Architectural TFs
Download
Report
Transcript Architectural TFs
Architectural TFs
MBV4230
Overview
DNA-binding
TFs
General principles
Architectural factors
MBV4230
Recognition of response elements
Activators vrs Architectural TFs
Ordinary activators with sequence specific DNA
binding
Key recruitment sites for assembly of transcription complexes
Architectural transcription factors playing a more
structural role in the assembly of transcription
complexes
MBV4230
Architectural TFs - brief history
Transcription activation - focus
on more and more dimentions
70-ties: 1-Dimentional understanding
80-ties: 2-Dimentional understanding
≥ RNAPII: TFs binding specific cis-elements
required for selective transcription
TFs mediate regulatory response
Promoters/enhancers: clusters of cis-elements
complex regulation - Several buttons have to be
pushed simultaneouly
Ptashnes simplification - mixed order OK
90-ties: 3-Dimentional understanding
Three-dimentional assembly of TFs required for
correct biological response
MBV4230
3D protein-promoter complexes
- factors dedicated architecture
some factors has a pure architectural
function
designated architectural transcription factors
They lack a transactivation domain (TAD)
Do not function out of their natural context (in contrast to ordinary
acitvators)
Their function is to confer a specific 3D structure on DNA
MBV4230
Classical HMG-proteins
non-histone chromatin proteins - original
defining criteria
high mobility in PAGE
soluble in 2-5% TCA
small < 30 kDa
High content of charged amino acids
abundant: 1 per. 10-15 nucleosomes
MBV4230
Classical HMG-proteins
Three classes of HMG DNA-binding proteins
HMGB
HMG-box family
HMG-AT-hook family
Eks: HMG 1 and HMG 2
Bends DNA substantially
Facilitators of nucleoprotein complexes
Eks.: HMGI(Y)
Antagonizing intrinsic distortions in the conformation of AT-rich DNA
HMG-nucleosome binding family
HMGA
HMGN
Eks.: HMG14 and 17
Mediates moderate destabilization of chromatin higher-order structure
Not present in yeast or fly
HMGB-proteins
MBV4230
HMG1 and 2
3 structural domains
A and B with high homology (80-90 aa)
acidic C-terminal
Interaction with DNA (and histones?)
A and B ≈ DNA
C-term ≈ histone H1 or unknown function
N
++++
A
++++
B
DNA
----
C
Histon H1?
MBV4230
HMG-boxes in architectural proteins
One or two
HMG-box
domains
30 Asp/Glu
acidic
basic
MBV4230
First eukaryotic architectural TF:
LEF1 (Grosschedl 1992)
LEF1: a cell type-specific TF
LEF1 contains an HMG-related domain
LEF1: a sequence-specific TF that binds
CCTTTGAAG
found in enhancer of TCR
LEF1 induces strong bending of DNA - about 130o
Induced bending brings nearly TFs in contact
MBV4230
LEF1 3D
MBV4230
LEF1 3D
MBV4230
A whole family of architectural TFs
with HMG-domains
UBF has repeated HMG-homologous repeats
4-6 ex dimer ≈ 10 HMG-like domains
activator of rRNA gener
UBF-DNA complex scaffold for SL-1 recruitment
Interaction with 180 bp that is packed into a distinct structure
DNA-motif in a series of TFs:
“HMG-box” designate the DNA-sequence-motif
“HMG-domain” designate the protein motif
MBV4230
Two subclasses of HMG-domain
proteins
Proteins with multiple HMG-domains
low sequence-specificity
Ubiquitous - found in all cell types
eks.: HMG1, HMG2, ABF-2, UBF
Proteins with single HMG-domain
(moderate) sequence-specificity
Cell type-specific
eks.: LEF-1, SRY, TCF-1, Sox, Mat-1, Ste11, Rox1
MBV4230
Characteristic DNA-binding
binds minor groove
induce bending of DNA
has high affinity for noncanonical DNA-structures
such as :
cruciform DNA
4-way junctions
cisplatin kinked DNA
+
MBV4230
NMR-structures
Examples
HMG1 B-domain
LEF-1
SRY
Yeast Nhp6p
Drosophila HMG-D
Common: 3 helix L-form
heliks II and III form an angle of about 80o
Conserved aromatic aa in kink
Basic concave side interact with DNA
MBV4230
