Interaction of Short Pentavalent Cationic Peptides with Negatively

Subscriber access provided by University of Kansas Libraries

B: Biomaterials and Membranes

Interaction of Short Pentavalent Cationic Peptides with Negatively Charged DPPG Monolayers and Bilayers: Influence of Peptide Modifications on Binding André Hädicke, and Alfred Blume J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.8b08667 • Publication Date (Web): 29 Oct 2018 Downloaded from http://pubs.acs.org on October 31, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 40 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 Journal of Physical Chemistry

1

Interaction of Short Pentavalent Cationic Peptides with Negatively Charged DPPG Monolayers and Bilayers: Influence of Peptide Modifications on Binding

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-Luther-University 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]

ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 2 of 40

2 Abstract The binding of oligopeptides with the structure (RX)4R and (KXX)4K, with X being the amino acid G or A, to lipid monolayers and bilayers of DPPG was studied and compared to the binding effects of peptides with the structure (KX)4K. The monolayer adsorption experiments again showed the superposition of condensation effects due to charge compensation and insertion of amino acid side chains leading to expansion of the monolayer. The latter effect was enhanced when glycine was replaced by arginine. The thermotropic phase behavior of dipalmitoyl-phosphatidylglycerol (DPPG) bilayer membranes and their mixtures with short cationic model peptides was investigated by differential scanning calorimetry (DSC) and infrared (IR) spectroscopy. Increasing the charge distance of the lysine residues in the series (K)5, (KG)4K and (KGG)4K results in an upshift of the main phase transition of DPPG up to 5 K as predicted for pure electrostatic binding. All peptides exhibit only unordered structures in bulk solution as well as bound to DPPG bilayers. (KGG)4K additionally shows a high propensity of turn structures due to its flexibility. The exchange of glycine by arginine in (KAA)4K leads only to a marginal increase in Tm, in contrast to the binding of (KA)4K where the formation of intervesicular antiparallel -sheets occurs leading to a much more pronounced stabilization of the gel phase. This shows that the sequence and flexibility of the oligopeptides has an important influence on the formation of secondary structures bound to the bilayers. Binding of (RX)4R peptides to DPPG bilayers has almost no influence on the lipid phase transition in bilayers. Here condensation and insertion effects almost compensate, as the results of monolayer experiments show. This is due to the higher propensity of arginine side chains to insert into the lipid headgroup region.


Page 3 of 40 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


3 1. Introduction Cationic peptides with a high proportion of lysines or arginines can be used as drugs to kill bacteria or as drug delivery systems to transport small molecules across cell membranes due to their ability to permeate through the membranes. Based on their effect on membranes or cells, these peptides are classified as antimicrobial peptides (AMP) or cell-penetrating peptides (CPP).1-3 Some examples for AMPs are defensins4 or the peptide LL375, and for CPPs the peptides penetratin, transportan, and TAT derivates from HIV-1.6-7 Both types of peptides have in common that the binding to the membrane is driven by electrostatic interactions of the unfolded peptide with the negatively charged lipid bilayers. However, the amino acid sequence of these peptides shows large variations. Consequently, they can also adopt different secondary structures. For some of these peptides the ability to penetrate into or through the lipid bilayer and to make the bilayer leaky8 is connected with the formation of secondary structures which are not present in aqueous solution, but are only formed upon interaction with the bilayer.9 The simplest model systems to determine the driving forces for interaction of either AMPs or CPPs with lipid bilayers are homopolymers of lysine or arginine10. It was found that polylysine is a good model system for AMPs to study the electrostatic interactions11 and the effects of the bound peptides on the behavior of the lipid membranes. A systematic study on the chain length dependence of bound polylysines on the thermotropic phase behavior of DPPG bilayers12 proved that short polylysines decrease the phase transition of phosphatidylglycerols13 whereas longer polylysines shift the phase transition to higher temperature.14 The binding of polyarginines affected the main transition temperature of negatively charged lipid bilayers only marginally. Depending on the chain length, the transition temperature was slightly increased or decreased. This finding was attributed to the superposition of two counteracting effects of



Page 4 of 40

4 peptide binding, namely charge compensation of the headgroup charges and hydrophobic interactions of the arginine side chains with the hydrophobic bilayer interior.15 For an effective insertion of the peptide into the bilayer, the number of arginines should be over 6. It was also found that the guanidinium moiety is more effectively inserted compared to other cationic side chains of amino acids. In addition, the distance of the alkyl spacer between the guanidinium group and the peptide backbone should be longer for more effective interactions and the peptide itself should contain enough conformational flexibility.16-17 The guanidinium moiety plays an important role, particularly its ability to bind to the lipid headgroup and to build up stable bidentate hydrogen bonds with the phosphate group.18-19 The side chain of arginine with its relative high hydrophobicity shows a 10-fold increase in efficiency for the translocation of the peptide with or without an attached carrier through the lipid bilayer than lysine analogues.20 Antimicrobial peptides bind to lipid membranes in a complex interplay via electrostatic interactions coupled with hydrophobic contributions. In this study, we focus on the influence of changes in peptide structure and tune the peptide parameters in a systematical manner. A set of tailor-made model peptides (see Figure 1) was used to study the binding of these peptides to DPPG bilayers and to evaluate changes of the bilayer properties. All peptides contain five positively charged amino acids lysine (K) or arginine (R) beside a neutral amino acid X, resulting in the sequence (KX)4K, (KXX)4K or (RX)4R. The interaction of these positively charged linear oligopeptides with model membranes was investigated to address the following questions: i) influence of the hydrophobicity of the uncharged amino acid in the sequence, ii) influence of the charge distance and periodicity of the peptide by introducing a second uncharged spacer amino acid, iii) influence of exchanging lysine by arginine. The influence of an increased peptide hydrophobicity of the amino acid X on the thermotropic phase behavior of DPPG model membranes was already published in previous papers.21-22


