Upload
trinhcong
View
233
Download
4
Embed Size (px)
Citation preview
1
1
2
Resurrecting Inactive Antimicrobial Peptides from Lipopolysaccharide (LPS) Trap 3
4
Harini Mohanram and Surajit Bhattacharjya# 5
6
School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, 7
Singapore 637551, Singapore 8
9
#Address correspondence to: Surajit Bhattacharjya, 60 Nanyang Drive, Singapore, 637551, 10
e-mail: [email protected], Fax: 65-6791-3856 11
12
Running Title: β-boomerang motif conquers LPS barrier 13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
AAC Accepts, published online ahead of print on 13 January 2014Antimicrob. Agents Chemother. doi:10.1128/AAC.02321-13Copyright © 2014, American Society for Microbiology. All Rights Reserved.
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
2
Abstract 28
Host defense antimicrobial peptides (AMPs) are a promising source of antibiotics for the 29
treatment of multiple drug-resistant pathogens. Lipopolysaccharide (LPS), the major 30
component of the outer leaflet of the outer membrane of Gram-negative bacteria, functions as 31
a permeability barrier against a variety of molecules including antimicrobial peptides (AMPs). 32
Further, LPS or endotoxin is the causative agent of sepsis killing 100,000 people per year in 33
the USA alone. LPS can restrict activity of AMPs inducing aggregations at the outer 34
membrane as observed for frog AMPs temporins and also in model AMPs. Aggregated 35
AMPs, as ‘trapped’ by the outer membrane, are unable to traverse through the cell wall 36
causing their inactivation. In this work, we show that these inactive AMPs can overcome LPS 37
induced aggregations while conjugated with a short LPS binding β-boomerang peptide motif 38
and become highly bactericidal. The generated hybrid peptides exhibit activity against Gram-39
negative and Gram-positive bacteria in high-salt and detoxify endotoxin. Structural and 40
biophysical studies establish mechanism of action of these peptides in LPS outer membranes. 41
Most importantly, this study provides a new concept for the development of potent broad 42
spectrum antibiotic with efficient outer membrane disruption as the mode of action. 43
44
45
46
47
48
49
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
3
Introduction 50
Survival of bacteria requires integrity of the outer layer or cell wall surrounding the plasma 51
membrane. The cell wall of Gram-positive bacteria contains a thick layer of a polymer, 52
termed peptidoglycan. The coarse meshwork of peptidoglycan provides a structural support 53
as well as balance osmotic pressure of cytosol and environment. By contrast, the cell wall of 54
Gram-negative bacteria is typified by an asymmetric outer membrane and a thin 55
peptidoglycan layer adjacent to the plasma or inner membrane. The inner leaflet of the outer 56
membrane contains phospholipids similar to cytoplasmic membrane, whereas, the outer 57
leaflet of the outer membrane is predominantly composed of lipopolysaccharide (LPS) [1, 2]. 58
The cell wall or peptidoglycan of Gram-positive bacteria does not maintain a permeability 59
barrier due to little resistance to the diffusion of antibiotics and antibacterial agents [3-5]. 60
While LPS faces the external environment, it acts as a permeability barrier for the outer 61
membrane of Gram-negative bacteria [3-5]. Molecules of <600 Da can freely diffuse through 62
outer membrane, however, higher molecular weight compounds are found to make a limited 63
access through the outer membrane [3-5]. LPS has an amphiphilic structure that can be 64
divided into three regions: a conserved lipid A part, a highly variable polysaccharide or O-65
antigen and a core oligosaccharide [1,2]. Lipid A domain, a hexa-acylated glucosamine based 66
phospholipid, anchors with the acyl chains of phospholipids of the inner leaflet of outer 67
membrane. The core oligosaccharide domain is centrally located that is covalently bonded 68
with lipid A and O-antigen. Apart from permeability barrier, LPS is also known as endotoxin, 69
a potent inducer of innate immune system in humans [6-8]. While in blood stream, LPS 70
recognizes a Toll-like receptor or TLR-4 in macrophages inducing production of variety of 71
inflammatory agents like TNF-α, IL1-β [9,10]. These inflammatory agents are necessary to 72
clear bacterial infection. However, a dysregulated immune system, often caused by a severe 73
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
4
Gram-negative infection, may result in over production of inflammatory molecules on-setting 74
lethal septic shock syndromes [9,10]. Septic shock accounts for nearly 100,000 deaths 75
annually in developed nations including USA [11,12]. At present treatment for septic shock is 76
rather limited. Consequently, there are intense interests to develop drugs against sepsis [13-77
16]. 78
Host defense antimicrobial peptides (AMPs), vital components of innate immune system, act 79
as a first line of defense against invading pathogens in most of the organisms [17-19]. AMPs 80
are considered to be highly important leads or potential therapeutics to control drug-resistant 81
bacterial pathogens [20-24]. Bacterial cell lysis caused by AMPs is an outcome of disruption 82
of the cytosolic membrane integrity. Cationic and hydrophobic characteristics of a vast 83
majority of AMPs are suitable for insertion and perturbation of anionic bacterial membranes 84
[25-27]. However, intra-cellular targets for AMPs are also identified [28-30]. Studies with 85
model membranes, resembling cytosolic membrane of bacteria, suggest that AMPs 86
mechanism of cell membrane perturbation may include barrel stave, torodial pore, carpet 87
mode or general interfacial mode of action [31-33]. It has also been realized that the cell wall 88
components of bacteria play important roles in insertion and translocation of AMPs towards 89
cytosolic membrane [34-36]. Several studies have demonstrated higher MIC values of AMPs 90
for Gram-negative bacteria in comparison to Gram-positive organisms [37-41]. In particular, 91
LPS of the outer membrane of Gram-negative bacteria has been found to be actively involved 92
in mode of action of AMPs [42-48]. Due to the LPS barrier there are less numbers of 93
antibiotic precursors against multiple drug resistant strains of Gram-negative in the drug 94
discovery pipelines [49, 50]. One of the mechanisms LPS-outer membrane employs toward 95
reduction of efficacy of AMPs through inducing self-association or aggregations of peptides 96
[51-54]. LPS induced aggregation of AMPs may result in their complete inactivation as 97
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
5
observed in temporins, a group of AMPs obtained from skin of European red frog and in 98
designed peptides [52-54]. Temporins are among the shortest AMPs demonstrating potential 99
in anti-bacterial therapeutics [54, 55]. However, a number of temporins including temporin-100
1Ta (TA) and temporin-1Tb (TB) are only active against Gram-positive bacteria [52-54]. 101
Similar observation has been made in designed Lys/Leu based AMP or KL12 whereby 102
peptide showed limited activity towards Gram-negative bacteria [51]. AMPs trapped in LPS 103
