Synthetic trimethyllysine receptors that bind histone 3, trimethyllysine 27 (H3K27me3) and disrupt its interaction with the epigenetic reader protein CBX7

Article abstract of DOI:10.1016/j.bmc.2013.09.024

Post-translational modifications act as 'on' or 'off' switches causing downstream changes in gene transcription. Modifications such as trimethylation of lysine 27 on histone H3 (H3K27me3) cause repression of transcription and stable gene silencing, and it

Full text of DOI:10.1016/j.bmc.2013.09.024

Bioorganic & Medicinal Chemistry 21 (2013) 7004–7010
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthetic trimethyllysine receptors that bind histone 3,
trimethyllysine 27 (H3K27me3) and disrupt its interaction
with the epigenetic reader protein CBX7
Sara Tabet, Sarah F. Douglas, Kevin D. Daze, Graham A. E. Garnett, Kevin J. H. Allen,
⇑
Emma M. M. Abrioux, Taylor T. H. Quon, Jeremy E. Wulff, Fraser Hof
Department of Chemistry, University of Victoria, Victoria, BC V8W 3V6, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 June 2013
Revised 30 August 2013
Accepted 7 September 2013
Available online 19 September 2013
Post-translational modifications act as ‘on’ or ‘off’ switches causing downstream changes in gene tran-
scription. Modifications such as trimethylation of lysine 27 on histone H3 (H3K27me3) cause repression
of transcription and stable gene silencing, and its presence is associated with aggressive cancers of many
types. We report here macrocyclic host-type compounds that can bind H3K27me3 preferentially over
unmethylated H3K27, and characterize their binding affinities and selectivities using a convenient
dye-displacement method. We also show that they can disrupt the protein–protein interaction of
H3K27me3 with the chromobox homolog 7 (CBX7), a methyllysine reader protein, using fluorescence
polarization. These results show that sub-micromolar potencies are achievable with this family of host
compounds, and suggest the possibility of their use as new tools to induce the disruption of methylly-
sine-mediated protein–protein interactions and to report on lysine methylation in vitro.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Post-translational modifications
Methylation
Epigenetics
Protein–protein interactions
Supramolecular chemistry
Calixarenes
1. Introduction
are generally associated with gene silencing by the combined action
of two multiprotein complexes, called polycomb repressive com-
Post-translational histone modifications play an important role
in many diseases, and the ‘histone code’ is a metaphor that explains
their functions in terms of the ‘writers,’ ‘erasers,’ and ‘readers’ of
each modification type. The diverse modifications of N-terminal
histone tails include acetylation (of Lys), phosphorylation (of Ser
and Thr), methylation (of Lys and Arg), lysine ubiquitylation and
SUMOylation, and even noncovalent post-translational modifica-
tion in the form of cis–trans proline isomerisation.1–5 Such histone
modifications act on chromatin by directly influencing the interac-
tions of histones with DNA and/or by acting as specific recruitment
sites for other regulatory proteins. Different modifications are di-
rectly responsible for a variety of downstream signalling outcomes
that include transcriptional activation or repression, chromatin
remodelling, and DNA repair and recombination.6–8 Lysine 27 of
histone 3 can be trimethylated, generating an epigenetic mark
(H3K27me3) that is the focus of intense current interest in biomed-
ical research due to its importance as a signalling element in multi-
ple metastatic cancers.9–12 Histones bearing the H3K27me3 mark
plexes 1 and 2 (PRC1/PRC2).13,14 The canonical description of the
polycomb silencing pathway starts with ‘writer’ proteins EZH1/
EZH2, which are histone methyltransferases and components of
PRC2, installing the H3K27me3 mark. Subsequent recruitment of
PRC1 to the locus is driven by an H3K27me3-‘reader’ protein that
is a modular component of PRC1. The ultimate result of PRC1/
PRC2 action is DNA methylation and stable gene silencing.13 The
Drosophila parent of the H3K27me3 reader module is called poly-
comb (the namesake of the entire pathway); in humans there are
five paralogs of polycomb called chromobox homolog (CBX) 2, 4,
6, 7, and 8. Each of these reader proteins is functionally dis-
tinct.13–17 The paralog whose function has been most carefully
studied is CBX7. CBX7 is specifically associated with the silencing
of the tumor suppressors p16INK4a and p14ARF that are upstream
controllers of Rb- and p53-mediated apoptosis, respectively.18–21
As with many epigenetic pathways, the functional outcome of sig-
nalling by CBX7 is highly context dependent.13 CBX7 is consistently
shown to be strongly proliferative in castration-resistant prostate
cancer cell lines, embryonic and adult stem cells, and in hematopoi-
esis/lymphomagenisis.15,17,20–23 CBX7 is upregulated in prostate
cancer upon progression from the androgen dependent state to
the more aggressive androgen-independent state.21
Abbreviations: CBX7, chromobox homolog 7; H3K27, histone H3 lysine 27;
H3K27me3, histone H3 lysine 27 trimethylated.
⇑
Corresponding author.
