Automated Design of Macrocycles for Therapeutic Applications: From Small Molecules to Peptides and Proteins

Article abstract of DOI:10.1021/acs.jmedchem.0c01500

Macrocycles and cyclic peptides are increasingly attractive therapeutic modalities as they often have improved affinity, are able to bind to extended protein surfaces, and otherwise have favorable properties. Macrocyclization of a known binder may stabili

Full text of DOI:10.1021/acs.jmedchem.0c01500

pubs.acs.org/jmc
Article
Automated Design of Macrocycles for Therapeutic Applications:
From Small Molecules to Peptides and Proteins
̈
Dan Sindhikara,* Michael Wagner, Paraskevi Gkeka, Stefan Gussregen, Garima Tiwari, Gerhard Hessler,
Engin Yapici, Ziyu Li, and Andreas Evers*
Cite This: https://dx.doi.org/10.1021/acs.jmedchem.0c01500
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Macrocycles and cyclic peptides are increasingly
attractive therapeutic modalities as they often have improved
affinity, are able to bind to extended protein surfaces, and
otherwise have favorable properties. Macrocyclization of a known
binder may stabilize its bioactive conformation and improve its
metabolic stability, cell permeability, and in certain cases oral
bioavailability. Herein, we present implementation and application
of an approach that automatically generates, evaluates, and
proposes cyclizations utilizing a library of well-established chemical reactions and reagents. Using the three-dimensional (3D)
conformation of the linear molecule in complex with a target protein as the starting point, this approach identifies attachment points,
generates linkers, evaluates their geometric compatibility, and ranks the resulting molecules with respect to their predicted
conformational stability and interactions with the target protein. As we show here with prospective and retrospective case studies,
this procedure can be applied for the macrocyclization of small molecules and peptides and even PROteolysis TArgeting Chimeras
(PROTACs) and proteins.
passive permeability by reducing polar surface area. In
addition, cyclization generally reduces degradation in the gut,
blood, and tissues by shielding residues from metabolic
proteases. Finally, covalent cyclization often enhances stability
against thermal and chemical stress (e.g., caused by elevated
temperature or presence of denaturants), which is often an
important goal in engineering studies for peptides or proteins/
enzymes for biotherapeutic or biotechnological applications.25
Medicinal chemistry allows further optimization of cyclic
peptides toward drug-likeness and enables access to the world
of nonpeptidic macrocycles in drug discovery. Thereby, despite
violating some or all of the “rule of five” (Ro5) parameters,
some naturally occurring, as well as synthetic, peptidic, and
nonpeptidic macrocycles have demonstrated the ability for cell
permeability and oral bioavailability.26−34 Indeed, a number of
macrocycles with a molecular weight (MW) ranging from 500
to 1500 Da have become successful drugs, with multiple
examples that can be administered orally.7,35−37
INTRODUCTION
■
During the past decade, macrocycles, defined as cyclic small
molecules or peptides typically with a molecular weight (MW)
of 500−2000 Da, have gained significant interest as therapeutic
agents, due to the introduction of new approaches for their
synthesis and screening.1−3 Moreover, there is an unmet need
for therapeutic agents that will be able to target a not yet well-
addressed type of disease-relevant surfaces and interfaces that
include, among others, extra- and intracellular protein−protein
interactions (PPIs) with large and shallow (featureless)
binding sites typically less amenable to small-molecule
drugs.4 Because of their size and complexity, macrocycles
have demonstrated that they can bind with antibody-like
affinity and specificity and successfully target PPIs.5−18 Hence,
macrocyclic molecules fill an important gap in the world of
drugs between small molecules and larger biologics.19,20
Significant progress has been made for peptides in terms of
synthesis, formulation, and delivery,21 but their therapeutic
value is partially limited due to their short in vivo half-life and
lack of oral bioavailability. Moreover, while in solution,
peptides span a large conformational space that increases the
entropic cost of binding. Cyclization is one of the ways to limit
this entropic cost of binding and thus improve its potency.
Importantly, it has been shown that constraining a peptide via
cyclization can also result in improved selectivity.22−24
Furthermore, cyclization is known to enhance the folding of
peptides into conformations that might allow the formation of
intramolecular hydrogen bonds, which in turn could improve
Meanwhile, a variety of chemical approaches for the
cyclization of peptides and small molecules have been
described in the literature (see refs 3, 6, 29, and 38−43).
Received: August 27, 2020
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
© XXXX American Chemical Society
J. Med. Chem. XXXX, XXX, XXX−XXX
A

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
One of the most popular strategies for cyclization is
lactamization. The first studies on intramolecular amide-bond
formation via side-chain stapling were reported by Felix et al.
for helix stabilization between i, i + 4 spaced amino acids.44
Since then, this approach has been further elaborated by many
others, exploring linkages with different chain lengths and
positioning, as described in detail in a number of reviews39,45
and references therein. The formation of carbon−carbon
bonds may be facilitated using a ring-closing metathesis
(RCM) reaction.46−48 This approach has been successfully
applied to several peptides binding to different biological
targets.49−55 Another strategy is the Cu(I)-catalyzed azide−
alkyne cycloaddition56,57 as a prototype of the popular “click
reaction”,58 which has been reported by Chorev et al. for an
analogue of the parathyroid hormone-related peptide59,60 and
by Wang et al. for B-cell CLL/lymphoma 9 (BCL9) α-helical
peptides.61 An alternative approach, photoinduced 1,3-dipolar
cycloaddition, the UV-induced reaction between tetrazoles and
alkenes, has been applied by Madden et al. to staple dual
peptide inhibitors of the p53-Mdm2/Mdmx interactions.62
Staples may also be introduced with thioethers by linking of
two cysteines via thiol−ene/−yne reactions63−66 or cysteine
arylation and alkylation.67−73 Cysteine alkylation with reagents,
containing three thiol-reactive groups, can also be used for
synthesis of bicyclic structures, when the precursor peptide
carries three cysteines.74,75 These approaches have been used
together with combinatorial chemistry or phage display to
screen structurally diverse monocyclic76−78 or bicyclic
peptides.79−81 Recently, it was demonstrated that this
chemistry can also be applied to the in situ cyclization of
proteins.82−84 For two proteins, Staphylococcus aureus sortase A
(SrtA) and the KIX domain from the human cAMP response
element-binding (CREB) protein, cysteine residues were
introduced in three surface-exposed positions and incubated
with a triselectrophilic cross-linker to generate bicyclic
enzymes with high tolerance toward thermal and chemical
stress.84
Given the three-dimensional (3D) structure of a linear
peptidic or nonpeptidic molecule as the starting point, the first
goal of cyclization is to maintain the conformation of the
bioactive region that is relevant for biological recognition and
affinity, and to introduce linkers where interactions are either
not affected or even further improved. Several studies
investigated the optimal positioning and linker lengths for
the stabilization of α-helical structures.46,59,61,68,85−91 However,
peptide−protein recognition is frequently mediated via non-
helical structures, such as loops, turns, or β-sheets,92 and might
therefore require specific optimization of the linker and its
positioning in the linear molecule for conformational
stabilization. Considering each residue in a peptide as potential
cyclization site and taking into account the diversity of
chemical strategies for cyclization might lead to a vast number
of molecules that could be synthesized. Implementation of
reliable in silico approaches for an accurate design is therefore
needed to guide experimental efforts toward the most
promising compounds.82,93−96 To perform rational design, a
reliable description of macrocycles’ conformational landscape
is essential. Unfortunately, the rugged conformational land-
scape limits the effectiveness of standard approaches for small
molecules to reliably predict macrocycle conformations.10,97,98
Consequently, significant work has been made into enhanced
sampling algorithms toward macrocycle conformation pre-
diction.97−109 Additional work has been put into docking or
otherwise calculating macrocycle binding free energies.110−116
More recently, a structure-based design of peptide macrocycles
targeting the interaction site of human adaptor protein 14-3-3
using free-energy perturbation (FEP) calculations was
proposed.117 Despite the above-mentioned efforts, to the
best of our knowledge, there is no computational method
available that combines automatic enumeration, evaluation,
and ranking of the macrocyclizations based on a list of
chemical linkers (derived from available reagents) and the 3D
structure of a linear reference.
The aim of the present study is to implement an approach
that, based on a given ligand 3D structure (ideally in complex
with a receptor), will provide synthesizable in silico designed
macrocyclic small molecules, peptides, or proteins targeting a
receptor−ligand or protein−protein interface (antagonists or
agonists). The presented approach automatically generates
proposals for synthesis by enumerating and screening
cyclizations. The screening is based on geometric constraints,
conformational rigidity, and optionally the predicted inter-
actions of the macrocycles to a target protein. The
enumeration uses linkers or breeding structures for cyclization
that can be synthesized with well-established chemical
reactions utilizing (commercially) available chemical reagents.
This list of chemical linkers can be modified by the user to
consider specific cyclization chemistries and reagents.
RESULTS AND DISCUSSION
■
Here, we present an overview of the general workflow for the
recognition and evaluation of sites for macrocyclization, the
identification of compatible chemical linkers, and their
conformational and enthalpic scoring. To demonstrate the
scope of this approach, prospective applications for a peptide
(dual glucagon-like peptide-1 (GLP-1)/glucagon receptor
agonists) and a small molecule (activated thrombin-activatable
fibrinolysis inhibitor (TAFIa)) project and retrospective
applications for a PROteolysis TArgeting Chimera (PROTAC)
and a protein will be shown. In addition, a further retrospective
peptide (β-catenin inhibitors) and small molecule (factor XIa
inhibitors) example can be found in the Supporting
Information (SI).
General Workflow. Our strategy involves several inte-
grated steps summarized in Figure 1. For technical details, see
the Experimental Section. Briefly, we start with the 3D
structure of the linear ligand in complex with the target
protein. Synthetic capabilities may suggest a set of cyclization
linkers and attachment points (“limbs”). The approach will
generate a full conformational ensemble of linkers, permute
pairs of attachment points within the rigid ligand, and then
geometrically filter cyclizations to eliminate highly strained or
otherwise unfavorable ones. The output are 3D structures of
complexed cyclized ligands. Additional scoring can reduce the
number of synthesis proposals. Since the goal of cyclization is
often stabilization of the bioactive conformation, here, we run
conformational sampling and rank the cyclized ligands
according to their calculated conformational propensity.
Conformational scoring for small molecules and small peptides
is performed using Prime Macrocycle Conformational
Sampling (Prime-MCS)106,107 and for larger peptides and
proteins with molecular dynamics (MD) simulations. An
“enthalpic” scoring of the protein−ligand complex can be done
using either the empirical scoring function GlideScore118,119 or
the molecular mechanics-generalized Born surface area (MM/
GBSA) method.120,121 In a standard project setting, it is
B
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
accurate correlation with experimental potencies but are
generally sufficient to provide an enrichment of active
molecules among the top scorers.
Identification of Attachment Points for Cyclization.
For the identification of attachment points (see Figure 1a), our
approach differentiates between the application for peptides/
proteins and small molecules. (i) For peptides, the default
attachment “vectors” for the introduction of cyclizations are
typically the Cα−Cβ bonds within the peptide sequence.
Hence, this approach allows us to consider any (commercially
or internally) available α-amino acid with side-chain groups
suitably reactive for cyclization, and the resulting design
proposals can be synthesized by classical stepwise (solid-
phase) peptide synthesis with a subsequent cyclization step. As
an additional option, it is possible to consider the Cα−Hα
vector to introduce D-amino acids for cyclization and
conformational stabilization. (ii) For small molecules, on the
other hand, the positioning of the attachment points for
cyclization generally depends on the specific synthetic route of
the target molecule. Therefore, our approach allows us to
define the most suited attachment vectors by specifying atomic
numbers or a user-defined substructure (defined as simplified
molecular-input line-entry system (SMILES) arbitrary target
specification (SMARTS)124) within the starting molecule.
With this procedure, it is also possible to introduce
“nonstandard” cyclizations into a peptide, such as side chain-
to-tail, head-to-side chain, head-to-tail, or other cyclizations.
Chemical Linkers for Cyclization. There are two options
to obtain chemical linkers for cyclization (see Figure 1b),
which can be either (i) provided as an explicit list derived from
reagents that allow cyclization along a known chemical route
or (ii) grown automatically by fusing a set of small spacer
fragments. Table 1 provides a list of example linkers which are
particularly suited for the side-chain cyclization of peptides.
