THE JOURNAL OF BIOLOGICAL CHEMISTRY
© 2003 by The American Society for Biochemistry and Molecular Biology, Inc.
Vol. 278, No. 11, Issue of March 14, pp. 8897–8903, 2003
Printed in U.S.A.
The Set2 Histone Methyltransferase Functions through the
Phosphorylated Carboxyl-terminal Domain of RNA Polymerase II*
Received for publication, November 29, 2002, and in revised form, January 2, 2003
Published, JBC Papers in Press, January 2, 2003, DOI 10.1074/jbc.M212134200
Bing Li‡¶, LeAnn Howe‡储, Scott Anderson§, John R. Yates III§, and Jerry L. Workman‡**
From the ‡Howard Hughes Medical Institute, Department of Biochemistry and Molecular Biology, The Pennsylvania State
University, University Park, Pennsylvania 16802-4500, and the §Department of Cell Biology, The Scripps Research
Institute, La Jolla, California 92037
The histone methyltransferase Set2, which specifically methylates lysine 36 of histone H3, has been shown
to repress transcription upon tethering to a heterologous promoter. However, the mechanism of targeting
and the consequence of Set2-dependent methylation
have yet to be demonstrated. We sought to identify the
protein components associated with Set2 to gain some
insights into the in vivo function of this protein. Mass
spectrometry analysis of the Set2 complex, purified using a tandem affinity method, revealed that RNA polymerase II (pol II) is associated with Set2. Immunoblotting
and immunoprecipitation using antibodies against subunits of pol II confirmed that the phosphorylated form
of pol II is indeed an integral part of the Set2 complex.
Gst-Set2 preferentially binds to CTD synthetic peptides
phosphorylated at serine 2, and to a lesser extent, serine
5 phosphorylated peptides, but has no affinity for unphosphorylated CTD, suggesting that Set2 associates
with the elongating form of the pol II. Furthermore, we
show that set2⌬ ppr2⌬ double mutants (PPR2 encodes
TFIIS, a transcription elongation factor) are synthetically hypersensitive to 6-azauracil, and that deletions in
the CTD reduce in vivo levels of H3 lysine 36 methylation. Collectively, these results suggest that Set2 is involved in regulating transcription elongation through
its direct contact with pol II.
Nucleosomal structure, once thought to be quite static, turns
out to be extremely dynamic and tightly regulated by various
nonhistone proteins (1). These protein modulators can be generally categorized into two groups: those that utilize the energy
released from ATP hydrolysis to alter histone-DNA contacts
(2), and those that are capable of modifying histones covalently.
To date, the known histone modifications include acetylation,
methylation, phosphorylation, ubquitinylation, and ADP-ribosylation (3, 4). The consequence of complex histone modification patterns has yet to be fully understood. However, based on
* This work was supported in part by Grant GM47867 from the
National Institute of General Medical Sciences (to J. L. W.) and by
National Center for Research Resources Yeast Center Grant RR11823.
The costs of publication of this article were defrayed in part by the
payment of page charges. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. Section 1734 solely to
indicate this fact.
¶ Postdoctoral associate of the Howard Hughes Medical Institute.
储 Supported by a postdoctoral fellowship from the Canadian Institutes of Health Research.
** Associate investigator of the Howard Hughes Medical Institute. To
whom correspondence should be addressed: Howard Hughes Medical
Institute, 306 Althouse Laboratory, Dept. of Biochemistry and Molecular Biology, The Pennsylvania State Univ., University Park, PA 16802.
Tel.: 814-863-8256; Fax: 814-863-0099; E-mail: jlw10@psu.edu.
This paper is available on line at http://www.jbc.org
a growing body of evidence, a “histone code” hypothesis has
been proposed suggesting that these modifications either directly alter chromatin structure (5) or create a series of molecular “bar codes” at the nucleosome surface for other proteins to
recognize (3, 6).
Most, if not all, chromatin modulators have a somewhat
weak intrinsic affinity for their nucleosomal substrates in vitro.
Nevertheless, the specific pattern of histone modifications and
the particular loci where chromatin remodeling occurs
throughout the genome strongly suggest that these activities
are primarily targeted to certain regions by either DNA bound
activators or repressors (7, 8). Acidic activators have been
shown to be able to target SWI/SNF or Spt-Ada-Gcn5-acetyltranferase to local promoters, thereby facilitating transcription
from nucleosomal templates (9 –12). Likewise, the repressor
Ume6 or co-repressor Ssn6/Tup1 can recruit histone deacetylase complexes to remove the acetylation marks from histone
tails, facilitating transcriptional repression (13–17). In addition to activator/repressor-dependent targeting of chromatinmodifying complexes, histone modifications, as proposed in the
“histone code” hypothesis, can also provide recognition sites for
protein complexes and enzymatic activities. For instance, the
binding of Spt-Ada-Gcn5-acetyltranferase or SWI/SNF to promoter nucleosomes is facilitated by bromodomain recognition of
acetylated histone tails (18, 19). Moreover, phosphorylation of
histone H3 serine 10 stimulates K14 acetylation by Gcn5 on the
same tail (20, 21), suggesting that cross-talk can occur between
different modifications. More strikingly, modifications of different histones can also influence each other in a specific manner.
Ubiquitinylation of histone H2B is required for both methylation of lysine 4 by Set1 and lysine 79 by Dot1, but not vice versa
(22–24).