E-mail address: fhof@uvic.ca (F. Hof).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.09.024

S. Tabet et al. / Bioorg. Med. Chem. 21 (2013) 7004–7010
7005
The molecular basis for targeting the H3K27me3–CBX7 com-
plex is most clearly demonstrated by mutagenesis studies that
show the complete abrogation of proliferative signal when a single
H3K27me3-binding residue of CBX7 is mutated.17,21,22 We have
been pursuing chemical agents for the disruption of the
H3K27me3–CBX7 complex. The histone’s trimethyllysine residue
is a perfectly defined and potent hot spot for this protein–protein
interaction, since CBX7 does not bind at all to unmethylated his-
tone 3.24 In the natural protein–protein complex, the trimethylly-
sine residue is recognized and bound by an aromatic cage motif
in CBX7, which is a rigid pocket defined by Phe11, Trp32, and
Trp35, (Fig. 1B, PDB code 2L1B).22 Multiple cation–pi contacts be-
tween these pi-rich side chains and the methylated ammonium
ion of Kme3 combine to drive complexation.25
We have previously shown that rigid, macrocyclic synthetic
compounds based on para-sulfonatocalix[4]arene (PSC) are potent
and selective binders of trimethyllysine as a free amino acid.26,27
These pocket-like macrocycles bind the methylated side chain of
Kme3 via multiple charge–charge and cation–pi contacts.27 Other
macrocycles can also bind Kme3 as the free amino acid28 and with-
in histone-tail peptide sequences29,30, representing an increasing
interest in using structured macrocycles to target post-transla-
tional modifications directly. We have also shown, using a FRET-
based protein biosensor, that some simple calixarene-based agents
can disrupt the interaction of H3K27me3 with CBX7.31 We report
here a set of macrocyclic compounds that constitute a new family
of H3K27me3-targeting compounds that, unlike previously re-
ported macrocycles, are easily modified to tune affinities and selec-
tivities. We report on a method for characterizing the direct
binding of such compounds to H3K27me3 using a competitive
fluorescence-based dye-displacement assay. Finally, we demon-
strate their disruption of in vitro H3K27me3–CBX7 binding using
by HPLC (or HPLC-MS) on a preparative Apollo C18 column (All-
tech, 5
m, 22 ꢀ 250 mm) or preparative Luna C-18 column (Phe-
nomenex,
Compounds were purified by running a gradient from 90:10 0.1%
TFA in H2O:0.1% TFA in MeCN to 10:90 0.1% TFA in H2O:0.1% TFA
in MeCN over 35 min.
l
5
lm, 21.2 ꢀ 250 mm), detecting at 280 nm.
2.2. Synthesis of calixarenes
2.2.1. 5-(4-Methylphenyl)-25,26,27,28-tetrahydroxy-11-17-23-
trisulfonatocalix[4]arene (5)
Compound 2 (0.1011 g, 0.1360 mmol), 4-methylphenylboronic
acid (0.0204 g, 1.1 equiv, 0.1496 mmol), Pd(OAc)2 (0.0061 g,
20 mol %) and sodium carbonate (0.0548 g, 3.8 equiv, 0.517 mmol)
were dissolved in 5 mL of deionized water inside a microwave vial,
sealed, and heated to 150 °C under microwave irradiation for 5 min
with cooling air and stirring on. HPLC purification and evaporation
of solvents in vacuo afforded a white powder in 47.5% yield
(0.0489 g). Mp: 240 °C (dec). IR (KBr pellet): 3350br, 1474s,
1457s, 1264w, 1211s, 1155s, 1113s, 1040s, 886w, 816w, 783 m,
668 m, 654 m, 626 m, 545 m. 1H NMR (500 MHz, D2O): d 7.83 (d,
J = 2.4 Hz, 2H), 7.75 (d, J = 2.4 Hz, 2H), 7.60 (s, 2H), 7.19 (s, 2H),
6.80 (d, J = 7.6 Hz, 2H), 5.70 (s, 2H), 4.06 (s, br, 8H), ꢁ0.68 (s, 3H).
13C NMR (125 MHz, D2O): d 152.1, 150.8, 146.5, 136.8, 136.4,
134.3, 133.9, 129.0, 128.8, 128.5, 128.4, 127.8 126.9, 126.8, 126.7,
126.5, 125.1, 30.7, 16.0. HR-ESI-MS: 753.07738 ([MꢁH]ꢁ, C35H29-
O13S3ꢁ; calcd 753.07758).
2.2.2. 5-(4-Methoxyphenyl)-25,26,27,28-tetrahydroxy-11-17-23-
trisulfonatocalix[4]arene (6)
Compound 2 (0.1080 g, 0.1454 mmol), 4-methoxyphenylboron-
ic acid (0.0244 g, 1.1 equiv, 0.1606 mmol), Pd(OAc)2 (0.0061 g,
20 mol %) and sodium carbonate (0.0551 g, 3.8 equiv, 0.519 mmol)
were dissolved in 5 mL of deionized water inside a microwave vial
and irradiated to 150 °C for 5 min with cooling air and stirring on.
HPLC purification and evaporation of solvents in vacuo afforded a
white powder in 43.9% yield (0.0491 g). Mp: 240 °C (dec). IR (KBr
pellet): 3245br, 1473s, 1457s, 1260w, 1239s, 1213s, 1180s,
1155s, 1114s, 1040s, 883w, 830w, 811w, 785m, 657m, 626m,
604m, 549m. 1H NMR (500 MHz, D2O): d 7.78 (d, J = 2.4 Hz, 2H),
7.71 (d, J = 2.0 Hz, 2H), 7.27 (s, 2H), 7.06 (d, J = 8.5 Hz, 2H), 6.03
(d, J = 7.7 Hz, 2H),), 4.06 (s, br, 8H), 1.60 (s, 3H). 13C NMR
(125 MHz, D2O): d 157.4, 151.8, 150.9, 146.9, 136.6, 136.2, 134.5,
131.6, 128.7, 128.5, 128.2, 128.0, 127.1, 127.0, 126.7, 126.5,
113.7, 52.8, 30.9, 30.6. HR-ESI-MS: 769.07107 ([MꢁH]ꢁ, C36H31O15-
S3ꢁ; calcd 769.07249).
a
fluorescence polarization assay that is amenable to high-
throughput screening. The resulting structure–activity relation-
ships uncover unanticipated aspects of molecular recognition for
these macrocyclic agents.
2. Materials and methods
2.1. Synthesis—general
All reagents for synthesis were purchased from Aldrich and
used as obtained. Lucigenin dye was purchased from Invitrogen
or Sigma and used as obtained. ESI-MS was performed on a Finni-
gan LCQ MS. The syntheses of compounds 1–4 and 8 and 9 have
been previously published.26,31,32 New hosts were made using 2
as starting material. All calixarenes and peptides were purified
Figure 1. (A) Unmethylated (K) and trimethylated (Kme3) states of lysine. (B) H3K27me3 bound to CBX7 (PDB ID: 2L1B). Aromatic residues that make cation–pi interactions
with the Kme3 side chain are colored yellow. (C) Trimethyllysine bound in the p-sulfonatocalix[4]arene cavity of Host 5; model generated using MMFFaq energy
minimization in Spartan.

