 2018

Figure 4. DHS Reprogramming in SCNT Is Independent of DNA Replication

A Cumulus cell G1 phase PCC

Activation

Aphidicolin

4h

+8h

SCNT Aphi

MII oocyte

B

C No treatment

PCA clustering of DHS profiles

Aphidicolin

12/12

lamin B1

11/11

PC2 (4.02%)

BrdU

100 Cumulus1 Cumulus2

0 Pat3

−100

Pat2 SCNT2

−200

20µm

Pat1

SCNT1

SC

_ NT

Ap

hi1

SCNT_Aphi2

−1500

−1000

−500

(A) Schematic illustration of the experiment to examine the effect of aphidicolin on SCNT-mediated DHS reprogramming. (B) Representative images of SCNT one-cell embryos immunostained with BrdU and lamin B1 antibodies. No BrdU signal was detected in aphidicolin-treated embryos. The number of SCNT embryos exhibiting the staining pattern and the total number of embryos analyzed are shown, respectively. Scale bar: 20 mm. (C) PCA of the genome-wide DHS profile of cumulus, IVF(Pat), SCNT, and aphidicolin-treated SCNT samples. Each dot represents an independent sample. (D) Heatmap showing liDNase-seq enrichment signal in the OC and CO DHSs after aphidicolin treatment. Each row represents a locus (DHS center ± 5 kb), and the red gradient color indicates the signal intensity. See also Figure S4.

0

PC1 (91.6%)

D

induced by M-phase-promoting factors in oocyte cytoplasm soon after microinCumulus SCNT IVF(Pat) SCNT Aphi Cumulus SCNT IVF(Pat) SCNT Aphi jection. The large amount of TFs stored in the oocyte cytoplasm, which is about 1,000 times larger than most somatic cells in volume (diameter, 70–80 mm in oocyte versus 5–10 mm in somatic cells), RPKM RPKM 15 15 10 10 may explain the fast and drastic replace5 5 0 0 ment of TFs in donor somatic nuclei. Once donor cell DHSs are lost upon SCNT, the majority of these loci remain 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb 5kb −5kb inaccessible and their associated genes are transcriptionally silenced during preimplantation development (Figure 3). TF and S4D). These results suggest that SCNT-mediated DHS re- motif analyses revealed that some maternal factors might be important for SCNT-mediated DHS reprogramming (Figure 2). programming is largely DNA replication independent. By profiling and comparing the chromatin accessibility of Future studies should test whether they are required for this prodonor cells, SCNT, and IVF one-cell embryos, we revealed that cess using oocyte-specific knockout approaches. DHS reprogramming is largely completed within 12 hr and that EXPERIMENTAL PROCEDURES this reprogramming takes place in a DNA replication-independent manner. This appears to be different from the cell fusion- Collection of Mouse Oocytes mediated reprogramming and the TF-mediated iPSC reprog- All animal studies were performed in accordance with guidelines of the Instituramming systems that require DNA replication or cell division, tional Animal Care and Use Committee of Harvard Medical School. The prosuggesting that SCNT reprogramming might be mechanistically cedures of oocyte collection and IVF were described previously (Inoue et al., different. The fast DHS reprogramming without a single-cell divi- 2017). The mouse strain used in this study was B6D2F1/J (BDF1) (The Jackson sion allows the separation of reprogramming events from devel- Laboratory; 100006). opmental processes such as major ZGA, which does not take SCNT place until the two-cell stage (Aoki et al., 1997). This finding is The procedures of SCNT using cumulus cells as donors were described preimportant as it identified the time window to which future mech- viously (Matoba et al., 2011). Briefly, MII oocytes were collected from superoanistic studies should be focused on. The global loss of DHSs is vulated 8- to 11-week-old BDF1 females. Cumulus cells were removed from likely due to the displacement of TFs from the transplanted so- oocytes by treatment with 300 U/mL bovine testicular hyaluronidase (Calbiomatic cell chromatin (Figure 2). Given that most TFs except the chem). MII oocytes were enucleated in HEPES-buffered KSOM containing 7.5 mg/mL cytochalasin B (CB) (Chalbiochem; 250233). Nuclei of cumulus cells pioneer factors are believed to be displaced from chromatin at were injected into the enucleated oocytes using a Piezo-driven micromaniputhe mitotic metaphase (Iwafuchi-Doi and Zaret, 2014), the lator (Primetech). After 1-hr incubation in KSOM, SCNT oocytes were activated SCNT-mediated loss of DHSs might be triggered by premature by Ca-free KSOM containing 3 mM strontium chloride (SrCl2) and 5 mg/ml CB chromosome condensation (PCC) of donor chromatin that is for 1 hr and further cultured in KSOM with CB for 4 hr. At 5 hpa, embryos were

