Reverse engineering truncations of an antimicrobial peptide dimer to identify the origins of potency and broad spectrum of action

Article abstract of DOI:10.1021/jm100483y

Antimicrobial peptides hold promise against antibiotic resistant pathogens. Here, to find the physicochemical origins of potency and broad spectrum antimicrobial action, we report the structure-activity relationships of synthetic intermediates (peptides A-D) of a potent lysine branched dimeric antibacterial peptide ΔFd. Our studies show that a tetracationic character in a weak helical fold (peptide C) elicits potent but narrow spectrum antimicrobial activity [Minimum inhibitory concentrations (MICs) E. coli 10 μM, S. aureus > 100 μM]. In contrast, a hexacationic character in a strong, amphipathic helix (ΔFd) confers potent and broad spectrum action [MICs E. coli 2.5 μM, S. aureus 5 μM]. While ΔFd caused rapid and potent permeabilization of the E. coli membranes, the less helical intermediates (peptides A-D) showed slow and weak to no responses. Two seminal findings that may aid future drug design are (a) at identical helicity, increasing charge enhanced outer membrane permeabilization, and (b) at identical charge, increasing helicity stimulated rate of outer membrane permeabilization and kill kinetics besides enhancing potency leading to broad spectrum action.

Full text of DOI:10.1021/jm100483y

J. Med. Chem. 2010, 53, 6079–6088 6079
DOI: 10.1021/jm100483y
Reverse Engineering Truncations of an Antimicrobial Peptide Dimer to Identify the Origins of Potency
and Broad Spectrum of Action
Aparna Anantharaman and Dinkar Sahal*
Malaria Research Laboratory, International Centre for Genetic Engineering and Biotechnology, Aruna Asaf Ali Road, New Delhi 110067, India
Received April 20, 2010
Antimicrobial peptides hold promise against antibiotic resistant pathogens. Here, to find the physico-
chemical origins of potency and broad spectrum antimicrobial action, we report the structure-activity
relationships of synthetic intermediates (peptides A-D) of a potent lysine branched dimeric anti-
bacterial peptide ΔFd. Our studies show that a tetracationic character in a weak helical fold (peptide C)
elicits potent but narrow spectrum antimicrobial activity [Minimum inhibitory concentrations (MICs)
E. coli 10 μM, S. aureus > 100 μM]. In contrast, a hexacationic character in a strong, amphipathic helix
(ΔFd) confers potent and broad spectrum action [MICs E. coli 2.5 μM, S. aureus 5 μM]. While ΔFd
caused rapid and potent permeabilization of the E. coli membranes, the less helical intermediates
(peptides A-D) showed slow and weak to no responses. Two seminal findings that may aid future drug
design are (a) at identical helicity, increasing charge enhanced outer membrane permeabilization, and
(b) at identical charge, increasing helicity stimulated rate of outer membrane permeabilization and kill
kinetics besides enhancing potency leading to broad spectrum action.
Introduction
better antibiotics. Such information is necessary to generate
antibiotic peptides that harbor the minimum functional core
with no extraneous features which enhance the cost of produc-
tion without a concurrent enhancement of their antibiotic
profile.
Due to widespread resistance to almost all conventional
antibiotics, there is an urgent need to discover or design novel
antibiotics.1 Naturally occurring antimicrobial2 and host
defense peptides are being developed as new anti-infective
therapeutic agents.3 Peptide antibiotics offer several advan-
tages over their nonpeptide counterparts. These include (a)
their lower tendency to induce resistance in bacteria (b)
specific action against bacterial membranes, (c) the ability to
target both the surface and the intracellular milieu of bacteria,
and (d) broad spectrum action against both Gram negative
and Gram positive bacteria. However, the utility of these
naturally occurring peptides has been curtailed by drawbacks
like high cost of production, poor potency, instability, and
poor selectivity.4 Therefore, to improve their performance,
efforts have been directed toward understanding the physico-
chemical code governing the activity of these peptides.5-7
A number of studies have elucidated the influence of
individual biochemical parameters like charge, hydrophobicity,
and conformation on the potency and selectivity of antimicro-
bial peptides.8-10 However, the composite physicochemical
code governing their antimicrobial properties like spectrum
of action, kill kinetics, and membrane permeabilizing abilities
has remained largely undefined. As action of antimicrobial
peptides (AMPsa) is a multipronged composite of several
parameters, elucidation of this code could help in designing
We have recently reported the seminal role of a lysine
branched dimeric motif in the de novo design of a potent,
broad spectrum antimicrobial peptide ΔFd.11 This branched
dimeric peptide has two cationic decapeptide amphipathic
helices (Ac-G-ΔF-R-K-ΔF-H-K-ΔF-W-A) linked to the R
and ε amino groups of a lysyl residue (Figure 1A). ΔFd is
characterized by the presence of R,β-didehydrophenylalanine
(ΔF), a nonproteinogenic amino acid which constrains the
peptides harboring it in a helical conformation12 and renders
them relatively more resistant to proteases than their phenyl-
alanine-containing counterpart. The chemical structures of
phenylalanine (F) and its R,β-didehydrophenylalanine analo-
gue (ΔF) are shown in parts B and C of Figure 1, respectively.
ΔFd shows potent, broad-spectrum activity against both
Gram negative and Gram positive bacteria and exhibits
strong selectivity for bacterial cells.11
To (a) identify the minimum functional module of ΔFd and
(b) dissect the origins of strong potency and broadspectrum of
action in ΔFd, we have examined the conformational and
antibiotic properties of four synthetic intermediates of ΔFd.
These intermediates have allowed us to unravel the effects of
charge and helicity on (a) antibiotic potency, (b) spectrum of
action, (c) kill kinetics, and (d) membrane permeabilization.
Even though it is well-known that the antimicrobial action of
a peptide is governed by several factors including charge,
helicity, and amphipathicity,6,13 the hierarchy and interplay
of these factors in the diverse actions of an antimicrobial pep-
tide are not known. Approaches that rely upon site specific sub-
stitutions are sequence of a peptide, to understand structure
*To whom correspondence should be addressed. Phone: 91-11-2674
1358. Fax: 91-11-26162316. E-mail: dinkar@icgeb.res.in.
a Abbreviations: AMPs, antimicrobial peptides; CD, circular dichro-
ism; CFU, colony forming units; ΔF, R,β-didehydrophenylalanine;
MRE, mean residue ellipticity; MBC, minimum bactericidal concentra-
tion; MIC, minimum inhibitory concentration; NPN, 1-N-phenyl-
naphthylamine; PI, propidium iodide; SDS, sodium dodecyl sulfate.
r
2010 American Chemical Society
Published on Web 08/03/2010
pubs.acs.org/jmc

6080 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16
Anantharaman and Sahal
Figure 1. (A) Chemical structure of lysine branched dimer ΔFd. Letters denote single letter codes for amino acid residues. It may be noted that
the two amino groups (marked as R and ε) of the boxed central lysine make amide bonds with the carboxy terminal alanyl residue of each
peptide chain. Solid lines demarcate the truncation boundaries of the dimers A through D. Acetyl-glycyl cap is common to all peptides shown.