Page 5 of 40 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


5

Figure 1 CPK models of the six linear peptides used in this study in an extended conformation.

The number of positively charged amino acid residues used in this study was fixed to 5. The number of charges in homopeptides of lysine or arginine determines their ability to bind to lipid membranes. For short homopeptides, the binding affinity to anionic bilayers composed of PS, PG or CL increases 10fold with each additional lysine (K) or arginine (R).23-24 To zwitterionic phosphatidylcholines (PC) only very weak binding was observed. In previous studies, we found a general increase of the phase transition temperature of DPPG by up to 20 K when (KX)4K peptides were bound to DPPG bilayers.21 The change in Tm depended on the charge ratio and the hydrophobicity of the uncharged amino acid X. FT-IR spectroscopy showed a stabilization of the gel phase, a higher acyl chain order, and a dehydration of the lipid headgroup region upon peptide binding. FT-IR spectroscopy also revealed that bound (KX)4K peptides can form intermolecular antiparallel β-sheets, their stability being determined by the hydrophobicity of the uncharged amino acid X. The hydrophilic (KG)4K a showed no stable sheet structure whereas for the more hydrophobic (KA)4K a stable β-sheet bound to gel phase DPPG bilayers was observed



Page 6 of 40

6 rearranging to an unordered structure above the lipid phase transition. For (KL)4K the β-sheet remained stably bound even when the DPPG bilayer is in the fluid state at higher temperature. In this paper we analyzed the influence of an increased charge distance and the exchange of lysine by arginine on the binding of peptides DPPG bilayers. In (KXX)4 peptides the distance of the charges is increased. Also the elongation of the chains increases the possibility to build up secondary structures. For short peptides, these possibilities are restricted as the stabilizing hydrogen bonds are too few and therefore a random conformation is more likely. The minimal peptide length for a stable -helix is 14 amino acids, the peptides used here have 13 amino acids at most and are thus below this value.25 Only the possibility for the formation of intermolecular β-sheet structures exists. The exchange of the charged amino acid K by R also alters the distance and the distribution of the positive charge in relation to the peptide backbone. The positive charge of the guanidinium moiety at the arginine side chain is more delocalized. Thus the arginine effect might play a role in peptide-lipid interaction due to its ability to build up stable hydrogen bonds with lipid phosphate groups, thus increasing the binding constant.18-19, 26,27. The peptide influence on the DPPG phase behavior was examined by differential scanning calorimetry (DSC) experiments. We then performed additional temperature dependent Fourier-transform infrared spectroscopy (FT-IR) experiments with these mixtures to obtain information on lipid acyl chain order, headgroup hydration, and induction of secondary structures of the peptides.

2. Experimental 2.1. Materials1,2-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 (KG)4K, (KA)4K, (KGG)4K, (KAA)4K, (RG)4R, and (RA)4R were custom-made via Fmoc solid phase ACS Paragon Plus Environment

Page 7 of 40 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


7 chemistry with a purity more than 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. 2.2. Methods 2.2.1. 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.28 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 defined volume of a freshly prepared 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.).



Page 8 of 40

8 2.2.2. Differential Scanning Calorimetry: Differential Scanning Calorimetry (DSC) was performed with a Microcal VP-DSC (MicroCal Inc., Northhampton, USA). In all experiments, the heating rate was 1 °C/min and a time resolution of 4 s/data point was used. Aqueous lipid and peptide samples were prepared separately. The pure lipid was suspended in 100 mM NaCl solution, followed by cyclic heating over the phase transition temperature and repeated vortexing. An Avanti Mini-Extruder (Avanti Polar Lipids Inc., Alabaster, USA) was used to prepare liposomes with a monodisperse size distribution by extruding the sample solution 15 times through a polycarbonate filter with a pore size of 100 nm. Lipid suspension and peptide solution were mixed and degassed directly before filling the samples into the calorimetric cell. The lipid concentration in the cell was always 2 mM and the peptide concentration varied from 0.04 mM up to 0.4 mM leading to charge ratios Rc = PG:K of 10:1, 5:1, 2:1, 1:1. The reference cell was filled with a 100 mM NaCl solution. At least three up- and down-scans were performed for each sample to check for reproducibility. All presented curves originate from the second heating scan. Transition enthalpy values H were determined after integration of the calorimetric peaks and were based on the weight concentration of DPPG in the lipid suspension. As the suspensions became turbid after addition of the peptide solution, this can lead to errors, as the suspension was not homogenous any more. We tried to minimize this error by reducing the volume of the prepared samples, so that almost the complete sample in the vial could be transferred into the calorimetric cell. 2.2.3. Attenuated Total Reflection Fourier-transformed Infrared Spectroscopy: For FT-IR measurements the TFA counter ion was exchanged by adding 0.5 mol·l-1 DCl and lyophilizing the solution three times before preparing the peptide solution in D2O. Attenuated Total Reflection Fourier-transformed Infrared (ATR-FT IR) spectra with a spectral resolution of 4 cm-1 were recorded using a Bruker Tensor 27 spectrophotometer equipped with an N2-


