Synthesis, antibacterial activity and mode of action of novel linoleic acid-dipeptide-spermidine conjugates

Article abstract of DOI:10.1039/c2ob26393a

Towards therapeutically viable mimics of host defense cationic peptides (HDCPs) here we report the design and synthesis of a small library, based on a novel hydrophobic-dipeptide-spermidine template. Lipidated sequences 11, 14, 15, 16, 18 and 19 exhibited potent activity against susceptible as well as drug resistant Gram-positive and Gram-negative bacterial strains. Structure-activity relationships of the template revealed a hydrophobicity window of 50-70% with minimum +2 charges to be crucial for activity and cell selectivity. Active sequences 14, 15 and 16 exhibited different modes of action based on dipeptide composition as revealed by studies on model membranes, intact bacterial cells and DNA. Further, severe damage to surface morphology of methicillin resistant S. aureus caused by 14, 15 and 16 at 10 × MIC was observed. The present study provides us two active sequences (14 and 16) with a membrane perturbing mode of action, cell selectivity to hRBCs and keratinocytes along with potent activity against clinically relevant pathogen MRSA. The designed template thus may prove to be a suitable probe to optimize sequences for better selectivity and potential to combat a wide range of drug resistant strains in further research. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2ob26393a

View Online / Journal Homepage
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: DOI: 10.1039/c2ob26393a
www.rsc.org/obc
PAPER
Synthesis, antibacterial activity and mode of action of novel linoleic
acid–dipeptide–spermidine conjugates†
Seema Joshi,‡a,b Rikeshwer P. Dewangan,‡a Shruti Yadav,a Diwan S. Rawatb and Santosh Pasha*a
Received 18th July 2012, Accepted 4th September 2012
DOI: 10.1039/c2ob26393a
Towards therapeutically viable mimics of host defense cationic peptides (HDCPs) here we report the
design and synthesis of a small library, based on a novel hydrophobic–dipeptide–spermidine template.
Lipidated sequences 11, 14, 15, 16, 18 and 19 exhibited potent activity against susceptible as well as drug
resistant Gram-positive and Gram-negative bacterial strains. Structure–activity relationships of the
template revealed a hydrophobicity window of 50–70% with minimum +2 charges to be crucial for
activity and cell selectivity. Active sequences 14, 15 and 16 exhibited different modes of action based on
dipeptide composition as revealed by studies on model membranes, intact bacterial cells and DNA.
Further, severe damage to surface morphology of methicillin resistant S. aureus caused by 14, 15 and 16
at 10 × MIC was observed. The present study provides us two active sequences (14 and 16) with a
membrane perturbing mode of action, cell selectivity to hRBCs and keratinocytes along with potent
activity against clinically relevant pathogen MRSA. The designed template thus may prove to be a
suitable probe to optimize sequences for better selectivity and potential to combat a wide range of drug
resistant strains in further research.
ceragenins,10,11 oligoacyllysines,12,13 arylamides,14 peptoid
Introduction
based scaffolds15 and lipopeptides16,17 that mimic HDCPs are
The emergence of multiple drug resistant bacterial strains causes
millions of deaths worldwide.1,2 The rise in morbidity and mor-
tality related to microbial infections has led to tremendous
pressure on health care systems.3 Host defense cationic peptides
(HDCPs) are known for their wide range of activity against
microorganisms including bacteria, fungi, viruses and even cancer-
ous cells. With multifaceted roles in innate immunity and a direct
cell lytic mode of action, it is difficult for a microorganism to
endure resistance against HDCPs.4,5 Therefore HDCPs are being
explored as promising alternatives to conventional drugs.6,7
However in spite of all these advantages only a handful of
antimicrobial peptides are under clinical trials due to their high
manufacturing costs, poor pharmacokinetic properties and
associated toxicity issues.8 To address these issues, efforts are
being made to explore the characteristic features of HDCPs such
as net positive charge at physiological pH and hydrophobic
bulk9 to develop more economic membrane active antimicrobial
peptidomimetics. Many classes of compounds such as
being developed as alternative antimicrobial agents with low sus-
ceptibility for development of resistance.
Cationic polyamines spermine, spermidine and putrescine are
ubiquitous components of eukaryotic and prokaryotic cells with
multiple roles in modulating functions of DNA, RNA and pro-
teins inside the cells.18 A number of modified/conjugated polya-
mines have been reported with various biological activities such
as LPS sequestration,19 anti-parasitic activity,20,21 anticancer
activity,22,23 as well as nucleic acid carriers for DNA transfec-
tion.24 Considering the medicinal potential of polyamines we
anticipated that incorporation of spermidine as a positively
charged moiety to hydrophobic–dipeptides may lead to small
peptidomimetics with better antibacterial activity.
Long-chain free fatty acids (FFAs) are known to be present at
skin surfaces, maintaining an acidic pH which helps to prevent
colonization of various microorganisms including methicillin-
resistant S. aureus and H. pylori.25–27 The mode of bacterial
killing for FFAs has not been unambiguously determined
however, the prime target is believed to be microbial membranes
where direct lysis, perturbation of the electron transport chain,
oxidative phosphorylation and inhibition of fatty acid meta-
bolism have been reported as some of the probable causes
leading to bacterial cell death.28,29 Therefore with multiple non-
specific modes of action, a broad range of activity and minimum
toxicity long chain FFAs were chosen to be incorporated into
designed peptidomimetics as the hydrophobic moiety.
aInstitute of Genomics and Integrative Biology, Mall Road, Delhi, India.
E-mail: spasha@igib.res.in; Fax: +91 11 27667471;
Tel: +91 11 27666156
bDepartment of Chemistry, University of Delhi, Delhi, India
†Electronic supplementary information (ESI) available: HPLC traces,
mass spectra and 1H NMR data of representative designed sequences.
See DOI: 10.1039/c2ob26393a
‡These authors have contributed equally to the work.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Online
Here we report the design and synthesis of a small library of
hydrophobic–dipeptide–spermidine template based sequences
consisting of cationic polyamine spermidine and FFAs (linoleic
acid and stearic acid). We dissected the template into two parts
to evaluate the role of the hydrophobic moiety–dipeptide or cat-
ionic–dipeptide part of the template in imparting activity and/or
selectivity to these sequences. We further evaluated interaction
of the active sequences with model membranes, intact bacterial
cells and DNA to probe into the molecular mode of action.
designed sequences due to its non-ribosomal origin. Based on
various biological activities associated with spermidine conju-
gates as outlined above, we incorporated spermidine at the C-ter-
minus of the dipeptides to give rise to sequences 1–3 (Table 1).
