Nanog and Oct4 associate with unique transcriptional ... - Nature

letters

Nanog and Oct4 associate with unique transcriptional repression complexes in embryonic stem cells Jiancong Liang1,6, Ma Wan1,6, Yi Zhang1, Peili Gu2,4, Huawei Xin1, Sung Yun Jung1, Jun Qin1,2, Jiemin Wong2,5, Austin J. Cooney2,3, Dan Liu1, and Zhou Songyang1,3,7 Nanog and Oct4 are essential transcription factors that regulate self-renewal and pluripotency of ES cells. However, the mechanisms by which Nanog and Oct4 modulate ES cell fate remain unknown. Through characterization of endogenous Nanog and Oct4 protein complexes in mouse ES cells, we found that these transcription factors interact with each other and associate with proteins from multiple repression complexes, including the NuRD, Sin3A and Pml complexes. In addition, Nanog, Oct4 and repressor proteins co-occupy Nanog-target genes in mouse ES cells, suggesting that Nanog and Oct4 together may communicate with distinct repression complexes to control gene transcription. To our surprise, of the various core components in the NuRD complex with which Nanog and Oct4 interact, Mta1 was preferred, whereas Mbd3 and Rbbp7 were either absent or present at sub-stoichiometric levels. We named this unique Hdac1/2- and Mta1/2-containing complex NODE (for Nanog and Oct4 associated deacetylase). Interestingly, NODE contained histone deacetylase (HDAC) activity that seemed to be comparable to NuRD, and retained its association with Nanog and Oct4 in Mbd3–/– ES cells. In contrast to Mbd3 loss-of-function, knockdown of NODE subunits led to increased expression of developmentally regulated genes and ES-cell differentiation. Our data collectively suggest that Nanog and Oct4 associate with unique repressor complexes on their target genes to control ES cell fate. Embryonic stem (ES) cells possess an unlimited potential for self-renewal while maintaining the capacity to produce ectodermal, endodermal and mesodermal germ layers and germ lines during embryogenesis1,2, making them ideal targets for regenerative medicine research. Several key ES-cell-specific transcription factors, such as Nanog and Oct4, are essential for self-renewal and pluripotency of ES cells3–5. In fact, co-expression of Oct4 or Nanog with other factors is sufficient to re-programme mouse or human fibroblasts into pluripotent cells6–10.

The activities of transcription factors such as Nanog and Oct4 are ultimately controlled by co-activators and co-repressors. Recent studies suggest that there is a link between the Mbd3-containing NuRD repression complex and ES-cell-specific transcription factors in cell fate determination11,12; however, the interaction between NuRD and Nanog has not been characterized and the function of the NuRD complex in relation to Nanog and Oct4 remains unclear. For example, expression or silencing of developmental genes targeted by Nanog and Oct4 remained largely intact in Mbd3–/– ES cells11, arguing against a global repression role of Mbd3 or the NuRD complex in ES cells. To understand how endogenous Nanog and Oct4 regulate self-renewal and pluripotency, we sought to identify their interaction partners in ES cells. We found that endogenous Nanog and Oct4 could interact with each other to form a complex. In addition, Oct4 and Nanog also associated with proteins from multiple repression complexes, including Sin3A and NuRD. Unexpectedly, although Nanog and Oct4 recruited multiple members of the NuRD complex, these repressor complexes differed significantly from the conventional NuRD complex. Our data suggest that Nanog and Oct4 interact and communicate with unique repression complexes to modulate transcription programmes necessary for determining ES cell fate. To identify the factors that form complexes with Nanog, endogenous Nanog and its associated proteins were affinity-purified from mouse ES cell nuclear extracts and sequenced by mass spectrometry (Fig. 1a). Several factors identified (for example, Oct4, Esrrb, RIF1 and Sall4) were previously shown to complex with epitope-tagged Nanog12,13 (Supplementary Information, Table S1). In addition, we identified a number of transcription factors and chromatin-associated proteins not yet shown to interact with Nanog (Supplementary Information, Table S1, in bold), including several proteins (SMARCA2, SMARCA4 and BAF180) in the nucleosome-remodelling SWI/SNF complex14,15. Interestingly, subunits of multiple co-repression complexes were also found, such as nuclear remodelling and histone deacetylation complex NuRD (Mta1, Mta2, Hdac1, Hdac2, p66α/Gatad2a

1 Verna and Marrs McLean Department of Biochemistry and Molecular Biology, and 2Department of Molecular and Cellular Biology, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030, USA. 3Stem Cells and Regenerative Medicine Center, Baylor College of Medicine, One Baylor Plaza, Houston, Texas 77030, USA. 4 Current address: Department of Cancer Genetics, University of Texas, M.D. Anderson Cancer Center, Houston, TX 77030, USA. 5Current address: The Institute of Biomedical Sciences, College of Life Sciences, East China Normal University, Shanghai, China. 6 These authors contributed equally to this work. 7 Correspondence should be addressed to Z. S. (e-mail: [email protected])

Received 20 February 2008; accepted 1 April 2008; published online 4 May 2008; DOI: 10.1038/ncb1736

nature cell biology volume 10 | number 6 | JUNE 2008 © 2008 Nature Publishing Group

731

letters

Nanog

SMARCA4, RIF1, BAF180

Sin3A Mta1 and Mta2 Gatad2a and 2b Hdac1/2

Mta1/2 NuRD Subunits

Mbd3

Antibody Rbbp7

3%

Sin3A

Nanog LSD1

49

D2

D2

TA

Fla

Fla

g–

GA

2

1 AC

t4

116

g–

Oc Anti-Flag IP

37

Fla

g–

116

Fla

Mr(K)

182

HD

IP P 1I

LS D

IP

2a

Ga

tad

P

3A

Sin

gI

ol

no Na

ntr

Inp

Co

3%

Mr(K)

IP

d ut

c

Mbd3

A

20

Mta1 Mta2

TA

30

B

Nanog, Oct4 Esrrb

g– TIN 2 Fla g– RA P1

Hdac2

GA

40

Antibody

g–

50

Pml

Fla

70

Sin3A

ac

90

Oct4

Hd

120

Nanog

Oct4

SMARCA2, Sin3A Sall4 Pml, Kap1, LSD1

160

Co

Mr(K)

g–

Mr(K)

