Toward proteomimetics: Terphenyl derivatives as structural and functional mimics of extended regions of an α-helix [18]

Full text of DOI:10.1021/ja0025548

5382
J. Am. Chem. Soc. 2001, 123, 5382-5383
Toward Proteomimetics: Terphenyl Derivatives as
Structural and Functional Mimics of Extended
Regions of an r-Helix
Brendan P. Orner, Justin T. Ernst, and Andrew D. Hamilton*
Department of Chemistry, Yale UniVersity
P.O. Box 208107, New HaVen, Connecticut 06510-8107
ReceiVed July 13, 2000
The design of synthetic structures that mimic large and
noncontiguous regions of a protein surface remains an elusive
goal.1 There has been considerable success in the field of small
peptidomimetics that reproduce features of short peptides in
extended2 or â-turn conformations.3 However, much less progress
has been made in the search for proteomimetics or nonpeptide
structures that mimic larger areas of the protein surface4 such as
an R-helix.5 This is remarkable given the ubiquitous role of
R-helical regions in mediating protein-protein interactions.6 The
difficulty clearly lies in the large and elongated surface area that
is presented by 2-4 turns of an R-helix. One strategy involves
the covalent or noncovalent stabilization of a 16-20-mer peptide
in a helical conformation either through side chain contacts,7 end
capping templation,8 specific folding9 or use of â-peptides.10 As
part of our interest in helix surface recognition,11,12 we sought an
entirely nonpeptidic scaffold that could be synthesized in a
modular fashion and project side chain functionality with similar
distance and angular relationships to those found in R-helices.
We herein report a new family of proteomimetics, based on a
functionalized terphenyl scaffold, that are structural mimics of
two turns of the myosin light chain kinase R-helix and show
functional analogy in binding with high affinity to calmodulin.
Figure 1. (A) Schematic representation of an R-helical 12-mer peptide
with i, i + 3, and i + 7 substituents, side view; (B) top view; (C) 3,2′,2"-
trisubstituted terphenyl, top view; (D) side view; (E) X-ray crystal
structure of 1.
terphenyl derivative.13 This is an attractive template for proteo-
mimetic design due to the simplicity of the structure and the
potential for an iterative synthesis. The alternating arrangement
of i, i + 3, and i + 7 groups through two turns in the helix
compares well with the 3,2′,2′′-substituents when the terphenyl
is in a staggered conformation with dihedral angles of 68° and
36° between the phenyl rings.14 In this easily accessible confor-
mation, the three subsituents project from the terphenyl core with
similar angular relationships and 4-25% shorter distances than
between the i, i + 3, and i + 7 â-carbons in an R-helix (Figure
1C and D).15
A modular synthesis of the 3,2′,2′′-tris-substituted-terphenyl
derivatives was developed on the basis of sequential Negishi
coupling reactions. The terminal 3-substituted phenyl triflate was
linked to the central 4-iodo-3-substituted phenyl silyl ether by
zinc transmetalation of the iodide and palladium-mediated
coupling. Deprotection and triflation of the resulting biphenyl was
followed by reaction with a second 4-iodo-3-substituted phenyl
silyl ether to give the terphenyl silyl ether. The final steps involved
deprotection and alkylation of the hydroxyl group with solubility
modulating groups such as acetate.
An X-ray crystal structure of 3,2′,2′′-trimethyl-4-nitro-4′′-
hydroxy-terphenyl derivative 1 (Figure 1E) showed the molecule
in a staggered conformation (dihedral angles, 59.1° and 120.7°)
with rings A and C projecting their Me substituents on the same
face of ring B.16 The distances between the Me groups are 5.10
(3,2′), 6.28 (2′,2′′), and 8.83 Å (3,2′′), in reasonable cor-
respondence to the i, i + 3, and i + 7 â-carbons in an R-helical
peptide. Other low-energy conformers will be present in solution;
however, the solid-state structure and the low rotational barrier
of related biphenyls17 point to the desired terphenyl conformation
(Figure 1C and D) being accessible, particularly in the presence
of a complementary recognition site.
