Amphiphilic cationic β3R3-peptides: Membrane active peptidomimetics and their potential as antimicrobial agents

Article abstract of DOI:10.1021/bm500101w

We introduce a novel class of membrane active peptidomimetics, the amphiphilic cationic β^3R3-peptides, and evaluate their potential as antimicrobial agents. The design criteria, the building block and oligomer synthesis as well as a detailed structure-activity relationship (SAR) study are reported. Specifically, infrared reflection absorption spectroscopy (IRRAS) was employed to investigate structural features of amphiphilic cationic β^3R3-peptide sequences at the hydrophobic/hydrophilic air/liquid interface. Furthermore, Langmuir monolayers of anionic and zwitterionic phospholipids have been used to model the interactions of amphiphilic cationic β^3R3-peptides with prokaryotic and eukaryotic cellular membranes in order to predict their membrane selectivity and elucidate their mechanism of action. Lastly, antimicrobial activity was tested against Gram-positive M. luteus and S. aureus as well as against Gram-negative E. coli and P. aeruginosa bacteria along with testing hemolytic activity and cytotoxicity. We found that amphiphilic cationic β^3R3-peptide sequences combine high and selective antimicrobial activity with exceptionally low cytotoxicity in comparison to values reported in the literature. Overall, this study provides further insights into the SAR of antimicrobial peptides and peptidomimetics and indicates that amphiphilic cationic β^3R3-peptides are strong candidates for further development as antimicrobial agents with high therapeutic index.

Full text of DOI:10.1021/bm500101w

Article
pubs.acs.org/Biomac
Amphiphilic Cationic β3R3-Peptides: Membrane Active
Peptidomimetics and Their Potential as Antimicrobial Agents
Simone Mosca,† Janos Keller,‡ Nahid Azzouz,† Stefanie Wagner,§ Alexander Titz,§ Peter H. Seeberger,†
Gerald Brezesinski,‡ and Laura Hartmann*,†
†Department of Biomolecular Systems and ‡Department of Interfaces, Research Campus Golm, Max Planck Institute of Colloids and
Interfaces, 14424 Potsdam, Germany
§Helmholtz Institute for Pharmaceutical Research Saarland (HIPS), Campus C 2.3, D-66123 Saarbrucken, Germany
̈
S
* Supporting Information
ABSTRACT: We introduce a novel class of membrane active peptidomimetics, the amphiphilic cationic β3R3-peptides, and
evaluate their potential as antimicrobial agents. The design criteria, the building block and oligomer synthesis as well as a detailed
structure−activity relationship (SAR) study are reported. Specifically, infrared reflection absorption spectroscopy (IRRAS) was
employed to investigate structural features of amphiphilic cationic β3R3-peptide sequences at the hydrophobic/hydrophilic air/
liquid interface. Furthermore, Langmuir monolayers of anionic and zwitterionic phospholipids have been used to model the
interactions of amphiphilic cationic β3R3-peptides with prokaryotic and eukaryotic cellular membranes in order to predict their
membrane selectivity and elucidate their mechanism of action. Lastly, antimicrobial activity was tested against Gram-positive M.
luteus and S. aureus as well as against Gram-negative E. coli and P. aeruginosa bacteria along with testing hemolytic activity and
cytotoxicity. We found that amphiphilic cationic β3R3-peptide sequences combine high and selective antimicrobial activity with
exceptionally low cytotoxicity in comparison to values reported in the literature. Overall, this study provides further insights into
the SAR of antimicrobial peptides and peptidomimetics and indicates that amphiphilic cationic β3R3-peptides are strong
candidates for further development as antimicrobial agents with high therapeutic index.
resistances,3,4 the development of novel antibiotics with new
modes of action, such as antimicrobial peptides, is required.
However, despite intense research efforts aimed at developing
INTRODUCTION
■
Bacterial resistance to conventional antibiotics has been
recognized as one of the greatest public health threats in the
21st century.1 Thus, alternative strategies and new molecules
AMPs as therapeutic agents, limited clinical outcome has
resulted so far.5,6 A major disadvantage of natural AMPs is their
for antimicrobial treatment are a major focus of synthetic and
protease liability related to the peptide formula per se that
pharmaceutical chemistry. Cationic antimicrobial peptides
(AMPs) form part of the innate immune response to microbial
dominates their unfavorable pharmacokinetic profile.7,8 In
addition, most of natural AMPs display either modest direct
and viral infection in many organisms and are considered a
antimicrobial activity or indiscriminant toxicity to prokaryotic
potential source of future antibiotics because of their particular
and eukaryotic cells.7,9 In order to overcome these major
mechanism of action compared to conventional antibiotics.
limitations and to realize the therapeutic potential of AMPs,
AMPs are a major class of membrane-active peptides (MAPs)
different classes of antimicrobial peptidomimetics have been
which strongly interact with negatively charged microbial
surfaces leading to perturbation and disruption of membranes.
developed with promising results.10−14 In particular, amphi-
Thereby, and in contrast to conventional antibiotics, AMPs can
rapidly kill a broad-spectrum of microbes with a decreased
likelihood of resistance development.2 Besides antivirulence
strategies that are pursued to avoid rapid developemt of
Received: January 21, 2014
Revised: March 30, 2014
Published: April 3, 2014
© 2014 American Chemical Society
1687
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
Figure 1. Synthetic strategy applied for the synthesis of β3R3-peptides. (a) β-diamine and β-diacid monomers were synthesized and coupled in
solution affording a small library of dimer building blocks (1−9). (b) These building blocks were employed for the solid phase synthesis of
amphiphilic cationic β3R3-peptide oligomers (10−19).
philic cationic β-peptides with helical conformation have
prompted expectations,13−15 as they can reproduce the
physicochemical features of AMPs but exhibit improved
proteolytic stability.16,17 Such β-peptides have proven good
antimicrobial activity,18−22 but still exhibit significant hemolytic
activity,18,20−22 that indicates indiscriminant cytotoxicity and
substantially hampers the development of these molecules as
therapeutics. Recently, we established the β3R3-peptides as a
novel class of peptidomimetics introducing alternating
directions of the amide bonds along β-peptide sequences and
extending the structural space available to β-peptides while
maintaining their enzymatic stability.23 Here, we extend this
class of peptidomimetics and present for the first time
amphiphilic cationic β3R3-peptides. The novel qualities of the
correlations between primary sequence, physicochemical and
biological properties of the amphiphilic cationic β3R3-peptides
and to derive a first set of design rules for the next generation of
antimicrobial peptidomimetics. Specifically, we report on the
design criteria, the building block and oligomer synthesis, and
the physicochemical and biological characterization of
amphiphilic cationic β3R3-peptide sequences and investigate
their potential as membrane active peptidomimetics and
antimicrobial agents.