Page 9 of 40 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


9 cooled MCT detector and a BioATR II unit (Bruker Optics, Ettlingen, Germany). A total of 256 scans were averaged. As a reference, spectra of NaCl solution (100 mM) at each corresponding temperature were used. The final absorbance spectra were calculated by -log (Isample/Ireference). The desired temperature was set by a computer-controlled circulating water bath (Haake C25P Phoenix II, Karlsruhe, Germany). To compare the results from DSC measurements with the IR experiments a similar sample preparation procedure was chosen. Aqueous lipid and peptide samples (10 mM) were prepared separately using a 100 mM NaCl solution in D2O as solvent. The pure lipid was suspended and held for 5 minutes at 50 °C in a water bath. Samples were directly prepared on the crystal surface by mixing aliquots of lipid suspension (15 µl), peptide solutions (3 µl) and 100 mM NaCl solution in D2O (12 µl) to obtain a charge ratio (Rc) of 1:1 and a lipid concentration of 5 mM. Before recording spectra, one heating and cooling scan was performed to ensure equilibration. Spectra were recorded in 2 °C intervals between 20 °C and 80 °C after temperature equilibration in a temperature interval (ΔT) of ± 0.1 °C for 15 min. The temperature was measured inside the cover plate of the sample holder by a Pt100 resistor (Omega Newport, Deckenpfronn, Germany). All absorbance spectra were shifted to a zero baseline in a spectral region where no vibrational peak occurred. To determine the position of the vibrational bands in a certain wavenumber interval second derivative spectra were calculated and the ‘peak picking’ function included in the Bruker OPUS software was applied. 3. Results and Discussion 3.1. Adsorption of peptides (KXX)4K and (RX)4R to DPPG monolayers at different initial surface pressure 3.1.1. (KGG)4K For the peptide (KGG)4K, the distance between the charged side chains is increased due to the insertion of a second glycine when compared with (KG)4K. In addition, all lysine side chains are pointing now towards opposite sides of the strand in the extended conformation of the peptide (see Figure 1). This ACS Paragon Plus Environment


Page 10 of 40

10 has clearly an influence on the binding properties (see section 3.2). Adsorption of (KGG)4K to a DPPG monolayer in the LE phase as well as the LC phase leads to a decrease of the surface pressure until the equilibrium surface pressure is reached after ca. 4 h (see Figure 2, top left panel). However, the higher molar mass of (KGG)4K slows down the binding kinetics compared to the binding of (KG)4K.28

3.1.2. (KAA)4K The peptide (KAA)4K has an increased hydrophobicity compared to (KGG)4K. For (KAA)4K, a distinct dependence of the adsorption kinetics on the phase state of the lipid was observed (see, Figure 2, top right panel). This was also seen before for the binding of (KAbu)4K.28 It seems that two alanines between the lysines have a similar effect on the adsorption kinetics as one amino butyric acid. For DPPG in the LE phase, first an increase in surface pressure is visible followed by a slower decrease of

. Thus, a reorientation of the peptide after binding to the lipid monolayer takes place to achieve a better electrostatic interaction. Compared to (KGG)4K, the adsorption kinetics of (KAA)4K are different when DPPG is in the LE phase For binding to the LC phase, only a decrease of the surface pressure is visible due to lipid condensation. Compared to (KGG)4K, the kinetics of peptide binding to the LC phase monolayer becomes faster. The total surface pressure decrease observed when adsorption equilibrium is reached, is lower for (KAA)4K compared to (KGG)4K. This is either due to a reduced film condensation by electrostatic adsorption or to partial insertion of (KAA)4K between the lipid headgroups due to its increased hydrophobicity. 3.1.3. (RG)4R (RG)4R is the arginine analogue of (KG)4K, also having an alternating sequence, but the positive charge of the peptide is delocalized over the guanidinium moiety and the whole arginine side chain is


Page 11 of 40

11 more hydrophobic compared to the lysine side chain. The dependence of the adsorption kinetics on the phase state of the lipid is comparable to the behaviour observed before for (KG)4K.28 Peptide binding to DPPG monolayers in the LE phase shows first a decrease in surface pressure followed by a slower increase of π. Lipid condensation over-compensates the effect of peptide incorporation, when the DPPG monolayer is in the LC phase with the result that only a decrease of the surface pressure is visible.

3.1.4. (RA)4R (RA)4R is the arginine analogue of (KA)4K. Independent of the phase state of the lipid monolayer, the adsorption kinetics shows first an increase of the surface pressure, followed by a slower decrease of the surface pressure also in the LC phase. Due to its enhanced hydrophobicity, (RA)4R shows an increased incorporation into the monolayer, the effect of lipid condensation is less pronounced leading to smaller values of Δπ.