Chemical structures are shown for (B) phenylalanine and (C) R,β-didehydrophenylalanine. All chemical structures have been made using
ChemDraw Ultra 11.0.
activity relationships,14-17 often do not allow firm conclu-
sions, as they affect more than one property of a peptide by a
single change. Here, we have used a reverse engineering
approach to identify how features of antimicrobial activity
such as potency, spectrum of action, kill kinetics, and per-
meabilization of the bacterial membrane evolve in a potent
antimicrobial peptide with progressive changes in its physico-
chemical characteristics like charge and helicity.
The resulting peptides differed in physicochemical proper-
ties such as length, charge, and the number of R,β-didehydro-
phenylalanine (ΔF) residues (Table 1). Peptides A and B
differed in length but had the same charge (þ2) and number
of ΔF residues (2 ΔFs). Peptides C and D with 4 ΔFs each
had a charge of þ4 and þ6, respectively. The full-length
dimer ΔFd had 6 ΔFs and a charge of þ6. The RPHPLC
retention times of peptides A-D and ΔFd were: 34, 45,
45, 39, and 39 min respectively (Supporting Information
Figure 1).
Results
Circular Dichroism Based Structures of Dimers of Increas-
ing Lengths. Circular dichroism (CD) spectroscopy was
used to determine the secondary structure of the peptides
in the membrane mimetic solution of sodium dodecyl sulfate
(SDS). The CD spectra of dimers A (11 mer) through ΔFd
(21 mer) (Figure 2A) indicate a progressive increase in
nucleation of stable secondary structures.
Dimer A (11 mer, 1 ΔF/chain), with very weak ellipticity at
∼280 nm, appears to be nearly random. Dimer B (13 mer,
1 ΔF/chain,) with significantly enhanced ellipticity of the
280 nm band, appears to be acquiring a more stable β-turn
structure.12 This is likely to be an effect of the increase in
Peptide Design. To determine the minimum peptide length
required for the antibiotic activity of ΔFd, we generated
its sequentially N-terminally truncated analogues viz peptide
A (5 residues/chain), peptide B (6 residues/chain), peptide
C (8 residues/chain), and peptide D (9 residues/chain)
(Table 1). As N-capping motifs and the nature of N terminal
residue have been shown to affect peptide structure and
activity,18,19 all the peptides listed above were uniformly
glycine capped and acetylated at the N terminus. The
chemical structure of ΔFd with the truncation boundaries
of its shortened dimer analogues has been shown in
Figure 1A.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16 6081
Table 1. Physicochemical Properties of the Peptides Used in This Study
a The progressive increments in chain length are indicated by the residues in boldface. The N terminus acetyl glycyl cap is common to all peptides.
Table 2. Antibacterial Activities of the Peptides Studieda
E. coli ML35p
S. aureus ATCC 700699
MIC
(μM)
MBC MBC/
MIC
MIC
(μM)
MBC
(μM)
MBC/
MIC
peptide
(μM)
peptide A >100
peptide B >100
>100
>100
>100
>100
5
peptide C
peptide D
ΔFd
10
5
15
10
5
1.5
2
2.5
2
7.5
1.5
a Standard error measurement (SEM) values of OD600 for maximum
growth and maximum CFU/mL were e1% of mean and 0.6% of mean,
respectively. All SEM values for end point OD600 (MIC) and CFU/mL
(MBC) of two independent experiments (each done in triplicates) have
been shown in Experimental Section.
It may be noted that the excitonic couplet seen with dimers C
and D is unequal, with 1.5 times greater negative intensity
centered at 298 nm than the positive intensity centered at
267 nm. In moving from dimers C/D (2 ΔFs/chain) to ΔFd
(3 ΔFs/chain), we observe (a) the 267-298 nm couplet
becomes significantly stronger in intensity and (b) the posi-
tive ellipticity (267 nm) becomes a little stronger than the
negative ellipticity (298 nm). It was observed that without
any change in the status of charge, the introduction of the
third ΔF residue per chain in ΔFd resulted in a 3-fold
increment in molar ellipticity of the helix associated excitonic
couplet (Figure 2A). Thus, it is clear that in the present series,
the dimer ΔFd makes the strongest helix (Figure 2B).
Figure 2. (A) Circular dichroism based conformational analysis of
dimeric peptides A-D and ΔFd (6 μM each) in 20 mM SDS/10 mM
sodium phosphate buffer pH 7.5. (B) Histogram showing mean
residue ellipticity (MRE) values for the dimeric peptides A-D and
ΔFd (for 267 nm band).
Antibacterial Activities of Dimers of Increasing Lengths.
The antibacterial activity of ΔFd and its truncated analogues
(peptides A-D), was tested against Escherichia coli ML35p
and Staphylococcus aureus ATCC 700699 (Table 2).
The antibacterial effect of the peptides against E. coli
ML35p was found to increase in potency with chain length
(peptides A-ΔFd) and the aforementioned features asso-
ciated with increasing chain length. The poorly structured
shorter dimers A and B (charge: þ2, 2 ΔFs) did not inhibit
the growth of either the Gram negative E. coli or the Gram
positive S. aureus [minimum inhibitory concentration (MIC)
> 100 μM]. Peptide C (charge: þ4 and 4 ΔFs, weak 310 helical
structure) was quite potent and bactericidal against E. coli
[MIC: 10 μM; minimum bactericidal concentration (MBC):
15 μM] but failed to act against S. aureus (MIC >100 μM).
chain length as a consequence of introducing one histidyl
residue per chain in dimer B. Interestingly, dimers C (17 mer)
and D (19 mer), each with two ΔF residues per chain, showed
an excitonic couplet at 267 nm (þ) and 298 nm (-). This
couplet is a signature of the formation of right handed 310
helices in ΔF containing peptides.20 The nearly overlapping
spectra of dimers C (8 mer/chain) and D (9 mer/chain)
suggest that the introduction of an additional ΔF residue
to each chain of dimer B (1 ΔF/6 mer chain) caused the
transition from β-turn (dimer B) to 310 helices (dimers C and D).

6082 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16
Anantharaman and Sahal
Figure 3. Plot showing inverse relationship between mean residue
ellipticity (MRE) (for 267 nm band) and MIC for peptides A-D
and ΔFd tested against E. coli ML35p and S. aureus ATCC 700699.
The number of positive charges and the number of ΔF residues in
each peptide are indicated by superscript and subscript, respec-
tively. *: MICs shown as 100 μM are in reality >100 μM.
Like peptide C, peptide D also had 4 ΔF residues but differed
inhaving a higher charge (þ6). Even thoughthe CDspectra of
peptides C and D (Figure 2A) were nearly overlapping in all
respects, peptide D showed a significantly improved antibac-
terial activity against E. coli (MIC: 5 μM; MBC: 10 μM) over
peptide C (Table 2). Thus, it can be concluded that the
improved potency of dimer D may be the result of its two
additional positive charges. However, both dimers C and D
failed to act against S. aureus up to a concentration of 100 μM
and hence were not broad spectrum in antibiotic action. In
sharp contrast, the introduction of the third ΔF residue to
each chain resulting in ΔFd was associated with dramatic
improvements in antibacterial potency and spectrum of ac-
tion. ThestronghelicalnatureofΔFd (Figure 2A) lednot only
to further potentiation of action against Gram negative E. coli
(MIC: 2.5 μM; MBC: 5 μM) but also a broadened spectrum of
action with potent bactericidal action against the Gram
positive S. aureus (MIC: 5 μM, MBC: 7.5 μM). The results
suggest that antibacterial potency and spectrum of action are
a function of increasing peptide length, charge and helicity.
