Design and high-resolution structure of a β3-peptide bundle catalyst

Article abstract of DOI:10.1021/ja5013849

Despite the widespread exploration of α-peptides as catalysts, there are few examples of β-peptides that alter the course of a chemical transformation. Our previous work demonstrated that a special class of β³-peptides spontaneously self-assembles in water into discrete protein-like bundles possessing unique quaternary structures and exceptional thermodynamic stability. Here we describe a series of β³-peptide bundles capable of both substrate binding and chemical catalysis ester hydrolysis. A combination of kinetic and high-resolution structural analysis suggests an active site triad composed of residues from at least two strands of the octameric bundle structure.

Full text of DOI:10.1021/ja5013849

Communication
pubs.acs.org/JACS
Design and High-Resolution Structure of a β3‑Peptide Bundle
Catalyst
Pam S. P. Wang,† Jennifer B. Nguyen,† and Alanna Schepartz*,†,‡
‡
†Department of Chemistry and Department of Molecular, Cellular and Developmental Biology, Yale University, New Haven,
Connecticut 06520-8107, United States
S
* Supporting Information
ABSTRACT: Despite the widespread exploration of α-
peptides as catalysts, there are few examples of β-peptides
that alter the course of a chemical transformation. Our
previous work demonstrated that a special class of β3-
peptides spontaneously self-assembles in water into
discrete protein-like bundles possessing unique quaternary
structures and exceptional thermodynamic stability. Here
we describe a series of β3-peptide bundles capable of both
substrate binding and chemical catalysisester hydrolysis.
A combination of kinetic and high-resolution structural
analysis suggests an active site triad composed of residues
from at least two strands of the octameric bundle structure.
eptides embody two molecular properties that engender
Pchemical catalysis. The propensity of a polyamide backbone,
even a short one, to occupy a restricted conformational space
facilitates the judicious placement of potential catalytic,
recognition, or stabilizing groups, while the chirality of amino
acid monomers and the structures they form can impart intrinsic
Figure 1. (A) Reactions evaluated in the presence or absence of the β3-
selectivity. Despite the widespread exploration of α-peptides as
peptides shown in panels B−D. Not all sequences assemble into β3-
catalysts for numerous reactions,1 there are only two reported
peptide bundles (see text). (B−D) Helical net diagrams of β3-peptides
studied herein.
examples of β-peptides that alter the course of a chemical
transformation.2 β-Peptides are polymers of β-amino acids,
which differ from natural α-amino acids by the addition of a
interaction.5 Previous work has also shown that histidine side
single backbone methylene unit per residue; the additional
methylene unit imparts structural and metabolic stability.3
chains are used extensively within the active sites of natural
Our previous work demonstrated that a special class of β3-
esterases, either as nucleophiles or, more frequently, as general
acids/bases.6 There is also an extensive biomimetic chemistry
peptides self-assembles in water into discrete helical bundles
literature7 to support combining binding and catalytic groups to
possessing a protein-like tertiary fold and exceptionally high
thermodynamic stability.4 Here we report a series of β3-peptide
facilitate chemical reactions in aqueous solution. We sought to
bundles capable of both substrate binding and chemical
catalysisester hydrolysis. A combination of kinetic and high-
resolution structural analysis suggests an esterase active site
composed of three functional groups positioned on separate
strands of the octameric bundle structure.
Our design began with the structure of Zwit-EYYK, the most
thermally and kinetically stable β3-peptide bundle characterized
to date.4d The Zwit-EYYK bundle folds cooperatively (TM = 78
°C at 25 μM) and is >90% octameric at this concentration.4d As a
model reaction, we chose the hydrolysis of 8-acetoxypyrene-
1,3,6-trisulfonate (1), which releases the fluorescent product
pyranine (2) upon ester hydrolysis (Figure 1A). Previous work
has shown that arginine side chains in natural enzymes can
interact favorably with sulfonate groups; the binding of
coenzyme M to hydrogenases is one example of such an
test the hypothesis that a β3-peptide bundle endowed with
judiciously positioned arginine and histidine side chains would
catalyze the hydrolysis of 1.