Nanog IP

Fla

MW

Inp ut ntr ol IP Na no gI P Hd ac 2I P

b

a

WB: Nanog

WB: Flag

WB: Nanog

WB: Flag

82 5% Input

Figure 1 Identification of the endogenous Nanog protein complexes from mouse ES cells. (a) Anti-Nanog antibodies were used for immunoprecipitation (IP) of the Nanog complex from mouse ES cell nuclear extracts. The IP products were resolved by SDS–PAGE and stained with Coomassie blue. The lanes were sectioned and digested with trypsin. Extracted peptides were sequenced by mass spectrometry. (b) Verification of the endogenous Nanog complex by co-immunoprecipitation. Mouse ES cell nuclear extracts were used for IP with anti-Nanog and anti-HDAC2 antibodies, and probed for indicated proteins by western blotting (WB). Rabbit IgG was used as a control. Arrows indicate the expected bands. The anti-Mta1/2 blotting antibody recognizes both Mta1 and Mta2. The top two bands are Mta1 as determined by RNAi (see Fig. 5a). Images in individual blot panels were from the same gel. (c) Mouse ES cell nuclear

extracts were immunoprecipitated with antibodies against Nanog, Sin3A, Gatad2a or LSD1. Western blots were performed using anti-Sina3A, antiNanog or anti-LSD1 antibodies. Rabbit IgG was used as a control. (d) Nanog can interact with Oct4, Hdac1, Hdac2 and GATAD2A. 293T cells were transiently transfected with expression vectors encoding YFPc–Nanog, together with constructs encoding FLAG-tagged Oct4, HDCA1, Hdac2, GATAD2A, GATAD2B, RAP1 or TIN2. Whole-cell extracts (WCE) were immunoprecipitated with anti-Flag antibodies and western blotting was performed with anti-Nanog or anti-FLAG antibodies. WCE input was also blotted with anti-Nanog or anti-FLAG antibodies. TIN2 and RAP1 are telomere binding proteins and were used as IP negative controls. Images in individual blot panels were from the same gel. Full scans of all gels are shown in Supplementary Information, Fig. S6.

and p66β/Gatad2b)16–18, Sin3A-HDAC complex (Sin3a, Hdac1/2, and Kap1; ref. 19), transcription repressor and demethylase LSD1(ref. 20), and the transcription factor Pml21. In this study, we focused on the role of Nanog and Oct4 associated transcription repression complexes. To confirm our mass spectrometry results, we carried out co-immunoprecipitation and western blotting analysis of the endogenous Nanog complexes in ES cells. Endogenous Oct4, Sin3A, Pml, Hdac2, Mta1 and Mta2 could indeed co-precipitate with Nanog, and reciprocal immunoprecipitations using anti-Hdac2, Sin3A, Gatad2a and LSD1 antibodies

also pulled down Nanog (Fig. 1b, c). Furthermore, FLAG-tagged Oct4, Hdac1, Hdac2 and Gatad2a, but not Mbd3, could be co-precipitated with exogenously expressed Nanog in 293T cells (Fig. 1d; Supplementary Information, Fig. S1). These results indicate that Nanog associates with components of multiple repression complexes. Nanog and Oct4 can co-occupy promoters of genes necessary for ES cell self-renewal and pluripotency22,23; however, it is not known whether Nanog and Oct4 interact directly with each other. FLAG-tagged Oct4 could co-precipitate with Nanog in 293T cells (Fig. 1d). Furthermore,

732


letters a

c

10 5

37

Nanog-CT

26 Coomassie stain

e

Mta1/2 Mbd3

Rbbp7

Inp

ut id Vo

Inp

Mr(K)

Mta1/2

Rbbp7

16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48

Nanog

Mta1 Mta2

Mr 90K

ut

Mr 670K

Co

1%

Mr(K) Oct4 Nanog Hdac2

Inp u ntr t ol Mt a2 IP Oc t4 IP

d

YF

F2 –Y

Nanog-FL

TR

S-Nanog-NT

FP

64 35

FP

+Y

YF

Pn

115 82

c Pn +Y Pn -N F P an c– og Oc +Y t4 YF F Pc Pn -O -N ct4 HD an og AC +Y 1– YF FP Pn c +M bd 3 YF –Y YF Pn FP Pn c +M –N bd an 3– og Y +M FP c bd 3– YF Pc

0

29 3T

Oc t4– GS T Ra p1 –G ST GS T

MW

10 % Inp ut GS T Ra p1 –G ST Oc t4– GS 10 T % Inp ut GS T Ra p1 – Oc GST t4– GS 10 T % Inp ut GS T Ra p1 Oc –GST t4– GS T

b

15

c

Nanog-CT

20

FP

Nanog-NT

25

YF

Nanog-FL

WR

2– Y

HD

305

TIN

256

n+

157

YFP-positive cells (percentage)

1

Hdac2

IgG Rbbp7

Oct4

IgG Rbbp7 Darker

Pml

Mr 1,200K

Figure 2 Nanog and Oct4 recruit multiple repressor complexes. (a) Schematic diagram of Nanog and its deletion mutants. HD, homeodomain. WR, tryptophan repeat. (b) The N terminus of Nanog is sufficient for in vitro interaction with Oct4. In vitro translated and 35S-labelled full-length (FL), N-terminal (NT), or C-terminal (CT) Nanog proteins were pulled down using bacterially expressed GST alone and GST-tagged Oct4 or RAP1. RAP1 was used as a negative control. (c) Nanog interacts with Oct4 but not Mbd3 in vivo. 293T cells were transiently transfected with expression vectors encoding Venus YFPn-tagged Nanog along with constructs encoding YFPc-tagged Oct4 or Mbd3 for fluorescence complementation analysis. The percentages of YFPpositive cells were compared using a flow cytometer. TRF2–YFPn/TIN2–YFPc

and HDAC1–YFPn/Mbd3–YFPc pairs were used as positive controls; YFPn– Nanog/YFPc- and YFPn/YFPc-tagged proteins were used as negative controls. Data are mean ± s.e.m. (n = 3). (d) Oct4 associates with NuRD subunits. Mouse ES cell nuclear extracts were immunoprecipitated with antibodies against Oct4 or Mta2, and probed for the indicated proteins. A goat anti-ICAM antibody was used as a control. Images in individual blot panels were from the same gel. (e) Mouse ES cell nuclear extracts were fractionated through a gel filtration column. Selected fractions from 10–48 were then resolved by SDS–PAGE and western blotting performed to identify the indicated proteins. Molecular weights of gel filtration standards are shown. Full scans of all gels are shown in Supplementary Information, Fig. S6

