Designing Simple Lipidated Lysines: Bifurcation Imparts Selective Antibacterial Activity

Article abstract of DOI:10.1002/cmdc.201600400

In the global effort to thwart antimicrobial resistance, lipopeptides are an important class of antimicrobial agents, especially against Gram-negative infections. In an attempt to circumvent their synthetic complexities, we designed simple membrane-active agents involving only one amino acid and two lipid tails. Herein we show that the use of two short lipid tails instead of a single long one significantly increases selective antibacterial activity. This study yielded several selective antibacterial compounds, and investigations into the properties of this compound class were conducted with the most active compound. Fluorescence spectroscopic studies revealed the capacity of the representative compound to cause depolarization and permeabilization of bacterial cell membranes. This membrane-active nature of the compound imparts superior activity against persister cells, biofilms, and planktonic cells. Topical application of the compound decreased bacterial burden in mice inflicted with burn-infections caused by Acinetobacter baumannii. We anticipate that the design principles described herein will direct the development of several antimicrobial agents of clinical importance.

Full text of DOI:10.1002/cmdc.201600400

DOI: 10.1002/cmdc.201600400
Communications
Designing Simple Lipidated Lysines: Bifurcation Imparts
Selective Antibacterial Activity
Chandradhish Ghosh, Mohini Mohan Konai, Paramita Sarkar, Sandip Samaddar, and
[a]
Jayanta Haldar*
In the global effort to thwart antimicrobial resistance, lipopep-
tides are an important class of antimicrobial agents, especially
against Gram-negative infections. In an attempt to circumvent
their synthetic complexities, we designed simple membrane-
active agents involving only one amino acid and two lipid tails.
Herein we show that the use of two short lipid tails instead of
a single long one significantly increases selective antibacterial
activity. This study yielded several selective antibacterial com-
pounds, and investigations into the properties of this com-
pound class were conducted with the most active compound.
Fluorescence spectroscopic studies revealed the capacity of
the representative compound to cause depolarization and per-
meabilization of bacterial cell membranes. This membrane-
active nature of the compound imparts superior activity
against persister cells, biofilms, and planktonic cells. Topical ap-
plication of the compound decreased bacterial burden in mice
inflicted with burn-infections caused by Acinetobacter bauman-
nii. We anticipate that the design principles described herein
will direct the development of several antimicrobial agents of
clinical importance.
tive activity against bacteria relative to that of the parent anti-
[6]
biotics. Herein we describe the development of simple lipi-
dated lysine-based membrane-active molecules and one possi-
ble approach toward achieving selective antibacterial activity.
Furthermore, the mechanism of antibacterial action, the ability
to stall resistance development in bacteria, along with anti-bio-
film properties of a representative compound were studied. Fi-
nally, the activity of the compound was also validated in
a murine model of Acinetobacter baumannii burn infection.
We envisioned that the structure of a simple lysine-based
membrane-active compound would be an alkyl chain conju-
gated to it. To have any significant activity, the alkyl chain
must be sufficiently long, which will facilitate interactions with
the bacterial membrane. In the first example, we chose to
couple dodecylamine to lysine using simple HBTU coupling
chemistry to obtain C K (1). Gradual increase in alkyl chain
12
length yielded compounds with tetradecyl (C K, 2), hexadecyl
14
(C K, 3), octadecyl (C K, 4) and eicosyl (C K, 5) chains. The ac-
16
18
20
ꢀ
tivity of the compounds (in their CF COO salt forms) were
3
evaluated against Staphylococcus aureus, Enterococcus faecium,
Escherichia coli, and A. baumannii (Table 1). The first three com-
pounds were found to possess good antibacterial activity.
However, in moving to C K and C K, the activity of the com-
The current predicament of antimicrobial resistance has not
only put world health at risk, but is also expected to severely
18
20
pounds was lost. Evaluation of their toxicity against erythro-
cytes, as indicated by their HC50 values (concentration at which
50% of erythrocytes are lysed), revealed that all the active
compounds possess substantial toxicity.
