Interactions between Surface-Immobilized ... - ACS Publications

Dec 12, 2017 - present the molecular behavior of surface-immobilized CP1 in the absence ... EG4, Creative PEGWorks, Winston Salem, NC, USA) solution i...
2 downloads 0 Views 945KB Size
Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Article

Interactions between Surface Immobilized Antimicrobial Peptides and Model Bacterial Cell Membranes Xiaofeng Han, Jingguo Zheng, Fengming Lin, Kenichi Kuroda, and Zhan Chen Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b03411 • Publication Date (Web): 12 Dec 2017 Downloaded from http://pubs.acs.org on December 14, 2017

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 free 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 accessible to all readers and 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.

Langmuir 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 29 1 2 3 4 5 6 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

Langmuir

Interactions between Surface Immobilized Antimicrobial Peptides and Model Bacterial Cell Membranes Xiaofeng Han1,*, Jingguo Zheng1, Fengming Lin1, Kenichi Kuroda2, Zhan Chen3,* 1

State Key Laboratory of Bioelectronics, School of Biological Science and Medical

Engineering, National Demonstration Center for Experimental Biomedical Engineering Education , Southeast University, Nanjing, China 210096 2

Department of Biologic and Materials Sciences, School of Dentistry, University of

Michigan, Ann Arbor, MI 48109 3

Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109

Corresponding authors: Xiaofeng Han ([email protected]), Zhan Chen ([email protected])

1 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

Abstract: Sum frequency generation vibrational spectroscopy (SFG) was used to study surface immobilization effects on the interactions between antimicrobial peptide (AMP) cecropin P1 (CP1) and model cell membranes. While free CP1 in solution interacted with a model cell membrane composed of a phosphatidylglycerole (PG) bilayer, electrostatic interaction led to the attachment of the CP1 molecules onto the PG surface, and the hydrophobic domain in the lipid bilayer enabled the peptides to insert into the bilayer and form α-helices from random coil structures. While CP1 molecules immobilized on a self-assembled monolayer (SAM) interacted with PG lipid vesicles, the intensity of SFG peak for the peptide α-helix decreased as the PG vesicle concentration increased. It was believed that when surface immobilized CP1 molecules interacted with lipid vesicles, they lay down on the surface or became random coils. When the immobilized CP1 interacted with a PG lipid monolayer on water, the strong interaction led to the lying down orientation of all the surface immobilized peptides as well. Differently, no significant interactions between surface immobilized CP1 with mammalian cell membrane model POPC bilayer were observed. Our results suggest that, instead of membrane insertion, the electrostatic interactions between the surface cationic charges of CP1 and anionic bacterial membranes may play an important role in the antimicrobial activity of the surface immobilized CP1 peptide. 2 ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29 1 2 3 4 5 6 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

Langmuir

Introduction: The therapeutic potential of antimicrobial peptides (AMPs) has been explored for the last decades owing to their efficacy against drug resistant bacteria.[1-10] In general, AMPs in solution act by disrupting bacterial cell membranes, causing membrane permembilization, breakdown of membrane potential, and ultimately cell death. The cationic charges of AMPs facilitate the binding to anionic bacterial membranes by electrostatic interactions over mammalian cell membranes which are less net negatively charged, imparting the selective activity of AMPs to bacteria. In addition to antibiotic use, AMPs have been immobilized onto solid surfaces as antimicrobial coatings for killing bacteria and preventing the bacterial attachment and biofilm formation.[10-16] However, surface immobilization of AMPs has been reported to directly affect their antimicrobial performance (e.g., efficacy, selectivity, and durability).[14-16] Many studies showed that tethering parameters including peptide surface density, peptides orientation, and surface functionality significantly influence the antimicrobial functions of surface-bound AMPs.[17-24] The molecular strategy of AMP immobilization remains challenging as to retain the antimicrobial activity of AMPs after immobilization. This is likely due to, at least in part, a gap in our knowledge on the molecular mechanism of surface-tethered AMPs for killing bacteria. As compared to those in solution, it has been difficult to characterize molecular conformations and orientations of AMPs at solid/solution interfaces in situ, which are critical parameters to determine ability of AMPs to capture and kill bacteria.[5-7] This is due to the lack of appropriate analytical methods which can provide sufficient 3 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

surface sensitivity and/or in situ capability to probe molecular structures of AMPs at solid/liquid interfaces. It is important to understand the molecular behaviors of immobilized AMPs to capture and kill bacteria in order for design and development of highly effective AMP-tethered antimicrobial coatings toward biomedical applications and beyond. In this study, we investigated the interactions between surface immobilized AMP Cecropin P1 (CP1) and model bacterial cell membranes using sum frequency generation (SFG) vibrational spectroscopy. SFG vibrational spectroscopy is a second order nonlinear optical spectroscopy and has been demonstrated to be a powerful technique to probe and characterize the structures and orientations of peptides/proteins at solid/liquid interfaces in situ.[17-22, 25-44] In our previous work, we have successfully applied this spectroscopic method to deduce interfacial orientations of a variety of peptides and proteins,[17-22, 37-44] providing detailed information on how peptides and proteins molecularly behave on surfaces and at interfaces. We will further extend our approach to study surface-immobilized CP1 peptides while interacting with lipid molecules. Cecropin P1 (CP1) is a cationic 31-amino acid peptide with a hydrophobic C terminus and a charged/amphiphilic N terminus.[45] While disordered in solution, CP1 adopts an α-helical conformation capable of membrane disruption upon binding to bacterial cell membranes.[46] The activity of CP1 is selective to bacteria over mammalian cells. Recently, surface-immobilized CP1 has been extensively studied as pathogen-selective capture molecules for the use in biosensor system and antimicrobial 4 ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29 1 2 3 4 5 6 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