MATERIALS
■
Commercial grade reagents and solvents were used as purchased
without further purification, except as indicated below. Fmoc-α-amino
acids, Boc-L-Asp-OBn, and Boc-D-Asp-OBn were purchased from Iris
Biotech, HATU, PyBOP and HOBt from Nova Biochem, and all other
reagents from Aldrich; all were used without further purification. All
solvents were HPLC grade. Regarding DCM, amylene-stabilized
HPLC grade was chosen. The solid-support resins were purchased
from Rapp Polymers. Tentagel S RAM resin (loading 0.23 mmol/g)
(Fmoc-protected) was used as purchased.
β
3R3-peptide backbone and the combination of proteinaceous
side chains allow for fine-tuning of the physicochemical
properties and thus also should allow for fine-tuning of their
membrane activity. Therefore, such amphiphilic cationic β3R3
-
peptides can be expected to have great potential as membrane
active peptidomimetics and antimicrobial agents. Complete
control over the building blocks’ chemical features as well as
oligomer assembly allows us to perform a detailed structure−
activity relationship (SAR) study aiming at disclosing
1688
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
RESULTS AND DISCUSSION
■
Building Block Design and Synthesis. Despite their
diversity in primary sequence and secondary structures, AMPs
generally share a cationic and globally amphiphilic character24
that is essential for the mode of action of AMPs and the ability
of membrane active peptidomimetics to interact with a
hydrophobic and negatively charged biomembrane.2,25 There-
fore, these physicochemical properties are also key features of
amphiphilic cationic β3R3-peptides. The synthesis of β3R3
-
peptides is generally based on the solid phase coupling of dimer
building blocks of β-diamines and β-diacids.23 Here, we
introduce a novel set of building blocks suitable for standard
solid phase peptide chemistry introducing both, cationic and
hydrophobic side chains. In total nine different dimer building
blocks were synthesized presenting variations on their side
chains and chiral configurations (Figure 1).
Figure 2. Amphiphilic cationic β3R3-peptide oligomers synthesized and
investigated in this study. First, 7 homochiral (S,S) oligomers made of
9 dimer building blocks and different primary sequence (10−16) have
been synthesized. The heterochiral (S,R) analogue (17) of sequence
14 was prepared. Lastly, shorter (18) and longer (19) analogues of
sequence 14 have been synthesized made of 5 and 13 dimer building
blocks, respectively.
Earlier we presented an efficient synthetic protocol to
produce Fmoc protected chiral β-diamines with proteinaceous
side chains.26 This route was applied using L-Ala, L-Val, and L-
Leu as starting materials.23 Here, this protocol was extended to
L-Phe. The nomenclature of β3R3-peptides monomers refers to
the v/s prefix system in which vXaa symbolizes the vicinal
diaminoalkyl analogue of the indicated amino acid residue Xaa
and sXaa symbolizes succinyl analogues. All diamine subunits
were synthesized in 30 g scale with overall yield ranging from
39% (vAla) to 44% (vPhe). In addition, the nonchiral analogue
of Gly (vGly) was prepared via a novel, convenient route that
afforded FmocNH-vGly-NH2·TFA in large scale (50 g) and
with high yield (75%). A major synthetic advantage of using β-
diamines and β-diacids as building blocks is the pool of
commercially available chiral β-diacids with functionalized side
chains, such as aspartic acid (Asp). Commercially available Boc-
L-Asp-OBn and Boc-D-Asp-OBn were exploited as precursors of
the succinyl analogues S-amino succinic acid ((S)sNH2) and R-
amino succinic acid ((R)sNH2), as Boc and Benzyl (Bn) groups
are orthogonal to the Fmoc-protecting group of the β-diamine
counterparts. Referring to our dimer strategy,26−29 the Fmoc
protected chiral β-diamines and the monoprotected diacid Boc-
Asp-OBn were combined in solution using a PyBOP/HOBt-
mediated coupling reaction in DMF with DIEA as base.
Notably, we developed a novel purification procedure based on
sequential precipitations that removed hydrophilic and hydro-
phobic impurities efficiently to provide the products in high
purity while avoiding time-consuming chromatographic purifi-
cation. Thus, this procedure allows for performing different
reactions in parallel and easy scale up. In the next step, the
benzyl ester (OBn) was deprotected via catalytic hydrogenation
with Pd on carbon (Pd/C) in DMF. Finally, precipitation in
Et2O/Hex mixture afforded the desired products. Conclusively,
all synthetic routes fulfill the requirements for the synthesis of
building blocks suitable for solid phase synthesis, employing
readily available starting materials and giving high yields on
multigram scale. These procedures rely exclusively on simple
and convenient purification methods and no chromatographic
purification was needed during the 8 steps synthesis. Finally, all
building blocks (1−9) were obtained in high purity (>99%), as
shown by NMR and HPLC (see the Supporting Information).
Oligomer Synthesis. The library of dimer building blocks
(1−9) was then directly employed for solid phase synthesis of
ten different amphiphilic cationic β3R3-peptide sequences (10−
19) (Figure 2) differing in their chain length, side chain pattern
and chiral configuration. The β3R3-peptides were synthesized via
automated solid-phase synthesis on a standard peptide
synthesizer and Fmoc amino acid coupling protocols using
double coupling with five equivalents building blocks and
coupling times of 1 h with HATU as an activation agent. At
first, the use of 25% piperidine in DMF resulted in inefficient
Fmoc deprotection, which could be attributed to a strong
tendency of the hydrophobic residues to aggregate on the solid
support.21 Extending the time of deprotection did also not
result in complete removal of Fmoc groups (see the Supporting
Information for the Fmoc cleavage UV-patterns and HPLC
traces). This challenge was successfully overcome by the use of
LiCl in 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), piperidine,
and DMF as mixture for Fmoc deprotection.30 In the last step,
the oligomers were cleaved from the resin using acidic
conditions (95% TFA, 2.5% triisopropylsilane, and 2.5%
water) releasing C-terminal carboxamides and simultaneously
deprotecting the Boc groups on the sNH2 units. All oligomers
of up to 13 dimer building blocks, a chain length corresponding
to 26 β-amino acids, were obtained in very high purity (>95%)
after precipitation in cold Et2O (see the Supporting
Information for HPLC traces). Noteworthy, for the parent
β3-peptides, the successful synthesis of peptides made up of 24
amino acids is considered to be a good indication that every
desired peptide sequence should be accessible.31 Thus, it is
remarkable that the developed synthetic strategy gives access to
long and highly pure β3R3-peptide sequences directly from
natural α-amino acids without the need for any chromato-
graphic purification along the complete synthetic route.