To test this hypothesis, we modified the sequence of the Zwit-
EYYK monomer to electrostatically guide substrate 1 into the
proximity of a single histidine side chain while minimally
perturbing the bundle structure. Substrate 1 is planar, with three
sulfonate groups whose structural relationship mimics that
between side chains at positions i and i + 6 on a single 14-helix
face. We reasoned that substituting arginine for the ornithine at
positions 3 and 9 of Zwit-EYYK (U and Z in Figure 1B) would
facilitate electrostatic guidance8 to a histidine side chain at
Received: February 9, 2014
Published: May 6, 2014
© 2014 American Chemical Society
6810
dx.doi.org/10.1021/ja5013849 | J. Am. Chem. Soc. 2014, 136, 6810−6813

Journal of the American Chemical Society
Communication
position 1 (X in Figure 1B) without severely compromising
bundle stability. A similar logic has been previously applied to
design a cyclic peptide catalyst for an analogous ester substrate.9
Based on this design rationale, we synthesized three variants of
Zwit-EYYK carrying a single α-histidine (αH) at position 1 and
one or two β3-homoarginine (β3R) residues at positions 3 and 9
(Figure 1B).10 Preliminary data showed that all three of these
first-generation peptidesβEst-1, βEst-2, and βEst-3cata-
lyzed the hydrolysis of 1 mM substrate 1 at a catalyst loading of
10 mol% in a solution buffered at pH 6, increasing the
background reaction rate by a factor of 20−30. Zwit-EYYK, as
expected, was inactive, while free histidine at 10 mol% enhanced
the reaction rate by <5-fold (Figure S1).
2, and βEst-3 assembled into bundles at high concentration, they
were primarily monomeric at 25 μM, the concentration chosen
for steady-state kinetics (Figure S2). Compared to Zwit-EYYK,
which was >90% octameric at 25 μM, βEst-1, βEst-2, and βEst-3
were <30% assembled at this concentration. The Zwit-EYYK X-
ray structure revealed that the β3E at position 1 (X in Figure 1B)
is involved in an interhelical salt-bridge interaction. This side
chain is substituted by αH in βEst-1−βEst-3, suggesting that the
observed destabilization could be due to loss of this acidic side
chain or inclusion of an α-amino acid at this position, or a
combination of these effects.4d
We pursued three strategies to recover bundle stability. To
evaluate whether the decrease in stability resulted from the
presence of an α-amino acid (αH) at position 1, we synthesized a
variant of βEst-2 that contained β3H at this position (βEst-2-βH
in Figure 1B). To evaluate whether the decrease was due to loss
of β3E at position 1, we restored this residue and appended the
αH to either the N- or C-terminus as a 13th residue (βEst-2N
and βEst-2C in Figure 1C). Finally, to evaluate whether entropic
effects could be harnessed to improve bundle stability, we
synthesized a covalent dimer containing two βEst-2 monomers
joined with a tetra-β-homoglycine (βG) linker, an analogue of
the highly stable Z28 bundle reported previously (βEst-28 in
Figure 1D).11
We next performed steady-state measurements to characterize
the reaction kinetics in greater detail. Incubation of 25 μM β3-
peptide (βEst-1, βEst-2 or βEst-3) with 0.15−2.5 mM 1 revealed
that hydrolysis followed Michaelis−Menten kinetics (Figure
2A). Kinetic constants derived from these data suggested kcat
The kinetic constants in Table 1 reveal dimerization as the
most effective strategy to regain bundle structure and improve
Table 1. Kinetic Constants Characterizing the Hydrolysis of 1
by β3-Peptide Catalysts
kcat
KM
kcat/KM
kcat
/
bundle at
a
peptide
βEst-1
(min−1
)
(μM)
(M−1 min−1
)
kuncat
25 μM?
