Binding of Short Cationic Peptides (KX)4K to Negatively Charged

Subscriber access provided by NEW YORK MED COLL

Article

Binding of short cationic peptides (KX)4K to negatively charged DPPG monolayers: Competition between electrostatic and hydrophobic interactions. André Hädicke, and Alfred Blume Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.5b02882 • Publication Date (Web): 19 Oct 2015 Downloaded from http://pubs.acs.org on October 19, 2015

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 38

Langmuir

1 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

Binding of short cationic peptides (KX)4K to negatively charged DPPG monolayers: Competition between electrostatic and hydrophobic interactions.

André Hädicke and Alfred Blume* Institute of Chemistry, MLU Halle-Wittenberg, von-Danckelmann-Platz 4, 06120 Halle/Saale, Germany *Corresponding author. Address: Institute of Chemistry–Physical Chemistry, Martin-LutherUniversity Halle-Wittenberg, von-Danckelmann-Platz 4, D-06120 Halle/Saale, Germany. Tel: +49 345 55 25850 Fax: +49 345 55 27157 E-mail address: [email protected] Keywords: cationic peptide, peptide binding, lipid monolayer, hydrophobic interaction, electrostatic interaction, surface pressure-area isotherm, fluorescence microscopy, adsorption kinetics

ACS Paragon Plus Environment

Langmuir

Page 2 of 38

2 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 The influence of the peptide sequence on the binding of short cationic peptides composed of five lysines alternating with uncharged amino acids within the series (KX)4K to negatively charged

monolayers

of

1,2-dipalmitoyl-sn-glycero-3-phosphoglycerol

(DPPG)

was

investigated by adsorption experiments in combination with epifluorescence microscopy. To evaluate the impact of electrostatic and hydrophobic contributions, different uncharged amino acids X with increasing hydrophobicity, where X = G (glycine), A (alanine), Abu (αaminobutyric acid), V (valine), or L (leucine), were introduced into the peptide sequence to tune the peptide hydrophobicity. The adsorption kinetics of these peptides to a DPPG monolayer showed always two superimposed processes, one leading to an increase and another one to a decrease of the surface pressure Π. Thus, the plots of the change in Π after peptide binding vs. initial surface pressure of the monolayer showed an unusual behaviour with maxima and negative changes in Π at high initial Π values. Epifluorescence microscopy confirmed that electrostatic binding of the peptides with a concomitant decrease in Π leads to a condensation of the lipid monolayer and the formation of liquid-condensed (LC) domains even at Π values where the monolayer is supposedly in the liquid-expanded (LE) state. An increase in hydrophobicity of the amino acid X was found to counteract the condensation and an increase in Π upon peptide binding is observed at low Π values, also concomitant with the formation of LC-domains. Compression of monolayers after peptide adsorption at low surface pressure for four hours leads to a change of the isotherms compared to pure DPPG isotherms. The phase transition of DPPG from LE to LC state is smeared out or is shifted to higher surface pressure. Considerable changes in the shapes of LC-domains were observed after peptide binding. Growth of the LC-domains was hindered in most cases and regular domain patterns were formed. Binding of (KL)4K lead to a decrease in line tension and the formation of extended filaments protruding from initially circular domains.


Page 3 of 38

Langmuir

3 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

Introduction Antimicrobial peptides show a high abundance (about 10% each) of the amino acids lysine, glycine, alanine and leucine1 leading to a cationic and amphipathic structure of these molecules.2-3 The preferential interaction of these peptides with bacterial membranes4 is a complex interplay of charge and hydrophobicity. Due to their positive charge they show preferred interactions with negatively charged lipid headgroups, but at the same time the hydrophobic residues interact with the lipid acyl chain region. To get further insights into the mode of interaction of antimicrobial peptides with membranes it is necessary to understand the very first binding step of these peptides and the resulting effects on lipid membrane properties. Upon binding, several processes may occur like changes of the secondary structure of the peptide, clustering of charged lipids, membrane thinning/thickening, membrane depolarization or even pore formation.5-6 To understand this complex behaviour, simplified models consisting of short cationic peptides and bilayer membranes with distinct lipid composition are preferentially studied first. The understanding of the relationship between function and structure might allow the design and production of cheaper and more efficient antimicrobial peptides.7-8 A lipid monolayer represents half of a bilayer9-11 and is particularly suitable to study the interaction or binding of molecules, such as peptides and proteins,12,13 polymers14 or amphiphilic substances15 to the lipid headgroup region and their penetration into the lipid acyl chain region. In monolayer experiments, the change in area or surface pressure can easily be measured and controlled. The combination of the monolayer experiments with epifluorescence microscopy,16-18 FT-IRRA spectroscopy,19 or x-ray techniques20 provides additional information about the lipid and peptide arrangement. The effect of positively charged amino acids on the electrostatically driven binding of cationic model peptides to lipid bilayers and monolayers was extensively tested before with ACS Paragon Plus Environment

Langmuir

4 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

homopeptides of lysine or arginine with different degrees of polymerisation.21-23 The contribution of the amino acid side chains to hydrophobic interactions with lipid acyl chains was investigated with a set of tailor-made pentapeptides having shorter side chains than lysine.24 In this previous study it was found that with decreasing side chain length the hydrophobic contribution to the binding affinity decreases, i.e. the binding affinity of Lys5 is not only due to the electrostatic attraction but has also a contribution from the hydrophobic CH2 groups in the side chain. In the present study we investigated the binding of linear nonapeptides containing five positively charged lysines and four uncharged amino acid X with the sequence (KX)4K, as shown in Figure 1, to negatively charged lipid monolayers. The hydrophobicity of the peptides was tuned by increasing the hydrophobicity of the side chains of the amino acid X using glycine (G), alanine (A), α-aminobutyric acid (Abu), valine (V), or leucine (L), respectively. Our aim was to study the contribution of the spacer amino acid X on the binding to negatively charged dipalmitoyl-phosphatidylglycerol (DPPG) monolayers at the air-water interface. Increasing the hydrophobicity of the amino acid X changed the balance between electrostatic and hydrophobic effects as evidenced from changes in surface pressure of the lipid monolayer after binding, thus indicating different depths of location of these cationic peptides in the monolayer.


Page 4 of 38

Page 5 of 38

Langmuir

5 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 1: CPK models of the five linear peptides (KX)4K with X = G, A, Abu, V, L in an extended conformation.

Experimental Materials 1,2-Dipalmitoyl-sn-glycero-3-phosphoglycerol (DPPG) was purchased from Lipoid GmbH (Ludwigshafen, Germany) and 1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine-N(lissamine rhodamine B sulfonyl) (Rh–DHPE) from Invitrogen (Karlsruhe, Germany). The peptides (KX)4K with X = G, A, Abu, V, L were custom-made via Fmoc solid phase chemistry with a purity above 98 % by GeneCust Europe (Dudelange, Luxemburg). Chloroform, Methanol (HPLC grade) and NaCl were purchased from Carl Roth GmbH&CO KG (Karlsruhe, Germany). All chemicals were used as received without further purification. Aqueous solutions were prepared with ultrapure water from a Milli-Q Advantage A10 system (Millipore S.A.S., Molsheim Cédex, France). Conductivity was lower than 0.055 µS/cm (25 °C) and TOC below 5 ppb.