The inverse correlation between helicity and MIC against
E. coli and S. aureus is shown in Figure 3.
Figure 4. (A) Kill kinetics of ΔFd (10 μM), peptide C (40 μM), and
peptide D (20 μM) (at 4ꢀ MIC each) against E. coli ML35p. (B) Plot
showing inverse relationship between mean residue ellipticity
(MRE) (for 267 nm band) and rate of cell death (kill kinetics)
caused by peptides C, D, and ΔFd against E. coli ML35p measured
as colony forming units (CFU)/mL. The number of positive charges
and the number of ΔF residues in each peptide are indicated by
superscript and subscript, respectively.
same charge but higher helicity than peptide D. These results
suggest that even though both charge and helicity are
determinants of antibacterial potency, helicity seems to play
a more dominant role than charge in imparting a higher rate
of bacterial killing to ΔFd.
In summary, two trends were apparent from the above
results: (a) as seen in case of peptide C, þ4 charges, 4 ΔFs and
weak 310 helical structure represent the minimum require-
ments for antibacterial activity, and (b) a broad spectrum of
action is conferred by the presence of þ6 charges, 6 ΔFs and
a strong helical fold (as seen in case of ΔFd).
Outer Membrane Permeabilization by Dimers C, D, and
ΔFd. Outer membrane permeabilization refers to disruption
of membrane integrity which facilitates the uptake of exo-
genous molecules.21 This is an important step in the mode of
action of antibacterial peptides, as these have been shown to
gain entry into the bacteria via a “self-promoted uptake”
mechanism involving displacement of native Ca2þ and Mg2þ
ions from the lipopolysaccharide (LPS) layer of the bacterial
membrane.22 To determine the physicochemical factors in-
fluencing membrane permeabilization, we studied the outer
and inner membrane permeabilizing abilities of the three
dimers (C, D, and ΔFd). We used 1-N-phenyl-naphthylamine
(NPN) as a fluorescent probe (NPN fluoresces strongly only
in a hydrophobic environment like the interior of a mem-
brane)22 to identify the kinetic traits of outer membrane
permeabilization associated with the three dimers. These
dimers, examined at 10 μM each, showed a characteristic
time dependent rise in fluorescence intensity due to increased
membrane permeabilization and, consequently, uptake of
NPN fluorescence (Figure 5A).
Comparative Kill Kinetics of Dimers C, D, and ΔFd. In
addition to its potency, the suitability of a peptide as an
antimicrobial agent is assessed also by the kinetics or the rate
at which it effects its killing action. At 4 times the respective
MIC, each dimer showed characteristic distinctions in the
rate of E. coli killing. At 5 min, while peptides C and D
showed a 2-fold decrease in colony forming units (CFU)/mL
(from 2 ꢀ 105 to 1 ꢀ 105 CFU/mL) (50% E. coli cells killed),
ΔFd showed a 25 fold decrease in CFU/mL (from 2 ꢀ 105 to
8 ꢀ 103 CFU/mL) (96% E. coli cells killed) (Figure 4A). An
approximate estimation suggests that 95% killing was
achieved by the three dimers in 40 min (peptide C), 20 min
(peptide D), and 5 min (ΔFd). Thus, among the three dimers,
the strongest helix ΔFd displayed the fastest kill kinetics,
which is suggestive of a good correlation between peptide
helicity and kill kinetics (Figure 4B). Although different in
charge, peptides C (þ4) and D (þ6) exhibited comparable
molar ellipticities (Figure 2A) and killed bacteria at compar-
able rates (Figure 4A). However, kill kinetics increased
significantly in case of ΔFd (charge: þ6), which had the
The rise in fluorescence signal was gradual and shallow
(peptide C), more rapid and high intensity (peptide D), and
most rapid with high intensity (ΔFd). Viewed in context of
the charge and helicities of these peptides, two clear trends in
the degree and rate of permeabilization are evident. First,
among the three peptides, peptide C (charge: þ4) showed the

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16 6083
shorter dimers C and D failed to permeabilize the inner
membrane. Consistent with these observations, the fluores-
cence microscopy based monitoring of Syto 9 (freely perme-
able to both wild type and membrane compromised cells)
and PI (permeable only into membrane compromised cells)
also indicated that only the ΔFd treated E. coli were PI
permeable (Figure 5C). It was interesting to observe that in
contrast to ΔFd and peptide D, peptide C gave rise to a thin
filamentous phenotype (Figure 5C) suggestive of inhibition
of cell division. Further studies are warranted to understand
the underlying mechanism of filamentation induced by this
peptide in E. coli.
Discussion
The de novo designed, lysine branched dimeric peptide ΔFd
is a potent broad spectrum antimicrobial {MIC (E. coli):
2.5 μM; MIC (S. aureus): 5 μM} without undue hemolytic
activity.11 To (a) define the minimum functionalantimicrobial
peptide module of ΔFd and (b) to find the origins of its
potency and broad spectrum of action, we designed a series of
N-terminally truncated analogues of ΔFd (peptides A-D)
(Table 1). These truncated shorter dimers differed in various
physicochemical factors like charge and helicity and thus
allowed us to investigate the roles of individual factors as well
as their interplay in regulating antibiotic activity. In the con-
text of antimicrobial peptides, such a reverse engineering
approach allows us to delineate the hierarchy and interplay
ofthe importantbiological parameterswithout beingimpeded
by confounding variables arising as a consequence of amino
acid substitution strategies. The guiding principles of anti-
biotic peptide design that have emerged from the present
study are discussed below.
Antibiotic Potency Improves with Charge and Helicity. The
shorter peptide dimers A (11 mer) and B (13 mer), with 2 ΔFs
were neither rich in charge (þ2) nor in folding propensity
(Figure 2). The lack of antibiotic activity (MIC > 100 μM) in
these two dimers indicates the specific requirements of
charge and conformation for antibiotic activity. Peptide C
(17 mer, charge: þ4, 4 ΔFs) marked the beginning of the
formation of a folded structure (as evidenced by CD spec-
troscopy, Figure 2) and the simultaneous introduction of two
additional positive charges. Interestingly, this was the shortest
peptide in the present series to show antibiotic activity
against E. coli (MIC: 10 μM; MBC: 15 μM). However, it
was narrow spectrum in action and did not show activity
against S. aureus (MIC > 100 μM). This led us to infer that in
the present series, a charge of þ4 and the presence of a weak
helical fold represented the minimum threshold require-
ments for antibiotic activity against E. coli. The hexacationic
peptide D (19 mer, charge: þ6, 4 ΔFs) was structurally at par
with peptide C (Figure 2). However, peptide D [MIC (E. coli):
5 μM; MBC: 10 μM] was more potent than peptide C against
E. coli and lacked activity against S. aureus (MIC > 100 μM).
This suggested that in moving from peptide C to peptide D,
the 2-fold potentiation of MIC was due to the introduction of
two additional positive charges. ΔFd (21 mer, charge: þ6, 6
ΔFs) formed the strongest helix (Figure 2) and showed the
highest potency against E. coli (MIC: 2.5 μM, MBC: 5 μM).
Figure 6A is a three-dimensional graphical representation of
the influence of charge and helicity on antibiotic potency, as
defined by the MIC. The graph shows that in all the dimers,
from peptide C to ΔFd, antibiotic potency against E. coli
shows a strong correlation with both charge and helicity.
