Biochem. J. (2012) 442, 527–538 (Printed in Great Britain)
527
doi:10.1042/BJ20111885
Solution structure of the Pdp1 PWWP domain reveals its unique binding
sites for methylated H4K20 and DNA
Yu QIU*, Wen ZHANG*, Chen ZHAO*, Yan WANG†, Weiwei WANG*, Jiahai ZHANG*, Zhiyong ZHANG*, Guohong LI†,
Yunyu SHI*, Xiaoming TU*1 and Jihui WU*1
*Hefei National Laboratory for Physical Sciences at Microscale and School of Life Sciences, University of Science and Technology of China, Hefei, Anhui, 230026, People’s Republic of
China, and †National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China
Methylation of H4K20 (Lys20 of histone H4) plays an important
role in the regulation of diverse cellular processes. In fission
yeast, all three states of H4K20 methylation are catalysed by
Set9. Pdp1 is a PWWP (proline-tryptophan-tryptophan-proline)
domain-containing protein, which associates with Set9 to regulate
its chromatin localization and methyltransferase activity towards
H4K20. The structure of the Pdp1 PWWP domain, which is
the first PWWP domain identified which binds to methyl-lysine
at the H4K20 site, was determined in the present study by
solution NMR. The Pdp1 PWWP domain adopts a classical
PWWP fold, with a five-strand antiparallel β-barrel followed
by three α-helices. However, it differs significantly from other
PWWP domains in some structural aspects that account, in part,
for its molecular recognition. Moreover, we revealed a unique
binding pattern of the PWWP domain, in that the PWWP domain
of Pdp1 bound not only to H4K20me3 (trimethylated Lys20 of
histone H4), but also to dsDNA (double-stranded DNA) via an
aromatic cage and a positively charged area respectively. EMSAs
(electrophoretic mobility-shift assays) illustrated the ability of the
Pdp1 PWWP domain to bind to the nucleosome core particle, and
further mutagenesis experiments indicated the crucial role of this
binding activity in histone H4K20 di- and tri-methylation in yeast
cells. The present study may shed light on a novel mechanism of
histone methylation regulation by the PWWP domain.
INTRODUCTION
are intensively regulated for various lysine residue methylations
across the genome. Some of them contain chromatin-binding
domains to regulate their localization or activities. For example,
the H3K9 methyltransferase Clr4 contains a chromodomain
which recognizes H3K9me (methylated Lys9 of histone H3) to
promote the spread of this modification across large chromosomal
domains [18]. In other cases, HMTs form complexes with
other chromatin-association proteins to gain access to histones
and perform methyltransferase activities. For instance, the
mammalian H4K20 methyltransferase Suv4-20 associates with
chromodomain protein HP1 (heterochromatin protein 1), which
binds H3K9me to facilitate further methylation of H4K20 [13].
Previously Wang et al. [19] reported that a PWWP (prolinetryptophan-tryptophan-proline) domain-containing protein from
S. pombe named Pdp1 associated with Set9 to regulate its
methyltransferase activity. It was indicated that through its PWWP
domain, Pdp1 was able to bind H4K20me1 in vivo. Moreover,
this binding was required for all three states of methylation of
H4K20 and Set9 chromatin localization for further methylation
of H4K20. Notably, both Pdp1 and Set9 were implicated in the
DSB (double-strand break) repair pathway. Intriguingly, Pdp2
(S. pombe GeneDB number SPBC215.07C) and Pdp3 (S. pombe
GeneDB number SPAC23D3.01), the homologues of Pdp1 in
fission yeast, were not involved in the regulation of H4K20
methylation and the DSB repair pathway, although they showed
strong homology with Pdp1, even outside the PWWP domain [19].
Post-translational modification of histones has emerged as a
key mechanism for controlling chromosome processes, such
as transcription. These modifications, including acetylation,
phosphorylation, methylation and ubiquitination, can affect the
interactions between chromosomes and regulatory proteins [1].
Methylation is an important histone modification implicated
in the regulation of gene transcription as well as the DNAdamage response [2]. As the only lysine residue which can
be methylated on the histone H4 tail, Lys20 is one of the
most extensively studied among all of the methylation sites
of histones [3]. Different degrees of H4K20 (Lys20 of histone
H4) methylation are involved in distinct cellular processes. For
example, H4K20me1 (monomethylated Lys20 of histone H4) is
responsible for cell-cycle progression and gene expression [4–
10], whereas H4K20me2 (dimethylated Lys20 of histone H4)
and H4K20me3 (trimethylated Lys20 of histone H4) are required
for DNA-damage checkpoint activation and maintenance of
heterochromatin structures respectively [11–13].
In mammalian cells, there are two groups of SET domaincontaining proteins that catalyse H4K20 methylation, i.e. PRSet7/Set8 for H4K20me1 and Suv4-20h for both H4K20me2 and
H4K20me3 [14–17], whereas in Schizosaccharomyces pombe,
Set9 is the only enzyme to establish the three states of
methylation of H4K20. The HMTs (histone methyltransferases)
Key words: DNA binding, histone methylation, nucleosome,
Pdp1, PWWP domain, Schizosaccharomyces pombe, Set9.
Abbreviations used: Brpf1, bromo- and PHD (plant homeodomain) finger-containing protein 1; Dnmt, DNA methyltransferase; DSB, double-strand
break; dsDNA, double-stranded DNA; EMSA, electrophoretic mobility-shift assay; FAM, 6-carboxyfluorescein; FPA, fluorescence polarization assay;
HDGF, hepatoma-derived growth factor; H3K9me, methylated Lys9 of histone H3; H3K36me3, trimethylated Lys36 of histone H3; H4K20, Lys20 of histone
H4; H4K20me1, monomethylated Lys20 of histone H4; H4K20me2, dimethylated Lys20 of histone H4; H4K20me3, trimethylated Lys20 of histone H4; HMT,
histone methyltransferase; HP1, heterochromatin protein 1; HSQC, heteronuclear single-quantum coherence; NOE, nuclear Overhauser effect; PHD,
plant homeodomain; PWWP domain, proline-tryptophan-tryptophan-proline domain; PWWP-L, long PWWP fragment; RMSD, root mean square deviation;
WHSC1, Wolf–Hirschhorn syndrome candidate 1; YES, yeast-extract sucrose.
1
Correspondence may be addressed to either of these authors (email xmtu@ustc.edu.cn or wujihui@ustc.edu.cn).
c The Authors Journal compilation c 2012 Biochemical Society
528
Y. Qiu and others
The PWWP domain is a weakly conserved module found in
a variety of eukaryotic proteins. It contains ∼ 70 amino acids
and was first characterized from the HDGF (hepatoma-derived
growth factor) [20] and WHSC1 (Wolf–Hirschhorn syndrome
candidate 1) genes [21]. In addition to the PWWP domain,
proteins in this family frequently contain other known chromatinassociation domains, such as chromodomains, bromodomains,
SET domains and PHD (plant homeodomain)-type zinc-finger
motifs, indicating a possible role of the PWWP domain in
chromatin regulation or modification.
Several structures of the PWWP domain have been solved in
recent years. Some of them were reported as DNA-binding folds,
such as PWWP domains in Dnmt3b (DNA methyltransferase 3b)
[22] and HDGF [23], whereas others were identified as methyllysine-recognizing modules, such as those in Brpf1 (bromoand PHD finger-containing protein 1) [24] and Dnmt3a [25]
recognizing trimethylated Lys36 on histone H3. Therefore the
identification of the interacting partner of the Pdp1 PWWP
domain remains an interesting problem. Is it DNA or histone with
methylated lysine, or both? Are these interactions responsible
for H4K20 methylation regulation by the PWWP domain? In the
present study, we solved the solution structure of the Pdp1 PWWP
domain by NMR, and confirmed it as an H4K20me3-recognizing
module through NMR perturbation experiments. Meanwhile, we
revealed the DNA-binding activity of the Pdp1 PWWP domain by
FPA (fluorescence polarization assay) and identified a positively
charged area responsible for this DNA binding by NMR titration
and mutagenesis experiments. As the results of FPA showed
that H4K20me3 and dsDNA (double-stranded DNA) bound
simultaneously to the Pdp1 PWWP domain, we further pointed out
the importance of the DNA-binding ability of the PWWP domain
for its nucleosome binding. Finally, mutagenesis experiments in
yeast cells confirmed the functional significance of this DNAbinding activity in the regulation of H4K20 di- and tri-methylation
and the DNA-damage checkpoint.
