Structural studies of β-turn-containing peptide catalysts for atroposelective quinazolinone bromination

Article abstract of DOI:10.1039/c6cc01428c

We describe herein a crystallographic and NMR study of the secondary structural attributes of a β-turn-containing tetra-peptide, Boc-Dmaa-d-Pro-Acpc-Leu-NMe₂, which was recently reported as a highly effective catalyst in the atroposelective bromination of 3-arylquinazolin-4(3H)-ones. Inquiries pertaining to the functional consequences of residue substitutions led to the discovery of a more selective catalyst, Boc-Dmaa-d-Pro-Acpc-Leu-OMe, the structure of which was also explored. This new lead catalyst was found to exhibit a type I′ β-turn secondary structure both in the solid state and in solution, a structure that was shown to be an accessible conformation of the previously reported catalyst, as well.

Full text of DOI:10.1039/c6cc01428c

View Article Online
View Journal
ChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. J. Metrano, N.
C. Abascal, B. Mercado, E. Paulson and S. J. Miller, Chem. Commun., 2016, DOI: 10.1039/C6CC01428C.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/chemcomm

Page 1 of 4
ChemComm
View Article Online
DOI: 10.1039/C6CC01428C
Journal Name
COMMUNICATION
Structural Studies of β-Turn-Containing Peptide Catalysts for
Atroposelective Quinazolinone Bromination†
A. J. Metrano,‡ N. C. Abascal,‡ B. Q. Mercado, E. K. Paulson, and S. J. Miller*
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
www.rsc.org/
We describe herein a crystallographic and NMR study of the
secondary structural attributes of a β-turn-containing tetra-
peptide, Boc-Dmaa-D-Pro-Acpc-Leu-NMe2, which was recently
reported as a highly effective catalyst in the atroposelective
bromination of 3-arylquinazolin-4(3H)-ones. Inquiries pertaining
to the functional consequences of residue substitutions led to the
discovery of a more selective catalyst, Boc-Dmaa-D-Pro-Acpc-Leu-
OMe, the structure of which was also explored. This new lead
catalyst was found to exhibit a type I’ β-turn secondary structure
both in the solid state and in solution, a structure that was shown
to be an accessible conformation of the previously reported
catalyst, as well.
Scheme 1: Atroposelective Bromination of 3-Arylquinazolin-4(3H)-ones (1)
D
-Pro-Acpc
O
i+1
Br
O
N
i+2
3 (10 mol%)
NBS (3.0 equiv)
O
N
Br
N
N
HN
OMe
(±)
N
OH
Me2N
O
PhMe/CHCl3 (9:1 v/v)
acetone (5 vol%)
0 ºC, 3 h
Br
Me
O
i
HN
i-Pr
i+3
Me
NH
then MeOH/TMSCHN2
1
2
86% yield, 97:3 er
Boc
O
3
• slow addition of NBS
over 2.5 h
NMe2
• 13 additional examples
• up to 93% yield, 99:1 er
transformations. One motif in particular has been very fruitful
in terms of catalyst development. The Pro-Xaa sequence,
where Pro is D- or L-proline and Xaa is an α,α-disubstituted,
achiral amino acid residue (often Aib§), reliably nucleates a
type II (or II’) β-turn secondary structure when located in the
loop region (i.e. the i+1 and i+2 positions, respectively) of a
tetrapeptide.7 We have reported a number of catalytic,
asymmetric methods that take advantage of the Pro-Xaa
structural motif.8 The atroposelective bromination of 3-
arylquinazolin-4(3H)-ones (1) is a recent example, in which we
In the field of asymmetric catalysis, certain catalyst
platforms have emerged as “privileged” chiral scaffolds due to
their ability to achieve high levels of enantioinduction across a
wide array of synthetically useful transformations.1 One
feature common to many of these scaffolds is a minimum of
rotatable bonds, which presumably allows for the formation of
rigid and well-defined transition states. Peptide-based catalysis
has also proven to be an effective strategy for the synthesis of
optically enriched compounds.2 Detailed analyses of structural
trends in proteins,3 as well as physical organic studies of oligo-
peptides in organic solution,4 have enabled the design of
specific peptide secondary structures with high precision.