both full-length and the amino-terminal region (Nanog–NT, residues 1–157) of Nanog could interact specifically with GST-tagged Oct4 in vitro (Fig. 2a, b). To investigate this interaction further, we used the bimolecular fluorescence complementation (BiFC) assay24. Nanog and Oct4 were tagged with the N-terminal half of Venus yellow fluorescent protein (YFPn–Nanog) and the C-terminal domain of YFP (YFPc–Oct4), respectively, and co-expressed in 293T cells. Nanog–Oct4 interaction would bring the YFP fragments together for fluorescence complementation to occur. Approximately 17% of YFPn–Nanog and YFPc–Oct4

co-expressing cells were YFP-positive, a result comparable to that of the positive control used in the study (~20% of cells co-expressing YFPn– TRF2 and YFPc–TIN2; ref. 24), whereas approximately 3% of the control cells were YFP-positive (Fig. 2c). These results suggest that Nanog and Oct4 are in close proximity in vivo and may interact directly. The data so far indicate that Nanog and Oct4 can interact with each other and form protein complex(es) with transcription repressors. To gain more insight into the protein interactions involving Nanog and Oct4, we also analysed the endogenous Oct4 complex using anti-Oct4


733

Hdac2

_

Oct4

Mta2

d

Control _

+

_

+

_ 15

Anti-H3

15

Oct4

Control

Nanog

Control

IP

Mta2 ( 20%)

Anti-H4K8AC

4 3.5 3 2.5 2 1.5 1 0.5 _

+

_

+

Control Nanog

Oct4 Nanog Hdac1/2 Actin

26 49 37 49 64 49 49 37

1.2 Relative HDAC activity

Mbd3

f

1/2 5

1/1 0

4.5

0

WT Mbd3–/– Mr(K) 37

Ab 2I P

37 26

5

WB: Hdac2

e

no g Ab

Relative specific HDAC activity

_

Nanog

Control

IP

Na

Rbbp7 49 Mbd3

TSA

IP

Ab

49 37

c

Mt a1

49

1/5

Nanog

49

82 64 64

Ab

Mbd3 26 Sox2

Co

Mr(K) Mta1/2

3%

Inp ut ntr ol IP

b

Co

Mr(K) 64 Hdac2 49

3%

a

Inp ut ntr ol IP Na no gA b1 IP Na no gA b1 IP Na (LS no ) gA b2 IP Mt a1 IP

letters

_

TSA

+

Mta2

IP

*

*

1 0.8 0.6 0.4 0.2 0 –0.1

_

_

+

+

_

+

_

+

Nanog

Nanog

Oct4

Oct4

WT

Mbd 3–/–

WT

Mbd3–/–

TSA IP ES cells

Figure 3 Nanog and Oct4 repression complexes contain HDAC activity but a low stoichiometric amount of Mbd3. (a) The endogenous Nanog complex was isolated from ES cells using two different antibodies (Nanog Ab1 and 2). Anti-Mta1 and larger-scale (LS, ×3) anti-Nanog immunoprecipitations were also performed in parallel. The immunoprecipitates were subjected to western blotting using the indicated antibodies. (b) Anti-Mta1 immunoprecipitates were serially diluted and compared with anti-Nanog immunoprecipitation by western blotting with anti-Mta1/2, anti-Hdac2, anti-Rbbp7 or anti-Mbd3 antibodies. (c) The HDAC activities of NODE and NuRD as assayed on nucleosomes. Immunoprecipitates of mouse ES cell nuclear extracts using various antibodies were used in HDAC assays in the presence or absence of TSA. The levels of acetylated H4K8 were subsequently examined by western blotting. TSA-treated nucleosomes were used as substrates. The protein levels of H3 and immunoprecipitated Hdac2 (bottom panel) were used as

controls. For IP controls, rabbit IgG (for Nanog) or a goat antibody against ICAM (for Mta2 and Oct4) was used. (d) Specific HDAC activities in different immunoprecipitates as in c were also measured using a Fluor de lys substrate. The measured activities were further normalized based on the amount of Hdac2 in the immunoprecipitates. Data are mean ± s.e.m. (n = 3). (e) Western blots of extracts from wild-type (WT) or Mbd3 null ES cells using different antibodies. Images in individual blot panels were from the same gel. (f) Nanog- and Oct4associated HDAC activities were not significantly affected by the absence of Mbd3. HDAC assays were performed using different immunoprecipitation products from WT or Mbd3–/– ES cell nuclear extracts either in the presence or absence of TSA. The background fluorescence was subtracted from the measurements. Data are mean ± s.e.m. (n = 3). *P values were greater than 0.05, indicating no significant difference between the samples. Full scans of all gels are shown in Supplementary Information, Fig. S6.

antibodies. Consistent with the model that Nanog and Oct4 form complexes with each other, there is substantial overlap in the Oct4 and Nanog protein complexes (Supplementary Information, Tables S1, S2). For example, the transcription factor Sall4 and NuRD components (Hdac1/2, Mta1/2, Gatad2a and Gatad2b) were also found in the Oct4 complex.

Furthermore, immunoprecipitation of endogenous Oct4 pulled down Nanog, Hdac2, Mta1/2 and Mta2 in ES cells (Fig. 2d), and reciprocal Hdac2 immunoprecipitation could pull down Oct4 (Fig. 1b). Next, we analysed the size of the Nanog and Oct4 complexes using chromatography (Fig. 2e). Consistent with our mass spectrometry

734


letters Inp IgGut Na IP n Oc og IP t4 Hd IP ac No 2 IP tem pla

te

a

Nanog region 1

Nanog region 1 Nanog region 2

Nanog region 2

Oct4 region 1

Oct4 region 1

Oct4 region 2

Oct4 region 2

Esrrb region 1

Esrrb region 1

80,889 bp Nanog

80,889 bp

Nanog 15,000 bp Oct4

15,000 bp

Oct4

219,000 bp 219,000 bp

Esrrb region 2

Esrrb region 2

Foxd3 region 1

Foxd3 region 1

50,000 bp

Foxd3 region 2

50,000 bp

Esrrb Esrrb

Foxd3 region 2

Foxd3 Foxd3

Jarid2 region 1

Jarid2 region 1

Jarid2 region 2

Jarid2 region 2

208,000 bp 208,000 bp Jarid2 Jarid2

Fgf5

Fgf5

Fgfbp1

Fgfbp1

25,700 bp

5,900 bp

Fgf5

Fgfbp1

Dmd

Dmd

2,513,750 bp

12 10

Nanog IP Oct4 IP Hdac2 IP IgG IP

8 6 4 2 Nanog

Pou5f1

d

Esrrb

Fgf5

8 6

Mta1 Ab1 Mta1 Ab2 IgG IP

4 2 0

Nanog

Pou5f1

Esrrb

Fgf5

Fgfbp1

Fgfbp1

te

0

Inp u IgG t I Na P no Oc g IP t4 Hd IP ac No 2 IP tem pla

Relative fold enrichment

c

Relative fold enrichment

Dmd

b

e 1000 Relative fold enrichment (log)