EXPERIMENTAL
Expression, purification and isotope labelling of Pdp1
The DNA fragments of the Pdp1 PWWP domain (residues 45–
150) were amplified from yeast genomic DNA (S. pombe) by
PCR. The DNA fragments were then ligated into the NdeI/
XhoI-cleaved plasmid pET-22b ( + ) (Novagen). All of the
recombinant Pdp1 PWWP proteins were produced in Escherichia
coli BL21(DE3) cells. Cells were grown in either LB (Luria–
Bertani) medium for unlabelled samples or minimal medium
supplemented with 15 NH4 Cl or 15 NH4 Cl and [13 C6 ]glucose for
15
N- or 15 N/13 C-samples. Generally, the protein expression was
induced at D600 = 0.8–1.0 with 0.5 mM IPTG (isopropyl βD-thiogalactopyranoside), prolonged for 24 h at 16 ◦ C. The
expressed recombinant protein was purified using a Ni2 + chelating column (GE Healthcare), followed by gel-filtration
chromatography using a Superdex 75 column. The purified protein
was exchanged to buffer A [20 mM potassium phosphate, 100 mM
NaCl, 2 mM EDTA, 0.1 % 2-mercaptoethanol and 0.2 % protease
inhibitor cocktail (Sigma), pH 4.8] and concentrated to 0.4–
0.6 mM for further NMR study.
Site-directed mutagenesis
A series of point mutations (K59D, R72D + R73D + K74D,
K104D + R105D and K124D + R125D + K126D) were introduced into the recombinant pET22b-Pdp1-PWWP vector.
The mutagenesis was performed using the QuikChange® sitedirected mutagenesis kit (Stratagene).
c The Authors Journal compilation c 2012 Biochemical Society
Peptide synthesis
Peptides H4K20, H4K20me1, H4K20me2, H4K20me3 (residues
15–25 of H4) and H3K36me3 (trimethylated Lys36 of histone H3;
residues 31–41 of H3) were synthesized by GL Biochem. They
were supplied with stringent analytical specifications (purer than
98 %), including HPLC and MS analysis. Before use, peptides
were weighed and dissolved in buffer A, the pH of each sample
was systematically controlled and, if necessary, adjusted to 4.8 by
adding dilute NaOH or HCl solution.
NMR spectroscopy
All spectra were recorded at 298 K on a Bruker DMX600
spectrometer with a cryoprobe. Both the purified 15 Nlabelled and 13 C/15 N-labelled proteins were dissolved to
a final concentration of 0.8 mM in 500 μl of buffer A
with 10 % 2 H2 O. The following spectra were recorded
to obtain backbone and side-chain resonance assignments
[26]: two-dimensional (1 H-15 N)-HSQC (heteronuclear singlequantum coherence), (1 H-13 C)-HSQC, HNCO, HN(CA)CO,
CBCA(CO)NH, CBCANH, HBHA(CO)NH, C(CO)NH-TOCSY,
H(CCO)NH-TOCSY, HCCH-TOCSY and HCCH-COSY. NOE
(nuclear Overhauser effect) distance restraints were obtained
from 15 N-edited three-dimensional NOESY and 13 C-edited threedimensional NOESY spectra with a mixing time of 110 ms. NMR
data were processed by NMRPipe and NMRDraw softwares
[27] and assigned with Sparky (T.D. Goddard and D.G. Kneller,
Sparky 3, University of California San Francisco, San Francisco,
CA, U.S.A.). The Talos + [28] program was used to obtain the
backbone dihedral angles ( and ψ) in secondary structures on
the basis of chemical-shift information. Hydrogen-bond restraints
were defined from slow-exchanging amide protons identified after
the exchange from the H2 O buffer to 2 H2 O.
Structure calculations
The structure calculation for the Pdp1 PWWP domain was
performed on the basis of proton–proton NOE restraints and
dihedral angle ( and ψ) restraints with a simulated annealing
protocol using the CNS version 1.2 program [29]. Hightemperature torsion angle dynamics were performed at 50 000
K for 15 ps (1000 steps) followed by a 15 ps cooling phase.
Initial structure calculations included only hydrogen bonds in
defined secondary structures from Talos + . In the following
refinement calculation, only hydrogen bonds whose donors could
be identified unambiguously were added. In the final calculations,
an ensemble of 200 structures [with no distance violations >0.3 Å
(1 Å = 0.1 nm) and no dihedral angle violations >5 ◦ ] was
generated from unambiguous NOEs previously determined. On
the basis of energetic criteria (low total energy, using the accept.inp routine) 20 models were selected to form a representative
ensemble of the calculated structures. All molecular drawings
were generated using PyMOL (http://pymol.sourceforge.net)
except Figure 2(A) which was generated using MOLMOL [30].
Chemical-shift perturbation
NMR titrations of the Pdp1 PWWP domain with different
methylation states of H4K20 and H3K36me3 peptides, as well
as the longer PWWP fragment (PWWP-L, residues 1–176)
with 14 bp of dsDNA were performed on 15 N-labelled proteins.
Peptides and DNA stock solutions in identical buffer were titrated
stepwise with a sample dilution of less than 10 %. The combined
chemical-shift perturbation was calculated using the following
equation:
Solution structure of the Pdp1 PWWP domain
δ ppm =
(δ HN )2 + (0.17δ N )2
δ HN and δ N are the chemical-shift variations in the proton
and nitrogen dimensions respectively. Dissociation constants were
estimated as described previously [31].
15
Table 1
Sequences of 5 FAM-labelled DNA probes
Probe name
Sequence
14 bp
5 FAM-CCTTACAGCAAAGC-3
5 FAM-GCTTTGCTGTAAGG-3
5 FAM-ATAATCGATATTTATTATGCTATTATACGTTAT-3
5 FAM-ATAACGTATAATAGCATAATAAATATCGATTAT-3
5 FAM-ATCGCCCGCGCACGCCGCTCCGCCGCAGCGCGT-3
5 FAM-ACGCGCTGCGGCGGAGCGGCGTGCGCGGGCGAT-3
AT-rich
GC-rich
N relaxation measurements
15
N relaxation experiments were carried out on a Bruker DMX600
NMR spectrometer at 298 K using previously published methods
[32]. The sample of the 15 N-labelled PWWP domain was dissolved
in buffer A as described above. With a 1 s recycle delay, the
T 1 (longitudinal relaxation time) and T 2 (transverse relaxation
time) were measured with eight relaxation delays [11.2, 61.4,
142, 242 (run twice), 363, 523, 754 and 1150 ms] and seven
relaxation delays [0, 17.6, 35.2, 52.8 (run twice), 70.4, 106 and
141 ms] respectively. The spectra measuring (1 H-15 N)-NOE were
acquired with a 2 s relaxation delay, followed by a 3 s period of
proton saturation. In the absence of proton saturation, the spectra
were recorded with a relaxation delay of 5 s. The exponential
curve-fitting and data analysis were carried out with the program
Sparky.
FPAs
FPAs were performed in buffer B [25 mM Hepes, 150 mM NaCl
and 2 mM EDTA (pH 7.0)] at 293 K using a SpectraMax M5
Microplate Reader system (Molecular Devices). The wavelengths
of fluorescence excitation and emission were 485 nm and 522 nm
respectively. Each 96-well contained 100 nM fluorescently
labelled [5 FAM (6-carboxyfluorescein)] DNA probe and
different amounts of Pdp1 PWWP or PWWP mutants, or the
PWWP–H4K20me3 complex (concentrations from 0 μM to
approximately 100 μM) with a final volume of 200 μl. For
each assay, DNA-free controls (PWWP or PWWP mutants,
or PWWP–H4K20me3 complex only) were included. The
fluorescence polarization data were analysed essentially as
described previously [33]. The sequences of 5 FAM-labelled
DNA probes are described in Table 1.
Nucleosome assembly and EMSAs (electrophoretic mobility-shift
assays)
Xenopus histone H3 used in the different methylation reactions carried a background mutation of C110A and histone H4
carried the K20C mutation (Kc20). The site-specific methylation
reactions were carried out as described previously [34] and the
efficiency of the reaction was analysed by MS. The octamers
were reconstituted by mixing and refolding the core histones
at an equimolar ratio and purified using a Superdex 200 sizeexclusion column [35]. For mononucleosome assembly, histone
octamers (Kc20me0 or Kc20me3) were assembled on to a biotinlabelled 177 bp ‘Widom-601’ DNA template at the ratio of
10 μg of octamer per 4 μg of DNA in a volume of 50 μl with
2 M NaCl, 10 mM Tris/HCl (pH 8.0), 1 mM EDTA and 5 mM
2-mercaptoethanol. Then the nucleosomes were successively
dialysed to reduce the salt from 2 M to 0.6 M at 4 ◦ C overnight.
The nucleosomes were pre-warmed at 37 ◦ C for 2 h. Then,
0.8 pmol of nucleosomes was incubated with 0–67 pmol of Pdp1
in 10 μl binding reactions [25 mM Hepes (pH 7.5), 150 mM NaCl
and 2 mM EDTA] for 2 h at 25 ◦ C. Samples were then loaded on to
5 % native polyacrylamide gels and detected by streptavidin–AP
(alkaline phosphatase).