Oligomer Design. Besides their general cationic and
amphiphilic character, other physicochemical and structural
factors determine the membrane activity and selectivity of
natural AMPs.2 These include sequence length, net charge,
hydrophobicity, presence of aromatic side chains, as well as
conformation and aggregation properties.22 All of these
parameters were also addressed for our β3R3-peptides by
choosing different primary sequences specifically varying, for
example, the number of aromatic side chains. The development
of membrane active amphiphilic cationic β3R3-peptides with
potential antimicrobial activity relies on a rational, hierarchical
1689
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
tuning of these parameters by varying the residue composition
of the primary sequence. Therefore, 10 different sequences of
amphiphilic cationic β3R3-peptides (10−19) were designed and
synthesized. All the oligomers feature a C-terminal carboxamide
moiety that generally enhances the activity of both natural and
synthetic antimicrobial peptidic agents.19,21 First, seven
homochiral (S,S) sequences (10−16) made of nine dimer
building blocks and the same number of net charges were
synthesized (Figure 2). Similar chain lengths are usually
targeted for short AMPs with potent antimicrobial activity
and β-peptides investigated as antimicrobial peptidomimet-
ics.18−22,24 Within these oligomers, the diamine units vGly,
vAla, vVal, vLeu, and vPhe were systematically interchanged to
produce amphiphilic cationic β3R3-peptide oligomers with
variegated combinations of different proteinaceous side chains,
aliphatic and aromatic, and thus different physicochemical
features. Primarily, this set of side chains features a fine scale of
hydrophobicity as one of the major requirements for the
effective interaction with membranes and resulting antimicro-
bial activity. However, excessive hydrophobicity can also lead to
unspecific cytotoxicity and increased hemolytic activity.19,20
Furthermore, this hydrophobicity may lead to aggregation in
aqueous media and result in reduced antimicrobial activity.10
Thus, it is crucial to truly fine-tune the hydrophobic properties
of the oligomers and find the right balance of hydrophobicity as
demonstrated by the detailed SAR studies of this first set of
oligomers.
determined via RP-HPLC as percent of acetonitrile required to
elute each oligomer from a C18 analytical column (Table 1).
Table 1. Chemical Features, Hydrophobicity, and
Hydrodynamic Diameter of the Amphiphilic Cationic β3R3
-
Peptide Oligomers
b
c
e
f
g
h
oligomer
.
*
Ar
H
Ø
πeq
amide I
d
10
11
12
13
14
15
16
17
18
19
9
9
sS
sS
sS
sS
sS
sS
sS
sR
sS
sS
0
2
2
0
2
0
4
2
1
3
0
//
0.2
1.1
1666.5 1.1
1656.1 1.6
1653.6 1.5
1661.2 1.6
1653.1 1.2
1661.3 1.2
1654.0 1.0
23.2
26.6
22.1
29.9
28.7
29.9
27.9
23.2
33.9
//
9
//
3.6
9
//
1.7
9
0.7
0.7
2.5
0.9
0.8
0.8
6.8
9
8.6
9
5.4
9
13.3
0.4
1648.2 0.4
i
5
1665.4
1.5
13
11.3
1654.0 0.9
a
b
n stands for number of building blocks. Asterisk (*) refers to the
absolute configuration of stereocenter on the diacid sNH2 units.
c
Hydrophobicity measure determined as percentage acetonitrile eluent
in RP-HPLC analysis 5% to 95% MeCN in 60 min, using a C18
d
column. Oligomer 10 is eluted within the injection peak.
e
f
Hydrodynamic diameter in nm measured by DLS. The oligomer
g
was not characterized. Equilibrium surface pressure in mN/m
obtained by surface tension measurements. Error bars: 1 mN/m
h
for all pressures. Amide I band position in cm−1 observed by IRRAS.
i
Note the very weak intensity of the band to be univocally
characterized.
The incorporation of aromatic residues is another important
factor that influences the membrane activity of peptides and
peptidomimetics. Therefore, sequences 10−16 also present
different numbers of vPhe units that can result in π−π
interactions with unsaturated fatty acids to potentially differ-
entiate microbial and human cytoplasmic membranes.32
In addition, structural factors can influence biological
properties of AMPs as their membrane activity requires nascent
or induced structural motifs producing amphiphilic macro-
molecular folds with spatial segregation of charges and
hydrophobic moieties.2,22 Recently, we have shown that the
This approach has been established to measure the lipophilicity
of various molecules,33 including antimicrobial β-peptides.21 In
general, within homochiral sequences of the same chain length
(10−16), the relative values achieved can be linearly correlated
to the hydrophobicity of the proteinaceous side chains on the
diamine residues vXaa. The most hydrophobic oligomers are
compounds 14 and 16 that feature the same values of
hydrophobicity (H = 29.9), sequence length, and net charge,
and exclusively differ in the number and density of aromatic
vPhe units, with three vLeu units of oligomer 14 replaced by
three vPhe in oligomer 16 (Figure 2). Therefore, the direct
comparison of compounds 14 and 16 allows us to evaluate the
effect of π−π interactions on the conformation and aggregation
properties, and thus on the membrane and biological activity of
amphiphilic cationic β3R3-peptides.
Interestingly, the RP-HPLC analysis showed higher hydro-
phobicity for the homochiral (vS,sS) oligomer 14 than the
heterochiral (vS,sR) analogue 17 presenting the same primary
sequence. Longer retention time generally correlates with a
greater propensity to adopt a globally amphiphilic conforma-
tion, as observed for α- as well as β-peptides.21,34 Thus, this
increase in retention time can be considered an indication that
homochiral sequences might fold to result in longer retention
times, while their heterochiral analogues might adopt a more
extended conformation.
