Bile acid amphiphiles with tunable head groups as highly selective antitubercular agents

Article abstract of DOI:10.1039/c4md00303a

Tuberculosis faces major challenges for its cure due to (a) long treatment period, (b) emergence of drug resistance bacteria, and (c) poor patient compliance. Disrupting the membrane integrity of mycobacteria as a therapeutic strategy has not been explored well as the rigid, waxy, and hydrophobic nature of mycobacterial lipids does not allow binding and penetration of charged amtimicrobial amphiphiles and peptides. Here, we present a new concept that fine-tuning of the charged head group modulates the specificity of amphiphiles against bacterial membranes. We show that hard-charged amphiphiles interact with mycobacterial trehalose dimycolates and penetrate through rigid mycobacterial membranes. In contrast, soft-charged amphiphiles specifically inhibit the growth of both E. coli and S. aureus via electrostatic interactions. These subtle variations between interactions of amphiphiles and bacterial membranes could be explored further to design more specific and potent antimycobacterial agents. This journal is

Full text of DOI:10.1039/c4md00303a

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Bile acid amphiphiles with tunable head groups as
highly selective antitubercular agents†
Cite this: DOI: 10.1039/c4md00303a
Sandhya Bansal,‡a Manish Singh,‡a Saqib Kidwai,‡b Priyanshu Bhargava,a
b
ac
Ashima Singh,a Vedagopuram Sreekanth,ac Ramandeep Singh and Avinash Bajaj
*
*
Tuberculosis faces major challenges for its cure due to (a) long treatment period, (b) emergence of drug
resistance bacteria, and (c) poor patient compliance. Disrupting the membrane integrity of mycobacteria
as a therapeutic strategy has not been explored well as the rigid, waxy, and hydrophobic nature of
mycobacterial lipids does not allow binding and penetration of charged amtimicrobial amphiphiles and
peptides. Here, we present a new concept that fine-tuning of the charged head group modulates the
specificity of amphiphiles against bacterial membranes. We show that hard-charged amphiphiles interact
with mycobacterial trehalose dimycolates and penetrate through rigid mycobacterial membranes. In
contrast, soft-charged amphiphiles specifically inhibit the growth of both E. coli and S. aureus via
electrostatic interactions. These subtle variations between interactions of amphiphiles and bacterial
membranes could be explored further to design more specific and potent antimycobacterial agents.
Received 11th July 2014
Accepted 31st August 2014
DOI: 10.1039/c4md00303a
www.rsc.org/medchemcomm
charge.4 Amphiphiles mimicking these AMPs with variable
cationic charge groups on hydrophobic moieties have been
explored for their antimicrobial activities.5,6 These amphiphiles
exert their antimicrobial activity through electrostatic interac-
tions with the lipid components of bacterial membranes.7,8
These membrane-disruptive amphiphiles have the ability to (a)
shorten the duration of treatment, and (b) eradicate drug-
resistant bacteria.
Introduction
Tuberculosis (TB) remains a global health burden with ꢀ8.6
million individuals infected with M. tuberculosis (Mtb) and
accounting for ꢀ2.0 million deaths in 2012.1 Major drawbacks
of the current TB regimen targeting essential mycobacterial
pathways is the 6 month long duration and poor compliance
among patients. This situation has further worsened due to the
emergence of various MDR/XDR TB strains, HIV–TB co-infec-
tion, and BCG vaccine failure to impart protection against
pulmonary TB.2 The advent of computational methods along
with whole-cell and HTS-based assays have led to the identi-
cation of various scaffolds that are currently in clinical trials.
The newly identied scaffolds should have (i) improved activity
against dormant bacteria, (ii) novel mechanism of action, (iii)
specicity against mycobacteria, (iv) good pharmacokinetics/
pharmacodynamics properties, and (v) compatibility with
current TB and retroviral therapies.3
Mycobacterial non-polar lipids present in the outer
membrane of mycobacteria may not interact with polar charged
amphiphiles/peptides, and hence do not allow their insertion
across rigid hydrophobic mycolic lipids. Hence, disrupting the
integrity of mycobacterial membranes has not been extensively
explored to combat TB.9 Therefore, the present study was con-
ducted to: (i) address the lack of specicity of AMPs and
amphiphiles against microorganisms, (ii) explore the disrup-
tion of membrane integrity as a mechanism to combat TB, and
(iii) better understand interactions between amphiphiles and
mycobacterial membranes.
