Design and characterization of novel antimicrobial peptides, R-BP100 and

RW-BP100, with activity against Gram-negative and Gram-positive bacteria

Inês M. Torcato, Yen-Hua Huang, Henri G. Franquelim, Diana Gaspar,

David J. Craik, Miguel A.R.B. Castanho, Sónia Troeira Henriques

PII: S0005-2736(12)00427-0
DOI: doi: 10.1016/j.bbamem.2012.12.002
Reference: BBAMEM 81132

To appear in: BBA - Biomembranes

Received date: 26 August 2012

Revised date: 29 November 2012
Accepted date: 4 December 2012

Please cite this article as: Inês M. Torcato, Yen-Hua Huang, Henri G. Franquelim, Diana
Gaspar, David J. Craik, Miguel A.R.B. Castanho, Sónia Troeira Henriques, Design and
characterization of novel antimicrobial peptides, R-BP100 and RW-BP100, with activity
against Gram-negative and Gram-positive bacteria, BBA - Biomembranes (2012), doi:
10.1016/j.bbamem.2012.12.002

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Design and characterization of novel antimicrobial peptides, R-BP100 and

RW-BP100, with activity against Gram-negative and Gram-positive
bacteria

T
Inês M. Torcatoa, Yen-Hua Huangb, Henri. G. Franquelima, Diana Gaspara, David J. Craikb,

IP
Miguel A.R.B. Castanhoa, Sónia Troeira Henriquesa,b,*

R
Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, 1649-028
Lisbon, Portugal

SC
b
Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD 4072,
Australia

NU
MA
D
TE

*Current Address: Institute for Molecular Bioscience, The University of Queensland, Brisbane,
QLD 4072, Australia, Tel.: +61-7-33462026; Fax: + 61-7-33462101; e-mail:
P

[email protected]
CE
AC

1
 ACCEPTED MANUSCRIPT

Abstract
BP100 is a short cationic antimicrobial peptide with a mechanism of action dependent on
peptide-lipid interactions and microbial surface charge neutralization. Although active against
Gram-negative bacteria, BP100 is inactive against Gram-positive bacteria. In this study we
report two newly designed BP100 analogues, RW-BP100 and R-BP100 that have the Tyr

T
residue replaced with a Trp and/or the Lys residues replaced with an Arg. The new analogues in

IP
addition to being active against Gram-negative bacteria, possesses activity against all tested

R
Gram-positive bacteria. Mechanistic studies using atomic force microscopy, surface plasmon
resonance and fluorescence methodologies reveal that the antibacterial efficiency follows the

SC
affinity for bacterial membrane. The studies suggest that the activity of BP100 and its analogues
against Gram-negative bacteria is mainly driven by electrostatic interactions with the

NU
lipopolysaccharide layer and is followed by binding to and disruption of the inner membrane,
whereas activity against Gram-positive bacteria, in addition to electrostatic attraction to the
MA
exposed lipoteichoic acids, requires an ability to more deeply insert in the membrane
environment, which is favoured with Arg residues and is facilitated in the presence of a Trp
residue. Knowledge on the mechanism of action of these antimicrobial peptides provides
D

information that assists in the design of antimicrobials with higher efficacy and broader spectra
of action, but also on the design of peptides with higher specificity if required.
P TE

Keywords: Antimicrobial peptide; Model membranes; Peptide-membrane interactions; Atomic

CE

force microscopy; Trp/Tyr fluorescence.

AC

2
 ACCEPTED MANUSCRIPT

Abbreviations: H-bond, Hydrogen bond; AMP, antimicrobial peptide; AFM, atomic force
microscopy; RP-HPLC, reversed phase high-performance liquid chromatography; MBC,
minimal bactericidal concentration; MIC, minimal inhibitory concentration; E. coli, Escherichia
coli; S. aureus, Staphylococcus aureus; K. pneumoniae, Klebsiella pneumonia; P. aeroginosa,
Pseudomonas aeroginosa; S. pneumonia, Streptococcus pneumonia; E. faecium, Enterococcus

T
faecium; ATCC, American type culture collection; MHB, Mueller Hinton broth; OD, optical

IP
density; PBS, phosphate buffered saline; NaCl, sodium chloride; Srms, bacterial shape; Rrms, cell

R
surface roughness; RBC, red blood cell; HC50, peptide concentration needed to achieve 50% of
haemolysis; IC50, peptide concentration needed to achieve 50% of cell death; REES, red-edge

SC
excitation shift; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; LUV, large
unilamellar vesicles; POPC, 1-palmitoyl-2-Oleoyl-sn-Glycero-3-Phosphocholine; POPG, 1-

NU
palmitoyl-2-Oleoyl-sn-Glycero-3-Phosphoglycerol; 5/16-NS, 5/16- doxyl stearic acid; CD,
circular dichroism; di-8-ANEPPS, 4-[2-[6-(dioctylamino)-2-naphthalenyl]ethenyl]-1-(3-
MA
sulfopropyl); PBMC, peripheral blood mononuclear cell; CF, carboxyfluorescein; LC50, peptide
concentration needed to induce 50% of leakage; T1/2, time necessary to achieve 50% of leakage;
LPS, lipopolysaccharide; LTA, lipoteichoic acid.
D
P TE
CE
AC

3
 ACCEPTED MANUSCRIPT

1. Introduction

Polycationic antimicrobial peptides (AMPs) are present in virtually all organisms as part
of the innate immune system and act as endogenous antibiotics. These peptides induce the direct
destruction of a wide diversity of microorganisms [1-4]. Owing to their ability to attack

T
different microorganisms, including bacteria, viruses and fungi, together with the growing

IP
problem of resistance to conventional antibiotics, AMPs have been regarded as promising

R
candidates for the development of novel antibiotics [1-4]. The fact that microorganisms are less
efficient in developing effective resistance mechanisms against AMPs than against classical

SC
antibiotics [1, 3, 5, 6] further supports the use of AMPs as novel therapeutics. In addition,
AMPs can act synergistically with conventional antibiotics [6]. Thus, even if not used on their

NU
own, AMPs can potentially be administered with classical antibiotics to improve their
therapeutic activity.
MA
Cationic AMPs are generally short and amphipathic and it has been proposed that they
primarily interact with bacteria by binding to the microbial negatively-charged membrane [7-9].
Membrane targeting is followed by permeabilization leading to the disruption of the bacterial
membrane structure [10, 11]. Alternatively, some AMPs seem to cross the membrane without
D

destroying it [11, 12] and exert their activity through interaction with intracellular targets (e.g.
TE

binding to and inhibition of nucleic acids) [13]. Their preference for the more negatively-
charged bacterial membranes over the eukaryotic membranes, which have an outer lipid leaflet
P

globally neutral, can be responsible for their selectivity [6, 11].

CE

Although having potential applications, naturally occurring AMPs are susceptible to

protease degradation and have low bioavailability. In an effort to overcome these disadvantages,
peptides with shorter and improved amino acid sequences and, therefore, with increased
AC

antimicrobial activity and decreased cytotoxicity have been designed and synthesized [2, 4, 11,
14]. BP100 (KKLFKKILKYL-NH2) is a successful example of a synthetic AMP. Obtained from
a combinatorial chemistry approach based on a cecropin A-melittin hybrid, BP100 is potent
against Gram-negative bacteria, possesses low cytotoxicity and has low susceptibility to
protease K degradation [14].
BP100 binds to the anionic bacterial membrane and neutralizes it [15]. Its ability to
efficiently disrupt anionic model membranes but not neutral model membranes, suggests that
BP100 selectively disrupts bacteria over mammalian cells [15]. More recently, BP100 was also
shown to be able to internalize into eukaryotic cells without disrupting them [16]. This finding
holds great promise for the use of BP100 as a means to treat eukaryotic cells that are infected
with bacteria.
Although BP100 has potential application against Escherichia coli (E. coli) [17] and
other Gram-negative bacteria [14], it does not have significant activity against tested Gram-

4
 ACCEPTED MANUSCRIPT

positive bacteria [17]. In theory, the efficiency of AMPs can be enhanced by further improving
their affinity for anionic membranes. Nevertheless, neither the role of the membrane
composition nor the structural features of peptides required for specificity are, as yet, fully
understood and predicting antimicrobial or cytotoxic activity from a given amino acid sequence
is difficult.

T
In the present study, based on previous studies showing that Arg and Trp residues

IP
enhance AMP activity [18, 19], we propose the development of new BP100 analogues with

R
putative improved membrane-binding properties: RW-BP100 (RRLFRRILRWL-NH2) and R-
BP100 (RRLFRRILRYL-NH2) in which the Trp residue was replaced with a Tyr or/and the Lys

SC
residues were replaced with Arg. The antimicrobial efficiency of the newly designed peptides
was tested against several Gram-negative and Gram-positive bacteria, including non-resistant

NU
and resistant strains. Their mechanism of action was evaluated by direct bacterial imaging with
atomic force microscopy (AFM) and by biophysical studies with model membranes. Our results
MA
show that the newly designed peptides are not only active against Gram-negative bacteria but
also against all tested Gram-positive bacteria. This study reveals that by replacing Lys with Arg
it is possible to gain activity against Gram-positive bacteria; this finding is very important as it
provides information for AMP design where a more specific or a broader activity is desired and
D

also provides information on the differences in AMP action against Gram-positive and Gram-
TE

negative bacteria.
P
CE

2. Materials and methods

2.1. Peptide synthesis

AC

BP100, R-BP100 and RW-BP100 were synthesized using microwave-assisted solid

phase using an automatic peptide synthesizer (CEM Liberty) following standard Fmoc
chemistry on a 2-chlorotrityl resin. The deprotected peptide was cleaved using trifluoroacetic
acid and purified by RP-HPLC. The synthetic peptides had ≥95% purity, as confirmed by
analytical RP-HPLC. Peptide concentrations were determined by absorbance at 280 nm
assuming ε = 1.49 × 103 M-1 cm-1 for BP100 and R-BP100 and ε = 5.50 × 103 M-1 cm-1 for RW-
BP100. The overall hydrophobicity of the three peptides was compared using their RP-HPLC
retention times [20]. Samples were run using a linear AB gradient (1% /min) at a flow rate of
0.3mL/min on a 0.3mL analytical column. Eluent A was 0.05% TFA (v/v) in water and eluent B
was 90% acetonitrile (v/v) and 0.05% TFA (v/v) in water.