These linkers are derived from common modified amino acids
that allow for cyclization via lactamization, click chemistry,
ring-closing metathesis, or cysteine alkylation. With respect to
our in silico method, these linkers are attached to the Cα−Cβ
or Cα−Hα vectors of the peptide to design cyclizations via L-
or D-amino acids. This list can be reduced or extended, for
example, based on other cyclization chemistry and/or the
availability of corresponding reagents, to assure the cyclization
proposals can be straightforwardly synthesized. Providing an
explicit list of linkers might also be used in a small-molecule
scenario where the synthetic route is already planned to
identify the most promising linkers from a list of available
chemical reagents.
Figure 1. Procedure for the automated design of macrocycles. (Step
1) The user prepares a linear starting molecule. The attachment
points may be either identified automatically or specified by the user.
(Step 2) Generation of chemical linkers. These might either be
provided as explicit lists, e.g., based on available chemical reagents, or
can be grown automatically by combining and enumerating small
breeding fragments. (Step 3) A full conformational ensemble of each
linker is generated and (step 4) mapped with all attachment points to
eliminate unfavorable cyclizations applying different geometric filter
criteria. (Step 5) Finally, the cyclized ligands are ranked applying
conformational and/or enthalpic scoring.
Alternatively, rather than specifying predefined cyclization
linkers, the linkers can be constructed in silico from a small set
of “breeding fragments” (see Table 2, for example, fragments),
combined and enumerated to fit within the geometrical
constraints of the ligand−receptor complex. As a result, the
algorithm will suggest a diverse set of linker chains that are
synthetically tractable. While the default spacers shown in
Table 2 are amide, poly(ethylene glycol) (PEG), or alkyl
moieties, this list of fragments can be expanded, for example, to
include aromatic rings, heterocycles, or other reactive groups.
Generation of Bicyclic Structures. As mentioned above,
it is often advantageous to transform linear peptides or
proteins to bicyclic structures, for example, via cysteine
alkylation using symmetric cyclization reagents that contain
three thiol-reactive groups (see Table 3).74,75,79−81 With our
approach, it is possible to generate such bicyclic structures in a
generally difficult to accurately dissect the enthalpic and
entropic contributions to the free energy of binding, in
particular if experimental data are obtained from cellular in
vitro assays. In terms of in silico prediction, rigorous free-
energy calculations, encapsulating both entropic and enthalpic
components, may be applied.116,122 However, since these
calculations are time-consuming and their predictivity depends
on the accuracy of the generated binding poses,123 here, for
high-speed enthalpic screening, we applied GlideScore for
small-molecule cyclizations and MM/GBSA otherwise. For the
selection of proposals for experimental synthesis in the
presented prospective case studies, we therefore ranked the
virtually designed macrocycles based on their predicted (1)
enthalpic scores and (2) subjected the top-ranked molecules to
conformational scoring. These scores do usually not provide an
C
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Table 1. Exemplary Chemical Linkers for Peptide
Cyclization
Table 1. continued
b
click chemistry. This list can be modified by the user based on specific
reagents and chemical cyclization routes.
Table 2. Example Breeding Fragments for the In Silico
Construction and Growth of Linkers for Small-Molecule
a
Cyclization
a
This list can be modified by the user.
Table 3. Example of Three-Armed Linkers for the
Generation of Bicyclic Structures Based on Common
a
Chemical Reagents Applying Cysteine Alkylation
a
This list can be modified by the user.
stepwise manner, as illustrated in Figure 2. First, the
trisymmetrical linker is converted to a two-armed linker with
the third limb considered as “nonreactive”. With this two-
armed linker, all geometrically reasonable (mono-)cyclized
structures are generated as outlined above. In the next step,
these cyclized structures are used as the starting points for a
second cyclization routine with an appropriate linker, where
the previously nonreactive limb is now considered as first
a
b
Ring-closing metathesis (RCM). Two-armed linkers for the
generation of monocycles based on common chemical reagents
applying cysteine alkylation, lactamization, ring-closing metathesis, or
D
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
mode of 4a in the full-length structure of the GLP-1 receptor
(pdb code 5VAI).127 In the first step, we automatically
generated cyclization proposals with all chemical linkers shown
in Table 1, allowing all residues to be considered as a
cyclization site. As a result, >2000 linkages were proposed by
the cyclization algorithm (see Figure 3b). Figure 3b illustrates
that the algorithm automatically identifies and excludes all
residues of the linear peptide from cyclization that establish
contacts with the GLP-1 receptor, for example, in the N-
terminal peptide tail where several residues are known to
establish interactions that are important for agonistic activity.
To reduce the number of synthesis proposals, we next (i)
focused the list of linkers to 1b, 1f, and 1k (see Table 1) for
cyclization via cysteine alkylation, lactamization, or RCM, and
(ii) based on structure−activity relationship (SAR) knowledge,
specified residues 10, 12, 14, 16, 18, 20, 21, 28, 33, 35, and 39
as possible attachment sites for cyclization. With these
parameters, 29 cyclization proposals were obtained (see Figure
3c). These proposals were subjected to molecular mechanical
minimization, ranked according to their calculated MM/GBSA
binding energy (see minimized poses in Figure 3d) and
subjected to conformational scoring with MD simulations (see
the Experimental Section for details).128 To assure that these
cyclizations do not induce conformations that are too distant
from the putative bioactive conformation, we considered only
variants with a favorable conformational score (root-mean-
square deviation (RMSD) <2.0) for further analysis. Finally,
we assessed the remaining design proposals with respect to
their predicted scores, diversity, and novelty. Using the linear
peptides 4a or 4g, which carries a D-serine (dSer) in position 2
for stabilization against dipeptidyl peptidase IV (DPP IV)-
mediated cleavage, as structural references, we synthesized 10
stapled peptides, applying RCM, lactamization, or cysteine
alkylation (see details in the Experimental Section and
analytical data in the Supporting Information). Subsequently,
we tested their potency on the GLP-1 and glucagon receptor in
a cAMP assay in receptor overexpressing human embryonic
kidney (HEK)-293 cells (see computed scores and exper-
imental potency data in Table 4).
Figure 2. Approach for modeling bicyclic peptide structures using a
trisymmetrical linker. In step 1, all geometrically reasonable
monocyclic structures are generated. In the second cyclization routine
(step 2), the unlinked (third) attachment point of the linker is
evaluated for attachment to the peptide backbone for construction of
the second cycle.
attachment point and each Cα atom of the peptide backbone
as second attachment point.
Application Examples. Peptide Example: Cyclization of
Dual Glucagon-like Peptide-1 (GLP-1)/Glucagon Receptor
Agonists (Prospective Case Study). We previously described
the discovery of dual agonists, which were identified by rational
design.125,126 Structural elements of glucagon were engineered
into the selective GLP-1 receptor agonist exendin-4, resulting
in hybrid peptides with potent dual GLP-1/glucagon receptor
activity.125 To probe the applicability of our cyclization
approach in a prospective manner, here, we used the previously
reported linear dual agonist 4a as the starting point for the
introduction of molecular cyclizations to optimize the potency
on the GLP-1 receptor. Figure 3a shows the predicted binding
As shown in Figure 3d, several i, i + 4 staples were identified
along the peptide helix, as in peptides 4b−d and 4h−i. Such
staples, obtained by lactamization or RCM, have been
described for other GLP-1 receptor agonists before.129−132
Here, introduction of these staples into the linear peptides 4a
and 4g resulted in improved potency, presumably due to
conformational stabilization and/or establishment of additional
interactions with the GLP-1 receptor. In detail, RCM from
positions 14−18 increased potency vs the linear reference
molecule by a factor of 45 (4d vs 4a) or even 63 (4h vs 4g),
whereas cyclization from positions 10−14 increased potency
by a factor of >2 (4c vs 4a). For the (K10, E14) lactamizations,
potency was slightly worsened (4b vs 4a), while the (K16,
E20) cyclization improved GLP-1R potency by a factor of 12
(4i vs 4g). These data validate that our approach is capable of
identifying classical cyclizations in helical peptide regions.
In addition to these i, i + 4 staples, we were interested to see
how far the method is able to design cyclizations within the
nonhelical peptide region. Therefore, we additionally selected
automatically generated cyclization proposals with attachment
points outside the peptide helix (i.e., residues 28−39) for
chemical synthesis. Applying lactamization, positions 28 and
33 were covalently linked via the side chains of Glu and Lys
residues, providing peptides 4j and 4k (see Figure 3d). We
Figure 3. Cyclization of dual agonists of the GLP-1 and glucagon
receptor. (a) Predicted binding mode of the dual agonist 4a at the
GLP-1R (pdb code 5VAI). (b) Automatically generated cyclization
proposals with all chemical linkers shown in Table 1. (c) Cyclization
proposals using a focused list of reagents and attachment positions.
(d) Predicted binding modes of synthesized peptides obtained from
MM/GBSA energy minimization.
E
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Table 4. Sequences, Computed Properties, and EC50 Values (Plus SEM Values; n = 2; Measured in pM in a cAMP Assay in
Overexpressing HEK-293 Cell Lines) of Stapled Peptidic Dual GLP-1/Glucagon Receptor Agonists and Their Linear
g
References
(a)
no
sequence
4a
HSQGTFTSDLSKQMDSRRAQDFIEWLKNGGPSSGAPPPS−NH2
(b)
computed properties
EC50 [pM] ( SEM, n = 2)
a
b
c
no
sequence modification vs 4a
conformational score
enthalpic score
GLP-1 receptor
glucagon receptor
4a
4b
4c
4d
4e
4f
4g
4h
4i
1.69
1.44
1.62
1.66
1.51
2.71
1.69
1.66
1.35
1.48
1.40
−193.1
−187.4
−192.9
−186.5
−207.3
−217.9
−217.6
−210.6
−219.0
−218.4
−247.9
18.3 1.0
47.5 3.6
7.4 0.5
0.4 0.0
24.9 1.2
194 14.5
75.4 3.7
1.2 0.1
6.4 0.2
14.7 0.8
8.13 0.3
28.6 1.2
119 5.0
2.3 0.1
369 20.5
12.9 0.9
578 32.0
122 5.5
8430 610
7.8 0.5
(K10,E14)-lactam
(Pal 10,Pal14)-RCM
d
(Pal14,Pal18)-RCM
e
(C21,C35,mXyl )-Cys-alk
f
(C21,C28,C35,Mes )-Cys-alk
dSer2
dSer2;(Pal14,Pal18)-RCM
dSer2;(K16,E20)-lactam
dSer2;(E28,K33)-lactam
dSer2;(K28,E33)-lactam
4j
4k
54.2 2.8
59.6 3.9
a
b
The conformational score was calculated as RMSD over the Cα residues. The enthalpic score was obtained from Prime MM/GBSA based on the
predicted peptide binding mode in the GLP-1 receptor. The enthalpic score was not calculated in the presence of the glucagon receptor. EC50
values are provided here as additional SAR information. Pal = 4-pentenyl-alanine. mXyl = meta-xylyl. Mes = mesitylene. The IUPAC names of
c
d
e
f
g
all molecules are provided in Table S4.
Figure 4. Macrocylization of a TAFIa inhibitor. (a) X-ray structure of anabaenopeptin C in complex with the surrogate protease carboxypeptidase
B (pdb code 5LRJ). (b) Anabaenopepin C was modified to obtain the starting structure for rescaffolding of the macrocycle. The attachment vectors
are indicated in magenta. (c) Breeding fragments that were combined and enumerated for cyclization. Predicted binding modes of (d)
automatically generated cyclization proposals and (e) four cyclization proposals that were synthesized and experimentally tested.
were delighted to see that these modifications improved
potency vs the linear reference peptide either by a factor of 5
(4j vs 4g) or even 9 (4k vs 4g). Finally, a covalent linkage
introduced between residues 21 and 35 by cysteine alkylation
via a meta-xylyl linker (peptide 4e, see Figure 3d) maintained
activity at the GLP-1 receptor. Such cyclizations as introduced
in these positions for peptides 4j, 4k, and 4e have, to our
knowledge, not been described for exendin-4 analogues in the
literature before and demonstrate that our approach is capable
of identifying novel cyclizations not only in helical but also in
looped regions of a peptide.
only one proposal was obtained (peptide 4f). Although it
revealed an unfavorable conformational score (RMSD = 2.71),
we synthesized 4f using 1,3,5-tris(bromomethyl)benzene
(TBMB) as cysteine cross-linker, with Cys residues introduced
into positions 21, 28, and 35. EC50 determination revealed,
however, a significant drop in potency (>factor 10 at both
receptors).
