Effective disruption of phosphoprotein-protein surface interaction using Zn(II) dipicolylamine-based artificial receptors via two-point interaction

Article abstract of DOI:10.1021/ja056585k

Protein phosphorylation is ubiquitously involved in living cells, and it is one of the key events controlling protein-protein surface interactions, which are essential in signal transduction cascades. We now report that the small molecular receptors beari

Full text of DOI:10.1021/ja056585k

Published on Web 01/21/2006
Effective Disruption of Phosphoprotein-Protein Surface
Interaction Using Zn(II) Dipicolylamine-Based Artificial
Receptors via Two-Point Interaction
Akio Ojida,† Masa-aki Inoue,‡ Yasuko Mito-oka,‡ Hiroshi Tsutsumi,§
Kazuki Sada,‡ and Itaru Hamachi*,†,‡,§
Contribution from the Department of Synthetic Chemistry and Biological Chemistry, Graduate
School of Engineering, Kyoto UniVersity, Katsura, Kyoto, 615-8510, Japan, Department of
Chemistry and Biochemistry, Graduate School of Engineering, Kyushu UniVersity, Fukuoka,
812-8581, Japan, and PRESTO (Organization and Function, JST)
Received October 1, 2005; E-mail: ihamachi@sbchem.kyoto-u.ac.jp
Abstract: Protein phosphorylation is ubiquitously involved in living cells, and it is one of the key events
controlling protein-protein surface interactions, which are essential in signal transduction cascades. We
now report that the small molecular receptors bearing binuclear Zn(II)-Dpa can strongly bind to a bis-
phosphorylated peptide in a cross-linking manner under neutral aqueous conditions when the distance
between the two Zn(II) centers can appropriately fit in that of the two phosphate groups of the phosphorylated
peptide. The binding property was quantitatively determined by ITC (isothermal titration calorimetry), induced
CD (circular dichroism), and NMR. On the basis of these findings, we demonstrated that these types of
small molecules were able to effectively disrupt the phosphoprotein-protein interaction in a phosphorylated
CTD peptide and the Pin1 WW domain, a phosphoprotein binding domain, at a micromolar level. The
strategy based on a small molecular disruptor that directly interacts with phosphoprotein is unique and
should be promising in developing a designer inhibitor for phosphoprotein-protein interaction.
Introduction
involved in a signaling complex. A significant feature of these
phosphoprotein-protein interactions is that the phosphate group
Protein-protein surface interactions play central roles in
controlling numerous biological processes in living cells.1,2
A variety of specific protein complexes are formed in a time-
and spatio-selective manner, producing the regulation and
maintenance of complicated cellular events. Protein surface
phosphorylation is, in particular, ubiquitously involved in signal
transduction cascades and is controlled by phosphorylation-
dephosphorylation processes catalyzed by the corresponding
protein kinases and phosphatases.3 In many cases, the phos-
phorylation of certain protein surfaces by the upstream kinase
creates a new binding site for the downstream proteins, in which
the phosphorylated protein surface is recognized by a phospho-
protein binding domain such as the SH2, PTB, FHA, and WW
domains.4 In general, these domains specifically interact with
short peptide motifs containing phosphorylated amino acid
residues (phospho-serine (pSer), -threonine (pThr), and -tyrosine
(pTyr)), resulting in the recruitment of the phosphoprotein
on a protein-protein interface acts as a pinpoint residue crucial
for the tight binding.5 That is, the phosphate group not only
contributes to the high-affinity binding via salt bridges and/or
hydrogen bonding but also induces the conformational re-
arrangement of surface amino acid residues necessary for the
complementary interaction with the phosphoprotein binding
domain. Thus, a small molecule that specifically interacts with
the phosphate group on a protein surface may potentially disrupt
the signaling complex and thus provide a potent modulator or
disruptor for the signal transduction cascades.
Because of the biological significance of protein surface
interactions, the design of small molecules that modulate or
interfere with certain protein-protein interactions has been
receiving considerable attention from the standpoint of funda-
mental molecular recognition6,7 and practical importance such
as pharmaceutical applications.8 However, many protein-protein
interfaces consist of a noncontinuous binding epitope on a
relatively wide surface area (generally > 1100 Å2), which often
makes it difficult to rationally design small molecule-based
modulators or inhibitors. To date, much effort has also been
† Kyoto University.
‡ Kyushu University.
§ PRESTO.
(1) Giot, T.; et al. Science 2003, 302, 1727.
(2) (a) Ball, L. J.; Ku¨hne, R.; Schneider-Mergener, J.; Oschkinat, H. Angew.
Chem., Int. Ed. 2005, 44, 2852. (b) Pawson, T.; Nash, P. Science 2003,
300, 445. (c) Chakrabarti, P.; Janin, J. Proteins 2002, 47, 334. (d) Pawson,
T.; Raina, M.; Nash, P. FEBS Lett. 2002, 513, 2.
(3) (a) Pawson, T. Scott, J. D. Trends Biochem. Sci. 2005, 30, 286-290. (b)
Johnson, S. A.; Hunter, T. Nature Methods 2005, 2, 17. (c) Sefton, B. M.;
Hunter, T. Protein Phosphorylation; Academic Press: New York, 1998.
(4) (a) Yaffe, M. B. Nat. ReV. Mol. Cell Biol. 2002, 3, 177. (b) Yaffe, M. B.;
Elia, A.. Curr. Opin. Cell Biol. 2001, 13, 131. (c) Pawson, T.; Gish, G. D.;
Nash, P. Trends Cell Biol. 2001, 11, 504. (d) Yan, K. S.; Kuti, M.; Zhou,
M.-M. FEBS Lett. 2002, 513, 67. (e) Durocher, D.; Jackson, S. P. FEBS
Lett. 2002, 513, 58. (f) Macias, M. J.; Wiesner, S.; Sudol, M. FEBS Lett.
