Transcript ES-cell specific enhanceosomes

June 13, 2008
Cited by 705
Authors
First author
• Xi Chen
Corresponding authors
• Chia Lin WEI, Ph.D.
• Huck-Hui Ng , Ph.D.
• Group Leader
• Executive Director
• Genome Institute of
Singapore
• National University of
Singapore
• Genome Institute of
Singapore
They focus on stem cell biology and are addressing two questions:
1) What makes a stem cell a stem cell?
2) How to make a non-stem cell a stem cell?
Background
Embryonic Stem(ES) Cell
• Pluripotency (differentiate into almost all lineages)
• Self-renewing ability
• Revolutionized biological research through the creation of genetically
altered animals.
• Human ES cell (promising)
Pluripotency
Self-renewal
Main Purpose
How is the ES cell pluripotency
maintained?
Maintenance of Pluripotency
•
•
•
Extrinsic: the LIF and BMP signaling pathways play a central role in the
maintenance of a pluripotential stem cell phenotype.
1. Leukemia Inhibitory Factor (LIF)
The binding of LIF to its receptor activates STAT3 through phosphorylation
•
•
2. Bone Morphogenetic Proteins (BMPs)
The binding of BMP4 to its receptors triggers the phosphorylation of Smad1 and
activates the expression of members of the Id(inhibitor of differentiation) gene
family
•
Intrinsic: factors such as transcription factors (TFs) are also essential for
specifying the undifferentiated state of ES cells.
– Oct4, Sox2, c-Myc, Klf4, Nanog, Esrrb, Zfx
Main Design
Gain insights into the transcriptional regulatory networks in ESC
Because Transcription factors (TFs) and their specific interaction
with targets are crucial for specifying gene expression programs
Objects
• 13 transcription factors: Nanog, Oct4, STAT3, Smad1, Sox2, Zfx,
c-Myc, n-Myc, Klf4, Esrrb, Tcfcp2I1, E2f1, CTCF
• 2 transcription regulators: p300, Suz12
Methods
• ChIP-seq
Outline
• ES-cell specific enhanceosomes




Mapping of Binding Sites of 13 TFs by Using ChIP-seq
Motif Analyses of TFBSs
A Subset of Multiple Transcription-Factor-Binding Loci
Nanog-Oct4-Sox2 cluster function as ES-Cell enhanceosomes
 p300 Is Recruited to the Nanog-Oct4-Sox2 Cluster
• ES-cell Regulatory Network
 Combinatorial Binding of TFs is correlated with ES-CellSpecific Expression
 Regulatory Network Defining ES-Cell Specific Expression
Mapping of Binding Sites of 13 TFs by Using ChIP-seq
ChIP-seq
Determine the threshold through Monte Carlo simulations
Remove peaks that were also found in the negative control library
Use ChIP-qPCR to further refine the threshold used
Valid
Motif Analyses of TFBSs
Matrices predicted by the de novo motif-discovery algorithm Weeder
A Subset of Multiple Transcription-Factor-Binding Loci
MTL: Multiple Transcription-Factor-Binding Loci
• Plot of the number of TFs bound per co-bound locus. The
distribution of randomly occurring co-bound loci is obtained by
simulation
A Subset of Multiple Transcription-Factor-Binding Loci
Distribution of clusters with different numbers of co-bound TFs.
(Promoter regions are defined ass sequences 2500 bp upstream and 500 bp
downstream of TSS)
MTL Associated with Nanog, Oct4, Sox2, Smad1, and STAT3
as ES-Cell Enhanceosomes
32.9%
43.4%
87.4%
56.8%
The convergence of the two key signaling pathways
(via Smad1 and STAT3) with the core circuitry
defined by Nanog, Oct4, and Sox2
MTL Associated with Nanog, Oct4, Sox2, Smad1, and STAT3
as ES-Cell Enhanceosomes
The binding sites of Nanog group are likely ES-cell specific enhancers
MTL Associated with Nanog, Oct4, Sox2, Smad1, and STAT3
as ES-Cell Enhanceosomes
The binding of Smad1 and STAT3 depend on Oct4, but otherwise is not
p300 Is Recruited to the Nanog-Oct4-Sox2 Cluster
the occurrence of p300 in different MTL types
• p300 was found to co-occur with
the Nanog-Oct4-Sox2 cluster
• Most p300-binding sites are
associated with 3–6 other TFs
• The composition of most p300containing clusters include Nanog,
Oct4, or Sox2
p300 Is Recruited to the Nanog-Oct4-Sox2 Cluster
ChIP-qPRC of p300
Binding of p300 to the genomic sites
depends on Oct4, Sox2 and Nanog
motif from p300-enriched sequences
resembles the sox-oct element
Outline
• ES-cell specific enhanceosomes