Langmuir

coatings.[47-49] To investigate the molecular action of immobilized CP1, we previously prepared CP1-immobilized surfaces through a cysteine residue added to the C-terminus (CP1c). The thiol group of cysteine was coupled with maleimide terminated self-assembled monolayer (SAM). E. coli cells attached on the surface were found stained by a LIVE/DEAD Viability dye kit, indicating that the immobilized CP1 killed bacteria on contact.[21] Our previous SFG study demonstrated that the CP1c helix had a single orientation tilting towards the surface normal (~30 degree vs. the surface normal) in water.[21] Molecular dynamics (MD) simulations have also supported the experimental results obtained from our SFG study.[23] However, these SFG and computational studies present the molecular behavior of surface immobilized CP1 in the absence of bacteria. It remains unclear how the surface immobilized α-helical CP1 peptides interact with bacteria for their antimicrobial activity. It is well known that antimicrobial peptides in solution kill bacteria by disrupting bacterial cell membranes.[4,5] It has been thought in the research field that immobilized AMPs also act by disrupting bacterial cell membranes in similar to those free in solution, but no direct evidence has been presented in literature. In this study, we examined interactions between CP1-immobilized surface and a model bacterial membrane as well as a model mammalian cell membrane using the SFG technique. Specifically, we wanted to determine (1) whether the surface immobilized CP1 molecules change their conformation and/or orientation upon contacting with a model cell membrane for antimicrobial action, (2) whether such changes in the conformation 5 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

and/or orientation of CP1 are selective to lipid compositions (bacteria-type vs. mammalian cell-type membrane) to exert their selective activity to bacteria over mammalian cells, and (3) whether the surface immobilized AMPs have a similar molecular behavior to free AMPs in solution or not. 2. Experimental 2.1 Preparation of SAM surfaces for CP1 immobilization We have extensively adopted a “near-total reflection experimental geometry” using CaF2 prisms as substrates to build lipid bilayers to collect SFG amide I signals of peptides and proteins associated with the lipid bilayers.[29-34] Here we want to collect amide I signals from surface immobilized CP1 on SAMs. SAMs are usually prepared on gold, SiO2, or Si surfaces. However, these materials block the SFG input IR beam in the amide I frequency range of 1500~1800 cm-1, therefore they cannot be used as substrates for near-total reflection SFG experiments. To solve this issue, CaF2 right angle prisms were chosen as the substrates, on which a thin layer of SiO2 was deposited for the growth of SAMs with silane. Right angle CaF2 prisms were purchased from Altos Photonics (Bozeman, MT, USA). These CaF2 prisms were soaked in toluene for 24 h and then sonicated in 1% ContrexTM AP solution from Decon Labs (King of Prussia, PA, USA) for 10 min. The prisms were thoroughly rinsed with Milli-Q water (18.2 MΩ·cm) and dried under N2. They were treated with oxygen plasma using a benchtop plasma cleaner (PE-25-JW) purchased from Plasma Etch (Carson City, NV, USA) for 4 min immediately before being coated with SiO2. A layer of 100 nm of SiO2 was deposited onto each CaF2 prism 6 ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29 1 2 3 4 5 6 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

Langmuir

by an electron-beam deposition process using a SJ-26 Evaporator system at a pressure below 10-5 Torr. The deposition rate is 5 A/s. The SiO2-coated CaF2 prisms were treated in the O2 plasma cleaner for 4 min to generate hydroxyl groups on SiO2 surfaces, which could enable the reaction with silane molecules. These prisms were placed into freshly made 1.0 mM maleimide-EG4-silane (Mal-EG4, Creative PEGWorks, Winston Salem, NC, USA) solution in anhydrous toluene for 48 h at room temperature. The functionalized prisms were rinsed with copious toluene followed by methanol and were dried with blowing nitrogen. 2.2 Immobilization of CP1 onto the EG4-Maleimide-SAM surface Potassium phosphate (monobasic and dibasic) solution

(1.0 M),

Tris(2-carboxyethyl) phosphine hydrochloride (TCEP) solution (0.5 M), toluene, and dichloromethane were purchased from Aldrich (Milwaukee, WI, USA). The wild-type CP1 and C-terminus modified CP1 (CP1c) were ordered from New England Peptide (Gardner, MA, USA). Ethylenediaminetetraacetic acid (EDTA) was obtained from Fisher Biotech. The thiol moiety in the CP1 cysteine residue has a strong affinity for the maleimide moiety on the SAM prepared on CaF2 with SiO2 coating (prepared as presented above), which promotes the covalent immobilization of the CP1 peptide to the maleimide terminated SAM surface. EDTA was added to the peptide solution in contact with the SAM as a chelating agent to sequester metal ions present in the phosphate buffer (PB) solution.

7 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

Page 8 of 29

For SFG data collection, the EG4-Maleimide SAM surface was in contact with a CP1 solution (with PB as solvent; the solution contains 5.0 mM phosphate, 0.03 mM TCEP, 0.01 mM EDTA, and 1.87 µM of CP1 or CP1c, and the pH is 7.2). SFG signal at 1650 cm-1 (contributed by the α-helical structure) was detected from the SAM/CP1 solution interface to monitor the peptide immobilization process. Once this amide I signal reached equilibrium (which took about 2 h), SFG spectra in the amide I frequency range were collected from the SAM/CP1 solution interface. We then replaced the CP1 solution with a PB solution (5 mM phosphate, 0.03 mM TCEP, and 0.01 mM EDTA, without CP1 in the solution) in a big dish for at least three times to wash the residue CP1 solution and physically loosely adsorbed CP1c on the SAM surface. We believe that the remaining CP1 molecules at the SMA/PB solution interface are the chemically immobilized CP1c on SAM. 2.3 Preparation of planar lipid monolayer, supported lipid bilayer and lipid vesicles 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine

(POPC)

and

1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (sodium salt) (POPG) were purchased from Avanti Polar Lipids. Using a published methodology,[50] a planar lipid monolayer was made at the water/air interface, and solid supported lipid bilayer at the CaF2 prism/water interface. Using a lipid extrusion method, solutions of POPC vesicles and POPG vesicles with a diameter of about 100 nm were prepared following procedures as published before.[51] 2.3 SFG measurements 8 ACS Paragon Plus Environment

Page 9 of 29 1 2 3 4 5 6 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

Langmuir

Details regarding SFG theories and setup have been reported previously.[37-46] In this article SFG ssp (s-polarized SF output, s-polarized visible input, and p-polarized infrared input) and ppp (p-polarized SF output, p-polarized visible input, and p-polarized infrared input) signals were detected from solid/liquid interfaces using the near total internal reflection geometry.[28] All of the SFG experiments were carried out at room temperature (22 ℃). The three experimental geometries to collect SFG spectra from the solid substrate supported lipid bilayer/free peptide buffer solution interface, surface immobilized peptide/lipid vesicle solution interface, and surface immobilized peptide/lipid monolayer on water interface are shown in Figure 1.

