Transcriptional control of lymphopoiesis

IMM1016: Recent Advances in Immunology
October 15, 2012
Michele K. Anderson
Sunnybrook Research Institute
[email protected]
The “Who Cares” Question
Developmental processes result in the establishment of cellular
Cell identity in a large part is dependent on the genes that are
expressed and those that are poised to be expressed in response to
external stimuli
During development genes are turned on that start the process of
establishing a certain identity, but other genes need to be turned off to
allow the locking in of that identity
Networks of interactions are required to keep the system stable even in
the absence of the signals that set the whole process in motion
Early stages of T cell development: setting the stage
Specification versus Commitment in T Cell Development
Specification: initiation of the T-cell gene expression program
Commitment: loss of the ability to become other cell types
These can be decoupled by genetic manipulation
Key players:
HEB (E-proteins)
Notch1 is critical for entry into the T-cell lineage
Ligand-activated receptor that complexes with RBPj to initiate transcription
Conditional deletion of Notch1: Mx-Cre, Lck
Ectopic expression of Notch1
In vitro expression of DL Notch ligands
OP9-DL1 or DL4 system (J.C. Zuniga-Pflucker)
Widely used as a readout of T cell potential
Can be used to generate large numbers of T cell precursors
Used to test impact of transcription factor deficiency
Developmental potential of early DN thymocytes cultured on OP9 bone marrow stromal cells
in the presence (OP9-DL1) or absence (OP9-control) of Dll1 expression.
Schmitt T M et al. J Exp Med 2004;200:469-479
© 2004 Rockefeller University Press
Two waves of transcription factor upregulation in
response to Notch signaling
Bcl11b enforces T cell commitment
Zinc-finger transcription factor that acts primarily as a repressor
Conditional deletion interrogated using OP9-DL1 co-culture
Deletion of Bcl11b allows development to DN3 stage
OP9-DL1 co-culture
Deletion of Bcl11b abolishes DN3 commitment
Deletion of Bcl11b favors NK and myeloid fates
Bcl11b primarily inhibits alternative lineage regulators
Distinct subcircuits drive specification vs. commitment
HEBAlt and HEBCan belong to the E-protein family of TFs
Network of interactions between T-cell specific
transcription factors
Student Presentation
How does commitment work?
How is the T cell program deployed?
How do these processes intersect?
Global examination of gene expression and chromatin status
Strategy: 1) Isolate five different stages of T cell development
Commitment as defined by OP9-DL1 co-culture assays
2) Use RNAseq to generate a complete catalogue of gene
expression at each developmental stage
Map onto the genome
Example of RNA seq data: HEBAlt in TCRgamma locus activation
Jg1 Cg1
Global changes in gene expression between each stage
All genes differentially expressed between at least two
populations by 2-fold
•Two biological replicates of
each population
•Good agreement between
•Few changes between DN1
and DN2a
•Large changes between DN2a
and DN2b
•FLDN2b and ThyDN3 very
similar to each other
Coordinated upregulation of pre-TCR genes at DN1 to DN2
Two major changes in TF usage: DN2a to DN2b and DN3 to DP
Bcl11b, Lef1, SpiB, Tcf1, Hes1,
PU.1, GATA2, C/EBPbeta, Bcl11a,
LMO2, Hhex, Lyl1
Broad downregulation of
regulators of alternative lineages
2) Use ChiPseq to generate a complete catalogue of chromatin
marks on cis-regulatory elements at each stage
Three histone modifications linked to activation status
•Histone3K(9,14) acetylation (H3Ac): linked to
activation at transcriptional start sites (TSSs)
•Histone3K4me3: linked to silencing
•Histone3K4me2: poised for activation or
silencing (developmentally labile)
Chromatin marks on expressed and silent genes at each stage
•H3Ac and me2 marks are on most expressed genes whereas me3 marks are not
•Most silent genes have no marks but those that do have me2 or me3
