Is the Mirror Image a True Reflection? Intrinsic Membrane Chirality Modulates Peptide Binding

Article abstract of DOI:10.1021/jacs.9b11194

Peptides with pharmaceutical activities are attractive drug leads, and knowledge of their mode-of-action is essential for translation into the clinic. Comparison of native and enantiomeric peptides has long been used as a powerful approach to discriminate membrane-or receptor-mediated modes-of-action on the basis of the assumption that interactions with cell membranes are independent of peptide chirality. Here, we revisit this paradigm with the cyclotide kalata B1, a drug scaffold with intrinsic membrane-binding activity whose enantiomer is less potent than native peptide. To investigate this chirality dependence, we compared peptide-lipid binding using mirror image model membranes. We synthesized phospholipids with non-natural chirality and demonstrate that native kalata B1 binds with higher affinity to phospholipids with chirality found in eukaryotic membranes. This study shows for the first time that the chiral environment of lipid bilayers can modulate the function of membrane-active peptides and challenges the view that peptide-lipid interactions are achiral.

Full text of DOI:10.1021/jacs.9b11194

Subscriber access provided by BOSTON UNIV
Article
Is the mirror image a true reflection? Intrinsic
membrane chirality modulates peptide binding
Sónia Troeira Henriques, Hayden Peacock, Aurelie Benfield, Conan K Wang, and David J Craik
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b11194 • Publication Date (Web): 25 Nov 2019
Downloaded from pubs.acs.org on December 1, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
Is the mirror image a true reflection? Intrinsic membrane chirality
modulates peptide binding
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Sónia Troeira Henriques†,‡,*, Hayden Peacock†, Aurélie H. Benfield†,‡, Conan K. Wang†, David J.
Craik†*
†
‡
Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland 4072, Australia
School of Biomedical Sciences, Faculty of Health, Institute of Health & Biomedical Innovation, Queensland University of
Technology, Translational Research Institute, Brisbane, QLD 4102, Australia
KEYWORDS Peptide-membrane interactions; membrane-active peptides; peptide therapeutics; model membranes;
cyclotides; membrane chirality; lipid chirality; peptide chirality; disulfide-rich peptides
ABSTRACT: Peptides with pharmaceutical activities are attractive drug leads and knowledge of their mode-of-action is essential
for translation into the clinic. Comparison of native and enantiomeric peptides has long been used as a powerful approach to
discriminate membrane- or receptor-mediated modes-of-action based on the assumption that interactions with cell membranes are
independent of peptide chirality. Here we revisit this paradigm with the cyclotide kalata B1, a drug scaffold with intrinsic membrane-
binding activity whose enantiomer is less potent than native peptide. To investigate this chirality dependence we compared peptide-
lipid binding using mirror image model membranes. We synthesised phospholipids with non-natural chirality and demonstrate that
native kalata B1 binds with higher affinity to phospholipids with chirality found in eukaryotic membranes. This study shows for the
first time that the chiral environment of lipid bilayers can modulate the function of membrane-active peptides and challenges the view
that peptide-lipid interactions are achiral.
INTRODUCTION
enantiomer has lower activity, the general consensus is that the
activity of a given peptide is dependent on its binding to a
protein. Alternatively, if L and D-forms of the peptide have
similar activity it is assumed that their mechanism does not
involve chirality and lipid bilayers may be the target. This
approach has been successfully used to investigate the mode-
of-action of a range of membrane-active peptides, including
An increasing number of natural and designed peptides have
been approved as therapeutics and peptides have accordingly
gained credibility as
representing an emerging new class of pharmaceuticals.
Peptides uniquely combine the high specificity of biologics and
a valuable therapeutic modality,
1
-2
1
the cell permeable properties of small molecule drugs. With
1
5
helical antimicrobial peptides, disulfide-rich toxin peptides
this combination of properties, peptides are attracting great
interest as candidates to inhibit intracellular protein-protein
16
able to inhibit ion channels, beta-sheet antimicrobial
17
18-19
3
-4
peptides, cell-penetrating peptides,
and orally available
interactions; a therapeutic space traditionally assumed to be
20
5
peptides.
‘
undruggable’ by either small molecule drugs or biologics.
Many peptide-based drug leads interact with cell membranes,
either directly affecting membrane components, or by crossing
the cell membrane. For example, many antimicrobial peptides
act by selectively targeting negatively-charged cell
6-7
membranes, whereas cell-penetrating peptides bind to and
cross cell membranes before reaching an intracellular target of
8
-9
interest. In addition, orally available peptide drugs cross the
intestinal epithelium through cell tight junctions, or permeate
1
0-11
Figure 1. Sequence and structure of the prototypic cyclotide, kalata
B1. (a) Sequence of kalata B1. The peptide backbone is cyclised
between Gly-1 and Asn-29; the black arrow indicates the direction
of the peptide chain. The six cysteines are labelled with Roman
numerals and the disulfide connectivity is shown in yellow. (b)
Surface representation of the three-dimensional structure of kalata
B1 (PDB ID: 1nb1) shown in two views. The residues that form the
hydrophobic patch are shown in green and those that form the
bioactive face are shown in red.
epithelium cells by crossing their membranes
Thus, the
molecular details of how peptides interact with components of
cell membranes (e.g., the lipid bilayer or embedded protein
receptors) is essential to progress peptide drug candidates
towards clinical applications.
Comparison of the efficacy of a peptide and its enantiomer
(i.e., its all-D-amino acid form) is a commonly used approach to
characterise mode-of-action and to distinguish peptide-lipid
12-14
from peptide-protein interactions.
It is well-known that
Cyclotides have
a range of intrinsic pharmaceutical
functional peptide-protein interactions are stereospecific and
hence critically dependent on peptide chirality. Thus, if a D-
bioactivities and a well-defined mode-of-action involving
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
membrane binding.21-22 These peptides are expressed in certain
plant species and are characterised by a cyclic backbone and six
cysteines linked in a knot (Figure 1a),23 a patch of hydrophobic
residues exposed on their surface, and a set of conserved
-1
-1
assuming 280= 5875 M .cm for L-kalata B1, D-kalata B1,
[T16S]kB1, and 280 = 5040 M-1.cm-1 for [W23W-OH]kB1.27
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Circular Dichroism (CD) spectroscopy. CD specta of L-
kalata, D-kalata B1 and mixed in a 1:1 ratio were obtained with
samples containing 50 M peptide in water. Spectra were
recorded (five scans, 190-260 nm) in a 1 mm quartz cuvette
with a Jasco J-810 spectropolarimeter. Blank was subtracted.
2
4
residues that form the so-called bioactive face (Figure 1b).
They exhibit diverse bioactivities, including anti-HIV and
cancer cell toxicity, which are all dependent on their ability to
target phosphatidylethanolamine (PE)-phospholipids at the cell
2
5-27
Isolation of human red blood cells (RBCs) and of
peripheral blood mononuclear cells (PBMCs). Fresh human
blood samples from three healthy donors were collected into
BD vacutainer® EDTA-tubes following protocols and ethics
approved by the Human research Ethics Department at the
University of Queensland. RBCs were isolated by centrifuging
the blood at 1500 g for 1 min and washed three times with
phosphate buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl,
membrane surface.
A vast collection of experimental
evidence from biophysical studies with model and biological
membranes, and comparison with bioassay data, suggests that
kalata B1, the prototypic and the most studied cyclotide, targets
cell membranes by a bi-modal mechanism that involves
cooperativity between distinct parts of the molecule: the
bioactive face targets the PE-headgroup, whereas the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4-30
hydrophobic patch inserts into lipid bilayers.