Effect of Aphidicolin on OC DHS

Effect of Aphidicolin on CO DHS

Cell Reports 23, 1939–1947, May 15, 2018 1945

washed in KSOM. To block DNA replication, embryos were transferred to KSOM containing 3 mg/mL aphidicolin (Sigma-Aldrich) at 4 hpa until sample collection at 12 hpa.

DECLARATION OF INTERESTS

Whole-Mount Immunostaining The procedure for BrdU labeling was described previously (Shen et al., 2014). Briefly, SCNT embryos were cultured in 100 mM BrdU from 5 hpa and fixed with 3.7% paraformaldehyde for 20 min at 9 hpa. After permeabilization with 0.5% Triton X-100 for 15 min, embryos were treated with 4 N HCl for 30 min followed by neutralization with 100 mM Tris-HCl (pH 8.0) for 15 min. The primary antibodies against BrdU (1/200; Roche Diagnostic) and lamin B1 (1/2,000; Santa Cruz; sc-6217) were incubated for 1 hr in 1% BSA/PBS at room temperature. The secondary antibodies used were Alexa Flour 488 donkey anti-mouse IgG (Life Technologies) and Alexa Flour 568 donkey anti-goat IgG. The embryos were mounted on a glass slide in Vectashield anti-bleaching solution with DAPI (Vector Laboratories). Fluorescence was detected under Zeiss LSM800.

Received: February 20, 2018 Revised: March 26, 2018 Accepted: April 6, 2018 Published: May 15, 2018

liDNase-Seq The procedures of liDNase-seq were described previously (Inoue et al., 2017).

The authors declare no competing interests.

REFERENCES Aoki, F., Worrad, D.M., and Schultz, R.M. (1997). Regulation of transcriptional activity during the first and second cell cycles in the preimplantation mouse embryo. Dev. Biol. 181, 296–307. Boyle, A.P., Davis, S., Shulha, H.P., Meltzer, P., Margulies, E.H., Weng, Z., Furey, T.S., and Crawford, G.E. (2008). High-resolution mapping and characterization of open chromatin across the genome. Cell 132, 311–322. Buenrostro, J.D., Giresi, P.G., Zaba, L.C., Chang, H.Y., and Greenleaf, W.J. (2013). Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213–1218.

Data Analysis of liDNase-Seq Reads were trimmed then mapped to the mm9 genome. PCR duplicates and multi-mapped reads were removed. The IDR method was used to select replicable DHS (Li et al., 2011). DHS detected in cumulus, one-cell SCNT, paternal pronuclei, and the SCNT+aphidicolin were merged and used in the downstream analysis.

Buenrostro, J.D., Wu, B., Litzenburger, U.M., Ruff, D., Gonzales, M.L., Snyder, M.P., Chang, H.Y., and Greenleaf, W.J. (2015). Single-cell chromatin accessibility reveals principles of regulatory variation. Nature 523, 486–490.

Additional Public Data Publicly available ATAC-seq data (Wu et al., 2016), RNA-seq data (Matoba et al., 2014; Wu et al., 2016; Xue et al., 2013), and H3K9me3 ChIP-seq data (Liu et al., 2016) were mapped to the mm9 genome (Supplemental Experimental Procedures).

Gurdon, J.B. (1962). The developmental capacity of nuclei taken from intestinal epithelium cells of feeding tadpoles. J. Embryol. Exp. Morphol. 10, 622–640.

