Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The role of histone H4 lysine 20 monomethylation in gene expression and differentiation
(USC Thesis Other)
The role of histone H4 lysine 20 monomethylation in gene expression and differentiation
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
THE ROLE OF HISTONE H4 LYSINE 20 MONOMETHYLATION IN REGULATING
GENE EXPRESSION AND DIFFERENTIATION
by
Jennifer Kae Sims
__________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOCHEMISTRY AND MOLECULAR BIOLOGY)
August 2008
Copyright 2008 Jennifer Kae Sims
ii
Dedication
I did it Grandpa.
iii
Acknowledgements
To my husband, Cory Sims, for everything he has done for me. For believing in
me and holding my hand every step of the way. Without you, I couldn’t have done this.
To my family, for their love and support, it’s been a long and hard road in
California and your encouragement has meant everything. Thank you from the bottom
of my heart.
To all of the members of the Rice lab, I can never thank you guys enough for all
of the suggestions and support that you have given me. I feel incredibly lucky to have lab
mates that are also like family. I appreciate everything you have done to help me both
professionally and personally. Thank you for putting up with me!
To my mentor, Dr. Judd Rice, thank you for believing in me and always being on
my side. You have given me confidence in my science and myself and I appreciate that
more than I could ever say.
iv
Table of Contents
Dedication
Acknowledgements
List of Tables
List of Figures
Abstract
Introduction
Transcriptional Activation
Transcriptional Silencing
Dual Roles for Methylated Lysine Residues?
Histone Methyl Lysine Binding Proteins
Combinatorial Modifications
Histone Methylation and Development
Chapter 1: Defining a novel repressive trans-tail histone code
mediated by PR-Set7
H4K20 methylation is found in transcriptionally inactive
regions of the genome
Methylated H4K20 and H3K9 are enriched within the same
silent nuclear regions
Monomethylation of H3K9 requires the H4K20
monomethyltransferase, PR-Set7
PR-Set7 interacts with the H3K9 methyltransferase,
Riz1/PRDM2
PR-Set7 alone is not sufficient to repress transcription
L3MBTL1 is required for H4K20 monomethyl mediated
transcriptional repression
Discussion and future directions
Chapter 2: Regulation of differentiation by H4K20
monomethylation
Loss of PR-Set7 results in an increase in gene expression
ii
iii
vii
viii
xi
1
3
9
18
19
22
25
28
29
34
42
46
57
61
65
70
70
v
PR-Set7 monomethyltransferase activity is required to repress
RUNX1
Monomethylated H4K20 and H3K9 are selectively targeted to
the RUNX1 promoter
Decreased monomethylated H4K20 results in reduced
L3MBTL1 at RUNX1
Monomethylated H4K20 is required for L3MBTL1 recruitment
to repress RUNX1
Decreased monomethylated H4K20 is specifically associated
with megakaryocytic differentiation
Loss of monomethylated H4K20 and L3MBTL1 at the RUNX1
promoter is an early event of megakaryopoiesis
Catalytically active PR-Set7 is required to prevent
megakaryopoiesis
Discussion and Future Directions
Chapter 3: The Role of Histone Methylation in Stem Cell
Differentiation
Global changes in histone methylation accompany mouse
embryonic stem cell differentiation progression
Isolation and standardization of ChIP assays in pluripotent hES
cells
Changes in H3K36 trimethylation mirror changes in gene
expression in hES cells undergoing commitment
ChIP sequencing provides a viable option for mapping histone
modifications within stem cells
Global changes in specific histone modifications in pluripotent
versus committed HES2 cells
Discussion and Future Directions
Chapter 4: Methods
Immunofluorescence Studies
Chromatin Immunoprecipitation Studies
Western Blot Studies
FACS Studies
Differentiation Studies
Luciferase Studies
Transfections
Quantitative Real Time PCR Studies
Immunoprecipitation Studies
71
76
79
81
84
88
94
96
110
110
114
118
127
130
134
139
139
146
163
169
172
173
174
193
vi
Bibliography
197
vii
List of Tables
Table 1: List of genes upregulated ≥10 fold in the absence of PR-Set7
Table 2: Gene expression patterns in sorted HES2 hES cells
Table 3: Antibodies used for immunofluorescence experiments
Table 4: Cell Line Sonication Conditions
Table 5: Antibodies used for chromatin immunoprecipitiations
Table 6: Antibodies used for Western blot analysis
Table 7: Antibodies used for FACs analysis
Table 8: Sequences of siRNA duplexes used for knockdown
experiments in HeLa cells
Table 9: Concentration of drugs used for selection of tissue culture cells
Table 10: Primers used in SYBR green Real Time PCR experiments
Table 11: Primers used for TaqMan based Real Time PCR
Table 12: Antibodies used for immunoprecipitations
72
123
145
148
149
167
171
178
181
188
191
196
viii
List of Figures
Figure 1: Known mammalian histone modifying enzymes and their
specific covalent modification are indicated in relation to the primary
amino acid sequence of the human core histones (H3, H4, H2A and
H2B).
Figure 2. Specificity of H4K20 methyl antibodies for Western analysis
Figure 3: Characterization of H4K20 methyl specific antibodies for
immunofluorescence
Figure 4: H4K20 methyl modifications are differentially enriched
within distinct silent nuclear compartments
Figure 5: Specific methylated states of H4K20 and H3K9 co-localize to
the same silent nuclear compartments
Figure 6: Histone H4K20 and Histone H3K9 patterns of methylation
are found in similar genomic regions along extended chromatin fibers
Figure 7: Preferential and selective enrichment of monomethylated
H4K20 and H3K9 on the same nucleosomal core particle in vivo.
Figure 8: H4K20 monomethylation is not dependent on H3K9
monomethylation
Figure 9: H3K9 monomethylation requires PR-Set7
Figure 10: PR-Set7 does not interact with known H3K9
monomethyltransferases
Figure 11: PR-Set7 preferentially interacts with Riz1/PRDM2
Figure 12: Depletion of Riz1/PRDM2 does not affect global levels of
H3K9 monomethylation
Figure 13: H3K9 monomethylation is reduced within specific genomic
locations in the absence of Riz1/PRDM2
4
30
32
35
37
40
43
45
47
49
51
54
56
ix
Figure 14: PR-Set7 is not sufficient for transcriptional repression or
activation
Figure 15: L3MBTL1 interacts with monomethylated H4K20 to repress
transcription
Figure 16: Monomethylated H4K20 and L3MBTL1 at the RUNX1
promoter is associated with RUNX1 repression
Figure 17: Slope calculations for semi-quantitative ChIP analysis
Figure 18: Monomethylated H4K20 is required for L3MBTL1
recruitment and RUNX1 repression
Figure 19: Megakaryocytic induction leads to an increase in RUNX1
expression in K562 cells
Figure 20: Decreased monomethylated H4K20 is selectively associated
with megakaryopoiesis
Figure 21: Reduction of monomethylated H4K20 and L3MBTL1 at the
RUNX1 promoter is an early event in megakaryopoiesis
Figure 22: Overexpression of RUNX1 sensitizes cells towards
megakaryopoiesis
Figure 23: Depletion of monomethylated H4K20 induces spontaneous
megakaryopoiesis
Figure 24: PR-Set7 does not interact with L3MBTL1
Figure 25: RUNX1 expression is not dependent on Riz1/PRDM2
Figure 26: Changes in global patterns of histone methylation during
mES differentiation
Figure 27: FACs analysis of HES2 hES cell line using GCTM2 and
TG30 as markers of pluripotency
Figure 28: Standardization of lowest number of cells used for ChIP
59
62
73
77
82
85
89
92
95
97
101
105
112
116
119
x
Figure 29: Standardization of H3K36 trimethyl ChIPs in HES2 hES
cells
Figure 30: Enrichment of H3K36 trimethylation is found at active
genes in GCTM2
high
/TG30
high
cells
Figure 31: H4K20 monomethyl targets can be identified by ChIP seq
Figure 32: Characterization of HES2V cell line
Figure 33: Global patterns of histone methylation change upon
commitment
121
125
128
132
135
xi
Abstract
In developing multi-cellular organisms, cell fate decisions are largely determined by
epigenetic programs that activate and repress specific sets of genes. For eukaryotic gene
regulation, the post-translational modifications of the DNA-associated histone proteins
are critical in transcriptional control. Here, we describe a novel trans-tail histone code
where the monomethylation of histones H4 lysine 20 (H4K20) and H3 lysine 9 (H3K9)
defines specific genomic regions of transcriptionally repressed chromatin. Global
analysis of this code revealed a defined unidirectional temporal sequence of methyl-
modifications in which monomethylated H3K9 depends on the H4K20
monomethyltransferase, PR-Set7. Importantly, we have identified a repressor protein,
L3MBTL1, which is recruited to and binds monomethylated H4K20 and that this
interaction is required to initiate transcriptional repression in vivo. Using expression
microarray analysis, we have identified numerous cell type-specific genes involved in
certain growth and differentiation pathways whose expression is directly regulated by the
H4K20 monomethylation silencing pathway. By focusing on one of these genes,
RUNX1/AML1, we found that monomethylation of H4K20 by PR-Set7 was required for
chromatin condensation and transcriptional repression but that H3K9 monomethylation
was dispensable. We also confirmed that the H4K20 monomethyl-binding protein,
L3MBTL1, was required but not sufficient for RUNX1 repression. Importantly, we
demonstrated that the lack of monomethylated H4K20 at the RUNX1 promoter results in
xii
increased RUNX1 protein expression and the spontaneous differentiation of BFU-E/MK
precursor cells specifically to the megakaryocytic lineage. We have further extended
these studies to examine the role of histone methylation patterns during embryonic stem
cell commitment. Using both Western analysis as well as high throughput microarray
analysis, we demonstrate clear changes in histone methylation patterns as stem cells
move from pluripotency to commitment phase. Collectively, these studies demonstrate
that histone methylation plays a critical role in human development.
1
Introduction
The eukaryotic genome is packaged and functionally organized into chromatin; a
structure composed of DNA and chromosomal-associated proteins. The most
fundamental repeating unit of chromatin is the nucleosome, which consists of 146 base
pairs of DNA wrapped around an octamer of the core histone proteins H2A, H2B, H3 and
H4. The crystal structure of the nucleosome core particle indicates that the N-terminal
tails of the histones extend from the nucleosome to interact with the nuclear environment
(Luger et al. 1997). Decades worth of research has demonstrated that specific amino
acids on the histone tails are targets of various post-translational modifications including
acetylation, phosphorylation, ubiquitination, poly(ADP-ribosylation) and methylation,
and that specific modified histone residues are associated with various biological
processes (Peterson and Laniel 2004). These observations led to the “histone code”
hypothesis where the modified histone tails, either alone or in combination, play a direct
role in regulating processes such as transcription, DNA damage and repair,
recombination and replication (Strahl and Allis 2000). Increasing evidence indicates that
specific modified histone residues recruit and bind certain types regulatory proteins that
are involved in initiating the particular DNA-templated process (Taverna et al. 2007)
Methylation of lysine residues within histone proteins was discovered over 40
years ago and is conserved in higher eukaryotes (Shilatifard 2006). This post-
translational modification was originally described as a stable modification that was able
to block the progress of transcription (DeLange and Smith 1971). As the individual
2
lysine residues that are modified were discovered, it was determined that lysine
methylation does not simply function just to block transcription but also plays a role in
other DNA template mediated processes such as recognition of DNA damage, cell cycle
control, X inactivation and also marks regions of chromatin that are undergoing active
transcription (Zhang and Reinberg 2001). Mass spectrometry studies have shown that ε-
amino groups of lysine residues can accept one, two or three methyl groups which
changes the hydrophobicity of the tail but does not affect the overall charge, leading to
further patterns of lysine methylation that can each play a different role in chromatin
templated processes (Rice and Allis 2001). The enzymes responsible for modifying
lysine residues are known as histone methyltransferases (HMTs) and are highly specific
in the residue that they will methylate as well as in the degree of methylation. HMTs that
target lysine residues have a common protein domain, the SET domain, which provides
catalytic function. The SET domain was originally found as a conserved sequence in the
Drosophilia malagaster proteins Suppressor of variegation 3-9 (a position effect
variegation modifier), Enhancer of zeste (a polycomb chromatin regulator) and Trithorax
(a trithorax group chromatin regulator) (Jenuwein 2001). The 130 amino acid SET
domain is found in all eukaryotic HMTs except for DOT1 (Feng et al. 2002; Dillon et al.
2005). SET enzymes function by transferring a methyl group from S-adenosyl-L-
methionine to the ε-amino group of a lysine residue, resulting in methyl-lysine and the
cofactor byproduct, S-adenosyl-L-homocysteine (Dillon et al. 2005). Within the two
conserved sequence motifs of the SET domain, there are two phenylalanine residues that
3
have been recently shown to act as a “switch” that controls the number of methyl groups
that are added to a lysine residue (Collins et al. 2004). Therefore, each HMT, depending
on the sequence within the catalytic pocket, is highly specific for the number of methyl
groups that it is able to add onto a lysine residue (Figure 1). The number and position of
the methyl groups added along the histone N terminal tail or within the core determines
the biological role that it plays, especially in transcriptional regulation.
Transcriptional Activation
Transcriptional activation is most often associated with acetylation of histone
tails. However, there is a great deal of evidence for methylation of histone tails also
providing a marker of activation. All three of these modifications reside on the histone
H3 tail: Lysine 4, 36 and 79.
Histone H3 Lysine 4 (H3K4)
The SET domain containing protein, MLL, was first identified as a protein
involved in translocations leading to hematological malignancies including acute myloid
leukemia (AML) and acute lymphoblastic leukemia (ALL) (Shilatifard 2006). More
recent experiments have shown that it and its yeast homolog, Set1, have HMT activity
specifically towards H3K4. H3K4 methylation is associated specifically with
transcriptional activation as the majority of H3K4 methylation is found at or within 1kb
regions of transcription start sites and within the body of actively transcribed genes
(Santos-Rosa et al. 2002; Scheider et al. 2004, Bernstein et al. 2005). Chromatin
4
Figure 1. Known mammalian histone modifying enzymes and their specific covalent
modification are indicated in relation to the primary amino acid sequence of the
human core histones (H3, H4, H2A and H2B).
These enzymes can methylate (Me) arginine or lysine residues, phosphorylate (P) serine
or threonine residues, acetylate (Ac) lysine residues or ubiquitylate (Ub) lysine residues.
Each enzyme and its respective histone modification are associated with specific DNA-
templated processes including transcription, replication, recombination and repair. Sims
et al. 2008.
5
immunoprecipitation assays (ChIPs) have shown that H3K4 dimethylation is enriched
within areas of the promoters of both inactive and active euchromatic genes whereas
trimethylation is specifically enriched within active promoters. This leads to a model of
H3K4 dimethylation acting as marker of inactive yet “poised” chromatin and H3K4
trimethylation enriching actively transcribed regions (Scheider et al. 2004). Consistent
with this model, MLL and Set1 associate with a larger multi-protein complex which is
required for its HMT activity, and contains the phosphorylated, or active, form of RNA
polymerase II (RNA pol II) (Santos-Rosa et al. 2002; Shilatifard 2006).
ChIP-microarray experiments (ChIP-chip) were performed to examine the
distribution of H3K4 di- and trimethylation across human chromosomes 21 and 22 as
well as various well conserved gene clusters (Bernstein et al. 2005). The data shows that
in agreement with previous studies, H3K4 trimethylation associates with RNA pol II as
well as transcription start sites (Bernstein et al. 2005). Interestingly, most H3K4
methylation was found in punctate regions near 5’ ends of genes except in the Homeotic
box (Hox) gene cluster where it spanned the entire cluster (Bernstein et al. 2005). This
occurred in a tissue and lineage specific manner where it is hypothesized to play a role in
keeping the necessary Hox cluster active for that specific tissue or lineage of cells
(Bernstein et al. 2005). These experiments also showed that methylated H3K4 associates
with regions enriched for H3K9/14 acetylation, both marks of transcriptional activation.
Together with previous data demonstrating that the Set1/MLL complex HMT activity is
6
stimulated by H3 acetylated peptides, it suggests that these modifications are found in
concert to contribute to transcriptional activation (Shilatifard 2006).
Recent ChIP-sequencing experiments (ChIP-seq) confirmed the presence of
H3K4 di- and trimethylation surrounding the transcription start site of active genes
(Barski et al. 2007). Monomethylated H3K4 was also enriched in genomic regions
surrounding the transcription start site further demonstrating that all three forms of
methylated H3K4 are found within actively transcribed genes. These experiments also
demonstrate that the degree of methylation correlates with relative position to the
transcriptional start site (i.e. trimethylated H3K4 is enriched closest to the transcriptional
start site).
Histone H3 Lysine 36 (H3K36)
H3K36 methylation has best been characterized in yeast as a mark of gene
activation (Shilatifard 2006). The enzyme responsible for methylating H3K36, Set2, is
directly associated with the Serine 2 phosphorylated form of RNA polymerase II (Ser2P
RNA pol II) in the body of transcribed genes, suggesting that methylated H3K36 is a
mark of transcription elongation (Joshi et al. 2005; Sun et al. 2005; Shilatifard 2006).
Genome wide mapping of H3K36 methylation shows that it is enriched towards the 3’
end of transcribed genes, consistent with Set2 interaction with Ser2P RNA pol II
(Bannister et al. 2005; Rao et al. 2005, Barski et al. 2007). Further analysis of the
interaction between RNA pol II and Set2 revealed that these two proteins must interact in
order for H3K36 methylation to be placed on chromatin (Shilatifard 2006). Histone
7
acetyl transferase (HAT) activity is also associated with Ser2P RNA pol II, which is
thought to aid the elongation process by creating a more “open” chromatin environment
(Kristjuhan et al., 2002; Kristjuhan and Svejstrup, 2004). This however presents a
potential problem during transcription as this loosened chromatin can allow for spurious
transcription of cryptic start sites (Kaplan et al. 2003). Recent studies have demonstrated
that methylation of H3K36 plays an important role in preventing spurious transcription
by regulating acetylation levels within coding regions (Carrozza et al. 2005; Keogh et al.
2005; Li et al. 2007). Methylation of H3K36 by Set2 with the coding region recruits the
histone deacetylase complex, Rpd3S via a H3K36 methyl binding protein, Eaf3 (Kadosh
and Struhl 1997; Kadosh and Struhl 1998; Keogh et al 2005). The Rpd3 complex then
reduces H3 acetylation within the coding region, which is critical for preventing spurious
transcription (Carrozza et al. 2005; Keogh et al. 2005).
Comparative analysis has been performed between di- and trimethylated H3K36
and demonstrated that there was little to no difference in distribution between the two
forms of H3K36 methylation, suggesting that di- and trimethylation play a similar role in
transcription (Bannister et al. 2005).
Histone H3 Lysine 79 (H3K79)
While most post-translational modifications of histones occur on the N terminal
tails, the crystal structure of the nucleosome indicates that K79 resides within the core
domain of histone H3 on the outer face. This position makes it accessible to the nuclear
environment, allowing for interaction with other proteins and the opportunity to be
8
methylated (Luger et al. 1997). The discovery of the HMT that methylates K79 revealed
the first HMT that does not contain a SET domain, Dot1 (Feng et al. 2002). Instead,
Dot1 contains a SAM binding domain, which is required for its methyltransferase
activity. Dot1 was first identified in a screen for proteins that, when overexpressed,
cause disruption of telomeric silencing leading to the hypothesis that it plays a role in
gene activation (Ng et al. 2002a). Methylated H3K79 has been shown to be critical in
preventing spreading of histone deacetylases into active euchromatic regions, potentially
preventing heterochromatin spreading (Ng et al. 2003). This mirrors the activity of
H3K4 methylation, further strengthening the hypothesis that H3K79 plays a role in gene
activation (Shilatifard 2006). Furthermore, trimethylated H3K79 is enriched within the
transcription start site of active promoters, although there is no preferential enrichment of
mono-or dimethylated H3K79 within the transcriptional start sites of either active or
inactive promoters (Barski et al. 2007).
Another possible function of H3K79 methylation was recently revealed in a study
that methylated H3K79 serves as a binding site for 53BP1, which recognizes DNA
damage, linking DNA damage to H3K79 methylation (Huyen et al. 2004). This work
however has come into question, as these results have not been able to be repeated (Kim
et al. 2006).
9
Transcriptional Silencing
Methylated lysine residues along histone tails also have been shown to function as
a marker for transcriptional repression. Three lysine residues within histone H3 and H4
are linked to gene repression: Histone H3 Lysine 9 and 27 and Histone H4 Lysine 20.
Histone H3 Lysine 9 (H3K9)
Screens performed in Drosophilia malagaster searching for mutations in the
silencing phenomenon, position effect varigation (PEV), lead to isolation of Suv3-9
which was later determined, along with its human homolog, Suv39h1, to have HMT
activity specifically for H3K9 (Reuter and Spierer 1992; Tschiersch et al. 1994; Rea et al.
2000). Since this discovery, H3K9 methylation has become one of the most thoroughly
characterized histone modifications.
Using methyl specific antibodies which are able to discriminate between one, two
or three methyl additions to K9, it was shown that H3K9 mono- and dimethylation are
found in silent regions of euchromatin whereas trimethylation is localized to pericentric
regions of heterochromatin (Peters et al. 2003). Knockout mouse studies showed that
Suv39h1 functions specifically as a trimethylase as Suv39h1-/- mice have a decrease in
global levels of H3K9 trimethylation while global levels of H3K9 mono-and
dimethylation remain unchanged (Rice et al. 2003). These mice also have reduced levels
of H3K9 trimethylation at pericentric regions, developmental defects and gross defects in
chromosomal segregation and genomic stability demonstrating the critical role of
10
trimethylated H3K9 in proper heterochromatin structure and chromosome segregation
(Peters et al. 2001). A more detailed analysis of the localization of H3K9 trimethylation
indicates that it is enriched in satellite regions and other repetitive elements that must be
kept silent in order to maintain genomic stability (Peters et al. 2003; Kondo and Issa
2003; Martens et al. 2005).
Many enzymes have been identified that contribute to global levels of mono-and
dimethylated H3K9 in the genome. G9a knockout mouse studies have shown that G9a
contributes the majority of H3K9 dimethylation as well as a significant amount of H3K9
monomethylation to the genome (Rice et al. 2003). These mice also show developmental
defects as well as inappropriate expression of the tumor-specific antigen MAGE genes
suggesting that H3K9 dimethylation also plays a role in gene silencing in the euchromatic
regions of the nucleus instead of pericentric regions (Tachibana et al. 2002). This
hypothesis was confirmed upon staining using methyl specific H3K9 antibodies showing
that both H3K9 mono- and dimethylation were excluded from RNA pol II enriched
regions (Rice et al. 2003). Further in vivo studies were performed showing that G9a
mediated H3K9 methylation is important for silencing developmentally regulated and
tissue specific genes (Nishio and Walsh 2004; Roopra et al. 2004). A G9a like protein,
GLP1/Eu-HMTase I, has high sequence and structural similarity to G9a as well as similar
HMT activity towards H3K9 (Tachibana et al. 2005). GLP1-/- mice were unable to
survive demonstrating its necessity for proper development (Tachibana et al. 2005).
Global levels of dimethylated H3K9 were reduced in these embryos showing that GLP1
11
also functions as an H3K9 dimethylase in vivo (Tachibana et al. 2005). Interestingly,
biochemical studies have shown that G9a and GLP form homo- and heterodimers, which
provides stability for the proteins as well as promotes maximum HMT activity towards
H3K9 (Tachibana et al. 2005). Similar to G9a, ESET/SETDB1 is described as another
HMT that methylates H3K9 within euchromatic regions (Wang et al. 2003).
ESET/SETDB1 is unique however, in its ability to both di- and trimethylate H3K9
depending on its association with an ATFa-associated factor, mAM (Wang et al. 2003).
ESET/SETDB1 alone acts as a H3K9 dimethylase that can repress transcription whereas
the complex of ESET/SETDB1 and mAM trimethylates H3K9 and acts as a potent
transcriptional inhibitor in a reconstituted chromatin transcription assay (Wang et al.
2003).
Recent studies have shown that other epigenetic pathways work in concert with
H3K9 methylation such as DNA methylation and the RNAi silencing pathway.
Suv39h1/h2, G9a and ESET/SETDB1 have all been shown to interact with DNA
methyltransferase enzymes linking DNA methylation to H3K9 methylation, possibly
creating a reinforced silent state that is transmitted during replication (Lehnertz et al.
2003; Estève et al. 2006; Li et al. 2006). Most recently, mice deficient for Dicer, a
member of the RNAi pathway, have been characterized and show a decrease in levels of
both H3K9 di-and trimethylation, suggesting that the RNAi pathway plays a role in
directing HMTs to methylate appropriate regions of the genome (Kanellopoulou et al.
12
2005). These results are similar to the RITS pathway described in S. pombe (Grewal and
Rice 2004).
Histone H3 Lysine 27 (H3K27)
The major function of H3K27 methylation is to regulate critical developmental
genes such as the Hox gene cluster during early embryonic development (Simon and
Tamkun 2002). The Polycomb group (PcG) acts antagonistically to the trithorax group
(trxG) to control Hox gene expression throughout development. Generally, PcG
functions as repressors of Hox gene transcription while trxG functions as transcriptional
activators (Cao and Zhang 2004). The HMT complex responsible for H3K27
methylation is the E(Z) PcG complex in mammalian cells, which is well conserved
throughout all multicellular organisms including plants (Cao and Zhang 2004).
Purification of the SET domain containing protein, EZH2 surprisingly revealed that the
enzyme on its own does not contain HMT activity towards any histone substrates
suggesting that the entire complex is important for HMT activity (Cao and Zhang 2004).
Many different complexes have been purified with 4 common components, the SET
domain containing protein EZH2, as well as ESC, NURF-55 and SUZ12 (Cao and Zhang
2004; Shilatifard 2006).
The specificity of the complex towards only H3K27 is questionable as multiple
groups have reported activity of the E(Z) PcG complex for H3K9 through in vitro
experiments (Czermin et al. 2002; Kuzmichev et al. 2002; Cao and Zhang 2004). The H3
tail has a repeating sequence surrounding both K9 and K27 (ARKS), which could result
13
in some amount of promiscuity for H3K9 and H3K27 HMTs. However, the multiprotien
complex could use a larger part of the surrounding sequence to confer specificity to the
HMTs. In vitro studies using G9a support this hypothesis, as it also methylates H3K27
although to a lesser extent (Tachibana et al. 2001). The content of the complex may also
determine if methylation will occur in a nucleosomal or core histone context, as both
types of specificity have been reported (Cao et al. 2002; Muller et al. 2002; Czermin et al.
2002; Kuzmichev et al. 2002).
Using methyl specific antibodies, it has been shown that H3K27 mono-, di- and
trimethylation partition to different regions of the nucleus. Monomethylated H3K27 is
enriched within pericentric regions of heterochromatin while di- and trimethylated
H3K27 are diffused through silent regions of euchromatin (Peters et al. 2003).
Furthermore, H3K27 monomethylation is enriched at the major and minor satellite
repetitive elements much like H3K9 and H4K20 trimethylation (Peters et al. 2003).
Trimethylated H3K27, however, is enriched at the promoter and within the body of the
Hox gene Ubx when it is in the “off” state and only found upstream of the promoter when
Ubx is in the “on” state consistent with the role of H3K27 methylation silencing Hox
gene expression (Papp and Müller 2006). Global analysis of the localization of
methylated H3K27 revealed enrichment of di-and trimethylation within the promoters of
inactive genes (Barski et al 2007). An additional members of the E(z) PcG complex that
aid in gene silencing are the DNA methyltransferase enzymes which have been shown to
directly interact with the complex. Knockdown of EZH2 or other components of the E(z)
14
PcG complex leads to a loss of DNA methylation (Viré et al. 2006). Recent evidence
further links trimethylated H3K27 and DNA methylation by demonstrating that genes
predisposed to cancer specific DNA methylation often are marked by H3K27
trimethylation in embryonic stem cells (ESCs) (Widschwendter et al. 2007). This
abarrent PRC complex and DNA methyation crosstalk offers a mechanistic approach of
the role that H3K27 trimethylation and DNA methylation plays in creating a cancer stem
cell by improperly methyating genes involved in differentiation. The silencing of these
genes through DNA methylation prevents proper differentiation, leading to a
predisposition to malignancy later in development.
Histone H4 Lysine 20 (H4K20)
Several studies have shown that H4K20 mono-, di- and trimethylation associate
with inactive regions of chromatin suggesting that this modification is important for
silencing and proper chromatin structure (Nishioka et al. 2002; Schotta et al. 2004; Sims
et al. 2006; Karachentsev et al. 2005). Immunofluorescence studies using methyl specific
antibodies demonstrate that mono- and dimethylated H4K20 is localized to RNA pol II
depleted regions of euchromatin and trimethylated H4K20 is enriched in pericentric
regions of heterochromatin (Figure 2; Schotta et al. 2004; Sims et al. 2006).
The enzymes responsible for H4K20 di- and trimethylation, Suv4-20h1/h2,
localize to pericentric regions of heterochromatin and when disrupted show loss of PEV.
This is consistent with the hypothesis that di- and trimethylated H4K20 function to
silence transcription (Schotta et al. 2004; Yang et al. 2008). Depletion of Suv4-20 h1 and
15
h2 in combination results in a global increase in monomethylated H4K20 but a decrease
in levels of both di-and trimethylated H4K20, confirming the specificity of the Suv4-20
enzymes (Yang, et al 2008). Unlike Suv39h1/h2, Suv4-20h1/h2 prefers nucleosomes to
core histones as a substrate for HMT activity (Schotta et al. 2004). Further experiments
reveal that in Suv39h-/- cells, H4K20 trimethylation is no longer detectable suggesting
that H4K20 trimethylation depends on H3K9 trimethylation (Schotta et al. 2004).
Consistent with its role as a repressive modification, dimethylated H4K20 is enriched at
specific Charlie repeat elements within the mouse genome (Martens et al. 2005).
Interestingly, mono-and dimethylated H4K20 is absent from other repetitive elements,
suggesting a targeted role of these modifications in gene repression programs.
Loss of histone H4K20 trimethylation has been described as a “hallmark of
human cancer” (Fraga et al. 2005; Tryndyak et al. 2006). Analysis was performed on 25
different cancer cell lines and a common factor in cancer progression, as modeled by cell
lines, was a decrease in the global levels of H4K20 trimethylation as well as protein
levels of Suv4-20 (Fraga et al. 2005; Tryndyak et al. 2006). The loss of H4K20
trimethylation also associated with loss of DNA methylation at repetitive elements, a
common feature of genomic instability associated with cancer progression.
The crystal structure of PR-Set7/Set8 demonstrates that it is the enzyme
responsible for monomethylating H4K20 (Xiao et al. 2005; Couture et al. 2005). PR-
Set7 functions to monomethylate K20 by interacting with the surrounding residues and
16
requires the presence of DNA in order to methylate, as PR-Set7 is a nucleosome specific
HMT (Nishioka et al. 2002; Fang et al. 2002, Yin et al. 2005).
Interestingly, throughout the cell cycle there is a fluctuation of bulk methylation
of histone H4 that is attributed to H4K20 as the other methylated residue on H4, arginine
3 remains constant throughout the cell cycle (Rice et al. 2002). PR-Set7 protein levels
are also cell cycle regulated, peaking during G2/M phase, which would allow the methyl
modification to be made on silent regions of sister chromosomes before segregation to
allow proper transmission of the epigenetic code to daughter cells (Rice et al. 2002).
Loss of PR-Set7 results in improper S phase progression as well as a G2/M arrest, due to
a collapse of replication forks as well as an increase in DNA damage; indicating that
monomethylated H4K20 and PR-Set7 are critical for cell cycle progression (Jorgensen et
al. 2007; Tardat et al. 2007). Studies performed in our laboratory have demonstrated that
loss of monomethylated H4K20 leads to decondensation of chromatin and confirms other
findings of a G2/M arrest and an increase in DNA damage, suggesting that the catalytic
function of PR-Set7 is required for genomic stability, proper chromatin structure and
possibly plays a role in condensation of chromatin prior to M phase (Houston et al.
2008).
As described above, the loss of monomethylated H4K20 leads to cell cycle arrest
as well as gross changes in genomic stability and chromatin structure. Therefore it was
reasonable to hypothesize that the disruption of PR-Set7 and the consequent depletion of
H4K20 monomethylation in animals would have a profound effect. As predicted, in both
17
Drosophilia and mouse, gene disruption of PR-Set7 leads to embryonic lethality
(Nishioka et al. 2002; Fang et al. 2002; Karachentsev et al. 2005; Huen et al 2008). The
attempt to make a knockout mouse revealed a severe lethality. The viable heterozygous
mice were crossed to create a homozygous PR-Set7 knockout mouse and through
genotyping experiments, they discovered that while heterozygous mice were viable (63%
of the population), homozygous PR-Set7 mice were not (0% of the population) (Huen et
al 2008).
