Key residues in octyl-tridecaptin A1 analogues linked to stable secondary structures in the membrane

Article abstract of DOI:10.1002/cbic.201402024

Tridecaptin A₁ is a linear antimicrobial lipopeptide comprised of 13 amino acids, including three diaminobutyric acid (Dab) residues. It displays potent activity against Gram-negative bacteria, including multidrug-resistant strains. Using solid-phase peptide synthesis, we performed an alanine scan of a fully active analogue, octyl-tridecaptin A₁, to determine key residues responsible for activity. The synthetic analogues were tested against ten organisms, both Gram-positive and Gram-negative bacteria. Modification of D-Dab8 abolished activity, and marked decreases were observed with substitution of D-allo-Ile12 and D-Trp5. Circular dichroism showed that octyl-tridecaptin A₁ adopts a secondary structure in the presence of model phospholipid membranes, which was weakened by D-Dab8-D-Ala, D-allo-Ile12-D-Ala, and D-Trp5-D-Ala substitutions. The antimicrobial activity of the analogues is directly correlated to their ability to adopt a stable secondary structure in a membrane environment. Just tri it! Analysis of tridecaptin A₁ and its octyl analogue have identified key residues responsible for the formation of a stable secondary structure in a model membrane environment. A combination of alanine scanning and CD spectroscopy showed that modification of these residues prevented formation of the secondary structure and abolished antimicrobial activity.

Full text of DOI:10.1002/cbic.201402024

CHEMBIOCHEM
FULL PAPERS
DOI: 10.1002/cbic.201402024
Key Residues in Octyl-Tridecaptin A1 Analogues Linked to
Stable Secondary Structures in the Membrane
Stephen A. Cochrane,[a] Brandon Findlay,[a] John C. Vederas,*[a] and Elaref S. Ratemi*[b]
Tridecaptin A1 is a linear antimicrobial lipopeptide comprised
of 13 amino acids, including three diaminobutyric acid (Dab)
residues. It displays potent activity against Gram-negative bac-
teria, including multidrug-resistant strains. Using solid-phase
peptide synthesis, we performed an alanine scan of a fully
active analogue, octyl-tridecaptin A1, to determine key residues
responsible for activity. The synthetic analogues were tested
against ten organisms, both Gram-positive and Gram-negative
bacteria. Modification of d-Dab8 abolished activity, and
marked decreases were observed with substitution of d-allo-
Ile12 and d-Trp5. Circular dichroism showed that octyl-tri-
decaptin A1 adopts a secondary structure in the presence of
model phospholipid membranes, which was weakened by d-
Dab8-d-Ala, d-allo-Ile12-d-Ala, and d-Trp5-d-Ala substitutions.
The antimicrobial activity of the analogues is directly correlated
to their ability to adopt a stable secondary structure in a mem-
brane environment.
Introduction
The golden era of antibiotics is disappearing, and recent years
have seen a drastic decline in the number of new clinically ap-
proved compounds. In the past 40 years, only a few structural-
ly and mechanistically distinct antibiotics have entered the
market.[1] One of these examples is the lipopeptide daptomy-
cin, whose success has led to a recent surge in interest in the
use of lipopeptides as antibiotics, with several compounds
now in various clinical trials.[2] Lipopeptides generally exert
their bactericidal effect through disruption of the bacterial
membrane, and as it is difficult for bacteria to reorganize their
membranes, this limits the development of resistant strains. As
such, lipopeptides are attractive candidates for tackling multi-
drug-resistant organisms, and the discovery of new classes of
lipopeptides might lead to new antibiotic candidates.
tive bacteria, including the foodborne pathogen Campylobact-
er jejuni and a multidrug-resistant Klebsiella pneumoniae strain.
The cytotoxicity and haemolytic activities of tridecaptin A1 are
low, making it a good candidate for food preservation and clin-
ical applications.
Although tridecaptin A1 is a well-characterized example of a
linear cationic lipopeptide, little is known about how it exerts
its bactericidal effect. Understanding the mode of action of
octyl-tridecaptin A1 is important, as it could enable the rational
design of more active analogues or unveil new antibiotic tar-
gets. By sequentially substituting each amino acid residue with
alanine, we were able to determine the key residues responsi-
ble for biological activity. The change in activity was then
linked to conformational changes in the peptides, which were
shown by circular dichroism (CD) to be disordered in aqueous
media and trifluoroethanol but adopt a defined secondary
structure in the presence of 50 nm phospholipid large uni-
lamellar vesicles (LUV). The alanine substitutions that had the
greatest detrimental effect on antimicrobial activity in our
panel of ten Gram-positive and Gram-negative bacteria,
namely d-Dab8-d-Ala, d-allo-Ile12-d-Ala, and d-Trp5-d-Ala, also
significantly reduced peptide secondary structure. There is
therefore a direct relationship between the secondary structure
of tridecaptin A1 and the observed antimicrobial activity.
