I NUCLEOSOME POSITIONING

1
I) NUCLEOSOME POSITIONING II) NUCLEOSOME
REMODELING BCH 6415 SPRING 2006 Dr.
YANG (Part II)
2
Human SWI/SNF interconverts a nucleosome between
its base state and a stable remodeled state G.
Schnitzler, S. Sif, RE Kingston Cell 9417 1998
added before (1) or after SWI/SNF(2)
32P-labelled reconstituted mononucleosomes
incubated w/ SWI/SNF, then subjected to
gel-shift analysis Conclusion formation of a
novel gel-shift band that is SWI/SNF, ATP, salt
dependent (added after addition of SWI/SNF)
Western blot of gel-shift gel using anti-BRG1
Ab Conclusion Novel band does not contain
SWI/SNF unlikely it is nucleosome bound to
SWI/SNF
3
Human SWI/SNF interconverts a nucleosome between
its base state and a stable remodeled state G.
Schnitzler, S. Sif, RE Kingston Cell 9417 1998
Silver-stained 2-D gels
DNase I enhanced - DNase I protected
Preparative Gel-shift
Silver-stained 2-D gels of nucleosome cores and
SWI/SNF-treated cores. The novel band (panel
D), as well as nucleosome cores (panels C D),
contain the 4 core histones and no other
detectable proteins (e.g., SWI/SNF proteins,
TFs)
1) Novel band isolated by preparative
gel-shift 2) Eluted novel band subjected to DNase
I cleavage Result complex in novel band (lane
3) has altered DNase I cleavage pattern compared
to standard nucleosome core (lanes 1 2) , and
resembles cleavage pattern of SWI/SNF-modified
nucleosome (ln 7). Conclusion novel band
does not resemble std. core, but is more similar
to SWI/SNF-remodeled nucleosome.
4
Novel band seidments at S-value of approx. a
dinucleosome
Novel band (isolated from prep. gel-shift)
sedimented in glycerol gradient novel band
sediments faster than mononucs.
Novel band subjected to gel filtration column
elutes faster than mononucs. and at dinucs.
Conclusion The SWI/SNF-remodeled structure in
the novel band may be some form of a nucleosome
dimer. Also, the DNA histone ratios are similar
to normal nucleosome cores (data not shown).
-
-
-


B bare DNA N novel band C mononucleos. cores
End-labeled gradient-isolated novel band and
cores digested w/ restriction enzymes or MNase.
PstI, HindIII, and MNase showed increased
digesttion/accessibility of novel band vs.
core. Conclusion DNA in the novel band (i.e.,
nucleosome dimer) has altered accessibilty to
both MNase and restriction enzymes compared to
standard mononucleosome cores.
G. Schnitzler, S. Sif, RE Kingston Cell 9417
1998
5
A) Incubated purified novel band w/ SWI/SNF
converted novel band to cores (lane 5) in
gel-shift. B) DNase I cleavage of isolated
novel band and cores with or w/out
SWI/SNFtreatment 1) novel band has altered
cleavage pattern compared to cores (lanes
1 vs 4) 2) Treating novel band w/ SWI/SNF
generates DNase I cleavage pattern
identical to cores treated w/ SWI/SNF
(lanes 3 vs 6)
Conclusion SWI/SNF (ATP) can convert
novel/remodeled nucleosomal structure back to
normal nucleosomal cores SWI/SNF creates
equilibrium between standard nucleosomes and
novel nucleosomal structure
1) End-labeled DNA fragment containing GAL4
binding site reconstituted and isolated as novel
band (or cores) after SWI/SNF treatment. 2)
Added GAL4 (1-94) and treated w/ DNAse 3)
Fractionated DNA on DNA seq. gel 4) Comparing
lanes 1 2 vs 8 9, 10mM GAL4 shows
significantly greater DNase protection on
novel band compared to cores.
Conclusion Gal4 binds with higher affinity to
the novel nucleosomal structure compared to
standard nucleosomes -gt SWI/SNF-mediated
remodeling facilitates TF binding 3.2-fold
Model of SWI/SNF action 2 normal nucleosome
cores (A) are bound by SWI/SNF and converted via
the energy of ATP hydrolysis to SWI/SNF-bound
novel species (C) before release (D). The
reaction also works in reverse, establishing an
equilibrium between species (A) and (D).
6
For similar study and result for RSC remodeling
complex, see Activated RSC-nucleosome complex
and persistently altered Form of the nucleosome.
Y Lorch, BR Cairns, M Zhang, RD Kornberg Cell
9429-34 1998
7
Histone octamer transfer by a chromatin
remodeling complex Y Lorch, M Zhang, RD
Kornberg Cell 96389-92 1999
Sedimentation analysis of RSC reaction product in
maltose gradient
mononucleosomes markers sediment in fraction 22
Conclusion RSC reac. product is a nucleosome
Nucleosome
uncut
DNase footprint of RSC reaction product
reac. product shows 10 bp DNase cleavage
ladder characteristic of a nucleo.
cut
Restriction enz. digestion of RSC reac. product
reac. product protected from digestion by DraI
Naked DNA
1) Incubated 32P-labeled 154 bp DNA frag. w/
nucleo. unlabeled core particles, and RSC with
or w/out ATP or RSC 2) Fractionated incubation
mix. by gel-shift