Recent data has shown that H4K20 monomethylation is not localized to repetitive
elements within the genome and is preferentially enriched within the 5’ region of genes
(Martens et al. 2005; Barski et al 2007). The discovery of specific gene targets of
monomethylated H4K20 is critical for understanding the role of this modification in
development. Our laboratory has begun to define these target genes and has found that
monomethylated H4K20 is preferentially found in genes compared to intergenic regions
and repetitive elements and loss of H4K20 monomethylation through depletion of PR-
Set7 or its catalytic activity results in ~2 fold increase in gene expression at target loci
defining a role for H4K20 monomethylation in gene silencing (Houston et al,
unpublished data and chapter 2 studies here). Furthermore, we have identified some of
the target genes as critical regulators of major differentiation pathways (Table 1 and
chapter 2 studies), strengthening the findings of the important role monomethylated
H4K20 plays in development.
18
Dual Roles for Methylated Lysine Residues?
Recent evidence challenges the model that H3K9 and H4K20 methylation only function
as markers of transcriptionally inactive chromatin. These modifications play a role in a
variety of other chromatin templated processes depending on the context of the
modification. This hypothesis seems to extend to transcriptional regulation as well.
Using conventional ChIPs, ChIP-chip and ChIP-seq, several labs have demonstrated that
all methylated forms of H3K9 and monomethylated H4K20 are enriched within
promoters and throughout the body of actively transcribed genes, indicating that these
modifications that have historically been described as marks of repressed chromatin can
also be considered as marks of activation (Vakoc et al. 2005; Talasz et al. 2005; Vakoc et
al. 2006). These genome wide studies have also found an enrichment of monomethylated
H3K27 within active promoters as well as enrichment of trimethylated H3K79 within
silent genes, both of which are unexpected results (Barski et al 2007). H3K36
methylation also seems to have a dual role in transcriptional activity. Traditionally, this
modification has been described as a mark of transcriptional activation but recent data
has shown that the H3K36 HMTs, Set2 and SMYD2 associate with histone deacetylases
and participate in silencing gene expression (Strahl et al. 2002; Brown, et al. 2006).
These dual roles for lysine methylation within histone residues could be resolved
if a model is considered in which methylation is directed to specific genomic locations
based on the members of the multi-protein complex associated with the HMT. In this
model, methylation would partition to active versus repressed genes, potentially in a
19
tissue specific manner, dependent on the components of the HMT multi-protein complex.
This model is supported by examining the Suv39h1 multi-protein complex and the
localization of H3K9 trimethylation within the genome. Immunofluorescence studies and
ChIP assays have found H3K9 trimethylation enriched in repetitive elements surrounding
the pericentric regions (Rice et al. 2003; Martens et al. 2005). However, H3K9
trimethylation is also targeted to genes in euchromatin (Martin and Zhang 2005).
Suv39h1 binds to the retinoblastoma protein, Rb, which targets H3K9 trimethylation to
promoters of S phase specific genes (Ait-Si-Ali et al. 2004). Therefore, it is possible for
modifications to be targeted to various regions of the genome, depending on the proteins
associated with the specific HMT for that residue. A second possibility is that the
modification recruits different binding proteins to carry out different transcriptional
functions.
Histone Methyl Lysine Binding Proteins
Consistent with the histone code hypothesis, it has been shown that methyl marks
are necessary but not sufficient for a specific chromatin templated process but rather
serve as binding sites for other proteins that carry out the biological response (Strahl and
Allis 2000; Taverna et al. 2007). There are two general classes of histone methyl lysine
binding domains, the Royal family consisting of chromo-, Tudor and MBT domain
containing proteins and the plant homeodomain (PHD) finger family (Maurer-Stroh et al.
2003; Taverna et al. 2007). A shared feature of both families is formation of an aromatic
20
cage within the binding domain that is required for recognition of and proper interactions
with the methylated lysine residue (Taverna et al. 2007). Due to the complexity of
methylated lysine residues (mono-, di- or trimethyl states), the size of the cage as well as
the surrounding residues that interact with the binding protein are critical for recognition
of each methyl modification and its degree of methylation.
A second critical component for targeting methyl binding proteins to unique
regions of the genome is through association with different multiprotein complexes. The
most well characterized member of the Royal family is the chromodomain containing
protein, heterochromatin protein 1 (HP1), which binds to di- and trimethylated H3K9
(Lachner et al. 2001; Bannister et al. 2001; Jacobs et al. 2001). HP1 is targeted to various
regions of the genome enriched for di- and trimethylated H3K9 through interactions with
different H3K9 HMTs. Suv39h1 is found in complex with Rb, which also binds HP1,
targeting it to E2F genes (Nielson et al. 2001). In contrast, another H3K9 HMT,
SETDB1 is a binding partner of the transcriptional repressor protein, KAP1, which also
interacts with HP1, targeting it to genes regulated by the Zn finger protein, KRAB
(Schultz et al. 2002; Sripathy et al. 2006). Other members of the chromodomain family
play critical roles in development and transcriptional regulation for di- and trimethylated
H3K36 (MRG15), H3K4 (CHD1) and H3K27 (Polycomb) (Zhang et al. 2006; Flanagan
et al. 2005; Sims et al. 2005; Fischle et al. 2003).
In addition to the chromodomain, the Tudor domain containing protein, 53BP1,
and the MBT containing protein, L3MBTL1, both bind to lower states of methylated
21
H4K20 (mono- and dimethyl) but carry out vastly different biological functions (Sanders
et al. 2004; Botuyan et al. 2006; Li et al. 2007; Min et al. 2007; Yang et al. 2008). The
p53 binding protein, 53BP1 binds to mono- and dimethylated H4K20 and plays a role in
foci formation at double stranded DNA breaks, suggesting that H4K20 mono- and
dimethylation are critical for genomic integrity (Sanders et al. 2004; Botuyan et al. 2006;
Yang et al. 2008). Indeed, recent evidence has demonstrated that in the absence of
monomethylated H4K20, cells are unable to enter G2/M phase and display defects in
replication (Jorgensen et al. 2007; Tardat et al. 2007). In addition, loss of dimethylated
H4K20 also impairs 53BP1 mediated foci formation in the presence of DNA damaging
agents (Yang et al. 2008).
In contrast, recent structural findings demonstrate that the tandem MBT repeats of
the L3MBTL1 repressor protein preferentially binds monomethylated H4K40 suggesting
a role for this combination of histone modification and methyl binding protein in gene
regulation (Kim et al. 2006; Li et al. 2007; Min et al. 2007). Consistent with this, it was
found that the binding of L3MBTL1 to monomethylated H4K20 creates a
transcriptionally non-permissive chromatin structure in vitro and that L3MBTL1
negatively regulates the expression of a subset of E2F target genes (Trojer et al. 2007).
The majority of identified PHD finger domain proteins interact with methylated
H3K4 and are involved in regulation of transcription. The large subunit of the NURF
chromatin remodeling complex, BPTF binds to trimethylated H3K4 to promote a
permissive chromatin structure to maintain expression of key developmental target genes
22
such as the Hox genes (Wysocka et al. 2006). In addition, H3K4 trimethylation is also
bound by TAF3 subunit, a member of the basal transcription factor TFIID, highlighting
the role of H3K4 trimethylation in transcriptional activation (Gangloff et al. 2001;
Vermeulen et al. 2007). In contrast to the role of H3K4 trimethylation in gene activation,
the methyl mark can also be targeted by the human inhibitor of growth, ING2 leading to
repression of target genes such as cyclin D1 (Shi et al. 2006; Pena et al. 2006). These
contrasting roles of H3K4 methylation due to the methyl binding protein associated with
it reinforce the importance of dual roles of modifications within the genome, providing
the cell with a mechanism to alter transcriptional states by exchanging binding proteins
rather than shifting the histone tail modifications (Taverna et al. 2007).
Combinatorial Modifications
The extension of the “histone code” into a broader “nucleosome code”
hypothesizes that higher order chromatin structure depends on combinations of
modifications on different histone tails either on the same or neighboring nucleosomes
(Jenuwein and Allis 2001). Combinations of these modifications could be ordered,
wherein one mark can lead to the recruitment of other modifying enzymes to specific
genomic locations to alter the local chromatin environment and potentially global gene
expression. The link between lysine methylation and ubiquitination for transcriptional
activation is an example of modifications working in concert to promote transcription.
Both H3K4 and H3K79 methylation are dependent on H2BK120/K123
monoubiquitination, possibly through interactions with the RNA pol II elongation
23
complex (Briggs et al. 2001; Dover et al. 2002; Ng et al. 2002b; Wood et al. 2003). A
series of elegant experiments have demonstrated that the ubiqutin E3 ligase complex,
Rad6/Bre1 is required for monoubiquitination of histone H2BK123 and upon
ubiquitination, a signal is propagated to activate the enzymes COMPASS and Dot1 to
methylate H3K4 and H3K79 respecitvely (Briggs et al. 2001; Dover et al. 2002; Ng et al.
2002b; Wood et al. 2003). This pathway has also been show to be negatively regulated
by deubiquitination of H2BK123 (Shilatifard 2006).
The active marks of chromatin are methylation of histone H4 Arginine 3 (H4R3)
and acetylation of lysine residues on histone H3 and H4. It was recently shown that upon
knockdown of PRMT1, the H4 arginine methyltransferase, methylation of H4R3 was lost
as well as acetylation of lysine residues on histone H3 and H4 (Huang et al. 2005). These
data suggests there is an interaction between these modifications found on different
histone tails but potentially on the same nucleosome, predicting that PRMT1 must first
methylate arginine residues before acetylation of histone H3 lysine 9 and 14 and histone
H4 can take place. The authors hypothesize that PRMT1 could recruit histone
acetyltransferases and work upstream of histone H3/H4 acetylation (Huang et al 2005).
This type of ordered process also holds true for heterochromatin formation
surrounding pericentric regions and repetitive elements that are enriched in H3K9 and
H4K20 trimethylation. Colocalization experiments show that H3K9 and H4K20
trimethylation staining completely overlap (Schotta et al. 2004; Sims et al. 2006).
Biochemical studies have shown that heterochromatin protein 1 (HP1) binds to H3K9
24
trimethylated histone tails and Suv4-20h1/h2 directly bind to HP1 creating a model in
which there is a distinct timing to trimethylation events on histones H3 and H4 where
H3K9 is trimethlyated by Suv39h1/h2, followed by binding of HP1 which recruits Suv4-
20 to trimethylate H4K20 (Schotta et al. 2004). Consistent with this model, repetitive
elements enriched with H3K9 trimethylation are also enriched for H4K20 trimethylation
and loss of HP1 in Drosphila results in a global loss of H4K20 trimethylation (Martens et
al. 2005; Yang et al. 2008). Interestingly, recent data has shown that H4K20 and H3K9
monomethlyation are found in similar regions of the genome, suggesting that H3K9
monomethylation could serve as a template for where H4K20 monomethylation should
be placed within the genome after cell division, much like the interaction between H3K9
and H4K20 trimethylation (Sims et al. 2006; Barski et al. 2007).
Inactivation of one of the X chromosomes in female mammalian cells is partially
controlled by histone modifications including combinations of lysine methylation and
ubiquitination. Immunofluorescence studies have shown H3K9 dimethylation, H3K27
trimethylation and H4K20 monomethylation are all localized to the inactive X
chromosome (Peters et al. 2003; Schotta et al. 2004). In order for H3K27 trimethylation
to be placed on the inactive X, H2AK119 must first be ubiquitinated (Wang et al. 2004).
The function of enrichment of these modifications on the inactive X has not been
elucidated and requires further experiments.
25
Histone Methylation and Development
Recent reports demonstrate that multi-protein complexes that modify histones and
those that bind histone modifications play critical roles in developmental pathways, in
part, by regulating the expression of genes involved in lineage specification (Bernstein et
al. 2007; Mikkelsen et al. 2007; Rice et al. 2007). One well-defined example of this is
the Drosophila ESC-E(Z) complex that specifically trimethylates H3K27, which binds
the chromodomain of Polycomb (PC). (Cao et al. 2002; Kuzmichev et al. 2002; Muller et
al. 2002; Fischle et al. 2003; Min et al. 2003). This binding event and subsequent
repression of Hox gene cluster is a critical step in establishing positional identity within a
developing embryo (Papp and Muller 2006).
Interestingly, embryonic stem (ES) cells display an unusual chromatin structure
within the promoters of key developmental transcription factors and highly tissue specific
genes (Mikkelsen et al. 2007). This structure consists of both active (trimethylated
H3K4) and repressive (trimethylated H3K27) modifications, termed “bivalent” chromatin
marks (Bernstein et al. 2006; Mikkelsen et al. 2007). These bivalent modifications are
resolved as the ES cells differentiate into tissue specific progenitor cells, leaving
trimethylated H3K4 enriched within active genes and trimethylated H3K27 within
repressed genes. The bivalent state of ES chromatin is thought to be critical for plasticity
of differentiation. Consistent with this, progenitor cells that still have further
developmental potential retain the bivalent state until terminal differentiation has
26
occurred (Mikkelsen et al. 2007). Furthermore, recent studies have shown that RNA pol
II is present at low levels in a stalled or “poised” state at the promoters of bivalent genes
and transcribes these genes at low levels. This poised state is dependent on
ubiquitination of H2A by the H3K27 methyltransferase complex PRC (Stock et al 2007).
The presence of poised RNA pol II within the bivalent ES promoters can aid in the
plasticity of these genes as they can be turned on rapidly to promote differentiation.
These data are supported by other studies demonstrating that most promoters in ES cells
contain RNA pol II as well as are enriched in H3K4 methylation (Guenther et al 2007).
This study further reveals that the majority of genes initiate transcription but are unable to
complete elongation which aids in regulation of transcription (Guenther et al 2007).
In vivo experiments also highlight the importance of methylated lysine residues in
proper development. Loss of mono- and dimethylated H3K9 in G9a and GLP1 knockout
mice result in developmental defects and embryonic lethality, demonstrating that the
H3K9 HMTs as well as H3K9 mono- and dimethylation are required for proper
development (Tachibana et al. 2002; Tachibana et al. 2005). In addition, disruption of
PR-Set7 and H4K20 monomethylation also is lethal to both Drosphila and mice due to
genomic instability and improper development (Nishioka et al. 2002; Fang et al. 2002;
Karachentsev et al. 2005; Huen et al. 2008).
To further characterize the role of histone methylation in development, it is
important to map the position of these modifications across the genome. These
epigenomic maps will provide tools to define the location of the marks as well as to
27
discover the target genes of each modification. Once the target genes have been
identified, the role of histone methylation in regulation of their transcription can be
studied and placed into the context of their role in development.
Here, we provide evidence that monomethylated H4K20 and H3K9 cooperate in a
novel “trans-tail histone code” and that monomethylated H4K20 plays a critical role in
regulation of gene expression as well as prevention of spontaneous megakaryocytic
differentiation through repression of the hematopoietic master regulator, RUNX1/AML1.
We also extend these studies to examine the dynamics of histone methylation through
early stages of differentiation by comparing pluripotent human embryonic stem cells to
those undergoing early commitment.
28
Chapter 1. Defining a novel repressive trans-tail histone code mediated by PR-Set7
Structural studies have demonstrated that H4K20 is positioned within a basic
patch region of the H4 tail (amino acids 16-25, Figure 1) that interacts with an acidic
patch located on the surface of H2A/B dimer of the neighboring nucleosome (Luger et al.
1997; Luger, K. and Richmond, T.J. 1998). The nucleosome structure predicts that
modifications of this residue could have a direct effect on chromatin compaction due its
position. Further evidence to support this hypothesis has come from in vitro studies
demonstrating that another lysine residue within the basic patch, H4K16, can be
acetylated and this modification directly prevents chromatin compaction (Shogren-Knaak
et al. 2006). Furthermore, it has been shown in vitro that methylated H4K20 and
acetylated H4K16 antagonize each other, providing a “switch” in the chromatin state
(Nishioka et al. 2002). Consistent with this, many studies have described H4K20
methylation as a mark associated with silenced chromatin (Fang et al. 2002; Nishioka et
al. 2002; Karachentsev et al. 2005; Sims et al. 2006).
Regulation of transcriptional silencing by H4K20 methylation becomes more
complex due to a unique property of lysine methylation. Unlike other modifications such
as acetylation, the ε-amino group of the lysine residue can accept one, two or three
methyl groups, allowing it to be mono-, di- or trimethylated (Rice, J.C. and Allis, C.D.
2001). Therefore, to study the function of H4K20 methylation, we needed to develop
tools that would allow us to discriminate between different degrees of methylated
H4K20. To address this, our lab has developed a panel of highly specific antibodies that
29
were able to differentiate between mono-, di-, and trimethylated H4K20. The specificity
of the antibodies was tested by peptide competition assays. Each methyl-specific
antibody was incubated with 1 μg of unmodified or each methyl-specific peptide for 1
hour prior to Western analysis. If the antibody is specific for a methyl modification, then
the antibody will be sequestered by the peptide, resulting in an absence of signal by
Western analysis (Figure 2). Using this assay, we were able to confirm that each
antibody was indeed highly specific.
H4K20 methylation is found in transcriptionally inactive regions of the genome
This novel panel of antibodies was used to further examine the subcellular
localization of mono-, di- and trimethylated H4K20. Based on previous reports, we
hypothesized that each modification would partition to different regions of the nucleus,
and that these nuclear localizations would be found within transcriptionally inactive
chromatin (Schotta et al 2004).
Immunofluorescence studies were performed using female mouse embryonic
fibroblasts (MEFs). An advantage of using MEFs is their large nuclei as well as a well-
characterized pattern of heterochromatic foci that can be visualized by enrichment of
DAPI staining (Peters et al. 2001; Rice et al. 2003; Schotta et al. 2004). To ensure that
the methyl specific H4K20 antibodies can be used for immunofluorescence staining and
to confirm that they were indeed specific, peptide competition assays were performed.
Each antibody was incubated with 1 μg of either unmodified, mono-, di-or trimethylated
30
Figure 2. Specificity of H4K20 methyl antibodies for Western analysis
Peptide competition followed by Western analysis on HeLa whole cell lysates, using the
different H4 Lys-20 methyl-specific antibodies with synthetic peptides that were
unmodified or mono-, di-, or trimethylated at K20. Each antibody is highly specific as
only the appropriate corresponding peptide can eliminate the signal for each of the
antibodies. Sims et al. 2006
31
peptide prior to immunofluorescence staining. If the antibody is specific, then the
corresponding peptide will sequester it, preventing the antibody from binding to its
epitope within the cell. As shown in Figure 3, each antibody (shown in red) was
competed away specifically by its corresponding peptide, indicating that these antibodies
are indeed highly specific and can be used for further immunofluorescence studies. The
entire nucleus was counterstained with DAPI (green).
The nuclear localization of each methyl modification was examined at a higher
magnificiation as show in Figure 4. H4K20 mono- and dimethylation were dispersed
throughout the interphase nucleus with punctate spots showing its enrichment within
specific genomic regions. Monomethylated H4K20 was enriched more toward the
nuclear periphery, excluded from nucleoli and was heavily enriched within the inactive X
chromosome (Figure 4A, X
i
). In contrast, H4K20 dimethylation was more evenly
distributed throughout the nucleus with some enrichment for nucleolar regions and absent
from X
i
. Both mono- and dimethylated H4K20 were excluded from the heterochromatic
DAPI dense foci where the majority of trimethylated H4K20 was found (Figure 4A). To
determine if H4K20 mono-and dimethylation partition to similar nuclear compartments,
co-staining experiments were performed. As show in Figure 4B, mono- (green) and
dimethylated (red) H4K20 are mutually exclusive as determined by a lack of overlap
(yellow).
32
Figure 3. Characterization of H4K20 methyl specific antibodies for
immunofluorescence
Each H4K20 antibody (Cy3; red) was incubated with 1 μg of each of the indicated
peptides for one hour at room temperature in 5% goat serum/PBS prior to incubation with
the fixed MEF cells. Cells were counterstained with DAPI (green). The data indicate
that only the correct corresponding peptide can effectively compete away the signal of
the appropriate H4K20 antibody. Sims et al. 2006
33
34
To determine if methylated H4K20 was associated with transcriptionally engaged
chromatin, cells were co-stained with the methyl-specific antibodies (red) as well as an
anti-RNA polymerase II antibody (RNA pol II, green). RNA pol II is the major
polymerase associated with mRNA transcription and therefore its position within the
nucleus is a marker of actively transcribed regions. As previously described, RNA pol II
staining was equally distributed throughout the nucleus but excluded from the nucleoli
(Figure 4A). H4K20 trimethylation clearly overlapped with heterochromatic foci (purple)
indicating that it is not found within transcriptionally active chromatin, however, the
dispersed staining patterns of H4K20 mono- and dimethylation suggested that these
modifications might associate with active chromatin. As seen in figure 4A, there is some
overlap (yellow) between mono-or dimethylated H4K20 and RNA pol II as seen in the
merged images, indicating that all three forms of methylated H4K20 are found in distinct
silent compartments of the nucleus.
Methylated H4K20 and H3K9 are enriched within the same silent nuclear regions
The staining patterns of methylated H4K20 were strikingly similar to that of the
previously described localization of H3K9 methylation (Rice et al. 2003). Based on this,
we hypothesized that similar methyl states would be targeted to the same silent nuclear
compartments. To test this hypothesis, MEFs were co-stained with various combinations
of the H4K20 and H3K9 methyl specific antibodies. Consistent with previous findings,
the majority of trimethylated H4K20 and H3K9 were enriched within the
35
Figure 4. H4K20 methyl modifications are differentially enriched
within distinct silent nuclear compartments.
A, immunofluorescence staining of MEFs using the H4K20 methyl-specific antibodies
(Ab). H4K20 methylation is visualized as red (Cy3), RNA pol II is green (FITC), and
DNA stained with DAPI is blue. The minor overlap between monomethylated H4K20
and RNA pol II is indicated by the yellow regions in the merged image. The co-
localization of trimethylated H4K20 with the DAPI-dense pericentric heterochromatic
regions is visualized as purple in the merged image. B, dual staining of monomethylated
(mono, red) and dimethylated (di, green) H4K20 indicates that they are mutually
exclusive, as demonstrated by the lack of overlap (yellow) in the merged image. The
unique staining patterns of each methylated form of H4K20 indicate that they partition to
distinct transcriptionally silent nuclear compartments within the mammalian nucleus. Bar
= 5μm. Sims et al. 2006
36
heterochromatic DAPI dense foci (Figure 5C, yellow; Schotta et al. 2004; Rice et al.
2003). Similarly, the bulk of H4K20 and H3K9 monomethylation were found in the
same nuclear compartments (Figure 5A, yellow). To a lesser extent than the monomethyl
patterns, H4K20 monomethylation and dimethylated H3K9 were enriched within the
same compartments with the exception of colocalization at the X
i
(Figure 5A).
Dimethylated H4K20 seemed to partition with dimethylated H3K9 as compared to
monomethylated H3K9, however, there were distinct regions within the nucleus where
the dimethylated residues were found separately (Figure 5B). Taken together, these
results suggest that the methylated states of H3K9 and H4K20 partition to the same
nuclear compartments.
To confirm that the colocalization of these modifications were indeed in the same
regions of chromatin, we prepared extended chromatin fibers from HeLa cells. The
extended chromatin fibers allow us to examine the colocalization of the modifications at
a lower state of chromatin compaction (Sullivan, B.A. and Karpen, G.H. 2004). Co-
staining experiments were performed using the various methyl specific antibodies and
similar to the findings in MEF interphase cells, the monomethylated forms of H4K20 and
H3K9 were enriched in the same punctated regions along the fibers (Figure 6A). Upon
closer examination, we determined that H4K20 monomethylation was nearly always
found enriched within the same regions as H3K9 monomethylation. Interestingly, the
reverse was not always true, as there were regions enriched solely for H3K9
monomethylation along the chromatin fiber suggesting that while monomethylated
37
Figure 5. Specific methylated states of H4K20 and H3K9 co-localize to the
same silent nuclear compartments.
A, MEFs co-stained for monomethylated H4K20 (Cy3; red) and either mono- (top
panels) or dimethylated (bottom panels) H3K9 (FITC, green). Monomethylated H4K20
and H3K9 are preferentially enriched within the same silent nuclear compartments as
observed by the large amount of yellow in the merged image. In contrast,
monomethylated H4K20 is mostly excluded from dimethylated H3K9 regions except on
the inactive X chromosome (Xi). B, MEFs co-stained for dimethylated H4K20 (Cy3; red)
and either mono- (top panels) or dimethylated (bottom panels) H3K9 (FITC; green).
Dimethylated H4K20 and H3K9 co-localize to similar silent nuclear compartments
(yellow in the merged image), although there are also regions where they are mutually
exclusive. Dimethylated H4K20 does not colocalize to monomethylated H3K9 nuclear
compartments. C, MEFs co-stained for trimethylated H4K20 (Cy3; red) and H3K9
(FITC; green). Both modifications are dramatically enriched within DAPI-dense
pericentric heterochromatin. Bar = 5 μm. Sims et al 2006
38
39
H4K20 was found with monomethylated H3K9, monomethylated H3K9 could be
targeted to different regions of the chromatin alone (Figure 6A).
Dimethlyated H4K20 was more dispersed along the chromatin fiber in
comparison with the punctate staining of H4K20 monomethylation, and similar to
findings in the MEFs, it was found enriched in the same regions as dimethylated H3K9
(Figure 6B). Although it seemed that H4K20 dimethylation colocalized with H3K9
monomethylation, upon closer inspection, the regions of dimethylated H4K20 actually
flanked regions enriched for monomethylated H3K9 (Figure 6B and 6D). As expected,
H4K20 and H3K9 trimethylation were found together but unexpectedly, H3K9 mono-and
dimethylation also colocalized with H4K20 trimethylation although at a lower frequency
(Figure 6C). Collectively, these data support the hypothesis that all three forms of
methylated H4K20 and H3K9 preferentially localize to similar silent regions of the
nucleus.
Our findings further suggested that monomethylated H4K20 and H3K9 would be
enriched within the same nucleosome particle. To test this hypothesis, we digested
chromatin extracted from HeLa cells into mononucleosomes using micrococcal nuclease
and used them as input for immunoprecipitations with the H4K20 monomethyl antibody.
As shown in figure 7A, Western analysis was performed on the input (I) and bound (B)
fractions using the H4K20 methyl specific antibodies to confirm the specificity of the
antibodies as well as to demonstrate that each of the methylated forms of H4K20 are
found in distinct regions of the nucleus (Sims et al. 2006). Furthermore, Western
40
Figure 6. Histone H4K20 and Histone H3K9 patterns of methylation are found in
similar genomic regions along extended chromatin fibers.
A, extended chromatin fibers from HeLa cells were co-stained for monomethylated
H4K20 (Cy3; red) and either mono- (left panels) or dimethylated (right panels) H3K9
(FITC; green). Monomethylated H4K20 preferentially co-localizes to monomethylated
H3K9 regions along chromatin fibers as seen by yellow in the merged image. In contrast,
monomethylated H4K20 is largely excluded from regions enriched in dimethylated
H3K9. B, chromatin fibers co-stained for dimethylated H4K20 (Cy3; red) and either
mono- (left panels) or dimethylated (right panels) H3 Lys-9 (FITC; green). Dimethylated
H4K20 preferentially co-localizes to dimethylated H3K9 regions along the chromatin
fibers (yellow in the merged image) and is excluded from regions enriched in
monomethylated H3K9 (see panel D). C, HeLa chromatin fibers co-stained for
trimethylated H4K20 (Cy3; red) and the different methylated states of H3K9 (FITC;
green). Trimethylated H4K20 and H3K9 are enriched within the same regions along the
length of the fiber, whereas mono- and dimethylated H3K9 co-localize less frequently. D,
enlargement of panel B demonstrating that dimethylated H4K20 and monomethylated
H3K9 flank each other along the chromatin fiber (arrows).
Bar = 2.5 μm. Sims et al 2006
41
42
analysis using methyl specific H3K9 antibodies confirmed that nucleosomes enriched for
monomethylated H4K20 were also enriched specifically for monomethylated H3K9
(Figure 7B; Sims et al. 2006). These data further demonstrate that specific methylated
states of H4K20 partition to distinct regions of the nucleus and suggest that a trans-tail
histone code exists whereby H4K20 and H3K9 monomethylation mark distinct silent
regions of chromatin within the mammalian nucleus on the same nucleosome.
Monomethylation of H3K9 requires the H4K20 monomethyltransferase PR-Set7
The identification of a trans-tail histone code between monomethylated H4K20 and
H3K9 was reminiscent of a recent study examining the relationship between
trimethylated H4K20 and H3K9 in which a trans-tail histone code regulating pericentric
heterochromatin was identified. The study demonstrated that lack of the Suv39 HMTs
led to the predicted loss of H3K9 trimethylation but also the unexpected loss of
trimethylated H4K20, suggesting that H4K20 trimethylation is dependent on the presence
of H3K9 trimethylation within pericentric regions (Schotta et al. 2004). Therefore, we
hypothesized that the global decrease in monomethylated H3K9 would result in a
concomitant global decrease in monomethylated H4K20. Mouse embryonic stem cells
lacking the G9a H3K9 methyltransferase were used to test this hypothesis (Tachibana et
al. 2002). Western analysis of the cells confirms our previous findings that the lack of
G9a results in the abolition of detectable dimethylated H3K9 and a significant reduction
43
Figure 7. Preferential and selective enrichment of monomethylated H4K20 and
H3K9 on the same nucleosomal core particle in vivo.
A, mononucleosomes prepared from HeLa cells were immunoprecipitated (IP) with the
H4K20 monomethyl-specific antibody. Western analysis (WB) of 2% of the input
material and 5% of the eluted bound material indicates that the immunoprecipitated
nucleosomes are specifically enriched for the monomethylated form of H4K20. NS
represents nonspecific signal most likely generated by performing both the
immunoprecipitation and the Western analysis with rabbit polyclonal antibodies. B,
Western analysis of the H4K20 monomethyl-enriched nucleosomes indicates that they
are also selectively enriched for monomethylated H3K9. Sims et al. 2006
44
in monomethylated H3K9 (Figure 8A) (Rice et al. 2003). However, no global changes in
the methylated forms of H4K20 were detected. Due to extensive culturing of these cells
it was possible that the loss of monomethylated H4K20 may have been overcome by
selective pressures. To exclude this possibility, HeLa cells were transiently transfected
with siRNA duplexes to specifically decrease the G9a H3K9 methyltransferase or lamin
A/C as the negative control (Figure 8B). Consistent with our previous findings (Rice et
al. 2003), Western analysis demonstrated the selective global reduction of mono- and
dimethylated H3K9 in the G9a siRNA cells compared to the lamin A/C siRNA cells; no
apparent changes in trimethylated H3K9 were observed. Similar to the G9a knockout
cells, there was no changes in global levels of methylated H4K20 (Figure 8A and 8B).
These results indicate that global levels of methylated H4K20 are not dependent on G9a
or global levels of mono- or dimethylated H3K9.
These findings raised the converse possibility: that monomethylated H3K9 may
be dependent upon monomethylated H4K20. Since PR-Set7 is the predominant H4K20
monomethyltransferase (Couture et al. 2005; Xiao et al. 2005), we hypothesized that the
depletion of monomethylated H4K20 via RNAi for PR-Set7 would result in decreased
monomethylated H3K9. Western analysis of HeLa cells transfected with two different
PR-Set7 shRNA plasmids demonstrated significant reductions of PR-Set7 and the
monomethylated form of H4K20; di- and trimethylated levels were unchanged (Figure
45
Figure 8. H4K20 monomethylation is not dependent on H3K9 monomethylation
A, Western analysis of mouse embryonic stem cell whole cell lysates that are lacking one
(+/-) or both copies of G9a (-/-) or with G9a reintroduced to the cells (-/-, rescue) to
detect global changes in H3K9 and H4K20 methylation states. B, Western analysis of
HeLa whole cell lysates treated with siRNA duplexes specific for lamin A/C or G9a to
detect global changes in the different methylated states of H4K20 and H3K9. Sims and
Rice 2008
46
9A). Remarkably, the same cells displayed a near complete global loss of
monomethylated H3K9 without any detectable alterations in di- or trimethylated H3K9.