Bile acids are inherently facial amphiphilic in nature due to
the stereochemical orientation of hydroxyl groups.10 Savage and
co-workers have synthesized cholic acid derived cationic anti-
microbials that possessed a higher affinity for lipid A of bacte-
rial membranes as compared to polymyxin B.11 In this
manuscript, we propose that ne-tuning of charged head
groups on bile acid amphiphiles modulate their specicity
against mycobacteria. We unraveled that hard-charged amphi-
philes specically kills mycobacteria as the hydrophobic, rigid,
waxy outer membranes of the mycobacteria can allow penetra-
tion of these hydrophobic amphiphiles (Fig. 1). Contrastingly,
so-charged amphiphiles specically kill Gram-positive/Gram-
Antimicrobial peptides (AMPs) induce non-receptor medi-
ated disruptions in the target membranes of microorganisms by
virtue of their amphiphilic nature with a discrete cationic
aLaboratory of Nanotechnology and Chemical Biology, Regional Centre for
Biotechnology, 180 Udyog Vihar, Phase I, Gurgaon-122016, Haryana, India. E-mail:
bajaj@rcb.res.in
bVaccine and Infectious Disease Research Centre, Translational Health Science and
Technology Institute, 496 Udyog Vihar Phase III, Gurgaon-122016, India. E-mail:
ramandeep@thsti.res.in
cManipal University, Manipal, India
† Electronic supplementary information (ESI) available: Fig. S1–S6 and Table S1,
synthesis of amphiphiles. See DOI: 10.1039/c4md00303a
‡ These authors contributed equally.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Fig. 1 Schematic presentation of the study showing the selectivity of hard-charged amphiphiles for mycobacteria, and soft-charged amphi-
philes for Gram-positive and Gram-negative bacteria.
negative bacteria through electrostatic interactions with polar more potent as compared to so-charged amphiphiles; (c) in
bacterial lipids.
general, DMAP-derived amphiphiles (CDCA-DMAP2, DCA-
DMAP2, and CA-DMAP3) were most potent against Mtb inhib-
iting mycobacterial growth in the range of 0.78–6.25 mM; (d) CA
derived amphiphiles followed the order of DMAP > PYR > PIP >
TMA > AMM in terms of their anti-tuberculosis activity; (e)
ammonium head group-derived so-charged amphiphiles
possess no activity against mycobacteria.
Results and discussion
Design and synthesis of amphiphiles
We engineered 20 amphiphiles derived from bile acids with
different charged head groups using four bile acids, lithocholic
acid (LCA), chenodeoxycholic acid (CDCA), deoxycholic acid
(DCA), and cholic acid (CA) (Fig. 2). The polarity of that attached
head groups varied from so-charged ammonium (AMM) to
hard-charged trimethyl ammonium (TMA), N-methylpiperidine
(PIP), pyridine (PYR), and dimethylaminopyridine (DMAP).
Amphiphiles were synthesized by chloroacetylation of the cor-
responding bile acid methyl ester10a followed by quaternization
with the respective tertiary amines (Scheme 1, ESI†), and char-
To unravel the selectivity of amphiphiles, we determined
their activities against Gram-positive (S. aureus) and Gram-
negative (E. coli) bacteria (Table 2). MIC99 determination studies
suggested that: (a) single-charged LCA amphiphiles showed no
growth inhibition up to 512 mM; (b) among multiple-charged
amphiphiles, so-charged amphiphiles (DCA-AMM2 and CA-
AMM3) inhibited 99% growth at 16–64 mM; (c) hard-charged
multiple-headed amphiphiles are less potent than so-charged
amphiphiles; (d) CA-derived hard-charged amphiphiles showed
no growth inhibition even at 512 mM. Interestingly, the so-
charged CA-AMM3 amphiphile was highly specic in its ability
to inhibit growth of E. coli/S. aureus, whereas the hard-charged
CA-DMAP3 amphiphile was highly specic for mycobacterial
species. CA-DMAP3 possesses MIC99 of 1.56–6.25 mM for
mycobacteria, and a selectivity index of ꢀ40–320 over E. coli/S.
aureus. The presence of 50% plasma also maintains its high
potency (MIC99 ¼ 6.25 mM) against mycobacteria.
1
acterized by H-NMR, 13C-NMR and HRMS (ESI†).