Nanog Esrrb

Jarid2 Foxd3 Fgfbp1

10 1 0.1

Dmd

Nanog IP Oct4 IP Mta1 IP Mta2 IP Hdac1 IP Hdac2 IP IgG IP

100

Nanog Pou5f1

Mbd3–/– ES cells

Esrrb

Fgf5

Fgfbp1

Mbd3–/– ES cells

Figure 4 Co-occupancy of Nanog, Oct4 and Hdac2 on Nanog target genes is independent of Mbd3. (a) ChIP experiments were performed on ES cells using total rabbit IgG or antibodies against Nanog, Oct4 or Hdac2. Left panels show the results of PCR amplification reactions from the immunoprecipitates. Right panels illustrate the corresponding amplified regions with red bars in genomic context. Exons are depicted as blue boxes. Arrows indicate the gene bodies and orientation, pointing from first to last exon. The numbers on top represent

the genomic length shown in each plot. (b) Real-time PCR quantification of ChIP data for Nanog, Oct4 or Hdac2 from a. Data are mean ± s.e.m. (n = 3). (c) Real-time PCR quantification of ChIP data for Mta1. Two different anti-Mta1 antibodies were used. Data are mean ± s.e.m. (n = 3). (d) ChIP experiments as described in a were performed using Mbd3-null ES cells. (e) Real-time PCR quantification of ChIP data for rabbit IgG, Nanog, Oct4, Hdac1, Hdac2, Mta1 and Mta2. Data are mean ± s.e.m. (n = 3).

and immunoprecipitation experiments, the majority of Nanog and a fraction of Oct4 were eluted together as a complex with a relative molecular mass of approximately 1,200,000 (Mr 1,200K). Notably, Oct4 (but not Nanog) was found predominantly in the low molecular weight fractions (Mr 70% knockdown for all except Mta2; Fig. 5a, b; Supplementary Information, Fig. S4). Knockdown of Mta1 did not affect the level of Mta2 (Fig. 5a). In Mbd3–/– ES cells, the mRNA level of Dppa3 was reduced, whereas pramel6 expression was activated11,26,27. Similarly, we found that Mbd3 knockdown ES cells showed reduced Dppa3 expression, whereas pramel6 was upregulated (Fig. 5b). However, Mta1 or Mta2 knockdown resulted in distinct gene expression patterns, compared with Mbd3 loss-of-function (Fig. 5b). For example, the level of Dppa3 increased in both Mta1 and Mta2 knockdown cells. These results support the suggestion that NODE represses Nanog and Oct4 target genes, and that NODE is functionally different from the canonical NuRD complex. As key regulators of ES cell self-renewal and pluripotency, reducing either Nanog or Oct4 levels in ES cells triggers differentiation1. When cultured under feeder-free conditions, wild-type ES cells require leukaemia inhibitory factor (LIF) to maintain self-renewal and will differentiate when LIF is removed1. To further define the role of NODE in ES cells, we examined the ability of ES cells treated with short hairpin (sh) RNA (to knockdown Mta1, Mta2 or Hdac2) to self-renew and differentiate in the presence or absence of LIF. Although inhibition of Mbd3 by RNA interference (RNAi) promoted ES cell self-renewal in the absence of LIF11, Hdac2 or Mta2 knockdown had minimal effects (Fig. 5c, d). It is possible that Hdac1 and Mta1, which are also present in the NODE complex, may compensate for the loss of Hdac2 and Mta2 under these conditions. In contrast, Mta1 knockdown by two different RNAi sequences induced ES cell differentiation even in the presence of LIF (Fig. 5c, d). As with Nanog knockdown cells, Mta1 RNAi cells showed much lower alkaline phosphatase activity, consistent with our finding that Mta1 is the major MTA homologue in NODE (Figs 1b, 2d, 3b). Interestingly, Mta1 knockdown led to increased expression of lineage markers (Foxa2 and Gata6) that associate with endoderm differentiation, similarly to Nanog knockdown (Fig. 5e); Gata6 was found to be a NODE-target gene (Supplementary Information, Fig. S5). These data suggest that Mta1, and probably NODE, may have a crucial role in ES cell self-renewal or inhibition of differentiation. Our data revealed that the pluoripotent transcription factors Nanog and Oct4 interact with each other and chromatin remodelling proteins (Fig. 1). Most of the Nanog- and Oct4-associated proteins identified here are in fact transcription repressors, suggesting that Nanog and Oct4 may be primarily involved in gene repression. Consistently, Mta1 knockdown moderately increased the expression of several Nanog or Oct4 target genes (Fig. 5b). Native NuRD complexes lacking both Mbd3 and Rbbp7 have not been reported previously. The NODE complexes we identified contain a subset of the subunits found in repression complexes such as NuRD and Sin3A. Although Mbd3 and Rbbp7 were present at very low levels, NODE still retained comparable HDAC activity. NODE primarily uses Mta1 (not Mta2 or Mbd3; Figs 1b, 2d, 3a, 5a), a protein implicated in Caenorhabditis elegans embryonic development and human cancers28. Knockdown of Mta1 caused upregulation of Dppa3 and triggered ES cell differentiation in the presence of LIF; association of NODE subunits with Nanog target genes was maintained in Mbd3–/– ES cells (Fig. 4), suggesting that Nanog and Oct4 form unique HDAC-repression complexes to control ES cell self-renewal. Mbd2 and Mbd3 could be recruited to the Nanog-target genes, such as Oct4, during ES cell differentiation in response to retinoic acid29. In addition, Mbd3–/– ES cells failed to silence pluripotent marker genes