2.2. Antimicrobial susceptibility assay

5
 ACCEPTED MANUSCRIPT

The antimicrobial activity of BP100 and its analogues was examined by evaluation of
the bacterial growth inhibition and bactericidal activity. A microtiter broth dilution method was
employed [21]. Gram-negative Escherichia coli ATCC 25922, E. coli CGSC 5167, Klebsiella
pneumoniae ATCC 13883, K. penumoniae ATCC 700603, Pseudomonas aeroginosa ATCC

T
10145, P. aeroginosa ATCC 27653 and Gram-positive Staphylococcus aureus ATCC 25923, S.

IP
aureus ATCC 33591, Streptococcus pneumonia ATCC 33400, S. pneumonia ATCC700677,

R
Enterococcus faecium ATCC 35667 and E. faecium ATCC 51559 were tested. Briefly, bacterial
suspensions (5  105 cells/mL) were incubated with two-fold dilutions of peptides (solubilized

SC
in sterile water) in 96-well non-binding surface plates (NBS, corning) for 24h at 37ºC and
compared with antibiotics as controls (collestin for Gram-negative bacteria, and kanamycin or

NU
tetracycline against Gram-positive bacteria). The minimal inhibitory concentration (MIC) was
the lowest concentration showing no visible growth. To confirm if the peptides were killing the
bacteria, the minimal bactericidal concentration (MBC) was determined by adding 30 L of
MA
resazurin dyes (0.01% (w/v)) to each well. The plates were incubated at 37ºC for further 18 h.
Wells with blue coloration indicate dead microorganism, whereas wells with pink coloration
D

indicate live microorganism. The MBC value is the lowest concentration of the wells with blue
coloration. The assays were done with Mueller Hinton broth (MHB) and repeated four times.
TE

2.3. Atomic force microscopy imaging

P
CE

Direct observation of the effects induced by BP100, R-BP100 and RW-BP100 in the
morphology of bacterial cells was obtained by AFM imaging as previously described [17]. Cell
suspensions with 2 × 107 cells/mL were spun down at 13,000 rpm (8 min for E. coli and 30 min
AC

for S. aureus) and washed twice with 10 mM phosphate buffer solution (PBS) containing 150
mM NaCl, pH 7.4 to remove the media. The peptides were incubated with the bacteria and the
slides prepared as before [17]. The final peptide concentrations were 1, 10 or 100 µM. On
average, five individual bacterial cells from two different zones of each slide with a total area of
4 × 4 µm2 were imaged. Height, error and phase-shift images were recorded. Height images
were treated with JPK data processing 4.0.23 and orthogonal projections were done in
Gwyddion 2.24 (Czech Metrology Institute, Czech Republic). Cross sections on five bacterial
cells were done using Gwyddion 2.24 to determine the average dimension of E. coli and S.
aureus cells. The average dimensions of the non-treated bacterial cells are in agreement with
previous observations [17, 22, 23] confirming the viability of the controls. E. coli and S. aureus
cells had respectively 0.20 ± 0.01 µm and 0.40 ± 0.08 µm of height, 1.16 ± 0.07 µm and 1.06 ±
0.11 µm of width and 2.73 ± 0.58 µm and 1.66 ± 0.34 µm of length.

6
 ACCEPTED MANUSCRIPT

2.4. Bacterial surface roughness and swelling analysis of AFM images

The bacterial cell surface roughness (Rrms) was determined through root-mean-square
calculations of flattened AFM height images as previously described [17]. Changes in bacterial
shape (Srms) were analysed following a protocol similar to the one used for roughness [17] but

T
instead of using a flattened representation of the bacterial cell surface, a mean-filter treated

IP
AFM image of the general shape of the bacteria was used. With this treatment it was possible to

R
remove the superficial bacterial roughness and analyse the general shape and volume of the
bacteria. The Srms was calculated using equation 1 [17]. Non-treated and peptide-treated bacteria

SC
were compared.

(1)

2.5. Haemolytic assay NU

MA
The haemolytic activity of the peptides was determined using human red blood cells
(RBCs) as described before [24]. Briefly, peptides were solubilised in 10 mM HEPES buffer
D

containing 150 mM NaCl, pH 7.4 and tested in triplicates (two-fold dilutions starting with 50
TE

µM). Melittin, a peptide with known haemolytic properties was also included as a positive
control. Samples with HEPES buffer or 0.1% (v/v) Triton X-100 were used to establish 0% or
P

100% of haemolysis, respectively. The experiment was repeated three times. The peptide
CE

concentration needed to achieve 50% of haemolysis (HC50) was determined as before [24].

2.6. Cytotoxicity against human cells

AC

Cervical cancer cells (HELA) were cultured in T175 flasks at 37ºC in an atmosphere of
5% CO2 with DMEM containing 10% FBS. Cells were seeded (5  103 cells/ well) in 96-well
plate flat-bottomed plate for 24 h before assay. BP100 and its analogues were incubated for 2 h
in concentrations ranging from 60 to 0.5 M, final volume 100 L in media without serum, to
avoid degradation. Blank without peptide, and cells with Tx-100 were included to establish 0%
and 100% of cell death. After incubation, peptides were washed with PBS, fresh media (100
L) containing serum and 10 L resazurin dye (0.02% (w/v)) were added to the cells and
incubated for 18 h. Absorbance signal was monitored at 540 and 620 nm and the percentage of
death cells was quantified by calculating the absorbance ratio, R, (R=A620/A540nm) and
considering 100% of cell death the R obtained with TX-100 and 0% of death the R obtained
with cells without compound, using the equation:

7
 ACCEPTED MANUSCRIPT

(2)

2.7. Binding affinity for lipopolysaccharide and lipotheicoic acid.

T
IP
The endpoint Chromogenic Limulus Amebocyte Lysate (LAL) test kit (QLC-1000,

R
Lonza) was employed to examine the ability of BP100 and its analogues to neutralize

SC
lipopolysaccharide (LPS) and lipoteichoic acid (LTA). This test is a sensitive indicator of the
presence of free, non-neutralized endotoxin (either LPS or LTA) [25]. LPS from E. coli
O111:B4 (standard provided with the LAL assay kit) or LTA from S. aureus (L2515 from

NU
Sigma) were incubated with various concentrations of BP100 and its analogues. Briefly, peptide
and LPS (1 EU/mL), or LTA (500 ng/mL, response equivalent to LPS 1 EU/mL), were
MA
incubated for 30 min to allow binding. LAL reagent was added and incubated for 10 min at
37ºC, followed by addition of the colourless substrate and incubation for further 6 min as per
manufacture instructions. The reaction was stopped with 25% (v/v) glacial acetic acid and the
D

release of p-nitroaniline followed by absorbance at 405 nm. Controls with water or peptide only
were included to confirm that all the samples were endotoxin-free. After blank subtraction, the
TE

concentration of free endotoxin was calculated by using calibration curves obtained with LPS
(0.125 – 1 EU/mL) or LTA (62.5-1000 ng/mL). The values were normalized and converted into
P

bound endotoxin (bound endotoxin = 1- free endotoxin) and plotted as a function of peptide
CE

concentration. The experiment was repeated three times.

AC

2.8. Preparation of lipid vesicles

Large unilamellar vesicles (LUVs) with a diameter of 100 nm, or small unilamellar
vesicles (SUVs) with a diameter of 50 nm were used as membrane model and prepared by
extrusion method as before [26]. LUVs were used in fluorescence studies whereas SUVs were
used to prepare the bilayers for surface plasmon resonance studies [27]. Vesicles composed of
1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC; Avanti Polar Lipids, USA) or with
mixtures of POPC and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG) were
prepared for comparison. All the lipid systems were prepared in 10 mM HEPES buffer, pH 7.4
containing 150 mM NaCl.

2.9. Binding of BP100 and analogues with lipid bilayers followed by SPR

8
 ACCEPTED MANUSCRIPT

The ability of BP100 and its analogues to bind to phospholipid lipid bilayers was
examined using SPR as described for other peptides [27]. L1 sensor chips and a Biacore 3000
instrument were used. The binding of the three peptides was compared for POPC, POPC/POPG
(4:1 molar ratio) and POPC/POPG (1:1 molar ratio). Each measurement was repeated three
times. The obtained sensorgrams were normalized for lipid deposition and peptide mass and

T
represented as peptide-to-lipid ratio (P/L). The relative affinity was compared based on the P/L

IP
obtained at a reporting point at the end of the association phase (t= 170 s) [28].

R
2.10. Changes in membrane dipole potential

SC
The membrane dipole potential can change upon insertion of peptide into the lipid

NU
membrane. Changes in model membranes dipolar potential were followed with fluorescence
excitation spectra of the membrane dye (4-[2-[6-(dioctylamino)-2-naphthalenyl]ethenyl]-1-(3-
MA
sulfopropyl) (di-8-ANEPPS; Invitrogen, USA)) as before [27]. The dye (4 µM) was added to
200 µM POPC or POPC/POPG (1:1 molar) LUVs and incubated overnight. The fluorescence
excitation spectra (λem= 613 nm for POPC and λem= 617 nm for POPC/POPG vesicles) were
recorded before and after peptide addition (12.5, 25 and 50 µM, incubated for ~ 5 min).
D

Dipole potential variations in Human cells membrane were also monitored using RBCs and
TE

peripheral blood mononuclear cells (PBMCs) and the same peptide concentrations. Cell
isolation and dye-labelling followed protocols previously described [29].
P
CE

2.11. Vesicle aggregation

Vesicle aggregation induced by BP100, R-BP100 or RW-BP100 was followed by

AC

optical density at 436 nm (OD436) [26]. The peptide (6.25, 12.5, 20, 25 and 50 µM) was added to
POPC or POPC/POPG (4:1 or 1:1 molar) vesicles (500 µM final lipid concentration) and the
OD followed as function of time.