Tridecaptin A1 is a linear antimicrobial lipopeptide produced
by Paenibacillus polymyxa and Paenibacillus terrae.[3,4] It is a 13-
residue peptide containing a mixture of d- and l-amino acids
that is N-terminally acylated with (3R)-hydroxy-(6S)-methylocta-
noic acid. Our group recently showed that the chiral lipid tail
can be substituted with octanoic acid to give octyl-tridecaptin
A1, which retains full activity and enables rapid and facile syn-
thesis of tridecaptin A1 analogues.[5] Octyl-tridecaptin A1 dis-
plays potent activity in the low-mm range against Gram-nega-
[a] S. A. Cochrane, Dr. B. Findlay, Dr. J. C. Vederas
Department of Chemistry, University of Alberta
11227 Saskatchewan Drive, Edmonton, Alberta, T6G 2G2 (Canada)
E-mail: john.vederas@ualberta.ca
Results and Discussion
Peptide synthesis and antimicrobial testing
[b] Dr. E. S. Ratemi
Department of Chemical and Process Engineering Technology
Jubail Industrial College
Jubail Industrial City, 31961, P.O. Box 10099 (Kingdom of Saudi Arabia)
E-mail: ratemi_e@jic.edu.sa
A complete alanine scan was performed on octyl-tridecaptin
A1. Peptides were synthesized on Wang resin by using Fmoc
solid-phase peptide synthesis (Scheme 1). l- and d-amino acids
were replaced with l-Ala and d-Ala, respectively. The N termi-
nus was acylated on-resin with octanoic acid. The synthesized
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201402024.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 1295 – 1299 1295

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
Scheme 1. Fmoc solid-phase peptide synthesis of octyl-tridecaptin A1 and alanine scan analogues. a) piperidine (20% in DMF), 3ꢁ5 min; b) Fmoc-X-OH, 1-
[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate, N,N-diisopropylethylamine, DMF, 1 h; c) continue SPPS; d) tri-
fluoroacetic acid/triisopropylsilane/H2O (95:2.5:2.5).
peptides were cleaved from solid support and purified by C18
reversed-phase (RP) HPLC prior to biological testing.
the exception of Ala10, which retained the moderate activity
of the parent peptide against Listeria monocytogenes upon l-
Glu10 substitution, no analogues showed Gram-positive activi-
ty. Analysis of the activity of the alanine scan analogues
against the Gram-negative bacteria was more revealing. In-
spection of the sequence of octyl-tridecaptin A1 suggests it is
a membrane-lysing peptide, given the combination of cationic
Dab residues and hydrophobic amino acids, as well as the acy-
Octyl-tridecaptin A1 displays strong activity against Gram-
negative bacteria but has weak activity against Gram-positive
bacteria. To probe the effect of amino acid substitutions, the
minimum inhibitory concentration (MIC) of all analogues
against six Gram-negative bacteria and four Gram-positive bac-
teria was determined by broth dilution assays (Table 1). With
Table 1. Minimum inhibitory concentrations (MICs [mgmLÀ1]) of alanine scan analogues against Gram-negative and Gram-positive bacteria.