The carboxyl-terminal domain (CTD)1 of Rpb1, the largest
subunit of the RNA polymerase II, plays an important role in
transcription regulation (25–27). It contains multiple repeats of
a heptapeptide sequence with the consensus of YSPTSPS (27)
and is responsible for the integrity of the RNA polymerase II
holoenzyme (28), preinitiation complex formation (29, 30) and
binding of various mRNA processing factors (31–33). Phosphorylation of the CTD is one of the major regulatory events during
the transcription cycle. It is acutely controlled by four reported
CTD kinases: Kin28 (34, 35), Srb10/Srb11 (36, 37), CTD kinase
I (with Ctk1 being the catalytic subunits) (38), Bur1/Bur2 (39),
and only one known CTD phosphatase, Fcp1 (40). Unphosphorylated pol II interacts with the mediator complex and is assembled into the preinitiation complex during transcription
initiation (30). The CTD is then phosphorylated by CTD ki1
The abbreviations used are: CTD, carboxyl-terminal domain; 6-AU,
6-azauracil; TAP, tandem affinity purification; pol II, RNA polymerase
II.
8897
8898
Set2 Is Associated with pol II
nases to start promoter clearance and elongation, causing dissociation of some factors responsible for preinitiation complex
formation (41) and allowing other factors necessary for elongation and termination to bind (25). Phosphorylation sites predominantly reside at serine 5 and serine 2 of the CTD repeats.
Chromatin immunoprecipitation experiments have shown that
serine 5 phosphorylation is localized to promoter regions at
initiation/early elongation stages, whereas serine 2 phosphorylation is associated with the coding region during the elongation phase (42). Interestingly, recent studies have shown
that the pol II holoenzyme possesses certain intrinsic chromatin-modifying activities that may help pol II elongate through
chromatin. Elp3, a subunit of pol II elongating holoenzyme, has
histone acetyltransferase activity (43). Human RNA polymerase II complex directly interacts with the histone acetyltransferases p300 and PCAF. p300 specifically recognizes the
unphosphorylated form and p300/CBP associated factor
(PCAF) associates with the phosphorylated form (44).
Histone methylation occurs at many residues within all four
histones, implying that it could have a critical impact on chromatin structural regulation (5). Recent identification of novel
histone methyltransferases have implicated histone methylation in many important biological processes, such as heterochromatic silencing, transcriptional activation, and transcriptional repression (7, 45). The catalytic domain of SUV39, the
first identified histone methyltransferase, locates within an
evolutionarily conserved “SET” domain and is flanked by two
other cysteine-rich domains (46). These structural motifs have
been used as a standard to search for other histone methyltransferase candidates. In the yeast, Saccharomyces cerevisiae,
there are 6 SET domain proteins. Set1 and Set2 can methylate
histone H3 lysine 4 and lysine 36 (47–51). In addition, a highcopy disruptor of silencing, Dot1, which has very weak homology with other methyltransferases, has also been found to
methylate histone H3 at lysine 79 in a nucleosomal context
(52–55). The methyltransferase activities of these enzymes are
essential for their related biological function, but how these
chromatin modifying enzymes are tethered to specific loci remains largely unknown. To investigate the potential targeting
mechanism of histone methyltransferases, we chose to first
identify the proteins that are associated with Set2 in vivo.
Mass spectrometry analysis of the Set2 complex purified using
a tandem affinity method revealed that Set2 physically associates with RNA polymerase II. Immunoblotting and immunoprecipitation with specific antibodies against different phosphorylated states of the pol II CTD showed that Set2
methyltransferase activity primarily associates with the elongating form of the RNA polymerase II that is hyperphosphorylated. This implies that methylation mediated by Set2 may be
involved in regulating transcription and elongation. Genetic
interaction between Set2 and transcription elongation factorTFIIS further strengthened this argument. Moreover, the CTD
plays an important role in H3 K36 methylation in vivo. These
data are consistent with the results of three other groups who
independently discovered the interaction of Set2 with the phosphorylated form of RNA polymerase II (56).2,3
MATERIALS AND METHODS
Yeast Strains and Genetic Manipulations—YBL102 (Set2TAP::TRP1) and YBL139 (⌬SET2::URA3) were made by PCR-based
tagging strategy with the vector pBS1479 (59, 60). The TAP tag was
integrated at the carboxyl terminus of the endogenous SET2 loci and
2
N. J. Krogan, M. Kim, A. Tong, A. Gorshani, G. Cagney, V. Canadien, D. Richards, B. Beattie, A. Emili, C. Boone, A. Shilatifard, S.
Buratowski, and J. Greenblatt, submitted for publication.
3
T. Xiao, H. Hall, K. Kizer, Y. Shibata, M. C. Hall, C. H. Borchers,
and B. D. Strahl, submitted for publication.
confirmed by PCR and Western blotting with peroxidase-anti-peroxidase (Sigma), which recognizes the protein A module of the tag. Yeast
strain CMKy80 (ppr2⌬::URA3) is a gift from Dr. Kane and has been
described (61). YBL155 (set2⌬ ppr2⌬) is a haploid strain generated from
a cross between YBL150 (set2⌬::LEU2) and CMKy80. RNA polymerase
II CTD truncation mutants were provided by J. Corden (62). Strain
RPB1, rpb1 9CTD, 10 CTD, and 14 CTD are constructed by introducing
individual plasmid pY1 (containing wild-type CTD), Y1WT (9), pY1WT
(10), or pY1WT (14) into Z26 (MAT alpha his3⌬200 ura3–52 leu2–3,112
rpb1⌬187::HIS3 GAL⫹(pRP112)), then shuffling out plasmid pRP112.