-1

4

5 0

1

2

(KAA)4K

LC

30

 / mN m

3

3

4

5 35

LC

30

25

25

20

20

15

15

10

10

5

LE

35

(RG)4R

LC

30

5

LE

0

-1

t/h

(RA)4R

-1

(KGG)4K

t/h

0

35

LC

30

25

25

20

20

15

15

10

10

5

5 LE

LE

0 0

1

2

3

t/h

4 5 0 1 2 ACS Paragon Plus Environment

3

t/h

 / mN m

2

0 4

5

-1

35

1

 / mN m

0

 / 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



Page 12 of 40

12

Figure 2: Change in surface pressure of a DPPG monolayer as a function of time after injection of different peptides (KXX)4K (top) and (RX)4R (bottom) with X = G or A 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 in the subphase was 3.0 µM. The LE-LC phase transition region is highlighted as grey hatched area. Monolayer states above and below this surface pressure are indicated.

3.2. Kinetic Analysis of (KXX)4K Adsorption to a DPPG Monolayer Similar to our previous kinetic analysis of the adsorption process of peptides (KX)4K to a DPPG monolayer28 we also analyzed the binding kinetics by fitting the experimental curves with a biexponential equation:   A1e  t / t  A2 e  t / t   0 1

2

(1)

In the case of (KGG)4K binding, only the slow condensation effect was visible. Therefore, the curve was fitted with a mono-exponential equation. In this case, both processes have probably similar relaxation times so that they cannot be separated. Figure 3, top panel, 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 - 6 mN m-1, where the monolayer is in the LE phase. The simulation parameters are given in Table 1. The amplitude A2 assigned to the lipid condensation effect should be similar for all peptides, because all peptides have the same overall charge. However, this is not the case. For instance, for the adsorption of (KGG)4K, the amplitude A2 is larger compared to (KG)4K. Apparently, the more flexible peptide chain of (KGG)4K enables an orientation of the charged lysine side chains with a better fit to the distances between lipid phosphate groups. As a consequence, a more effective lipid condensation effect ACS Paragon Plus Environment

Page 13 of 40

13 takes place. An opposite effect is seen, when the amino acid X becomes more hydrophobic. The amplitude A2 for the adsorption of (KAA)4K is the smallest of all studied peptides.

6

(KG)4K,

fit

(RG)4R,

fit

(KAA)4K,

fit

(RA)4R,

fit

(KA)4K,

fit

(KGG)4K,

fit

 / mN m

-1

5

4

3

0

1

2

3

4

5

t/h

LE phase

LC phase

-1

0

 / 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


-3

(KGG)4K (KAA)4K

-6

(RG)4R (RA)4R (KG)4K

-9

(KA)4K 0

5

10

15

ini / mN·m

-1 20

25

30

Figure 3: Top: Time dependent development of the surface pressure of a DPPG monolayer after injection of different solutions of the peptides (KXX)4K and (RX)4R into the subphase at a starting surface pressure of 5 mN m-1 (T = 20 °C, LE phase, 100 mM NaCl). The final concentration of the peptide in the subphase was 3.0 µM. Bottom: Difference of surface pressure Δπ after 5 h of adsorption of the peptides (KXX)4K and (RX)4R to a DPPG monolayer with different initial surface pressure πini, the line at πini = 10 mN m-1 is drawn to distinguish



Page 14 of 40

14 between the different phases of the DPPG monolayer. At πini > 10 mN m-1, the film is in the LC-phase. Data for (KG)4K and (KA)4K were taken from ref. 28.

For (KGG)4K, the curve can be fitted by a single exponential describing only the condensation effect. As the amplitude A2 is very high, it is probable that (KGG)4K is not able to insert into the monolayer. For (KAA)4K, both processes are present. The observed amplitude A1 has a relatively low value compared to the one observed for other peptides, whereas the relaxation time t1 is comparable with the most hydrophobic peptide (KL)4K. Thus, it can be concluded that already alanine with its relatively low hydrophobicity leads to an increase in insertion. When the amino acid K is replaced by R the total charge does not change. However, the positive charge in R is delocalized in the guanidinium moiety leading to a decrease in polarity. Comparing the parameters for (RG)4R and (KG)4K, only marginal changes are seen due to the replacement of K by R, as both relaxation times for lipid condensation and peptide incorporation are similar. Although the arginine side chain should be more hydrophobic, the amplitude A2 of lipid condensation is slightly enhanced, whereas the amplitude of peptide incorporation A1 is slightly decreased. An explanation might be the ability of the guanidinium moiety to bridge two lipids,18-19 which is impossible for lysine residues. (RA)4R shows a similar behaviour as found for peptides with the structure (KX)4K tested before.28 The incorporation of a more hydrophobic amino acid than glycine slows down the relaxation time t2 of the lipid condensation and concomitantly increases the amplitude A2. On the other hand, t1 becomes faster and the amplitude of A1 of peptide incorporation is increased in its strength. Although these general statements are true for (RA)4R, the amplitude for lipid condensation is larger than the one for peptide incorporation, leading to the observed small total decrease of the surface pressure. The obtained parameters for (RA)4R are similar to those obtained for (KAbu)4K, confirming the proposed increased