2002, 513, 30.
(5) Johnson, L. N.; Lewis, R. J. Chem. ReV. 2001, 101, 2209.
9
2052
J. AM. CHEM. SOC. 2006, 128, 2052-2058
10.1021/ja056585k CCC: $33.50 © 2006 American Chemical Society

Disruption of Phosphoprotein
−
Protein Interaction
A R T I C L E S
Chart 1. Molecular Structure of the Artificial Receptors, and
devoted to the development of small molecular inhibitors for
the surface interaction of the phosphoprotein binding domain
such as the SH2 domain.9 Most of them are peptidemimetics
of the phosphoprotein segment that can interact with a shallow
surface pocket (surface groove) of the phosphoprotein binding
domain. On the other hand, no small molecular inhibitors that
directly interact with a phosphoprotein have yet been reported.
This should be an alternative and promising strategy for
developing unique inhibitors for these phosphoprotein-protein
interactions.
Amino Acid Sequences of the Pin1WW Domain (6-39) and the
CTD Phosphopeptides Discussed Here
We have recently discovered the bis(Zn(II)-dipicolylamine
(Dpa)) derivatives as the first artificial receptors for multiply
phosphorylated peptides.10 Under neutral aqueous conditions,
a binuclear Zn(II)-Dpa-bipyridyl derivative strongly (Kapp > 105
M-1) binds to bis-phosphorylated peptides via a two-point
interaction, based on coordination chemistry between the Zn(II)
sites and the phosphate groups (i.e., the cross-linking strategy).
The preliminary results encouraged us to attempt the disruption
of the phosphoprotein-protein interaction using the Zn(II)-Dpa
complexes. Since the multiple phosphorylation serves as a
common mechanism regulating protein function and constitutes
a complicated regulatory program for signaling pathways like
a dynamic “molecular barcode” in living cells,11 it is expected
that the targeting multiphosphorylated proteins by a synthetic
small molecule will open a new possibility for artificial
regulation of a signal transduction cascade. We now report that
some of the artificial receptors can serve as a potent disruptor
for the phosphoprotein-protein surface interaction via a cross-
(6) (a) Yin, H.; Hamilton, A. D. Angew. Chem., Int. Ed. 2005, 44, 4130. (b)
Roy, B. C.; Banerjee, A. L.; Swanson, M.; Jia, X. G.; Haldar, M. K.; Mallik,
S.; Srivastava, D. K. J. Am. Chem. Soc. 2004, 126, 13206. (c) Banerjee,
A. L.; Swanson, M.; Roy, B. C.; Jia, X.; Haldar, M. K.; Mallik, S.;
Srivastava, D. K. J. Am. Chem. Soc. 2005, 126, 10875. (d) Jain, R.; Ernst,
J. T.; Kutzki, O.; Park, H. S.; Hamilton, A. D. Mol. DiVersity 2004, 8, 89.
(e) Salvatella, X.; Martinell, M.; Gair´ı, M.; Mateu, M. G.; Feliz, M.;
Hamilton, A. D.; de Mendoza, J.; Giralt, E. Angew. Chem., Int. Ed. 2004,
43, 196. (f) Gradl, S. N.; Felix, J. P.; Isacof, E. Y.; Garcia, M. L.; Trauner,
D. J. Am. Chem. Soc. 2003, 125, 12668. (g) Fazal, M. A.; Roy, B. C.; Sun,
S.; Mallik, S.; Rodgers, K. R. J. Am. Chem. Soc. 2001, 123, 6283. (h)
Peczu, M. W.; Hamilton, A. D. Chem. ReV. 2000, 100, 2479.
(7) (a) Ojida, A.; Miyahara, Y.; Kohira, T.; Hamachi, I. Biopolymers (Peptide
Science) 2004, 76, 177. (b) Ojida, A.; Mito-oka, Y.; Sada, K.; Hamachi, I.
J. Am. Chem. Soc. 2004, 126, 2454. (c) Ojida, A.; Mito-oka, Y.; Inoue,
M.; Hamachi, I. J. Am. Chem. Soc. 2002, 124, 6256. (d) Mito-oka, Y.;
Tsukiji, S.; Hiraoka, T.; Kasagi, N.; Shinkai, S.; Hamachi, I. Tetrahedron
Lett. 2001, 42, 7059. (e) Takashima, H.; Shinkai, S.; Hamachi, I. Chem.
Commun. 1999, 2345.
linking fashion. The detailed binding studies and the inhibition
assay using the phosphorylated CTD peptide and the WW
domain as a model pair revealed that the intramolecular
juxtaposition of the two Zn(II)-Dpa sites at a suitable distance
is crucial for the strong molecular recognition of the phospho-
rylated CTD peptide so as to produce the effective disruption
for their surface interaction.