N-Terminal hydrophobic tagging of dipeptide sequences was
done to evaluate the role of FFAs such as linoleic acid and
stearic acid in imparting activity leading to sequences 4–11. To
evaluate the role of lipidation as a control an aromatic moiety
3-(4-hydroxyphenyl)-propionic acid (HPPA) was also conju-
gated. The HPPA was used based on our observations that
covalent hybridization of HPPA to a tetra-peptide template led to
the discovery of some potent tetra-peptidomimetics.35 To have
an appropriate balance between charge and hydrophobicity, the
designed complete template consisting of a hydrophobic tag at
the N-terminal with cationic spermidine at the C-terminus result-
ing in sequences 12–19 was also synthesized. With at least +2
charge and varied degrees of hydrophobicity these conjugates
were further evaluated for antibacterial activity and mode of
action studies.
Results and discussion
1. Design and synthesis
Clinically approved lipopeptide antibiotics such as polymyxin B,
daptomycin and echinocandins have spurred the research for
novel analogues of these classes of lipopeptides with clinical
relevance.30–32 A large number of membrane active lipopeptides
have been designed to achieve net positive charge (at least +2)
and hydrophobic bulk making use of varied chemical
moieties.15–17 However, cationic lipopeptides with saturated fatty
acids are difficult to optimize for cell selectivity.33 Realizing the
importance of charge and amphiphilicity for membrane active
antimicrobial peptidomimetics, we designed a novel short tem-
plate that comprised a hydrophobic moiety attached at the N-
terminus to dipeptide–spermidine. The sequences presented in
the current study are unique where inherently active unsaturated
fatty acids were coupled to dipeptides which were further tagged
with organic polyamine spermidine, prompting membrane
activity as well as cellular translocation to bind DNA. For the
dipeptide portion we made use of three different combinations of
tryptophan (Trp) and ornithine (Orn) amino acids i.e. Trp–Trp,
Trp–Orn and Orn–Orn. Hydrophobic Trp amino acid was chosen
because of its well documented membrane anchoring property.34
Cationic residue Orn was used to impart protease stability to
The sequences were synthesized using a combination of solid
phase and solution phase strategy as presented in Scheme 1.
Structures of representative sequences are shown in Fig. 1.
2. Antibacterial activity
Antibacterial activity of designed sequences against a range of
Gram-positive and Gram-negative bacterial strains was deter-
mined using the serial broth dilution method (Table 2). Out of
the sequences 1–3, sequence 1 exhibited MIC at 113.6 μg mL−1
against S. aureus. Out of sequences 4–11, 4, 5 and 9 were
devoid of antibacterial activity, whereas sequences 6 and 7
showed MIC at 113.6 μg mL−1 against S. aureus and B. subtilis.
Sequence 8 with +1 charge showed moderate activity against
tested strains with MIC in the range of 28.4–113.6 μg mL−1
Sequence 11 with +1 charge and stearic acid tagging showed
.
Table 1 Sequences, molecular mass, charge and % of acetonitrile at RP-HPLC elution
Mass [M]+
Sequences
Composition
Calc.
Obser.
Charge
% acetonitrile at RP-HPLC elution
1
2
3
4
5
6
7
8
NH2-WW-spermidine
NH2-WO-spermidine
NH2-OO-spermidine
HPPA-WW-COOH
HPPA-WO-COOH
LIN-WW-COOH
LIN-WO-COOH
LIN-OO-COOH
STER-WW-COOH
STER-WO-COOH
STER-OO-COOH
HPPA-WW-spermidine
HPPA-WO-spermidine
LIN-WW-spermidine
LIN-WO-spermidine
LIN-OO-spermidine
STER-WW-spermidine
STERWO-spermidine
STER-OO-spermidine
518.3238
446.3238
374.3238
539.2289
467.2289
653.4061
581.4061
509.4061
657.4374
585.4374
513.4374
666.3762
594.3762
780.5535
708.5535
636.5535
784.5848
712.5848
640.5848
518.3228
446.3226
374.3235
539.2277
467.2278
653.62a
581.4063
509.4054
657.68a
585.4366
513.4368
666.3762
594.3760
780.5537
708.5544
636.5540
784.5858
712.5847
640.5853
+3
+4
+5
−1
0
−1
0
+1
−1
0
+1
+2
+3
+2
+3
+4
+2
+3
+4
26
22
19
61
46
84
79
59
88
89
70
52
27
70
62
51
76
68
63
9
10
11
12
13
14
15
16
17
18
19
HPPA: 3-(4-hydroxy phenyl)-propionic acid; LIN: linoleic acid; STER: stearic acid.a ESI-MS data.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Online
Scheme 1 Reagents and conditions: (a) Fmoc-NH-X1-COOH, DIPEA, DCM : DMF (50 : 50), 3 h, MeOH for 30 min, (b) 20% piperidine in DMF,
(c) Fmoc-NH-X2-COOH, HOBt, DIPCDI, DCM : DMF (50 : 50), 1.5 h, (d) R-COOH, HOBt, DIPCDI, DCM : DMF (50 : 50), 14 h, (e) 50% TFA–
DCM, (f) TFE : acetic acid : DCM (1 : 1 : 8), 1.5 h, (g) N1,N4-bis(Boc)-spermidine, DIPCDI, HOBt, THF, 0 °C for 30 min, rt 18 h, (h) (Boc)2O,
DIPEA–DCM, 1.5 h.
good antibacterial activity against Gram-positive bacterial strains
with MIC in the range of 7.1–14.2 μg mL−1. Sequence 12 with
HPPA tagging showed moderate activities against almost all the
be in the range of 3.5–28.4 μg mL−1 and 7.1–14.2 μg mL−1
respectively (except sequence 17). Against the four Gram-posi-
tive strains tested in the study, all the sequences showed MIC
values below 15 μg mL−1. Interestingly MIC of these sequences
against MRSA was found to be comparable or even better as
compared to S. aureus. Sequence 17 with 76% acetonitrile at
RP-HPLC elution was above the hydrophobicity threshold of
70%, above which the sequences were found to be less active
against tested strains. It is important to note that in comparison
to standard antibiotics tetracycline and polymyxin B, sequences
14–19 showed better MIC against MRSA (except 17), where 14
tested strains with MIC in the range of 56.8–113.6 μg mL−1
.
Sequence 13 showed activity against S. aureus and B. subtilis
with MIC at 113.6 μg mL−1. Percentage of acetonitrile from
RP-HPLC elution profile data as an indicator for hydrophobicity
along with MIC data showed that the designed sequences with a
threshold of acetonitrile (50%) with a single positive charge (8
and 11) showed good antibacterial activity (Table 2). Therefore
similar to the previously reported concept of a hydrophobicity
window for HDCPs,36 here we propose a window of 50% to
70% acetonitrile at RP-HPLC elution (Table 1), below or above
this threshold the sequences were either less active or inactive.