737

letters (for example, Nanog, Oct4 and Rex1) during differentiation and early embryogenesis11, highlighting the importance of Mbd3 in ES cell differentiation. The NuRD-related repression complex NODE does not require Mbd3 to control ES cell self-renewal, and may represent the building block for further assembly of higher-order regulatory complexes on Nanog or Oct4 target genes, once the differentiation process is in motion. Recent findings of histone H3 bivalent modification in undifferentiated ES cells highlight the complex array of signals and markers that ES cells utilize as they undergo differentiation30. Our data indicate that there is an additional level of control for gene transcription in ES cells through selective association with active subunits of repression complexes. Analogous to bivalent marks, the assembly of ES-cell-specific repression complexes (such as NODE) may be necessary to maintain Nanog and Oct4 target gene expression in a semi-active or -repressive state. It will be of great interest to determine the relationship between bivalent modification domains and the unique repression complexes we have identified. METHODS Cell lines. AB2.2 mouse ES cells (passage number 18, provided by Darwin Core facility, Baylor College of Medicine, Houston, TX). The Mbd3–/– ES cell line was provided by Brian Hendrich11. All ES cells were grown in ES medium with 15% fetal bovine serum (FBS), either on irradiated mouse embryonic fibroblasts as feeder cells or with LIF4. Affinity purification of Nanog and Oct4 complexes. Nuclear extracts (approximately 10 mg ml–1) of mouse ES cells (1–2 × 109) were incubated with 20 µg of anti-Nanog (BL1662, Bethyl Laboratories) or anti-Oct3/4 (sc-9081, Santa Cruz) antibodies at 4 °C, with rocking, for 2 h and then with100 µl protein A beads (Santa Cruz) for another 1 h. The bound proteins were washed four times with NETN (20 mM Tris at pH 8.0, 100 mM NaCl, 1 mM EDTA and 0.5% NP-40). Proteins were eluted in Laemli loading buffer, resolved by SDS–PAGE and stained with Coomassie blue. The gel was subsequently sectioned and sequenced on a liquid chromatography tandem mass spectrometer, as described previously31. Peptide identification was performed using the Bioworks software. The parameters we used were Xcorr 1.5, 2.0 and 2.5 for the three charge states. Peptide and protein probability cut-off were both 0.001. Peptide spectra were individually checked manually. Co-immunoprecipitation and western blotting. ES cell nuclear extracts (150 µl for each immunoprecipitation) were immunoprecipitated with appropriate antibodies (3 µg for each immunoprecipitation) followed by western blotting. As controls, either rabbit IgG (for Nanog and HDAC2) or a goat anti-ICAM-1 (for MTA2 and Oct4) antibody was used. For interaction studies, 293T cells were cotransfected with plasmids encoding various potential Nanog-associated proteins in different combinations using the calcium phosphate method. At 48 h after transfection, cells were lysed in cell lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 10% glycerol with protease inhibitor mixture) for 30 min. Whole-cell extracts were collected and incubated with anti-FLAG M2 antibody. Immunoprecipitation with an anti-M2 antibody and western blotting with an anti-Nanog antibody were carried out as described previously32. As controls, whole-cell extracts were analysed by blotting with anti-Nanog and anti-FLAG-tag antibodies. GST pull down. Bacterial GST–Oct4 fusion proteins were purified using glutathione agarose beads. For in vitro expression of Nanog constructs TNT Quick Coupled Transcription/Translation System (Promega) was used in the presence of 35S-Met. The reaction mixture was used for GST pull down. Expression constructs and antibodies. For expression in 293T cells, full-length Oct4, PML, human HDAC1, mouse Hdac2, Mbd3 and GATA2Da/b were cloned into the pCL vector (FLAG-tagged) and Nanog was TOPO-cloned into pcDNA3.1 (Invitrogen). Nanog, Hdac2, Mta1, Mbd3 and human TIN2 and HDAC1 were also cloned into BiFC vectors24. The following antibodies were used: anti-MTA1/2

738

(sc-9447, recognizes both MTA1 and MTA2), anti-MTA1 (Bethyl Laboratories), anti-Oct4 (sc-8628), anti-Oct3/4 (sc-9081), anti-PML (sc-18423), anti-HDAC2 (sc-9959), anti-HDAC1 (sc-6298 and 7872), anti-MBD3 (sc-9402), anti-RbAp46/ Rbbp7 (sc-8272), anti-mSinA (sc-767), and anti-ICAM-1 (sc-1510) (Santa Cruz Biotechnology), anti-Nanog (BL1662), anti-Nanog (BL1663) and anti-HDAC2 (BL2667) (Bethyl Laboratories) antibodies. Anti-FLAG M2 antibody-conjugated agarose beads (Sigma) were used for immunoprecipitation and western blotting, respectively. Bimolecular fluorescence complementation (BiFC). BiFC was performed as described previously24. Briefly, Venus YFP N-terminal domain (residues 1–155) was fused to TRF2, Nanog or HDAC1. YFP C-terminal domain (YFPc, residues 156–239) was fused to TIN2, Oct4 or Mbd3. These vectors were cotransfected in 293T cells. One day after transfection, the cells were collected for flow cytometry analysis on a cytometer (Guava PCA). RNAi knockdown and RT-quantitative PCR. To knockdown Nanog, Mta1, Mta2, Hdac2 and Mbd3, retroviral vectors expressing different shRNA sequences driven by mouse U6 promoter were used as described previously33. For RNA interference, the following sequences were used: shMta2: GGTTGTCTGTCTTTTCCGGttcaagagaCCGGAAAAGACAGACAACC; shHdac2: CCAATGAGTTGCCATATAATttcaagagaATTATATGGCAACTCATTGG; shMbd3: GAAGCTAAGTGGATTGAGTttcaagagaACTCAATCCACTTAGCTTC; shNanog: GTAGCTCAGATATGCAATAttcaagagaTATTGCATATCTGAGCTAC; shMta1-1: GAAATATGGTGGCTTGAAAttcaagagaTTTCAAGCCACCATATTTC; shMta1-2: GGACATATTGGAAGAAATAttcaagagaTATTTCTTCCAATATGTCC; EGFP-1: CACAAGCTGGAGTACAACTttcaagagaAGTTGTACTCCAGCTTGTG; EGFP-2: GCACAAGCTGGAGTACAACTttcaagagaAGTTGTACTCCAGCTTGTG; Scramble: GTATTATCATGCGCGGATTttcaagagaAATCCGCGCATGATAATAC. Total RNA was isolated using an RNeasy Mini Kit (Qiagen). Reverse transcription was performed using iScript Select cDNA Synthesis Kit (BioRad). Real-time PCR was performed using an ABI PRISM 7300 Sequence Detection System and SYBR Green Master Mix as described previously34. Colony formation and alkaline phosphatase assays. ES cells were plated at 100–200 cells per well in 6-well plates in the presence of LIF (103 U ml–1). After overnight incubation, the cells were cultured in the presence or absence of LIF for three or five days and then stained with the alkaline phosphatase detection kit (Chemicon)11. Gel filtration and western blotting. ES cell nuclear extracts (0.5 ml) were applied to gel filtration columns (Superose-6 HR) and protein complexes were separated into different fractions according to their size, as described reviously31. Fractions 9–31 from the gel filtration were resolved by SDS–PAGE and probed for Oct4, Nanog, PML, Hdac2, Mta1, Mta2 and Sin3A. Chromatin immunoprecipitation (ChIP). Mouse ES cells were grown to 80–90% confluence and were chemically crosslinked by the addition of fresh formaldehyde solution (37%) to a final concentration of 1% for 10 min at room temperature with gently shaking. Cells were rinsed twice with cold 1 × PBS followed by addition of 2 M glycine to stop crosslinking and were collected using a silicon scraper. Cells were lysed and sonicated to solublize and shear crosslinked DNA, as described previously, with a minor modification35. Briefly, we used a Virsonic 600 and sonicated at power 4 for 10 × 10 s pulses (30 s pause between pulses) at 4 °C while samples were immersed in an ice bath. The resulting wholecell extract was precleared with 50 µl protein A beads, 10 µl IgG, 10 µl 5% BSA and 5 µg of sheared salmon sperm DNA for each sample. After centrifugation, one fifth of the supernatant was incubated overnight at 4 °C with 30 µl of Protein A agarose beads and 3 µg of the appropriate antibodies, 1 µl BSA (5%) and 25 µg of sheared salmon sperm DNA. Beads were washed four times with ChIP buffer and once with TE containing 1 mM dithiothreitol. Bound complexes were eluted from the beads and crosslinking was reversed by overnight incubation at 65 °C. Whole-cell extract DNA (reserved from the sonication step) was also treated to reverse crosslinking. Immunoprecipitated DNA and whole-cell extract DNA were then purified by Qiagen PCR purification kit (Qiagen). HDAC assay and western blotting. Trichostatin A (TSA)-treated hyperacetylated nucleosomes were used as substrates for the deacetylase assay, as