Figure 1 SFG experimental geometries for studying substrate supported lipid bilayer/free peptide solution interface (left), surface immobilized peptide/lipid vesicle solution interface (middle), and surface immobilized peptide/lipid monolayer on water interface (right).

3. Results and Discussions 3.1 The interactions between free CP1 and lipid bilayers Prior to studying the molecular behaviors of immobilized CP1, we first investigated the interactions between CP1 free in solution and the CaF2 supported 9 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

POPC and POPG lipid bilayers that were prepared by the published method.[29-34] PG lipids are anionic lipids abundant in bacterial cell membranes, while the major component of mammalian cell membranes is zwitterionic PC lipids. Therefore, the POPG and POPC lipid bilayers present model bacterial and mammalian cell membranes, respectively. These lipid bilayers were placed in contact with CP1 buffer solutions with a peptide concentration of 3.74 µM. SFG spectra were collected from the lipid bilayer/CP1 solution interfaces. A strong amide I signal of CP1 was observed from the POPG lipid bilayer/CP1 solution interface (Figure 2a). This result suggests that CP1 molecules were bound to the bilayer surface with order. This is consistent with our previous results on CP1-DPPG bilayer interactions.[34] Cationic CP1 was likely to bind to anionic POPG lipids by electrostatic interactions. The amide I signal shown in Figure 2a is dominated by a peak centered at ~1650 cm-1, indicating that the interfacial CP1 molecules associated with the POPG bilayer adopted an α-helical conformation. For an α-helix peptide, the χppp/χssp signal strength ratio (taken after fitting the ppp and ssp SFG spectra) can be used to determine its orientation as shown in our previous publications.[27] In Figure 2a, the fitted χppp/χssp ratio was determined to be 1.1, but it is outside of the possible range for a delta or Gaussian orientation distribution of a helical peptide, indicating that the orientation of CP1 α-helices has a multiple orientation distribution.[29] A previous study reported that CP1 α-helices accumulated on bacterial cell membrane surfaces in a “carpet-like” way to increase membrane permeation [2] rather than vertically inserting into the lipid bilayer to form pores. Our result indicated that the CP1 α-helices did not adopt a single orientation. 10 ACS Paragon Plus Environment

Page 10 of 29

Page 11 of 29

On the other hand, the SFG spectrum collected from the POPC bilayer/CP1 solution interface showed no discernable SFG signal in the amide I frequency region (Figure 2b). This result suggests that either CP1 molecules did not bind to the POPC lipid bilayer at all or they bound to the lipid bilayer surface, but did not form any secondary structures or non-lying down ordered orientations to generate SFG amide I signal in the SFG spectrum. The binding of cationic CP1 peptides to the zwitterionic POPC lipid bilayer may not be favorable, resulting in rather weak interactions as compared to binding to the anionic POPG lipid bilayer. Because the POPC lipid bilayer presents a mammalian cell membrane model, the weak interaction of CP1 with the POPC lipid bilayer appears to reflect low hemolytic activity of CP1.[2, 5] 120

ppp ssp

100 Intensity (a.u.)

80

120

ppp ssp

100 Intensity (a.u.)

1 2 3 4 5 6 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

Langmuir

(a)

60 40 20

80 60

(b)

40 20

0

0

1500 1550 1600 1650 1700 1750 1800

1500 1550 1600 1650 1700 1750 1800

.

-1

-1

Wavenumber(cm )

Wavenumber(cm )

Figure 2 SFG amide I spectra collected from (a) the supported POPG bilayer/CP1 solution interface and (b) the POPC bilayer/CP1 solution interface. Dots are experimental data and solid lines are fitted SFG spectra.

3.2 The interaction between the immobilized CP1 and lipid vesicle Next, we investigated the interaction between the surface immobilized CP1 with model cell membranes. For previous SFG studies on interactions between model 11 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

membranes and peptides free in solution, substrate supported lipid bilayers have been frequently used [29-34], and SFG spectra were collected from the lipid bilayer/peptide solution interfaces. Here, we used lipid vesicles in solution which could freely interact with the surface immobilized CP1 to mimic the situation in which the immobilized peptides capture the bacteria from solution and kill them on contact. The lipid extrusion method was utilized to prepare POPC or POPG lipid vesicles in a dimeter of ~100 nm.[51] It should be noted that the size of CP1c peptide is only a few nanometers, which is much smaller than the size of lipid vesicles used in this study. Therefore, the curvature effect of the lipid vesicles on the interaction between CP1c and the lipid vesicles could be neglected (or most of the peptides “saw” flat lipid bilayers).

SFG spectrum in the amide I range of the immobilized CP1c was firstly collected from the SAM (with peptide)/water (without lipid vesicles) interface (Figure 3a). A strong amide I band with peak centering at 1650 cm-1 was detected, indicating that the immobilized CP1c spontaneously adopted an α-helical secondary structure without interacting with lipids, as we demonstrated previously.[34] SFG amide I spectra were then collected from the interfaces between surface immobilized CP1c and POPG vesicle solutions with different vesicle concentrations (Figure 3a). The SFG amide I signals had reduced signal intensities compared to that collected from the CP1c/water interface without vesicles. Also, the signal intensity was decreased more significantly as the vesicle concentration increased (Figure 3a). The SFG signal is still centered at ~1650 cm-1, indicating that the majority of the surface immobilized 12 ACS Paragon Plus Environment

Page 12 of 29

Page 13 of 29

CP1c retained α-helical secondary structure while in contact with POPG vesicle solutions.