529
Yeast strains, plasmids, medium and genetic methods
The Pdp1Δ yeast strain was bought from Bioneer. Plasmids
used in the present study are listed in Supplementary Table
S1 (at http://www.BiochemJ.org/bj/442/bj4420527add.htm). All
plasmids were inserted via a Myc tag at the N-terminus and
constructed according to a protocol described previously [36].
The K104D/R105D mutant of Pdp1 was generated with the
QuikChange® site-directed mutagenesis kit. Media such as EMM
(Edinburgh minimal medium) and YES (yeast-extract sucrose)
were made according to standard procedures. Yeast cells were
transformed using lithium acetate [37]. For the growth studies on
plates, cells in exponential phase were diluted. Five concentrations
with a D600 of 10 − 1 , 10 − 2 , 10 − 3 , 10 − 4 and 10 − 5 were made. Cells
were grown on YES plates at 30 ◦ C for 3 days.
Western blot analysis and antibodies
Protein extracts were prepared using standard procedures
[38]. The following antibodies were used for Western blot
analyses: anti-H4K20 (Abcam, ab16483), anti-H4K20me1
(Abcam, ab9051), anti-H4K20me2 (Abcam, ab9052), antiH4K20me3 (Abcam, ab9053) and anti-Myc (Sigma, M4439).
RESULTS
Sequence analysis of the Pdp1 PWWP domain
Pdp1, 359 residues in length, can be divided into two parts: a
PWWP domain at the N-terminus, and an unconserved region
at the C-terminus (Figure 1A). Amino acid sequence analysis
for the PWWP domains showed that they contain a few conserved
residues and share a similar secondary structure pattern composed
of five β-strands and three α-helices (Figure 1B), implying
that the Pdp1 PWWP domain might perform a similar function
to that of other PWWP domains.
Solution structure and dynamic analysis of the Pdp1 PWWP domain
The solution structure of the Pdp1 PWWP domain corresponding
to amino acids 45–150 was determined by multidimensional
heteronuclear NMR spectroscopy. The assembly of the 20 lowestenergy structures and the best representative structure are shown
in Figures 2(A) and 2(B) respectively. Table 2 lists the structural
statistics for the 20 deposited NMR structures, which signifies
a high-quality structure of the Pdp1 PWWP domain. The final
co-ordinates and the structure factors have been deposited in
the Protein Data Bank (PDB code 2L89). The RMSD (root
mean square deviation) of the well-defined regions in the
secondary structures of the 20 structures was 0.56 Å for the
backbone and 1.05 Å for the heavy atoms. A PROCHECK [39]
analysis of the 20 NMR structures indicated that >99 % of the
residues lay in the most favoured region and additionally allowed
region of the Ramachandran plot. The residues in the disallowed
c The Authors Journal compilation c 2012 Biochemical Society
530
Figure 1
Y. Qiu and others
The Pdp1 PWWP domain
(A) Diagram of the Pdp1 domain architecture. (B) Sequence alignment of Pdp1 with other PWWP domains. sp, S. pombe ; m, mouse; h, Homo sapiens . Partially and fully conserved residues are
highlighted with a light grey and dark grey background respectively. The hexagons indicate the aromatic residues involved in trimethylated lysine recognition.
Figure 2
Solution structure of the Pdp1 PWWP domain
(A) Ensemble of 20 lowest-energy structures for residues 49–143 of the Pdp1 PWWP domain. (B) Ribbon representation of the lowest-energy structure of the Pdp1 PWWP domain. (C) Molecular
electronic potential surface of the PWWP domain. The view in (C) on the right is oriented similarly to that in (A) and (B).
regions were those in the terminal parts or in the loops due to the
paucity of inter-residual NOEs.
The Pdp1 PWWP domain adopts a classical PWWP fold, with
a five-strand antiparallel β-barrel followed by three α-helices at
the C-terminus (Figure 2B). The β-barrel consisted of strand 1
(residues 53–59), strand 2 (residues 65–77), strand 3 (residues
c The Authors Journal compilation c 2012 Biochemical Society
86–94), strand 4 (residues 99–103) and strand 5 (residues 108–
110). All of the β-strands are linked by tight turns, except that
β2 and β3 are joined by a flexible loop. The β-barrel is folded
with a hydrophobic core involving Val58 , Val92 , Phe94 , Ala101 and
Val103 . β2 and β3 form an extended antiparallel β-sheet which is
distinct from those of other PWWP domains. The turn between
Solution structure of the Pdp1 PWWP domain
Table 2
Summary of structure statistics
Structural statistics for the 20 NMR structures of the Pdp1 PWWP domain. 1 kcal = 4.184 kJ.
531
ing stronger backbone flexibilities in these regions, consistent with
their location near the N terminus, C-terminus or within the loops.
Measurement
NMR distance and dihedral constraints
Distance constraints
Total
Intraresidue
Inter-residue
Sequential (i − j | = 1)
Medium-range (1<|i − j |<5)
Long-range (|i − j |>5)
Hydrogen bonds
Total dihedral angle restraints*
ϕ
ψ
Structure statistics
Mean energies (kcal·mol − 1 )
E total
E vdw
E noe
E angle
E bond
E improper
E dihedral
Violations (mean +
− S.D.)
Distance constraints (Å)
Dihedral angle constraints (◦ )
Deviations from idealized geometry
Bond lengths (Å)
Bond angles (◦ )
Impropers (◦ )
PROCHECK Ramachandran plot analysis (%)†
Residues in most favoured regions
Residues in additionally allowed regions
Residues in generously allowed regions
Residues in disallowed regions
Structural RMSD for secondary-structure regions (Å)‡
Backbone heavy atom (N, Cα and C )
Heavy atom
Structural comparison of the Pdp1 PWWP with other PWWP
domains
0.0029 +
− 0.00016
0.3749 +
− 0.0071
0.1972 +
− 0.0138
Pdp1 and Pdp2 are homologues in S. pombe and their PWWP
domains share 31 % amino acid sequence identity (Figure 3A).
A DALI [40] search using the Pdp1 PWWP domain as the query
sequence also showed that the Pdp2 PWWP domain is the one with
the highest structural similarity in PDB entries. The Cα RMSD
between the Pdp1 and Pdp2 PWWP domains is 2.4 Å, with a Zscore of 10.4. The Pdp1 PWWP domain adopts a similar fold to
that of the Pdp2 PWWP domain, except for the region between β2
and β3. In Pdp1, an extended antiparallel β-sheet formed by
β2 and β3 is connected by a flexible loop, whereas in Pdp2, the
corresponding β-sheet is attached by an extra short α-helix and a
310 helix (Figure 3B).
A similar difference was also observed when comparing the
structure of the Pdp1 PWWP domain with that of the Brpf1
PWWP domain which was first reported in a complex structure
with methylated histone [24]. Unlike the Pdp1 PWWP domain,
the antiparallel β-sheet formed by β2 and β3 of the Brpf1 PWWP
domain is joined by an α-helix and a β-sheet. Moreover, the last
α-helix in the Brpf1 PWWP domain is longer than that in the Pdp1
PWWP domain (Figure 3B). The structural variation of PWWP
domains is consistent with their different primary sequences in this
region (Figure 3A) and might be closely related to their distinct
molecular recognition patterns.
85.9
13.2
0.8
0.1
The PWWP domain of Pdp1 binds to H4K20me3 through its
conserved aromatic cage
1599
522
1077
443
204
430
32
56
57
− 335.25 +
− 14.09
− 441.65 +
− 12.25
15.99 +
− 1.53
69.54 +
− 2.64
14.92 +
− 1.62
5.57 +
− 0.78
0.38 +
− 0.18
0.0114 +
− 0.0006
0.2285 +
− 0.0564
0.56
1.05
*The ϕ and ψ angle restraints are generated from secondary structures by Talos + .
†All non-glycine residues, ϕ/ψ of most favoured and additionally allowed regions are given
by PROCHECK [39].
‡Atoms of well-defined secondary structure regions: residues 50–76 and 87–138.
β3 and β4 contains a cis-proline at residue 96, and β4 and β5
are linked by a helical turn. The C-terminal portion of the PWWP
domain consists of two α-helices (α1 and α2) linked by a sixresidue turn and a short α-helix (α3) at the very C-terminus.
The first two α-helices are intimately associated with each other
via hydrophobic residues (Iso116 , Phe119 , Lys120 , Lys130 , Iso131 and
Tyr134 ). The second α-helix is in extensive contact with residues
on one side of the β-barrel.