Nonetheless, the many bond rotations available to short
peptides provides the possibility that these peptidic catalysts
may exist as an equilibrial mixture of conformers under the
reaction conditions, both in low energy ground states and in
competing transition states.4,5 Deciphering which of these
conformers, if any, are kinetically-competent has been an on-
going challenge in the field.6
identified the D
-Pro-Acpc§-containing peptide 3 as an effective
catalyst (Scheme 1).9 The catalytic Dmaa§ residue was
embedded at the i-position of the turn for functional group
cooperativity with the backbone amides. Employing optimized
conditions (slow addition of NBS§ over 2.5 hours), 3 provided
good yields and excellent enantioselectivities over a broad
substrate scope.9
Catalyst 3 emerged from a systematic optimization of the
peptide sequence, in which Acpc-containing peptides were
found to be superior to other i+2 variants, providing
tribrominated quinazolinone 2 in 35–40% ee§ higher than the
corresponding Aib-containing peptides.9 The decision to
examine Acpc as an i+2-residue in our catalyst optimization
studies was motivated by the search for different Aib-variants
for this crucial turn position in pursuit of higher ee values.
While we anticipated that the geometric constraints imposed
by the cyclopropyl ring would influence the secondary
structure in some way, we were unsure of what those effects
might be. With excellent catalyst performance in hand, we
turned our attention to structural studies of this catalyst
family, aspiring to better understand, and perhaps eventually
Our laboratory has often capitalized on the secondary
structural preferences of certain peptide sequences in
optimizing lead catalysts for various asymmetric
Department of Chemistry, Yale University, New Haven, CT 06520-8107, USA
E-mail: scott.miller@yale.edu
† Electronic Supplementary Information (ESI) available: synthetic procedures,
characterization, crystallographic information, and NMR/computational data are
provided. CCDC 1453124 (3(c)) and 1453125 (4l). See DOI: 10.1039/x0xx00000x.
‡ A.J.M. and N.C.A. contributed equally.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

ChemComm
Page 2 of 4
COMMUNICATION
Journal Name
turn tendency of 3(a) manifests in the comparatively large
difference (25.7º) between the ψ(i+1) dihedrals of 3(a) and
View Article Online
DOI: 10.1039/C6CC01428C
(a)
ψ(i+1)
3(b). Another prominent difference between 3(a) and 3(b) is
the degree of backbone bending. We were able to quantify
this observation by measuring the angle between two
intersecting planes, one defined by the α-carbons of residues i,
φ(i+2)
ψ(i+2)
φ(i+1)
3.029(6) Å
2.937(6) Å
2.952(6) Å
2.906(6) Å
2.968(5) Å
i+1, i+2, and i+3, and the other by the α-carbons of i–1 (the 3º-
carbon of the Boc§-group), i, i+3, and i+4 (the trans-Me of the
NMe2 end cap).† The backbone of conformer 3(a) was found to
3(a)
3(b)
type II' β-hairpin
type II' β-hairpin
bend 66
3(b) is more extended and bends only 41
structure of a related -Pro-Aib-containing type II’ β-turn,
previously reported,12 also exhibits this bent structure, but to a
slightly lesser degree (63 bend).
º
, making for a more compact structure, while that of
φ(i+1) = 63.6(6)º, ψ(i+1) = –113.0(6)º
φ(i+2) = –77.2(7)º, ψ(i+2) = –3.7(8)º
φ(i+1) = 62.9(7)º, ψ(i+1) = –138.7(5)º
φ(i+2) = –78.1(7)º, ψ(i+2) = –5.3(8)º
º
. The X-ray crystal
(b)
(c)
D
idealized type I' β-turn
φ(i+1) = 60º, ψ(i+1) = 30º
φ(i+2) = 90º, ψ(i+2) = 0º
º
2.871(2) Å
3.022(2) Å
Conformer 3(c) was unanticipated based on our previous
studies of Pro-Xaa turns in peptide-based catalysts (Figure 1c).