Page 15 of 40 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


15 hydrophobicity of the arginine side chain, as well as the increased condensation ability due to the bidentate guanidinium moiety. Table 1: Fit values for the adsorption kinetics of the peptides (KXX)4K and (RX)4R injected into the subphase underneath a DPPG monolayer at 5 - 6 mN m-1. 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.*Data for (KG)4K and (KA)4K were taken from ref.28.

peptide

peptide incorporation

lipid condensation

A1 / mN m-1 t1 / h

A2 / mN m-1 t2 / h

(KGG)4K --

--

π0 / mN m-1

2.20 ± 0.01

0.74 ± 0.01 2.77 ± 0.10

(KAA)4K -0.74 ± 0.01 0.11 ± 0.01 0.69 ± 0.01

0.98 ± 0.03 5.03 ± 0.01

(RG)4R

-1.11 ± 0.01 2.83 ± 0.02 1.14 ± 0.03

0.45 ± 0.01 5.54 ± 0.01

(RA)4R

-1.68 ± 0.02 0.38 ± 0.02 2.41 ± 0.02

1.05 ± 0.04 4.04 ± 0.01

(KG)4K*

-1.55 ± 0.02 2.48 ± 0.14 1.49 ± 0.03

0.28 ± 0.01 6.17 ± 0.03

(KA)4K*

--

1.25 ± 0.02 3.64 ± 0.01

--

1.66 ± 0.01

3.3. Changes of Surface Pressure as a Function of πini The change in surface pressure  = 0 - ini observed after adsorption of a protein or peptide to a lipid monolayer as a function of the initial surface pressure ini shows in many cases a linear relationship, from which the maximum insertion pressure MIP is determined, i.e. the value of ini where  = 0 mN m-1.29-30 However, in our previous study we had already found a much more complicated behavior due to the competition between electrostatic effects leading to condensation (decrease in ) and insertion leading to expansion of the film (increase in ). At high initial surface pressure condensation effects due to electrostatic binding usually dominated so that negative changes in  were observed.28 We therefore analyzed our data of the surface pressure after reaching equilibrium to see whether the additional spacer amino acid X in peptides (KXX)4K had any effect. Only data of π vs. πini at high



Page 16 of 40

16 surface pressure, where the film is in the LC phase, were fitted as depicted in Figure 3 (bottom panel). The value of πsp, the so-called superposition surface pressure, where condensation and incorporation effects on π cancel each other, was obtained from the linear fit at the point where  was zero. Adsorption of (KGG)4K causes a decrease in surface pressure for all values of πini. The value of πsp of ~ 1 mN m-1 is lower than the πsp value for (KG)4K and is in accordance with the notion that the increase in charge spacing leads to a more flexible peptide chain and a better fit between the distance of the lipid headgroup and peptide side chain charges, as already noted from the analysis of the kinetic curves above. For the binding of (KAA)4K to the DPPG monolayer, a different behaviour for both phase states is seen. Up to πini = 5 mN m-1 incorporation of the peptide into the lipid monolayer slightly dominates and a slightly positive  is observed decreasing to zero with increase in . The superposition surface pressure πsp = 4.8 ± 1.8 mN m-1 found for (KAA)4K is comparable to the value found for (KA)4K, but higher than the value found for (KGG)4K, indicating the increasing dominance of the insertion effect. The dependence of  from πini in the LC phase shows a linear relationship with a slope of m = -0.23. This value is less negative than all determined values before, indicating weak interaction. Due to an increased charge distance, the impact of lipid condensation is lowered compared to (KX)4K. The plot Δπ vs. πini for (RG)4R is very similar to the plot for (KG)4K. For adsorption of (RG)4R at low initial surface pressure to DPPG monolayer in the LE phase 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 so that finally  = 0 mN m-1. For the LC phase, the surface pressure decreases linearly with increasing πini and thus the lipid condensation becomes more prominent (see Figure 3, bottom panel). The break in the plot corresponds to surface pressure of the LE/LC transition region of


Page 17 of 40 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


17 the DPPG monolayer at 10 mN m-1 and the fitted value for πsp. The determined m value of -0.38 suggests strong attractive interaction between the peptide comparable to (KG)4K. Adsorption of (RA)4R causes only a small decrease in surface pressure at all initial surface pressure values. There is no difference in behaviour in the LE and LC phase. As explained above, both processes nearly compensate each other with the result that the extrapolated superposition of incorporation and condensation is extrapolated πsp ≈ 0 mN m-1. However, the low value of πsp coupled with the large error, allows no definite conclusions. Table 2: Values of n, m, πsp and R2 obtained from the fitting of data for binding of the peptides (KXX)4K and (RX)4R to a DPPG monolayer in the LC phase presented in Figure 3. 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