8
2 4 2 4
mM Na HPO , 2 mM KH PO ). Isolated RBCs were used for
haemolytic assay as below. PBMCs were isolated by density
gradient centrifugation using Ficoll-Paque PREMIUM (GE
The mode-of-action of kalata B1 challenges the premise of
using enantiomers to assess the importance of interaction with
cell membrane components. In previous studies, we have shown
that the synthetic enantiomer of kalata B1 (D-kalata B1, Figure
27
healthcare; protocol 28-4039-56 AD). The layer containing
PBMCs was collected, and cells were washed three times with
PBS, counted and resuspended in RPMI supplemented with
penicillin/streptomycin (1%), L-glutamine (20 mM) and FBS
(10%). PBMCs were used in cytotoxicity assay as detailed
below.
2
4
a), although active, is consistently less potent (by around 20 to
0%) in all activities tested (e.g. hemolysis, insecticidal, anti-
3
1-32
HIV) than natural kalata B1 (L-kalata B1).
The lower
activity of D-kalata B1, compared to L-kalata B1, correlated
32
with a lower affinity for tested lipid bilayers. Although these
Hemolytic assay. Hemolysis of RBCs induced by L- or D-
kalata B1 was quantified by following release of haemoglobin
into the supernatant (absorbance at 415 nm) as previously
described. Briefly, RBCs (0.5% v/v) were resuspended in
phosphate buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl,
HPO , 2 mM KH PO ) and incubated with L- or D-
2 4 2 4
kalata B1 at various concentrations (two-fold dilutions, starting
at 64 M) for 1 h at 37C. The assay was done in triplicates.
studies supported a mechanism dependent on peptide-lipid
binding, they contradict the general assumption of peptide-lipid
interactions being essentially independent of chirality and
raised the question of a potential yet-unidentified membrane
receptor.
3
6
8
mM Na
Lipid membranes of eukaryotic and prokaryotic cells are
chiral owing to a chiral centre within the glycerol moiety of the
3
6
3
3
phospholipids. In the current study, we were interested in
investigating whether the chiral environment created by
biological membranes could discriminate between kalata B1
enantiomers, and therefore explain the lower potency of D-
kalata B1. To achieve this aim, we synthesised non-natural
phospholipid enantiomers, and compared L- and D-kalata B1 in
their ability to bind to model membranes composed of either
natural or non-natural phospholipid enantiomers. In addition,
we explored the molecular details of the interactions and
identified the regions within the peptide and the lipid molecules
that are important for the chiral recognition. Together, these
studies confirm that the bioactive patch of kalata B1 binds to
PE-headgroups in a chirality-independent manner, whereas
insertion of the hydrophobic path into the lipid bilayers is
dependent on the chiral environment. Thus, differences in the
potency of L-kalata B1 and D-kalata B1 are modulated by the
chirality of biological membranes.
Cell culture. Adherent human foreskin fibroblast cells (HFF-
), human cervix epitheloid carcinoma cells (HeLa), and human
melanoma cells (MDA-MB-435S) were cultured in DMEM
supplemented with penicillin/streptomycin (1%) and with foetal
bovine serum (FBS; 15% (HFF-1) and 10% (HeLa, MDA-MB-
435S)). Cells were grown in a humidified atmosphere with 5%
1
CO at 37C in T75 flasks till reaching 90% confluence, and
2
sub-cultured with a 1:4 ratio two-to-three times a week.
Cytotoxicity assay. The day before the assay, cultured cells
were detached with trypsin-EDTA, counted, resuspended in
fresh media and dispensed onto
a
96-well plate (5x
0 cells/well). On the day of the assay, cells were incubated
with L- and D-kalata B1 with two-fold dilutions (starting at 64
M) in fresh media without serum and incubated for 24h.
3
1

Blanks without peptide, and controls with TX-100 (0.1% v/v)
were included. Resazurin was added to cells (0.02% w/v) 2 h
after peptide addition and incubated for 22 h. Cell death was
quantified by resazurin fluorescence emission (excitation = 565
nm / emission = 584 nm) using cells incubated with TX-100 to
establish 100% of cell death, and cells samples without
METHODS
Peptide extraction and synthesis. The native cyclotide L-
kalata B1 was isolated from the above-ground parts of
3
4
Oldenlandia affinis as previously detailed. L-kalata B1
7
treatment to establish 0% of cell death, as before . Three
mutants, [T16S]kB1 and [W23W-OH]kB1 were synthesised
27
independent replicates were conducted. This assay was repeated
to test toxicity against PBMCs, with the exception that freshly
isolated PBMCs were resuspended and distributed in a 96-well
and characterised as before. The all D-kalata B1 was
32
synthesised using the protocol previously detailed. The purity
of the native and synthetic peptides was examined by analytical
reverse phase-HPLC. All the synthetic peptides retained the
4
plate (10 cells/well) on the day of the assay.
1
fold of native peptide, as confirmed by H NMH spectroscopy
Insertion of peptides into cell membranes followed by
3
5
using methodologies previously described. Peptide sample
concentration was determined following absorbance at 280 nm,
variations in membrane dipole. In the day before the
4
experiment, HeLa cells were plated (2 x 10 cells/well) onto a
9
6-well plate (polystyrene black plate with flat clear bottom).
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
On the day of the experiment, 8 M di-8-ANEPPS (Thermo
3 min) over deposited lipid bilayers to follow association; after
peptide injection stopped, dissociation was followed for 10 min.
The chip surface was regenerated as previously detailed.
Freshly prepared HEPES buffer filtered with a 0.2 m pore size
filter was used as running buffer and to prepare peptide
solutions and SUV suspensions. The effect of pH was examined
by preparing all the samples in HEPES buffer with the pH
adjusted.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Fisher Scientific) was added to cells in fresh media and
incubated at 37C for 2 h to allow the dye to partition to the
cells; as this dye is non-fluorescent in aqueous environment,
non-incorporated dye does not need to be removed. Peptides at
various concentrations were added to cells and incubated for 20
min. Fluorescence excitation spectrum (emission = 592 nm;
frequency 400 Hz, number of flashes 50, integration time 2 s)
of cells incubated with 32 M L- or D-kalata B1, and of cells
without peptide, were acquired at 37C in a Tecan Infinite
M1000Pro. Alterations in cell membrane dipole potential upon
insertion of peptides was detected by spectral variations on the
di-8-ANEPPS: all excitation spectra were normalised to total
4
0
Membrane disruption induced by L- and D-kalata B1. The
ability to disrupt lipid bilayers was compared by examining the
leakage of carboxyfluorescien (CF) from LUVs in 96-well
black plates. Briefly, LUVs (5 M lipid) loaded with CF at self-
quenching concentrations (50 mM) and dispersed in HEPES
buffer were incubated with two-fold dilutions of L-, D-kalata, or
with mixtures of the two enantiomers (i.e., L-/D-kalata B1 at
25:75, 50:50, and 75:25) for 20 mins. Blanks without peptide,
and controls with melittin (a known membrane-disruptive
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
37
integrated area; excitation spectrum obtained with cells
incubated with 32 M peptide were subtracted from spectrum
obtained with cells without peptide. A blue shift in the
differential excitation spectrum indicates an increase, whereas
a red shift indicates a decrease in the membrane dipolar
2
5
peptide), and with 0.05% (v/v) Triton X-100 were included .
The percentage of vesicles with leaky membranes was
quantified by following the fluorescence emission intensity of
CF (excitation = 594 nm / emission = 515 nm; Tecan Infinite
M1000pro) considering fluorescence signal from sample
incubated with Triton X-100 as 100% of leakage and buffer as
3
8
potential.
Dose-response effects were examined by
calculating the ratio of the excitation at 420 and 495 nm with
emission at 592 nm39 with varying concentrations of L- and D-
kalata B1. These experiments were repeated on three
independent days.
4
1
0
% of leakage, as before .
Synthesis of phospholipids with non-natural chirality.
Synthesis was conducted as summarised in scheme 1 in the
results section. Full details of the synthesis and NMR
characterization of each intermediary is shown in
Supplementary information.
Insertion of L- and D-kalata B1 into model membranes.