Antibacterial activities against different microorganisms
We then evaluated activities of amphiphiles against mycobac-
terial species, M. smegmatis (Msm), M. tuberculosis H37Rv (Mtb),
and M. bovis BCG (Table 1). From structure–activity studies, we
concluded that (a) multiple-charged amphiphiles are more
potent as compared to single-charged amphiphiles; (b) among
multiple-charged amphiphiles, hard-charged amphiphiles are
Fig. 2 Molecular structures of lithocholic acid (LCA)-, chenodeoxycholic acid (CDCA)-, deoxycholic acid (DCA)-, and cholic acid (CA)-based
amphiphiles; different charged head groups were synthesized and studied.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Table 1 Antimycobacterial activities, hemolytic activities and therapeutic index of amphiphiles
MIC99a (mM)
MIC99a (mM)
Amphiphile
BCG
mc2155
H37Rv
MHC50b (mM)
TIc
Amphiphile
BCG
mc2155
H37Rv
MHC50b (mM)
TIc
LCA-AMM1
LCA-TMA1
LCA-PIP1
>50
50
50
>50
50
37.5
4.68
6.25
9.375
3.12
0.78
>50
6.25
6.25
25
12.5
>50
3.12
1.56
3.12
0.78
>50
>50
50
50
>50
50
>50
12.5
12.5
25
>1
>1
>1
>1
>1
>1
>1
>1
>1
>1
<20
>20
>20
<20
>20
<20
>80
>80
>40
>320
DCA-AMM2
DCA-TMA2
DCA-PIP2
DCA-PYR2
DCA-DMAP2
CA-AMM3
CA-TMA3
CA-PIP3
CA-PYR3
CA-DMAP3
Levooxacin
25
25
>50
>50
>50
6.25
12.5
4.68
>50
50
>1
>1
>1
>1
>1
>1
>1
>1
>1
>1
<20
<20
>160
>80
>213
<20
>20
>40
>26
12.5
0.78
1.56
0.78
>50
12.5
6.25
3.12
1.56
0.58
3.12
6.25
2.34
18.75
6.25
6.25
6.25
3.12
0.39
LCA-PYR1
LCA-DMAP1
CDCA-AMM2
CDCA-TMA2
CDCA-PIP2
CDCA-PYR2
CDCA-DMAP2
INH
25
37.5
6.25
0.78
3.12
0.78
>160
d
d
d
d
—
—
—
—
a
b
MIC99 of amphiphiles against three mycobacterial strains. MHC50 is minimum hemolytic conc. at which 50% hemolysis is observed.
c
Therapeutic index for M. tuberculosis H37Rv as ratio of MHC50/MIC99 (M. tuberculosis H37Rv). d Not determined.
Table 2 Antibacterial activities of bile acid amphiphiles against Gram-negative (E. coli) and Gram-positive (S. aureus) bacteria; and toxicities of
bile acid amphiphiles against lung epithelial (A549) and macrophage (THP-1) cell lines
a
a
MIC99 (mM)
IC50b (mM)
MIC99 (mM)
IC50b (mM)
Amphiphile
E. coli
S. aureus
A549
THP-1
Amphiphile
E. coli
S. aureus
A549
THP-1
LCA-AMM1
LCA-TMA1
LCA-PIP1
>512
>512
>512
>512
>512
64
256
128
128
128
>512
>512
>512
>512
>512
128
512
256
256
128
24.6 ꢁ 8.8
69 ꢁ 2.5
42 ꢁ 6.5
68 ꢁ 1.3
112 ꢁ 5.9
14.8 ꢁ 9.5
144 ꢁ 7.8
79 ꢁ 7.1
168 ꢁ 0.1
154 ꢁ 4.8
14.0 ꢁ 0.2
62 ꢁ 2.6
77 ꢁ 6.0
89 ꢁ 5.0
86 ꢁ 1.5
13.7 ꢁ 0.8
142 ꢁ 1.7
164 ꢁ 1.7
>200
DCA-AMM2
DCA-TMA2
DCA-PIP2
DCA-PYR2
DCA-DMAP2
CA-AMM3
CA-TMA3
CA-PIP3
16
64
14.7 ꢁ 5.6
157 ꢁ 2.26
50 ꢁ 8.1
78 ꢁ 5.0
50 ꢁ 9.6
75.9 ꢁ 1.3
>200
>200
>200
>200
—
13.96 ꢁ 0.3
105 ꢁ 7.3
77 ꢁ 8.3
124 ꢁ 8.7
63 ꢁ 5.4
38.2 ꢁ 5.1
80 ꢁ 6.02
>200
256
>512
128
64
256
128
>512
64
LCA-PYR1
LCA-DMAP1
CDCA-AMM2
CDCA-TMA2
CDCA-PIP2
CDCA-PYR2
CDCA-DMAP2
Polymyxin
32
64
>512
>512
>512
>512
—
>512
512
>512
256
—
CA-PYR3
CA-DMAP3
—
>200
>200
—
137 ꢁ 2.1
c
c
0.5
1.0
—
—
a
MIC99 of amphiphiles against Gram-positive (S. aureus) and Gram-negative (E. coli) bacterial strains. b IC50 is minimum conc. at which 50% cell
death is observed against lung epithelial cells (A549) and macrophages (THP-1). c Not determined.