β
3R3-peptide backbone has an intrinsic propensity for an
extended conformation at the air/water interface.23 However,
homochiral amphiphilic cationic β3R3-peptides cannot segregate
cationic and hydrophobic moieties in an extended conforma-
tion and are expected to fold in order to adopt globally
amphiphilic macromolecular folds typical for cationic AMPs. In
contrast, a heterochiral sequence is expected to adopt a globally
amphiphilic structure in extended conformation. Therefore, the
heterochiral (S,R) analogue (17) of sequence 14 was prepared.
Lastly, shorter (18) and longer (19) analogues of sequence 14,
made of 5 and 13 dimer building blocks, respectively, have been
synthesized to evaluate the influence of chain length on the
biological activity of amphiphilic cationic β3R3-peptides.
Physicochemical Characterization of Amphiphilic
Cationic β3R3-Peptides. After the synthesis and chemical
characterization of the different amphiphilic cationic β3R3
-
peptide sequences, we investigated their physicochemical
properties. Specifically physicochemical parameters such as
hydrophobicity, hydrodynamic radius, aggregation behavior and
membrane activity will be determined and then correlated with
In addition, the chain length proved to have a strong
influence on the hydrophobicity of amphiphilic cationic β3R3
-
peptides as increasing the chain length leads to an increase in
hydrophobicity (compare oligomers 18 (H = 23.2), 14 (H =
29.9) and 19 (H = 33.9) featuring the same side chain pattern
and exclusively differing in the number of repeats).
Furthermore, DLS was employed to determine the size
distribution profile of the oligomers in solution (Table 1)
and their potential aggregation in aqueous media. Oligomer 14
the antimicrobial properties of the amphiphilic cationic β3R3
-
peptides in dependence of their primary sequence.
Hydrophobicity is perhaps the most critical parameter for
membrane activity and selectivity of AMPs.2,25 Therefore, the
hydrophobicity of the amphiphilic cationic β3R3-peptides was
1690
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
Table 2. Surface Tension Experiments with Oligomers 14, 17, and 19 and Model Membranes
a
oligomer
14
17
19
c
bare
πeq
6.8
11.4
7.9
13.3
16.6
13.9
11.3
14.0
12.5
b
d
POPG πinj0
POPC πinj0
Δπ
Δπ
e
POPG πinj30
POPC πinj30
ΔA
ΔA
11.8 (12.7%)
13.8 (14.4%)
20.3 (20.3%)
1.6 (1.8%)
−0.7 (−0.8%)
−1.0 (−1.1%)
a
b
Injected into the subphase to obtain a concentration of 500 nM. πinjX stands for the initial surface pressure at X mN/m of the lipid monolayer
c
d
before the peptide injection. πeq stands for surface pressure in mN/m of the equilibrated oligomer monolayer. Δπ is the surface pressure change.
e
ΔA stands for the change in surface area (cm2) and percent of the equilibrated mixed oligomer−lipid monolayer to the lipid monolayer before the
peptide injection. Error bars: 1mN/m for all pressures; 2% for areas.
served as reference sequence with proper hydrophobicity and
altered pattern of side chains. In comparison, oligomers 18 and
19 present the same side chain pattern but different chain
length. In contrast, oligomer 15 features the same chain length
but different side chains. Lastly, oligomer 17 (vS,sR) differs
from oligomer 14 (vS,sS) by the chiral configuration of the
diacid sNH2 units. Regardless of the chain length and the side
chains, the oligomers 14, 15, 17, 18, and 19 produce similar
profiles with average hydrodynamic diameters of around 0.7−
0.8 nm. In contrast, oligomer 16 displays a significantly
increased hydrodynamic diameter of around 2.5 nm that can be
attributed to the formation of aggregates. Such aggregation in
aqueous media could derive from the sequence density of
aromatic vPhe side chains that can generate both hydrophobic
and π−π interactions. Oligomer 14 features the same value of
hydrophobicity (H = 29.9) as oligomer 16 (Table 1) but
contains a lower number and density of aromatic vPhe units
(Figure 2), therefore we mainly attribute the aggregation
behavior to π−π interactions of oligomer 16. Lastly, surface
tension measurements and IRRAS were employed to evaluate
quantitative and qualitative physicochemical determinants of
the amphiphilic cationic β3R3-peptides at the air/water interface
(Table 1). Indeed, the air/water interface as the simplest
hydrophobic/hydrophilic interface can mimic the interface of
the cell membrane and thus gives a first indication on the
potential membrane activity of the β3R3-peptides.35−47
First, the equilibrium surface pressure of a 500 nM solution
in buffer (10 mM PBS at pH 7.4 with 150 mM NaCl) was
measured as an alternative mean to estimate the hydrophobicity
of the oligomers 10−19 (Table 1). A general trend parallel to
the values of hydrophobicity obtained by RP-HPLC is found.
However, a stronger influence of the chain length and number
of charges is observed compared to the hydrophobicity values
achieved by RP-HPLC. In particular, the short oligomer 18 is
not surface active. In addition, oligomer 16 shows a reduced
equilibrium surface pressure. This could correlate with the
tendency of oligomer 16 to form aggregates in solution, as seen
in DLS experiments that are less surface active. In contrast, the
heterochiral (vS,sR) oligomer 17 displays an extraordinary high
equilibrium surface pressure. This observation can be explained
by the potential to adopt an extended conformation segregating
charges and hydrophobic moieties. Therefore, oligomer 17
could form strands that can assemble and align at the air/water
interface and increase the surface concentration.