in oligomeric states are unable to traverse or translocate through the outer membrane possibly 104
causing their inactivation [54, 56, 57]. In this work, we demonstrate that a short LPS binding 105
peptide motif or β-boomerang motif, GWKRKRFG, designed in our laboratory [58,59] 106
rescues these inactive AMPs including TA, TB and KL12 peptide from LPS outer membrane. 107
Towards this, synthetic peptides are made containing β-boomerang motif at the C-termini and 108
investigated for bactericidal activities, cell lysis and mode of action. In the on-set of drug 109
resistance and LPS permeability, current study suggests that β-boomerang LPS binding motif 110
could be useful for the development of potent, salt resistant, non-toxic and broad spectrum 111
antimicrobial agents. 112
Materials and Methods 113
Peptides and rhodamine labeled peptides were synthesized commercially by GL Biochem 114
(Shanghai, China) and further purified by a reverse phase HPLC, Waters, using a C18 column 115
(300 Å pore size, 5 μm particle size). A linear gradient of acetonitrile/water mixture was used 116
for the purification and the peak eluting with highest purity was collected and freeze-dried. 117
LPS of Escherichia. coli 0111:B4, FITC-LPS of E. coli 055:B5, acrylamide, NPN (1-N-118
phenylnaphthylamine) were purchased from Sigma. The bacterial strains were obtained from 119
American Type Culture Collection, VA, USA. SYTOX green was obtained from Invitrogen. 120
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
6
Antibacterial assay: The minimum inhibitory concentration (MIC) of hybrid peptides was 121
determined using broth dilution method against four Gram-negative bacteria (Escherichia. 122
coli, Pseudomonas. aeruginosa ATCC 27853, Klebsiella. pneumoniae ATCC 13883, 123
Salmonella. enterica ATCC 14028) and four Gram-positive strains (Bacillus. subtilis, 124
Staphylococcus. aureus ATCC 25923, Streptococcus. pyogenes ATCC 19615 and 125
Enterococcus. faecalis ATCC 29212). Mid log phase cultures of these bacterial strains were 126
obtained either in Mueller-Hinton (MH) or LB broth and diluted to an OD600 of 0.01 127
(~106cfu/ml). Two-fold serial dilutions of the peptides were carried out in 96-well 128
polypropylene microtiter plates in a total volume of 50 µL at concentrations from 200 µM to 129
0.1 µM. To this 50 µL of the diluted cells in MH or LB broth were added and incubated for 130
18 hrs at 37oC. Water in the place of peptides acted as negative control. The MIC values were 131
computed spectrophotometrically and the concentration at which there is complete inhibition 132
of growth of bacteria was considered as MIC of the peptide. In order to dissect bacterial 133
killing and bacterial growth inhibiting activity, the wells before and after MIC values were 134
streaked onto MH agar plates and incubated overnight at 37oC. The well that does not show 135
visible bacterial growth was compared to the wells that were considered as MIC values. 136
Hemolytic assay: Blood was collected from healthy mice in a tube containing EDTA. RBCs 137
with EDTA were centrifuged at 800 × g for 10 min to remove the buffy plasma coat layer. 138
Then the resulting RBCs were resuspended in PBS (35 mM phosphate buffer, 150 mM NaCl, 139
pH 7.0) and washed three times. Around 50 μL of suspended RBCs were added to equal 140
volume of two-fold dilution of the peptides in 96-well micro titer plates and incubated for one 141
hour. The final erythrocyte concentration is 4% (V/V). After one hour, the mixture was 142
centrifuged and the release of hemoglobin in the supernatant was determined 143
spectrophotometrically at OD540. Buffer and 1% triton-X in the place of peptides served as 144
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
7
negative control and positive control respectively. The percentage of hemolysis was 145
calculated using the following formula: 146
Percentage of Hemolysis = (OD Peptide-OD buffer)/(OD triton-X-OD buffer)*100. 147
LPS neutralization assay: The strength of the peptides to neutralize LPS was assayed using 148
commercially available LAL chromogenic kit (QCL 100 Cambrex). The protocol explained 149
in the manufacturer’s instructions was strictly adhered. The endotoxic principle, LPS in 150
Gram-negative bacteria activates a proenzyme in Limulus amoebocyte lysate (LAL). This 151
activated enzyme in turn catalytically splits colored product para-nitroanilide (pNA) from the 152
colorless substrate Ac-Ile-Glu-Ala-Arg-para-nitroanilide and is detected 153
spectrophotometrically at OD410. The peptides were dissolved in the pyrogen free water 154
supplied with the kit and pH was adjusted to 7.0 with 1N HCl or 1N NaOH (which is 155
prepared in pyrogen free water). The increasing concentrations of the peptides were 156
incubated with 1.0 EU (Endotoxin units) in a total volume of 50 μL for 30 minutes at 37 oC. 157
About 50 μL of LAL reagent was added to peptide-EU complex and further incubated for 10 158
minutes followed by addition of 100 μL substrate. After incubation of six minutes for the 159
reaction, the release of colored product was recorded at OD410. Water in the place of peptides 160
served as negative control (blank) that is considered as 0% inhibition and percentage of LPS 161
neutralization was calculated by 162
% of LPS neutralization = [(OD blank- OD peptide)/OD blank]*100. 163
Fluorescence studies: All of the fluorescence studies were carried out in Cary Eclipse 164
fluorescence spectrophotometer (Varian, Inc) in 10 mM phosphate buffer, pH=7.0 unless 165
otherwise specified. 166
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
8
Aggregation state of the peptides using rhodamine fluorescence: 2 μM of the purified 167
rhodamine labeled peptides were added to increasing concentrations of LPS and the 168
fluorescence was monitored at excitation of 485 nm and emission at 550-620 nm in 10 mM 169
phosphate buffer, pH=7.0. 170
Membrane permeabilization assays: E. coli BL21 (DE3) cells were grown to mid 171
logarithmic phase in LB broth and diluted to an OD600 of 0.5. For outer membrane 172
permeabilization using NPN dye, about 500 μL of the cells were added to 10 μM of NPN (1-173
N-phenylnaphthylamine) dissolved in acetone) and basal fluorescence was recorded with 174
excitation at 350 nm and emission maximum at 420 nm. An increasing concentration of 175
hybrid peptides was added to cells with NPN and fluorescence intensity after addition of each 176
concentration was recorded. For membrane permeability using SYTOX green dye, 0.5 OD600 177
cells were added to 1 μM dye and incubated with shaking at 37 oC for 15 min. Basal 178
fluorescence was recorded by excitation at 485 nm and emission at 520 nm followed by the 179
addition of increasing concentration of peptides. 180
Membrane depolarization: The assay to measure membrane depolarization ability was 181
performed in intact E. coli cells and its spheroplast. For intact cells, the bacteria were grown 182
to mid log phase, centrifuged and suspended in 5 mM HEPES and 20 mM glucose at pH=7.4. 183
After a brief wash, the cell pellets were suspended in the same buffer containing 100 mM 184
KCl to an OD600 nm of 0.05. Spheroplast was prepared by suspending the log phase grown 185