Mapping of Binding Sites of 13 TFs by Using ChIP-seq
Motif Analyses of TFBSs
A Subset of Multiple Transcription-Factor-Binding Loci
Nanog-Oct4-Sox2 cluster function as ES-Cell enhanceosomes
 p300 Is Recruited to the Nanog-Oct4-Sox2 Cluster
• ES-cell Regulatory Network
 Combinatorial Binding of TFs is correlated with ES-CellSpecific Expression
 Regulatory Network Defining ES-Cell Specific Expression
Combinatorial Binding of Transcription Factors Is Associated with ESCell-Specific Expression
Class I genes are enriched in
binding sites for Nanog, Oct4,
Sox2, Smad1
Class II genes are bound
heavily by c-Myc and n-Myc
Combinatorial Binding of Transcription Factors Is Associated with ESCell-Specific Expression
the expression level of classI,
class II and III are higher than
genes in class IV and class V
60% of genes upregulated in ES
cells are from class I and class II
Combinatorial binding patterns of
TFs have predictive power for EScell-specific expression
Regulatory Network Defining ES-Cell Specific Expression
• Aim:
– Construct a network that specifies ES-cells
• Dataset:
– 2 public undifferentiated v.s differentiated gene expression datasets.
– Chip-Seq results generated by this study
Regulatory Network Defining ES-Cell Specific Expression
Workflow (1)
Gene
• Two tables
TSS
– Genes ranked by their expression fold change
– Genes ranked by their binding score to a TF
K= 10 kb
K= K+ 10 kb
Expression-ranked
genes list
Big
Binding-ranked
genes list
Yes
No
Chip-Seq peaks
in TSS ± k ?
Yes
K <= 1Mb ?
No
Fold-Change
Score = 0
Score = # Tags
associated with
called peak
Small
+k
-k
TSS
Regulatory Network Defining ES-Cell Specific Expression
Workflow (2)
Note:
A responder analysis is one in which each subject is classified as either a
‘responder’ or a ‘non-responder’.
Expression-ranked
genes list
Binding-ranked
genes list for a TF
Responder Analysis
TF
Gene
ESC regulatory network
Summary
•
TFs are wired to the ES-cell genome in two major ways.
– Nanog, Oct4, Sox2, Smad1, STAT3
– c-Myc, n-Myc, Zfx, E2f1
•
Highly dense binding loci involving these factors have characteristic features
of enhanceosomes
•
The coactivator p300 is predominantly recruited to dense binding loci
involving proteins found in the first cluster
•
Constructed a transcriptional regulartory network model that integrates the
two key signaling pathways with the intrinsic factors in ES cells.
Discussion (1) -- comparison
Chen et al
•
•
The authors did a concurrent survey of the location of multiple TF
in a single cell type.
Shares some similarities to the work of Kim et al.
Chen et al
Kim et al
Organism
Mouse
Mouse
TF
13 TF+ 2 TR
9 TF
Platform
Chip-Seq
promoter DNA microarrays
Anti-body
Target endogenous proteins
Target biotin-tagged proteins
Network
constructio
n
Chip-Seq + Gene expression
Chip-Seq + PPI
Number of
networks
1
3
Kim et al
STAT3,
Smad1,
Oct4, Sox2,
Zfx,
Klf4,c-Myc,
n-Myc,
Nanog
Esrrb,
Tcfcp2l1,
E2f1,CTCF
Dax1,
Rex1,
Zpf281
, Nac1
Discussion (2) -- enhancersome
Enhanceosome : is a nucleoprotein complex composed of distinct sets of TFs bound directly or
indirectly to enhancer DNA
•
The study showed that the Nanong-Oct4-Sox2 cluster exhibits features of enhencesome:
– The binding sites are densely clustered within relatively compact genomic segments.
– These regions act as enhancer when placed downstream of the luciferase reporter.
– Associated with active region marks (H3K4me3).
– P300 (enhancer marker) is recruited to the Nanog-Oct4-Sox2 cluster.
Daniel,P. et al, Cell, 2007
Thanos,D. and Maniatis,T., Cell, 1995
Discussion (3) -- weakness
• To further confirm the enhancersome's function genome wide and the
relationship between the different TFs
- STARR-seq
- biochemical verification
• The network might involve more regulatory factors.
Acknowledgements
Pro Ren
Group members:
Zehua Liu, Jingyi Wu, Mohamed Nadhir Djekidel
All the audience