MIC data on these sequences thus ensured that lipidation with a
single positive charge may lead to activity (8 and 11) however;
charge alone (1–3) without lipidation is not sufficient to impart
appreciable activity.
and 15 were very promising with MIC at 0.88 μg mL−1
.
3. Toxicity evaluation of designed sequences
The efficacy of designed sequences as safe antibacterial agents
was established based on their interaction with enucleated
hRBCs and human keratinocytes (HaCaT cells). Most of the
sequences (1–7, 9, 12, 13, 16 and 17) were found to be non-
hemolytic up to 62.5 μg mL−1. Sequences 8 and 11 were found
to be lytic to hRBCs, however on conjugation of spermidine,
Sequences 14–19 showed a broad range of activity against
Gram-positive as well as Gram-negative bacterial strains with
MIC values in the range of 0.88–28.4 μg mL−1. Against E. coli
and P. aeruginosa, MIC values of these sequences were found to
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Online
reduction in hemolysis was observed for N-terminal tagged
dipeptides (Table 2). Sequences 14 and 16 showed negligible
hemolytic activity whereas sequence 15 caused 40% hemolysis
at concentrations corresponding to 18–78 times its MIC values
against Gram-positive bacterial strains tested in the study.
Sequences with stearic acid tagging (18 and 19) were found to
be more lytic to hRBC than the linoleic acid counter sequences
(15 and 16). Similar to our results, experimental evidence is
present in the literature that reports cationic lipopeptides with
saturated fatty acid chain length to be more lytic as compared to
their unsaturated counterparts.33
Since HDCP mimics are considered more suitable as topical
agents, we evaluated the effects of designed sequences on
human keratinocytes. Results of the MTT assay on HaCaT cells
showed 80% and 100% cell survival for sequences 14 and 16 at
62.5 μg mL−1 respectively (Fig. 2). Even up to a high concen-
tration of 250 μg mL−1 65% and 78% cell survival was observed
for sequences 14 and 16 respectively. These data thus confirm no
detrimental effects of these sequences on human keratinocytes
even upon incubation for 18 h.
4. Bactericidal kinetics
HDCPs with a predominant membrane active mode of action are
endowed with the potential to kill bacterial cells within minutes
at concentrations higher than MIC. In order to monitor the rapid-
ity of mode of action of designed sequences, we incubated
sequences 14, 15 and 16 with log-phase MRSA at 37 °C and
monitored the course of change in optical density (OD600) at
different time intervals. The results showed inhibition of bac-
terial growth after 2 h at MIC for sequences 15 and 16 (Fig. 3).
Fig. 1 Representative structures of designed antibacterial sequences.
Table 2 Antibacterial activity and hemolysis of designed sequences against Gram-positive and Gram-negative bacterial strains
MIC (μg mL−1
)
Gram-negative bacteria
Gram-positive bacteria
Sequences
E. coli
P. aeruginosa
A. baumannii
S. aureus
E. faecalis
B. subtilis
MRSA
% hemolysis (at 62.5 μg mL−1
)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
>227.2
>227.2
>227.2
>227.2
>227.2
>227.2
>227.2
28.4
>227.2
>227.2
113.6
56.8
ND
ND
ND
ND
ND
ND
>227.2
>227.2
>227.2
>227.2
>227.2
>227.2
ND
56.8
ND
>227.2
28.4
56.8
113.6
>227.2
>227.2
>227.2
>227.2
113.6
113.6
56.8
227.2
113.6
7.1
56.8
113.6
7.1
3.5
3.5
ND
ND
ND
ND
ND
ND
>227.2
>227.2
113.6
113.6
ND
ND
ND
ND
ND
ND
>227.2
>227.2
>227.2
>227.2
56.8
ND
ND
7.1
56.8
>227.2
0.8
0.8
3.5
0
0
0
0
0
0
0
80
0
>227.2
>227.2
>227.2
>227.2
113.6
ND
113.6
ND
113.6
>227.2
7.1
>227.2
>227.2
>227.2
>227.2
28.4
ND
ND
14.2
ND
23
84
0
ND
56.8
113.6
3.5
3.5
7.1
>227.2
22.7
7.1
>227.2
14.2
3.5
0
16
40
4
14.2
7.1
3.5
56.8
7.1
17
18
19
>227.2
28.4
7.1
0.3
0.7
113.6
14.2
14.2
0.3
56.8
28.4
28.4
ND
7.1
3.5
3.5
0.3
>56.8
14.2
14.2
14.2
28.4
ND
3.5
7.1
ND
28.4
3.5
3.5
>50
28.4
0
74
51
ND
ND
Tetracycline
Polymyxin B
0.7
ND
15.5
14.2
ND: not determined.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Online
Fig. 2 Percentage viability of cells upon treatment with different con-
centrations of sequences based on the MTT assay. Error bars represent
mean SD.
Fig. 4 Concentration-dependent leakage of encapsulated calcein dye
from bacterial mimic POPC/POPG LUVs (180 μM) at pH 7.2 measured
after 5 min of incubation with different concentrations of designed
sequences. Sequences are presented as (
) 17, ( ) 18, ( ) 19 and (
) 14, (
) tetracycline.
) 15, ( ) 16,
(
was observed where at the highest concentration tested (7.87 μg
mL−1) partial leakage of vesicle contents was observed (Fig. 4).
For 15 and 18 a rapid increase in fluorescent intensity due to a
preferential pore forming ability was observed, where 18 caused
a burst release of calcein dye at 1.99 μg mL−1 causing almost
72% leakage instantly. For sequences 17 and 19 moderate levels
of leakage were observed where the maximum extent of leakage
reached up to 60% and 51% respectively, at the highest concen-
tration tested. These data make it clear that sequences with
stearic acid conjugation lead to a better pore forming ability as
compared to linoleic acid tagged sequences. Consistent with pre-
vious reports on HDCPs, the Trp–Trp dipeptide sequences 14
and 17 showed a preference to reside at the membrane inter-
phase, causing relatively lower levels of calcein leakage.34
Overall out of the active sequences, sequences 15 and 18 caused
formation of large enough pores for complete leakage of encap-
sulated calcein dye whereas sequences 14 and 16 showed partial
leakage and sequences 17 and 19 caused moderate levels of
leakage. Standard antibiotic tetracycline showed less than 15%
leakage up to the highest concentration tested.
Fig. 3 Time dependent killing of MRSA upon treatment with different
sequences at (A) MIC and (B) 4 × MIC. Sequences are presented as 14
(green bar), 15 (red bar), 16 (purple bar) and control (orange bar).