2.12. Vesicle leakage studies

Leakage of vesicle contents induced by BP100, R-BP100 and RW-BP100 was

quantified by carboxyfluorescein (CF) fluorescence dequenching (lipid vesicles were prepared
with 50 mM of CF solubilised in 10 mM HEPES buffer containing 150 mM NaCl). Vesicle
preparation, lipid quantification and determination of leakage percentage followed protocols
previously described [30]. Different peptide concentrations with two-fold dilutions starting from
100 µM were incubated with POPC or POPC/POPG (1:1 molar) LUVs (50 µM, final lipid
concentration). The fluorescence intensity (λexc = 492 nm, λem= 518 nm) was measured after five

9
 ACCEPTED MANUSCRIPT

minutes incubation and LC50 (peptide concentration needed to induce 50% of leakage)
calculated as before [30]. The leakage kinetic was followed for 150 s after addition of peptide
(25 µM) to POPC or POPC/POPG (1:1 molar) vesicles (50 µM lipid concentration) and the time
necessary to achieve half of the response (T1/2) was determined considering a pseudo-first order
kinetic.

T
IP
2.13. The fluorescence properties of BP100 and its analogues in aqueous solution and in

R
membrane environment

SC
The peptides included in this study have intrinsic fluorescence due to the presence of a
Tyr (BP100 and R-BP100) or Trp residue (RW-BP100) in their sequence. Unless otherwise

NU
stated, BP100 and R-BP100 were excited at 274 nm and RW-BP100 was excited at 280 nm.
Fluorescence emission spectra were scanned as a function of peptide concentration, pH and urea
MA
concentration. The existence of red-edge excitation shift effect (REES) was studied for RW-
BP100: fluorescence emission spectrum was followed upon excitation with different
wavelengths in the range 270 to 320 nm. Exposition of Tyr or Trp residue to the aqueous
environment was evaluated by fluorescence emission quenching upon titration with acrylamide,
D

as before [31]. To minimize the relative quencher/fluorophore light absorption ratio BP100 and
TE

R-BP100 were excited at 274 nm whereas RW-BP100 was excited at 290 nm. Fluorescence
spectra with a band width of 5/10 nm for BP100 and R-BP100, or 3/6 nm for RW-BP100, were
P

acquired at 25ºC in a spectrofluorimeter Edinburgh FS900 with 0.5 cm length quartz cuvettes.
CE

Peptides were solubilised in 10 mM HEPES buffer, pH 7.4 containing 150 mM NaCl.

To evaluate if the Trp or Tyr residue inserts in the lipid bilayer, the partition of BP100
and its analogues was followed by peptide fluorescence emission. Peptides were titrated with
AC

POPC or POPC/POPG (4:1 or 1:1 molar ratio) vesicle suspensions (final lipid concentrations in
the range 0 - 4 mM). All the samples were incubated for five minutes before spectra scanning.
To have negligible inner filter effect, the experiment was performed with 50 µM BP100 or R-
BP100 and 15 µM of RW-BP100 (absorbance < 0.1). The integrated area of each fluorescence
spectrum was corrected for dilution and light dispersion, was normalized for the signal in buffer
(I/IW) and plotted as a function of lipid concentration [32]. The partition coefficient, Kp, and the
ratio of the fluorescence emission obtained when all the peptide molecules are partitioned in the
lipid (IL) to the fluorescence intensity when all the peptide molecules are in water environment
(IW), IL/IW, were determined by fitting a non-linear equation as previously detailed [32].

2.14. Peptide in-depth location in lipidic membranes

10
 ACCEPTED MANUSCRIPT

The membrane in-depth location of the Trp or Tyr residue in the tested peptides was
determined by a differential fluorescence quenching approach as previously described [33].
Briefly, quenching of peptide fluorescence emission was evaluated in the presence of quenchers
(acrylamide, 5- and 16- doxyl stearic acid (5- and 16-NS)) that localize with different depths in
the membrane [33]. 50 µM of BP100 and R-BP100 or 15 µM of RW-BP100 were incubated

T
with 3 mM POPC or POPC/POPG (1:1) vesicles and titrated with increasing concentrations of

IP
acrylamide, 5- or 16-NS (Sigma-aldrich, USA) as detailed elsewhere for other peptides [31].

R
Data were analysed using the Stern-Volmer equation and quenching efficiency quantified with
KSV (Stern-Volmer constant) [31]. When a positive deviation to the Stern-Volmer representation

SC
was observed data were analysed as sphere-of-action quenching [34].

NU
2.15. Secondary structure followed by Circular dichroism (CD) spectroscopy
MA
CD spectra were obtained on a Jasco J-815 instrument (Jasco Co, Tokyo, Japan) using a
1 mm path length cuvette. Briefly, spectra were collected from 195 to 260 nm at a speed of 50
nm/min. Peptide and lipid samples were prepared in 10 mM phosphate buffer with 150 mM
NaF, pH 7.4. The CD signal of the peptides (50 µM) was recorded in the absence/presence of
D

POPC or POPC/POPG (1:1) vesicles (peptide-to-lipid ratio was 1:9 molar or 1:19 molar
TE

depending on the signal intensity and vesicles dispersion contribution). To evaluate the effect of
low pH and of a denaturising agent in the peptides structure, the peptide CD signal was also
P

evaluated at pH 2 and in the presence of 5 mM urea. Spectra were corrected by subtracting the
CE

appropriate blank (buffer, lipid in buffer, urea in buffer or buffer at pH 2) and converted to
mean residue molar ellipticity. Spectra were smoothed using the noise reduction routines
provided with the J-815 spectropolarimeter software (using Savitzky-Golay method). A
AC

quantitative analysis of the percentage of helicity was performed using three deconvolution
software packages (SELCON3, CONTINLL and CDSSTR). No agreement was obtained
between the three packages, and therefore a quantitative analysis was not included.

3. Results

3.1. Design of BP100 derivatives and their hydrophobic and basic properties

BP100 [14], like many other antimicrobial peptides (e.g. cecropins, magainins,
cathelicidins and defensins) [35], kills bacteria through targeting membranes. The aim of the
present study was to improve the activity of BP100 by modulating its membrane-binding

11
 ACCEPTED MANUSCRIPT

properties and to gain more insights into the importance of basic and hydrophobic residues for
antimicrobial efficacy and spectrum of action. Therefore, two new BP100 analogues, RW-
BP100 and R-BP100 (Fig.1), in which the Tyr residue is replaced with a Trp residue and/or the
Lys residues were replaced with Arg residues, were designed. The global charge is +6 for the
three peptides and the overall hydrophobicity, as measured with RP-HPLC retention times (36.6

T
min for BP100, 37.7 min for R-BP100 and 40.4 min RW-BP100), follows the order BP100 < R-

IP
BP100 < RW-BP100.

R
The putative membrane-binding affinity trend of the three peptides was estimated using
the Wimley & White membrane interface hydrophobic scale [36]. Most hydrophobic scales

SC
focus on the membrane hydrophobic bulk phase [37], whereas the Wimley & White scale
predicts the partition of the whole residue from water to the phospholipid membrane interfaces

NU
[38], which is particularly relevant to predict the ability of a peptide to target membranes. The
membrane-binding partition as estimated by Wimley & White scale suggests the following
MA
trend: BP100 < R-BP100 < RW-BP100 (see Fig 1).

3.2. Newly designed AMPs have higher antimicrobial activity than BP100
D

The antimicrobial activities of BP100, R-BP100 and RW-BP100 were tested against
TE

Gram-negative and Gram-positive bacterial strains (Table 1). These bacteria have pathological
interest and possess strains with antibiotic resistance and therefore they were chosen as models
P

of Gram-negative and Gram-positive bacteria. Resistant and control strains were included for
CE

comparison. BP100 was active at low micromolar concentrations against all the Gram-negative
bacteria tested (control and resistant strains) but has lower efficacy against Gram-positive
bacteria, which is in agreement with previous reports [17, 39]. The newly designed peptides
AC

were more active against Gram-negative bacteria than BP100, but most importantly are active at
low micromolar concentration against all tested Gram-positive bacteria. RW-BP100 is the most
active derivative, particularly against Gram-positive bacteria. It is worth noting that RW-BP100
has a MIC below 1 M against tested resistant strains of E. coli, S. aureus, K. pneumonia and S.
pneumonia. The determined MBC values are close to the MIC, revealing that the peptides are
bactericidal and inhibit the bacterial growth by killing the bacteria.
Direct observation of the effects induced by BP100 and the two analogues was obtained
by AFM imaging (Fig. 2). E. coli and S. aureus were used as models of Gram negative and
Gram positive bacteria, respectively. BP100 and its analogues show identical effects on E. coli
morphology. E. coli cells treated with 1 or 10 µM of peptide have their volume increased (Fig.
2E) and develop round-shaped structures on the membrane surface (Fig. 2A), which increase the
overall surface roughness (Fig. 2C). When treated with 100 µM of BP100, or the analogues, the

12
 ACCEPTED MANUSCRIPT

E. coli membrane is disrupted and the intracellular contents leak extensively (see Fig. 2A).
These effects are in agreement with a previous report on BP100 [17].
Effects on S. aureus cell morphology, roughness and volume (Fig. 2B, D, F) were
evident with R-BP100 and RW-BP100 at concentrations ≥ 10 µM, whereas BP100 only has
visible effects at 100 µM. Membrane disruption of S. aureus is evident for all the peptides at the

T
highest concentration tested.