15

H2O 0.5 g/L POPG vesicle solution 2.5 g/L POPG vesicle solution

10

ssp

(a) 5

H2O

15

Intensity (a.u.)

Intensity (a.u.)

1 2 3 4 5 6 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

Langmuir

0

0.5 g/L POPC vesicle solution 2.5 g/L POPC vesicle solution

10

5

ssp

(b)

0

1500 1550 1600 1650 1700 1750 1800

1500

1550

1600

1650

1700

1750

1800

_1

-1

Wavenumber (cm )

Wavenumber(cm )

Figure 3 SFG amide I spectra collected from (a) interfaces between surface immobilized CP1c and water as well as POPG vesicle solutions and (b) interfaces between surface immobilized CP1c and water as well as POPC vesicle solutions.

The studies using CP1 free in solution above indicated that there was no apparent interaction between CP1 and POPC lipids or no formation of ordered structures. We therefore should not expect to see any strong interactions between surface immobilized CP1c and POPC vesicles. Indeed, POPC lipid vesicles did not change any SFG spectra collected from the interfaces with immobilized peptides in water (Figure 3b). This result indicates that the α-helical structure of CP1 remained intact, and thus, POPC vesicles did not bind to the peptides on the surface at all. This result clearly shows that the surface immobilized CP1c interacted with POPG bilayer

13 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

(model bacterial cell membrane) and POPC bilayer (model mammalian cell membrane) differently.

We further analyzed the SFG spectra collected from the surface immobilized CP1c/POPG vesicle solution interface. SFG ssp and ppp spectra from the CP1c/POPG vesicle solution interfaces provided the fitted χppp/χssp ratios of 1.54, 1.54 and 1.52 at 0, 0.5 and 2.5 lipid g/L of POPG vesicle solutions, respectively (Figure 4a, 4b, 4c). These results are very reproducible and the error bars for the measured SFG signal strengths and ratios are smaller than 5%. The similar ratios (1.52-1.54, corresponding to

orientation angles of around 35 degree between the α-helix principal axis and the surface normal) indicate that there was no molecular orientation change when the surface immobilized CP1 interacted with POPG lipid vesicles. So, what then caused the SFG intensity reduction when POPG lipid vesicles interacted with the immobilized CP1 peptides? According to the SFG selection rule, a random coil structure of peptide chains generates very weak or no SFG signal. In addition the amide I signal intensity is zero when the α-helix is lying down on the lipid bilayer surface. Because the interaction with the POPG vesicles did not change the molecular orientation of CP1 helices, the intensity reduction was likely due to the mixed binary population of intact CP1 helices standing up (tilting at ~35o vs. the surface normal) on the surface (high intensity) and those in a random coli structure or lying down (no intensity). One explanation to generate such two different peptide states is that the lipid vesicles bound to only a part of the surface. When contacted with the POPG lipid vesicle, the 14 ACS Paragon Plus Environment

Page 14 of 29

Page 15 of 29 1 2 3 4 5 6 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

Langmuir

immobilized CP1 might completely lie down on the surface or unfold to a random coli, which generated no SFG signal and thus reduced the overall signal intensity. On the other hand, the immobilized CP1 molecules not in contact with the lipid vesicles were still tilting on the surface and generated strong SFG signals. As the vesicle concentration increased, the lipid vesicles could cover a larger surface area or more CP1 peptides, leading to lower overall SFG signal intensity. This is distinctively different from the results of CP1 free in solution. For the case of free CP1 in solution, due to the electrostatic attraction, positively charged CP1 molecules could bind to the negatively charged POPG bilayer surface. Since the lipid bilayer has a hydrophobic core composed of alkyl chains, the C-termini of the CP1 peptides could favorably interact with the hydrophobic inner core of the lipid bilayer, enabling the peptides to insert into the lipid bilayer partially or even entirely. Our SFG results above on the free peptide – lipid bilayer interactions indicated that peptides adopted multiple orientations. This is compatible with the fact that peptides may insert into the bilayer with different orientations. However, when the CP1c was immobilized onto the EG4-Mal SAM surface through the C-terminus, it was no longer available for CP1c to insert to the lipid bilayer. In turn, the cationic amphiphilic N-terminus of CP1 was exposed to the solution and interacted with the POPG lipid bilayer. Therefore, the binding of POPG lipid vesicles would not be favorable for peptide insertion, but caused the helices to lie down or possibly denature. Such peptides would not generate SFG signal, therefore the total SFG intensity observed in the presence of lipid vesicles decreased. 15 ACS Paragon Plus Environment

Langmuir

It should be noted that our previous research indicated that CP1c immobilized onto a maleimide SAM covered 34% of the surface.[22] Therefore, there was adequate space for all the peptides to lie down on the surface while interacting with the PG lipid molecules. 30

ppp ssp

15

(a) 10 5

20

10 5 0

1550

1600

1650

1700

1750

-5 1500 1550 1600 1650 1700 1750 1800 -1 Wavenumber(cm ) 30 ppp 25 ssp

1800

_1

Wavenumber (cm ) ppp ssp

20 15

Intensity (a.u.)

Intensity (a.u.)

(d)

15

0 1500

ppp ssp

25 Intensity (a.u.)

Intensity (a.u.)

20

(b) 10 5

20

(e)

15 10 5 0

0 1500

1550

1600

1650

1700

1750

-5 1500

1800

_1

Wavenumber (cm )

1600

1650

1700

1750

(c)

10 5

1800

_1

Wavenumber (cm )

ppp ssp

25

Intensity (a.u.)

15

1550

30

ppp ssp

20

Intensity (a.u.)