Sequence 14 was found to be less effective at MIC even up to
5 h. At 4 × MIC all three sequences inhibited bacterial growth
from 2 h onwards keeping growth arrested till 5 h. However
complete eradication of bacterial growth by any of the tested
sequences was not observed up to 5 h. At 4 × MIC 16 exhibited
the most potent inhibition of growth compared to 14 or 15. Thus
in line with HDCPs the designed sequences are capable of inhi-
biting bacterial growth within hours of initial interactions.
6. Membrane depolarization
To study the mode of action on intact MRSA cells, a membrane
depolarization experiment was performed. In this experiment if
the sequences are able to alter membrane potential as a result of
pore formation/membrane destabilization an increase in fluor-
escence intensity is observed. The data show concentration
dependent effects and no perfect co-relation between MIC
values and extent of membrane depolarization could be observed
(Fig. 5). The changes in fluorescence were instant with
maximum increase within 2 min after treatment at all concen-
trations. Only a 48% increase in fluorescence intensity was
caused by active sequences 14 and 17 even up to a concentration
of 19.2 μg mL−1. Sequences 15, 16, 18 and 19 lead to significant
depolarization of membrane potential leading to an almost
5. Calcein leakage
To examine the role of membrane interactions in the mode of
action for designed sequences 14–19, we first compared their
membrane pore forming ability using artificial LUVs comprising
bacterial mimic membranes with composition POPC : POPG
[7 : 3, w/w]. For sequences 14 and 16, graded release of encapsu-
lated calcein dye upon increasing the concentration of sequences
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Online
Fig. 6 Gel retardation assay, binding was assayed by the inhibitory
effect of conjugates on the migration of DNA bands. Various amounts of
conjugates were incubated with 100 ng of plasmid DNA at room temp-
erature for 1 h and the reaction mixtures were applied to 1% agarose gel
electrophoresis.
Fig. 5 Membrane depolarization ability of designed sequence. MRSA
was grown to log phase (OD600 = 0.05) and treated with different con-
centrations of desired sequences. Sequences are presented as (
) 15, ( ) 16, ( ) 17, ( ) 18, ( ) 19 and (
tetracycline.
) 14,
(
)
78–96% increment in fluorescence at a concentration of
14.5 μg mL−1. Concomitant with the depolarization experiment
a PI uptake experiment was set up to evaluate if depolarization
was a lethal event. The PI uptake data showed loss of viability
upon treatment of MRSA with sequences at 19.2 μg mL−1
(data not shown here).
7. DNA binding
Antimicrobial potency of various classes of DNA binding agents
is well reported in the literature.37,38 We accessed DNA binding
ability of sequences 14–19 as the role of spermidine in mem-
brane translocation as well as DNA complexation is well
known.39 Sequences 14 and 16 showed binding with DNA
(100 ng) causing retardation at 12.5 μg mL−1 where for 16 slight
retardation was observed even at 6.25 μg mL−1 (Fig. 6).
Sequence 15 showed excellent DNA binding ability with com-
plete retardation at 3.12 μg mL−1. Sequences 18 and 19
showed good DNA retardation ability with complete binding at
6.25 μg mL−1. Noticeably Trp–Trp sequences 14 and 17 showed
poor DNA complexation which may in part be ascribed to lower
charge density in these sequences. However it was intriguing
that DNA binding was influenced by overall structure of the
sequences as Trp–Orn dipeptide containing sequences 15 and 18
showed the most potent DNA retardation ability as compared to
Orn–Orn analogues 16 and 19. Recently it was shown that a
novel class of DNA minor groove binders based on benzophe-
none tetra-amide scaffolds showed strong DNA binding ability
with membrane active bactericidal mode of action.40 Therefore
unexpectedly DNA binding was not directly involved with mode
of action of this class of compounds.41 On similar lines in the
present study upon comparing DNA binding and antimicrobial
potency (sequence 14 showed lesser DNA binding though was
equipotent as 15) it was evident that there were no direct corre-
lations between DNA retardation and antimicrobial potency/
mode of action.
Fig. 7 Scanning electron microscopic images of MRSA (ATCC
33591) treated with sequences at 10 × MIC for 30 min. The bar in the
image represents 1 μm.
8. Scanning electron microscopy
To have visual evidence for the membrane active mode of
action, we incubated sequences 14, 15 and 16 with MRSA at
concentrations higher than MIC. Disintegration of bacterial
membranes due to pore formation or surface swelling have been
microscopically observed for membrane active HDCPs at con-
centrations higher than MIC.42,43 Control MRSA cells exhibit a
bright smooth appearance with an intact cell membrane
(Fig. 7A).
Membrane damage and cellular debris as small and round
structures were apparent upon treatment of MRSA with sequence
14 (Fig. 7B). In sequence 15 treated S. aureus cells, cellular pro-
trusions as well as flattened cells due to complete leakage of cel-
lular contents were visible (Fig. 7C). For sequence 16 deformed
outer membranes were observed where surface blabbing was
visible without much leakage of cellular contents (Fig. 7D). The
appearance of such protrusions on the surface of S. aureus
caused by HDCPs Gramicidin S and PGLa has recently been
reported as well.44 These data are in agreement with calcein
leakage and membrane depolarization experiments where
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Online
sequence 15 caused cellular damage in the form of membrane
disruption whereas 14 and 16 were found to show lesser leakage
of encapsulated dye.
membrane perturbing mode of action at concentrations higher
than MIC. These sequences were also able to alter the electro-
phoretic mobility of DNA which although was not directly
related to activity may as well be responsible for further
enhanced potency of these sequences due to intracellular mode of
action. The structure–activity work in this study paves the way for
the design of optimized FFAs based peptidomimetics with spermi-
dine/spermine conjugated with different combinations of amino
acids which is currently in progress in our laboratory.
Overall the present study affords us sequences 14–19 with
potent activity against a broad range of bacterial strains including
MRSA. Upon dissecting the designed template to identify
features responsible for activity and selectivity, we found that the
cationic charge imparted to dipeptides by spermidine in
sequences 1–3 was not sufficient per se to show bactericidal
properties. Lipidation alone with neutral or negative charge in
sequences 6, 7, 9 and 10 also led to low activity. With unit posi-
tive charge, lipidated sequences 8 and 11 exhibited improved
activity though with compromised cell selectivity (Table 2).
However conjugation of spermidine in sequences 14–19
improved potency as well as cell selectivity. With minimum +2
charges, lipidated sequences 14–19 showed a broad range of
antibacterial activity.