R IP
3.3. Cytotoxicity and therapeutic index of BP100 and its analogues

SC
The cytotoxicity of the peptides was assessed in HELA cells (Fig. 3A), revealing that
BP100 is the least toxic peptide and RW-BP100 is the most toxic. Although being the most

NU
cytotoxic peptide, RW-BP100 has the highest therapeutic index (TI, i.e., the ratio of toxic dose
to therapeutic dose (IC50/MIC)) against Gram positive bacteria, whereas BP100 has the lowest
MA
TI (e.g. the TI of BP100, R-BP100 and RW-BP100 against the resistant S. aureus strain ATTC
33591 is 3, 6 and 48, respectively). When comparing the TI for Gram negative bacteria the three
peptides have a therapeutic index  10 for all the resistant bacteria strains tested.
D

Haemolytic studies were also carried and showed that RW-BP100 is the most
haemolytic peptide whereas BP100 has low haemolytic activity in the concentration range
TE

tested (see HC50 in Fig. 3B).

P

3.4. Binding to LPS and LTA

CE

The outer membranes of Gram-positive and Gram-negative bacteria differ significantly

AC

in their composition, which could explain the selectivity of BP100 towards Gram-negative
bacteria. In addition to a thicker peptidoglycan layer, Gram-positive bacteria have LTA exposed
in the outer membrane, whereas Gram negative bacteria have a thin peptidoglycan layer and an
outer lipid bilayer covered with LPS [40]. These negatively-charged molecules activate multiple
signal transduction pathways and constitute the first physical barrier which need to be
transversed by antimicrobial peptides [41]. Therefore, the ability of BP100 and its analogues to
bind to LPS and LTA was examined and compared using a LAL assay. Figure 4 shows that the
three peptides are able to bind and neutralize LPS and LTA, but BP100 has the weakest activity.
R-BP100 and RW-BP100 have identical activity to neutralize both molecules.

3.5. BP100 and its analogues have preference for negatively-charged membranes as identified
by SPR
The ability of BP100 and its analogues to bind to phospholipid bilayers was examined
using SPR to gain insights into their mechanism of action (Fig. 5). Membranes composed of

13
 ACCEPTED MANUSCRIPT

zwitterionic POPC, which mimics the bulk fluid phase in eukaryotic cell membranes, were
compared with POPC/POPG mixtures to mimic anionic bacterial cell membranes. SPR studies
confirmed that BP100 has the lowest affinity for the membrane whereas RW-BP100 has the
highest affinity. The three peptides have a preference for more anionic membranes relative to
zwitterionic membranes (Fig. 5C). The preference of BP100 for more anionic membranes in

T
relation to neutral membranes is in agreement with a previous study [15]. As predicted using the

IP
Wimley & White amino acid hydrophobic scale (see Fig.1), the new derivatives have higher

R
affinity for membranes than BP100.

SC
3.6. Changes in the dipolar potential induced by BP100 and its analogues

NU
Changes in the membrane dipole potential might occur upon insertion of peptides into
phospholipid bilayers and can be monitored through spectral shifts in the fluorescence
MA
excitation spectra of di-8-ANEPPS probe [42]. The three peptides induced a larger change in the
dipolar potential of POPC/POPG (1:1) than in POPC membranes (Fig. 6A&B), confirming a
preference for negatively-charged membranes in respect to zwitterionic membranes. In addition,
changes in dipolar potential reveal that these peptide molecules are inserting in the anionic lipid
D

membranes rather than just being adsorbed at the phospholipid headgroup region. RW-BP100
TE

induced the largest change in the dipole potential in both lipidic mixtures studied (POPC and
POPC/POPG 1:1). Significant alterations in the RBC’s or PBMC’s membrane dipole potential
P

were not detected for any of the analogues (data not shown) at concentrations higher than the
CE

MIC (12.5, 25 and 50 µM), suggesting that these peptides do not insert significantly into human
cells.
AC

3.7. Vesicle aggregation and leakage induced by BP100 and its analogues

The membrane surfaces of vesicles can be brought in close proximity due to membrane-
peptide-membrane interactions, resulting in vesicle aggregation [43]. The stability of vesicles
dispersed in solution is mainly maintained by repulsive electrostatic forces and a repulsive
hydration shell [43]. The interactions of positively-charged peptides with the negatively-charged
membranes can cancel repulsive electrostatic forces and induce dehydration of the lipid polar
groups. Such alterations make the surface more unstable and facilitate the aggregation of the
vesicles [43]. The three peptides are able to induce aggregation of POPC/POPG (1:1) vesicles
but not of the vesicles composed with POPC or POPC/POPG (4:1) (up to P/L = 0.1, Fig. 6C).
These results support the finding that all three peptides prefer more negatively-charged
membranes over neutral membranes and show that the membrane surface is disturbed upon
peptide binding. BP100 showed the highest ability to induce aggregation of POPC/POPG (1:1

14
 ACCEPTED MANUSCRIPT

molar) vesicles at 25 M, suggesting that this peptide more efficiently disturbs the membrane
surface than the other two peptides. RW-BP100 showed the weakest vesicle surface
destabilization ability.
It was previously reported that BP100 makes lipid vesicles permeable to Co2+ ions [15].
In the present study we evaluated the leakage of contents from POPC or POPC/POPG (1:1)

T
vesicles induced by BP100 and its analogues (Fig. 6D and Table 2). Vesicles loaded with CF

IP
(final lipid concentration of 50 µM) were incubated with varying concentrations of peptide (up

R
to P/L of two). BP100 and its analogues induce leakage of the anionic POPC/POPG (1:1)
vesicles with identical efficiency (Table 2; LC50 ~ 5 µM, P/L ~ 0.1). RW-BP100 induces vesicle

SC
leakage at a faster rate than the other peptides, (see Fig. 6D and T1/2 values in Table 2). A faster
leakage rate might correlate with a faster membrane-binding rate.

NU
Leakage from POPC vesicles occurs at a lower extension (Fig. 6D) and the efficiency is
peptide dependent, being BP100 the least and RW-BP100 the most effective (Table 2). This
MA
trend is expected considering the haemolytic profile of the three peptides.

3.8. Fluorescence properties of BP100 and its analogues in buffer and in the presence of the
D

lipid bilayer
TE

BP100 and its analogues are intrinsically fluorescent due to Tyr or Trp residues in their
sequence. The fluorescence of these aromatic residues is sensitive to its environment; therefore,
P

the intrinsic steady-state fluorescence properties can give insights into the structure and
CE

organization of the peptides in the aqueous environment. BP100 and R-BP100 have excitation
and emission spectra similar to L-Tyr amino acid (excitation maximum = 275 nm and emission
AC

maximum = 302 nm), whereas RW-BP100 has excitation and emission spectra identical to L-
Trp amino acid in buffer (excitation = 280 nm and emission = 347 nm). These spectral
characteristics suggest that BP100 and its analogues have the Tyr or Trp residue exposed to the
solvent [44]. This finding is supported by the observation that the fluorescence quantum yield is
independent of the peptide concentration (at least up to 100 µM) and independent of pH (as low
as 2.5, data not shown). Due to the high sensitivity of Trp to the surrounding environment, the
fluorescence of Trp-containing peptides might have REES as a result of peptide aggregation;
this phenomenon is not so easily detected with Tyr fluorescence [44]. RW-BP100 fluorescence
does not show REES, which supports a non-aggregated fluorophore population (data not
shown). Overall, the fluorescence results suggest that BP100 and its analogues are non-
structured and do not aggregate in aqueous solution.
The fluorescence emission quantum yield of the Tyr residue in BP100 or R-BP100 and
of the Trp residue in RW-BP100 increased upon titration with lipid vesicles. In addition, RW-
BP100 emission spectra have a blue shift characteristic of Trp residue transition from aqueous

15
 ACCEPTED MANUSCRIPT

environment to a more hydrophobic environment [44]. Such alterations in the Tyr/Trp

florescence suggest that these residues insert into the lipid bilayer. Therefore, the Tyr/Trp
fluorescence emission can be used to quantify the relative affinity of the these residues for lipid
membranes through the determination of the partition coefficient, Kp [32]. In the three peptides
the hydrophobic region binds with higher affinity to membranes with higher percentage of

T
negatively-charged phospholipids, relative to zwitterionic membranes, as shown with Kp values

IP
(Table 3). In addition, at a fixed lipid concentration the fluorescence emission spectra of RW-

R
BP100 has a larger blue-shift in POPC/POPG (1:1) relative to POPC membranes (e.g. 18 nm in
1 mM POPC/POPG (1:1) vs. 10 nm in 1 mM POPC). The trend observed for IL/IW (the ratio of

SC
the fluorescence emission obtained when all the fluorophores are partitioned in the lipid (IL) to
the fluorescence intensity when all the fluorophores are in water (IW)) is BP100 ~ R-BP100 <

NU
RW-BP100 suggesting that RW-BP100 has its Trp residue more deeply inserted in the
membrane, than the Tyr in R-BP100 or in BP100 peptide.
MA
3.9. Membrane in-depth location of BP100 and its analogues

The location of the Tyr/Trp residue when the peptides are bound to the membrane was
D

monitored by a differential quenching approach. 5- and 16-NS are stearic acid molecules
TE

derivatized (carbon 5 or 16) with a doxyl group, a quencher of the Tyr and Trp fluorescence.
With a differential depth location in the membrane, the doxyl group in 5-NS more efficiently
P

quenches fluorophores located in the water-membrane interface, whereas 16-NS is more

CE

efficient as a quencher of fluorophores more deeply inserted [33]. In addition, acrylamide, a

soluble quencher unable to partition extensively into lipidic membranes, was used to quench
fluorophores accessible to the aqueous environment [31] and its quenching efficiency was
AC

measured in the absence and presence of LUVs suspension.