Peptide E. coli
ATCC
Salmonella P. aeruginosa C. jejuni K. pneumoniae Acinetobacter Enterococcus Staphylococcus L. monocytogenes Enterococcus
enterica
ATCC 27853 NCTC
11168
ATCC 13883
baumannii
faecalis ATCC aureus ATCC
ATCC 15313
faecium
25922
ATCC 13311
ATCC 19606
29212
29213
ATCC 19434
Oct-
TriA1
Ala1
Ala2
Ala3
Ala4
Ala5
Ala6
Ala7
Ala8
Ala9
Ala10
Ala11
Ala12
Ent-
3.13
6.25
25
1.56
3.13
12.5
100
100
25
50
6.25
12.5
3.13
6.25
6.25
12.5
12.5
6.25
100
25
>100
100
25
100
1.56
12.5
6.25
6.25
3.13
6.25
6.25
6.25
50
25
>100
50
12.5
100
12.5
25
>100
100
25
50
>100
12.5
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
100
>100
>100
>100
>100
>100
>100
>100
>100
>100
25
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
100
25
3.13
12.5
>100
12.5
6.25
6.25
50
12.5
50
6.25
12.5
>100
6.25
25
100
6.25
12.5
>100
12.5
6.25
6.25
25
>100
>100
50
>100
12.5
6.25
12.5
25
6.25
3.13
3.13
12.5
12.5
12.5
25
50
25
>100
>100
100
12.5
12.5
Oct-
TriA1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 1295 – 1299 1296

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
lated N terminus. It is therefore not surprising that most of the
conservative substitutions of short side-chain amino acids, in-
cluding d-Val1, Gly3, d-Ser4, l-Ser6, and l-Val11 had little effect
on the activity of the peptide. Substitution of Glu10 with ala-
nine is a fairly dramatic substitution; however, it had little
effect on activity. Interestingly, replacing d-allo-Ile12 with d-Ala
decreased the activity eightfold or greater against most of the
Gram-negative organisms, correlating to the high hydropho-
bicity of this residue, which we hypothesized was key to dis-
ruption of the membrane. Similarly, a large decrease was ob-
served when the aromatic amino acids were replaced, with
a two- to fourfold decrease in activity upon l-Phe9Ala substitu-
tion and an eight- to 16-fold decrease in activity upon d-Trp5
substitution. However, although we expected that substitution
of any of the Dab residues would have a significant effect on
activity, replacement of d-Dab2 or l-Dab7 resulted in only
a fourfold decrease against most organisms. This could be ad-
vantageous for the development of tridecaptin A1 analogues
with lower toxicity, as removal of positive charges in the analo-
gous polymyxins has been shown to decrease the nephrotoxic-
ity of these cyclic cationic lipopeptides,[6] but it is hard to ex-
plain in the context of a nonspecific membrane-lytic interac-
tion. In contrast, substituting d-Dab8 with d-Ala abolished the
activity of octyl-tridecaptin A1 to the limit of concentrations
tested (100 mgmLÀ1), thus suggesting that this residue plays a
key role beyond that of a localized positive charge. We there-
fore concluded that Dab8 is vital for the activity of octyl-tri-
decaptin A1, whilst d-allo-Ile12 and d-Trp5 also appear to be
important, and that a defined sequence–activity relationship
exists.
we would expect it to form an ordered, likely amphiphilic,
structure in hydrophobic environments. Trifluoroethanol is
often used as a 50% solution in water to mimic such environ-
ments, but in our hands, it had no effect on the CD spectrum
of tridecaptin A1. Attempting to more closely mimic the in vivo
environment, we therefore decided to measure the CD spec-
trum of octyl-tridecaptin A1 in the presence of negatively
charged phospholipid LUVs.
A 3:1 mixture of 1-palmitoyl-2-oleoylphosphatidylcholine
(POPC) and 1-palmitoyl-2-oleoylphosphatidylglycerol (POPG)
was used to synthesize 50 nm LUVs. The net negatively
charged phospholipid bilayer is an effective mimic of the bac-
terial membrane.[7] Adding ten lipid-equivalents (based on
phosphate content) of 50 nm phospholipid LUVs induces a sec-
ondary structure in tridecaptin A1 with a strong negative mini-
mum at 196 nm and moderately positive maximum at 224 nm
in the CD spectrum. This type of curve has been attributed to
the poly-l-proline II helix adopted by positively charged poly-
proline chains and the extended helix structure observed in
collagen.[8] Tridecaptin A1 is polycationic and appears to only
adopt a stable secondary structure when it comes into contact
with the negatively charged membrane. Given the mixture of
l- and d-amino acids, it seems likely that tridecaptin A1 is
unable to adopt classical secondary structures like the a-helix
and b-sheet.