7006
S. Tabet et al. / Bioorg. Med. Chem. 21 (2013) 7004–7010
were treated as constants, at 0.5 lM and the maximum fluores-
cence of dye without presence of host, respectively.
2.2.3. 5-(2,3-Dimethoxyphenyl)-25,26,27,28-tetrahydroxy-11-
17-23-trisulfonatocalix[4]arene (7)
Compound 2 (0.041 g, 0.055 mmol), 2,3-dimethoxyphenylbo-
ronic acid (0.010 g, 1 equiv, 0.055 mmol), tetrabutylammonium
bromide (0.0089 g, 0.5 equiv, 0.028 mmol), Pd(OAc)2 (0.0025 g,
20 mol %) and sodium carbonate (0.026 g, 3.8 equiv, 0.209 mmol)
were dissolved in 5 mL of deionized water inside a microwave vial
and irradiated at 150 °C for 5 min with cooling air and stirring on.
The aqueous solution was washed with CH2Cl2 (2 ꢀ 20 mL) then
EtOAc (1 ꢀ 25 mL) and concentrated. HPLC purification and evapo-
ration of solvents in vacuo afforded a white powder in 42% yield
(0.018 g). Mp: 245 °C (dec). IR (KBr pellet): 3366br, 1465s,
1261w, 1213s, 1160s, 1118s, 1042s, 889w, 784m, 661m, 625m,
559m. 1H NMR (500 MHz, D2O): d 7.64 (d, J = 2.1 Hz, 2H), 7.62 (d,
J = 2.0 Hz, 2H), 7.37 (s, 2H), 7.10 (s, 2H), 6.96 (t, J = 7.8 Hz, 1H),
6.82 (d, J = 7.8 Hz, 1H), 6.76 (d, J = 6.9 Hz, 1H), 3.99 (m, br, 8H),
3.70 (s, 3H), 2.48 (s, 3H). 13C NMR (125 MHz, D2O): d 153.3,
152.3, 150.9, 147.8, 145.2, 135.4, 135.3, 134.4, 131.4, 130.0,
129.1, 128.5, 128.2, 128.0, 126.6, 126.5, 126.4, 124.9, 122.3,
112.1, 59.6, 56.0, 30.7, 30.5. HR-ESI-MS: 799.07935 ([MꢁH]ꢁ, C36-
H31O15S3ꢁ; calcd 799.08303).
2.6. Kd Determination—competition experiments for
calixarene–peptide affinities
Samples for the dye displacement assay were prepared in
NUNC-96 black well plates with optically clear bottom, and were
composed of 0.01 M of phosphate buffer at pH of 7.4, 500 nM of
lucigenin, 1.25
tions (H3K27 or H3K27me3, 0–2 mM) made up with distilled
water to a final volume of 200 L. Emission spectra were collected
lM of calixarene, and varying peptide concentra-
l
as above. All experiments were performed in duplicate. Calixa-
rene–peptide Kd values were determined by plotting emission
intensity (dFobs) as a function of peptide concentration [G]t and fit-
ting the data with the program Origin using an expression and
accompanying cubic equations initially derived by Nau and co-
workers.35–37 (These expressions were subsequently adapted and
shown to be useful also for UV–vis data.38
)
Equations used for fitting of dye displacement data:
dFobs ¼ Ei þ ððEhi ꢁ EiÞ ꢂ ððKi ꢂ HÞ=ð1 þ ðKi ꢂ HÞÞÞÞ
ð2Þ
Fitting was achieved by iterative nonlinear least squares regres-
sion using the cubic step equation:
2.3. Synthesis of peptides
All reagents used for peptides synthesis were purchased from
ChemImpex or Sigma Aldrich. All Histone 3 peptides (H3K27 = Ac-
ða ꢂ H ꢂ H ꢂ H þ b ꢂ H ꢂ H þ c ꢂ H þ dÞ=ð3 ꢂ a ꢂ H ꢂ H þ 2 ꢂ b ꢂ H þ cÞ
AARKSAPY-C(O)NH2,
H3K27me3 = Ac-AARKme3SAPY-C(O)NH2,
a ¼ Ki ꢂ Kg;
FITC-H3K27me3 = FITC-bAla-AARKme3SAPY-C(O)NH2) were syn-
thesized using the standard Fmoc solid-phase peptide synthesis
protocol33 as implemented on a CEM Liberty 1 microwave-based
peptide synthesizer on Rink amide resin (ChemImpex). All se-
quences had a tyrosine introduced at the C-terminus to facilitate
UV detection during HPLC purification. All peptides were purified
as described under Section 2.1 and used without desalting.
b ¼ Ki þ Kg þ Ki ꢂ Kg ꢂ It þ Ki ꢂ Kg ꢂ ꢀ ꢁ Ki ꢂ Ki ꢂ Ht;
c ¼ 1 þ Ki ꢂ It þ Kg ꢂ ꢀ ꢁ ðKi þ KgÞ ꢂ Ht;
d ¼ ꢁHt;
where y is dFobs = Fobs ꢁ Fmin (change in emission intensity) and ꢀ
(incorporated in the term H as shown above) equals the total guest
concentration [G]t, which was varied from 0–1.5 mM for H3K27 and
0–0.5 mM for H3K27me3. Ki (association constant for host–dye
complex) was determined by previous 1:1 direct titration (see
above) and held constant, and Ehi (the inherent emission intensity
of the calixarene–dye complex) was set to 0. It (total concentration
of dye) and Ht (total concentration of host) were held constant dur-
2.4. Protein expression
CBX7 chromodomain was expressed and purified as previously
reported,24 using Addgene plasmid 25241 deposited by C. Arrow-
smith, Structural Genomics Consortium, Toronto, Canada.
2.5. Kd Determination—direct titrations for calixarene–dye
affinities
ing the titration at [lucigenin]t = 0.5
lM and [calixarene]t =
1.25 M. The adjustable parameters during fitting were Kg (associ-
l
ation constant of the calixarene–guest complex) and Ei (emission
intensity of the uncomplexed dye). H (free concentration of host)
was defined as above and used for iterative fitting with an initial
guess of H = Ht.