•Chromatin marks are good indicators of activation status of genes in these cells
No drastic changes in overall chromatin status between stages
Distal elements are more labile than promoter elements
Each distally marked region was assigned to the closest gene
Some distal elements corresponded to known regulatory sites
Patterns of gene expression over developmental time
C Upregulated
V Transient
C Downregulated
Timing of chromatin changes at TSSs relative to transcription
H3Ac tightly correlated with RNA
Me2 often present before and
after RNA
Me3 associated with repressed
genes but most silent genes
don’t acquire this mark
Next….Zoom in on
hematopoietic genes…
Upregulation of T-cell specific genes in DN and DP stages
Lineage-inappropriate genes downregulated and marked
Lineage-inappropriate genes not expressed but marked
Activation at DN2a to DN2b transition: CD3 versus Bcl11b
CD3 enhancer
available even in
CD3 gradual
acquisition and loss
of HAc3 and me3
Bcl11b much more
abrupt acquisition
and loss of marks at
DN2a to DN2b
Histone marks on T-regulator GATA3 and B-regulator Pax5
(remember, not
on all silenced
Alternative lineage regulators downregulated at DN2a to DN2b
PU.1 in hematopoiesis and early T cell development
PU.1 is expressed at high levels in myeloid cells, intermediate levels in
B cells, and low levels in T cell precursors
PU.1 is high in DN1, low in DN2, very low in DN3, and absent from DP,
SP, and mature T cell precursors
Overexpression of PU.1 in either B cell or T cell precursors redirects
them to the myeloid fate in the presence of the myeloid growth factors
Overexpression of PU.1 in T cell precursors in FTOC blocks T cell
development at the DN stages progressively
However, T cells cannot develop without PU.1. Is this because of an
overall defect in pre-thymic precursors or is PU.1 playing a positive role
in the context of Notch signaling?
Examine binding of PU.1 in T cell precursors using ChIP-Seq
PU.1 binding in T cell precursors versus B cells and macrophages
B cells
PU.1 bound 34K sites in DN T cells, comparable to the number bound in B
cells and macrophages
However, these were primarily different sites than those bound in B cells or
The tighter the plots, the more similar the samples are
PU.1 binding decreases as DN stages progress but the binding
sites remain largely the same
DN1 by DN2a
DN1 by Me2
DN2a by DN2b
DN2b by me2
PU.1 binding is tightly
correlated with H3K4me2
(poised) status throughout the
early stages of T cell
PU.1 binding and chromatin
mark status on Tal1 (SCL) locus
during T cell development
Me2 persists after mRNA halts
Me3 increases and spreads as T cell
development proceeds and mRNA
PU.1 decreases but remains bound to
the same sites as T cell development
PU.1 binding and chromatin
mark status on IL7R during T
cell development
Me2 precedes mRNA expression
Not many me3 marks
PU.1 stays bound to the same site
before and during T cell development
PU.1 is associated with expressed genes during early T cell
development prior to commitment
GATA3 is essential for several stages of T cell development
GATA3 is expressed throughout T cell development at fairly constant
Lack of GATA3 inhibits even the earliest ETP stage of T cell
Overexpression of GATA3 in precursors, however, blocks T cell
development at the DN2 stage
Addition of IL3 and SCL and withdrawal of DL Notch ligands can
however rescue cells into the mast cell lineage
GATA3 plays different roles at different stages of T cell development: bselection, CD4 lineage, Th1 lineage
Does GATA3 bind the same sites throughout T cell development? ChIPseq analysis
GATA3 binds different sites at different stages of T cell development
GATA3 sites on Ly1-1 (expressed early) and Ets2 (expressed in DP)
GATA3 sites on Itk (expressed late) and Ets2 (expressed throughout)
Network of interactions between T-cell specific
transcription factors
What are the target genes?
How are they regulated?
ChIP-seq + RNAseq
Need better antibodies to ChIP
seq more of these factors!

similar documents