A positively charged surface enriched in basic residues (Lys59 ,
Arg72 , Arg73 , Lys74 , Lys104 , Arg105 , Lys109 , Lys124 and Arg125 )
indicates that the PWWP domain may interact with negatively
charged molecules such as DNA (Figure 2C). On the other side of
this protein is a negatively charged surface mainly formed by the
acid residues (Glu129 , Glu132 , Glu135 , Asp142 , Asp145 and Glu146 ) of
the second α-helix and the C-terminal tail.
The dynamic characteristics of the secondary-structure
elements and the loops were investigated by NMR
relaxation experiments. The heteronuclear NOE data showed
relatively small values for residues 46–50, 78–85, 95–
96, 124–127 and 142–150 (Supplementary Figure S1
at http://www.BiochemJ.org/bj/442/bj4420527add.htm), suggest-
A previous study on the Pdp1 PWWP domain revealed its
interaction with H4K20me1 [19]. To better understand the
structural basis of this interaction, a chemical-shift perturbation
experiment was employed. Meanwhile, in order to explore
the binding specificity of the Pdp1 PWWP domain towards
methylated H4K20, the interactions of the PWWP domain with
all types of methylated as well as non-methylated H4K20
peptides were investigated. Surprisingly, different from the
previous observation [19], our results indicated that the Pdp1
PWWP domain could bind all types of methylated H4K20
peptides, albeit weakly (Supplementary Figures S2A–S2C at
http://www.BiochemJ.org/bj/442/bj4420527add.htm). Analysis
of the ligand concentration-dependence of the chemical-shift
changes induced by H4K20me3 peptide gave a dissociation
constant (K d ) for the interaction of 6.0 +
− 1.7 mM (Figures 4A
and 4B), which is comparable with the K d (2.7 +
− 0.2 mM)
for the interactions between the Brpf1 PWWP domain and
H3K36me3. The K d decreased along with the increase in
degree of methylation of H4K20 (Supplementary Figure S3 at
http://www.BiochemJ.org/bj/442/bj4420527add.htm), indicating
that the interaction between the Pdp1 PWWP domain and histone
H4 is methyl-group-mediated with a preference for H4K20me3.
This is consistent with the argument of other reports that PWWP
domains can recognize trimethylated lysine modifications on
histones [24,25,41]. We also tested whether the Pdp1 PWWP
domain could bind to H3K36me3 under the same conditions.
However, no apparent interaction was observed, suggesting that
the binding of the Pdp1 PWWP domain to trimethylated lysine is
highly specific to H4K20 (Supplementary Figure S2D).
Recently, some structures of PWWP domain complexes
have been reported, including the crystal structures of the
c The Authors Journal compilation c 2012 Biochemical Society
532
Figure 3
Y. Qiu and others
Structure comparison of the Pdp1 PWWP domain with the Pdp2 PWWP domain and the Brpf1 PWWP domain
(A) Sequence alignment of the PWWP domains of Pdp1, Pdp2 and Brpf1. The grey box encloses the structural variation parts. sp, S. pombe ; h, Homo sapiens . (B) Structural (above) and topological
(below) comparisons of the PWWP domains of Pdp1 (left-hand panel), Pdp2 (middle panel; PDB code 1H3Z) and Brpf1 (right-hand panel, PDB code 2X4X). The grey boxes enclose the structural
variation parts corresponding to those in (A).
Brpf1 PWWP domain in complex with H3K36me3 and the
HDGF2 PWWP domain in complex with H3K79me3 and
H4K20me3 [41]. These structures show that the trimethyl
group fits into a pocket at one end of the β-barrel formed
by three aromatic residues. The side chains of these residues
are perpendicular to each other and form a cage around the
trimethyl ammonium group [41]. The structural comparison
between the Pdp1 PWWP domain and other PWWP domains
in complex with methylated peptides reveals that all of these
PWWP domains share high structural similarities in the βbarrel core and that the aromatic residues for methylated
lysine recognition are conserved (Figures 1B and 4E). The
trimethylated lysine residues in these complex structures all
adopt the same orientation, suggesting that the Pdp1 PWWP
domain might recognize methylated lysine in a similar manner.
The result of our NMR perturbation experiment showed that
most of the residues with obvious chemical shift changes
(δ>0.1 p.p.m.), including Lys59 , Ala60 , Gly62 , Phe94 , Asn99 ,
Phe100 and Ala101 , were located near the conserved aromatic cage
on the surface of the Pdp1 PWWP domain (Figures 4C and 4D).
This is consistent with the structural comparison result: Phe94 ,
Asn99 , Phe100 and Ala101 are located near the methylated lysine
residue (Figure 4E), whereas Lys59 , Ala60 and Gly62 might interact
with residues beside the methylated lysine residue on histone
peptide. On the basis of these results, we proposed a model
of the Pdp1 PWWP domain in complex with H4K20me3. The
model shows that the Pdp1 PWWP domain utilizes the aromatic
cage formed by Tyr63 , Trp66 , Phe94 and cation–π interactions to
recognize the trimethylated Lys20 , and two residues (Asp97 and
Asn99 ) from the loop between β3 and β4 to form an extensive
c The Authors Journal compilation c 2012 Biochemical Society
network of hydrogen bonds with the residues (Arg19 , Lys20
and Val21 ) on the histone H4 tail (Supplementary Figure S4 at
http://www.BiochemJ.org/bj/442/bj4420527add.htm).
The PWWP domain of Pdp1 binds non-specifically to DNA through
positively charged residues
The PWWP domain of Dnmt3b has been reported to associate with
DNA [22], whereas the PWWP domain of Pdp2, a homologue of
Pdp1, does not bind DNA in vitro [42]. Our analysis of the solution
structure of the Pdp1 PWWP domain indicated that there is a
positively charged surface enriched in basic residues (Figure 2C).
A question was therefore raised from this observation of whether
the Pdp1 PWWP domain is able to bind DNA molecules, in
addition to H4K20me3. To answer this question, we used the
FPA to test the DNA-binding ability of the Pdp1 PWWP domain.
As shown in Figure 5(A), fluorescence polarization anisotropy (in
units of mP) increased with greater protein/DNA ratios. The K d
value of the interaction between the Pdp1 PWWP domain and
DNA was determined to be 4.2 +
− 0.2 μM by fitting the data of
FPA. These results revealed the DNA–binding ability of the Pdp1
PWWP domain.
We next performed the NMR titration experiment to investigate
the DNA-binding surface of the PWWP domain. However, the
protein precipitated considerably at the concentration suitable
for NMR study during the DNA titration process. In order to
avoid the precipitation, we extended the boundary of the PWWP
domain and finally obtained a longer PWWP fragment (PWWPL) corresponding to residues 1–176 which was stable for the
Solution structure of the Pdp1 PWWP domain
Figure 4
533
The PWWP domain of Pdp1 binds to H4K20me3
(A) (1 H-15 N)-HSQC NMR spectra of the 15 N-labelled Pdp1 PWWP domain in the absence (black) and presence (red) of 6.97 mM H4K20me3 peptide, showing chemical-shift perturbations of a
number of residues. (B) Plots of the chemical-shift changes of seven well-resolved amide resonances against H4K20me3 peptide concentrations. Each dissociation constant is determined by fitting
1 15
the data to a single-site ligand-binding model, and the overall K d is 6.0 +
− 1.7 mM by averaging the individual ones. (C) The histogram displays H- N chemical-shift changes observed in the
corresponding spectra of the Pdp1 PWWP domain shown in (A). (D) The molecular surface of the Pdp1 PWWP domain in the same orientation as in Figure 2(A) with residues coloured according
to the magnitude of the observed chemical-shift change upon addition of H4K20me3. The colour-coding is according to the following scheme: blue, δ>0.1 p.p.m.; marine, 0.1 p.p.m.>δ>
0.05 p.p.m.; and light blue, 0.05 p.p.m.>δ>0.025 p.p.m. The positions of Tyr63 , Trp66 and Phe94 are shown. (E) Methylated lysine recognition by the PWWP domains including that of Brpf1 (PDB
code 3MO8), HDGF2 (PDB code 3QJ6 for the complex with H3K79me3, and 3QBY for the complex with H4K20me3) and Pdp1.
DNA titration experiment. The comparison of the (1 H-15 N)-HSQC
spectrum of PWWP-L with that of the PWWP domain showed that
most peaks of these two spectra overlapped well, which enabled
us to assign a portion of the peaks of PWWP-L. On the basis
of these assignments, we monitored DNA binding by following
the changes produced in the (1 H-15 N)-HSQC NMR spectrum
of PWWP-L. Both marked changes in chemical shift, such as
the amide proton of Gly53 , Val58 , Val103 , Arg105 , Lys128 and the
indole amide protons of residue Trp65 , and a reduction in signal
intensity, such as Leu57 , Ala68 and Glu129 were observed upon
addition of the DNA molecules (Figure 5B). Some of these
perturbed residues are located in the hydrophobic core, such as
Leu57 , Val58 and Val103 , whereas others reside in positively charged
patches on a domain surface, such as Trp65 , Arg105 and Lys128
(Figure 5C). This result confirmed the fact that the Pdp1 PWWP
domain binds dsDNA, whereas the slow exchange property of the
binding prevented us from calculating the binding affinity by this
NMR titration result. Moreover, the perturbed residues within the
positively charged surface are implicated in the DNA binding.