(KGG)4K

0.4 ± 1.0

-0.35 ± 0.04

1.1 ± 2.9

0.913

(KAA)4K

1.1 ± 0.4

-0.23 ± 0.03

4.8 ± 1.8

0.910

(RG)4R

3.9 ± 0.5

-0.38 ± 0.02

10.3 ± 1.4

0.978

(RA)4R

0.0 ± 0.8

-0.08 ± 0.04

0.0 ± 10.0

0.440

peptide



Page 18 of 40

18 3.4. Interaction of peptides with DPPG bilayers The monolayer experiments described above were used to test the ability of the peptides to insert into the monolayer at low surface pressure, i.e. into monolayers in the LE-phase. For all peptides a decrease in surface pressure was observed when the monolayer was in the LC-phase, indicating condensation due to charge compensation. In lipid bilayers, the area per lipid cannot be varied but the lipid vesicles can undergo a phase transition at constant pressure into the liquid-crystalline phase at higher temperature connected with an increase in molecular area and fluidization of the chains. A further difference between the two systems is their dimensionality. Lipid monolayers are pseudo-2-D systems neglecting the thickness of the monolayer. In lipid bilayer systems, however, the system can undergo a transition into the third dimension into well-ordered stacked bilayers with intercalated peptide layers after binding of peptides. This we have shown before for the binding of different peptides of the structure (KX)4K to DPPG and other negatively charged lipids

21, 31.

The peptides with alternating

sequences (KX)4K and a periodicity of 2 are predestined to form β-sheets, because an uncharged and/or hydrophobic side opposes a highly positively charged side. Data from ATR-IR spectroscopy showed that the stability of the proposed β-sheets is determined by the hydrophobicity of the spacer X.21 The peptides with a sequence of (KXX)4K studied here exhibit a periodicity of 3 and therefore a 310helix is the favored secondary structure.32-33 With further increased periodicity, the probability to form helices is enhanced, a periodicity of 3.6 would lead to α-helices, but glycine as the smallest amino acid is known as helix breaker due to its flexibility and the missing side chain 34. The minimal length for a stable helix is predicted to be 14 amino acids.25 The peptides used here with maximal 13 amino acids are below this value. Therefore, it is expected that these peptides do not form stable secondary structures in solution. However, it is still possible that these structures or other secondary structures can


Page 19 of 40 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


19 be formed after binding to a bilayer surface. This depends on the one hand on the interaction partner, and on the other hand also on the peptide length.35 3.4.1. Thermotropic behavior of DPPG-peptide mixtures studied by DSC 3.4.1.1. DPPG-(KXX)4K Figure 4 shows the thermotropic behavior of DPPG-(KXX)4K mixtures at a charge ratio of Rc=1, i.e. at charge saturation. Both peptides, (KGG)4K and (KAA)4K, slightly shift the phase transition temperature Tm to higher values. The Tm shift is larger for (KAA)4K than for (KGG)4K. The whole transition spans the temperature range from 40 °C up to 50 °C for both peptides (KXX)4K and for all other peptides shown in Table 3. For comparison, the DSC curves for mixtures with (KG)4K and (KA)4K are also included. Whereas the mixture with (KA)4K shows a very large upshift in Tm due to the formation of -sheets, all three other DSC curves are very similar. The appearance of different pronounced peaks for these DPPG-peptide mixtures indicates possible domain formation into immiscible lipid-rich and peptide-rich aggregates. The transition enthalpies for 1:1 mixtures of DPPG(KXX)4K are increased compared to pure DPPG vesicles. Table 3 shows a detailed comparison of the obtained parameters. The transition temperature and enthalpy increases in the order pure DPPG, mixtures with (K)536 and mixtures with (KG)4K and decreases again for (KGG)4K. Thus, (KG)4K stabilizes the gel phase of the DPPG more effectively, but has the broadest transition range. Data for DPPG-(K)14 and mixtures DPPG-(K)5 were added to Table 3 for comparison. (K)5 has the same number of charges, but contains no intercalated uncharged amino acids. (K)14 is one amino acid longer than the peptides (KXX)4K with 13 amino acids but carries 14 positive charges. Apparently, the chain length of the added peptide and the number of charges seems to be less important as results from (K)5 and (K)14 are similar concerning the phase transition temperature and enthalpy.12, 36 Based on these results, one can conclude that the ACS Paragon Plus Environment


20 different phase transition temperatures for DPPG mixtures with (K)5, (KG)4K and (KGG)4K are more related to the different charge separation and not to the total length of the peptide. 10

DPPG

8

+(KA)4K

6

+(KG)4K

4

+(KAA)4K

10

+(KGG)4K

0

pure DPPG

DPPG-(RX)4R mixtures

8

-1

Cp / kcal·mol ·K

2

-1

-1 -1

Cp / kcal·mol ·K

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 20 of 40

6 4

+(RA)4R

2

+(RG)4R pure DPPG

0

30

40

50 T / °C

60

70

20

30

40

50

60

70

T / °C

Figure 4: Thermograms of pure DPPG vesicles and different DPPG-peptide mixtures with a charge ratio Rc =1:1. The 2nd heating scan is shown, because equilibrium was only reached after passing the phase transition. Samples were prepared in aqueous 100 mM NaCl with a fixed lipid concentration of 2 mM. Data for (KG)4K and (KA)4K were taken from ref. 28.

