Unconventional Secondary Structure Mimics: Ladder-Rungs

Article abstract of DOI:10.1002/anie.202002639

Secondary structures tend to be recognizable because they have repeating structural motifs, but mimicry of these does not have to follow such well-defined patterns. Bioinformatics studies to match side-chain orientations of a novel hydantoin triazole chemotype (1) to protein-protein interfaces revealed it tends to align well across parallel and antiparallel sheets, like rungs on a ladder. One set of these overlays was observed for the protein-protein interaction uPA?uPAR. Consequently, chemotype 1 was made with appropriate side-chains to mimic uPA at this interface. Biophysical assays indicate these compounds did in fact bind uPAR, and elicit cellular responses that affected invasion, migration, and wound healing.

Full text of DOI:10.1002/anie.202002639

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Unconventional Secondary Structure Mimics: Ladder-rungs
Authors: Chen-Ming Lin, Maritess Arancillo, Jonathan Whisenant, and
Kevin Burgess
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202002639
Angew. Chem. 10.1002/ange.202002639
Link to VoR: http://dx.doi.org/10.1002/anie.202002639
http://dx.doi.org/10.1002/ange.202002639

10.1002/anie.202002639
Angewandte Chemie International Edition
COMMUNICATION
Unconventional Secondary Structure Mimics: Ladder-rungs
Chen-Ming Lin,a,† Maritess Arancillo,a,† Jonathan Whisenant,a and Kevin Burgess*a
a
Dr. Chen-Ming Lin, Maritess Arancillo, Jonathan Whisenant, and Prof. Kevin Burgess
Department of Chemistry
Texas A & M University
Box 30012, College Station, TX 77842, USA
E-mail: burgess@tamu.edu
†
These authors contributed equally to this work
Supporting information for this article is given via a link at the end of the document.
Abstract: Secondary structures tend to be recognizable because
they have repeating structural motifs, but mimicry of these does not
have to follow such well-defined patterns. Bioinformatics studies to
match side-chain orientations of a novel hydantoin triazole chemotype
(1) to protein-protein interfaces revealed it tends to align well across
parallel and antiparallel sheets, like rungs on a ladder. One set of
these overlays was observed for the protein-protein interaction
uPA•uPAR. Consequently, chemotype 1 was made with appropriate
side-chains to mimic uPA at this interface. Biophysical assays
indicate these compounds did in fact bind uPAR, and elicit cellular
responses that affected invasion, migration, and wound healing.
the orientation of the side-chains than comparable peptides; and,
(iii) have similar spacings of amino acid side-chains to those in
peptides. Consistent with the first two of these criteria, structures
1 probably could be assembled from amino acids, and they have
only three degrees of freedom that govern side-chain orientations.
Furthermore, the N-substituted hydantoin contains a peptide-like
N-Cα-CO-N-Cα-CO periodicity, and the 1,4-triazole fragment has
a wider separation between consecutive side-chains.
Conformations of 1 with R1 – R3 = methyl were simulated in
a continuous dielectric 80 using quenched molecular dynamics,[9-
10]
and any that were 3 kcal/mol above the lowest energy were
discarded. Remaining conformers were clustered on their six Cα
– Cβ coordinates to give families that present the amino acid side-
chains in similar orientations. This data can be used in several
ways.
Secondary structure mimicry is appealing for its simplicity.
Elements of secondary structure at protein-protein interaction
(PPI) interfaces are conspicuous in structural studies, and provide
starting points for molecular design. This “top-down” strategy
begins with a secondary structure and progresses towards
molecular designs that fit it. Those designs tend to be predictable;
for instance they may present amino acid side-chains on the face
of a helix that binds a receptor protein, or above and below one
side of a strand-turn-strand that similarly binds. This is true for
mostly peptidic systems (eg stapled helices[1]), and for minimalist
mimics comprising of a non-peptidic core frameworks presenting
amino acid side-chains (eg sheet mimics[2-4]).
In the first application of this conformational data, the six Cα
– Cβ coordinates of the preferred conformations were overlaid on
every combination of three amino acids in each of the ideal
secondary structures (Exploring Key Orientations on Secondary
structures, EKOS, see[11-12] for details). Figure 1a shows the root
mean square deviation (RMSD) for the best fitting conformer for
each secondary structure. Rainbow coding in Fig 1a shows that
the fit is better on the extended secondary structures (blue –
violet), than on the helical ones (red – green), ie 1 is a better mimic
of sheets and strands than turns and helices.