bacterial pellets in 10 mM Tris, 25 % sucrose pH=7.4. The pellets were washed twice and 186
resuspended in the same buffer containing 1 mM EDTA. The cells were incubated for 15 min 187
with shaking and centrifuged. The pellets were immediately dissolved in ice-cold water and 188
further incubated for 10 min at 4oC. The cells devoid of outer membrane i.e. spheroplasts 189
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
9
were collected by centrifugation and the pellets were dissolved to an OD600 of 0.05 in same 190
buffer (5 mM HEPES, 20 mM glucose, 100 mM KCl) as intact cells. The depolarization 191
detecting dye DiS-C3-5 (3,3`-diethylthiodicarbocyanine iodide) dye was added to either intact 192
cells or spheroplasts and basal fluorescence was recorded by exciting the dye at 622 nm and 193
emission collected at 630-700 nm. This was incubated for about 45-60 minutes until the 194
decrease in fluorescence in stable i.e. the dye partitions into the membrane. This was then 195
followed by the addition of increasing concentration of different peptides and the 196
depolarizing ability was recorded as the measurement of increase in fluorescence intensity. 197
Measurement of dissociation of LPS micelles: The ability of the conjugated peptides to 198
dissociate LPS micelles was studied fluorometrically using FITC conjugated LPS and also by 199
dynamic light scattering (DLS) measurements. In FITC-LPS, the fluorescence of FITC is 200
self-quenched in LPS micelles which upon dissociation of micelles would dequench. Basal 201
fluorescence of 0.5 μM of FITC-LPS was recorded at an excitation of 480 nm and emission 202
at 520 nm. The dissociation of LPS micelles was recorded as a function of increase in 203
fluorescence intensity with addition of increasing concentration of peptides. In DLS 204
measurements, the distribution of various sizes of LPS micelles was determined with 0.5 μM 205
of LPS. The change in this distribution was observed after addition of peptides in different 206
ratios e.g. 1:0.5, 1:1 and 1:2 (LPS: Peptide). The scattering was measured with Dynamic 207
Light Scattering software provided with the instrument (Brookhaven Instruments Corp., 208
Holtsville, NY) and the scattering data was analyzed with CONTIN method. 209
Intrinsic tryptophan fluorescence and acrylamide quenching: The binding of hybrid peptides 210
to LPS was determined using tryptophan fluorescence. Tryptophan is an intrinsic fluorescent 211
probe and is extremely sensitive to polarity of the environment. About 5 μM of the peptide 212
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
10
was taken with 500 μL buffer and the emission of tryptophan fluorescence was noted in free-213
state with excitation at 280 nm and emission at 300-400 nm. Each peptide sample was titrated 214
with increasing concentrations of LPS and fluorescence emission was recorded from 300-400 215
nm. Fluorescence quenching of tryptophan residue was carried out by adding increasing 216
concentrations of acrylamide from stock of 5 M to solution containing only peptide and also 217
to peptide-LPS (1:4) complexes. The Stern-Volmer constant (Ksv) values were then 218
calculated using F0/F = 1+Ksv [Q] where F0 and F are the fluorescence intensities before and 219
after addition of quencher and [Q] is the quencher concentration. 220
Circular dichroism spectroscopy: Conformational change of the peptides upon binding to 221
membranes was detected using CD spectroscopy. Data were collected using a Chirascan CD 222
spectrometer (Applied Photophysics Ltd., UK). The peptide and peptide/micelle complex 223
were scanned from 190 to 240 nm wavelengths in a 0.01 cm path length cuvette for an 224
average of 3 scans. Baseline scans were acquired using 10 mM phosphate buffer, pH=7.0. 225
Samples containing only LPS were also acquired with same settings. About 25 μM of 226
peptides were used with 30 μM of LPS to obtain secondary structure of the peptides. The 227
appropriate baselines were used to subtract the data and the corrected data were converted to 228
molar ellipticity (deg.cm2.dmol-1). 229
Isothermal titration calorimetry (ITC): Binding of peptides with LPS micelles were 230
determined with ITC using VP-ITC Micro calorimeter (Microcal Inc, Northampton, MA). 231
Peptides and LPS were dissolved in 10 mM phosphate buffer, pH=7.0 and filtered. LPS at a 232
concentration of 10 μM was loaded into the sample cell and the reference cell was filled with 233
the buffer. The syringe was filled with 1mM peptide stock. Typically 25 injections of 3.5 μL 234
of the peptides were made into the sample cell at 25oC. The sample cell was stirring 235
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
11
continuously at 300 rpm. Raw data was collected and fitted using single site binding model in 236
Microcal Origin 5.0 software (Origin Lab Corporation, Northampton, MA). Association 237
constant (Ka) and enthalpy change (ΔH) were directly obtained from the software used. ΔG 238
and TΔS were calculated using the fundamental equations of thermodynamics, ΔG = -RTlnKa 239
and TΔS=(ΔH-ΔG) respectively. 240
Electron microscopy: Mid log phase grown E. coli cells were incubated with various 241
concentrations of LG21 peptide (3 μM, 8 μM and 15 μM) and 50 μM of LG21R19A for 2 242
hours. The cells were centrifuged and the pellets were dissolved in 10 μL 10 mM Phosphate 243
buffer, pH 7.0 and a drop containing the bacteria were loaded onto carbon-coated EM grids. 244
The grids were then negatively stained with 2% phosphotungstic acid and examined using 245
Jeol JEM-1230 electron microscope. 246
247
Results and Discussion 248
Peptide Design: In previous studies, we have designed de novo a series of 12-amino acid long 249
cationic/hydrophobic peptides with antimicrobial and anti-endotoxic activity [58, 59]. A 250
sequence motif WKRKRF located at the center of the primary structures of the designed 251
peptide has been demonstrated to be critical for the structure in LPS and activity. An octa-252
peptide or GG8, G-WKRKRF-G, interacted with LPS and adopted boomerang-like 253
conformation in LPS. Amphipathic structure of the boomerang motif is demarcated by close 254
packing of the aromatic side chains of residues W and F whereas the side chains of the four 255
basic amino acids are distally located. GG8 peptide or boomerang motif is devoid of 256
antimicrobial and antiendotoxic activities, but its atomic resolution structure in LPS indicated 257
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
12
specific interactions. We have surmised that inclusion of boomerang motif sequence in TA, 258
TB and KL12 based peptide may abolish LPS induced aggregation yielding broad spectrum 259
AMPs (Fig. 1). We have obtained synthetic hybrid peptides of TA, TB, KL12 containing the 260
boomerang motif at the C-terminus (Table 1). Two analog peptides containing Ala 261
replacements are also synthesized to understand role of the β-boomerang motif sequence in 262
activity (Table 1). 263
Antimicrobial Activity and Red Blood Cell Lysis by Hybrid Peptides: Table 2 summarizes 264
toxicity of the hybrid peptides, LG21, FG21, KG20 and LG21R19A and LG21W15AF20A 265