Statistical Methods DHS IDR was estimated using the IDR method. Motif p values were calculated using HOMER (Heinz et al., 2010) with background regions having similar GCbias. GO and pathway p values were estimated using the binomial test in GREAT. Wilcoxon rank sum test (Figures 3C and S3A), paired t test (Figures 3D and S3B), and Pearson’s correlation were done using R (http://www. r-project.org/). DATA AND SOFTWARE AVAILABILITY The accession number for the datasets reported in this study (summarized in Table S4) is GEO: GSE110851. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, and four tables and can be found with this article online at https://doi.org/10.1016/j.celrep.2018.04.036. ACKNOWLEDGMENTS We thank Dr. Luis Tuesta for critical reading of the manuscript. This project was supported by NIH (R01HD092465) and the Howard Hughes Medical Institute (HHMI). Y.Z. is an Investigator of the Howard Hughes Medical Institute. AUTHOR CONTRIBUTIONS Y.Z. conceived the project. A.I., S.M., T.S., C.Z., and F.L. performed the experiments. M.N.D. and L.J. analyzed sequencing datasets. M.N.D., A.I., and Y.Z. interpreted the data and wrote the manuscript.

1946 Cell Reports 23, 1939–1947, May 15, 2018

Chung, Y.G., Matoba, S., Liu, Y., Eum, J.H., Lu, F., Jiang, W., Lee, J.E., Sepilian, V., Cha, K.Y., Lee, D.R., and Zhang, Y. (2015). Histone demethylase expression enhances human somatic cell nuclear transfer efficiency and promotes derivation of pluripotent stem cells. Cell Stem Cell 17, 758–766.

Heinz, S., Benner, C., Spann, N., Bertolino, E., Lin, Y.C., Laslo, P., Cheng, J.X., Murre, C., Singh, H., and Glass, C.K. (2010). Simple combinations of lineagedetermining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol. Cell 38, 576–589. Hormanseder, E., Simeone, A., Allen, G.E., Bradshaw, C.R., Figlmuller, M., Gurdon, J., and Jullien, J. (2017). H3K4 methylation-dependent memory of somatic cell identity inhibits reprogramming and development of nuclear transfer embryos. Cell Stem Cell 21, 135–143.e6. Hussein, S.M., Puri, M.C., Tonge, P.D., Benevento, M., Corso, A.J., Clancy, J.L., Mosbergen, R., Li, M., Lee, D.S., Cloonan, N., et al. (2014). Genomewide characterization of the routes to pluripotency. Nature 516, 198–206. Inoue, K., Oikawa, M., Kamimura, S., Ogonuki, N., Nakamura, T., Nakano, T., Abe, K., and Ogura, A. (2015). Trichostatin A specifically improves the aberrant expression of transcription factor genes in embryos produced by somatic cell nuclear transfer. Sci. Rep. 5, 10127. Inoue, A., Jiang, L., Lu, F., Suzuki, T., and Zhang, Y. (2017). Maternal H3K27me3 controls DNA methylation-independent imprinting. Nature 547, 419–424. Iwafuchi-Doi, M., and Zaret, K.S. (2014). Pioneer transcription factors in cell reprogramming. Genes Dev. 28, 2679–2692. Jin, W., Tang, Q., Wan, M., Cui, K., Zhang, Y., Ren, G., Ni, B., Sklar, J., Przytycka, T.M., Childs, R., et al. (2015). Genome-wide detection of DNase I hypersensitive sites in single cells and FFPE tissue samples. Nature 528, 142–146. Jullien, J., Pasque, V., Halley-Stott, R.P., Miyamoto, K., and Gurdon, J.B. (2011). Mechanisms of nuclear reprogramming by eggs and oocytes: a deterministic process? Nat. Rev. Mol. Cell Biol. 12, 453–459. Knaupp, A.S., Buckberry, S., Pflueger, J., Lim, S.M., Ford, E., Larcombe, M.R., Rossello, F.J., de Mendoza, A., Alaei, S., Firas, J., et al. (2017). Transient and permanent reconfiguration of chromatin and transcription factor occupancy drive reprogramming. Cell Stem Cell 21, 834–845.e6. Koche, R.P., Smith, Z.D., Adli, M., Gu, H., Ku, M., Gnirke, A., Bernstein, B.E., and Meissner, A. (2011). Reprogramming factor expression initiates widespread targeted chromatin remodeling. Cell Stem Cell 8, 96–105.