To test the idea of R-helix mimicry by terphenyl derivatives,
we focused on the interaction between calmodulin (CaM) and an
R-helical domain of smooth muscle myosin light-chain kinase
(smMLCK).18 CaM also represents an interesting target in our
continuing search19 for molecules that influence cell cycle events.20
In R-helix-protein complexes critical interactions are often
found along one face of the helix, involving side chains from the
i, i + 3, and i + 7 residues.6 The relative positions of these groups
in an all-Ala R-helix are shown in Figure 1 A-D and compared
to the projection of substituents in a tris-functionalized 3,2′,2′′-
(1) Andrews, M. J. I.; Tabor, A. B. Tetrahedron 1999, 55, 11711-11743.
(2) Smith, A. B.; Knight, S. D.; Sprengeler, P. A.; Hirschmann, R. Bioorg.
Med. Chem. 1996, 4, 1021-1034.
(3) Sasaki, T.; Lieberman, M. Protein Mimetics. In ComprehensiVe
Supramolecular Chemistry; Murakami, Y., Ed.; Pergamon Press: Oxford,
1996, Vol. 4, pp193-242.
(4) For a different approach to this problem see; Park, H. S.; Lin, Q.;
Hamilton, A. D. J. Am. Chem. Soc. 1999, 121, 8-13.
(5) For recent exceptions see; O’Donnell, M.; Garippa, R. J.; O’Neill, N.
C.; Bolin, D. R.; Cotrell, J. M. J. Biol. Chem. 1991, 266, 6389-6392. Nolan,
W. P.; Ratcliffe, G. S.; Rees, D. C. Tetrahedron Lett. 1992, 33, 6879-6882.
Xuereb, H.; Maletic, M.; Gildersleeve, J.; Pelczer, I.; Kahne, D. J. Am. Chem.
Soc. 2000, 122, 1883-1890.
(6) For a recent review see; Fairlie, D. P.; West, M. L.; Wong, A. K. Curr.
Med. Chem. 1998, 5, 29-62.
(7) Jackson, D. Y.; King, D. S.; Chmielewski, J.; Singh, S.; Schultz, P.
G.J. Am. Chem. Soc. 1991, 113, 9391-9392. Ghadiri, M. R.; Choi, C. J. Am.
Chem. Soc. 1990, 112, 1630-1632. O¨ spay, G.; Taylor, J. W. J. Am. Chem.
Soc. 1992, 114, 6966-6973. Albert, J. S.; Hamilton, A. D. Biochemistry 1995,
34, 984-990.
(8) Kemp, D. S.; Allen, T. J.; Oslick, S. L.; Boyd, J. G. J. Am. Chem. Soc.
1996, 118, 4240. Austin, R. E.; Maplestone, R. A.; Sefler, A. M.; Lui, K.;
Hruzewicz, W. N.; Lui, C. W.; Cho, H. S.; Wemmer, D. E.; Bartlett, P. A. J.
Am. Chem. Soc. 1997, 119, 6461.
(13) For an example of 2,3′-disubstituted biphenyls as constrained turn
mimics see; Nesloney, C. L.; Kelly, J. W. J. Am. Chem. Soc. 1996, 118, 5836-
5845.
(14) Using the MacroModel program, Still, W. C. Columbia University.
(15) Stereoisomerism will occur if one of the rings rotates by 180°.
(16) For covalent control of terphenyl conformational isomerism see:
Kiupel, B.; Niederalt, C.; Nieger, M.; Grimme, S.; Vo¨gtle, F. Angew. Chem.,
Int. Ed. 1998, 37, 3031.
(17) Mislow, K.; Glass, M. A.; O’Brien, R. E.; Rutkin, P.; Steinberg, D.