Protein Purification—The TAP purification was performed essentially as described elsewhere (59, 60) with minor modification. Briefly,
6 liters of yeast cells expressing TAP-tagged Set2 were grown in yeast
extract-peptone-dextrose medium at 30 °C to an optical density of about
2 at 600 nm. The cell pellet was resuspended in an extraction buffer (E
buffer) (40 mM HEPES pH 7.5; 350 mM NaCl; 10% glycerol; 0.1% Tween
20, 1 mM sodium orthovanadate; 10 mM NaF; and 1 mM ␤-glycerophosphate, supplemented with complete sets of fresh proteinase inhibitors)
and disrupted in a bead beater (Biospec). Crude whole-cell extracts
were then clarified by ultracentrifugation and directly applied to IgGSepharose and calmodulin resin as described previously (59, 60), with
the exception that all buffers contained 10% glycerol. The purified
complex was resolved on a 10% SDS-PAGE gel and visualized with
silver staining. Excised protein bands were subjected to mass spectrometry as described in previously (63).
Histone Methyltransferase (HMT) Assay—In vitro histone methylation reactions were carried out for 1 h at 30 °C in a 25-␮l system
containing 50 mM Tris-HCl (pH 8.0); 50 mM NaCl; 1 mM EDTA; 1 mM
MgCl2; 1 mM dithiothreitol, 1 ␮l of [3H]methyl S-adenosyl-methionine
(80 Ci/mmol, Amersham Biosciences) with 2 ␮g of HeLa oligonucleosomes or recombinant yeast histone octamers as substrates. Half of the
reaction was then spotted on Whatman P-81 filter paper for liquid
scintillation counting and the other half was separated on an 18%
SDS-PAGE gel, stained with Coomassie Blue, and fluorographed with
En3hance (PerkinElmer Life Sciences) as described previously (64).
Peptide Pull-down Assay—The biotinylated CTD peptides (about 1.5
␮g) were bound to 0.5 mg of streptavidin-coated Dynabeads M280) in 50
␮l of high salt binding buffer (25 mM Tris-HCl pH 8.0; 1 M NaCl; 1 mM
dithiothreitol; 5% glycerol; 0.03% Nonidet P-40) at 4 °C for 2 h. The
protein-bound beads were washed once with high salt binding buffer
and twice with CTD binding buffer (25 mM Tris-HCl pH 8.0; 50 mM
NaCl; 1 mM dithiothreitol; 5% glycerol; 0.03% Nonidet P-40) and finally
resuspended in 50 ␮l of CTD binding buffer. Five hundred nanograms
of bacterially overexpressed GST-Set2 protein was mixed with the
beads and incubated at 4 °C for 1 h on a Dyna-Mixer (Dynal). Following
washing three times with CTD binding buffer, beads were then subjected to an in vitro HMT assay.
6-AU Sensitivity Assay—Wild-type and set2⌬ (YBL150) strains were
transformed with the ARS/CEN plasmid pRS316 prior to the test. All
yeast strains were grown in the synthetic media lacking uracil to an
OD600 of 1, then plated onto ⫺Ura synthetic drop-out plates supplemented with 25 ␮g/ml, 50 ␮g/ml, or 100 ␮g/ml of 6-azauracil (Sigma).
Cells were allowed to grow 2– 4 days at 30 °C before photographs were
taken.
RESULTS
Histone Methyltransferase Set2 Physically Associates with
RNA Polymerase II—Histone methylation and HMTs have
been extensively investigated by different research groups,
however, the mechanism regarding how histone methyltransferases are targeted to specific loci has yet to be carefully
addressed. In an effort to determine a potential targeting
mechanism of histone methyltransferases, we decided to identify proteins associated with a known histone methyltransferase and gain some insights from these protein partners.
Histone methyltransferase Set2 was chosen to start, because
Set2 possesses very robust HMT activity on nucleosomal substrates (48, 55) and has been shown to regulate transcription
on tethering to a heterologous promoter (48).
We took advantage of a tandem affinity purification (TAP)
technique (59, 60) to isolate potential partners of Set2 in vivo.
TAP-tag with a TRP1 marker was PCR-amplified from vector
pBS1479 and integrated to the carboxyl termini of the chromosomal copy of SET2. TAP-tagged Set2 strain was confirmed by
PCR and Western blotting with PAP (Sigma). Yeast whole-cell
Set2 Is Associated with pol II
8899
FIG. 1. Identification of the proteins associated with the histone methyltransferase Set2. A, the Set2 complex was purified from the
yeast strain (YBL102) expressing carboxyl-termini-TAP-tagged Set2. The purified complex was separated on a 10% SDS-PAGE and visualized with
silver staining. A similar purification from an untagged strain (w303a) was performed in parallel and we did not observe any significant bands with
silver staining (data not shown). The excised bands were subjected to mass spectrometry analysis. Identified peptides from indicated bands are
listed in the boxes. The Rpb3 subunit is solely identified by Western blotting using an antibody against yeast Rpb3 (see Fig. 2A). The estimate
molecular mass of the bands labeled with asterisks are matched with the migration pattern of previously purified RNA polymerase II and are
presumably corresponding subunits of RNA polymerase II. B, yeast whole-cell extract (YBL102) and TAP-purified Set2 complex were assayed by
immunoblotting with the monoclonal antibodies 8WG16 (against unphosphorylated CTD of Rpb1), H5 (against serine 2 phosphorylated CTD), and
H14 (against serine 5 phosphorylated CTD), as well as polyclonal antibodies against the Rpb3 subunit, and Taf6 (as a negative control).
extracts from the Set2-TAP strain were prepared and subsequently applied to IgG-Sepharose and calmodulin resin. Purified Set2 complex was then separated on a SDS-PAGE gel and
subjected to protein fingerprinting by mass spectrometry (Fig.
1A). Protein identification of the Set2 complex revealed that
two of the largest components in the Set2 complex are Rpb1
and Rpb2, the two largest subunits of RNA polymerase II.
Moreover, the silver staining pattern of the complex excluding
Set2 (as labeled in Fig. 1A) is very similar to that of previously
purified RNA polymerase II, suggesting that the complex we
isolated by this approach is likely a Set2 bound RNA polymerase II complex.