To determine if these changes were dependent upon H4K20 monomethylation or the
presence of PR-Set7 protein, cells were transfected with a catalytically dead (CD) PR-
Set7 R265G mutant that acts as a dominant negative by depleting cells of
monomethylated H4K20 without reducing levels of PR-Set7 (Figure 9B) (Nishioka et al.
2002). Surprisingly, Western analysis demonstrated that monomethylated H3K9 was
retained in these cells despite the loss of monomethylated H4K20. These findings
indicate that H3K9 monomethylation is dependent upon PR-Set7 but is independent of its
catalytic function.
PR-Set7 interacts with the H3K9 histone methyltransferase, Riz1/PRDM2
While we predicted that H3K9 monomethylation would be required for H4K20
monomethylation, we surprisingly discovered the opposite; that global H3K9
monomethylation was dependent upon the presence of the PR-Set7 H4K20
monomethyltransferase but independent of the catalytic function of PR-Set7. Since we
have shown that monomethylation of these two histone residues are enriched at the same
genomic regions, these findings predict that an unidentified H3K9
monomethyltransferase is targeted to these regions by interacting with PR-Set7 (Figures
8 and 9) (Sims et al. 2006). The obvious candidates for this interaction, G9a and GLP1
(Tachibana et al. 2002; Tachibana et al. 2005), the known H3K9
47
Figure 9. H3K9 monomethylation requires PR-Set7
A, HeLa cells were transfected with two different PR-Set7-specific shRNA constructs and
Western analysis was performed to detect global changes in the methylated states of
H4K20 and H3K9 compared to the empty vector (null). B, Similar Western analysis was
performed in HEK293 cells transfected with a FLAG-tagged control plasmid (null) or a
dominant negative PR-Set7 R256G catalytically dead point mutant (PR-Set7 CD) that
depletes cells of monomethylated H4K20. Sims and Rice 2008
48
monomethyltransferases, do not bind to PR-Set7 (Figure 10).
Recent reports have described a newly characterized H3K9 HMT, Riz1/PRDM2
that was identified as a transcriptional repressor linked to tumorigenesis (Steele-Perkins
et al. 2001; He et al. 1998; Chadwick et al. 2000; Xie et al. 1997). The methyltransferase
activity of Riz1/PRDM2 has been confirmed in vitro to be specific for H3K9 but the
degree of methylation that it achieves has yet to be determined (Kim et al. 2003).
Because of its role as a transcriptional repressor and its unknown methyltransferase
status, we considered it a candidate HMT that could bind to PR-Set7 and monomethylate
H3K9.
To test this hypothesis, we performed in vitro immunoprecipitations by incubating
HeLa nuclear extracts that were overexpressing FLAG-HA tagged PR-Set7 (FLHA Set7)
with in vitro transcribed/translated,
35
S-labelled Riz1/PRDM2. PR-Set7 was
immunoprecipitated from the lysates using anti-FLAG M2 agarose matrix and the bound
fractions were examined for the presence of
35
S-Riz1/PRDM2. As shown in figure 11A,
there is preferential binding between PR-Set7 and Riz1/PRDM2 as compared to the
negative controls, GFP with Riz1/PRDM2 and G9a with Riz1/PRDM2. To further
confirm the interaction between PR-Set7 and Riz1/PRDM2, in vivo co-
immunoprecipitations were performed to determine if the interaction occurs in cells. The
co-immunoprecipitation experiments were performed in the presence of
bismaleimidohexane (BMH), which functions by irreversibly crosslinking sulfhydryl
49
Figure 10. PR-Set7 does not interact with known H3K9 monomethyltransferases
HEK 293 cells were transfected with either A, HA-G9a and myc-PR-Set7 or B, FLAG-
GLP1 and myc-PR-Set7. Twenty-four hours after transfection, protein extracts were
collected using co-IP buffer (50mM Tris pH 7.5, 150mM NaCl, 0.5mM DTT, 1% NP-40,
1μg/mL pepstatin A, 1 μg/mL leupeptin/aprotinin, 1mM PMSF) and
immunoprecipitations were performed at 4°C using α-HA or FLAG-conjugated agarose
beads. 5% of input (I) and 25% of bound fractions (B) were examined by Western
analysis using α-myc antibody to determine binding of PR-Set7 to the indicated proteins.
Sims and Rice 2008
50
groups. Western analysis on the bound fractions demonstrated that PR-Set7 and
Riz1/PRDM2 indeed interact in vivo (Figure 11B). The interaction between PR-Set7 and
Riz1/PRDM2 was only detected in the presence of BMH, indicating that the interation is
weak or quite transient. Importantly, the interaction between PR-Set7 and Riz1/PRDM2
was specific as BMH crosslinking did not lead to non-specific binding between the
control, FLHA p53 and Riz1/PRDM2 (Figure 11B).
The interaction between PR-Set7 and Riz1/PRDM2 predict that Riz1/PRDM2
functions as a H3K9 monomethyltransferase. To test this hypothesis, siRNA duplexes
were used to deplete HeLa cells of either Lamin A/C or Riz1/PRDM2 over a period of 6
days during which protein and RNA extracts were collected to examine both
Riz1/PRDM2 expression as well as monitor levels of monomethylated H3K9 and
H4K20. Quantitative RT-PCR revealed that at days 2 and 6, there was appreciable
knockdown of Riz1/PRDM2 mRNA expression (Figure 12A). However, we were unable
to detect any changes in the global levels of monomethylated H3K9 at either time point
by Western analysis; no changes were observed in monomethylated H4K20 or
dimethylated H3K9 (Figure 12B). It is plausible that the global levels of H3K9
monomethylation will not change as HeLa cells also express G9a and we predict that
depletion of Riz1/PRDM2 will not affect G9a expression. However, monomethylated
H3K9 may be reduced at specific genomic locations in the absence of Riz1/PRDM2. To
test this hypothesis, we examined the enrichment of monomethylated H3K9 at specific
genomic targets in the absence of Riz1/PRDM2.
51
Figure 11. PR-Set7 preferentially interacts with Riz1/PRDM2
A, Nuclear extracts prepared from HeLa cells overexpressing FLHA-PR-Set7 or
FLHA-GFP were incubated with 5 μL in vitro transcribed/translated,
35
S labeled
Riz1/PRDM2 or G9a and α-FLAG-conjugated agarose beads overnight at 4°C. 2% of
input (I), 4% of unbound (U) and 10% of bound (B) fractions were examined by Western
analysis using α-HA antibody to determine immunoprecipitation efficiency. 5% of input
(I), 2% of unbound (U) and 30% of bound (B) fractions were analyzed by
autoradiography to determine binding of Riz1/PRDM2 or G9a to the indicated proteins.
B, HeLa cells were transfected with either FLHA-PR-Set7 and Riz1/PRDM2 or FLHA-
p53 and Riz1/PRDM2. Twenty-four hours after transfection, protein extracts were
collected using co-IP buffer (50mM Tris pH 7.5, 150mM NaCl, 0.5mM DTT, 1% NP-40,
1μg/mL pepstatin A, 1 μg/mL leupeptin/aprotinin, 1mM PMSF) and
immunoprecipitations were performed at 4°C using α-FLAG-conjugated agarose beads.
3% of input (I) and 50% of bound (B) fractions were examined by Western analysis for
immunoprecipitation efficiency using α-HA antibody or for Riz1 binding using α-Riz1
antibody.
52
53
Chromatin immunoprecipitations (ChIPs) were performed in HeLa cells lacking Lamin
A/C or Riz1 using antibodies specific for monomethylated H4K20 and H3K9, a general
histone H3 antibody as the positive control or preimmune rabbit IgG as the negative
control. PCR reactions were performed for each sample using increasing amounts of
ChIP template for amplification of an intronic region of PWP2, which was previously
defined in our laboratory as a genomic region enriched for monomethylated H3K9
(Houston and Rice, unpublished data). Input DNA was used as the positive control for
successful PCR amplification. Interestingly, we observed a decrease in the enrichment of
monomethylated H3K9 within these genomic targets, suggesting that Riz1/PRDM2 does
play a role in monomethylating H3K9 (Figure 13). Surprisingly, we also observed a
significant enrichment of monomethylated H4K20 upon Riz1 knockdown suggesting that
PR-Set7 does not require Riz1/PRDM2 for targeting to specific genomic locations. This,
however, does not explain the consistent increase in H4K20 monomethylation at these
regions.
Collectively, these data support our hypothesis that PR-Set7 is found in complex
with a H3K9 monomethyltransferase, which we identify as Riz1/PRDM2. However,
these results do not address the biological relevance of the trans-tail histone code, does it
function to repress transcription in vivo?
54
Figure 12. Depletion of Riz1/PRDM2 does not affect global levels of H3K9
monomethylation
A, HeLa cells were treated with siRNA duplexes specific for Riz1/PRDM2 or Lamin
A/C. Quantitative RT-PCR was performed to determine levels of Riz1/PRDM2 or lamin
A/C expression, normalized to GAPDH expression, and plotted as the fold increase
relative to mock (y-axis). Experiment was performed once in triplicate. B, H3K9 mono-
and dimethylation and H4K20 monomethylation levels for the mock, lamin A/C and
Riz1/PRDM2 samples were determined by Western analysis. Coomassie staining of
histones was used as the loading control.
55
56
Figure 13. H3K9 monomethylation is reduced within specific genomic locations in
the absence of Riz1/PRDM2
ChIPs were performed in HeLa cells transfected with siRNA duplexes specific for lamin
A/C or Riz1/PRDM2 using either an H4K20 monomethyl-specific antibody, an H3K9
monomethyl-specific antibody, an L3MBTL1 antibody, a general H3 antibody (positive
control) or rabbit preimmune serum (negative control). Increasing amounts of the final
ChIPed material (0.15%, 0.5% and 1.5%; black triangle) were used as the template in a
thirty cycle PCR amplification using primer sets specific to the PWP2 intronic region.
Input DNA (0.005%, 0.0015% and 0.05%) served as the positive control for PCR.
57
PR-Set7 alone is not sufficient to repress transcription
To test our prediction that PR-Set7 and Riz1/PRDM2 and therefore H4K20 and H3K9
monomethylation function to repress transcription, we utilized a luciferase based
transcriptional repression assay. The GAL 4 DNA binding domain was fused to PR-Set7
(DBD-PR-Set7) which allows a protein to be targeted to a plasmid containing 5 copies of
the GAL4 upstream activating sequence followed by a constitutively active SV-40
promoter driving luciferase expression (5xUAS SV40 Luc) (Yu et al. 2003). When
transfected into cells, the 5xUAS SV40 Luc plasmid expresses high levels of luciferase.
If PR-Set7 functions to repress transcription, then co-transfection of DBD-PR-Set7 and
5xUAS SV40 Luc will result in a decrease in luciferase expression (Figure 14A).
Increasing concentrations of DBD-PR-Set7 were co-transfected with 5xUAS SV40 Luc
into HEK293 cells and luciferase levels were measured. As a positive control for
transcriptional repression, increasing amounts of the potent transcriptional repressor,
SMRT was co-transfected with the 5xUAS SV40 Luc (Chen et al. 1995; Yu et al. 2003).
As predicted, we saw ~5 fold decrease in luciferase expression upon co-transfection of
DBD-SMRT and 5xUAS SV40 Luc (Figure 14B). Surprisingly, the luciferase expression
increased as the amount of DBD-PR-Set7 transfected into cells increased, suggesting that
PR-Set7 may function to activate transcription (Figure 14B). One potential explanation
for these unexpected results was that the 5xUAS SV40 Luc plasmid might not have
proper nucleosomal assembly, making it a poor substrate for H4K20 monomethylation by
PR-Set7. To address this issue, we obtained a HEK293 cell line that stably expresses 5
58
copies of the UAS followed by a constitutively active thymidine kinase (TK) promoter
and repeated the experiments with similar results (Ishizuka, T. and Lazar, M.A. 2003).
To test the theory that PR-Set7 may function as a transcriptional activator, we used a
luciferase based activation assay, in which the promoter driving luciferase expression was
a minimal TATA promoter that does not express luciferase unless activated (Webb et al.
1998; Chen et al. 1999). In these experiments, we did not see an activation of luciferase
expression as compared to the positive control, the known transcriptional activator,
CARM1 (Figure 14C) (Chen et al. 1999). The experiments were also repeated using a
HeLa cell line that stably expresses the TATA driven luciferase gene with similar results
to rule out nucleosomal assembly as a potential limitation to the experiment (data not
shown).
Taken together, these data suggest that PR-Set7 and H4K20 monomethlylation
are not sufficient to repress transcription, raising the possibility that a co-factor or other
portion of the complex necessary for repression was missing from our experiments.
Consistent with this, the histone code hypothesis suggests that histone modifications
serve as a binding platform for effector proteins that are responsible for carrying out the
biological function of the modification (Strahl, B.D. and Allis, C.D. 2000; Taverna et al.
2007). One well-defined example of this is di- and trimethylated H3K9, which are
known to be marks of heterochromatin but the modification itself is not sufficient for
silencing. Instead, heterochromatin protein 1 (HP1) must first interact with the
modifications to repress transcription (Lachner et al. 2001).
59
Figure 14. PR-Set7 is not sufficient for transcriptional repression or activation
A, Schematic of luciferase repression assay used to examine the effect of PR-Set7 on
gene expression. A Gal4 DBD construct is either co-transfected with 5xUAS SV40 Luc
into HEK293 cells or the Gal4 DBD construct is transfected into HEK293 cells that are
stably expressing 5xUAS TK Luc. Once targeted to the luciferase vector, PR-Set7 can
monomethylated H4K20, leading to luciferase repression. B, 100ng of 5xUAS SV40 Luc
was co-transfected with increasing concentrations of the indicated Gal4 DBD constructs
(25ng, 50ng, 100ng). Luciferase expression was normalized to Renilla and graphed as
arbitrary light units (y-axis). Three independent biological replicates were performed to
generate standard deviations. C, 100ng of the indicated Gal4 DBD constructs was
transfected into HeLa cells stably expressing 5xUAS TATA Luc. Luciferase expression
was normalized to Renilla and graphed as arbitrary light units (y-axis). Three
independent biological replicates were performed to generate standard deviations.
60
61
L3MBTL1 is required for H4K20 monomethyl mediated transcriptional repression
The tudor family protein, L3MBTL1, was first described as a tumor suppressor
protein in Drosophilia, deletion of which leads to brain malignancies (Gateff et al. 1993;
Wismar et al. 1995; Boccuni et al. 2003). Further studies on the human L3MBTL1
proteins have demonstrated that it functions as a potent transcriptional repressor through
its three MBT repeats (Figure 15A and B) (Boccuni et al. 2003). Several recent studies
have found that L3MBTL1 binds to monomethylated methylated H3K9 and H4K20
peptides, leading us to hypothesize that L3MBTL1 may play a critical role in
transcriptional repression by binding to H4K20 monomethylation (Li et al. 2007; Min et
al. 2007; Kalakonda et al. 2008).
To test this hypothesis, we obtained a non-targeted HA tagged L3MBTL1
construct from Dr. Stephen Nimer’s laboratory to co-transfect with DBD-PR-Set7 into
HEK293 cells stably expressing 5xUAS TK Luc. We predicted that L3MBTL1 would be
targeted to the TK promoter due to an enrichment of monomethylated H4K20 provided
by DBD-PR-Set7. Consistent with this, we observed a dose dependent decrease in
luciferase expression upon transfecting cells with increasing amounts of HA-L3MBTL1
and a constant amount of DBD-PR-Set7, suggesting that L3MBTL1 indeed plays a role
in repressing transcription through binding monomethylated H4K20 (Figure 15B)
(Kalakonda et al. 2008). To confirm that the MBT repeats are critical for repression, we
transfected in a mutant L3MBTL1 that lacks the MBT repeats and found that the
repression of luciferase expression was relieved (Figure 15B) (Kalakonda et al. 2008).
62
Figure 15. L3MBTL1 interacts with monomethylated H4K20 to repress
transcription
A, Schematic representation of the tumor suppressor, L3MBTL1. L3MBTL1 contains 3
conserved MBT repeat domains (termed p1, p2, p3) a zinc finger doimain (Zn) and a
SPM domain thought to be important for protein-protein interactions. B, HEK 293 cells
were co-transfected with the indicated Gal4-DBD constructs and increasing amounts of
HA-tagged L3MBTL1. These data show that PR-Set7 and H4K20 monomethylation are
not sufficient to repress gene expression. However, when co-expressed with L3MBTL1,
there is a dose dependent decrease in luciferase expression. Co-transfection experiments
with the PR-Set7 R265G catalytically dead mutant demonstrates that repression is
dependent on H4K20 monomethylation. Furthermore, the interaction between H4K20
methylation and L3MBTL1 is specific for monomethylation, as co-transfection with the
H4K20 trimethyltransferase, Suv 4-20, has no effect on repression.
Kalakonda et al. 2008
63
64
One explanation for repression of luciferase expression could be that repression is
mediated through interaction of PR-Set7 and L3MBTL1 and not the interaction of
L3MBTL1 with monomethylated H4K20. To test this hypothesis, we fused the dominant
negative, catalytically dead point mutant of PR-Set7 (R265G; CD) to Gal4 DBD and co-
transfected this construct with increasing amounts of L3MBTL1 (Nishioka et al. 2002).
No repression was observed, confirming that the critical players for repression are
H4K20 monomethylation and L3MBTL1 (Figure 15B) (Kalakonda et al. 2008). A recent
study reported that the MBT repeats can bind to dimethylated H4K20 (Kim et al. 2006).
This raises the question of whether L3MBTL1 could bind to any methylated lysine 20
residue and induce repression. To exclude this possibility, we co-transfected the known
H4K20 di-and trimethyltransferase, Suv4-20 (DBD-Suv4-20) with increasing amounts of
L3MBTL1 and did not observe any change in luciferase expression (Figure 15B)
(Kalakonda et al. 2008). Collectively, these data demonstrate that interaction of
L3MBTL1 with monomethylated H4K20 is critical for transcriptional repression, through
interaction of the monomethyl modification with the MBT domains. The crystal
structure of the MBT repeats binding to the H4K20 peptides have been solved,
demonstrating that the 2
nd
MBT repeat is the critical domain for interaction with
methylated H4K20 residues (Li et al. 2007; Min et al. 2007; Trojer et al. 2007).
65
Discussion and Future Directions
In these studies, we have demonstrated that H4K20 methylation is localized to
transcriptionally inactive regions of the nucleus and that each level of methylation
(mono-, di- or trimethylation) partitions to a separate compartment of the nucleus.
Furthermore we have found that each modification is enriched within the same genomic
locations as another mark of silent chromatin, H3K9 methylation, defining a novel
repressive trans-tail histone code (Sims et al 2006).
We further define the repressive trans-tail histone code by examining the
relationship between monomethylated H4K20 and H3K9. Surprisingly, we found that
global levels of H3K9 monomethylation are dependent on the presence of PR-Set7
protein, predicting that an unknown H3K9 monomethyltransferase interacts with PR-Set7
either directly or indirectly. We tested for binding of the known H3K9
monomethyltransferases, G9a and GLP1 and could not detect binding between these
HMTs and PR-Set7. We have identified Riz1/PRDM2 as a candidate H3K9
methyltranferase that could interact with PR-Set7 and demonstrated binding to PR-Set7
through in vitro and in vivo immunoprecipitation experiments. These experiments
however, do not address whether interaction between Riz1/PRDM2 and PR-Set7 is direct
or indirect. It is possible that other members of the PR-Set7 multi-protein complex could
interact with or recruit Riz1/PRDM2 to specific genomic regions to monomethylate
H3K9. This question can be addressed through in vitro binding experiments using
recombinant PR-Set7 and Riz1/PRDM2.
66
Riz1/PRDM2 has been described as an H3K9 methyltransferase but the
specificity of Riz1/PRDM2 is still unclear (Kim et al. 2003). We were unable to detect a
change in global levels of either mono-or dimethylated H3K9 upon knockdown of
Riz1/PRDM2 in HeLa cells (Figure 12). This was surprising and unexpected based on
our findings that lack of PR-Set7 resulted in a global loss of H3K9 monomethylation
(Figure 9). However, knockdown of G9a, the most well characterized H3K9
monomethyltransferase also does not completely deplete cells of monomethylated H3K9,
suggesting that more than one H3K9 HMT can compensate for global levels of
monomethylated H3K9. To address this issue, knockdown of Riz1/PRDM2 would need
to be performed in the absence of G9a to examine the residual levels of H3K9
monomethylation. To date, there are no available G9a knockdown or knockout cell lines
and so knockdown of both HMTs simultaneously would have to be standardized. In an
opposite approach, overexpression of Riz1 in the absence of G9a could be performed to
examine if Riz1 can return H3K9 monomethylation levels to WT levels. An alternative
approach to identify the methyl state that Riz1/PRDM2 achieves would be to perform in
vitro HMT assays using either histone H3 peptides or nucleosomes as a substrate
followed by Western analysis to examine the levels of H3K9 methylation. A third
method to identify the methyl state specificity of Riz1/PRDM2 would be to overexpress
Riz1/PRDM2 in the MES G9a knockout cell lines to determine if overexpression of
Riz1/PRDM2 is sufficient to rescue the loss of H3K9 monomethylation in these cells.
Until these experiments have been performed, it will remain unclear as to whether or not
67
Riz1/PRDM2 is truly an H3K9 monomethyltransferase and if it contributes to global
levels of monomethylated H3K9. To our knowledge, there are no further known
candidate proteins that could function as a H3K9 monomethyltransferase as all other
H3K9 HMTs such as SETDB1 and ESET have been defined biochemically as H3K9 di-
or trimethyltransferases (Wang et al. 2003; Li et al. 2006)
In a separate attempt to define the methyl state of Riz1/PRDM2, we performed
ChIP experiments in HeLa cells depleted of Riz1/PRDM2 within a defined H3K9
monomethyl genomic region. We observed a clear loss of H3K9 monomethylation from
the PWP2 intronic region suggesting that Riz1/PRDM2 functions as a H3K9
monomethyltransferase albeit a minor HMT. Interestingly, we also observed an increase
in monomethylated H4K20 within the target gene in the absence of Riz1/PRDM2 as well
as a clear depletion of histone H3 general, possibly indicating a change in chromatin
structure within the region. This suggests that PR-Set7 can be targeted to genomic
locations without Riz1/PRDM2 but it remains unclear as to why there is an increase in
monomethylated H4K20 and what the role of H4K20 monomethylation plays in gene
regulation and chromatin structure within this gene. These data also suggest that
Riz1/PRDM2 could negatively regulate PR-Set7, monomethylated H4K20 and
L3MBTL1 enrichment at target genes possibly by methylating PR-Set7 itself to regulate
its localization to target genes. This hypothesis can be tested by in vitro methylation
assays examining the ability of PR-Set7 to be methylated by Riz1/PRDM2. If PR-Set7 is
methylated, how does this methylation affect function and targeting? This can be
68
addressed through creating a point mutant of the methylation site and overexpressing the
mutant PR-Set7 and examine its function in cells through Western or ChIP analysis.
Alternatively, a dominant negative, catalytically dead point mutant of Riz1/PRDM2
could be overexpressed in cells and the enrichment of monomethylated H4K20 can be
examined at target genes to determine if the catalytic function of Riz1/PRDM2 is critical
for PR-Set7 targeting and function.
These results open many more questions about the trans-tail histone code such as
if Riz1/PRDM2 is a H3K9 monomethyltransferase but only contributes a minor portion
of monomethylation to the genome, then why is the majority of monomethylated H3K9
depleted in the absence of PR-Set7, what is the significance of the increase in
monomethylated H4K20 in the absence of Riz1/PRDM2 and most importantly what is
the biological relevance of these paired histone modifications?
To begin to address the biological relevance of the trans-tail histone code, we
examined the role of PR-Set7 in gene expression. We demonstrated that PR-Set7 alone is
not sufficient for either transcriptional repression or activation and that the transcriptional
repressor, L3MBTL1 is a critical component of the silencing pathway as its MBT repeats
specifically bind to monomethylated H4K20 to repress transcription. Consistent with
this, we also observed an increase in L3MBTL1 within the PWP2 intronic region in a
similar manner as the increase in H4K20 monomethylation. We found that in the
absence of PR-Set7 catalytic activity, there was no detectable gene repression.
Presumably, the nucleosomes within the TK promoter would be enriched for
69
monomethylated H3K9 as Riz1/PRDM2 would be in complex with PR-Set7, suggesting
that monomethylated H3K9 is not sufficient for interaction with L3MBTL1 to repress
transcription. This result has been confirmed in vivo through examination of L3MBTL1
localization to the H4K20 monomethyl target gene, RUNX1, upon overexpression of PR-
Set7 CD.
A recent report analyzed the role of L3MBTL1 in altering chromatin structure and
demonstrated that L3MBTL1 and its MBT domains are critical for chromatin compaction
in the presence of nucleosomes monomethylated at H4K20 residues (Trojer et al. 2007).
Compaction within monomethylated H4K20 target genes could provide a mechanism for
gene silencing. This hypothesis can be examined using chromatin accessibility assays in
well-defined targets of L3MBTL1 and H4K20 monomethylation. These experiments
would provide great insights into the role H4K20 monomethylation and L3MBTL1 play
in regulating gene expression in vivo.
70
Chapter 2. Regulation of Differentiation by H4K20 Monomethylation
To further characterize the biological importance of the trans-tail histone code we
needed to identify target genes. To our knowledge, there were no reports of specific
genes enriched for monomethylated H4K20 and H3K9 with L3MBTL1. Based on
experiments described above, we predicted that loss of PR-Set7 and therefore H4K20 and
H3K9 monomethylation would lead to an increase in gene expression. We also
hypothesized that the genes regulated by the trans-tail histone code would play an
important role in development as the loss of PR-Set7 in Drosophila and mouse results in
severe developmental defects (Nishioka et al. 2002; Karachentsev et al. 2005; Huen et al.
2008).
Loss of PR-Set7 results in an increase in gene expression
To determine the target genes regulated by PR-Set7 and H4K20
monomethylation, we extracted RNA from HeLa cells depleted of PR-Set7 as well as
mock transfected cells. The RNA samples were labeled and applied to an Affymetrix
expression microarray to compare gene expression patterns between mock and PR-Set7
knockdown cells. We identified a total of 2100 probes that were altered (either 2 fold up
or downregulated) in our comparison studies. Of these probes, 67% of them were
upregulated in the absence of PR-Set7, suggesting that PR-Set7 and monomethylated
H4K20 does function to repress transcription. As shown in table 1, approximately 50%
of the genes upregulated by greater than 10 fold were either cell type specific genes or
involved in differentiation (denoted by *). Quantitative real time PCR (RT-PCR)
71
analysis was performed to verify the genes identified by microarray. During this
analysis, we observed a high false positive rate as only ~30% of the genes consistently
were upregulated in the absence of PR-Set7 and H4K20 monomethlyation (Table 1,
denoted by #, false positives denoted by ^).
PR-Set7 monomethyltransferase activity is required to repress RUNX1
One of the identified target genes was AML1/RUNX1; a master regulator of
hematopoietic differentiation and a gene commonly translocated in leukemias (Kurokawa
2006). Quantitative RT-PCR was performed on the PR-Set7 shRNA and control shRNA
HeLa cells to confirm these findings (Figure 16A). As predicted, there was a >2-fold
increase in RUNX1 expression in the absence of PR-Set7 and monomethylated H4K20.
RUNX1 expression was also analyzed in HeLa cells transfected with the PR-Set7 CD
plasmid to determine if the increase in expression was directly correlated to a loss of
monomethylated H4K20. As we had previously observed with several other genes
enriched in monomethylated H4K20 (Table 1; Houston et al, unpublished results), the
absence of this histone modification resulted in a >3-fold increase in RUNX1 expression.
Concomitant with an increase in RUNX1 mRNA levels, we also observed a significant
increase in RUNX1 protein levels in the PR-Set7 shRNA cells (Figure 16B). These
findings strongly suggest that the monomethylation of H4K20 by PR-Set7 plays a key
upstream regulatory role in RUNX1 expression. Furthermore, these findings suggest that
the presence of PR-Set7 protein, itself, and monomethylated H3K9 are not sufficient for
RUNX1 repression.
72
Gene Name Fold Increase Function
ZBTB1
#
71 Zinc finger, transcription factor
TRDN^ 42 Mitochondrial kinase
DNAJC15^ 37 Drug resistance
ZNF560^ 34 Zinc finger; KRAB repressive domain
CHES1 32 DNA damage-inducible checkpoint
DSG2^ 29 Cell adhesion
PLAU
#
29 Serine protease; degrades extracellular matrix
PDGFD* 27 Platelet derived growth factor
RTN4IP1^ 26 Mitochondrial; inhibits cell regeneration
CECR5* 26 Cat eye syndrome-like protein
Klotho*
#
26 Aging, bone morphogenesis
HTN* 26 Salivary-specific protein
RUNX1*
#
23 Hematopoietic differentiation
RSN
#
23 Proliferation; cell survival
ZNF677^ 23 Zinc finger; DNA binding
SKIL* 21 Differentiation; repressor protein
TIG3*
#
21 Retinoic receptor responder
TCF12* 16 Regulator of lineage specific genes
erythropoietin* 10 Regulator of red cell production
Table 1. List of genes upregulated ≥10 fold in the absence of PR-Set7
Expression of RNA extracted from HeLa cells depleted of PR-Set7 was compared to
RNA extracted from mock-transfected HeLa cells using an Affymetrix expression
microarray. Shown above is a partial list of genes that were upregulated ≥10 fold.
* denotes genes that are involved in differentiation, development or are cell type specfic
#
denotes genes that were confirmed to be upregulated in the absence of H4K20
monomethylation by quantitative RT-PCR.
^ denotes genes that were found to be false positives by quantitative RT-PCR.
73
Figure 16. Monomethylated H4K20 and L3MBTL1 at the RUNX1 promoter is
associated with RUNX1 repression.
A, HeLa cells were transfected with a control expression vector (mock), a vector
expressing full length PR-Set7 (WT) or the R265G catalytically dead mutant (CD), or
shRNA vectors that specifically deplete cells of PR-Set7 or L3MBTL1. Quantitative RT-
PCR was performed to determine levels of RUNX1 expression, normalized to GAPDH
expression, and plotted as the fold increase relative to mock (y-axis). Three independent
biological replicates were performed to generate standard deviation. B, RUNX1 protein
levels for the mock and PR-Set7 shRNA samples were determined by Western analysis.
A general histone H4 antibody was used as the loading control. C, ChIPs were
performed in HeLa cells transfected with an empty vector (null), a PR-Set7 shRNA
vector or the PR-Set7 CD vector using either an H4K20 monomethyl-specific antibody,
an H3K9 monomethyl-specific antibody, an L3MBTL1 antibody, a general H3 antibody
(positive control) or rabbit preimmune serum (negative control). Increasing amounts of
the final ChIPed material (0.15%, 0.5% and 1.5%; black triangle) were used as the
template in a thirty cycle PCR amplification using primer sets specific to the RUNX1
promoter or upstream region (negative control). Input DNA (0.005%, 0.0015% and
0.05%) served as the positive control for PCR. D, Semi-quantitative analysis was
performed by first calculating the density of each PCR band using Quantity One
(BioRad). The resultant values were plotted and the slope for each sample was
determined and then graphed relative to the slope of the input (y-axis) to reflect the
74
degree of enrichment for each histone modification or protein. Three independent
biological replicates were performed to generate standard deviation. Sims and Rice 2008
75
76
Monomethylated H4K20 and H3K9 are selectively targeted to the RUNX1 promoter
Since we had identified that monomethylated H4K20 was enriched at the RUNX1
promoter in HeLa cells (Houston et al, unpublished data), we predicted that this putative
silencing pathway was selectively targeted to this region to repress RUNX1 expression.
To determine this, ChIPs were performed in HeLa cells (null) using antibodies specific
for monomethylated H4K20, a general histone H3 antibody as the positive control or
preimmune rabbit IgG as the negative control. PCR reactions were performed for each
sample using increasing amounts of ChIP template for amplification of the RUNX1
promoter or a region 160 kb upstream that is devoid of monomethylated H4K20 (Figure
16C). Input DNA was used as the positive control for successful PCR amplification.
Visual inspection of the PCR reactions verify that monomethylated H4K20 was enriched
at the RUNX1 promoter compared to the RUNX1 upstream region in HeLa cells (null).