Similar structures of HMG domains
MBV4230
Minor groove binding,
intercalation and bending
Objective: shorten the distance
between cis-elements facilitating
interaction between bound factors
DNA <500bp relatively stiff induced bending required
Mechanism for induced bending of
DNA
Protein scaffold
HMG B-domain: L-shaped protein
TBP: sadle
Minor groove binding
DNA-binding face = hydrophobic surface that
conforms to a wide, shallow minor groove
4 residues inserted deep into the minor groove
Full or partial intercalation (“kile”)
MBV4230
Intercalation
in protein-induced DNA-bending
Partial intercalation in the DNA helix of a protein side
chain introduces a kink in the DNA enhancing the
bend
Large hydrophobic residues (N-term helix I) partially intercalates between two
base pairs
The A-box HMG domain has only an Ala in the X position not large enough to
intercalate,
Intercalation linked to bending also seen in other
factors
Partial (TBP)
Full (ETS1)
Inserted side chain unstacks two basepairs
side chain as stacking-partner
side chain penetrates into the helix
side chain (Trp) as new stacking-partner
Result: helix axis direction altered
MBV4230
Two points of intercalation, X and Y
Basic tail
Binds
Major
groove
X only
X and Y
Y only
X = major kink and intercalation site, Y=second kink due to partial intercalation
MBV4230
Cooperation with TFs
A major role of non-seq.spec.
architectural factors is to facilitate
formation of complex nucleoprotein
assemblies
Need interaction with sequence specific TF to be
directed to precise locations
An introduced bend could facilitate binding of one
factor, and this could subsequently assist a second
factor
The seq.spec. architectural factors is
known to participate in the formation
of complex nucleoprotein assemblies
like enhanceosomes
TCR and Interferon b
MBV4230
Are all TFs architectural?
A large number of publications “TFx bends DNA”
positive reports “TFx bends DNA”
negative reports “TFx does not bend DNA”
All TFs that bind on one side of DNA will induce
bending due to one-sided neutralization of charge
Degree of bending will depend on ionic condition
Uncertain if biologically relevant
The term “Architectural TFs“ should be reserved for
factors with a particularly developed bending
mechanism
MBV4230
The charge neutralization model
- - - - - - - - - - -- -- - - + - - - ++
--+
+
+
-+
+
+
+
--+
+
2
--3
+
-1
- - --Sp1
Asymmetrical charge neutralization
Bending
+ - -+-+- - - -+ +
+
- - - - -+-+
+ + - - -- +
- -- -+
+ +- - - - - ++- -+ + + ++ - - - +
-+ +
--- + + 2
--+
3
1
Sp1
Bending effect of charge neutralization
reduced in the pr esence of multvalent cations
2. subgruppe: HMGA
.. First described by Søren Laland,
an almost forgotten discovery
MBV4230
HMGA - proteins with AT-hook
The mammalian HMGI/Y (HMGA) proteins participate
in a wide variety of cellular processes
All members have multiple copies of a DNA-binding
motif called the `AT hook'
including regulation of gene trx and induction of neoplastic transformation and
promotion of metastatic progression.
that binds to the narrow minor groove of stretches of AT-rich sequence.
The proteins have little secondary structure in
solution but assume distinct conformations when
bound to DNA or other proteins
Their flexibility allows the HMGI/Y proteins to induce both structural changes
in chromatin substrates and the formation of stereospecific complexes called
`enhanceosomes'. Reciprocal conformational changes occur in both the
HMGI/Y proteins themselves and in their interacting substrates.
MBV4230
Members
4 known members
Alternatively splicing gives rise to two isoform proteins, HMGA1a (HMGI)
and HMGA1b (HMGY). These two are identical in sequence except for a
deletion of 11 residues between the the first and second AT hook in the latter.
Alternative splicing also produces HMGA1c.