H.; Weiss, J.; Djerassi, C. J. Am. Chem. Soc. 1961, 84, 1455. Bott, G.; Field,
L. D.; Sternhell, S. J. Am. Chem. Soc. 1980, 102, 5618.
(18) Meador, W. E.; Means, A. R.; Quicho, F. A. Science 1992, 257, 1251-
1255.
(19) Sebti, S. M.; Hamilton, A. D. Methods Enzymol. 2000, 325, 381-
388.
(9) Zondlo, N. J.; Schepartz, A. S., J. Am. Chem. Soc. 1999, 121, 6938-
6939. Struthers, M. D.; Cheng, R. P.; Imperiali, B., J. Am. Chem. Soc. 1996,
118, 3073-3081.
(10) Gellman, S. H. Acc. Chem. Res. 1998, 31, 173. Hinterman, T.;
Gademann, K.; Jaun, B.; Seebach, D. HelV. Chim. Acta, 1998, 81, 983.
Hamuro, Y.; Geib, S. J.; Hamilton, A. D., J. Am. Chem. Soc. 1997, 119,
10587-10593.
(11) Peczuh, M. W.; Hamilton, A. D.; Sanchez-Quesada, J.; deMendoza,
J.; Haack, T.; Giralt, E. J. Am. Chem. Soc. 1997, 119, 9327-9328.
(12) Albert, J. S.; Peczuh, M. W.; Hamilton, A. D. Bioorg. Med. Chem.
1997, 5, 1455-1467.
10.1021/ja0025548 CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/11/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 22, 2001 5383
elution with 5 M biotin. The control experiment with an
underivatized avidin resin resulted in no observable retention of
2. In addition, polyacrylamide gel permeation chromatography
(6 kD cutoff) showed that 2 alone (70 µM in buffer) was retained
on the column but passed through when mixed with 1 equiv of
CaM, due presumably to the formation of a CaM:2 complex. To
amplify the binding event between CaM and 2, an enzymatic assay
was also employed. CaM activates the enzyme 3′-5′-cyclic
nucleotide phosphodiesterase (PDE) through an interaction that
is thought to involve the same hydrophobic region that binds to
smMLCK.23 Addition of 2 to a solution of CaM and PDE caused
a dose-dependent reduction in the ability of CaM to activate the
enzyme for the hydrolysis of substrate (Figure 3).24 The inhibitory
potency of 2 (IC50 ) 800nM) in this PDE assay is only 10-fold
less than the 20-mer R-helical peptide RS20 (IC50 ) 80nM) and
much stronger than the trimethylterphenyl 4 (IC50 > 20µM)25 that
Figure 2. Design of the smMLCK mimetic.
lacks the trio of binding substituents or monophenyl 5 (IC50
)
150µM). This result suggests that 2 acts as a functional mimic of
a natural CaM substrate by antagonizing the binding interaction
between CaM and PDE.
Evidence that 2 also acts as a structural mimic of smMLCK,
and binds to the same region in the C-terminal domain of CaM,
came from competition experiments using the helical peptide
C20W (Figure 2). This is a helical region of the plasma membrane
calcium pump protein that binds exclusively to the C-terminal
domain of CaM, in the same area as smMLCK.26 Addition of
dansylated C20W to CaM results in an increase in the fluorescence
emission intensity due to complex formation. This effect is
reversed on titration of an excess of 2 into a solution of the
complex CaM:C20W-dansyl, clearly suggesting that 2 and C20W
are competing for the same binding site on CaM.27 If this model
is correct, we should be able to enhance the binding of 2 by
optimizing the fit between the terphenyl substituents and the
binding pockets on CaM. To test this we prepared, using
modifications of the synthetic route above, 1- and 2-naphthyl
derivatives 3a and 3b as improved analogues of the Trp800 indole
side chain in smMLCK. Figure 3 shows that both 3a and 3b are
very potent inhibitors of CaM activation of PDE enzyme activity
with IC50 values of 9 nM and 20 nM, respectively. For 3a this
potency corresponds to an 8-fold improvement over the helical
peptide RS20 from smMLCK and renders it among the most
active CaM antagonists known. However, the full extent of helix
mimicry by 2, 3a, and 3b in terms of their precise conformations
and modes of binding to CaM awaits high-resolution structures
of these complexes.