LUVs composed of L/L-POPC/POPE, or D/D-POPC/POPE
(
final lipid concentration of 200 M) with 8 M of di-8-
37
ANEPPS vesicles were incubated with varying concentrations
of peptides in a 96-well plate format, and the fluorescence
excitation spectra (emission = 596 nm; frequency 400 Hz, number
of flashes 50, integration time 2 s) recorded in a Tecan Infinite
M1000pro. Fluorescence excitation differential spectra were
determined as above, described with cells.
Preparation of lipid vesicles. Synthetic (R)-phospholipids
(i.e. 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine ((R)-
POPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-
phosphoethanolamine ((R)-POPE), referred to in the text as l-
POPC and l-POPE) were obtained commercially from Avanti
polar lipids. The (S)-phospholipids (i.e. 2-oleoyl-3-palmitoyl-
sn-glycero-1-phosphocholine ((S)-POPC) and 2-oleoyl-3-
Binding of L- and D-kalata B1 to dodecylphosphocholine
(DPC) micelles followed by NMR spectroscopy. DPC
palmitoyl-sn-glycero-1-phosphoethanolamine
((S)-POPE)
(dodecylphosphocholine-d38, 98%) was purchased from
referred to in the text as d-POPC and d-POPE) were synthesised
in this study, as detailed in the results section. Each
phospholipid was solubilised in spectrophotometric grade
chloroform and mixed at the required ratio and amount;
Novachem. 16-doxylstearate was purchased from Sigma
Aldrich. Both L- and D-kalata B1 were prepared by dissolving
2 2
lyophilized peptide in H O (10% D O) to a concentration of 0.4
-1
-1
mM (quantified using an ε280 of 5500 M cm ). The pH was
2
chloroform was evaporated under a stream of N to yield a lipid
adjusted to 2.8 via addition of concentrated HCl. One- and two-
film; trace amounts of solvent were evaporated in a desiccator
under vacuum overnight. Small unilamellar vesicles (SUVs; 50
nm diameter) or large unilamellar vesicles (LUVs; 100 nm
diameter) were prepared by resuspending lipid films in buffer,
fracturing by freeze-thaw cycles, and sizing by extrusion, as
1
1
dimensional NMR spectra ( H, H TOCSY, NOESY) were
recorded on a Bruker Avance 500 MHz spectrometer at the
indicated temperatures. Mixing times of 80 and 100 ms were
used for TOCSY and NOESY experiments, respectively.
Translational diffusion measurements were recording using the
stebpgp1s19 pulse program, which uses bipolar gradient pulses
for diffusion and the 3-9-19 pulse sequence for water
suppression. An exponential window function with 2 Hz line
4
0
previously detailed. SUVs were used in SPR experiments and
LUVs were used in experiments employing fluorescence
methodologies, as previously explained.37 Unless otherwise
stated, vesicle suspensions were prepared in HEPES buffer (10
mM HEPES, 150 mM NaCl, pH 7.4).
broadening was applied before Fourier transformation,
42
followed by baseline correction.
Data analyses were
Binding of kalata B1 analogues to lipid bilayers followed
by surface plasmon resonance (SPR). Peptide-lipid
interactions were investigated at 25C by SPR using L1 sensor
chips and a Bioacore 3000 instrument (GE healthcare) and
accomplished using the relaxation T1/T2 routine to measure
signal attenuation of methyl proton peaks. Spectra were
processed with TopSpin (Bruker), phased and calibrated, and
then assigned with CCPNMR software v2.4.2. Chemical shifts
2
5-26
in the
1
H dimension were referenced to internal 4,4-dimethyl-4-
or tetramethylsilane.
following methodologies previously optimised.
SUV suspensions (0.5 mM) were injected over an L1 chip (2
L/min for 40 min) reaching a steady signal, which suggests
Briefly,
silapentane-1-sulfate
Dodecylphosphocholine and 16-doxylstearate was added to
peptide samples using lyophilized aliquots of stock solutions in

coverage of the whole chip surface and deposition of stable lipid
bilayers. A pulse of 10 mM NaOH (50 L/min, 36 s) was
injected to remove loosely bound membrane portions. Peptide
samples at various concentrations were injected (5 L/min for
H
2
O and methanol/H
relaxation enhancement induced by 16-doxylstearate was
characterized by cross-peak attenuation in NOESY (τ
2
O, respectively. The paramagnetic
m
=
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
probe
1
00ms) spectra: Relative Signal Intensity (%) = 100% × I
/
towards all the tested cells, as shown by dose-response curves
and quantified by the concentration required to achieve 50%
0
probe
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
I ; where I
and I are the Hα-HN (Hα-Hδ for prolines) cross-
peak intensities in the spectra of sample with and without 16-
doxylstearate, respectively.
cell death (Figure 3). These results confirm that L- and D-kalata
B1 are active, but native kalata B1 is more potent than its
enantiomer. The slope of the dose-response curves for any
given cell type is identical for the two isomers, suggesting that
they act using the same mode-of-action.
4
3-44
RESULTS
Comparison of L- and D-kalata B1 bioactivity. Native kalata
34
B1 was extracted from Oldenlandia affinis and enantiomeric
3
2
D-kalata B1 was synthesised as before. The peptides were
purified to high purity (>95%) and had identical retention times
on reverse-phase HPLC. The three-dimensional structure of D-
kalata B1 has been previously determined by solution-state
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
2
45
NMR spectroscopy and racemic crystallography, and
confirmed to be the mirror image of native kalata B1 (Figure
2
a). CD spectra obtained with pure solutions of L- and D-kalata
B1 show equal molar ellipticity but opposite CD signal (Figure
b), whereas the CD signal of the two enantiomers mixed in
Figure 2. L- and D-kalata B1 enantiomers. (a) Cartoon
representation of the three-dimensional structure of L-kalata B1
2
equal proportions is nullified, further confirming that native and
D-kalata B1 are mirror images of each other.
(red; PDB ID: 1nb1) and D-kalata B1 (blue; PDB ID: 2jue), as
determined by NMR. Disulfide bonds are shown in yellow. (b) CD
spectra of L-kalata (red), D-kalata B1 (blue), and both mixed in a
1:1 ratio (orange); obtained with 50 M peptide.
The toxicities of the two enantiomers towards cultured cancer
cells and fresh human blood cells isolated from healthy donors
were compared. L-kalata B1 was more toxic than D-kalata B1
Figure 3. Cytotoxicity of L-kalata B1 (L-kB1) and D-kalata B1 (D-kB1). The two enantiomers were tested for their toxicity towards cultured
cell lines (HeLa, HFF-1, and MDA-MB-435S) and isolated blood cells (peripheral blood mononuclear cells, PBMCs, and red blood cells,
RBCs). Cell death of cultured cells and PBMCs was determined via a resazurin assay. Toxicity towards RBCs was examined by haemolysis
as measured by the absorbance of haemoglobin released into solution. The concentration required to achieve 50% cell death (CC50) was
calculated by fitting data points to a sigmoidal dose-response with variable slope in Graphpad Prism 7.0d. The results confirm that L-kalata
B1 is more potent across all cell lines tested than its enantiomer.
Insertion of L- and D-kalata B1 into cell membranes.
Insertion of peptides into membranes induces changes in
membrane dipolar potential, and these changes can be
detected by fluorescent membrane dyes sensitive to the
4b). The membrane dipolar potential results from water
molecules in the proximity of the headgroup region, the ester
groups between the headgroup to the glycerol and the fatty
46
acid carbonyl groups of the lipids. Thus, alteration in the
2
5
surrounding electric field, such as Di-8-ANEPPS. To
investigate whether L- and D-kalata B1 insert into cell
membranes, Di-8-ANEPPS was incorporated into HeLa cells
and its fluorescence excitation spectrum examined in cells
without peptide, and in those treated with L- or D-kalata B1.
Differential spectra (Figure 4a) show that both peptides
induce a blue shift in the excitation (i.e. maximum at 420 nm
with a minimum at 495 nm), indicating an increase in the
dipolar potential. This is more evident when the ratio of
excitation at 420 nm and 495 nm relative to emission at 592
nm is plotted as a function of peptide concentration (Figure
membrane dipolar potential suggest that both peptides insert
into HeLa cells, but L-kalata B1 does so to a larger extent.