The relative quenching efficiency can be compared with the Stern-Volmer constant, KSV
(Table 4). All of the peptides tested were more efficiently quenched by acrylamide in the
absence of model membranes than in their presence, suggesting that the peptides are inserted
into the lipid bilayer and less accessible to the aqueous environment. In addition, the
fluorescence of all the peptides is more efficiently quenched in the presence of POPC/POPG
(1:1 molar) relative to POPC model membranes, which confirms better insertion into
negatively-charged than into zwitterionic membranes.
In POPC membranes all of the peptides have the fluorescent residue more accessible to
5-NS than to 16-NS, suggesting a location in the interface. In POPC/POPG (1:1) membranes,
RW-BP100 is more efficiently quenched by 16-NS than 5-NS, suggesting a deeper insertion,
whereas the fluorescence of BP100 or R-BP100 is quenched by 5- and 16-NS with identical
efficiency. A deeper insertion of RW-BP100 into POPC/POPG (1:1) membrane is consistent

16
 ACCEPTED MANUSCRIPT

with the larger change in the dipolar potential (see Fig. 6B) and lower surface destabilization
observed by aggregation studies, when compared with the other two analogues (see Fig. 6C).
Altogether these results show that the Tyr/Trp region of the molecule can insert in the
hydrophobic region of the phospholipid membrane.

T
3.10. CD spectroscopy in the presence of membrane vesicles

R IP
To analyse the structure of the peptides in aqueous solution and in the presence of lipid
membrane we performed CD measurements (Fig. 7). In buffer all peptides had very low

SC
ellipticity above 210 nm and a strong negative band near 195 nm, which is typical of a random
coil structure [45]. A lack of structure was confirmed in the presence of urea, a denaturizing

NU
agent. No alterations in fluorescence or CD signals were detected for any of the peptides, even
with concentrations as high as 5 M urea (data not shown). In addition, all the peptides
MA
maintained a random coil conformation in buffer with pH 2 (data not shown).
In the presence of anionic vesicles the three peptides show a transition from
unstructured to helical form, revealing a change in conformation upon membrane insertion.
With POPC no alterations in the CD signal were detected, suggesting the requirement of anionic
D

phospholipids for an alteration in the conformation. A helical wheel conformation suggests an

TE

amphipathic structure with the nonpolar residues in one side and the positive side chains in the
other face of the helix.
P
CE

4. Discussion

In this study we designed two new AMPs, R-BP100 and RW-BP100, and compared
AC

their antimicrobial activities and toxic properties with the parent analogue BP100. Atomic force
microscopy with bacterial cells and biophysical studies with model membranes were employed
to gain insights on the mechanism of action of BP100 and its analogues.
MIC and MBC assays revealed that the newly designed AMPs are more active than
BP100, previously reported to be active against Gram-negative bacteria [14] but inactive against
the Gram-positive S. aureus [17]. Here we show that R-BP100 and RW-BP100 were not only
active against Gram-negative bacteria, but also had activity against Gram-positive bacteria at
low micromolar concentrations (see Table 1). Although RW-BP100 is the most toxic analogue,
it has the highest TI against the tested Gram-negative bacteria. To decrease the toxic properties
without compromising the amphipathic character, shifting the position of the Trp residue within
the hydrophobic part of the amphipathic helix is a possibility [46].
AFM imaging suggests that BP100, R-BP100 and RW-BP100 exert their action by
targeting the anionic bacterial surface and disrupting the membrane integrity. A mechanism

17
 ACCEPTED MANUSCRIPT

dependent on electrostatic attractions between the polycationic peptides and the anionic
microbial surface, followed by insertion into and permeabilization of the cytoplasmic
membrane, is supported by biophysical results obtained with model membranes (see Fig. 4-7).
Our studies revealed that the peptides in this study are able to bind and neutralize both LTA and
LPS, the negatively-charged molecules exposed at the Gram-negative and Gram-positive

T
bacteria, respectively. Furthermore, SPR and fluorescence studies revealed that the peptides are

IP
able to bind and insert into lipid bilayers, preferring anionic over neutral membranes. Upon

R
insertion into anionic membranes, the three peptides acquire an amphipathic helical
conformation (see Fig. 7B), insert into the lipid bilayer, as confirmed with the fluorescence

SC
properties of the Tyr/Trp residue (see Tables 3 and 4), and promote local perturbations, as
revealed by changes in the dipolar potential and vesicle aggregation (see Fig. 6). The

NU
amphipathic helical conformation maximizes both electrostatic and hydrophobic interactions
with the membrane as the positive face promotes binding to the anionic headgroups, whereas
MA
the nonpolar face favours contact with the hydrophobic part of the membrane [47] and allows
insertion of the molecule into the hydrophobic core of the lipid bilayer [11]. As a result of
binding, insertion and accumulation into the phospholipid bilayer, BP100 and its analogues
induce membrane disturbances that result in membrane permeabilization (see Fig. 6D).
D

A mechanism of action dependent on binding to the anionic bacterial surface, followed

TE

by peptide insertion into the membrane, and permeabilization is in agreement with previous
studies with BP100 [15, 17]. Biophysical studies with model membranes and E. coli cells
P

revealed that BP100 binds to the anionic surface of E. coli cells, inserts into the phospholipid
CE

bilayer and induces the leakage of cytoplasmic content from E. coli cells [17].
Electrostatic interactions and amphipathic structure are clearly important for the activity
of BP100, and its analogues, but they cannot explain the differences in the relative efficiency of
AC

the three peptides against Gram-positive bacteria. All the tested peptides acquire a helical
amphipathic structure upon interaction with negatively-charged membranes, have identical
ability to induce leakage from negatively-charged membranes, have the same global net charge
(+ 6) and are able to bind to LTA, but BP100 has weak activity against Gram-positive bacteria,
whereas R-BP100 and RW-BP100 are active. The activity of the three analogues against Gram-
positive bacteria tracks with their membrane-binding affinity. Even with the same charge and
preference for anionic membranes, R-BP100 and RW-BP100, are better adapted to establish
interactions with membranes, than BP100. The preference of these newly designed peptides for
the membrane environment is in agreement with their higher overall hydrophobicity, but might
also be the result of particular features of the Arg and Trp residues.
The higher membrane-binding affinity of R-BP100 over BP100 results from the
substitution of the Lys residues with Arg. Although Lys and Arg have the same charge, their
membrane-binding properties are different, suggesting that not only the charge but some

18
 ACCEPTED MANUSCRIPT

specific properties of the guanidinium side chain in Arg are responsible for this behaviour [48].
One possible explanation is the pronounced ability of the guanidinium group to establish strong
bidentate H-bonds with phosphate moieties from two lipid headgroups, whereas the Lys
ammonium side chain can only interact with a single lipid headgroup (Fig. 8A&B) [49]. In
addition of being better H-bond donors, guanidinium side chains also establish stronger

T
electrostatic cation- interactions with aromatic residues, than Lys residues [50]. Cation-

IP
interactions occur between basic residues (Arg and Lys) and aromatic residues (Phe, Tyr and

R
Trp) [50] and are important for peptide self-association within membranes and facilitate deeper
insertion into membranes by shielding cationic side chains [50, 51]. Furthermore, the Arg side

SC
chain is able to form H-bonds even when involved in cation- interactions, whereas the Lys side
chain cannot [52].

NU
RW-BP100, besides having the Lys mutated with Arg, also possesses a Trp instead of a
Tyr residue, which confers higher affinity and deeper insertion into the membrane relative to the
MA
analogue R-BP100 (see Fig. 6 and Table 4). These results are in agreement with previous
studies showing that the Trp residue not only partitions more favourably in the membrane
interface, but is also more hydrophobic and has higher affinity for bulk hydrophobic phases,
D

than Tyr [38]. In addition, the Trp residue facilitates the insertion of the cationic Arg into the
membrane hydrophobic region through cation- interactions [19] as above mentioned. The
TE

importance of Trp residues on the antimicrobial activity of peptides has been previously
correlated with their ability to bind to lipid membranes [18, 19, 53] and is further supported in
P

this study.
CE

Based on our findings we propose that to efficiently kill Gram-positive bacteria by a

membrane-dependent mechanism a peptide should be able to: target anionic membranes through
AC

electrostatic interactions; partition into the water-lipid interface regions of membranes through
hydrophobic interactions and H-bonds; and be able to more deeply insert into the hydrophobic
membrane environment, where the ability to establish cation- interactions seems to be
important to shield the charge of the cationic residues and disrupt the phospholipid membrane.
The three peptides are highly active against the tested Gram-negative bacteria, showing
that these bacteria can be targeted and disrupted by peptides containing either Arg or Lys
residues. We propose that to kill Gram-negative bacteria, H-bond formation and cation-
interactions are not as important as for Gram-positive bacteria. The Gram-negative bacterial
surface is covered with LPS molecules, which form a very compact and impermeable layer due
to non-covalent associations with divalent (calcium and magnesium) cations [54]; the activity of
cationic peptides against Gram-negative bacteria has been correlated with their ability to bind
and cross the LPS barrier by displacing divalent cations from their binding sites [1, 41, 54] and
eventually disrupting the cytoplasmic membrane. A mechanism of action dependent on binding

19
 ACCEPTED MANUSCRIPT

and crossing the LPS barrier before disrupting the inner membrane is consistent with the E. coli
strain CGSC 5167 (with rough LPS, and therefore more penetrable) being more susceptible to
all the peptides, than the control strain ATCC 25922 (with smooth LPS) (see Table 1). The
importance of Arg and Trp residues for activity against Gram-positive, but not Gram-negative
bacteria, could be the result of a larger peptidoglycan layer in the Gram-positive outer

T
membrane, relative to Gram-negative bacteria, which requires the unique hydrogen binding

IP
geometry of Arg and is facilitated by the complex amphipathic properties of Trp that together

R
promote stable and deep insertion into the cell wall of Gram-positive bacteria.