Figure 3. Antagonism of CaM by monitoring PDE hydrolysis of
mant-cGMP. [ ) 5, 9 ) 2, b ) RS-20, 1 ) 3b 2 ) 3a. 13.9 nM
CaM, 50 mu/mL PDE, 8 µΜ mant-cGMP, 10 mM Mops, 0.5 mM MgCl2,
90 mM KCl, 0.73 mM CaCl2, 0.2% DMSO, pH 7.0.
Moreover, earlier work from DeGrado has shown that CaM
provides an effective recognition surface for a variety of peptides
in an R-helical conformation.21 The sequence of a 20-mer
fragment (RS20) in smMLCK is shown in Figure 2, and
mutational studies have established a key role for three i, i + 3,
and i + 7 residues (Trp800, Thr803, and Val807) in binding to
the C-terminal domain of CaM in a complex that also involves
the collapsed N-terminal region.22 The hydrophobic side chains
of this key trio of residues can be mimicked by the corresponding
3,2′,2′′-terphenyl 2 which, in a staggered conformation, should
project them with a similar rise and angle to that in smMLCK.
For synthetic simplicity we changed the indole of Trp800 to
phenyl and removed the hydroxyl of Thr803. Using Negishi
couplings of differently substituted phenyltriflates, we prepared
terphenyl 2 as a mimic of the calmodulin binding face of
smMLCK. The free hydroxyl at the end of the iterative terphenyl
synthesis was alkylated with benzyl bromoacetate, the ester was
hydrolyzed, and the resulting carboxylic acid was converted to
the ammonium salt of 2, which proved to be surprisingly soluble
in buffer with <1% DMSO.
Acknowledgment. This paper is dedicated to the memory of Professor
John Osborn, scholar, wit and friend to many. We thank the National
Institutes of Health for partial support of this work.
Supporting Information Available: Details of the characterization
of the binding interaction of calmodulin with the terphenyl derivatives,
the phosphodiesterase activity assay, the X-ray crystallographic analysis
of 1, and the syntheses of 2, 3a and 3b (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
JA0025548
(23) Yuan, T.; Walsh, M. P.; Sutherland, C.; Fabian, H.; Vogel, H. J.
Biochemistry 1999, 38, 1446-1455.
(24) Johnson, J. D.; Walters, J. D.; Mills, J. S., Anal. Biochem. 1987, 162,
291-295.
Passage of 2 through an avidin-based affinity resin derivatized
with biotinylated CaM led to retention of the terphenyl on the
column. Discrete complex formation to CaM was indicated by
the release of 2 from the column, as monitored by HPLC, on
(25) A precise IC50 value could not be achieved due to poor solubility above
20 µM.
(26) Elshorst, B.; Hennig, M.; Forsterling, H.; Diener, A.; Maurer, M.;
Schulte, P.; Schwalbe, H.; Griesinger, C.; Krebs, J.; Schmid, H.; Vorherr, T.;
Carafoli, E. Biochemistry 1999, 38, 12320-12332 In the case of smMLCK
there is a major conformational change in CaM and the N-terminal regions
clamps down on the opposite face of the helix.
(20) Taule´s, M.; Rius, E.; Talaya, D.; Lo´pez-Girona, A.; Bachs, O.; Agell,
N. J. Biol. Chem. 1998, 273, 33279-33286.
(21) O’Neil, K. T.; DeGrado, W. F. TIBS 1990, 15, 59-64.
(22) Meador, W. E.; Means, A. R.; Quiocho, F. A. Science 1992, 257,
1251-1255.
(27) Assuming a 1:1 stoichiometry for the CaM:2 complex, a KD value of
1.6 µM can be calculated (compared to 39 nM for RS20 under the same
conditions).