Cytotoxicity studies against mammalian cells
Mechanism of action
Cytotoxicity studies of amphiphiles against mammalian A549 The above studies concluded that so-charged CA-AMM3
(human lung epithelial cell line) and THP-1 (human monocyte selectively inhibits the growth of E. coli/S. aureus, whereas hard-
cell line) cells showed that single-charged LCA amphiphiles are charged CA-DMAP3 is highly selective for mycobacteria. We
the most cytotoxic, and so-charged amphiphiles are more toxic observed that the killing of E. coli/S. aureus by CA-AMM3 was
as compared to hard-charged amphiphiles (Table 2, Fig. S1 and dose-dependent and bactericidal in nature, whereas no growth
S2, ESI†). A highly sensitive propidium-iodide-based cell-cycle inhibition of E. coli/S. aureus was observed on exposure to CA-
analysis also showed no change in the cell-cycle phase of A549 DMAP3 (Fig. 3a and S5, ESI†). Incubation of Msm with CA-
cells on treatment with 100 mM of CA based-amphiphiles DMAP3 inhibited bacterial growth by 10 000–100 000 fold
(Fig. S3, ESI†). Hemolytic studies against chicken and sheep (Fig. 3b), whereas no such growth inhibition was observed on
blood RBCs showed that all the amphiphiles possessed MHC50 exposure to CA-AMM3. Similarly, we observed perturbations in
values greater than 1 mM (Table 1 and Fig. S4, ESI†). Thera- E. coli/S. aureus membranes only on incubation with CA-AMM3
peutic index (MHC50/MIC99) calculations suggested that hard- (Fig. 3c), whereas CA-DMAP3 specically inhibited the ability of
charged CA amphiphiles have a high therapeutic index over Msm to generate proton motive force (Fig. 3d).
so-charged amphiphiles, and DMAP-derived multiple-charged
Next, we performed AFM studies to determine both
amphiphiles are the most potent with the highest therapeutic morphological and topological changes in E. coli and M. bovis
index.
BCG on treatment with CA-AMM3 and CA-DMAP3. Incubation of
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
Fig. 3 (a) Time-dependent killing of E. coli and S. aureus suggesting a bactericidal effect in the presence of CA-AMM3. (b) Time-dependent killing
of Msm showing selective bactericidal activity by CA-DMAP3. (c) Membrane permeabilization of E. coli and S. aureus by propidium iodide
showing selective permeabilization of E. coli and S. aureus by CA-AMM3. (d) Effect of CA-AMM3 and CA-DMAP3 on the membrane potential of
Msm suggests selective activity of CA-DMAP3.
E. coli with CA-AMM3 induced surface indentations, micelle-like Mechanistic studies showed that the presence of EDTA
structures, and leakage of a large amount of cytoplasmic enhances the activity of CA-DMAP3 by 64-fold and by 16-fold
contents (Fig. 4b and Table S1, ESI†). Incubation of M. bovis against E. coli and S. aureus, respectively, whereas only ꢀ2-fold
BCG with CA-DMAP3 led to pore and groove formation with a increase in activity was observed for CA-AMM3. These results
rugged surface (Fig. 4f), whereas the morphology and topology conclude that that presence of LPS and proteoglycans inhibits
of CA-DMAP3-treated E. coli (Fig. 4c) and CA-AMM3 treated M. the interactions of CA-DMAP3 with E. coli/S. aureus bacterial
bovis BCG (Fig. 4e) were similar to their respective untreated membranes, thereby accounting for lack of its activity against
samples (Fig. 4a and d).
them.
To understand the mechanism of this differential specicity,
we next determined the activity of these amphiphiles against
both E. coli and S. aureus in the presence of EDTA, a known
destabilizer of lipopolysaccharide (LPS) and proteoglycans.12
Amphiphile–membrane interactions
We speculated that presence of hydrophobic mycolic lipids like
trehalose dimycolate (TDM)13 might be responsible for the
Fig. 4 (a–f) Atomic force micrographs of E. coli (a–c) and BCG (d–f) on incubation with CA-AMM3 (b and e) and CA-DMAP3 (c and f) amphiphiles
suggesting selective disruptions of E. coli membranes by CA-AMM3 (b) and BCG membranes by CA-DMAP3 (f) respectively.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Fig. 5 (a–c) Surface hydration of model membranes suggesting the dehydrated nature of PE : PG : TDM membranes; and maximum dehy-
dration of PE : PG : LPS and PE : PG membranes on interactions with amphiphiles due to electrostatic interactions. (d–f) Membrane rigidity of
model bacterial membranes showing high rigidity of PE : PG : TDM membranes; and induction of membrane fluidity of PE : PG : TDM
membranes on interactions with CA-DMAP3.