0.013
0.018
0.019
0.008
0.011
0.014
0.020
0.014
0.026
0.015
0.012
447
345
487
351
3467
147
4
29
54
413
588
612
255
356
460
649
458
820
476
386
no
βEst-2
no
βEst-3
39
no
βEst-2-βH
βEst-2N
βEst-2C
23
yes
yes
yes
yes
yes
yes
yes
no
Figure 2. (A,B,D) Plots of observed initial reaction rate (V) vs substrate
1 concentration in the absence or presence of the indicated β3-peptide.
Data were fit to the Michaelis−Menten equation; panel B also shows the
fit to the Haldane equation (cyan, βEst-28), which accounts for substrate
inhibition. Reactions were performed in a 10 mM Bis-Tris buffer (pH 6)
with [β3-peptide] = 25 μM. (C) Plots of kobs vs [β3-peptide] measured
under pre-steady-state conditions.
3
98
b
βEst-28
5102
1028
6446
73
b
βEst-28-1H
14
b
βEst-28-2R
βEst-2C-V
βEst-2C-A
4
204
1602
8
values of 0.013, 0.018, and 0.019 min−1 and KM values of 447,
345, and 487 μM for βEst-1, βEst-2, and βEst-3, respectively.
βEst-2, containing two β3R residues, displayed a more favorable
KM and the highest specificity (kcat/KM = 54 M−1 min−1). Since
KM reflects binding affinity, this observation supports a model in
which substrate binding is mediated by electrostatic interactions
between guanidinium groups on the peptide and sulfonate
groups on the substrate. Relative to the buffer reaction, βEst-1,
βEst-2, and βEst-3 enhanced the rate of ester hydrolysis (kcat/
a
b
kuncat = 3.144 × 10−5 min−1. Constants calculated using the Haldane
equation to account for substrate inhibition. To aid comparisons,
kinetic constants were calculated on the basis of moles of β3-peptide,
not moles of β3-peptide bundle.
catalytic activity. While βEst-2-βH was more structured than
βEst-2 (almost 80% bundle at 25 μM), its esterase activity was
compromised, with kcat/KM = 23 M−1 min−1 (Figure S3). In a
similar way, βEst-2N, βEst-2C, and βEst-28 all exhibited higher
degrees of association (>80% bundle at 25 μM) than βEst-2
(Figure S4). However, the catalytic activities of these peptides
varied drastically: βEst-2C was more efficient than βEst-2, βEst-
2N was virtually inactive, and βEst-28 displayed very rapid initial
rates but did not obey Michaelis−Menten kinetics (Figure 2B).
The dependence of catalytic activity on the relative positions of
αH and β3R residues implies that peptide−substrate interactions
are highly specific. The nearly 2-fold increase in the catalytic
efficiency of βEst-2C (kcat/KM = 98 M−1 min−1) over βEst-2
(kcat/KM = 54 M−1 min−1) is a result of its improved affinity for
the substrate (KM = 147 μM) and perhaps the enhanced helicity
k
uncat) by factors of 413, 588, and 612. These kinetic parameters
are comparable to those of a similarly sized dendritic peptide,
RM-G2, which catalyzes the hydrolysis of 1 with a rate
enhancement of 340 and kcat/KM = 120 M−1 min−1 at pH
5.5.7e In a similar way, a designed 4-helix α-peptide bundle,
MNKR, catalyzes p-nitrophenyl fumarate ester hydrolysis with
kcat/KM = 10.2 M−1 min−1 at pH 5.7f The only other 14-helical β-
peptide catalyst reported, whose structure is unknown, catalyzes
the retroaldol cleavage of a β-hydroxyketone with kcat/KM = 26
M−1 min−1 despite a rate acceleration of kcat/kuncat = 3000.2a
The favorable kinetic constants notwithstanding, subsequent
circular dichroism experiments revealed that while βEst-1, βEst-
6811
dx.doi.org/10.1021/ja5013849 | J. Am. Chem. Soc. 2014, 136, 6810−6813

Journal of the American Chemical Society
Communication
of βEst-2C at 25 μM. The nearly 100-fold increase in the catalytic
efficiency of βEst-28 is also the result of a greatly improved
substrate affinity (KM = 4 μM).