Comparison of the computed enthalpic scores with the
experimental GLP-1 receptor potency data (see data in Table
4) does not reveal a linear correlation, possibly due to the fact
that the modeled binding modes are not based on accurate
experimental structures (the resolution of the GLP-1 receptor
structure pdb 5VAI is 4.1 Å). With respect to their
conformational scores, the monocyclic peptides do not
significantly differ. Interestingly, the bicyclic peptide 4f reveals
Based on these findings, we also applied our algorithm for
the identification of bicyclic structures (see above) using the
three-armed linker 3a as analogue of the two-armed linker 1b.
Due to the geometric constraints of the peptide’s 3D structure,
F
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
c
Table 5. Macrocyclic Derivatives with Different Linkers and Their Computed Conformational and Enthalpic Scores
a
b
c
The conformational score was calculated by Prime-MCS. The enthalpic score was obtained from GlideScore. 5a−c were synthesized and
subjected to IC50 determination.
the lowest potency and has the worst conformational score,
suggesting that the introduced bicycle forces the peptide
backbone too far away from the bioactive conformation.
In terms of SAR, it is interesting that the introduced
cyclizations do not always result in proportional potency shifts
at the GLP-1 and glucagon receptor. Most remarkably,
introduction of cyclizations at positions 14−18 by RCM
results in a significant potency increase at the GLP-1 receptor
compared to the linear reference (factor 46 for 4d vs 4a and
factor 63 for 4h vs 4g) but a significant potency decrease at the
glucagon receptor (factor 29 for 4a vs 4d and factor 69 for 4g
vs 4h). These results confirm that cyclization of peptides might
not only be useful for the optimization of potency or metabolic
stability but also to modulate selectivity between similar
targets.22−24
within noncanonical peptide regions based on the 3D structure
of the linear reference peptide. Although we did not observe a
perfect correlation between computed scores and experimental
potency data, the approach was able to suggest highly active
compounds for synthesis. Indeed, all synthesized design
proposals (except for 4f) were able to activate the GLP-1
receptor in the one- or two-digit picomolar range. Six out of
these 10 molecules are even more potent than the linear
reference peptides. Since the in silico approach was designed to
consider only chemical linkers that are derived from available
chemical reagents, all molecules could be straightforwardly
synthesized.
Small Molecule Example 1: Macrocyclic Inhibitors of
Activated Thrombin-Activatable Fibrinolysis Inhibitor
(TAFIa) from Natural Product Anabaenopeptin (Prospective
Case Study). We previously described the identification of
anabaenopeptins, cyclic peptides produced by cyanobacteria,
as potent inhibitors of TAFIa.133 Starting from the X-ray co-
In summary, this prospective case study illustrates that our
approach identifies classical motifs for the stabilization of α-
helices, but in addition is able to identify favorable cyclizations
G
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
crystal structure of anabaenopeptin C bound to the surrogate
protease carboxypeptidase B (pdb code 5LRJ; see Figure 4a),
drug design efforts led to simplified noncyclic small molecules
with increased ligand efficiency.134 In addition, we started a
structure-based rational approach toward macrocycles to
reduce the size and polar peptidic character of the peptidic
ring. Here, we used our automated approach for the
prospective design of macrocyclic synthesis proposals. Starting
from the bioactive conformation of anabaenopeptin C (pdb
code 5lrj), we sketched a precursor structure with the
attachment points for cyclization at the exit vectors of the
macrocyclic ring (see Figure 4b and Table 5). To identify
suited linkers for cyclization, a set of five reactive “breeding
fragments” (see Figure 4c) was provided to the in silico
algorithm. Combinatorial enumeration of these fragments
would theoretically result in >300 000 different linker chains.
By automatic consideration of the geometrical constraints
imposed by the linker attachment vectors and the protein
environment, this number was automatically reduced to only
68 and further filtered out for unwanted structural motifs (see
the Experimental Section for technical details), resulting in 45
cyclization proposals. These were docked into the protease
binding pocket, ranked according to their enthalpic scores
(GlideScore; see Figure 4d), filtered by their conformational
scores (RMSD < 2.0) and inspected with respect to their
diversity, novelty, and synthetic accessibility. Here, we
synthesized four macrocyclic analogues and subsequently
tested their potency in an in vitro assay (see the Experimental
Section and the Supporting Information for all experimental
details).
In conclusion, this prospective study demonstrates how
molecules with larger peptidic motifs can be generally
converted into leadlike small macrocycles with favorable ligand
efficiency by shrinking and rescaffolding those regions that do
not establish key interactions with the target protein.
Cyclization of a PROTAC (Retrospective Case Study).
PROteolysis TArgeting Chimeras (PROTACs) have become
an appealing technology for modulating a protein of interest by
degradation.136 PROTACs are bifunctional molecules where
one ligand binds a protein of interest and the other one a
ubiquitin ligase (E3). Both ligands are connected with an
optimal linker. Degradation is initiated when PROTACs bind
the protein of interest and E3 to form a ternary complex.
Recently, Testa et al. demonstrated that macrocyclization can
also be introduced into PROTACs.137 A crystal structure of
PROTAC MZ1 (6a) in complex with the E3 ligase von
Hippel−Lindau protein (VHL) and its target, the second
bromodomain (BD2) of the bromodomain and extra-terminal
(BET) motif protein (see Figure 5a), was used as a starting
As shown in Table 5, these four macrocycles cover three
different chemical classes that were synthesized via different
chemical routes (see SI). Although the docking modes suggest
similar 3D conformations, the macrocycles are considerably
different with respect to their two-dimensional (2D) structures
(except for 5b and 5d). IC50 determination (see Table 5)
revealed inhibition for all four macrocycles in the two-digit
nanomolar range, with the most potent compound (5d)
showing an IC50 value of 10 nM, which is in a similar range
compared to other potent small-molecule inhibitors of TAFIa
that have been reported before.134,135 In comparison to the
reference molecule anabaenopeptin C, this represents a
potency loss of factor 5, but therefore a significant improve-
ment in terms of molecular weight (808 vs 483 Da). Thus, the
obtained small-molecule macrocycles represent interesting
starting points for further multiparameter compound opti-
mization toward improved lead likeness, for example, by
replacing the basic lysine residue by other less polar amino
acids.134,135
As for the previous case study, there is no linear correlation
between the computed scores and the experimental affinities
(see Table 5). Interestingly, the least potent macrocycle also
reveals the weakest conformational score. But since this dataset
is small and does not significantly differ in terms of scores and
pIC50 data, it is not finally possible to conclude in how far the
scoring scheme was able to efficiently enrich active molecules
among the top scorers. Nevertheless, we would like to
emphasize that the approach was able to suggest a small set
of chemically feasible synthesis proposals from a large
theoretical chemical space, with all synthesized molecules
representing novel, potent, and interesting starting points for
further optimization.
Figure 5. (a) X-ray structure of the linear PROTAC MZ1 (8a) in
complex with Brd4BD2 and VHL-EloB-EloC (pdb code 5T35). (b)
The PROTAC structure was modified to obtain the starting structure
for cyclization. The attachment vectors are indicated in magenta. (c)
Cyclized conformations comprising PEG units of different length,
minimized in the protein environment. (d) Comparison of the
predicted binding mode of 8b with the experimental binding mode
(pdb 6SIS).
point for macrocyclization by introducing a PEG linker at
attachment sites that are pointing away from both proteins
(Figure 5b). Despite a loss of binding affinity for BET family
member bromodomain-containing protein 4 (Brd4), the
macro-PROTAC 6b, comprising three PEG units, exhibited
cellular activity comparable to the linear PROTAC 6a. Here,
we investigated whether this cyclization can be reproduced
with our automated approach, using the precursor structure
shown in Figure 5b as the starting point. For the in silico
construction and growth of linkers, the PEG fragment 2a
(Table 2) was provided as breeding fragment. This fragment
was combined and enumerated to fit within the geometrical
constraints of the ligand−protein binding sites. Six linkers
comprising three to eight PEG units were identified, including
the three-PEG linker, which was synthesized (6b) and
H
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
successfully tested by Testa et al. All macrocyclic PROTAC
proposals were subjected to MM/GBSA binding energy
calculations with the Prime software (see minimized poses in
Figure 5c). Comparison of 6b with the X-ray structure (pdb
code 6SIS) reveals that the experimental binding pose could be
well reproduced (see Figure 5d).
Due to their high molecular weight, a major optimization
parameter of PROTACs is often their cellular activity.136,138
This case study demonstrates the applicability of our in silico
macrocyclization approach for PROTACs and might therefore
represent an interesting strategy to enhance their drug-likeness
by improving cell permeability and cellular stability.
Protein Example: Generation of Bicyclic Structures of the
KIX Domain from the Human CREB Binding Protein for
Thermal Stabilization (Retrospective Case Study). The KIX
domain (kinase-inducible domain (KID) interaction domain)
is an adaptor domain of transcriptional coactivator cAMP
response element-binding (CREB) protein with two different
protein binding partners and is composed of a central three α-
helix bundle (α1, α2, α3). The junction between this bundle
and the C-terminal 310 helix (G1) is crucial for structural
integrity (Figure 6a).139 Recently, Pelay-Gimeno et al.
In this case study, we investigated whether our in silico
cyclization approach is able to automatically generate this
bicyclic construct. In analogy to the procedure described by
Pelay-Gimeno et al., residues around the C-terminal G1 helix
were selected as attachment sites for the generation of
trivalently linked structures (see Figure 6a) using linker 3c
(Table 3). Applying our procedure for the automatic
generation of bicyclic structures (see above), 122 proposals
were obtained, including the construct designed by Pelay-
Gimeno et al. Both the cyclized construct and the natural
protein were subjected to a 100 ns unrestrained MD
simulation at an elevated temperature (340 K) in explicit
water to rationalize the experimentally observed improvement
of the cyclized variant in thermal stability. Although the
conformational fluctuation in the lower unconstrained region
of the helix bundle is similar between the cyclized and
noncyclized variants (see Figure 6c,d), the region where helix
G1 is covalently linked to α1 and α2 isin agreement with the
improved thermal stability data reported by Pelay-Gimeno et
al.conformationally more stable (RMSDCα 1.04 vs 1.72 Å).
This case demonstrates the applicability of in silico
cyclization as a suitable method for the engineering of proteins
for increased tolerance toward thermal and chemical
denaturation.84
SUMMARY AND CONCLUSIONS
■
Peptidic and nonpeptidic macrocycles represent an underex-
plored class of drugs with properties that could address current
deficiencies of peptide and small-molecule drugs.96 Therefore,
cyclization of molecules from small molecules to peptides and
proteins has become an important approach in pharmaceutical
design. Here, we have designed, implemented, and applied a
computational algorithm to automatically generate cyclized
molecules and screen them using conformational and enthalpic
scoring. Through case studies on both retrospective and
prospective data, we have shown that the method is able to
generate numerous reasonable cyclizations, including the ones
generated from simple breeding units. The method incorpo-
rates user-dictated synthetic capabilities and constraints to
ensure proposals that can be straightforwardly synthesized.
As we have demonstrated in two prospective case studies
(peptidic dual GLP-1/glucagon receptor agonists and small-
molecule TAFI inhibitors), the method was able to generate
synthesis proposals that resulted in novel molecules that
provided new SAR insights. Although all synthesis proposals
resulted in highly potent molecules, we observed that there was
no linear correlation between the predicted scores and the
experimentally determined in vitro potencies, probably due to
still existing inaccuracies of scoring functions and docking
tools, as, for example, recently reviewed by Guedes et al.140
Nevertheless, since several studies have demonstrated the
general efficiency of scoring functions in enriching active
compounds among top scorers in virtual screening applica-
tions, we recommend to use enthalpic scoring in a realistic
project scenario to reduce the number of virtual hits to a
tractable number (e.g., ≤100). These hits should then be
subjected to conformational scoring and/or free-energy
calculations or the application of other (e.g., tailor-made)
scoring functions. Finally, we recommend to inspect the
resulting hits with respect to novelty, diversity, and availability
of reagents to select the most promising candidates for
synthesis.