Results and Discussion
Molecular Design of BisZn(II)-Dpa as Disruptor for
Phosphopeptide Interaction of WW Domain. WW domains
are a small binding motif for the proline-rich sequence and found
in many different signaling or structural proteins.2a,4d,12 They
are composed of approximately 40 amino acids and folded as
a stable, triple-stranded â-sheet. Among them, a WW domain
derived from the peptidyl-prolyl isomerase Pin1 is a class IV
WW domain, which specifically interacts with pSer-Pro or pThr-
Pro motifs.13 In this study, a recognition pair of the Pin1 WW
domain (6-39) and the CTD (C-terminal domain of RNA
polymerase II) doubly phosphorylated peptide was employed
as a model system for evaluating the inhibitory activity of our
artificial receptors (Chart 1), because of the rich structural basis
of their surface complex. An X-ray crystallographic analysis
(8) (a) Baberjee, A. L.; Tobwala, S.; Haldar, M. K.; Swanson, M.; Roy, B. C.;
Mallik, S.; Srivastava, D. K. Chemm. Commun. 2005, 2549. (b) Yin, H.;
Lee, G.-I.; Sedey, K. A.; Kutzki, O.; Park, H. S.; Orner, B. P.; Ernst, J. T.;
Wang, H.-G.; Sebti, S. M.; Hamilton, A. D. J. Am. Chem. Soc. 2005, 127,
10191. (c) Yin, H.; Lee, G.-I.; Sedey, K. A.; Rodriguez, J. M.; Wang, H.-
G.; Sebti, S. M.; Hamilton, A. D. J. Am. Chem. Soc. 2005, 127, 5463. (d)
Walensky, L. D.; Kung, A. L.; Escher, I.; Malia, T. J.; Barbuto, S.; Wright,
R. D.; Wagner, G.; Verdine, G. L.; Korsmeyer, S. J. Science 2004, 305,
1466. (e) Arkin, M. R.; Wells, J. A. Nat. ReV. Drug DiscoVery 2004, 3,
301. (f) Berg, T. ChemBioChem 2004, 5, 1051. (g) Horswill, A. R.; Savinov,
S. N.; Benkovic, S. J. PNAS 2004, 44, 15591. (h) Vassilev, L. T.; Vu, B.
T.; Graves, B.; Carvajal, D.; Podlaski, F.; Filipovic, Z.; Kong, N.; Kammlot,
U.; Lukacs, C.; Klein, C.; Fotouhi, N.; Liu, E. A. Science 2004, 303, 844.
(i) Berg, T. Angew. Chem., Int. Ed. 2003, 42, 2462. (j) Cochran, A. G.
Curr. Opin. Chem. Biol. 2001, 5, 654. (k) Cochran, A. G. Chem. Biol.
2000, 7, R85. (l) Blaskovich, M. A.; Lin, Q.; Delarue, F. L.; Sun, J.; Park,
H. S.; Coppola, D.; Hamilton, A. D.; Sebti, S. M. Nat. Biotechnol. 2000,
18, 1065.
(9) (a) Song, Y.-L.; Roller, P. P.; Long, Y.-Q. Bioorg. Med. Chem. Lett. 2004,
14, 3205. (b) Sundaramoorthi, R.; Kawahata, N.; Yang, M. G.; Shakespeare,
W. C.; Metcalf, C. A., III; Wang, Y.; Merry, T.; Eyermann, C. J.; Bohacek,
R. S.; Narula, S.; Dalgarno, D. C.; Sawyer, T. K. Biopolymers (Peptide
Science) 2003, 71, 717. (c) Mu¨ller, G. Top. Curr. Biol. 2000, 211, 19. (d)
Cody, W. L.; Lin, Z.; Panek, R. L.; Rose, D. W.; Rubin, J. R. Curr. Pharm.
Des. 2000, 6, 59. (e) Nguyen, J. T.; Turck, C. W.; Cohen, F. E.;
Zuckermann, R. N.; Lim, W. A. Science 1998, 282, 2088.
(12) (a) Sudol, M.; Sliwa, K.; Russo, T. FEBS Lett. 2001, 490, 190. (b) Zarrinpar,
A.; Lim, W. A. Nat. Struct. Biol. 2000, 7, 611.
(13) (a) Verdecia, M. A.; Bowman, M. E.; Lu, K. P.; Hunter, T.; Noel, J. P.
Nat. Struct. Biol. 2000, 7, 639. (b) Xu, Y.-X.; Hirose, Y.; Zhou, X. Z.; Lu,
K. P.; Manley, J. L. Genes & DeVelopment 2003, 17, 2765. (c) Lu, P.-J.;
Zhou, X.-Z.; Liou, Y.-C.; Noel, J.-P.; Lu, K.-P. J. Biol. Chem. 2002, 277,
2381. (d) Wintjens, R.; Wieruszeski, J.-M.; Drobecq, H.; Rousselot-Pailley,
P.; Bue´e, L.; Lippens, G.; Landrieu, I. J. Biol. Chem. 2001, 276, 25150.
(10) Ojida, A.; Inoue, M.; Mito-oka, Y.; Hamachi, I. J. Am. Chem. Soc. 2003,
125, 10184.
(11) (a) Yang, X.-J. Oncogene 2005, 24, 1653. (b) Holmberg, C. I.; Tran, S. E.
F.; Erikson, J. E.; Sistonen, L. Trends Biochem. Sci. 2002, 27, 619.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2053

A R T I C L E S
Ojida et al.
Figure 1. ORTEP drawing (50% probability ellipsoids) of the biphenyl
compound 1 (C38H36N6‚Zn2Cl4‚Et2O). Hydrogen atoms and disordered
solvent molecules are omitted for clarity.