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
Figure 5. Structure of phospholipids used in this study. Natural
1
2
3
4
5
6
7
(left) and non-natural (right) enantiomers of the phospholipids
POPE and POPC. The head-groups are highlighted in blue, the
glycerol units are highlighted in red, and the stereocentres are
indicated with asterisks.
To examine whether the chirality of the membrane plays a
role in the activity of kalata B1, we synthesised POPC and
POPE phospholipids with non-natural chirality (i.e. the (S)-
enantiomers), which are not available commercially. The
headgroups of both POPC and POPE are achiral, and these
phospholipids only have one chiral center, which is located in
the glycerol moiety (Figure 5).
Figure 4. Insertion of L- and D-kalata B1 into HeLa cells.
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Changes in membrane dipolar potential in HeLa cells induced by
L-kalata B1 and D-kalata B1 were detected by the spectral shift of
di-8-ANEPPS fluorescence excitation spectra. (a) Differential
fluorescence spectra of di-8-ANEPPS obtained when cells were
incubated with 32 M L- or D-kalata B1 and subtracted from the
spectrum obtained in the absence of peptide. The fluorescence
difference obtained with L- and D-kalata B1 show a blue shift,
which indicates an increase in the dipolar potential. (b) Ratio of
excitation at 420 and 495 nm with emission at 592 nm plotted as
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
As shown in Scheme 1, the non-natural enantiomers were
synthesised in six steps from commercially available (R)-(−)-
2
,2-dimethyl-1,3-dioxolane-4-methanol. The hydroxyl group
of the starting material was protected as 2-
a
methoxyethoxymethyl (MEM) ether, and the acetal was
48
hydrolysed to yield diol 4. Regioselective acylation of the
a
function of peptide concentration normalized for the
primary hydroxyl group with palmitoyl chloride, followed by
secondary acylation with oleoylchloride, yielded the protected
fluorescence obtained without peptide. Error bars are SD of three
independent experiments. L-kalata B1 induces a larger shift,
suggesting that it inserts deeper than its enantiomer into HeLa
cells.
4
8
diacylglycerol 6. The MEM group was cleaved with
4
9
iodotrimethylsilane (generated in situ), and subsequent
5
0-51
phosphorylation, substitution and hydrolysis
yielded (S)-
Synthesis of phospholipids with non-natural chirality. In
previous studies, we have shown that both L- and D-kalata B1
bind to lipid bilayers that possess PE-phospholipids, and bind
with only weak affinity for membranes without PE-
POPE (2-oleoyl-3-palmitoyl-sn-glycero-1-
phosphoethanolamine) (1) and (S)-POPC (2-oleoyl-3-
1
palmitoyl-sn-glycero-1-phosphocholine) (2). The H NMR
spectra of 1 and 2 overlay with those of the corresponding
natural (R) enantiomers, which were commercially obtained
3
2
phospholipids. The binding affinity of D-kalata B1 was
lower than that of L-kalata B1 for membranes composed of 1-
palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine ((R)-POPC)
(
see spectra in Supplementary Information), and the specific
22
−1
−1
optical rotations ([]
D
−1
= –6.0 deg·mL·g ·dm for 1 and –
mixed
with
1-palmitoyl-2-oleoyl-sn-glycero-3-
−
1
5.9 deg·mL·g ·dm for 2) are of the same magnitude, but
opposite signs, of the corresponding natural enantiomers.
Commercially obtained (R)-POPC and (R)-POPE, and
synthesised (S)-POPC and (S)-POPE were used throughout
this study to prepare model membranes. For the sake of
simplicity, the natural (R) enantiomers will be referred to as
L-phospholipids (e.g. L-POPE) and the non-natural (S)
enantiomers will be referred to as D-phospholipids (e.g. D-
POPE).
phosphoethanolamine ((R)-POPE) in a 4:1 molar ratio,
corresponding to the proportion of PE-phospholipids present
4
7
in mammalian plasma membranes. In the current study, we
chose POPC and POPE as phospholipids for model
membranes for comparison with previous studies with kalata
25-26, 32
B1.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 11
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of non-natural phospholipid enantiomers. MEM = 2-methoxyethoxymethyl, DIPEA = N,N-diisopropylethylamine,
DCM = dichloromethane, TEA = triethylamine, DMAP = 4-dimethylaminopyridine, THF = tetrahydrofuran.
Binding of L- and D-kalata B1 to mirror image model
membranes. The ability of L- and D-kalata B1 to bind to
model membranes composed of phospholipids of non-natural
chirality was examined using SPR. Peptide samples were
injected over lipid bilayers deposited onto an L1 sensor chip
molar ratio) of natural (L) or non-natural (D) chirality. L-kalata
B1 has higher affinity towards L- than D-only bilayers,
whereas D-kalata B1 prefers membranes composed of D-
phospholipids. Of note, the affinity of L-kalata B1 for L/L-
POPC/POPE is identical to that of D-kalata B1 for D/D-
POPC/POPE, as indicated by the dose-response curves
(Figure 6b and Table S1), showing that kalata B1 enantiomers
prefer membranes of their own chirality.
2
5-26, 32
using methodologies previously optimised
. Both
peptide enantiomers have weak affinity for membranes
composed of D-POPC and their affinity improves with
increasing proportions of D-POPE, as exemplified by
sensorgrams and dose response curves obtained with D-kalata
B1 (Figure 6a, Figure S1 and fitted parameters in Table S1).
These results show that L- and D-kalata B1 retain selectivity
for PE-phospholipids regardless of lipid chirality, and indicate
that lipid chirality does not affect their general mode of
membrane-binding.
We further explored the importance of specific
phospholipid chirality by examining the activity of L- and D-
kalata B1 in membranes with mixed chirality (i.e. L/D- and
D/D-(POPC/POPE) (Figure 6a,b, Figure S1 and Table S1).
The binding affinity of D-kalata B1 to model membranes
follows the trend: D/D- > L/L- > D/L- > L/L-POPC/POPE,
whereas L-kalata B1 binding affinity has the opposite trend.
These results suggest the chirality of the PE-phospholipid and
of the membrane environment are both important for the
overall activity of the peptide.
The relevance of membrane chirality to the binding of
kalata B1 was investigated by comparing binding affinities for
model membranes composed of mixtures of POPC/POPE (4:1
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
Figure 6. Binding to and disruption of model membranes induced by L- and D-kalata B1. (a, b) Membrane-binding was followed using SPR:
peptides were injected over bilayers deposited onto an L1 chip for 180 s (association phase) and the dissociation was followed for 600 s.