The six Cα – Cβ coordinates of the preferred conformations
were also overlaid on a huge database (>250k) of side-chains on
crystallographically characterized PPI interfaces (Exploring Key
Orientations, EKO, see[13-14] for details). Chemotype 1 has
preferred conformations that overlay more often on secondary
structures (rather than on regions that have no discernible
secondary structures) than most of the minimalist mimics
analyzed in a previous study. Moreover, those overlays are most
frequently on extended structures (hairpins, sheets, strands) than
on turns or helices, consistent with findings from EKOS outlined
above.
We are interested in bottom-up approaches to interface
mimicry, ie beginning with non-peptidic structures that present
amino acid side-chains (minimalist mimics[5-6]), and exploration of
where they fit on PPI interfaces. From the onset it was clear this
approach provides insights into structures that resemble interface
segments that have no recognizable secondary structures.[7-8]
However, we did not anticipate finding minimalist mimics that
consistently overlay on interface secondary structures in an
unpredictable way. Here we share one such case of “ladder-rung”
mimics, generalized by structure 1.
Visualization of how chemotype 1 best overlays on side-
chains of parallel and anti-parallel sheets indicated two amino-
acid side chains tend to align on adjacent edge residues, the
triazole spans the sheet, which places the third residue on an
amino acid on the other side. This observation extrapolated to
the sheet overlays on interface segments; three examples
(uPA•uPAR,[15] Ku70•Ku80,[16] complement C4-A•C4-B[17]) are
shown in Fig 1, and 10 more are given in Figure S1b. We coined
the term “ladder-rung mimic” because chemotype 1 frequently
tends to fit amino acid side-chains well when it lays across sheet
like a rung on a ladder.
R1
R2
O
O
H
R3
O
HN
O
H2N
N
N
OH
N
OH
H
N
N
R1
R3
O
O
R2
N
similar peptides : more flexible
1
Chemotypes 1 satisfy three fundamental requirements;
they: (i) are potentially easy to prepare with substituents that
correspond to amino acids, including the less tractable
functionalized ones; (ii) have less freely rotating bonds that alter
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202002639
Angewandte Chemie International Edition
COMMUNICATION
310-helix
α-helix
γ-turn
inverse
γ-turn
parallel antiparallel
β-sheet
π-helix
sheet-turn β-strand
sheet
β-sheet
helical
extended
a
b
c
Ku80
uPA
Ku70
Complement C4-A
uPAR
Complement C4-B
Figure 1. a RMSD (Å) of the best overlays of 1aaa (all stereochemistries) on each of the ideal secondary structures. b Distribution of secondary structures at PPI
interfaces where an overlay of <0.3 Å RMSD was deduced (516 interfaces) as derived by DSSP[18-20] calculations (these do not differentiate strand-turn-strand,
parallel- and antiparallel-sheets). c Overlay of 1aaa on uPA in uPA•uPAR (PDB 3bt1, uPA is urokinase plasminogen activator, RMSD 0.14 Å), on Ku70 in Ku70•Ku80
(PDB 1jeq, RMSD 0.19 Å), and on complement C4-A in C4-A•C4-B (PDB 4fxk; RMSD 0.21 Å).
Scheme 1 outlines a solid phase synthesis of chemotypes
1. Briefly, this route features coupling azido acids[21] to the resin,
copper-mediated cycloaddition of amino acid-derived alkynes,
Nα-deprotection, coupling of an amino acid and hydantoin
formation.[22] Mimics in this paper are described using lower case
one-letter codes to relate the side-chains R1 – R3 to the closest
amino acids {eg k’ for –(CH2)4NHBoc of Lys (k), side-chains from
tBu-protected Ser is denoted s’ until deprotected when the
residue becomes s, etc. A complete guide to the protection used
is given in the SI.
R3
R3
H
.