For mixtures with (KA)4K and (KAA)4K, a completely different thermotropic behavior was seen. For DPPG-(KA)4K mixtures a much higher Tm with a reduced transition enthalpy was observed. Data obtained from FT-IR measurements showed a completely different secondary structure for (KA)4K when bound to the DPPG bilayers, namely β-sheets intercalated between gel phase DPPG bilayers, whereas all other peptides having mostly unordered structures. As the mixture with (KAA)4K shows only a slight upshift in Tm it can already be concluded that the formation of β-sheets is unlikely for this peptide. The general stabilization effect of the gel phase after binding is visible for added peptides and can be attributed to the charge neutralization upon peptide binding, because all peptides exhibit the same charge. Peptides with one uncharged spacer show a greater upshift than pentalysine and peptides with two uncharged spacers.


Page 21 of 40 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


21 3.4.1.2. DPPG-(RX)4R The exchange of the amino acid K against R should have a distinct effect on the binding behavior as the positive charge of the amino acid side chain is delocalized in the guanidinium moiety making the side chain more hydrophobic. The DSC curves of DPPG mixtures with (RG)4R and (RA)4R at a charge ratio of 1:1 are shown in Figure 4, right panel. As known from previous studies, arginine homopolymers are capable to cross bilayers even at low temperatures due to their increased hydrophobicity.17, 37 When additional spacer amino acids are introduced between the arginine amino acids, the length of the spacer plays an inportant role for the ability ob the peptide to penetrate a lipid bilayer. An echange of glycine by the much longer artificial amino acid 6-amino-caproic acid shows an 2.5 fold increase of the cellular uptake measured by fluorescence spectroscopy.38 The question arises, whether the enhanced permeation of R containing peptides through lipid bilayers is also connected with a change in thermotropic behavior. As can be seen from Figure 4, the effect of added peptides on the phase transition temperature is marginable. At a charge ratio of Rc = 1, the peak for the main transition remains at 40.3 °C, the same temperature as for pure DPPG. The transition is only slightly broader and exhibits a small shoulder at the high temperature side shifting the Tmid by 0.5 °C to higher temperature (see Table 3). However, an unchanged phase transition temperature does not necessarily mean that the peptide is not bound. It can be a result of two competing processes which overlay and compensate each other.. As the lipid vesicle sample shows aggregation induced by peptide binding and we have also observed binding to DPPG monolayers (see above), the compensation effect is the more likely explanation.



Page 22 of 40

22 Table 3: Overview of the thermodynamic data obtained for DPPG-peptide mixtures with a charge ratio of 1:1. The peptides had different sequences and thus different distances between the charged lysines. Tm is the main phase transition and marked by the peak with the highest heat capacity and T1/2 the full width of this peak at half maximum. Especially for the asymmetric peaks with several shoulders Tmid obtained from the midpoint of the integral of the curves, is more suitable for a comparison. ΔTbase is the width of the peak at its base and ΔH is the transition enthalpy.

Tm / °C

Tmid / °C

T1/2 / °C

ΔTbase / K

ΔH / kcal mol-1

pure DPPG

40.3

40.2

1.5

3.5

10.4

+ (K)5 36

42.2

42.2

+ (KG)4K

41.3

44.2

5.6

10.7

13.2

+ (KGG)4K

40.7

42.9

4.8

9.0

10.8

+ (KA)4K

60.0

59.4

4.3

7.7

6.3

+ (KAA)4K

42.1

44.0

2.7

8.7

11.4

+ (K)14 12

42

42

+ (RG)4R

40.4

40.7

1.8

3.7

8.4

+ (RA)4R

40.3

40.6

1.0

2.9

9.6

10.7

10.9

3.4.2. Thermotropic behavior of DPPG-peptide mixtures studied by ATR-FT-IR spectroscopy 3.4.2.1. CH2 stretching bands of DPPG 3.4.2.1.1. (KXX)4K mixtures The frequency of the antisymmetric (as(CH2)) and symmetric (s(CH2)) methylene stretching vibrations are an indicator for the order of the acyl chains in the hydrophobic membrane region. When the acyl chains are highly ordered in all-trans conformations in the gel phase, the band maxima are positioned at a low wavenumber. When the lipid passes into the liquid-crystalline phase, the absorption maxima are shifted to higher wavenumber caused by the increasing fraction of gauche conformers in the acyl chains.21, 39-40


Page 23 of 40 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


23 Figure5 shows the wavenumber of the symmetric (s(CH2)) methylene stretching vibration as a function of temperature. Mixtures of DPPG with (KGG)4K show lower frequencies for the νs(CH2) band in the gel phase as well as in fluid phase compared to pure DPPG (see Figure 5, left panel). Such a decrease in the νs(CH2) band wavenumber is caused by a better ordering of the acyl chains.40 The phase transition of the DPPG-(KGG)4K mixture occurs at 43 °C and is in good agreement with the Tm value obtained from the DSC experiments. In the IR data, a hint of a second increase in wavenumber is visible at 59 °C. Such a two-step behavior was also observed by Schwieger et al.12 for DPPGpolylysine complexes and was discussed as a proof for domain formation of DPPG with bound polylysine. In mixtures of DPPG with (KAA)4K, an increase in chain order in the gel phase is not seen, but in the fluid phase the decrease in wavenumber is significant so that the overall increase in wavenumber at the phase transition is less pronounced. Introducing a spacer amino acid between the lysines changes the ability of the peptide to increase the chain order of the lipid upon binding. Comparing the peptides (K)5, (KG)4K and (KGG)4K as well as (K)5, (KA)4K and (KAA)4K for their influence on the wavenumber of the νs(CH2), the effect of the better acyl chain ordering is most pronounced for (K)5 (Δ%= - 1.08 cm-1),36 lower for the peptides with glycine spacers (KG)4K and (KGG)4K and for (KA)4K (Δ% = -0.5 cm-1), and absent for (KAA)4K. Reasons might be the increased binding constants for peptides with no or only one spacer to negatively charged lipids27 or a better charge compensation due to geometrical reasons.