Experimental section
Materials
Fmoc-protected amino acids and resins were purchased from
Novabiochem. Dimethylamino pyridine (DMAP), N,N-diisopro-
pylcarbodiimide (DIPCDI), N-hydroxybenzotrizole (HOBt), di-
isopropyl ethylamine (DIPEA), N-methyl pyrrolidinone (NMP),
piperidine, trifluoroethanol (TFE), trifluoroacetic acid (TFA), tri-
isopropyl silane (TIS), calcein and 3,3′-dipropylthiadicarbocya-
nine iodide (DiSC35) were obtained from Sigma Chemical Co.
1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and
1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol)
(sodium salt) (POPG) were purchased from Avanti Polar Lipids.
Dulbecco’s modified Eagles’ medium-high glucose (DMEM),
antibiotic/antimycotic solutions, heat inactivated fetal bovine
serum (FBS), trypsin from porcine pancreas and 3-(4,5-
Since the sequences show potent activity against MRSA, to
have better insights into the mode of action of designed
sequences we characterized their interactions with S. aureus
mimic artificial membranes, intact MRSA and DNA.
Mode of action studies revealed a predominant role of the
dipeptide sequence in initial binding, bactericidal kinetics and
membrane disrupting abilities of designed sequences. For
sequence 14, low leakage causing ability, lower levels of mem-
brane depolarization as well as reduced DNA binding ability
make membrane destabilization a less probable mode of action at
MIC. A slower bactericidal kinetics of 14 at MIC might as well
be due to different modes of action operative at low concen-
trations. Slower bactericidal kinetics has previously been
reported for HDCP mimics interfering with vital functions in
bacterial cells other than membrane disruption.45 However in
SEM studies at concentrations 10 × MIC cellular debris and dif-
fused outer bacterial membranes were evident for 14 potentiating
a membrane disruptive mode of action (Fig. 7B). Sequence 15
showed faster leakage of calcein along with membrane depolar-
ization, rapid killing kinetics and excellent DNA binding ability.
Therefore this sequence showed clean membrane perturbing
mode of action at MIC as well as higher concentrations as was
evidenced in SEM images of the treated MRSA (Fig. 7C).
Sequence 16 with rapid bactericidal kinetics, good DNA binding
ability caused appreciable damage to membrane potential in
MRSA at the tested concentrations, however moderate levels of
leakage causing ability showed that either transient pores were
formed or the pores were not large enough to cause leakage of
calcein which is evident by the surface blabbing observed in
SEM studies. A low leakage causing ability in spite of potent
membrane depolarization has previously been reported for ana-
logues of HDCP indolicidin.46
dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium
bromide
(MTT) were obtained from Sigma Aldrich Chemical Company.
All solvents used for the purification were of HPLC grade and
obtained from Merck, Germany. Dimethylformamide (DMF) and
dichloromethane (DCM) were obtained from Merck India. DMF
was double distilled prior to use.
Synthesis and purification of sequences
The dipeptides were synthesized on 2-chlorotrityl chloride resin
as a solid support using Fmoc chemistry as reported pre-
viously.47 The terminal amino group of dipeptides was Boc pro-
tected on a solid support before cleavage from the resin under
mild conditions (TFE : CH3COOH : CH2Cl2 cocktail, 1 : 1 : 8) to
retain Boc groups. The cleaved dipeptides were coupled with
N1,N4-bis(boc)spermidine (SIGMA) using HOBt and DIPCDI in
dry tetrahydrofuran under a N2 atmosphere at 0 °C for 30 min
followed by 18 h at rt as reported previously.48 The obtained
product was dissolved in CHCl3 (15 mL) and washed with 1%
aqueous NaHCO3 (50 mL), 1% aqueous HCl (50 mL), and brine
(50 mL). The organic layer was dried over anhydrous Na2SO4
and concentrated under reduced pressure to give crude dipeptide
spermidine conjugates. Boc groups were removed from the con-
jugates using 50% TFA in CH2Cl2 to give sequences 1–3. On a
solid support N-terminal end tagging of dipeptides was achieved
by coupling 4 equiv. of 3-(4-hydroxyphenyl)-propionic acid
(HPPA)/linoleic acid/stearic acid overnight with HOBt and
DIPCDI. The Kaiser test was performed to check completion of
reactions on a solid support.49 The N-terminal tagged di-peptido-
mimetics were cleaved from a solid support under two different
conditions. For synthesis of 4–11, cleavage was effectuated
using 50% TFA in DCM. For synthesis of 12–19, cleavage was
Conclusion
In summary, using simple chemistry and economically viable
building blocks FFAs (linoleic acid/stearic acid), Trp, Orn and
spermidine, we obtained 6 active sequences with a broad range
of activity against Gram-positive as well as Gram-negative bac-
terial strains including clinically relevant pathogen MRSA.
Sequences 14 and 16 showed excellent cell selectivity and
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Online
performed under mild conditions (TFE : CH3COOH : CH2Cl2
cocktail 1 : 1 : 8). Further Boc protected N-terminal tagged con-
jugates were coupled with N1,N4-bis(boc) spermidine as
described for dipeptides earlier in the text. Finally Boc groups
were removed with 50% TFA resulting in sequences 12–19. Syn-
thesized sequences were purified using an RP-HPLC column
(7.8 × 300 mm, 125 Å, 10 μm particle size) with either gradients
of 10 to 90% buffer 2 where buffer 1 was water (0.05% TFA)
and buffer 2 was acetonitrile (0.05% TFA) over 45 min or 30 to
100% buffer 2 gradients were run over 45 min where buffer 1
was water (0.1% TFA) and buffer 2 was acetonitrile (0.1% TFA).
The correct sequences after purification were confirmed by
LC-MS/MS (Quattro micro API, Waters), LC-ESI-HRMS on
UHPLC (Dionex, Germany) and LTQ Orbitrap XL (Thermo
where A represents absorbance of sample wells at 540 nm and
A0 and At represent zero percent and 100% hemolysis determined
in PBS and 1% Triton X-100, respectively.
Cytotoxicity
To assess cell viability, the MTT assay was performed as
described previously.52 HaCaT keratinocytes, 3000 cells per
well, were seeded in 96-well plates in DMEM HAMS
F12 media supplemented with 10% serum (FBS) to grow over-
night. The next day media were aspirated and fresh incomplete
media were added (50 μL per well). To the wells serial two-fold
dilutions of different test sequences (50 μL) were added and the
plates were incubated at 37 °C with 5% CO2 for 18 h. After 18 h
the media were aspirated and 100 μL of MTT solution was
added to each well. The plates were further incubated for 4 h in
CO2 at 37 °C. After 4 h the MTT-containing medium was
removed by aspiration. The blue formazan product generated
was dissolved by the addition of 100 μL of 100% DMSO per
well. The plates were then gently swirled for 2–3 min at room
temperature to dissolve the precipitate. The absorbance was
monitored at 540 nm. Percentage viability was calculated based
on the following formula:
1
Fisher Scientific, USA) mass determination and H NMR. Mass
1
spectra, analytical HPLC traces and H NMR data of representa-
tive sequences are provided in supplementary files.