of the complex of the Pin 1 WW domain with the doubly
phosphorylated CTD peptide revealed that the phosphate group
of the Ser9 of the CTD peptide tightly interacts with the Pin1
WW domain through the hydrogen bonding with the side chains
of Ser16, Arg 17, and Tyr23 and the backbone amide of Arg17,
whereas the phosphate group Ser6 does not participate in the
specific binding. The complex structure also indicates that the
distance of the two phosphate groups is 9.7 Å.13a
To obtain a potent disruptor for the phosphopeptide/WW
domain interaction using the cross-linking strategy, we prepared
a family of binuclear Zn(II) complexes (1-7), which are
designed to possess two Zn(II)-Dpa centers at a distinct distance
by changing the spacer moiety (Chart 1). The receptor 5
possessing a phenyl group as the spacer has the shortest distance
between the two Zn(II) centers (4-6 Å), whereas the longest
one is 6 or 7 (14-17 Å), which possesses a 3,3′-stylyl or
terphenyl group as the spacer. The distance between the two
Zn(II) centers is also adjustable by changing the substitution
position of the Dpa sites on the biaryl spacer unit as shown in
the biphenyl (1 and 3) and the bipyridyl types of artificial
receptors (2 and 4). Among them, we successfully obtained the
detail structural information about the biphenyl receptor of 1
by an X-ray crystallographic analysis (Figure 1). The binuclear
Zn complex of 1 with Cl- counterions crystallized from
CH3CN-Et2O to afford a single crystal suitable for an X-ray
analysis. It was revealed in the crystal state that the dihedral
angle of the two phenyl groups of 1 is 37° and the distance of
the two Zn(II) centers is 11.5 Å, which is relatively close to
that of the two phosphate groups of the doubly phosphorylated
CTD peptide (9.7 Å). Thus, it might be expected that the
biphenyl type of receptor might be a potent inhibitor candidate
for the phosphoprotein/WW domain interaction. All artificial
receptors were prepared from the corresponding halomethyl
compound by the substitution reaction with 2,2′-dipicolylamine
and then by the metal complexation with Zn(NO3)2. As a
monodentate control compound, the mononuclear receptor 8 was
also prepared by the same procedure. The structures of these
receptors were fully characterized by 1H NMR, mass spectros-
copy, and elemental analysis (see Supporting Information).
Binding Studies between Artificial Receptor and Phos-
phorylated Peptide by Isothermal Titration Calorimetry.
Prior to evaluating the inhibitory activity of these receptors, the
binding affinity of the receptors with the phosphorylated peptides
was quantitatively determined by isothermal titration calorimetry
(ITC). Figure 2 shows the typical titration data of 1 with the
Figure 2. Typical ITC titration curve and processed data (for the titration
of 1 with pS-6,9). Measurement conditions: [1] ) 20 µM, [pS-6,9] ) 300
µM (24 × 10 µL injections), 50 mM HEPES buffer, pH 7.2, 25 °C.
doubly phosphorylated CTD fragment, pS-6,9 peptide. The
resultant binding isotherm was fitted with a curve derived from
the 1:1 binding algorithm (n ) 1.13) to yield a binding constant
of (8.11 ( 0.46) × 106 M-1. Interestingly, the present interaction
was an entropy-driven endothermic process (∆S > 0, ∆H >
0), presumably due to the release of water molecules from the
solvent sphere fixed around the receptor and/or the phospho-
peptide. Table 1 summarized the thermodynamic parameters,
as obtained by fitting the binding isotherm to a 1:n interaction
model. Almost the same strong binding affinity was obtained
in the case of the bipyridyl receptor 2 (K ) (2.80 ( 0.36) ×
106 M-1, Table 1). The binding of 1 or 2 with the pS-6,9 peptide
was much stronger than those of 8 with the pS-6,9 peptide or
the singly phosphorylated pS-9 peptide, in which the binding
affinities were estimated to be in the range of 103 M-1 by
analyzing with a 1:2 binding model (n ) 0.5) or with a 1:1
binding model (n ) 1.0), respectively (Figure S1). These results
suggest that 1 or 2 interacts with the pS-6,9 peptide via a two-
point interaction so as to exhibit the remarkably strong binding
affinity in aqueous solution. This is also supported by the ITC
data between 1 and the pS-9 peptide, in which the slope of the
titration curve was gentle even in the measurement with a
relatively high concentration of 1 (100 µM), indicating that the
binding with the pS-9 peptide is much weaker than that with
the doubly phosphorylated pS-6,9 peptide (Figure S1).14 It is
(14) The rather largely deviated n value from 1 (n ) 1.35) and the relatively
large heat formation (7.89 kcal/mol) compared to the other bindings suggest
that the complex formation between 1 and pS-9 peptide is not a single
binding mode but a mixture of several binding modes containing a 1:2 or
higher order complexes.
9
2054 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Disruption of Phosphoprotein−Protein Interaction
A R T I C L E S
Table 1. Stoichiometry (n), Binding Constant (K, M-1), Enthalpy (∆H, kcal mol-1), and Entropy (T∆S, kcal mol-1) for the Interactions of the
Artificial Receptors with the Phosphorylated Peptidesa
peptide
1
2
3
4
5
6
7
8
pS-6,9
n
K
1.14
0.99
0.69
0.70
1.00
0.57
0.50
(8.11 ( 0.46) × 106 (2.80 ( 0.36) × 106
(4.26 ( 0.30) × 105
b
(8.49 ( 0.51) × 103
4.60 ( 0.12
9.95
∆H
T∆S 1.54
n
K
∆H
T∆S
6.01 ( 0.02
4.53 ( 0.04
13.3
9.67 ( 0.23 6.79 ( 0.26 4.45 ( 0.05
9.34 ( 0.17
12.2
1.16
pS-9
1.35
1.00
b
b
b
(4.24 ( 0.42) × 103
4.20 ( 0.66
9.14
b
b
(1.11 ( 0.11) × 103
3.09 ( 0.22
7.23
7.89 ( 0.13
a The themodynamic parameters were obtained by fitting the binding isotherm to a 1:n interaction model, assuming a single type of binding mode. b Not
measured.
to be noted that the strong binding of 1 or 2 with the pS-6,9
peptide is due to the favorable positive ∆S value that is
significantly larger than those observed in the weak interactions
of the mononuclear complex 8. The large entropic gain of 1 or
2 observed here may be partially ascribed to the cooperativity
of the two Zn(II)-Dpa sites in the binding to the pS-6,9 peptide.15
The ∆H value in the binding between 8 and the pS-9 peptide
((3.09 ( 0.22) kcal/mol) is almost half of that of the binding
between 1 with the pS-6,9 peptide (∆H ) 6.01 ( 0.02 kcal/
mol). In addition, the bindings between 2, 5, and 8 with the
pS-6,9 peptide are also endothermic processes with 4.45-4.60
kcal/mol as a positive ∆H value.16 On the basis of these data,
the average heat for the binding of a single Zn(II)-Dpa site per
phosphate site is reasonably determined to be 2-3 kcal/mol.