To confirm the mass spectrometry identification, we performed immunoblotting with some of the available antibodies
against different forms of pol II to determine whether the
corresponding antibodies can recognize the components identified by mass spectrometry. Western blotting using antibody
against the Rpb3 subunit of pol II demonstrated that Rpb3 is
an integral part of the Set2 complex. Taf6 antibodies were used
as a negative control (Fig. 1B). The Rpb1 subunit of pol II has
different isoforms in regard to the phosphorylation status of its
CTD. Monoclonal antibodies H5 (against CTD-serine 2-Pi) and
H14 (against CTD-serine 5-Pi) were able to pick up the low
abundant signals from whole-cell extracts. However, they react
strongly with the TAP purified fraction. Conversely, using antibody 8WG16 (which recognizes unphosphorylated CTD), we
found that although the unphosphorylated CTD is relatively
more abundant in crude extracts, it was not enriched by our
purification (Fig. 2A). We believe that the faint signal in the
Set2-TAP fraction with 8WG16 antibody is probably due to
weak cross-reactivity of this antibody with phosphorylated
forms of CTD (65).
Previously, RNA polymerase II has been purified through a
method that combines conventional chromatograph and immunoaffinity column using 8WG16 antibody (66, 67). This method
efficiently enriches unphosphorylated RNA polymerase II, but
it inevitably loses some protein components that only interact
with phosphorylated polymerase II. This may be the reason
why Set2 was not found in previously isolated pol II. To further
characterize the Set2 methylase-bound RNA polymerase, and
rule out the possibility that Set2 might bind separately to a
subset of pol II, we thus subjected the purified Set2 complex
directly onto a gel filtration column. Eluted fractions were
either subjected to Western blotting to detect the presence of
proteins (Fig. 2A) or assayed for histone methyltransferase
activity (Fig. 2B). The peak fraction containing the bulk of
Rpb1 (H5, serine 2-Pi-CTD) and Rpb3 co-fractionate with histone methyltransferase activity, suggesting that Set2 is an
integral component of a phosphorylated RNA polymerase II
complex. Based on the elution profile of the gel filtration column, we estimate that molecular mass of this complex is about
700 kDa, which is consistent with the previously reported molecular mass of pol II (about 550 – 600 kDa) plus the mass of
Set2-CBP (about 95 kDa) (66, 68, 69). Collectively, these results indicate that Set2 physically associates with the hyperphosphorylated form of RNA polymerase II.
Set2 Specifically Recognizes the Phosphorylated CTD—The
histone methyltransferase Set2 primarily associates with the
phosphorylated form of RNA polymerase II but not the unphosphorylated form, implying that the CTD likely contributes to
this specific interaction. We next sought to determine whether
the CTD is directly involved in this interaction and if the CTD
alone is sufficient to support Set2-pol II interaction. Using a
defined synthetic CTD peptide with different combinations of
phosphorylation at serine 2 and serine 5, Ho and Shuman have
successfully demonstrated the interaction between mammalian
guanylytransferase and the CTD phosphopeptide (70). We thus
adopted a similar system where biotinylated synthetic peptides
consisting of 4 tandem repeats of the CTD heptad sequence
(YSPTSPS) are used. The CTD peptide only contains 4 repeats of
heptapeptides, whereas CTD-serine 2-Pi and CTD-serine 5-Pi
have a phosphate group on all 4 specific serines in each peptide.
These peptides were bound to streptavidin-coated magnetic
beads in a high salt buffer and washed extensively to remove
unbound proteins. Peptide-bound beads were then adjusted to a
low salt buffer and incubated with purified, bacterially overexpressed GST-Set2 protein. Measurement of histone methyltransferase activity was subsequently taken for bound fractions and
8900
Set2 Is Associated with pol II
FIG. 2. Set2 is an integral component of an RNA polymerase II complex. TAP-purified Set2 complex was subjected to gel filtration
chromatography on a mini-Superose 6 column using a Smart system (Amersham Biosciences). All fractions were assayed by immunoblotting (A)
and in vitro HMT assays (B). Fractions not shown in the figure did not contain significant signals in either assay. Sizes of the complexes were
estimated by calibrating the column with molecular mass standards. A, immunoblotting of column fraction with the indicated antibodies. B, HMT
assays with HeLa oligonucleosomes as substrates. Both recombinant GST-Set2 and the Set2 complex can efficiently use HeLa oligonucleosomes
and recombinant yeast histone octamers as substrates, but neither can methylate HeLa core histones (data not shown).
unbound fraction (supernatant) (Fig. 3). Fluorography showed
that both the serine 2-Pi and serine 5-Pi CTD peptide beads can
efficiently pull down HMT activity from solution, whereas unmodified CTD fails to do so (Fig. 3A). Using a more quantitative
liquid scintillation counting assay (Fig. 3B), we demonstrated
that the serine 2-Pi CTD peptide displays relatively stronger
affinity to GST-Set2 than the serine 5-Pi CTD peptide, which is
consistent with the Western blot in Fig. 1B. Therefore, we conclude that phosphorylation of the CTD is sufficient to determine
the binding of Set2 to RNA polymerase II in vitro.
To determine whether the specific interaction between Set2
and RNA polymerase II can serve as a signal to regulate the
recruitment of histone methyltransferase activity of Set2 to its
target in vivo, we measured the relative methylation levels of
histones in RNA polymerase II CTD truncation mutants. Although removal of all the CTD repeats are lethal, deletion of a
significant number of repeats can be tolerated by cells (62, 71).