To further investigate whether the Pdp1 PWWP domain binds
DNA via its positively charged area, we engineered a series of
mutations within this region and tested the DNA-binding abilities
c The Authors Journal compilation c 2012 Biochemical Society
534
Figure 5
Y. Qiu and others
The PWWP domain of Pdp1 interacts with dsDNA
(A) FPAs of the wild-type PWWP domain, with a 5 FAM-labelled 14 bp dsDNA. The data were fitted according to the equation in the Experimental section. (B) Superimposition of (1 H-15 N)-HSQC
NMR spectra showing PWWP-L resonances in the absence (red) and presence (blue) of the equivalent of 14 bp dsDNA molecules. The assigned peaks, which grow weak (black-box-enclosed) or
substantially change their positions (arrows) upon DNA binding, are marked. The asterisk and the broken lines indicate that the resonances are from the indole amide proton of the tryptophan residue
and from the amide protons which are unable to be assigned respectively. (C) The molecular electronic potential surface which is the same as in Figure 2(C). The red circle encloses the residues
perturbed in the NMR titration experiment. (D) FPAs of the wild-type PWWP domain and the DNA-binding mutants, with a 5 FAM-labelled 14 bp dsDNA. (E) FPAs of the Pdp1 PWWP domain
interaction with AT-rich and GC-rich dsDNA, showing the Pdp1 PWWP domain binding to AT-rich and GC-rich dsDNA without an obvious preference.
of these mutants. The disassociation constants for the DNA
interaction of the mutants are summarized in Figure 5(D) and
Table 3. The results showed that the mutations of Arg72 , Arg73 ,
Lys74 , Lys104 and Arg105 reduced the DNA-binding activity of the
Pdp1 PWWP domain significantly, suggesting an important role
for the corresponding positively charged area in DNA binding
(Figure 5C).
AT- and GC-rich DNA sequences were further used to test
whether the Pdp1 PWWP domain binds to DNA with any sequence preference. As shown in Figure 5(E), no obvious sequence
preference was observed. The K d values calculated for the
PWWP binding to AT- and GC-rich DNA were 4.9 +
− 0.5 μM and
5.3 +
− 0.5 μM respectively. This result is consistent with the report
that the HDGF PWWP domain binds to DNA without sequence
preference [23].
c The Authors Journal compilation c 2012 Biochemical Society
H4K20me3 and dsDNA could bind simultaneously to the Pdp1
PWWP domain
Our results indicate that the H4K20me3 peptide and dsDNA bind
to the Pdp1 PWWP domain in separated regions, suggesting
that they might be able to bind the Pdp1 PWWP domain
simultaneously. To confirm this, the PWWP–H4K20me3 complex
formed by adding excessive H4K20me3 to the Pdp1 PWWP
domain was used to titrate the 5 FAM-labelled dsDNA. The
FPA result showed that the PWWP–H4K20me3 complex retained
the DNA-binding activity, with a K d value of 4.5 +
− 0.4 μM,
similar to that of the free Pdp1 PWWP domain (Figure 6A).
The similar DNA-binding affinities for the Pdp1 PWWP domain
free and in complex with H4K20me3 indicated that H4K20me3
does not affect dsDNA binding to the Pdp1 PWWP domain, and
Solution structure of the Pdp1 PWWP domain
Figure 6
535
The Pdp1 PWWP domain binds to Xenopus nucleosome core particles with H4K20me3 modification
(A) FPAs of the Pdp1 PWWP–H4K20me3 complex interaction with 5 FAM-labelled 14 bp dsDNA, showing the Pdp1 PWWP domain binding to H4K20me3 and dsDNA simultaneously. (B) EMSAs
of H4Kc20me0 mononucleosome core particles from Xenopus and its derivative H4Kc20me3 mononucleosome core particles with wild-type PWWP (above) and the K104D/R105D mutant (below)
respectively. The probe and PWWP–nucleosome complex positions are shown. (C) Quantification of the EMSAs in (B). The ratio of shifted/unshifted mononucleosome core particles at 3.3 μM
PWWP is shown. (D) Overview of the computational model of the Pdp1 PWWP domain in complex with a nucleosome core particle with H4K20me3 modification. The Pdp1 PWWP domain is shown
in green; the histones H3, H4, H2A and H2B and DNA components of the nucleosome core particle are shown in cornflower blue, purple, wheat, pink and light blue respectively. The H4K20me3
group is shown in red.
Table 3 Dissociation constants of the interactions between the Pdp1 PWWP
domain and 14 bp dsDNA determined by FPAs
Values are means +
− S.D.
Protein
K d value (μM)
Wild-type
K59D
R72D + R73D + K74D
K104D + R105D
K124D + R125D + K126D
4.2 +
− 0.2
29.1 +
− 2.2
161.1 +
− 39.5
43.6 +
− 2.1
24.7 +
− 2.1
DNA and H4K20me3 are able to bind the Pdp1 PWWP domain
simultaneously.
The Pdp1 PWWP domain can bind to the nucleosome
Many regulatory proteins play their roles in regulating chromatin
functions through specific binding to nucleosomes. On the basis
of our accumulated data, we speculated that the Pdp1 PWWP
domain might bind to histone and DNA at the nucleosome level
with a preference for H4K20me3. To test this, we first prepared
Xenopus mononucleosome core particles bearing H4Kc20me0
or H4Kc20me3 modifications in vitro. The wild-type or the
K104D/R105D mutant of the Pdp1 PWWP domain was then
added to the two kinds of mononucleosomes for EMSAs
respectively. The EMSA results showed that the binding of the
wild-type PWWP domain to both kinds of nucleosomes is much
stronger than that of the mutated PWWP domain (K104D/R105D)
whose DNA-binding activity is severely reduced (Figure 6B),
suggesting that DNA binding contributes the major energy to
the PWWP domain binding with the nucleosome. The free
DNA bands which appeared in the lane might result from
the disassociation of mononucleosome core particles in the
process of electrophoresis, and this disassociation could be
prevented by the addition of more Pdp1 PWWP domain.
Although the interactions between the Pdp1 PWWP domain
and H4K20me3 peptide are relatively weak (K d = 6.0 +
− 1.7 mM),
the nucleosomes with the H4Kc20me3 modification showed
enhanced binding to both the wild-type and the mutant PWWP
domains in EMSAs. The trimethylation of H4K20 enhances
PWWP domain binding to mononucleosome core particles
by nearly 15 % (Figure 6C), indicating a binding preference
of Pdp1 for nucleosomes with H4K20me3 modifications in
the chromatin environment. On the basis of the results
described above, we proposed a model of the Pdp1
PWWP domain in complex with a modified nucleosome
by using HADDOCK and molecular dynamics simulation
(described in the Supplementary Experimental section at
c The Authors Journal compilation c 2012 Biochemical Society
536
Y. Qiu and others
methylation [12,19], suggested that the Pdp1 PWWP domain is
involved in DNA-damage checkpoint activation. To prove this,
the growth rates of the rescued and the mutant cells after UV
treatment were compared. As shown in Figure 7(B), the strain with
Pdp1 transformation was not sensitive to UV, whereas the pdp1
and PWWP strains showed growth defects, indicating that the
Pdp1 PWWP domain might play important roles in DNA-damage
checkpoint activation. However, the K104D/R105D mutant strain
was little affected after UV treatment, probably owing to the
residual H4K20me2 (Figure 7A).
DISCUSSION
Figure 7 Interactions between the PWWP domain and DNA are necessary
for H4K20 high-degree methylation and are involved in the DNA-damage
checkpoint
(A) Western blot analysis of cell extracts from the indicated strains was performed with
anti-H4K20me and anti-H4 antibodies. (B) Serial dilution plating assays were performed to
measure the survival of yeast strains after treatment with the doses of UV indicated.
http://www.BiochemJ.org/bj/442/bj4420527add.htm). The model
showed that, via its β-barrel, the Pdp1 PWWP domain binds
to nucleosome core particles through the histone H4 tail and
nucleosomal DNA; no interactions with other histones occur
(Figure 6D).