ITC experiments were also conducted for the other binuclear
receptors 3, 4, and 6, which have the two Zn(II) centers with
different distances (Table 1). The rapid cease of the heat
formation before the addition of 1 equiv of the pS-6,9 peptide
gave the discrepancy in the 1:1 binding stoichiometry (n )
0.57-0.70), suggesting that the complex formation with 3, 4,
or 6 is not 1:1 and may contain several different binding modes.
In the case of the terphenyl derivative 7 that was not sufficiently
soluble in water, we were not able to evaluate the binding
property. These results imply that the strong affinity with the
1:1 binding with the doubly phosphorylated pS-6,9 peptide is
achieved only in the case when the distance between the two
Zn(II)-Dpa sites of the artificial receptor is suitable for the cross-
linking interaction with the two phosphate groups. Since the
distance of the two Zn(II) centers in 1 and 2 can be varied by
the rotation of the biaryl unit, it may be reasonably assumed
that they flexibly adjust the two Zn(II) distances to fit the
distance between the two phosphate sites of the CTD peptide.
On the other hand, such a flexible distance modulation does
not operate for the corresponding para-substituted isomer 3 and
4, resulting in the non-1:1 interaction.
Figure 3. (a) CD spectra of the pS-6,9 peptide (20 µM) (O) and the
complex between 1 (20 µM) and the pS-6,9 peptide (20 µM) (b). (b) CD
Job titration between 1 and the pS-6,9 peptide, [1] + [pS-6,9] ) 40 µM.
Measurement conditions: 10 mM borate buffer, pH 8.0, 25 °C.
Cross-Linking Binding Mode of Artificial Receptor with
Phosphopeptide. The cross-linking binding mode of the recep-
tors 1 or 2 with the phosphorylated peptides was explored by
the circular dichroism (CD) studies. As shown in Figure 3a,
the strong Cotton peaks at 200, 230, and 265 nm appeared in
the complex 1 with the doubly phosphorylated pS-6,9 CTD
peptide (20 µM in 10 mM borate buffer, pH 8.0). Similar strong
Cotton peaks were also observed during the interaction with
the bipyridyl receptor 2 (Figure S2). Since the peak maxima at
230 and 265 nm correspond to the absorption maxima of the
receptor 1 or 2, it is reasonably considered that the strong
induced CDs are originated from the receptors by complexation
with the pS-6,9 CTD peptide. On the other hand, no CD peaks
were induced in the case of the interaction between the
monophosphorylated pS-9 CTD peptide and 1 (data not shown).
The induced CDs observed in the case of 1 or 2 upon the pS-
6,9 peptide binding may reflect a chirally defined rigid
(15) (a) Tobey, S. L.; Anslyn, E. V. J. Am. Chem. Soc. 2003, 125, 10963. (b)
Williams, D. H.; Westwell, M. S. Chem. Soc. ReV. 1998, 27, 57. (c) Jencks,
W. P. Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 4046.
(16) We previously revealed that the receptor 5 binds to a single phosphate
group of a phosphopeptide using the two Dpa sites in a chelation-like
fashion.7 However, the much stronger binding of 5 with the pS-6,9 peptide
(K ) 4.26 × 105 M-1) compared to that with the pS-9 peptide (K ) 4.24
× 103 M-1) in the present case implies that 5 interacts with the two
phosphate groups of pS-6,9 peptide with a cross-linking fashion like the
binding mode of 1 or 2.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2055

A R T I C L E S
Ojida et al.
conformation of the receptors via a two-point fixation within
the complex with the pS-6,9 peptide. The job plot between 1
and the pS-6,9 peptide was also examined using the induced
CD peak. The maximum CD change was observed when the
molar ratio of 1 and the pS-6,9 peptide is equivalent, clearly
indicating that the binding stoichiometry is 1:1 (Figure 3b). The
1:1 binding stoichiometry was also confirmed by the Job plot
in the complex formation between 2 and the pS-6,9 peptide
(Figure S2).
The two-point interaction of 1 with the pS-6,9 peptide is also
supported by a 31P NMR study. A broad single peak at 6.40
ppm corresponding to the two phosphate groups of pS-6,9
without 1 changed into the two separated and upfield shifted
peaks at 4.53 and 3.68 ppm upon the addition of 1 equiv of 1
(Figure S3). The upfield shifts of the 31P signal by complexation
with the Zn(II)-Dpa complex was also observed in the case of
the complexation with the 1,8-bisZn(II)-Dpa-anthracene com-
pound as previously reported.3c This result indicates that the
cross-linking receptor 1 directly interacts with the two phosphate
sites of the pS-6.9 peptide. In a 1H NMR study, the four protons
due to the C6 positions of the four pyridine rings of 1 were
observed as a doublet at 8.50 ppm which is indistinguishable
from each other, suggesting that 1 has a highly symmetrical
structure in aqueous solution (Figure S4). However, this peak
split into the four distinct peaks at 8.89, 8.78, 8.72, and 8.52
ppm upon addition of the pS-6,9 peptide. Other pyridine protons
also changed into the different peaks in the complex with the
pS-6,9 peptide. These results imply that the two Zn(II)-Dpa sites
of 1 are involved in the interaction with the two phosphate sites
of the pS-6,9 peptide so that four pyridine rings are located in
the different environments.
Figure 4. Fluorescent anisotropy change of the rhodamine-labeled phos-
phopeptide Rho-pS-6,9 (5 µM) upon the addition of the various artificial
receptors (0-25 µM) in the presence of the WW domain (25 µM). (a) pS-
6,9 peptide with 1 (9), 3 (b), 5 (2), 6 ([), and 8 (0). (b) pS-9 peptide
with 1 (2) and 2 (b). Measurement conditions: 50 mM HEPES buffer,
pH 7.2, 20 °C, λex ) 550 nm.