• The RSC-mediated transfer product is a nucleosome
  because
• the product migrates in gel shifts as a
  nucleosome (C)
• the product sediments in density gradients as a
  nucleosome (A)
• the product shows DNase 10 bp periodicity
  cleavage pattern
• of a nucleosome (B)
• 4) the product shows protection of
  restriction sites like a nucleo.
• 5) the product is similar in stability at
  higher ionic strength like a
• nucleosome (data not shown here)
• However, it is not the same as the altered
  nucleosome formed by RSC action.

Conclusion RSC can catalyze the transfer of a
histone octamer from nucleosomes to free DNA
8
Kinetics of histone octamer transfer a)
dependence of octamer transfer rate on core
particle b) dependence of transfer rate on
free DNA
RSC reaction cycle and pathway for histone
octamer transfer to free DNA
Alt nuc stably altered nucleosome nuc
normal nucleosome RSC-nuc activated
RSC-nucleosome complex trans nuc product of
histone octamer transfer
9
See also Persistent site-specific remodeling
of a nucleosome array by transient action of the
SWI/SNF complex. T. Owen-Hughes, RT Utley, J
Cote, CL Peterson, JL Workman Science 273513
1999. Evidence for octamer transfer
10
EVIDENCE FOR NUCLEOSOME SLIDING MEDIATED BY
CHROMATIN REMODELING COMPLEXES
ATP-dependent histone octamer sliding mediated by
the chromatin remodeling complex NURF A Hamiche,
R Sandaltzopoulos, DA Gdula, C Wu Cell
97833-42 1999
Nucleosome mobilization catalyzed by the yeast
SWI/SNF complex I Whitehouse, A Flaus, BR Cairns,
MF White, JL Workman, T Owens-Hughes Nature
400784 1999
11
Chromatin Remodeling In Vivo Evidence for a
Nucleosome Sliding Mechanism Fazzio et al.
Molec. Cell 121333-1340 expts. done in
yeast
1) Used Isw2- mutant yeast strain grown on
raffinose, where 3 nuclesomes in promoter region
appear to be shifted, leading to partial
deregulation of gene. 2) Constructed
gal-inducible allele of ISW2 w/ FLAG
tag integrated into normal ISWI locus. 3)
Analyzed POT1 locus, a previously identified
target of Isw2. 4) Performed ChIP assays w/
anti-FLAG Ab for association of POT1 locus with
Iswi-FLAG (i.e, Iswi complex). 4) Showed Iswi2
complex increased assoc. with POT1 locus during
course of gal induction of Iswi.
1) Induced Iswi expression with galactose. 2)
At different times after gal induction, treated
wild-type and Iswi- mutant cells with
DNA-cleaving agents in vivo (i.e., treated
spheroplasts). 3) Each of the 3 cleavage
agents has preference for cleaving chromatin in
linker regions. 4) Purified DNA at each time
point, digested DNA with appropriate restriction
enzyme, and Southern blotted DNA. 5) Used
32P-labelled probe appropriate for
indirect end-labelling. 6) Autorads show a
gradual shift of nucleosome positions upstream
toward promoter until WT positions are reached.
7) Consistent with sliding of 3 nucleosomes
toward promoter.
12
Fazzio et al. (cont.)
1) Performed same experiment as before, but
analyzed MNase cleavage pattern at high
resolution. 2) Used primer extension assay to
detect MNase cleavage sites at single nucleotide
resolution in DNA seq. gel therefore, each band
in gel represents a single MNase cleavage site.
3) Gel shows shift in positions of linker
regions during course of gal induction and
Iswi2-catalyzed nucleosome repositioning. Also
demonstrated that Isw2 complex slides nucleosomes
without disrupting integrity of nucleosomes (data
not shown here).
Overall conclusions 1) ISWI complex catalyzes
nucleosome sliding in vivo 2) sliding is
unidirectional toward the promoter 3) sliding
is localized to a few nucleosomes.
13