Quantitative analysis was performed by first plotting the intensity of each band of a
sample to determine a slope (Figure 17). The slope of each ChIP was then normalized to
the slope of the input DNA and plotted with standard error generated from three
independent biological replicates (Figure 16D). Consistent with Figure 16C, the
quantitative analysis indicates an ~12-fold enrichment of monomethylated H4K20 at the
RUNX1 promoter compared to the upstream region in the HeLa cells (null). Based on the
findings above, we predicted that monomethylated H3K9 would also be specifically
77
Figure 17. Slope calculations for semi-quantitative ChIP analysis
A, ChIP analysis was performed as described in Figure 13C. B, Example of linear plots
of density measurements collected from ChIP PCRs in panel A. The values were plotted
as A, B and C for each part of the gradient shown on the agarose gels. Sims and Rice
2008
78
targeted to this region. ChIPs were performed with an H3K9 monomethyl-specific
antibody and visual inspection of the resultant PCR amplifications revealed an
enrichment of monomethylated H3K9 at the RUNX1 promoter as compared to the
upstream region (Figure 16C). Quantitative analysis confirmed an ~5-fold increase of
monomethylated H3K9 at the RUNX1 promoter providing further evidence of the
selective targeting of this trans-tail histone code (Figure 16D).
To further verify these findings, ChIPs were performed in HeLa cells transfected
with the PR-Set7 shRNA plasmid that depletes cells of global levels of monomethylated
H4K20 (Figure 9B). Visual inspection of the PCR amplifications from the PR-Set7
shRNA cells revealed a significant reduction in monomethylated H4K20 at the RUNX1
promoter when compared to the null cells (Figure 16C). Quantitative analysis indicates
an ~10-fold decrease in monomethylated H4K20 at the RUNX1 promoter; similar to
levels observed at the upstream region (Figure 16D). ChIP analysis performed with the
monomethyl-specific H3K9 antibody in the PR-Set7 shRNA cells also revealed an ~12-
fold reduction of this modification at the RUNX1 promoter when compared to null cells;
similar to levels observed at the upstream region. These findings indicate that both
H4K20 and H3K9 monomethylation at the RUNX1 promoter are dependent on PR-Set7.
Additional ChIP analysis was performed in HeLa cells transfected with the PR-
Set7 CD plasmid, which serves as a dominant negative by reducing monomethylated
H4K20 in the cells without decreasing PR-Set7 protein levels (Figure 9C). As predicted,
visualization of the PCR reactions demonstrates a marked reduction in monomethylated
79
H4K20 at the RUNX1 promoter compared to the null cells (Figure 16C). Quantitation of
the samples confirmed the reduction, however, the levels of monomethylated H4K20 at
the RUNX1 promoter were consistently higher when compared to the PR-Set7 shRNA
cells (Figure 16D). This is most likely due to the inherent competition between wild type
and mutant PR-Set7 in the transfected cells resulting in the observed residual
monomethylation of H4K20 at the RUNX1 promoter. Consistent with the findings above,
ChIP analysis demonstrated no observable change in monomethylated H3K9 at the
RUNX1 promoter in the PR-Set7 CD cells compared to null (Figure 16C and 16D).
These findings provide further evidence that both histone modifications are dependent
upon the targeting of PR-Set7 to specific regions in the genome. It is also important to
note that the continued presence of monomethylated H3K9 at the RUNX1 promoter in the
PR-Set7 CD cells does not inhibit RUNX1 derepression when compared to the PR-Set7
shRNA cells, providing further evidence that monomethylated H3K9 is not sufficient for
RUNX1 repression (Figure 16A).
Decreased monomethylated H4K20 results in reduced L3MBTL1 at RUNX1
We and others have shown that the MBT repeats of the L3MBTL1 repressor
protein selectively binds monomethylated H4K20 in vitro (Figure 15B; Kim et al. 2006;
Li et al. 2007; Min et al. 2007). Due to this association, we theorized that L3MBTL1 was
directly binding monomethylated H4K20 in vivo to repress RUNX1 expression.
Consistent with this theory, depletion of L3MBTL1 by RNAi resulted in a >4-fold
80
increase in RUNX1 expression, similar to the increase observed in the absence of
monomethylated H4K20, indicating a role for L3MBTL1 in regulating RUNX1 (Figure
16A). Based on these findings, we predicted that L3MBTL1 was targeted specifically to
the RUNX1 promoter. To test this, ChIP analysis using an L3MBTL1 antibody was
performed in HeLa (null) cells. Visualization and quantitation of the PCR amplifications
confirmed a significant enrichment of L3MBTL1 at the RUNX1 promoter compared to
the upstream region (Figure 16C and 16D). Importantly, the depletion of PR-Set7 and
monomethylated H4K20 in the PR-Set7 shRNA HeLa cells resulted in a dramatic
reduction of L3MBTL1 from the RUNX1 promoter when compared to null cells; similar
to the levels observed in the upstream region. The loss of L3MBTL1 from the promoter
was coincident with the derepression of RUNX1 (Figure 16A). ChIP analysis in the PR-
Set7 CD cells revealed a marked reduction of L3MBTL1 enrichment at the RUNX1
promoter compared to the null cells (Figure 16C and 16D). However, the decreased
levels of L3MBTL1 did not reach those observed in the PR-Set7 shRNA cells, most
likely due to persistence of low levels of monomethylated H4K20 by the competition
between the mutant and endogenous wild type PR-Set7. Collectively, these findings
indicate that the presence of L3MBTL1 at the RUNX1 promoter is associated with
RUNX1 repression and that the recruitment of L3MBTL1 to this region is most likely due
to its interaction with monomethylated H4K20.
81
Monomethylated H4K20 is required for L3MBTL1 recruitment to repress RUNX1
To verify that PR-Set7 was targeted to the RUNX1 promoter, a myc-tagged
eukaryotic expression vector encoding full length PR-Set7 was transfected into HeLa
cells for ChIP analysis using myc antibodies, since a reliable ChIP-grade PR-Set7
antibody is not yet available. While we anticipated that over-expression of the PR-Set7
monomethyltransferase would result in global increases in monomethylated H4K20,
Western analysis demonstrated a significant reduction of this histone modification
compared to myc-null cells (Figure 18A). This reduction was proportional to a dramatic
global increase in trimethylated H4K20 but with no observable changes in dimethylated
H4K20. ChIP analysis was performed with various antibodies at the RUNX1 promoter in
HeLa cells transfected with the myc-PR-Set7 plasmid or the myc-null control plasmid
(Figure 18B). ChIPs performed with the myc antibody demonstrated an enrichment of
PR-Set7 at the RUNX1 promoter compared to myc-null cells confirming that PR-Set7 is
targeted to the RUNX1 promoter (Figure 18B). Consistent with the Western analysis,
ChIPs in cells over-expressing myc-PR-Set7 demonstrated a significant reduction in
monomethylated H4K20 at the RUNX1 promoter and a concomitant increase in
trimethylated H4K20. No observable changes in monomethylated H3K9 were detected at
the RUNX1 promoter. Since these cells were co-transfected with an HA-tagged
eukaryotic expression plasmid encoding full length L3MBTL1, ChIP analysis was
performed using an HA antibody to assess its enrichment at the RUNX1 promoter in these
different backgrounds. The results demonstrate that HA-L3MBTL1 was selectively
82
Figure 18. Monomethylated H4K20 is required for L3MBTL1 recruitment and
RUNX1 repression.
A, HeLa cells were co-transfected with an HA-tagged L3MBTL1 plasmid and either a
myc-PR-Set7 plasmid or myc-null plasmid as the negative control. Western analysis of
cell lysates was performed with methyl-specific H4K20 antibodies or a general H4
antibody as the loading control. B, ChIP analysis was performed in these cells as
described in Figure 13C. Sims and Rice 2008
83
84
reduced in the myc-PR-Set7 cells containing decreased monomethylated H4K20 and
increased trimethylated H4K20 at the RUNX1 promoter compared to the myc-null cells.
Importantly, the reduction of monomethylated H4K20 and L3MBTL1 at the RUNX1
promoter in the myc-PR-Set7 cells is coincident with increased RUNX1 expression
(Figure 16A). Collectively, these findings indicate that PR-Set7 and H4K20
monomethylation are targeted to the RUNX1 promoter which strongly suggest that
RUNX1 repression is mediated by the specific binding of L3MBTL1 to monomethylated
H4K20.
Decreased monomethylated H4K20 is specifically associated with megakaryocytic
differentiation
The RUNX1/AML1 transcription factor plays a critical role in mammalian
hematopoiesis, in part, by regulating many hematopoietic lineage-specific genes
(Ichikawa et al. 2004a). RUNX1 expression is also tightly regulated during the
establishment and maintenance of lineage-committed cells in adult megakaryopoiesis.
Consistent with this, the absence of RUNX1 results in the loss of definitive
hematopoiesis and is associated with defects in platelet production (Okuda et al. 1996;
Ichikawa et al. 2004b). In precursor blast-forming unit erythroid/megakaryocte (BFU-
E/MK) cells, RUNX1 expression is repressed and remains absent during erythropoiesis
(Figure 19A) (North et al. 2004). In contrast, enhanced expression of RUNX1 in
85
Figure 19. Megakaryocytic induction leads to an increase in RUNX1 expression in
K562 cells
A, The precursor blast-forming unit erythrocyte/megakaryocyte (BFU-E/MK) cells are
RUNX1 negative. During erythropoiesis, the cells remain RUNX1 negative whereas
RUNX1 expression is required for megakaryopoiesis. The human K562 multipotent cell
line mimics BFU-E/MK cells and can be induced to selectively differentiate depending
on treatment with hemin or TPA. B, Efficiency of K562 megakaryocytic differentiation
was determined by FACS analysis for the percentage of CD41 positive cells after
treatment with 10nM TPA for 5 days compared to vehicle control (left panel). Standard
methods of benzidine staining and spectrophotomic measurement of soluable hemoglobin
were used to determine erythrocytic differentiation efficiency after treatment with 50mM
hemin for 4 days compared to vehicle control (right panel). Three independent biological
replicates were used to generate standard deviation. C, Whole cell lysates of K562 cells
treated with vehicle and either hemin or TPA were analyzed for RUNX1 protein levels.
A general H4 antibody was used to determine equal loading. Sims and Rice 2008
86
87
precursor cells is an early event in megakaryopoiesis and recent findings demonstrate that
expression of RUNX1 sensitizes precursor cells towards megakaryopoiesis (Elagib et al.
2003; Ichikawa et al. 2004b). Based on these reports and our findings above, we
predicted that PR-Set7 and monomethylated H4K20 would be reduced in cells committed
to megakaryopoiesis compared to precursor cells. Since the human K562 multipotent
cell line mimics precursor cells (Tani et al. 1996), we used these as the model system to
determine if the components of the silencing pathway were also present at the RUNX1
promoter. ChIP analysis revealed that, identical to the findings in HeLa cells, all
components of the pathway were enriched at the RUNX1 promoter compared to the
upstream region in the wild type K562 cells (Figure 20A). The presence of this pathway
at the RUNX1 promoter is coincident with the repression of RUNX1 and the absence of
protein product (Figure 19C) (Elagib et al. 2003). Consistent with our data in HeLa cells,
these findings indicate that monomethylated H4K20 and L3MBTL1 are associated with
RUNX1 repression in K562 cells.
The K562 multipotent cells can be chemically treated with hemin or phorbol
esters to induce erythrocytic or megakaryocytic differentiation, respectively (Figure 19A)
(Rutherford et al. 1979; Tetteroo et al. 1984). We confirmed these differentiation
pathways using increased expression of the cell surface marker, CD41, as a marker of
megakaryocyte differentiation, and an increase in hemoglobin production measured by
soluable hemoglobin levels as well as benzidine staining as a measure of erythrocytic
differentiation (Figure 19B).
88
Since RUNX1 is expressed specifically during megakaryopoiesis, we predicted
that K562 cells treated with phorbol esters would display decreased levels of PR-Set7 and
monomethylated H4K20 compared to vehicle treated cells. Consistent with this
hypothesis, Western analysis of K562 cells treated with the phorbol ester, TPA, displayed
reduced levels of PR-Set7 and monomethylated H4K20 compared to control cells and
this was coincident with increased RUNX1 (Figure 20C). Although we predicted a
decrease in monomethylated H3K9 due to the reduction of PR-Set7, the levels remained
relatively unchanged in TPA-treated K562 cells. Also unexpected was the observed
decrease in PR-Set7 in K562 cells induced to the erythrocytic lineage by treatment with
hemin. However, RUNX1 was not expressed in these cells consistent with the relatively
elevated global levels of both monomethylated H4K20 and H3K9. These findings
strongly suggest that the specific reduction of monomethylated H4K20 is required for
megakaryopoiesis in K562 cells.
Loss of monomethylated H4K20 and L3MBTL1 at the RUNX1 promoter is an early
event of megakaryopoiesis
Based on the findings above, we predicted that the components of this new
silencing pathway would be absent at the RUNX1 promoter specifically in the TPA-
treated K562 cells compared to vehicle control or hemin-treated cells. ChIP analysis was
performed in the different heterogeneous cell backgrounds where ~42% of hemin-treated
89
Figure 20. Decreased monomethylated H4K20 is selectively associated with
megakaryopoiesis.
A, ChIP analysis was performed in K562 cells as described in Figure 13C. B,
Quantitative real time PCR analysis was used to determine levels of RUNX1, PR-Set7 and
L3MBTL1 in K562 cells treated with vehicle, hemin or TPA. Values were normalized to
GAPDH and then plotted relative to vehicle control cells (y axis). Three independent
biological replicates were used to generate standard deviation.
C, K562 cells were treated with vehicle or either hemin or TPA to induce erythropoiesis
or megakaryopoiesis, respectively. Western analysis was performed on the cell lysates
using the indicated antibodies. A general H4 antibody was used as the loading control.
Sims and Rice 2008
90
91
cells were positively stained for benzidine and ~33% of the TPA-treated cells were CD41
positive (Figure 19B). Visualization of the subsequent PCR amplifications confirmed the
enrichment of monomethylated H4K20, monomethylated H3K9 and L3MBTL1 at the
RUNX1 promoter in both the vehicle control K562 cells and those treated with hemin
when compared to the RUNX1 upstream region (Figure 21A). As predicted, there was a
marked reduction in monomethylated H4K20 and L3MBTL1 at the RUNX1 promoter in
the TPA-treated K562 cell population. In contrast to the global analysis, there was an
observable decrease in monomethylated H3K9 at the RUNX1 promoter. Quantitation of
the PCR products from three independent biological replicates, as described above,
confirmed these observations (Figure 21B). Therefore, the loss of monomethylated
H4K20 and L3MBTL1 at the RUNX1 promoter is associated with increased RUNX1 and
the selective differentiation toward the megakaryocytic lineage.
Increased RUNX1 is critical during early stages of megakaryocytic differentiation
as megakaryopoiesis fails to initiate without RUNX1 (Ichikawa et al. 2004b; North et al.
2004). Based on our findings we predicted that the loss of monomethylated H4K20 and
L3MBTL1 from the RUNX1 promoter was a critical upstream event required for RUNX1
activation during the commitment of precursor cells to megakaryocytes. Since
expression of the CD41 cell surface marker occurs early in megakaryopoiesis, prior to
increased DNA ploidy and cell size (Ichikawa et al. 2004b), TPA-treated K562 cells in
the early stages of megakaryocytic differentiation were isolated by FACs based on both
92
Figure 21. Reduction of monomethylated H4K20 and L3MBTL1 at the RUNX1
promoter is an early event in megakaryopoiesis.
A, ChIP analysis was performed in K562 cells treated with vehicle, hemin or TPA as
described in Figure 13C. B, Semi-quantitative analysis of the ChIP data was performed
as described in Figure 13D. C, TPA-treated K562 cells in the early stages of
megakaryopoiesis were isolated by FACs based on small cell size and expression of the
CD41 cell surface marker. ChIP analysis was performed on these cells or vehicle treated
cells using either an H4K20 monomethyl-specific antibody, an H3K9 monomethyl-
specific antibody, a L3MBTL1 antibody, a general H3 antibody (positive control) or
rabbit preimmune serum (negative control). Thirty cycles of PCR amplifications were
performed using primers for the RUNX1 promoter and 1.5% of the ChIPed material as the
template. Input DNA (0.05%) was used as the positive control for PCR. D, Semi-
quantitative analysis of the ChIP data was performed by determining the density of each
PCR band using Quantity One (BioRad). The resulting value for each ChIP sample was
then plotted relative to the value of the input signal to reflect the degree of enrichment of
each histone modification or protein. Three independent biological replicates were
performed to generate standard deviation.
Sims and Rice 2008
93
94
their small size and high levels of CD41 expression. ChIP analysis was performed in
these cells compared to K562 cells treated with vehicle. Visualization of the subsequent
PCR amplifications demonstrated a reduction of monomethylated H4K20 and L3MBTL1
at the RUNX1 promoter in the early differentiating K562 cells (Figure 21C). Quantitative
analysis confirmed these observations and also revealed decreased monomethylated
H3K9 at the RUNX1 promoter; this change was not apparent by visualization (Figure
21D). These findings demonstrate that the reduction of monomethylated H4K20 and
L3MBTL1 at the RUNX1 promoter is an early event in megakaryopoiesis.
Catalytically active PR-Set7 is required to prevent megakaryopoiesis
It was previously reported that the expression of RUNX1 sensitizes K562 cells
towards megakaryocytic differentiation upon TPA treatment (Elagib et al. 2003).
Therefore, our findings predicted that the depletion of monomethylated H4K20 and the
associated activation of RUNX1 would have a similar sensitizing effect. To test this
hypothesis, K562 cells were first co-transfected with a plasmid that constitutively
expresses GFP and either a FLAG-tagged RUNX1 plasmid or an empty FLAG-null
plasmid as the negative control prior to treatment with TPA. FACs analysis was
performed in these different cells five days post-transfection to determine the number of
CD41 positive cells within the GFP positive cell population (Figure 22). Consistent with
the previous report, we observed an ~50% increase in the number of CD41 positive cells
in the FLAG-RUNX1 TPA-treated cells compared to control (Figure 22). Similar
95
Figure 22. Overexpression of RUNX1 sensitizes cells towards megakaryopoiesis
K562 cells were co-transfected with a plasmid expressing GFP and either a FLAG-null (control) or RUNX1 overexpressing
plasmid and treated with 10nM TPA for 2 days. FACS analysis was performed in these cells to determine the number of
CD41 positive cells within the population of GFP positive cells. The results were plotted as the fold increase relative to FL-
null cells (y-axis). Four biological replicates for each sample were performed to generate standard deviation. Sims and Rice
2008
96
experiments were performed using the FLAG-null plasmid or the FLAG-PR-Set7 CD
plasmid that acts as a dominant negative by depleting cells of monomethylated
H4K20without reducing levels of PR-Set7. Similar to the results in HeLa cells, Western
analysis confirmed the specific decrease of monomethylated H4K20 and increased
RUNX1 in the FLAG-PR-Set7 CD cells compared to control cells (Figure 23A). These
cells were treated with a vehicle control or increasing amounts of TPA and FACs analysis
was performed as described above. As predicted, we observed a dose-dependent increase
in the number of CD41 positive cells in the FLAG-null cells with increasing amounts of
TPA (Figure 23C). In contrast, the FLAG-PR-Set7 CD cells displayed a >50% increase
in the number of CD41 positive cells compared to the FLAG-null control cells when
treated with only the vehicle (Figure 23B and 23C). Importantly, the number of CD41
positive cells did not increase with increasing amounts of TPA in the FLAG-PR-Set7 CD
cells strongly suggesting that they had already achieved the maximal differentiation
potential (Figure 23C). Collectively, these findings indicate that the absence of a
catalytically active PR-Set7 and monomethylated H4K20 results in the spontaneous
megakaryocytic differentiation of K562 cells.
Discussion and Future Directions
We report here, that monomethylated H4K20, monomethylated H3K9 and the
L3MBTL1 repressor protein converge at specific genomic regions in vivo and that they
97
Figure 23. Depletion of monomethylated H4K20 induces spontaneous
megakaryopoiesis.
A, K562 cells were co-transfected with a plasmid expressing GFP and either a FLAG-null
plasmid (control) or the FLAG-PR-Set7 R265G catalytically dead (CD) mutant that
depletes cells of monomethylated H4K20. Western blot analysis was performed in these
cells with the indicated antibodies. A general H3 antibody was used as a loading control.
B, FACs analysis was performed to examine levels of both CD41 (PE) and GFP in cells
co-transfected as described above. C, Quantitative analysis of FACs of cells treated with
increasing amounts of TPA to determine the number of CD41 positive cells within the
population of GFP positive cells. The results were plotted as the fold increase relative to
the vehicle treated FLAG-null cells (y-axis). Four biological replicates for each sample
were performed to generate standard deviation. Sims and Rice 2008
98
99
function cooperatively to repress transcription. These findings are consistent with the
results presented in Chapter 1, strongly suggesting that the interaction between
monomethylated H4K20 and L3MBTL1 is essential for the observed repressive effect as
the depletion of either result in the derepression of RUNX1 in different cell lines. It is
possible that L3MBTL1 is recruited to these regions by an interaction with PR-Set7,
however, consistent with our luciferase studies, we have been unable to detect such an
interaction by co-immunoprecipitation (Figure 15 and 24). While the expression of myc-
PR-Set7 let to an increase in PR-Set7 at the RUNX1 promoter, we consistently detected a
decrease in both monomethylated H4K20 and L3MBTL1 at this region, further
suggesting that PR-Set7 and L3MBTL1 do not interact in vivo (Figure 18). Importantly,
we demonstrate that the replacement of monomethylated H4K20 with trimethylated
H4K20 at the RUNX1 promoter upon overexpression of PR-Set7 WT results in the
displacement of the L3MBTL1 repressor protein and subsequent expression of the gene
(Figure 16 and 18). These findings are consistent with recent reports demonstrating that
L3MBTL1 only binds lower lysine methylated states and does not bind trimethylated
states (Kim et al. 2006; Li et al. 2007; Min et al. 2007). In addition, these results are
identical to experiments where the localization of the PR-Set7 CD mutant to the RUNX1
promoter leads to the loss of monomethylated H4K20 which results in the displacement
of L3MBTL1 and increased expression of RUNX1. (Figure 16) Collectively, these
findings indicate that the specific monomethylation of H4K20 by PR-Set7 is essential for
the localization of L3MBTL1 to the RUNX1 promoter to repress its transcription. The
100
residual presence of L3MBTL1 in these experiments could be due to the persistence of
low levels of monomethylated H4K20 at the RUNX1 promoter. However, recent in vitro
binding studies indicate that the MBT repeats of L3MBTL1 can also bind
monomethylated H3K9 and, since monomethylated H3K9 remains present at the RUNX1
promoter in these experiments, these observations suggest that this modification may play
a role in recruiting or stabilizing L3MBTL1 to this region (Li et al. 2007). Consistent
with this, in the PR-Set7 CD HeLa cells where monomethylated H4K20 is dramatically
reduced, we observed residual enrichment of monomethylated H3K9 and L3MBTL1 at
the RUNX1 promoter, again suggesting a possible in vivo interaction between them
(Figure 16C). However, the continued presence of monomethylated H3K9 at the RUNX1
promoter was not sufficient to repress RUNX1 transcription suggesting that this
modification does not play a direct role in gene repression (Figure 16A). These findings
also indirectly reinforce that the interaction between monomethylated H4K20 and
L3MBTL1 appears to be the critical step for repression.
PR-Set7 and monomethylated H4K20 were originally identified as being
associated with repressed chromatin and our findings here and those of another study
indicate that PR-Set7 and monomethylated H4K20 clearly participate in a gene
repression pathway by recruiting and binding the L3MBTL1 repressor protein in vivo
(Fang et al. 2002; Nishioka et al. 2002; Trojer et al. 2007). However, several recent
reports document an association of monomethylated H4K20 with actively transcribed
genes (Talasz et al. 2005; Vakoc et al. 2006; Barski et al. 2007). One possible
101
Figure 24. PR-Set7 does not interact with L3MBTL1
HEK 293 cells were transfected with HA-L3MBTL1 and DBD-PR-Set7. Twenty four
hours after transfection, protein extracts were collected using co-IP buffer (50mM Tris
pH 7.5, 150mM NaCl, 0.5mM DTT, 1% NP-40, 1 μg/mL pepstatin A, 1 μg/mL
leupeptin/aprotinin, 1mM PMSF) and immunoprecipitations were performed at 4°C using
α-HA-conjugated agarose beads. 5% of input (I) and 25% of bound fractions (B) were
examined by Western blot analysis using α-DBD antibody to determine binding of PR-
Set7 to L3MBTL1
These experiments were performed in the absence of the crosslinking reagent, BMH.
Sims and Rice 2008
102
explanation for these apparent differences is that this histone modification could
participate in both transcriptional activation and repression pathways depending on the
specific gene, similar to what has been observed for di- and trimethylated H3K9 (Vakoc
et al. 2005). This could be achieved by the recruitment of distinct H4K20 monomethyl-
binding regulatory proteins that could activate or repress transcription (Rice and Allis
2001). However, a more likely possibility is that the underlying function of this pathway
is to fine-tune the dosage of the corresponding gene rather than completely ablating its
transcription. This theory is consistent with both observations: that monomethylated
H4K20 is found within several active genes and that the depletion of this modification
consistently results in the increased expression of these genes (Houston and Rice, data
not shown).
An additional piece of evidence supporting this hypothesis stems from
examination of RUNX1 expression in the absence of PR-Set7 compared to Riz1/PRDM2.
In cells depleted of PR-Set7, monomethylated H4K20 is decreased within the RUNX1
promoter, leading to an increase in expression (Figure 16A and 25A). However, in the
absence of Riz1/PRDM2, monomethylated H4K20 is enriched within the RUNX1
promoter, decreasing its expression, supporting the hypothesis that levels of H4K20
monomethylation are critical for fine-tuning expression (Figure 25A). Comparison
studies involving target genes, both actively transcribed and repressed, enriched for
H4K20 monomethylation will help to further untangle the role of monomethylated
H4K20 and PR-Set7 in gene regulation.
103
It has been postulated that the monomethylation of histone lysines is required
prior to the di- and trimethylation by different sets of enzymes (Pesavento et al. 2008).
However, our data strongly suggests that this is not the case, as the depletion of
monomethylated H3K9 did not visually alter changes in the di- or trimethylated state
(Figure 8A and 8B). Similarly, the depletion of monomethylated H4K20 did not visually
alter global levels of di- or trimethylated H4K20 (Figure 9A and 9B). While this analysis
was performed on a global scale, it is possible that at genes, the methylation patterns may
be altered in the absence of monomethylation. Additionally, in a variety of
differentiating cell types, a global decrease in PR-Set7 and monomethylated H4K20 is
observed with a concomitant increase in H4K20 trimethylation, excluding the possibility
that monomethylation is serving as a substrate for higher degrees of methylated H4K20
(Biron et al. 2004, Figure 33). These results do not discount the hypothesis that the
preferred substrate for the HMTs is previously methylated histones, but rather that the
enzymes can achieve high degrees of methylation in the absence of lower levels of
methylation. This is consistent with our previous report (Rice et al. 2003) where the
depletion of mono- and dimethylated H3K9 does not alter trimethylation.
Surprisingly, this same relationship seems to exist between mono-and
trimethylated H4K20 in other cell types as overexpression of wild type PR-Set7 in HeLa
cells resulted in the loss of its enzymatic monomethyl-specificity and induced an increase
to the trimethylated form of H4K20 both globally and at target loci (Figure 18A and
18B). The increase in the degree of lysine methylation for several histone
104
methyltransferases is not an uncommon event when they are present in excess either in
vitro or in vivo. For example, in vitro studies examining the methyltransferase activity of
G9a, a mono-and dimethyltransferase in vivo, demonstrated that G9a can trimethylate
H3K9 if the reaction is allowed to proceed for longer periods of time or excess G9a is
used in the reaction (Tachibana et al. 2001). Furthermore, over-expression of G9a or the
G9a-like protein (GLP1) in cells results in a global shift of H3K9 from lower methylated
states to the trimethylated state (unpublished observations).
It is also interesting to note that the loss of monomethylated H4K20 and
L3MBTL1 is associated with a slight, but consistent, decrease in the levels of histone H3
at the RUNX1 promoter when compared to null (Figure 16C and 20A). These findings
suggest that in the absence of the components of this silencing pathway that there is a
decrease in nucleosomal occupancy consistent with previous observations that changes in
nucleosome occupancy are correlated with an increase in transcription (Barski et al.
2007). This is in contrast to results presented in figure 13, as there is a clear decrease in
histone H3 occupancy but an increase in monomethylated H4K20 and L3MBTL1 in the
absence of Riz1/PRDM2, suggesting that enrichment of H4K20 monomethylation and
L3MBTL1 even in an altered chromatin environment is sufficient for transcriptional
repression. Consistent with this hypothesis, in the absence of Riz1/PRDM2, RUNX1
expression decreases compared to mock transfected cells or to cells depleted of G9a
(Figure 25A). At the same time, there is an increase in the enrichment of both H4K20
monomethylation and L3MBTL1 within the RUNX1 promoter region; again reinforcing
105
Figure 25. RUNX1 expression is not dependent on Riz1/PRDM2
A, HeLa cells were treated with siRNA duplexes specific for Riz1/PRDM2, PR-Set7 or
G9a. Quantitative RT-PCR was performed to determine levels of Riz1/PRDM2 or
RUNX1 expression, normalized to GAPDH expression, and plotted as the fold increase
relative to mock (y-axis). It is important to note that Riz1/PRDM2 expression was not
measured in cells lacking PR-Set7 in this experiment. B, ChIPs were performed in HeLa
cells transfected with siRNA duplexes specific for lamin A/C or Riz1/PRDM2 using
either an H4K20 monomethyl-specific antibody, an H3K9 monomethyl-specific
antibody, an L3MBTL1 antibody, a general H3 antibody (positive control) or rabbit
preimmune serum (negative control). Increasing amounts of the final ChIPed material
(0.15%, 0.5% and 1.5%; black triangle) were used as the template in a thirty cycle PCR
amplification using primer sets specific to the RUNX1 promoter region. Input DNA
(0.005%, 0.0015% and 0.05%) served as the positive control for PCR.
106
107
the hypothesis that monomethylated H4K20 and L3MBTL1 are required for silencing
(Figure 25B). These data further suggest that the presence of H3K9 monomethylation
could regulate the amount of H4K20 monomethylation and L3MBTL1 that can be
enriched within specific genomic locations to prevent overcondensation of the chromatin
and therefore controlling proper gene dosage. One possible way to test this hypothesis is
to examine active genes that are enriched for both H4K20 and H3K9 monomethylation
and see if their expression changes in the absence of Riz1/PRDM2. If Riz1/PRDM2 and
H3K9 monomethylation are helping to regulate H4K20 monomethyl mediated
condensation and therefore silencing, an active gene could then be turned off in the
absence of Riz1/PRDM2 due to the increase in H4K20 monomethylation.
RUNX1 is an important transcription factor that controls the expression of a
variety of lineage-specific genes and, therefore, itself must be tightly regulated to prevent
premature differentiation (Ichikawa et al. 2004a). We have demonstrated in the
multipotent K562 cells that the newly described repressive trans-tail histone code
operates at RUNX1 to control its expression and, therefore, megakaryocytic
differentiation. We discovered that the loss of PR-Set7 and monomethylated H4K20 was
an early event in differentiation, occurring prior to DNA ploidy and cell morphology
changes, indicating that this repressive pathway participates in regulating
megakaryopoiesis (Figure 21C).
Contrary to our earlier observations, global levels of monomethylated H3K9 were
retained in the TPA-treated K562 cells despite the overall reduction of PR-Set7, although
108
monomethylated H3K9 was reduced at the RUNX1 promoter (Figure 21A). It is
interesting to note that hemin-treated cells also displayed reduced levels of PR-Set7 but
retained both monomethylated H4K20 and H3K9 (Figure 20C). How could this be
possible if PR-Set7 is the major H4K20 monomethyltransferase? These observations
imply that these histone modifications are relatively stable in a separate differentiation
pathway, even in the absence of PR-Set7. It also implies that specific reduction of
monomethylated H4K20 during megakaryopoeisis is a defining mechanistic feature of
this pathway when compared to erythropoiesis. It remains unclear how the selective
reduction of this histone modification is achieved, although it is tempting to speculate
that there is an unknown H4K20 demethylase that operates specifically in the precursor
cells to initiate megakaryocytic differentiation, in part, by activating RUNX1. A second
possibility is that L3MBTL1 is more stable during erythrocytic differentiation, protecting
H4K20 monomethyltion from removal during cell cycle progression or by an unknown
H4K20 demethylase. Indeed, ChIP analysis of the RUNX1 promoter in hemin-treated
cells demonstrates an increase in enrichment of L3MBTL1 as compared to vehicle-
treated cells (Figure 21B). This increase could be sufficient to keep the chromatin
condensed within the RUNX1 promoter, protecting the methyl modifications within this
region, allowing for selective retention of the modifications even in the absence of their
respective HMTs.