Inhibitory Activity of Artificial Receptors Evaluated by
Fluorescence Polarization Assay. The inhibitory activity of
the artificial receptors toward the interaction between the Pin1
WW domain and the CTD peptide was evaluated by a
fluorescence polarization assay using the rhodamine-labeled
doubly phosphorylated peptide (Rho-pS-6,9). The fluorescence
anisotropy value of Rho-pS-6,9 (50 mM HEPES, pH 7.2)
increased to ∼0.12 by the addition of the Pin1 WW domain,
indicating the complex formation of Rho-pS-6,9 with the WW
domain. The curve-fitting analysis of the obtained saturation
curve afforded the affinity constant of 1.8 × 105 M-1 (Figure
S5). When the receptor 1 was added to the solution containing
the Pin 1 WW domain/Rho-pS-6,9 complex, a sharp decrease
in the anisotropy value took place (Figure 4a). This change
indicated that the receptor 1 releases Rho-pS-6,9 from the
complex by the competitive displacement of the WW domain
with 1. A curve fitting analysis of the displacement titration
the monophosphorylated peptide with the WW domain (Figure
4b). Such a sharp contrast suggests that the second phosphate
site, which is not masked by the WW domain, is crucial for the
effective disruption of the interaction. This is quite consistent
with the ITC results, which shows the much weaker binding
affinity of 1 with pS-9 rather than that with pS-6,9 (Table 1).
As another control experiment, the mononuclear Zn(II) complex
8 was also used in the displacement assay for Rho-pS-6,9
(Figure 4a). As expected from the ITC result, 8 did not induce
any anisotropy change. On the other hand, in the displacement
assay using the receptors 3-6 (Figure 4a and Figure S6), which
cannot form a tight 1:1 binding complex with the pS-6,9 peptide,
the anisotropy value gently decreased upon addition of them,
indicating that these receptors are moderate inhibitors, less potent
than 1. Thus, it can be concluded that the strong binding affinity
due to the two-point metal-ligand interaction is essential for
the effective disruption of the surface interaction of the WW
domain with the phosphopeptide (Figure 5). It is noteworthy
that the effective inhibition potency of 1 and 2 may enable us
to propose that the direct and tight binding of the small artificial
receptor to the phosphate group on a protein surface becomes
an alternative promising strategy for the inhibition of the
phosphoprotein-protein interaction.
gave the Ki value of 1 to be 1.84 × 106 M-1 17
. The addition of
the bipyridyl type receptor 2 also caused a sharp decrease in
the anisotropy value like 1 (Ki ) 1.38 × 106 M-1, Figure S6),
showing that both are effective disruptors for the interaction
between the pS-6,9 CTD peptide and the WW domain. The
displacement assay was also conducted for the monophos-
phorylated peptide (Rho-pS9), the affinity of which for the Pin1
WW domain (Kapp ) 1.8 × 105 M-1) was comparable to the
case of Rho-pS-6,9 (Figure S5). Interestingly, in this case, the
anisotropy value did not decrease at all with the addition of 1,
indicating that 1 did not effectively inhibit the interaction of
(17) Kirk, S. R.; Luedtke, N. W.; Tor, Y. J. Am. Chem. Soc. 2000, 122, 980.
9
2056 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Disruption of Phosphoprotein−Protein Interaction
A R T I C L E S
Figure 5. Schematic representation of the disruption of the phosphopeptide/WW domain interaction by the Zn(II) dipicolylamine-based artificial receptors.
3,3′-Bis[(2,2′-dipicolylamino)methyl]biphenyl 2Zn(NO3)2 Com-
plex (1). To a solution of 3,3′-bis[(2,2′-dipicolylamino)methyl]biphenyl
(201 mg, 0.35 mmol) in CH3CN (2 mL) was added dropwise an aqueous
solution of 0.596 M Zn(NO3)2 (1.05 mL, 0.63 mmol), and the mixture
was stirred for 30 min at rt. After removal of the solvent in vacuo, the
residue was diluted with distilled H2O, filtered through a cellulose
acetate filter (pore size 0.45 µm), and lyophilized. The obtained solid
was filtered and washed with AcOEt to give the Zn(II) complex (220
mg, 66%) as a colorless powder: 1H NMR(400 MHz, D2O) δ 3.77
(4H, s), 3.87 (4H, d, J ) 15.6 Hz), 4.27 (4H, d, J ) 16.0 Hz), 7.15
(4H, d, J ) 8.0 Hz), 7.43-7.47 (6H, m), 7.50-7.56 (6H, m), 7.95
(4H, t, J ) 7.6 Hz), 8.59 (4H, d, J ) 4.8 Hz). 13C NMR (150 MHz,
D2O + CD3OD) δ 56.8, 56.9, 126.0, 126.4, 128.8, 130.9, 131.1, 132.0,
133.4, 142.3, 142.6, 149.4, 155.8. FAB-MS m/e 893.15 [M - NO3]+.
Anal. Calcd for C38H36N6‚2Zn(NO3)2 : C, 47.76; H, 3.80; N, 14.66.
Found: C, 47.42; H, 3.75; N, 14.25.
Conclusion
We have demonstrated that the artificial receptors bear-
ing the two Zn(II)-Dpa sites can strongly bind to the bis-
phosphorylated peptide in a cross-linking manner under the
neutral aqueous conditions, when the distance between the two
Zn(II) centers can appropriately fit that of the two phosphate
groups of the doubly phosphorylated peptide. On the basis of
this finding, we have successfully carried out the effective
disruption of the CTD phosphopeptide/Pin1 WW domain
interaction using these types of small molecules. The cross-
linking mode for inhibition by the artificial receptors presented
herein, i.e., the selective recognition of a multiphosphorylated
peptide depending on the number and the distance of the
phosphate groups, should be a valuable strategy for disrupting
other complexes involving various multiphosphorylated proteins.