SC
5. Conclusions

NU
In this study we designed two BP100 analogues, which in addition to being active
MA
against the tested Gram-negative were also active against Gram-positive bacteria. Identification
of peptides with high activity against Gram-negative and Gram-positive bacteria is of major
relevance due to the development of strains with resistance to common antibiotics. In fact, the
novel peptides were here shown to be active against several pathogenic bacteria, including
D

strains with antibiotic resistance.

TE

Our studies revealed that whereas the antimicrobial activity of BP100 and its analogues
against Gram-negative bacteria seems to be mainly dependent on an ability to bind to LPS layer
P

through electrostatic interactions and membrane disruption through hydrophobic interactions.

CE

The activity against Gram-positive, although dependent on electrostatic attractions, and

phospholipid membrane disruption, requires further interactions with the cell wall components
in which H-bonds and cation- interactions might play a role. Overall, the results obtained in
AC

this study provide valuable information on the correlation between amino acid sequence and
antimicrobial activity that can guide the development of new AMPs with either specific activity
or a broader spectrum of action.

Acknowledgements

IMT work was supported by a grant (PTDC/SAU-BEB/099142/2008), HGF and DG held a

scholarship from the Fundação para a Ciência e Tecnologia, Portugal, SFRH/BD/39039/2007
and SFRH/BPD/73500/2010, respectively. STH is the recipient of a Discovery Early Career
Researcher Award (DE120103152), awarded by Australian Research Council. DJC is a National
Health and Medical Research Council Fellow. We thank Marta Ribeiro (IMM, Lisbon
University) for the valuable scientific discussions. We thank Dr Mark Blaskovich (IMB, UQ)

20
 ACCEPTED MANUSCRIPT

for kindly providing the bacterial strains employed in this study and Angela Kavanagh (IMB,
UQ) for the help in establishing the antimicrobial assays.

References

T
[1] M. Zasloff, Antimicrobial peptides of multicellular organisms, Nature, 415 (2002) 389-395.

IP
[2] M.L. Mangoni, Y. Shai, Short native antimicrobial peptides and engineered ultrashort

R
lipopeptides: similarities and differences in cell specificities and modes of action, Cell Mol Life
Sci, 68 (2011) 2267-2280.

SC
[3] M. Pasupuleti, A. Schmidtchen, M. Malmsten, Antimicrobial peptides: key components of
the innate immune system, Crit Rev Biotechnol, 32 (2012) 143-171.

NU
[4] M.L. Mangoni, Y. Shai, Temporins and their synergism against Gram-negative bacteria and
in lipopolysaccharide detoxification, Biochim biophys acta, 1788 (2009) 1610-1619.
MA
[5] W.C. Wimley, Describing the mechanism of antimicrobial peptide action with the interfacial
activity model, ACS Chem Biol, 5 (2010) 905-917.
[6] M.R. Yeaman, N.Y. Yount, Mechanisms of antimicrobial peptide action and resistance,
Pharmacol Rev, 55 (2003) 27-55.
D

[7] K. Matsuzaki, K. Sugishita, M. Harada, N. Fujii, K. Miyajima, Interactions of an

TE

antimicrobial peptide, magainin 2, with outer and inner membranes of Gram-negative bacteria,
Biochim biophys acta, 1327 (1997) 119-130.
P

[8] R. Bessalle, H. Haas, A. Goria, I. Shalit, M. Fridkin, Augmentation of the antibacterial

CE

activity of magainin by positive-charge chain extension, Antimicrob Agents Chemother, 36

(1992) 313-317.
[9] M. Dathe, H. Nikolenko, J. Meyer, M. Beyermann, M. Bienert, Optimization of the
AC

antimicrobial activity of magainin peptides by modification of charge, FEBS Lett, 501 (2001)
146-150.
[10] C. Friedrich, M.G. Scott, N. Karunaratne, H. Yan, R.E. Hancock, Salt-resistant alpha-
helical cationic antimicrobial peptides, Antimicrob Agents Chemother, 43 (1999) 1542-1548.
[11] I. Zelezetsky, A. Tossi, Alpha-helical antimicrobial peptides--using a sequence template to
guide structure-activity relationship studies, Biochim biophys acta, 1758 (2006) 1436-1449.
[12] C.B. Park, K.S. Yi, K. Matsuzaki, M.S. Kim, S.C. Kim, Structure-activity analysis of
buforin II, a histone H2A-derived antimicrobial peptide: the proline hinge is responsible for the
cell-penetrating ability of buforin II, Proc Natl Acad Sci U S A, 97 (2000) 8245-8250.
[13] M. Cudic, L. Otvos, Jr., Intracellular targets of antibacterial peptides, Curr Drug Targets, 3
(2002) 101-106.

21
 ACCEPTED MANUSCRIPT

[14] E. Badosa, R. Ferre, M. Planas, L. Feliu, E. Besalu, J. Cabrefiga, E. Bardaji, E.

Montesinos, A library of linear undecapeptides with bactericidal activity against
phytopathogenic bacteria, Peptides, 28 (2007) 2276-2285.
[15] R. Ferre, M.N. Melo, A.D. Correia, L. Feliu, E. Bardaji, M. Planas, M. Castanho,
Synergistic effects of the membrane actions of cecropin-melittin antimicrobial hybrid peptide

T
BP100, Biophys J, 96 (2009) 1815-1827.

IP
[16] K. Eggenberger, C. Mink, P. Wadhwani, A.S. Ulrich, P. Nick, Using the peptide BP100 as

R
a cell-penetrating tool for the chemical engineering of actin filaments within living plant cells,
Chembiochem : a European journal of chemical biology, 12 (2011) 132-137.

SC
[17] C.S. Alves, M.N. Melo, H.G. Franquelim, R. Ferre, M. Planas, L. Feliu, E. Bardaji, W.
Kowalczyk, D. Andreu, N.C. Santos, M.X. Fernandes, M.A. Castanho, Escherichia coli cell

NU
surface perturbation and disruption induced by antimicrobial peptides BP100 and pepR, J Biol
Chem, 285 (2010) 27536-27544.
MA
[18] K. Hilpert, R. Volkmer-Engert, T. Walter, R.E. Hancock, High-throughput generation of
small antibacterial peptides with improved activity, Nat Biotechnol, 23 (2005) 1008-1012.
[19] D.I. Chan, E.J. Prenner, H.J. Vogel, Tryptophan- and arginine-rich antimicrobial peptides:
structures and mechanisms of action, Biochim biophys acta, 1758 (2006) 1184-1202.
D

[20] Y.B. Huang, X.F. Wang, H.Y. Wang, Y. Liu, Y. Chen, Studies on mechanism of action of
TE

anticancer peptides by modulation of hydrophobicity within a defined structural framework,

Mol Cancer Ther, 10 (2011) 416-426.
P

[21] I. Wiegand, K. Hilpert, R.E. Hancock, Agar and broth dilution methods to determine the
CE

minimal inhibitory concentration (MIC) of antimicrobial substances, Nat Protoc, 3 (2008) 163-
175.
[22] P. Eaton, J.C. Fernandes, E. Pereira, M.E. Pintado, F. Xavier Malcata, Atomic force
AC

microscopy study of the antibacterial effects of chitosans on Escherichia coli and

Staphylococcus aureus, Ultramicroscopy, 108 (2008) 1128-1134.
[23] B. Fang, S. Gon, M. Park, K.N. Kumar, V.M. Rotello, K. Nusslein, M.M. Santore,
Bacterial adhesion on hybrid cationic nanoparticle-polymer brush surfaces: ionic strength tunes
capture from monovalent to multivalent binding, Colloids Surf B Biointerfaces, 87 (2011) 109-
115.
[24] Y.H. Huang, M.L. Colgrave, R.J. Clark, A.C. Kotze, D.J. Craik, Lysine-scanning
mutagenesis reveals an amendable face of the cyclotide kalata B1 for the optimization of
nematocidal activity, J Biol Chem, 285 (2010) 10797-10805.
[25] Z. Zidek, E. Kmonickova, P. Kostecka, P. Jansa, Microfiltration method of removal of
bacterial contaminants and their monitoring by nitric oxide and Limulus assays, Nitric oxide :
biology and chemistry / official journal of the Nitric Oxide Society, 28C (2012) 1-7.

22
 ACCEPTED MANUSCRIPT

[26] S.T. Henriques, M.A. Castanho, Consequences of nonlytic membrane perturbation to the
translocation of the cell penetrating peptide pep-1 in lipidic vesicles, Biochemistry, 43 (2004)
9716-9724.
[27] S.T. Henriques, L.K. Pattenden, M.I. Aguilar, M.A. Castanho, PrP(106-126) does not
interact with membranes under physiological conditions, Biophys J, 95 (2008) 1877-1889.

T
[28] S.T. Henriques, M.A. Castanho, L.K. Pattenden, M.I. Aguilar, Fast membrane association

IP
is a crucial factor in the peptide pep-1 translocation mechanism: a kinetic study followed by

R
surface plasmon resonance, Biopolymers, 94 (2010) 314-322.
[29] P.M. Matos, M.A. Castanho, N.C. Santos, HIV-1 fusion inhibitor peptides enfuvirtide and

SC
T-1249 interact with erythrocyte and lymphocyte membranes, PLoS One, 5 (2010) e9830.
[30] Y.H. Huang, M.L. Colgrave, N.L. Daly, A. Keleshian, B. Martinac, D.J. Craik, The

NU
biological activity of the prototypic cyclotide kalata b1 is modulated by the formation of
multimeric pores, J Biol Chem, 284 (2009) 20699-20707.
MA
[31] S.T. Henriques, M.A. Castanho, Environmental factors that enhance the action of the cell
penetrating peptide pep-1 A spectroscopic study using lipidic vesicles, Biochim biophys acta,
1669 (2005) 75-86.
[32] N.C. Santos, M. Prieto, M.A.R.B. Castanho, Quantifying molecular partition into model
D

systems of biomembranes: an emphasis on optical spectroscopic methods Biochim biophys

TE

acta, 1612 (2003) 123-135.