[
1]
affect the global economy. Natural antimicrobial peptides are
[2]
an inspiration for the design of next-generation drugs. The
advent of synthetic membrane-active agents has bolstered the
[
3]
antimicrobial pipeline with a new class of drugs. Although
much remains to be achieved in terms of clinical success, their
importance as future antibiotics cannot be overstated. Lipidat-
ed peptidomimetics have gained popularity as a successful
Table 1. Antibacterial and hemolytic activity of the compounds.
[a]
Compound
MIC [mm]
HC50
[4]
[
b]
[c]
[d]
[e]
method for producing broad-spectrum antibacterial agents.
We have designed several membrane-active agents in the
SA
EF
EC
AB
C
12-K (1)
23
11
5
>50
>50
23
22
5
46
44
21
>50
>50
46
22
10.5
10
6
23
11
5
>50
>50
23
22
5
23
11
10.5
>50
>50
>50
22
10.5
10
6
120
105
73
114
185
390
172
110
108
88
[
5]
past. Lipidation of antibiotics in current use has also been re-
C -K (2)
1
4
[
6]
ported. Although there is no doubt about the efficacy of
such designs, there is an argument for making a simpler and
more cost-effective lipopeptide. More importantly, lipidation of
antibiotics leads to compounds that generally lose their selec-
C
C
C
C
16-K (3)
18-K (4)
20-K (5)
6
6
-K-C (6)
C -K-C (7)
6
8
8
8
C
C
C
-K-C
8
(8)
-K-C10 (9)
5
6
5
6
[
a] C. Ghosh, M. M. Konai, P. Sarkar, Dr. S. Samaddar, Prof. Dr. J. Haldar
Chemical Biology and Medicinal Chemistry Laboratory, New Chemistry Unit,
Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Jakkur,
Bengaluru 560064, Karnataka (India)
10-K-C10 (10)
colistin
vancomycin
CTAB
25
0.5
8.5
N.D.
0.5
8.5
0.4
>100
8.5
0.4
>100
8.5
N.D.
N.D.
82
Fax: (+91)80-2208-2627
E-mail: jayanta@jncasr.ac.in
[
a] Values are the average of at least two experiments, each done in tripli-
cate (error <5%); N.D.: not determined. [b] S. aureus (MTCC 737).
c] E. faecium (ATCC 19634). [d] E. coli (MTCC 443). [e] A. baumannii (MTCC
37).
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cmdc.201600400: synthesis and characterization of
compounds and description of experimental protocols.
[
7
ChemMedChem 2016, 11, 1 – 6
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
Although it is well accepted that optimum amphipathicity is
necessary for potent antimicrobial activity, increase in alkyl
The activities of bifurcated compounds 6–10 were then
tested against S. aureus, E. faecium, E. coli, A. baumannii, and
human erythrocytes. Colistin and vancomycin were used as ref-
erence drugs for assays against Gram-negative and Gram-posi-
tive bacteria, respectively (Table 1). The classical surfactant ce-
trimonium bromide (CTAB) was also tested against the bacte-
ria. Compound 6, which is the bifurcated analogue of 1, re-
tained activity against S. aureus (MIC: 23 mm), E. faecium (MIC:
46 mm), and E. coli (MIC: 23 mm), but lost activity against A. bau-
mannii (MIC>50 mm). However, with an HC50 value of 390 mm,
it was significantly less toxic toward erythrocytes than 1 (HC :
120 mm). Compound 7, the asymmetric bifurcated analogue of
2, consists of an octyl chain and a hexyl chain. With the excep-
tion of E. faecium, the parent compound 2 was active at 11 mm.