exists a direct relationship between bundle stoichiometry and β3-
peptide sequence; specifically, β3-peptides with β3L residues at
positions i, i+3, i+6, and i+9 assemble into octamers, those with
β3V or β3I at these positions assemble into tetramers, and those
with β3A at these positions are constitutively monomeric.15
Based on this relationship, we synthesized two stoichiometric
variants of βEst-2C, one containing an all-valine face and another
containing an all-alanine face (βEst-2C-V and βEst-2C-A in
Figure 1C, respectively). Steady-state kinetics measurements
revealed that the monomeric βEst-2C-A was the least effective
catalyst of this series (Figure 2D), while βEst-2C and βEst-2C-V
(80% and 73% bundle at 25 μM) exhibited similar levels of
activity. These results support the conclusions that bundle
formation contributes to catalysis and that octamers and
tetramers, but not monomers, can assemble a functional
esterolytic active site.
To understand the differences in activity between bundle-
forming and monomeric β3-peptide catalysts, we obtained a high-
resolution X-ray structure of βEst-2C. As predicted, βEst-2C self-
associates into a quaternary assembly whose backbone skeleton is
virtually superimposable with that of the Zwit-EYYK octamer
(RMSD = 0.171; SDM at 5.0 Å cutoff = 3.425; Q-score = 0.920)
(Figure 3A). The refined model of the βEst-2C bundle at 1.81 Å
We designed βEst-28 using a previously reported strategy that
recapitulates the characteristic β3-peptide octamer fold with 4
subunits instead of 8.11 Although the relative positions of the αH
and the β3R residues in βEst-28 mimic those in βEst-2, the kinetic
profile of βEst-28 was entirely unanticipated. Instead of initial
velocity (V) increasing as a function of substrate concentration, V
reached a maximum at [1] = 200 μM and then steadily
descended toward an asymptote. This behavior is diagnostic of
substrate inhibition, a well-known phenomenon that occurs in
∼20% of natural enzymes, often to avoid excessive production or
degradation of important metabolic intermediates.12 We used
the Haldane equation13a modified version of the Michaelis−
Menten equation that includes an additional equilibrium
constant, Kito fit the hydrolysis kinetics observed in the
presence of βEst-28. The Haldane model calculates kcat/KM
=
5102 M−1 min−1 for βEst-28, almost 2 orders of magnitude
greater than that for βEst-2. βEst-28 is 85% bundle at 25 μM
concentration, emphasizing the benefit of a catalyst possessing
higher order structure and multiple potential catalytic sites.
To provide additional support for the substrate inhibition
model, we conducted kinetic measurements under pre-steady-
state conditions. These measurements were performed using
excess βEst-28 (5- to 40-fold over substrate 1), which allowed us
to monitor a single substrate turnover. Because the substrate was
present in such small amounts relative to βEst-28, the possibility
of substrate inhibition was effectively excluded. The observed
rate constants, kobs, were extracted from fits of the data at each
catalyst concentration to single-exponential curves (Figure S5).