• All three families can change position of
  nucleosomes on DNA
• (sliding nucleosomes)
• All can also form regularly spaced nucleosomes in
  a nucleosomal array
• ISWI-containing complexes appear to act primarily
  by nucleosome sliding
• SWI/SNF can change conformation of nucleosomes to
  expose nucleosomal
• DNA on surface of histone octamer
  without sliding nucleosome.
• dMi-2 moves histone octamer toward center of DNA
  fragment in vitro
• ISWI moves octamer toward ends of DNA fragment

SNF2, not ISWI
SNF2 subfamily
SNF2, not ISWI
Nucleosome Sliding
ISWI family
Disassembly of nucleosomes?
Summary of SNF2 ISWI family remodeling complexes
SNF2 subfamily
Various activities of ATP-dependent chromatin
remodeling complexes (Not all complexes carry
out all activities)
14
Mechanisms of chromatin remodeling
Fr. Flaus Owen-Hughes 2004
Fr. Lusser Kadonaga BioEssays 2512 2003
15
Hypothetical structure of altered dinucleosome
product after SWI/SNF remodeling
octamer displacement
Hypothetical Nucleosome Remodeling Pathways (Fr.
Becker Horz Ann Rev Biochem 2002) essentially
based on bulge diffusion model
Nucleosome sliding in cis
OR, Perhaps SWI/SNF stabilizes or freezes
nucleosomes in conformations with obstructed
sites exposed
16
ATP-Driven Exchange of Histone H2AZ Variant
Catalyzed by SWR1 Chromatin Remodeling Complex
Mizuguchi et al. Science 303343-348 2004.
H2A.Z (Htz1 in yeast) 1) a variant of histone
H2A associates w/ H2B 2) enriched in euchromatic
regions 3) acts as boundary element to prevent
spread of heterchromatin into euchromatin4)
positive regulator of transcription 5) inhibits
chromatin condensation promotes global
decondensation of chromatin.
Immunopurified complexes containing Htz1 (using
Flag-tagged Htz1) and found SWR1.
1) SWR1- a Swi/Snf-related ATPase of unknown
function. 2) Expressed Flag-tagged SWR1 in
yeast. 3) Immunopurified SWR1 complex from
yeast whole cell extracts w/ anti-Flag Ab. 4)
Fractionated on SDS gels and found over 12
subunits. 5) Identified gel bands by peptide
microsequencing and mass spec. 6) Found histones
among SWR1-associated proteins, including Htz1.
Conclusion Htz1 contained in SWR1 complex.
17
Mizaguchi et al. (cont.)
1) Performed ChIP assay with anti-Htz1-Flag Ab
on wild-type swr1- mutant cells. 2) Used PCR
primers specific to promoter region of RDS1 gene.
3) Found loss of Htz1 in RDS1 promoter in swr1-
cells. Conclusion Htz1 localization to
promoter regions is dependent upon SWR1.
18
Mizaguchi et al. (cont.)
1) Assembled nucleosomal arrays in vitro onto DNA
immobilized on magnetic beads using a
recombinant yeast nucleosome assembly system. 2)
Incubated assembled nucleosome arrays with
purified SWR1 complex, Htz1-Flag H2B dimers,
and ATP. 3) Washed off unbound molecules. 4)
Immobilized nucleosomes stripped of histones
and subjected to Western blot analysis with
anti-Flag Ab on SDS gel. 5) Found that Htz1 is
incorporated into nucleosomes. 6) Also
demonstrated that ATP is required, Htz1-H2B is
preferentially transferred over H2A-H2B, Htz1
transfer is specific to SWR1 complex vs SWI/SNF,
ISWI, RSC, INO80 (data not shown here).
(Htz1 in nucleosomes)