against bacterial strains, in MH broth, and red blood cells. Remarkably, hybrid peptides, 266
LG21, FG21 and KG20, demonstrate potent antibacterial activity, with low MIC values, 267
including Gram-negative strains in comparison to the parent peptides (Table 2). It may be 268
noted that the parent peptides, TA, TB and KL12, were found to be largely ineffective against 269
Gram-negative bacteria [51-53, 60]. Interestingly, the mutated analogs of LG21, LG21R19A 270
and LG21W15AF20A exhibited largely limited bacterial cell killing activity (Table 2). The 271
role of outer membrane towards bactericidal activity of the parent peptides and hybrid 272
peptides correlate well with the membrane depolarization, studied with a florescent dye DiS-273
C3-5, of E. coli cells and spheroplast i.e. E. coli cells lacking outer membrane [51]. The 274
hybrid peptides efficiently depolarizes E. coli cells and spheroplast, whereas, parent peptides 275
show lower depolarization of E. coil cells and more depolarization of the spheroplast (Fig. 276
S1). The LG21R19A peptide was least active to depolarize the cells and spheroplast (Fig. S1). 277
These data indicate that the outer membrane, as a permeability barrier, plays essential roles in 278
insertion of the peptides. 279
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
13
In hemolysis assays, the hybrid peptides, LG21, FG21 and KG20, demonstrate very low 280
activity (Table 2). At 100 µM concentration of peptides, lysis of red blood cells has been 281
estimated to be only <10%. However, a somewhat higher hemolysis has been detected for the 282
mutated analogs LG21R19A and LG21W15AF20A peptides (Table 2). 283
We have also tested antibacterial activity of the active hybrid peptides LG21, FG21 and 284
KG20 in LB medium containing 150 mM NaCl. As can be seen, LG21 and FG21 peptides are 285
able to exert antimicrobial activity even in the presence of salt, with MIC value ranging from 286
8 to 10 µM, against test bacteria (Table 3). However, KG20 peptide has been observed to be 287
salt sensitive showing a rather high MIC values (Table 3). A number of AMPs are known to 288
be inhibited under physiological salt concentrations including β-defensins, magainins, 289
indolicidin etc. [61-63]. Taken together, these results demonstrate that the hybrid peptides, 290
LG21, FG21 and KG20 are bestowed with broad spectrum of bactericidal activity and 291
notably activities are retained, for LG21 and FG21, even in the presence of salt. Further, it 292
may be noted that TA and TB were found to be poorly hemolytic whereas KL12 peptide 293
exhibited strong hemolysis, ~80% at 100 µM concentration [51, 55]. In other words, LG21 294
and FG21 retain low hemolytic activity akin to TB and TA peptides. Further, the inclusion of 295
the boomerang motif in KL12 peptide has significantly lowered the hemolytic property of the 296
hybrid KG20 peptide. The lowered hemolytic activity of the KG20 peptide, in comparison to 297
KL12, may be arising due to the preferential interactions, as demonstrated in previous studies 298
[58,59], of the boomerang motif of the hybrid peptide with negatively charged lipids of 299
bacterial membranes over zwitterionic lipids in mammalian cells. The inability of the mutated 300
peptides, LG21R19A and LG21W15AF20A, to exert bactericidal effect implicates the vital 301
role of the cationic residues and aromatic residues W15 and F20 in the short boomerang 302
motif sequence for the activity of the hybrid peptides. It may be noted that despite the same 303
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
14
cationicity of the LG21W15AF20A peptide akin to LG21, the mutated peptide lacks 304
bactericidal activity. Most strikingly, a single mutation of R19A in LG21R19A peptide also 305
severely reduces the bactericidal activity. The poor bactericidal activity and relatively higher 306
hemolytic activity of the mutated peptides, LG21W15AF20A and LG21R19A, may be 307
related the altered interactions and structures in the context of different cell types. To gain 308
potential correlation in structure-activity, we have utilized LG21R19A peptide along with 309
LG21, FG21 and KG20 peptides for further studies as described below. 310
Endotoxin Neutralization by Hybrid Peptides: LPS or endotoxin neutralization by the hybrid 311
peptides are examined using LAL assays [64]. LAL assay is highly sensitive in detecting free 312
LPS at concentration as low as 1pM [65]. LPS neutralizing proteins and peptides can 313
sequester LPS reducing free endotoxin in solution [66,67]. All three hybrid peptides, LG21, 314
FG21 and KG20, demonstrate inhibitory activity in neutralizing LPS (Fig. S2). The LG21 315
peptide shows a higher potency in endotoxin neutralization in comparison to FG21 and KG20 316
(Fig. S2). LG21 exhibits inhibition of LPS even at 3 µM concentration. At 10 µM 317
concentration of LG21 >75% inhibition of LPS has been detected. Overall, there has been an 318
increase of inhibition with increasing concentrations of LG21, FG21 and KG20 peptides (Fig. 319
S2). The mutated peptide LG21R19A does not show LPS inhibitory activity in a significant 320
way (Fig. S2). Note, previous study showed that LPS neutralization ability of TB and TA 321
peptides is highly limited [53]. These hybrid peptides, LG21, FG21 and KG20, contain 322
ability to neutralize toxicity of LPS. It may be worthwhile to consider that the incorporation 323
of the boomerang motif in other AMPs may enhance their antimicrobial and anti-endotoxic 324
activities with potentially lowering toxicity. 325
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
15
Self-association of Hybrid Peptides in LPS: Rhodamine labeled peptides are utilized to 326
assess self-associations of hybrid peptides in presence of LPS. The fluorescence emission 327
intensity of rhodamine is highly sensitive to molecular aggregations whereby aggregations of 328
rhodamine labeled peptides would show a decrease in fluorescence intensity of the 329
fluorophore due to self-quenching [52]. Fig. 2 shows a plot indicating differences in 330
fluorescence intensity (ΔF) of rhodamine, measured at emission maxima of 590 nm, for 331
rhodamine labeled peptides in the absence of LPS and in presence of LPS, at 1 µM, 2 µM, 4 332
µM, 8 µM and 10 µM concentrations. As can be seen, there has been an increase in 333
fluorescence intensity of rhodamine for LG21, FG21 and KG20 hybrid peptides with 334
increasing concentrations of LPS, demonstrating plausible absence of aggregations in LPS for 335
these peptides. By contrast, fluorescence intensity of rhodamine has been found to be 336
decreased for LG21R19A peptide, implying aggregation of the mutated peptide in complex 337
with LPS (Fig. 2). Collectively, aforementioned results suggest that the incorporation of LPS 338
binding boomerang motif abolish LPS induced aggregations in hybrid peptides. The ability to 339
kill Gram-negative bacterial strains by the hybrid peptides correlate well with their lack of 340
aggregations in LPS. The self-association of LG21R19A peptide in LPS appears to be 341
responsible for its reduced bactericidal activity. 342
Permeation of E. coli Cell Membrane and Membrane Disruption by Hybrid Peptides: In 343