(i) 20% piperidine/DMF
H
(i) 20% piperidine/DMF
FmocHN
R2
N
Fmoc-alkyne, CuSO4 5H2O
FmocHN
N
N
N
N3
O
sodium ascorbate, DMF, 25 °C, 16 h
(ii) azido acid, HATU
iPr2NEt, DMF, 25 °C, 2 h
(ii) Fmoc-AA, HATU
iPr2NEt, DMF, 70 °C, µW
N
O
L
D
-i;
-n;
L
D
-k’;
-f;
L
D
-y’
-k’
LL-s’i; DL-fk’; LL-n’k’; LL-k’y’
DD-fn; LD-nf; DD-fk’
R1
R1
R1
FmocHN
R3
H
R3
O
R3
O
HN
HN
O
H
HN
R2
(i) 20% piperidine/DMF
N
N
N
N
NH2
95:2.5:2.5 TFA:Et3SiH:H2O
4 h, 25 °C
N
N
N
N
N
O
O
N
O
O
R2
O
R2
N
(ii) 4-nitrophenyl chloroformate
iPr2NEt, MeCN, 15 min, 25 °C
(iii) 5% iPr2NEt/ NMP
N
N
LLL-fs’i; LLL-n’s’i; DLL-n’s’i; DDL-s’fk’
LDD-s’fk’; DLL-s’n’k’; DLL-n’k’y’
LDD-s’fn; DLD-s’n’f; DDD-s’fk’
LLL-fs’i; LLL-n’s’i; DLL-n’s’i; DDL-s’fk’
LDD-s’fk’; DLL-s’n’k’; DLL-n’k’y’
LDD-s’fn; DLD-s’n’f; DDD-s’fk’
LLL-fsi; LLL-nsi; DLL-nsi; DDL-sfk
LDD-sfk; DLL-snk; DLL-nky
LDD-sfn; DLD-snf; DDD-sfk
Scheme 1. Solid phase syntheses of the featured compounds 1.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202002639
Angewandte Chemie International Edition
COMMUNICATION
Having access to the compounds, the next step was to
establish if chemotypes 1 could disrupt a PPI. For this it was
necessary to prepare compounds 1 with appropriate amino acid
side-chains. Of the many potential PPI targets that emerged from
the EKO screen, uPA•uPAR was selected because of its
relevance to endothelial-to-mesenchymal transitions in cancer
metastases.[23-27] Consideration of all the various overlays below
a threshold RMSD (0.5 Å was chosen here) was necessary to
determine which side-chains and stereochemistries should be
designed into structures 1. Full data is given in Table S1, and one
example is shown in Figure 2. In Fig 2 for instance, mimic 1aaa
(lemon) in the DLL-configuration overlaid on 22Asn, 21Ser, 28Ile.
The docking pose shown in Fig 2 was modified to include these
side-chains, then the system was minimized in Glide; the energy
of the system was not abnormally high in the series of all overlays,
implying the backbone of DLL-1nsi had no unfavorable
interactions with uPAR. Overall, there were ten overlays like this.
in the range (7.5 – 22 μM), where LLL-1fsi was retained most
strongly (7.5 ± 0.1 μM).
Compounds that bind uPAR are not necessarily expected to
be cytotoxic. Cytotoxicities of LLL-1fsi, LDD-1sfk, and DDL-1sfk
on MDA-MB-231 (uPAR+ metastatic, triple negative breast cancer
cells;[35] Fig S2) were checked; only LLL-1fsi decreased in viability
significantly at 200 μM. From this point forward, only the
compound with highest affinity for uPAR, LLL-1fsi, was
considered.
a
a
b
Ser 21
Ile 28
Asn 22
b
Figure 2. a Mimic DLL-1aaa overlaid on uPA at 22Asn, 21Ser, and 28Ile without
unfavorable interactions of the backbone with uPAR, implying DLL-1nsi should
be prepared and tested. b Overlay of a preferred conformer of DDD-1aaa on
an ideal strand-turn-strand illustrating how the mimic projects two side-chains
above and one below the plane of equator of the secondary structure.
In fact, the description of this work does not exactly follow
the chronology of the experiments; the ten compounds prepared
in Scheme 1 have amino acid side-chains that correspond to uPA
residues where 1aaa overlaid on the uPA•uPAR interface in the
EKO analysis (checked in Glide as described above).
Assays used to monitor the effects of small molecules on
uPA•uPAR tend to test for parameters such as cell migration and
invasion to reflect influence on metastatic spread; the work of
Meroueh and co-workers on testing small molecules discovered
by high-throughput screening on the same target was a good
model for our studies.[28-33]
c
A competitive ELISA (uPAR anchored, uPA in solution) was
first used to monitor binding to uPAR. At 50 μM, all ten candidate
compounds from the EKO evaluation inhibited binding of uPA-
HRP (5 nM; HRP is horse radish peroxidase) to some extent. The
fact that all ten compounds passed the EKO evaluation speaks to
the validity of this method for identification of binders. Three of
these (LLL-1fsi, LDD-1sfk, DDL-1sfk) were selected for
determination IC50 values; the values obtained were 8.35 ± 0.05,
13.5 ± 0.1, 25.4 ± 0.1 μM, respectively (Fig 2b).