Samples for the direct titration were prepared in NUNC 96
black-well plates with an optically clear bottom, and were com-
posed of 0.01 M of phosphate buffer at pH 7.4, 500 nM of lucigenin,
and varying concentrations of hosts (0–5
lM) made up with dis-
tilled water to a final volume of 200
lL. Emission spectra from
2.7. Fluorescence polarization assay
445–645 nm using a SpectraMaxÒ M5/M5e Microplate Reader
were collected at kexc 369 nm. Host 5 required the use of a smaller
concentration of lucigenin of 250 nM and therefore a smaller range
The conditions for FP displacement assay were adapted from
those used in an earlier report of a direct peptide-protein titration
(see Supplementary information).24 The peptides FITC–H3K27me3,
H3K27, H3K27me3 and bovine serum albumin (Sigma) were re-
suspended in deionized H2O and the calixarenes were dissolved
in water containing 4 equiv of NaOH to neutralize residual acid
from HPLC purifications. The competitive assay was performed in
black, 96-well plates (NUNC-black well plates with optically clear
bottom) in a buffer containing 20 mM Tris–HCl pH 8.0, 250 mM
NaCl, 1 mM DTT, 1 mM benzamidine, 1 mM PMSF, and 0.01%
Tween. Constant concentrations of CBX7 and FITC–K27me3 were
of concentrations of hosts (0–1.5 lM) due to very strong binding
and inability to fit curves at high concentrations. All experiments
were performed in duplicate. Calixarene-LCG KInd values were
determined by plotting emission intensity (dFobs) as a function of
calixarene concentration (H) and fitting the data to the following
expression34 using origin:
dFobs ¼ ðFmax ꢁ FminÞ ꢂ ððD þ Ht þ ð1=KiÞ ꢁ sqrtððD þ Ht
^
þ ð1=KiÞÞð2Þ ꢁ ð4 ꢂ Ht ꢂ HtÞÞÞ=ð2 ꢂ DÞ
ð1Þ
where; y equals the change in fluorescence (dFobs = Fobs ꢁ Fmin) and
used at 8.68
tions were varied from 0–10 mM, and samples were made up to
a final volume of 100 L. Plates were incubated for 15 min in
darkness prior to reading with a SpectraMax M5 plate reader
lM and 500 nM, respectively. Calixarene concentra-
ꢀ equals the total host concentration Ht ([calixarene]t = 0–5
lM).
Parameters, Fmin and Ki where adjustable where Fmin equals the
minimum fluorescence of dye when saturated with host. D and Fmax
l

S. Tabet et al. / Bioorg. Med. Chem. 21 (2013) 7004–7010
7007
(Molecular Devices) with kexc 450 nm, kobs 530 nm, and an instru-
ment cutoff of 515 nm. The parallel and perpendicular intensities
of emission were adjusted for background of the buffer, giving mil-
lipolarization (mP) values for each reading that were determined
and normalized to percentage of complex formed. Values were
graphed using XLfit (IDBS) and fitted using a sigmoidal curve func-
tion from which IC50 values were determined. Experiments were
performed in triplicate for two independent trials and the IC50 val-
ues reported are the averages of all values.
representing H3K27me3 (Ac-AARKme3SAPY-C(O)NH2) and unme-
thylated H3K27 (Ac-AARKSAPY-C(O)NH2). Previous studies have
shown that the dye lucigenin is a promiscuous binder of many dif-
ferent sulfonated calixarenes,39,40 and we found it to be generally
well suited to use in a dye-displacement format with each of these
compounds. We sought to create a rapid, quantitative assay for
H3K27me3 (and H3K27) binding according to the schematic in
Fig. 2. The first step is a direct titration of calixarene into lucigenin;
upon binding, fluorescence decreases and the dissociation constant
of the dye for host (KInd) is determined by curve fitting of emission
intensity as a function of concentration (Fig. 2B and Supplementary
Fig. in SI). The second step is a titration in which a starting solution
of calixarene and dye is titrated with a histone peptide of interest.
The addition of the peptide causes dye displacement, and the fluo-
rescence increase upon release of dye is fitted to determine the cal-
ixarene–peptide dissociation constant Kd (Fig. 2C and Figures in SI).
Each of these titration steps was carried out rapidly for multiple
compounds at a time using 96-well microplates.
3. Results and discussion
3.1. Synthesis of compounds
Our earlier trimethyllysine-targeting compounds31 were based
on a synthetic pathway that created calixarene derivatives func-
tionalized at the ‘lower rim.’ We found these compounds difficult
to make in a manner that allows the rapid buildup of diversity.
They are also inherently poor at tuning affinities for trimethylly-
sine-containing peptides and proteins, simply because the lower
rim is distal to the actual binding surface of the macrocycle. We re-
cently developed a new and facile synthetic route to functionaliza-
tion at the ‘upper rim’ and showed using simple amino acids that
groups introduced at this position are in direct contact with the
bound Kme3 residue, and therefore have a strong influence on
binding affinities and selectivities.26 These syntheses rely on inter-
mediates 2 (aryl bromide) and PSC-NH2 (aniline), that have been
desymmetrized using selective, high-yielding reactions, and that
are well suited to further elaboration by Pd-mediated cross cou-
pling or formation of sulfonamides. The syntheses of compounds
studied as H3K27me3-binders in this report are shown in
Scheme 1. Biaryl-substituted compounds, such as 4, were prepared
by Suzuki coupling in water using ‘ligand-free’ reaction conditions
that also included Bu4N+ Brꢁ as a phase-transfer agent. The sulfon-
amide derivative 9 was prepared from PSC-NH2 using optimized,
buffered reaction conditions as reported previously.26 All of the
compounds prepared and tested here are characterized by having
a desymmetrized upper rim bearing three sulfonates and a single,
appended aromatic substituent that can modify interactions with
the H3K27me3 target.