Figure 6. Generation of bicyclic variants of KIX domain from the
human CREB binding protein. (a) NMR structure of KIX (PDB code
2AGH). The junction between the helical bundle (α1, α2, α3) and
the C-terminal 310 helix (G1) that was subjected to covalent
stabilization is indicated. (b) Cyclized construct that links residues
594, 599, and 646. (c, d) Three-dimensional alignment of snapshots
from 100 ns simulations of the linear and cyclized proteins.
presented an in situ protein stabilization procedure (IN-
CYPRO) that involves the introduction of three surface-
exposed cysteine residues placed at different secondary
structure elements while still being in spatial proximity.84
The recombinantly produced construct with cysteine residues
in three different positions is then reacted with a triselec-
trophile for covalent cyclization. With this approach, a bicyclic
variant of the KIX domain was designed that was cyclized via
positions 594, 599, and 646 and exhibited strongly increased
thermal stability while still showing functional binding activity.
I
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Figure 7. Workflow for (a) peptides and (b) small molecules.
trifluoroacetic acid (TFA)). Purity and chemical identity of the
product were assessed by ultra performance liquid chromatography
(UPLC) and liquid chromatography-mass spectrometry (LC-MS)
(for details, see Table S1) and confirmed to have ≥95% purity for all
key compounds.
The presented cyclization approach can be applied to
different modalities and consider peptidic and nonpeptidic
structural motifs. This aspect might, for example, be
particularly interesting in a project where an inhibitor of a
PPI is searched. After identification of protein residues
interacting with the target, the approach can first be used for
the design of smaller cyclic peptides. In the next step, this
cyclic peptide can be converted to a smaller macrocycle with
less peptidic character to optimize cell permeability and oral
bioavailability. Potential future advances in predicting perme-
ability141 and pharmacokinetic properties of macrocycles might
further strengthen the applicability of the presented method
for the multiparameter optimization of macrocycles.
In Vitro Cellular Assays for GLP-1 and Glucagon Receptor
Potency. Agonism of peptides for the two receptors was determined
by functional assays measuring cAMP response of HEK-293 cell lines
stably expressing human GLP-1 or glucagon receptor. The cells were
grown in a T-175 culture flask placed at 37 °C to near confluence in
medium (Dulbecco’s modified Eagle’s medium (DMEM)/10% fetal
bovine serum (FBS)) and collected in 2 mL vials in cell culture
medium containing 10% dimethyl sulfoxide (DMSO) in concen-
tration of 10−50 million cells/mL. Each vial contained 1.8 mL of
cells. The vials were slowly frozen to −80 °C in isopropanol and then
transferred in liquid nitrogen for storage. Prior to their use, frozen
cells were thawed quickly at 37 °C and washed (5 min at 900 rpm)
with 20 mL of cell buffer (1× Hank’s balanced salt solution (HBSS);
20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
(HEPES), plus 0.1% human serum albumin (HSA)). The cells were
resuspended in assay buffer (cell buffer plus 2 mM isobutylmethyl-
xanthine (IBMX)) and adjusted to a cell density of 1 million cells/
mL. For measurement of cAMP generation, 5 μL of cells (final 5000
cells/well) and 5 μL of test compound were added to a 384-well plate,
followed by incubation for 30 min at room temperature. The cAMP
generated was determined using a kit from Cisbio Corp. based on
homogeneous time-resolved fluorescence (HTRF). The cAMP assay
was performed according to manufacturer’s instructions (Cisbio).
After addition of HTRF reagents diluted in lysis buffer (kit
components), the plates were incubated for 1 h, followed by
measurement of the fluorescence ratio at 665/620 nm. In vitro
potency of agonists was quantified by determining the concentrations
that caused 50% activation of the maximal response (EC50). Reported
values are mean ( SEM) EC50 (pM) from n = 2 values within a single
experiment.
Taken together, we think that this approach marks a large
step forward in accelerated pharmaceutical design via efficient
molecular cyclization.
EXPERIMENTAL SECTION
■
Synthesis of Peptides. Solid-phase synthesis was carried out on a
typical Rink resin (e.g., from Agilent Technologies with a loading of
0.38 mmol/g, 75−150 μm). The Fmoc synthesis strategy was applied
with O-(benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluoro-
phosphate (HBTU)/N,N-diisopropylethylamine (DIPEA) activation.
The peptide was cleaved from the resin with King’s cocktail.142 For
the synthesis of cyclic lactam containing peptides, orthogonal
protecting groups were used (allyl/alloc for the cyclization residues).
For the synthesis of hydrocarbon stapled peptides via ring-closing
metathesis (RCM), the corresponding amino acids with alkene side
chains were used. On resin, cyclization was achieved by treatment of
the resin with the Hoveyda−Grubbs second-generation catalyst from
Aldrich. Finally, for the synthesis of cysteine alkylated peptides, cross-
linking was achieved by addition of the cross-linker (1,3-bis-
(bromomethyl)benzene for 4e and 1,3,5-tris(bromomethyl)benzene
for 4f) to the Cys deprotected peptides. The crude product was
purified via preparative high-performance liquid chromatography
(HPLC) on a Waters column (e.g., XBridge, BEH130, Prep C18, 5
μM) using an acetonitrile/water gradient (both buffers with 0.1%
Synthesis of TAFIa Inhibitors. All solvents used were
commercially available and were used without further purification.
Reactions were typically run using anhydrous solvents (unless
otherwise noted) under an inert atmosphere of argon. Starting
J
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
materials used were available from commercial sources. 1H NMR
spectra were recorded in the indicated deuterated solvent at 500
MHz. Purity of all compounds tested in biological assays was
determined to be ≥95% by LC-MS. Detailed synthesis procedures
and analytical data are provided in the Supporting Information.
In Vitro Cellular Assays for TAFIa Inhibition. The prepared
substance was tested for TAFIa inhibition using the Actichrome
plasma TAFI Activity Kit from American Diagnostica (Pr. No. 874).
This entailed adding 28 μL of assay buffer (20 mM HEPES, 150 mM
NaCl, pH 7.4) and 10 μL of TAFIa (American Diagnostica Pr. No.
874TAFIA; 2.5 μg/mL) to 2 μL of 2.5 mM DMSO solution of the
substance and incubating in a 96 half-well microtiter plate at room
temperature for 15 min. The enzyme reaction was started by adding
10 μL of TAFIa developer (prediluted 1:2 with assay buffer). The
time course of the reaction was followed at 420 nm in a microtiter
plate reader (SpectraMax plus 384; Molecular Devices) for 15 min.
The IC50 value was calculated from the averaged values (duplicate
determination) of serial dilutions of the substance with the aid of the
Softmax Pro software (version 4.8; Molecular Devices). Reported
values are mean ( SEM) IC50 from n = 2 values within a single
experiment.
Algorithmic Details of the Cyclizer. The cyclization method
proceeds from a complexed ligand to a series of cyclized ligands,
which have been geometrically screened. There are two “modes” for
the workflow: “small molecule” and “peptide”. Though there are many
similarities, some key differences made splitting of the two modes
more efficient. Either mode will attempt to construct geometrically
stable and chemically reasonable linkers to span the cyclization length
in a manner that does not negatively affect the complex structure (e.g.,
by introducing strain, clashes, or breaking good interactions). For
small molecules, there are typically fewer potential attachment points,
and a series of “spacers” to span the cyclization length, which can be
pieced together to fill the space. For peptides, the attachment points
are fixed but numerous (e.g., all side chains), and the linkers tend to
be of fixed length (e.g., known side-chain bridges). Figure 7 shows a
breakdown of the two modes.
For specific cases, it is possible to “mix” the peptide and small-
molecule approach. In a small-molecule scenario, for example, the
user can specify an explicit list of linkers (e.g., based on a list of
available chemical reagents) that will not be enumerated and pieced
together, but will be directly mapped onto the user-specified
attachment points of the ligand. For a peptide, on the other hand,
it is possible to evaluate other attachment points than the Cα−Cβ or
Cα−Hα vector, for example, for the introduction of head-to-side
chain, side chain-to-tail, and head-to-tail cyclizations or for a
rescaffolding of the backbone of a peptide within a macrocycle.
Detailed Description of the Workflow. Input. The user must
specify a complexed structure for cyclization, and it may be required
to identify the ligand (this is sometimes nontrivial for peptide
complexes as peptide ligands are chemically similar to proteins). The
user must specify a mode; for the small molecule, mode must specify
chemical linkers using SMILES patterns, and for the peptide, mode
must specify side-chain bridges either by name or by SMILES. All of
these inputs are sanity-checked to confirm no obvious problems.
Limb Recognition. For peptide cyclizations, typically all side-chain
Cα−Cβ and Cα−Hα vectors are considered as potential limbs. This
can be further controlled by specifying a subregion to target using
Schrodinger’s atom selection language (ASL). This may be especially
useful if there is a larger peptide system and there is some knowledge
of the system that would preclude some regions from cyclization. For
small molecules, users specify criteria for identifying potential branch
points (such as ASL, SMARTS, or manually selecting limbs for
cyclization).
Limb Pair Filtering. All combinations of limb pairs are enumerated
and filtered hierarchically to eliminate unphysical constructions. The
pairs are filtered if they either have too few bonds between them or
too few bonds per distance143 spanned to eliminate inefficient
cyclization. Then, the linker is conformationally sampled. Sub-
sequently, using the conformations, the remaining limbs are filtered
whether they match the limb distance and orientations seen in the
conformational ensemble.
Attach Linker Precursors. For small molecules, the entire
constructed linker is attached to one of the limbs. For peptides,
each limb side chain is mutated to either an L- or D-alanine (according
to the specified chirality), then portions of the linker are attached to
either side chain, leaving “dangling” side chains.
Close Linker. The dangling linkers are then closed by slowly
bringing the closing bond atoms closer to each other and finally
forming the bond. This process can be time-consuming as the bond
closing often runs into steric or strain issues.
Post Minimize and Filter. Finally, the system is minimized and
strain filters are applied to assure no highly strained linkers have been
built. The cyclized molecules are then output for further analysis.
Implementation. The macrocyclization script has been imple-
mented via command line in Schrodinger software as of release 2020-
1. To run, use $SCHRODINGER/run -FROM psp macrocyclize -h.
Molecular Dynamics Simulations. MD simulations were carried
out using an explicit solvent MD package, Desmond program (version
4.7, Desmond Molecular Dynamics System; D. E. Shaw Research,
New York, NY and version 3.1, Maestro-Desmond Interoperability
̈
Tools; Schrodinger) with inbuilt optimized potentials for liquid
simulation (OPLS 2.1) force field.144−146 The proteins were prepared
for simulation by first checking their correctness using the protein
preparation wizard tool and Epik module was used for deriving the
protonation states of the proteins at neutral pH. The system was
prepared by placing the proteins in a cubic box with periodic
boundary conditions specifying the shape and size of box as 10 Å × 10
Å × 10 Å distance. Predefined TIP3P water model was used as a
solvent, and the systems were neutralized by adding an appropriate
number of ions. The solvated systems were relaxed by implementing
steepest descent and the limited-memory Broyden−Fletcher−Gold-
farb−Shanno algorithms in a hybrid manner. The simulation was
performed under the NPT ensemble for 100 ns implementing the
Berendsen thermostat and barostat methods. The Nose−Hoover
thermostat algorithm147,148 was used to maintain a constant
temperature, and the Martyna−Tobias−Klein barostat algorithm149
was employed for maintaining 1 atm of pressure throughout the
simulation, respectively. The short-range coulombic interactions were
analyzed using the short-range method with a cutoff value of 9.0 Å.
The particle mesh Ewald (PME) method150 was used for treating the
long-range electrostatic interactions. All bonds involving hydrogen
atoms were constrained using the SHAKE algorithm.151
Enthalpic Scoring. For small molecules, binding poses of cyclized
molecules were predicted with Glide using integrating macrocycle ring
sampling.111 Enthalpic scoring of the protein−ligand complex was
done using the empirical scoring function GlideScore.118,119 For larger
systems, the cyclized molecules were minimized in the protein
environment, and the molecular mechanics-generalized Born surface
area (MM/GBSA) method120,121 was used as enthalpic score.
Conformational Scoring. Conformational scoring was per-
̈
formed using Schrodinger’s macrocycle conformational stability
script.107 Briefly, the script scores compounds on their propensity
to maintain the specified bioactive substructure given a sampled
ensemble. Here, macrocyclizations of small molecules were scored as
described in the paper, using Prime-MCS to generate a conforma-
tional ensemble, then using the stability script to estimate the
expected RMSD in the bioactive substructure using Boltzmann
weighting at 298.13 K. For peptide and protein systems, simple NPT
MD was run for 100 ns. The stability script here does not need to
Boltzmann weight since the conformers already obey the NPT
distribution. The bioactive substructure was defined uniquely for each
system as is recommended and the conformational score is provided
as average RMSD over the Cα atoms of the bioactive region.