selective killing of mycobacteria by CA-DMAP3. We therefore hydrophobic membranes. We did not observe this enhanced
probed the interactions of CA-AMM3 and CA-DMAP3 with model membrane uidity on incubation of PE : PG and PE : PG : LPS
bacterial membranes to quantify changes in the surface with CA-DMAP3. These observations suggest that CA-DMAP3 has
hydration and rigidity of membranes. Using a Laurdan probe,14 the ability to interact with rigid mycolic lipids and form pore-
we observed that TDM-doped mycobacterial model membranes, like structures in mycobacterial membranes.18
PE : PG : TDM (1 : 1 : 2), are more dehydrated than E. coli
modeled membranes, PE : PG : LPS (5 : 3 : 1) and PE : PG (1 : 1)
(Fig. 5a–c).
Conclusions
Incubation of CA-AMM3 induced a higher amount of dehy-
dration in PE : PG : LPS and PE : PG membranes (Fig. 5a–c) due
to strong electrostatic interactions, thereby accounting for the
selective activity of CA-AMM3 against E. coli/S. aureus. Similarly,
the smaller amount of dehydration of PE : PG : TDM
membranes by CA-AMM3 accounts for poor interactions and the
inability of CA-AMM3 to kill mycobacteria. We did not observe
any dehydration of PE : PG : TDM membranes upon incubation
with CA-DMAP3 (Fig. 5c), thereby suggesting the presence of
In summary, we demonstrated that ne-tuning charged head
groups on a bile acid scaffold modulates their specicity against
bacteria. So-charged primary amine amphiphiles interact with
Gram-positive and Gram-negative bacterial membranes,
whereas hard-charged head groups provide effective and selec-
tive interactions with hydrophobic mycobacterial membranes.
In-depth mechanistic studies revealed that this specicity in
mechanism of action for these amphiphiles was due to molec-
ular differences in the cell wall architecture of mycobacterial
and Gram-positive/Gram-negative bacteria. The present study
will help us in understanding the molecular basis of specic
amphiphile–membrane interactions and in the design of more
potent 2nd-generation amphiphiles that are highly specic for a
particular bacterial species, which might be useful to combat
the problem of drug resistance.
hydrophobic
interactions
between
CA-DMAP3
and
PE : PG : TDM.
Membrane uidity studies using a DPH probe15 suggested
that mycobacterial modeled membranes are more rigid in
comparison to other membranes (Fig. 5d–f). Incubation of
PE : PG : LPS and PE : PG with CA-AMM3 does not induce any
uidity, suggesting poor interactions of CA-AMM3 with the
hydrophobic regions of membranes. Therefore the observed
dehydration without any alteration in the rigidity of
PE : PG : LPS and PE : PG conrms a carpet-like mechanism for
the activity of CA-AMM3 against E. coli/S. aureus.16 Contrast-
ingly, the presence of hydrophobic, rigid mycolic lipids like
TDM prevents interactions of CA-AMM3 with mycobacterial
membranes. Interestingly, CA-DMAP3 increases the uidity of
mycobacterial (PE : PG : TDM) membranes as observed in case
of alamethicin17 by virtue of the insertion of CA-DMAP3 in
Experimental section
Materials and methods
Bile acids were purchased from Sigma-Aldrich. CCCP, INH, and
DiCO2 were purchased from Invitrogen corporation. All the
synthesized compounds were puried using Combi-ash chro-
1
matography with a 230–400 mesh size silica gel. H NMR and
13C NMR spectra were recorded using a Brucker 400 MHz
spectrometer. Chemical shis (d) are reported in ppm with
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
tetramethylsilane as the internal standard. High-resolution
mass spectra were measured on an AB-SCIEX-5600 mass spec-
trometer. E. coli (MTCC 443) and S. aureus (MTCC 737) were
purchased from MTCC. The A549 cell line was purchased from
Sigma Aldrich and THP-1 cells were a kind gi from NCCS,
Pune. M. smegmatis and M. tuberculosis H37Rv were a kind gi
from Dr Clion E. Barry, NIH, and M. bovis BCG was kind gi
from Professor Anil K. Tyagi (Department of Biochemistry,
UDSC, India).