The φ,ψ(i+1) and φ,ψ(i+2) dihedrals are consistent with a type
I’ β-turn, with some clear distortions from ideality. The ψ(i+1)
• related by amide plane-
flip in loop region
idealized type II' β-turn
φ(i+1) = 60º, ψ(i+1) = –120º
φ(i+2) = –80º, ψ(i+2) = 0º
3(c)
type I' β-turn
dihedral is 14.8
while ψ(i+2) is 14.9
φ(i+2) dihedral of 3(c) is 19.1
dihedral deviations perhaps may be explained by the existence
of a strong intramolecular H-bond between NHAcpc and OBoc. It
seems plausible that a favourable H-bonding interaction could
provide the driving force for modest distortion of the
º
closer to syn-coplanar than in the ideal case,
larger than the ideal. Moreover, the 79.1
smaller than expected. These
φ(i+1) = 64.3(2)º, ψ(i+1) = 15.2(2)º
φ(i+2) = 71.9(2)º, ψ(i+2) = 14.9(3)º
º
º
º
Figure 1: Structural diversity in the X-ray crystal structures of 3. (a)
Conformers 3(a) and 3(b) are both consistent with type II’ β-hairpin
geometries.9 (b) Conformer 3(c) is consistent with a distorted type I’ β-
turn. (c) Idealized type I’ and II’ β-turns as classified by the φ and ψ
dihedral angles of the loop residues.3,10
predict, the effects of residue substitutions during catalyst backbone conformation. The ten-membered ring β-turn H-
optimization. bond is also present in 3(c), though it is slightly longer than in
In our previous report,9 we were able to grow a single 3(a) and 3(b). The two intramolecular H-bonds observed in this
crystal of 3 using slow vapour diffusion of pentane into a conformer provide a sequential double β-turn geometry
nearly saturated solution of 3 in ethyl acetate. X-Ray reminiscent of a helix. Though unanticipated, this is not
diffraction provided a unit cell containing two distinct surprising given the well-documented helical tendencies of
conformers, 3(a) and 3(b) (Figure 1a).9 In an attempt to assess Acpc and its derivatives.13 Nonetheless, our studies reinforce
the reproducibility of the crystallization, we re-grew crystals of the notion that the i+2 Acpc-residue may not be an innocent
3 under nearly identical conditions, and we fortuitously Aib-surrogate, as it appears to induce conformational
observed a polymorphic unit cell containing yet another heterogeneity in the ground-state catalyst structures.
conformer (3(c), Figure 1b). The observation of three distinct
Mechanistic questions arising from these intriguing
solid-state conformations of 3 was striking and unexpected. As structural observations stimulated an expanded study of
a harbinger of the conformational diversity of peptide-based
catalysts, we were eager to expand our inquiry into their
structures.
i+1
O
Br
i+2
O
N
R'
NBS (3.0 equiv)
3 (10 mol%)
O
N
Br
N
R"
N
HN
OMe
(±)
N
OH
Me2N
O
PhMe/CHCl3 (9:1 v/v)
23 ºC, 1 h
then MeOH/TMSCHN2
Br
Me
O
i
HN
R
Me
NH
The X-ray crystal structures of 3 revealed two different turn
motifs in the solid state. β-Turn types are typically classified by
the φ and ψ dihedral angles of the loop residues, in this case
φ,ψ(i+1) and φ,ψ(i+2).3 Idealized dihedrals for types I’ and II’
are shown in Figure 1c. By this analysis, conformers 3(a) and
3(b) are consistent with type II’ β-turns, while 3(c) most closely
resembles a type I’ β-turn (Figure 1a,b).
Both 3(a) and 3(b) were found to exhibit strong
intramolecular H-bonds between NHLeu and ODmaa and
between NHDmaa and OLeu, consistent with those expected for
β-hairpin structures (Figure 1a).4,7 An additional seven-