A plot of kobs against catalyst concentration (Figure 2C) was then
fit to a hyperbolic function to obtain the rate constant for the
chemical step, kchem (the horizontal asymptote), and the apparent
Kd (the peptide concentration corresponding to half of the
asymptote). The kinetic parameters obtained from pre-steady-
state studies of βEst-28 (kchem = 0.083 min−1; Kd,app = 14 μM)
agreed well with those obtained from steady-state measurements
(kcat = 0.020 min−1; KM = 4 μM), providing support for the
substrate inhibition model. kchem is expected to be equal to or
greater than kcat, since the latter is reflective of the rate-limiting
step of the reaction. On the other hand, Kd,app should closely
match KM, since both reflect the affinity of the catalyst for the
substrate.14
One explanation for the substrate inhibition observed with
βEst-28 is that, as a covalently linked dimer of βEst-2, it contains
4 β3R and 2 αH residues, increasing the likelihood of alternative,
nonproductive catalyst−substrate interactions. To investigate
this possibility, we synthesized two βEst-28 variants, one
containing a single αH per βEst-28 monomer and another
containing a single pair of β3R residues per βEst-28 monomer
(βEst-28-1H and βEst-28-2R in Figure 1D, respectively). βEst-
28-1H was 5-fold less active than βEst-28, with changes in both
kcat and KM. Surprisingly, βEst-28-2R is (slightly) more active
than βEst-28. Substrate inhibition was observed in both cases
(Figure S6), suggesting that further studies will be necessary to
completely understand the origins of this effect. Pre-steady-state
analyses reveal, nevertheless, that the additional αH and β3R
residues enhance catalyst efficiency in a single substrate turnover;
as assessed by the metrics kchem and Kd,app, neither of the two 28-
mer variants was as active as the parent βEst-28 (Figure 2C).
We next investigated the dependence of catalytic activity on
β3-peptide bundle stoichiometry. As previously reported, there
Figure 3. Structure of the βEst-2C bundle at 1.81 Å resolution. Shading
corresponds to helix orientation. (A) Ribbon diagram of a single
octamer showing the αH and β3R side chains (gray). (B−D)
Representative interhelical active sites at the (B) parallel, (C)
antiparallel, and (D) tetramer−tetramer interfaces. Distances between
histidine and arginine side chains are highlighted in red.
resolution (R/Rfree = 21.1/24.6%) consists of eight 14-helices
organized into two tetramers related by a two-fold rotation axis,
with each tetramer comprising four helices arranged in a parallel/
antiparallel/parallel array. Substituting arginine for ornithine side
chains retained the salt-bridge interactions, and interestingly,
some interhelical arginine residues are parallel-stacked, a
configuration that is often present within highly polar networks
in natural proteins.16 Additionally, the C-terminal histidines π-
stack with adjacent tyrosine side chains at the tetramer−tetramer
interface.
6812
dx.doi.org/10.1021/ja5013849 | J. Am. Chem. Soc. 2014, 136, 6810−6813

Journal of the American Chemical Society
Communication
But how many potential active sites does the βEst-2C bundle
contain? The dimensions of substrate 1 (7.4 Å long, 5.4 Å wide)
suggest that an active site on the βEst-2C bundle would be
characterized by one αH and two β3R residues located within
roughly 15−20 Å. Taken with the structure, this analysis suggests
that the βEst-2C bundle contains three fundamentally different
active sites. The first occurs at the parallel interhelical interface
and consists of a β3R9−αH13 side-chain pair from one helix and a
β3R9 side chain from an adjacent helix (Figure 3B). The second
potential active site, which occurs at an antiparallel interhelical
interface, contains the same β3R9−αH13 pair but includes β3R3
from the neighboring helix (Figure 3C). The third potential
active site is located at the tetramer−tetramer interface,
consisting of αH13 from one helix and a β3R3−β3R9 pair from
another (Figure 3D). Considering that each bundle comprises 4
parallel, 2 antiparallel, and 4 tetramer−tetramer helical contacts,
and there are 2 active sites per interhelical interface, there are
theoretically 20 intermolecular active sites per bundle. This
analysis could explain the observation that bundle assembly
enhances catalysis, even assuming low active-site occupancy.
Finally, we asked whether intrinsic chirality would endow β3-
peptide bundles with the ability to effect enantioselective
catalysis. Indeed, βEst-2C exhibited significant chiral discrim-
ination between the enantiomers of the 2-phenylpropionate ester
substrate (R)-3 and (S)-3, catalyzing the hydrolysis of (R)-3 4
times faster than that of (S)-3 at 10 mol% catalyst loading (Figure
S7). Although the selectivity of βEst-2C is modest in comparison
to that of natural enzymes, it compares favorably with other
synthetic esterases demonstrating activity on similar substrates.