Langmuir

6 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

Methods Monolayer adsorption measurements Adsorption experiments were performed with a home-built circular trough with a diameter of 6 cm and a depth of 0.3 cm covered by a plastic hood to keep the temperature and humidity constant. The surface pressure was recorded using a microbalance (Riegler and Kirstein GmbH, Berlin, Germany) equipped with a Wilhelmy plate. Prior to each experiment the trough was thoroughly rinsed and filled with 100 mM NaCl solution. The pressure sensor was calibrated using the surface pressure of ultrapure water (72 mN m-1) and that of air (0 mN m-1) as reference points. The subphase temperature was maintained at 20.0 ± 0.1 °C by a circulating water bath (Thermostat F3, Haake, Karlsruhe, Germany). The subphase was stirred during the experiment using a small stirring magnet to accelerate diffusion of the added solutes. First a freshly prepared defined volume of a lipid solution in chloroform (1 mM) was spread onto the subphase forming a phospholipid monolayer to obtain an initial surface pressure Πini. After waiting for 15 min for complete solvent evaporation the measurement was started and the surface pressure of the pure lipid monolayer was recorded for at least 10 min. Then an aqueous peptide solution was injected below the phospholipid monolayer through a channel above the bottom of the trough. The injection volume was adjusted to give a peptide concentration of 3 µM in the subphase. Analysis and fitting of the kinetic adsorption curves were performed with routines of the software Origin 8.0 (Origin Lab Corp.). Epifluorescence microscopy of lipid monolayers Fluorescence microscopy imaging of monolayers at the air-water interface was performed using an Axio Scope A1 Vario epifluorescence microscope (Carl Zeiss MicroImaging, Jena, Germany). Underneath the microscope a home-build Langmuir Teflon trough with a maximal area of 20.8 cm2 and a moveable computer-controlled Teflon barrier (Riegler & Kirstein,


Page 6 of 38

Page 7 of 38

Langmuir

7 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

Berlin, Germany) was positioned on an x–y stage (Märzhäuser, Wetzlar, Germany) to be able to move the film surface with respect to the objective lens to any desired position. The x–y–z motion control was managed by a MAC5000 system (Ludl Electronic Products, Hawthorne, NY, USA). The trough was enclosed by a homebuilt Plexiglas hood to ensure a dust-free environment and to minimize evaporation of water. The temperature of 20.0 ± 0.1 °C was maintained with a circulating water bath and the whole setup was placed on a vibrationdamped optical table (Newport, Darmstadt, Germany). The air-water surface was illuminated using the following set up from Carl Zeiss MicroImaging (Jena, Germany): a 100 W mercury arc lamp (HXP 120 C), a long working distance objective (LD EC Epiplan-NEOFLUAR 50x) and a filter/beam splitter combination (Filter Set 81HE), to select appropriate wavelengths for the excitation and detection of Rh–DHPE. Images were recorded using an EMCCD camera (ImageEM C9100-13, Hamamatsu, Herrsching, Germany). Image analysis and data acquisition were done using the AxioVision software (Carl Zeiss MicroImaging, Jena, Germany). All presented images show areas of individually contrast-adjusted raw data. For combined monolayer adsorption measurements with subsequent compression, monolayer films of DPPG were prepared by spreading from a stock solution with a concentration of 1 mM in chloroform containing only 0.01 mol% fluorescently labelled Rh–DHPE. This low concentration was possible due to the high quantum yield of Rh–DHPE and the high sensitivity of our CCD camera. Rh–DHPE prefers the liquid-expanded state of a lipid monolayer so that liquid-condensed domains appear dark.25 Monolayer adsorption measurements were conducted as described above. After reaching the adsorption equilibrium the monolayer was compressed using a compression speed of 2 Å2·molecule-1·min-1. Microscopy images were taken during adsorption of the peptide and compression of the monolayer.


Langmuir

8 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

Results and Discussion

Adsorption of (KL)4K and (KG)4K to a DPPG monolayer In experiments performed without a lipid monolayer the surface pressure remains constant at 0 mN m-1 (see Figure 3 in panels a-e (black curves)). The high abundance of charged lysines ensure good water solubility and although hydrophobic amino acids like valine and leucine are incorporated into peptides their presence does not lead to a measurable surface activity. We then determined adsorption isotherms of (KG)4K and of (KL)4K to a DPPG monolayer by injecting different amounts of a peptide solution into the subphase underneath the DPPG monolayer with an initial surface pressure of 25 mN m-1 for (KG)4K and 5 mN m-1 for (KL)4K. These values were chosen to obtain significant changes in Π (see Figure 3). Depending on the initial surface pressure the total amount of spread lipids differ and the theoretical lipid concentration at the air-water surface of the trough ranges from 0.6 up to 1.1

µM. Binding of (KG)4K to the DPPG monolayer decreases and of (KL)4K increases the surface pressure as a function of concentration of the peptide (see Figure 2). Above a peptide concentration of 1-3 µM saturation was reached and the change in surface pressure remained constant (see Figure 2, upper panel). According to the Π–A isotherm a DPPG molecule at 5 mN m-1 covers an area of 90 Å2 at the monolayer the beginning plateau value of ∆Π for (KL)4K coincides with a lipid to peptide ratio of 1 at the given trough dimensions. Charge compensation is reached at even lower concentrations because each peptide carries five positive charges. At the used peptide concentration of 3 µM an excess of peptide is provided to ensure that the adsorption maximum is reached for all different peptides and initial surface pressure values, although most of the peptide stays in the subphase.


Page 8 of 38

Page 9 of 38

Langmuir

9

10

5 -1

-1

DPPG monolayer at 5 mN m , injection of (KL)4K -1

∆Π / mN m

DPPG monolayer at 25 mN m , injection of (KG)4K

0

-5

-10

0

2

4

6

8

10

c / µM 10

Injection of (KL) 4K to a DPPG monolayer at 5 mN m

-1

-1

8 ∆Π / mN m

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

theoretical concentration of DPPG -1 2 at 5 m N m (90 Å ) and -1 2 at 30 mN m (50 Å )

6

4

excess of peptide R c =[K]/[lipid]>1

2

0 0

1

2

3

c / µM

Figure 2: Top: Adsorption isotherms of (KG)4K and (KL)4K determined by difference in surface pressure obtained after 5 h of adsorption to a DPPG monolayer of Πini = 25 mN m-1 ((KG)4K) or Πini = 5 mN m-1 ((KL)4K) as a function of their subphase concentration. Bottom: Adsorption isotherm of (KL)4K. Red curve is drawn to guide the eye. Depending on the initial surface pressure the total amount of spread lipids differ, the theoretical lipid concentration at the air-water interface of the trough thus ranges from 0.6 up to 1.1 µM. For the peptide concentration of 3 µM, which has been used in all experiments the amount of peptide is in excess compared to that of the lipid. For pure electrostatic binding of a peptide with five cationic charges one would estimate an apparent binding constant to purely negatively charged lipid bilayers of 1-2 105 M-1, a value supported by experimental results obtained by isothermal titration calorimetry using different


Langmuir

10 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

pentapeptides.24, 26-27 Separating the charged lysine residues by introducing alanine into the peptides sequence reduces the binding constant 3-fold.28 Except for (KG)4K all other peptides have amino acids more hydrophobic than alanine in their sequence. Therefore, the binding constants can be assumed to be higher due to the additional hydrophobic interaction. For a binding constant of 106 M-1 one would calculate that under our experimental conditions 75 % of all lipids have bound peptides assuming a stoichiometry of 5 lipids per bound peptide solely due to electrostatic effects. The adsorption curves presented in Figure 2 show that saturation is almost reached at a concentration of 3 µM peptide, thus the binding constant must be higher than estimated above and must have a value of ca. 107 M-1 for (KG)4K and even higher for (KL)4K.