A fluorescence polarization assay was used to determine
dissociation constants of the three lead compounds with respect
to solubilized uPAR (Fig 3c; Ploug’s uPAR binding peptide
AE105[34] was used as a competitor). The values obtained were
Figure 3. a Competitive ELISA of all compounds at 50 μM. b Dose response of
three (lead) compounds to obtain IC50 values. c Competitive fluorescence
polarization to determine the binding, Ki, of the lead compounds 1.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202002639
Angewandte Chemie International Edition
COMMUNICATION
(M1603497). The NMR instrumentation at Texas A&M
a
University was supported by a grant from the National Science
Foundation (DBI-9970232).
Keywords: protein-protein interactions • peptidomimetics •
uPAR • peptide • cancer
b
c
Figure 4. a Dose response wound healing assay (peptide AE105 at 0.5 μM as
a positive control). Data processed using ibidi software (available online).[36]
Normalized migration and c invasion assays using MDA-MB-231 compared
with a DMSO control.
b
Mimic LLL-1fsi was evaluated in assays for wound healing,
cell migration, and invasion; MDA-MB-231 cells were used
throughout. Dose dependent effects were observed in all three
assays (images in Fig S3 and S4; AE105 was used as positive
control;[37] the test compounds had no effect on invasion or
migration of uPAR-negative HEK293 kidney embryonic cells[35]).
Thus LLL-1fsi suppresses cell movement into a scratch, migration
in response to a chemical gradient, and invasion across an
extracellular matrix-analog; all characteristics of suppression of
metastases.
In conclusion, bioinformatic analyses using the EKO suite of
approaches led to the identification of an unconventional
secondary structure mimic, the ladder-rung system 1.
Incorporation of side-chains on 1 corresponding to one potential
PPI target, uPA•uPAR, led to ten compounds that bound in this
chemotypes series. These were proven to competitively bind
uPAR and be functional in cellular assays designed to detect
suppression of metastatic spread.
Acknowledgements
We thank Dr Adam Mazar for a sample of uPAR, and the
following grant agencies for funding: DoD BCRP Breakthrough
Award (BC141561), CPRIT (RP170144 and RP180875), and
The Robert A. Welch Foundation (A-1121), Texas A&M
University (RP180875), NIH/NIBIB (R01EB022230) and NSF
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202002639
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents
Insert graphic for Table of Contents here.
Steps towards new peptidomimetics. A bottom up approach was used to design a ladder-rung mimic chemotype overlaying
across β-hairpins to perturb protein-protein interactions. The chemotype was evaluated on a databank of over 250,000 PDBs
followed by biological assays to probe the inhibition of uPA•uPAR protein interactions.
[1]
[2]
[3]
[4]
D. A. Guarracino, J. A. Riordan, G. M. Barreto, A. L. Oldfield, C. M. Kouba, D. Agrinsoni, Chem. Rev. 2019, 119, 9915-9949.
K. D. Stigers, M. J. Soth, J. S. Nowick, Curr. Opin. Chem. Biol. 1999, 3, 714-723.
J. M. Davis, L. K. Tsou, A. D. Hamilton, Chem. Soc. Rev. 2007, 36, 326-334.
J. S. Nowick, M. Abdi, K. A. Bellamo, J. A. Love, E. J. Martinez, G. Noroha, E. M. Smith, J. W. Ziller, J. Am. Chem. Soc. 1995, 117,
89-99.
[5]
H. Yin, A. D. Hamilton, in Chemical Biology, Vol. 1-3 (Eds.: S. L. Schreiber, T. M. Kapoor, G. Wess), Wiley-VCH-VCH Verlag, 2008,
pp. 250 - 269.
[6]
[7]
E. Ko, J. Liu, K. Burgess, Chem. Soc. Rev. 2011, 40, 4411-4421.
J. Taechalertpaisarn, R.-L. Lyu, M. Arancillo, C.-M. Lin, L. M. Perez, T. R. Ioerger, K. Burgess, Org. Biom. Chem. 2019, 17, 3267-
3274.
[8]
J. Taechalertpaisarn, R.-L. Lyu, M. Arancillo, C.-M. Lin, Z. Jiang, L. M. Perez, T. R. Ioerger, K. Burgess, Org. Biomol. Chem. 2019,
17, 908-915.