Three CTD truncation mutants containing 9, 10, and 14 repeats, respectively and an isogenic wild-type strain (with 26
repeats) (62) were tested for the histone methylation level by
Western blotting using specific antibodies against the modified
histones (Fig. 4). We found that there exists a good correlation
between the relative level of methylation of H3 at lysine 36 and
the degree of CTD truncation. In particular, the CTD mutant
containing 9 repeats displays a significant reduction of K36
methylation. This correlation is consistent with growth phenotype of these mutants, where only the mutant containing 9
repeats shows temperature sensitivity (62). More importantly,
this methylation defect occurs specifically at lysine 36, which is
mediated by Set2 (48), but not at lysine 4 (Fig. 4A) where its
corresponding methyltransferase Set1 does not seem to stably
interact with RNA polymerase II (49 –51). This result suggests
that CTD truncation does not cause a general histone methylation defect, and the compromised methylation level of H3 at
lysine 36 is likely caused by reduced recruitment of the histone
methyltransferase Set2 to its chromatin substrates through
the truncated RNA polymerase II CTDs.
Set2 Genetically Interacts with TFIIS and Participates in
Regulating Transcription Elongation—Set2 specifically binds
to hyperphosphorylated RNA polymerase II, which is believed
to be an elongating form of pol II, suggesting that Set2 mediated histone methylation may be involved in transcription elongation. A genetic approach was taken to address this issue in
vivo. Transcription factor TFIIS plays an important role in
regulating elongation. TFIIS can stimulate RNA polymerase II
to cleave its nascent transcripts whenever pol II stalls at the
blocks generated by either DNA or DNA-binding proteins,
thereby facilitating RNA polymerase II to proceed through the
DNA templates (72, 73). Moreover, TFIIS has been shown to
genetically interact with many chromatin-modifying factors,
such as the Spt4/Spt5 elongation complex, the elongator complex (with Elp3 being a histone acetyltransferase), the Spt16/
Cdc68 component of FACT (facilitates chromatin transcription), and SWI/SNF (see review in Ref. 73). This strongly
suggests that chromatin remodeling is involved in the regulation of transcription elongation. Unlike SWI/SNF, where deletion of its subunits is synthetically lethal with deletion of TFIIS
(ppr2⌬), deletion of SET2 in combination with deletion of TFIIS
(ppr2⌬) are viable and grow normally on synthetic media (Fig.
5) and rich media (data not shown).
The drug, 6-AU has been widely used in a phenotypical test
for transcription elongation defects. When metabolized, 6-AU is
Set2 Is Associated with pol II
FIG. 3. Set2 specifically binds to the hyperphosphorylated
CTD peptide. Biotinylated synthetic CTD peptides (1.5 ␮g) were adsorbed to 500 ␮g of streptavidin-coated magnetic beads (Dynal M280).
About 500 ng of recombinant GST-Set2 was incubated either with
magnetic beads alone (Mock) or beads with immobilized peptides, as
indicated, for 1 h at 4 °C. Unbound fraction (Sup) and bound fraction (B)
were subjected by HMT assay with 2 ␮g of HeLa oligonucleosomes.
Reactions were incubated at 30 °C for 1 h with occasional gently flicking
the tubes. Half of the reactions were loaded on an 18% SDS-PAGE gel
for fluorography (A) and the remainder were spotted on Whatman P-81
filters for liquid counting (B).
8901
FIG. 5. Set2 is involved in transcription elongation in vivo.
Yeast strains were grown in ⫺Ura synthetic drop-out media to an OD600
of 1. Cultures were collected and diluted to identical optical densities.
Four microliters of serial dilutions from each culture were spotted on
SD ⫺URA plates or SD ⫺URA supplemented with different concentration of 6-AU. Photographs of the plates were taken after 2 days for
control plates and up to 4 days for strains on 6-AU. Wild-type and set2⌬
were transformed with pRS316, an ARS/CEN plasmid to make them
URA⫹, all other strains contain the URA3 marker. SD, yeast synthetic
drop-out medium.
than the wild-type on 6-AU plates. Nevertheless, ppr2⌬set2⌬
double mutants are synthetically hypersensitive to 6-AU, even
at lower concentrations (Fig. 5, 25 ␮g/ml), indicating that SET2
genetically interacts with PPR2. Very similar observations
were made in the case of the elongator complex in that deletion
of elongator subunits alone only cause marginal phenotypes,
but in combination with deletion of TFIIS result in a more
severe sensitivity to 6-AU (75). The genetic interaction between
TFIIS and Set2 further solidified the argument that Set2 plays
an important role in transcription elongation, presumably
through its association with RNA polymerase II.
DISCUSSION
FIG. 4. Methylation of H3 at lysine 36 is compromised in RNA
polymerase II CTD truncation mutants. A, whole-cell extracts were
prepared from yeast strains indicated. RPB1 contains a wild-type copy
of the RNA polymerase II CTD, whereas rpb1 9 CTD, 10 CTD, and 14
CTD harbor only the plasmid carrying 9, 10, and 14 CTD repeats,
respectively. Western blots were probed with antibodies against dimethylated K36 of histone H3, dimethylated K4 of histone H3, and Taf6
(as a loading control). B, the relative levels of K36 methylation in the
RNA polymerase II CTD truncation mutants are quantified based on
three experiments.
converted to 6-azaUMP, which inhibits the enzyme of de novo
nucleotide synthesis and reduces the level of GTP and UTP
(74). Nucleotide depletion creates an additional obstacle for the
crippled transcription elongation machinery to overcome,
thereby making elongation mutants more sensitive to the drug.