membered ring, γ-turn-like H-bond was observed between
NHAcpc and ODmaa in 3(a). The bifurcated nature of the H-bonds
to ODmaa suggests that, even in the solid state, there is a
conformational equilibrium between β- and γ-turn geometries,
and that 3(a) is sampling both to some extent (perhaps
favouring the β-turn on the basis of bond length).6,11 The γ-
i+3
1
2
Boc
O
• NBS added all at once
X
4
Peptide Sequences
4a Boc-Dmaa-DD--PPrroo--AAiibb--PPhhee--NOMMee
4b BBoocc--DDmmaaaa--DD--PPrroo--AAibib--PPhhee--ONMHeMe
4c Boc-Dmaa-DD--PPrroo--AAiibb--LPehue--NNMMee2
4d BBoocc--DDmmaaaa--DD--PPrroo--AAibib--LPehue-O-PMhe2
4e BBoocc--DDmmaaaa--DL-Pro-Aib-LPehue--ODM-Pehe2
4f BBoocc--DDmmaaaa--DD-Pro-Aib-DVa-Pl-hNeM-Pehe-
4g BBoocc--DDmmaaaa--DD--PPrroo--ACible--VPahl-eO-NMMee2
4h BBoocc--DDmmaaaa--DD-Pro-Acpc-Phe-NMe2
2
2
2
2
4i BBooc-cD-Dmmaaa-D-D-P-Proro-P-Ahcep-cP-hPeh-eN-MOeM2e
4j BBoocc--DDmmaaaa--DD-Pro-Aiccp-Pc-hGel-yN-NMMe2e2
4k Boc-Dmaa-DD--PPrroo--AAccppcc--GPlhye-O-OMMee
3
BBooc-cD-Dmmaaa-D-D-P-Proro-A-Accppcc-P-Lheeu--(NRM)-e2
4l BBoocc-D-Dmmaaaa-D-D--PPrroo-A-APchpec(5-LNeMue-O2Me
4m BBoocc-D-Dmmaaaa-L-D-P-Pror-oA-Acpccp-cL-eDu--POhMe2e
4n BBoocc--DDmmaaaa--DD-Pro-Acpc-GValyl-NOMMee2
4o BBoocc--DDmmaaaa--DD--PPrroo--AAccppcc--VGally-O-NMMee2
4p BBoocc--DDmmaaaa--DD--PPrroo--AAic-pPch-Lee-NuMe2
2
4q BBoocc--DDmmaaaa--DD-Pro-Acicp-cP-hVea-lO-NMMee2
4r BBoocc--DDmmaaaa--DD-Pro-Aiccp-Lc-eNul-eN-NMMe2e2
4s Boc-Dmaa-DD--PPrroo--AAicp-Lce-Nu-pOgM-Mee2
4t Boc-Dmaa-L-Pro-Aic-Leu-OMe
4u Boc-Dmaa-D-Pro-Phe-Phe-NMe2
4v Boc-Dmaa-D-Pro-Phe-Phe-OMe
4wmBaoc-Dmaa-D-Pro-Phe-Leu-NMe2
4xD-BPoroc-ADcmpaca-C-Dh-gP-rNoM-Pehe-Leu-OMe
2
Figure 2: Peptide structure-function studies led to the discovery of an
improved catalyst (4l). Peptides were examined on a 0.05 mmol scale
using the above conditions. Some data from ref. 9 are reproduced for
clarity.
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 4
ChemComm
Journal Name
COMMUNICATION
peptide-based catalysts (4) for the bromination of 1 (Figure 2).
In particular, we sought to better understand the effects of
seemingly subtle changes to the peptide sequence on
enantioselectivity. A number of interesting trends emerged
from this data set. In our original report,9 optimization of
catalyst 3 led to the observation that alkyl-substituted
residues, such as Leu and Val, were preferred over Phe-type
residues at the i+3 position,† but only within the optimal Acpc-
series (4h vs. 3 & 4n). This i+3 alkyl effect proved to be quite
general across the different i+2-series. Modest enantio-
selectivity increases of 7–10% ee were observed in the Aib-
series (4a vs. 4c & 4f), but more drastic increases of about 25%
View Article Online
DOI: 10.1039/C6CC01428C
(a)
(b)
2.964(3)
Å
3.096(3) Å
4l
type I' β-turn
φ(i+1) = 62.4(3)º, ψ(i+1) = 17.7(3)º
φ(i+2) = 72.7(4)º, ψ(i+2) = 11.9(5)º
Legend
3(c) 4l
• RMSD = 0.03 Å
for β-turn backbone overlay
Figure 3: (a) The X-ray crystal structure of 4l was also found to exhibit a
type I’ β-turn geometry.10 (b) Structural overlay of the X-ray crystal
structures of conformer 3(c) and 4l.
ee were observed in the L-Phe-series (4u vs. 4w). The low-
selectivity Aic§-series did not benefit from Leu-substitution to mol% with full retention of enantioselectivity (97:3 er) at the
any significant extent (4p vs. 4r). We were also intrigued by the expense of moderately diminished yield (80% isolated).
subtle effects of the C-terminal end cap on enantioselectivity.