Adsorption of (KX)4K peptides to DPPG monolayers at different initial surface pressures Adsorption experiments of (KX)4K peptides to a DPPG monolayer at different initial surface pressures were performed to study the effect of the peptide on physical state of the monolayer, to monitor the kinetics of binding and a possible incorporation of the peptide into the DPPG monolayer. After injection of a peptide solution into the subphase underneath the DPPG monolayer the surface pressure changed due to interaction of the peptides with the lipid molecules (see Figure 3). The change in surface pressure depends on the initial surface pressure Πini of the lipid monolayer and also on the sequence of the peptide. Adsorption of basic peptides to negatively charged lipids is mainly driven by electrostatic interactions leaving zwitterionic lipids mostly unaffected.29-31 A decrease in surface pressure is observed as a result of condensation of the monolayer due to electrostatic interactions leading to a shielding of the repulsive head group charges when the total area of the monolayer is constant.24 Lipid condensation is also seen for electrostatic binding of divalent cations to PG monolayer.32 The net charge of all tested peptides is +5 at neutral pH and due to length and flexibility of lysine ACS Paragon Plus Environment

Page 10 of 38

Page 11 of 38

Langmuir

11 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

side chains several lipid molecules can be bridged upon binding, the oligopeptide thus functioning as a pentadentate ligand. That an oligopeptide with several basic residues can act as a bridging agent binding to more than one lipid was shown theoretically and proven experimentally. It was found that the apparent free energy of binding becomes more negative by ca. 1 kcal mol-1 with every additional basic residue.26-27 The efficacy of bridging depends also on the charge distance between the cationic charges in the side chain of the peptide compared to the average distance of the lipid molecules. A mean distance of 7.2 Å between two α-carbon atoms in the peptide backbone of neighbouring lysine residues was calculated for the sequence (KX)4K when in an extended conformation. As DPPG molecules carry their negative charge at the phosphate group in close proximity to the air-water interface, one could expect that the positively charged peptide might access the negatively charged phosphate group and, as a consequence build up hydrogen bonds with the ammonium group of its lysine residues. In a DPPG monolayer in the liquid-expanded state and a surface pressure of 5 mN m-1 one lipid occupies roughly 90 Å2 and thus two headgroups are 9.5 Å apart. Based on this simple calculation, lipid condensation in the liquid-expanded state is likely to occur optimize lipid-peptide contacts, as a result of the neighbouring distances between lysine residues. Compression of the monolayer to 30 mN m-1 reduces the area per lipid and in the LC state the molecules now are closer together having a distance of 7.1 Å. However, a condensation effect caused by electrostatic shielding of negative lipid headgroup charges is still possible as the side chains of the peptide can adapt to smaller distances. As the DPPG isotherm has a very steep slope at this surface pressure in the LC state, the resulting surface pressure decrease can be very significant even if the area change is small. To study the effect of additional hydrophobic interactions we used peptides with amino acids of increased hydrophobicity. Peptides interacting with the acyl chain region can be located between the lipid molecules occupying space at the air-water interface thus the surface


Langmuir

12 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

pressure will increase at constant total area.33 Other hydrophobic residues leading to an increase in surface pressure are long alkyl chains. Many membrane bound proteins contain membrane anchors, for instance, acyl or farnesyl moieties.34-35 These residues insert into the lipid monolayer leading to a concomitant increase of surface pressure. This occurs not only for charged monolayers but also for monolayers containing only zwitterionic lipids as the driving force is the hydrophobic effect.36 Although lysine as charged residue preferentially acts via electrostatics the lysine side chain can also contribute to hydrophobic interactions.23-24 A superposition of electrostatic and hydrophobic interaction is observed when peptides are charged on the one hand and have hydrophobic side chains on the other hand.23 The effect that the injected positively charged peptide is first electrostatically attracted to the lipid monolayer driven by interactions between phosphate moieties and lysine residues and then bound strongly to the headgroup region via additional hydrophobic interaction obtained by a clustering of leucine side chains into the more hydrophobic region of the bilayer is known from studies with the α-helical antimicrobial peptides KLAL37 and (LKKL)4.38 To tune the contributions of additional hydrophobic interactions we therefore changed the amino acid X systematically from G to A, Abu, V, and L. Adsorption kinetics of (KG)4K to a DPPG monolayer (KG)4K has the least hydrophobic amino acid glycine G. The adsorption kinetics of (KG)4K to a DPPG monolayer differ below and above the LE/LC phase transition region at 8-10 mN m-1 (see Figure 3a). After the injection of (KG)4K under a DPPG monolayer in the LE state with a surface pressure Π lower than 10 mN m-1 first a decrease in Π over a period of half an hour is detected followed by a much slower increase in Π. This indicates that at least two consecutive processes take place, the initial fast one reducing Π, followed by a slower process leading to an increase in Π.


Page 12 of 38

Page 13 of 38

Langmuir

13

4

50

a

1

2

t/h

3

4

5

b 35

(KA)4K

30

30

25

25

20

20

15

15

10

10

5

5

-1

3

0

30

30

25

25

20

20

15

15

10

10

5

5

0 35

0 35

e

(KL)4K

f

30

30

25

25

20

20

15

15

10

10

5

5

0

0 80 100 120 140

0

1

2

3 t/h

4

5 40

60

2

-1

d 35

(KV)4K

Π / mN m

c

(KAbu)4K

-1

-1

Π / mN m

t/h

(KG)4K

0 35

-1

2

Π / mN m

Π / mN m

-1

35

1

Π / mN m

0

Π / mN m

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

-1

A / Å molecule

Figure 3: Change in surface pressure of a DPPG monolayer as a function of time after injection of different peptides into the subphase (T = 20 °C, 100 mM NaCl) at various initial surface pressures. The lipid monolayer remains at a constant surface pressure prior to peptide injection at t = 600 s (marked with the dashed line). The final concentration of the peptides (panel a-e) in the subphase was 3.0 µM. Adsorption equlibrium was reached within 5h; panel f: surface pressure area isotherm of a pure DPPG monolayer, the LE/LC phase transition region between 8 mN m-1 and 10 mN m-1 is highlighted by shading.


Langmuir

14 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

A possible, still speculative explanation for the observed two processes is the following: the electrostatic adsorption of the highly charged peptide to the lipid headgroup region first causes a shielding of the headgroup charges and a concomitant reduction in repulsive forces. The resulting increase in chain order and the reduction of the area per lipid molecule with possible bridging of more than one lipid due to binding of a peptide molecule decreases the surface pressure. The subsequent slower increase in surface pressure is then possibly caused by a reorientation of the peptides so that partial insertion of the side chains in between the lipid headgroups can occur. A two-step model for binding of cationic peptides to a lipid layer composed of negatively charged molecules has been before suggested and experimentally proven for monolayers24 as well as for bilayer systems.39 Injections of (KG)4K into the subphase of a DPPG monolayer with an initial surface pressure Πini above 10 mN m-1, i.e. in the LC state, lead only to a decrease of the surface pressure without any second process. The decrease in Π becomes higher with increasing Πini of the film. This can be explained by the form of the monolayer surface pressure-area isotherm of DPPG (see Figure 3f) which becomes steeper with increasing compression of the film. The shielding of the repulsion between the headgroups by electrostatic adsorption of the peptides induces a chain ordering and condensation of the film although the lipid order in the LC state is already higher compared to the LE state. Incorporation of the peptide into the densely packed lipid monolayers is less likely so that the condensation effect dominates. This interpretation is supported by infrared reflection adsorption spectroscopy experiments. We found a slight decrease in wavenumber of the CH2-stretching vibrational bands after binding of the peptides to LC state DPPG monolayers. An example for the peptide (KL)4K is shown in Figure SI1of the supplementary information. The binding of (KG)4K leads to a similar decrease of ν(CH2)asym by ca. 0.6 cm-1 indicating a further ordering of the acyl chains in the LC state. From these findings it can be concluded that the peptide (KG)4K is not hydrophobic