A yeast strain lacking SET2 displays a very modest 6-AU
sensitivity even at relatively high drug concentration (Fig. 5, 50
␮g/ml and 100 ␮g/ml) whereas ppr2⌬ clearly grows more slowly
We have presented biochemical evidence that the histone
methyltransferase Set2 physically associates with the RNA
polymerase II. This interesting observation provides a novel
mechanism for the targeting of histone methyltransferase activity to genes by interaction with the basal transcription machinery. More importantly, this association is highly regulated,
because Set2 binds to the serine 2 phosphorylated CTD, and
the serine 5 phosphorylated CTD to a lesser extent, but shows
little interaction with the unphosphorylated CTD. CTD phosphorylation is well known to be coordinately controlled by
multiple CTD kinases and phosphatases during a transcription
cycle (27). CTD enters preinitiation complex as an unphosphorylated form at the transcription initiation stage (28). Serine 5
of the CTD is subsequently phosphorylated at the promoter
clearance and early elongation stages around promoter regions
(42). With pol II reading along the coding region, serine 2
becomes gradually phosphorylated whereas the phosphate
group at serine 5 is removed (42). Set2 preferentially binds to
the elongating form (serine 2-Pi-CTD) of RNA polymerase, suggesting that it might play a role in pol II elongation through
nucleosomal templates. Its relatively weak interaction with the
serine 5-CTD might also make Set2 able to enter the transcription cycle at an early stage. Furthermore, Set2 genetically
interacts with transcription elongation factor TFIIS, providing
in vivo evidence for the involvement of histone methyltransferases in elongation. This conclusion is also supported by the
earlier observation that K4 methylated histone H3 is primarily
distributed in the coding regions of active genes (76).
The question of how Set2-mediated histone methylation in-
8902
Set2 Is Associated with pol II
fluences transcription elongation still remains unresolved.
Based on recent studies and data presented here, we propose
some working models regarding the function of Set2 in transcription elongation. First, it was noticed that when yeast RNA
polymerase III transcribes through nucleosomes, it creates an
intranucleosomal DNA loop, which causes polymerase to intermittently stall along the nucleosomal DNA templates (77, 78).
In this sense, methylation of histone tails may alter the nucleosomal structure in such a way that RNA polymerase II can
read through its template smoothly and efficiently. Second,
methylation marks can also stimulate the binding of other
chromatin remodeling factors and elongation factors so as to
facilitate elongation. A Brm-containing chromatin remodeling
complex can recognize the histone methylation pattern generated by the Drosophila epigenetic regulator-histone methyltransferase, Ash1, and binds to the methylated peptide in vitro
(79). It would be interesting to determine whether SWI/SNF
prefers to bind to methylated nucleosomes, because its function
has been tightly linked to transcription elongation in an in vivo
study (61). Third, Set2-mediated methylation might also block
the addition of other histone modifications or prevent other
chromatin-modifying complexes from accessing nucleosomes. It
has been demonstrated that methylation of K4 of histone H3
can disrupt the interaction between the histone deacetylase
NuRD and the H3 tail (80, 81). In addition, genome-wide hypoacetylation of chromatin overlaps with H3 K4 hypomethylation (76). Future experiments to address these possibilities will
certainly improve our understanding on the effects of methylation on transcription elongation.
Histone acetylation is dynamically regulated by histone
acetyltransferases and histone deacetylases (82). However, to
date, a histone demethylase has yet to be discovered, which
leaves a controversial debate over whether histone methylation
is dynamic or static (45, 83, 84). It has been proposed that
removal of methylation marks from chromatin is either though
proteolytic digestion of histone tail or total replacement of the
methylated histone (3). Accumulating evidence suggests that
methylation might also be dynamically regulated, based on a
series of observations that methylation is involved in many
transcription system in which genes need to be turned on or off
in a rapid fashion or within one cell cycle (47, 57, 58, 85, 86).
The data presented here also point to the fact that at least some
methylation might be reversible, considering the association of
Set2 with the basal transcription machinery, which often needs
to transcribe the same nucleosomal DNA region for multiple
rounds. It would be inefficient to erase those marks by cleaving
the tail or reassembling nucleosomes. To this end, investigations aimed at finding out the potential mechanism for the
turnover of methylation are an important priority.
Acknowledgments—We gratefully acknowledge Steve Buratowski for
kindly providing the synthetic CTD peptides. We thank Drs. C. Kane
and Joseph Reese for yeast strains and reagents. We also thank the
members of Simpson, Reese, and Tan laboratories for comments and
advice on this study.
REFERENCES
1. Workman, J. L., and Kingston, R. E. (1998) Annu. Rev. Biochem. 67, 545–579
2. Vignali, M., Hassan, A. H., Neely, K. E., and Workman, J. L. (2000) Mol. Cell.
Biol. 20, 1899 –1910
3. Jenuwein, T., and Allis, C. D. (2001) Science 293, 1074 –1080
4. Berger, S. L. (2002) Curr. Opin. Genet. Dev. 12, 142–148
5. Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F., and Richmond, T. J.