The related HMGA2 (HMGI-C) protein is coded for by a separate gene.
Conserved
Homologues of the mammalian HMGA proteins have been found in yeast,
insects, plants and birds, as well as in all mammalian species examined.
MBV4230
HMGA - AT-hook binding to DNA
Each HMGA protein
possesses 3 similar, but
independent, AT hooks
which have an invariant peptide
core motif of Arg-Gly-Arg-Pro
(”palindromic” consensus
PRGRP) flanked on either side by
other conserved positively
charged residues.
The HMGA proteins
bind, via the AT hooks,
to the minor groove
of stretches of AT-rich DNA but
recognize substrate structure,
rather than nucleotide sequence.
MBV4230
HMGA proteins heavily modified
The HMGA proteins are among the most highly phosphorylated
proteins in the mammalian nucleus.
Cell cycle-dependent phosphorylation pga cdc2 activity in the G2/M phase of the cycle.
Sites: T53 and T78 situated at the N-terminal ends of the 2. and 3. AT-hook.
Phosphorylation significantly reduces (>20-fold) DNA binding.
HMGA proteins are the downstream targets of a number of
signal transduction pathways that lead to phosphorylation.
HMGA proteins are also acetylated
at Lys65 by CBP and at Lys71 by PCAF
…as well as methylated and poly-ADP ribosylated
Hypothesis: Modifications may alter DNA-binding specificity?
MBV4230
Architectural effects
Architectural effects
Binding of full-length HMGA proteins can bend, straighten, unwind and induce
loop formation in linear DNA molecules in vitro.
Multiple contact points with DNA may alter
conformation of DNA
A single AT-hook preferentially binds to stretches of 4-6 bp of AT-rich sequence,
and partially neutralizes the negatively charged backbone phosphates on only
one face of the DNA helix.
The number and spacing of AT-rich binding sites in DNA influences the
conformation of bound DNA and the biological effects elicited.
HMGA may also induce conformational change in
proteins
HMGA forms protein-protein interactions with other transcription factors,
which alters the 3D structure of the factors resulting in enhanced DNA binding
and transcriptional activation.
MBV4230
Maniatis: HMGI(Y) contributes to
formation of enhanceosomes
virus-inducible enhancer in the
IFN-b gene (human interferon b)
cis-elements for NF-kB, IRF-1, ATF-2-c-Jun
Synthetic (multiple cis-elements)
enhancer ≠ natural
Too high basal transcription
Less induction
Responds to several stimuli, while natural
enhancer only responds to virus
Biological function depends of
HMGI(Y) as architectural
component
HMG I(Y)
First described by Lund and Laland
binds AT-rich DNA in minor groove (“AT-hook”)
MBV4230
Recent
verision
MBV4230
Other functions of HMGA proteins
HMGA and cancer
HMGI/Y proteins are also involved in a
diverse range of other cellular processes
including pathologic processes such as
neoplastic transformation and metastatic
progression.
Chromosomal translocations in
a long 3.intron
Intron 3 of the HMGA2 genes is extremely
long (>25 kb in human and >60 kb in mouse)
and separates the three exons that contain the
AT hook motifs from the remainds of the 3´untranslated tail region of the gene.
Translocation within the exceptionally long
third intron are commonly observed in benign
mesenchymal tumors.
3. subgruppe: HMGN
MBV4230
HMGN proteins
Three functional domains of the HMGN proteins:
a bipartite nuclear localization signal (NLS),
a nucleosomal binding domain (NBD)
and a chromatin-unfolding domain (CHUD). The CHUD domain has a net
negative charge.
Binding of HMGN proteins to nucleosomes decreases
the compactness of chromatin, and facilitates trx
MBV4230
HMGN: architectural elements
reducing compactness of chromatin
Model of the binding of
HMGN proteins to
chromatin
HMGNs interact with
both the DNA and the
histone component of
the nucleosome
The CHUD domain interacts with
the amino terminus of histone H3.
May also affect H1 binding
Incorporation of HMGN
proteins into chromatin
is believed to reduce
the compactness of the
chromatin fiber.