Peptide-to-lipid ratio (P/L mol/mol) was calculated to normalise the response to the peptide molecular weight and total amount of lipid
deposited. (a) Sensorgrams obtained with binding of 32 M D-kalata B1 to model membranes composed of mixtures of POPC with POPE,
with L- or D-chirality. Full lines were obtained over a D-POPC bilayer, and with increasing molar ratios of D-POPE (D-POPC/ D-POPE at 9:1
and 4:1); dashed and dotted lines were obtained with membranes composed of 20% POPE: L/L-POPC/POPE, L/D-POPC/POPE, D/L-
POPC/POPE (4:1 molar ratio). (b) Dose-response curves obtained with L- or D-kalata B1 injected over L/L-POPC/POPE (4:1, left) or D/D-
POPC/POPE (4:1, right) membranes. Curves show P/L obtained at the end of injection (t = 170 s) plotted versus concentration of peptide
sample. Parameters of fitted dose response are in Table S1. (c) Percentage of vesicle leakage induced by L-kalata B1, D-kalata B1, or varying
mixtures of the two enantiomers. Melittin was included as a control. Peptides (0–10 M) were incubated with large unilamellar vesicles (5
M) loaded with carboxyfluorescein and leakage calculated by measuring the fluorescence of the dye after 15 min of incubation. The
concentration required to disrupt 50% of the vesicles (LC50) were determined by fitting curves using Graphpad Prism 7.0d
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Disruption of mirror image model membranes by L- and
D-kalata B1. In previous studies we have shown that L-kalata
2
5
B1 disrupts bilayers of model membranes. In the current
study we compared L-, D-kalata, and mixtures of the two
enantiomers (i.e., L-/D-kalata B1 at 25:75, 50:50, and 75:25)
in their efficiency to disrupt large unilamellar vesicles (LUVs)
composed of L/L-POPC/POPE or D/D-POPC/POPE (Figure
6
c). LUVs were incubated with increasing concentrations of
L-, D-kalata B1, or mixtures of the enantiomers and the
percentage of vesicle leakage was followed by
25
carboxyfluorescein dequenching. Lipid bilayer disruption of
L/L-POPC/POPE vesicles by the peptide mixtures (see peptide
concentration required to induce leakage of vesicles in Figure
6c) follows the same trend as observed for the membrane-
3
2
binding affinity and toxicity, i.e. L-kalata B1> L/D-kalata B1
(
(
75:25) > L/L-kalata B1 (50:50) > D-kalata B1 > L/D-kalata B1
25:75). This trend in disruption efficiency is mirrored by D-
kalata B1 and D/D-POPC/POPE vesicles, i.e. D-kalata B1>
L/D-kalata B1 (25:75)> L/D-kalata B1 (50:50) > L-kalata B1 >
L/D-kalata B1 (75:25). These results not only show that L- and
D-kalata B1 have the same membrane-binding properties,
ability to disrupt membranes and identical mode of action, but
also emphasise the importance of lipid chirality in the activity
of kalata B1.
Figure 7. Insertion and interaction of kalata B1 analogues with
model membranes. (a) Changes in membrane dipolar potential of
liposomes composed of L/L-POPC/POPE or D/D-POPC/POPE in
the presence of 50 M L- or D-kalata B1 as followed with
differential fluorescence spectra of di-8-ANEPPS. (b) Binding of
L-kalata B1 injected obtained over L-POPC/L-POPE (4:1)
membranes tested at different pH values as followed using SPR.
(c) Binding of L-kalata B1, [W23W-OH]kB1 and [T16S]kB1 to
bilayers composed of L/L-POPC/POPE or D/D-POPC/POPE
Insertion of L- and D-kalata B1 into mirror image model
membranes. To examine insertion of L- and D-kalata B1 into
bilayers composed of L- or D-phospholipids, we prepared
LUVs composed of L/L-POPC/POPE or D/D-POPC/POPE and
2
5
the Di-8-ANEPPS label. The fluorescence excitation of Di-
-ANEPPS dye was monitored in the absence and presence of
8
(4:1).
peptide, as was done with cells (explained above). As shown
in Figure 7a, L-kalata B1 induces a blue shift in the excitation
spectrum of the dye when incorporated into L/L-POPC/POPE
vesicles, whereas D-kalata B1 has a weak effect when added
to these model membranes. When the experiment was
conducted with D/D-POPC/POPE vesicles, the analogues have
the opposite effect: D-kalata B1 induces a larger blue shift in
the fluorescence excitation of Di-8-ANEPPS than L-kalata
B1. These results indicate that L-kalata B1 inserts deeper into
lipid bilayers composed of L-phospholipids compared to D-
kalata B1, whereas D-kalata B1 inserts deeper into bilayers
composed of D-phospholipids. These results suggest kalata B1
insertion into lipid bilayers involves interactions with the
chiral region of the phospholipid.
Targeting of PE-headgroups and insertion of kalata B1
into chiral lipid membranes. In previous studies, we
proposed that binding of kalata B1 to lipid membranes
requires the concomitant ability to specifically target PE-
headgroups (by residues in its bioactive patch) and to insert
into the hydrophobic region of the membrane (by residues in
its hydrophobic patch, see Figure 1).
To establish the importance of the bioactive face of kalata
B1 in targeting the PE-headgroup, we compared the
membrane-binding affinity of L-kalata B1 to L/L-POPC/POPE
across a range of pH values: lowering the pH protonates the
side chain of Glu7, which is in the center of the bioactive face
(Figure 1b). Results from previous studies conducted at pH
7
.4 suggest that the positively charged Arg28 and the
negatively charged Glu7 side chains of kalata B1 coordinate
electrostatic interactions with the phosphate and ammonium
2
5, 30
moieties of PE-lipids, respectively.
The membrane-
binding affinity of L-kalata B1 did not change significantly
when tested at pH 7.4, 5.5 or 4, but was reduced at pH 3.5 and
below, and no binding was detected at pH 2 (Figure 7b and
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
Figure S2a). Kalata B1 is structurally stable at acidic pH, as effect on the affinity of [T16S]kB1 (see Figure 7c and Table
5
2-53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
shown by previous studies conducted at pH <1,
and the
S2). These results support the hypothesis that the bioactive
face interacts with the PE-headgroup and its interactions are
achiral, whereas insertion of the hydrophobic patch is
dependent on membrane chirality.
mixture POPC/POPE forms a stable lipid bilayer in the
conditions and pH range tested, as suggested by the observed
deposition levels onto the L1 sensor chip (see Figure S2b).
Thus, alterations in the membrane-binding properties are
Binding and insertion of L- and D-kalata B1 into achiral
micelles. To further examine whether membrane chirality is
important for the insertion of the hydrophobic patch and
explain differences in the activity of L- and D-kalata B1, we
used NMR spectroscopy to study the interaction between
kalata B1 enantiomers and DPC micelles (Figure 8), an achiral
lipid-like system. Both L- and D-kalata B1 were found to bind
to DPC micelles in an identical manner, involving surface-
exposed hydrophobic residues, and in particular Trp23 located
in the hydrophobic patch, as previously indicated in a study
likely due to protonation of the side chain of Glu (pK
a
= 4.25)
as the pH drops below 4, combined with the protonation of the
5
4
a
PE phosphate group (pK = 1.7) at pH 2, which render kalata
B1 unable to recognise the PE headgroup and therefore unable
to bind to lipid membranes.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To examine whether chirality is involved in either
recognition of PE-headgroups or insertion into membranes,
we selected two bioactive L-kalata B1 analogues, known to
possess lower membrane-binding affinity than L-kalata B1
and with a single conservative mutation in either the bioactive
patch, [T16S]kB1, or the hydrophobic patch, [W23W-
4
4
on L-kalata B1. Here, changes of the Trp23 Hɛ1 chemical
1
shift, translational diffusion coefficient, and backbone
H
2
7
OH]kB1 (see Figure1b); each of these parts of the peptide
bind to a different region of the PE-phospholipid. As the PE-
headgroup is achiral, and the chirality of the membrane results
from the glycerol center (see Figure 5), we expected lipid
chirality to affect their membrane-binding properties
differently. We examined the membrane-binding affinity of
chemical shifts with changes in DPC concentration were
essentially identical between L- and D-kalata B1 (Figure 8b-
d). Furthermore, paramagnetic relaxation measurements to
monitor depth of insertion confirm that the binding orientation
of L- and D-kalata B1 are identical (Figure 8e). These results
reveal that when the binding partner of kalata B1 is achiral
and does not contain PE-head groups, changes in chirality of
kalata B1 do not affect insertion of its hydrophobic patch,
reiterating the point that membrane chirality is responsible for
the differences in lipid binding and bioactivity between kalata
B1 enantiomers.