(Htz1 not assoc. w/ nucleosomes and washed off)
Before SWR1 After SWR1
1) Assembled immobilized nucleosomal arrays as
above. 2) Incubated nucleosomes with SWR1
complex, Htz1-Flag H2B dimers, and ATP. 3)
Washed eluted histones from immobilized DNA. 4)
Fractionated eluted histones SDS gel and stained
gel with silver. 5) Found loss of H2A after
incubation with SWR1. Conclusions The SWR1
complex catalyzes the exchange of H2A for Htz1,
likely as Htz1-H2B for H2A-H2B dimers, in intact
nucleosomes.
19
How do remodeling complexes know where and when
to go?
Three potential mechanisms for directing SWI/SNF
to correct genes 1) Non-targeting model
SWI/SNF may introduce transient changes in
chromation structure at random sites throughout
genome, and stabilization of remodeled chromatin
structure occurs only in the presence of
DNA-binding transcription factor(s). 2) Pol II
association SWI/SNF may be targeted to specific
genes by association with RNA pol II holoenzyme
(which is recruited by transcriptional
activators). But studies have shown only a
fraction of all pol II holoenzyme contains
SWI/SNF. 3) Targeting by transcription factors
SWI/SNF may be targeted to the correct genes by
direct interaction and recruitment by
gene-specific activators (and repressors). 4)
Other mechanisms
20
Nucleosome remodeling complexes
All contain an ATPase subunit exhibit
ATP-dependent chromatin perturbation activity
ATPase subunits alone can remodel chromatin
structure in vitro Other subunits may have
roles in modulating activity of ATPase subunit or
targeting of complexes to correct genes and
regulatory regions SWI/SNF best
characterized regulates 5-6 of yeast genes
can act as a or regulator of
transcription RSC Sth1 ATPase subunit similar
to Swi2/Snf2, but more global effect on
transcription than SWI/SNF SWI2 ATPase
stimulated by both nucleosomes and naked DNA,
but ISWI ATPase preferentially stimulated by
nucleosomes over naked DNA Mi-2 ATPase only
induced by nucleosomal DNA Mi-2 Swi2 ATPase do
not require N-term. tails of histones ISWI
requires N-term tail of H4 for full ATPase
stimulation
21
Other Mechanisms for Targeting Remodeling
Complexes
Chromatin Remodeling Complex
a) Targeting by DNA-binding factors b)
Targeting by methylated DNA (for silencing) c)
Targeting by interaction w/ nuclear matrix
and actin-like proteins d) Targeting and
stabilization by modification pattern of
histones
Fr. Becker Horz Ann Rev Biochem 2002
22
Recruitment of the SWI/SNF chromatin remodeling
complex by transcriptional activators N
Yudkovsky, C Logie, S Hahn, CL Peterson Genes
Devel. 132369-74 1999
?Srb2 prevents pol II recruitment to
promoter rSrb recombinant Srb2 TBP1143N TBP
mutation rTBP recombinant TBP
SWI/SNF components
Immobilized templates used in assay
pol II components
1) Used template of His4 promoter w/ TATA box (
promoterless control) attached to magnetic
beads 2) Formed PICs w/ various yeast nuclear
extracts GAl4 washed away uncomplexed
proteins (using templates immobilized by magnetic
beads) 3) Detached promoter/PIC from beads by
PstI digestion 4) Western blotted PIC w/ Abs to
Swi3, Snf5, Rpb3, Gal11, TBP, Gal4