indicating intramolecular hydrogen bond networks, such as
helices. In contrast, the heterochiral (vS,sR) oligomer 17
displays the amide I band at around 1648 cm−1, typical for
intermolecular hydrogen bond networks at the air/water
interface.38,42,44,48,49 Indeed, similar values have been observed
for the hydrophobic β3R3-peptide oligomers forming strands
that self-assemble into amorphous β-sheet-like structures.23
Interaction of Amphiphilic Cationic β3R3-Peptides with
Model Membranes. Both surface tension and IRRAS shed
light on the interactions with model membranes and thus the
ability of the amphiphilic cationic β3R3-peptides to penetrate
lipid membranes.50 As model membranes, the anionic 1-
palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol)
(POPG) and the zwitterionic 1-palmitoyl-2-oleoyl-sn-glycero-3-
phosphocholine (POPC) lipids were employed to mimic
prokaryotic and eukaryotic membranes, respectively. Both
lipids have one unsaturated acyl chain and thus form stable
monolayers in the liquid disordered state even at high surface
pressures. Two sets of experiments were performed at different
surface pressures using different amounts of lipid. The
experiments were carried out on 10 mM PBS at pH 7.4 with
150 mM NaCl as aqueous subphase. In the first set of
experiments, 25 μL of a 1 mM lipid solution in CHCl3 were
spread on the air/buffer interface and the monolayer was
compressed to 30 mN/m. This surface pressure is discussed as
the internal pressure in biological membranes.51 Thereafter, the
surface pressure was kept constant and changes in the surface
area were monitored. In the second set of experiments, only 13
μL of the lipid solutions were spread and compressed to the
same trough surface as in the previous experiment but with the
surface pressure still at 0 mN/m close to the lift-off point of the
isotherm. The area was kept constant and changes in the
surface pressure were monitored. The amphiphilic cationic
β
3R3-peptide oligomers were injected under the lipid film to
obtain a concentration of 500 nM in the subphase and the
resulting surface pressure and changes in surface area were
detected, respectively (Table 2). Injection of the same amount
of pure buffer did not create any changes in the monolayer
characteristics.
In general, the equilibrium surface pressure after the injection
of the oligomers below the uncompressed zwitterionic
eukaryotic-like POPC monolayer at 0 mN/m is almost equal
to the equilibrium surface pressure of the oligomers alone
showing that the oligomer adsorption is solely driven by their
surface activity. Moreover, the area of the compressed POPC
monolayer at 30 mN/m does not change after the injection of
the oligomers, which means that no insertion of the amphiphilic
cationic β3R3-peptide into the lipid layer or adsorption to the
hydrophilic lipid head groups occurs. Differently, in the case of
To further confirm this observation and provide qualitative
structural information, the position of the amide I band in the
IRRA spectra was analyzed (Table 1). All homochiral (vS,sS)
oligomers 10−16, 18, and 19 present the amide I band at
around 1660 ( 6) cm−1 that is characteristic for conformations
1691
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
Table 3. MIC Values of the Amphiphilic Cationic β3R3
Peptide Oligomers 10−19
-
a POPG monolayer at 0 mN/m, the equilibrium surface
pressure observed after injection of the oligomers is
significantly higher than the value previously registered at the
bare buffer surface pointing to specific interactions between the
oligomers and POPG. Additionally, the amphiphilic cationic
a
oligomer
M. luteus
E. coli
10
11
12
13
14
15
16
17
18
19
no activity
no activity
no activity
no activity
10.7 μg/mL
16.6 μg/mL
25.3 μg/mL
29.2 μg/mL
no activity
5.3 μg/mL
no activity
no activity
no activity
87.4 μg/mL
β
3R3-peptides also insert into the lipid monolayer at a surface
pressure of 30 mN/m and the area of the equilibrated mixed
peptide-lipid monolayer is increased. Specifically, oligomer 19
gives the largest effect with an overall increase in surface area of
20%. Thus, the amphiphilic cationic β3R3-peptides and
especially oligomer 19 produce strong interactions with the
negatively charged prokaryotic model membrane. In contrast,
no interaction was observed with the zwitterionic eukaryotic
101.9 μg/mL
64.2 μg/mL
108.8 μg/mL
130.6 μg/mL
no activity
membrane model. Therefore, the amphiphilic cationic β3R3
-
43.2 μg/mL
a
The MIC values have been extrapolated from the plots of
peptides can be expected to display high and selective
antibacterial activity without cytotoxicity versus eukaryotic
host cells.
antimicrobial activity using linear fit (for plots of antimicrobial activity,
see the SI).
In addition, the position of the amide I band in the IRRA
spectra in the presence of the model membranes was analyzed.
Interestingly, an exclusive shift toward a higher wavenumber is
observed in the amide I band of the heterochiral (vS,sR)
oligomer 17 in the presence of the bacterial membrane
mimicking POPG indicating a change in conformation (see the
Supporting Information). Thus, similar to several natural
AMPs,2 lipid−peptide interactions seem to change the
conformational state of the heterochiral amphiphilic cationic
Gram-positive M. luteus than 14 (Tables 1 and 3). However,
oligomer 15 shows higher activity against Gram-negative E. coli
(Table 3). Thus, the aromatic vPhe units seem to play an
important role driving the membrane selectivity of amphiphilic
cationic β3R3-peptides.
From the physicochemical characterization of the oligomers
we know that also the chiral configuration of the sNH2 diacid
units seems to influence the conformation as well as the
effective hydrophobicity of the oligomers. We expect to see a
difference in antibacterial activity. Indeed, the heterochiral (vS,
sR) oligomer 17 is less active than its homochiral (vS, sS)
analogue 14 against both, M. luteus and E. coli (Table 3). This
is again in agreement with a reduced hydrophobicity and could
be also attributed to intermolecular sheet assembly of strands of
oligomer 17 resulting in reduced antibacterial activity. We
believe that this finding is in agreement with the differences
observed for helical or non helical antimicrobial peptides and
the differences in their mode of action. To draw further
conclusions, a more detailed structural analysis will be required.
We observed that the chain length and number of charges of
the amphiphilic cationic β3R3-peptides extensively influence
their antimicrobial activity. Generally antimicrobial activity
increases with increasing numbers of repeating units and
charges (Tables 1 and 3). Indeed, oligomers 14, 18, and 19 that
feature the same side chain pattern and differ exclusively in the
number of repeats (Figure 2) show very different antimicrobial
activity. While the shortest oligomer 18 is not active, the
longest oligomer 19 is the most active amphiphilic cationic
β
3R3-peptide. For the homochiral peptide of the same sequence,
no shift of the amide I band was observed. Current studies are
further evaluating this finding.
Antibacterial Activity of Membrane Active β3R3
-
Peptides. After the comprehensive physicochemical character-
ization proving the general potential of amphiphilic cationic
β
3R3-peptides for high and selective activity toward prokaryotic
membranes, antibacterial growth assays have been performed
for all oligomers (10−19). The Minimal Inhibitory Concen-
tration (MIC) values were determined against E. coli (strain
BL21) and M. luteus (type strain DSM 20030) as representative
strains of both, Gram-negative and Gram-positive bacteria.