Page 14 of 38

Page 15 of 38

Langmuir

15 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

enough to penetrate tightly packed DPPG monolayers in the LC state and is only electrostatically adsorbed to the lipid headgroup region, whereas it can more readily penetrate a less dense monolayer in the LE state. Adsorption kinetics of (KA)4K to a DPPG monolayer The peptide (KA)4K contains the more hydrophobic amino acid alanine instead of glycine. The adsorption kinetics of the peptide (KA)4K are independent of the initial surface pressure Πini and no influence of the phase state of the lipid monolayer was detected. All kinetic curves show a decrease of the surface pressure over time (see Figure 3b), the decrease becoming more prominent at higher Πini. At the highest and 2nd highest value of Πini the slow kinetics prevents the determination of the equilibrium value. Adsorption kinetics of (KAbu)4K to a DPPG monolayer The unnatural amino acid α-amino butyric acid (Abu) was used to bridge the gap between the natural amino acids alanine and valine, having an intermediate hydrophobicity. For (KAbu)4K a distinct dependence of the adsorption kinetics on the phase state of the DPPG monolayer was observed. For DPPG in the LE state, first an increase in surface pressure is visible followed by a slower decrease of Π. Compared to the adsorption kinetics of (KG)4K, a reversed sequence of surface pressure increase and decrease is thus observed. The following assumptions to explain these observations are difficult to prove. One could assume that the initial electrostatic adsorption immediately leads to hydrophobic interactions with concomitant insertion of (KAbu)4K into the lipid monolayer and thus to an increase in Π. The following slower decrease in Π after 30 min could then be due to a reorientation of (KAbu)4K with concomitant better electrostatic interaction, overcompensating the hydrophobic interactions. The final equilibrium surface pressure is then below Πini, resulting in an overall negative ∆Π. It could be that the peptides change their secondary structure from an unordered conformation in solution to a β-sheet structure when bound to the membrane ACS Paragon Plus Environment

Langmuir

16 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. In this structure the electrostatic interactions might be increased compared to an unordered structure, as all side chains are oriented to one side and the charge distances between the cationic side chains have a better match with the lipid headgroup separation. Preliminary IR-experiments with DPPG vesicles with bound (KAbu)4K indicate that β-sheet structure formation is indeed possible (unpublished results). For binding to the LC state only the decrease of the surface pressure is visible due to lipid condensation by electrostatic interactions. Adsorption kinetics of (KV)4K to a DPPG monolayer The amino acid valine in (KV)4K has an even larger hydrophobic side chain compared to the peptides presented before. Therefore, immediately an incorporation of the peptide into the lipid monolayer with an increase in Π can be detected when (KV)4K is adsorbed to a DPPG monolayer up to an initial surface pressure Πini of 16 mN m-1. The condensation effect leading to a surface pressure decrease is probably present but overcompensated by the surface pressure increase due to insertion. Only at low surface pressure an overshoot of the surface pressure above the final equilibrium surface pressure is visible with a consecutive slight decrease of Π. The comparison with the kinetics of the (KAbu)4K incorporation reveals that the adsorption becomes faster with increasing hydrophobicity of the peptide. Above Πini values of 16 mN m-1, a similar behaviour was detected as observed before for (KAbu)4K, namely only a decrease of the surface pressure after adsorption. Adsorption kinetics of (KL)4K to a DPPG monolayer Leucine is the amino acids with the highest hydrophobicity used here. The kinetics of the adsorption of (KL)4K to a DPPG monolayer is clearly dominated by the insertion of the hydrophobic peptide into the monolayer. Immediately after injection into the subphase Π increases and at intermediate values of Πini an overshoot is visible followed by the slower condensation of the monolayer coupled with the decrease of Π. Independent of Πini in the LC


Page 16 of 38

Page 17 of 38

Langmuir

17 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

state, the final surface pressure reached is 17-19 mN m-1. A likely explanation for the constancy of these values is the observation that the plateau for the LE/LC phase transition with bound (KL)4K is in this surface pressure range as seen in the isotherm obtained with a different setup (see epifluorescence experiments below and Figure 8). Due to the high affinity of the peptide to the lipid monolayer lipid condensation becomes stronger in the LC state. Nevertheless the peptide incorporation still counteracts the lipid condensation. As both processes have different kinetics a shoulder in the adsorption experiments is observed.

Kinetic analysis The kinetics of the two observed effects, the decrease of Π upon lipid condensation and the increase of Π upon incorporation into the monolayer, change their relative order by increasing the hydrophobicity of the added peptide. Lipid condensation shows faster kinetics during adsorption of (KG)4K followed by a slower process of peptide incorporation. We tried to quantitatively analyze the kinetics of adsorption by fitting the experimental adsorption curves obtained for the five different peptides at a specific initial surface pressure using the following bi-exponential equation:

Π = A1e − t / t + A2 e − t / t + Π 0 1

2

(1)

The parameters A1 and t1 provide information on the amplitude and rate constant for the peptide incorporation and A2 and t2 for the condensation effect, the value Π0 being the final surface pressure at equilibrium after 5 hours. The amplitude of both effects corresponds to the properties of the peptides describing the relative strength of the electrostatic and hydrophobic interaction with the DPPG monolayer. Figure 4 shows the experimental and fitted curves for the adsorption of the different peptides to a DPPG monolayer starting from a surface pressure of ca. 5-7 mN m-1, where the monolayer is in the LE state. Fitting the adsorption kinetics of peptides to LC state monolayers was omitted because only a decrease of surface pressure is


Langmuir

Page 18 of 38

18 observed with the exception of the most hydrophobic peptide (KL)4K. An overview of the obtained parameters is given in Table 1. 18 16

-1

14

Π / mN m

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

12

2

(KG)4K,

fit, R =0,99

(KA)4K,

fit, R =0,99

(KAbu)4K,

fit, R =0,99

(KV)4K,

fit, R =0,93

(KL)4K,

fit, R =0,95

2 2 2 2

10 8 6 4 0

1

2

3

4

5

time / h

Figure 4: Time dependent development of the surface pressure of a DPPG monolayer after injection of different peptide solutions into the subphase at a initial surface pressure of 5-6 mN m-1 in the LE state (T = 20 °C, 100 mM NaCl). The final concentration of the peptide in the subphase was 3 µM. Adsorption equlibrium was reached within 5 h. Table 1: Fit values for the adsorption kinetics of an injected peptide into the subphase of a DPPG monolayer at 5-7 mN m-1. Values for (KA)4K in parenthesis are for a monoexponential fit as a bi-exponential fit could not reliably be performed. A1, A2: amplitudes of the two processes, t1, t2: relaxation rates, Π0: final surface pressure after 5 h. Data obtained by least square fit with the given bi-exponential equation. Errors are from numerical fitting routine of Origin 8.0. peptide incorporation

lipid condensation

peptide

A1 / mN m-1

t1 / h

A2 / mN m-1

t2 / h

Π0 / mN m-1

(KG)4K

-1.55 ± 0.02

2.48 ± 0.14

1.49 ± 0.03

0.28 ± 0.01

6.17 ± 0.03

(1.66 ± 0.01)

(1.25 ± 0.02)

3.64 ± 0.01

(KA)4K (KAbu)4K

-1.66 ± 0.03

0.15 ± 0.01

2.45 ± 0.02

1.08 ± 0.01

3.90 ± 0.01

(KV)4K

-8.41 ± 0.61

0.18 ± 0.01

2.94 ± 0.64

0.65 ± 0.09

9.82 ± 0.02

(KL)4K

-15.00 ± 0.49

0.08 ± 0.01

3.53 ± 0.50

0.23 ± 0.04

16.44 ± 0.08


Page 19 of 38

Langmuir

19 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

From these data certain trends are visible. The amplitude A2 describing the lipid condensation effect should be similar for all peptides because all peptides have the same overall charge. Indeed, the value for A2 increases only slightly with the hydrophobicity of the peptide as stronger hydrophobic interactions lead to a closer location of the charged lysine side chains at the surface and lead to more effective lipid condensation. The relaxation time t2 shows first an increase starting from (KG)4K and then a significant decrease going from (KA)4K to (KL)4K. In the case of (KA)4K binding only the slow condensation effect was visible. It could be that both processes have similar relaxation times so that they cannot be separated. The relaxation time t1 describing the effect of peptide incorporation becomes shorter the more hydrophobic the peptide is, the relaxation times for (KAbu)4K and (KV)4K being very similar and for (KL)4K being shorter by a factor of only 2. It could be that for the peptide with the highest hydrophobicity the diffusion of the peptide to the air-water interface is the rate limiting step so that only small differences between the most hydrophobic peptides are observed. The amplitude A1 of this effect correlates with the hydrophobicity and increases from (KG)4K to (KL)4K due to a possibly deeper incorporation of the peptides into the lipid headgroup region. Comparison with literature data shows that similar adsorption kinetics were observed for the binding of poly-L-lysine or poly-L-arginine to a DPPG monolayer23 with no additional intervening hydrophobic amino acids. For poly-L-arginine, however, the arginine side chain is more hydrophobic so that even for DPPG in the LC state a surface pressure increase is observed due to an insertion of the side chain into the headgroup region. For the pentapeptide Lys5 which is less hydrophobic, this surface pressure increase is reduced. Further reduction of spacer length enhances the effect of condensation.24