Previous studies have shown that an increment of charge8 or
Figure 5. (A) Outer membrane permeabilization of E. coli ML35p
by peptides (10 μM each) measured by increase in NPN fluorescence
for 10 min. (B) Inner membrane permeabilization of E. coli ML35p
by peptides (10 μM each) measured by increase in PI fluorescence
for 10 min. (C) Varied cell morphologies induced by the full length
dimer ΔFd and its shorter versions D and C. Fluorescence micro-
scopic images of E. coli ML35p treated with peptides at the
respective MICs of each: ΔFd (2.5 μM), peptide D (5 μM), peptide
C (10 μM) for 18 h, and stained by PI/Syto9. Control represents cells
not treated with peptide. The same section has been viewed under
Syto filter (top panel) and PI filter (bottom panel). Scale bars
(10 μm) have been indicated as white bars.
least degree and rate of outer membrane permeabilization.
Second, even though the degree of permeabilization was the
same for peptide D (charge: þ6) and ΔFd (charge: þ6), the
rate of permeabilization was higher for the more helical ΔFd.
These results suggest that while the degree of permeabiliza-
tion is influenced by charge, the rate of permeabilization
appears to be determined by the helicity of the peptide.
Inner Membrane Permeabilization by Dimers C, D, and
ΔFd. Kinetics of bacterial inner membrane permeabilization
by the dimers was studied by measuring the uptake of the
fluorescent probe propidium iodide (PI). This dye enters
only membrane compromised cells and fluoresces upon
binding to nucleic acids.23 Figure 5B shows that ΔFd, the
strongest helix and the most potent antibiotic in the present
series, showed strong ability to permeabilize the inner mem-
brane. However, unlike outer membrane permeabilization
(Figure 5A), the permeabilization of inner membrane by
ΔFd showed a lag period of ∼2.5 min. A closer look at the
trends of the outer versus the inner membrane permeabiliza-
tion induced by ΔFd suggests that the time (∼230 s) at which
the outer membrane permeabilization signal shows a small
fall coincides with the time (∼150-300 s) when inner per-
meabilization shows the first rise (Figure 5B). In contrast, the

6084 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16
Anantharaman and Sahal
Figure 6. Multiparameter physicochemical correlates of antibacterial activity against E. coli. Combined influence of charge and helicity
(MRE) on (A) antimicrobial potency, (B) kill kinetics (CFU/mL) (C) degree (Pmax), and (D) rate (time of Pmax in s) of outer membrane
permeabilization (OMP). MRE: mean residue ellipticity; MIC: minimum inhibitory concentration; Pmax: permeabilization maximum
(maximum permeabilization achieved in the duration of the experiment).
helicity16,24 in isolation improves antimicrobial potency. Our
results suggest that the interplay of both these physicochem-
ical factors in the same template may be necessary to elicit
potent antimicrobial activity.
are isomeric in sequence, charge, and hydrophobicity
(Supporting Information Table 1) but differ only in their
helicity.11 In contrast to ΔFd {MIC (E. coli): 2.5 μM; MIC
(S. aureus): 5 μM}, D-Lys-ΔFd showed a narrow spectrum
of antibiotic action {MIC (E. coli): 5 μM; MIC (S. aureus):
55 μM}.11 Thus, even though the helical ΔFd and the non-
helical D-Lys-ΔFd share the same charge and hydrophobicity,
the reduced helicity of D-Lys-ΔFd is solely responsible for
its narrow spectrum of antibiotic action. Therefore, in terms
of spectrum of antibiotic action, D-Lys-ΔFd (which is non-
helical like peptide D), does not benefit from its increased
hydrophobicity over peptide D. This suggests that the narrow
spectrum of action and kill kinetics of peptide D result
primarily from its reduced helicity and not hydrophobicity.
Although in the context of our present study, the influence
of hydrophobicity cannot be completely ruled out, our results
suggest that helicity may exert overriding effects on hydro-
phobicity in influencing spectrum of antibiotic action. The
inverse relationship between MIC and MRE (Figure 3),
observed in our current study, confirms the strong require-
ment of helicity for the design of broad spectrum antimicro-
bial peptides. It is well-known that Gram positive and Gram
negative bacterial membranes differ in composition due to
presence or absence of components like lipopolysaccharide (in
gram negative) or teichoic acids (in gram positive), which are
also believed to act as receptors for antibacterial peptides.26,27
Further studies are required to determine whether conforma-
tion of antimicrobial peptides influences their interaction with
these receptors and if it is this interaction that defines their
spectrum of action.
In the Background of Constant Charge, Increment in Heli-
city Broadens the Spectrum of Action and Enhances the Kill
Kinetics. While potency against E. coli was influenced by
both charge and helicity, spectrum of action was largely
governed by the helicity of the cationic peptide. As seen
previously in Figure 3, unlike the parent peptide ΔFd, none
of the shorter dimers was broad spectrum in action. It may be
noted that although both peptide D (weak helix) and the full
length ΔFd (strong helix) are hexacationic, peptide D targets
only E. coli (MIC (E. coli): 5 μM; MIC (S. aureus): >100 μM),
whereas ΔFd acts against both E. coli and S. aureus (MIC
(E. coli): 2.5 μM; MIC (S. aureus): 5 μM). This suggests that
even though charge is essential for antimicrobial activity, it
must be introduced in a conformational context for the
peptide to become broad spectrum in action.
It may however be noted that in addition to differences in
helicity, peptide D (4 ΔFs) and ΔFd (6 ΔFs) differ also
in hydrophobicity because ΔFd has two additional ΔF
residues over peptide D (Table 1). As the positive influence
of hydrophobicity on antimicrobial activity is well docu-
mented,5,25 it seems reasonable to suppose that the improved
antibacterial activity of ΔFd, in comparison to the less
hydrophobic peptide D, could be attributed, in part, to its
increased hydrophobic character. Previously, we have
dissected the role of helicity in ΔFd by constructing its
less helical counterpart, D-Lys-ΔFd.11 ΔFd and D-Lys-ΔFd

Article
In addition to potency of killing, another important para-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16 6085
helicity in enhancing the rate of membrane permeabilization.
Interestingly, this trend of the rate of membrane permeabi-
lization mirrors the rate of bacterial killing (Figure 6B). This
suggests that a correlation exists between the helical content
of an antimicrobial peptide, rate of membrane permeabiliza-
tion, and the rate of bacterial killing. Friedrich et al. (1999),
have shown that the highly helical CP29 exhibits better kill
kinetics and membrane permeabilizing activities as com-
pared its less helical analogue CP26 (both Cecropin-melittin
(CEME) hybrid peptide analogues).29 While individually,
charge30 and helicity31 have been shown to influence mem-
brane permeabilization, our findings specifically pinpoint
that charge may be the determinant of Pmax, while helicity
may determine the rate of outer membrane permeabilization.