We demonstrated that the ectopic expression of RUNX1 sensitized K562 cells to
the differentiation effects of TPA but itself was not sufficient to induce megakaryopoiesis
109
(Figure 22). Surprisingly, in cells lacking monomethylated H4K20, TPA was not
required to induce megakaryopoiesis; the cells tended to spontaneously differentiate
(Figure 23B and 23C). While RUNX1 is likely to be an important component of this
effect, we hypothesize that other unidentified genes that contribute to megakaryopoeisis
are also regulated by this repressive pathway and that the lack of monomethylated
H4K20 results in their activation culminating in megakaryocytic differentiation. It is
important to note that the differentiation-associated changes in monomethylated H4K20
is not restricted to hematopoiesis. A previous report demonstrated that global levels of
monomethylated H4K20 were highest in mouse neuroblasts and myoblasts and that these
levels were significantly reduced during their differentiation, similar to what we observed
during megakaryopoiesis (Biron et al. 2004). Collectively, these findings strongly
suggest that this repressive pathway plays a role in preserving a multipotent phenotype by
repressing the expression of certain lineage-specific genes. The identification of the
genes regulated by this repressive pathway will provide critical insights into this complex
biological problem.
110
Chapter 3. The Role of Histone Methylation in Stem Cell Differentiation
Our studies thus far have defined a role for H4K20 monomethylation in
differentiation of multipotent progenitor cells (BFU-E/MK) to committed precursors
(erythrotic lineage and megakaryocytic lineage). Recent reports have demonstrated a role
for histone methylation in alterations of transcriptional processes during the transition
from pluripotency to multipotent progenitor cells (Guenther et al. 2007; Mikkelsen et al.
2007). Indeed, many studies comparing the global histone methylation patterns within
gene promoters of embryonic stem cells and fully differentiated adult cells have shown
that there is a clear shift between active and repressive histone modifications at genes
critical for developmental regulation and that these modifications correlate with changes
in gene expression. Based on our studies examining early megakaryocytic
differentiation, we predict that these changes in histone modifications will occur quickly
to accommodate rapid changes in transcription that are critical for defining cellular
identity. To test this hypothesis, we will use mouse and human embryonic stem cells,
which will allow us to map changes in histone modifications during the first stages of
development.
Global changes in histone methylation accompany mouse embryonic stem cell
differentiation progression
In pilot studies performed in collaboration with Robin Wesselschmidt of the Stem
Cell Core Facility at USC, we utilized an established cell culture system to induce
111
neuronal cell differentiation (Bain et al., 1994). The 129/S6/SvEv murine ES (mES) cell
line was induced to differentiate through removal from feeder layer support, and
subsequent culture on nonadherent substrate. At day four (d4) embryoid bodies (EBs)
were split and maintained for an additional eight days in either the presence or absence of
all-trans retinoic acid (RA), an inducer of neuronal cell differentiation. Additionally, one
group of cells were grown for 4 more days after removal of RA (d8+) to examine further
differentiation. The cells that were maintained in the absence of RA differentiated into a
mixed population of cardiomycytes and various stages of dermal cells. Cells were
harvested at various intervals along the differentiation time course and examined for
changes in various histone modifications (Figure 26).
We chose to analyze the methylation patterns of H4K20, H3K9 and H3K27 as
work from others has demonstrated an involvement for these modifications in
differentiation (Mikkelsen et al. 2007; Guenther et al. 2007; Bernstein et al. 2007). As a
control for both terminally differentiated cells as well as to control for contamination of a
feeder layer, we included protein extracts collected from ionized mouse embryonic
fibroblasts (MEF feeder layer). Based on previous studies, we anticipated changes in
methylation patterns throughout differentiation, especially in levels of mono-and
trimethylated H4K20 and trimethylated H3K27 (Mikkelsen et al. 2007; Biron et al.
2004). We observed a global decrease in levels of monomethylated H4K20 and a
concomitant increase in trimethylated H4K20. This is consistent with
immunofluorescence studies conducted in developing mouse embryos in which levels of
112
Figure 26. Changes in global patterns of histone methylation during mES
differentiation
Undifferentiated mES cells were induced to differentiate into embryoid bodies (EB d2
and d4) through removal from feeder layer support. After four days, the EBs were plated
and maintained in the presence or absence of RA (-RA, +RA) for up to 8 days. An
additional group of cells were grown in the presence or absence of RA for 4 more days
(d8+ EB-RA, d8+ EB+RA). Core histones were extracted from the cells at the indicated
timepoints and Western analysis was performed using the indicated antibodies.
Coomassie stain of core histones was used as a loading control.
113
114
H4K20 trimethylation were increased as cells became terminally differentiated within the
neuronal tube (Biron et al. 2004). Interestingly, we were unable to detect any global
changes in methylated H3K9. This is contrary to what we would have predicted but
similar to results seen in differentiated K562 cells in which global levels of
monomethylated H3K9 were constant in the presence of decreased monomethylated
H4K20 (Figure 20C). These data suggest that during differentiation, the two
modifications may serve different functions or that H3K9 monomethylation could serve
as a “marker” for changes in H4K20 monomethylation. Surprisingly, levels of H3K27
trimethylation were largely undetectable in mES cells throughout the entire
differentiation time course. The small amounts detected were only visualized after
incubation with >5 times the normal amount of antibody used. Overall, there was also no
change between cells treated with or without RA, suggesting that the lineage the cells
differentiate towards is not the critical step for changes in histone methylation but rather
the differentiation process as a whole.
Isolation and standardization of ChIP assays in pluripotent hES cells
Our preliminary studies in the mES differentiation culture model demonstrated
that global changes in histone methylation patterns could be detected during
differentiation. Therefore, we predicted that we would be able to detect alterations in
histone methylation when examining specific genomic regions involved in regulating
115
development. To study this further, we moved from well-characterized mES cell lines to
human embryonic stem (hES) cells.
It has recently been reported that the majority of the work performed in hES cells
has not taken into consideration that the “pluripotent” starting material consists of many
different cell types due to spontaneous differentiation under culture conditions (Hu et al.
1997; Niwa et al. 2005; Smith, A 2005; Laslett et al. 2007). The traditional expressed
cell surface markers and transcription factors such as Oct3/4 and SSEA-3, used to define
pluripotency can also describe cells that have begun early stages of commitment,
resulting in studies that have not taken into consideration this heterogeneous population
when describing their results (Guenther et al. 2007; Laslett et al. 2007). Furthermore,
recent data suggests that lineage choices (i.e. changes in transcriptional regulation of gene
expression) are made prior to changes in stem cell markers, highlighting the importance
of isolating a pure undifferentiated population and comparing it to cells undergoing early
stages of commitment (Laslett et al. 2007). The pluripotent specific cell surface markers,
GCTM2 and TG30 (CD9), have been used to purify the most pluripotent population of
cells from cells that have begun commitment. The expression patterns of these markers
correlate well with expression levels of the well-established transcription factor involved
in pluripotency maintenance, Oct3/4, allowing for confident sorting of hES cells (Laslett
et al. 2007). We double stained the HES2 hES cell line with GCTM2 and TG30 and
were able to successfully sort out GCTM2
high
/TG30
high
(pluripotent, Figure 27 R6
116
Figure 27. FACs analysis of HES2 hES cell line using GCTM2 and TG30 as
markers of pluripotency
HES2 cells were maintained in culture conditions suitable for retaining an
undifferentiated state were collected and incubated with α-TG30 alone, α-GCTM2 alone,
a combination of α-TG30 and α-GCTM2 or no primary antibody as a negative control.
After incubation with the proper fluorescently labeled secondary antibodies, the cells
were sorted using the MoFlo cell sorter and collected in hES medium. Quadrant R6 and
R7 in the double positive cells were collected for further experiments.
117
118
quadrant of double stained cells) cells and also collected GCTM2
low
/TG30
low
(committed,
Figure 27, R7 quadrant of double stained cells) cells.
Upon collecting the two populations of cells, we realized that the starting material
obtained was very small and was not sufficient for the ChIP protocol that we had
previously used. Therefore it was necessary to take a step back and alter our protocol to
accommodate a small sample amount. To examine the smallest number of cells that we
could use reliably for ChIP assays, we used dilutions of HeLa ChIP lysate to perform
immunopreciptiations with an H3 general antibody and a non-specific rabbit serum
(preimmune) as a negative control, followed by DNA extraction and PCR of the RUNX1
promoter, which has been previously characterized (see Chapter 2). We determined that
we could confidently reproduce ChIP assays using as few as 30,000-50,000
cells/immunoprecipitation compared to 5,000,000 cells/immunoprecipitation commonly
used for most protocols (Figure 28 and 29). This two-fold decrease in cell number allows
the amount of material collected from FACs sorting to be sufficient for ChIP assays.
Changes in H3K36 trimethylation mirror changes in gene expression in hES cells
undergoing commitment
To begin our genome wide studies using the sorted hES cells, we decided to
investigate the well-established modification, trimethylated H3K36, which is associated
with gene activation and spans the body of actively transcribed genes (Bannister et al.
119
Figure 28. Standardization of lowest number of cells used for ChIP
HeLa cells were collected and counted for cell number before resuspending cells in ChIP
lysis buffer. ChIPs were performed using an H3 general antibody or a nonspecific rabbit
serum (preimmune) as a negative control. 1.5% and 5% of bound material was used as
template for a thirty cycle PCR reaction. 0.05% and 0.15% of input was used as template
for a positive control for PCR amplification.
120
2005; Edmunds et al. 2008; Mikkelsen et al. 2007). We predicted that H3K36
trimethylation would be a good modification to use to predict genes that were up- or
downregulated during differentiation as it is relatively easy to detect because of the large
abundance of the modification within genes and therefore easy to monitor. As a pilot
study to confirm that our techniques would be suitable for detecting H3K36
trimethylation, we decided to perform ChIP assays using 30,000 sorted cells/IP from both
GCTM2
high
/TG30
high
(diploid +/+)
and
GCTM2
low
/TG30
low
(diploid -/-) cells. The
specificity of the ChIPs was confirmed by examining trimethylated H3K36 enrichment
across the body of the constuatively active GAPDH gene. Using primers specific for the
promoter and 3kb downstream region, we predicted that trimethylated H3K36 would be
enriched specifically within the 3kb downstream region and would be absent from the
promoter region in both the GCTM2
high
/TG30
high
and
GCTM2
low
/TG30
low
cells (Bannister
et al. 2005). As shown in figure 29B, we could detect specific enrichment of
trimethylated H3K36 within the 3kb downstream region of GAPDH in both
GCTM2
high
/TG30
high
(diploid +/+)
and
GCTM2
low
/TG30
low
(diploid -/-) cells that was
absent in the promoter region; genomic DNA (gDNA) was used as a positive control for
PCR.
To further examine the changes in enrichment in trimethylated H3K36 across
genes, we applied the ChIP’d DNA to a NimbleGen ChIP-chip microarray spanning
chromosomes 8-14. These regions of the genome were chosen based on the number of
positive and negative controls for pluripotent gene expression that we could find within a
121
Figure 29. Standardization of H3K36 trimethyl ChIPs in HES2 hES cells
A, 10ng of input or bound ChIP material was used as input for WGA2 amplification. The
resultant amplified DNA was quantified and 10ng of the amplified DNA was used as
input for a second round of WGA2 amplification. 6% of this DNA was run on a 1%
agarose gel to identify any PCR bias and confirm the integrity of the DNA samples. The
reactions were performed in quadruplicate as shown; amplified genomic DNA (gDNA)
was run as a positive control. B, 0.3% of H3K36 trimethyl enriched, amplified DNA and
0.4% of amplified input DNA was used as template for a thirty-five cycle PCR reaction;
genomic DNA (gDNA) was used as a positive control for PCR. The input sample for the
diploid -/- promoter lane did not PCR. Subsequent PCRs showed that this sample could
be PCR’d but this is not depicted in this figure.
122
123
Expressed in
GCTM2
high
/TG30
high
cells
Gene Name Accession Number Chromosome Reference
TERF1 NM_017489 8 Laslett, A.L. et al. 2007
CALB1 NM_004929 8 Laslett, A.L. et al. 2007
BNC2 AK001099 9 Laslett, A.L. et al. 2007
HELLS AK021443 10 Laslett, A.L. et al. 2007
ADM NM_001124 11 Laslett, A.L. et al. 2007
Nanog NM_024865 12 O'Neill, L.P. et al. 2006
Expressed in
GCTM2
low
/TG30
low
cells
Gene Name Accession Number Chromosome Reference
Slit-1 NM_003061 10 Laslett, A.L. et al. 2007
Pax-2 NM_003988 10 Laslett, A.L. et al. 2007
Pax-6 NM_001604 11 Laslett, A.L. et al. 2007
CDX2 NM_001265 13 O'Neill, L.P. et al. 2006
FoxA1 NM_004496 14 Laslett, A.L. et al. 2007
Table 2. Gene expression patterns in sorted HES2 hES cells
List of genes that are either up or downregulated cells either expressing GCTM2
high
/TG30
high
or
GCTM2
low
/TG30
low
that can be
used as positive and negative controls for ChIP-chip.
124
single microarray chip (Table 2). We predicted that the genes that were highly expressed
in the GCTM2
high
/TG30
high
cells would have preferential enrichment of H3K36
trimethylation within the body of the gene and there would be a reduction in enrichment
of trimethylated H3K36 for these same genes in the GCTM2
low
/TG30
low
cells and that the
opposite would be true for genes expressed in the GCTM2
low
/TG30
low
cells. As shown in
figure 30, we detected enrichment of H3K36 trimethylation within the TERF1 gene in
GCTM2
high
/TG30
high
cells (HES2+/+ H3K36me3 ChIP NimbleGen). We were unable to
detect peaks of trimethylated H3K36 within the nearby non-pluripotent genes KCNB2
and RPESP, demonstrating the specificity of the results. As yet, we have not analyzed
the differences in trimethylated H3K36 enrichment between GCTM2
high
/TG30
high
and
GCTM2
low
/TG30
low
cells.
These experiments served as a proof of principle that our techniques would be
viable and that we could obtain clear results from the experiments. We were also able to
confirm that amplification of the ChIP’d DNA for NimbleGen ChIP-chip was a viable
option as we were unable to detect a PCR bias in any of our samples (Figure 29A).
Recent genome wide studies of histone modifications have begun to use a new,
more high-throughput technique, ChIP sequencing (ChIP-seq), which allows for direct
sequencing of ChIP’d DNA (Barski et al. 2007; Mikkelsen et al. 2007). This technique
eliminates the need for many rounds of PCR amplification to obtain enough DNA for
microarray analysis, potentially removing a bias from the experiment, as well as
125
Figure 30. Enrichment of H3K36 trimethylation is found at active genes in
GCTM2
high
/TG30
high
cells
ChIPs were performed on 30,000 HES2 GCTM2/TC30 double positive (+/+) cells using
an H3K36 trimethyl specific antibody. Part of the ChIPed material was used on a
NimbleGen tiling array containing a portion of chromosome 7, plotted as the log2 ratio of
H3K36me3 – input DNA control (ChIP-NimbleGen). Another part of the ChIPed
material was used to generate a DNA library for cluster formation and sequencing using
the Illumina Genome Analyzer, as described, resulting in ~7.5 x 10^6 sequences uniquely
aligned to the human genome (ChIP-Seq). Two negative controls were also used: HES2
DNA isolated after crosslinking and sonication (ChIP input) and random sequencing of
HES2 genomic DNA (shotgun). All data was aligned on the UCSC genome browser at
the indicated region of chromosome 8.
126
127
providing genome wide analysis at a lower cost than applying the ChIP’d DNA to many
different microarray chips. ChIP-seq also can provide more analysis of repetitive
elements compared to ChIP-chip, allowing for the examination of more methylated lysine
residues that may be enriched in repetitive sequences (Barski et al. 2007).
ChIP sequencing provides a viable option for mapping histone modifications within
stem cells
As a pilot study for ChIP-seq in collaboration with Tanya Spektor, we used
ChIP’d DNA enriched for either monomethylated H4K20 or H3K9 from HeLa cells.
These experiments would allow us to confirm known targets of the trans-tail histone code
by comparing the ChIP-seq data to previously characterized genes as well as to identify
additional targets that will provide a more detailed analysis of the biological role of
H4K20 and H3K9 monomethylation. Triplicate biological replicates were performed and
enriched DNA was sequenced by Illumina/Solexa. As shown in figure 31, we confirmed
previously described targets of H4K20 monomethylation, PTTG1P and ADARB1
(Houston and Rice, unpublished data) as well as identify novel targets, SUMO3 and
POFUT2. These results suggest that ChIP-seq is a more sensitive method to define gene
targets of H4K20 monomethylation compared to ChIP-chip.
In a separate study, we sequenced the ChIP’d DNA from sorted stem cells enriched for
H3K36 trimethylation to cross-validate the NimbleGen results. As shown in figure 30,
we identified a large region enriched for H3K36 trimethylation within the
128
Figure 31. H4K20 monomethyl targets can be identified by ChIP seq
Three independent ChIPs were performed in HeLa cells using an H4K20 monomethyl
specific antibody, as described. The quality and quantity of the ChIPed DNA was
determined before generating the DNA library for cluster formation and sequencing using
the Illumina Genome Analyzer, as described. Each replicate produced ~14 x 10^6
sequences of ~32 bp in length; ~7.5 x 10^6 (52%) per replicate was uniquely aligned to
the human genome (ChIP-Seq). The ChIP-Seq data was compared to previous ChIP-chip
data from three independent biological replicates of H4K20me1 in HeLa cells using
NimbleGen whole genome tiling arrays, plotted as the log2 ratios of H4K20me1 – input
DNA control (ChIP-NimbleGen). All data was aligned on the UCSC genome browser at
the indicated region of chromosome 21.
129
130
TERF1 gene, but absent from the KCNB2 and RPESP similar to the results of the
NimbleGen ChIP-chip (HES2 +/+ H3K36me3 ChIP-seq). These results suggest that
ChIP-seq can be used to confidently map methylation patterns across the genome using
small amounts of cells and that NimbleGen ChIP-chip can be used to validate the results.
Global changes in specific histone modifications in pluripotent versus committed
HES2 cells
Thus far, we have standardized experiments to sort hES cells as well as
developed methods to analyze histone methylation patterns on a global scale using both
ChIP-chip and ChIP-seq techniques from these cells. We also have examined enrichment
of H3K36 trimethylation within pluripotent genes and shown that our techniques are
working properly. We next asked if after sorting the hES cells we could detect global
changes in histone methylation patterns by Western analysis. This would allow us to
predict what modifications would be most interesting for further study using ChIP-seq.
The hES cell line, HES2, that we have used thus far is difficult to grow and therefore it is
difficult to obtain enough material for Western analysis. To circumvent this problem, we
used a variant HES2 cell line (HES2V), which has a growth advantage in culture but
retains the majority of the pluripotent properties of HES2 cells. Karyotypes of the
HES2V cells indicate that the major change in the chromosomal structure is a duplication
of chromosome 1 as well as the presence of a “marker chromosome A” that was not fully
characterized (Figure 32A). We also examined the staining patterns and FACs analysis
131
of the HES2V cells and found a similar number of cells were present in the
GCTM2
high
/TG30
high
and GCTM2
low
/TG30
low
cells (Figure 32B). Therefore, we
determined that these cells were suitable for Western analysis but not for ChIP-seq
experiments.
Surprisingly, we detected dramatic changes in global levels of various histone
methylation patterns when comparing the GCTM2
high
/TG30
high
and GCTM2
low
/TG30
low
cells (Figure 33). We observed a substantial decrease in monomethylated H4K20,
trimethylated H3K4 and trimethylated H3K36 and a concomitant increase in
trimethylated H3K9 and H4K20 in GCTM2
low
/TG30
low
cells compared to
GCTM2
high
/TG30
high
cells; no detectable changes were observed in dimethylated H3K9
and H4K20 or monomethylated H3K9 and H3K27 levels. The changes in H4K20 mono-
and trimethylation were similar to those observed in mES differentiation experiments
(Figure 26) suggesting that the changes in these two histone modifications could play a
significant role in the differentiation process across species. Much like the mES cells, we
were also unable to detect the presence of H3K27 trimethylation on a global scale (data
not shown). This was unexpected, as we predicted that trimethylated H3K27 would not
only be detected but most likely would not change on a global scale due to recent reports
indicating that H3K27 trimethylation has a critical role in controlling gene expression of
pluripotent and lineage specific genes during differentiation (Bernstein et al. 2006;
Mikkelsen et al. 2007; Guenther et al. 2007).
132
Figure 32. Characterization of HES2V cell line
A, Karyotype of HES2V cell line indicates that all chromosomes appear cytologically
normal except chromosome 1, which contains a duplication (black arrows). The cells
also contain an extra “marker chromsome”, A, which has not been further characterized.
B, HES2V cells were maintained in culture conditions suitable for retaining an
undifferentiated state were collected and incubated with α-TG30 alone, α-GCTM2 alone,
a combination of α-TG30 and α-GCTM2 or no primary antibody as a negative control.
After incubation with the proper fluorescently labeled secondary antibodies, the cells
were sorted using the MoFlo cell sorter and collected in hES medium. Quadrant R6 and
R7 in the double positive cells were collected for further experiments. The results
indicate that the cells contain a similar percentage of GCTM2
high
/TG30
high
and
GCTM2
low
/TG30
low
cells as compared to HES2 cells.
133
134
Discussion and Future Directions
Here, we have described several methods for studying changes in histone
methylation patterns throughout embryonic stem cell differentiation. We determined the
lowest number of cells that can be reliably used for ChIP analysis after sorting and that
can be used for both ChIP-chip and ChIP-seq with reliable results. We have identified
two histone modifications, H4K20 mono-and trimethylation, that have opposite patterns
of global enrichment during all stages of differentiation in both mES and hES cell lines
suggesting that the regulation of these modifications during differentiation could be
important for setting up lineage specification.
Upon standardization of ChIP-seq, it will be important to continue to examine
various histone modifications in both pluripotent (GCTM2
high
/TG30
high
) as well as
committed cells (GCTM2
low
/TG30
low
). An important consideration in comparison studies
however, is that the committed cells will be a mixed population as there is no control for
the specific lineage that the cell are beginning to commit towards. To combat this
problem, the cells could be further sorted using a third marker that specifies a specific
lineage such as Pax-6, which will identify cells committed towards the neuronal pathway,
a common pathway that hES cells tend to spontaneously differentiated into in culture
(Reubinoff et al. 2001). An alternative approach could be to compare the
GCTM2
high
/TG30
high
cells to hES cells that have been chemically differentiated or to
fully differentiated cells to ensure that the committed cells are more of a homogeneous
population.
135
Figure 33. Global patterns of histone methylation change upon commitment
HES2V cells were sorted into GCTM2
high
/TG30
high
(+/+) and GCTM2
low
/TG30
low
(-/-)
cells. Western analysis was performed on nuclear extracts using the indicated antibodies.
An H4 general antibody was used as a loading control.
136
137
Once gene targets are identified, their expression patterns can be examined using
expression mircroarray technology to correlate changes in histone methylation patterns
with changes in gene expression. Using this combined approach, we will be able to
identify known and new target genes involved in maintaining pluripotency as well as
genes that are critical for transitioning to a committed lineage.
Our preliminary data indicates that we will be able to map the dynamic changes
that occur during commitment. Based on Western analysis, we predict that H4K20
mono-and trimethylation will show a dramatic shift in their abundance and perhaps
localization within the genome. It will be very interesting to compare the gene targets
between mono-and trimethylated H4K20 to determine if the genes that are decreasing in
monomethylated H4K20 are gaining H4K20 trimethylation as well as examining their
expression profile for changes in gene expression. This would provide direct evidence of
a switch between a mono-and trimethylated state that plays an important role in
development (Biron et al. 2004).
The dramatic changes that we observed by Western analysis between the two
populations of cells was surprising because we predicted that a global change would not
necessarily be detected as the abundance of the modification would remain the same but
the position of the modification within the genome would be altered. This suggests that
there might be further regulation of the HMTs responsible for methylation during
differentiation. This can be examined either by expression or Western analysis of the
enzymes. Another possibility is that as the hES cells are differentiating, the chromatin
138
structure within the nucleus is also changing, making the substrates of the HMTs less or
more available, altering the global levels of methylation patterns. This hypothesis can be
tested by examining changes in chromatin accessibility within target genes identified by
ChIP-seq experiments.
Taken together, these studies will shed light on the growing field of stem cell
differentiation and help to create maps of changes in histone modifications throughout
development. This will allow for further study of the biological relavence of histone
methylation in differentiation and development.
139
Chapter 4. Methods
Immunofluorescence Studies
Chromatin fibers from mammalian tissue culture cells (adapted from Susan Forsburg’s
protocol) (See Table 1 for antibodies and dilutions; make all solutions fresh)
1. Trypsinize cells, wash in PBS and resuspend in NIB buffer. Spin down the cells
600xg 7 minutes at 4 deg. Wash nuclear pellet twice with PBS+1mM PMSF.
2. Resuspend nuclear pellet in chromatin lysis buffer and pipet lines of chromatin
down supercharged slides. Let dry for ~5 minutes. **I usually draw two lines
down the width of the slide using a liquid blocker marker (PAP PEN) before
streaking the chromatin onto the slides. This prevents the antibody from moving
beyond where the chromatin fibers should be**
3. Immerse slides in chromatin lysis buffer in a Coplin jar for 10 minutes.
4. Slowly lift the slides straight out of the jar and immerse in PBS for ~2 minutes
5. Immerse slides in 4% formaldehyde/PBS for 20 minutes
6. Slowly lift the slides straight out of the jar and immerse in PBS for ~10 minutes
7. Immerse slides in blocking solution for 1 hour at room temperature
8. Slowly lift the slides straight out of the jar and blot the ends of the slides to
remove excess moisture. Add 100uL primary antibody, diluted in blocking
solution, to the slides and incubate overnight at 4 deg.
9. Immerse slides in wash buffer 3x 5 minutes at room temperature
140
10. Add 100uL secondary antibody, diluted in blocking solution, to the slides and
incubate at room temperature for 1 hour. Make sure to cover the slides with foil
to protect from the light.
11. Immerse slides in wash buffer 3x 5 minutes at room temperature
12. Add mounting medium to the slides and a coverslip.
NIB Buffer Chromatin Lysis Buffer
150mM NaCl 25mM Tris, pH 7.5
10mM HEPES, pH 7.4 0.5M NaCl
1.5mM MgCl
2
1% Triton X-100
10mM KCl 0.5M Urea (diluted from 1M
0.5% NP-40 urea made up immediately
0.5mM DTT prior to use)
1mM PMSF
1ug/mL pepstatin
1ug/mL aprotinin/leupeptin
Blocking Solution (in PBS) Wash Solution
1% BSA 0.05% Tween 20 in PBS
0.5% Triton X-100
0.02% NaN
3
141
**For more detecting more loosely associated proteins, use Protein Lysis Buffer (200mM
Tris, pH 7.5, 50mM EDTA, and 0.5% SDS) instead of chromatin lysis buffer.
Immunofluorescence of whole cells plated on coverslips (See Table 1 for antibodies and
dilutions; make all solutions fresh)
1. Sterilize coverslips by rinsing in 100% EtOH and quickly passing through a flame
to dry. Place sterilized coverslips into 6 well plates.
2. Plate cells onto sterilized coverslips to a density of 10
5
cells/mL (2mL/well) in a 6
well plate.
3. Fix cells 10 minutes RT in 1% paraformaldehyde/PBS when cells reach ~60-80%
confluency. (Make 1% paraformaldehyde by mixing with PBS and heating to
60
0
C and adding 10M NaOH dropwise until paraformaldehyde goes into
solution.)
4. Wash cells 3x10 minutes PBS
5. Permeabilize cells 5 minutes RT with 0.2% Triton X-100/PBS
6. Remove permeabilization solution and wash cells 3x5 minutes PBS
7. Block cells 1 hour RT in 5% donkey serum/PBS
8. Remove blocking solution and add primary antibody (125 μL/coverslip, diluted in
5% donkey serum/PBS). Try to move coverslip to the center of the dish to
prevent the antibody from wicking off of the coverslip. Incubate the primary
142
antibodies at 37
0
C for 1 hour in a humidified chamber (Tupperware + damp paper
towel lining the edges).
9. Add 2mL 5% donkey serum/PBS and wash cells, repeat 2 more times.
10. Remove serum and add secondary antibodies (125 μL/coverslip, diluted in 5%
donkey serum/PBS). Make sure that the coverslips are in the center of the dish
and incubate at 37
0
C for 1 hour in a humidified chamber (Tupperware + damp
paper towel lining the edges).
11. Wash cells 3x10 minutes in 5% donkey serum/PBS (Incubate in the dark to
prevent decreasing fluorescence strength).
12. Mount coverslips onto slides using mounting medium containing DAPI
(VectaShield) by adding 1 drop of mounting medium to the slide and rinsing
coverslips briefly in dH
2
O before placing cell side down onto mounting medium.
Blot off the edges of the slides by tilting onto kimwipes.
13. If saving the slides, make sure to seal the edges with fingernail polish to prevent
the slides from drying out. Store slides at 4
0
C in the dark.
**For dual staining cells (also true for fibers): if the different primary antibodies
being used are from different species (ex. mouse and rabbit) it is fine to mix them
together and add. The same is true for different species of secondary antibodies.
However, if the primary antibodies are from the same species (ex. rabbit and rabbit) it
is critical to incubate them separately by adding one primary, washing and adding the
143
secondary, washing and then adding the second primary antibody, washing and
adding the second secondary, washing and then mounting. It is also critical to do a
control well where the 1
st
primary antibody is added, followed by its secondary and
then the 2
nd
secondary antibody is added to make sure that only one color is picking
up the 1
st
primary antibody. Finally, it is important to use the two primary antibodies
in both orders (ex. H4K20 mono, FITC, H3K9 mono, Cy3 and also H3K9 mono,
FITC, H4K20 mono, Cy3) to make sure that the patterns are the same.**
Immunofluorescence of stem cells (protocol from Robin Wesselschmidt) (See Table 1 for
antibodies and dilutions; make all solutions fresh)
**These IFs were done on chamber slides instead of coverslips.**
1. Remove media from the cells and dry them 1hr at RT
2. Fix cells 15 minutes at RT in 4% paraformaldehyde/PBS (500 μL/well)
3. Wash cells 2x500 μL PBS (-Ca, -Mg) **The wells can now be filled with 600 μL
PBS and the slides can be stored at 4
0
C until ready to be stained.
4. Permeabilize cells 10 minutes 0.1% Triton X-100/PBS (500 μL/well) at RT
5. Wash cells 2x500 μL PBS (-Ca, -Mg)
6. Block cells 1.5h 6% goat serum/PBS (500 μL/well) at RT
7. Add 100 μL primary antibody/well (dilute in 6% goat serum/PBS). Incubate 1.5h
RT
8. Wash 3x500 μL PBS (-Ca, -Mg)
144
9. Add 100 μL secondary antibody/well (dilute in 6% goat serum/PBS) and incubate
1.5h RT
10. Wash 3x500 μL PBS (-Ca, -Mg)
11. Add 1 drop of mounting medium and a coverslip. Store at 4
0
C in the dark until
ready to image. It’s best to seal the coverslip with nail polish to prevent the cells
from drying out.
145
Antibody Species Whole Cell
Dilutions
Chromatin Fiber
Dilutions
H4K20 monomethyl
(linear)
Rabbit 1:1000 1:100
H4K20 dimethyl
(branched)
Rabbit 1:500 1:100
H4K20 trimethyl
(branched)
Rabbit 1:1000 1:100
H3K9 monomethyl Rabbit 1:2000 1:200
H3K9 dimethyl Rabbit 1:2000 1:200
H3K9 trimethyl Rabbit 1:1000 1:100
RNA pol II
(covance)
Mouse 1:1000
H3K27 monomethyl Rabbit 1:3000
H3K27 dimethyl Rabbit 1:3000
H3K27 trimethyl Rabbit 1:500
GCTM2 mouse IgM no dilution
TG30 mouse IgG2a no dilution
Oct3/4 (santa cruz) mouse IgG2b 1:200
Pax6 (santa cruz) rabbit 1:100
FITC secondary Rabbit or Mouse 1:200 1:200
Cy3 secondary Rabbit or Mouse 1:150 1:150
Alexa Fluor 488
(invitrogen)
mouse IgG2a
and mouse IgM
1:200
Alexa Fluor 568
(invitrogen)
mouse IgG2b
and rabbit
1:200
Table 3. Antibodies used for immunofluorescence experiments
Antibodies used for immunofluorescence studies, species of each antibody and the
dilutions used for either whole cell immunofluorescence or chromatin fibers.