Changes of surface pressure as a function of Πini The change in surface pressure ∆Π = Π0 - Πini observed after adsorption of a protein or peptide to a lipid monolayer as a function of the initial surface pressure Πini shows in many


Langmuir

Page 20 of 38

20 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

cases a linear relationship, from which the maximum insertion pressure MIP is determined, i.e. the value of Πini where ∆Π = 0 mN m-1. Per definition MIP is the value above which no incorporation into the lipid bilayer occurs anymore.11, 40-42 However, we had already observed before for the binding of cationic pentapeptides to DPPG monolayers that this relation and the interpretation cannot be used for the electrostatic binding of peptides, as negative surface pressure changes and also changes in slope were observed.24 This is again found for the peptides used in this study (see Figure 5). For (KG)4K ∆Π values are very small for an adsorption to a monolayer in the LE state but become more negative with increasing Πini in the LC state. For (KA)4K and (KAbu)4K with low hydrophobicity all values for ∆Π are negative and a linear behaviour of ∆Π vs. Πini is seen for the adsorption to a monolayer in the LC state. For the more hydrophobic peptides (KV)4K and (KL)4K a different behaviour is observed. At low initial surface pressure Πini an increase of ∆Π up to a certain maximum value of Πini is followed by a decrease in ∆Π which becomes negative at higher values of Πini. Thus the adsorption of the peptides studied here leads to a completely different behaviour of ∆Π vs. Πini as observed before for proteins, where a linear behaviour and only positive ∆Π values were observed.13,40-42 The observed negative values of ∆Π are, in our opinion, caused by the condensation of the lipid monolayer due to electrostatic shielding of charges. As shown in Figure 5 ∆Π decreases linearly with increasing Πini only at initial surface pressures Πini > 10 mN m-1 where the DPPG monolayer is in the LC state according to:

∆Π = mΠ ini + n

(2)

The values for m (slope of the regression line), n (intercept with y), and Πsp (superposition surface pressure) as shown in Table 2 were all obtained at a given subphase concentration of peptides and for binding to a DPPG monolayer in the LC state. ACS Paragon Plus Environment

Page 21 of 38

Langmuir

21 The observed value of ∆Π is determined by a superposition of condensing and insertion effects. When ∆Π = 0 mN m-1 both effects exactly compensate. This occurs at a value of Πini = Πsp = -n/m which we call here the superposition surface pressure Πsp.

(KG)4K

10

(KA)4K (KAbu)4K (KV)4K

5 -1

(KL)4K ∆Π / mN m

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

0

-5

-10 LE state

LC state

-15 0

5

10

15 Πini / mN m

20

25

30

-1

Figure 5: Difference of surface pressure ∆Π after 5 h of adsorption of a peptide solution to a DPPG monolayer with different initial surface pressure Πini, the lines at ∆Π = 0 and Πini = 10 mN m-1 are drawn to guide the eye and to distinguish between the different states of the DPPG monolayer. The final concentration of the peptide in the subphase was was 3 µM. Table 2: Values of m, n, Πsp, and R2 obtained from the fitting of data for binding of the peptide to a DPPG monolayer in the LC state presented in Figure 5. Errors of the parameters m, n, and Πsp from least square fit of the regression lines calculated with Origin 8.0. intercept n / mN m-1

slope m

superposition surface pressure, Πsp,/ mN m-1

R2

(KG)4K

4.9 ± 1.6

-0.44 ± 0.08

11.0 ± 4.2

0.849

(KA)4K

2.0 ± 0.9

-0.40 ± 0.04

4.9 ± 2.2

0.967

(KAbu)4K

2.5 ± 1.5

-0.39 ± 0.07

6.3 ± 3.9

0.910

(KV)4K

8.9 ± 1.4

-0.56 ± 0.08

15.9 ± 3.4

0.904

(KL)4K

17.5 ± 0.9

-1.00 ± 0.04

17.5 ± 1.1

0.990

peptide


Langmuir

22 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

For adsorption of (KG)4K at low initial surface pressure to DPPG monolayer in the LE state below 10 mN m-1 no notable changes are visible. As the kinetics show two distinct processes of peptide insertion and lipid condensation they compensate each other fully so that ∆Π = 0 mN m-1. For the LC state, the surface pressure decreases linearly with increasing Πini and thus the lipid condensation becomes more prominent. Although there are differences in the adsorption kinetics between (KA)4K and (KAbu)4K the dependence of ∆Π on Πini is similar. Adsorption of these peptides causes a decrease in surface pressure at all initial surface pressure values. There is no difference in behaviour in the LE and LC state. The superposition of incorporation and condensation is reached at Πsp = ~5-6 mN m-1. For the binding of (KV)4K to the DPPG monolayer a different behaviour is seen. Up to Πini = 16 mN m-1 incorporation of the peptide into the lipid monolayer dominates and a positive ∆Π is observed which reaches a maximum at Πini = 10 mN m-1. The superposition surface pressure Πsp of the peptide (KV)4K is much higher with ~16 mN m-1 indicating the increasing dominance of the insertion effect. Adsorption of (KL)4K to a DPPG monolayer in the surface pressure range up to 10 mN m-1 first leads to an increase in surface pressure and only in the range of Πini from 10 mN m-1 up to 30 mN m-1 a linear decrease in ∆Π is seen. The superposition surface pressure Πsp of 17.5 ± 1.1 mN m-1 is the highest observed for all peptides. The slope m seems to be related to the extent of the hydrophobic contribution, as it increases for peptides with more hydrophobic amino acids, such as (KV)4K, and (KL)4K compared to (KG)4K. When the values m, n and Πsp were analyzed a coupling of these three parameters was observed as shown in Figure 6. All three parameters have minimal values for peptides with intermediate hydrophobicity, i.e. (KA)4K or (KAbu)4K.


Page 22 of 38

Page 23 of 38

Langmuir

23

20 18

intercept n superposition surface pressure

16

1,0

Πsp

0,8

12 10 0,6

8

-m

-1

14 n, Πsp / mN m

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

6 4

0,4

2

slope -m:

0 -H

-CH3

-CH2CH3

(KG)4K

(KA)4K

(KAbu)4K

0,2 -CH(CH3)2 -CH2CH(CH3)2 (KV)4K

(KL)4K

Figure 6: Graphical representation of the values m, n and Πsp for peptides (KX)4K with different side chain residues of the uncharged amino acid X.