The bifurcated compound 7 displayed two-fold lower activity
(MIC: 22 mm) against S. aureus, E. coli, and A. baumannii, but it
was more active than 2 against E. faecium. The HC50 value of
compound 7 was 172 mm, while that of compound 2 was
105 mm. Like its single-chain counterpart 3, compound 8 dis-
played good activity against all bacteria (MIC range: 5–
10.5 mm), but was more active against E. faecium and relatively
less toxic. Compound 4 was found to be inactive against all
the bacteria tested up to 50 mm. In comparison, 9 turned out
to be a very active compound in this series. It displayed MIC
values of 5 mm against both S. aureus and E. coli, while against
E. faecium and A. baumannii the MIC value was 10.5 mm. Com-
pound 10 was as active as 9 (MIC: 6 mm) against S. aureus,
E. faecium, and E. coli, but slightly more active against A. bau-
mannii. Although the single-chain analogue 5 was less toxic
[7]
chain length also leads to a compromise in selectivity. We hy-
pothesized that by decreasing the chain length but not com-
promising much on overall amphiphilicity, further selectivity
could be achieved. One simple approach to do so was to bifur-
cate the alkyl chain, for example, by splitting the hexadecyl
long chain to two octyl chains. With this in mind, we designed
a set of new molecules containing two short alkyl chains in-
stead of one long chain.
We symmetrically and asymmetrically varied the length of
the two chains from hexyl to decyl. We deliberately avoided
using chains with an odd number of methylene units to obtain
asymmetry in the designs. The synthetic strategy is outlined in
Scheme 1. In the first step of synthesis, alkanals were first re-
acted with alkylamines in dry methanol and then reduced by
sodium borohydride to obtain dialkyl amines. These dialkyl
amines were then coupled to Boc-Lys(Boc)-OH in a chloroform/
N,N-dimethylformamide mixture using 2-(1H-benzotriazol-1-yl)-
5
0
1,1,3,3-tetramethyluronium
hexafluorophosphate
(HBTU)
chemistry. In compounds 1–5, single long-chain amines were
directly coupled to Boc-Lys(Boc)-OH. After purifying the com-
pounds by column chromatography, the Boc groups were sub-
sequently deprotected with 50% trifluoroacetic acid in di-
chloromethane to yield the final compounds. They were puri-
fied by HPLC to >95% purity and were subsequently charac-
1
13
terized by H NMR, C NMR, IR and HRMS.
(
HC : 185 mm), it was inactive against all bacteria. The classical
50
surfactant CTAB was active at 8.5 mm against all the bacteria
tested. Because selective antibacterial activity is the most de-
sired condition for membrane-active agents, it could be envi-
sioned from this study that bifurcation of long chains into any
membrane-active antibiotics containing such a feature should
substantially increase the selectivity of the resulting com-
pound.
To determine the potential of such compounds as antibacte-
rial agents, we chose to carry out further studies with com-
pound 9. The minimum bactericidal concentration (MBC) of 9
was determined to be 2ꢁMIC. The compound was also ob-
served to be nontoxic toward human embryonic kidney (HEK)
cells at its MBC, which emphasizes the selectivity of the com-
pound at bactericidal concentrations (Figure 1). The kinetics of
the bactericidal action of compound 9 were then studied.
Within minutes of treatment at concentrations only three-fold
its MIC value, 9 was able to lyse cells of both Gram-negative
(
Figure 2A) and Gram-positive bacteria (Supporting Informa-
tion Figure S1). The capacity of the compound to lyse bacterial
cells could be attributed to its membrane-active nature.
Because it is widely known that bacteria have difficulty in
developing resistance against membrane-active agents, the
ability of the compound to induce resistance development
was studied (Figure 2B). Serial passage of bacterial culture
Scheme 1. Reagents and conditions: a) NaN
MeOH, 808C, 24 h, 2. HCl; c) Boc-Lys(Boc)-OH, HBTU, DIPEA, DMF/CHCl
2:1), RT, 24 h; d) CF COOH (50%), CH Cl , RT, 6 h; e) 1. R’NH /MeOH, RT, 8 h,
. NaBH
, 08C!RT, 12 h, 3. HCl.
3 3
, MeOH, RT, 24 h; b) 1. Ph P/
(
both S. aureus and E. coli) at sub-MIC concentrations of com-
3
pound 9 yielded no resistant mutants. In comparison, norfloxa-
cin (comparator drug for S. aureus) and colistin (comparator
(
2
3
2
2
2
4
&
ChemMedChem 2016, 11, 1 – 6
www.chemmedchem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
drug for E. coli), yielded resistant mutants within six passages,
and in fourteen passages a 400-fold increase in MIC was ob-
served for norfloxacin and an 800-fold increase was observed
for colistin.
The membrane-disruptive properties of the compound was
demonstrated by performing fluorescence studies with E. coli
cells. Experiments with the membrane-potential-sensitive dye
DiSC3(5) showed that 9 at 3ꢁMIC is able to dissipate the po-
tential of the E. coli membrane (Figure 2C). Studies with N-phe-
nylnaphthylamine (NPN) showed that compromise of the
E. coli outer membrane was achieved within minutes of treat-
ment with 9 (Figure 2D). Treatment with 9 also allowed the
entry of propidium iodide (PI) into E. coli cells, as evident by
the increase in dye fluorescence intensity upon compound
treatment (Figure 2E).
Next, to gain insight into the potential areas of application
of compound 9, we chose to study its antibacterial efficacy at
various pH values (pH 5.5–8.5) and salinity (1–3%). Compound
Figure 1. Toxicity studies: fluorescence microscopy images of HEK cells after
treatment with or without Triton X and compound 9 for 24 h. Staining was
done with calcein AM and propidium iodide (PI). (A–C) Untreated cells (nega-
tive control); (D–F) cells treated with 0.1% Triton X (positive control);
9
was found to retain its activity against both S. aureus and
E. coli under different physiological conditions (Supporting In-
formation Table S1). At various percentages of salinity, 9 main-
(G–I) cells treated with compound 9 at MBC (2ꢁMIC)
ꢀ
1
Figure 2. A) Bactericidal time–kill kinetics of compound 9 (concentrations at 3ꢁ and 6ꢁMIC) against E. coli (*: <50 CFUmL ). B) Ability to withstand resist-
ance development (SA: S. aureus, EC: E. coli). The ability of compound 9 to act on E. coli membrane: C) membrane depolarization and D) outer and E) inner
membrane permeabilization. SYTO-9 staining of A. baumannii biofilms: F) untreated control, and treatment with compound 9 at G) 5ꢁMIC and H) 10ꢁMIC.
I) Decrease in bacterial viability within the biofilm. J) Decrease in biofilm mass as determined by crystal violet staining. K) In vivo activity against A. baumannii
ꢀ
1
ꢀ1
burn infection: the concentration of 9 was 40 mgkg and that of colistin was 5 mgkg . Statistical analysis was performed using Student’s t-test. Differences
are considered statistically significant from the untreated group with a value of p<0.05 with 95% confidence intervals (***p<0.001).
ChemMedChem 2016, 11, 1 – 6
www.chemmedchem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
tained its MIC value of 5 mm against both S. aureus and E. coli.
At pH 5.5, 9 was found to be active at 10 mm against S. aureus
and at 5 mm against E. coli. No variation in MIC was observed
against either bacteria at any other pH value. We also tested
the activity of the compound in the presence of 4% bovine
serum albumin (BSA). Compound 9 was active at 10 mm
against both S. aureus and E. coli in the presence of BSA. The
stability of the compound at such broad ranges of conditions
suggests that this class of compounds is suited for use in treat-
ing bacterial infections at various regions in the body.
dose). The mice were observed for irritation, tremors, convul-
sions, salivation, or diarrhea for 14 days. Although the mice
showed no sign of irritation or unnatural behavior, visual ob-
servation of the skin showed development of corrosion on the
first and second days. However, the mice appeared to heal,
and normal generation of fur was observed henceforth (within
a week), proving that 9 possesses little dermal toxicity. To dem-
onstrate its antibacterial efficacy, burn wounds of mice inflicted
with A. baumannii were continuously treated for six days at
ꢀ
1
a concentration of 40 mgkg (no toxicity was observed at this
concentration). Subsequently, the wounded portions were sev-
ered, homogenized and plated. Figure 2K shows that treat-
ment with compound 9 decreased the bacterial burden by
2 log units relative to the untreated case in six days (p value:
0.0003). Colistin, which was used as a positive control, com-
Given the promising antibacterial efficacy displayed by com-
pound 9 and its capacity to infiltrate bacterial membranes, the
potential of 9 to treat metabolically inactive cells and biofilms
was tested. Persister cells, which are known to down-regulate
their metabolism to evade antibiotic stress, are often associat-
ed with chronic infection. Targeting their cell membrane is an
ꢀ
1
pletely cleared the bacterial burden even at 5 mgkg . Al-
though the efficacy of the compounds in this study falls short
of colistin, it should be kept in mind that colistin is prone to
triggering resistance development in bacteria, unlike 9 (Fig-
ure 2B). The rapid development of resistance against colistin
could be attributed to its lipopolysaccharide-mediated mecha-
[
8]
efficient way of lysing such cells. Owing to the inability of
conventional antibiotics to treat such cells, a compound that
can lyse persister cells would be medically very important. Per-
sister cells of S. aureus and E. coli (the number of bacteria were
5
5
log CFU in each case) succumbed to lysis (limit of detection:
ꢀ
1
[9]
0 CFUmL ) when treated with 9 at 5ꢁMIC, whereas in the
nism of action. Moreover, colistin is used only against Gram-
control no change in number of cells were observed in either
case. Upon performing the same spectroscopic experiments
with E. coli persister cells, it was shown that 9 is able to carry
out its lytic function by depolarization and disruption of the
membranes of persister cells (Figure S2).
negative bacteria, but 9 is active against both Gram-positive
and Gram-negative bacteria.
In summary, this study describes the development of simple
membrane-active agents using two long chains and one
amino acid (l-lysine). An important conclusion of the study is
that the use of two short chains instead of a single long chain
can effectively increase the selective antibacterial activity of
the resulting compounds. Through the study, one potent
membrane antimicrobial agent was discovered which acts on
planktonic cells, persister cells, and bacterial biofilms. Active in
murine models of infection, the compound retains activity
under various physiological conditions and does not select re-
sistant mutants rapidly. Although many more studies need to
be done, initial data reveal the potential of this compound
class for the treatment of topical Gram-negative infections.
The ability of compound 9 to disrupt preformed biofilms of
S. aureus and E. coli was then determined. Treatment of 24-
hour-old preformed biofilms of S. aureus and 72-hour-old pre-
formed biofilms of E. coli at 10ꢁMIC were completely disrupt-
ed relative to biofilms left untreated (Figure S3). Because many
antibiotics are intrinsically ineffective against Gram-negative
pathogens, treatment of Gram-negative biofilms is extremely
difficult. A. baumannii is a notorious example of an aggressive
biofilm-forming bacteria. The capacity of 9 to obliterate pre-
formed biofilms of a clinical isolate of A. baumannii (R674) was
studied in detail (Figure 2F–J). The MIC value of 9 and colistin
against this stain was 10 and 0.7 mm, respectively. Staining with
SYTO-9 revealed that the thickness of the biofilms in the un-
treated case was 9.6 mm (Figure 2F). Compound 9 was effec-
tive in disrupting biofilms at 5ꢁMIC, leaving behind a monolay-
er of cells (thickness: 2.4 mm), whereas at 10ꢁMIC, almost no
bacterial cells were observed (Figure 2G–H). Furthermore,
treatment with compound 9 (5ꢁMIC) also brought down the
bacterial burden embedded within the biofilm by 3 log units,
while a ~4 log decrease was observed upon treatment with 9
at 10ꢁMIC (Figure 2I). Apart from that, decrease in biofilm
mass was also observed upon treatment with compound 9, as
interpreted from the optical density (OD) values of the crystal
violet (CV)-stained biofilms left after or without treatment with
compound (Figure 2J).
Acknowledgements
We thank Prof. C. N. R. Rao, F.R.S. (JNCASR) for his constant sup-
port and encouragement.
Keywords: antibiotics · antimicrobial resistance · biofilms ·
drug design · peptidomimetics
[
1] a) Antimicrobial Resistance: Global Report on Surveillance, World Health
Organization, 2014: www.who.int/drugresistance/documents/surveillan-
cereport/en/; b) S. Gandra, D. M. Barter, R. Laxminarayan, Clin. Microbiol.
Infect. 2014, 20, 973–980.
[
2] a) C. Ghosh, J. Haldar, ChemMedChem 2015, 10, 1606–1624; b) S. Rotem,
A. Mor, Biochim. Biophys. Acta Biomembranes 2009, 1788, 1582–1592;
c) M. Zasloff, Nature 2002, 415, 389–395; d) R. E. W. Hancock, H. G. Sahl,
Nat. Biotechnol. 2006, 24, 1551–1557; e) B. Findlay, G. G. Zhanel, F.
Schweizer, Antimicrob. Agents Chemother. 2010, 54, 4049–4058.
3] a) B. Mensa, G. L. Howell, R. Scott, W. F. DeGrado, Antimicrob. Agents Che-
mother. 2014, 58, 5136–5145; b) J. Isaksson, B. O. Brandsdal, M. Engqvist,
G. E. Flaten, J. S. M. Svendsen, W. Stensen, J. Med. Chem. 2011, 54, 5786–
Finally, to validate the antimicrobial potency of compound
9
, we tested its efficacy in a murine model of infection. Prior to
that, acute dermal toxicity of the compound was done. The
[
2
skin on the back of mice were shaved (~2 cm ), and the
ꢀ
1
shaved regions were treated with 9 at 200 mgkg (single
&
ChemMedChem 2016, 11, 1 – 6
www.chemmedchem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
5
795; c) H. D. Thaker, A. Som, F. Ayaz, D. H. Lui, W. X. Pan, R. W. Scott, J.
Angew. Chem. 2012, 124, 5750–5754; c) Y. Berkov-Zrihen, I. M. Herzog, M.
Feldman, A. Sonn-Segev, Y. Roichman, M. Fridman, Bioorg. Med. Chem.
2013, 21, 3624–3631; d) N. Thamban Chandrika, K. D. Green, J. L.
Houghton, S. Garneau-Tsodikova, ACS Med. Chem. Lett. 2015, 6, 1134–
1139; e) M. P. Mingeot-Leclercq, J. L. Decout, MedChemComm 2016, 7,
586–611.
Anguita, G. N. Tew, J. Am. Chem. Soc. 2012, 134, 11088–11091; d) B. P.
Mowery, S. E. Lee, D. A. Kissounko, R. F. Epand, R. M. Epand, B. Weisblum,
S. S. Stahl, S. H. Gellman, J. Am. Chem. Soc. 2007, 129, 15474–15476;
e) H. Mohammad, S. Thangamani, M. N. Seleem, Curr. Pharm. Des. 2015,
2
1, 2073–2088.
[
4] a) I. S. Radzishevsky, S. Rotem, D. Bourdetsky, S. Navon-Venezia, Y. Carme-
li, A. Mor, Nat. Biotechnol. 2007, 25, 657–659; b) Y. G. Hu, M. N. Amin, S.
Padhee, R. S. E. Wang, Q. Qiao, G. Bai, Y. Q. Li, A. Mathew, C. H. Cao, J. F.