146
Chromatin Immunoprecipitation Studies
Chromatin Immunoprecipitations using Chelex (See Table 5 for antibodies and dilutions;
Protocol adapted from Nelson et al. 2006)
ChIP Buffers: To all buffers (except Chelex buffer) add 1mg/mL aprotinin/leupeptin,
1mg/mL pepstatin, 1mM PMSF before using.
Cell Lysis Buffer (**I have found that this buffer works best if made fresh**)
5mM PIPES pH 8
85mM KCl
0.5% NP-40
Nuclear Lysis Buffer
50mM Tris pH 8
10mM EDTA
1% SDS
ChIP Dilution Buffer
150mM NaCl
16.7mM Tris pH 7.5
3.3mM EDTA
1% Triton X-100
0.1% SDS
147
0.5% Na-Doc
RIPA/LiCL Buffer
50mM HEPES
1mM EDTA
1% NP-40
0.7% Na-Doc
0.5M LiCl
TE Buffer
10mM Tris pH 7.5
1mM EDTA
10% Chelex Buffer
1g Chelex (Bio-Rad)
10mL dH
2
O
**chelex will NOT stay in solution and must be vortexed before each use**
148
Cell line Pulse length Output
HeLa 1x20 sec 4.0
293T/TK22 1x20 sec 4.0
K562 1x15 sec 4.0
Dami 1x10 sec 4.0
CHRF288-11 1x20 sec 4.0
HES2 or variant cells (less
than 10
5
cells/500 μL)
1x15 sec 4.0
HeLa or 293T (less than 10
7
cells/mL)
1x15 sec 4.0
Table 4. Cell Line Sonication Conditions
Standardized sonication conditions for each cell type used for chromatin
immunoprecipitations. All sonications were performed at 4
0
C.
149
Antibody Amount Used
H3 general (ChIP grade, Abcam) 4 μL
H4K20 monomethyl (BRANCHED) 4 μL
H4K20 trimethyl (BRANCHED) 13 μL
H3K9 monomethyl (BRANCHED) 9 μL
HA (mouse monoclonal, Roche) 1 μg
L3MBTL1 (Rice Lab or LPBio) 20 μL
myc (mouse monoclonal, Roche) 2 μg
H3K36 trimethyl (Abcam) 4 μL
Table 5. Antibodies used for chromatin immunoprecipitiations
The antibody and amounts used for chromatin immunopreciption. All antibodies are
rabbit unless specified.
150
Preparation and formaldehyde fixation of cells:
Adherent cell lines:
1. Remove media from confluent 15 cm plates and wash cells with PBS.
2. Trypsinize cells and count
3. Resuspend cells in complete growth media (ex: DMEM/10%FBS) to no more
than 1.5*10
8
cells/10mL
4. Add 1% formaldehyde to the media (270 μL 37% formaldehyde/10mL) and rotate
cells 8 minutes room temperature.
5. Quench fixation reaction by adding 1M glycine to a final concentration of 0.125M
and rotate 5 minutes room temperature.
Suspension cell lines:
1. Using confluent T175 flasks, make sure cells are in a single cell suspension and
are shaken off of the plastic
2. Count cells and resuspend cells in complete growth media (ex:
RPMI1640/10%FBS) to no more than 1.5*10
8
cells/10mL
3. Add 1% formaldehyde to the media (270 μL 37% formaldehyde/10mL) and rotate
cells 8 minutes room temperature.
4. Quench fixation reaction by adding 1M glycine to a final concentration of 0.125M
and rotate 5 minutes room temperature.
**Note: for the rest of the protocol, treat adherent and suspension cells the same**
151
Harvesting and preparation of soluable chromatin
1. Pellet formaldehyde fixed cells 8 min 1200 rpm at 4
0
C
2. Remove media and resuspend cell pellet in 10mL PBS and spin cells down 8 min
1200 rpm at 4
0
C
3. Remove PBS and resuspend cell pellet in 10mL cell lysis buffer, incubate cells on
ice 10 minutes and spin down 5 min 5000 rpm at 4
0
C to pellet nuclei
4. Remove cytoplasmic fraction and resuspend nuclear pellet in nuclear lysis buffer
to a final concentration of 10
8
cells/mL.
5. Aliquot nuclear lysate into 500 μL aliquots and incubate on ice 10 minutes.
Lysates can be stored at -80
0
C for up to 1 year. **Do NOT freeze/thaw chromatin
more than once**
6. Sonicate lysates according to optimized protocols for each cell line. The DNA
should optimally be about 500 to 1000bp in length. This can be checked by
sonicating the chromatin, adding chelex buffer to extract the DNA and running it
on an agarose gel.
7. Centrifuge sonicated lysate at 4
0
C, 14K rpm for 10 minutes to pellet debris.
Chromatin Immunoprecipitation
1. Dilute 50 μL soluable chromatin sample in 150 μL cold ChIP dilution buffer
(5*10
6
cells/IP)
152
**Make sure to save 50uL for Input (undiluted!), store at -40
0
C overnight until IP
washes are finished**
2. Prepare protein A beads by washing beads in ChIP dilution buffer and resuspend
as 50% slurry in ChIP dilution buffer
3. Preclear chromatin by adding 40 μL 50% protein A slurry and rotating 1 hour at
4
0
C (**up to 4 hours is okay**)
4. Spin down 4000 rpm 5 min at 4
0
C.
5. Transfer supernatent to a new tube, add antibody and rotate overnight at 4
0
C.
6. The next day, prepare and wash more protein A beads, resuspending them as a
50% slurry in ChIP dilution buffer
7. Add 20 μL 50% slurry of beads to each IP sample and rotate at 4
0
C for 1 hour
(**incubating longer with the beads seems to increase the background**)
8. Pellet the beads 4000 rpm 5 min at 4
0
C, save supernatent in case the IP didn’t
work.
9. Wash the beads 2x500 μL cold ChIP dilution buffer
10. Wash the beads 4x500 μL cold RIPA/LiCl buffer
11. Wash the beads 2x500 μL cold TE buffer
153
DNA extraction using Chelex buffer
1. Add 100 μL 10% chelex buffer (make sure to vortex before adding to each
sample) to beads, vortex and boil 10 minutes. Use pipette tips that have been cut
to the 10 μL mark to add chelex.
2. Cool to room temperature and add 1 μL Proteinase K. Incubate at 55
0
C for 30
minutes, vortexing intermittently.
3. Boil 10 minutes, spin down full speed 30 sec and transfer supernatent to a new
tube.
4. Add 100 μL dH
2
O, vortex and collect bound material by poking a hole in the
bottom of the tube and the cap. Place the tube with the holes in it into a fresh tube
and centrifuge 1 minute full speed. Combine this supernatent with the 1
st
supernatent.
5. Store DNA at -40
0
C until analyzing by PCR.
PCR analysis of ChIP’d DNA
• All PCR conditions need to be standardized for each specific primer set.
• In general, I use a 30 cycle PCR with 60
0
C annealing temperature.
• For each PCR reaction, 1 μL of collected DNA is 0.5% of bound fraction
and I usually use 1-3 μL depending on my primer sets for each bound
fraction. For input PCRs, I dilute my DNA 1:100 and use 10uL for the
PCR reactions (0.05%)
• Alternatively, gradient template PCRs can be set up using the following
amounts of DNA:
154
Input Bound
0.005% 0.15%
0.015% 0.5%
0.05% 1.5%
0.15% 5%
Chromatin Immunoprecipitation for ChIP sequencing (using new HeLa cells)—
Experiments performed in conjunction with Tanya Spektor (See Table 5 for antibodies
and dilutions)
ChIP Buffers: To all buffers add 1mg/mL aprotinin/leupeptin, 1mg/mL pepstatin, 1mM
PMSF before using.
Farnham Nuclei Lysis Buffer
50 mM Tris pH 8
10 mM EDTA
1% SDS
RIPA Buffer Nuclear Isolation Buffer
50mM HEPES pH 8 150 mM NaCl
1mM EDTA 10 mM HEPES pH7.4
1% NP-40 1.5 mM MgCl
2
0.7% Na-Doc 10 mM KCl
0.5 M LiCl 0.5% NP-40 (This may need to be
decreased, depending on cell type)
155
0.5 mM DTT
Isolate Chromatin
1. Trypsinize HeLa cells
2. Wash cells 1 X in PBS.
3. Resuspend the cell pellet in media with 1 % formaldehyde and rotate at RT for
8 min.
4. Quench reaction by adding 1M glycine to a final concentration of 0.125 M and
rotate at room temp for 5 minutes.
5. Spin down cells 1200 rpm 8 minutes 4
0
C
6. Wash 2 X with PBS (spin down the same as in step 5).
7. Resuspend cell pellet in nuclear isolation buffer and incubate on ice for 10 min.
8. Spin 8 min at 600 g.
9. Resuspend nuclei pellet in Farnham nuclei lysis buffer to a concentration of 10
8
nuclei/mL and incubate on ice for 10 min. **Nuclear lysate can be stored after
this step at -80
0
C for up to 1 year, do not freeze and reuse sonicated chromatin**
10. Sonicate 500 μL chromatin for 2 X 20 sec at output level of 4.0 (40%) (to obtain
approx 500-1000bp sonication conditions must be determined for each cell type).
11. Spin sonicated chromatin at 14000 rpm for 10 min at 4
0
C. Keep supernatant.
Immunoprecipitation
156
1. Place 600 μL of pro-A Dynabeads in 1.5 mL tube and wash 3 times with PBS (-
Ca, -Mg) + 5 mg/ml BSA.
2. Resuspend in 600 μL PBS (-Ca, -Mg) + 5mg/mL BSA. Split beads as follows:
• 400 μL beads plus Specific Antibody (approx 40 micrograms = ~100 μLserum)
plus 500 μL PBS (-Ca, -Mg) +5mg/mL BSA
• 100 μL beads plus 300 μL PBS (-Ca, -Mg) + 5mg/mL BSA
• 100 μL beads plus 15 μL non-specific IgG plus 300 μL PBS + 5mg/mL BSA
3. Incubate rotating O/N at 4
0
C
4. Wash beads 3 X with PBS (-Ca, -Mg).
5. Resuspend beads plus antibody in 400 μL of Farnham IP dilution buffer.
Resuspend No Ab and IgG beads in 100 μL of IP dilution buffer each.
6. Set up IPs as follows:
• Beads plus (or minus) Ab: 100 μL (We used 100 μL of H4K20 monomethyl or
H3K9 monomethyl/IP)
• Sonicated Chromatin: 100 μL
• IP dilution Buffer: 1000 μL
1200μL total
7. Rotate O/N at 4
0
C
Washing and Eluting
157
1. Use magnet to pull beads away from the supernatant and save 1st supernatant as
unbound fraction.
2. Wash beads 8 times with 1 mL cold RIPA buffer
3. Wash beads once with 1 mL cold TE
4. Spin for 3 min at 3000 rpm RT and remove remaining supernatant.
5. Add 60 μL of TE + 1% SDS and incubate at 65
0
C for 10 min.
Vortex every two minutes. Remove supernatant and save.
6. Add 120 μL of TE + SDS and incubate at 65
0
C for 10 min.
7. Pool supernatants and add an additional 200 μL of TE to the pooled samples. Add
15μL 5 M NaCl and 1 μL 10 mg/mL RNase A. (Take 100 μL of total input and
treat similarly). Incubate at 65
0
C O/N to reverse cross-links.
8. Add 2.5V EtOH, 0.1V 3M NaOAc, 5 μL 5M NaCl and precipitate O/N -40
0
C
9. Spin samples for 15 min at 14000 rpm at 4
0
C. Let pellets dry.
10. Dissolve each pellet in 135 μL TE. Add 15μL of 10 X ProK buffer and 1.5 μL of
Pro K. Incubate in 37
0
C waterbath for 2 hours.
11. Adjust concentration to 400 μL with TE and phenol:chloroform extract samples.
12. Add 5 μg of glycogen and 1mL EtOH to each sample and precipitate O/N -40
0
C
13. Spin down for 30 minutes at 14000 rpm at 4
0
C and resuspend pellets in 30 μL Tris
pH 7.5
**This DNA is now ready to be given to the Epigenomics Core or to Illumina for
processing for ChIP-seq. If possible, it is best to give them ~50-100ng of DNA.
158
Test PCR
• Do test PCR of experimental sample vs. IgG of a known binding site. Do
not proceed if experimental sample does not show much more
amplification than IgG. If you do not have a known binding site, look to
see that you have more overall amplification with your experimental
samples than with IgG.
Chromatin Immunoprecipitation for ChIP-chip (using stem cells) (See Table 5 for
antibodies and dilutions)
ChIP Buffers: To all buffers add 1mg/mL aprotinin/leupeptin, 1mg/mL pepstatin, 1mM
PMSF before using.
Cell Lysis Buffer (**I have found that this buffer works best if made fresh**)
5mM PIPES pH 8
85mM KCl
0.5% NP-40
Nuclear Lysis Buffer
50mM Tris pH 8
10mM EDTA
1% SDS
ChIP Dilution Buffer
150mM NaCl
159
16.7mM Tris pH 7.5
3.3mM EDTA
1% Triton X-100
0.1% SDS
0.5% Na-Doc
RIPA/LiCL Buffer
50mM HEPES
1mM EDTA
1% NP-40
0.7% Na-Doc
0.5M LiCl
TE Buffer
10mM Tris pH 7.5
1mM EDTA
**For all of the experiments that I did, I first sorted the stem cells and have found that the
protocol can be used with as little as 20,000 cells/IP reaction, sonicating 1x15 sec, output
4.0. I would not trust much less than that**
1. Add 1% formaldehyde to the media (270 μL 37% formaldehyde/10mL) and
rotate cells 8 minutes room temperature.
160
2. Quench fixation reaction by adding 1M glycine to a final concentration of
0.125M and rotate 5 minutes room temperature.
3. Spin down 8 minutes 1200 rpm 4
0
C.
4. Wash pellets with PBS and respin
5. Resuspend cells in up to 10mL cell lysis buffer (for smaller numbers of cells,
1mL is sufficient.). Incubate 10 minutes on ice
6. Spin down 5000 rpm 5 minutes 4
0
C.
7. Resuspend nuclear pellet in 500 μL nuclear lysis buffer (this is for smaller
numbers of cells, up to 100,000 cells total, the entire lysate will then be used
for 1 ChIP set). Lysates can be stored at this point for up to 1 year at -80
0
C.
Do not store after sonication.
8. Sonicate 1x15 sec pulse, output 4.0 (40%), spin down 10 minutes 14000 rpm
4
0
C.
9. Aliquot 250 μL for input and freeze.
10. Dilute the rest to 500 μL using ChIP dilution buffer and add 40 μL washed,
equilibrated 50% slurry protein A beads and rotate 1hr at 4
0
C.
11. Spin down 4000 rpm 5 minutes 4
0
C, transfer supernatent to a new tube.
12. Add antibody and rotate O/N 4
0
C.
13. The next day, add 45 μL 50% slurry protein A beads (washed, equilibrated)
and rotate 1hr 4
0
C.
14. Spin down 4000 rpm for 5 minutes at 4
0
C.
161
15. Wash beads 8x1mL RIPA/LiCl buffer
16. Wash beads 2x1mL TE buffer
17. Add 60 μL TE/1% SDS, incubate 10 minutes 65
0
C, vortex every 2 minutes.
18. Spin down, transfer supernatent to a new tube and save. Add 120 μL TE/1%
SDS to the beads, incubate 10 minutes 65
0
C, vortex every 2 minutes.
19. Spin down and collect supernatent. Pool with 1
st
supernatent and add 1 μL
RNase A, 15 μL 5M NaCl, and 200μL TE. Incubate O/N at 65
0
C to reverse
crosslinks.
20. The next day, add 1mL EtOH and precipitate O/N at -40
0
C
21. The next day, spin down samples 15 minutes 14000 rpm at 4
0
C and dry pellets
to remove residual EtOH.
22. Resuspend pellets in 135 μL TE. Add 15 μL 10x Proteinase K buffer and
1.5μL proteinase K. Incubate at 37
0
C for 2h.
23. Add 248.5 μL TE for a final volume of 400 μL and phenol:chloroform extract.
24. Add 2.5 μL 1:10 glycogen (diluted in TE) to extracted sample (top layer of
phenol:chloroform) and 1mL EtOH. Precipitate O/N -40
0
C.
25. Spin down precipitated DNA 30 minutes 14000 rpm 4
0
C
26. Resuspend pellet in 30 μL dH
2
O.
27. Quantitate DNA and dilute out to 1ng/ μL (do NOT dilute the whole sample at
once, only make ~40 μL stock) for WGA2 amplifications
162
WGA2 amplification (Sigma kit)—treat input and bound fractions the same!
1. Add 1 μL 10x fragmentation buffer to 10ng (10μL) DNA. Incubate at 95
0
C 4
minutes and cool on ice
2. Add 2 μL 1x library preparation buffer, 1 μL library stabilization solution,
vortex and spin down. Incubate 2 min at 95
0
C, cool on ice
3. Add 1 μL 1x library preparation enzyme, vortex and PCR using WGA2
Library protocol:
16
0
C 20 min
24
0
C 20 min
37
0
C 20 min
75
0
C 5 min
4
0
C hold
4. Add 7.5 μL 10x amplification master mix, 47.5 μL dH
2
O, 5 μL WGA DNA
polymerase to the PCR’d material. Vortex and PCR using WGA Amplify
program
95
0
C 3 min
95
0
C 15 sec
65
0
C 5 min
**Cycle 2
nd
two steps 14x**
5. Quantify DNA and use 10ng of amplified DNA as input for a 2
nd
step
amplification. Set up the following reaction:
163
7.5μL 10x amplification master mix, 5 μL WGA DNA polymerase, 10ng
amplified DNA, dH
2
O (final volume of reaction 75 μL) and reamplify using
WGA Amplify program.
6. Purify DNA using PCR Purification Kit (Qiagen) and quantify DNA.
Test PCRs
Set up test PCRs using primers that should amplify a positive and negative region
to ensure that ChIPs worked properly. I used 0.3 μL template for input reactions and
0.2μL for bound fractions and amplified for 35 cycles. These conditions may have to be
changed depending on your primer sets.
We sent Nimblegen 4 μg total DNA from Input and Bound samples for ChIP-chip
analysis.
Western Blot Studies
Western Blot Analysis
Preparation of Lysates
1. Collecting cells: Trypsinize cells off of petri dishes and spin down 500xg 5 min
Resuspend pellet in PBS and respin to wash out all media
2. Resuspend the cell pellets in 2x Laemmli to lyse the cells (final concentration of
10^7 cells/mL).
164
3. Boil the lysate for 10 minutes and homoginize using a 21-gauge needle (10-15x)
after lysate has cooled.
4. For western blot analysis, we usually load the gels by cell number/lane using ~10
5
cells/lane or ~10uL/lane.
2xLaemmli Buffer
0.2% SDS
3mM Tris
0.4M glycine
Gel Transfer
1. Make “sandwich” of blot paper, membrane, and gel in the following order
(bottom to top):
• 9 pieces of blot paper soaked in 1x Tobin buffer+MeOH
• Membrane soaked in MeOH
• Gel soaked in 1x Tobin buffer+MeOH
• 6 pieces of blot paper soaked in 1x Tobin buffer+MeOH
2. Remove all air bubbles that may be trapped between the pieces of blot paper and
to smooth out the gel using a pipet to ensure good transfer.
165
3. Run transfer for 120 min at 54mA/gel (ex: 2 gels=106mA), place a 1L bottle
filled up at least 500mL on top of the apparatus to ensure that the sandwich is held
together tightly.
4. Disassemble “sandwich” and place membrane into MeOH for ~5min. To check
for good transfer, the gel can be coomassie stained for 1hr and destained for 1 hr
after transfer.
5. Stain membrane with Ponceau S to look at transfer quality for 10 min (rocking)
and rinse with dH
2
O.
Blot (See Table 4 for antibody conditions and dilutions)
1. Block for 30 min in 5%milk/TBS pH 7.5
2. Seal the membrane in plastic pouch after adding primary antibody (diluted in 1%
milk/TBS, 4mL/whole membrane), incubate on Nutator for 1 hr at room
temperature. (The incubation time and temperature can be adjusted as the
antibody may require)
3. Wash membrane 3x10 min TBS/0.1% Tween-20
4. Seal membrane in plastic pouch after adding secondary antibody (diluted in 1%
milk/TBS, 4mL/whole membrane), incubate on Nutator for 1 hr at room
temperature. (The incubation time and temperature can be adjusted as the
antibody may require)
5. Wash membrane 3x10 min TBS/0.1% Tween-20
6. Develop using ECL plus kit:
166
• Dilute 25 μL of Solution B into 1mL of Solution A/blot (this solution is
light sensitive, so make in a foil wrapped 15mL conical tube)
• Add dropwise to membrane, evenly covering the entire surface—
especially around your MW of interest.
• Incubate 5 min, moving foil around to spread out ECL.
• Develop using film, it’s best to do a 1 min and 3 min exposure to ensure
that you are seeing your bands of interest. A shorter or longer exposure
may also be required.
167
Antibody Species Dilution Incubation
Conditions
H4K20 monomethyl
(linear)
Rabbit 1:8000 1.5h RT
H4K20 dimethyl
(branched)
Rabbit 1:15000 1h RT
H4K20 trimethyl
(branched)
Rabbit 1:10000 1.5h RT
H3K9 monomethyl Rabbit 1:40000 1.5h RT
H3K9 dimethyl Rabbit 1:60000 1.5h RT
H3K9 trimethyl Rabbit 1:6000 1.5h RT
H4 general Rabbit 1:40000 1h RT
H3 general Rabbit 1:500000 1h RT
RUNX1 Rabbit 1:250 O/N 4
0
C
RIZ1 Rabbit 1:250 2h RT
PR-Set7 2A Rabbit 1:1000 2h RT (better than
2169B)
PR-Set7 2169B Rabbit 1:1000 2h RT
FLAG Mouse 1:2500 1h RT
HA Rabbit 1:1000 1h RT
integrin a2 Rabbit 1:250 2h RT (cross reacts
with marker a lot!)
L3MBTL1 (Rice
Lab)
Rabbit 1:1000 2h RT
L3MBTL1 (LPBio) Rabbit 1:1000 2h RT (cross reacts
with marker a little)
L3MBTL1 (Nimer
Lab)
Rabbit 1:1000 O/N 4
0
C
H3K27 monomethyl Rabbit 1:50000 1hr RT
H3K27 dimethyl Rabbit 1:10000 1h RT
H3K27 trimethyl Rabbit 1:5000 1h RT
H4K16 Acetyl Rabbit 1:15000 1h RT
β-actin Mouse 1:40000 1h RT
Table 6. Antibodies used for Western blot analysis
Antibodies, their species and the conditions used for western blot are listed
168
Ammonium Sulfate Precipitation of IgG fraction of Serum
Solution needed:
~3.8-4.1 M Ammonium Sulfate (NH
4
)
2
SO
4
—pH between 7.4 to 7.8. Always check pH
of the solution before use. Adjust using H
2
SO
4
or NaOH.
1. Add equal volume of PBS/1mM PMSF to serum
2. Spin at 300xg for 10 minutes RT
3. Transfer the supernatent to a beaker and stir. Add (NH
4
)
2
SO
4
dropwise until the
final concentration of (NH
4
)
2
SO
4
reaches 2.05M (equal volume (NH
4
)
2
SO
4
to
starting amount) and stir for 10 minutes (if small volume, prepare in 1.5mL epi
tube and rotate 10 minutes)
4. Incubate solution at RT for 30 minutes to allow for precipitation
5. Spin down 500x g 10 minutes RT
6. SAVE THE PELLET and resuspend in PBS/1mM PMSF (equal volume to
starting volume)
7. Repeat steps 3-6 two more times.
8. Resuspend the final pellet in 1mL PBS/1mM PMSF and add solution to a dialysis
bag with MWCO 10-12kDa (double the bag size to allow for expansion)
9. Dialyze solution against PBS/1mM PMSF for 1 hour (repeat two more times,
leaving the third change overnight at 4
0
C.
169
10. Add glycerol to a final concentration of 10% and NaN
3
to a final concentration of
0.05%. Aliquot into ~100 μL aliquots and store at -80
0
C.
FACS studies
Analysis/Sorting of CD41 positive K562 cells
1. Collect cells, wash with PBS and resuspend in FACS buffer to a final
concentration of no more than 1.5*10
7
cells/mL for sorting or 10
6
cells/mL for
analyzing.
2. Add 20 μL α-CD41-FITC or PE/10
6
cells, vortex and incubate at RT in the DARK
for at least 30 minutes.
FACS buffer
2% FBS
0.1% NaN
3
PBS (-Ca, -Mg)
Analysis/Sorting of Stem Cells (See Table 5 for antibodies and dilutions)
1. Collect cells and wash 1x PBS (-Ca, -Mg).
2. Remove PBS and wash 3x 1mL 0.1%BSA/PBS (-Ca, -Mg)
3. Resuspend cells in 1mL 0.1% BSA/PBS (-Ca, -Mg) and aliquot 250 μL into 4
tubes for each antibody condition, spin down cells.
170
4. Remove 0.1% BSA/PBS and resuspend pellets in 300 μL either α-GCTM2, α-
TG30 or both α-GCTM2 and α-TG30, or 0.1% BSA/PBS (negative control) and
incubate on ice 30 minutes.
5. Spin down cells 5 minutes 600xg RT and wash cells 2x1mL 0.1% BSA/PBS (-Ca,
-Mg).
6. Resuspend cell pellets in 100 μL secondary antibodies diluted in 0.1% BSA/PBS
and incubate on ice 30 minutes in the dark.
7. Spin down cells 5 minutes 600xg RT and wash cells 3x1mL 0.1% BSA/PBS (-Ca,
-Mg).
8. Resuspend cells in 1mL 0.1% BSA/PBS (-Ca, -Mg) and strain cells using a cell
strainer (BD Falcon, REF 352360) to remove clumps of cells before FACS.
171
Antibody Species Dilution
GCTM2 mouse IgM no dilution
TG30 (CD9) mouse IgG2a no dilution
Alexa Fluor 488 anti-mouse IgG2a 1:500
Alexa Fluor 647 anti-mouse IgM 1:500
CD41-FITC
(eBioSciences)
mouse anti-human,
preconjugated
20μL/10
6
cells
CD41-PE (eBioSciences) mouse anti-human,
preconjugated
20μL/10
6
cells
Table 7. Antibodies used for FACs analysis
172
Differentiation Studies
TPA differentiation of K562 cells
• The stock solution of TPA is made in sterile DMSO. Treat 2.5*10
5
cells/mL cells
with a final concentration of 10nM TPA for 5 days. The stock of TPA will have
to be diluted to reach this concentration and so I made dilutions in RPMI 1640
media and then added the diluted TPA to the cells. This will allow for ~30%
differentiation as measured by CD41 positive cells by FACS analysis as well as a
~2 fold increase in RUNX1 gene expression. As a vehicle control, it’s best to
dilute DMSO in the same way and then add that to the cells.
Hemin differentiation of K562 cells and analysis of differentiation efficiency
• The stock solution of hemin has to be made in the following way:
Dilute 13mg of hemin in 200 μL 0.5M NaOH, then add 250 μL 1M Tris pH
7.5. Vortex and then dilute to 5mL with dH
2
O and filter sterilize. This
will result in a working stock of 4mM. Use this stock to treat 2.5*10
5
cells/mL with a final concentration of 50 μM hemin for 96 hours. This will
allow for ~45% differentiation but the cells will not have ejected their
nuclei. As a vehicle control, I made up a stock solution the same way but
without adding the hemin powder and treated cells with it.
173
• To measure the % differentiation, benzidine staining and measurement of soluable
hemoglobin are performed exactly as described in the following papers:
Gopalakrishnan and Anderson, 1979 and Wanda et al., 1981.
Luciferase Studies
1. Transfect 6 well plates of cells using lipo2000 or lipo/plus protocol, using 5ng of
Renilla vector as a transfection control with the your experimental DNA.
2. 48 hours after transfection, lyse cells in 1x Passive Lysis Buffer (PLB, 5x stock
provided by Promega Luciferase kit at –40) by either adding 350uL 1xPLB to
plates and rock for 15 minutes, room temperature and collecting cells by scraping
or trypsinizing cells off of plastic and resuspending in 1xPLB.
3. Freeze/thaw the lysates 2x alternating between –80 and room temperature (it
usually takes about 15 minutes to freeze)
4. Spin down the lysates 2 minutes full speed at room temperature.
5. Aliquot 20uL cleared lysate per well (x4 wells) into black luminometer plates.
6. Add 100uL LAR II reagent/well to read luciferase values, this is a time sensitive
process so make sure to add just before putting the plate in the TopCount
7. After reading is finished, add 100uL Stop&Glo reagent (to read renilla values),
this is a time sensitive process so make sure to add just before putting the plate in
the TopCount
174
***LAR II reagent is prepared by mixing 10mL Luciferase Assay Buffer II with
Luciferase assay substrate. 10mL LAR II is sufficient for 1 full 96 well plate. This
buffer can be stored for 1 year at –80.
***Stop&Glo reagent is prepared by mixing 10mL Stop&Glo buffer with 200uL 50x
Stop&Glo substrate. 10mL Stop&Glo reagent is sufficient for 1 full 96 well plate. This
reagent must be made fresh every time.
Both LAR II and Stop&Glo are light sensitive and must be kept on ice when not in use.
Transfections (Lentiviral experiments, siRNA studies, electroporations, kill curve
information for different cell lines)
Lentiviral production and infection
Virus Production
DAY 1: Plate 293FT cells in 10cm plate so that they will be ~90% confluent on the day
of transfection (~5*10
6
cells in 10mL)
**DO NOT USE 293FT cells that have been passed more than 20 times**
**DO NOT COAT PLATES WITH POLY LYSINE**
DAY 2: Transfect cells
175
1. Dilute 9ug of ViraPower Packaging mix and 3ug of pLenti6 expression vector
(USE VECTORS THAT HAVE BEEN MIDI- OR CsCl PREPPED ONLY) in
1.5mL OMEM in a 15mL conical tube
2. In a separate tube, dilute 36uL Lipo2000 in 1.5 mL OMEM. Incubate 5 min RT
3. Combine diluted DNA with diluted Lipo2000 and incubate 20 min RT
4. Add DNA-Lipo mix DROPWISE to each plate of cells and incubate cells
overnight
DAY3: Remove media containing Lipo/DNA and replace with 10mL complete media
DAY 4
***FROM THIS POINT ON, YOU ARE WORKING WITH LIVE VIRUS, SO BE
SURE TO DOUBLE GLOVE AND WEAR A LAB COAT TO PROTECT YOURSELF
BECAUSE THE VIRUS CAN INFECT YOU! MAKE SURE TO PUT EVERYTHING
THAT TOUCHES VIRUS IN A SOLUTION OF 70% EtOH/1% SDS TO
INACTIVATE THE VIRUS. LABEL EVERYTHING THAT YOU STORE FOR
OTHERS TO KNOW WHERE THE LENTIVIRUS IS!***
Collect supernatent from plates and store in foil wrapped conical tube at 4 deg. Add
10mL fresh media
DAY 5:
1. Collect supernatent from plates and combine with supernatent from day 4 samples
2. Centrifuge supernatants at 1000 rpm for 5 min at 4 deg to pellet cell debris
3. Filter the supernatants through a 0.45 μm low protein-binding filter.
176
4. Add filtered supernatent , up to 30mL/ tube to 25x89 mm ultratubes (open top,
thick walled, polycarbonate). Spin for 90 min in SW 28 swinging bucket rotor at
25000 rpm, 4 deg.
5. Aspirate supernatent exhaustively until the tubes seem dry. Invert tubes over a
Kimwipe to drain remaining supernatent for 5 min. Aspirate the media residue at
the opening of the tubes
6. Add 100uL PBS+Ca and Mg with NO bicarbonate to the tubes and seal with
parafilm. Nutate the tubes at 4 deg overnight to resuspend viral pellet. Use virus
immediately or store at –80 for up to 1 year.
**This protocol was standardized using 20mL collected supernatent and 100uL of
concentrated virus is enough to infect one well of a 12 well plate with 1:10 dilution of the
virus (Vf=500uL). At this dilution, 30% of suspension cells (K562) were infected.**
The drug selection marker on the lentiviral plasmids is blasticidin. 3.5ug/mL is the
optimal amount to 100% kill K562 cells that are not blasticidin resistant, 2ug/mL for
HeLa S3 cells. The normal killing time is 10-14 days.
siRNA duplex transfections (adapted from Julien and Herr, 2004, see Table 6 for siRNA
duplex sequences)
1. Plate HeLa cells (or other desired cell line) at a density of 10
5
cells/well
(2mL/well) in a 6 well dish and grow overnight.