Antibacterial activity
Antibacterial activity of designed sequences was evaluated using
a modification of the serial broth dilution method as reported
previously.50 Bacterial strains used in this study were as follows,
E. coli (ATCC 11775), P. aeruginosa (ATCC 25668), A. bau-
mannii (ATCC 19606), S. aureus (ATCC 29213), E. faecalis
(ATCC 7080), B. subtilis (ATCC 6633) and methicillin resistant
S. aureus (ATCC 33591). The inoculums were prepared from
mid-log phase bacterial cultures. Each well of the first 11
columns of a 96-well polypropylene micro titre plate (SIGMA)
was inoculated with 100 μL of approximately 105 CFU mL−1 of
bacterial suspension per mL of Mueller-Hinton broth (MHB,
DIFCO). Then 11 μL of serially diluted test sequences in
0.001% acetic acid and 0.2% bovine serum albumin (SIGMA)
over the desired concentration range was added to the wells of
micro titre plates. The micro titre plates were incubated overnight
with agitation (200 rpm) at 37 °C. After 18 h absorbance was
% cell viability ¼ ðA=AcontrolÞ ꢂ 100
where A represents sample absorbance at a given concentration
and Acontrol represents untreated cells. The experiment was
repeated thrice and results are given as mean SD.
Bactericidal kinetics
Overnight cultures of methicillin resistant S. aureus (ATCC
33591) were grown in fresh MHB up to log phase. For determin-
ing the time course of killing activity 100 μL of fresh MHB was
added to all wells of the 96 well-plate. Then 90 μL of approxi-
mately 105 CFU mL−1 were added to the wells of the 96-well
plate (4 wells for a single concentration). Then 10 μL of appro-
priate concentrations of test sequences corresponding to MIC
and 4 × MIC were added to the wells. The plates were incubated
at 37 °C at 200 rpm. Absorbance of the plates was read at
600 nm at various time points at 0, 1, 2, 3, 4 and 5 h. The exper-
iment was repeated on three different days and values are plotted
as mean SD.
read at 630 nm. Cultures (approximately 105 CFU mL−1
)
without test sequences were used as a positive control. Un-inocu-
lated MHB was used as a negative control. Tests were carried out
in duplicate on three different days. Minimum inhibitory concen-
tration (MIC) is defined as the lowest concentration of test
sequences that completely inhibits growth. For comparison
peptide antibiotic polymyxin B and tetracycline were also
assayed under identical conditions.
Calcein leakage
Hemolytic activity
The ability of designed sequences 14–19 to cause leakage from
artificial LUVs composed of bacterial mimic membrane (POPC/
POPG) was accessed as described previously.53,54 Briefly,
desired mixtures of the lipids POPC/POPG (7 : 3, w/w) were dis-
solved in a 2 mL chloroform–methanol mixture in a 150 mL
round bottom flask. The solvent was removed under a stream
of nitrogen and the lipid film obtained was lyophilized
overnight to remove any traces of organic solvent. The dry
lipid films were rehydrated with 10 mM Tris-HCl [70 mM
calcein, 150 mM NaCl, 0.1 mM EDTA]. The liposome
suspension obtained after rehydration was freeze thawed for
five cycles and extruded 16 times through two stacked
Hemolytic activity assay was done as described previously.51
Briefly, 100 μL of fresh hRBC suspension 4% v/v in PBS
(35 mM phosphate buffer, 150 mM NaCl) was placed in a 96-
well plate. After incubation of the test sequences (100 μL) in the
erythrocyte solution for 1 h at 37 °C, the plates were centrifuged
and the supernatant (100 μL) was transferred to fresh 96-well
plates. Absorbance was read at 540 nm using an ELISA plate
reader (Molecular Devices). Percent hemolysis was calculated
using the following formula:
% hemolysis ¼ 100½ðA ꢀ A0Þ=ðAt ꢀ A0Þꢁ
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012

View Online
DNA binding assay
polycarbonate filters (Mini extruder, Avanti Polar Lipids).
Free calcein was removed by passing the liposome suspension
through a Sephadex G-50 column at 23 °C and eluting with a
buffer containing 10 mM Tris-HCl [150 mM NaCl, 0.1 mM
EDTA]. After passing the liposome through a Sephadex G-50,
liposome diameter was measured by dynamic light scattering
using a Zetasizer Nano ZS (Malvern Instruments). The average
diameter of LUVs was found to be in the range of 90–110 nm.
Different concentrations of test sequences were incubated
with POPC/POPG LUVs for 5 min before exciting the
samples. Leakage was monitored by measuring the fluorescence
intensity at an emission wavelength of 520 nm upon excitation
at 490 nm on a model Fluorolog (Jobin Yuvon, Horiba) spec-
trofluorimeter. A slit width of 3 nm was used for both excitation
and emission. Percentage dye leakage was calculated using the
formula
Gel retardation experiments were performed as described pre-
viously.54,57 Briefly, 100 ng of plasmid DNA (pBluescript II
SK+) was mixed with increasing amounts of test sequences
14–19 in 20 μL of binding buffer (5% glycerol, 10 mM Tris-
HCl, pH 8.0, 1 mM EDTA, 1 mM dithiothreitol, 20 mM KCl,
and 50 μg mL−1 bovine serum albumin). Reaction mixtures were
incubated at room temperature for 1 h. Subsequently, 4 μL of
native loading buffer was added (10% Ficoll 400, 10 mM Tris-
HCl, pH 7.5, 50 mM EDTA, 0.25% bromophenol blue, and
0.25% xylene cyanol), and a 20 μL aliquot subjected to 1%
agarose gel electrophoresis in 0.5× Tris borate–EDTA buffer
(45 mM Tris-borate and 1 mM EDTA, pH 8.0). The gels were
run for 1.5 h at 80 V and visualized with ethidium bromide. The
plasmid DNA was purchased from Stratagene and was used as
such without further purification.