To assess the relevance of the membrane-induced structure
to the antimicrobial activity of octyl-tridecaptin A1, the CD
spectra of the alanine scan analogues were recorded. All ana-
logues were completely disordered in aqueous buffer and TFE,
but many retained the structure of the natural peptide in the
presence of LUVs (Figure 2). In most cases, the magnitude of
CD spectroscopy
To probe the decrease in activity upon substitution of certain
amino acids, CD spectroscopy was used to analyze the struc-
ture of the alanine analogues. As expected from the short se-
quence and mixture of l- and d-amino acids, the natural pep-
tide is unstructured in aqueous buffer (Figure 1). If tridecaptin
A1 exerts its bactericidal effect through membrane interactions,
Figure 2. CD spectra of alanine scan analogues in the presence of phospho-
lipid LUVs. l: wavelength (nm), [q]: molar ellipticity (degcm2 molÀ1).
the negative band at 194 nm can be directly correlated to the
activity of the peptide. Ala10 had the largest negative band,
which could be due to the overall increase in positive charge
when l-Glu10 is replaced with alanine. This would provide
a stronger electrostatic interaction between the peptide and
the negatively charged LUVs and bacterial membrane, propor-
Figure 1. CD spectra of tridecaptin A1 with A) HEPES buffer; B) 50% TFE; and
C) phospholipid LUVs. l: wavelength (nm), [q]: molar ellipticity
(degcm2 molÀ1).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 1295 – 1299 1297

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
tionate with the sustained antimicrobial activity. Ala1, Ala4,
and Ala6, in which d-Val1, d-Ser4, and l-Ser6 have been re-
placed, have a similar spectrum to the natural peptide and
also have similar antimicrobial activity. The next group of ana-
logues, Ala3, Ala11, and Ala7, have similar spectra with a greatly
reduced negative band, which correlates to a roughly two- to
fourfold reduction in activity against sensitive Gram-negative
organisms. The CD spectra of the remaining analogues were
more revealing. Substituting d-Dab8 with d-Ala abolished both
the antimicrobial activity and the secondary structure, whereas
d-Trp5-d-Ala and d-allo-Ile12-d-Ala substitutions both led to
an average eightfold decrease in antimicrobial activity. The re-
maining analogues, Ala2 and Ala9, also showed a substantial
decrease in activity against some Gram-negative bacteria. The
overall trend in these results is that the magnitude of the neg-
ative band at 194 nm in the CD spectra of the tridecaptin A1
analogues correlates directly to their antimicrobial activity. Res-
idues d-Trp5, d-Dab8, and d-allo-Ile12 are required for effective
folding of octyl-tridecaptin A1 in a phospholipid bilayer, likely
directly affecting the ability of the peptides to kill Gram-nega-
tive bacteria.
proportional. The least active analogues did not adopt a secon-
dary structure in the presence of the phospholipid LUVs.
This study shows that a combination of alanine scanning
and CD spectroscopy are an effective tool for analyzing the
structures of peptides as they pertain to antimicrobial activity.
The results suggest tridecaptin A1 exerts its bactericidal effect
through an interaction with or in the bacterial membrane, pos-
sibly involving a chiral receptor. Further studies are underway
to probe the nature of this interaction.
Experimental Section
General information: NMR spectra were recorded on a Varian
1
Inova 500 or 600 spectrometer. For H NMR spectra, d values were
referenced to D2O (4.79 ppm) or [D6]DMSO (2.50 ppm). HRMS spec-
tra were recorded on an Agilent Technologies 6220 oaTOF by elec-
trospray ionization (ESI). Matrix-assisted laser desorption ionization
(MALDI) MS/MS analysis was performed on a Bruker Ultraflextreme
MALDI TOF/TOF. 4-Hydroxy-a-cyanocinnamic acid was used as
a matrix. Peptides were purified on a a Gilson Preparative HPLC
system equipped with a model 322 HPLC pump, a GX-271 liquid
handler, a 156 UV/Vis detector, and a 10 mL sample loop. A Phe-
nomenex C18 column (5 mm, 21.2ꢁ250 mm) was used for prepara-
tive scale purification. HPLC solvents were filtered with a Millipore
filtration system under vacuum before use.
To account for the selectivity of octyl-tridecaptin A1 for bac-
teria over human cells, we also measured the CD spectrum in
the presence of neutral LUVs, which mimic the mammalian
membrane.[9,10] The peptide adopted the same secondary
structure (see the Supporting Information), which suggests the
bactericidal effect might be due to more than membrane dis-
ruption. To further corroborate this theory, the enantiomer of
octyl-tridecaptin A1, Ent-Oct-TriA1, was synthesized and tested
against the organisms in Table 1. Although activity against the
Gram-positive organisms was mostly unaffected, there was a
marked decrease in activity against the Gram-negative organ-
isms; this suggests there could also be a chiral receptor in-
volved in the mode of action. Studies are currently underway
to undercover the full mode of action of tridecaptin A1.