The affinities of each compound for H3K27 and H3K27me3 are
reported in Table 1. Data were reproducible from replicate to rep-
licate, and satisfactory fits for all curves were obtained (see SI), ex-
cept for those of compound 9, which is a very weak binding
compound (see below), and compound 3, which itself contains a
p-nitrophenol (PNP) chromophore in its structure. The Kd values
for the target histone tail H3K27me3 range from 0.34 to 10 lM.
For comparison, the reported in vitro binding affinities of naturally
evolved Kme3-reader proteins for trimethyllysine histone ele-
ments are in the range of 0.7–110 l
M.24,25,41 All compounds
showed a preference for the methylated peptide H3K27me3 over
unmethylated H3K27, with the magnitude of selectivity ranging
from 3- to 26fold. Selectivity data for natural proteins is hard to ob-
tain from the literature because affinities of reader proteins for
unmethylated peptides are often not reported, or are too weak to
measure under the conditions of a given biological assay. Some re-
ported trimethylated/unmethylated selectivities exist, and seem to
point to a general limit of >100fold for trimethyllysine/lysine selec-
tivities.25,42 One exception is the PHD domain of reader protein
CHD4, which binds trimethylated and unmethylated H3K9 part-
ners with nearly equal affinities.43 In any case, we can say with
some confidence that most of the naturally evolved proteins have
better selectivities for methylated over unmethylated partners
than do these synthetic compounds. The slight erosion of selectiv-
ity of the aryl-substituted compounds relative to PSC seems to
arise more generally from an increased affinity for unmethylated
H3K27 (relative to PSC) than from any general change in affinities
3.2. Dye displacement assay for histone tail binding
A dye-displacement assay was used to measure the binding of
each compound with unmethylated and trimethylated peptides
Scheme 1. Synthesis of sulfonato-calix[4]arenes 1–9. Reagents and conditions: (a) Ar-B(OH)2, Na2CO3, TBAB, Pd(OAc)2, lw, 2 h, 150 °C b) NaOH then RaNi, MeOH/H2O (1:1),
H2, overnight (c) TsCl, H2O, 100 mM Na2HPO4, pH 8, overnight.

7008
S. Tabet et al. / Bioorg. Med. Chem. 21 (2013) 7004–7010
Figure 2. Indicator displacement assay for calixarene-peptide binding. (A) Schematic of the indicator displacement assay showing first the quenching of the dye upon
addition of the host (governed by the dissociation constant of the host to dye (Kind)) followed by the competitive addition of an unmethylated or trimethylated peptide (guest)
causing the release of the dye (governed by the host-guest dissociation constant (Kd)). (B) Direct fluorescence titration of LCG (0.5
at pH 7.4, kex = 369 nm. The inset shows the fitted titration curve for the calculation of the dissociation constant Kind at 485 nm. (C) Exemplary dye-displacement assay curves
done using 0.01 M phosphate buffer at pH 7.4, kex = 369 nm, kem = 485 nm. 500 nM lucigenin dye, 1.25 M of host 6 Ç, host 8 d and varying concentrations of unmethylated
lM) with host 1 in 0.01 M phosphate buffer
l
(red) or trimethylated peptide (blue) showing the different selectivities of each host to a particular peptide. For clarification a semi-log plot of these data can be found in
Supplementary information.
Table 1
Activities of trimethyllysine-targeting compounds as determined by dye displacement assay and FP protein–protein disruption assaya
b
c
c
Calix[4]arene
Substituents
KInd for LCG
M)
Kd for H3K27
M)
Kd for H3K27me3
M)
H3K27me3/H3K27
selectivity
IC50 for H3K27me3-CBX7 disruption
M)
(
l
(
l
(
l
(l
PSCd
X= SO3
X= Ph
X= Br
n.d.
0.055 0.011
220 7d
19.0 4.4
13.3 7.6
30 60
11.3 2.9
2.7 1.0
20 9.3
25 14
23 16
>100
5.4 0.1d
0.75 0.18
0.70 0.17
10 13
0.88 0.12
0.34 0.03
0.78 0.22
2.13 0.53
1.86 0.64
>100
40
25
19
3
13
8
26
12
12
n.d.
2500 900
510 50
470 60
>5000
370 40
230 20
360 30
1700 700
590 40
>2000
ꢁ
Host 1
Host 2
Host 3
Host 4
Host 5
Host 6
Host 7
Host 8
Host 9
0.085 0.025
0.460 0.044
0.050 0.020
0.0132 0.013
0.120 0.009
0.103 0.017
0.124 0.019
1.91 0.29
X= NO2
X= Ph(CN)
X= Ph(Me)
X= Ph(OMe)
X= Ph(2,3-OMe)
X= Ph(CONH2)
X= NHSO2Ph(Me)
a
All Kd values are averages from duplicate determinations. IC50 values from fluorescence polarization assay are averages of triplicate determinations.
b
Conditions for direct titration of calixarene into LCG: [LCG] = 0.5
lM with host at varying concentrations between 0–5
lM in 0.01 M phosphate buffer at pH 7.4, kex
369 nm, kem 485 nm done in duplicates.
c
Conditions for the competitive titration of peptide into host/LCG solution: [LCG] = 0.5
and [H3K27me3] = 0–0.5 mM in 0.01 M sodium phosphate buffer at pH 7.4, kex 369 nm, kem 485 nm.
lM, [host] = 1.25 lM with varying peptide concentrations of [H3K27] = 0–1.5 mM
d
31
Data for PSC taken from reference.
(IC50 value).
Literature values were determined by isothermal titration calorimetry (Kd values) or FRET assay for H3K27me3-CBX7 disruption
for the targeted H3K27me3. The compounds all make complexes
with Kme3 in which the quaternary ammonium ion is buried deep
within the macrocycle’s binding pocket, and the appended aro-
matic rings are in close contact with the trimethyllysine side
chain’s methylene groups, as previously confirmed for representa-
tive members of this class by 1H NMR.26 This structural model can
explain the increased affinities for H3K27 upon introduction of aryl
groups, because the aryl groups make similar favourable contacts
with the methylenes of both trimethyllysine and unmethylated
lysine.