% dye leakage ¼ 100½ðF ꢀ F0Þ=ðFt ꢀ F0Þꢁ
Scanning electron microscopy
where F is the fluorescence intensity achieved by addition of
different concentrations of sequences. F0 and Ft are fluorescence
intensities in buffer and with Triton X-100 (20 μL of 10%
solution) respectively. All measurements were made in duplicate
and less than 4% deviation was obtained in the data points. A
phosphate assay was performed to determine the concentration
of lipids for the leakage experiment.55
For electron microscopy samples were prepared as described
previously.58 Briefly, freshly inoculated methicillin resistant
S. aureus (ATCC 33591) was grown on MHB up to an OD600 of
0.5 (corresponding to 108 CFU mL−1). Bacterial cells were then
spun down at 4000 rpm for 15 min, washed thrice in PBS
(20 mM, 150 mM NaCl) and re-suspended in an equal volume
of PBS. The cultures were then incubated with test sequences
14, 15 or 16 at 10 × MIC for 30 min. Controls were run in the
absence of sequences. After 30 min, the cells were spun down
and washed with PBS thrice. For cell fixation the washed bac-
terial pallet was re-suspended in 1 mL of 2.5% glutaraldehyde in
PBS and was incubated at 4 °C for 4 h. After fixation, cells were
spun down and washed with PBS twice. Further the samples
were dehydrated in series of graded ethanol solutions (30% to
100%), and finally dried in desiccators under a vacuum. An auto-
matic sputter coater (Polaron OM-SC7640) was used for coating
the specimens with 20 nm gold particles. Then samples were
viewed via a scanning electron microscope (EVO 40, Carl Zeiss,
Germany).
Membrane depolarization
For the evaluation of membrane depolarization a previously
defined method was used.56 Briefly, overnight grown MRSA was
subcultured into MHB for 2–3 h at 37 °C to obtain midlog phase
cultures. The cells were centrifuged at 4000 rpm for 10 min at
25 °C, washed, and re-suspended in respiration buffer (5 mM
HEPES, 20 mM glucose, pH 7.4) to obtain a diluted
suspension of OD600 ≈ 0.05. A membrane potential-sensitive
dye, 3,3′-dipropylthiadicarbocyanine iodide (DiSC35), 0.18 μM
(prepared in DMSO) was added to a 500 μL aliquot of the
re-suspended cells and allowed to stabilize for 1 h. Baseline
fluorescence was acquired using a Fluorolog (Jobin Yuvon,
Horiba) spectrofluorometer by excitation at 622 nm and
emission at 670 nm. A bandwidth of 5 nm was employed
for excitation and emission. Subsequently, increasing
concentrations of test sequences between 2 and 19.2 μg mL−1
were added to the stabilized cells and the increase of
fluorescence on account of the dequenching of DiSC35 dye
was measured after every 2 min to obtain the maximal
depolarization. Percent depolarization was calculated by using
the formula
Acknowledgements
This work was financially supported by a Council of Scientific
and Industrial Research (CSIR) network project NWP-013. We
thank Dr Souvik Maiti and Dr Kausik Chakraborty for useful
discussions and instrumentation facility. We acknowledge the
Analytical Instrumental Research Facility at Jawahar Lal Nehru
University, Delhi, India for providing us scanning electron
microscopic facility. Dr V. Sabareesh and Richa Guleria are
acknowleged for help in ESI-HR-MS data acquisition. The
author S.J. is thankful to CSIR, New Delhi, India, for the award
of SRF.
% depolarization ¼ ðF ꢀ F0Þ=ðFm ꢀ F0Þ ꢂ 100
where F is the fluorescence intensity 2 min after addition of
sequences, F0 is the initial basal fluorescence intensity, and Fm is
the maximum fluorescence intensity obtained after addition of
10 μg mL−1 gramicidin. Percent depolarization mean SD of
two independent experiments was plotted versus increasing con-
centrations of different sequences.
References
1 S. B. Levy and B. Marshall, Nat. Med., 2004, 10, 122.
2 J. Davies and D. Davies, Microbiol. Mol. Biol. Rev., 2010, 74, 417.
3 A. Opar, Nat. Rev. Drug Discovery, 2007, 6, 943.
4 R. E. W. Hancock and H. G. Sahl, Nat. Biotechnol., 2006, 24, 1551.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem.

View Online
5 M. Zasloff, Nature, 2002, 415, 389.
6 J. He, D. K. Yarbrough, J. Kreth, M. H. Anderson, W. Shi and R. Eckert,
Antimicrob. Agents Chemother., 2010, 54, 2143.
7 S. E. Blondelle and K. Lohner, Curr. Pharm. Des., 2010, 28, 3204.
8 P. W. Latham, Nat. Biotechnol., 1999, 17, 755.
9 M. B. Strom, B. E. Haug, M. L. Skar, W. Stensen, T. Stiberg and
J. S. Svendsen, J. Med. Chem., 2003, 46, 1567.
10 B. Ding, N. Yin, Y. Liu, J. C. Garcia, R. Evanson, T. Orsak, M. Fan,
G. Turin and P. B. Savage, J. Am. Chem. Soc., 2004, 126, 13642.
11 J. N. Chin, M. J. Rybak, C. M. Cheung and P. B. Savage, Antimicrob.
Agents Chemother., 2007, 51, 1268.
33 H. Sarig, S. Rotem, L. Ziserman, D. Danino and A. Mor, Antimicrob.
Agents Chemother., 2008, 52, 4308.
34 D. I. Chan, E. J. Prenner and H. J. Vogel, Biochim. Biophys. Acta, 2006,
1758, 1184.
35 G. S. Bisht, D. S. Rawat, A. Kumar, R. Kumar and S. Pasha, Bioorg.
Med. Chem. Lett., 2007, 17, 4343.
36 Y. Chen, M. T. Guarnieri, A. I. Vasil, M. L. Vasil, C. T. Mant and
R. S. Hodges, Antimicrob. Agents Chemother., 2007, 51, 1398.
37 J. A. Kaizerman, M. I. Gross, Y. Ge, S. White, W. Hu, J. X. Duan,
E. E. Baird, K. W. Johnson, R. D. Tanaka, H. E. Moser and R. W. Burli,
J. Med. Chem., 2003, 46, 3914.
12 I. S. Radzishevsky, S. Rotem, D. Bourdetsky, S. Navon-Venezia,
Y. Carmeli and A. Mor, Nat. Biotechnol., 2007, 25, 657.
13 F. Zaknoon, H. Sarig, S. Rotem, L. Livne, A. Ivankin, D. Gidalevitz and
A. Mor, Antimicrob. Agents Chemother., 2009, 53, 3422.
14 N. Srinivas, K. Moehle, K. A. Hadeed, D. Obrecht and J. A. Robinson,
Org. Biomol. Chem., 2007, 5, 3100.
38 A. I. Khalaf, A. H. Ebrahimabadi, A. J. Drummond, N. G. Anthony,
S. P. Mackay, C. J. Suckling and R. D. Waigh, Org. Biomol. Chem.,
2004, 2, 3119.