Krijger, P.H.L., Di Stefano, B., de Wit, E., Limone, F., van Oevelen, C., de Laat, W., and Graf, T. (2016). Cell-of-origin-specific 3D genome structure acquired during somatic cell reprogramming. Cell Stem Cell 18, 597–610.

Mitalipov, S., and Wolf, D. (2009). Totipotency, pluripotency and nuclear reprogramming. Adv. Biochem. Eng. Biotechnol. 114, 185–199.

Li, Q., Brown, J.B., Huang, H., and Bickel, P.J. (2011). Measuring reproducibility of high-throughput experiments. Ann. Appl. Stat. 5, 1752–1779.

Sharma, S.C., and Richards, J.S. (2000). Regulation of AP1 (Jun/Fos) factor expression and activation in ovarian granulosa cells. Relation of JunD and Fra2 to terminal differentiation. J. Biol. Chem. 275, 33718–33728.

Li, D., Liu, J., Yang, X., Zhou, C., Guo, J., Wu, C., Qin, Y., Guo, L., He, J., Yu, S., et al. (2017). Chromatin accessibility dynamics during iPSC reprogramming. Cell Stem Cell 21, 819–833.e6.

Shen, L., Song, C.X., He, C., and Zhang, Y. (2014). Mechanism and function of oxidative reversal of DNA and RNA methylation. Annu. Rev. Biochem. 83, 585–614.

Liu, W., Liu, X., Wang, C., Gao, Y., Gao, R., Kou, X., Zhao, Y., Li, J., Wu, Y., Xiu, W., et al. (2016). Identification of key factors conquering developmental arrest of somatic cell cloned embryos by combining embryo biopsy and single-cell sequencing. Cell Discov. 2, 16010.

Sridharan, R., Gonzales-Cope, M., Chronis, C., Bonora, G., McKee, R., Huang, C., Patel, S., Lopez, D., Mishra, N., Pellegrini, M., et al. (2013). Proteomic and genomic approaches reveal critical functions of H3K9 methylation and heterochromatin protein-1g in reprogramming to pluripotency. Nat. Cell Biol. 15, 872–882.

Lu, F., Liu, Y., Inoue, A., Suzuki, T., Zhao, K., and Zhang, Y. (2016). Establishing chromatin regulatory landscape during mouse preimplantation development. Cell 165, 1375–1388.

Stadhouders, R., Vidal, E., Serra, F., Di Stefano, B., Le Dily, F., Quilez, J., Gomez, A., Collombet, S., Berenguer, C., Cuartero, Y., et al. (2018). Transcription factors orchestrate dynamic interplay between genome topology and gene regulation during cell reprogramming. Nat. Genet. 50, 238–249.

Matoba, S., Inoue, K., Kohda, T., Sugimoto, M., Mizutani, E., Ogonuki, N., Nakamura, T., Abe, K., Nakano, T., Ishino, F., and Ogura, A. (2011). RNAimediated knockdown of Xist can rescue the impaired postimplantation development of cloned mouse embryos. Proc. Natl. Acad. Sci. USA 108, 20621–20626.

Tsubouchi, T., Soza-Ried, J., Brown, K., Piccolo, F.M., Cantone, I., Landeira, D., Bagci, H., Hochegger, H., Merkenschlager, M., and Fisher, A.G. (2013). DNA synthesis is required for reprogramming mediated by stem cell fusion. Cell 152, 873–883.

Matoba, S., Liu, Y., Lu, F., Iwabuchi, K.A., Shen, L., Inoue, A., and Zhang, Y. (2014). Embryonic development following somatic cell nuclear transfer impeded by persisting histone methylation. Cell 159, 884–895.

Wu, J., Huang, B., Chen, H., Yin, Q., Liu, Y., Xiang, Y., Zhang, B., Liu, B., Wang, Q., Xia, W., et al. (2016). The landscape of accessible chromatin in mammalian preimplantation embryos. Nature 534, 652–657.

McLean, C.Y., Bristor, D., Hiller, M., Clarke, S.L., Schaar, B.T., Lowe, C.B., Wenger, A.M., and Bejerano, G. (2010). GREAT improves functional interpretation of cis-regulatory regions. Nat. Biotechnol. 28, 495–501.