We also looked for the expected connections between ring elec-
tronics and strength of cation–pi interactions; that is, that more
electron-rich rings (e.g. methoxy-substituted 6) would form stron-
ger cation–pi interactions with the lysine/trimethyllysine side
chains of the histone peptides than would a more electron-poor
ring (e.g. cyano-substituted 4). These trends are not present in
the data for any aryl-appended host. For example, MeO-substituted
6 and CN-substituted 4 have equal affinities for H3K27me3, and
similar affinities for H3K27. But strong effects do exist for substit-
uents on the main macrocycle—Compound 3 (nitro-) differs from
compound 2 (bromo-) by only a single group, but has ꢃ15fold low-
er affinity for H3K27me3 and has the lowest H3K27me3/H3K27
selectivity of any compound in this group. This suggests that a
strong cation–pi interaction does exist for the aryl groups of the
main macrocycle that does not exist for the aryl groups appended
to the upper rim. Further, compound 7 (2,3-dimethoxy–) is signif-
icantly weaker than is compound 6 (4-methoxy–), which is in this
case best explained by the conformational differences induced by
the presence of an ortho-substituted biaryl linkage in 7 that is
not present in 6. Compound 9—the only member that bears a sul-
fonamide linked aryl group—has very low affinities for either
methylated or unmethylated target. One possibility is that the sul-
fonamide linker disrupts binding. But we also see evidence of self-
inclusion in the 1H NMR spectrum of 9 (a strongly upfield-shifted
CH3 resonance that indicates inclusion of the methyl group in the
binding pocket of another copy of the calixarene) that offers an
alternative explanation for the poor activity of 9.
3.3. Fluorescence polarization assay for H3K27me3–CBX7
disruption
Fluorescence polarization (FP) assays have been used exten-
sively to measure the affinities of histone peptides for their reader
proteins.24,25,42 They are also amenable for testing for protein
interaction disruptors.44 In our implementation, we used a direct
titration of recombinant CBX7 chromodomain into FITC-labeled
H3K27me3 to establish selective protein–peptide binding, and to
determine the concentration of protein at which 90% of full FP

S. Tabet et al. / Bioorg. Med. Chem. 21 (2013) 7004–7010
7009
Figure 3. (A) Schematic of fluorescence polarization assay showing a dye-labeled H3K27me3 peptide and a CBX7 protein with high fluorescence polarization (see
Supplementary information). Addition of a synthetic host that binds the peptide disrupts the complex and causes a return to low fluorescence polarization. (B) Determination
of specificity for FP competition assay. FITC-H3K27me3 Peptide and CBX7 constant at concentrations of 500 nM and 8.68 lM were exposed to nonfluorescently labeled
H3K27me3 (blue) and H3K27 (green) and BSA (red) at concentrations ranging from 0–250 mM in buffer (20 mM Tris–HCl pH 8.0, 250 mM NaCl, 1 mM DTT, 1 mM
benzamidine, 1 mM PMSF, 0.01% Tween). (C) FP competition assay with calixarenes. Competition assays were performed with FITC-H3K27me3 peptide and CBX7 constant at
concentrations of 500 nM and 8.68
lM. Calixarenes were re-suspended in 4 equiv NaOH and were tested from 5 mM (calixarene 3-blue) or 10 mM (calixarenes 6 green and 8
red) to 0 mM in buffer (20 mM Tris–HCl pH 8.0, 250 mM NaCl, 1 mM DTT, 1 mM benzamidine, 1 mM PMSF, 0.01% Tween).
response was achieved. Titration of inhibitor compounds into a
pre-formed complex of CBX7 and FITC-H3K27me3 caused a de-
crease in FP that demonstrates disruption of the reader protein–
H3K27me3 complex. It is unconventional that this particular assay
involves disruption of the protein–peptide complex with an agent
that binds to the dye-labelled peptide, as opposed to an agent that
binds to the unlabelled protein. But the FP signal of the peptide–
calixarene complex is significantly lower than that arising from
the peptide–protein complex, due to the fact that the calixarenes
(ꢃ1 kDa) are lower molecular weight than the CBX7 protein do-
main (7.8 kDa), so the assay functions well. Control titrations with
unlabeled H3K27me3 (positive control), as well as with unmethy-
lated H3K27 and BSA (negative controls), demonstrate the specific-
ity of the observed disruption. Fig. 3B and C show exemplary data
for the disruption of the interaction between CBX7 and the peptide
upon back-titration with different control compounds (B) and with
calixarenes of differing potency (C). Again, all compounds were
conveniently tested in 96-well plate format with good run-to-run
reproducibility. In competition assays such as these, the IC50 values
depend strongly on the concentrations of fluorescently labelled
peptide and/or the protein competing with the inhibitor for the
peptide’s binding site, and their magnitudes only approximate Kd
values. Satisfyingly, the IC50 values determined using this disrup-
tion FP assay fall into three categories—low, medium, and high—
that agree reasonably well with the rank ordering of direct
H3K27me3 binding determined by dye displacement assay.
modification pathways. The structure–function relations identified
here have provided us with a basic understanding of the molecular
recognition determinants for these synthetic receptors. Also, the
modular synthesis that we use to produce them suggests that more
analogs, designed based on those lessons, can be made and tested
in order to create more potent and more site-selective analogs. The
flexible synthesis also offers the chance to identify agents with
selectivities for other post-translational modification states,30 and
that would potentially disrupt other classes of reader protein inter-
actions. Given the extreme rarity of chemical agents of any kind
that can disrupt the complexes of methyllysine–reader pro-
teins,45–49 these compounds, though unconventional in their ap-
proach, represent a jumping off point for new bioorganic and
biochemical studies of post-translational methylation pathways.
Acknowledgments
This work was supported by Genome B.C., West Coast Ride to
Live, Prostate Cancer Canada, and NSERC. K.D.D. thanks Ride to
Live—Vancouver Island and the Prostate Cancer Foundation of BC
for fellowship support. J.E.W. and F.H. are both Canada Research
Chairs and Scholars of the Michael Smith Foundation for Health
Research.