1 2 3 4 5 6 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

Page 16 of 29

20

(f)

15 10 5 0

0 1500

1550

1600

1650

1700

1750

-5 1500

1800

1550

1600

1650

1700

1750

1800

_1

_1

Wavenumber (cm )

Wavenumber (cm )

Figure 4 SFG amide I spectra collected from the interfaces between surface immobilized CP1c and POPG vesicle solutions at different concentrations of (a) 0 g/L, (b) 0.5 g/L, (c) 2.5 g/L as well as the interfaces between surface immobilized CP1c and POPC vesicle solutions at different concentrations of

16 ACS Paragon Plus Environment

Page 17 of 29 1 2 3 4 5 6 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

Langmuir

(d) 0 g/L,

(e) 0.5 g/L, (f) 2.5 g/L. Dots are experimental data and solid lines are the fitted SFG

spectra.

Figures 4d, 4e, and 4f show the fitting results of the SFG ssp and ppp spectra collected from the surface immobilized CP1c/water and immobilized CP1c/POPC vesicle solutions (with different concentrations) interfaces. All the spectra are the same, indicating there was no interaction between surface immobilized CP1c and POPC. 3.3 The interaction between immobilized CP1c with a lipid monolayer on water In the lipid vesicle experiment described above, the vesicles freely interacted with the immobilized CP1 peptides. Therefore, the electrostatic interactions between CP1 (cationic) and POPG lipids (anionic) played an important role in peptide binding. We discussed that the immobilized CP1c might lie down or be in a random coil upon interaction with POPG lipid vesicles due to the unfavorable interaction between the cationic N-terminus and the hydrophobic core of lipid bilayer. To examine the role of hydrophobic interactions played in the peptide insertion, we designed an experiment in which the surface-immobilized CP1c molecules were brought into contact with a POPG lipid monolayer formed at the air/water interface to force surface immobilized CP1c molecules to interact with the hydrophobic tails of POPG lipid monolayer (Figure 1). Specifically, the CaF2 prism modified with surface immobilized CP1 was removed from water and then placed from the air side on top of a monolayer of POPG with a surface pressure of 34 mN/m. In the lipid monolayer, the hydrophilic lipid head 17 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

groups were located on water and the hydrophobic tails were sticking out into air. The CaF2 prism approached from the air side, and therefore, the immobilized CP1 peptides interacted with the hydrophobic tails of lipids. When the CP1 was in contact with the POPG lipid monolayer, the SFG signal in the amide I frequency range disappeared completely, showing that immobilized CP1c molecules lost ordering at the interface (Figure 5b). This may be caused by the adoption of a random orientation distribution of CP1 helices or a lying down orientation. Alternatively, the CP1 peptides may be in a random coil. The result was also different from the partial reduction in the SFG signal intensity of the peptides when interacted with lipid vesicles in solution discussed above. This could be explained by the difference in two experiments. Here the surface immobilized with peptides was forced to contact with the lipid monolayer, all the peptides on the surface could interact with the lipids, while the vesicles in solution might not cover all the surface as discussed above. On the other hand, no change was found in the spectrum when the POPC monolayer on water was in contact with the immobilized CP1 (Figure 5c). Although the peptides were forced to contact with the POPC lipids, the interaction did not alter the peptide conformation and orientation, indicating no interaction. Even though here the CP1 peptides interacted with the hydrophobic acyl chains of lipids first, but the response of immobilized CP1 was selective to the head groups of the lipids. These results suggest that the charge of lipid head groups is a dominant factor to determine the molecular conformation and orientation of immobilized CP1 during their interactions. Although we expect the interaction between the charged CP1c N-terminus and the hydrophobic tails of POPG 18 ACS Paragon Plus Environment

Page 18 of 29

Page 19 of 29

would be unfavorable, the electrostatic interactions between cationic CP1 molecules and anionic POPG lipids finally decide the molecular behaviors of immobilized CP1.

30

ppp ssp

30

Intensity (a.u.)

25

ppp ssp

25 Intensity (a.u.)

20 15

20 15

(b)

10 5 0

(a)

1500 1550 1600 1650 1700 1750 1800 -1

10

Wavenumber(cm )

5

30

0

25

1500 1550 1600 1650 1700 1750 1800

Intensity (a.u.)

1 2 3 4 5 6 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

Langmuir

-1

Wavenumber(cm )

ppp ssp

20 15

(c)

10 5 0 1500 1550 1600 1650 1700 1750 1800 -1

Wavenumber(cm )

Figure 5 SFG amide I spectra collected from (a) the surface immobilized CP1c/water interface, (b) the interface between the surface immobilized CP1c and a monolayer POPG on water, and (c) the interface between the surface immobilized CP1c and a monolayer POPC on water.

19 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

The conclusions obtained above from the SFG amide I signals of the immobilized peptide/monolayer lipid on water interface can be further validated by our SFG spectra collected in the high frequency range. The SFG spectra of the immobilized CP1c in contact with different environments including pure water, a POPG monolayer on water, and a POPC monolayer on water were collected (Figure 6a) in the range of 2600~3700 cm-1. Such signals contained contributions from interfacial water O-H stretches (centered at ~3200 and 3400 cm-1), the amide A group of CP1c (~3300 cm-1), and C-H stretching modes (2800-3000 cm-1). The broad band of the immobilized CP1c contacting with pure water was the same as that of the peptides contacting with POPC monolayer on water, centered at ~3300 cm-1. We believe that for both cases surface immobilized CP1c molecules were contacting with water, because if POPC monolayer was still present on water, the water signal should be detected. POPC left the surface because they dissolved into water. The 3300 cm-1 peak was mainly contributed by the CP1c amide A group, showing that for both cases CP1 molecules are very ordered at the interface. Furthermore, the spectral features in the C-H stretching frequency region (2800 ~ 3000 cm-1) from the CP1c/water and CP1c/POPC monolayer interfaces are similar. Such C-H signals could be contributed by peptides, SAM, and POPC lipids. Since the C-H spectra were similar, such C-H signals detected from the two interfaces here should be dominated by the contributions from peptides and SAM, not the POPC monolayer on water. Therefore the removal of the POPC monolayer from the surface was proved by both the O-H/N-H stretching and C-H stretching signals. 20 ACS Paragon Plus Environment

Page 20 of 29

Page 21 of 29 1 2 3 4 5 6 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

Langmuir

Differently, the broad band generated from immobilized CP1c in contact with POPG monolayer was significantly different from that of immobilized CP1c contacting pure water or POPC monolayer. The SFG spectrum from the CP1c/POPG monolayer interface was dominated by two water O-H peaks at 3200 and 3400 cm-1, showing that water molecules were ordered at the interface. However, the 3300 cm-1 signal was absent, indicating that CP1c molecules lost ordering at this interface due to strong peptide-lipid interactions, agreeing well with the results obtained from the SFG amide I studies reported above regarding the peptide/POPG monolayer studies. For further validating our above discussions, we also collected SFG spectra in the high frequency range between SAM surface (without CP1c immobilized) and different media, including water, POPC monolayer on water, and POPG monolayer on water (Figure 6b). The three SFG spectra were almost the same. We believe that for both the POPC and POPG monolayer cases, the interaction between the SAM surface and lipid monolayer surface destabilized the lipid monolayers and either dissolved both the POPC and POPG monolayers or made them random, otherwise the lipid C-H signal should be very strong. The same interfacial water signals for three cases indicated that more likely the removal of the lipid monolayer from the surface occurred. Therefore for all the three cases the SFG spectra were contributed from the same SAM/water interface. This again shows that the strong favorable interaction presented above between positively charged CP1c and negatively charged POPG did not remove the POPG monolayer from the surface.

21 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

Figure 6

High frequency range SFG spectra collected from (a) interfaces between surface

immobilized CP1c and water, a POPC monolayer on water, and a POPG monolayer on water and (b) interfaces between SAM (without CP1c immobilization) and water, a POPC monolayer on water, and a POPG monolayer on water.

22 ACS Paragon Plus Environment

Page 22 of 29

Page 23 of 29 1 2 3 4 5 6 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

Langmuir

3.4.

Relationship

between

the

peptide

conformation/orientation

and

antimicrobial activity.

Figure 7 Schematics showing interactions between peptides and lipids at (top left) substrate supported lipid bilayer/free CP1 solution interface, (bottom left) surface immobilized CP1/lipid vesicle solution interface, and (right) surface immobilized CP1/lipid monolayer on water interface

Figure 7 displays schematics of the results at the substrate supported lipid bilayer/free CP1 solution interface, surface immobilized CP1/lipid vesicle solution interface, and surface immobilized CP1/lipid monolayer on water interface obtained in this research. These results suggest that the immobilized CP1 helices lay down (or became random coils) upon binding to the anionic POPG lipid bilayer which is a bacterial cell membrane model. The interactions between CP1c peptides and model bacterial cell membranes are different for the free peptides in solution (insertion and tilting into the membrane) and surface immobilized peptides (lying down or random coil). It has been reported that surfaces coated with cationic synthetic polymers showed on-contact killing of bacteria.[52,53] A study proposed that the surface 23 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

density of positive charges determined the bacterial killing efficacy, rather than the ability of polymers to insert into and disrupt bacterial membranes. Therefore, the electrostatic interactions between the surface cationic charges of CP1 and anionic bacterial membranes (instead of membrane insertion) may play an important role in the antimicrobial activity of peptides, which is in good agreement with our data presented here 4. Conclusion In this work, the molecular interactions between surface immobilized CP1 and model cell membranes were investigated using SFG. For comparison purpose, the interactions between free CP1 molecules in solution and model cell membranes were also studied. Regarding the three questions raised above in the introduction section, the following conclusions can be drawn: (1) Surface immobilized CP1 molecules did change their orientation (to lie down) or conformation (to denature) upon contacting with a model bacterial cell membrane which may be related to antimicrobial action. (2) It was found that the surface immobilized CP1 actively interacted with the model bacteria membrane (POPG monolayer and vesicle) but not the model mammalian cell membrane (POPC monolayer and vesicle), which shows the selectivity of the surface immobilized AMPs. (3) The surface immobilized AMPs (lying down or denaturing) have different molecular behaviors from free AMPs in solution (membrane insertion). We believe that the bacterial inhibition effect of the surface immobilized peptides comes from the strong electrostatic interaction between the cationic CP1 peptide chains and anionic bacteria cell membranes. To generalize this conclusion, it is 24 ACS Paragon Plus Environment

Page 24 of 29

Page 25 of 29 1 2 3 4 5 6 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

Langmuir

necessary to perform systematic investigation on interactions including kinetics between model cell membrane and other antimicrobial peptides or mutated cecropin P1 with different charges, which will be performed in the future. A lipid vesicle may be too simple to fully represent a cell to understand the antimicrobial peptide – cell membrane interactions. More

detailed correlations between the antimicrobial activity of AMPs and surface immobilized AMP orientation/conformation is under the current investigation on surface immobilized peptide – cell membrane interactions in situ using live bacteria. Acknowledgment This work is supported by the National Natural Science Foundation of China (Grant: 21773028), the Priority Academic Program Development of Jiangsu Higher Education Institutions (1107037001). It is also supported by US Army Research Office (W911NF-11-1-0251) and US National Science Foundation (CHE-1505385).

Reference: 1.

Boman, H.G., B. Agerberth, and A. Boman, Mechanisms of Action on Escherichia coli of Cecropin P1 and PR-39, Two Antibacterial Peptides from Pig Intestine. Infect Immun, 1993. 61(7): p. 2978-2984.

2.

Gazit, E., et al., Structure and Orientation of the Mammalian Antibacterial Peptide Cecropin P1 within Phospholipid Membranes. Journal of Molecular membranes, 1996. 258: p. 860-870.

3.

Yeaman, M.R. and N.Y. Yount, Mechanisms of antimicrobial peptide action and resistance. Pharmacological Reviews, 2003. 55(1): p. 27-55.

4.

Zasloff, M., Antimicrobial peptides of multicellular organisms. Nature, 2002. 415(6870): p.

5.

Shai, Y., Mechanism of the binding, insertion and destabilization of phospholipid bilayer

389-395. membranes by α-helical antimicrobial and cell non-selective membrane-lytic peptides. Biochimica et Biophysica Acta, 1999. 1462: p. 55-70. 6.

Patricia Lesmes, L., et al., A C-terminal cationic fragment derived from an arginine-rich peptide exhibits in vitro antibacterial and anti-plasmodial activities governed by its secondary structure properties. Peptides, 2009. 30(12): p. 2150-2160.

25 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

Page 26 of 29

7.

Paulsen, V.S., et al., Structure-Activity Relationships of the Antimicrobial Peptide Arasin

8.

Kawai, T. and S. Akira, The role of pattern-recognition receptors in innate immunity: update

1-And Mode of Action Studies of the N-Terminal, Proline-Rich Region. Plos One, 2013. 8(1). on Toll-like receptors. Nature Immunology, 2010. 11(5): p. 373-384. 9.

Hetru, C., et al., Androctonin, a hydrophilic disulphide-bridged non-haemolytic anti-microbial peptide: a plausible mode of action. Biochemical Journal, 2000. 345: p. 653-664.

10.

Hancock, R.E.W. and H.-G. Sahl, Antimicrobial and host-defense peptides as new

11.

Fulmer, P.A., J.G. Lundin, and J.H. Wynne, Development of antimicrobial peptides (AMPs)

anti-infective therapeutic strategies. Nature Biotechnology, 2006. 24(12): p. 1551-1557. for use in self-decontaminating coatings. ACS Appl Mater Interfaces, 2010. 2(4): p. 1266-70. 12.

Gregory, K. and C.M. Mello, Immobilization of Escherichia coli cells by use of the antimicrobial peptide cecropin P1. Appl Environ Microbiol, 2005. 71(3): p. 1130-4.

13.

Traba, C. and J.F. Liang, Bacteria responsive antibacterial surfaces for indwelling device

14.

Wu, X., et al., Effect of immobilization on the antimicrobial activity of a cysteine-terminated

infections. Journal of Controlled Release, 2015. 198: p. 18-25. antimicrobial Peptide Cecropin P1 tethered to silica nanoparticle against E. coli O157:H7 EDL933. Colloids and Surfaces B-Biointerfaces, 2017. 156: p. 305-312. 15.

Chen, R., et al., Characterization of chemoselective surface attachment of the cationic peptide melimine and its effects on antimicrobial activity. Acta Biomater, 2012. 8(12): p. 4371-9.

16.

Hilpert, K., et al., Screening and characterization of surface-tethered cationic peptides for antimicrobial activity. Chem Biol, 2009. 16(1): p. 58-69.

17.

Ye, S., et al., Orientation difference of chemically immobilized and physically adsorbed biological molecules on polymers detected at the solid/liquid interfaces in situ. Langmuir, 2009. 26(9): p. 6471-6477.

18.

Han,

X.,

et

al.,

Solvent

Effect

and

Time-Dependent

Behavior

of

C-Terminus-Cysteine-Modified Cecropin P1 Chemically Immobilized on a Polymer Surface. Langmuir, 2011. 27(11): p. 7042-7051. 19.

Li, Y., et al., Room temperature freezing and orientational control of surface-immobilized peptides in air. Chemical Communications, 2015. 51(55): p. 11015-11018.

20.

Li, Y., et al., Effects of Peptide Immobilization Sites on the Structure and Activity of Surface-Tethered Antimicrobial Peptides. Journal of Physical Chemistry C, 2015. 119(13): p. 7146-7155.

21.

Han, X., et al., Different Interfacial Behaviors of Peptides Chemically Immobilized on Surfaces with Different Linker Lengths and via Different Termini. Journal of Physical Chemistry B, 2014. 118(11): p. 2904-2912.

22.

Wang, Q., et al., Interfacial Behaviors of Antimicrobial Peptide Cecropin P1 Immobilized on Different Self-Assembled Monolayers. Journal of Physical Chemistry C, 2015. 119(39): P. 22542-22551.

23.

Wang, Z., et al., Molecular Structures of C- and N-Terminus Cysteine Modified Cecropin P1 Chemically

Immobilized

onto

Maleimide-Terminated

Self-Assembled

Mono

layers

Investigated by Molecular Dynamics Simulation. Journal of Physical Chemistry B, 2014. 118(21): p. 5670-5680. 24.

Wang, Z., et al., Environmental Effect on Surface Immobilized Biological Molecules. Journal of Physical Chemistry B, 2014. 118(42): p. 12176-12185.

26 ACS Paragon Plus Environment

Page 27 of 29 1 2 3 4 5 6 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

Langmuir

25.

Wang, J., et al., Detection of amide I signals of interfacial proteins in situ using SFG. Journal

26.

Nguyen, K.T., J.T. King, and Z. Chen, Orientation determination of interfacial β-sheet

of the American Chemical Society, 2003. 125(33): p. 9914-9915. structures in situ. The Journal of Physical Chemistry B, 2010. 114(25): p. 8291-8300. 27.

Nguyen, K.T., et al., Orientation Determination of Protein Helical Secondary Structures Using Linear and Nonlinear Vibrational Spectroscopy. The Journal of Physical Chemistry B, 2009. 113(36): p. 12169-12180.

28.

Wang, J.; Mark, A.; Chen, X.; Schmaier, A. H.; Waite, J. H.; Chen, Z., Detection of amide I signals of interfacial proteins in situ using SFG. Journal of the American Chemical Society, 2003. 125(33): P. 9914-9915.

29.

Chen, X., et al., Multiple orientation of melittin inside a single lipid bilayer determined by combined vibrational spectroscopic studies. Journal of the American Chemical Society, 2007. 129(5): p. 1420-1427.

30.

Ye, S., K.T. Nguyen, and Z. Chen, Interactions of alamethicin with model cell membranes investigated using sum frequency generation vibrational spectroscopy in real time in situ. The Journal of Physical Chemistry B, 2010. 114(9): p. 3334-3340.

31.

Chen, X., et al., Real-time structural investigation of a lipid bilayer during its interaction with melittin using sum frequency generation vibrational spectroscopy. Biophysical journal, 2007. 93(3): p. 866-875.

32.

Chen, X. and Z. Chen, SFG studies on interactions between antimicrobial peptides and supported lipid bilayers. Biochimica et Biophysica Acta (BBA)-Biomembranes, 2006. 1758(9): p.

33.

1257-1273.

Nguyen, K.T., et al., Molecular interactions between magainin 2 and model membranes in situ. The Journal of Physical Chemistry B, 2009. 113(36): p. 12358-12363.

34.

Wang, T., et al., Single Lipid Bilayers Constructed on Polymer Cushion Studied by Sum Frequency Generation Vibrational Spectroscopy. Journal of Physical Chemistry C Nanomater Interfaces, 2011. 115(15): p. 7613-7620.

35.

Fu, L.; Liu, J.; Yan, E. C. Y., Chiral sum frequency generation spectroscopy for characterizing protein secondary structures at interfaces. Journal of the American Chemical Society 2011, 133(21): P. 8094-8097.

36.

Fu, L.; Xiao, D.; Wang, Z.; Batista, V. S.; Yan, E. C. Y., Chiral sum frequency generation for in situ probing proton exchange in antiparallel β-sheets at interfaces. Journal of the American Chemical Society, 2013. 135(9): P. 3592-3598.

37.

Weidner, T.; Breen, N. F.; Li, K.; Drobny, G. P.; Castner, D. G., Sum frequency generation and solid-state NMR study of the structure, orientation, and dynamics of polystyrene-adsorbed peptides Proceedings of the National Academy of Sciences USA, 2010. 107(30): P. 13288-13293.

38.

Chen, X., et al., In situ investigation of heterotrimeric G protein βγ subunit binding and orientation on membrane bilayers. Journal of the American Chemical Society, 2007. 129(42): p. 12658-12659.

39.

Engel, M. F. M.; vandenAkker, C. C.; Schleeger, M.; Velikov, K. P.; Koenderink, G. H.; Bonn, M., The polyphenol EGCG inhibits amyloid formation less efficiently at phospholipid

27 ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 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

interfaces than in bulk solution. Journal of the American Chemical Society, 2012. 134(36): P. 14781-14788. 40.

Weidner, T.; Apte, J. S.; Gamble, L. J.; Castner, D. G., Probing the orientation and conformation of α-helix and β-strand model peptides on self-assembled monolayers using sum frequency generation and NEXAFS spectroscopy. Langmuir, 2010. 26: P. 3433-3440.

41.

York, R. L.; Browne, W. K.; Geissler, P. L.; Somorjai, G. A., Peptides adsorbed on hydrophobic surfaces-a sum frequency generation vibrational spectroscopy and modeling study. Israel Journal of Chemistry, 2007. 47(1): P. 51-58.

42.

Baio, J. E., et al., eversible activation of pH-sensitive cell penetrating peptides attached to gold surfaces, Chemical Communications, 2015. 51: P. 273-275.

43.

Boughton, A.P., et al., Heterotrimeric G protein β1γ2 subunits change orientation upon complex formation with G protein-coupled receptor kinase 2 (GRK2) on a model membrane. Proceedings of the National Academy of Sciences, 2012. 108(37): p. E667-E673.

44.

Ye, S., et al., Observing a Model Ion Channel Gating Action in Model Cell Membranes in Real Time in Situ: Membrane Potential Change Induced Alamethicin Orientation Change. Journal of the American Chemical Society, 2012. 134(14): p. 6237-6243.

45.

Lee, J., et al., Antibacterial peptides from pig intestine: Isolation of a mammalian cecropin. Proceedings of the National Academy of Sciences USA, 1989. 86: p. 9159-9162.

46.

Gazit, E., et al., Interaction of the Mammalian Antibacterial Peptide Cecropin P1 with Phospholipid Vesicles. Biochem., 1995. 34: p. 11479-11488.

47.

Uzarski, J.R. and C.M. Mello, Detection and Classification of Related Lipopolysaccharides via a Small Array of Immobilized Antimicrobial Peptides. Analytical Chemistry, 2012. 84(17): p. 7359-7366.

48.

Gregory, K. and C.M. Mello, Immobilization of Escherichia coli cells by use of the antimicrobial peptide cecropin P1. Appl Environ Microbiol, 2005. 71(3): p. 1130-4.

49.

Strauss, J., et al., Binding, inactivation, and adhesion forces between antimicrobial peptide cecropin P1 and pathogenic E. coli. Colloids Surf B Biointerfaces, 2010. 75(1): p. 156-64.

50.

Li, B., et al., Sum Frequency Generation of Interfacial Lipid Monolayers Shows Polarization

51.

Olson, F., et al., Preparation of liposomes of defined size distribution by extrusion through

Dependence on Experimental Geometries, Langmuir, 2016. 32 (28), pp 7086–7095. polycarbonate membranes. Biochimica Et Biophysica Acta, 1979. 557(1): p. 9-23. 52.

Kügler, R., et al., Evidence of a charge-density threshold for optimum efficiency of biocidal cationic surfaces, Microbiology, 2005. 151: P. 1341–1348.

53.

Murata, H., et al., Permanent, non-leaching antibacterial surfaces - 2: How high density cationic surfaces kill bacterial cells. Biomaterials, 2007. 28(32): p. 4870-4879.

28 ACS Paragon Plus Environment

Page 28 of 29

Page 29 of 29 1 2 3 4 5 6 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

Langmuir

TOC graphic

ACS Paragon Plus Environment