Small-Molecule Linker Construction. To limit the combinatorics
of permuting linkers, a minimum and maximum linker length is
calculated using the recognized limbs. Then, linkers are enumerated
using combinations of the specified spacers that could potentially span
the distance within the range. These constructed linkers are then
filtered using simple chemical rules to avoid unchemical linkers (e.g.,
avoid −OO− or −NN− linkages).
K
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
ubiquitin ligase; Elo, elongin; FBS, fetal bovine serum; FEP,
free-energy perturbation; Fmoc, fluorenylmethoxycarbonyl;
GLP-1, glucagon-like peptide-1; GLP-1R, glucagon-like pep-
tide-1 receptor; HBSS, Hank’s balanced salt solution; HBTU,
O-(benzotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluoro-
phosphate; HEK, human embryonic kidney; HEPES, 4-(2-
hydroxyethyl)-1-piperazineethanesulfonic acid; HPLC, high-
performance liquid chromatography; HSA, human serum
albumin; HTRF, homogeneous time-resolved fluorescence;
IBMX, isobutylmethylxanthine; LC-MS, liquid chromatogra-
phy−mass spectrometry; KIX domain, kinase-inducible
domain (KID) interacting domain; MD, molecular dynamics;
Mes, mesitylene; MM/GBSA, molecular mechanics-general-
ized Born surface area; MW, molecular weight; mXyl, meta-
Xylyl; NMR, nuclear magnetic resonance; PEG, poly(ethylene
glycol); Prime-MCS, Prime macrocycle conformational sam-
pling; PROTACs, proteolysis targeting chimeras; RCM, ring-
closing metathesis; Ro5, rule-of-five; RMSD, root-mean-square
deviation; SAR, structure−activity relationship; SEM, standard
error of the mean; SMARTS, SMILES arbitrary target
specification; SMILES, simplified molecular-input line-entry
system; SrtA, Staphylococcus aureus sortase A; TAFI, thrombin-
activatable fibrinolysis inhibitor; TBMB, 1,3,5-tris-
(bromomethyl)benzene; TFA, trifluoroacetic acid; THF,
tetrahydrofuran; UPLC, ultraperformance liquid chromatog-
raphy; VHL, von Hippel−Lindau protein
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c01500.
■
sı
Analytical data with RP-HPLC retention times and
molecular masses of the synthesized peptides; IUPAC
names of the synthesized peptides; detailed synthesis
procedures and analytical data for TAFIa inhibitors; two
additional retrospective case studies: a peptide example
(hydrocarbon-stapled peptides targeting β-catenin) and
a small molecule example (macrocyclic factor XIa
inhibitors) (PDF)
Molecular formula strings (CSV)
AUTHOR INFORMATION
Corresponding Authors
■
Dan Sindhikara − Schrodinger, Inc., New York, New York
10036, United States; Phone: +1 212 548 2371;
Email: Dan.Sindhikara@schrodinger.com
Andreas Evers − Integrated Drug Discovery, Sanofi-Aventis
Deutschland GmbH, 65926 Frankfurt am Main, Germany;
orcid.org/0000-0003-4643-1941; Email: Andreas.Evers@
merckgroup.com
Authors
Michael Wagner − Integrated Drug Discovery, Sanofi-Aventis
Deutschland GmbH, 65926 Frankfurt am Main, Germany
Paraskevi Gkeka − Integrated Drug Discovery, Sanofi R&D,
91385 Chilly-Mazarin, France; orcid.org/0000-0002-
0752-3539
REFERENCES
■
(1) Dougherty, P. G.; Qian, Z.; Pei, D. Macrocycles as Protein-
Protein Interaction Inhibitors. Biochem. J. 2017, 474, 1109−1125.
(2) Yudin, A. K. Macrocycles: Lessons from the Distant Past, Recent
Developments, and Future Directions. Chem. Sci. 2015, 6, 30−49.
(3) Zaretsky, S.; Yudin, A. K. Recent Advances in the Synthesis of
Cyclic Pseudopeptides. Drug Discovery Today Technol. 2017, 26, 3−
10.
Stefan Gussregen − Integrated Drug Discovery, Sanofi-Aventis
̈
Deutschland GmbH, 65926 Frankfurt am Main, Germany;
orcid.org/0000-0002-1868-9614
Garima Tiwari − Integrated Drug Discovery, Sanofi-Aventis
Deutschland GmbH, 65926 Frankfurt am Main, Germany
Gerhard Hessler − Integrated Drug Discovery, Sanofi-Aventis
Deutschland GmbH, 65926 Frankfurt am Main, Germany
Engin Yapici − Schrodinger, Inc., New York, New York 10036,
United States
(4) Mabonga, L.; Kappo, A. P. Protein-Protein Interaction
Modulators: Advances, Successes and Remaining Challenges. Biophys.
Rev. 2019, 11, 559−581.
(5) Abdalla, M. A.; McGaw, L. J. Natural Cyclic Peptides as an
Attractive Modality for Therapeutics: A Mini Review. Molecules 2018,
23, No. 2080.
Ziyu Li − Integrated Drug Discovery, Sanofi-Aventis Deutschland
GmbH, 65926 Frankfurt am Main, Germany
(6) Ali, A. M.; Atmaj, J.; Van Oosterwijk, N.; Groves, M. R.;
̈
Domling, A. Stapled Peptides Inhibitors: A New Window for Target
Complete contact information is available at:
Drug Discovery. Comput. Struct. Biotechnol. J. 2019, 17, 263−281.
(7) Giordanetto, F.; Kihlberg, J. Macrocyclic Drugs and Clinical
Candidates: What Can Medicinal Chemists Learn from Their
Properties? J. Med. Chem. 2014, 57, 278−295.
(8) Lee, A. C.-L.; Harris, J. L.; Khanna, K. K.; Hong, J.-H. A
Comprehensive Review on Current Advances in Peptide Drug
Development and Design. Int. J. Mol. Sci. 2019, 20, No. 2383.
(9) Naylor, M. R.; Bockus, A. T.; Blanco, M.-J.; Lokey, R. S. Cyclic
Peptide Natural Products Chart the Frontier of Oral Bioavailability in
the Pursuit of Undruggable Targets. Curr. Opin. Chem. Biol. 2017, 38,
141−147.
(10) Poongavanam, V.; Danelius, E.; Peintner, S.; Alcaraz, L.; Caron,
G.; Cummings, M. D.; Wlodek, S.; Erdelyi, M.; Hawkins, P. C. D.;
Ermondi, G.; Kihlberg, J. Conformational Sampling of Macrocyclic
Drugs in Different Environments: Can We Find the Relevant
Conformations? ACS Omega 2018, 3, 11742−11757.
https://pubs.acs.org/10.1021/acs.jmedchem.0c01500
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors thank R. Loder and R. Dharanipragada for peptide
■
̈
synthesis; H. Kohler and J. Czech for in vitro potency
characterization; and C. Poeverlein, E. Feyfant, K. Borrelli, T.
Day, and J. Knight for helpful discussions.
ABBREVIATIONS
■
3D, three-dimensional; alk, alkyl; ASL, atom selection lanuage;
BCL9, B-cell CLL/lymphoma 9; BD, bromodomain; BET,
bromodomain and extra-terminal domain; Brd4, bromodomain
4; cAMP, cyclic adenosine monophosphate; CREB, cAMP
response element-binding protein; DCM, dichloromethane;
DIPEA, N,N-diisopropylethylamine; DMEM, Dulbecco’s
modified Eagle’s medium; DMSO, dimethyl sulfoxide; dOal,
D-7-octenyl-alanine; DPP IV, dipeptidyl peptidase IV; E3, E3
(11) Qian, Z.; Dougherty, P. G.; Pei, D. Targeting Intracellular
Protein-Protein Interactions with Cell-Permeable Cyclic Peptides.
Curr. Opin. Chem. Biol. 2017, 38, 80−86.
(12) Robertson, N. S.; Spring, D. R. Using Peptidomimetics and
Constrained Peptides as Valuable Tools for Inhibiting Protein−
Protein Interactions. Molecules 2018, 23 (4), 959−976.
L
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
(13) Rubin, S.; Qvit, N. Cyclic Peptides for Protein-Protein
Interaction Targets: Applications to Human Disease. Crit. Rev.
Eukaryotic Gene Expression 2016, 26, 199−221.
(37) Whitty, A.; Viarengo, L. A.; Zhong, M. Progress towards the
Broad Use of Non-Peptide Synthetic Macrocycles in Drug Discovery.
Org. Biomol. Chem. 2017, 15, 7729−7735.
(38) Dharanipragada, R. New Modalities in Conformationally
Constrained Peptides for Potency, Selectivity and Cell Permeation.
Future Med. Chem. 2013, 5, 831−849.
(14) Russo, A.; Aiello, C.; Grieco, P.; Marasco, D. Targeting
“Undruggable” Proteins: Design of Synthetic Cyclopeptides. Curr.
Med. Chem. 2016, 23, 748−762.
(39) Lau, Y. H.; de Andrade, P.; Wu, Y.; Spring, D. R. Peptide
Stapling Techniques Based on Different Macrocyclisation Chem-
istries. Chem. Soc. Rev. 2015, 44, 91−102.
(15) Stone, T. A.; Deber, C. M. Therapeutic Design of Peptide
Modulators of Protein-Protein Interactions in Membranes. Biochim.
Biophys. Acta, Biomembr. 2017, 1859, 577−585.
(40) Ngambenjawong, C.; Pineda, J. M. B.; Pun, S. H. Engineering
an Affinity-Enhanced Peptide through Optimization of Cyclization
Chemistry. Bioconjugate Chem. 2016, 27, 2854−2862.
́
(16) Wojcik, P.; Berlicki, Ł. Peptide-Based Inhibitors of Protein-
Protein Interactions. Bioorg. Med. Chem. Lett. 2016, 26, 707−713.
(17) Yamagishi, Y.; Shoji, I.; Miyagawa, S.; Kawakami, T.; Katoh, T.;
Goto, Y.; Suga, H. Natural Product-like Macrocyclic N-Methyl-
Peptide Inhibitors against a Ubiquitin Ligase Uncovered from a
Ribosome-Expressed de Novo Library. Chem. Biol. 2011, 18, 1562−
1570.
(41) Skowron, K. J.; Speltz, T. E.; Moore, T. W. Recent Structural
Advances in Constrained Helical Peptides. Med. Res. Rev. 2019, 39,
749−770.
(42) Tang, J.; He, Y.; Chen, H.; Sheng, W.; Wang, H. Synthesis of
Bioactive and Stabilized Cyclic Peptides by Macrocyclization Using
C(Sp3)-H Activation. Chem. Sci. 2017, 8, 4565−4570.
(43) White, C. J.; Yudin, A. K. Contemporary Strategies for Peptide
Macrocyclization. Nat. Chem. 2011, 3, 509−524.
(18) Zorzi, A.; Deyle, K.; Heinis, C. Cyclic Peptide Therapeutics:
Past, Present and Future. Curr. Opin. Chem. Biol. 2017, 38, 24−29.
(19) Ermert, P. Design, Properties and Recent Application of
Macrocycles in Medicinal Chemistry. Chimia 2017, 71, 678−702.
(20) Ermert, P.; Luther, A.; Zbinden, P.; Obrecht, D. Frontier
Between Cyclic Peptides and Macrocycles. Methods Mol. Biol. 2019,
2001, 147−202.
(21) Lau, J. L.; Dunn, M. K. Therapeutic Peptides: Historical
Perspectives, Current Development Trends, and Future Directions.
Bioorg. Med. Chem. 2018, 26, 2700−2707.
(22) Hruby, V. J. Peptide Science: Exploring the Use of Chemical
Principles and Interdisciplinary Collaboration for Understanding Life
Processes. J. Med. Chem. 2003, 46, 4215−4231.
(23) Hruby, V. J. Designing Peptide Receptor Agonists and
Antagonists. Nat. Rev. Drug Discovery 2002, 1, 847−858.
(24) Kessler, H. Conformation and Biological Activity of Cyclic
Peptides. Angew. Chem., Int. Ed. 1982, 21, 512−523.