Supplementary data
Supplementary data (titration curves, characterization data, and
supplementary isothermal titration calorimetry experiments)
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.bmc.2013.09.024.
4. Conclusion
The calixarenes that we report have a unique ability to bind to
post-translational modifications, as opposed to binding within the
ordered binding pockets of protein targets. This novel feature offers
an effective route to the disruption of protein–protein interactions
that are triggered by post-translational modifications. It also opens
the door to their further development as agents for the character-
ization and analysis of post-translational modification pathways.
The compounds presented here could, in principle, be active
against many different trimethyllysine-containing targets—a fea-
ture that distinguishes them from small molecules that target con-
cave binding pockets with some inherent selectivity. The broad
selectivity of the calixarenes will in fact facilitate their use as bio-
chemical reagents for the study of multiple pathways, and the
promise of using supramolecular reagents for the construction of
novel biochemical assays has already been established.36,37,40
There is also hope for improving their specificity further to the
point of being useful for studies into the chemical biology of
References and notes
1. Nelson, C. J.; Santos-Rosa, H.; Kouzarides, T. Cell 2006, 126, 905.
2. Dilworth, D.; Gudavicius, G.; Leung, A.; Nelson, C. J. Biochem. Cell Biol. 2012, 90,
55.
3. Musselman, C. A.; Lalonde, M. E.; Cote, J.; Kutateladze, T. G. Nat. Struct. Mol. Biol.
2012, 19, 1218.
4. Chi, P.; Allis, C. D.; Wang, G. G. Nat. Rev. Cancer 2010, 10, 457.
5. Ruthenburg, A. J.; Li, H.; Patel, D. J.; David Allis, C. Nat. Rev. Mol. Cell. Biol. 2007,
8, 983.
6. Kouzarides, T. Cell 2007, 128, 693.
7. Jenuwein, T.; Allis, C. D. Science 2001, 293, 1074.
8. Sawan, C.; Herceg, Z. In Adv. Genet.; Zdenko Herceg and Toshikazu, U., Ed.;
Academic Press, 2010; Vol. 70, pp. 57.
9. Qi, W.; Chan, H.; Teng, L.; Li, L.; Chuai, S.; Zhang, R.; Zeng, J.; Li, M.; Fan, H.; Lin,
Y.; Gu, J.; Ardayfio, O.; Zhang, J.-H.; Yan, X.; Fang, J.; Mi, Y.; Zhang, M.; Zhou, T.;
Feng, G.; Chen, Z.; Li, G.; Yang, T.; Zhao, K.; Liu, X.; Yu, Z.; Lu, C. X.; Atadja, P.; Li,
E. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 21360.
10. Verma, S. K.; Tian, X.; LaFrance, L. V.; Duquenne, C.; Suarez, D. P.; Newlander, K.
A.; Romeril, S. P.; Burgess, J. L.; Grant, S. W.; Brackley, J. A.; Graves, A. P.;

7010
S. Tabet et al. / Bioorg. Med. Chem. 21 (2013) 7004–7010
Scherzer, D. A.; Shu, A.; Thompson, C.; Ott, H. M.; Aller, G. S. V.; Machutta, C. A.;
29. Ingerman, L. A.; Cuellar, M. E.; Waters, M. L. Chem. Commun. 2010, 46, 1839.
30. James, L. I.; Beaver, J. E.; Rice, N. W.; Waters, M. L. J. Am. Chem. Soc. 2013, 135,
6450.
31. Daze, K. D.; Pinter, T.; Beshara, C. S.; Ibraheem, A.; Minaker, S. A.; Ma, M. C. F.;
Courtemanche, R. J. M.; Campbell, R. E.; Hof, F. Chem. Sci. 2012, 3, 2695.
32. Daze, K. D.; Hof, F. Acc. Chem. Res. 2013, 46, 937.
Diaz, E.; Jiang, Y.; Johnson, N. W.; Knight, S. D.; Kruger, R. G.; McCabe, M. T.;
Dhanak, D.; Tummino, P. J.; Creasy, C. L.; Miller, W. H. ACS Med. Chem. Lett.
2012, 3, 1091.
11. McCabe, M. T.; Ott, H. M.; Ganji, G.; Korenchuk, S.; Thompson, C.; Van Aller, G.
S.; Liu, Y.; Graves, A. P.; Iii, A. D. P.; Diaz, E.; LaFrance, L. V.; Mellinger, M.;
Duquenne, C.; Tian, X.; Kruger, R. G.; McHugh, C. F.; Brandt, M.; Miller, W. H.;
Dhanak, D.; Verma, S. K.; Tummino, P. J.; Creasy, C. L. Nature 2012, 492, 108.
12. Knutson, S. K.; Wigle, T. J.; Warholic, N. M.; Sneeringer, C. J.; Allain, C. J.; Klaus,
C. R.; Sacks, J. D.; Raimondi, A.; Majer, C. R.; Song, J.; Scott, M. P.; Jin, L.; Smith, J.
J.; Olhava, E. J.; Chesworth, R.; Moyer, M. P.; Richon, V. M.; Copeland, R. A.;
Keilhack, H.; Pollock, R. M.; Kuntz, K. W. Nat. Chem. Biol. 2012, 8, 890.
13. Gieni, R. S.; Hendzel, M. J. Biochem. Cell Biol. 2009, 87, 711.
33. Amblard, M.; Fehrentz, J.-A.; Martinez, J.; Subra, G. Mol. Biotechnol. 2006, 33,
239.
34. Hirose, K. J. Incl. Phenom. Macro. Chem. 2001, 39, 193.
35. Bakirci, H.; Nau, W. M. Adv. Funct. Mat. 2006, 16, 237.
36. Hennig, A.; Bakirci, H.; Nau, W. M. Nat. Meth. 2007, 4, 629.
37. Bailey, D. M.; Hennig, A.; Uzunova, V. D.; Nau, W. M. Chem. Eur. J. 2008, 14,
6069.