Xue, Z., Huang, K., Cai, C., Cai, L., Jiang, C.Y., Feng, Y., Liu, Z., Zeng, Q., Cheng, L., Sun, Y.E., et al. (2013). Genetic programs in human and mouse early embryos revealed by single-cell RNA sequencing. Nature 500, 593–597.

Cell Reports 23, 1939–1947, May 15, 2018 1947

Cell Reports, Volume 23

Supplemental Information

Reprogramming of Chromatin Accessibility in Somatic Cell Nuclear Transfer Is DNA Replication Independent Mohamed Nadhir Djekidel, Azusa Inoue, Shogo Matoba, Tsukasa Suzuki, Chunxia Zhang, Falong Lu, Lan Jiang, and Yi Zhang

A

B SCNT samples DHS correlation

SCNT Rep2 (RPKM) 5 10 15 20

40 30 20 10

r=0.79

0

r=0.92

0

Cumulus Rep2 (RPKM)

Cumulus samples DHS correlation

0

10 20 30 40 Cumulus Rep1 (RPKM)

C

0

5 10 15 SCNT Rep1 (RPKM)

IVF(Pat) SCNT Cumulus

OO Pdia4 1

350

200

0

0

Hmgb1

OC

350

Mtmr4

350

0

Mxd1

350

0

Exd2

350

0

Pds5a

0

350

Stk17b

0

350

Ogfrl1

0

2 1 2 1 2 3 kb

2 kb

3 kb

2 kb

3 kb

3 kb

350

1

Cmtm6

0

350

Pdcl3

Snrnp25

350

200

0

0

0

3 kb

2 kb

rCO

rOC IVF(Pat) SCNT Cumulus

20

Mcm10

150

Gpat2

Phyhipl

0

150

150

0

0

Meioc

150

Piwil1

0

2 1 2 1 2 2 kb

2 kb

2 kb

2.5 kb

2 kb

2 kb

2 kb

2 kb

CO Ddx19b

IVF(Pat) SCNT Cumulus

4930547E08Rik 1

Zfp42

Rec8

150

150

150

150

0

0

0

0

2 1 2 1 2 2 kb

2 kb

2 kb

2 kb

Figure S1. liDNase-Seq data reproducibility and DHS examples, Related to Figure 1. (A) Scatterplot showing the correlation of liDNase-seq signal of two biological replicates of cumulus cells. (B) Scatterplot showing the correlation of liDNase-seq signal of two biological replicates of 1-cell stage SCNT embryos. (C) Genome browser view of representative liDNase-seq loci for the different DHS categories.

A

GO enrichment of genes in proximity of OO DHS

protein folding DNA replication ncRNA metabolic process nuclear export regulation of mitotic spindle organization DNA-dependent DNA replication ncRNA processing DNA-dependent DNA replication initiation response to topologically incorrect protein RNA localization

0

1

2

3

4

5

6

7

8

9

10 11

8.66 7.36 6.03 5.99 5.95 5.80 5.56 5.43 5.39

B

11.73

GO enrichment of genes in proximity of OC DHS

cell junction organization neg. regulation of phosphate metabolic process actin filament organization cell-cell junction organization negative regulation of protein phosphorylation cell junction assembly mitochondrial transport negative regulation of MAP kinase activity response to cadmium ion JAK-STAT cascade

-log 10 (Binom ial p-value)

C

0 2 4 6 8 10 12 14 16 18 20 22 24 26

12.91 12.28 11.10 10.56

27.05 25.11 21.25 18.70 18.36 17.37

-log 10 (Binom ial p-value)

de novo motifs enriched in rOC DHS Motif

TGACTCACC CCGGAAGT TG CTA C ATTGGT G A

C

C

T TA G C

A

G T A

C

G T

C G

A

TT

G A AG

G C T A

G

T

A

A G T T T A

G

A

C

C

T

A G C

T

CC

T G A G C T

T

G

A

C A

G C

T

G C

A

G C

A

C

T

A

C

T

A

G

A

CC G

A

G T

Best match TF

Sívalue % target % background

$3íE=,3

Hí

16.98

5.59

Elf1(ETS)

Hí

20.23

9.23

Hí

17.56

5.51

CCAA7íEox (NFY)

Figure S2. Functional and motif analysis of different DHS categories, Related to Figure 2. (A) Bar plot showing the top 10 enriched GO terms of OO-associated genes using GREAT enrichment. (B) Bar plot showing the top 10 enriched GO terms of COassociated genes using GREAT enrichment. (C) TFs with binding motif enriched in the rOC DHSs.