When the sufficient binding selectivity and affinity is acquired
by combining a variety of interaction forces, this approach may
lead to effective small molecular modulators or disruptors for
signal cascades in living cells in future. We are currently
evaluating this line of research.
Peptide Synthesis. The peptide coupling was performed with an
automated peptide synthesizer (ABI 433A, Applied Biosystems).
Purification of peptide was carried out with a Hitachi L-7100 HPLC
system. The molecular weight of the peptide was confirmed by mass
spectrometer (MALDI-TOF, Perseptive Biosystems Voyager DE-RP)
using 3,5-dimethoxy-4-hydroxycinnmaic acid as a matrix.
(i) Synthesis of Pin1 WW Domain. The Pin1 WW domain (6-39)
was synthesized using the standard Fmoc-based FastMoc coupling
chemistry (0.25 mmol scale) on Alko-PEG Resin preloaded with Fmoc-
Glycine (0.1 mmol, loading rate: ∼0.22 mmol/g, purchased from
Watanabe Chemical Industries, Ltd.). The coupling reactions were
performed using 10 equiv of amino acid and 10 equiv of 1-hydroxy-
benzotriazole (HOBt). The peptide cleavage from resin and side-chain
deprotection were carried out by treatment with 10 mL of TFA
containing m-cresol (0.25 mL), thioanisole (0.75 mL), and ethanedithiol
(0.75 mL) over 1 h at rt. After removal of TFA in vacuo, the crude
peptide was precipitated in cold diethyl ether and purified by reversed-
phase HPLC (column; YMC-pack ODS-A, 20Φ × 250 mm). Purifica-
tion conditions were as follows: mobile phase, CH3CN (0.1% TFA)/
H2O (0.1% TFA) ) 23/77 f 30/70 (linear gradient over 40 min); flow
rate, 9.9 mL/min; detection, UV (220 nm). MALDI-TOF mass: calcd
for [M + H]+ 4025.1, found 4024.6.
Experimental Section
Synthesis of the Artificial Receptors. All zinc complexes (1-8)
were synthesized from the corresponding halomethyl compound by the
nucleophilic substitution with 2,2′-dipicolylamine and the subsequent
zinc complexation. Characterizations of these compounds were per-
1
formed by H NMR, mass spectroscopy, and elemental analysis (see
Supporting Information). Typical synthetic procedures are as follows:
3,3′-Bis[(2,2′-dipicolylamino)methyl]biphenyl. A solution of 3,3′-
bis(bromomethyl)biphenyl18 (700 mg, 2.06 mmol), 2,2′-dipicolylamine
(0.82 g, 4.12 mmol), and potassium carbonate (1.13 g, 8.23 mmol) in
anhydrous DMF (10 mL) was stirred for 1.5 h at rt. After dilution with
1 N HCl, the resulting mixture was washed with AcOEt. The aqueous
phase was alkalized with 2 N NaOH and extracted with AcOEt (×2).
The combined organic layers were washed with water and brine
followed by drying over MgSO4. After removal of the solvent in vacuo,
the residue was purified by column chromatography (silica gel, CHCl3/
MeOH/NH3aq ) 60/1/1) to give the title compound (773 mg, 65%) as
a pale yellow oil: 1H NMR (400 MHz, CDCl3) δ 3.77 (4H, s), 3.86
(8H, s), 7.17 (4H, t, J ) 6.0 Hz), 7.37-7.46 (6H, m), 7.61-7.67 (10H,
m), 8.52 (4H, d, J ) 4.8 Hz). 13C NMR (150 MHz, CDCl3) δ 58.6,
60.1, 122.0, 122.8, 126.0, 127.6, 127.7, 128.7, 136.4, 139.5, 141.2,
149.0, 159.7. FAB-MS m/e 577.34 [M + H]+.
(ii) Synthesis of Phophopeptide. The peptides were synthe-
sized using the standard Fmoc-based FastMoc coupling chemistry
(0.1 mmol scale) on Fmoc-Amide Resin (Applied Biosystems). Fmoc-
Ser[PO(Obzl)OH]-OH (Watanabe Chemical Industries, Ltd.) was used
as a phosphorylated amino acid unit for the peptide coupling. The
peptide cleavage from resin and side-chain deprotection were carried
out by the treatment with TFA-m-cresol-thioanisole (86:2:12) over
1 h at rt. After removal of TFA in vacuo, the crude peptide was
precipitated in tert-butyl-methyl ether and purified by reversed-phase
HPLC (column; YMC-pack ODS-A, 20Φ × 250 mm). Purification
(18) Tashiro, M.; Yamato, T. J. Org. Chem. 1985, 50, 2939.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 2057

A R T I C L E S
Ojida et al.
conditions were as follows: mobile phase, CH3CN (0.1% TFA)/H2O
(0.1% TFA) ) 5/95 f 24/76 (linear gradient over 25 min); flow rate,
9.9 mL/min; detection, UV (220 nm). MALDI-TOF mass pS-6,9: calcd
for C47H70N12O25P2 [M - H]- 1265.1, found 1265.8; pS-9: calcd for
C47H71N12O22P [M - H]- 1187.1, found 1186.1.
(iii) Synthesis of Rhodamine-Conjugated Peptide. A purified
phosphopeptide, tetramethylrhodamine isothiocyanate (2 equiv, a
mixture of 5-, 6-isomer purchased from Molecular Probes, Inc.,), and
diisopropylethylamine (4 equiv) were dissolved in DMF, and the
mixture was stirred 5 h at rt. After completion of the reaction (checked
by HPLC analysis), the obtained peptide was purified by HPLC
(column; Desovil ODS-UG-5, 10Φ × 250 mm). Purification conditions
were as follows: mobile phase, CH3CN/100 mM AcNH4 in H2O )
15/85 f 40/60 (linear gradient over 40 min); flow rate, 3 mL/min;
detection, UV (220 nm). MALDI-TOF mass: calcd for C72H90N14O29P2S
[M - H]- 1708.6, found 1711.23.