Inner Membrane Permeabilization May Not Be Essential
for Antibacterial Activity. The requirements of inner mem-
brane permeabilization were found to be more stringent than
that of outer membrane permeabilization. This is substan-
tiated by the fact that only the full length ΔFd permeabilized
the inner membrane (Figure 5B). In contrast, peptides C and
D were unable to permeabilize the inner membrane even
after 18 h (Figure 5C). Antimicrobial peptides such as
Buforin II, which have the ability to inhibit bacterial growth,
also fail to permeabilize the bacterial inner membrane to
PI.24 Viewed in conjunction with the inability of the potent
dimers C and D to permeabilize the inner membrane, this
suggests that inner membrane permeabilization may not be a
stringent requirement for antimicrobial activity. As peptide
D (4 ΔFs, charge: þ6) is much less helical and hydrophobic
when compared to ΔFd (6 ΔFs, charge: þ6), the selective
acquisition of the ability to permeabilize the inner membrane
by ΔFd can be intuitively attributed to: (a) hydrophobicity or
(b) helicity. To find which of these parameters was the most
relevant, we examined the inner membrane permeabilization
properties of D-Lys-ΔFd. As shown in Supporting Informa-
tion Figure 2, while the degree of inner membrane permea-
bilization achieved by ΔFd and D-Lys-ΔFd was identical,
ΔFd (6 ΔFs) showed a distinct edge over the nonhelical
D-Lys-ΔFd (6 ΔFs) in the rate of permeabilization. The results
suggest that helicity exerts overriding effects on hydrophobi-
city to influence membrane permeabilization. Further, our
data (Supporting Information Figure 2) suggests that while
the extent of inner membrane permeabilization is influenced
by hydrophobicity, the rate of inner membrane permeabiliza-
tion is governed primarily by helicity.
meter of the performance of an antibiotic is the rate at which
it kills the target bacterium. A faster killing rate of an
antibiotic is expected to result in a faster clearance of
bacterial load in a patient. Thus, in peptides of comparable
antibiotic potencies, a faster acting antibiotic would be more
desirable. While the physicochemical code governing anti-
biotic potency has been investigated extensively, the factors
affecting the rate of bacterial killing have so far remained
largely undefined. In this study, we have found that helicity
has more prominent role than charge in imparting faster kill
kinetics to a peptide. This conclusion is based on the obser-
vations that (a) peptides C (charge: þ4) and D (charge: þ6)
with comparable helicities, albeit different charges, exhibit
nearly similar kill kinetics and (b) both peptide D (weakly
helical) and ΔFd (strongly helical) are hexacationic and yet
the kill kinetics of peptide D is an order of magnitude slower
than is the case with ΔFd (Figure 4). In addition, we have
previously shown that against E. coli, the nonhelical D-Lys-
ΔFd exhibits significantly poorer kill kinetics than ΔFd.11
Taken together, the results suggest that helicity has a stron-
ger influence on bacterial kill kinetics than charge and
hydrophobicity. Figure 6B is a graphical representation of
the combined influence of charge and helicity on kill kinetics.
The graph shows that the identically charged peptides D and
ΔFd (charge: þ6), owing to their differing helicities, exhibit
differing kill kinetics. Therefore, strong helicity in a peptide
confers not only potent, broad spectrum antimicrobial
activity but also makes it a rapidly killing antimicrobial agent.
Charge Determines Degree while Helicity Governs the Rate
of Outer Membrane Permeabilization. The permeabilization
of bacterial membranes by antibacterial peptides represents
the first line of attack on the bacterial intruders causing
disruption of transmembrane potentials crucial to survival of
bacteria.27 This membrane permeabilizing action of AMPs
enables them to sensitize bacteria to nonpeptide antibiotics
for which the LPS layer acts as a formidable barrier.28 Thus,
it is important to delineate the physicochemical parameters
of a peptide that facilitate membrane permeabilization. We
studied the degree and rate of outer membrane permeabili-
zation of E. coli by measuring the uptake of the fluorescent
dye NPN (Figure 5A). The results (Figure 6C) show the
influence of charge and helicity on the maximum permeabi-
lization (Pmax). The graph shows that among the three
peptides, peptide C (charge: þ4, moderate helicity) showed
slow, minimal permeabilization and peptides D (charge: þ6,
moderate helicity) and ΔFd (charge: þ6, maximal helicity)
both permeabilized the E. coli outer membrane maximally.
The different helicities of peptides D and ΔFd did not influ-
ence the Pmax achieved by these peptides (Figure 6C). There-
fore, in contrast to the pivotal role of helicity in potency of an
antimicrobial peptide (Figure 6A), it appears that charge
plays a more important role than helicity in the degree of
permeabilization achieved by such a peptide.
The inner membrane permeabilization by ΔFd showed a
lag period of ∼150 s. This coincided with the time at which
the outer membrane permeabilization by ΔFd shows a fall in
intensity (Figure 5A). This sequential phenomenon seems to
suggest that ΔFd may be migrating from the outer to the
inner membrane at ∼150 s and causing heightened permea-
bilization of the inner membrane (Figure 5A and B). How-
ever, the fact that ΔFd is most potent may suggest that
peptides that permeabilize both the outer and inner bacterial
membranes may be more effective antimicrobials than those
that permeabilize the outer membrane alone. It is interesting
to note that dimers C and D can potentially be used to
selectively permeabilize the outer membrane without influ-
encing the integrity of the inner membrane. In this way, they
are expected to sensitize Gram negative bacteria and provide
an adjuvant effect to the action of antibiotics like novobiocin
and erythromycin.21
However, while peptides D and ΔFd exhibited the same
Pmax, the rate of attainment of Pmax was three times faster for
ΔFd than peptide D (Figure 5A). Figure 6D shows the
influence of charge and helicity on the rate of bacterial outer
membrane permeabilization. The graph clearly shows that
rate of permeabilization is influenced by helicity and not
charge. The hexacationic peptide D (less helical) and ΔFd
(strongly helical) showed maximum permeabilization at 600
and 150 s, respectively. The slower rate of permeabilization
by the less helical peptide D suggests the important role of
In summary, our results show that in a dimeric scaffold, a
tetracationic character in a weak helical fold is sufficient to

6086 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16
Anantharaman and Sahal
elicit potent but narrow spectrum antibiotic action (E. coli
(Minimum inhibitory concentration (MIC): 10μM); S. aureus
(MIC > 100 μM)). However, in the same dimeric scaffold, a
hexacationic character embedded in a strong, amphipathic
helix is necessaryforpotentandbroadspectrumaction(E. coli
(MIC 2.5 μM) and S. aureus (MIC 5 μM)). We report that: (a)
at identical helicity, incremental changes in charge enhance
outer membrane permeabilization, (b) at identical charge,
incremental changes in helicity enhance the rate of outer
membrane permeabilization, initiate inner membrane per-
meabilization, enhance antibiotic potency, and kill kinetics
and lead to broad spectrum of action across the Gram
negative-Gram positive divide. Taken together, our results
demonstrate how the interplay of charge and helicity reg-
ulates the activity of antibacterial peptides.
(60F-254) were from Merck, Germany; Mueller Hinton (MH)
broth and Bacto agar were from Difco; propidium iodide (PI)
and Syto 9 were from Molecular Probes.
Preparation of Fmoc-X-DL-threo-β-phenylserine. Fmoc-X-DL-
threo-β-phenylserine (X=Gly, Lys (Boc), Ala) were synthesized
in a salt coupling using a mixed anhydride method. Fmoc amino
acid (15 mmol) (dissolved in 15 mL of sodium refluxed and
distilled THF) was activated at -15 °C for 10 min with IBCF
and NMM (15 mmol each). A solution of 15 mmol of DL-threo-
β-phenylserine made in 1 equiv of NaOH (15 mL) was added to
the mixed anhydride, and the reaction mixture was stirred at
room temperature overnight. Following evaporation of THF,
citric acid was added to the aqueous solution to attain pH ∼ 2.