[33] M.X. Fernandes, J. Garcia de la Torre, M.A. Castanho, Joint determination by Brownian
P

dynamics and fluorescence quenching of the in-depth location profile of biomolecules in

CE

membranes, Anal Biochem, 307 (2002) 1-12.

[34] M.A. Castanho, M.J. Prieto, Fluorescence quenching data interpretation in biological
systems. The use of microscopic models for data analysis and interpretation of complex
AC

systems, Biochim biophys acta, 1373 (1998) 1-16.

[35] Y. Shai, Mode of action of membrane active antimicrobial peptides, Biopolymers, 66
(2002) 236-248.
[36] W.C. Wimley, S.H. White, Experimentally determined hydrophobicity scale for proteins at
membrane interfaces, Nat struct biol, 3 (1996) 842-848.
[37] J.L. MacCallum, D.P. Tieleman, Hydrophobicity scales: a thermodynamic looking glass
into lipid-protein interactions, Trends Biochem Sci, 36 (2011) 653-662.
[38] S.H. White, W.C. Wimley, Hydrophobic interactions of peptides with membrane
interfaces, Biochim biophys acta, 1376 (1998) 339-352.
[39] M.N. Melo, R. Ferre, L. Feliu, E. Bardaji, M. Planas, M.A. Castanho, Prediction of
antibacterial activity from physicochemical properties of antimicrobial peptides, PLoS One, 6
(2011) e28549.

23
 ACCEPTED MANUSCRIPT

[40] T.J. Silhavy, D. Kahne, S. Walker, The bacterial cell envelope, Cold Spring Harb perspect
biol, 2 (2010) a000414.
[41] Y. Rosenfeld, Y. Shai, Lipopolysaccharide (Endotoxin)-host defense antibacterial peptides
interactions: role in bacterial resistance and prevention of sepsis, Biochim biophys acta, 1758
(2006) 1513-1522.

T
[42] P.M. Matos, H.G. Franquelim, M.A. Castanho, N.C. Santos, Quantitative assessment of

IP
peptide-lipid interactions. Ubiquitous fluorescence methodologies, Biochim biophys acta, 1798

R
(2010) 1999-2012.
[43] A. Mulgrew-Nesbitt, K. Diraviyam, J. Wang, S. Singh, P. Murray, Z. Li, L. Rogers, N.

SC
Mirkovic, D. Murray, The role of electrostatics in protein-membrane interactions, Biochim
biophys acta, 1761 (2006) 812-826.

NU
[44] N.C. Santos, M.A.R.B. Castanho, Fluorescence spectroscopy methodologies on the study
of proteins and peptides. On the 150th anniversary of protein fluorescence Trends in Applied
MA
Spectroscopy, 4 (2002) 113-125.
[45] N.J. Greenfield, Using circular dichroism spectra to estimate protein secondary structure,
Nat Protoc, 1 (2006) 2876-2890.
[46] O. Rekdal, B.E. Haug, M. Kalaaji, H.N. Hunter, I. Lindin, I. Israelsson, T. Solstad, N.
D

Yang, M. Brandl, D. Mantzilas, H.J. Vogel, Relative spatial positions of tryptophan and cationic
TE

residues in helical membrane-active peptides determine their cytotoxicity, J Biol Chem, 287
(2012) 233-244.
P

[47] J. Seelig, Thermodynamics of lipid-peptide interactions, Biochim biophys acta, 1666

CE

(2004) 40-50.
[48] Y. Su, T. Doherty, A.J. Waring, P. Ruchala, M. Hong, Roles of arginine and lysine residues
in the translocation of a cell-penetrating peptide from (13)C, (31)P, and (19)F solid-state NMR,
AC

Biochemistry, 48 (2009) 4587-4595.

[49] J.B. Rothbard, T.C. Jessop, P.A. Wender, Adaptive translocation: the role of hydrogen
bonding and membrane potential in the uptake of guanidinium-rich transporters into cells, Adv
Drug Deliv Rev, 57 (2005) 495-504.
[50] J.P. Gallivan, D.A. Dougherty, Cation-pi interactions in structural biology, Proc Natl Acad
Sci U S A, 96 (1999) 9459-9464.
[51] S. Mecozzi, A.P. West, Jr., D.A. Dougherty, Cation-pi interactions in aromatics of
biological and medicinal interest: electrostatic potential surfaces as a useful qualitative guide,
Proc Natl Acad Sci U S A, 93 (1996) 10566-10571.
[52] M.P. Aliste, J.L. MacCallum, D.P. Tieleman, Molecular dynamics simulations of
pentapeptides at interfaces: salt bridge and cation-pi interactions, Biochemistry, 42 (2003) 8976-
8987.

24
 ACCEPTED MANUSCRIPT

[53] A.A. Stromstedt, M. Pasupuleti, A. Schmidtchen, M. Malmsten, Evaluation of strategies

for improving proteolytic resistance of antimicrobial peptides by using variants of EFK17, an
internal segment of LL-37, Antimicrob Agents Chemother, 53 (2009) 593-602.
[54] L. Leive, The barrier function of the gram-negative envelope, Ann N Y Acad Sci, 235
(1974) 109-129.

T
R IP
SC
NU
MA
D
P TE
CE
AC

25
 ACCEPTED MANUSCRIPT

Figure captions

Figure 1. Theoretical prediction of the affinity of the peptides for the membrane environment.
The free transfer energy of (A) BP100, (B) R-BP100 and (C) RW-BP100 from water to POPC
membranes interface (ΔGwater-membrane interface) is represented based on Wimley & White amino

T
acid interfacial hydrophobic scale [36]. Amino acids with ΔGwater-membrane interface < 0 have higher

IP
affinity for the membrane interface region than for the aqueous environment. The new
derivatives, R-BP100 and RW-BP100, have the Lys residues mutated with Arg residues (white

R
bars) and RW-BP100 also has the Tyr residue replaced with a Trp residue (shaded bar). The

SC
residues that were not mutated are represented with black bars.

Figure 2. The effect of BP100, R-BP100 and RW-BP100 on E. coli and S. aureus cells as

NU
monitored by AFM. (A, B) Orthogonal tridimensional projections of E. coli (A) and S. aureus
(B) cells incubated for 2 hours in the absence (control) or in the presence of 1, 10 or 100 µM of
peptide. Images have a total area of 4×4 µm2. (C, D) Differences in the superficial roughness
MA
and (E, F) in bacterial shape of E. coli and S. aureus cells were determined through root-mean-
square calculation (Rrms and Srms, respectively). For each peptide concentration five bacterial
cells were analysed. Asterisks represent the significance level of the difference between treated
D

and non-treated bacterial cells. A one-way ANOVA and a Bonferroni test were employed. *
TE

0.01< p-value < 0.05; ** 0.001 < p-value < 0.01 and *** p-value < 0.001.
P

Figure 3. Cytotoxicity and haemolytic activity of BP100 and its analogues. (A) Cytotoxicity
dose response curves obtained against HELA cells as measured with a resazurin assay. Peptides
CE

were incubated with HELA cells for 2h at 37ºC in medium without serum. The % of dead cells
was determined by absorbance of the ratio of reduced/oxidized dye. The concentration needed
AC

to achieve 50% of cell death (IC50) as obtained by a non-linear fit of the data was 49.2 ± 1.4,
17.5 ± 0.6 and 14.4 ± 0.4 μM for BP100, R-BP100 and RW-BP100, respectively. (B)
Haemolysis dose response curves obtained after incubation of 0.25% (v/v) RBCs with peptide
for 1h at 37ºC; HC50 is > 100 μM for BP100, 50.9 ± 25.0 μM for R-BP100 and 4.9 ± 1.0 μM for
RW-BP100. Error bars represent standard error of the mean for three replicates.

Figure 4. Neutralization of LTA and LPS in the presence of BP100 or its analogues. Peptides
were incubated with endotoxin for 30 min and the amount of free endotoxin quantified by LAL
assay. The fraction of bound endotoxin as a function of peptide is shown. K D is the peptide
concentration required to neutralize 50% of endotoxin and was determined by non-linear fit of
the data. Error bars represent standard error of the mean of three replicates.

Figure 5. Membrane-binding affinity of BP100 and its analogues as monitored by surface

plasmon resonance. The response to the total amount of lipid deposited on the chip surface was

26
 ACCEPTED MANUSCRIPT

normalized by calculating the peptide-to-lipid ratio (P/L mol/mol) (1RU = 1 pg.mm-2).

Sensorgrams obtained upon injection of 10 M peptide over (A) POPC or (B) POPC/POPG
(1:1) bilayers. (C) Comparison of the P/L obtained for the three peptides at a reporting point at
the end of the association phase over POPC, POPC/POPG (4:1) or POPC/POPG (1:1) bilayers.
Error bars represent standard error of the mean of three replicates.