Cell cycle analysis
Cell cycle analyses against A549 cells for bile acid amphiphiles
were performed according to published protocol.21
Hemolytic assay22
Hemolytic assays were performed using chicken and sheep
blood. For RBC isolation, blood was centrifuged at 4000 rpm for
5 min. For hemolytic assays, 195 mL of erythrocyte suspension
(5% or 20% RBC) were added per well in a 96-well plate and
incubated with facial amphiphiles at the desired concentra-
tions. Aer incubation for 1 h, the plate was centrifuged at
1200g for 15 min. The supernatant was diluted 1 : 100 in 1ꢄ PBS
and absorption at 413 nm was measured. All our assay plates
included positive (1% PBST), buffer (1ꢄ PBS) and solvent
(MeOH) controls. The percentage of hemolysis was determined
from {(A ꢂ A0)/(Atotal ꢂ A0)} ꢄ 100, where A is the absorbance of
the test well, A0 is the absorbance of the negative controls, and
Atotal is the absorbance of 100% hemolysis wells at 413 nm. All
the experiments were performed at least twice in duplicates.
Antibacterial activity
We determined the antibacterial activity of bile acid amphi-
philes using a slight modication of a method mentioned by
Hancock et al.19 The bacterial strains were grown for 6 h and
diluted to 105 cfu mLꢂ1. 150 mL of this bacterial suspension was
added to a 96-well plate containing the required concentration
of amphiphiles. The plate was then incubated at 37 ꢃC with
continuous shaking for 12 h, and OD600 nm was measured
using a molecular probe M5 microplate reader. MIC99 values
were determined by taking the average of triplicate values for
each concentration and were performed in duplicate. Polymyxin
was used as the positive control in our assays. For experiments
with EDTA, both the bacterial cultures were grown in nutrient
broth containing 10 mM EDTA and MIC99 was determined as
described above in the presence of 10 mM EDTA. The anti-
mycobacterial activity of these amphiphiles was determined
using an inverted plate reader method as described previously.20
The plates were incubated at 37 ꢃC and MIC99 values were read
microscopically using an inverted plate reader aer 14 days for
M. bovis BCG, Mtb, and aer 2 days for Msm. Each reading was
made three independent times. Standard drugs such as isoni-
azid and levooxacin were used as positive controls in our
assays. For killing curves early-log phase cultures were incu-
bated with CA-AMM3 and CA-DMAP3 at designated time points.
Bacterial enumeration was performed by plating 10-fold serial
dilutions on MB7H10 plates, and the plates were incubated at
Membrane permeabilization studies23
E. coli and S. aureus cells were grown till mid log phase. Cells
were harvested by centrifugation, washed with 1ꢄ PBS and re-
suspended in 1ꢄ PBS. These washed E. coli and S. aureus bacilli
suspensions were pre-incubated with amphiphiles at
a
concentration of 80 mM or 160 mM, respectively, followed by the
addition of 15 mM propidium iodide (PI). The uptake of PI was
measured by the increase in uorescence of PI for 10 min as a
measure of membrane permeabilization using an excitation
wavelength of 535 nm and an emission wavelength of 617 nm.
Membrane potential studies24
Msm was grown till an OD600 nm value of 1.0 and bacteria were
pre-incubated with amphiphiles or INH or CCCP. Bacterial cells
were immediately exposed to 15 mM of DiCO2 (3,30-diethylox-
acarbocyanine iodide) at room temperature. Aer 30 minutes of
labeling, cells were washed twice with 1ꢄ PBS, and green uo-
rescence (Ex480 nm/Em530 nm) and red uorescence (Ex488 nm/
Em610 nm) were measured in a 96-well plate reader (Biotek
synergy Hr). Cells treated with no drug were kept as the control
for background uorescence. Membrane potential was calcu-
lated as the ratio of red uorescence to green uorescence using
Biotek Synergy Hit and Gene5 soware.
ꢃ
37 C.
Cytotoxicity assay21
The THP-1 cell line was maintained in RPMI media as per
standard protocol and differentiated in macrophages by the
overnight addition of 50 ng mLꢂ1 of PMA (phorbol 12-myristate
13-acetate). For the cytotoxicity assay, 5 ꢄ 103 cells were seeded
per well in a 96-well plate. Aer 24 h, cells were overlayed with
Atomic force microscopy studies (sample preparation and
imaging)25
medium containing various concentrations of amphiphiles. Poly-L-lysine coated (1.0 mm uniform thickness) microscopic
Aer overlaying macrophages for 48 h, 20 mL of 5 mg mLꢂ1 MTT slides were obtained from Polysciences Inc. All the AFM-
{3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide} imaging experiments were done on a JPK NanoWizard® AFM
was added to each well. Aer incubation for 4 h, cells were lysed head using an AC air mode cantilever. The pyramidal tip
by the addition of 200 mL of a 1 : 1 mixture of DMSO and MeOH cantilever used was purchased from ACTA made of silicon with
to dissolve the formazen crystals and the absorption at 540 nm a spring constant of 40 N mꢂ1. All the data was processed using
was measured. Cell viability was calculated using the equation JPKSPM data-processing soware. For AFM studies bacteria
[{A540(treated cells) ꢂ background}/{A540(untreated cells) ꢂ were grown till an OD600 nm value of 1.0, harvested, washed
background}] ꢄ 100.