The interactions between the Pdp1 PWWP domain and DNA are
necessary for high-degree H4K20 methylation in vivo
The present study has shown that the Pdp1 PWWP domain could
bind simultaneously to the H4K20me3 marker and DNA at the
nucleosome level. However, the functional significance of the
interactions between the Pdp1 PWWP domain and nucleosomal
DNA remains to be investigated. To better understand this
interaction, we constructed three pJK148 plasmids expressing
intact Pdp1, Pdp1 with the PWWP domain deleted (PWWP),
and Pdp1 with the K104D/R105D mutation respectively
(Supplementary Table S1). These plasmids were then transformed
into pdp1 haploid strains and analysed by Western blotting
using anti-H4K20me antibodies. The results show that pdp1
and PWWP strains lost H4K20me2/3 states totally, whereas
transforming Pdp1 plasmid into the pdp1 strain could restore
the H4K20me2/3 states (Figure 7A). These results suggest
that the PWWP domain is crucial for the maintenance of
H4K20me2 and H4K20me3 states in vivo. Meanwhile, the
mutation K104D/R105D, which affected the PWWP–DNA
interactions, severely reduced H4K20me2 and H4K20me3 levels
without affecting the H4K20me1 level (Figure 7A), indicating
that the interactions between the Pdp1 PWWP domain and DNA
are important for high-degree H4K20 methylation.
The Pdp1 PWWP domain is involved in DNA-damage checkpoint
activation
The results of the present study, combined with the previous
studies which indicated that H4K20me2 is involved in the DNAdamage checkpoint and that Pdp1 affects Set9-mediated H4K20
c The Authors Journal compilation c 2012 Biochemical Society
Pdp1 has been identified as an important regulatory protein of
Set9, which is a histone lysine methyltransferase responsible
for H4K20 methylation in fission yeast [19]. The structure of
the Pdp1 PWWP domain, the first PWWP domain identified to
bind to methyl-lysine H4K20, was determined by solution NMR
in the present study. Moreover, we revealed that the PWWP
domain of Pdp1 binds not only to H4K20me3, but also to
dsDNA, which has never been reported for other PWWP domains.
Intriguingly, the DNA binding is crucial for PWWP–nucleosome
complex formation, and the defect of DNA-binding activity
within the Pdp1 PWWP domain severely reduces H4K20 trimethylation and affects H4K20 dimethylation, as demonstrated
in the K104D/R105D mutant.
The structural comparison revealed that the Pdp1 PWWP
domain adopts a similar fold to that of other PWWP domains.
However, the structure of the Pdp1 PWWP domain in the region
between β2 and β3 is significantly different from those of other
PWWP domains. The region is folded into a helix followed by a
β-sheet in Brpf1 [24], whereas short helices in Dnmt3b [22] and
Pdp2 [42]. However, in Pdp1, the corresponding region is folded
into a long flexible loop connecting β2 and β3. This structural
variation might explain the distinct molecular recognition patterns
of these PWWP domains.
Several methylated H4K20 recognition motifs have so far been
reported, such as Crb2 or 53BP1 tandem tudor domains recognizing H4K20me2 [11], and JMJD2A tudor domains recognizing
H4K20me3 [43]. Both the PWWP and tudor domains belong to
the ‘Royal family’ whose members form hydrophobic cavities
via aromatic residues to interact with methylated histones
or chromatin proteins. It is remarkable that the β-barrel of
the Pdp1 PWWP domain and the first tudor domain of
53BP1 share a significant structural similarity with a backbone
RMSD of 1.4 Å, despite their low sequence identity of 18 %
(Supplementary Figure S5A at http://www.BiochemJ.org/bj/442/
bj4420527add.htm). Compared with the tudor domain of 53BP1,
the Pdp1 PWWP domain possesses a more open binding cavity
with fewer direct interactions with the methylated lysine side
chain, indicating that there is sufficient space to accommodate a
trimethyl group (Supplementary Figure S5B). However, despite
the fold topology of the JMJD2A tudor domain being radically
different from those of the 53BP1 tudor domain and the Pdp1
PWWP domain, its trimethyl-lysine-binding pocket is similar to
that of the Pdp1 PWWP domain in terms of the aromatic triplet
and an aspartic acid residue [43]. From the structural aspect, these
variations corresponding to different methylation levels and the
common features shared by H4K20me3 recognition modules may
partly explain why the Pdp1 PWWP domain preferentially binds
to H4K20me3.
The interactions between the PWWP domain and methylated
histones are often very weak. In the present study, the Pdp1
PWWP domain interacts with H4K20me3 with a K d of
Solution structure of the Pdp1 PWWP domain
∼ 6 mM, which is comparable with the results of previous
studies: the Brpf1 PWWP domain binds H3K36me3 with a
K d of ∼ 3 mM [24] and the HDGF2 PWWP domain binds
H3K36me2 with a K d of ∼ 1 mM [41]. On the other hand,
other known methyl-lysine-recognition domains bind methylated
histone peptides with relatively high affinities, such as the PHD
finger of ING2 binding H3K4me3 with a K d of ∼ 1.5 μM
[44] and the chromodomain of HP1 binding H3K9me3 with
a K d of ∼ 2.5 μM [45]. Compared with PWWP domains,
these methyl-lysine-recognition domains rely on other forces
to interact with the peptides besides the hydrophobic and
cation–π interactions. Using the HP1 chromodomain as an
example, residues of the H3 tail form β-sheet interactions
with residues in the chromodomain, and the side chains of the
histone peptide form complementary van der Waals contacts
and hydrogen-bonded interactions with the chromodomain.
However, in our case, merely the trimethylated lysine forms
hydrophobic and cation–π interactions with the aromatic
residues of the PWWP domain, which dominate the histone
binding to PWWP domain, although histone peptide residues also
make a few hydrogen bonds with residues in the PWWP domain.
This may explain why Pdp1 shows very weak binding affinity for
H4K20me3.
A previous study has shown that the Pdp1 PWWP domain
binds to methylated H4K20, and mutations within the PWWP
domain that disrupt this interaction result in the delocalization of
Set9 from chromatin and loss of H4K20me3 [19]. In the present
study, the Pdp1 PWWP domain was shown to interact physically
with dsDNA, and this interaction plays an important role in the
PWWP–nucleosome interaction. Our results further reveal that
the DNA binding is essential for H4K20me2/3 states and is
involved in DNA-damage checkpoint activation. On the basis
of these results, we propose that, through the interaction of the
PWWP domain with DNA and methylated H4K20, Pdp1 is able
to localize Set9 on chromatin, which then either performs H4K20
methylation de novo or maintains the high-degree methylation
of H4K20 within the chromatin context. Considering the weak
interactions between the Pdp1 PWWP domain and H4K20me3,
and the fact that H4K20me3 does not affect Pdp1 PWWP
domain binding to DNA, the interactions between the Pdp1
PWWP domain and DNA probably play the determining role
in the recognition process, whereas those between PWWP
and H4K20me might provide some kind of selectivity towards
nucleosomes with different H4K20 methylation states.
In summary, the findings of the present study provide structural
and biochemical insight into the unique simultaneous recognition
of methylated H4K20 and DNA by the Pdp1 PWWP domain.
This may suggest a new role for the Pdp1 PWWP domain in
maintaining the local Set9 concentration via binding nucleosomes
with methylated H4K20, which further facilitates methylation
of Lys20 on surrounding histone H4 by Set9 during replicationdependent chromatin assembly.
AUTHOR CONTRIBUTION
Yu Qiu conceived the study, solved the structure, performed the biochemical experiments
and wrote the paper. Wen Zhang performed the experiments in yeast. Chen Zhao purified the
mutant proteins. Yan Wang reconstituted the mononucleosome. Weiwei Wang and Jiahai
Zhang recorded the NMR spectrum. Zhiyong Zhang performed the molecular dynamics
simulation. Guohong Li, Xiaoming Tu and Jihui Wu edited the paper before submission.
ACKNOWLEDGMENTS
We thank Dr Qingguo Gong and Dr Xuecheng Zhang for a critical reading of the paper
prior to submission, Dr Changwen Jin and Dr Hongwei Li for their help in partial data
537
collection processing at the Beijing NMR Center, and Dr Annick Dejaegere and Dr Cédric
Grauffel for help in the methylated lysine force-field parameters.
FUNDING
This work was supported by the National Basic Research Program of China (973 Program)
[grant numbers 2011CB966302, 2009CB918804]; the Chinese National Natural Science
Foundation [grant number 30830031]; the Chinese National High Tech R&D Program
[grant number 2006AA02A315]; and the Knowledge Innovation Program of the Chinese
Academy of Science [grant number KSCX2-EW-Q-4].
REFERENCES
1 Allis, C., Jenuwein, T. and Reinberg, D. (2007) Epigenetics, Cold Spring Harbor
Laboratory Press, Cold Spring Harbor
2 Kouzarides, T. (2007) SnapShot: histone-modifying enzymes. Cell 128, 802
3 Wang, Y. and Jia, S. (2009) Degrees make all the difference: the multifunctionality of
histone H4 lysine 20 methylation. Epigenetics 4, 273–276
4 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.