T
IP
Figure 6. Changes in membrane properties due to interaction with BP100 and its analogues. (A,
B) Membrane dipole potential variation induced by 50 µM of BP100, R-BP100 or RW-BP100

R
in 200 µM of (A) POPC or (B) POPC/POPG (1:1) vesicles. The final concentration of the di-8-

SC
ANNEPS probe was 4 µM. All the peptides induce a larger dipolar potential variation in
POPC/POPG (1:1 molar) than in POPC vesicles. RW-BP100 induces the largest dipolar

NU
potential variation. (C) Vesicle aggregation studies. OD436nm was monitored after addition of 25
µM of BP100, R-BP100 or RW-BP100 to 500 µM of POPC/POPG (1:1 molar) vesicles, or 50
µM of BP100 to POPC or POPC/POPG (4:1 molar) vesicles. The OD values were normalized
MA
for the baseline. No aggregation and, therefore, no increase in OD was observed for POPC or
POPC/POPG (4:1) vesicles. The three peptides induce aggregation of POPC/POPG (1:1)
vesicles. (D) Kinetics of vesicle leakage induced by 25 µM of BP100, R-BP100 and RW-BP100
D

in the presence of 50 µM POPC/POPG (1:1 molar) and POPC vesicles, as followed by

TE

dequenching of CF. The fluorescence signal was converted into % of leakage; TX-100 0.1%
(v/v) was added at the end of each kinetics (not shown) to establish 100% of leakage in each
P

assay.
CE

Figure 7. CD spectra of BP100 and its analogues in the absence or presence of lipid vesicles.
CD spectra of 50 µM of BP100 (A), R-BP100 (B) and RW-BP100 (C) in 10 mM phosphate
AC

buffer containing 150 mM NaF or in the presence of POPC (500 µM) or in the presence of
POPC/POPG (1:1) (500 µM for R-BP100 and RW-BP100 and 1mM for BP100). Insert in panel
B shows the top view of predicted helical three-dimensional structure of R-BP100 in anionic
membrane environment putting in evidence the amphipathic character of the helix. Hydrophilic
amino acid residues are coloured in blue, basic amino acids are coloured in green and Tyr
residue is coloured in light green. The side chains of Phe and Tyr residues are represented by
ball and stick. The image was modulated using PyMOL Molecular Graphics System, version
1.2r3pre (Schrödinger, LLC).

Figure 8. Representation of the predicted helical three-dimensional structure of BP100 and its
analogues in anionic membrane environment. (A) Side view of BP100 showing the ability of
Lys residues to establish monodentate H-bonds. (B) Side view of RW-BP100 showing the
ability of Arg residues to establish bidentate H-bonds with phosphate groups from two different

27
 ACCEPTED MANUSCRIPT

phospholipids. The colour scheme is the same as in Fig. 7B; the two images were modulated
using PyMOL Molecular Graphics System, version 1.2r3pre (Schrödinger, LLC).

T
R IP
SC
NU
MA
D
P TE
CE
AC

28
 ACCEPTED MANUSCRIPT

PT
Tables

RI
Table 1. Antimicrobial activity of BP100, R-BP100 and RW-BP100.a

SC
MIC (M)b MBC (M)c
Bacterial strain and species Relevant feature
BP100 R-BP100 RW-BP100 BP100R-BP100 RW-BP100
Gram negative

NU
Escherichia coli ATCC 25922 control 2.00.0 0.90.4 0.50.2 3.00.6 1.10.3 1.30.3
Escherichia coli CGSC 5167 resistant, LPS mutant 0.70.2 0.40.1 0.20.1 1.30.5 0.50.0 0.30.1
Klebsiella pneumoniae ATCC 13883 control 1.00.0 0.30.2 0.10.0 3.50.5 0.40.1 0.40.1

MA
Klebsiella pneumoniae ATCC 700603 multiresistant 4.00.0 0.60.3 0.30.0 4.00.0 2.10.7 0.90.1
Pseudomonas aeruginosa ATCC 10145 control 8.00.0 2.00.0 2.50.5 8.00.0 4.51.3 3.50.5
Pseudomonas aeruginosa ATCC 27853 resistant 4.00.0 1.50.3 1.50.3 4.00.0 1.50.3 1.80.3

ED
Gram positive
Staphylococcus aureus ATCC 25923 control 32.00.0 6.01.2 0.80.1 32.00.0 8.00.0 1.00.0
Staphylococcus aureus ATCC 33591 resistant 16.00.0 3.00.6 0.30.1 28.04.0 4.00.0 0.60.1

PT
Streptococcus pneumoniae ATCC 33400 control 32.00.0 5.01.0 0.80.1 32.00.0 7.01.0 1.00.0
Streptococcus pneumoniae ATCC 700677 resistant 16.00.0 4.00.0 0.40.1 28.04.0 6.01.2 0.90.1
Enterococcus faecium ATCC 35667
Enterococcus faecium ATCC 51559
CE control
multiresistant
32.00.0
16.00.0
4.00.0
2.00.0
1.00.0
1.00.0
32.00.0
16.00.0
4.00.0
2.00.0
1.10.3
1.00.0
a
The antimicrobial activity was determined in MHB. b MIC is the lowest concentration showing no visible growth. c MBC is the lowest concentration
AC

required to kill all the cells as measured with resazurin dye. The values shown are the average and standard error of four replicates.

29
 ACCEPTED MANUSCRIPT

Table 2. Vesicle leakage induced by BP100, R-BP100 and RW-BP100.

Peptide Lipid mixture LC50 ± SE (μM) b T1/2 ± SE (s)c
POPC > 100 6.21± 0.30
BP100
POPC/POPG (1:1) 3.71 ± 0.11 4.85 ± 0.04
POPC 83.28 ± 8.89 3.67 ± 0.23

T
R-BP100
POPC/POPG (1:1) 5.06 ± 1.66 0.37 ± 0.04

IP
POPC 10.71 ± 0.67 0.63± 0.07
RW-BP100
POPC/POPG (1:1) 5.16 ± 0.34 0.14 ±0.01

R
a
Peptides were incubated with vesicles suspensions containing 50 M of
total lipid concentration. b The peptide concentration needed to achieve

SC
50% leakage (LC50) and the standard errors were determined by fitting a
non-linear regression equation in which the maximum response was fixed
to 100% leakage. c Time required to achieve half of the maximum response
upon addition of 25 M of peptide to lipid suspension containing 50 M of
lipid.

NU
MA
D
P TE
CE
AC

30
 ACCEPTED MANUSCRIPT

Table 3. Affinity of BP100, R-BP100 and RW-BP100 for different

lipid mixtures.a
Peptide Lipid mixture Kp (×103) IL/IW
POPC 0.26 ± 0.9 2.11 ± 1.86
BP100 POPC/POPG (4:1) 1.19 ± 0.17 2.12 ± 0.67
POPC/POPG (1:1) 3.97 ±1.05 2.27 ± 1.27

T
POPC 1.51 ± 0.11 2.51 ±0.42

IP
R-BP100 POPC/POPG (4:1) 3.54 ± 0.36 2.86 ± 0.62
POPC/POPG (1:1) 6.26 ± 0.96 2.72 ± 0.88

R
POPC 1.12 ± 0.06 4.30 ± 0.59
RW-BP100 POPC/POPG (4:1) 3.09 ± 0.56 3.98 ± 1.62

SC
POPC/POPG (1:1) 27.9 ± 7.90 3.41 ± 1.99
a
Peptides were titrated with LUV suspensions and the fluorescence
emission signal (I/IW) plotted as a function of lipid concentration.

NU
The Kp and IL/IW parameters were obtained from fitting a non-linear
equation (see Fig. 5).
MA
D
P TE
CE
AC

31
 ACCEPTED MANUSCRIPT

Table 4. Emission fluorescence quenching of BP100, R-BP100 and RW-BP100 induced by

5-NS, 16-NS or acrylamide.a
Peptide Lipid mixture KSV, 5 NS (M-1) KSV, 16 NS (M-1) KSV, acrylamide (M-1)

T
POPC 0.80 ± 0.16 0.55 ± 0.23 9.60 ± 0.82

IP
BP100 POPC/POPG (1:1) 2.47 ± 0.06 2.02 ± 0.06 1.94 ± 0.29
Buffer 34.10 ± 1.57

R
b
POPC 1.36 ± 2.10 0.21 ± 0.22 4.88 ± 0.30

SC
R-BP100 POPC/POPG (1:1) 1.88 ± 0.09 1.72 ± 0.15 1.37 ± 0.26
Buffer 8.80 ± 0.22
POPC 7.48 ± 1.47 2.86 ± 0.34 4.79 ± 0.34

NU
RW-BP100 POPC/POPG (1:1) 0.68 ± 0.24 2.19 ± 0.27 3.23 ± 0.21
Buffer 18.40 ± 0.63
a
Each peptide was titrated with a quenching agent (5-NS, 16-NS or acrylamide) in the
MA
presence or absence of LUVs and quantified by Stern Volmer constant, KSV. KSV and
standard errors were determined by fitting the Stern Volmer equation. b Stern-Volmer
representation shows a positive deviation, therefore KSV was determined by fitting a
quenching with sphere-of-action.
D
P TE
CE
AC

32
 ACCEPTED MANUSCRIPT

T
IP
Fig. 1

R
SC
NU
MA
D
PTE
CE
AC

33
 ACCEPTED MANUSCRIPT

T
IP
R
SC
Fig. 2
NU
MA
D
PTE
CE
AC

34
 ACCEPTED MANUSCRIPT

T
IP
R
SC
Fig. 3

NU
MA
D
PTE
CE
AC

35
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
Fig. 4 MA
D
PTE
CE
AC

36
 ACCEPTED MANUSCRIPT

T
IP
R
Fig. 5

SC
NU
MA
D
PTE
CE
AC

37
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D

Fig. 6
PTE
CE
AC

38
 ACCEPTED MANUSCRIPT

T
IP
R
SC
NU
MA
D
PTE
CE
AC

Fig. 7

39
 ACCEPTED MANUSCRIPT

T
IP
R
SC
Fig. 8

NU
MA
D
PTE
CE
AC

40
 ACCEPTED MANUSCRIPT

T
IP
R
SC
Graphical Abstract

NU
MA
D
P TE
CE
AC

41
 ACCEPTED MANUSCRIPT

Highlights

 Two novel antimicrobial peptides (AMPs) active against Gram- and Gram+ bacteria
 Arg instead of Lys residues improve AMP activity against Gram+ bacteria.


T
AMP mode of action against Gram+ requires electrostatic binding but also deeply
insertion in the membrane

IP
 Lys vs Arg residues relevant for AMP specificity vs broad spectra of action.

R
SC
NU
MA
D
P TE
CE
AC

42