Conclusion SWI/SNF is recruited to promoters
independent of pol II holoenzyme and TBP
(i.e., SWI/SNF not brought to
promoters as part of pol II holo. or complexed
w/TPB)
Figure 1. SWI/SNF is recruited to promoters in
the immobilized template assay independently of
holoenzyme and TBP. (A)
Immobilized templates used in this study.
(Top) The wild-type template contains the HIS4
core promoter and transcription start sites
(bottom) the core
promoter, including the TATA box, was deleted and
replaced by downstream nonpromoter sequences to
create the
Promoter template. (B) Recruitment of Swi3p and
Snf5p in wild-type and mutant NEs. PIC assembly
was performed with the nuclear
extracts indicated at top and
analyzed by Western blotting using antibodies
against the components indicated at right. All
reactions
included the activator Gal4-AH. rSrb2 (200 ng)
and rTBP (400 ng) were added where indicated.
23
G Gal4 DNA-binding domain only A Gal4-AH
(weak activation domain) V Gal4-VP16 (strong
activation domain) W His4 promoter template ?
promoterless template
1) Assembled PIC as before Gal4, Gal4-AH, or
Gal4-VP16 using wt or TBP-defective
(TBP1143N) nuclear extracts 2) Western blotted
isolated PIC using Abs to components of
SWI/SNF, pol II, general TFs. Results 1)
Gal4-AH and Gal4-VP16 increased recruitment of
SWI/SNF gt13- and gt21-fold,
respectively 2) Activator-dependent
SWI/SNF recruitment did not
require promoter sequences (only Gal4 binding
site) 3) Recruitment of RSC complex
(i.e., STH1) did not require
activators (or TBP)
STH1 component of RSC remodeling
complex
Conclusion SWI/SNF is recruited to DNA by
activators independent of promoter sequences
but, RSC complex
recruitment to PIC apparently does not require
activators
24
Templates of reconstituted nucleosomal
arrays (arrays contain nucleosome positioning
seqs.)_
1) 32P-labeled arrays mixed w/ 15X unlabeled
arrays (in four combinations of labeled and
unlabeled arrays) that do or do not compete
away GAL4-VP16 from labeled template 2)
Mixtures treated with HincII or HincII SWI/SNF
w/ or w/out GAL4-VP16 3) Cleaved arrays w/
Hinc II to see whether SWI/SNF remodeled
nucleosomes and exposed Hinc II site 4) Cleavage
products run on gel 5) 32P-labeled bands from
Hinc I cleavage quantitated Results 1) SWI/SNF
can remodel nucleosomal arrays
in absence of GAL4-VP-16 binding
2) GAL4-VP16 binding to array enhances
SWI/SNF remodeling of
nucleosomal array

cold unlabeled array does not compete away
GAL4-VP16 maximal effect of GAL4-VP16 on
labeled array
Conclusion SWI/SNF remodeling activity can be
recruited/targeted by a transcriptional activator

25
Phosphorylation of linker histones regulates
ATP-dependent chromatin remodeling enzymes PJ
Horn, LM Carruthers, C Logie, DA Hill, MJ
Solomon, PA Wade, AN Imbalzano, JC Hansen, CL
Peterson Nature Struc. Biol. 9263-267 2002

• Remodeling activities of SWI/SNF, Mi-2, ACF
  complexes are inhibited by
• incorporation of linker histones (histone H1)
  into nucleosomal arrays prior
• to addition of remodeling complexes.
• But, phosphorylation of the linker histone can
  rescue remodeling by SWI/SNF.
• Therefore, linker histones may contribute to
  regulation of remodeling complexes,
• and there may be a functional relationship
  between phosphorylation of linker
• histones and chromatin remodeling complexes.

26
Transcription through nucleosomes by Pol II
Hypothetical mechanism of transcription through
chromatin by polymerase II (Pol II) in vivo. As
Pol II (pink) travels along a eukaryotic gene, it
converts DNA-bound octamers (blue) in its path
to hexamers (green a smaller green shape is the
displaced H2AH2B dimer). Facilitation of this
process could be achieved with the aid of
elongation factors such as FACT (facilitates
chromatin transcription), histone modifications
and/or histone variants. Transient loss of the
H2AH2B dimer from the nucleosome could create a
window of opportunity for chromatin remodeling
factors or DNA-binding proteins. Unless the
disrupted state is stabilized, complete
nucleosomes are eventually restored.
Transcription might also result in nucleosome
redistribution and partial nucleosome depletion.
mechanism different for pol III