14. Sauvageau, M.; Sauvageau, G. Cell Stem Cell 2010, 7, 299.
38. Hargrove, A. E.; Zhong, Z.; Sessler, J. L.; Anslyn, E. V. J. New Chem. 2010, 34, 348.
39. Minaker, S. A.; Daze, K. D.; Ma, M. C. F.; Hof, F. J. Am. Chem. Soc. 2012, 134,
11674.
40. Guo, D.; Uzunova, V. D.; Su, X.; Liu, Y.; Nau, W. M. Chem. Sci. 2011, 2, 1722.
41. Nielsen, P. R.; Nietlispach, D.; Mott, H. R.; Callaghan, J.; Bannister, A.;
Kouzarides, T.; Murzin, A. G.; Murzina, N. V.; Laue, E. D. Nature 2002, 416, 103.
42. Bernstein, E.; Duncan, E. M.; Masui, O.; Gil, J.; Heard, E.; Allis, C. D. Mol. Cell Biol.
2006, 26, 2560.
15. O’Loghlen, A.; Muñoz-Cabello, A. M.; Gaspar-Maia, A.; Wu, H.; Banito, A.;
Kunowska, N.; Racek, T.; Pemberton, H. N.; Beolchi, P.; Lavial, F.; Masui, O.;
Vermeulen, M.; Carroll, T.; Graumann, J.; Heard, E.; Dillon, N.; Azuara, V.;
Snijders, A. P.; Peters, G.; Bernstein, E.; Gil, J. Cell Stem Cell 2012, 10, 33.
16. Morey, L.; Pascual, G.; Cozzuto, L.; Roma, G.; Wutz, A.; Benitah, S. A.; Di Croce,
L. Cell Stem Cell 2012, 10, 47.
´
17. Klauke, K.; Radulovic, V.; Broekhuis, M.; Weersing, E.; Zwart, E.; Olthof, S.;
Ritsema, M.; Bruggeman, S.; Wu, X.; Helin, K.; Bystrykh, L.; de Haan, G. Nat. Cell.
Biol. 2013, 15, 353.
43. Mansfield, R. E.; Musselman, C. A.; Kwan, A. H.; Oliver, S. S.; Garske, A. L.;
Davrazou, F.; Denu, J. M.; Kutateladze, T. G.; Mackay, J. P. J. Biol. Chem. 2011,
286, 11779.
44. Roehrl, M. H. A.; Wang, J. Y.; Wagner, G. Biochem. 2004, 43, 16056.
45. Gao, C.; Herold, J. M.; Kireev, D.; Wigle, T.; Norris, J. L.; Frye, S. J. Am. Chem. Soc
2011, 133, 5357.
46. Herold, J. M.; Wigle, T. J.; Norris, J. L.; Lam, R.; Korboukh, V. K.; Gao, C.;
Ingerman, L. A.; Kireev, D. B.; Senisterra, G.; Vedadi, M.; Tripathy, A.; Brown, P.
J.; Arrowsmith, C. H.; Jin, J.; Janzen, W. P.; Frye, S. V. J. Med. Chem. 2011, 54,
2504.
18. Aguilo, F.; Zhou, M.-M.; Walsh, M. J. Cancer Res. 2011, 71, 5365.
19. Maertens, G. N.; El Messaoudi-Aubert, S.; Racek, T.; Stock, J. K.; Nicholls, J.;
Rodriguez-Niedenführ, M.; Gil, J.; Peters, G. PLoS One 2009, 4, e6380.
20. Gil, J.; Bernard, D.; Martinez, D.; Beach, D. Nat. Cell. Biol. 2004, 6, 67.
21. Bernard, D.; Martinez-Leal, J. F.; Rizzo, S.; Martinez, D.; Hudson, D.; Visakorpi,
T.; Peters, G.; Carnero, A.; Beach, D.; Gil, J. Oncogene 2005, 24, 5543.
22. Yap, K. L.; Li, S.; Muñoz-Cabello, A. M.; Raguz, S.; Zeng, L.; Mujtaba, S.; Gil, J.;
Walsh, M. J.; Zhou, M.-M. Mol. Cell 2010, 38, 662.
23. Bracken, A. P.; Helin, K. Nat. Rev. Cancer 2009, 9, 773.
24. Kaustov, L.; Ouyang, H.; Amaya, M.; Lemak, A.; Nady, N.; Duan, S.; Wasney, G.
A.; Li, Z.; Vedadi, M.; Schapira, M.; Min, J.; Arrowsmith, C. H. J. Biol. Chem. 2011,
286, 521.
25. Hughes, R. M.; Wiggins, K. R.; Khorasanizadeh, S.; Waters, M. L. Proc. Natl. Acad.
Sci. U.S.A. 2007, 104, 11184.
26. Daze, K. D.; Ma, M. C.; Pineux, F.; Hof, F. Org. Lett. 2012, 14, 1512.
27. Beshara, C. S.; Jones, C. E.; Daze, K. D.; Lilgert, B. J.; Hof, F. ChemBioChem 2010,
11, 63.
47. James, L. I.; Barsyte-Lovejoy, D.; Zhong, N.; Krichevsky, L.; Korboukh, V. K.;
Herold, J. M.; MacNevin, C. J.; Norris, J. L.; Sagum, C. A.; Tempel, W.; Marcon, E.;
Guo, H.; Gao, C.; Huang, X. P.; Duan, S.; Emili, A.; Greenblatt, J. F.; Kireev, D. B.;
Jin, J.; Janzen, W. P.; Brown, P. J.; Bedford, M. T.; Arrowsmith, C. H.; Frye, S. V.
Nat. Chem. Biol. 2013, 9, 184.
48. Kireev, D.; Wigle, T. J.; Norris-Drouin, J.; Herold, J. M.; Janzen, W. P.; Frye, S. V. J.
Med. Chem. 2010, 53, 7625.
49. Wagner, E. K.; Nath, N.; Flemming, R.; Feltenberger, J. B.; Denu, J. M.
Biochemistry 2012, 51, 8293.
28. Gamal-Eldin, M. A.; Macartney, D. H. Org. Biomol. Chem. 2013, 11, 488.