A

log2(FPKM + 0.1)

OO

Cumulus

&HOO

Hí

Hí

OC

2-Cell

4-Cell

&HOO

Hí

Hí

ICM

 

Hí

Hí

 5 

CO-associated genes activated during ZGA

 2



 

'G[E Zfp42



3UUD Zm\QG

Ubr4

TrPW

(5LN &SQH

PDWM

PDESFO %FDV

(16086*

FPKM=2



5

Nfyb +VSK

+foo

4



&FQH Terb2



Hí

)&

Hí

)&



Hí

0HG )&

Q 

Hí

í&HOOORJ2(FPKM+0.1))

log2(FPKM + 0.1)

25 

C 

expression of OC-associated genes (TSS±10kb)

FPKM=2

B

Q 

í



Cumulus 2-Cell 4-Cell

&HOO

ICM

í



2

4



í&HOOORJ2(FPKM+0.1))





Figure S3. Reprogrammed DHS are associated with loss of somatic cell transcription program and activation of ZGA genes, Related to Figure 3. (A) Averaged gene expression levels of the OO and OC-associated genes (TSS±1 kb) in cumulus and preimplantation embryos. The p-values were calculated using the Wilcoxon rank sum test. (B) Averaged gene expression levels of the OC-associated genes (TSS±10 kb) in cumulus and preimplantation embryos. The p-values were calculated using the paired t-test. (C) Dot plot showing activation of CO-associated ZGA genes [red, FC>3, mean FPKM(2-Cell) >2] as well as the 2-Cell down regulated genes [blue, FC2].

SCNT Aphidicolin replicates DHS correlation

B

C

5 10 SCNT Rep1 (RPKM)

15

Aphidicolin effect on OC

D

SCNT1C 2

SCNT1C 1

IVF (Pat) 2

Aphidicolin effect on CO Hí

Hí



IVF (Pat) 1

IVF (Pat) 3

SCNT1C Aphi 1

0

SCNT1C Aphi 2

0

r=0.63

Cumulus 2

Cumulus 1

SCNT Rep2 (RPKM) 5 10

15

A

Hí

Hí

15

Hí

Hí

20

RPKM

RPKM

10

10

5

0

0 Cumulus

SCNT

IVF(Pat)

SCNT Aphi

Cumulus

SCNT

IVF(Pat) SCNT Aphi

Figure S4. DHS reprogramming in SCNT is DNA replication independent, Related to Figure 4. (A) Scatterplot showing the correlation of liDNase-seq signal between two biological replicates of aphidicolin-treated SCNT samples. (B) Dendrogram showing the clustering of DHS profiles. (C, D) Box plot showing the normalized read per kilobase million (RPKM) intensities of the OC (C) and CO (D) with aphidicolin treatment. The p-values were calculated using paired t-test. In (C) the p-values were equal to 0 (due to the machine limitation), thus we showed them as 1 in at least one of the replicates. The PCA analysis was performed using all the merged peaks.

ATAC-seq analysis The preimplantation embryos ATAC-seq data were from a previous publication (Wu et al., 2016). Raw reads were first trimmed using Trimmomatic (Bolger et al., 2014)(v.0.36). 4-bp sliding window was used to scan the reads and trim when the window mean quality dropped below 15. The trimmed pair-end reads were then mapped to the mm9 genome using Bowtie2 (v2.2.9) with parameter -X 2000 --mm -1. Unmapped reads, reads with mapping quality less than 10 were removed using samtools (v1.3.1-38g9df6321). Reads mapping to the mitochondrial genome were discarded. PCR duplicates were removed using the MarkDuplicates command from Picard tools (v2.6.0). The bam files of the replicates were then merged together using samtools.

RNA-seq analysis All of the RNA-seq data used in this study were download from public sources. Cumulus cell RNA-seq data were from a previous study (Matoba et al., 2014) under the GEO accession numbers GSM1426195 and GSM1426196. The mouse oocyte data were from the public data of a previous study (Xue et al., 2013) available under GEO accession numbers GSM1080195 and GSM1080196. The raw data of preimplantation embryos were from a previous study (Wu et al., 2016) under the GEO accession number GSE66582. The downloaded raw reads were trimmed using Trimmomatic(v.0.36) and then mapped to the mm9 genome. Reads with a length of at least 50-bp were mapped to the genome using STAR (Dobin et al., 2013) (v.2.5.2b) with the following parameters: --outSAMtype BAM SortedByCoordinate –outSAMunmapped Within –outFilterType BySJout --outSAMattributes NH HI AS NM MD --outFilterMultimapNmax 20 --outFilterMismatchNmax 999 --quantMode TranscriptomeSAM GeneCounts. Gene expression was then estimated using RSEM (Li and Dewey, 2011) given the Ensemble NCBIM37.67 transcriptome assembly.

H3K9me3 ChIP-seq data analysis The cumulus and 2-cell embryo SCNT H3K9me3 ChIP-seq data were from a previous publication (Liu et al., 2016) under GEO accession numbers GSM1811770, GSM1811771, GSM1811775, and GSM1811776. Reads were trimmed using Trimmomatic (Bolger et al., 2014) and then mapped to the mm9 genome using Bowtie1 (Langmead et al., 2009) (v0.12.9) with parameters ‘--best -strata --sam -m 1’. Multi-mapped and unmapped and low-quality reads were removed using samtools (Li et al., 2009) (v1.3.1-38g9df6321) and PCR duplicates were removed using the MarkDuplicates command from Picard tools (v2.6.0).

Genomic annotation of DHS The Ensemble NCBIM37.67 genomic annotation was used to associate DHS peaks with the different genomic regions. Briefly, the makeTxDbFromGFF function from the GenomicFeatures (Lawrence et al., 2013) R/Bioconductor package was used to load the GTF file, then we called the annotatePeak function from the R/Bioconductor ChIPpeakAnno (Zhu et al., 2010) package for genomic annotation. Promoters were considered to be 1Kb from TSS and all the regions that did not fall within exons, introns, or UTRs were classified as distal intergenic regions.

Functional enrichment analysis The GREAT utility (McLean et al., 2010) was used to perform functional enrichment analysis of DHS. All the functional analysis were performed under the “Basal plus extension” mode in which the gene regulatory domain was defined as ±10kb around the TSS with a distal domain up to 1Mb. Only the Gene Ontology terms or pathways with an FDR < 0.05 were considered.

Motif analysis All the de-novo motifs in this study were detected using HOMER (Heinz et al., 2010) (v4.9) with parameters mm9 and -size 500. Denovo motifs that had a p-value < 0.0001 and at least a two-fold enrichment between foreground and background were considered for downstream analysis.

Heatmap generations The DHS peaks were extended by ±5kb from their peak center to form 10kb windows, each window was then split into 100 bin using the binner function from the R/Bioconductor package genomation (Akalin et al., 2015). The number in each bin were counted using the summarizeOverlaps function from the GenomicAlignments R/Bioconductor package (Lawrence et al., 2013). A matrix was then generated and displayed using the EnrichedHeatmap R/Bioconductor package (Zuguang, 2017).

PCA analysis The DHS peaks detected in Cumulus, SCNT, SCNT+ Aphidicolin treatment and IVF(Pan) samples were merged together, their RPKM values were calculated in each biological replicate then the ‘prcomp’ function available in R (http://www.r-project.org/) was used with parameters ‘center = FALSE, scale. = FALSE’.

Genomic tracks generation Smoothed bigwig tracks for each sample were generated using the bamCoverage utility from the deeptools suite (Ramirez et al., 2014) (v2.5.1) with parameters ‘--binSize 25 --extendReads 200 --normalizeUsingRPKM --outFileFormat bigwig --scaleFactor 1’. All genomic track visualization was performed using IGV (Thorvaldsdottir et al., 2013).

Statistical analyses Statistical analyses were implemented with R (http://www.r-project.org/). Pearson’s r coefficient was calculated using the ‘cor’ function with default parameters.

Pipeline automation All of the RNA-seq, ChIP-seq, and ATAC-seq mapping pipelines were automated using Bpipe (Sadedin et al., 2012).