One dendritic peptide, for example, favors the enantiomer (S)-3
with an enantiomeric ratio E = 2.8.7h This result, together with
the kinetic parameters of the peptides we evaluated, suggests that
β3-peptide bundles are no less desirable than α-peptides as
scaffolds for catalyst development and may have unique
advantages due to the combined attributes of structural
predictability, stability, and metabolic orthogonality.
ACKNOWLEDGMENTS
■
We are grateful to the W. M. Keck Foundation for support of this
work and to Professor Scott Miller, Professor Anna Marie Pyle,
and Dr. Clarissa Melo Czekster for helpful discussions. We are
especially thankful to Professor Richard Baxter for providing
laboratory space for peptide crystallization.
REFERENCES
■
(1) (a) Davie, E. A. C.; Mennen, S. M.; Xu, Y. J.; Miller, S. J. Chem. Rev.
2007, 107, 5759. (b) Jacobsen, E. N.; MacMillan, D. W. C. Proc. Natl.
Acad. Sci. U.S.A. 2010, 107, 20618. (c) Knowles, R. R.; Jacobsen, E. N.
Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 20678.
(2) (a) Muller, M. M.; Windsor, M. A.; Pomerantz, W. C.; Gellman, S.
H.; Hilvert, D. Angew. Chem., Int. Ed. 2009, 48, 922. (b) Coffey, P. E.;
Drauz, K. H.; Roberts, S. M.; Skidmore, J.; Smith, J. A. Chem. Commun.
2001, 2330.
(3) (a) Appella, D. H.; Barchi, J. J.; Durell, S. R.; Gellman, S. H. J. Am.
Chem. Soc. 1999, 121, 2309. (b) Hart, S. A.; Bahadoor, A. B. F.;
Matthews, E. E.; Qiu, X. Y. J.; Schepartz, A. J. Am. Chem. Soc. 2003, 125,
4022. (c) Kritzer, J. A.; Stephens, O. M.; Guarracino, D. A.; Reznik, S. K.;
Schepartz, A. Bioorg. Med. Chem. 2005, 13, 11. (d) Seebach, D.;
Gardiner, J. Acc. Chem. Res. 2008, 41, 1366.
(4) (a) Daniels, D. S.; Petersson, E. J.; Qiu, J. X.; Schepartz, A. J. Am.
Chem. Soc. 2007, 129, 1532. (b) Goodman, J. L.; Petersson, E. J.;
Daniels, D. S.; Qiu, J. X.; Schepartz, A. J. Am. Chem. Soc. 2007, 129,
14746. (c) Petersson, E. J.; Craig, C. J.; Daniels, D. S.; Qiu, J. X.;
Schepartz, A. J. Am. Chem. Soc. 2007, 129, 5344. (d) Craig, C. J.;
Goodman, J. L.; Schepartz, A. Chembiochem 2011, 12, 1035.
(5) Clark, D. D.; Boyd, J. M.; Ensign, S. A. Biochemistry 2004, 43, 6763.
(6) (a) Ghosh, D.; Sawicki, M.; Lala, P.; Erman, M.; Pangborn, W.;
Eyzaguirre, J.; Gutierrez, R.; Jornvall, H.; Thiel, D. J. J. Biol. Chem. 2001,
276, 11159. (b) Osterlund, T.; Contreras, J. A.; Holm, C. Febs Lett. 1997,
403, 259.
(7) (a) Breslow, R. Acc. Chem. Res. 1995, 28, 146. (b) Li, X. Y.; Liu, D.
R. J. Am. Chem. Soc. 2003, 125, 10188. (c) Cleij, M. C.; Drenth, W.;
Nolte, R. J. M. Recl. Trav. Chim. Pays-Bas 1992, 111, 459. (d) Cleij, M.