Gram-positive bacteria have a thick peptidoglycan layer on the
outside of the cell membrane while Gram-negative bacteria
have a thick or thin peptidoglycan layer that is located between
two cell membranes. Thus, differences can be expected for the
membrane interactions with AMPs in dependence of the
bacterial strain. Indeed, we see a strong dependency of the
monomer sequence as well as the targeted organism on the
resulting antibacterial activity (Table 3).
In general, the amphiphilic cationic β3R3-peptides show
higher activity against Gram-positive M. luteus. Within
sequences of the same chain length of nine repeating units
(10−17), differences in hydrophobicity showed a strong effect
on the antibacterial activity. The most hydrophobic oligomers
are 14, 15, and 16 (Table 1). Indeed, 14 is the most active
antibacterial oligomer with a MIC of 10.7 μg/mL against M.
luteus. In contrast, oligomerS 15 and 16 display reduced activity
with MIC against M. luteus of 16.6 and 25.3 μg/mL,
respectively. Obviously, the number of aromatic side chains
has to be considered as well. The higher MIC value of oligomer
16 could derive from the aggregation in aqueous media related
to the higher density of aromatic side chains which was
observed in DLS experiments (Table 1). Indeed, self-assembly
in aqueous media may result in decreased antimicrobial
activity.10 In contrast, oligomer 15 is slightly less hydrophobic,
presents no aromatic side chains and is less active against
β
3R3-peptide against both, M. luteus and E. coli, with MIC of 5.3
μg/mL (1.9 μM) and 43.2 μg/mL (15.4 μM), respectively.
Such values are comparable to biologically important natural
AMPs, such as mammalian defensins52 or the most active
antimicrobial β-peptides known from literature.12 Therefore,
this data clearly shows that amphiphilic cationic β3R3-peptides
can display potent antimicrobial activity in correlation to their
primary sequence.
In order to further investigate the antimicrobial activity also
in dependence of the targeted bacterial strain, antimicrobial
peptides that showed activity against Gram-negative E. coli and
Gram-positive M. luteus were then evaluated for their
antibacterial activity against two additional, clinically relevant
bacterial pathogens of both classes, Pseudomonas aeruginosa
PAO1 and Staphylococcus aureus Newman, respectively (Figure
3). Furthermore, the antimicrobial peptide MAX153 was used as
a control. Here, none of the tested peptides showed measurable
1692
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
Figure 3. Antimicrobial peptide MAX1 and β3R3-peptide 19 show antimicrobial activity against P. aeruginosa but not against S. aureus. Growth of
Gram-negative P. aeruginosa PAO1 and Gram-positive S. aureus Newman was measured after growing 14 h in the presence of different
concentrations of AMP MAX1 and β3R3-peptides 14, 15, 17, and 19.
antibacterial activity up to 100 μg/mL against Gram-positive S.
aureus. However, amphiphilic peptidomimetic 19 as well as the
control MAX1 showed good antimicrobial activity against P.
aeruginosa. The other amphiphilic β3R3- peptides 14, 15, and 17
did not show activity against P. aeruginosa up to 130 μg/mL.
These results highlight the potential of peptidomimetic 19 and
control MAX1 as potential therapeutics against infections with
P. aeruginosa. In contrast to the increased antibacterial activity
of 19 for the Gram-positive M. luteus, certain mechanisms of
resistance in S. aureus seem to prevent toxicity against this
Gram-positive organism. S. aureus is known to covalently
modify its cell envelope with L-lysine or D-alanine catalyzed by
the bacterial enzymes MprF and DltA, respectively.54 These
mechanisms introduce positive charges on the surface, which
leads to electrostatic repulsion of cationic antimicrobial
peptides, such as the ones reported here, rendering them
ineffective.
membranes and do not produce hemolysis at the concentration
required for antimicrobial activity (Table 4) regardless of their
Table 4. MIC, HD50, and Selectivity Index of the
Amphiphilic Cationic β3R3-Peptide Oligomers 14, 15, 17,
and 19
a
a
ab
GM
a
c
oligomer
M. luteus
E. coli
HD50
SI
14
15
17
19
10.7
16.6
29.2
101.9
64.2
33.0
3065.2
3223.7
2563.2
4013.7
92
98
32.6
61.7
15.1
130.6
43.2
41
5.3
265
a
b
Expressed μg/mL. Geometric mean (GM) of MIC values for the M.
c
luteus and E. coli. Selectivity index (SI) is calculated as HD50/GM.
physicochemical differences (Table 1) as well as their different
activities against prokaryotic cell membranes (Tables 2 and 3).
For all compounds, the concentration required for the lysis of
50% of the red blood cells (HD50) is significantly higher than
the lowest reported HD50 values for β-peptides with
antimicrobial activity so far and range from 1315.5 μM (17)
to 1714.4 μM (15).18−22 In addition, only the heterochiral
sequence 17 shows hemolytic activity at a concentration of 200
μM (390 μg/mL). Thus, by comparing the oligomers 14
(vS,sS) and 17 (vS,sR) that differ exclusively in the
configuration of their chiral centers on the sNH2 units, we
can conclude that homochiral sequences (vS,sS) are superior as
antimicrobial peptidomimetics combining high antibacterial
activity with extremely low cytotoxicity.
Cytotoxicity of Antibacterial Amphiphilic Cationic
β
3R3-Peptides. Both, natural AMPs and β-peptides with high
antimicrobial activity often display indiscriminant cytotoxicity
and high hemolytic activity,7,9,18,20−22 which substantially
hampers the development of these systems as therapeutics.
Therefore, the hemolytic activity of the amphiphilic cationic
β
3R3-peptide sequences has been evaluated by testing
hemoglobin release as a result of human red blood cell
(hRBC) lysis upon incubation with the oligomers. Moreover,
the integrity of the hRBC was controlled by microscopic
analysis before and after treatments with amphiphilic cationic
β
3R3-peptide oligomers. In addition, the toxicity toward human
Conclusively, the amphiphilic cationic β3R3-peptides can
display an extremely high selectivity index (Table 4) that is
defined as the ratio between the HD50 and the MIC values.