letters described previously17. Briefly, different antibodies (rabbit IgG, Nanog, ICAM-1, Oct4 and MTA2; 4 µg) were used for immunoprecipitation with ES cell nuclear extracts (200 µl). Protein A/Protein G beads (15 µl) were used to pull down complexes for each immunoprecipitation. Beads were washed 4 times with NETN and once with deacetylase buffer (75 mM Tris-HCl, pH 7.0, 10 0 mM NaCl, 2 mM β-mercaptoethanol and 0.1% EDTA). For each reaction, washed immunoprecipitated beads (5 µl) were incubated with hyperacetylated nucleosomes (1 µg) with or without TSA (300 nM, final concentration) at 30 °C with shaking for 1 h. Samples were centrifuged briefly and supernatants were used for western blotting to examine changes in the levels of acetylated H4K8. The protein level of histone H3 was used as a loading control. For HDAC fluorimetric assay, anti-Nanog, MTA2 and rabbit IgG immunoprecipitates from ES cell nuclear extracts were washed once with assay buffer provided with an HDAC fluorimetric assay kit (BIOMOL). The immunoprecipitates were then incubated with fluorescence-producing substrate (Fluor de lys substrate) with or without TSA for 1 h at room temperature and treated with developer solution for 10 mins. Fluorescence signals were detected in a 96-well plate using Victor3 V 1420 Multilabel Counter (Perkin Elmer). Note: Supplementary Information is available on the Nature Cell Biology website. Acknowledgements We thank Amin Safari, Xueping Xu and Ok-Hee Lee for technical assistance. We also thank Yaoyun Liang and Xinhua Feng for helping us with quantitative PCR analysis. We thank Brian Hendrich for providing the Mbd3–/– ES cells. This work was supported by NIH grants GM69572 and GM81627 to Z.S and DK73524 to A.J.C. D.L. is supported by American Heart Association. Z.S. is a Leukemia and Lymphoma Society scholar. Author contributions J.L., M.W., Y.Z., P.G., H.X., S.Y.J., J.Q. and D.L. performed the experiments; J.W. and A.J.C. provided reagents; Z.S. provided advice on the experimental design and wrote the manuscript. All authors commented on the manuscript. Competing financial interests The authors declare no competing financial interests. Published online at http://www.nature.com/naturecellbiology/ Reprints and permissions information is available online at http://npg.nature.com/ reprintsandpermissions/ 1. Chambers, I. & Smith, A. Self-renewal of teratocarcinoma and embryonic stem cells. Oncogene 23, 7150–7160 (2004). 2. Rossant, J. Stem cells from the mammalian blastocyst. Stem Cells 19, 477–482 (2001). 3. Nichols, J. et al. Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95, 379–391 (1998). 4. Chambers, I. et al. Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells. Cell 113, 643–655 (2003). 5. Mitsui, K. et al. The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell 113, 631–642 (2003). 6. Takahashi, K. & Yamanaka, S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663–676 (2006). 7. Okita, K., Ichisaka, T. & Yamanaka, S. Generation of germline-competent induced pluripotent stem cells. Nature 448, 313–317 (2007).