twice with 1ꢄ PBS and subsequently exposed to CA-AMM3 or
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
CA-DMAP3 in a 96-well plate. Aer incubation for 24 h, samples
were applied on poly-L-lysine coated slides and dried using a
slow stream of nitrogen gas. Imaging was done using the AC air/
tapping mode with 15 mm z-scale size. All images were obtained
with a scan speed of 0.5 Hz and a resolution of 1024 ꢄ 1024
pixels. The height, width, and 3D-toplogical information was
acquired and processed with JPKSPM data-processing
soware.
2011, 2, 378; (d) J. D. Guzman, P. N. Mortazavi, T. Munshi,
D. Evangelopoulos, T. D. McHugh, S. Gibbons,
J. Malkinson and S. Bhakta, MedChemComm, 2014, 5, 47.
4 R. E. W. Hancock and H. Sahl, Nat. Biotechnol., 2006, 24,
1551.
5 (a) H. D. Thaker, A. Cankaya, R. W. Scott and G. N. Tew, ACS
Med. Chem. Lett., 2013, 4, 481; (b) C. W. T. Chang and
J. Y. Takemoto, MedChemComm, 2014, 5, 1048; (c)
I. M. Herzog and M. Fridman, MedChemComm, 2014, 5,
1014; (d) S. Yadav, S. Joshi, M. A. Qadar Parsha and
S. Pasha, MedChemComm, 2013, 4, 874; (e) I. M. Herzog,
M. Feldman, A. Eldar-Boock, R. Satchi-Fainaro and
M. Fridman, MedChemComm, 2013, 4, 120; (f)
D. S. S. M. Uppu, P. Akkapeddi, G. B. Manjunath,
V. Yarlagadda, J. Hoque and J. Haldar, Chem. Commun.,
2013, 49, 9389.
6 (a) R. D. Jahnsen, A. Sandberg-Schaal, K. J. Vissing,
J. O. Neill, I. Chopra and R. E. Lee, Nat. Rev. Microbiol.,
2011, 9, 62; (b) E. W. Sugandhi, R. V. Macri, A. A. Williams,
B. L. Kite, C. Slebodnick, J. O. Falkinham, A. R. Esker and
R. D. Gandour, J. Med. Chem., 2007, 50, 1645; (c) R. Kapoor,
P. R. Eimermann, J. W. Hardy, J. D. Cirillo, C. H. Contag
and A. E. Barron, Antimicrob. Agents Chemother., 2011, 55,
3058.
7 Y. Hu, M. N. Amin, S. Padhee, R. E. Wang, Q. Qiao, G. Bai,
Y. Li, A. Mathew, C. Cao and J. Cai, ACS Med. Chem. Lett.,
2012, 3, 683.
8 J. Haldar, P. Kondaiah and S. Bhattacharya, J. Med. Chem.,
2005, 48, 3823.
9 E. W. Sugandhi, R. V. Macri, A. A. Williams, B. L. Kite,
C. Slebodnick, J. O. Falkinham, A. R. Esker and
R. D. Gandour, J. Med. Chem., 2007, 50, 1645.
10 (a) M. Singh, A. Singh, S. Kundu, S. Bansal and A. Bajaj,
Biochim. Biophys. Acta, Biomembr., 2013, 1828, 1926; (b)
L. Huang, Y. Sun, H. Zhu, Y. Zhang, J. Xu and Y. Shen,
Steroids, 2009, 74, 701; (c) A. Gioiello, F. Venturoni,
S. Tamimi, C. Custodi, R. Pellicciari and A. Macchiarulo,
MedChemComm, 2014, 5, 750.
Amphiphile–membrane interactions
Membrane vesicles with the desired lipid ratios were prepared
as described previously.26 Changes in the surface hydration of
vesicles were studied aer incubation of the vesicles with 10
weight percent of facial amphiphiles at 25 ꢃC for 4 h. We
recorded the generalized polarization of Laurdan in a 96-well
plate in a Molecular Devices M5 instrument. Fluorescence of
Laurdan was recorded using an excitation wavelength of 350
nm and emission wavelengths of 440 nm and 490 nm. The
generalized polarization (GP) was calculated using the equa-
tion GP ¼ (I440 ꢂ I490)/(I440 + I490). Similarly, we measured
changes in steady-state anisotropy of DPH in a 96-well plate
using lex at 350 nm and lem of 45ꢃ2 nm aer incubation of these
vesicles with amphiphiles at 25 C in a Molecular Devices M5
instrument.