C.; Drenth, W.; Nolte, R. J. M. Recl. Trav. Chim. Pays-Bas 1993, 112, 1.
(e) Javor, S.; Delort, E.; Darbre, T.; Reymond, J. L. J. Am. Chem. Soc.
2007, 129, 13238. (f) Broo, K. S.; Nilsson, H.; Nilsson, J.; Flodberg, A.;
Baltzer, L. J. Am. Chem. Soc. 1998, 120, 4063. (g) Broo, K. S.; Nilsson,
H.; Nilsson, J.; Baltzer, L. J. Am. Chem. Soc. 1998, 120, 10287. (h) Douat-
Casassus, C.; Darbre, T.; Reymond, J. L. J. Am. Chem. Soc. 2004, 126,
7817. (i) Esposito, A.; Delort, E.; Lagnoux, D.; Djojo, F.; Reymond, J. L.
Angew. Chem., Int. Ed. 2003, 42, 1381. (j) Kofoed, J.; Darbre, T.;
Reymond, J. L. Org. Biomol. Chem. 2006, 4, 3268. (k) Delort, E.; Darbre,
T.; Reymond, J. L. J. Am. Chem. Soc. 2004, 126, 15642. (l) Bolon, D. N.;
Mayo, S. L. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 14274. (m) Wei, Y. N.;
Hecht, M. H. Protein Eng. Des. Sel. 2004, 17, 67. (n) Baumeister, B.;
Sakai, N.; Matile, S. Org. Lett. 2001, 3, 4229.
In summary, here we describe a structurally characterized β3-
peptide bundle possessing measurable catalytic function. Unlike
two previously reported β-peptide catalysts,2 the molecules
described here self-assemble into discrete, unique, themostable
quaternary structures and are capable of both substrate
recognition and chemical catalysis. The dependence of catalytic
activity on the geometric arrangement of histidine and arginine
residues, as well as bundle assembly, points to the existence of
substrate-specific active sites that could be optimized using
structure-guided design.
(8) Getzoff, E. D.; Cabelli, D. E.; Fisher, C. L.; Parge, H. E.; Viezzoli, M.
S.; Banci, L.; Hallewell, R. A. Nature 1992, 358, 347.
(9) Vial, L.; Dumy, P. J. Am. Chem. Soc. 2007, 129, 4884.
(10) For this initial series, we avoided using β3-homohistidine due to
challenges associated with its synthesis.
(11) Petersson, E. J.; Schepartz, A. J. Am. Chem. Soc. 2008, 130, 821.
(12) Reed, M. C.; Lieb, A.; Nijhout, H. F. Bioessays 2010, 32, 422.
(13) Haldane, J. B. S. Enzymes; Longmans, Green: London and New
York, 1930.
(14) Uter, N. T.; Perona, J. J. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
14396.
(15) (a) Goodman, J. L.; Molski, M. A.; Qiu, J.; Schepartz, A.
Chembiochem 2008, 9, 1576. (b) Wang, P. S. P.; Craig, C. J.; Schepartz,
A. Tetrahedron 2012, 68, 4342.
ASSOCIATED CONTENT
■
S
* Supporting Information
Detailed descriptions of peptide synthesis and characterization,
kinetics, CD, and structure determination. Coordinates of the
βEst-2C bundle have been deposited in the Cambridge
Crystallographic Data Centre as entry CCDC 1000723. These
data can be obtained free of charge at www.ccdc.cam.ac.uk/data_
request/cif. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
(16) Lee, D.; Lee, J.; Seok, C. Phys. Chem. Chem. Phys. 2013, 15, 5844.
Corresponding Author
alanna.schepartz@yale.edu
Notes
The authors declare no competing financial interest.
6813
dx.doi.org/10.1021/ja5013849 | J. Am. Chem. Soc. 2014, 136, 6810−6813