We were also able to grow a single crystal of the new lead
While the dimethyl amides were indeed found to be superior catalyst, 4l, using the same slow vapour diffusion technique
in lower-selectivity regimes (4a vs. 4b, 4c vs. 4d, 4f vs. 4g, & 4p employed in the crystallization of 3. Interestingly, the crystal
vs. 4q), this trend did not necessarily hold within the more was quite fragile and fractured during X-ray diffraction at
selective catalyst series (4j vs. 4k & 4w vs. 4x). In fact, cryogenic temperatures.† We were eventually able to collect
examination of the methyl ester variant of 3 led to the diffraction data at –40 ºC without damaging the crystal, though
discovery of a new lead catalyst, 4l, which provided tribromide the thermal ellipsoids are quite large as a result. Nonetheless,
2 in 95:5 er§ without recourse to slow addition. Thus, it seems the unit cell contained only one conformer of 4l, which clearly
that the methyl esters may be beneficial above some adopted the same distorted type I’ β-turn as 3(c) (Figure 3a). In
selectivity threshold, especially when paired with Leu at the fact, the two structures overlay remarkably well, with an RMS
i+3 position (4l, & 4x). Dimethyl amides are generally deviation of only 0.03 Å in the β-turn backbone (Figure 3b).†
presumed to be stronger H-bond acceptors than methyl The intramolecular H-bonding network is nearly identical, as
esters.14 It seems plausible that a weakened NHDmaa to OLeu well, exhibiting the same sequential bis-β-turn that renders
(hairpin) H-bond could produce a significant geometric change, the structure helical.
especially when paired with backbone conformational driving
To obtain structural insight under conditions more
analogous to the reactions conditions, we complemented the
forces, such as the more helical tendencies of Acpc.13
Perhaps the most intriguing result to emerge from the solid-state analysis with solution-phase NMR studies of 3 and
expanded peptide data set involved the substitution of
-Pro 4l in d6-benzene. It is clear from a simple overlay of the two 1H-
for -Pro in each i+2 series (Figure 2). This change is expected NMR spectra that they exhibit, on average, in different
L
D
to induce a conformational shift from the “mirror image” β- structures (Figure 4a,b). Notably, the NHBoc signal of 3 is 1.28
turns (types I’ or II’) to the canonical β-turns (types I or II) via ppm further downfield than that of 4l, while the NHAcpc signal
an amide plane-flip in the loop region.15 Historically, we have of 4l is 0.93 ppm further downfield than that of 3. These data
often found that such changes to the catalyst turn sense are consistent with a difference in the H-bonded state of the
provided enantioselectivities in favour of the opposite
C(O)NMe2
(a) Peptide 3 (600 MHz, 0.01 M in C6D6, 25 ºC)
enantiomer with significant erosion of magnitude, especially
NHLeu
NHAcpc
αDmaa
αLeu
δ'D
-Pro
αD
δD
-Pro
-Pro
NHBoc
with Pro-Xaa-containing catalysts.16 Such is the case here in
CO2Me
(b) Peptide 4l (600 MHz, 0.01 M in C6D6, 25 ºC)
NHLeu
NHAcpc
the Aib- and Aic-series; the change from
Pro-Aib (4e) produced a drop from 35% ee to –11% ee, and the
change from -Pro-Aic (4s) to -Pro-Aic (4t) followed suit
(Figure 2). However, the change from -Pro-Acpc (4l) to -Pro-
D-Pro-Aib (4d) to L-
αLeu
αDmaa
αD
δD
δ'D
-Pro
-Pro
-Pro
NHBoc
D
L
D
L
(c) NOE-Derived Solution Structures
O
O
H
N
Acpc (4m) maintained a high level of enantioselectivity in the
opposite direction (–77% ee), which may suggest that absolute
enantiocontrol is largely dictated by the turn sense with
relatively little contribution from the flanking residues. In our
experience, this degree of enantio-divergence is notable and
unprecedented, especially for an epimeric catalyst, and it
warrants future investigation.