Adsorption of peptides to DPPG monolayers studied with epifluorescence microscopy Epifluorescence microscopy experiments were performed to investigate whether the adsorption of the peptides to the lipid monolayers and their possible incorporation have an effect on domain formation and their shape. In the LE state of pure DPPG monolayer at 3.4 mN m-1 the fluorescence marker Rh–DHPE is homogeneously distributed and the fluorescence micrograph shows a uniform bright monolayer as the onset of the LE/LC phase transition occurs at a surface pressure of about 810 mN m-1 (see Figure 7). Upon peptide addition the formation of liquid-condensed domains is triggered in all cases despite the fact that Π can increase or decrease, depending on the added peptide. The LC-domains appear dark as the fluorescence marker molecule is excluded from them.25 Within each image all domains have similar sizes, except for the DPPG monolayer with adsorbed (KG)4K showing a distribution of the domain sizes. After 4h of adsorption ca. 17% of the total monolayer area is covered with circular domains in the case of adsorption of (KG)4K, (KA)4K and (KAbu)4K, although the surface pressure is lower compared to that of the pure DPPG monolayer before peptide injection and is also still way ACS Paragon Plus Environment

Langmuir

24 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

below the phase transition pressure of pure DPPG (see Figure 7). This confirms the assumption stated above that upon electrostatic binding of the peptides to the DPPG monolayer lipid condensation occurs and the acyl chains become more ordered with the result of the formation of liquid-condensed domains.43 After adsorption of (KV)4K and (KL)4K, only 6 % and 11 %, respectively, of the monolayer show condensed domains. After binding of these two peptides Π increases indicating that hydrophobic parts of the peptide are incorporated into the headgroup region of the lipid monolayer overcompensating the effect of electrostatic condensation and thus leading to a smaller number of LC-domains.

Figure 7: Fluorescence microscopy images of a DPPG monolayer at the air-water interface before peptide adsorption and after 4-5 h of adsorption of the linear peptides reaching final surface pressures as indicated. The final concentration of the peptide in the subphase was 3 µM. Lipid dye: 0.01 mol% Rh–DHPE. The scale bar given in the picture of pure DPPG holds for all other pictures in this figure. After 5 h of adsorption of the peptides to the lipid monolayer, a slow compression of the monolayer was started, the surface pressure was monitored and epifluorescence microscopy


Page 24 of 38

Page 25 of 38

Langmuir

25 images were acquired during the compression process. The compression isotherms and the calculated compressibility values for pure DPPG and for DPPG with adsorbed peptide from the subphase are shown in Figure 8 (see Figure SI5 for complete isotherms). The epifluorescence microscopy images recorded during the compression of the monolayer are depicted in Figure 9. pure DPPG monolayer DPPG monolayer with adsorbed peptide (KG)4K

60 50

(KAbu)4K

40

Π / mN⋅m

-1

(KA)4K (KV)4K (KL)4K

30 20 10 0 50

60

70 · Å

A/

0,25

κ / m mN

-1

0,20

2

molecule

80

90

-1

pure DPPG monolayer DPPG monolayer with adsorbed peptide (KG)4K (KA)4K (KAbu)4K (KV)4K

0,15

(KL)4K

0,10 0,05 0,00 50

60

70 Å

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

A/

2

molecule

80

90

-1

Figure 8: Top: Surface pressure area isotherm of a pure DPPG monolayer and DPPG monolayer with adsorbed peptide from the subphase. Bottom: Compressibility of a pure DPPG monolayer and DPPG monolayers with adsorbed peptide. The observed compression isotherm of pure DPPG with its characteristic plateau starting at about 8 mN m-1 for the phase transition from the LE to the ordered LC state is in good


Langmuir

26 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

agreement with literature data.44-46 Micrographs of pure DPPG monolayers (Figure 9, upper row) show a homogeneous bright monolayer in the LE state. The appearance of the first irregular dark domains of condensed lipids fits with the LE/LC phase transition with its high compressibility between 80 and 65 Å2/molecule (Figure 8, lower panel). Further compression increases the area covered with condensed lipids and the dye is excluded from these ordered domains and forced to accumulate at domain boundaries, giving a higher contrast in the image. The compression isotherm obtained after binding of (KG)4K is similar compared to the DPPG isotherm, only the phase transition region seems to be smeared out after (KG)4K adsorption leading to a reduced compressibility compared to pure DPPG (Figure 8, lower panel). Adsorption of (KA)4K and (KAbu)4K first lowers the surface pressure of the DPPG monolayer and the compression isotherm therefore starts below the one of pure DPPG at high molecular areas. During compression of the mixed films also only a smeared-out plateau is visible as seen before for (KG)4K, which is slightly more prominent for binding of (KAbu)4K. The adsorption of the more hydrophobic peptides (KV)4K and (KL)4K increase the surface pressure due to their insertion into the monolayer in the LE state at high molecular areas. The largest effect is seen for (KL)4K. Compression of the DPPG/(KV)4K film shows an onset of the phase transition pressure at 15 mN m-1. For the DPPG/(KL)4K the phase transition pressure is even higher with 19 mN m-1. These two peptides with their larger hydrophobic side chains insert in between the lipid headgroups in the LE state thus increasing the phase transition pressure. The compressibility in the transition region is reduced for mixed monolayers compared to pure DPPG. A direct dependence of compressibility of the mixed monolayers with the hydrophobicity of the added peptide is seen for (KA)4K to (KL)4K (Figure 8, lower panel). The fluorescence micrographs taken during the compression of the DPPG monolayer after peptide binding are quite similar for the adsorbed peptides (KG)4K, (KA)4K, (KAbu)4K and ACS Paragon Plus Environment

Page 26 of 38

Page 27 of 38

Langmuir

27 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

(KV)4K. Dark condensed domains are visible even at a surface pressure below the LE/LC phase transition region of pure DPPG and the plateau surface pressures of the film with bound peptides (see Figure 9). The epifluorescence images strongly suggest that lipid condensation occurs after adsorption of the peptide to the monolayer in the LE state (see Figure SI2, SI3, supplementary information). Compression of these mixed monolayers from 90 Å2/molecule to 55 Å2/molecule also increases the area covered by the condensed domains as expected. However, compared to the domains seen for pure DPPG films, the distribution of the LCdomains is not random but becomes highly ordered. Also, all domains have similar size and are almost round at low surface pressure and form hexagonal shapes at higher surface pressure and lower surface area. A closer look at the organization of the domains reveals a six-fold symmetry in the organization. This becomes much clearer after a Fouriertransformation of the image (see Figure SI4, supplementary information). The domains are stable and seem not to fuse upon compression. Thus it can be concluded that binding of peptides lowers the line tension between LE state and LC-domains. Electrostatic repulsion between the single domains then leads to a quasi-hexagonal arrangement of domains. The occurrence of these hexagonal patterns results thus from an equilibrium between electrostatic repulsion and line tension as discussed before.47,48 A significantly different behaviour is observed for mixed monolayers of DPPG after adsorption of the peptide with the highest hydrophobicity, (KL)4K (see Figure 9, bottom row). At the beginning of the compression isotherm at high areas per molecule also condensed domains are visible due to the adsorbed peptide. The plateau in the compression isotherm region appears at 19 mN m-1 and corresponds to the onset of changes in the domain shape seen in the epifluorescence micrographs. At the beginning of this plateau at 17 mN m-1 thin filaments of condensed lipid grow from the existing domains acting as nucleation sites. Further compression increases the number and the length of these filaments. Over the whole compression range the initially formed round domains are still present. ACS Paragon Plus Environment

Langmuir

28 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 9: Epifluorescence microscopy images of a pure DPPG monolayer and a DPPG monolayer with adsorbed peptide at the air-water interface (T = 20°C, 100 mM NaCl subphase) recorded at different surface pressure. Lipid dye: 0.01 mol% Rh–DHPE. The scale bar given in the image for a pure DPPG monolayer is the same for all other images. ACS Paragon Plus Environment