(25) Purkayastha, A.; Kang, T. J. Stabilization of Proteins by
Covalent Cyclization. Biotechnol. Bioprocess Eng. 2019, 24, 702−712.
(26) Chu, Q.; Moellering, R. E.; Hilinski, G. J.; Kim, Y.-W.;
Grossmann, T. N.; Yeh, J. T.-H.; Verdine, G. L. Towards
Understanding Cell Penetration by Stapled Peptides. Med. Chem.
Commun. 2015, 6, 111−119.
(27) Dougherty, P. G.; Sahni, A.; Pei, D. Understanding Cell
Penetration of Cyclic Peptides. Chem. Rev. 2019, 119, 10241−10287.
(28) Drucker, D. J. Advances in Oral Peptide Therapeutics. Nat. Rev.
Drug Discovery 2019, 19, 277−289.
(44) Felix, A. M.; Heimer, E. P.; Wang, C. T.; Lambros, T. J.;
Fournier, A.; Mowles, T. F.; Maines, S.; Campbell, R. M.; Wegrzynski,
B. B.; Toome, V.; et al. Synthesis, Biological Activity and
Conformational Analysis of Cyclic GRF Analogs. Int. J. Pept. Protein
Res. 1988, 32, 441−454.
(45) Taylor, J. W. The Synthesis and Study of Side-Chain Lactam-
Bridged Peptides. Biopolymers 2002, 66, 49−75.
(46) Blackwell, H. E.; Grubbs, R. H. Highly Efficient Synthesis of
Covalently Cross-Linked Peptide Helices by Ring-Closing Metathesis.
Angew. Chem., Int. Ed. 1998, 37, 3281−3284.
(47) Kim, Y.-W.; Grossmann, T. N.; Verdine, G. L. Synthesis of All-
Hydrocarbon Stapled α-Helical Peptides by Ring-Closing Olefin
Metathesis. Nat. Protoc. 2011, 6, 761−771.
(48) Verdine, G. L.; Hilinski, G. J. All-Hydrocarbon Stapled Peptides
as Synthetic Cell-Accessible Mini-Proteins. Drug Discovery Today
Technol. 2012, 9, e41−e47.
(49) Bernal, F.; Tyler, A. F.; Korsmeyer, S. J.; Walensky, L. D.;
Verdine, G. L. Reactivation of the P53 Tumor Suppressor Pathway by
a Stapled P53 Peptide. J. Am. Chem. Soc. 2007, 129, 2456−2457.
(50) Bird, G. H.; Madani, N.; Perry, A. F.; Princiotto, A. M.; Supko,
J. G.; He, X.; Gavathiotis, E.; Sodroski, J. G.; Walensky, L. D.
Hydrocarbon Double-Stapling Remedies the Proteolytic Instability of
a Lengthy Peptide Therapeutic. Proc. Natl. Acad. Sci. U.S.A. 2010, 107,
14093−14098.
(29) Jing, X.; Jin, K. A Gold Mine for Drug Discovery: Strategies to
Develop Cyclic Peptides into Therapies. Med. Res. Rev. 2019, 40,
753−810.
(30) Nielsen, D. S.; Shepherd, N. E.; Xu, W.; Lucke, A. J.; Stoermer,
M. J.; Fairlie, D. P. Orally Absorbed Cyclic Peptides. Chem. Rev. 2017,
117, 8094−8128.
(31) Nielsen, D. S.; Lohman, R.-J.; Hoang, H. N.; Hill, T. A.; Jones,
A.; Lucke, A. J.; Fairlie, D. P. Flexibility versus Rigidity for Orally
Bioavailable Cyclic Hexapeptides. ChemBioChem 2015, 16, 2289−
2293.
(51) Danial, N. N.; Walensky, L. D.; Zhang, C.-Y.; Choi, C. S.;
Fisher, J. K.; Molina, A. J. A.; Datta, S. R.; Pitter, K. L.; Bird, G. H.;
Wikstrom, J. D.; Deeney, J. T.; Robertson, K.; Morash, J.; Kulkarni,
A.; Neschen, S.; Kim, S.; Greenberg, M. E.; Corkey, B. E.; Shirihai, O.
S.; Shulman, G. I.; Lowell, B. B.; Korsmeyer, S. J. Dual Role of
Proapoptotic BAD in Insulin Secretion and Beta Cell Survival. Nat.
Med. 2008, 14, 144−153.
́
(52) Lawrence, J.; Jourdan, M.; Vallee, Y.; Blandin, V. Peptide
Cyclization via Ring-Closing Metathesis: The N-Alkenoxy Peptide
Approach. Org. Biomol. Chem. 2008, 6, 4575−4581.
̈
̈
(32) Rader, A. F. B.; Weinmuller, M.; Reichart, F.; Schumacher-
Klinger, A.; Merzbach, S.; Gilon, C.; Hoffman, A.; Kessler, H. Orally
Active Peptides: Is There a Magic Bullet? Angew. Chem., Int. Ed. 2018,
57, 14414−14438.
(53) Sviridov, D.; Ikpot, I.; Stonik, J.; Drake, S.; Amar, M.; Osei-
Hwedieh, D.; Piszczek, G.; Turner, S.; Remaley, A. Helix Stabilization
of Amphipathic Peptides by Hydrocarbon Stapling Increases
Cholesterol Efflux by the ABCA1 Transporter. Biochem. Biophys.
Res. Commun. 2011, 410, 446−451.
(33) Villar, E. A.; Beglov, D.; Chennamadhavuni, S.; Porco, J. A.;
Kozakov, D.; Vajda, S.; Whitty, A. How Proteins Bind Macrocycles.
Nat. Chem. Biol. 2014, 10, 723−731.
(34) Vinogradov, A. A.; Yin, Y.; Suga, H. Macrocyclic Peptides as
Drug Candidates: Recent Progress and Remaining Challenges. J. Am.
Chem. Soc. 2019, 141, 4167−4181.
(54) Walensky, L. D.; Kung, A. L.; Escher, I.; Malia, T. J.; Barbuto,
S.; Wright, R. D.; Wagner, G.; Verdine, G. L.; Korsmeyer, S. J.
Activation of Apoptosis in Vivo by a Hydrocarbon-Stapled BH3 Helix.
Science 2004, 305, 1466−1470.
(55) Zhang, H.; Zhao, Q.; Bhattacharya, S.; Waheed, A. A.; Tong,
X.; Hong, A.; Heck, S.; Curreli, F.; Goger, M.; Cowburn, D.; Freed, E.
O.; Debnath, A. K. A Cell-Penetrating Helical Peptide as a Potential
HIV-1 Inhibitor. J. Mol. Biol. 2008, 378, 565−580.
(56) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A
Stepwise Huisgen Cycloaddition Process: Copper(I)-Catalyzed
(35) Driggers, E. M.; Hale, S. P.; Lee, J.; Terrett, N. K. The
Exploration of Macrocycles for Drug Discovery–an Underexploited
Structural Class. Nat. Rev. Drug Discovery 2008, 7, 608−624.
(36) Mallinson, J.; Collins, I. Macrocycles in New Drug Discovery.
Future Med. Chem. 2012, 4, 1409−1438.
M
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
Regioselective “Ligation” of Azides and Terminal Alkynes. Angew.
Chem., Int. Ed. 2002, 41, 2596−2599.
(57) Tornøe, C. W.; Christensen, C.; Meldal, M. Peptidotriazoles on
Solid Phase: [1,2,3]-Triazoles by Regiospecific Copper(I)-Catalyzed
1,3-Dipolar Cycloadditions of Terminal Alkynes to Azides. J. Org.
Chem. 2002, 67, 3057−3064.
(58) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Click Chemistry:
Diverse Chemical Function from a Few Good Reactions. Angew.
Chem., Int. Ed. 2001, 40, 2004−2021.
(59) Cantel, S.; Isaad, A. L. C.; Scrima, M.; Levy, J. J.; DiMarchi, R.
D.; Rovero, P.; Halperin, J. A.; D’Ursi, A. M.; Papini, A. M.; Chorev,
M. Synthesis and Conformational Analysis of a Cyclic Peptide
Obtained via i to I+4 Intramolecular Side-Chain to Side-Chain Azide-
Alkyne 1,3-Dipolar Cycloaddition. J. Org. Chem. 2008, 73, 5663−
5674.
(60) Scrima, M.; Chevalier-Isaad, A. L.; Rovero, P.; Papini, A. M.;
Chorev, M.; D’Ursi, A. M. CuI-Catalyzed Azide−Alkyne Intra-
molecular i-to-(I+4) Side-Chain-to-Side-Chain Cyclization Promotes
the Formation of Helix-Like Secondary Structures. Eur. J. Org. Chem.
2010, 2010, 446−457.
Backbone Conformations in Bicyclic Peptides. ChemBioChem 2012,
13, 1032−1038.
(75) Timmerman, P.; Beld, J.; Puijk, W. C.; Meloen, R. H. Rapid and
Quantitative Cyclization of Multiple Peptide Loops onto Synthetic
Scaffolds for Structural Mimicry of Protein Surfaces. ChemBioChem
2005, 6, 821−824.
(76) Diderich, P.; Bertoldo, D.; Dessen, P.; Khan, M. M.; Pizzitola,
I.; Held, W.; Huelsken, J.; Heinis, C. Phage Selection of Chemically
Stabilized α-Helical Peptide Ligands. ACS Chem. Biol. 2016, 11,
1422−1427.
(77) Kale, S. S.; Villequey, C.; Kong, X.-D.; Zorzi, A.; Deyle, K.;
Heinis, C. Cyclization of Peptides with Two Chemical Bridges Affords
Large Scaffold Diversities. Nat. Chem. 2018, 10, 715−723.
(78) Kong, X.-D.; Moriya, J.; Carle, V.; Pojer, F.; Abriata, L. A.;
Deyle, K.; Heinis, C. De Novo Development of Proteolytically
Resistant Therapeutic Peptides for Oral Administration. Nat. Biomed.
Eng. 2020, 4, 560−571.
(79) Baeriswyl, V.; Heinis, C. Phage Selection of Cyclic Peptide
Antagonists with Increased Stability toward Intestinal Proteases.
Protein Eng., Des. Sel. 2013, 26, 81−89.
(61) Kawamoto, S. A.; Coleska, A.; Ran, X.; Yi, H.; Yang, C.-Y.;
Wang, S. Design of Triazole-Stapled BCL9 α-Helical Peptides to
Target the β-Catenin/B-Cell CLL/Lymphoma 9 (BCL9) Protein-
Protein Interaction. J. Med. Chem. 2012, 55, 1137−1146.
(62) Madden, M. M.; Muppidi, A.; Li, Z.; Li, X.; Chen, J.; Lin, Q.
Synthesis of Cell-Permeable Stapled Peptide Dual Inhibitors of the
P53-Mdm2/Mdmx Interactions via Photoinduced Cycloaddition.
Bioorg. Med. Chem. Lett. 2011, 21, 1472−1475.
(63) Tian, Y.; Li, J.; Zhao, H.; Zeng, X.; Wang, D.; Liu, Q.; Niu, X.;
Huang, X.; Xu, N.; Li, Z. Stapling of Unprotected Helical Peptides via
Photo-Induced Intramolecular Thiol-Yne Hydrothiolation. Chem. Sci.
2016, 7, 3325−3330.
(80) Bernhagen, D.; Jungbluth, V.; Quilis, N. G.; Dostalek, J.; White,
P. B.; Jalink, K.; Timmerman, P. Bicyclic RGD Peptides with
Exquisite Selectivity for the Integrin Avβ3 Receptor Using a “Random
Design” Approach. ACS Comb. Sci. 2019, 21, 198−206.
(81) Heinis, C.; Rutherford, T.; Freund, S.; Winter, G. Phage-
Encoded Combinatorial Chemical Libraries Based on Bicyclic
Peptides. Nat. Chem. Biol. 2009, 5, 502−507.
(82) Dang, B.; Wu, H.; Mulligan, V. K.; Mravic, M.; Wu, Y.;
Lemmin, T.; Ford, A.; Silva, D.-A.; Baker, D.; DeGrado, W. F. De
Novo Design of Covalently Constrained Mesosize Protein Scaffolds
with Unique Tertiary Structures. Proc. Natl. Acad. Sci. U.S.A. 2017,
114, 10852−10857.