283, 19478–19488
5 Kalakonda, N., Fischle, W., Boccuni, P., Gurvich, N., Hoya-Arias, R., Zhao, X., Miyata, Y.,
Macgrogan, D., Zhang, J., Sims, J. K. et al. (2008) Histone H4 lysine 20 monomethylation
promotes transcriptional repression by L3MBTL1. Oncogene 27, 4293–4304
6 Oda, H., Okamoto, I., Murphy, N., Chu, J., Price, S. M., Shen, M. M., Torres-Padilla,
M. E., Heard, E. and Reinberg, D. (2009) Monomethylation of histone H4-lysine 20 is
involved in chromosome structure and stability and is essential for mouse development.
Mol. Cell. Biol. 29, 2278–2295
7 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, 2225–2230
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, 12760–12766
9 Sims, J. K. and Rice, J. C. (2008) PR-Set7 establishes a repressive trans-tail histone code
that regulates differentiation. Mol. Cell. Biol. 28, 4459–4468
10 Trojer, P., Li, G., Sims, III, R. J., Vaquero, A., Kalakonda, N., Boccuni, P., Lee, D.,
Erdjument-Bromage, H., Tempst, P., Nimer, S. D. et al. (2007) L3MBTL1, a
histone-methylation-dependent chromatin lock. Cell 129, 915–928
11 Botuyan, M. V., Lee, J., Ward, I. M., Kim, J. E., 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, 1361–1373
12 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, 603–614
13 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, 1251–1262
14 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,
1455–1465
15 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,
1086–1099
16 Nishioka, K., Rice, J. C., Sarma, K., Erdjument-Bromage, H., Werner, J., Wang, Y.,
Chuikov, S., Valenzuela, P., Tempst, P., Steward, R. et al. (2002) PR-Set7 is a
nucleosome-specific methyltransferase that modifies lysine 20 of histone H4 and is
associated with silent chromatin. Mol. Cell 9, 1201–1213
17 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. et al. (2005) Specificity and mechanism of the
histone methyltransferase Pr-Set7. Genes Dev. 19, 1444–1454
18 Zhang, K., Mosch, K., Fischle, W. and Grewal, S. I. (2008) Roles of the Clr4
methyltransferase complex in nucleation, spreading and maintenance of heterochromatin.
Nat. Struct. Mol. Biol. 15, 381–388
19 Wang, Y., Reddy, B., Thompson, J., Wang, H., Noma, K., Yates, III, J. R. and Jia, S. (2009)
Regulation of Set9-mediated H4K20 methylation by a PWWP domain protein. Mol. Cell
33, 428–437
c The Authors Journal compilation c 2012 Biochemical Society
538
Y. Qiu and others
20 Izumoto, Y., Kuroda, T., Harada, H., Kishimoto, T. and Nakamura, H. (1997)
Hepatoma-derived growth factor belongs to a gene family in mice showing significant
homology in the amino terminus. Biochem. Biophys. Res. Commun. 238, 26–32
21 Stec, I., Wright, T. J., van Ommen, G. J., de Boer, P. A., van Haeringen, A., Moorman, A. F.,
Altherr, M. R. and den Dunnen, J. T. (1998) WHSC1, a 90 kb SET domain-containing
gene, expressed in early development and homologous to a Drosophila dysmorphy gene
maps in the Wolf-Hirschhorn syndrome critical region and is fused to IgH in t(4;14)
multiple myeloma. Hum. Mol. Genet. 7, 1071–1082
22 Qiu, C., Sawada, K., Zhang, X. and Cheng, X. (2002) The PWWP domain of mammalian
DNA methyltransferase Dnmt3b defines a new family of DNA-binding folds. Nat. Struct.
Biol. 9, 217–224
23 Lukasik, S. M., Cierpicki, T., Borloz, M., Grembecka, J., Everett, A. and Bushweller, J. H.
(2006) High resolution structure of the HDGF PWWP domain: a potential DNA binding
domain. Protein Sci. 15, 314–323
24 Vezzoli, A., Bonadies, N., Allen, M. D., Freund, S. M., Santiveri, C. M., Kvinlaug, B. T.,
Huntly, B. J., Gottgens, B. and Bycroft, M. (2010) Molecular basis of histone H3K36me3
recognition by the PWWP domain of Brpf1. Nat. Struct. Mol. Biol. 17, 617–619
25 Dhayalan, A., Rajavelu, A., Rathert, P., Tamas, R., Jurkowska, R. Z., Ragozin, S. and
Jeltsch, A. (2010) The Dnmt3a PWWP domain reads histone 3 lysine 36 trimethylation
and guides DNA methylation. J. Biol. Chem. 285, 26114–26120
26 Clore, G. M. and Gronenborn, A. M. (1994) Multidimensional heteronuclear nuclear
magnetic resonance of proteins. Methods Enzymol. 239, 349–363
27 Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J. and Bax, A. (1995) NMRPipe: a
multidimensional spectral processing system based on UNIX pipes. J. Biomol. NMR 6,
277–293
28 Shen, Y., Delaglio, F., Cornilescu, G. and Bax, A. (2009) TALOS + : a hybrid method for
predicting protein backbone torsion angles from NMR chemical shifts. J. Biomol. NMR
44, 213–223
29 Brunger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Grosse-Kunstleve, R.
W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S. et al. (1998) Crystallography &
NMR system: a new software suite for macromolecular structure determination. Acta
Crystallogr. Sect. D Biol. Crystallogr. 54, 905–921
30 Koradi, R., Billeter, M. and Wüthrich, K. (1996) MOLMOL: a program for display and
analysis of macromolecular structures. J. Mol. Graphics 14, 51–55, 29–32
31 Shen, W., Xu, C., Huang, W., Zhang, J., Carlson, J. E., Tu, X., Wu, J. and Shi, Y. (2007)
Solution structure of human Brg1 bromodomain and its specific binding to acetylated
histone tails. Biochemistry 46, 2100–2110
Received 21 October 2011/12 December 2011; accepted 13 December 2011
Published as BJ Immediate Publication 13 December 2011, doi:10.1042/BJ20111885
c The Authors Journal compilation c 2012 Biochemical Society
32 Farrow, N. A., Muhandiram, R., Singer, A. U., Pascal, S. M., Kay, C. M., Gish, G.,
Shoelson, S. E., Pawson, T., Forman-Kay, J. D. and Kay, L. E. (1994) Backbone dynamics
of a free and phosphopeptide-complexed Src homology 2 domain studied by 15N NMR
relaxation. Biochemistry 33, 5984–6003
33 Liu, Y., Huang, H., Zhou, B. O., Wang, S. S., Hu, Y., Li, X., Liu, J., Zang, J., Niu, L., Wu, J.
et al. (2010) Structural analysis of Rtt106p reveals a DNA binding role required for
heterochromatin silencing. J. Biol. Chem. 285, 4251–4262
34 Simon, M. D., Chu, F., Racki, L. R., de la Cruz, C. C., Burlingame, A. L., Panning, B.,
Narlikar, G. J. and Shokat, K. M. (2007) The site-specific installation of methyl-lysine
analogs into recombinant histones. Cell 128, 1003–1012
35 Luger, K., Rechsteiner, T. J. and Richmond, T. J. (1999) Expression and purification of
recombinant histones and nucleosome reconstitution. Methods Mol. Biol. 119, 1–16
36 Keeney, J. B. and Boeke, J. D. (1994) Efficient targeted integration at leu1-32 and
ura4-294 in Schizosaccharomyces pombe . Genetics 136, 849–856
37 Gietz, R. D., Schiestl, R. H., Willems, A. R. and Woods, R. A. (1995) Studies on the
transformation of intact yeast cells by the LiAc/SS-DNA/PEG procedure. Yeast 11,
355–360
38 Matsuo, Y., Asakawa, K., Toda, T. and Katayama, S. (2006) A rapid method for protein
extraction from fission yeast. Biosci. Biotechnol. Biochem. 70, 1992–1994
39 Laskowski, R. A., Rullmannn, J. A., MacArthur, M. W., Kaptein, R. and Thornton, J. M.
(1996) AQUA and PROCHECK-NMR: programs for checking the quality of protein
structures solved by NMR. J. Biomol. NMR 8, 477–486
40 Holm, L. and Sander, C. (1993) Protein structure comparison by alignment of distance
matrices. J. Mol. Biol. 233, 123–138
41 Wu, H., Zeng, H., Lam, R., Tempel, W., Amaya, M. F., Xu, C., Dombrovski, L., Qiu, W.,
Wang, Y. and Min, J. (2011) Structural and histone binding ability characterizations of
human PWWP domains. PLoS ONE 6, e18919
42 Slater, L. M., Allen, M. D. and Bycroft, M. (2003) Structural variation in PWWP domains.
J. Mol. Biol. 330, 571–576
43 Lee, J., Thompson, J. R., Botuyan, M. V. and Mer, G. (2008) Distinct binding modes
specify the recognition of methylated histones H3K4 and H4K20 by JMJD2A-tudor. Nat.