Page 28 of 38

Page 29 of 38

Langmuir

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

Apparently, the adsorption of (KL)4K lowers the line tension between LC-domains and the surrounding LE state much more compared to the other peptides due to the bulkier side chain of leucine. Summary and Conclusions In the present study we used different nonapeptides containing five lysine residues separated by intervening uncharged amino acids X leading to general the sequence (KX)4K. The impact of an increased hydrophobicity of X on the binding behaviour of the peptide to negatively charged DPPG monolayers was investigated. The spacer amino acid X leads to a distance of the α-carbon atoms of the charged amino acid K of ~7.2 Å, a distance close to the separation of headgroups of DPPG monolayer in the LE state, so that optimal electrostatic interactions can be expected. Our results using five different peptides (KX)4K show that a very subtle interplay between electrostatic and hydrophobic effects influences the binding properties of these cationic nonapeptides. The equilibrium values of the surface pressure obtained after binding are difficult to interpret. The analysis of the kinetics of adsorption of the peptides to DPPG monolayers gives additional information. The time dependence of the surface pressure showed two processes with different kinetics, a condensation of the lipid monolayer decreasing the surface pressure, and an incorporation reaction increasing the surface pressure. The observed surface pressure changes ∆Π as a function of initial surface pressure Πini showed a complicated behaviour, quite different from previously observed effects when peptides and proteins are incorporated into lipid monolayers.11,13,40,41 For the nonapeptides (KX)4K, the behaviour of ∆Π vs. Πini depends on the hydrophobicity of the intervening amino acid X. At Πini values below 10 mN m-1 the incorporation effect dominates for the more hydrophobic peptides (KV)4K and (KL)4K, whereas condensation and incorporation effects more or less compensate for the other three peptides (KG)4K, (KA)4K and (KAbu)4K. Above an initial surface pressure of 10 mN m-1 the slopes of the ∆Π vs. Πini plots are negative and the


Langmuir

30 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

∆Π values are all negative at high initial surface pressure of the monolayer indicating the dominance of the electrostatic condensation effect. The epifluorescence microscopy supports the results obtained by the adsorption experiments. An induction of LC-domains after peptide adsorption to a monolayer in the LE state was observed for all peptides. Upon compression of DPPG monolayers with pre-adsorbed peptides the size of the domains remained almost constant and a regular quasi-hexagonal pattern of LC-domains at higher surface pressure is formed indicating stabilization of the domains by electrostatic repulsion. In the case of binding of the most hydrophobic peptide (KL)4K, a significantly different domain shape was observed. Due to a stronger decrease in line tension the domains formed longer filament-like extensions. The bound peptide alters the surface pressure-area isotherms but seems to be squeezed out of the monolayer at higher surface pressure so that only the condensing effect remained. The results obtained show that for the binding of these peptides to charged lipid monolayers electrostatic as well as hydrophobic interactions are important. The different contributions can be separated by time dependent adsorption measurements. The observation of equilibrium values of the surface pressure alone might lead to wrong conclusions. For instance, we could clearly show by epifluorescence spectroscopy that the decrease in surface pressure of LE state DPPG found after binding of some of the peptides is connected with formation of LCdomains, though the isotherms seem to indicate that the monolayer is still completely in the LE state. Independent measurements using infrared reflection adsorption spectroscopy (IRRAS) confirmed these observations as a small decrease in the frequency of the CH2stretching bands was observed due to the formation of condensed domains though the majority of the lipids remained in the LE state (see Figure SI1 in supplementary information).


Page 30 of 38

Page 31 of 38

Langmuir

31 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

Acknowledgements This work was supported by the Deutsche Forschungsgemeinschaft (GRK 1026 Conformational Transitions in Macromolecular Interactions, Project A1). Supporting Information Available: Additional information on percentage of LC-domains after peptide binding, Fourier analysis of hexagonal domain patterns, complete DPPG monolayer isotherms with and without bound peptides. This material is available free of charge via Internet at http://pubs.acs.org.


Langmuir

32 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

References 1.

Wang, G.; Li, X.; Wang, Z. APD2: the updated antimicrobial peptide database and its

application in peptide design. Nucleic Acids Res. 2009, 37, D933-D937. 2.

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

infective therapeutic strategies. Nat Biotech 2006, 24, 1551-1557. 3.

Zasloff, M. Antimicrobial peptides of multicellular organisms. Nature 2002, 415, 389-

395. 4.

Matsuzaki, K. Control of cell selectivity of antimicrobial peptides. Biochim. Biophys.

Acta 2009, 1788, 1687-1692. 5.

Nguyen, L. T.; Haney, E. F.; Vogel, H. J. The expanding scope of antimicrobial

peptide structures and their modes of action. Trends Biotechnol. 2011, 29, 464-472. 6.

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

bilayer membranes by K-helical antimicrobial and cell non-selective membrane-lytic peptides. Biochim. Biophys. Acta 1999, 1462, 55-70. 7.

Bechinger, B. Structure and Functions of Channel-Forming Peptides: Magainins,

Cecropins, Melittin and Alamethicin. The Journal of Membrane Biology 1997, 156, 197-211. 8.

Hwang, P. M.; Vogel, H. J. Structure-function relationships of antimicrobial peptides.

Biochem. Cell Biol. 1998, 76, 235-246. 9.

Brezesinski, G.; Möhwald, H. Langmuir monolayers to study interactions at model

membrane surfaces. Adv. Colloid Interface Sci. 2003, 100–102, 563-584. 10.

Brockman, H. Lipid monolayers: why use half a membrane to characterize protein-

membrane interactions? Curr. Opin. Struct. Biol. 1999, 9, 438-443. 11.

Maget-Dana, R. The monolayer technique: a potent tool for studying the interfacial

properties of antimicrobial and membrane-lytic peptides and their interactions with lipid membranes. Biochim. Biophys. Acta 1999, 1492, 109-140.


Page 32 of 38

Page 33 of 38

Langmuir

33 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

12.

Arouri, A.; Kerth, A.; Dathe, M.; Blume, A. The Binding of an Amphipathic Peptide

to Lipid Monolayers at the Air/Water Interface Is Modulated by the Lipid Headgroup Structure. Langmuir 2011, 27, 2811-2818. 13.

Giehl, A.; Lemm, T.; Bartelsen, O.; Sandhoff, K.; Blume, A. Interaction of the GM2-

activator protein with phospholipid–ganglioside bilayer membranes and with monolayers at the air–water interface. Eur. J. Biochem. 1999, 261, 650-658. 14.

Hädicke, A.; Blume, A. Interactions of Pluronic block copolymers with lipid

monolayers studied by epi-fluorescence microscopy and by adsorption experiments. J. Colloid Interface Sci. 2013, 407, 327-338. 15.

Schwieger, C.; Chen, B.; Tschierske, C.; Kressler, J.; Blume, A. Organization of T-

shaped facial amphiphiles at the air/water interface studied by infrared reflection absorption spectroscopy. J. Phys. Chem. B 2012, 116, 12245-12256. 16.

Alhakamy, N. A.; Kaviratna, A.; Berkland, C. J.; Dhar, P. Dynamic Measurements of

Membrane Insertion Potential of Synthetic Cell Penetrating Peptides. Langmuir 2013, 29, 15336-15349. 17.

Ishitsuka, Y.; Pham, D. S.; Waring, A. J.; Lehrer, R. I.; Lee, K. Y. C. Insertion

selectivity of antimicrobial peptide protegrin-1 into lipid monolayers: Effect of head group electrostatics and tail group packing. Biochim. Biophys. Acta 2006, 1758, 1450-1460. 18.

Travkova, O. G.; Brezesinski, G. Adsorption of the antimicrobial peptide arenicin and

its linear derivative to model membranes – A maximum insertion pressure study. Chem. Phys. Lipids 2013, 167–168, 43-50. 19.

Kerth, A.; Erbe, A.; Dathe, M.; Blume, A. Infrared Reflection Absorption

Spectroscopy of Amphipathic Model Peptides at the Air/Water Interface. Biophys. J. 2004, 86, 3750-3758.


Langmuir

34 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

20.

Travkova, O. G.; Andrä, J.; Möhwald, H.; Brezesinski, G. Influence of Arenicin on

Phase Transitions and Ordering of Lipids in 2D Model Membranes. Langmuir 2013, 29, 12203-12211. 21.