Me
Me
Me
N
OMe
Me
O
H
H
N
N
O
O
N
H
Me
simulated annealing/
DFT optimization
N
O
Me
O
simulated annealing/
DFT optimization
O
N
BocHN
N
Me
Me
Me
BocHN
N
Me
NOE Map of 4l
NOE Map of 3
type I' β-hairpin
type I' β-turn
(helical)
Though more reactive and selective than 3 under the
screening conditions (Figure 2), catalyst 4l provided 2 in 92%
isolated yield and 97:3 er using the optimized slow addition
protocol described in Scheme 1, the same er as observed for
3.† However, we were able to decrease the loading of 4l to 1
Solution Structure of 3
Solution Structure of 4l
Figure 4: Overlay of the 1H-NMR spectra of (a) 3 and (b) 4l showing
distinct differences in the hydrogen-bonding networks. (c) NOESY data
was used to compute solution structures of 3 and 4l.†
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

ChemComm
Page 4 of 4
COMMUNICATION
Journal Name
two peptides that is consistent with the crystallographic data.† Faculty of Arts and Sciences High Performance Computing
View Article Online
To investigate these trends at higher resolution, we modelled Center, and by the NSF under grantD#OCI:N10S.100389-/2C161C3C2014t2h8aCt
the solution structures of 3 and 4l using 2D-NMR and partially funded acquisition of the facilities.
1
computational techniques. We acquired H-1H NOESY data on
Notes and references
both peptides and integrated the non-sequential NOESY cross-
peaks to derive distance restraints,17 which were then used in
§ Abbreviations: Aib, aminoisobutyric acid; Aic, 2-aminoindane-
a simulated annealing protocol.† Geometry optimization of the
2-carboxamide; Acpc, 1-aminocyclopropyl-1-carboxamide; Boc,
annealing outputs using DFT§ at the B3LYP/6-31G** level of
N-tert-butoxy carbamate; DFT, density functional theory; Dmaa,
theory provided NMR-derived solution structures (Figure 4c).†
L-β-dimethylaminoalanine; ee, enantiomeric excess; er,
enantiomer ratio; NBS, N-bromosuccinimide; NOE, nuclear
In peptide 3, we were able to observe NOE§ interactions
Overhauser effect.
between NHLeu↔NHAcpc, NHLeu↔NHDmaa, βDmaa↔βLeu, NHLeu
↔
1
T. P. Yoon and E. N. Jacobsen, Science, 2003, 299, 1691; Q.-L.
Zhou (Ed.), Privileged Chiral Ligands and Catalysts, Wiley
VCH, Weinheim, Germany, 2011.
B. Lewandowski and H. Wennemers, Curr. Opin. Chem. Biol.,
2014, 22, 40; H. Wennemers, Chem. Commun., 2011, 47,
12036; E. A. Colbie Davie, S. M. Mennen, Y. Xu, and S. J.
Miller, Chem. Rev., 2007, 107, 5759.
α
D-Pro, and NHAcpc↔αD-Pro. A type I’ β-hairpin structure was
apparent in the computed structure (Figure 4c), which seems
to possess certain characteristics of all three solid-state
conformers 3(a–c). It shows the loop region dihedrals of 3(c),
the backbone bending of 3(b), and the NHBoc to OLeu hairpin H-
bond observed in both 3(a) and 3(b) in place of the NHAcpc to
OBoc H-bond of 3(c). The hybrid nature of the NMR-derived
solution structure may reflect a time average of multiple
equilibrating conformers in solution. On the other hand, the
solid- and solution-state structures of 4l coincided more
closely. Particularly relevant through-space interactions are
observed between NHLeu↔NHAcpc, NHLeu ↔t-BuBoc, NHAcpc↔αD-
Pro, NHAcpc↔δD-Pro, δD-Pro↔βAcpc, and δ'D-Pro↔βAcpc.. These turn-
consistent features were also corroborated in the calculated
structure, which produced a type I’ β-turn similar to the solid-
state structure of 4l (Figure 4c). These results reflect the subtle
consequences of swapping the C-terminal functional group in
an otherwise identical sequence. It seems as though the
favourable driving force to form a β-hairpin H-bond to the
dimethyl amide acceptor of 3 is counterbalanced by the helical
2
3
4
C. M. Wilmot and J. M. Thornton, J. Mol. Biol., 1988, 203,
221; E. G. Hutchinson and J. M. Thornton, Protein Sci., 1994,
3, 2207.
T. S. Haque, J. C. Little, and S. H. Gellman, J. Am. Chem. Soc.,
1996, 118, 6975; T. S. Haque, J. C. Little, and S. H. Gellman, J.
Am. Chem. Soc., 1994, 116, 4105; G.-B. Liang, C. J. Rito, and
S. H. Gellman, J. Am. Chem. Soc., 1992, 114, 4440.
M. B. Fierman, D. J. O’Leary, W. E. Steinmetz, and S. J. Miller,
J. Am. Chem. Soc., 2004, 126, 6967; C. E. Jakobsche, G. Peris,
and S. J. Miller, Angew. Chem. Int. Ed., 2008, 47, 6707.