8. Wernig, M. et al. In vitro reprogramming of fibroblasts into a pluripotent ES-cell-like state. Nature 448, 318–324 (2007). 9. Yu, J. et al. Induced Pluripotent stem cell lines derived from human somatic cells. Science 318, 1917–1920 (2007). 10. Takahashi, K. et al. Induction of Pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861–872 (2007). 11. Kaji, K. et al. The NuRD component Mbd3 is required for pluripotency of embryonic stem cells. Nature Cell Biol. 8, 285–292 (2006). 12. Wang, J. et al. A protein interaction network for pluripotency of embryonic stem cells. Nature 444, 364–368 (2006). 13. Wu, Q. et al. Sall4 interacts with Nanog and co-occupies Nanog genomic sites in embryonic stem cells. J. Biol. Chem. 281, 24090–24094 (2006). 14. Smith, C. L. & Peterson, C. L. ATP-dependent chromatin remodeling. Curr. Top. Dev. Biol. 65, 115–148 (2005). 15. Workman, J. L. Nucleosome displacement in transcription. Genes Dev. 20, 2009–2017 (2006). 16. Xue, Y. et al. NURD, a novel complex with both ATP-dependent chromatin-remodeling and histone deacetylase activities. Mol. Cell 2, 851–861 (1998). 17. Zhang, Y. et al. Analysis of the NuRD subunits reveals a histone deacetylase core complex and a connection with DNA methylation. Genes Dev. 13, 1924–1935 (1999). 18. Toh, Y., Pencil, S. D. & Nicolson, G. L. A novel candidate metastasis-associated gene, mta1, differentially expressed in highly metastatic mammary adenocarcinoma cell lines. cDNA cloning, expression, and protein analyses. J. Biol. Chem. 269, 22958–22963 (1994). 19. Ahringer, J. NuRD and SIN3 histone deacetylase complexes in development. Trends Genet. 16, 351–356 (2000). 20. Shi, Y. et al. Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119, 941–953 (2004). 21. Wu, W. S. et al. The growth suppressor PML represses transcription by functionally and physically interacting with histone deacetylases. Mol. Cell Biol. 21, 2259–2268 (2001). 22. Boyer, L. A. et al. Core transcriptional regulatory circuitry in human embryonic stem cells. Cell 122, 947–956 (2005). 23. Loh, Y. H. et al. The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nature Genet. 38, 431–440 (2006). 24. Chen, L. Y., Liu, D. & Songyang, Z. Telomere maintenance through spatial control of telomeric proteins. Mol. Cell Biol. 27, 5898–5909 (2007). 25. Feng, Q. & Zhang, Y. The NuRD complex: linking histone modification to nucleosome remodeling. Curr. Top. Microbiol. Immunol. 274, 269–290 (2003). 26. Saitou, M., Barton, S. C. & Surani, M. A. A molecular programme for the specification of germ cell fate in mice. Nature 418, 293–300 (2002). 27. Bortvin, A. et al. Incomplete reactivation of Oct4-related genes in mouse embryos cloned from somatic nuclei. Development 130, 1673–1680 (2003). 28. Manavathi, B. & Kumar, R. Metastasis tumor antigens, an emerging family of multifaceted master coregulators. J. Biol. Chem. 282, 1529–1533 (2007). 29. Gu, P., Le Menuet, D., Chung, A. C. & Cooney, A. J. Differential recruitment of methylated CpG binding domains by the orphan receptor GCNF initiates the repression and silencing of Oct4 expression. Mol. Cell Biol. 26, 9471–9483 (2006). 30. Bernstein, B. E. et al. A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125, 315–326 (2006). 31. Liu, D. et al. PTOP interacts with POT1 and regulates its localization to telomeres. Nature Cell Biol. 6, 673–680 (2004). 32. O’Connor, M. S., Safari, A., Xin, H., Liu, D. & Songyang, Z. A critical role for TPP1 and TIN2 interaction in high-order telomeric complex assembly. Proc. Natl Acad. Sci. USA 103, 11874–11879 (2006). 33. Xin, H. et al. TPP1 is a homologue of ciliate TEBP-β and interacts with POT1 to recruit telomerase. Nature 445, 559–562 (2007). 34. Chew, J. L. et al. Reciprocal transcriptional regulation of Pou5f1 and Sox2 via the Oct4/ Sox2 complex in embryonic stem cells. Mol. Cell Biol. 25, 6031–6046 (2005). 35. Yoon, H. G., Chan, D. W., Reynolds, A. B., Qin, J. & Wong, J. N-CoR mediates DNA methylation-dependent repression through a methyl CpG binding protein Kaiso. Mol. Cell 12, 723–734 (2003).


739

s u p p l e m e n ta r y i n f o r m at i o n

Fl

ag Fl -T ag IN Fl -R 2 ag A -H P1 d Fl ac2 ag -M bd 3

Supplemental Fig. 1 Liang e

Anti-Flag IP

49

WB: Nanog

82 64 49

WB: Flag

49

10% Input

82 64 49

WB: Nanog

WB: Flag

1X 3X 9X

Supplemental Figure 1. Nanog can interact with Hdac2 but not Mbd3. 293T cells were transiently transfected with expression vectorswestern encoding Nanog together with constructs blotted (WB) with anti-Nanog or FLAG antibodies. Increasing amounts Figure S1 Nanog can interact with Hdac2 but not Mbd3. 293T cells were of Mbd3 vectors were Whole co-transfected cell to show that Mbd3 failed to bind Nanog were transiently transfected with expression vectors encoding Nanog together with encoding FLAG-tagged Hdac2, Mbd3, RAP1, or TIN2. extracts (WCE) even at high concentrations. WCE input was also blotted with anti-Nanog or constructs encoding FLAG-tagged Hdac2, Mbd3, RAP1, or TIN2. Whole cell immunoprecipitated (IP) with anti-Flag antibodies and western blotted (WB) with anti-Nanog FLAG antibodies. TIN2 and RAP1 were used as IP negative controls. extracts (WCE) were immunoprecipitated (IP) with anti-Flag antibodies and or FLAG antibodies. Increasing amounts of Mbd3 vectors were co-transfected to show that Mbd3 failed to bind Nanog even at high concentrations. WCE input was also blotted with anti-Nanog or FLAG antibodies. TIN2 and RAP1 were used as IP negative controls.

www.nature.com/naturecellbiology

1


Relative acetylated protein level

Supplemental Fig. 2 Li

1.2 1 0.8 0.6 0.4 0.2 0 -

og

n Na

A TS

og

n Na

+

A TS

A- A+ SAS S T T 4 t4 T a2 t c t c O M O

Supplemental Figure 2. Nanog and Oct4 associate with HDAC activity. The data in Fig total histoneexperiments. H3 amount. Relative acetylated H3 protein levels after Figure S2 quantified Nanog and Oct4 associate with HDAC activity. The data in 3c were by averaging two independent The HDAC activity associa treatment were compared. The signals in TSA treated samples were Figure 3c were quantified by averaging two independent experiments. considered asH3 1. HDAC activity Nanog or Oct4 was normalized to withTheNanog orassociated Oct4 with was normalized to total histone amount. Relative acetylated H3 pro levels after treatment were compared. The signals in TSA treated samples were considered a

2



Relative enrichment

Supplemental Fig. 3 Liang et.al

2 1.5

Mbd3 IP Rbbp7 IP IgG IP

1 0.5 0

Esrrb

Nanog

Pou5f1

Supplemental Figure 3. Mbd3 and Rbbp7 are not enriched on Nanog target genes. ES cell Figure ChIP experiments were performed usingagainst total rabbit IgG, or antibodies against Mbd3 or Rbbp7. ChIP data were quantified by real-time PCR. Error S3 Mbd3 and Rbbp7 are not enriched on Nanog target genes. ES bars indicatePCR. standard errors (n=3). bars indicate standard ChIP experiments were performed using totalquantified rabbit IgG, or antibodies Mbd3 orcellRbbp7. ChIP data were by real-time Error errors (n=3).


3



Relative expression (folds)

1.2 1 0.8 0.6 0.4 0.2 0

g P c2 d3 a1-1 a1-2 ta2 ano GF Hda Mb t t M h s sh shN sh shM shM sh

Supplemental Figure 4. RNAi knockdown of Nanog, Mta1, Mta2, Hdac2, and Mbd3 mouse ES cells. Individual shRNA sequences against different genes were introduced mouse ES cells by retroviral infection. Three days after drug selection, mRNA levels of Hda Mta1, Mta2, Mbd3, and Nanog were examined by quantitative RT-PCR. Error bars indic standard errors (n=3). Figure S4 RNAi knockdown of Nanog, Mta1, Mta2, Hdac2, and Mbd3 in mouse ES cells. Individual shRNA sequences against different genes were introduced into mouse ES cells by retroviral infection. Three

4

days after drug selection, mRNA levels of Hdac2, Mta1, Mta2, Mbd3, and Nanog were examined by quantitative RT-PCR. Error bars indicate standard errors (n=3).




Relative enrichment (folds)

Gata6 20 IgG IP Nanog IP Oct4 IP Mta1 IP Hdac2 IP

15

10

5

0 -3K

-1K

0 (TSS)

+1K

+3K

3' UTR

Nanog

Supplemental Figure 5. Co-occupancy of NODE components on Gata6. ES cell gene IgG, locus. Nanog ChIP was used as a positive control, and 3’ UTR as a Figure S5 Co-occupancy of NODE components on using Gata6. ES cell ChIP rabbit experiments were performed total or antibodies against Nanog, Oct4, Mta negative control. Transcription start site (TSS) was designated as 0. 1K experiments were performed using total rabbit IgG, or antibodies against Hdac2. ChIP quantified by Real-time PCR using sets and 3K, 1 kb and 3kb awayprimer from TSS. Error bars corresponding indicate standard errors to diffe Nanog, Oct4, Mta1, data or Hdac2.were ChIP data were quantified by Real-time (n=3). PCR using primer sets corresponding to different regions of the Gata6 regions of the Gata6 gene locus. Nanog ChIP was used as a positive control, and 3’ UTR negative control. Transcription start site (TSS) was designated as 0. 1K and 3K, 1 kb and away from TSS. Error bars indicate standard errors (n=3).


5

s u p p l e m e n ta r y i n f o r m at i o n Fl a Fl g-G a Fl g- AT a G A Fl g-T AT D2 ag IN AD A ve -R 2 2 B ct A or P1

Hd

ac

2

IP

b 3% Co Inp n u Na tro t no l IP g IP

a

Fl a Fl g-O a Fl g-Hct4 ag D -H AC da 1 c2

Supplemental Fig. 6ab Liang et.al

82 WB: Nanog 64 49

82

Mta1/2 64

Anti-Flag IP

116 82

Pml

82 64 WB: Flag 49

64

Hdac2

49 49

Oct4

82 64 WB: Nanog 49

37

5% Input 37

Mbd3

82 64 WB: Flag 49

26

Fig. 1b

Supplemental Fig. 6cd Liang et.al

IP

t4

In Co pu nt t ro Na l IP no g Na Ab no 1 I P g Ab Na no 1 IP g (L Ab S) 2 IP M ta 1 IP

Oc

In Co put nt ro lI M P ta 2 IP

d scan gel images. Supplemental Figure 6. Key full

1%

c

Fig. 1d

49

Nanog

3%

37

64 49

Hdac2

82

Mta1/2

64

26

Mbd3

64

Hdac2

49

49 37

Nanog

37 26

Mbd3

Fig. 3a

Fig. 2d Figure S6 Key full scan gel images.

Supplemental Figure 6. Key full scan gel images. 6



Supplemental Fig. 6ef Liang et.al

82

Mta1/2

-2 d ac 2 sh M bd sh M 3 ta sh M 2 ta 1 sh M 1 ta 1 -2 sh N an o g

-1

sh H

FP

FP sh G

sh G

Ab 2

f

Na no g

IP ta 1 M

3%

In Co put nt ro lI P

IP

e

82

Mta1 Mta2

64

64

Hdac2 49 64

Hdac2

49

Rbbp7 49

64

Tubulin Mbd3

26

49

Nanog 37

Fig. 3b

Fig. 5a

Supplemental Figure 6. Key full scan gel images.


7

Protein

Complex

Transcriptional Function

SMARCA2

SWI/SNF

Chromatin remodeling

SMARCA4

SWI/SNF


BAF180

SWI/SNF


Rif1

?

?

Sall4

Nanog

Transactivator

Err2

Nanog

Transcription factor

Oct4

Nanog


Mta1 and Mta2

NuRD

Repressor

Gatad2a and 2b

NuRD

Repressor

Hdac1 and Hdac2 NuRD, SIN3A, CoREST

Repressor

Sin3A

SIN3A

Repressor

Pml

?

Repressor?

Kap1

SIN3A

Repressor

LSD1

CoREST ect.

Repressor

Supplemental Table I Large-scale identification of Nanog-associated proteins by mass spectrometry. Bold, putative Nanog associated proteins identified in this study only.

Protein

Complex

Transcriptional Function

Sall4

Nanog

Transactivator

Oct4

Nanog


Mta2

NuRD

Repressor

Gatad2a, 2b

NuRD

Repressor

Hdac2

NuRD, SIN3A, CoREST

Repressor

Hdac1

NuRD, SIN3A, CoREST

Repressor

Supplemental Table II Large-scale identification of Oct4-associated proteins by mass spectrometry. Bold, proteins previously not shown to associate with Oct4.

Supplemental Table III PCR primers (in a separate excel file).