Furthermore, oligomers 17 and 19 as exemplary structures for
amphiphilic cationic β3R3-peptides showed high stability against
enzymatic degradation (see the SI). Overall, this indicates the
great potential of amphiphilic cationic β3R3-peptides for the
future development as novel antimicrobial agents. In particular,
oligomer 19 is the best candidate for further research and a
good starting point for the development as antimicrobial agent
due to its high antimicrobial activity and unique selectivity.
foreskin fibroblasts (HFF) was tested by monitoring fibroblast
growth in the presence or absence of the oligomers after 24, 48,
and 72 h. All oligomers 10−19 show no cytotoxicity toward
HFF and hemolytic activity at a concentration of 80 μM, which
is in the order of magnitude of the MIC values (see the
Supporting Information). In addition, the hemolytic activity of
selected oligomers was tested in a concentration range between
0 to 2000 μM. The oligomers 14, 15, and 19 were chosen as
they display the highest antimicrobial activity. Furthermore, the
hemolytic activity of the heterochiral (vS,sR) oligomer 17 was
investigated due to its peculiar conformational properties,
which could potentially influence the hemolytic activity of
antimicrobial peptidomimetics. In general, all oligomers tested
show negligible hemolytic activity. In agreement with the
IRRAS and surface tension experiments carried out in the
presence of the model eukaryotic membrane POPC, all four
tested oligomers do not interact with eukaryotic cell
CONCLUSIONS
■
In summary, we have introduced amphiphilic cationic β3R3
-
peptides as a novel class of membrane active peptidomimetics
and evaluated their potential as antibacterial agents. Novel
building blocks were synthesized combining hydrophobic
1693
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
(2) Fjell, C. D.; Hiss, J. A.; Hancock, R. E. W.; Schneider, G. Nat. Rev.
Drug Discovery 2012, 11, 37−51.
diamine units with proteinaceous side chains and aspartic acid
as chiral β-diacid units providing cationic charges in α-position.
Different amphiphilic cationic β3R3-peptide sequences were
prepared varying primary sequence, chirality and chain length.
All oligomers were characterized for their physicochemical
properties via HPLC, DLS, surface tension, and IRRAS
experiments. Additionally, the interactions of three selected
oligomers with model prokaryotic and eukaryotic membranes
have been tested by surface tension and IRRAS measurements.
Thereby, the relationship between the building block and
backbone features was elucidated, as well as the primary
sequences and the overall physicochemical properties that
conclusively determine membrane activity and selectivity of the
oligomers.
(3) Hauck, D.; Joachim, I.; Fommmeyer, B.; Varrot, A.; Philipp, B.;
Moller, H. M.; Imberty, A.; Exner, T. E.; Titz, A. ACS Chem. Biol.
̈
2013, 8, 1775−1784.
(4) Sommer, R.; Joachim, I.; Wagner, S.; Titz, A. Chimia 2013, 67,
286−290.
(5) Eckert, R. Future Microbiol. 2011, 6, 635−651.
(6) Vaara, M. Curr. Opin. Pharmacol. 2009, 9, 571−576.
(7) Hancock, R. E. W.; Sahl, H. G. Nat. Biotechnol. 2006, 24, 1551−
1557.
(8) Nguyen, L. T.; Chau, J. K.; Perry, N. A.; de Boer, L.; Zaat, S. A. J.;
Vogel, H. J. PLoS One 2010, 5, e12684.
(9) Khandelia, H.; Langham, A. A.; Kaznessis, Y. N. Biochim. Biophys.
Acta-Biomembr. 2006, 1758, 1224−1234.
The amphiphilic cationic β3R3-peptides were tested for their
potential as antimicrobial peptidomimetics. Cell viability
(10) Rotem, S.; Mor, A. Biochim. Biophys. Acta, Biomembr. 2009,
1788, 1582−1592.
(11) Giuliani, A.; Rinaldi, A. C. Cell. Mol. Life Sci. 2011, 68, 2255−
2266.
experiments showed that specific amphiphilic cationic β3R3
-
peptides with appropriate physicochemical properties have very
high antimicrobial activity especially against Gram-positive
bacteria. In addition, the amphiphilic cationic β3R3-peptides
show an extraordinary selectivity index (SI up to 265) with high
antimicrobial and yet extremely low hemolytic activity. This
study provided new insights into the SAR of AMPs and proved
that the amphiphilic cationic β3R3-peptides are strong
candidates for further development as antimicrobial agents
with high therapeutic index. Future studies will further elucidate
the mechanisms of selectivity targeting different bacteria and
cell types.
(12) Godballe, T.; Nilsson, L. L.; Petersen, P. D.; Jenssen, H. Chem.
Biol. Drug Des. 2011, 77, 107−116.
(13) Tew, G. N.; Scott, R. W.; Klein, M. L.; Degrado, W. F. Acc.
Chem. Res. 2010, 43, 30−39.
(14) Robinson, J. A. Curr. Opin. Chem. Biol. 2011, 15, 379−386.
(15) Cheng, R. P.; Gellman, S. H.; DeGrado, W. F. Chem. Rev. 2001,
101, 3219−3232.
(16) Seebach, D.; Gardiner, J. Acc. Chem. Res. 2008, 41, 1366−1375.
(17) Horne, W. S.; Gellman, S. H. Acc. Chem. Res. 2008, 41, 1399−
1408.
(18) Porter, E. A.; Weisblum, B.; Gellman, S. H. J. Am. Chem. Soc.
2002, 124, 7324−7330.
(19) Liu, D. H.; DeGrado, W. F. J. Am. Chem. Soc. 2001, 123, 7553−
7559.
ASSOCIATED CONTENT
■
(20) Hamuro, Y.; Schneider, J. P.; DeGrado, W. F. J. Am. Chem. Soc.
S
* Supporting Information
1999, 121, 12200−12201.
(21) Raguse, T. L.; Porter, E. A.; Weisblum, B.; Gellman, S. H. J. Am.
Chem. Soc. 2002, 124, 12774−12785.
(22) Arvidsson, P. I.; Frackenpohl, J.; Ryder, N. S.; Liechty, B.;
Petersen, F.; Zimmermann, H.; Camenisch, G. P.; Woessner, R.;
Seebach, D. ChemBioChem 2001, 2, 771−773.