Reuter, M.; Schwieger, C.; Meister, A.; Karlsson, G.; Blume, A. Poly-l-lysines and

poly-l-arginines induce leakage of negatively charged phospholipid vesicles and translocate through the lipid bilayer upon electrostatic binding to the membrane. Biophys. Chem. 2009, 144, 27-37. 22.

Schwieger, C.; Blume, A. Interaction of poly(L-lysines) with negatively charged

membranes: an FT-IR and DSC study. Eur. Biophys. J. 2007, 36, 437-450. 23.

Schwieger, C.; Blume, A. Interaction of Poly(l-arginine) with Negatively Charged

DPPG Membranes: Calorimetric and Monolayer Studies. Biomacromolecules 2009, 10, 21522161. 24.

Hoernke, M.; Schwieger, C.; Kerth, A.; Blume, A. Binding of cationic pentapeptides

with modified side chain lengths to negatively charged lipid membranes: Complex interplay of electrostatic and hydrophobic interactions. Biochim. Biophys. Acta 2012, 1818, 1663-1672. 25.

Dyck, M.; Lösche, M. Interaction of the Neurotransmitter, Neuropeptide Y, with

Phospholipid Membranes: Film Balance and Fluorescence Microscopy Studies. J. Phys. Chem. B 2006, 110, 22143-22151. 26.

Ben-Tal, N.; Honig, B.; Peitzsch, R. M.; Denisov, G.; McLaughlin, S. Binding of

small basic peptides to membranes containing acidic lipids: theoretical models and experimental results. Biophys. J. 1996, 71, 561-575. 27.

Kim, J.; Mosior, M.; Chung, L. A.; Wu, H.; McLaughlin, S. Binding of peptides with

basic residues to membranes containing acidic phospholipids. Biophys. J. 1991, 60, 135-148. 28.

Mosior, M.; McLaughlin, S. Binding of basic peptides to acidic lipids in membranes:

effects of inserting alanine(s) between the basic residues. Biochemistry 1992, 31, 1767-1773.


Page 34 of 38

Page 35 of 38

Langmuir

35 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

29.

Dufourcq, J.; Faucon, J. F.; Maget-Dana, R.; Pileni, M. P.; Hélène, C. Peptide-

membrane interactions A fluorescence study of the binding of oligopeptides containing aromatic and basic residues to phospholipid vesicles. Biochim. Biophys. Acta 1981, 649, 6775. 30.

Murray, D.; Arbuzova, A.; Hangyás-Mihályné, G.; Gambhir, A.; Ben-Tal, N.; Honig,

B.; McLaughlin, S. Electrostatic Properties of Membranes Containing Acidic Lipids and Adsorbed Basic Peptides: Theory and Experiment. Biophys. J. 1999, 77, 3176-3188. 31.

Murray, D.; Arbuzova, A.; Honig, B.; McLaughlin, S. The role of electrostatic and

nonpolar interactions in the association of peripheral proteins with membranes. In Current Topics in Membranes, Sidney A. Simon, T. J. M., Ed.; Academic Press, 2002; Vol. Volume 52, pp 277-307. 32.

Garidel, P.; Blume, A. 1,2-Dimyristoyl-sn-glycero-3-phosphoglycerol (DMPG)

monolayers: influence of temperature, pH, ionic strength and binding of alkaline earth cations. Chem. Phys. Lipids 2005, 138, 50-59. 33.

Papahadjopoulos, D.; Moscarello, M.; Eylar, E. H.; Isac, T. Effects of proteins on the

thermotropic phase transitions of phospholipid membranes. Biochim. Biophys. Acta 1975, 401, 317-335. 34.

Casey, P. J.; Solski, P. A.; Der, C. J.; Buss, J. E. p21ras is modified by a farnesyl

isoprenoid. Proc. Natl. Acad. Sci. USA 1989, 86, 8323-8327. 35.

Zhang, F. L.; Casey, P. J. Protein Prenylation: Molecular Mechanisms and Functional

Consequences. Annu. Rev. Biochem 1996, 65, 241-269. 36.

Meister, A.; Nicolini, C.; Waldmann, H.; Kuhlmann, J.; Kerth, A.; Winter, R.; Blume,

A. Insertion of Lipidated Ras Proteins into Lipid Monolayers Studied by Infrared Reflection Absorption Spectroscopy (IRRAS). Biophys. J. 2006, 91, 1388-1401. 37.

Erbe, A.; Kerth, A.; Dathe, M.; Blume, A. Interactions of KLA Amphipathic Model

Peptides with Lipid Monolayers. ChemBioChem 2009, 10, 2884-2892. ACS Paragon Plus Environment

Langmuir

36 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

38.

Mangavel, C.; Maget-Dana, R.; Tauc, P.; Brochon, J.-C.; Sy, D.; Reynaud, J. A.

Structural investigations of basic amphipathic model peptides in the presence of lipid vesicles studied by circular dichroism, fluorescence, monolayer and modeling. Biochim. Biophys. Acta 1998, 1371, 265-283. 39.

Hitz, T.; Iten, R.; Gardiner, J.; Namoto, K.; Walde, P.; Seebach, D. Interaction of α-

and β-Oligoarginine-Acids and Amides with Anionic Lipid Vesicles: A Mechanistic and Thermodynamic Study. Biochemistry 2006, 45, 5817-5829. 40.

Calvez, P.; Bussières, S.; Demers, E.; Salesse, C. Parameters modulating the

maximum insertion pressure of proteins and peptides in lipid monolayers. Biochimie 2009, 91, 718-733. 41.

Calvez, P.; Demers, E.; Boisselier, E.; Salesse, C. Analysis of the Contribution of

Saturated and Polyunsaturated Phospholipid Monolayers to the Binding of Proteins. Langmuir 2011, 27, 1373-1379. 42.

Boisselier, É.; Calvez, P.; Demers, É.; Cantin, L.; Salesse, C. Influence of the Physical

State of Phospholipid Monolayers on Protein Binding. Langmuir 2012, 28, 9680-9688. 43.

Trommeshauser, D.; Galla, H.-J. Interaction of a basic amphipathic peptide from the

carboxyterminal part of the HIV envelope protein gp41 with negatively charged lipid surfaces. Chem. Phys. Lipids 1998, 94, 81-96. 44.

Sacré, M. M.; Tocanne, J. F. Importance of glycerol and fatty acid residues on the

ionic properties of phosphatidylglycerols at the air—water interface. Chem. Phys. Lipids 1977, 18, 334-354. 45.

Tocanne, J. F.; Ververgaert, P. H. J. T.; Verkleij, A. J.; van Deenen, L. L. M. A

monolayer and freeze-etching study of charged phospholipids I. Effects of ions and pH on the ionic properties of phosphatidylglycerol and lysylphosphatidylglycerol. Chem. Phys. Lipids 1974, 12, 201-219.


Page 36 of 38

Page 37 of 38

Langmuir

37 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

46.

van Dijck, P. W. M.; de Kruijff, B.; Verkleij, A. J.; van Deenen, L. L. M.; de Gier, J.

Comparative studies on the effects of pH and Ca2+ on bilayers of various negatively charged phospholipids and their mixtures with phosphatidylcholine. Biochim. Biophys. Acta 1978, 512, 84-96. 47.

Andelman, D.; Broçhard, F.; Joanny, J. F. Phase transitions in Langmuir monolayers

of polar molecules. J. Chem. Phys. 1987, 86, 3673-3681.42 48.

McConnell, H. M. Theory of hexagonal and stripe phases in monolayers. Proc. Natl.

Acad. Sci. USA 1989, 86, 3452-3455.


Langmuir

38 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

Graphical abstract


Page 38 of 38

Binding of Short Cationic Peptides (KX)4K to Negatively Charged

Recommend Documents