Struct. Mol. Biol. 15, 109–111
44 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, 100–103
45 Jacobs, S. A. and Khorasanizadeh, S. (2002) Structure of HP1 chromodomain bound to a
lysine 9-methylated histone H3 tail. Science 295, 2080–2083
Biochem. J. (2012) 442, 527–538 (Printed in Great Britain)
doi:10.1042/BJ20111885
SUPPLEMENTARY ONLINE DATA
Solution structure of the Pdp1 PWWP domain reveals its unique binding
sites for methylated H4K20 and DNA
Yu QIU*, Wen ZHANG*, Chen ZHAO*, Yan WANG†, Weiwei WANG*, Jiahai ZHANG*, Zhiyong ZHANG*, Guohong LI†,
Yunyu SHI*, Xiaoming TU*1 and Jihui WU*1
*Hefei National Laboratory for Physical Sciences at Microscale and School of Life Sciences, University of Science and Technology of China, Hefei, Anhui, 230026, People’s Republic of
China, and †National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, Beijing 100101, China
EXPERIMENTAL
Atomistic molecular dynamics simulation of the complex model
The nucleosome structure was obtained from PDB code 3C1B
[1], and the Pdp1 PWWP domain was docked into the structure
by HADDOCK [2,3] to construct an initial structural model for
the complex. The standard molecular dynamics simulation was
performed with a parallel implementation of the GROMACS4.5.4 package [4–7], using the CHARMM27 force field [8,9]. The
system contains the protein–DNA complex, 214 Na + and 92515
waters, with 301 292 atoms in total. A 3 ns production run was
Figure S1
conducted by using the Verlet integration scheme with a 2 fs timestep [10]. The simulation was performed in a constant NPT ensemble, and the system was coupled to a temperature bath of 300 K
through use of an velocity rescaling thermostat [11]. The pressure
was adjusted to 1 bar ( = 100 kPa) with a relaxation time of 0.5 ps,
and the compressibility was 4.5×10 − 5 bar − 1 [12]. Covalent bonds
were constrained using the LINCS algorithm [13,14], while the
cut-off distances for the Coulomb and van der Waals interactions
were chosen to be 0.9 and 1.4 nm respectively. The long-range
electrostatic interactions were treated by the PME algorithm [15],
with a tolerance of 10 − 5 and an interpolation order of 4.
Pdp1 PWWP domain backbone NMR relaxation data
15
N longitudinal (T1) and transversal (T2) relaxation time, and (1 H-15 N)-heteronuclear NOE are represented for residues of the Pdp1 PWWP domain. The red box encloses the flexible loop between
β2 and β3.
1
Correspondence may be addressed to either of these authors (email xmtu@ustc.edu.cn or wujihui@ustc.edu.cn).
c The Authors Journal compilation c 2012 Biochemical Society
Y. Qiu and others
Figure S2
NMR perturbation of the 15 N-labelled Pdp1 PWWP domain with different histone peptides
Pdp1 PWWP domain binding to H4 peptides with different degrees of methylation on Lys20 (A–C), as demonstrated by changes of the protein backbone resonances in the two-dimensional
(1 H-15 N)-HSQC spectra. NMR perturbation of the 15 N-labelled Pdp1 PWWP domain compared with (A) monomethyl, (B) dimethyl or (C) trimethyl Lys20 on H4. No interactions between the Pdp1
PWWP domain and (D) H3K36me3 peptide were found.
Figure S3 Different disassociation constants (K d ) of the Pdp1 PWWP domain interacting with monomethyl (A), dimethyl (B) or trimethyl (C) Lys20 on H4
peptides which derived from the fitting of NMR perturbation data
The K d value of each interaction is given above each curve.
c The Authors Journal compilation c 2012 Biochemical Society
Solution structure of the Pdp1 PWWP domain
Figure S4 A model of Pdp1–H4K20me3 interactions in ribbon
representation
Pdp1 employs side chains in its loop between β3 and β4 to form hydrogen bonds with residues
on the H4 tail. Hydrogen bonds are depicted as yellow broken lines.
Figure S5 Structural comparison between the Pdp1 PWWP domain and the 53BP1 tandem tudor domain (PDB code 2IG0) shows that they share a significant
structural similarity
Pdp1 possess a more open binding cavity than 53BP1.
c The Authors Journal compilation c 2012 Biochemical Society
Y. Qiu and others
Table S1
Plasmids used to transform S. pombe
UTR, untranslated region
Protein
Plasmid
Pdp1
PWWP
Pdp1 (including 5 UTR and 3 UTR); inserted into pJK148
Pdp1 (including 5 UTR and 3 UTR) with the PWWP domain (residues
1–150) deleted; inserted into pJK148
Pdp1 with two amino acids mutated (K104D/R105D); inserted into
pJK148
K104D/R105D
REFERENCES
1 Lu, X., Simon, M. D., Chodaparambil, J. V., Hansen, J. C., Shokat, K. M. and Luger, K.
(2008) The effect of H3K79 dimethylation and H4K20 trimethylation on nucleosome and
chromatin structure. Nat. Struct. Mol. Biol. 15, 1122–1124
2 Dominguez, C., Boelens, R. and Bonvin, A. M. (2003) HADDOCK: a protein-protein
docking approach based on biochemical or biophysical information. J. Am Chem. Soc.
125, 1731–1737
3 de Vries, S. J., van Dijk, A. D., Krzeminski, M., van Dijk, M., Thureau, A., Hsu, V.,
Wassenaar, T. and Bonvin, A. M. (2007) HADDOCK versus HADDOCK: new features and
performance of HADDOCK2.0 on the CAPRI targets. Proteins 69, 726–733
4 Berendsen, H. J. C., Vanderspoel, D. and Vandrunen, R. (1995) GROMACS: a
message-passing parallel molecular dynamics implementation. Comput. Phys. Commun.
91, 43–56
Received 21 October 2011/12 December 2011; accepted 13 December 2011
Published as BJ Immediate Publication 13 December 2011, doi:10.1042/BJ20111885
c The Authors Journal compilation c 2012 Biochemical Society
5 Lindahl, E., Hess, B. and van der Spoel, D. (2001) GROMACS 3.0: a package for
molecular simulation and trajectory analysis. J. Mol. Model. 7, 306–317
6 Van der Spoel, D., Lindahl, E., Hess, B., Groenhof, G., Mark, A. E. and Berendsen,
H. J. C. (2005) GROMACS: fast, flexible, and free. J. Comput. Chem. 26,
1701–1718
7 Hess, B., Kutzner, C., van der Spoel, D. and Lindahl, E. (2008) GROMACS 4: algorithms
for highly efficient, load-balanced, and scalable molecular simulation. J. Chem. Theory
Comput. 4, 435–447
8 MacKerell, A. D., Bashford, D., Bellott, M., Dunbrack, R. L., Evanseck, J. D., Field, M. J.,
Fischer, S., Gao, J., Guo, H., Ha, S. et al. (1998) All-atom empirical potential for
molecular modeling and dynamics studies of proteins. J. Phys. Chem. B. 102,
3586–3616
9 Grauffel, C., Stote, R. H. and Dejaegere, A. (2010) Force field parameters for the
simulation of modified histone tails. J. Comput. Chem. 31, 2434–2451
10 Verlet, L. (1967) Computer experiments on classical fluids. I. Thermodynamical
properties of Lennard-Jones molecules. Phys. Rev. 159, 98–103
11 Bussi, G., Donadio, D. and Parrinello, M. (2007) Canonical sampling through velocity
rescaling. J. Chem. Phys. 126, 014101
12 Berendsen, H. J. C., Postma, J. P. M., Vangunsteren, W. F., Dinola, A. and Haak, J. R.
(1984) Molecular dynamics with coupling to an external bath. J. Chem. Phys. 81,
3684–3690
13 Hess, B., Bekker, H., Berendsen, H. J. C. and Fraaije, J. (1997) LINCS: a linear constraint
solver for molecular simulations. J. Comput. Chem. 18, 1463–1472
14 Hess, B. (2008) P-LINCS: a parallel linear constraint solver for molecular simulation. J.
Chem. Theory Comput. 4, 116–122
15 Essmann, U., Perera, L., Berkowitz, M. L., Darden, T., Lee, H. and Pedersen, L. G. (1995)
A smooth particle mesh Ewald method. J. Chem. Phys. 103, 8577–8593