(23) Mosca, S.; Dannehl, C.; Moginger, U.; Brezesinski, G.;
Hartmann, L. Org. Biomol. Chem. 2013, 11, 5399−5403.
(24) Wang, G. S.; Li, X.; Wang, Z. Nucleic Acids Res. 2009, 37,
D933−D937.
(25) Giuliani, A.; Pirri, G.; Bozzi, A.; Di Giulio, A.; Aschi, M.; Rinaldi,
A. C. Cell. Mol. Life Sci. 2008, 65, 2450−2460.
(26) Mosca, S.; Wojcik, F.; Hartmann, L. Macromol. Rapid Commun.
2011, 32, 197−202.
(27) Wojcik, F.; Mosca, S.; Hartmann, L. J. Org. Chem. 2012, 77,
4226−4234.
Analytical data of building blocks and oligomers, NMR spectra
of intermediates and building blocks, HPLC traces of building
blocks and oligomers, CD and IRRAS spectra of all oligomers,
experimental details on antibacterial growth and hemolysis
assays. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: laura.hartmann@mpikg.mpg.de.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
(28) Ponader, D.; Wojcik, F.; Beceren-Braun, F.; Dernedde, J.;
Hartmann, L. Biomacromolecules 2012, 13, 1845−1852.
(29) Wojcik, F.; O’Brien, A. G.; Gotze, S.; Seeberger, P. H.;
Hartmann, L. Chem.Eur. J. 2013, 19, 3090−3098.
(30) Seebach, D.; Namoto, K.; Mahajan, Y. R.; Bindschadler, P.;
Sustmann, R.; Kirsch, M.; Ryder, N. S.; Weiss, M.; Sauer, M.; Roth, C.;
Werner, S.; Beer, H. D.; Munding, C.; Walde, P.; Voser, M. Chem.
Biodiversity 2004, 1, 65−97.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the Max Planck Society as well as the
German Research Foundation (DFG, Emmy Noether program
HA5950/1-1) for financial support. A.T. acknowledges the
Helmholtz Association for financial support. We thank Uwe
(31) Seebach, D.; Beck, A. K.; Bierbaum, D. J. Chem. Biodiversity
2004, 1, 1111−1239.
(32) Yeaman, M. R.; Yount, N. Y. Pharmacol. Rev. 2003, 55, 27−55.
(33) Du, C. M.; Valko, K.; Bevan, C.; Reynolds, D.; Abraham, M. H.
Anal. Chem. 1998, 70, 4228−4234.
Moginger and Dr. Daniel Kolarich for support with the
̈
enzymatic stability assays and Olaf Niemeyer for technical
support.
(34) Blondelle, S. E.; Houghten, R. A. Biochemistry 1992, 31, 12688−
12694.
REFERENCES
(35) Cherepanov, D. A.; Feniouk, B. A.; Junge, W.; Mulkidjanian, A.
■
Y. Biophys. J. 2003, 85, 1307−1316.
(36) Teschke, O.; de Souza, E. F. Chem. Phys. Lett. 2005, 403, 95−
101.
(1) Boucher, H. W.; Talbot, G. H.; Bradley, J. S.; Edwards, J. E., Jr.;
Gilbert, D.; Rice, L. B.; Scheldl, M.; Spellberg, B.; Bartlett, J. Clin.
Infect. Dis. 2009, 48, 1−12.
1694
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695

Biomacromolecules
Article
(37) Jiang, D. L.; Dinh, K. L.; Ruthenburg, T. C.; Zhang, Y.; Su, L.;
Land, D. P.; Zhou, F. M. J. Phys. Chem. B 2009, 113, 3160−3168.
(38) Olak, C.; Muenter, A.; Andrae, J.; Brezesinski, G. J. Pept. Sci.
2008, 14, 510−517.
(39) Weiner, S.; Traub, W. FEBS Lett. 1980, 111, 311−316.
(40) Tu, R. S.; Tirrell, M. Adv. Drug Delivery Rev. 2004, 56, 1537−
1563.
(41) Leon, L.; Logrippo, P.; Tu, R. Biophys. J. 2010, 99, 2888−2895.
(42) Maltseva, E.; Kerth, A.; Blume, A.; Mohwald, H.; Brezesinski, G.
ChemBioChem 2005, 6, 1817−1824.
(43) Wang, C. S.; Zheng, J. Y.; Zhao, L.; Rastogi, V. K.; Shah, S. S.;
DeFrank, J. J.; Leblanc, R. M. J. Phys. Chem. B 2008, 112, 5250−5256.
(44) Hoernke, M.; Koksch, B.; Brezesinski, G. Biophys. Chem. 2010,
150, 64−72.
(45) Stefaniu, C.; Vilotijevic, I.; Santer, M.; Silva, D. V.; Brezesinski,
G.; Seeberger, P. H. Angew. Chem., Int. Ed. 2012, 51, 12874−12878.
(46) Segman, S.; Lee, M. R.; Vaiser, V.; Gellman, S. H.; Rapaport, H.
Angew. Chem.-Int. Ed. 2010, 49, 716−719.
(47) Segman-Magidovich, S.; Lee, M. R.; Vaiser, V.; Struth, B.;
Gellman, S. H.; Rapaport, H. Chem.Eur. J. 2011, 17, 14857−14866.
(48) Jackson, M.; Mantsch, H. H. Crit. Rev. Biochem. Mol. 1995, 30,
95−120.
(49) Tamm, L. K.; Tatulian, S. A. Q. Rev. Biophys. 1997, 30, 365−
429.
(50) Deshayes, S.; Morris, M. C.; Divita, G.; Heitz, F. J. Pept. Sci.
2006, 12, 758−765.
(51) Marsh, D. Biochim. Biophys. Acta 1996, 1286, 183−223.
(52) Sahl, H. G.; Pag, U.; Bonness, S.; Wagner, S.; Antcheva, N.;
Tossi, A. J. Leukocyte Biol. 2005, 77, 466−475.
(53) Salick, D. A.; Kretsinger, J. K.; Pochan, D. J.; Schneider, J. P. J.
Am. Chem. Soc. 2007, 129, 14793−14799.
(54) Peschel, A. Trends Microbiol. 2002, 10 (4), 179−186.
1695
dx.doi.org/10.1021/bm500101w | Biomacromolecules 2014, 15, 1687−1695