The precipitate obtained was dissolved in 100 mL of ethyl
acetate and transferred to a separating funnel. Following the
removal of the lower aqueous layer, the ethyl acetate layer was
washed extensively with water to remove citric acid up to its last
traces while monitoring the pH of washings on a pH paper. The
ethyl acetate layer was washed with brine and allowed to pass
through a bed of anhydrous sodium sulfate. Evaporation of
ethyl acetate on a rotary evaporator gave solid dipeptide acids.
Preparation of Fmoc-X-ΔPhe Azalactone. Fmoc-X-DL-threo-
β-phenylserine was mixed with recrystallized anhydrous sodium
acetate (obtained by fusing the salt and allowing it to cool in a
desiccator) in freshly distilled acetic anhydride and stirred over-
night. The thick slurry obtained was mixed with ice and stirred in
a cold room. Following trituration, the yellow dipeptide aza-
lactone was filtered on sinter funnel and dried to constant
weight. The authenticity and purity of the azalactones was
assessed by thin layer chromatography (TLC), mass spectros-
copy, and UV-visible spectroscopy.
Peptide Synthesis. Peptides were synthesized as C terminal
amides using standard Fmoc (fluorenylmethyloxycarbonyl)
chemistry on Rink Amide MBHA (methylbenzhydrylamine)
resin in the manual mode using DIPCDI/HOBt as coupling
agents. Fmoc-Lys(Fmoc)-OH was used to make a branching
core in the dimers. The side chain protections used were: 2,2,-
5,7,8-pentamethylchroman-6-sulfonyl, Pmc (Arg), and t-butyl-
oxycarbonyl, Boc (Lys, Trp). Couplings were done in DMF at
4-fold molar excess at concentrations ∼500 mM. Removal of
Fmoc was by 20% piperidine in DMF. Both coupling of amino
acid and Fmoc deprotection were monitored by Kaiser test.32
ΔPhe was introduced into peptides as Fmoc-X-ΔPhe azalactone
(X = Gly, Lys (Boc), Ala) dipeptide block,20 which was allowed
to couple overnight in DMF. At the completion of assembly of
the peptides, their amino termini were acetylated using 20%
acetic anhydride in DCM. After acetylation of peptides, the
resin was washed extensively with DMF, DCM, and methanol
and dried in a desiccator under vacuum. Peptides were cleaved
by stirring the resin in a cleavage mixture (95% TFA, 2.5%
water, and 2.5% TIS) for 2 h. The suspension was filtered using a
sinter funnel, TFA was rotary evaporated, and peptide was
precipitated in cold dry ether. Ether was filtered through a sinter
funnel, and peptide on the funnel was dissolved in 5% acetic acid
and lyophilized.
Conclusions
The term “antimicrobial activity” is not restricted to
potencies alone but is a comprehensive term, encompassing
kill kinetics, spectrum of action, and membrane perturbing
effects. In the present study, we have observed that each of
these parameters is governed by a fine balance of different
physicochemical factors. The results suggest that the compo-
nents of the physicochemical code, e.g. charge and helicity,
governing antibacterial activity of AMPs, do not act alone but
in concert to influence antibiotic action. This emphasizes the
need to study these not justin part but also in unison with each
other. Such studies are necessary to understand the interac-
tions between these factors so that a more comprehensive
design is achieved. These results can serve as guidelines to
generate improved analogues of existing antimicrobial pep-
tides or de novo design “tailor-made” templates which are
more competent to fight life threatening bacterial infections.
Experimental Section
General. All solvents and reagents were used as received
unless specified. Reverse phase high performance liquid chro-
matography (RPHPLC) peptide purification was performed on
C18 PRC-ODS column (Shimadzu, 2 cm ꢀ 15 cm, 15 μm;
detection: 214 and 280 nm). The purity of the final compounds
was determined by RPHPLC (column: Novapak C18, 4 μm,
4.6 mm ꢀ 250 mm; flow rate: 1 mL/min; solvent system: water-
acetonitrile, 0.1% TFA; gradient: 5% acetonitrile to 75%
acetonitrile in 70 min (1% acetonitrile/minute); detection:
214 nm (Supporting Information Figure 1)). The peptides were
found to have >95% purity. OD600 was measured in a micro-
titer plate reader (VERSA max tunable, Molecular Devices,
Sunnyvale, CA). Fluorescence assays were performed on
Perkin-Elmer LS 50 fluorimeter, and fluorescence microscopic
analysis was done on Nikon Fluorescence microscope. Circular
dichroism spectroscopic analysis was performed on Jasco J-810
spectropolarimeter.
Materials. Fmoc (Fluorenylmethoxycarbonyl) amino acid
derivatives and Rink Amide methylbenzhydrylamine (MBHA)
resin for peptide synthesis were from Nova Biochem; N,N0-
diisopropylcarbodiimide (DIPCDI), piperidine, dimethylform-
amide (DMF), dichloromethane (DCM), N-hydroxybenzotria-
zole (HOBt), isobutylchloroformate (IBCF), trifluoroacetic
acid (TFA), triisopropylsilane (TIS), D,L-threo-β-phenylserine,
sodium hydroxide (NaOH), citric acid, glucose, HEPES, and
bovine serum albumin (BSA) were from Sigma Chemical Company;
N-methylmorpholine (NMM) and 1-N-phenyl-naphthylamine
(NPN) were from Aldrich; sodium chloride, acetic anhydride
and tetrahydrofuran (THF) were from Qualigens; acetic acid,
ethyl acetate, sodium acetate, and sodium sulfate were from
Merck, India; silica gel thin layer chromatography (TLC) plates
Peptide Purification and Mass Spectrometry. Crude peptide
was purified by reverse phase high performance liquid chroma-
tography (RPHPLC) using water-acetonitrile gradient (HPLC
conditions: C18 PRC-ODS column (Shimadzu, 2 cm ꢀ 15 cm,
15 μm; flow rate: 5 mL/min); gradient: 5-75% ACN, 0.1% TFA
in 70 min; detection at 214 and 280 nm). The identity of peptides
was confirmed by electrospray ionization mass spectrometry at
ICGEB, New Delhi (Supporting Information Figure 3).
Antibiotic Susceptibility Testing. The minimum inhibitory
concentrations (MICs) were determined against Escherichia coli
ML35p and the vancomycin/methicillin resistant Staphylococcus
aureus ATCC 700699 according to the modified MIC method for
cationic antimicrobial peptides.33 Overnight grown bacterial cells
were diluted in Mueller Hinton(MH) broth to a celldensity of 105
CFU/mL. Then100 μL of thisculture was aliquoted intothe wells

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16 6087
of a 96-wells flat bottom microtiter plate (Costar) and 11 μL of
10ꢀ stock of each peptide (in 0.2% BSA and 0.01% acetic acid)
was added. Thismixture was incubated at37°C ina rotaryshaker
incubator (Kuhner, Switzerland) set at 200 rpm. After 18 h of
incubation, OD600 was measured in a microtiter plate reader.
MIC is defined as the lowest concentration of a drug that inhibits
measurable growth of an organism after overnight incubation.