X-ray Crystallography. X-ray diffraction data were collected on a
Rigaku R-AXIS RAPID diffractometer with a 2D area detector using
graphite-monochromatized Cu K radiation (λ ) 1.5418 Å) at ca. 120
K. Lattice parameters were obtained by least-squares analysis from
reflections for three oscillation images. Direct methods (SIR92) were
used for the structure solution. All calculations were performed using
the TEXSAN19 crystallographic software package. All non-hydrogen
atoms are found by Fourier syntheses. The structure was refined by a
full matrix least-squares procedure using observed reflections (>2.0 σ
(I)) based on F2, and all non-hydrogen atoms were refined with
anisotropic displacement parameters and hydrogen atoms were placed
in idealized positions with isotropic displacement parameters relative
to the connected non-hydrogen atoms and not refined. Crystallographic
parameters are summarized in the Supporting Information.
Isothermal Titration Calorimetry (ITC). ITC titration was per-
formed on an Isothermal Titration Calorimeter from MicroCal Inc. All
measurements were conducted at 298 K. In general, a solution of the
peptide (1-2 mM) in 50 mM HEPES buffer (pH 7.2) was injected
stepwise (10 µL × 24) to a solution of the artificial receptor (20-100
µM) dissolved in the same solvent system. The concentration of the
phosphorylated peptide was determined based on the UV absorbance
of the tyrosine residue (ꢀ ) 1530 M-1 cm-1 at 274 nm20). The measured
heat flow was recorded as a function of time and converted into
enthalpies (∆H) by integration of the appropriate reaction peaks.
Dilution effects were corrected by subtracting the results of a blank
experiment with a solution of 50 mM HEPES solution (pH 7.2) in place
of peptide solution under the identical experimental conditions. The
binding parameters (Kapp, ∆H, ∆S, n) were evaluated by applying a
one-site model using the software Origin (MicroCal Inc.).
with smoothing to reduce noise. In a Job plot, a titration of continuous
variation was conducted where the concentration of both the receptor
1 or 2 and the pS-6,9 peptide was varied between 0 and 40 µM such
that the number of moles in solution stayed constant.
Nuclear Magnetic Resonance (NMR). 31P NMR spectra were
recorded on a Brucker DRX-600 (243 MHz) at 25 ( 1 °C. 85% H3PO4
was used as an external reference. Samples were prepared using a
concentration of 0.5 mM pS-6,9 peptide in 50 mM HEPES (pH 7.2)/
D2O (90:10) in the absence and presence of 0.5 mM receptor 1.
1H NMR spectra were recorded on a JEOL JNM A-400 (400 MHz)
at 25 ( 1 °C. Samples were prepared using a concentration of 0.5 mM
of 1 in D2O in the absence and presence of 0.5 mM of the pS-6,9
peptide. The observed peaks were individually assigned by the COSY
experiments.
Fluorescence Anisotropy Experiment. Fluorescence anisotropy
experiments were performed with a Perkin-Elmer LS55 spectrometer.
A stock solution of the artificial receptor (1-2 mM in distilled H2O)
was titrated into an aqueous solution (0.5 mL) containing the rhodamine-
conjugated peptide (5 µM) and the Pin1 WW domain (25 µM) in 50
mM HEPES (pH 7.2). The fluorescence emission of the rhodamine-
conjugated peptide (Rho-pS-6,9 or Rho-pS-9) was measured (λex
)
550 nm) at 20 °C upon titration of the receptors, and the fluorescence
anisotropy value (r) was calculated using the emission intensity at 576
nm by eq 1,21
IV - G ‚ IH
IV + 2G ‚ IH
r )
(1)
where IV and IH are the fluorescence intensities observed through
polarizers parallel and perpendicular to the polarization of the exciting
light, respectively, and G is a correction factor accounting for
instrumental differences in the fluorescence detection. Experimental
data were fitted into eq 2 to determine the IC50 value, which was in
turn used to determine the Ki value (M-1) according to the reported
method,17
X ‚ rmin + rmax ‚ IC50
r )
(2)
X + IC50
where X is the concentration of the artificial receptor.
Acknowledgment. M.I. thanks JSPS for his JSPS predoctoral
fellowship. This work is partially supported by the Nagase
Science Foundation.
Supporting Information Available: Synthesis and charac-
terization of the artificial receptors, selected crystallographic
parameters for 1, selected data of the ITC experiments,
experimental data of the CD study of 2, 31P and 1H NMR studies
of 1 and the pS-6,9 peptide, and selected data of the fluorescent
anisotropy experiments (PDF); X-ray crystallographic data of
1 (CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
Circular Dichroism (CD). CD spectra were recorded using a
JASCO J-720W spectropolarimeter. The aqueous solution of the
phosphopeptide (20 µM) in 10 mM borate buffer (pH 8.0) in the absence
and the presence of the artificial receptor (20 µM) was prepared, and
CD spectra of the solutions were measured from 340 to 200 nm (scan
speed; 50 nm/min) at 4 °C using a water-jacketed quartz cell (1 mm
path length). Each spectrum represents the average of 16 time scans
(19) TEXAN, X-ray structure analysis package; Molecular Strucuture Corp.: The
Woodland, TX, 1985.
(20) Nishio, H.; Kosaka, A.; Hembury, G. A.; Matsushima, K.; Inoue, Y. J.
Chem. Soc., Perkin Trans. 2002, 582.
JA056585K
(21) Weber, G. AdV. Protein Chem. 1953, 8, 415.
9
2058 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006