(64) Wang, Y.; Bruno, B. J.; Cornillie, S.; Nogieira, J. M.; Chen, D.;
Cheatham, T. E.; Lim, C. S.; Chou, D. H.-C. Application of Thiol-
Yne/Thiol-Ene Reactions for Peptide and Protein Macrocyclizations.
Chem. −Eur. J. 2017, 23, 7087−7092.
(83) Neubacher, S.; Saya, J. M.; Amore, A.; Grossmann, T. N. In
Situ Cyclization of Proteins (INCYPRO): Cross-Link Derivatization
Modulates Protein Stability. J. Org. Chem. 2019, 85, 14761476−1483.
(84) Pelay-Gimeno, M.; Bange, T.; Hennig, S.; Grossmann, T. N. In
Situ Cyclization of Native Proteins: Structure-Based Design of a
Bicyclic Enzyme. Angew. Chem., Int. Ed. 2018, 57, 11164−11170.
(85) Blackwell, H. E.; Sadowsky, J. D.; Howard, R. J.; Sampson, J.
N.; Chao, J. A.; Steinmetz, W. E.; O’Leary, D. J.; Grubbs, R. H. Ring-
Closing Metathesis of Olefinic Peptides: Design, Synthesis, and
Structural Characterization of Macrocyclic Helical Peptides. J. Org.
Chem. 2001, 66, 5291−5302.
(65) Wang, Y.; Chou, D. H.-C. A Thiol-Ene Coupling Approach to
Native Peptide Stapling and Macrocyclization. Angew. Chem., Int. Ed.
2015, 54, 10931−10934.
(66) Zhao, B.; Zhang, Q.; Li, Z. Constructing Thioether-Tethered
Cyclic Peptides via on-Resin Intra-Molecular Thiol-Ene Reaction. J.
Pept. Sci. 2016, 22, 540−544.
(67) Fadzen, C. M.; Wolfe, J. M.; Cho, C.-F.; Chiocca, E. A.; Lawler,
S. E.; Pentelute, B. L. Perfluoroarene-Based Peptide Macrocycles to
Enhance Penetration Across the Blood-Brain Barrier. J. Am. Chem.
Soc. 2017, 139, 15628−15631.
(68) Jo, H.; Meinhardt, N.; Wu, Y.; Kulkarni, S.; Hu, X.; Low, K. E.;
Davies, P. L.; DeGrado, W. F.; Greenbaum, D. C. Development of α-
Helical Calpain Probes by Mimicking a Natural Protein-Protein
Interaction. J. Am. Chem. Soc. 2012, 134, 17704−17713.
(69) Peraro, L.; Siegert, T. R.; Kritzer, J. A. Conformational
Restriction of Peptides Using Dithiol Bis-Alkylation. Methods
Enzymol. 2016, 580, 303−332.
(70) Rojas, A. J.; Pentelute, B. L.; Buchwald, S. L. Water-Soluble
Palladium Reagents for Cysteine S-Arylation under Ambient Aqueous
Conditions. Org. Lett. 2017, 19, 4263−4266.
(71) Rojas, A. J.; Zhang, C.; Vinogradova, E. V.; Buchwald, N. H.;
Reilly, J.; Pentelute, B. L.; Buchwald, S. L. Divergent Unprotected
Peptide Macrocyclisation by Palladium-Mediated Cysteine Arylation.
Chem. Sci. 2017, 8, 4257−4263.
(72) Spokoyny, A. M.; Zou, Y.; Ling, J. J.; Yu, H.; Lin, Y.-S.;
Pentelute, B. L. A Perfluoroaryl-Cysteine S(N)Ar Chemistry
Approach to Unprotected Peptide Stapling. J. Am. Chem. Soc. 2013,
135, 5946−5949.
(86) Galande, A. K.; Bramlett, K. S.; Trent, J. O.; Burris, T. P.;
Wittliff, J. L.; Spatola, A. F. Potent Inhibitors of LXXLL-Based
Protein-Protein Interactions. ChemBioChem 2005, 6, 1991−1998.
́
(87) Harrison, R. S.; Shepherd, N. E.; Hoang, H. N.; Ruiz-Gomez,
G.; Hill, T. A.; Driver, R. W.; Desai, V. S.; Young, P. R.; Abbenante,
G.; Fairlie, D. P. Downsizing Human, Bacterial, and Viral Proteins to
Short Water-Stable Alpha Helices That Maintain Biological Potency.
Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 11686−11691.
(88) Jackson, D. Y.; King, D. S.; Chmielewski, J.; Singh, S.; Schultz,
P. G. General Approach to the Synthesis of Short .Alpha.-Helical
Peptides. J. Am. Chem. Soc. 1991, 113, 9391−9392.
(89) Leduc, A.-M.; Trent, J. O.; Wittliff, J. L.; Bramlett, K. S.; Briggs,
S. L.; Chirgadze, N. Y.; Wang, Y.; Burris, T. P.; Spatola, A. F. Helix-
Stabilized Cyclic Peptides as Selective Inhibitors of Steroid Receptor-
Coactivator Interactions. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
11273−11278.
(90) Phelan, J. C.; Skelton, N. J.; Braisted, A. C.; McDowell, R. S. A
General Method for Constraining Short Peptides to an α-Helical
Conformation. J. Am. Chem. Soc. 1997, 119, 455−460.
(91) Schafmeister, C. E.; Po, J.; Verdine, G. L. An All-Hydrocarbon
Cross-Linking System for Enhancing the Helicity and Metabolic
Stability of Peptides. J. Am. Chem. Soc. 2000, 122, 5891−5892.
(92) Pelay-Gimeno, M.; Glas, A.; Koch, O.; Grossmann, T. N.
Structure-Based Design of Inhibitors of Protein−Protein Interactions:
Mimicking Peptide Binding Epitopes. Angew. Chem., Int. Ed. 2015, 54,
8896−8927.
(73) Vinogradova, E. V.; Zhang, C.; Spokoyny, A. M.; Pentelute, B.
L.; Buchwald, S. L. Organometallic Palladium Reagents for Cysteine
Bioconjugation. Nature 2015, 526, 687−691.
(74) Chen, S.; Morales-Sanfrutos, J.; Angelini, A.; Cutting, B.;
Heinis, C. Structurally Diverse Cyclisation Linkers Impose Different
N
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
(93) Bhardwaj, G.; Mulligan, V. K.; Bahl, C. D.; Gilmore, J. M.;
Harvey, P. J.; Cheneval, O.; Buchko, G. W.; Pulavarti, S. V. S. R. K.;
Kaas, Q.; Eletsky, A.; Huang, P.-S.; Johnsen, W. A.; Greisen, P.;
Rocklin, G. J.; Song, Y.; Linsky, T. W.; Watkins, A.; Rettie, S. A.; Xu,
X.; Carter, L. P.; Bonneau, R.; Olson, J. M.; Coutsias, E.; Correnti, C.
E.; Szyperski, T.; Craik, D. J.; Baker, D. Accurate de Novo Design of
Hyperstable Constrained Peptides. Nature 2016, 538, 329−335.
(94) Est, C. B.; Mangrolia, P.; Murphy, R. M. ROSETTA-Informed
Design of Structurally Stabilized Cyclic Anti-Amyloid Peptides.
Protein Eng., Des. Sel. 2019, 32, 47−57.
(112) Martin, S. J.; Chen, I.-J.; Chan, A. W. E.; Foloppe, N.
Modelling the Binding Mode of Macrocycles: Docking and
Conformational Sampling. Bioorg. Med. Chem. 2020, 28, No. 115143.
̈
̈
(113) Glide Schrodinger Release 2019-3 Glide; Schrodinger, LLC:
New York, NY, 2019 (accessed Jan 31, 2018).
(114) Ugur, I.; Schroft, M.; Marion, A.; Glaser, M.; Antes, I.
Predicting the Bioactive Conformations of Macrocycles: A Molecular
Dynamics-Based Docking Procedure with DynaDock. J. Mol. Model.
2019, 25, No. 197.
(115) Wagner, V.; Jantz, L.; Briem, H.; Sommer, K.; Rarey, M.;
Christ, C. D. Computational Macrocyclization: From de Novo
Macrocycle Generation to Binding Affinity Estimation. ChemMed-
Chem 2017, 12, 1866−1872.
(116) Yu, H. S.; Deng, Y.; Wu, Y.; Sindhikara, D.; Rask, A. R.;
Kimura, T.; Abel, R.; Wang, L. Accurate and Reliable Prediction of
the Binding Affinities of Macrocycles to Their Protein Targets. J.
Chem. Theory Comput. 2017, 13, 6290−6300.
(117) Wallraven, K.; Holmelin, F. L.; Glas, A.; Hennig, S.; Frolov, A.
I.; Grossmann, T. N. Adapting Free Energy Perturbation Simulations
for Large Macrocyclic Ligands: How to Dissect Contributions from
Direct Binding and Free Ligand Flexibility. Chem. Sci. 2020, 11,
2269−2276.
(118) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: A New
Approach for Rapid, Accurate Docking and Scoring. 1. Method and
Assessment of Docking Accuracy. J. Med. Chem. 2004, 47, 1739−
1749.
(119) Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.;
Frye, L. L.; Pollard, W. T.; Banks, J. L. Glide: A New Approach for
Rapid, Accurate Docking and Scoring. 2. Enrichment Factors in
Database Screening. J. Med. Chem. 2004, 47, 1750−1759.
(120) Kollman, P. A.; Massova, I.; Reyes, C.; Kuhn, B.; Huo, S.;
Chong, L.; Lee, M.; Lee, T.; Duan, Y.; Wang, W.; Donini, O.; Cieplak,
P.; Srinivasan, J.; Case, D. A.; Cheatham, T. E. Calculating Structures
and Free Energies of Complex Molecules: Combining Molecular
Mechanics and Continuum Models. Acc. Chem. Res. 2000, 33, 889−
897.
(121) Srinivasan, J.; Miller, J.; Kollman, P. A.; Case, D. A.
Continuum Solvent Studies of the Stability of RNA Hairpin Loops
and Helices. J. Biomol. Struct. Dyn. 1998, 16, 671−682.
(122) Paulsen, J. L.; Yu, H. S.; Sindhikara, D.; Wang, L.; Appleby, T.
C.; Villasenor, A. G.; Schmitz, U.; Shivakumar, D. Evaluation of Free
̃
Energy Calculations for the Prioritization of Macrocycle Synthesis. J.
Chem. Inf. Model. 2020, 60, 3489−3498.
(123) Cappel, D.; Jerome, S.; Hessler, G.; Matter, H. Impact of
Different Automated Binding Pose Generation Approaches on
Relative Binding Free Energy Simulations. J. Chem. Inf. Model.
2020, 60, 1432−1444.
(124) Daylight Theory: SMARTSA Language for Describing
Molecular Patterns. https://www.daylight.com/dayhtml/doc/theory/
theory.smarts.html (accessed April 6, 2020).
(125) Evers, A.; Haack, T.; Lorenz, M.; Bossart, M.; Elvert, R.;
Henkel, B.; Stengelin, S.; Kurz, M.; Glien, M.; Dudda, A.; Lorenz, K.;
Kadereit, D.; Wagner, M. Design of Novel Exendin-Based Dual
Glucagon-like Peptide 1 (GLP-1)/Glucagon Receptor Agonists. J.
Med. Chem. 2017, 60, 4293−4303.
́
(95) Gueret, S. M.; Thavam, S.; Carbajo, R. J.; Potowski, M.;
́
Larsson, N.; Dahl, G.; Dellsen, A.; Grossmann, T. N.; Plowright, A.
T.; Valeur, E.; Lemurell, M.; Waldmann, H. Macrocyclic Modalities
Combining Peptide Epitopes and Natural Product Fragments. J. Am.
Chem. Soc. 2020, 142, 4904−4915.
(96) Mulligan, V. K. The Emerging Role of Computational Design
in Peptide Macrocycle Drug Discovery. Expert Opin. Drug Discovery
2020, 15, 833−852.
(97) Bonnet, P.; Agrafiotis, D. K.; Zhu, F.; Martin, E. Conforma-
tional Analysis of Macrocycles: Finding What Common Search
Methods Miss. J. Chem. Inf. Model. 2009, 49, 2242−2259.
(98) Chen, I.-J.; Foloppe, N. Tackling the Conformational Sampling
of Larger Flexible Compounds and Macrocycles in Pharmacology and
Drug Discovery. Bioorg. Med. Chem. 2013, 21, 7898−7920.