For example: N. C. Abascal, P. A. Lichtor, M. W. Giuliano, and
S. J. Miller, Chem. Sci., 2014, 5, 4504.
A. Ravi, B. V. Venkataram Prasad, and P. Balaram, J. Am.
Chem. Soc., 1983, 105, 105; A. Ravi and P. Balaram,
Tetrahedron, 1984, 40, 2577; I. L. Karle, S. K. Awasthi and P.
Balaram, Proc. Nat. Acad. Sci. U.S.A., 1996, 93, 8189.
For relevant examples: C. R. Shugrue and S. J. Miller, Angew.
Chem. Int. Ed., 2015, 54, 11173; C. T. Mbofana and S. J.
Miller, J. Am. Chem. Soc., 2014, 136, 3285; K. T. Barrett and
S. J. Miller, J. Am. Chem. Soc., 2013, 135, 2963.
5
6
7
8
preferences of the D-Pro-Acpc turn motif observed in the solid-
state, while there is less of a tendency to hairpin formation to
the weaker methyl ester acceptor of 4l.
9
M. E. Diener, A. J. Metrano, S. Kusano, and S. J. Miller, J. Am.
Chem. Soc., 2015, 137, 12369.
In conclusion, we have shown that short sequence peptide-
based catalysts exhibit greater conformational diversity than
one might anticipate on the basis of design principles. Our
results provide evidence that there may be some tolerance of,
or even benefit to, catalyst flexibility, as we are able to achieve
high enantioselectivities in the bromination of 1 using these
peptides. However, this structural flexibility also complicates
the process of mechanistic analysis. Of course, it is possible
that the kinetically competent species is an all-together
different conformer, or that conformational changes occur in a
Curtin-Hammett-type fashion to access the active catalyst. In
any case, the experimental observation of multiple
conformations and turn types speaks to the challenges
associated with studying conformationally dynamic catalysts.
10 CYLview was used to render X-ray crystal data: CYLview,
1.0b; C. Y. Legault, Université de Sherbrooke, 2009
(http://www.cylview.org).
11 A. I. Jiménez, C. Cativiela, and M. Marraud, Tetrahedron
Lett., 2000, 41, 5353; G. Némethy and M. P. Printz,
Macromolecules, 1972, 5, 755.
12 J. T. Blank and S. J. Miller, Biopolymers (Pept. Sci.), 2006, 84,
38.
13 C. Toniolo, M. Crisma, F. Formaggio and C. Peggion
Biopolymers, 2001, 60, 396; E. Benedetti, B. Di Blasio, V.
Pavone, C. Pedone, A. Santini, V. Barone, F. Fraternelli, F.
Lelj, A. Bavoso, M. Crisma, and C. Toniolo, Int. J. Macromol.,
1989, 11, 353.
14 P. Gilli, L. Pretto, V. Bertolasi, and G. Gilli, Acc. Chem. Res.,
2009, 42, 33.
15 S. Hayward, Protein Sci., 2001, 10, 2219.
Experimentally restrained computational approaches may 16 For example: B. J. Cowen, L. B. Saunders, and S. J. Miller, J.
assist in the path forward.18
We would like to extend our sincerest thanks to Dr.
Byoungmoo Kim for helpful discussions and Matthew E. Diener
Am. Chem. Soc., 2009, 131, 6105; E. R. Jarvo, G. T. Copeland,
N. Papaioannou, P. J. Bonitatebus, and S. J. Miller, J. Am.
Chem. Soc., 1999, 121, 11638.
17 S. Macura, B. T. Farmer, II, and L. R. Brown, J. Mag. Res.,
early contributions to this project. We also thank the National
1986, 70, 493.
Institute of General Medical Sciences of the NIH (GM-068649) 18 R.-Z. Liao, S. Santoro, M. Gotsev, T. Marcelli, and F. Himo,
ACS Catal., 2016, 6, 1165; N. Mittal, K. M. Lippert, C. K. De, E.
G. Klauber, T. J. Emge, P. R. Schreiner, and D. Seidel, J. Am.
Chem. Soc., 2015, 137, 5748.
for financial support. A.J.M. gratefully acknowledges financial
support from the NSF Graduate Research Fellowship Program.
Computational work was supported by the Yale University
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins