Deciphering the Role of Intramolecular Networking in Cholic Acid-Peptide Conjugates on the Lipopolysaccharide Surface in Combating Gram-Negative Bacterial Infections

Article abstract of DOI:10.1021/acs.jmedchem.8b01357

The presence of lipopolysaccharide and emergence of drug resistance make the treatment of Gram-negative bacterial infections highly challenging. Herein, we present the synthesis and antibacterial activities of cholic acid-peptide conjugates (CAPs), demonstrating that valine-glycine dipeptide-derived CAP 3 is the most effective antimicrobial. Molecular dynamics simulations and structural analysis revealed that a precise intramolecular network of CAP 3 is maintained in the form of evolving edges, suggesting intramolecular connectivity. Further, we found high conformational rigidity in CAP 3 that confers maximum perturbations in bacterial membranes relative to other small molecules. Interestingly, CAP 3-coated catheters did not allow the formation of biofilms in mice, and treatment of wound infections with CAP 3 was able to clear the bacterial infection. Our results demonstrate that molecular conformation and internal connectivity are critical parameters to describe the antimicrobial nature of compounds, and the analysis presented here may serve as a general principle for the design of future antimicrobials.

Full text of DOI:10.1021/acs.jmedchem.8b01357

Subscriber access provided by MIDWESTERN UNIVERSITY
Article
Deciphering the Role of Intramolecular Networking in Cholic
Acid-Peptide Conjugates (CAPs) at Lipopolysaccharide
Surface in Combating Gram-negative Bacterial Infections
Kavita Yadav, Sandeep Kumar, Deepakkumar Mishra, Mohammad Asad, Madhurima Mitra, Prabhu
Srinivas Yavvari, Siddhi Gupta, Madhukar Vedantham, Pavit Ranga, Varsha Komalla, Sanjay Pal, Priyanka
Sharma, Arti Kapil, Archana Singh, Nirpendra Singh, Aasheesh Srivastava, Lipi Thukral, and Avinash Bajaj
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01357 • Publication Date (Web): 28 Jan 2019
Downloaded from http://pubs.acs.org on January 29, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
Deciphering the Role of Intramolecular Networking in Cholic Acid-Peptide Conjugates
(CAPs) at Lipopolysaccharide Surface in Combating Gram-negative Bacterial
Infections
7
8
9
Kavita Yadav,1,2,a Sandeep Kumar,1,2,a Deepakkumar Mishra,1 Mohammad Asad,1
Madhurima Mitra,1 Prabhu S. Yavvari,3 Siddhi Gupta,1 Madhukar Vedantham,1 Pavit
Ranga,1 Varsha Komalla,1 Sanjay Pal,1,4 Priyanka Sharma,5 Arti Kapil,5 Archana Singh,6
Nirpendra Singh,7 Aasheesh Srivastava,3 Lipi Thukral,6,* and Avinash Bajaj1,*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1. Laboratory of Nanotechnology and Chemical Biology, Regional Centre for Biotechnology,
3rd Milestone Faridabad-Gurgaon Expressway, NCR Biotech Science Cluster, Faridabad-
121001, Haryana, India.
2. Manipal Academy of Higher Education, Manipal-576104, Karnataka, India.
3. Department of Chemistry, Indian Institute of Science Education and Research, Bhopal-
462030, Madhya Pradesh, India.
4. Kalinga Institute of Industrial Technology, Bhubaneswar-751024, Odisha, India.
5. Department of Microbiology, All India Institute of Medical Sciences, Ansari Nagar, New
Delhi-110029, India.
6. CSIR-Institute of Genomics and Integrative Biology, Mathura Road, New Delhi-110025,
India.
7. Regional Centre for Biotechnology, 3rd Milestone Faridabad-Gurgaon Expressway, NCR
Biotech Science Cluster, Faridabad-121001, Haryana, India.
a: authors contributed equally.
Corresponding Author:
Lipi Thukral, Email: lipi.thukral@igib.res.in
Avinash Bajaj, Email: bajaj@rcb.res.in
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 38
1
2
3
4
ABSTRACT
5
6
7
8
Presence of lipopolysaccharide and emergence of drug resistance make the treatment of
Gram-negative bacterial infections highly challenging. Herein, we present the synthesis and
antibacterial activities of Cholic Acid-Peptide conjugates (CAPs) demonstrating that Valine-
Glycine dipeptide-derived CAP 3 is the most effective antimicrobial. MD simulations and
structural analysis revealed that precise intramolecular network of CAP 3 is maintained in
the form of evolving edges suggesting intramolecular connectivity. Further, we found high
conformational rigidity in CAP 3 that confers maximum perturbations in bacterial membranes
relative to other small molecules. Interestingly, CAP 3-coated catheters did not allow the
formation of biofilms in mice, and treatment of wound infections with CAP 3 was able to
clear the bacterial infection. Our results demonstrate that molecular conformation and
internal connectivity are critical parameters to describe antimicrobial nature of compounds,
and analysis presented here may serve as a general principle for design of future
antimicrobials.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 38
Journal of Medicinal Chemistry
1
2
3
4
INTRODUCTION
5
6
7
8
Microbial infections caused by Gram-negative bacteria pose a serious healthcare challenge
due to emergence of multi-drug resistance towards existing antibiotics.1 Escherichia coli (E.
coli) is one of the major causative agent for skin and soft tissue infections especially in case
of neonatal omphalitis, surgical site infections and infections after burn injuries.2 E. coli-
mediated nosocomial catheter-associated infections caused by drug resistant bacteria are
a serious burden in medical settings.3 Ability of the E. coli to form biofilms on skin or soft
tissues and catheters make the treatment more difficult due to ability of biofilms to evade
components of host immune response, stability of biofilms against mechanical forces,
repeated infections and inability of the existing antibiotics to penetrate the enriched matrix
of biofilms.4,5
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Antimicrobials targeting key components of protein machinery may not adequately control
the evolving drug-resistance as bacteria has the propensity to become resistant through
accumulation of genetic mutations and evolutionary selection.6 In contrast, bacterial
membranes provide a suitable target for engineering of antimicrobial agents as targeting of
membranes does not allow the bacteria to acquire drug resistance.7 Membrane targeting of
Gram-negative bacteria is more challenging over Gram-positive bacterial membranes due
to presence of extra lipopolysaccharide (LPS) coated outer membranes8 where lipid A-
mediated bridge crosslinking creates a tough barrier for any toxic material.9,10
Lipopeptides (LPPs) and antimicrobial peptides (AMPs) are naturally occurring metabolites
produced by host organisms as a part of innate immune system that can act against different
pathogenic viral, bacterial and fungal infections.11-12 Naturally isolated LPPs have a single
hydrophobic chain attached to cyclic peptides making them amphiphilic in nature.11 In
contrast, -helix and -sheet-based natural AMPs provide required facial amphiphilic
character with clear segregation of charge and hydrophobic amino acids.12 Amphiphilic
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 38
1
2
3
4
5
6
7
8
9
nature of these LPPs and AMPs allow them to fold into certain structural conformations that
is essential for executing antimicrobial activities. Therefore, numerous LPPs, AMPs and their
derivatives have been engineered with different amino acid variations and hydrophobic
tails,13-16 but only few studies have stressed on the role of different structural conformations
in antimicrobial activity.17-20
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cholic Acid (CA) scaffold, like AMPs, provides a facial amphiphilic character with three
hydroxyl groups on its concave side.21 Many CA-derived steroidal antimicrobials called
Ceragenins have designed where CA was modified with charged amino groups along with
different hydrophobic tails at carboxyl terminal.22-24 However, the impact of natural amino
acid-derived CA-peptide conjugates on antibacterial properties and their interactions with
bacterial membranes at atomistic level has never been studied in detail.25 Therefore, we
undertook a systematic study to probe the antibacterial effect of all natural amino acid-
derived peptides appended at CA scaffold against Gram-negative bacteria. In-depth
molecular dynamics (MD) simulations followed by network analysis witnessed that tethering
of Valine-Glycine dipeptides on CA (CAP 3) provides required conformational rigidity for
executing the effective interactions with bacterial membranes. Biophysical and biochemical
assays validated that specific conformation stabilized by intramolecular interactions allows
CAP 3 in executing membrane perturbations through specific contacts and in combating
drug resistant, persistent, wound and catheter infections in in vitro and in vivo model
systems.
RESULTS AND DISCUSSION
Design, synthesis and structure-activity relationship. We synthesized twenty CAPs with
general chemical formula of CA-(G-X)3 (referred as CA-X3) where X is any natural amino
acid conjugated to CA through a glycine linker (Figure 1A). For synthesis, CA (21) was first
esterified with benzyl bromide in basic conditions to give benzyl cholate (22) in quantitative
4
ACS Paragon Plus Environment

Page 5 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
yields (Figure S1). Boc protected glycines were then conjugated to three hydroxyl groups of
benzyl cholate (22) using DCC/DMAP coupling followed by Boc deprotection using 4M HCl
in dioxane. Different amino acids with suitable protecting groups were then attached to tri-
glycine conjugated benzyl cholate (23) using diimide-based coupling reagents followed by
deprotection as mentioned in Supplementary Information. All CAPs were characterized by
1H NMR and HRMS and purity of CAPs (>95%) was confirmed by HPLC.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We then tested the antibacterial activity of CAPs against E. coli using broth dilution assay to
determine the minimum inhibitory concentration at which 99% bacterial killing (MIC99) was
observed.26 SAR studies revealed that basic and acidic amino acid-derived CAPs like 14
(CA-D3), 15 (CA-E3), 16 (CA-R3), 18 (CA-K3), 19 (CA-N3) and 20 (CA-Q3), and aromatic
amino acid-derived CAPs like 8 (CA-F3), 9 (CA-Y3) and 10 (CA-W3) are not active (Table
S1). In contrast, CAPs appended with aliphatic amino acid residues like 1 (CA-G3), 2 (CA-
A3), 3 (CA-V3), 4 (CA-I3), 5 (CA-L3) and 7 (CA-M3) are active in the range of 8-128 M.
Valine-derived CAP 3 (CA-V3) and Isoleucine-derived CAP 4 (CA-I3) are most active with
MIC99 of 8 M (Table S1). SAR against other Gram-negative bacteria Klebsiella pneumoniae
and Acinetobacter baumannii witnessed that CAP 3 and CAP 4 are active at 16 M (Table
S1). Hemolytic activity of CAPs against human red blood cells (RBCs) and cytotoxicity
against epithelial cells (A549) revealed that CAP 4 is highly toxic towards RBCs and
epithelial cells without any selectivity for bacterial membranes (Table S1). In contrast, CAP
3 was ~6-7 fold selective for E. coli over RBCs and epithelial cells. To overrule the detergent-
mediated effect of CAP 3, we studied the self-assembled properties of CAP 3 and
determined the critical miceller concentration (CMC) of CAP 3. CAP 3 did not show any
aggregation up to 100 M (CMC > 100 M) confirming the non-detergent-mediated
antibacterial effect of amphiphile.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 38
1
2
3
4
5
6
7
8
9
Natural AMPs and LPPs are usually rich in basic amino acids like lysine, arginine or polar
hydrophobic amino acids like tryptophan that allows them to bind effectively with bacterial
membranes.27 In contrast, above results witnessed that lysine, arginine and tryptophan-
derived CAPs did not show any antibacterial activity. To understand this SAR, we selected
most active hydrophobic valine-derived CAP 3 and three inactive CAPs based on polar
hydrophobic tryptophan ( CAP 10), acidic glutamic acid (CAP 15) and basic lysine (CAP 18)
amino acids (Figure 1B). We compared their interactions with LPS-derived membranes
using computational, biophysical and biochemical approaches.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Biomolecular simulations. We performed all-atom molecular dynamics simulations to
study the differential binding modes of the selected CAPs with model Gram-negative
bacterial membranes.28 Four independent molecular systems containing complex LPS of
Gram-negative bacteria were generated that mimic the experimentally tested molecular
process. Gram-negative outer membrane model consists of LPS molecules in the outer
leaflet and varied phospholipids in inner leaflet to simulate the physiologically relevant
bacterial membrane interface (Figure S2, Table S2).29,30 The starting structure
representation was prepared with no prior membrane contacts (Figure S3A), and after 300
ns, subtle changes in membrane architecture were observed (Figure S3B). Towards the end
of the simulation at 500 ns, CAPs displayed varied membrane permeabilization activity. In
particular, CAP 3 was completely inserted within the outer leaflet comprising of LPS
molecules (Figure 2A). CAP 15 also showed partial insertion, whereas CAP 10 and CAP
18 could execute relatively few membrane interactions (Figure 2A). The contour maps
shown in Figure 2B depict top view of the simulation box after 500 ns with each grid value
showing membrane thickness that represents the distance between upper and lower
membrane atoms. Significant membrane thinning in CAP 3-membrane complex was
observed with scattered blue color patches of ~2 nm thickness representing compressed
6
ACS Paragon Plus Environment

Page 7 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
membranous structures. CAP 10 and 18 did not cause any thinning of membranes, whereas
we observed CAP 15 induced membrane thinning to some extent (Figure 2B).
Above results revealed a clear trend, where CAP 3 causes maximum disruptions to the
membrane architecture and CAP 18 showed poor interactions with membranes in spite of
being appended to basic charged amino acids. As molecular crowding is a key factor in
antimicrobial activity, we ran additional simulations with higher concentrations of CAP 3
where two independent structures were generated containing membranes with 5 and 20
molecules of CAP 3 in water environment (Figure S4). As a consequence, severe membrane
defects were observed as compared to control (only membrane) trajectories (Figure S4).
These findings demonstrate the ability of CAP 3 to integrate into the bacterial LPS groups
and cause significant structural variations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We then characterised four segments of bacterial membranes, O-Antigen, outer and Inner
polysaccharides, lipid A and phospholipids to depict the precise binding site differences of
CAPs with bacterial membranes. We found that CAP 3, most active antibacterial, favourably
interacts with inner polysaccharides of bacterial membranes (Figure S5A). In contrast,
inactive CAPs 10, 15 and 18 reside on the surface of bacterial membranes and interact with
the outer core polysaccharides (Figure S5B-D). Remarkably, the core fold of CAP 3 was
inserted completely to interact with O-antigen segment through polar interactions (Figure
S5A). In contrast, position of other CAPs were not able to form an inward or membrane
facing binding pocket and therefore could not perform same interactions as that of CAP 3
(Figure S5B-D).
To understand the toxicity differences observed between CAP 3 (less toxic) and CAP 4
(more toxic) against mammalian cells, we also performed simulations of these molecules
(CAP 3 and CAP 4) with model mammalian cell membranes (Table S2). The atomistic
trajectories of these two molecules with DPPC-cholesterol molecules (model membranes)
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 38
1
2
3
4
5
6
7
8
9
were generated to mimic the mammalian membranes. CAP 4 showed immediate membrane
penetration as compared to CAP 3 that was residing only at the surface of membrane without
any insertion (Figure S6A-B). The kinetic calculations of the distance showed increased
membrane contacts of CAP 4 as compared to CAP 3 (Figure S6C). Number of contacts in
case of CAP 4 significantly increased from 40 to ~80 as compared to CAP 3 that did not
show any significant increase in number of contacts (Figure S6D). Therefore, favorable
interactions of CAP 4 with mammalian cell membranes make it more toxic than CAP 3.
Savage’s group has performed fluorescence-based studies to decipher the interactions of
CA-derived amphiphiles (Ceragenins) with Gram-negative membranes, and proposed that
lipid A component of LPS is responsible for interactions of these antimicrobials with bacterial
membranes.22,23 They also synthesized water soluble derivative of Lipid A and fluorophore-
derivatives of cationic steroids and compared the binding abilities of Ceragenins with lipid
A.22,23 In contrast, our studies provided atomistic level interactions of CA-derived
amphiphiles with bacterial membranes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Structural analysis of CAPs. We then undertook a detailed survey of molecular
conformations adopted by these CAPs. Values of root mean square fluctuations (RMSF)
were calculated and mapped onto chemical structures as shown in Figure 3A. CAP 10 was
the most flexible with highest RMSF of 0.73 nm (Figure 3A, S7A). Interestingly, CAP 3
showed rigid conformational character with lowest RMSF value of 0.42 nm. These values
were then decomposed depending on the atoms of cholic acid and side-chain where cholic
acid atoms, a common denominator across all molecules, showed considerable changes in
all CAPs except for CAP 3 (Figure S7B). In contrast, side-chain atoms for all CAPs were
highly mobile and dynamic during interactions with membrane groups (Figure S7C).
To determine the most commonly occurring conformations, we performed clustering on all
structures within each trajectory and selected top clusters (Figure S8). Top clusters for CAPs
8
ACS Paragon Plus Environment

Page 9 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10 and 15 trajectories showed only ~11.8 and ~22.4% due to varying number of
conformations in these CAPs (Figure S8). CAP 3 exhibited similar conformation as the top
cluster was present in ~94.56% probability whereas CAP 18 existed in a conformation with
90% probability. Superimpositions of all the top clusters revealed structural variability in each
CAP that clearly witnessed the rigidity of CAP 3 over other CAPs (Figure 3B). Therefore,
structural analysis of CAPs confirmed that chemical space of CAP 3 is highly rigid and this
unique interfacial conformational state may be directly coupled to its potent activity towards
membrane permeabilization.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Intra-molecular network-based analysis. We then quantified the number of membrane
contacts and hydrogen bonds formed by each CAP with membrane surface, and observed
that both membrane contacts and hydrogen bonds increased as a function of time for all the
CAPs (Figure S9). CAPs 3 and 15 formed minimal membrane contacts (Figure S9A). Kinetic
plots demonstrated the ability of these molecules to form strong interactions with
membranes where CAP 3 is perturbing the membranes with relatively few non-covalent
interactions (Figure S9B).
In order to probe the differential interactions executed by these CAPs, we studied the
dynamics of intra-molecular interactions in CAPs at LPS membrane surface. We utilized the
network approach analogous to Residue Interaction Network (RIN) that is computed widely
for intra-protein interactions.31-32 In this network analysis, interactions among different atoms
within a molecule correspond to “edges” and number of atoms within molecule correspond
to “nodes”. Specifically, the nodes stay constant and edges are dynamically formed by non-
covalent interactions during the length of the simulations. In order to probe geometric
arrangements between nodes and edges, we computed networks for CAPs at 0 and 500 ns.
Global network topology derived from structures of CAPs shown in Figure S10-S13 suggests
that networks are dense with disparate nodes towards the periphery. We then computed
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 38
1
2
3
4
5
6
7
8
9
general network parameters about centrality and connectedness of the networks and
witnessed that several edges are contributed from CA backbone atoms (Figure 4).
Further, we quantified the dynamic parameters of networks and noticed that CA backbone
presented “hub” features across all the CAPs. The hubs are defined as nodes that have
stable and more number of interactions. Comparison of all nodes (in particular 20-64 nodes)
representing CA backbone atoms (or hub) of CAPs suggested that only CAP 3 execute new
interactions within molecule after 500 ns with 885 edges whereas only 868 interactions were
observed at 0 ns (Figure 4, Figure S14). In contrast, significantly lesser number of edges
are formed in other three CAPs (Figure 4, Table S3). In addition, side-chain atoms of CAP
3 also contributed to this increase where benzene ring interacts heavily with the backbone
atoms (Figure S10). We observed that interactions between side-chain atoms and CA
backbone in other CAPs are highly dynamic (Figure S11-S13, Figure S14).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Our network analysis therefore suggests that transition from starting conformational state to
other structures occurs through dynamic changes within CAP 3. In particular, new edges
(interactions) are formed within the molecule in different parts of chemical space. These
findings suggested that intra-molecular interactions compensated with low membrane
contacts formed by CAP 3 may directly aid in membrane perturbation behavior. To test how
the order parameters calculated from MD simulations can rank the molecules based on
activity, we ran additional simulations of CAP 4 (antibacterial like CAP 3) and CAP 9 (inactive
against bacteria) with model bacterial membranes. CAP 4 showed full insertion into the
bacterial membranes (Figure S15A-B) as compared to interfacial contacts made by CAP 9
(Figure S15C-D). Quantification of membrane thickness witnessed CAP 4-mediated
enhanced thinning of bacterial membranes as compared to CAP 9 (Figure S15E) and RMSF
calculations confirmed more rigidity of CAP 4 over CAP 9 amphiphile (Figure S15F). Time
calculations of membrane contacts and hydrogen bonding could not capture any molecular
10
ACS Paragon Plus Environment

Page 11 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
differences (Figure S16A, S16B). In contrast, intramolecular networking of CAP 4 revealed
higher number of edges with increased intramolecular networking as compared to CAP 9 on
interactions with bacterial membranes (Figure 16C-D). Therefore, these simulations validate
that intramolecular networking of small molecules may accurately predict the activity of small
molecules with high precision.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Validation of CAP-LPS Interactions. We then validated the differential interactions of the
selected CAPs with model membranes as depicted by MD simulations using biophysical
and biochemical assays. To assess the impact of CAPs on membrane rigidity, we first
prepared diphenylhexatriene (DPH)-doped model Gram-negative bacterial vesicles using
LPS, DPPE and DMPG lipids,33 and measured the change in DPH anisotropy on incubation
with CAPs. We observed ~2-fold increase in fluidity of membranes on incubation with CAP
3 whereas insignificant change in rigidity was observed on incubation of other CAPs (Figure
5A). We then evaluated the comparative binding affinities of these CAPs with LPS using
Dansyl-Polymyxin B displacement assay where complexes of LPS with Dansyl-labelled
Polymyxin B were titrated with CAPs and change in fluorescence was quantified (10 was
insoluble in water and not tested further).34 Increase in fluorescence of Dansyl-Polymyxin B
due to its displacement from LPS complexes confirmed the strongest affinity of CAP 3 for
LPS over other CAPs (Figure 5B). Similarly, increase in fluorescence on titrations of CAPs
with fluorescent boron-dipyromethane conjugated LPS (BODIPY-LPS) established that CAP
3 can bind and disintegrate the LPS aggregates more effectively than other CAPs (Figure
5C).35
We then assessed the relative membrane binding affinities of CAPs using Surface Plasmon
Resonance (SPR)36 where CAP 3 showed highest binding response with irreversible binding
among all the CAPs (Figure 5D-5F). Analysis of overall affinity constant (KA) witnessed
highest binding of CAP 3 over CAP 15 and CAP 18.37 Comparison of K1 and K2 revealed
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 38
1
2
3
4
5
6
7
8
9
that second step involving integration of antimicrobials into bacterial membranes is
responsible for higher KA for CAP 3 (Figure 5G). This strong association of CAP 3 in second
step might be due to presence of balanced charge and hydrophobicity in valine-derived CAP
3 that allows electrostatic interactions followed by hydrophobic integration in bacterial
membranes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bactericidal effect of CAP 3. To further validate the differential interactions of CAPs with
bacterial membranes, we compared the abilities of CAPs to permeabilize the outer bacterial
membranes in E. coli using N-phenyl naphthylamine (NPN) fluorescent dye as its
fluorescence gets enhanced on penetration in disrupted hydrophobic membranes.38
Comparison of permeation assay witnessed that CAP 3 is more effective in permeabilization
of outer bacterial membranes over other CAPs (Figure 6A). Interactions of membrane
targeting antimicrobials in general induce depolarization of the membranes and release of
accumulated quenched fluorescent dyes like 3',3'-diethylthiadicarbocyanine DiSC2(5) from
depolarized membranes results in enhanced fluorescence.39 Comparative analysis of CAPs
confirmed CAP 3-mediated increase in fluorescence of DiSC2(5) thereby making it most
effective in depolarization of bacterial membranes (Figure 6B). Quantification of the uptake
of membrane impermeable dye propidium iodide (PI) by bacteria on treatment with different
CAPs revealed CAP 3-induced significant increase in number of PI positive cells (Figure
6C). Microscopy studies showed the uptake the PI confirming the membrane lytic nature of
the CAP 3 (Figure S17A). To confirm the LPS-mediated interactions of CAP 3 with bacterial
membranes, we tested the antibacterial activity of CAP 3 against E. coli in presence of LPS
and measured the percentage of PI positive cells. We observed LPS-mediated dose
dependent inhibition of PI uptake by E. coli on CAP 3 treatment thereby confirming the
effective CAP 3-LPS interactions (Figure S17B). Therefore, above studies validated the SAR
and our observations from MD simulations where valine-derived CAP 3 was found most
12
ACS Paragon Plus Environment

Page 13 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
effective in binding with LPS membranes over other CAPs that is responsible for its
antibacterial effect.
7
8
9
We then performed time dependent killing assay where bacteria (E. coli) were treated with
CAP 3 for different time and CFUs were quantified. We did not observe any colony after 6h
of CAP 3 treatment at 1X MIC99 and 90 min of CAP 3 treatment at 4X MIC99 was sufficient
to kill most of the bacteria (Figure 6D). Impact of CAP 3 on membrane morphology was then
assessed using transmission electron microscopy where untreated bacteria showed a
smooth rod-shaped morphology and CAP 3 treatment induced distinct morphological
changes with distorted shapes and ‘kinks’ in the bacterial membrane (Figure 6E).
Major challenge for use of any antimicrobial is the ability of bacteria to develop resistance
against antimicrobials.40 Therefore, we tested the ability of E. coli to develop resistance
against CAP 3 and observed that E. coli was unable to develop resistance against the CAP
3 whereas there was multi-fold increase in MIC99 of neomycin (Figure 6F). Antimicrobial
activity of CAP 3 against stationary and persistent E. coli cells established that CAP 3 was
able to kill the stationary and persistent bacteria where ampicillin was ineffective (Figure
6G). Antibacterial activities at different CFUs of E. coli validated that CAP 3 was also able
to clear the bacterial growth even at CFUs of 1012/mL, making it highly potent antimicrobial
(Figure 6H).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Activity against biofilms. Biofilms usually respond poorly to antibiotic therapy and are
responsible for inducing antibiotic resistance due to increased mutation rates, high
expression of efflux pumps and trapping of antibiotics in exopolysaccharide matrix.41
Therefore, we assessed the ability of CAP 3 to disrupt the E. coli biofilms. Pre-formed E. coli
biofilms were treated with different concentrations of CAP 3 for 24h and were quantified by
CFU analysis. A dose dependent decrease in CFUs was observed on CAP 3 treatment and
CAP 3 at 64 M (8X MIC99) resulted in a ~8-log fold decrease in CFU (Figure 7A). Crystal
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 38
1
2
3
4
5
6
7
8
9
violet staining witnessed ~90% decrease in biofilm mass after treatment with CAP 3 at 8X
MIC99 (Figure 7B). Although the concentrations for biofilm degradation are much higher than
the toxic concentration of the amphiphile for mammalian cells, it should be noted that the
experimental biofilms used here are very thick in nature with high bacterial load that is not
usually observed in clinical settings. Bactericidal nature of CAP 3 on biofilms was then
established using SYTO9-PI staining where SYTO9 can permeabilize and stain all the
bacteria and PI can only permeabilize and stain dead bacteria. Confocal laser scanning
micrographs (CLSM) of untreated biofilms stained with SYTO9-PI showed thick biofilm mass
of viable SYTO9-stained green fluorescent E. coli bacteria without any visible PI-stained
red bacteria (Figure 7C, upper panel). CAP 3 treatment resulted in increase of PI-stained E.
coli establishing the bactericidal and biofilm disrupting effect of CAP 3 (Figure 7C, lower
panel). Quantification of the biofilm thickness confirmed significant reduction in biofilm
formation on CAP 3 treatment (~ 3 m) as compared to untreated biofilms (>8 m) (Figure
7D). Scanning electron microscopy (SEM) was then performed to understand the effect of
CAP 3 on architecture of E. coli biofilms. Untreated biofilms presented a thick biomass with
rod shape (E. coli) bacteria forming an extensive network of pores and channels that helps
in distribution of nutrients (Figure 7E). CAP 3 treatment unsettles this network and
breakdowns the extracellular architecture by disrupting the bacterial membranes and by
creating isolated lytic bacteria (Figure 7E).42 Therefore, these results establish the ability of
the CAP 3 to interact with Gram-negative bacteria submerged in thick biofilms and clear
them.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In vivo activity. As E. coli-mediated catheter and wound infections are very common, we
first tested the ability CAP 3-coated catheters to prevent the biofilm formation during in vitro
conditions. Sterilized catheter pieces (~1 cm) were first coated with CAP 3 where catheters
were dipped in solution of CAP 3 in dichloromethane (DCM) and DCM was evaporated
14
ACS Paragon Plus Environment

Page 15 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
providing CAP 3-coated catheters. These coated catheters were dipped in E. coli
suspension for three days and CFU analysis confirmed the absence of any adhered bacteria
on CAP 3-coated catheters whereas significant amount of bacterial coating was observed
on uncoated and control (DCM)-coated catheters (Figure 8A). To test the efficacy of CAP
3-coated catheters in preventing biofilm formation in mice, uncoated, DCM (control)- and
CAP 3-coated catheter pieces (~1cm, 1 catheter/mice, 4 mice/group) were inserted
subcutaneously in an incision on the flank of mice43 and these incised sites were infected
with E. coli. CFU analysis confirmed that there was no adherence of any bacteria on CAP
3-coated catheters after three days unlike uncoated and control catheters establishing the
ability of CAP 3 in preventing bacterial growth in murine models (Figure 8B).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To assess the efficacy of CAP 3 in clearing the wound infections,44 we first created the
wounds on BALB/c mice and infected them with E. coli strain. After 6h of infection, mice
were randomized into four groups (3 mice/group) and were either left untreated (group 1) or
treated with ampicillin (40mg/kg) (group 2) or CAP 3 (40mg/kg) (group 3) or neomycin
(40mg/kg) (group 4) thrice daily for four days. CFU analysis on day 5 witnessed significant
reduction in bacterial load on wound where ampicillin was less effective (Figure 8C). We
then performed bioluminescence imaging of mice infected with E. coli bioluminescent strain
(Xen14) and observed significant decrease in bioluminescence on CAP 3 treatment
confirming its bactericidal effect (Figure 8D). These results therefore establish that CAP 3
can act as bactericidal in clearing wound infections.
To validate the therapeutic efficacy of CAP 3 against multi-drug resistant strains,
antibacterial activity of CAP 3 was tested against different multidrug-resistant clinical strains
of E. coli, A. baumannii and K. pneumoniae. CAP 3 was active in range of 4-16 M (MIC99)
against all clinical strains and activity was in range of 4-8 M (MIC99) for E. coli strains
(Figure 8E).
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 38
1
2
3
4
CONCLUSIONS
5
6
7
8
In this work, we presented the design of CA-peptide conjugates to decipher the impact of
all-natural amino acids on antibacterial activities against Gram-negative bacterial species.
Twenty CAPs were engineered where different natural amino acids were appended on
hydroxyl groups of benzylated cholic acid using a glycine linker. SAR witnessed that basic
charged amino acid like lysine, arginine and histidine, and polar hydrophobic amino acid
likes tryptophan and proline-derived CAPs are inactive in spite of their charged nature,
whereas valine-derived CAP 3 is the most active. Comparative MD simulations with LPS
model membranes revealed atomistic features with CAP 3, mediating maximum thinning of
bacterial membranes over other CAPs. Structural and network analysis witnessed stable
conformation of CAP 3 and enhanced intramolecular networking that allowed it to perform
specific interactions with bacterial membranes. Mechanistic studies confirmed LPS-
mediated bactericidal nature of CAP 3 that allowed it to kill drug resistant bacteria and
degrade/prevent the biofilms in murine models. Therefore, this report provides a mechanistic
understanding of membrane perturbations in context of detailed structural and molecular
information of CAPs and explain how intramolecular network of a given antimicrobial is the
distinguishing factor.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
1. Synthesis of CA-peptide (CAPs) conjugates. Detailed synthesis, characterization and
spectral data of all the molecules is described in Supporting information. Purity of the
compounds was tested by HPLC and found to be more than 95%.
2. Other experimental details. Computational methods used, microorganisms and culture
conditions, antibacterial assay, hemolytic assay, cytotoxicity assay, biophysical assay for
studying amphiphile model membrane interactions, membrane permeabilization studies,
activities against biofilms, In vitro catheter assay, in vivo catheter infection studies, in vivo
16
ACS Paragon Plus Environment

Page 17 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
bioluminescence imaging and activity assay against clinical strains are detailed in supporting
information.
7
8
9
3. Ethics statement. All animal experiments were performed after due ethical approval from
Institutional Animal Ethics Committee of Regional Centre for Biotechnology
(RCB/IAEC/2016/001). All experiments with human blood samples and clinical bacterial
isolates were performed after due ethical approval from Institute Ethics Committee of All
India Institute of Medical Sciences (IEC/NP-433/09/10.2015) and Regional Centre for
Biotechnology (RCB-IEC-H-7).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SUPPORTING INFORMATION
Supporting figures S1-S17, supporting tables S1-S4, molecular formula strings, materials
and methods, detailed synthesis of molecules and other experimental details like
computational methods used, microorganisms and culture conditions, antibacterial assay,
hemolytic assay, cytotoxicity assay, biophysical assay for studying amphiphile model
membrane interactions, membrane permeabilization studies, activities against biofilms, in
vitro catheter assay, in vivo catheter infection studies, in vivo bioluminescence imaging and
activity assay against clinical strains are available in supporting information.
AUTHOR CONTRIBUTIONS
K.Y. performed bacterial culture and cell culture-based studies. D.M., P.R., V.K. and S.K.
synthesized and characterized all the CAPs. K.Y., P.S.Y., S.P. and M.V. performed animal
experiments. M.A. performed computational studies. M.M. performed SPR and other
biophysical experiments. S.G., M.V. and P.S.Y. performed in vitro biofilm experiments. S.G.
and A.S. processed samples for TEM studies. P.S. and A.K. helped in screening of clinical
strains. P.S.Y. performed SEM experiments and A.S. supervised it. L.T. supervised the
computational studies. A.B. conceived the idea, designed the experiments and supervised
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 38
1
2
3
4
5
6
the whole project. Manuscript was written by K.Y., S.K., L.T. and A.B. and approved by all
the authors.
7
8
9
ACKNOWLEDGEMENTS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We are grateful to RCB, IGIB, IISER and AIIMS for intramural funding. Work in A.B.
laboratory is supported by the following grants: BT/PR12297/MED/29/895/2014 (DBT) and
BT/PR17525/MED/29/1021/2016 (DBT). P.S.Y and S.P. thank UGC, and K.Y. and S.K.
thank RCB for research fellowships. Financial support from the DBT-RA Program in
Biotechnology and Life Sciences is gratefully acknowledged. Small animal facility of
Regional Centre for Biotechnology is supported BT/PR5480/INF/22/158/2012 (DBT). A.S. is
supported by BT/PR17525/MED/29/1021/2016 (DBT). L.T. thanks support from CSIR
intramural funding for this project under MLP 1805 grant. We thank CSIR-4PI for
supercomputing resources, and TEM facility at CSIR-IGIB. We also thank the support of
DBT e-Library Consortium (DeLCON) for providing access to e-resources.
AUTHOR INFORMATION
*Corresponding Author: Email: bajaj@rcb.res.in; Tel: 0091-129-2848831.
ORCID
Avinash Bajaj: 0000-0002-1333-9316
Notes: Authors declare no competing financial interests.
ABBREVIATIONS USED
CA, Cholic Acid; LPPs, lipopeptides; AMPs, antimicrobial peptides; CAP, cholic acid-peptide
conjugate; CMC, Critical miceller concentration; DPPC, Dipalmitoylphosphatidylcholine;
RMSF, root mean square fluctuations, DPH, diphenylhextatriene, DPPE, 1,2-Dipalmitoyl-
sn-glycero-3-phosphoethanolamine
DMPG,
1,2-Dimyristoyl-sn-glycero-3-
phosphorylglycerol sodium salt; NPN, N-phenyl naphthylamine, DiSC2(5), 3',3'-
diethylthiadicarbocyanine; PI, Propidium Iodide, CFUs, colony forming units.
18
ACS Paragon Plus Environment

Page 19 of 38
Journal of Medicinal Chemistry
1
2
3
4
REFERENCES
5
6
7
8
1. Kaper, J. B.; Nataro, J. P.; Mobley, H. L. T. Pathogenic Escherichia coli, Nat. Rev.
Microbiol. 2004, 2, 123-140.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. Petkovšek, Ž.; Eleršič, K.; Gubina, M.; Žgur-Bertok, D.; Erjavec, M. S. Virulence Potential
of Escherichia coli Isolates from Skin and Soft Tissue Infections. J. Clin. Microbiol. 2009, 47,
1811-1817.
3. Jacobsen, S. M.; Stickler, D. J.; Mobley, H. L. T.; Shirtliff, M. E. Complicated Catheter-
Associated Urinary Tract Infections Due to Escherichia coli and Proteus mirabilis. Clin.
Microbiol. Rev. 2009, 21, 26-59.
4. Bjarnsholt, T.; Alhede, M.; Alhede, M.; Eickhardt-Sørensen, S. R.; Moser, C.; Kühl, M.;
Jensen, P. Ø.; Høiby, N. The in vivo Biofilm. Trends Microbiol. 2013, 21,466-474.
5. Donlan, R. M.; Costerton, J. W. Biofilms: Survival Mechanisms of Clinically Relevant
Microorganisms. Clin. Microbiol. Rev. 2002, 15, 167–193.
6. Toprak, E.; Veres, A.; Michel, J. B.; Chait, R.; Hartl, D. L.; Kishony, R. Evolutionary Paths
to Antibiotic Resistance Under Dynamically Sustained Drug Stress. Nat. Genet. 2011, 44,
101-105.
7. Hurdle, J. G.; O’Neill, A. J.; Chopra, I.; Lee, R. E. Targeting Bacterial Membrane Function:
An Underexploited Mechanism for Targeting Persistent Infections. Nat. Rev. Microbiol.
2011, 62, 62-75.
8. Silhavy, T. J.; Kahne, D.; Walker, S. The Bacterial Cell Envelope. Cold Spring Harb.
Perspect. Biol. 2010, 2, a000414.
9. Papo, N.; Shai, Y. A Molecular Mechanism for Lipopolysaccharide Protection of Gram-
Negative Bacteria from Antimicrobial Peptides. J. Biol. Chem. 2005, 280, 10378-10387.
10. Rosenfeld, Y.; Papo, N.; Shai, Y.; Endotoxin (Lipopolysaccharide) Neutralization by
Innate Immunity Host-Defense Peptides. J. Biol. Chem. 2006, 281, 1636-1643
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 38
1
2
3
4
5
6
11. Hamley, I. W. Lipopeptides: From Self-Assembly to Bioactivity. Chem Commun. 2015,
51, 8574-8583.
7
8
9
12. Hancock, R. E. W.; Diamond, G. The Role of Cationic Antimicrobial Peptides in Innate
Host Defences. Trends Microbiol. 2000, 8, 402–410
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13. Stach, M.; Siriwardena, T. N.; Köhler, T.; Delden, C. van Delden; Darbre, T.; Reymond,
J. L. Combining Topology and Sequence Design for The Discovery of Potent Antimicrobial
Peptide Dendrimers Against Multidrug-Resistant Pseudomonas aeruginosa. Angew. Chem.
Int. Ed., 2014, 53, 12827-12831.
14. Konai, M. M.; Adhikary, U.; Samaddar, S.; Ghosh, C.; Haldar, J. Structure-Activity
Relationship of Amino Acid Tunable Lipidated Norspermidine Conjugates: Disrupting
Biofilms with Potent Activity against Bacterial Persisters. Bioconjug. Chem. 2015, 26, 2442-
2453.
15. Liu, D.; Choi, S.; Chen, B.; Doerksen, R. J.; Clements, D. J.; Winkler, J. D.; Klein, M. L.;
DeGrado, W. F. Nontoxic Membrane-Active Antimicrobial Arylamide Oligomers. Angew.
Chem. Int. Ed. 2004, 43, 1158-1162.
16. Moon, S. H.; Zhang, X.; Zheng, G.; Meeker, D. G.; Smeltzer, M. S.; Huang, E. Novel
Linear Lipopeptide Paenipeptins with Potential for Eradicating Biofilms and Sensitizing
Gram-Negative Bacteria to Rifampicin and Clarithromycin. J. Med. Chem. 2017, 60, 9630–
9640.
17. Rončević, T.; Vukičević, D.; Ilić, N.; Krce, L.; Gajski, G.; Tonkić, M.; Goić-Barišić, I.;
Zoranić, L.; Sonavane, Y.; Benincasa, M.; Juretić, D.; Maravić, A.; Tossi, A. Antibacterial
Activity Affected by The Conformational Flexibility in Glycine-Lysine Based Α-Helical
Antimicrobial Peptides. J. Med. Chem. 2018, 61, 2924–2936.
20
ACS Paragon Plus Environment

Page 21 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
18. Tang, H.; Doerksen, R. J.; Jones, T. V.; Klein, M. L.; Tew, G. N. Biomimetic Facially
Amphiphilic Antibacterial Oligomers with Conformationally Stiff Backbones. Chem. Biol.
2006, 13, 427-435.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19. Uppu, D. S. S. M.; Konai, M. M.; Baul, U.; Singh, P.; Siersma, T. K.; Samaddar, S.;
Vemparala, S.; Hamoen, L. W.; Narayana, C.; Haldar, J. Isosteric Substitution in Cationic-
Amphiphilic Polymers Reveals An Important Role For Hydrogen Bonding in Bacterial Membrane
Interactions. Chem. Sci. 2016, 7, 4613-4623.
20. Siriwardena, T. N.; Capecchi, A.; Gan, B. H.; Jin, X.; He, R.; Wei, D.; Ma, L.; Köhler, T.;
van Delden, C.; Javor, S.; Reymond, J. L. Optimizing Antimicrobial Peptide Dendrimers in
Chemical Space. Angew. Chem. Int. Ed. 2018, 57, 8483-8487.
21. Lai, X. Z.; Feng, Y.; Pollard, J.; Chin, J. N.; Rybak, M. J.; Bucki, R.; Epand, R. F.; Epand,
R. M.; Savage, P. B. Ceragenins: Cholic Acid-Based Mimics of Antimicrobial Peptides. Acc.
Chem. Res. 2008, 41, 1233–1240.
22. Ding, B.; Yin, N.; Liu, Y.; Cardenas-Garcia, J.; Evanson, R.; Orsak, T.; Fan, M.; Turin,
G.; Savage, P. B. Origins of Cell Selectivity of Cationic Steroid Antibiotics. J. Am. Chem.
Soc. 2004, 126, 13642-13648.
23. Bansal, S.; Singh, M.; Kidwai, S.; Bhargava, P.; Singh, A.; Sreekanth, V.; Singh, R.;
Bajaj, A. Bile Acid Amphiphiles with Tunable Head Groups as Highly Selective
Antitubsercular Agents. MedChemComm 2014, 5, 1761-1768.
24. Savage, P. B. Design, Synthesis and Characterization of Cationic Peptide and Steroid
Antibiotics. Eur. J. Org. Chem. 2002, 5, 759-768.
25. Singla, P., Dalal, P., Kaur, M., Arya, G., Nimesh, S., Singh, R., Salunke, D. B. Bile Acid
Oligomers and Their Combination with Antibiotics to Combat Bacterial Infections. J. Med.
Chem. 2018, 61, 10265–10275.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 38
1
2
3
4
5
6
7
8
9
26. Wiegand, I.; Hilpert, K.; Hancock, R. E. W. Agar and Broth Dilution Methods to
Determine the Minimal Inhibitory Concentration (MIC) of Antimicrobial Substances. Nat.
Protoc. 2008, 3, 163-175.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27. Chan, D. I.; Prenner, E. J.; Vogel, H. J. Tryptophan and Arginine-Rich Antimicrobial
Peptides: Structures and Mechanisms of Action. Biochim. Biophys. Acta-Biomembr. 2006,
1758, 1184-1202.
28. Boags, A.; Hsu, P. C.; Samsudin, F.; Bond, P. J., Khalid, S. Progress in Molecular
Dynamics Simulations of Gram-Negative Bacterial Cell Envelopes. J. Phys. Chem. Lett.
2017, 8, 2513-2518.
29. Jefferies, D.; Hsu, P. C.; Khalid, S. Through The Lipopolysaccharide Glass: A Potent
Antimicrobial Peptide Induces Phase Changes in Membranes. Biochemistry 2017, 56, 1672-
1679.
30. Castillo, N.; Monticelli, L.; Barnoud, J.; Tieleman, D. P. Free energy of WALP23 dimer
association in DMPC, DPPC, and DOPC bilayers. Chem. Phys. Lipids. 2013, 169, 95-105.
31. Nagpal, S.; Tiwari, S.; Mapa, K.; Thukral, L. Decoding Structural Properties of a Partially
Unfolded Protein Substrate: En Route to Chaperone Binding. PLoS Comput. Biol. 2015, 11,
e1004496.
32. Doncheva, N. T.; Klein, K.; Domingues, F. S.; Albrecht, M. Analyzing and Visualizing
Residue Networks of Protein Structures. Trends Biochem. Sci. 2011, 36, 179-182.
33. Singh, M.; Bansal, S.; Kundu, S.; Kidwai, S.; Bhargava, P.; Singh, A.; Motiani, R. K.;
Shyam, R.; Sreekanth, V.; Sengupta, S.;
Bajaj, A. Synthesis, Structure-Activity
Relationship, and Mechanistic Investigation of Lithocholic Acid Amphiphiles for Colon
Cancer Therapy. MedChemComm. 2015, 6, 192-201.
22
ACS Paragon Plus Environment

Page 23 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
34. Moore, R. A.; Bates, N. C.; Hancock, R. E. W. Interaction of Polycationic Antibiotics
with Pseudomonas Aeruginosa Lipopolysaccharide and Lipid a Studied by Using Dansyl-
Polymyxin. Antimicrob. Agents Chemother. 1986, 29, 496–500.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
35. Uppu, D. S. S. M.; Haldar, J. Lipopolysaccharide Neutralization by Cationic-Amphiphilic
Polymers through Pseudoaggregate Formation. Biomacromolecules. 2016, 17, 862-873.
36. Mozsolits, H.; Wirth, H. J; Werkmeister, J.; Aguilar, M. I. Analysis of Antimicrobial
Peptide Interactions with Hybrid Bilayer Membrane Systems Using Surface Plasmon
Resonance. Biochim. Biophys. Acta-Biomembr. 2001, 1512, 64-76.
37. Epand, R. F.; Maloy, W. L.; Ramamoorthy, A. ; Epand, R. M. Probing The “Charge
Cluster Mechanism” in Amphipathic Helical Cationic Antimicrobial Peptides. Biochemistry.
2010, 49, 4076–4084.
38. Loh, B.; Grant C.; Hancock, R. E. Use of the Fluorescent Probe 1-N-
Phenylnaphthylamine to Study the Interactions of Aminoglycoside Antibiotics with the Outer
Membrane of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 1984, 4, 546–551.
39. Sims, P. J.; Waggoner, A. S.; Wang, C. H.; Hoffman, J. F. Studies on the Mechanism by
Which Cyanine Dyes Measure Membrane Potential in Red Blood Cells and
Phosphatidylcholine Vesicles. Biochemistry. 1974, 13, 3315–3330.
40. Lewis, K. Platforms for Antibiotic Discovery. Nat. Rev. Drug Discov. 2013, 12, 371-387.
41. Bjarnsholt, T.; Ciofu, O.; Molin, S.; Givskov, M.; Høiby, N. Applying Insights from Biofilm
Biology to Drug Development - Can A New Approach be Developed? Nat. Rev. Drug
Discov., 2013, 12, 791-808.
42. Yu, K.; Lo, J. C.Y.; Yan, M.; Yang, X.; Brooks, D. E.; Hancock, R. E.W.; Lange, D.;
Kizhakkedathu, J. N. Anti-Adhesive Antimicrobial Peptide Coating Prevents Catheter
Associated Infection in A Mouse Urinary Infection Model. Biomaterials. 2017, 116, 69-81.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 38
1
2
3
4
5
6
7
8
9
43. Fisher, L. E.; Hook, A. L.; Ashraf, W.; Yousef. A.; Barrett, D.A.; Scurr, D. J.; Chen, X.;
Smith, E. F.; Fay, M.; Parmenter, C. D. J.; Parkinson, R.; Bayston. R. Biomaterial
Modification of Urinary Catheters with Antimicrobials to Give Long-Term Broad Spectrum
Antibiofilm Activity. J. Control Release 2015, 202, 57–64.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
44. Uppu, D. S. S. M.; Samaddar, S.; Ghosh, C.; Paramanandham, K.; Shome, B. R.; Haldar,
J. Amide Side Chain Amphiphilic Polymers Disrupt Surface Established Bacterial Bio-Films
and Protect Mice from Chronic Acinetobacter baumannii Infection. Biomaterials. 2016, 74,
131-143.
24
ACS Paragon Plus Environment

Page 25 of 38
Journal of Medicinal Chemistry
1
2
3
4
Figure Legends
5
6
7
8
Figure 1. (A) General molecular structure of Cholic Acid Peptide conjugates (CAPs) (1-20)
where three hydroxyl termini were derivatized with dipeptides of X-G- motif (X = any natural
amino acid, G = glycine) and carboxyl terminal is modified with benzyl group. (B) Molecular
structures of Valine-Glycine- (CAP 3), Tryptophan-Glycine- (CAP 10), Glutamic Acid-
Glycine-(CAP 15), and Lysine-Glycine- (CAP 18) derived CAPs.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. (A) Representative snapshot of MD simulations-derived structures in the
presence of model LPS membranes and different CAPs at 500 ns. (B) Membrane thickness
is plotted as a function of X-Y dimension of simulation box (top view) to quantify geometric
differences. The color bar indicated the thickness values with blue (lower) values towards
the spectrum shows membrane thinning.
Figure 3. (A) Mapping of root mean square fluctuation (RMSF) as a function of each atom
in CAPs 3, 10,15 and 18. (B) Superimposed conformations of four CAPs across different
time points in grey and representative coloured line conformation shows the top cluster in
each trajectory.
Figure 4. Global intra-molecular network for CAPs at 0 and 500 ns showing the CA
backbone and number of edges formed by each CAP at 0 and 500 ns. The
circles/edges/atom numbers are coloured in orange with different intensities (low to high),
with dark orange nodes connected to multiple atoms.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 38
1
2
3
4
5
6
7
8
9
Figure 5. (A) Change in anisotropy of Diphenylhexatriene (DPH) confirm the ability of CAP
3 to disrupt the membranes more effectively than other CAPs. DPH-doped model Gram-
negative bacterial membranes were incubated with CAPs and change in anisotropy was
measured by fluorescence. Data is presented as Mean  SD of three replicates and
statistical analysis was performed by unpaired two-tailed Student’s t-test. (B) LPS binding
abilities of CAPs show strongest binding of CAP 3 over other CAPs. Dansyl Polymyxin B-
LPS complexes were incubated with different CAPs and change in fluorescence of Dansyl-
Polymyxin was measured. Data is presented as Mean  SD of three replicates and statistical
analysis was performed by unpaired two-tailed Student’s t-test. (C) Change in fluorescence
of BODIPY show the strong interactions of CAP 3 with LPS causing its disintegration.
BODIPY-LPS aggregates were incubated with different CAPs and change in fluorescence
was measured. Data is presented as Mean  SD of three replicates and statistical analysis
was performed by unpaired two-tailed Student’s t-test. (D-F) Surface Plasmon Resonance
based sensorgrams of CAPs 3 (D), 15 (E) and 18 (F) confirm irreversible and strong binding
of CAP 3 over other CAPs. (G) Binding constants calculated from sensorgrams of 3, 15 and
18 after binding with bacterial membranes show strong association of CAP 3. Data is
presented as an average of three replicates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. (A) Change in fluorescence of N-phenyl naphthylamine (NPN) show enhanced
ability of CAP 3 to perturb the outer bacterial membranes as compared to other CAPs. NPN
stained E. coli were incubated with different CAPs (8 M) and change in fluorescence
intensity was measured with time. Data is presented as an average of three replicates. (B)
Time dependent change in DISC2(5) fluorescence in E. coli show better ability of CAP 3 to
permeabilize the inner bacterial membranes over other CAPs. DISC2(5)-labelled E. coli were
treated with different CAPs (8 M) and change in fluorescence intensity was measured with
26
ACS Paragon Plus Environment

Page 27 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
time. Data is presented as an average of three replicates. (C) Percentage of Propidium
iodide (PI) positive E. coli cells show maximum number of dead cells on CAP 3 treatment.
E. coli were treated with different CAPs at 32 M, stained with PI and analyzed by flow
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
cytometry. Data is presented as Mean  SD of four replicates and statistical analysis was
performed by unpaired two-tailed Student’s t-test. (D) Time kill assay confirm the bactericidal
effect of CAP 3 on E. coli as no colonies observed on treatment. E. coli was treated with 1X
and 4X MIC99 of CAP 3 and CFU analysis was performed at different time points. Data is
presented as Mean  SD of four replicates and statistical analysis was performed using two-
way ANOVA. (E) Transmission electron micrographs of untreated and CAP 3 (8 M) treated
E. coli at different time show membrane disruption in bacteria on CAP 3 treatment. (F) Fold
change in MIC99 of CAP 3 and neomycin on pre-treated E. coli show inability of the bacteria
to develop drug resistance. MIC99 of CAP 3 and neomycin was measured on pre-treated
samples using broth-dilution assay in four replicates. (G) Bactericidal effect of CAP 3 and
ampicillin treatment on persistent and stationary bacteria. Data is presented as Mean  SD
of two replicates and statistical analysis was performed by unpaired two-tailed Student’s t-
test. (H) Fold change in MIC99 of CAP 3 and Polymyxin at different CFUs show only 2-fold
increase in MIC99 of CAP 3 at 1012 CFU/mL. MIC99 of CAP 3 and Polymyxin was measured
at different CFUs using broth-dilution assay in four replicates.
Figure 7. (A-B) Dose-dependent effect of CAP 3 on colony forming units (A) and biofilm
biomass (B) of E. coli biofilms confirm bactericidal effect of CAP 3. Pre-formed biofilms were
treated with different doses of CAP 3 followed by CFU and biomass quantification using
crystal violet. Data is presented as Mean  SD of three replicates and statistical analysis
was performed by unpaired two-tailed Student’s t-test. (C) Representative fluorescence
micrographs of untreated and CAP 3-treated E. coli biofilms confirm bactericidal nature of
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 38
1
2
3
4
5
6
7
8
9
CAP 3 with dead bacteria shown in red stained by PI and live bacteria shown in green
stained by SYTO9. (D) Quantification of thickness of untreated and CAP 3-treated biofilms
shows significant reduction in thickness on CAP 3 treatment. Data is presented as Mean 
SD of three replicates and statistical analysis was performed by unpaired two-tailed
Student’s t-test. (E) Representative scanning electron micrographs of untreated and CAP 3
treated E. coli biofilms show disintegration of the biofilm structure on CAP 3 treatment.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. (A) CFU analysis showing the ability of CAP 3-coated catheters to prevent E. coli
biofilm formation during in vitro conditions. Untreated, DCM- and CAP 3-coated catheters
were incubated with bacteria for three days, and bacterial load on catheters was calculated
by CFU analysis. Data is presented as Mean  SD of three replicates and statistical analysis
was performed by unpaired two-tailed Student’s t-test. # no colonies were observed. (B)
CFU analysis showing the effect of CAP 3-coated catheters in BALB/c mice to prevent
biofilm formation by E. coli. Untreated, DCM and CAP 3-coated catheters were implanted in
BALB/c mice (n = 3/group) followed by E. coli infection, and CFU analysis was performed
after three days. Data is presented as Mean  SD of three mice and statistical analysis was
performed by unpaired two-tailed Student’s t-test. # no colonies were observed. (C) Effect
of CAP 3 in reducing the bacterial burden on wounds in BALB/c mice. E. coli infected wounds
were treated with Ampicillin, CAP 3 and Neomycin for four days and bacterial load was
quantified by CFU analysis. Data is presented as Mean  SD of three mice and statistical
analysis was performed by unpaired two-tailed Student’s t-test. # no colonies were
observed. (D) Bioluminescence images showing significant reduction in bioluminescence of
Xen 14 E. coli wound infections on CAP 3 treatment. Xen 14 E. coli infected wounds were
treated with CAP 3 for four days and bioluminescence imaging was performed. (E)
28
ACS Paragon Plus Environment

Page 29 of 38
Journal of Medicinal Chemistry
1
2
3
4
5
6
Antibacterial activities of CAP 3 against different clinical strains of E. coli, K. pneumoniae,
A. baumannii and its comparison with standard antibiotics. # no colonies were observed.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
29
ACS Paragon Plus Environment