(99) Coutsias, E. A.; Lexa, K. W.; Wester, M. J.; Pollock, S. N.;
Jacobson, M. P. Exhaustive Conformational Sampling of Complex
Fused Ring Macrocycles Using Inverse Kinematics. J. Chem. Theory
Comput. 2016, 12, 4674−4687.
(100) Geng, H.; Jiang, F.; Wu, Y.-D. Accurate Structure Prediction
and Conformational Analysis of Cyclic Peptides with Residue-Specific
Force Fields. J. Phys. Chem. Lett. 2016, 7, 1805−1810.
(101) Gil-Ley, A.; Bussi, G. Enhanced Conformational Sampling
Using Replica Exchange with Collective-Variable Tempering. J. Chem.
Theory Comput. 2015, 11, 1077−1085.
(102) Hosseinzadeh, P.; Bhardwaj, G.; Mulligan, V. K.; Shortridge,
M. D.; Craven, T. W.; Pardo-Avila, F.; Rettie, S. A.; Kim, D. E.; Silva,
D.-A.; Ibrahim, Y. M.; Webb, I. K.; Cort, J. R.; Adkins, J. N.; Varani,
G.; Baker, D. Comprehensive Computational Design of Ordered
Peptide Macrocycles. Science 2017, 358, 1461−1466.
(103) Kamenik, A. S.; Lessel, U.; Fuchs, J. E.; Fox, T.; Liedl, K. R.
Peptidic Macrocycles - Conformational Sampling and Thermody-
namic Characterization. J. Chem. Inf. Model. 2018, 58, 982−992.
(104) Labute, P. LowModeMDImplicit Low-Mode Velocity
Filtering Applied to Conformational Search of Macrocycles and
Protein Loops. J. Chem. Inf. Model. 2010, 50, 792−800.
́
(105) Olanders, G.; Alogheli, H.; Brandt, P.; Karlen, A. Conforma-
tional Analysis of Macrocycles: Comparing General and Specialized
Methods. J. Comput.-Aided Mol. Des. 2020, 34, 231−252.
(106) Sindhikara, D.; Spronk, S. A.; Day, T.; Borrelli, K.; Cheney, D.
L.; Posy, S. L. Improving Accuracy, Diversity, and Speed with Prime
Macrocycle Conformational Sampling. J. Chem. Inf. Model. 2017, 57,
1881−1894.
(107) Sindhikara, D.; Borrelli, K. High Throughput Evaluation of
Macrocyclization Strategies for Conformer Stabilization. Sci. Rep.
2018, 8, No. 6585.
(108) Watts, K. S.; Dalal, P.; Tebben, A. J.; Cheney, D. L.; Shelley, J.
C. Macrocycle Conformational Sampling with MacroModel. J. Chem.
Inf. Model. 2014, 54, 2680−2696.
(126) Evers, A.; Bossart, M.; Pfeiffer-Marek, S.; Elvert, R.;
Schreuder, H.; Kurz, M.; Stengelin, S.; Lorenz, M.; Herling, A.;
Konkar, A.; Lukasczyk, U.; Pfenninger, A.; Lorenz, K.; Haack, T.;
Kadereit, D.; Wagner, M. Dual Glucagon-like Peptide 1 (GLP-1)/
Glucagon Receptor Agonists Specifically Optimized for Multidose
Formulations. J. Med. Chem. 2018, 61, 5580−5593.
(109) Witek, J.; Keller, B. G.; Blatter, M.; Meissner, A.; Wagner, T.;
Riniker, S. Kinetic Models of Cyclosporin A in Polar and Apolar
Environments Reveal Multiple Congruent Conformational States. J.
Chem. Inf. Model. 2016, 56, 1547−1562.
(110) Allen, S. E.; Dokholyan, N. V.; Bowers, A. A. Dynamic
Docking of Conformationally Constrained Macrocycles: Methods and
Applications. ACS Chem. Biol. 2016, 11, 10−24.
(127) Zhang, Y.; Sun, B.; Feng, D.; Hu, H.; Chu, M.; Qu, Q.;
Tarrasch, J. T.; Li, S.; Sun Kobilka, T.; Kobilka, B. K.; Skiniotis, G.
Cryo-EM Structure of the Activated GLP-1 Receptor in Complex
with a G Protein. Nature 2017, 546, 248−253.
́
(111) Alogheli, H.; Olanders, G.; Schaal, W.; Brandt, P.; Karlen, A.
Docking of Macrocycles: Comparing Rigid and Flexible Docking in
̈
̈
(128) Prime Schrodinger Release 2019-3 Prime; Schrodinger, LLC:
New York, NY, 2019 (accessed Jan 31, 2018).
Glide. J. Chem. Inf. Model. 2017, 57, 190−202.
O
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
pubs.acs.org/jmc
Article
(129) Day, J. W.; Ottaway, N.; Patterson, J. T.; Gelfanov, V.; Smiley,
D.; Gidda, J.; Findeisen, H.; Bruemmer, D.; Drucker, D. J.;
Chaudhary, N.; Holland, J.; Hembree, J.; Abplanalp, W.; Grant, E.;
Ruehl, J.; Wilson, H.; Kirchner, H.; Lockie, S. H.; Hofmann, S.;
Woods, S. C.; Nogueiras, R.; Pfluger, P. T.; Perez-Tilve, D.; DiMarchi,
(144) Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. Develop-
ment and Testing of the OPLS All-Atom Force Field on Conforma-
tional Energetics and Properties of Organic Liquids. J. Am. Chem. Soc.
1996, 118, 11225−11236.
(145) Jorgensen, W. L.; Tirado-Rives, J. The OPLS [Optimized
Potentials for Liquid Simulations] Potential Functions for Proteins,
Energy Minimizations for Crystals of Cyclic Peptides and Crambin. J.
Am. Chem. Soc. 1988, 110, 1657−1666.
̈
R.; Tschop, M. H. A New Glucagon and GLP-1 Co-Agonist
Eliminates Obesity in Rodents. Nat. Chem. Biol. 2009, 5, 749−757.
(130) Miranda, L. P.; Winters, K. A.; Gegg, C. V.; Patel, A.; Aral, J.;
Long, J.; Zhang, J.; Diamond, S.; Guido, M.; Stanislaus, S.; Ma, M.; Li,
(146) Shivakumar, D.; Harder, E.; Damm, W.; Friesner, R. A.;
Sherman, W. Improving the Prediction of Absolute Solvation Free
Energies Using the Next Generation OPLS Force Field. J. Chem.
Theory Comput. 2012, 8, 2553−2558.
́
H.; Rose, M. J.; Poppe, L.; Veniant, M. M. Design and Synthesis of
Conformationally Constrained Glucagon-like Peptide-1 Derivatives
with Increased Plasma Stability and Prolonged in Vivo Activity. J.
Med. Chem. 2008, 51, 2758−2765.
(147) Hoover, W. G. Canonical Dynamics: Equilibrium Phase-Space
Distributions. Phys. Rev. A 1985, 31, 1695−1697.
(131) Murage, E. N.; Gao, G.; Bisello, A.; Ahn, J.-M. Development
of Potent Glucagon-like Peptide-1 Agonists with High Enzyme
Stability via Introduction of Multiple Lactam Bridges. J. Med. Chem.
2010, 53, 6412−6420.
́
(148) Nose, S. A Unified Formulation of the Constant Temperature
Molecular Dynamics Methods. J. Chem. Phys. 1984, 81, 511−519.
(149) Martyna, G. J.; Tobias, D. J.; Klein, M. L. Constant Pressure
Molecular Dynamics Algorithms. J. Chem. Phys. 1994, 101, 4177−
4189.
(132) Murage, E. N.; Schroeder, J. C.; Beinborn, M.; Ahn, J.-M.
Search for Alpha-Helical Propensity in the Receptor-Bound
Conformation of Glucagon-like Peptide-1. Bioorg. Med. Chem. 2008,
16, 10106−10112.
(150) Darden, T.; York, D.; Pedersen, L. Particle Mesh Ewald: An
N·log(N) Method for Ewald Sums in Large Systems. J. Chem. Phys.
1993, 98, 10089−10092.
(151) Ryckaert, J.-P.; Ciccotti, G.; Berendsen, H. J. C. Numerical
Integration of the Cartesian Equations of Motion of a System with
Constraints: Molecular Dynamics of n-Alkanes. J. Comput. Phys. 1977,
23, 327−341.
̈
(133) Schreuder, H.; Liesum, A.; Lonze, P.; Stump, H.; Hoffmann,
H.; Schiell, M.; Kurz, M.; Toti, L.; Bauer, A.; Kallus, C.; Klemke-Jahn,
C.; Czech, J.; Kramer, D.; Enke, H.; Niedermeyer, T. H. J.; Morrison,
̈
V.; Kumar, V.; Bronstrup, M. Isolation, Co-Crystallization and
Structure-Based Characterization of Anabaenopeptins as Highly
Potent Inhibitors of Activated Thrombin Activatable Fibrinolysis
Inhibitor (TAFIa). Sci. Rep. 2016, 6, No. 32958.
̈
(134) Halland, N.; Bronstrup, M.; Czech, J.; Czechtizky, W.; Evers,
A.; Follmann, M.; Kohlmann, M.; Schiell, M.; Kurz, M.; Schreuder, H.
A.; Kallus, C. Novel Small Molecule Inhibitors of Activated Thrombin
Activatable Fibrinolysis Inhibitor (TAFIa) from Natural Product
Anabaenopeptin. J. Med. Chem. 2015, 58, 4839−4844.
(135) Halland, N.; Czech, J.; Czechtizky, W.; Evers, A.; Follmann,
M.; Kohlmann, M.; Schreuder, H. A.; Kallus, C. Sulfamide as Zinc
Binding Motif in Small Molecule Inhibitors of Activated Thrombin
Activatable Fibrinolysis Inhibitor (TAFIa). J. Med. Chem. 2016, 59,
9567−9573.
(136) Sun, X.; Gao, H.; Yang, Y.; He, M.; Wu, Y.; Song, Y.; Tong, Y.;
Rao, Y. PROTACs: Great Opportunities for Academia and Industry.
Signal Transduct. Target. Ther. 2019, 4, No. 64.
(137) Testa, A.; Hughes, S. J.; Lucas, X.; Wright, J. E.; Ciulli, A.
Structure-Based Design of a Macrocyclic PROTAC. Angew. Chem.,
Int. Ed. 2020, 59, 1727−1734.
(138) Hughes, S. J.; Ciulli, A. Molecular Recognition of Ternary
Complexes: A New Dimension in the Structure-Guided Design of
Chemical Degraders. Essays Biochem. 2017, 61, 505−516.
(139) De Guzman, R. N.; Goto, N. K.; Dyson, H. J.; Wright, P. E.
Structural Basis for Cooperative Transcription Factor Binding to the
CBP Coactivator. J. Mol. Biol. 2006, 355, 1005−1013.
(140) Guedes, I. A.; Pereira, F. S. S.; Dardenne, L. E. Empirical
Scoring Functions for Structure-Based Virtual Screening: Applica-
tions, Critical Aspects, and Challenges. Front. Pharmacol. 2018, 9,
No. 1089.
́
(141) Le Roux, A.; Blaise, E.; Boudreault, P.-L.; Comeau, C.;
̈
Doucet, A.; Giarrusso, M.; Collin, M.-P.; Neubauer, T.; Kolling, F.;
̈
Goller, A. H.; Seep, L.; Tshitenge, D. T.; Wittwer, M.; Kullmann, M.;
Hillisch, A.; Mittendorf, J.; Marsault, E. Structure−Permeability
Relationship of Semipeptidic MacrocyclesUnderstanding and
Optimizing Passive Permeability and Efflux Ratio. J. Med. Chem.
2020, 63, 6774−6783.
(142) King, D. S.; Fields, C. G.; Fields, G. B. A Cleavage Method
Which Minimizes Side Reactions Following Fmoc Solid Phase
Peptide Synthesis. Int. J. Pept. Protein Res. 1990, 36, 255−266.
(143) Cummings, M. D.; Sekharan, S. Structure-Based Macrocycle
Design in Small-Molecule Drug Discovery and Simple Metrics To
Identify Opportunities for Macrocyclization of Small-Molecule
Ligands. J. Med. Chem. 2019, 62, 6843−6853.
P
https://dx.doi.org/10.1021/acs.jmedchem.0c01500
J. Med. Chem. XXXX, XXX, XXX−XXX