177
2. Dilute 100 pmol duplexes in 200 μL OMEM and dilute 5 μL Oligofectamine in
48μL OMEM (per well transfected). Incubate 5 minutes RT
3. Mix oligofectamine/OMEM with siRNA/OMEM and incubate 20 minutes RT
4. Add 258 μL mixture and 2mL DMEM/well.
5. Repeat steps 2-4 twelve hours later.
6. 48 hours after the 2
nd
transfection, collect RNA/WCL and replate some cells to
transfect again if needed 24 hours later.
178
Gene Sequence
PR-Set7 AUCGCCUAGGAAGACUGAUC
L3MBTL1 GUUCAGUCAUAGUAAAGAA
G9a UAAGAAUCAUCCUCUCUCAUU
Lamin A/C Dharmacon catalog# D-001050-01
Table 8. Sequences of siRNA duplexes used for knockdown experiments in HeLa
cells
179
Electroporation of K562 cells (Standardized by Shumin Wu)
1. Collect K562 cells and spin down 6 min 600xg RT
2. Wash cells in PBS and respin
3. Resuspend cells to a final concentration of 10
7
cells/mL in electroporation buffer
(100mM HEPES pH 7.4, 10 μg/mL DEAE/Dextran, Opti-MEM medium).
4. Mix 10 μg DNA with 200 μL of cells and transfer to 0.2mm electroporation cuvette
and incubate at 37°C for 2 minutes.
5. Electroporate at 1000 μF, 150V.
6. Incubate electroporated cells at 37°C for 10 minutes.
7. Transfer cells to one well of a 6 well plate containing 1.8mL RPMI 1640/10%
FBS media, then rinse out cuvette with media to ensure a good transfer of cells to
the plate. For this step, use a pastuer pipet to transfer the cells and try to avoid
foam of dead cells.
8. Change media 24-48 hours after electroporation to remove cell debris.
**This will result in ~10-25% transfection efficiency. The transfection efficiency can
be measured by electroporating a GFP construct into the cells and measuring GFP
expression by FACS analysis. I have also co-transfected 1 μg of GFP plasmid with
9μg of other plasmids (ex FLHA Set7 CD) and measured transfection efficiency by
FACS analysis.**
180
Kill curve information for various cell lines
Kill curves were performed by plating cells at 8*10
5
cells/mL in 6 well plates. 24 hours
after plating the cells should be ~80-90% confluent. Increasing concentrations of the
drug was added to the cells and the cells were examined daily for killing efficiency either
by trypan blue staining and counting or visual inspection (Table 7).
181
Cell Line Puromycin
(2d)
Zeocin (14d) Hygromycin
(7d)
Blasticidin
(14d)
HeLa tetR 1 μg/mL 400 μg/mL
MCF7 tet off 2 μg/mL 500 μg/mL 300μg/mL
old HeLa 6 μg/mL 400 μg/mL
K562 3 μg/mL 3.5 μg/mL
HEK 293T 3 μg/mL 400 μg/L
HeLa S3 2 μg/mL 400 μg/mL 2 μg/mL
Table 9. Concentration of drugs used for selection of tissue culture cells
Each cell line was plated as described and carried in drug containing media until
approximately 95% of the cells were killed by the indicated time (show in parethesis).
182
Quantitative real time PCR studies
cDNA conversion from mRNA
**RNA is extracted using Qiagen RNeasy extraction kit (Cat#: 74104)
cDNA conversion kit (ABI TaqMan RT reagent Cat# N808-0234)
Using this kit, you can RT 2 μg in 100 μL but I have always RT’d 500ng in 50 μL.
All reagents except DEPC H
2
O are kept at -40
0
C. Before starting, thaw 10x RT buffer,
MgCl
2
, dNTPs, random hexamers and RNA. Once items have thawed, store on ice. DO
NOT allow RNase or RT enzyme to come to RT, always store on ice.
RT Reaction Master Mix 1X RXN
10x RT buffer 5 μL
MgCl
2
11μL
dNTPs 10 μL
Random Hexamers 2.5μL
RNase Inhibitor 1 μL
Reverse Transcriptase 1.25 μL
DEPC H
2
O and RNA (combined total volume) 19.25 μL
183
Reverse Transcription PCR Conditions
25
0
C for 10 minutes
48
0
C for 30 minutes
95
0
C for 5 minutes
4
0
C hold
We assume 100% conversion of RNA to cDNA and based on this we amplify 5-15ng of
cDNA
qPCR using SYBR green
Expression Primer Design for Real Time PCR using SYBR green
Look up the gene of interest on UCSC genome browser (http://genome.ucsc.edu/) and
select the gene under the RefSeq Genes (these are the sequences that have been verified).
If your gene is not here, then select from known genes. This will bring up a map of the
gene within the genome and you want to click on the gene again under RefSeq. Then
select mRNA/Genomic Alignments. The cDNA sequence will be the first sequence that
you see and this is the sequence that you want to use to generate expression primers.
Select any portion of the sequence that is shown in blue as this is the portion that is the
coding region of the cDNA. Make sure that the portion that you select contains some
base pairs that are light blue because these are the exon-exon junctions and are important
for designing good expression primers. **Take note of where the exon-exon junction is
184
located, i.e. how many base pairs away it is from the start of the sequence that you
highlighted, and try to highlight at least 300 base pairs**
Copy and paste your sequence into Primer3 program (http://frodo.wi.mit.edu/cgi-
bin/primer3/primer3_www.cgi). It is not necessary to remove the numbers on the outside
of the sequence. Make sure that the boxes are checked for Pick left primer and Pick right
primer. Set the product size to be 150-250 and change the %GC Min to be 30%
and then press “Pick Primers”. The next page will contain a list of 5 primer sets that are
optimal for your region of interest. The first set will be displayed at the top of the page
and the other 4 sets will be at the bottom of the page. The 1
st
set is the most optimal set.
It would be best to copy and paste all 5 primer sets into either word or excel so that you
have a record of the primer sequences in case the 1
st
set doesn’t work well so that you
don’t have to repeat the entire process. It is also a good idea to write down the length of
each primer because you will need this information later.
Open the USC/Norris Cancer Center DNA Core Facility oligo synthesis website
(http://uscnorriscancer.usc.edu/Core/DNA/OSynth.aspx). Select Judd Rice from the P.I.
list and fill in the rest of the information with your information (contact person, email
etc.). Make sure that DNA is selected from the Type of primers that you want to make.
Name your primers (the 1
st
primer is the forward primer so I usually call it gene name_F,
ex: myoD_F) and paste in the sequence into the box labeled 5’. After you have finished
185
adding in your primer sequences, make sure that 40nmole is selected from the scale and
that desalted (no charge) is selected from the purification and that None is selected from
the modifications. Select USC account number from the payment methods and enter in
the account number (provided by Judd). Select an Anantomical Site (required, you can
always put unsure or special studies, it doesn’t matter) and then press Refresh
Costs/Count Bases to double check that there are no errors in the information that you
added. If there is an error, it will show up in red. If everything seems okay, then press
submit. The oligos will be ready within 1-2 days after ordering and can be picked up
from NOR 5338 between 8am-5pm. You will also get an email when the oligos are
ready.
Testing Primers for Use with SYBR green and setting up Real Time using SYBR green
**It is critical to determine the melting temperature of primers to be used for SYBR
green as primer dimers can interfere with interpretation of the results because the SYBR
green will intercalate into the dimers as well as the true PCR product.
1. Set up 2 reactions, one using the cDNA of interest and one without cDNA
template.
186
PCR Master Mix 1X RXN
2X SYBR green 10μL
10 pmol F primer 1 μL
10 pmol R primer 1 μL
cDNA 1 μL
PCR H
2
O 7 μL
2. Run program JennMeltCurve:
95
0
C for 3 minutes
40-50 cycles of:
95
0
C for 10 seconds
56
0
C for 30 seconds**fluorescence read here**
95
0
C for 1 minute
56
0
C for 1 minute
71 cycles starting at 60
0
C, temperature increases 0.5
0
C every 10 seconds
and ending at 95
0
C**fluorescence read here**
3. Analyze data by comparing the melt curves ± cDNA. The optimal primer set will
not have any curves for the primers only sample but if there is a curve, it’s best to
find a temperature where the primers only curve will be completely melted away
without effecting the cDNA sample. It is important to retest the melt temperature
every time a new set of primers are ordered.
187
4. Once the melt temperature has been established, set up at least 3 reactions/sample
using the following master mix:
PCR Master Mix 1X RXN
2X SYBR green 10μL
10 pmol F primer 1 μL
10 pmol R primer 1 μL
cDNA 1 μL
PCR H
2
O 7 μL
**Make sure to include a –cDNA template control in duplicate
5. Run program JennSYBRgreen:
95
0
C for 3 minutes
40-50 cycles of:
95
0
C for 10 seconds
56
0
C for 30 seconds
melt spec temp for 10 sec **fluorescence read here**
hold at 20
0
C
188
Table 10. Primers used in SYBR green Real Time PCR experiments
Primer names, melt temperatures, and sequences (F on top, R on bottom)
189
Primers Melt
Temperature
Sequence
PR-Set7 83
0
C 5’ ATTGCCACCAAGCAGTTCTC 3’
5’ CGATGAGGATGAGGTGAGGT 3’
L3MBTL1 85
0
C 5’ GAGCTCCTCAAACCCATGAA 3’
5’ GGCCTTCTGCTCCTCTAGGT 3’
RUNX1 80
0
C 5’ ACTTCCTCTGCTCCGTGCT 3’
5’ GCGGTAGCATTTCTCAGCTC 3’
GAPDH 82
0
C 5’ CAGCCGAGCCACATCGCTCAGACA 3’
5’ TGAGGCTGTTGTCATACTTCTC 3’
Lamin A/C 82
0
C 5’ CAAGCTTGAGGCAGCCCTAG 3’
5’ CTCACGCAGCTCCTCACTGTA 3’
myoD 80
0
C 5’ AGCACTACAGCGGCGACT 3’
5’ AGGCAGTCTAGGCTCGACAC 3’
Riz1 specific 80
0
C 5’ AATTTGGGATGGATGTGCATTG 3’
5’ GGCGCGATTGGCTTTAAAGT 3’
All Riz
isoforms
80
0
C 5’ CTGGATCCCCAAGAGCCGGAAAGGGAAGAA 3’
5’ TCTGGCTCACTTGTCTTCAGTTGT 3’
DNAJC15 77
0
C 5’ TGGATCTGCGAGAAGAAACC 3’
5’ CTGCAACACCCAGTCCTACA 3’
CHES1 80
0
C 5’ GCACCTACTGGGTGGAAAAA 3’
5’ GAGGACAGGTGGGAGGTGTA 3’
CECR5 80
0
C 5’ CTGCAGCTGATCATGGATGT 3’
5’ GCCCGTCACTTTCTGGTAAA 3’
KL 80
0
C 5’ ATCACCATCGACAACCCCTA 3’
5’ ACACCTGACCTCCCTGAGTG 3’
macro H2A 85
0
C 5’ GGCAGGTGTCATCTTTCCAG 3’
5’ GGCGATGGTCACTCCTTTTA 3’
SKIL 79
0
C 5’ CTGGGGCTTTGAATCAGCTA 3’
5’ CATGGTCACCTTCCTGCTTT 3’
ANKRD15 80
0
C 5’ CCCTGACTTCCAGAAAACCA 3’
5’ TGCTGATTGGCTTTCCTTCT 3’
ZNF560 80
0
C 5’ ACTGGACCCAGCTCAGAGAA 3’
5’ CCTTTGGTTTTCAGGCACAT 3’
ZNF677 74
0
C 5’ GCCTTGTACAGGGACGTGAT 3’
5’ GTCCCACAGGCTGTCAAACT 3’
Histone H3.3 80
0
C 5’ CCACTGAACTTCTGATTCGC 3’
5’ GCG TGC TAG CTG GAT GTC TT 3’
RSN 85
0
C 5’ AGGAGAAGCAGCAGCACATT 3’
5’ TCCATTTTGGCTTCCAATTC 3’
190
qPCR using TaqMan primers
**PCR Master Mix (ABI# 4304437)
PCR Master Mix 1X RXN
2X PCR Master Mix 10 μL
20X Primer/Probe 1 μL
DEPC H
2
O and cDNA 1 μL cDNA + 8 μL H
2
O
Note: 20 μL is the minimal reaction volume
BioRad Real Time PCR Conditions
50
0
C 2 minutes
95
0
C 10 minutes
40-50 cycles of
95
0
C 15 sec
60
0
C 1 minute
191
Gene Name TaqMan Assay Number
PR-Set7 Hs01029945_m1
Suv4-20 (CGI-85) Hs00211300_m1
G9a Hs00198710_m1
EuHMTase1 (GLP1) Hs00226978_m1
GAPDH Hs99999905_m1
Lamin A/C Hs00153462_m1
Histone H3.3 Hs00855159_g1
L3MBTL1 Hs00210032_m1
DNAJD1 Hs00387763_m1
TCF12 Hs00175295_m1
CD44 Hs00153304_m1
SMYD1 Hs00400855_m1
PTK6 Hs00178742_m1
EEF1A2 Hs00157325_m1
PTPN2 Hs00747429_mH
NSE1 Hs00376942_m1
ATM Hs00175892_m1
SOX17 Hs00751752_s1
DUSP16 Hs00411837_m1
SIX3 Hs00193667_m1
ADD3 Hs00249890_m1
MYL6 Hs00819642_m1
RXRA Hs00172565_m1
Table 11. Primers used for TaqMan based Real Time PCR
192
Analysis of Real Time PCR Data
1. Average all of the replicates for each sample and calculate the standard deviation.
The standard deviation should not exceed 1 cycle. If so and quadruplicate
samples were set up, take out the outlier sample cycle threshold value. If only
triplicate samples were run, take into consideration that it might be best to re-run
the PCR reactions as a standard deviation greater than 1 is too large.
2. Plug the average values of the housekeeping gene and the gene of interest into the
following formula:
2
-(average gene of interest-average housekeeping gene)
3. Take the resultant values and normalize them by mock treated expression data:
mock/mock=1
treated/mock=expression relative to mock treated (normalized to
housekeeping gene)
193
Immunoprecipitation studies
IP lysis buffer
50mM Tris pH 7.5
150mM NaCl
0.5mM DTT
1% NP-40
protease inhibitors (1 μg/mL pepstatin A, 1 μg/mL leupeptin/aprotinin, 1mM PMSF)
Preparation of agarose beads for immunoprecipitation
Protein A or G beads (use pipette tips cut at 10 μL mark to transfer beads)
1. Beads are stored in 20% EtOH solution as a 50% slurry. Aliquot the amount
of beads needed for IPs and spin down 30 seconds 14000 rpm to remove
EtOH
2. Wash beads 2x1mL IP lysis buffer to equilibrate
3. Resuspend beads in 50% slurry of IP lysis buffer
Conjugated beads (use pipette tips cut at 10 μL mark to transfer beads)
1. Beads are stored in solution as a 50% slurry. Aliquot the amount of beads
needed for IPs.
2. Add 10 volumes 100mM glycine pH 2.5 to remove excess antibody. Spin
down 30 seconds 14000 rpm to remove supernatent
194
3. Add 10 volumes 200mM Tris pH 8. Spin down 30 seconds 14000 rpm to
remove supernatent
4. Add 10 volumes IP lysis buffer + protease inhibitors. Spin down 30 seconds
14000 rpm to remove supernatent
5. Resuspend beads as 50% slurry in IP lysis buffer.
Immunoprecipitations from tissue culture cells
1. Trypsinize cells from 6 well plate and pellet cells 5 minutes 600xg RT
2. Wash pellet 1x PBS and re-spin
3. Resuspend cells in 300-500 μL cold IP lysis buffer + protease inhibitors
4. Spin down 10 minutes 14000 rpm 4
0
C to pellet insoluable fraction
5. Separate supernatent from the pellet and use the supernatent for IPs.
**If using protein A or G beads, preclear lysates 4h-overnight using 40 μL of 50%
slurry of appropriate beads. Spin down lysates 5 minutes 4000 rpm 4
0
C. Transfer
lysates to a new tube.
6. Keep ~20-50 μL of supernatent at -40
0
C for input for Western blot analysis. Use
the remaining supernatent to set up IP reactions.
7. After adding beads and antibody or pre-conjugated beads, rotate overnight at 4
0
C.
8. Spin down 5 minutes 4000 rpm at 4
0
C and save supernatent as unbound fraction.
9. Wash beads 2x500 μL cold IP lysis buffer.
195
10. Resuspend beads in 35 μL 6x SDS dye and either boil 10 minutes or elute at 55
0
C
for 10 minutes. Elution at 55
0
C allows for separation of the bound fraction
without getting too much of the heavy and light chain of the antibody to reduce
background.
11. Collect bound fraction by poking a hole in the top of the epi tub and in the bottom
and placing in a clean epi tube to spin down. This allows for collection of the
bound material without getting any of the beads. Proceed with Western analysis
or save bound and unbound fractions at -40
0
C.
Immunoprecipitations using IP lysates and TNT’d material
1. Prepare IP lysates as described above and add 3-10 μL of TNT’d material to the
lysates.
2. Proceed with IP reactions as above.
**The amount of TNT’d material that you will use depends on how efficient the
binding is. This will have to be standardized for each TNT’d protein.
196
Antibody Amount Used Beads Used Preclear? How
long?
α-DBD 2.5 μg Protein G Yes, 4 h 4
0
C
α-FLAG 20 μL 50% slurry pre-conjugated No
α-HA 20 μL 50% slurry pre-conjugated No
α-PR-Set7 2A 20 μL Protein A Yes, overnight 4
0
C
Table 12. Antibodies used for immunoprecipitations
Shown above are the amounts of antibody used for immunoprecipitations using a 300-
500μL starting volume for IPs.
197
Bibliography
Ait-Si-Ali, S., Guasconi, V., Fritsch, L., Yahi, H., Sekhri, R., Naguibneva, I., Robin, P.,
Cabon, F., Polesskaya, A. and Harel-Bellan, A. 2004. A Suv39h-dependent
mechanism for silencing S-phase genes in differentiating but not cycling cells.
EMBO J. 23: 605-15.
Bain, G., Ray, W.J., Yao, M. and Gottlieb, D.I. 1994. From embryonal carcinoma cells
to neurons: the P19 pathway. Bioessays 16(5): 343-8.
Barski, A., Cuddapah, S., Cui, K., Roh, T.Y., Schones, D.E., Wang, Z., Wei, G.,
Chepelev, I. and Zhao, K. 2007. High-resolution profiling of histone methylations
in the human genome. Cell 129(4): 823-37.
Bannister, A.J., Zegerman, P., Partridge, J.F., Miska, E.A., Thomas, J.O., Allshire, R.C.,
and Kouzarides, T. 2001. Selective recognition of methylated lysine 9 on histone
H3 by the HP1 chromo domain. Nature. 410(6824): 120-4.
Bannister, A.J., Schneider, R., Myers, F.A., Thorne, A.W., Crane-Robinson, C. and
Kouzarides, T. 2005. Spatial distribution of di- and tri-methyl lysine 36 of histone
H3 at active genes. J Biol Chem. 280(18): 17732-6.
Bernstein, B.E., Kamal, M., Lindblad-Toh, K., Bekiranov, S., Bailey, D.K., Huebert,
D.J., McMahon, S., Karlsson, E.K., Kulbokas, E.J.3rd Gingeras, T.R., et al. 2005.
Genomic maps and comparative analysis of histone modifications in human and
mouse. Cell 120: 169–81.
Bernstein, B.E., Mikkelsen, T.S., Xie, X., Kamal, M., Huebert, D.J., Cuff, J., Fry, B.,
Meissner, A., Wernig, M., Plath, K., Jaenisch, R., Wagschal, A., Feil, R.,
Schreiber,S.L. and Lander, E.S. 2006. A bivalent chromatin structure marks key
developmental genes in embryonic stem cells. Cell 125(2): 315-26.
Bernstein, B.E., Meissner, A. and Lander, E.S. 2007. The mammalian epigenome. Cell
128(4): 669-81.
Biron, V.L., McManus, K.J., Hu, N., Hendzel, M.J. and Underhill, D.A. 2004. Distinct
dynamics and distribution of histone methyl-lysine derivatives in mouse
development. Dev Biol. 276(2): 337-51.
Botuyan, M.V., Lee,J., Ward, I.M., Kim, J., Thompson, J.R., Chen,J. and Mer, G.
2006. Structural Basis for the Methylation State-Specific Recognition of Histone
H4-K20 by 53BP1 and Crb2 in DNA Repair. Cell 127: 1361-73.
198
Briggs, S.D., Xiao, T., Sun, Z.W., Caldwell, J.A., Shabanowitz, J., Hunt, D.F., Allis, C.D.
and Strahl, B.D. 2001. Gene silencing: Trans-histone regulatory pathway in
chromatin. Nature 418: 498.
Brown, M.A., Sims 3
rd
, R.J., Gottlieb, P.D. and Tucker, P.W. 2006. Identification and
characterization of Smyd2: a split SET/MYND domain-containing histone H3
lysine 36-specific methyltransferase that interacts with the Sin3 histone
deacetylase complex. Mol Cancer 5: 26-36.
Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, P., Jones, R.S.
and Zhang, Y. 2002. Role of histone H3 lysine 27 methylation in Polycomb-group
silencing. Science 298(5595): 1039-43.
Cao, R. and Zhang, Y. 2004. The functions of E(Z)/EZH2-mediated methylation of
lysine 27 in histone H3. Curr Opin Genet Dev. 14: 155–64.
Carrozza, M.J., Li, B., Florens, L., Suganuma, T., Swanson, S.K., Lee, K.K., Shia, W.J.,
Anderson, S., Yates, J., Washburn, M.P., and Workman, J.L. 2005.
Histone H3 Methylation by Set2 Directs Deacetylation of Coding Regions by
Rpd3S to Suppress Spurious Intragenic Transcription. Cell. 123: 581–92
Chadwick, R.B., Jiang, G.-L., Bennington, G.A., Yuan, B., Johnson, C.K., Stevens,
M.W., Niemann, T.H., Peltomaki, P., Huang, S. and de la Chapelle, A. 2000.
Candidate tumor suppressor RIZ is frequently involved in colorectal
carcinogenesis. Proc Natl Acad Sci. 97: 2662–7.
Chen, D., Ma, H., Hong, H., Koh, S.S., Huang, S.M., Schurter, B.T., Aswad, D.W. and
Stallcup, M.R. 1999. Regulation of transcription by a protein methyltransferase.
Science 284: 2174–7.
Chen, J.D. and Evans, R.M. 1995 A transcriptional co-repressor that interacts with
nuclear hormone receptors. Nature. 377: 454–7.
Collins, R.E., Tachibana, M., Tamaru, H., Smith, K.M., Jia, D., Zhang, X., Selker, E.U.,
Shinkai, Y. and Cheng, X. 2005. In vitro and in vivo analysis of a Phe/Tyr switch
controlling product specificity of histone lysine methyl transferases. J Biol Chem.
280(7): 5563-70.
Couture, J.F., Collazo, E., Brunzelle, J.S. and Trievel, R.C. 2005. Structural and
functional analysis of SET8, a histone H4 Lys-20 methyltransferase. Genes Dev.
19(12): 1455-65.
199
Czermin, B., Melfi, R., McCabe, D., Seitz, V., Imhof, A. and Pirrotta, V. 2002.
Drosophila enhancer of Zeste/ESC complexes have a histone H3
methyltransferase activity that marks chromosomal polycomb sites. Cell 111:
185–196.
DeLange, R.J. and Smith, E.L. 1971. Histones: structure and function. Annu Rev
Biochem. 40:279-314.
Dillon, S.C., Zhang, X., Trievel, R.C. and Cheng, X. 2005. The SET-domain protein
superfamily: protein lysine methyltransferases. Genome Biol. 6(8): 227.1-.10
Dover, J., Schneider, J., Tawain-Boateng, M.A., Wood, A., Dean, K., Johnston, M. and
Shilatifard, A. 2002. Methylation of histone H3 by COMPASS requires
ubiquitination of H2B by Rad6. J Biol Chem. 277: 28368-71.
Elagib, K.E., Racke, F.K., Mogass, M., Khetawat, R., Delehanty, L.L. and Goldfarb,
A.N. 2003. RUNX1 and GATA-1 coexpression and cooperation in
megakaryocytic differentiation. Blood 101(11): 4333-41.
Edmunds, J.W., Mahadevan, L.C. and Clayton, A.L. 2008. Dynamic histone H3
methylation during gene induction: HYPB/Setd2 mediates all H3K36
trimethylation. EMBO J. 27(2): 406-20.
Esteve, P.O., Chin, H.G., Smallwood, A., Feehery, G.R., Gangisetty, O., Karpf, A.R.,
Carey, M.F. and Pradhan, S. 2006. Direct interaction between DNMT1 and G9a
coordinates DNA and histone methylation during replication. Genes Dev. 20:
3089–3103.
Fang, J., Feng, Q., Ketel, C.S., Wang, H., Cao, R., Xia, L., Erdjument-Bromage, H.,
Tempst, P., Simon, J.A. and Zhang, Y. 2002. Purification and functional
characterization of SET8, a nucleosomal histone H4-lysine 20-specific
methyltransferase. Curr Biol. 12(13): 1086-99.
Feng, Q., Wang, H., Ng, H.H., Erdjument-Bromage, H., Tempst, P., Struhl, K. and
Zhang, Y. 2002. Methylation of H3-lysine 79 is mediated by a new family of
HMTases without a SET domain. Curr Biol. 12(12): 1052-8.
Fischle, W., Wang, Y., Jacobs, S.A., Kim, Y., Allis, C.D. and Khorasanizadeh, S. 2003.
Molecular basis for the discrimination of repressive methyl-lysine marks in
histone H3 by Polycomb and HP1 chromodomains. Genes Dev. 17(15): 1870-81.
200
Flanagan, J.F., Mi, L.Z., Chruszcz, M., Cymborowski, M., Clines, K.L., Kim, Y., Minor,
W., Rastinejad, F. and Khorasanizadeh, S. 2005. Double chromodomains
cooperate to recognize the methylated histone H3 tail. Nature 438: 1181–85.
Fraga, M.F., Ballestar, E., Villar-Garea, A., Boix-Chornet, M., Espada, J., Schotta, G.,
Bonaldi, T., Haydon, C., Ropero, S., Petrie, K., Iyer, N.G., Perez-Rosado, A.,
Calvo, E., Lopez, J.A., Cano, A., Calasanz, M.J., Colomer, D., Piris, M.A., Ahn,
N., Imhof, A., Caldas, C. and Jenuwein, T. 2005. Loss of acetylation at Lys16 and
trimethylation at Lys20 of histone H4 is a common hallmark of human cancer.
Nat Genet. 37(4): 391-400.
Gangloff, Y.G., Pointud, J.C., Thuault, S., Carre, L., Romier, C., Muratoglu, S., Brand,
M., Tora, L., Couderc, J.L. and Davidson, I. 2001. The TFIID components human
TAF(II)140 and Drosophila BIP2 (TAF(II)155) are novel metazoan homologues
of yeast TAF(II)47 containing a histone fold and a PHD finger. Mol Cell Biol.
21(15): 5109-21.
Gateff, E., Loffler, T. and Wismar, J. 1993. A temperature-sensitive brain tumor
suppressor mutation of Drosophila melanogaster: developmental studies and
molecular localization of the gene. Mech Dev. 41(1): 15-31.
Gerber, M. and Shilatifard, A. 2003. Transcriptional elongation by RNA polymerase II
and histone methylation. J Biol Chem. 278(29): 26303-6.
Gopalakrishnan, T.V. and Anderson, W.F. 1979. Mouse erythroleukemia cells. Methods
Enzymol. 58: 506-11.
Grewal, S.I. and Rice, J.C. 2004. Regulation of heterochromatin by histone methylation
and small RNAs. Curr Opin Cell Biol. 16: 230–38.
Guenther, M.A., Levine, S.S., Boyer, L.A., Jaenisch, R. and Young, R.A. 2007. A
chromatin landmark and transcription initiation at most promoters in human cells.
Cell 130: 77-88.
He, L., Yu, J.X., Liu, L., Buyse, I.M., Wang, M.-S. Yang, Q.-C., Nakagawara, A.,
Brodeur, G.M., Shi, Y.E. and Huang, S. 1998. RIZ1, but not the alternative RIZ2
product of the same gene, is underexpressed in breast cancer, and forced RIZ1
expression causes G2–M cell cycle arrest and/or apoptosis. Cancer Res. 58:
4238–44.
201
Houston, S.I., McManus, K.J., Adams, M.M., Sims, J.K., Carpenter, P.B., Hendzel, M.J.,
and Rice, J.C. 2008. Catalytic function of the PR-Set7 histone H4 lysine 20
monomethyltransferase is essential for mitotic entry and genomic stability. J Biol
Chem. (in press)
Hu, M., Krause, D., Greaves, M., Sharkis, S., Dexter, M., Heyworth, C. and Enver, T.
1997. Multilineage gene expression precedes commitment in the hemopoietic
system. Genes Dev. 11: 774-85.
Huang, S., Litt, M. and Felsenfeld, G. 2005. Methylation of histone H4 by arginine
methyltransferase PRMT1 is essential in vivo for many subsequent histone
modifications. Genes Dev. 19: 1885-93.
Huen, S., Sy, S.M., van Deursen, J.M. and Chen, J. 2008. Direct interaction between
Set8 and PCNA couples H4-K20 methylation with DNA replication. J Biol Chem.
(in press).
Huyen, Y., Zgheib, O., Ditullio, R.A. Jr., Gorgoulis, V.G., Zacharatos, P., Petty, T.J.,
Sheston, E.A., Mellert, H.S., Stavridi, E.S. and Halazonetis, T.D. 2004.
Methylated lysine 79 of histone H3 targets 53BP1 to DNA double-strand breaks.
Nature. 432(7015): 406-11.
Ichikawa, M., Asai, T., Chiba, S., Kurokawa, M. and Ogawa, S. 2004a. Runx1/AML-1
ranks as a master regulator of adult hematopoiesis. Cell Cycle 3(6): 722-4.
Ichikawa, M., Asai, T., Saito, T., Seo, S., Yamazaki, I., Yamagata, T., Mitani, K., Chiba,
S., Ogawa, S., Kurokawa, M. and Hirai, H. 2004b. AML-1 is required for
megakaryocytic maturation and lymphocytic differentiation, but not for
maintenance of hematopoietic stem cells in adult hematopoiesis. Nat Med. 10(3):
299-304.
Ishizuka, T. and Lazar, M.A. 2003. The N-CoR/Histone Deacetylase 3 complex is
required for repression by thyroid hormone receptor. Mol Cell Biol. 23(15): 5122-
31.
Jacobs, S.A., Taverna, S.D, Zhang, Y., Briggs, S.D., Li, J., Eissenberg, J.C. and Allis,
C.D. 2001. Specificity of the HP1 chromo domain for the methylated N-terminus
of histone H3. EMBO J. 20(18); 5232-41.
Jenuwein, T. 2001. Re-SET-ting heterochromatin by histone methyltransferases.
Trends Cell Biol. 11(6): 266-73.
202
Jenuwein, T. and Allis, C.D. 2001. Translating the Histone Code. Science 293: 1074-80
Jorgensen, S., Elvers, I., Trelle, M.B., Menzel, T., Eskilden, M., Jensen, O.N., Helleday,
T., Helin, K. and Sorensen, C.S. 2007. The histone methyltransferase SET8 is
required for S-phase progression. J Cell Biol. 179(7): 1337-45.
Joshi, A.A. and Struhl, K. 2005. Eaf3 chromodomain interaction with methylated H3-
K36 links histone deacetlyation to Pol II elongation. Mol Cell 20: 971-78.
Julien, E. and Herr, W. 2004. A Switch in Mitotic Histone H4 Lysine 20 Methylation
Status Is Linked to M Phase Defects upon Loss of HCF-1. Mol Cell 14(6): 713-
25.
Kadosh, D., and Struhl, K. 1997. Repression by Ume6 involves recruitment of a complex
containing Sin3 corepressor and Rpd3 histone deacetylase to target promoters.
Cell. 89: 365–71.
Kadosh, D., and Struhl, K. 1998. Targeted recruitment of the Sin3-Rpd3 histone
deacetylase complex generates a highly localized domain of repressed chromatin
in vivo. Mol. Cell. Biol. 18: 5121–27
Kalakonda, N., Fischle, W., Boccuni, P., Gurvich, N., Hoya-Arias, R., Zhao, X., Miyata,
Y., MacGrogan, D., Zhang, J., Sims, J.K., Rice, J.C. and Nimer, S.D. 2008.