Peptide concentrations were determined spectrophotometrically
at 280 nm (ε280: 19000 M-1cm-1 for ΔF, 5050 M-1cm-1 for
tryptophan). Each experiment was done in triplicate and was
repeated at least twice. Standard error measurements (SEM)
values have been calculated for end point OD600 (for each of
the MIC values determined) from triplicates of two independent
experiments. The OD600 values ((SEM) for E. coli were: ∼0.7 (
0.007 (untreated control), 0.011 ( 0.0048 (ΔFd), 0.009 ( 0.002
(peptide C), and 0.008 ( 0.0025 (peptide D). The OD600 values
((SEM) for S. aureus were: ∼0.7 ( 0.006 (untreated control),
0.068 ( 0.003 (ΔFd).
Minimum bactericidal concentrations (MBCs) were deter-
mined in accordance with the guidelines of the Clinical and
Laboratory Standards Institute (CLSI)34 against E. coli ML35p
and S. aureus ATCC 700699 by plating 100 μL from each clear
well of the MIC experiment on MH Agar plates in triplicates.
After incubation for 18 h, MBC was identified as the lowest
concentration that did not permit growth of 99.9% bacteria on
the agar surface. Standard error measurements (SEM) values
have been calculated for end point CFU/mL (for each of the
MBC values determined) from triplicates of two independent
experiments. The CFU/mL values ((SEM) for E. coli were:
∼105 ( 632 CFU/mL (untreated control, at the start of the
experiment), 109 ( 3 CFU/mL (ΔFd), 115 ( 1 CFU/mL
(peptide C), and 108 ( 3 CFU/mL (peptide D). The CFU/mL
values ((SEM) for S. aureus were: ∼105 ( 614 CFU/mL
(untreated control, at the start of the experiment), 100 ( 1
CFU/mL (ΔFd).
Determination of Kill Kinetics. Overnight grown E. coli
ML35p cells were diluted in MH broth to a cell density of 105
CFU/mL. One mL of this cell suspension was incubated with
each peptide at 4X MIC (peptide C (40 μM), peptide D (20 μM),
ΔFd (10 μM), or water (in case of control)) and incubated
at 37 °C, 200 rpm in a rotary shaker incubator (Kuhner,
Switzerland). At different time points, a 100 μL aliquot was
withdrawn, diluted, and plated on MH agar plates. The plates
were kept at 37 °C for 20 h and colonies counted.
Outer Membrane Permeabilization Assay. The outer mem-
brane permeabilization activity of the peptides was determined
by the NPN (N-phenylnapthylamine) assay.22 Midlog phase
E. coli ML35p cells were harvested (4000 rpm, 4 °C, 10 min),
washed, and resuspended in 5 mM glucose/5 mM HEPES buffer
pH 7.2. Then 10 μL of 250X concentration of peptides in water
was added to a cuvette containing 2.5 mL of cells and 10 μM
NPN (50 μL from a 500 μM stock in acetone). Excitation
wavelength: 350 nm (slit width: 5 nm); emission wavelength:
420 nm (slit width: 10 nm). The uptake of NPN as a measure
of outer membrane permeabilization was monitored by the
increase in fluorescence of NPN for 10 min.
suspension was incubated (37 °C, 200 rpm) in 135 μL of 5 mM
HEPES buffer pH 7.2 containing peptides at their MICs over-
night. The samples were incubated with PI (2.7 μM) and Syto 9
(6 μM) for 15 min. A smear was made, heat fixed, and visualized
under a fluorescence microscope. Cells without peptide served
as control.
Circular Dichroism (CD) Spectroscopy. CD experiments were
performed on a spectropolarimeter with a 1 mm path length
cuvette. Spectra were acquired between 190 and 340 nm at 25 °C
(scan speed 200 nm/min, response time 4 s, bandwidth 1 nm) in
20 mM SDS/10 mM sodium phosphate buffer pH 7.5. Five
spectra were collected and averaged.
Acknowledgment. Research Fellowship of the Council of
Scientific and Industrial Research, Government of India to
A.A. is acknowledged. This research was supported by a grant
BT/PR3325/BRB/10/283/2002 to D.S. from the Department
of Biotechnology, Government of India. Gifts of E. coli
ML35p from Dr. Liam Good, Karolinska Institutet, Stockholm,
Sweden, and S. aureus ATCC 700699 from Dr. R. P. Roy,
National Institute of Immunology, New Delhi, India, are grate-
fully acknowledged. We thank the anonymous reviewers for
their constructive criticism that has enriched our work.
Supporting Information Available: Physicochemical proper-
ties of ΔFd and D-Lys-ΔFd, reverse phase HPLC (RPHPLC)
chromatograms, inner membrane permeabilization by peptide
D, ΔFd and D-Lys-ΔFd, and electrospray ionization mass
spectra of the peptides studied. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Zasloff, M. Antimicrobial peptides of multicellular organisms.
Nature 2002, 415, 389–395.
(2) Bulet, P.; Stocklin, R.; Menin, L. Antimicrobial peptides: from
invertebrates to vertebrates. Immunol Rev. 2004, 198, 169–184.
(3) Hancock, R. E. W.; Chapple, D. S. Peptide Antibiotics. Antimi-
crob. Agents Chemother. 1999, 43 (6), 1317–1323.
(4) Hancock, R. E. W.; Sahl, H. Antimicrobial and host defense
peptides as new anti-infective therapeutic strategies. Nature. Bio-
technol. 2006, 24 (12), 1551–1557.
(5) Dathe, M.; Wieprecht, T. Structural features of helical antimicro-
bial peptides: their potential to modulate activity on model
membranes and biological cells. Biochim. Biophys. Acta 1999,
1462, 71–87.
(6) Giangaspero, A.; Sandri, L.; Tossi, A. Amphipathic alpha helical
antimicrobial peptides. A systematic study of the effects of struc-
tural and physical properties on biological activity. Eur. J. Bio-
chem. 2001, 268, 5589–5600.
(7) Makovitzki, A.; Avrahami, D.; Shai, Y. Ultrashort antibacterial
and antifungal lipopeptides. Proc. Natl. Acad. Sci. U.S.A. 2006,
103, 15997–16002.
(8) Dathe, M.; Nikolenko, H.; Meyer, J.; Beyermann, M.; Bienert, M.
Optimization of the antimicrobial activity of magainin peptides by
modification of charge. FEBS Lett. 2001, 501, 146–150.
(9) Dathe, M.; Wieprecht, T.; Nikolenko, H.; Handel, L.; Maloy,
W. L.; MacDonald, D. L.; Beyermann, M.; Bienert, M. Hydro-
phobicity, hydrophobic moment and angle subtended by charged
residues modulate antibacterial and haemolytic activity of amphi-
pathic helical peptides. FEBS Lett. 1997, 403, 208–212.
(10) Deslouches, B; Phadke, S. M.; Lazarevic, V.; Cascio, M.; Islam, K.;
Montelaro, R. C.; Mietzner, T. A. De novo generation of cationic
antimicrobial peptides: influence of length and tryptophan sub-
stitution on antimicrobial activity. Antimicrob. Agents Chemother.
2005, 49, 316–322.