39 J. G. Delcros, S. Tomasi, S. Duhieu, M. Foucault, B. Martin, M. Roch,
V. E i fler-Lima, J. Renault and P. Uriac, J. Med. Chem., 2006, 49,
232.
15 N. P. Chongsiriwatana, T. M. Miller, M. Wetzler, S. Vakulenko,
A. J. Karlsson, S. P. Palacek, S. Mobashery and A. E. Barron, Antimicrob.
Agents Chemother., 2011, 55, 417.
16 A. Makovitzki, D. Avrahami and Y. Shai, Proc. Natl. Acad. Sci. U. S. A.,
2006, 103, 15997.
17 W. Kamysz, C. Silvestri, O. Cirioni, A. Giacometti, A. Licci, A. Della
Vittoria, M. Okroj and G. Scalise, Antimicrob. Agents Chemother., 2007,
51, 354.
40 S. K. Vooturi, C. M. Cheung, M. J. Rybak and S. M. Firestine, J. Med.
Chem., 2009, 52, 5020.
41 S. K. Vooturi, M. B. Dewal and S. M. Firestine, Org. Biomol. Chem.,
2011, 9, 6367.
42 I. L. Exposito, L. Amigo and I. Recio, Biochim. Biophys. Acta, 2008,
1778, 2444.
43 A. J. Hyde, J. Parisot, A. McNichol and B. B. Bonev, Proc. Natl. Acad.
Sci. U. S. A., 2006, 103, 19896.
18 K. Igarashi and K. Kashiwagi, Biochem. Biophys. Res. Commun., 2000,
271, 559.
19 A. Shrestha, D. Sil, S. S. Malladi, H. J. Warshakoon and S. A. David,
Bioorg. Med. Chem. Lett., 2009, 19, 2478.
44 M. Hartmann, M. Berditsch, J. Hawecker, M. F. Ardakani, D. Gerthsen
and A. S. Ulrich, Antimicrob. Agents Chemother., 2010, 54, 3132.
45 S. Rotem, I. S. Radzishevsky, D. Bourdetsky, S. N. Venezia, Y. Carmeli
and A. Mor, FASEB J., 2008, 22, 2652.
20 X. Bi, C. Lopez, C. J. Bacchi, D. Rattendi and P. M. Woster, Bioorg.
Med. Chem. Lett., 2006, 16, 3229.
46 Y. H. Nan, J. K. Bang and S. Y. Shin, Peptides, 2009, 30, 832.
47 K. Barlos, O. Chatzi, D. Gatos and G. Stavropoulos, Int. J. Pept. Protein
Res., 1991, 37, 513.
48 W. H. Chen, X. B. Shao, R. Moellering, C. Wennersten and S. L. Regen,
Bioconjugate Chem., 2006, 17, 1582.
21 B. K. Verlinden, J. Niemand, J. Snyman, S. K. Sharma, R. J. Beattie,
P. M. Woster and L. M. Birkholtz, J. Med. Chem., 2011, 54, 6624.
22 T. Yildirim, K. Bilgin, G. Y. Ciftci, E. T. Ecik, E. Senkuytu, Y. Uluda,
L. Tomak and A. Kilic, Eur. J. Med. Chem., 2012, 52, 213.
23 R. A. Casero Jr. and P. M. Woster, J. Med. Chem., 2009, 52, 4551.
24 A. Liberska, A. Lilienkampf, A. U. Broceta and M. Bradley, Chem.
Commun., 2011, 47, 12774.
25 D. R. Drake, K. A. Brogden, D. V. Dawson and P. W. Wertz, J. Lipid
Res., 2008, 49, 4.
26 M. Farrington, N. Brenwald, D. Haines and E. Walpole, J. Med. Micro-
biol., 1992, 36, 56.
49 E. Kaiser, R. L. Colescott, C. D. Bossinger and P. I. Cook, Anal.
Biochem., 1970, 34, 595.
50 M. A. Wikler, D. E. Low, F. R. Cockerill, D. J. Sheehan, W. A. Craig,
F. C. Tenover and M. L. Dudley, Methods for dilution antimicrobial sus-
ceptibility tests for bacteria that grow aerobically: approved standard-
seventh edition (2006) CLSI (formerly NCCLS) M7-A7.
51 M. Sharma, P. Joshi, N. Kumar, S. Joshi, R. K. Rohilla, N. Roy and
D. S. Rawat, Eur. J. Med. Chem., 2011, 46, 480.
27 C. Q. Sun, C. J. O’Connor and A. M. Roberton, FEMS Immunol. Med.
Microbiol., 2003, 36, 9.
28 P. A. Desbois and V. J. Smith, Appl. Microbiol. Biotechnol., 2010, 85,
1629.
52 L. Schmidtchen, G. Ringstad, H. Kasetty, M. W. Mizuno, M. Rutland and
M. Malmsten, Biochim. Biophys. Acta, 2011, 1808, 1081.
53 L. Zhang, A. Rozek and R. E. W. Hancock, J. Biol. Chem., 2001, 276,
35714.
29 C. J. Zheng, J. S. Yoo, T. G. Lee, H. Y. Choc, Y. H. Kim and W. G. Kim,
FEBS Lett., 2005, 579, 5157.
30 M. Vaara, J. Fox, G. Loidl, O. Siikanen, J. Apajalahti, F. Hansen,
N. Frimodt-Moller, J. Nagai, M. Takano and T. Vaara, Antimicrob. Agents
Chemother., 2008, 52, 3229.
54 S. Joshi, G. S. Bisht, D. S. Rawat, A. Kumar, R. Kumar, S. Maiti and
S. Pasha, Biochim. Biophys. Acta, 2010, 1798, 1864.
55 G. R. Bartlett, J. Biol. Chem., 1959, 234, 466.
56 P. N. Domadia, A. Bhunia, A. Ramamoorthy and S. Bhattacharjya, J. Am.
Chem. Soc., 2010, 132, 18417.
31 J. N. Steenbergen, J. Alder, G. M. Thorne and F. P. Tally, J. Antimicrob.
Chemother., 2005, 55, 283.
57 S. T. Yang, J. Y. Lee, H. J. Kim, Y. J. Eu, S. Y. Shin, K. S. Hahm and
J. I. Kim, FEBS J., 2006, 273, 4040.
32 M. P. C. Mulder, J. A. W. Kruijtzer, E. J. Breukink, J. Kemmink,
R. J. Pieters and R. M. J. Liskamp, Bioorg. Med. Chem., 2011, 19, 6505.
58 M. Singh and K. Mukhopadhyay, Antimicrob. Agents Chemother., 2011,
55, 1920.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2012