Histone H4 lysine 20 monomethylation promotes transcriptional repression by
L3MBTL1. Oncogene. (in press)
Kanellopoulou, C., Muljo, S.A., Kung, A.L., Ganesan, S., Drapkin, R., Jenuwein, T.,
Livingston, D.M. and Rajewsky, K. 2005. Dicer-deficient mouse embryonic stem
cells are defective in differentiation and centromeric silencing. Genes Dev. 19:
489–501.
Kaplan, C.D., Laprade, L., and Winston, F. 2003. Transcription elongation factors repress
transcription initiation from cryptic sites. Science. 301: 1096–9.
Karachentsev, D., Sarma, K., Reinberg, D. and Steward, R. 2005. PR-Set7 dependent
methylation of histone H4 Lys20 functions in repression of gene expression and is
essential for mitosis. Genes Dev. 19(4): 431-5.
203
Keogh, M.C., Kurdistani, S.K., Morris, S.A., Ahn, S.H., Podolny, V., Collins, S.R.,
Schuldiner, M., Chin, K., Punna, T., Thompson, N.J., Boone, C., Emili, A.,
Weissman, J.S., Hughes, T.R., Strahl, B.D., Grunstein, M., Greenblatt, J.F.,
Buratowski, S. and Krogan, N.J. 2005. Cotranscriptional set2 methylation of
histone H3 lysine 36 recruits a repressive Rpd3 complex. Cell. 123(4): 593-605
Kim, J., Daniel, J., Espejo, A., Lake, A., Krishna, M., Xia, L., Zhang, Y. and Bedford,
M.T. 2006. Tudor, MBT and chromo domains gauge the degree of lysine
methylation. EMBO Rep. 7(4): 397-403.
Kim, K.-C., Geng, L. and Huang, S. 2003. Inactivation of a histone methyltransferase by
mutations in human cancers. Cancer Res. 63: 7619-23.
Kondo Y. and Issa, J-P.J. 2003. Enrichment for Histone H3 Lysine 9 methylation at Alu
repeats in human cells. J Biol Chem. 278(30): 27658-62.
Kristjuhan, A., and Svejstrup, J.Q. 2004. Evidence for distinct mechanisms facilitating
transcript elongation through chromatin in vivo. EMBO J. 23: 4243–52.
Kristjuhan, A., Walker, J., Suka, N., Grunstein, M., Roberts, D., Cairns, B.R., and
Svejstrup, J.Q. 2002. Transcriptional inhibition of genes with severe histone h3
hypoacetylation in the coding region. Mol. Cell. 10: 925–33.
Kurokawa, M. 2006. AML1/Runx1 as a versatile regulator of hematopoiesis: regulation
of its function and a role in adult hematopoiesis. Int J Hematol. 84(2): 136-42.
Kuzmichev, A., Nishioka, K., Erdjument-Bromage, H., Tempst, P., and Reinberg, D.
2002. Histone methyltransferase activity associated with a human multiprotein
complex containing the Enhancer of Zeste protein. Genes Dev. 16(22): 2893-
2905.
Laslett, A., Grimmond, S., Gardiner, B., Stamp, L., Lin, A., Hawes, S.M., Wormald, S.,
Nikolic-Paterson, D., Haylock, D., and Pera, M.F. 2007. Transcriptional analysis
of early lineage commitment in human embryonic stem cells. BMC Dev Biol. 7:12
Lachner, M., O’Carroll, D., Rea, S., Mechtler, K. and Jenuwein, T. 2001. Methylation of
histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410: 116-20.
Lehnertz, B., Ueda, Y., Derijck, A.A., Braunschweig, U., Perez-Burgos, L., Kubicek, S.,
Chen, T., Li, E., Jenuwein, T. and Peters, A.H. 2003. Suv39h-mediated histone
H3 lysine 9 methylation directs DNA methylation to major satellite repeats at
pericentric heterochromatin. Curr Biol. 13:1192–1200.
204
Li, H., Rauch, T., Chen, Z.X., Szabo, P.E., Riggs, A.D. and Pfeifer, G.P. 2006. The
histone methyltransferase SETDB1 and the DNA methyltransferase DNMT3A
interact directly and localize to promoters silenced in cancer cells. J Biol Chem.
281: 19489–19500.
Li, B., Gogol, M., Carey, M., Pattenden, S.G., Seidel, C. and Workman, J.L. 2007.
Infrequently transcribed long genes depend on the Set2/Rpd3S pathway for
accurate transcription. Genes Dev. 21(11):1422-30.
Li, H., Fischle, W., Wang, W., Duncan, E.M., Liang, L., Murakami-Ishibe, S., Allis, C.D.
and Patel, D.J. 2007. Structural basis for lower lysine methylation state-specific
readout by MBT repeats of L3MBTL1 and an engineered PHD finger. Mol Cell
28(4): 677-91.
Luger, K., Mader, A.W., Richmond, R.K., Sargent, D.F. and Richmond, T.J. 1997.
Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature
389(6648): 251-60.
Luger, K. and Richmond, T.J. 1998. The histone tails of the nucleosome. Curr Opin
Genet Dev. 8(2): 140-6.
Martens, J.H., O’Sullivan, R.J., Braunschweig, U., Opravil, S., Radolf, M., Steinlein, P.
and Jenuwein, T. 2005. The profile of repeat-associated histone lysine
methylation states in the mouse epigenome. EMBO J. 24: 800–12.
Martin, C. and Zhang, Y. 2005. The diverse functions of histone lysine methylation. Nat
Rev Mol Cell Biol. 6: 838-49.
Maurer-Stroh, S., Dickens, N.J., Hughes-Davies, L., Kouzarides, T., Eisenhaber, F. and
Ponting, C.P. 2003. The Tudor domain 'Royal Family': Tudor, plant Agenet,
Chromo, PWWP and MBT domains. Trends Biochem Sci. 28(2): 69-74.
Mikkelsen, T.S., Ku, M., Jaffe, D.B., Issac, B., Lieberman, E., Giannoukos, G., Alvarez,
P., Brockman, W., Kim, T.K., Koche, R.P., Lee, W., Mendenhall, E., O'Donovan,
A., Presser, A., Russ, C., Xie, X., Meissner, A., Wernig, M., Jaenisch, R.,
Nusbaum, C., Lander, E.S. and Bernstein, B.E. 2007. Genome-wide maps of
chromatin state in pluripotent and lineage-committed cells. Nature 448 (7153):
553-60.
Min, J., Allali-Hassani, A., Nady, N., Qi, C., Ouyang, H., Liu, Y., MacKenzie, F.,
Vedadi, M. and Arrowsmith, C.H. 2007. L3MBTL1 recognition of mono- and
dimethylated histones. Nat Struct Mol Biol. 14(12): 1229-30.
205
Min, J., Zhang, Y. and Xu, R.M. 2003. Structural basis for specific binding of Polycomb
chromodomain to histone H3 methylated at Lys 27. Genes Dev. 17(15): 1823-28.
Muller, J., Hart, C.M., Francis, N.J., Vargas, M.L., Sengupta, A., Wild, B., Miller, E.L.,
O'Connor, M.B., Kingston, R.E. and Simon, J.A. 2002. Histone methyltransferase
activity of a Drosophila Polycomb group repressor complex. Cell 111(2): 197-
208.
Nielsen, S.J., Schneider, R., Bauer, U.M., Bannister, A.J., Morrison, A., O’Carroll, D.,
Firestein, R., Cleary, M., Jenuwein, T., Herrera, R.E. and Kouzarides, T. 2001. Rb
targets histone H3 methylation and HP1 to promoters. Nature. 412(6846): 561-5.
Nelson, J.D., Denisenko, O., Sova, P. and Bomsztyk, K. 2006. Fast chromatin
immunoprecipitation assay. Nucleic Acids Res. 34(1): e2.
Ng, H.H., Feng, Q., Wang, H., Erdjument-Bromage, H., Tempst, P., Zhang, Y., and
Struhl, K. 2002a. Lysine methylation within the globular domain of histone H3 by
Dot1 is important for telomeric silencing and Sir protein association. Genes Dev.
16: 1518–27.
Ng, H.H., Xu, R-M., Zhang, Y. and Struhl, K. 2002b. Ubiquitination of histone H2B is
required for efficient Dot1-mediated methylation of histone H3 lysine 79. J Biol
Chem. 277: 34655-7.
Ng, H.H, Ciccone, D.N., Morshead, K.B. Oettinger, M.A. and Struhl, K. 2003. Lysine-
79 of histone H3 is hypomethylated at silenced loci in yeast and mammalian cells:
a potential mechanism for position-effect variegation. Proc Natl Acad Sci. 100(4):
1820-5.
Nishio, H., and Walsh, M.J. 2004. CCAAT displacement protein/cut homolog recruits
G9a histone lysine methyltransferase to repress transcription. Proc Natl Acad Sci.
101: 11257–62.
Nishioka, K., Rice, J.C., Sarma, K., Erdjument-Bromage, H., Werner, J., Wang, Y.,
Chuikov, S., Valenzuela, P., Tempst, P., Steward, R., Lis, J.T., Allis, C.D. and
Reinberg, D. 2002. PR-Set7 is a nucleosome-specific methyltransferase that
modifies lysine 20 of histone H4 and is associated with silent chromatin. Mol Cell
9(6): 1201-13.
Niwa, H., Toyooka, Y., Shimosato, D., Strumpf, D., Takahashi, K., Yagi, R. and Rossant,
J. 2005. Interaction between Oct3/4 and Cdx2 determines trophectoderm
differentiation. Cell 123: 917-29.
206
North, T.E., Stacy, T., Matheny, C.J., Speck, N.A. and de Bruijn, M.F. 2004. Runx1 is
expressed in adult mouse hematopoietic stem cells and differentiating myeloid
and lymphoid cells, but not in maturing erythroid cells. Stem Cells 22(2): 158-68.
Okuda, T., van Deursen, J., Hiebert, S.W., Grosveld, G. and Downing, J.R. 1996. AML1,
the target of multiple chromosomal translocations in human leukemia, is essential
for normal fetal liver hematopoiesis. Cell 84(2): 321-30.
Pena, P.V., Davrazou, F. Shi, X., Walter, K.L., Verkhusha, V.V., Gozani, O., Zhao, R.
and Kutateladze, T.G. 2006 . Molecular mechanism of histone H3K4me3
recognition by plant homeodomain of ING2. Nature. 442(7098): 100-3.
Papp, B. and Muller, J. 2006. Histone trimethylation and the maintenance of
transcriptional ON and OFF states by trxG and PcG proteins. Genes Dev. 20(15):
2041-54.
Pesavento, J.J., Yang, H., Kelleher, N.L. and Mizzen, C.A. 2008. Certain and progressive
methylation of histone H4 at lysine 20 during the cell cycle. Mol Cell Biol. 28(1):
468-86.
Peters, A.H., O’Carroll, D., Scherthan, H., Mechtler, K., Sauer, S., Schofer, C.,
Weipoltshammer, K., Pagani, M., Lachner, M., Kohlmaier, A., Opravil, S., Doyle,
M., Sibilia, M. and Jenuwein, T. 2001. Loss of the Suv39h histone
methyltransferases impairs mammalian heterochromatin and genome stability.
Cell 107(3): 323-37.
Peterson, C.L. and Laniel, M.A. 2004. Histones and histone modifications. Curr Biol.
14(14): R546-51.
Rao, B., Shibata, Y., Strahl, B.D. and Lieb, J.D. 2005. Dimethylation of histone H3 at
lysine 36 demarcates regulatory and nonregulatory chromatin genomewide. Mol
Cell Biol. 25: 9447–59.
Rea, S., Eisenhaber, F., O’Carroll, D., Strahl, B.D., Sun, Z.W., Schmid, M., Opravil, S.,
Mechtler, K., Ponting, C.P., Allis, C.D. and Jenuwein, T. 2000. Regulation of
chromatin structure by site-specific histone H3 methyltransferases. Nature 406:
593–99.
Reubinoff, B.E., Itsykson, P., Turetsky, T., Pera, M.F., Reinhartz, E., Itzik, A. and Ben-
Hur,T. 2001. Neural progenitors from human embryonic stem cells. Nat
Biotechnol. 19: 1134-40.
207
Reuter, G. and Spierer, P. 1992. Position effect variegation and chromatin proteins.
Bioessays 14: 605–12.
Rice, J.C. and Allis, C.D. 2001. Histone methylation versus histone acetylation: new
insights into epigenetic regulation. Curr Opin Cell Biol. 13(3): 263-73.
Rice, J.C., Briggs, S.D., Ueberheide, B., Barber, C.M., Shabanowitz, J., Hunt, D.F.,
Shinkai, Y. and Allis, C.D. 2003. Histone methyltransferases direct different
degrees of methylation to define distinct chromatin domains. Mol Cell 12(6):
1591-8.
Rice, J.C., Nishioka, K., Sarma, K., Steward, R., Reinberg, D. and Allis, C.D. 2002.
Mitotic-specific methylation of histone H4 Lys 20 follows increased PR- Set7
expression and its localization to mitotic chromosomes. Genes Dev. 16(17): 2225-
30.
Rice, K.L., Hormaeche, I. and Licht, J.D. 2007. Epigenetic regulation of normal and
malignant hematopoiesis. Oncogene 26(47): 6697-6714.
Roopra, A., Qazi, R., Schoenike, B., Daley, T.J. and Morrison, J.F. 2004. Localized
domains of G9a-mediated histone methylation are required for silencing of
neuronal genes. Mol Cell 14: 727–38.
Rutherford, T.R., Clegg, J.B. and Weatherall, D.J. 1979. K562 human leukaemic cells
synthesise embryonic haemoglobin in response to haemin. Nature 280(5718):
164-5.
Sanders, S.L., Portoso, M., Mata, J., Bahler, J., Allshire, R.C. and Kouzarides, T. 2004.
Methylation of histone H4 lysine 20 controls recruitment of Crb2 to sites of DNA
damage. Cell. 119(5): 603-14.
Santos-Rosa, H., Schneider, R., Bannister, A.J., Sherriff, J., Bernstein, B.E., Emre, N.C.,
Schreiber, S.L., Mellor, J. and Kouzarides, T. 2002. Active genes are tri-
methylated at K4 of histone H3. Nature 419: 407-11.
Schneider, R., Bannister, A.J., Myers, F.A., Thorne, A.W., Crane-Robinson, C. and
Kouzarides, T. 2004. Histone H3 lysine 4 methylation patterns in higher
eukaryotic genes. Nat Cell Biol. 6: 73–77.
Schotta, G., Lachner, M., Sarma, K., Ebert, A., Sengupta, R., Reuter, G., Reinberg, D.
and Jenuwein, T. 2004. A silencing pathway to induce H3-K9 and H4-K20
trimethylation at constitutive heterochromatin. Genes Dev. 18(11): 1251-62.
208
Schultz, D.C., Ayyanathan, K., Negorev, D., Maul, G.G. and Rauscher, F.J. 3
rd
2002.
SETDB1: a novel KAP-1-associated histone H3, lysine 9-specific
methyltransferase that contributes to HP1-mediated silencing of euchromatic
genes by KRAB zinc-finger proteins. Genes Dev. 16(8): 919-32.
Shi, X., Hong, T., Walter, K.L., Ewalt, M., Michishita, E., Hung, T., Carney, D., Pena, P.,
Lan, F., Kaadige, M.R., Lacoste, N., Cayrou, C., Davrazou, F., Saha, A., Cairns,
B.R., Ayer, D.E., Kutateladze, T.G., Shi, Y., Cote, J., Chua, K.F. and Gozani, O.
2006. ING2 PHD domain links histone H3 lysine 4 methylation to active gene
repression. Nature 442(7098): 96-9.
Shilatifard, A. 2006. Chromatin modifications by methylation and ubiquitination:
implications in the regulation of gene expression. Annu Rev Biochem.75: 243-69.
Shogren-Knaak, M., Ishii, H., Sun, J.M., Pazin, M.J., Davie, J.R. and Peterson, C.L.
2006. Histone H4-K16 acetylation controls chromatin structure and protein
interactions. Science. 311(5762): 844-7.
Smith A. 2005. The battlefield of pluripotency. Cell 123: 757-60.
Simon, J.A. and Tamkun, J.W. 2002. Programming off and on states in chromatin:
mechanisms of Polycomb and trithorax group complexes. Curr Opin Gene Dev.
12: 210-8.
Sims, J.K., Houston, S.I., Magazinnik, T. and Rice, J.C. 2006. A trans-tail histone code
defined by monomethylated H4 Lys-20 and H3 Lys-9 demarcates distinct regions
of silent chromatin. J Biol Chem. 281(18): 12760-6.
Sims, J.K., Magazinnik, T., Houston, S.I., Wu, S., and Rice, J.C. "Histone Modifications
and Epigenetics." Epigenetics. Ed. Jorg Tost. Horizon Scientific Press, 2008.
Sims, J.K. and Rice, J.C. 2008. PR-Set7 establishes a repressive trans-tail histone code
that regulates differentiation. Mol. Cell. Biol. (in press).
Sims, R.J., 3rd Chen, C.F., Santos-Rosa, H., Kouzarides, T., Patel, S.S. and Reinberg, D.
2005. Human but not yeast CHD1 binds directly and selectively to histone H3
methylated at lysine 4 via its tandem chromodomains. J Biol Chem. 280: 41789–
41792.
209
Sripathy, S.P., Stevens, J. and Schultz, D.C. 2006. The KAP1 corepressor functions to
coordinate the assembly of de novo HP1-demarcated microenvironments of
heterochromatin required for KRAB zinc finger protein-mediated transcriptional
repression. Mol Cell Biol. 26(22): 8623-38.
Strahl, B.D. and Allis, C.D. 2000. The language of covalent histone modifications.
Nature 403(6765): 41-45.
Strahl, B.D., Grant, P.A., Briggs, S.D., Sun, Z-W., Bone, J.R., Caldwell, J.A., Mollah, S.,
Cook, R.G, Shabanowitz, J., Hunt, D.F. and Allis, C.D. 2002. Set2 is a
nucleosomal histone H3-selective methyltransferase that mediates transcriptional
repression. Mol Cell Biol. 22(5): 1298-1306.
Steele-Perkins, G., Fang, W., Yang, X., Van Gele, M., Carling,T., Gu, J., Buyse, I.M.,
Fletcher, J.A., Liu, J., Bronson, R., Chadwick, R.B., de la Chapelle, A., Zhang,
X., Speleman, F. and Huang, S. 2001. Tumor formation and inactivation of RIZ1,
an Rb-binding member of a nuclear protein–methyltransferase superfamily. Genes
Dev. 15: 2250–62.
Stock, J.K., Giadrossi, S., Casanova, M., Brookes, E., Vidal, M., Koseki, H., Brockdorff,
N., Fisher, A.G., and Pombo, A. 2007. Ring1-mediated ubiquitination of H2A
restrains poised RNA polymerase II at bivalent genes in mouse ES cells. Nat Cell
Biol. 9(12): 1428-35.
Sullivan, B.A. and Karpen, G.H. 2004. Centromeric chromatin exhibits a histone
modification pattern that is distinct from both euchromatin and heterochromatin.
Nat Struct Mol Biol. 11(11): 1076-83.
Sun, X-J., Wei, J., Wu, X-Y., Hu, M., Wang, L., Wang, H-H., Zhang, Q-H., Chen, S-J.,
Huang, Q-H. and Chen, Z. 2005. Identification and characterization of a novel
human histone H3 lysine 36-specific methyltransferase. J Biol Chem. 280(42):
35261-71.
Tachibana, M., Sugimoto, K., Fukushima, T. and Shinkai, Y. 2001. SET domain-
containing protein, G9a, is a novel lysine-preferring mammalian histone
methytransferase with hyperactivity and specific selectivity to Lysines 9 and 27 of
Histone H3. J Biol Chem. 276(27): 25309-17.
210
Tachibana, M., Sugimoto, K., Nozaki, M., Ueda, J., Ohta, T., Ohki, M., Fukuda, M.,
Takeda, N., Niida, H., Kato, H. and Shinkai, Y. 2002. G9a histone
methyltransferase plays a dominant role in euchromatic histone H3 lysine 9
methylation and is essential for early embryogenesis. Genes Dev. 16(14): 1779-
91.
Tachibana, M., Ueda, J., Fukuda, M., Takeda, N., Ohta, T., Iwanari, H., Sakihama, T.,
Kodama, T., Hamakubo, T. and Shinkai, Y. 2005. Histone methyltransferases G9a
and GLP form heteromeric complexes and are both crucial for methylation of
euchromatin at H3-K9. Genes Dev. 19(7): 815-26.
Talasz, H., Lindner, H.H., Sarg, B. and Helliger, W. 2005. Histone H4-lysine 20
monomethylation is increased in promoter and coding regions of active genes and
correlates with hyperacetylation. J Biol Chem. 280(46): 38814-22.
Tani, T., Ylanne, J. and Virtanen, I. 1996. Expression of megakaryocytic and erythroid
properties in human leukemic cells. Exp Hematol. 24(2): 158-168.
Tardat, M., Murr, R., Herceg, Z., Sardet, C. and Julien, E. 2007. PR-Set7-dependent
lysine methylation ensures genome replication and stability through S phase. J
Cell Biol. 179(7): 1413-26.
Taverna, S.D., Li, H., Ruthenburg, A.J., Allis, C.D. and Patel, D.J. 2007. How chromatin-
binding modules interpret histone modifications: lessons from professional pocket
pickers. Nat Struct Mol Biol. 14(11): 1025-40.
Tetteroo, P.A., Massaro, F., Mulder, A., Schreuder-van Gelder, R. and von dem Borne,
A.E. 1984. Megakaryoblastic differentiation of proerythroblastic K562 cell-line
cells. Leuk Res. 8(2): 197-206.
Trojer, P., Li, G., Sims, R.J., 3rd, Vaquero, A., Kalakonda, N., Boccuni, P., Lee, D.,
Erdjument-Bromage, H., Tempst, P., Nimer, S.D., Wang, Y.H. and Reinberg, D.
2007. L3MBTL1, a histone-methylation-dependent chromatin lock. Cell 129(5):
915-28.
Tryndyak, V.P., Kovalchuk, O. and Pogribny, I.P. 2006. Loss of DNA methylation and
histone H4 lysine 20 trimethylation in human breast cancer cells is associated
with aberrant expression of DNA methyltransferase 1, Suv4-20h2 histone
methyltransferase and methyl-binding proteins. Cancer Biol Ther. 5(1): 65-70.
211
Tschiersch, B., Hofmann, A., Krauss, V., Dorn, R., Korge, G. and Reuter, G. 1994. The
protein encoded by the Drosophila position-effect variagation suppressor gene
Su(var)3-9 combines domains of homeotic gene complexes. EMBO J. 13: 3822-
31.
Vakoc, C.R., Mandat, S.A., Olenchock, B.A. and Blobel, G.A. 2005. Histone H3 lysine 9
methylation and HP1gamma are associated with transcription elongation through
mammalian chromatin. Mol Cell 19(3): 381-91.
Vakoc, C.R., Sachdeva, M.M., Wang, H. and Blobel, G.A. 2006. Profile of histone lysine
methylation across transcribed mammalian chromatin. Mol Cell Biol. 26(24):
9185-95.
Vermeulen, M., Mulder, K.W., Denissov, S., Pijnappel, W.W., van Schaik, F.M., Varier,
R.A., Baltissen, M.P., Stunnenberg, H.G., Mann, M. and Timmers, H.T. 2007.
Selective anchoring of TFIID to nucleosomes by trimethylation of histone H3
lysine 4. Cell 131(1): 58-69.
Vire, E., Brenner, C., Deplus, R., Blanchon, L., Fraga, M., Didelot, C., Morey, L., Van
Eynde, A., Bernard, D., Vanderwinden, J.M., et al. 2006. The Polycomb group
protein EZH2 directly controls DNA methylation. Nature. 439: 871–4.
Wanda, P.E., Lee, L.T. and Howe, C. 1981. A spectrophotometric method for measuring
hemoglobin in erythroleukemic cells (K562). J Histochem Cytochem. 29(12):
1442-4.
Wang, H., An, W., Cao, R., Xia, L., Erdjument-Bromage, H., Chatton, B., Tempst, P.,
Roeder, R.G. and Zhang, Y. 2003. mAM facilitates conversion by ESET of
dimethyl to trimethyl lysine 9 of histone H3 to cause transcriptional repression.
Mol Cell. 12: 475–87.
Wang, H., Wang, L., Erdjument-Bromage, H., Vidal, M., Tempst, P. and Zhang, Y.
2004. Role of H2A ubiquitination in Polycomb silencing. Nature 431: 873-8.
Webb, P., Nguyen, P., Shinsako, J., Anderson, C., Feng, W., Nguyen, M.P., Chen, D.,
Huang, S., Subramanian, S., McKinerney, E., Katzenellenbogen, B.S., Stallcup,
M.R. and Kushner, P.J. 1998. Estrogen Receptor Activation Function 1 Works by
Binding p160 Coactivator Proteins Mol Endocrinol. 12(10): 1605-18.
Widschwendter, M., Fiegl, H., Egle, D., Mueller-Holzner, E., Spizzo, G., Marth, C.,
Weisenberger, D.J., Campan, M., Young, J., Jacobs, I. and Laird, P.W. 2007.
Epigenetic stem cell signature in cancer. Nat Gen. 39(2): 157-8.
212
Wismar, J., Loffler, T., Habtemichael, N., Vof, O., Geissen, M., Zirwes, R., Altmeyer,
W., Sass, H. and Gateff, E. 1995. The Drosophila melanogaster tumor suppressor
gene lethal(3)malignant brain tumor encodes a proline-rich protein with a novel
zinc finger. Mech Dev. 53(1): 141-54.
Wood, A., Krogan, N.J., Dover, J., Schneider, J., Heidt, J., Boateng, M.A., Dean,K.,
Golshani, A., Zhang, Y., Greenblatt, J.F., Johnston, M., and Shilatifard, A. 2003.
Bre1, an E3 Ubiquitin Ligase Required for Recruitment and Substrate Selection of
Rad6 at a Promoter. Mol Cell. 11: 267-74
Wu, R., Terry, A.V., Singh, P.B. and Gilbert, D.M. 2005. Differential subnuclear
localization and replication timing of histone H3 lysine 9 methylation states. Mol
Biol Cell. 16(6): 2872-81.
Wysocka, J., Swigut, T., Xiao, H., Milne, T.A., Kwon, S.Y., Landry, J., Kauer, M.,
Tackett, A.J., Chait, B.T., Badenhorst, P., Wu, C. and Allis, C.D. 2006. A PHD
finger of NURF couples histone H3 lysine 4 trimethylation with chromatin
remodeling. Nature. 442(7098): 86-90.
Xiao, B., Jing, C., Kelly, G., Walker, P.A., Muskett, F.W., Frenkiel, T.A., Martin, S.R.,
Sarma, K., Reinberg, D., Gamblin, S.J. and Wilson, J.R. 2005. Specificity and
mechanism of the histone methyltransferase Pr-Set7. Genes Dev. 19(12): 1444-54.
Xie, M., Shao, G., Buyse, I.M. and Huang, S. 1997. Transcriptional repression mediated
by the PR domain zinc finger gene RIZ. J Biol Chem. 272(42): 26360-6.
Yang, H., Pesavento, J.J., Starnes, T.W., Cryderman, D.E., Wallrath, L.L., Kelleher,
N.L. and Mizzen, C.A. 2008. Preferential dimethylation of histone H4-lysine 20
by Suv4-20. J Biol Chem. (in press)
Yin, Y., Liu, C., Tsai, S.N., Zhou, B., Ngai, S.M. and Zhu, G. 2005. SET8 recognizes the
sequence RHRK20VLRDN within the N terminus of Histone H4 and mono
methylates Lysine 20. J Biol Chem. 280(34): 30025-31.
Yu, J., Li, Y., Ishizuka, T., Guenther, M.G. and Lazar, M.A. 2003. A SANT motif in the
SMRT corepressor interprets the histone code and promotes histone deacetylation.
EMBO J. 22(13): 3403-10.
Zhang, P., Du, J., Sun, B., Dong, X., Xu, G., Zhou, J., Huang, Q., Liu, Q., Hao, Q. and
Ding, J. 2006. Structure of human MRG15 chromo domain and its binding to
Lys36-methylated histone H3. Nucleic Acids Res. 34: 6621–28.
213
Zhang, Y. and Reinberg, D. 2001. Transcription regulation by histone methylation:
interplay between different covalent modifications of the core histone tails. Genes
Dev. 15: 2343-60.
Abstract (if available)
Abstract
In developing multi-cellular organisms, cell fate decisions are largely determined by epigenetic programs that activate and repress specific sets of genes. For eukaryotic gene regulation, the post-translational modifications of the DNA-associated histone proteins are critical in transcriptional control. Here, we describe a novel trans-tail histone code where the monomethylation of histones H4 lysine 20 (H4K20) and H3 lysine 9 (H3K9) defines specific genomic regions of transcriptionally repressed chromatin. Global analysis of this code revealed a defined unidirectional temporal sequence of methylmodifications in which monomethylated H3K9 depends on the H4K20 monomethyltransferase, PR-Set7. Importantly, we have identified a repressor protein, L3MBTL1, which is recruited to and binds monomethylated H4K20 and that this interaction is required to initiate transcriptional repression in vivo. Using expression microarray analysis, we have identified numerous cell type-specific genes involved in certain growth and differentiation pathways whose expression is directly regulated by the H4K20 monomethylation silencing pathway.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Studies of the biological relevance of Histone H4 Lysine 20 monomethylation: discovery of its role in the cell cycle and localization within the human genome
PDF
Identification and characterization of PR-Set7 and histone H4 lysine 20 methylation-associated proteins
PDF
Multiple functions of the PR-Set7 histone methyltransferase: from transcription to the cell cycle
PDF
Epigenetic plasticity of cultured female human embryonic stem cells and regulation of gene expression and chromatin by PR-SET7 mediated H4K20me1
PDF
Determinination of the causal potential of histone modifications on transcription and chromatin structure
PDF
Mixed lineage leukemia proteins (MLLs), their effect as coregulators on target gene expression and global histone methylation
PDF
Using epigenetic toggle switches to repress tumor-promoting gene expression
PDF
Developing new tools to discover and investigate histone H3 proteases
PDF
Regulation of Aurora kinase B and its effect on phosphorylation of G9a/GLP
PDF
DNA hypermethylation: its role in colorectal tumorigenesis and potential clinical applications
PDF
Determining the mechanism and necessity of matrix metalloproteinase 2 (MMP-2) nuclear localization for proficient skeletal muscle differentiation
PDF
Characterization of Drosophila longevity and fecundity regulating genes
PDF
Investigation of a causal role of transposable element activation in vertebrate aging
PDF
Differential role of two coactivators, CCAR1 and CARM1, for dysregulated beta-catenin activity in colorectal cancer cell growth and gene expression
PDF
Investigating the role of SASH1 gene located on chromosome 6 in ovarian cancer
PDF
Molecular mechanism of the recruitment of SWI/SNF chromatin remodeling complex and histone acetyltransferase to estrogen-responsive promoters
PDF
Targeting chromatin modification in human cancer: SMYD3 mediated ERα transcription regulation. Cooperative role between H3.3 and HP1ϒ
PDF
Histone 3 lysine 4 tri-methylation in the regulation of epicardial epithelial-to-mesenchymal transformation
PDF
Cancer-recurrent missense mutations in SET domain alter KMT2A methyltransferase activity
PDF
The mechanism of recruitment of Tip60 to ER target genes
Asset Metadata
Creator
Sims, Jennifer Kae
(author)
Core Title
The role of histone H4 lysine 20 monomethylation in gene expression and differentiation
School
Keck School of Medicine
Degree
Doctor of Philosophy
Degree Program
Biochemistry and Molecular Biology
Degree Conferral Date
2008-08
Publication Date
07/14/2010
Defense Date
04/21/2008
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
differentiation,gene expression,histone H4 lysine 20,OAI-PMH Harvest
Language
English
Advisor
Rice, Judd C. (
committee chair
), An, Woojin (
committee member
), Aparicio, Oscar Martin (
committee member
), Laird, Peter W. (
committee member
)
Creator Email
stemple@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m1343
Unique identifier
UC1317708
Identifier
etd-Sims-20080714 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-88881 (legacy record id),usctheses-m1343 (legacy record id)
Legacy Identifier
etd-Sims-20080714.pdf
Dmrecord
88881
Document Type
Dissertation
Rights
Sims, Jennifer Kae
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
differentiation
gene expression
histone H4 lysine 20