Cai, ACS Med. Chem. Lett. 2012, 3, 683–686; c) N. P. Chongsiriwatana,
T. M. Miller, M. Wetzler, S. Vakulenko, A. J. Karlsson, S. P. Palecek, S. Mo-
bashery, A. E. Barron, Antimicrob. Agents Chemother. 2011, 55, 417–420;
d) S. S. Shankar, S. N. Benke, N. Nagendra, P. L. Srivastava, H. V. Thulasir-
am, H. N. Gopi, J. Med. Chem. 2013, 56, 8468–8474.
[7] a) C. Ghosh, G. B. Manjunath, P. Akkapeddi, V. Yarlagadda, J. Hoque, D. S.
Uppu, M. M. Konai, J. Haldar, J. Med. Chem. 2014, 57, 1428–1436;
b) D. S. S. M. Uppu, P. Akkapeddi, G. B. Manjunath, V. Yarlagadda, J.
Hoque, J. Haldar, Chem. Commun. 2013, 49, 9389–9391.
[8] a) C. Ghosh, G. B. Manjunath, M. M. Konai, D. S. S. M. Uppu, J. Hoque, K.
Paramanandham, B. R. Shome, J. Haldar, PLoS One 2015, 10, e014409;
b) J. G. Hurdle, A. J. O’Neill, I. Chopra, R. E. Lee, Nat. Rev. Microbiol. 2011,
9, 62–75; c) C. Ghosh, G. B. Manjunath, M. M. Konai, D. S. S. M. Uppu, K.
Paramanandham, B. R. Shome, R. Ravikumar, J. Haldar, ACS Infect. Dis.
2016, 2, 111–122.
[
[
5] a) M. M. Konai, C. Ghosh, V. Yarlagadda, S. Samaddar, J. Haldar, J. Med.
Chem. 2014, 57, 9409–9423; b) M. M. Konai, J. Haldar, ACS Infect. Dis.
2
015, 1, 469–478; c) M. M. Konai, U. Adhikary, S. Samaddar, C. Ghosh, J.
[9] J. H. Moffatt, M. Harper, P. Harrison, J. D. Hale, E. Vinogradov, T. Seemann,
R. Henry, B. Crane, F. St Michael, A. D. Cox, B. Adler, R. L. Nation, J. Li, J. D.
Boyce, Antimicrob. Agents Chemother. 2010, 54, 4971–4977.
Haldar, Bioconjugate Chem. 2015, 26, 2442–2453; d) J. Hoque, M. M.
Konai, S. Gonuguntla, G. B. Manjunath, S. Samaddar, V. Yarlagadda, J.
Haldar, J. Med. Chem. 2015, 58, 5486–5500.
6] a) V. Yarlagadda, P. Akkapeddi, G. B. Manjunath, J. Haldar, J. Med. Chem.
2
014, 57, 4558–4568; b) I. M. Herzog, K. D. Green, Y. Berkov-Zrihen, M.
Feldman, R. R. Vidavski, A. Eldar-Boock, R. Satchi-Fainaro, A. Eldar, S. Gar-
Received: August 9, 2016
neau-Tsodikova, M. Fridman, Angew. Chem. Int. Ed. 2012, 51, 5652–5656;
Published online on && &&, 0000
ChemMedChem 2016, 11, 1 – 6
www.chemmedchem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

COMMUNICATIONS
C. Ghosh, M. M. Konai, P. Sarkar,
S. Samaddar, J. Haldar*
How simple but selective antibacterial
membrane-active agents can be de-
signed is presented herein. Bifurcation
of long alkyl chains into two short
chains leads to selective antibacterial
activity. A representative bifurcated
compound exhibits broad-spectrum
bactericidal activity and disrupts bio-
films. It also treats burn wounds in mice
infected with Acinetobacter baumannii.
This principle could be applied to fur-
ther designs of membrane-active
agents.
&& – &&
Designing Simple Lipidated Lysines:
Bifurcation Imparts Selective
Antibacterial Activity
&
ChemMedChem 2016, 11, 1 – 6
www.chemmedchem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!