Peptide synthesis: Solid-phase peptide synthesis was performed
on a 100 or 50 mmol scale by using Fmoc chemistry on preloaded
Fmoc-Ala-Wang resin (100–200 mesh, Novabiochem, #856001).
Reactions were performed in 12 mL polypropylene solid-phase
extraction tubes, fitted with a 20 mm polyethylene frit (Aldrich,
#57176). The resin was gently stirred with a magnetic stirrer. The
resin was preswollen by addition of DMF (5 mL, 10 min) while stir-
ring. Between deprotections and couplings, the vessel was drained
and washed with DMF (3ꢁ5 mL). The Fmoc group was removed
by stirring with 20% piperidine in DMF (2ꢁ5 mLꢁ2 min; 1ꢁ5 mLꢁ
10 min). The deprotection steps were monitored by spotting each
filtrate on a silica gel TLC plate and visualizing the benzofulvene
piperidine adduct with a 254 nm UV lamp. The appropriate amino
acid (5 equiv) was preactivated by shaking with HATU (5 equiv)
and DIPEA (10 equiv) in DMF (5 mL) for 5 min. The resin was stirred
in the coupling solution for 1 h, drained, and washed with DMF
(3ꢁ5 mL). The deprotection and coupling steps were continued to
complete the peptide synthesis. The resin-bound peptide was
washed with CH2Cl2 (3ꢁ5 mL) and dried under argon for
20 min.The resin was transferred to a screw-top vial containing
TFA/TIPS/H2O (95:2.5:2.5, 5 mL) and was shaken for 2 h. The cleav-
age mixture was filtered and concentrated in vacuo, and the pep-
tide was precipitated with ether. The crude peptide was dissolved
in H2O/MeCN (1:1, 5 mL) and purified by preparative-scale C18-RP-
HPLC: Phenomenex C18 column, flow rate 10 mLminÀ1, detected at
220 nm. Gradient: starting from 20% MeCN (0.1% TFA) and 80%
water (0.1% TFA) for 5 min, increasing to 55% MeCN over 30 min,
then increasing to 95% MeCN over 3 min, remaining at 95% MeCN
for 3 min, decreasing to 20% MeCN over 2 min, then remaining at
20% MeCN for 5 min. Pure product-containing fractions were
pooled, concentrated, frozen, and lyophilized to yield the product
as a white powder. Peptide sequences and structure were con-
firmed by HRMS, NMR, and/or MS/MS analysis.
Conclusions
An alanine scan of the lipopeptide octyl-tridecaptin A1 showed
that the residues d-Trp5, d-Dab8, and d-allo-Ile12 are necessary
for activity against Gram-negative bacteria, whereas substitu-
tion of all other amino acids was well tolerated. Of particular
note is d-Dab8, which was essential for antimicrobial activity.
The short linear sequence of tridecaptin A1 makes it well suited
for chemical synthesis, and the information derived from this
alanine scan could guide the synthesis of more active and/or
less toxic analogues.
Tridecaptin A1 and its octyl analogue are disordered in aque-
ous media and 50% trifluoroethanol, with no discernable sec-
ondary structure. However, the peptides adopt a definite sec-
ondary structure in the presence of 50 nm phospholipid LUVs.
Although trifluoroethanol is often the solvent of choice for in-
ducing structures in membrane-associated peptides and pro-
teins, our results suggest that this is not always an effective
model for membrane-interacting peptides. The activity of the
alanine scan analogues and their ability to adopt a secondary
structure in the model membrane environment are directly
Peptide characterization: See the Supporting Information.
Bacterial growth conditions: All cultures were grown from glycer-
ol stocks. Organisms (excluding C. jejuni) were grown in Mueller
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 1295 – 1299 1298

CHEMBIOCHEM
FULL PAPERS
www.chembiochem.org
Hinton (MH) broth at 378C and 225 rpm. C. jejuni was grown on
MH agar plates at 378C under 10% CO2, 5% O2, and 85% N2.
are grateful for the assistance of Mark Miskolzie with NMR
experiments and Christine Syzmanski and Bernadette Beadle for
assistance with Campylobacter testing. This work was supported
by the Natural Sciences and Engineering Research Council of
Canada (NSERC), Griffith Laboratories Canada, the Alberta Glyco-
mics Centre, and the Canada Research Chair in Bioorganic and
Medicinal Chemistry (J.C.V.).