[9]
E. Ko, A. Raghuraman, L. M. Perez, T. R. Ioerger, K. Burgess, J. Am. Chem. Soc. 2013, 135, 167-173.
D. Xin, Andreas Holzenburg, Kevin Burgess, Chemical Science, 5, 4914-4921.
M. J. Perez de Vega, M. Martin-Martinez, R. Genzalez-Muniz, Current Topics in Medicinal Chemistry 2007, 7, 33-62.
D. Xin, E. Ko, L. M. Perez, T. R. Ioerger, K. Burgess, Org. Biomol. Chem. 2013, 11, 7789-7801.
O. Sperandio, M. A. Miteva, B. O. Villoutreix, Methods and Principles in Medicinal Chemistry 2011, 48, 435-465.
L. M. C. Meireles, G. Mustata, Current Topics in Medicinal Chemistry 2011, 11, 248-257.
S. L. Gonias, J. Hu, Frontiers in Pharmacology 2015, 6, 154/151 - 154/156.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
J. R. Walker, R. A. Corpina, J. Goldberg, Nature 2001, 412, 607-614.
R. T. Kidmose, N. S. Laursen, J. Dobo, T. R. Kjaer, S. Sirotkina, L. Yatime, L. Sottrup-Jensen, S. Thiel, P. Gal, G. R. Andersen,
Proc. Natl. Acad. Sci. 2012, 109, 15425-15430, S15425/15421-S15425/15413.
D. Frishman, P. Argos, Proteins: Structure, Function, and Genetics 1995, 23, 566-579.
M. Heinig, D. Frishman, Nucleic Acids Res. 2004, 32, W500-W502.
G. Touw Wouter, C. Baakman, J. Black, E. Krieger, G. Vriend, A. H. te Beek Tim, P. Joosten Robbie, Nucleic acids research 2015,
43, D364-D368.
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
P. B. Alper, S.-C. Hung, C.-H. Wong, Tetrahedron Lett. 1996, 37, 6029-6032.
M. Meusel, M. Guetschow, Org. Prep. Proced. Int. 2004, 36, 391-443.
L. Gonias Steven, J. Hu, Frontiers in pharmacology 2015, 6, 154.
R. D. Lester, M. Jo, V. Montel, S. Takimoto, S. L. Gonias, J. Cell Biol. 2007, 178, 425-436.
H. Noh, S. Hong, S. Huang, Theranostics 2013, 3, 487-495.
A. P. Mazar, Clinical Cancer Research 2008, 14, 5649-5655.
K. Dass, A. Ahmad, A. S. Azmi, S. H. Sarkar, F. H. Sarkar, Cancer Treatment Reviews 2008, 34, 122-136.
D. Xu, K. Bum-Erdene, Y. Si, D. Zhou, M. K. Ghozayel, S. O. Meroueh, ChemMedChem 2017, 12, 1794-1809.
T. Mani, F. Wang, W. E. Knabe, A. L. Sinn, M. Khanna, I. Jo, G. E. Sandusky, G. W. Sledge, D. R. Jones, R. Khanna, K. E. Pollok,
S. O. Meroueh, Bioorg. Med. Chem. 2013, 21, 2145-2155.
[30]
[31]
T. Mani, D. Liu, D. Zhou, L. Li, W. E. Knabe, F. Wang, K. Oh, S. O. Meroueh, ChemMedChem 2013, 8, 1963-1977.
F. Wang, W. E. Knabe, L. Li, I. Jo, T. Mani, H. Roehm, K. Oh, J. Li, M. Khanna, S. O. Meroueh, Bioorg. Med. Chem. 2012, 20,
4760-4773.
[32]
[33]
[34]
F. Wang, J. Li, A. L. Sinn, W. E. Knabe, M. Khanna, I. Jo, J. M. Silver, K. Oh, L. Li, G. E. Sandusky, G. W. Sledge, Jr., H. Nakshatri,
D. R. Jones, K. E. Pollok, S. O. Meroueh, J. Med. Chem. 2011, 54, 7193-7205.
M. Khanna, F. Wang, I. Jo, W. E. Knabe, S. M. Wilson, L. Li, K. Bum-Erdene, J. Li, G. W. Sledge, R. Khanna, S. O. Meroueh, ACS
Chem. Biol. 2011, 6, 1232-1243.
T. J. D. Jorgensen, H. Gaardsvoll, K. Dano, P. Roepstorff, M. Ploug, Biochemistry 2004, 43, 15044-15057.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202002639
Angewandte Chemie International Edition
COMMUNICATION
[35]
[36]
V. D'Mello, S. Singh, Y. Wu, R. B. Birge, J. Biol. Chem. 2009, 284, 17030-17038.
"Wound Healing ACAS Image Analysis", can be found under https://ibidi.com/software-and-image-analysis/187-wound-healing-
acas-image-analysis.html, 2020
[37]
M. Ploug, S. Ostergaard, H. Gaardsvoll, K. Kovalski, C. Holst-Hansen, A. Holm, L. Ossowski, K. Dano, Biochemistry 2001, 40,
12157-12168.
6
This article is protected by copyright. All rights reserved.