(1997) Nature 389, 251–260
6. Strahl, B. D., and Allis, C. D. (2000) Nature 403, 41– 45
7. Kouzarides, T. (2002) Curr. Opin. Genet. Dev. 12, 198 –209
8. Peterson, C. L., and Workman, J. L. (2000) Curr. Opin. Genet. Dev. 10,
187–192
9. Natarajan, K., Jackson, B. M., Zhou, H., Winston, F., and Hinnebusch, A. G.
(1999) Mol Cell 4, 657– 664
10. Brown, C. E., Howe, L., Sousa, K., Alley, S. C., Carrozza, M. J., Tan, S., and
Workman, J. L. (2001) Science 292, 2333–2337
11. Neely, K. E., Hassan, A. H., Wallberg, A. E., Steger, D. J., Cairns, B. R.,
Wright, A. P., and Workman, J. L. (1999) Mol. Cell 4, 649 – 655
12. Yudkovsky, N., Logie, C., Hahn, S., and Peterson, C. L. (1999) Genes Dev. 13,
2369 –2374
13. Watson, A. D., Edmondson, D. G., Bone, J. R., Mukai, Y., Yu, Y., Du, W.,
Stillman, D. J., and Roth, S. Y. (2000) Genes Dev. 14, 2737–2744
14. Vogelauer, M., Wu, J., Suka, N., and Grunstein, M. (2000) Nature 408,
495– 498
15. Kadosh, D., and Struhl, K. (1998) Mol. Cell. Biol. 18, 5121–5127
16. Rundlett, S. E., Carmen, A. A., Suka, N., Turner, B. M., and Grunstein, M.
(1998) Nature 392, 831– 835
17. Kadosh, D., and Struhl, K. (1997) Cell 89, 365–371
18. Hassan, A. H., Neely, K. E., and Workman, J. L. (2001) Cell 104, 817– 827
19. Hassan, A. H., Prochasson, P., Neely, K. E., Galasinski, S. C., Chandy, M.,
Carrozza, M. J., and Workman, J. L. (2002) Cell 111, 369 –379
20. Lo, W. S., Duggan, L., Tolga, N. C., Emre, Belotserkovskya, R., Lane, W. S.,
Shiekhattar, R., and Berger, S. L. (2001) Science 293, 1142–1146
21. Lo, W. S., Trievel, R. C., Rojas, J. R., Duggan, L., Hsu, J. Y., Allis, C. D.,
Marmorstein, R., and Berger, S. L. (2000) Mol. Cell 5, 917–926
22. Ng, H. H., Xu, R. M., Zhang, Y., and Struhl, K. (2002) J. Biol. Chem. 277,
34655–34657
23. Briggs, S. D., Xiao, T., Sun, Z. W., Caldwell, J. A., Shabanowitz, J., Hunt, D. F.,
Allis, C. D., and Strahl, B. D. (2002) Nature 418, 498
24. Sun, Z. W., and Allis, C. D. (2002) Nature 418, 104 –108
25. Kobor, M., and Greenblatt, J. (2002) Biochim. Biophys. Acta 1577, 261–275
26. Conaway, J. W., Shilatifard, A., Dvir, A., and Conaway, R. C. (2000) Trends
Biochem. Sci. 25, 375–380
27. Carlson, M. (1997) Annu. Rev. Cell. Dev. Biol. 13, 1–23
28. Kim, Y. J., Bjorklund, S., Li, Y., Sayre, M. H., and Kornberg, R. D. (1994) Cell
77, 599 – 608
29. Usheva, A., Maldonado, E., Goldring, A., Lu, H., Houbavi, C., Reinberg, D., and
Aloni, Y. (1992) Cell 69, 871– 881
30. Lu, H., Flores, O., Weinmann, R., and Reinberg, D. (1991) Proc. Natl. Acad.
Sci. U. S. A. 88, 10004 –10008
31. Cho, E. J., Takagi, T., Moore, C. R., and Buratowski, S. (1997) Genes Dev. 11,
3319 –3326
32. Proudfoot, N. J., Furger, A., and Dye, M. J. (2002) Cell 108, 501–512
33. Corden, J. L., and Patturajan, M. (1997) Trends Biochem. Sci 22, 413– 416
34. Cismowski, M. J., Laff, G. M., Solomon, M. J., and Reed, S. I. (1995) Mol. Cell.
Biol. 15, 2983–2992
35. Feaver, W. J., Svejstrup, J. Q., Henry, N. L., and Kornberg, R. D. (1994) Cell
79, 1103–1109
36. Liao, S. M., Taylor, I. C., Kingston, R. E., and Young, R. A. (1991) Genes Dev.
5, 2431–2440
37. Liao, S. M., Zhang, J., Jeffery, D. A., Koleske, A. J., Thompson, C. M., Chao,
D. M., Viljoen, M., van Vuuren, H. J., and Young, R. A. (1995) Nature 374,
193–196
38. Lee, J. M., and Greenleaf, A. L. (1991) Gene Expr. 1, 149 –167
39. Murray, S., Udupa, R., Yao, S., Hartzog, G., and Prelich, G. (2001) Mol. Cell.
Biol. 21, 4089 – 4096
40. Kobor, M. S., Archambault, J., Lester, W., Holstege, F. C., Gileadi, O., Jansma,
D. B., Jennings, E. G., Kouyoumdjian, F., Davidson, A. R., Young, R. A., and
Greenblatt, J. (1999) Mol. Cell 4, 55– 62
41. Svejstrup, J. Q., Li, Y., Fellows, J., Gnatt, A., Bjorklund, S., and Kornberg,
R. D. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 6075– 6078
42. Komarnitsky, P., Cho, E. J., and Buratowski, S. (2000) Genes Dev. 14,
2452–2460
43. Wittschieben, B. O., Otero, G., de Bizemont, T., Fellows, J., ErdjumentBromage, H., Ohba, R., Li, Y., Allis, C. D., Tempst, P., and Svejstrup, J. Q.
(1999) Mol. Cell 4, 123–128
44. Cho, H., Orphanides, G., Sun, X., Yang, X. J., Ogryzko, V., Lees, E., Nakatani,
Y., and Reinberg, D. (1998) Mol. Cell. Biol. 18, 5355–5363
45. Zhang, Y., and Reinberg, D. (2001) Genes Dev. 15, 2343–2360
46. 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) Nature 406, 593–599
47. 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) Nature
419, 407– 411
48. 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)
Mol. Cell. Biol. 22, 1298 –1306
49. Roguev, A., Schaft, D., Shevchenko, A., Pijnappel, W. W., Wilm, M., Aasland,
R., and Stewart, A. F. (2001) EMBO J. 20, 7137–7148
50. Nagy, P. L., Griesenbeck, J., Kornberg, R. D., and Cleary, M. L. (2002) Proc.
Natl. Acad. Sci. U. S. A. 99, 90 –94
51. Miller, T., Krogan, N. J., Dover, J., Erdjument-Bromage, H., Tempst, P.,
Johnston, M., Greenblatt, J. F., and Shilatifard, A. (2001) Proc. Natl. Acad.