[T16S]kB1 and [W23W-OH]kB1 for membranes composed
of L/L-POPC/POPE, or of D/D-POPC/POPE, and confirmed
that both analogues have lower affinity for these membranes
(Figure 7c), compared to the parent L-kalata B1. Interestingly
lipid chirality has a more pronounced effect on the membrane-
binding affinity of [W23W-OH]kB1, and has only a weak
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
1
Figure 8. Interaction of L- and D-kalata B1 with dodecylphosphocholine (DPC). (a) H 1D NMR spectra of L- and D-kalata B1, each at
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2 2
0.4 mM in H O (10% D O), pH 2.8, 25°C. (b) The Hε1 proton peak of W23 of L- and D-kalata B1 at various ratios of DPC to peptide. (c)
Chemical shift of the W23 Hε1 proton (ppm) and translational diffusion coefficient (relative to original value) are shown versus the
DPC/peptide ratio for each enantiomer. (d) Changes in chemical shifts upon incorporation into DPC micelles. The black and grey bars for
each residue denote changes in Hα and HN chemical shifts, respectively. (e) Attenuation of intensities of selected cross-peaks (Hα-HN for
residues except prolines, Hα-Hδ for prolines) in 100 ms NOESY spectra of kB1/DPC complex by 0.4 mM 16-doxylstearate at 40°C.
The current study provides a paradigm shift in our
understanding
of
peptide-membrane
interactions,
DISCUSSION
demonstrating for the first time that the chirality of lipid
bilayers can modulate peptide-lipid interactions and modify
associated bioactivity. As lipid chirality might affect peptides’
ability to insert, disrupt and/or to cross cell membranes, it is
worth revisiting and reanalysing chiral effects on peptide-lipid
interactions for various pharmaceutically relevant activities.
This is particularly important because the use of all-D-
peptides and replacement of individual residues with D-amino
acids are strategies often used to improve activity and/or
Comparison of a native peptide (L-form) with its mirror
image (D-form) has long been thought to be an elegant
approach to differentiate membrane mediated versus receptor-
mediated modes-of-action. This approach assumes that
peptide-lipid interactions within cell membranes are
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
5
independent of chirality. Our previous studies with the D-
3
2
form of kalata B1 highlighted an unusual behaviour and led
us to investigate whether peptide-lipid interactions are indeed
chirality-independent. Of the numerous peptides that bind to
membranes, kalata B1 is an excellent test case because its
binding to phospholipids involves distinct parts of the
molecule and can be used to isolate chiral contributions and
investigate the importance of lipid chirality on the activity of
peptides.
5
6-58
stability of pharmaceutically relevant peptides.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website.
Here we compared the bioactivities of L- and D-kalata B1
towards a range of cell lines, and their membrane-binding
properties using chiral model membranes (see Figures 3, 4, 6,
7). We showed that L-kalata B1 binds, inserts and disrupts
model membranes composed of L-phospholipids, more
efficiently than membranes composed of D-phospholipids
Synthesis and characterization of phospholipids with non-natural
chirality, SPR sensorgrams and fitted parameters (PDF)
AUTHOR INFORMATION
(see Figures 6,7). These properties and potency were mirrored
Corresponding Authors
by D-kalata B1, which shows preference for model
membranes composed of D-phospholipids. These results
unequivocally show that the lower bioefficacy of D-kalata B1,
when compared to L-kalata B1, is modulated by the lipid
bilayer chirality.
* sonia.henriques@qut.edu.au
*
d.craik@uq.edu.au
Present Addresses
Hayden Peacock current address: Department of Chemistry,
Graduate School of Science, The University of Tokyo, Tokyo
113-0033, Japan.
Kalata B1 binds to lipid bilayers via two distinct regions of
the peptide that interact with different parts of the
25-26
phospholipid.
SPR and NMR studies with model
membranes and kalata B1 analogues revealed that recognition
of PE-headgroups is mediated by the bioactive face in a
process that is independent of chirality, whereas insertion into
the bilayer is dependent on the hydrophobic patch and is
modulated by the chiral environment created by the
phospholipids (see Figure 7b,c & Figure 8). These studies
support the notion that in addition to concomitant targeting of
a specific phospholipid headgroup and insertion into the
hydrophobic region of the lipid bilayer, kalata B1 inserts at
least as deep as the location of the chiral moiety within the
phospholipid bilayer (see Figure 5).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The project was supported by a National Health Medical
Research Council (NHMRC) project grant (APP1084965).
S.T.H. is an Australian Research Council (ARC) Future Fellow
(FT150100398), D.J.C. is an ARC Australian Laureate Fellow
(FL150100146). The Translational Research Institute is
supported by a grant from the Australian Government. The
authors thank Dr Thomas Durek (The University of Queensland)
for valuable discussions, Mr Olivier Cheneval (The University of
Queensland) for synthesis of d-kalata B1, Dr Yen-Hua Huang
(The University of Queensland) for help with the purification and
extraction of native kalata B1, and with synthesis of kalata B1
analogues, Dr Robert Reid (The University of Queensland) for
his support with the phospholipid synthetic chemistry, David
Edwards (The University of Queensland) for assistance with
HPLC analysis and Alun Jones (The University of Queensland)
for acquiring high-resolution mass spectra of the phospholipids
and synthetic intermediates.
After the landmark study by Merrifield and colleagues
15
comparing L- and D-forms of helical antimicrobials, it is
normally assumed that membrane-active peptides, and their
mirror images have identical activity. This was also observed
by us in a recent study with beta-sheet antimicrobial
1
7
peptides. However, a closer look to the literature reveals this
is not true for all the membrane-active peptides: for instance,
the human defensins HNP1 and HD5 and their mirror images
were found to have identical activity against E. coli, but not
against S. aureus: L-forms of HNP1 and HD5 were more
potent in killing S. aureus than their enantiomers. The authors
of that study proposed the existence of an alternative
membrane-independent mechanism in the activity against S.
ABBREVIATIONS
55
aureus involving an unidentified cellular chiral component.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
PE, phosphatidylethanolamine; CD, Circular Dichroism; RBCs,
human red blood cells; PBMCs, peripheral blood mononuclear
cells; PBS, phosphate buffered saline; TX-100, Triton X-100;
(16) Suchyna, T. M.; Tape, S. E.; Koeppe, R. E., 2nd; Andersen, O. S.;
Sachs, F.; Gottlieb, P. A., Bilayer-dependent inhibition of
mechanosensitive channels by neuroactive peptide enantiomers. Nature
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
004, 430, 235-40.
(R)-POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine;
(17) Wang, C. K.; King, G. J.; Conibear, A. C.; Ramos, M. C.;
L-POPC,
R)-POPE,
1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine;
1-palmitoyl-2-oleoyl-sn-glycero-3-
Chaousis, S.; Henriques, S. T.; Craik, D. J., Mirror Images of
(
Antimicrobial Peptides Provide Reflections on Their Functions and
Amyloidogenic Properties. J. Am. Chem. Soc. 2016, 138, 5706-13.
(18) Najjar, K.; Erazo-Oliveras, A.; Brock, D. J.; Wang, T. Y.; Pellois, J.
P., An l- to d-Amino Acid Conversion in an Endosomolytic Analog of
the Cell-penetrating Peptide TAT Influences Proteolytic Stability,
Endocytic Uptake, and Endosomal Escape. J. Biol. Chem. 2017, 292,
phosphoethanolamine; l-POPE, -palmitoyl-2-oleoyl-sn-glycero-
-phosphoethanolamine; (S)-POPC, 2-oleoyl-3-palmitoyl-sn-
3
glycero-1-phosphocholine; D-POPC, 2-oleoyl-3-palmitoyl-sn-
glycero-1-phosphocholine; (S)-POPE, 2-oleoyl-3-palmitoyl-sn-
glycero-1-phosphoethanolamine; D-POPE, 2-oleoyl-3-palmitoyl-
sn-glycero-1-phosphoethanolamine; L-kB1, L-kalata B1; D-kB1,
D-kalata B1; DPC, dodecylphosphocholine; SUVs, small
unilamellar vesicles; LUVs, large unilamellar vesicles; SPR,
surface plasmon resonance.
8
47-861.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(19) Wender, P. A.; Mitchell, D. J.; Pattabiraman, K.; Pelkey, E. T.;
Steinman, L.; Rothbard, J. B., The design, synthesis, and evaluation of
molecules that enable or enhance cellular uptake: peptoid molecular
transporters. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 13003-8.