Inner Membrane Permeabilization. Midlog phase E. coli
ML35p cells were harvested (4000 rpm, 4 °C, 10 min), washed,
and resuspended in 5 mM glucose in 5 mM HEPES buffer pH
7.2. Then 10 μL of 250X concentration of peptides in water was
added to a cuvette containing 2.5 mL of cells and 2.7 μM
propidium iodide (PI). Excitation wavelength: 535 nm (slit
width: 5 nm); emission wavelength: 617 nm (slit width: 10 nm).
The uptake of PI as a measure of inner membrane permeabiliza-
tion was measuredbythe increase influorescence of PIfor 10 min.
PI and Syto 9 Uptake Based Fluorescence Microscopic Anal-
ysis of Cell Permeabilization by Peptides. Overnight grown
E. coli ML35p cells were subcultured (OD600: 0.35), harvested
by spinning (4000 rpm, 10 min), washed, and resuspended in
5 mM HEPES pH 7.2 to get 108 cfu/mL. Then 15 μL of E. coli
(11) Dewan, P. C.; Anantharaman, A.; Chauhan, V. S.; Sahal, D.
Antimicrobial action of prototypic amphipathic cationic decapep-
tides and their branched dimers. Biochemistry. 2009, 48 (24), 5642–
5657.
(12) Mathur, P.; Ramakumar, S.; Chauhan, V. S. Peptide design using
alpha,beta-dehydro amino acids: from beta-turns to helical hair-
pins. Biopolymers 2004, 76, 150–161.
(13) Yeaman, M. R.; Yount, N. Y. Mechanisms of antimicrobial
peptide action and resistance. Pharmacol. Rev. 2003, 55, 27–55.

6088 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 16
Anantharaman and Sahal
(14) Kondejewskii, L. H.; Niaraki, M. J.; Farmer, S. W.; Lix, B.; Kay,
C. M.; Sykes, B. D.; Hancock, R. E. W.; Hodges, R. S. Dissociation
of Antimicrobial and Hemolytic Activities in Cyclic Peptide Dia-
stereomers by Systematic Alterations in Amphipathicity. J. Biol.
Chem. 1999, 274 (19), 13181–13192.
(15) Liu, Z.; Brady, A.; Young, A.; Rasimick, B.; Chen, K.; Zhou, C.;
Kallenbach, N. R. Length Effects in Antimicrobial Peptides of the
(RW)n Series. Antimicrob. Agents Chemother. 2007, 51 (2), 597–603.
(16) Pathak, N.; Salas-Auvert, R.; Ruche, G.; Janna, M. H.; McCarthy,
D.; Harrison, R. G. Comparison of the effects of hydrophobicity,
amphiphilicity and alpha-helicity on the activities of antimicrobial
peptides. Proteins 1995, 22, 182–186.
antimicrobial peptide: the proline hinge is responsible for the cell-
penetrating ability of buforin II. Proc. Natl. Acad. Sci. U.S.A. 2000,
97, 158245–8250.
(25) Juwadi, P.; Vunnam, V.; Merrifield, E. L.; Boman, H. G.; Merri-
field, R. B. Hydrophobic Effects on Antibacterial and Channel-
forming Properties of Cecropin A-Melittin Hybrids. J. Pept. Sci.
1996, 2, 223–232.
(26) Kristian, S. A.; Datta, V.; Weidenmaier, C.; Kansal, R.; Fedtke, I.;
Peschel, A.; Gallo, R. L.; Nizet, V. D-Alanylation of Teichoic Acids
Promotes Group A Streptococcus Antimicrobial Peptide Resis-
tance, Neutrophil Survival, and Epithelial Cell Invasion. J. Bacter-
iol. 2005, 187 (19), 6719–6725.
(17) Subbalakshmi, C.; Bikshapathy, E.; Sitaram, N.; Nagaraj, R.
Antibacterial and Hemolytic Activities of Single Tryptophan
Analogs of Indolicidin. Biochem. Biophys. Res. Commun. 2000,
274, 714–716.
(18) Chakrabartty, A.; Doig, A. J.; Baldwin, R. L. Helix capping
propensities in peptides parallel those in proteins. Proc. Natl. Acad.
Sci. U.S.A. 1993, 90, 11332–11336.
(19) Malina, A.; Shai, Y. Conjugation of fatty acids with different
lengthsmodulates antibacterial and antifungal activityof a cationic
biologically inactive peptide. Biochem. J. 2005, 390, 695–702.
(20) Ramagopal, U. A.; Ramakumar, S.; Sahal, D.; Chauhan, V. S. De
novo design and characterization of an apolar helical hairpin
peptide at atomic resolution: compaction mediated by weak inter-
actions. Proc. Natl. Acad. Sci. U.S.A. 2001, 98 (3), 870–874.
(21) Varra, M. Agents That Increase the Permeability of the Outer
Membrane. Microbiol. Rev. 1992, 56 (3), 395–411.
(22) Hancock, R. E. W.; Farmer, S. W.; Li, Z. S.; Poole, K. Interaction
of aminoglycosides with the outer membranes and purified lipo-
polysaccharide and OmpF porin of Escherichia coli. Antimicrob.
Agents Chemother. 1991, 35, 1309–1314.
(27) Brogden, K. A. Antimicrobial peptides: pore formers or metabolic
inhibitors in bacteria? Nature Rev. Microbiol. 2005, 3, 238–250.
(28) Nikaidol, H.; Vaara, M. Molecular Basis of Bacterial Outer
Membrane Permeability. Microbiol. Rev. 1985, 49 (1), 1–32.
(29) Friedrich, C.; Scott, M. G.; Karunaratne, N.; Yan, H; Hancock,
R. E. W. Salt-Resistant Alpha-Helical Cationic Antimicrobial
Peptides. Antimicrob. Agents Chemother. 1999, 43 (7), 1542–1548.
(30) Matsuzaki, K.; Sugishita, K.; Harada, M.; Fujii, N.; Miyajima, K.
Interactions of an antimicrobial peptide, magainin 2, with outer
and inner membranes of Gram-negative bacteria. Biochim. Bio-
phys. Acta 1997, 1327, 119–130.
(31) Hao., G.; Shi., Y.; Tang., Y.; Le, G. The membrane action
mechanism of analogs of the antimicrobial peptide Buforin 2.
Peptides 2009, 30, 1421–1427.
(32) Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I.
Color Test for Detection of Free Terminal Amino Groups in
the Solid-Phase Synthesis of Peptides. Anal. Biochem. 1970, 34,
595–598.
(33) Steinberg, D. A.; Hurst, M. A.; Fujii, C. A.; Kung, A. H. C.; Ho,
J. F.; Cheng, F. C.; Loury, D. J.; Fiddes, J. C. Protegrin-1: a broad-
spectrum, rapidly microbicidal peptide with in vivo activity. Anti-
microb. Agents Chemother. 1997, 41, 1738–1742.
(34) Clinical and Laboratory Standards Institute. Methods for deter-
mining bactericidal activity of antimicrobial agents; approved
guideline. Document M26-A. Clinical and Laboratory Standards
Institute: Wayne, PA, 1999.
(23) Boulos, L.; Prevost, M.; Barbeau, B.; Coallier, J.; Desjardins, R.
LIVE/DEAD BacLight: application of a new rapid staining method
for direct enumeration of viable and total bacteria in drinking water.
J. Microbiol. Methods 1999, 37, 77–86.
(24) Park, C. B.; Yi, K.; Matsuzaki, K.; Kim, M. S.; Kim, S. C.
Structure-activity analysis of buforin II, a histone H2A-derived