order to determine membrane damage caused by the hybrid peptides, we have carried out 344
fluorescence studies, with E. coli cells, employing membrane probes: NPN, SYTOX green 345
NPN and SYTOX green are unable to enter into an intact cell unless the membrane integrity 346
is compromised by additions of membrane disrupting compounds. The fluorescence intensity 347
of NPN delineate a drastic increase while bound to membrane lipid components whereas 348
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
16
fluorescence intensity of the cationic dye SYTOX green enhances in complex with the intra-349
cellular nucleic acids. 350
Fig. 3 (panels A,B) shows difference in fluorescence intensity (ΔF) of NPN (Fig. 3A), 351
SYTOX green (Fig. 3B) vs concentrations of hybrid peptides. As can be seen, fluorescence 352
intensities of NPN and SYTOX green are significantly increased upon additions of increasing 353
concentrations of LG21, FG21 and KG20 peptides. These results strongly demonstrate that 354
the hybrid peptides, LG21, FG21 and KG20, efficiently disrupt integrity of bacterial outer 355
and inner membranes causing an influx of fluorescent molecules NPN and SYTOX green 356
into the cells. Notably, in these assays, there are no significant changes of fluorescence 357
intensity of these probes upon treatment with inactive LG21R19A peptide (Fig. 3, panels A, 358
B). Further, EM images are obtained for E. coli cells either in absence or upon treated with 359
LG21 peptide at three different concentrations, 3 µM, 8 µM and 15 µM and LG21R19A at 50 360
μM concentration for 2 hours (Fig. 4). Morphology of the cells of the LG21 peptide treated 361
bacteria is distinctly modified in comparison to control (Fig. 4). Plausible membrane or cell 362
wall damage can be seen at 3 µM (close to MIC) concentration of LG21 (Fig. 4B), in 363
comparison to untreated cells (Fig. 4A). At higher concentrations, 8 µM and 15 µM, of LG21 364
cell shape and integrity changes and ghost like images are clearly observed (Fig. 4C and Fig. 365
4D). By contrast, the inactive LG21R19A peptide does not impart any discernable damage to 366
the E. coli cells (Fig. 4E). 367
Binding Affinity of Hybrid Peptides with LPS: In order to better understand superior activity 368
of hybrid peptides against Gram-negative bacteria, LPS binding parameters are obtained 369
using ITC experiments (Fig. S3). LPS-peptide interactions are endothermic in nature for the 370
active peptides, LG21, FG21 and KG20, as disclosed by upward trend to the ITC heat peaks 371
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
17
(Fig. S3 A-C, top panels). Usually, entropy driven or endothermic binding has been observed 372
for LPS-AMPs interactions with gel phase of LPS at 25oC [68,69]. Interestingly, the inactive 373
peptide LG21R19A shows an opposite trend whereby LPS interactions appear to be 374
exothermic in nature (Fig. S3 panel D, top). The apparent dissociation constant (Kd) values 375
and thermodynamic parameters of LPS-peptide interactions are estimated from ITC data 376
(Table 4). The hybrid peptides, LG21, KG21 and FG20, interacted with LPS with sub micro-377
molar affinity with Kd of 0.6 µM, 0.6 µM and 0.3 µM, respectively, whereas, LG21R19A 378
delineates a comparatively much weaker binding to LPS with Kd of 10 µM (Table 4). 379
Localization of Hybrid Peptides in LPS: We have further assessed insertion of hybrid 380
peptides in LPS micelles using intrinsic tryptophan fluorescence experiments. Fluorescence 381
emission spectra of Trp for LG21, FG21, KG20 and LG21R19A were obtained at various 382
concentrations of LPS (ranging from 1 µM to 20 µM) (Fig. S4). As seen, tryptophan residue 383
has experienced a marked blue shift, i.e. emission maximum shifted toward a shorter 384
wavelength in comparison to free peptide, for LG21, FG21 and KG20 peptides in LPS. In 385
marked contrast, tryptophan residue of the mutated peptide LG21R19A exhibited a highly 386
limited spectral shift in LPS (Fig. S4, Table 5). The solvent accessibility of the tryptophan 387
residue was judged by fluorescence quenching studies in presence of acrylamide either in free 388
solutions or in LPS containing solutions. The quenching constant (Ksv) values of LG21, 389
FG21 and KG20 peptides in LPS are estimated to be significantly lower in comparison to 390
Ksv values determined in free solution (Table 5). By contrast, LG21R19A peptide showed 391
comparable Ksv values for tryptophan residue in LPS and in buffer solution (Table 5). These 392
observations demonstrate that in active peptides, LG21, FG21 and KG20, tryptophan residue 393
is inserted into non-polar environment of LPS with restricted exposure to solvent whereas 394
tryptophan residue of the inactive peptide, LG21R19A, lacks non-polar environment of acyl 395
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
18
chains of LPS. In other words, the LG21R19A peptide is predominantly localized at the outer 396
surface or hydrophilic sugar region of LPS with higher degree of solvent exposure. 397
Disaggregation of LPS Aggregates by Hybrid Peptides: High-binding affinity and insertion 398
into hydrophobic lipid A domain of LPS of the active hybrid peptides may cause structural 399
changes of LPS. We have examined perturbation of LPS using dynamic light scattering (DLS) 400
and fluorescence of FITC labelled LPS. DLS experiments examine size distribution of LPS 401
aggregates in presence of peptides (Fig. 5). In free solution, LPS is highly polydisperse with 402
an average diameter of 806 nm (Fig. 5A). There has been a drastic reduction in size of LPS 403
aggregates with concomitant reduction in polydispersity in the presence of LG21 (Fig. 5B), 404
FG21 (Fig. 5C) and KG20 (Fig. 5D) peptides. By contrast, changes of LPS aggregates appear 405
to be largely limited; with an average diameter of 521 nm, in the presence of mutated 406
LG21R19A peptide (Fig. 5E). The average diameter of LPS micelles has been estimated, for 407
each of the active peptide, to be 85 nm, 87 nm and 134 nm for LG21, FG21 and KG20, 408
respectively. Hybrid peptide induced disaggregation of LPS micelles has also been observed 409
from fluorescence intensity of FITC in FITC conjugated LPS. Fluorescence intensity of FITC 410
in FITC-LPS is largely quenched in solution due to aggregation of LPS. Binding of proteins 411
or peptides causing perturbation of LPS aggregated structures may enhance fluorescence 412
intensity [59,70,71]. Proteins or peptides mediated dissociation of LPS have been correlated 413
with their anti-endotoxin activity [59,70,71]. Figure 5F shows change in fluorescence 414
intensity (ΔF) of FITC-LPS with increasing concentrations of LG21, FG21, KG20 and 415
LG21R19A peptides. As can be seen, there have been drastic increases in ΔF for active 416
hybrid peptides, LG21, FG21 and KG20, in a dose dependent manner, indicating potential 417
dissociation of LPS aggregates (Fig. 5F). On the other hand, the inactive peptide, LG21R19A, 418
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
19
does not cause dissociation of LPS as suggested by the lack of increase of fluorescence 419
intensity of FITC (Fig. 5F). 420
Secondary Structures of the Hybrid Peptides: CD spectroscopy, at far UV region (240 -190 421
nm), was carried out to assess global conformations of hybrid peptides in free solution and in 422
complex with LPS micelles. CD spectra, in free solution, of the hybrid peptides, LG21, FG21, 423
KG20 and LG21R19A, delineate a single band ~195 nm, indicating random conformations 424
(Fig. S5). CD spectra obtained for the hybrid peptides and mutant analog either in LPS or in 425
detergent micelles are characterized by two negative bands at ~ 220-225 nm and ~208-210 426
nm, indicating predominantly helical conformations. Intriguingly, active hybrid peptides, 427
LG21, FG21 and KG20 as well as the inactive mutant LG21R19A assume helical 428
conformations in LPS. It may be noted that TA, TB and KL peptides demonstrated helical 429
conformations in LPS and random conformations in water [51,53]. In other words, 430
irrespective of the activity, interactions of these peptides with LPS induce conformational 431
transitions into helical states. At present, it is not clear from the CD studies whether the LPS 432
binding motif assumes the boomerang-like conformations in LPS micelles. Atomic resolution 433
structures of these hybrid peptides and the mutated peptides in complex with LPS micelles 434
are required to be determined for specific structural features. 435
Conclusion 436
The lower activity of AMPs toward Gram-negative bacteria may arise from their self- 437
association in LPS outer membrane. In this work, we demonstrate that an LPS binding 438
peptide motif when conjugated with Gram-negative inactive AMPs temporins and KL12 439
abolished LPS-induced aggregations yielding broad spectrum salt resistant hybrid AMPs. 440
Hybrid peptides demonstrate neutralization of endotoxin and poor in hemolysis. As a mode of 441
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
20
action, hybrid peptides bind LPS with high-affinity and disrupt the higher order structures of 442
LPS. These traits of the hybrid peptides are correlated with efficient disruption of the outer 443
membrane and cell permeabilization. The boomerang motif is critically involved in broad 444
spectrum activity of the hybrid peptides as a single mutation LG21R19A in LG21 produced 445
an inactive peptide that shows self-association in LPS. The lack of bactericidal and endotoxin 446
neutralization activity of the LG21R19A peptide may stems from its low affinity binding to 447
LPS micelles and limited perturbation of LPS structural states. We surmise that hybrid 448
peptides obtained in this work and their mode of action through LPS outer membrane would 449
be useful to design new class of cell wall permeabilizing AMPs. Such AMPs can be tested 450
further for in vivo efficacy in animal models. 451
Acknowledgment: This work is supported by a grant from the Ministry of Education (MOE), 452
Singapore (RG11/12). 453
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
1
References 454
1. Raetz CRH. 1990. Biochemistry of endotoxins. Annu Rev Biochem 59: 129-170. 455
2. Nikaido H. 1999. Outer membrane in Escherichia coli and Salmonella typhimurium. In 456
Neidhardt FC. Editor. American Society for Microbiology, Washington, DC. pp 29-47. 457
3. Nikaido H. 1994. Prevention of drug access to bacterial targets: permeability barriers and 458
active efflux. Science 264: 382-388. 459
4. Nikaido H, Vaara M. 1985. Molecular basis of bacterial outer membrane permeability. 460
Microbiol. Rev 49: 1-32. 461
5. Snyder DS, McIntosh TJ. 2000. The lipopolysaccharide barrier: correlation of antibiotic 462
susceptibility with antibiotic permeability and fluorescent probe binding kinetics. 463
Biochemistry 39: 11777-11787. 464
6. Cohen J. 2002. The immunopathogenesis of sepsis. Nature 420: 885-891. 465
7. Beutler B, Rietschel ET. 2003. Innate immune sensing and its roots: the story of endotoxin. 466
Nat Rev Immunol 3: 169-176. 467
8. Ward PA. 2004. The dark side of C5a in sepsis. Nat Rev Immunol 4: 133-142. 468
9. Miller SI, Ernst RK, Bader MW. 2005. LPS, TLR4 and infectious disease diversity. Nat 469
Rev Microbiol 3: 36-46. 470
10. Oberholzer A, Oberholzer C, Moldawer LL. 2001. Sepsis Syndromes: Understanding the 471
Role of Innate and Acquired Immunity. Shock 16: 83–96. 472
11. Martin GS, Mannino DM, Eaton S, Moss M. 2003. The epidemiology of sepsis in the 473
United States from 1979 through 2000. N Engl J Med 348: 1546-1554. 474
12. Angus DC, Wax RS. 2001. Epidemiology of sepsis: an update. Crit Care Med 29: S109-475
S116. 476
13. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriguez A, 477
Steingrub JS, Garber GE, Helterbrand JD, Ely EW, Fisher Jr CJ. 2001. Efficacy and 478
safety of recombinant human activated protein C for severe sepsis. N Engl J Med 344: 699–479
709. 480
14. Oberholzer A, Oberholzer C, Moldawer LL. 2001. Sepsis syndromes: understanding the 481
role of innate and acquired immunity. Shock 16: 83–96. 482
15. Buras JA, Holzmann B, Sitkovsky M. 2005. Animal models of sepsis: setting the stage. Nat 483
Rev Drug Disc 4: 854–865. 484
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
2
16. Angus DC. 2011. The search for effective therapy for sepsis: back to the drawing board? 485
JAMA 306: 2614-2615. 486
17. Zasloff M. 2002 Antimicrobial peptides of multicellular organisms. Nature 415: 389–395. 487
18. Fjell CD, Hiss JA, Hancock RE, Schneider G. 2011. Designing antimicrobial peptides: 488
form follows function. Nat Rev Drug Disc. 11: 37-51. 489
19. Fox JL. 2013. Antimicrobial peptides stage a come back. Nature Biotechnol. 31: 379-382. 490
20. Peters BM, Shirtliff ME, Jabra-Rizk MA. 2010. Antimicrobial Peptides: Primeval 491
Molecules or Future Drugs? PLoS Pathogens 6: e1001067 492
21. Yount NY, Yeaman MR. 2013. Peptide antimicrobials: cell wall as a bacterial target Ann 493
NY Acad Sci 1277:127-38. 494
22. Hancock RE, Sahl HG. 2006. Antimicrobial and host-defense peptides as new anti-infective 495
therapeutic strategies. Nat Biotechnol. 24: 1551–1557. 496
23. Giuliani A, Pirri G, Nicoletto S. 2007. Antimicrobial peptides: an overview of a promising 497
class of therapeutics. Cent Eur J Biol. 2: 1–33. 498
24. Shai Y. 2002. From innate immunity to de-novo designed antimicrobial peptides. Curr Pharm 499
Des 8: 715–725. 500
25. Huang HW. 2000. Action of antimicrobial peptides: Two-state model. Biochemistry 501
39: 8347–8352. 502
26. Rakowska PD, Jiang H, Ray S, Pyne A, Lamarre B, Carr M, Judge PJ, Ravi J, Gerling 503
UI, Koksch B, Martyna GJ, Hoogenboom BW, Watts A, Crain J, Grovenor CR, 504
Ryadnov MG. 2013. Nanoscale imaging reveals laterally expanding antimicrobial pores in 505
lipid bilayers. Proc Natl Acad Sci U S A 10: 8918-8923. 506
27. Nguyen LT, Haney EF, Vogel HJ. 2011. The expanding scope of antimicrobial peptide 507
structures and their modes of action. Trends Biotechnol. 9: 464-472. 508
28. Harris M, Mora-Montes HM, Gow NA, Coote PJ. 2009. Loss of mannosylphosphate from 509
Candida albicans cell wall proteins results in enhanced resistance to the inhibitory effect of a 510
cationic antimicrobial peptide via reduced peptide binding to the cell surface. Microbiology 511
155: 1058–1070. 512
29. Nicolas P. 2009. Multifunctional host defense peptides: intracellular-targeting antimicrobial 513
peptides. FEBS J 276: 6483-6496. 514
30. Yeaman MR, Yount NY. 2003. Mechanisms of antimicrobial peptide action and 515
resistance. Pharmacol Rev 55: 27–55. 516
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
3
31. K. Matsuzaki. 1998. Magainins as paradigm for the mode of action of pore forming 517
polypeptides. Biochim. Biophys. Acta 1376: 391–400. 518
32. Oren Z, Shai Y. 1998. Mode of action of linear amphipathic alpha-helical antimicrobial 519
peptides, Biopolymers 47:451–463. 520
33. Wimley WC. 2010. Describing the mechanism of antimicrobial peptide action with the 521
interfacial activity model. ACS Chem Biol 5: 905–917. 522
34. Sochacki KA, Barns KJ, Bucki R, Weisshaar JC. 2011. Real-time attack on single 523
Escherichia coli cells by the human antimicrobial peptide LL-37. Proc Natl Acad Sci U S A. 524
108:e77-81. 525
35. Chunhong Li, Loren PB, Collin DD, Barry MW, Glenn WA, Paul BS. 1999 Incremental 526
conversion of outer-membrane permeabilizers into potent antibiotics for gram-negative 527
bacteria J Am Chem Soc. 121: 931-940. 528
36. Hancock, REW. 1984 Alterations in outer membrane permeability. Annu Rev Microbiol 38: 529
237-264. 530
37. Oren Z, Shai Y. 1997. Selective lysis of bacteria but not mammalian cells by diastereomers 531
of melittin: structure-function study. Biochemistry 36: 1826-1835. 532
38. Steiner H, Hultmark D, Engstrom A, Bennich H, Boman HG. 1981. Sequence and 533
specificity of two antibacterial proteins involved in insect immunity Nature 292: 246–248. 534
39. Boman, HG. 2000. Innate immunity and normal microflora. Immunol Rev 173: 5–16. 535
40. Biswajit M, Guangshun W. 2012. Ab initio design of potent anti-MRSA peptides based on 536
Database Filtering Technology. J Am Chem Soc. 134:12426-12429. 537
41. Raja Z, André S, Piesse C, Sereno D, Nicolas P, Foulon T, Oury B, Ladram A. 2013. 538
Structure, antimicrobial activities and mode of interaction with membranes of Bovel 539
Phylloseptins from the painted-belly leaf frog, Phyllomedusa sauvagii. PLoS One 8: e70782. 540
42. Snyder S, Kim D, McIntosh TJ. 1999. Lipopolysaccharide bilayer structure: effect of 541
chemotype, core mutations, divalent cations, and temperature. Biochemistry 38: 10758-542
10767. 543
43. Snyder DS, McIntosh TJ. 2000. The lipopolysaccharide barrier: correlation of antibiotic 544
susceptibility with antibiotic permeability and fluorescent probe binding kinetics. 545
Biochemistry 39: 11777-11787. 546
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
4
44. Labischinski H, Barnickel G, Bradaczek H, Naumann D, Rietschel ET, Giesbrecht P. 547
1985. High state of order of isolated bacterial lipopolysaccharide and its possible contribution 548
to the permeation barrier property of the outer membrane. J Bacteriol. 162: 9-20. 549
45. Allende D, McIntosh TJ. 2003. Lipopolysaccharides in bacterial membranes act like 550
cholesterol in eukaryotic plasma membranes in providing protection against melittin-induced 551
bilayer lysis. Biochemistry 42: 1101-1108. 552
46. Bhunia A, Ramamoorthy A, Bhattacharjya S. 2009. Helical hairpin structure of a potent 553
antimicrobial peptide MSI-594 in lipopolysaccharide micelles by NMR spectroscopy. 554
Chemistry 15: 2036-2040 555
47. Bhunia A, Domadia P, Torres J, Hallock KJ, Ramamoorthy A, Bhattacharjya S. 2010. 556
NMR structure of pardaxin, a pore-forming antimicrobial peptide, in lipopolysaccharide 557
micelles: mechanism of outer membrane permeabilization. J Biol Chem. 285: 3883-3895. 558
48. Domadia P, Bhunia A, Ramamoorthy A, Bhattacharjya S. 2012. Struture, Interactions 559
and antibacterial activities of MSI-594 derived mutant peptide MSI-594F5A in 560
lipopolysaccharide micelles: role of helical hairpin conformation in outer membrane 561
permeabilizaiton. J Am Chem Soc. 132:18417-18428. 562
49. Fischbach MA, Walsh CT. 2009. Antibiotics for emerging pathogens. Science 325: 1089-563
1093. 564
50. Taubes G 2008 The bacteria fight back. Science 321: 356-361. 565
51. Papo N, Shai Y. 2005 A molecular mechanism for lipopolysaccharide protection of gram-566
negative bacteria from antimicrobial peptides. J Biol Chem. 280: 10378–10387. 567
52. Rosenfeld Y, Barra D, Simmaco M, Shai Y, Mangoni ML. 2006. A synergism between 568
Temporins toward Gram-negative bacteria overcomes resistance imposed by the 569
Lipopolysaccharide protective layer. J Biol Chem. 281: 28565-28574. 570
53. Mangoni ML, Epand RF, Rosenfeld Y, Peleg A, Barra D, Epand RM, Shai Y. 2008. 571
Lipopolysaccharide, a key molecule involved in the synergism between Temporins in 572
inhibiting bacterial growth and in endotoxin neutralization. J Biol Chem. 283: 22907-22917. 573
54. Mangoni ML, Shai Y. 2009. Temporins and their synergism against Gram-negative bacteria 574
and in lipopolysaccharide detoxification. Biochim Biophys Acta 1788: 1610-1619. 575
55. Mangoni ML. 2006. Temporins, anti infective peptides with expanding properties. Cell Mol 576
Life Sci 63:1060-1069. 577
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
5
56. Bhunia A, Saravanan R, Mohanram H, Mangoni ML, Bhattacharjya S. 2011. NMR 578
structures and interactions of temporin-1Tl and temporin-1Tb with lipopolysaccharide 579
micelles: mechanistic insights into outer membrane permeabilization and synergistic activity. 580
J Biol Chem. 286: 24394-24406 581
57. Rathi Saravanan, Mangesh Joshi, Harini Mohanram, Anirban Bhunia, Maria Luisa 582
Mangoni, Surajit Bhattacharjya. 2013. NMR structure of temporin-1 Ta in 583
lipopolysaccharide micelles: mechanistic insight into inactivation by outer membrane. PLoS 584
ONE 8: e72718 585
58. Bhattacharjya S, Domadia PN, Bhunia A, Malladi S, David SA. 2007. High resolution 586
solution structure of a designed peptide bound to lipopolysaccharide: transferred nuclear 587
Overhauser effects, micelle selectivity and anti-endotoxic activity. Biochemistry 46: 5864-588
5874. 589
59. Bhunia A, Mohanram H, Domadia PN, Torres J, Bhattacharjya S. 2009. Designed beta-590
boomerang antiendotoxic and antimicrobial peptides: structures and activities in 591
lipopolysaccharide. J Biol Chem. 284: 21991-22004. 592
60. Maurizio S. Giuseppina M, Silvia C, Rossella M, Mangoni ML,Donatella B. 1996. 593
Temporins, antimicrobial peptides from the European red frog Rana temporaria. Eur J 594
Biochem. 242:788-792. 595
61. Goldman MJ, Anderson GM, Stolzenberg DE, Kari UP, Zasloff M, Wilson MJ. 1997. 596
Human β-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis. 597
Cell 88: 553–560. 598
62. Lee IH, Cho Y, Lehrer RI. 1997. Effects of pH and salinity on the antimicrobial properties 599
of clavanins. Infect Immun 65: 2898–2903. 600
63. Park IY, Cho JH, Kim KS, Kim YB, Kim MS, Kim SC. 2004. Helix stability confers salt 601
resistance upon helical antimicrobial peptides. J Biol Chem. 279:13896–13901. 602
64. Lindsay GK, Roslansky PF, NovitskyTJ. 1989. Single-step, chromogenic Limulus 603
amebocyte lysate assay for endotoxin. J Clin Microbiol. 27: 947-951. 604
65. Friberger P, Sorskog L, Nilsson K, Knos M. 1987. Detection of bacterial endotoxin with 605
the Limulus amebocyte lysate test. In Watson SW, Levin J, Novitsky TJ: editors. Alan R. Liss, 606
New York. pp. 49–169. 607
66. Jerala R, Porro M. 2004. Endotoxin neutralizing peptides. Curr Top Med Chem. 4: 1173-608
1184. 609
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
6
67. Bhattacharjya S. 2010. De novo designed lipopolysaccharide binding peptides: structure 610
based development of antiendotoxic and antimicrobial drugs. Curr Med Chem. 17: 3080-611
3093. 612
68. Srimal S, Surolia N, Balasubramanian S, Surolia A. 1996. Titration calorimetric studies to 613
elucidate the specificity of the interactions of polymyxin B with lipopolysaccharides and 614
lipidA Biochem J. 315: 679–686. 615
69. Howe J, Andra J, Conde R, Iriarte M, Garidel P, Koch MH, Gutsmann T, Moriyon I, 616
Brandenburg K. 2007. Thermodynamic analysis of the lipopolysaccharide dependent 617
resistance of Gram negative bacteria against polymyxin B Biophys J. 92: 2796–2805. 618
70. Tobias PS, Soldau K, Gegner JA, Mintz D, Ulevitch RJ. 1995. Lipopolysaccharide 619
binding protein mediated complexation of lipopolysaccahride with soluble CD14. J Biol 620
Chem. 270: 10482–10488. 621
71. Rosenfeld Y, Sahl HG, Shai Y. 2008. Parameters involved in antimicrobial and endotoxin 622
detoxification activities of antimicrobial peptides. Biochemistry 47: 6468–6478. 623
624
625
626
627
628
629
630
631
632
633
634
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
7
Table 1: Amino acid sequences of the hybrid peptides LG21, FG21, KG20 and mutated 635
analogs of LG21 with Ala replacement. 636
Peptide Amino acid sequence*
LG21 LLPIVGNLLKSLLGWKRKRFG
FG21 FLPLIGRVLSGILGWKRKRFG
KG20 KLLLKLKLKLLKGWKRKRFG
LG21R19A LLPIVGNLLKSLLGWKRKAFG
637
*The amino acid sequences from parent peptides, TB (in LG21), TA (in FG21) and 638
KL12 (in KG20) are highlighted in blue, whereas β-boomerang motif is in red. 639
Mutations to Ala are italicized in LG21R19A and LG21W15AF20A. All peptides are 640
amidated at their C-termini. 641
642
643
644
645
646
647
648
649
650
651
652
653
654
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
8
655
Table 2: Minimal inhibitory concentration (MIC, in μM) in MH broth and hemolysis of 656
hybrid peptides. MIC values are also determined for the parent peptides TA, TB and 657
KL12 peptides for comparison. 658
Parent Peptides Hybrid peptides and mutants
TB TA KL12 LG21 FG21 KG20 LG21
R19A
LG21
W15AF20A
Gram-negative
bacteria
Escherichia coli (lab
strain) 100 >200 50 2 2 4 100 100
Pseudomonas aeruginosa
(ATCC27853) 100 >200 100 4 2 4 200 100
Klebsiella pneumonia
(ATCC13883) 100 >200 >200 2 4 10 200 200
Salmonella enterica
(ATCC14028) 200 >200 >200 2 5 10 >200 >200
Gram-positive
bacteria
Bacillus subtilis (lab
strain)
25 25 50 2 1 4 50 100
Staphylococcus aureus
(ATCC25923)
25 25 50 2 0.5 4 50 100
Streptococcus pyogenes
(ATCC19615)
25 25 50 2 3 10 100 200
Enterococcus faecalis
(ATCC29212)
50 50 100 5 10 10 100 >200
% of hemolysis at
100 μM
8.5 9.9 2.4 25 60
659
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
9
Table 3: Minimal inhibitory concentration (MIC, in μM) in salt containing LB medium 660
of hybrid peptides. 661
LG21 FG21 KG20 Gram-negative
bacteria
Escherichia coli (lab
strain) 8 10 100
Pseudomonas
aeruginosa
(ATCC27853)
8 8 50
Klebsiella pneumonia
(ATCC13883) 8 8 100
Salmonella enterica
(ATCC14028) 8 10 200
Gram-positive
bacteria
Bacillus subtilis (lab
strain)
10 10 50
Staphylococcus aureus
(ATCC25923)
12.5 10 25
Streptococcus pyogenes
(ATCC19615)
8 12.5 100
Enterococcus faecalis
(ATCC29212)
12.5 10 50
662
663
664
665
666
667
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
10
Table 4: Thermodynamic parameters and binding affinity of the hybrid peptides with 668
LPS as determined from ITC. 669
Peptide-LPS
interaction
Ka
(µM-1)
ΔH
(kcal.mol-1)
TΔS
(kcal.mol-1deg-1)
ΔG
(kcal.mol-1)
Kd
(µM)
LG21 1.64 0.58 8.9 -8.4 0.6
FG21 1.6 4.4 12.8 -8.4 0.6
KG20 3.2 3.4 12.2 -8.8 0.3
LG21R19A 0.099 -1.7 5.1 -6.8 10
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
11
Table 5: Emission maximum wavelength (λmax) and Ksv values of Trp residue of hybrid 686
peptides in buffer (Free) and in LPS micelles. 687
688
Peptide λmax (nm) Ksv
Free LPS Free LPS
LG21 358 338 15.6 4.7
FG21 356 338 25.4 7.7
KG20 354 336 27.9 5.8
LG21R19A 356 354 17.9 10.9
689
690
691
692
693
694
695
696
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
12
Figure 1 697
Figure 1: Proposed model of activation of inactive AMPs aggregated in LPS by β-698
boomerang motif. Cartoon showing aggregation of AMP in LPS (left panel) and disruption 699
of LPS outer membrane by non-aggregated AMP upon conjugation with β-boomerang motif 700
(right panel). 701
702
703
704
705
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
13
Figure 2 706
707
708
Figure 2: Association of hybrid peptides with LPS by rhodamine fluorescence. Plot 709
shows changes in fluorescence intensity, at the emission maxima (ΔF590 nm), of rhodamine 710
labeled LG21 (rhoLG21), FG21 (rhoFG21) and KG20 (rhoKG20) and LG21R19A 711
(rhoLG21R19A) peptides as a function of concentration of LPS. Experiments are done in 10 712
mM sodium phosphate buffer, pH 7.0. 2 μM of rhodamine labeled peptides were individually 713
titrated with increasing concentrations of LPS. Fluorescence was monitored at excitation of 714
485 nm and emission at 550-620 nm. 715
716
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
14
Figure 3 717
718
Figure 3: Cell permeabilization by hybrid peptides. (panel A) Plot shows changes in 719
fluorescence intensity of NPN (1-N-phenylnaphthylamine) at emission 420 nm with 720
increasing concentrations of LG21, FG21, KG20 and LG21R19A peptides in presence of E. 721
coli cells. E. coli cells were incubated with 10 μM of NPN and increasing concentrations of 722
peptides were titrated, in individual experiments, measuring NPN fluorescence. (panel B) 723
Plot shows changes in fluorescence intensity of SYTOX green at emission 520 nm with 724
increasing concentrations of LG21, FG21, KG20 and LG21R19A peptides in presence of E. 725
coli cells. E. coli cells were incubated with 1 μM SYTOX green at 37 oC for 15 min. 726
Fluorescence emission of SYTOX green was recorded in the absence of peptides and with 727
increasing concentrations of peptides. 728
729
730
731
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
15
Figure 4 732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
Figure 4: Cell damage by LG21 in EM studies. Electron micrograph of negatively 751
stained E. coli cells in the absence (panel A) and at 3 µM (panel B), 8 µM (panel C) and 15 752
µM (panel D) concentrations of LG21 peptide and 50 μM (panel E) LG21R19A peptide. 753
754
755
756
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from
16
Figure 5 757
758
759
760
761
762
763
764
765
766
767
768
Figure 5: Disaggregation of LPS by hybrid peptides. (panels A-E) Bar diagrams 769
showing diameter versus intensity of scattered light for LPS alone (panel A) and in presence 770
of LG21 (panel B), FG21 (panel C), KG20 (panel D) and LG21R19A (panel E). (panel F) 771
Changes in fluorescence intensity of FITC-labeled LPS as a function of various 772
concentrations of LG21, FG21, KG20 and LG21R19A peptides. All experiments were carried 773
out in 10 mM sodium phosphate buffer, pH=7.0. 774
775
776
777
on April 12, 2018 by guest
http://aac.asm.org/
Dow
nloaded from