(20) Marelli, U. K.; Bezencon, J.; Puig, E.; Ernst, B.; Kessler, H.,
Enantiomeric cyclic peptides with different Caco-2 permeability suggest
carrier-mediated transport. Chemistry 2015, 21, 8023-7.
REFERENCES
(1) Lau, J. L.; Dunn, M. K., Therapeutic peptides: Historical
perspectives, current development trends, and future directions. Bioorg.
Med. Chem. 2018, 26, 2700-2707.
(21) Troeira Henriques, S.; Craik, D. J., Cyclotide Structure and
Function: The Role of Membrane Binding and Permeation.
Biochemistry 2017, 56, 669-682.
(
2) Wang, C. K.; Craik, D. J., Designing macrocyclic disulfide-rich
peptides for biotechnological applications. Nat. Chem. Biol. 2018, 14,
17-427.
3) Touti, F.; Gates, Z. P.; Bandyopadhyay, A.; Lautrette, G.; Pentelute,
(
22) Weidmann, J.; Craik, D. J., Discovery, structure, function, and
applications of cyclotides: circular proteins from plants. J. Exp. Bot.
016, 67, 4801-12.
4
(
2
(23) Craik, D. J.; Daly, N. L.; Bond, T.; Waine, C., Plant cyclotides: A
unique family of cyclic and knotted proteins that defines the cyclic
cystine knot structural motif. J. Mol. Biol. 1999, 294, 1327-36.
B. L., In-solution enrichment identifies peptide inhibitors of protein-
protein interactions. Nat. Chem. Biol. 2019, 15, 410-418.
(4) 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. Engl. 2015, 54, 8896-
(24) Huang, Y. H.; Colgrave, M. L.; Clark, R. J.; Kotze, A. C.; Craik, D.
J., Lysine-scanning mutagenesis reveals an amendable face of the
cyclotide kalata B1 for the optimization of nematocidal activity. J. Biol.
Chem. 2010, 285, 10797-805.
9
27.
(5) Cunningham, A. D.; Qvit, N.; Mochly-Rosen, D., Peptides and
peptidomimetics as regulators of protein-protein interactions. Curr.
Opin. Struct. Biol. 2017, 44, 59-66.
(25) Henriques, S. T.; Huang, Y. H.; Castanho, M. A.; Bagatolli, L. A.;
Sonza, S.; Tachedjian, G.; Daly, N. L.; Craik, D. J.,
Phosphatidylethanolamine binding is a conserved feature of cyclotide-
membrane interactions. J. Biol. Chem. 2012, 287, 33629-43.
(6) Mourtada, R.; Herce, H. D.; Yin, D. J.; Moroco, J. A.; Wales, T. E.;
Engen, J. R.; Walensky, L. D., Design of stapled antimicrobial peptides
that are stable, nontoxic and kill antibiotic-resistant bacteria in mice.
Nat. Biotechnol. 2019, 37, 1186-1197.
(26) Henriques, S. T.; Huang, Y. H.; Rosengren, K. J.; Franquelim, H.
G.; Carvalho, F. A.; Johnson, A.; Sonza, S.; Tachedjian, G.; Castanho,
M. A.; Daly, N. L.; Craik, D. J., Decoding the membrane activity of the
cyclotide kalata B1: the importance of phosphatidylethanolamine
phospholipids and lipid organization on hemolytic and anti-HIV
activities. J. Biol. Chem. 2011, 286, 24231-41.
(27) Troeira Henriques, S.; Huang, Y. H.; Chaousis, S.; Wang, C. K.;
Craik, D. J., Anticancer and toxic properties of cyclotides are dependent
on phosphatidylethanolamine phospholipid targeting. ChemBioChem
(7) Torcato, I. M.; Huang, Y. H.; Franquelim, H. G.; Gaspar, D.; Craik,
D. J.; Castanho, M. A.; Troeira Henriques, S., Design and
characterization of novel antimicrobial peptides, R-BP100 and RW-
BP100, with activity against Gram-negative and Gram-positive bacteria.
Biochim. Biophys. Acta 2013, 1828, 944-55.
(
8) Henriques, S. T.; Melo, M. N.; Castanho, M. A., Cell-penetrating
peptides and antimicrobial peptides: how different are they? Biochem. J.
006, 399, 1-7.
9) Herce, H. D.; Garcia, A. E.; Cardoso, M. C., Fundamental molecular
2
014, 15, 1956-65.
2
(
(28) Wang, C. K.; Wacklin, H. P.; Craik, D. J., Cyclotides insert into
lipid bilayers to form membrane pores and destabilize the membrane
through hydrophobic and phosphoethanolamine-specific interactions. J.
Biol. Chem. 2012, 287, 43884-98.
mechanism for the cellular uptake of guanidinium-rich molecules. J.
Am. Chem. Soc. 2014, 136, 17459-67.
(
10) Lian, W.; Jiang, B.; Qian, Z.; Pei, D., Cell-permeable bicyclic
peptide inhibitors against intracellular proteins. J. Am. Chem. Soc. 2014,
36, 9830-3.
(29) Cranfield, C. G.; Henriques, S. T.; Martinac, B.; Duckworth, P.;
Craik, D. J.; Cornell, B., Kalata B1 and Kalata B2 Have a Surfactant-
Like Activity in Phosphatidylethanolomine-Containing Lipid
Membranes. Langmuir 2017, 33, 6630-6637.
1
(11) 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. Engl.
(30) Grage, S. L.; Sani, M. A.; Cheneval, O.; Henriques, S. T.; Schalck,
C.; Heinzmann, R.; Mylne, J. S.; Mykhailiuk, P. K.; Afonin, S.;
Komarov, I. V.; Separovic, F.; Craik, D. J.; Ulrich, A. S., Orientation
and Location of the Cyclotide Kalata B1 in Lipid Bilayers Revealed by
Solid-State NMR. Biophys. J. 2017, 112, 630-642.
(31) Colgrave, M. L.; Kotze, A. C.; Huang, Y. H.; O'Grady, J.;
Simonsen, S. M.; Craik, D. J., Cyclotides: natural, circular plant
peptides that possess significant activity against gastrointestinal
nematode parasites of sheep. Biochemistry 2008, 47, 5581-9.
(32) Sando, L.; Henriques, S. T.; Foley, F.; Simonsen, S. M.; Daly, N.
L.; Hall, K. N.; Gustafson, K. R.; Aguilar, M. I.; Craik, D. J., A
Synthetic mirror image of kalata B1 reveals that cyclotide activity is
independent of a protein receptor. ChemBioChem 2011, 12, 2456-62.
2
018, 57, 14414-14438.
(12) Castle, M.; Nazarian, A.; Yi, S. S.; Tempst, P., Lethal effects of
apidaecin on Escherichia coli involve sequential molecular interactions
with diverse targets. J. Biol. Chem. 1999, 274, 32555-64.
(13) Dutta, S.; Foley, A. R.; Warner, C. J. A.; Zhang, X.; Rolandi, M.;
Abrams, B.; Raskatov, J. A., Suppression of Oligomer Formation and
Formation of Non-Toxic Fibrils upon Addition of Mirror-Image
Abeta42 to the Natural l-Enantiomer. Angew. Chem. Int. Ed. Engl. 2017,
5
6, 11506-11510.
(14) McGrath, D. M.; Barbu, E. M.; Driessen, W. H.; Lasco, T. M.;
Tarrand, J. J.; Okhuysen, P. C.; Kontoyiannis, D. P.; Sidman, R. L.;
Pasqualini, R.; Arap, W., Mechanism of action and initial evaluation of
a membrane active all-D-enantiomer antimicrobial peptidomimetic.
Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 3477-82.