Code availability Additional custom codes used for bioinformatics analysis are available upon request.

Supplemental References Akalin, A., Franke, V., Vlahovicek, K., Mason, C.E., and Schubeler, D. (2015). Genomation: a toolkit to summarize, annotate and visualize genomic intervals. Bioinformatics 31, 1127-1129. Bolger, A.M., Lohse, M., and Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114-2120. Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, M., and Gingeras, T.R. (2013). STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15-21. Heinz, S., Benner, C., Spann, N., Bertolino, E., Lin, Y.C., Laslo, P., Cheng, J.X., Murre, C., Singh, H., and Glass, C.K. (2010). Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell 38, 576-589. John, S., Sabo, P.J., Thurman, R.E., Sung, M.H., Biddie, S.C., Johnson, T.A., Hager, G.L., and Stamatoyannopoulos, J.A. (2011). Chromatin accessibility pre-determines glucocorticoid receptor binding patterns. Nat Genet 43, 264-268. Langmead, B., Trapnell, C., Pop, M., and Salzberg, S.L. (2009). Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10, R25. Lawrence, M., Huber, W., Pages, H., Aboyoun, P., Carlson, M., Gentleman, R., Morgan, M.T., and Carey, V.J. (2013). Software for computing and annotating genomic ranges. PLoS Comput Biol 9, e1003118. Li, B., and Dewey, C.N. (2011). RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12, 323. Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G., Abecasis, G., Durbin, R., and Genome Project Data Processing, S. (2009). The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078-2079. Li, Q., Brown, J.B., Huang, H., and Bickel, P.J. (2011). Measuring reproducibility of high-throughput experiments. Ann Appl Stat 5, 1752-1779. Liu, W., Liu, X., Wang, C., Gao, Y., Gao, R., Kou, X., Zhao, Y., Li, J., Wu, Y., Xiu, W., et al. (2016). Identification of key factors conquering developmental arrest of somatic cell cloned embryos by combining embryo biopsy and single-cell sequencing. Cell Discov 2, 16010. Matoba, S., Liu, Y., Lu, F., Iwabuchi, K., Shen, L., Inoue, A., and Zhang, Y. (2014). Embryonic Development following Somatic Cell Nuclear Transfer Impeded by Persisting Histone Methylation. Cell 159, 884-895. McLean, C.Y., Bristor, D., Hiller, M., Clarke, S.L., Schaar, B.T., Lowe, C.B., Wenger, A.M., and Bejerano, G. (2010). GREAT improves functional interpretation of cis-regulatory regions. Nat Biotechnol 28, 495-501. Ramirez, F., Dundar, F., Diehl, S., Gruning, B.A., and Manke, T. (2014). deepTools: a flexible platform for exploring deepsequencing data. Nucleic Acids Res 42, W187-191. Sadedin, S.P., Pope, B., and Oshlack, A. (2012). Bpipe: a tool for running and managing bioinformatics pipelines. Bioinformatics 28, 1525-1526. Thorvaldsdottir, H., Robinson, J.T., and Mesirov, J.P. (2013). Integrative Genomics Viewer (IGV): high-performance genomics data visualization and exploration. Brief Bioinform 14, 178-192. Wu, J., Huang, B., Chen, H., Yin, Q., Liu, Y., Xiang, Y., Zhang, B., Liu, B., Wang, Q., Xia, W., et al. (2016). The landscape of accessible chromatin in mammalian preimplantation embryos. Nature 534, 652-657. Xue, Z., Huang, K., Cai, C., Cai, L., Jiang, C.Y., Feng, Y., Liu, Z., Zeng, Q., Cheng, L., Sun, Y.E., et al. (2013). Genetic programs in human and mouse early embryos revealed by single-cell RNA sequencing. Nature 500, 593-597. Zhu, L.J., Gazin, C., Lawson, N.D., Pages, H., Lin, S.M., Lapointe, D.S., and Green, M.R. (2010). ChIPpeakAnno: a Bioconductor package to annotate ChIP-seq and ChIP-chip data. BMC Bioinformatics 11, 237. Zuguang, G. (2017). EnrichedHeatmap: Making Enriched Heatmaps. R package version 192.