Sci. U. S. A. 98, 12902–12907
52. Feng, Q., Wang, H., Ng, H. H., Erdjument-Bromage, H., Tempst, P., Struhl, K.,
and Zhang, Y. (2002) Curr. Biol. 12, 1052–1058
53. Lacoste, N., Utley, R. T., Hunter, J. M., Poirier, G. G., and Cote, J. (2002)
J. Biol. Chem. 277, 30421–30424
54. van Leeuwen, F., Gafken, P. R., and Gottschling, D. E. (2002) Cell 109,
745–756
55. Ng, H. H., Feng, Q., Wang, H., Erdjument-Bromage, H., Tempst, P., Zhang, Y.,
and Struhl, K. (2002) Genes Dev. 16, 1518 –1527
56. Li, J., Moazed, D., and Gygi, S. P. (2002) J. Biol. Chem. 277, 49383– 49388
57. Strahl, B. D., Briggs, S. D., Brame, C. J., Caldwell, J. A., Koh, S. S., Ma, H.,
Cook, R. G., Shabanowitz, J., Hunt, D. F., Stallcup, M. R., and Allis, C. D.
(2001) Curr. Biol. 11, 996 –1000
58. Ma, H., Baumann, C. T., Li, H., Strahl, B. D., Rice, R., Jelinek, M. A., Aswad,
D. W., Allis, C. D., Hager, G. L., and Stallcup, M. R. (2001) Curr. Biol. 11,
1981–1985
Set2 Is Associated with pol II
59. Puig, O., Caspary, F., Rigaut, G., Rutz, B., Bouveret, E., Bragado-Nilsson, E.,
Wilm, M., and Seraphin, B. (2001) Methods 24, 218 –229
60. Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M., and Seraphin, B.
(1999) Nat. Biotechnol. 17, 1030 –1032
61. Davie, J. K., and Kane, C. M. (2000) Mol. Cell. Biol. 20, 5960 –5973
62. West, M. L., and Corden, J. L. (1995) Genetics 140, 1223–1233
63. Grant, P. A., Duggan, L., Cote, J., Roberts, S. M., Brownell, J. E., Candau, R.,
Ohba, R., Owen-Hughes, T., Allis, C. D., Winston, F., Berger, S. L., and
Workman, J. L. (1997) Genes Dev. 11, 1640 –1650
64. Grant, P. A., Berger, S. L., and Workman, J. L. (1999) Methods Mol. Biol. 119,
311–317
65. Cho, E. J., Kobor, M. S., Kim, M., Greenblatt, J., and Buratowski, S. (2001)
Genes Dev. 15, 3319 –3329
66. Edwards, A. M., Darst, S. A., Feaver, W. J., Thompson, N. E., Burgess, R. R.,
and Kornberg, R. D. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 2122–2126
67. Myers, L. C., Leuther, K., Bushnell, D. A., Gustafsson, C. M., and Kornberg,
R. D. (1997) Methods 12, 212–216
68. Donaldson, I. M., and Friesen, J. D. (2000) J. Biol. Chem. 275, 13780 –13788
69. Mueller, C. L., and Jaehning, J. A. (2002) Mol. Cell. Biol. 22, 1971–1980
70. Ho, C. K., and Shuman, S. (1999) Mol. Cell 3, 405– 411
71. Nonet, M., Sweetser, D., and Young, R. A. (1987) Cell 50, 909 –915
72. Reines, D., Conaway, R. C., and Conaway, J. W. (1999) Curr. Opin. Cell Biol.
11, 342–346
73. Wind, M., and Reines, D. (2000) BioEssays 22, 327–336
8903
74. Exinger, F., and Lacroute, F. (1992) Curr. Genet. 22, 9 –11
75. Otero, G., Fellows, J., Li, Y., de Bizemont, T., Dirac, A. M., Gustafsson, C. M.,
Erdjument-Bromage, H., Tempst, P., and Svejstrup, J. Q. (1999) Mol. Cell 3,
109 –118
76. Bernstein, B. E., Humphrey, E. L., Erlich, R. L., Schneider, R., Bouman, P.,
Liu, J. S., Kouzarides, T., and Schreiber, S. L. (2002) Proc. Natl. Acad. Sci.
U. S. A. 99, 8695– 8700
77. Studitsky, V. M., Clark, D. J., and Felsenfeld, G. (1995) Cell 83, 19 –27
78. Studitsky, V. M., Kassavetis, G. A., Geiduschek, E. P., and Felsenfeld, G.
(1997) Science 278, 1960 –1963
79. Beisel, C., Imhof, A., Greene, J., Kremmer, E., and Sauer, F. (2002) Nature
419, 857– 862
80. Nishioka, K., Chuikov, S., Sarma, K., Erdjument-Bromage, H., Allis, C. D.,
Tempst, P., and Reinberg, D. (2002) Genes Dev. 16, 479 – 489
81. Zegerman, P., Canas, B., Pappin, D., and Kouzarides, T. (2002) J. Biol. Chem.
277, 11621–11624
82. Peterson, C. L. (2002) Mol. Cell 9, 921–922
83. Bannister, A. J., Schneider, R., and Kouzarides, T. (2002) Cell 109, 801– 806
84. Davie, J. K., and Dent, S. Y. (2002) Curr. Biol. 12, R59-R61
85. 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) Nature 412, 561–565
86. Bauer, U. M., Daujat, S., Nielsen, S. J., Nightingale, K., and Kouzarides, T.
(2002) EMBO J. 3, 39 – 44