(15) Wade, D.; Boman, A.; Wahlin, B.; Drain, C. M.; Andreu, D.;
Boman, H. G.; Merrifield, R. B., All-D amino acid-containing channel-
forming antibiotic peptides. Proc. Natl. Acad. Sci. U. S. A. 1990, 87,
(
33) Pereto, J.; Lopez-Garcia, P.; Moreira, D., Ancestral lipid
biosynthesis and early membrane evolution. Trends Biochem. Sci. 2004,
9, 469-77.
34) Plan, M. R.; Goransson, U.; Clark, R. J.; Daly, N. L.; Colgrave, M.
2
(
L.; Craik, D. J., The cyclotide fingerprint in oldenlandia affinis:
elucidation of chemically modified, linear and novel macrocyclic
peptides. ChemBioChem 2007, 8, 1001-11.
4
761-5.
ACS Paragon Plus Environment

Page 11 of 11
Journal of the American Chemical Society
(
35) Daly, N. L.; Clark, R. J.; Craik, D. J., Disulfide folding pathways of
(48) Duralski, A. A.; Spooner, P. J. R.; Watts, A., Synthesis of
Optically-Active Poly-Unsaturated Diacylglycerols. Tetrahedron Lett.
1989, 30, 3585-3588.
cystine knot proteins. Tying the knot within the circular backbone of the
cyclotides. J. Biol. Chem. 2003, 278, 6314-22.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
36) Sani, M. A.; Henriques, S. T.; Weber, D.; Separovic, F., Bacteria
(49) Heathcock, C. H.; Finkelstein, B. L.; Jarvi, E. T.; Radel, P. A.;
Hadley, C. R., 1,4-Stereoselection and 1,5-Stereoselection by Sequential
Aldol Addition to Alpha,Beta-Unsaturated Aldehydes Followed by
Claisen Rearrangement - Application to Total Synthesis of the Vitamin-
E Side-Chain and the Archaebacterial-C40 Diol. J. Org. Chem. 1988,
53, 1922-1942.
(50) Lebeau, L.; Oudet, P.; Mioskowski, C., Synthesis of New
Phospholipids Linked to Steroid-Hormone Derivatives Designed for 2-
Dimensional Crystallization of Proteins. Helvetica Chimica Acta 1991,
74, 1697-1706.
(51) Ravily, V.; Gaentzler, S.; Santaella, C.; Vierling, P., Synthesis of
highly fluorinated di-O-alk(en)yl-glycerophospholipids and evaluation
of their biological tolerance. Helv. Chim. Acta 1996, 79, 405-425.
(52) Colgrave, M. L.; Craik, D. J., Thermal, chemical, and enzymatic
stability of the cyclotide kalata B1: the importance of the cyclic cystine
knot. Biochemistry 2004, 43, 5965-75.
(53) Wang, C. K.; Clark, R. J.; Harvey, P. J.; Rosengren, K. J.; Cemazar,
M.; Craik, D. J., The role of conserved Glu residue on cyclotide stability
and activity: a structural and functional study of kalata B12, a naturally
occurring Glu to Asp mutant. Biochemistry 2011, 50, 4077-86.
(54) Marsh, D., Handbook of Lipid Bilayers. 2nd Edition.; CRC Press:
Boca Raton, FL, 2013.
(55) Wei, G.; de Leeuw, E.; Pazgier, M.; Yuan, W.; Zou, G.; Wang, J.;
Ericksen, B.; Lu, W. Y.; Lehrer, R. I.; Lu, W., Through the looking
glass, mechanistic insights from enantiomeric human defensins. J. Biol.
Chem. 2009, 284, 29180-92.
(56) Liu, D.; Guo, P.; McCarthy, C.; Wang, B.; Tao, Y.; Auguste, D.,
Peptide density targets and impedes triple negative breast cancer
metastasis. Nat. Commun. 2018, 9, 2612.
(57) Miles, J. J.; Tan, M. P.; Dolton, G.; Edwards, E. S.; Galloway, S.
A.; Laugel, B.; Clement, M.; Makinde, J.; Ladell, K.; Matthews, K. K.;
Watkins, T. S.; Tungatt, K.; Wong, Y.; Lee, H. S.; Clark, R. J.; Pentier,
J. M.; Attaf, M.; Lissina, A.; Ager, A.; Gallimore, A.; Rizkallah, P. J.;
Gras, S.; Rossjohn, J.; Burrows, S. R.; Cole, D. K.; Price, D. A.; Sewell,
A. K., Peptide mimic for influenza vaccination using nonnatural
combinatorial chemistry. J. Clin. Invest. 2018, 128, 1569-1580.
(58) Garton, M.; Nim, S.; Stone, T. A.; Wang, K. E.; Deber, C. M.;
Kim, P. M., Method to generate highly stable D-amino acid analogs of
bioactive helical peptides using a mirror image of the entire PDB. Proc.
Natl. Acad. Sci. U. S. A. 2018, 115, 1505-1510.
May Cope Differently from Similar Membrane Damage Caused by the
Australian Tree Frog Antimicrobial Peptide Maculatin 1.1. J. Biol.
Chem. 2015, 290, 19853-62.
(37) Henriques, S. T.; Pattenden, L. K.; Aguilar, M. I.; Castanho, M. A.,
The toxicity of prion protein fragment PrP(106-126) is not mediated by
membrane permeabilization as shown by a M112W substitution.
Biochemistry 2009, 48, 4198-208.
(38) Cladera, J.; O'Shea, P., Intramembrane molecular dipoles affect the
membrane insertion and folding of a model amphiphilic peptide.
Biophys. J. 1998, 74, 2434-42.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(39) Vitha, M. F.; Clarke, R. J., Comparison of excitation and emission
ratiometric fluorescence methods for quantifying the membrane dipole
potential. Biochim. Biophys. Acta 2007, 1768, 107-14.
(40) Henriques, S. T.; Pattenden, L. K.; Aguilar, M. I.; Castanho, M. A.,
PrP(106-126) does not interact with membranes under physiological
conditions. Biophys. J. 2008, 95, 1877-89.
(41) Huang, Y. H.; Colgrave, M. L.; Daly, N. L.; Keleshian, A.;
Martinac, B.; Craik, D. J., The biological activity of the prototypic
cyclotide kalata b1 is modulated by the formation of multimeric pores.
J. Biol. Chem. 2009, 284, 20699-707.
(42) Wang, C. K.; Northfield, S. E.; Swedberg, J. E.; Harvey, P. J.;
Mathiowetz, A. M.; Price, D. A.; Liras, S.; Craik, D. J., Translational
diffusion of cyclic peptides measured using pulsed-field gradient NMR.
J. Phys. Chem. B 2014, 118, 11129-36.
(43) Shenkarev, Z. O.; Nadezhdin, K. D.; Lyukmanova, E. N.; Sobol, V.
A.; Skjeldal, L.; Arseniev, A. S., Divalent cation coordination and mode
of membrane interaction in cyclotides: NMR spatial structure of ternary
complex Kalata B7/Mn2+/DPC micelle. J. Inorg. Biochem. 2008, 102,
1
246-56.
(44) Wang, C. K.; Colgrave, M. L.; Ireland, D. C.; Kaas, Q.; Craik, D.
J., Despite a conserved cystine knot motif, different cyclotides have
different membrane binding modes. Biophys. J. 2009, 97, 1471-81.
(45) Wang, C. K.; King, G. J.; Northfield, S. E.; Ojeda, P. G.; Craik, D.
J., Racemic and quasi-racemic X-ray structures of cyclic disulfide-rich
peptide drug scaffolds. Angew. Chem. Int. Ed. Engl. 2014, 53, 11236-41.
(46) Gross, E.; Bedlack, R. S., Jr.; Loew, L. M., Dual-wavelength
ratiometric fluorescence measurement of the membrane dipole potential.
Biophys. J. 1994, 67, 208-16.
(
47) van Meer, G.; Voelker, D. R.; Feigenson, G. W., Membrane lipids:
where they are and how they behave. Nat. Rev. Mol. Cell Biol. 2008, 9,
12-24.
1
TOC graphic
ACS Paragon Plus Environment