Angewandte
Communications
Chemie
the two diastereomers in the simulations, exhibit different
conformational preferences (Figure 3D; Supporting Infor-
mation). For the S-diastereomer 10a, the dominant calculated
structures are shown in Figures S5 and S6 (Supporting
Information), and demonstrate no significant secondary
structures, which is in excellent agreement with the CD
results. Further simulation of a peptide without the in-tether
R-substitution group (Ac-cyclo-CAAAS5(2-H)-NH2) indi-
cates that the polyproline-II (PII) conformation is intrinsi-
cally favored by the residues, and the representative structure
of the most populated cluster is not helical (Supporting
Information, Figure S5A). In this non-helical structure, an
R = CH3/Ph substitution with (S)-chirality can be added
without any steric interference (Supporting Information,
Figure S5B). However, the non-helical structure is signifi-
cantly destabilized when an R = CH3 substitution group is
placed in (R)-chirality, and is very comfortable when the
peptide backbone adopts an a-helical conformation
(Supporting Information, Figure S5C). A larger R = Ph
group in (R)-chirality lead to higher destabilization of non-
helical structures, and a stronger preference for an a-helical
conformation (Supporting Information, Figure S5D). More
detailed information can be found in Figures S6–S8 (Support-
ing Information).
The influence of conformation on the biochemical/bio-
physical properties of peptides remains unclear, largely
because of the absence of methods for constructing peptides
with minimal differences in chemical composition. Cell
permeability is the major limitation for peptide therapeutics
and is influenced by many aspects, including conformation.[23]
Scrambling the positions of a few amino acids in a peptide can
dramatically change permeability and other biophysical
properties. To date, our strategy provides the only method
for specifically investigating the sole influence of conforma-
tional differences. Firstly, peptide diastereomers 11a/
11b FITC-bA-[cyclo-CRARS5(2-Ph)]-NH2 and 12a/12b
FITC-bA-[cyclo-CRRRS5(2-Ph)]-NH2 (bA (beta alanine),
FITC (fluorescein isothiocyanate)) were synthesized and
separated. As shown in Figure 4, the helical diastereomers
11b and 12b could successfully penetrate HEK293T cells
within 2 h while the other diastereomers were much less
permeable (Figure 4A). This led us to consider whether
a helical conformation itself could make peptides permeable.
Figure 4. Cell permeability of pentapeptide diastereomers. A) Fluores-
cent confocal microscopy images of HEK293T cells incubated with
FITC labeled peptides 11a/b and 12a/b (5 mm) at 378C for 2 h (blue
(DAPI), green (FITC)). B) CD spectra of peptides 11a/b and 12a/b at
208C in 50% TFE buffer. C) Flow cytometry measurements of
HEK293T cells with peptide 11a/b, 12a/b, and 13a/b (5 mm) at 378C
for 2 h.
substitution groups at the tether chiral center on the target
binding affinity of the peptide. ER-1 a/b and ER-2 a/b were
synthesized based on their reported sequences (Fig-
ure 5A),[24a] which contain a methyl or phenyl group at the
chiral center, respectively. ER-1b and ER-2b showed a sig-
nificant increase in helicity compared to ER-1a and ER-2a
(Figure 5B). The binding affinity of ER-1b (ca. 1 nm) and
ER-2b (ca. 69 nm) is much better than that for ER-1a (ND)
and ER-2a (> 600 nm; Figure 5C,D). Interestingly, ER-1b
showed a significantly enhanced binding affinity compared to
all previously reported ER-a peptide ligands, which may be
caused by the additional interaction contributed by the
methyl group at the stereocenter in the tether with the ERa
protein. Peptides PDI-1a/b and PDI-2a/b were also synthe-
sized based on their reported sequences.[25] PDI-1b and 2b
showed a remarkable increase in helicity compared to PDI-1a
and 2a (Figure 5E). They also showed significantly better
binding affinities than PDI-1a and 2a (Figure 5F,G). How-
ever, PDI-2b showed unfavorable binding (ca. 504 nm)
compared to PDI-1b (ca. 165 nm), which may be caused by
the steric hindrance imposed by the bulky phenyl group and
MDM2. These results constitute the first direct evidence of
a direct relationship between helical enhancement and
peptide ligand/protein target binding. Moreover, these results
suggest that the substitution group also interacts directly with
the binding groove, providing a valuable modification site for
future applications, such as fragment-based peptide ligand
design.
Peptide
diastereomers
13a/13b
FITC-bA-[cyclo-
CAKAS5(2-Ph)]-NH2 were subsequently tested. Peptide
13b showed enhanced helicity over peptide 13a (Supporting
Information, Figure S9A). However, while peptide 13b out-
performed peptide 13a in terms of penetration of the cell
membrane, it only showed minimal penetrative efficacy
(Supporting Information, Figures S9B and S9C). These
results suggest that, although helical conformation itself
may not guarantee the permeability of peptides, it is
a determining factor. The structural elucidation of estrogen
receptor alpha (ERa) and mammal double minute 2 (MDM2)
with their constrained peptide ligands clearly showed the
interaction between the protein targets and the ligand tethers,
mostly in the flat hydrophobic region surrounding the target
ligand binding site.[24] Based on the results, we chose these two
model targets to study the influence of peptide helicity and
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!