Broth dilution assay: MIC values were determined according to
CLSI guidelines.[11] Polymyxin B was used as a positive control, with
MIC values against CLSI QC strains (E. coli ATCC 25922, Pseudomo-
nas aeruginosa ATCC 27853) within the range of published values.
LUV preparation: POPC and POPG were purchased as 25 mgmLÀ1
solutions from Avanti Polar Lipids. The mini-extruder and mem-
branes were also purchased from Avanti Polar Lipids. POPC
(228 mL, 7.5 mmol) and POPG (77 mL, 2.5 mmol) were mixed and
dried under an argon stream, followed by under high vacuum for
2 h. The film was hydrated with phosphate buffer (1 mL, 5 mm
KH2PO4, pH 7.0) and vortexed for 30 s. The mixture was frozen in
a dry ice bath, thawed with warm water, and vortexed for 30 s. The
freeze–thaw–vortex cycle was repeated twice. Passing the milky
suspension through a 50 nm polycarbonate membrane ten times
by using a mini-extruder formed 50 nm LUVs. The phosphate con-
centration of the LUV solution was determined by using the Stew-
art method.[12]
CD spectroscopy: CD spectra were recorded on an OLIS DSM 17
CD spectrometer. Samples were added to a 0.4 mL quartz cuvette
with a 0.1 cm path length, measured from 190–250 nm at 208C
and averaged over five scans. The peptide concentration remained
constant at 20 mm for all experiments. The CD spectrum of phos-
phate buffer was subtracted from all spectra. For the vesicle experi-
ments, ten lipid equivalents (by phosphate concentration) were
mixed thoroughly with the peptide solution before measuring. A
digital filter of 15 was applied to CD spectra, which were converted
to molar ellipticity units.
Keywords: alanine scan
·
antimicrobial peptides
·
lipopeptides · structure–activity relationships · tridecaptins
[1] T. Roemer, C. Boone, Nat. Chem. Biol. 2013, 9, 222–231.
[2] G. Pirri, A. Giuliani, S. F. Nicoletto, L. Pizzuto, A. C. Rinaldi, Cent. Eur. J.
Biol. 2009, 4, 258–273.
[3] T. Kato, H. Hinoo, J. Shoji, J. Antibiot. 1978, 31, 652–661.
[4] C. T. Lohans, M. J. van Belkum, S. A. Cochrane, Z. Huang, C. S. Sit, L. M.
McMullen, J. C. Vederas, ChemBioChem 2014, 15, 243–249.
[5] S. A. Cochrane, C. T. Lohans, J. R. Brandelli, G. Mulvey, G. D. Armstrong,
J. C. Vederas, J. Med. Chem. 2014, 57, 1127–1131.
[6] M. Vaara, J. Fox, G. Loidl, O. Siikanen, J. Apajalahti, F. Hansen, N. Fri-
modt-Møller, J. Nagai, M. Takano, T. Vaara, Antimicrob. Agents Chemother.
2008, 52, 3229–3236.
[7] S. Thennarasu, D. K. Lee, A. Tan, U. Prasad Kari, A. Ramamoorthy, Bio-
chim. Biophys. Acta Biomembr. 2005, 1711, 49–58.
[8] N. J. Greenfield, Nat. Protoc. 2007, 1, 2876–2890.
[9] Y. Jin, H. Mozsolits, J. Hammer, E. Zmuda, F. Zhu, Y. Zhang, M. I. Aguilar,
J. Blazyk, Biochemistry 2003, 42, 9395–9405.
[10] J. Blazyk, R. Wiegand, J. Klein, J. Hammer, R. M. Epand, R. F. Epand, W. L.
Maloy, U. P. Kari, J. Biol. Chem. 2001, 276, 27899–27906.
[11] J. C. Stewart, Anal. Biochem. 1980, 104, 10–14.
[12] CLSI M7 A9 Ed. 9, Methods for Dilution Antimicrobial Susceptibility
Tests for Bacteria that Grow Aerobically, 2011.
Acknowledgements
We would like to thank Dr. Randy Whittal, Jing Zheng, and Bela
Reiz for their assistance with ESI-MS and MS/MS experiments. We
Received: February 10, 2014
Published online on May 9, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2014, 15, 1295 – 1299 1299