Acknowledgements
AB conceived the idea. AB and RS designed the experiments.
MS, PB, AS synthesized the amphiphile molecules. SB per-
formed all the experiments with E. coli and S. aureus. SB, SK, RS
performed experiments with mycobacteria. MS performed the
Laurdan and DPH-based experiments with model membranes.
VS performed AFM experiments. SB, MS, VS, RS, and AB
analyzed the data. AB and RS wrote the manuscript. AB and RS
supervised the overall research. The authors thank RCB,
THSTI, DST and DBT for funding. SB thanks DBT for a
Research Fellowship. VS thanks RCB for a Research Fellowship.
RS and AB thank DBT and DST for their Ramalingaswamy and
Ramanujan Fellowships, respectively. The authors are grateful
to Prof. Anil K. Tyagi, Delhi University, for access to BSL3
facility.
11 X. Lai, Y. Feng, J. Pollard, J. N. Chin, M. J. Bybak, R. Bucki,
R. F. Epand, R. M. Epand and P. B. Savage, Acc. Chem. Res.,
2008, 41, 1233.
12 H. L. Alakomi, M. Sareela and I. M. Helande, Microbiology,
2003, 149, 2015.
13 C. W. Harland, D. Rabuka, C. R. Bertozzi and
R. Parthasarathy, Biophys. J., 2008, 94, 4718.
Notes and references
1 D. Young, M. Perkins, K. Duncan and C. E. Barry 3rd, J. Clin. 14 S. Bhattacharya and A. Bajaj, J. Phys. Chem. B, 2007, 111,
Invest., 2008, 118, 1255. 2463.
2 (a) A. E. Clatworthy, E. Pierson and D. T. Hung, Nat. Chem. 15 S. Bhattacharya and A. Bajaj, Langmuir, 2007, 23, 8988.
Biol., 2007, 3, 541; (b) J. N. Burrows, R. L. Elliot, T. Kaneko, 16 M. Zasloff, Nature, 2002, 415, 389.
´
C. E. Mowbray and D. Waterson, MedChemComm, 2014, 5, 17 H. Duclohier and H. Wroblewski, J. Membr. Biol., 2001, 184, 1.
688.
18 K. A. Brogden, Nat. Rev. Microbiol., 2005, 3, 238.
3 (a) C. Dye, Nat. Rev. Microbiol., 2009, 7, 81; (b) 19 I. Wiegand, K. Hilpert and R. E. W. Hancock, Nat. Protoc.,
G. A. I. Moustafa, S. Nojima, Y. Yamano, A. Aono, M. Arai, 2008, 3, 163.
S. Mitarai, T. Tanaka and T. Yoshimitsu, MedChemComm, 20 R. Singh, M. Singh, G. Arora, S. Kumar, P. Tiwari and
2014, 5, 720; (c) R. P. Tripathi, J. Pandey, V. Kukshal,
A. Ajay, M. Mishra, D. Dube, D. Chopra, R. Dwivedi,
V. Chaturvedi and R. Ramachandran, MedChemComm,
S. Kidwai, J. Bacteriol., 2013, 195, 2839.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun.

View Article Online
MedChemComm
Concise Article
21 V. Sreekanth, S. Bansal, R. K. Motiani, S. Kundu, 24 M. Berney, F. Hammes, F. Bosshard, H. U. Weilemann and
S. K. Muppu, T. D. Majumdar, K. Panjamurthy, T. Egli, Appl. Environ. Microbiol., 2007, 73, 3283.
S. Sengupta and A. Bajaj, Bioconjugate Chem., 2013, 24, 1468. 25 M. E. Nunez, M. O. Martin, P. H. Chan, L. K. Duong,
22 N. P. Chongsiriwatana, J. A. Patch, A. M. Czyzewski,
M. T. Dohm, A. Ivankin, D. Gidalevitz, R. N. Zuckermann
A. R. Sindhurakar and E. M. Spain, Methods Enzymol.,
2005, 397, 256.
and A. E. Barron, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 26 P. Bhargava, M. Singh, V. Sreekanth and A. Bajaj, J. Phys.
2794.
Chem. B, 2014, 118, 9341.
23 J. Y. Je and S. K. J. Kim, J. Agric. Food Chem., 2006, 54, 6629.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2014