24 2854.0

2852.5 2852.0

DPPG + (KGG)4K

0.7

DPPG DPPG + (KGG)4K DPPG + (KAA)4K DPPG + (KG)4K

-1

wavenumber / cm

-1

2853.0

CH2,s upscan

A(1725 cm ) -1 -1 A(1725 cm )+A(1740 cm )

2853.5

DPPG + (KA)4K

2851.5 2851.0 2850.5

0.6

+ (KAA)4K

0.5

+ (KA)4K

+ (KG)4K

0.4 0.3 0.2

2850.0 2849.5

0.1 20

40

T / °C

60

80

10

20

30

40 T / °C

50

60

70

2854.0

2852.5 2852.0

DPPG + (KG)4K

0.7

DPPG DPPG + (KG)4K DPPG + (KA)4K DPPG + (RG)4R

-1

-1

2853.0

CH2,s upscan

A(1725 cm ) -1 -1 A(1725 cm )+A(1740 cm )

2853.5

wavenumber / cm

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 24 of 40

DPPG + (RA)4R

2851.5 2851.0 2850.5

0.6

+ (KA)4K

0.5

+ (RA)4R

+ (RG)4R

0.4 0.3 0.2

2850.0 2849.5 20

40

T / °C

60

80

0.1 10

20

30

40 T / °C

50

60

70

Figure 5: Left: wavenumber of the maximum of the symmetric CH2 stretching band of pure DPPG and DPPG mixtures with peptides of increasing charge distance and hydrophobicity: (KG)4K, (KA)4K, (KGG)4K, (KAA)4K, respectively. Lines to guide the eye were constructed by an adjacent averaging of 5 points. Right: ratio of the area of the ν(CO) vibrational band at low wavenumber caused by hydrogen bonding to the whole area of the carbonyl vibrational peak. All samples had a charge ratio Rc = 1:1 and were prepared in D2O containing 100 mM NaCl at pD = 7.2. Data for (KG)4K and (KA)4K were taken from ref. 28.

3.4.2.1.2. (RX)4R mixtures For mixtures of DPPG with (RG)4R and (RA)4R, a clear shift of the phase transition to 48 °C is visible with only marginal differences between the two peptides (see Figure 5, left panel). The transition ranges from 40 - 54 °C. At low temperature below the phase transition, the acyl chains become slightly more ordered. This effect vanishes at high temperature when DPPG is in the liquid-crystalline phase. The shift of Tm is more pronounced in the samples used for IR than observed for the samples used in


Page 25 of 40 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


25 DSC measurements. The reason for this observation is probably the five-fold higher concentration of lipid and peptide used for the samples in the ATR measurements, as an increase in total concentration increases the degree of binding at a fixed charge ratio. 3.4.2.2. C=O bands of DPPG 3.4.2.2.1. (KXX)4K mixtures Characteristic vibrational bands of the lipid headgroup region are the carbonyl stretching vibrational band ν(C=O) and the phosphodiester bands. The vibrational frequency of the lipid ester C=O groups as well as the phosphodiester bands will be shifted to lower wavenumber due to hydrogen bonding of water molecules to the oxygen atoms.41-42 Peptide addition influences the hydration level of the lipid bilayer in both phases and an upshift of the transition temperature is visible (see Figure 5, right panel). We have observed before that the binding of lysine containing peptides reduces the hydration of the lipid headgroup.12,

21, 36

In the gel phase at 20 °C, the intensity of the C=O band at low frequency

representing the hydrogen bonded C=O groups is reduced after peptide binding in the order pure DPPG (37%) >DPPG/(KG)4K (25%) > DPPG/(KGG)4K (20%). Above the phase transition of the system a higher proportion of the hydrogen bonded fraction is visible but the headgroups are still less hydrated than the headgroups in pure DPPG. For DPPG mixed with (KAA)4K, hydration is also slightly below that of the pure lipid in the gel phase. An exception is the mixture with (KA)4K studied before. In this case a much stronger dehydration was seen combined with a large shift of Tm. This was explained by the formation of intercalated membrane bound -sheets between the bilayers. 3.4.2.2.2. (RX)4R mixtures



Page 26 of 40

26 The arginine peptides (RG)4R and (RA)4R bound to DPPG have similar effects compared to the corresponding lysine peptides. Both cause to the same extent a reduction of the lipid hydration in the gel phase (see Figure 5, right panel). At higher temperature the effect of (RG)4R vanishes and the system shows similar behavior as pure lipid vesicles. However, the slightly more hydrophobic (RA)4R also reduces the hydrogen bonded fraction in the fluid state. 3.4.2.3. Amide I bands of peptides 3.3.2.3.1. DPPG-(KXX)4K mixtures The position of the amide-I band gives information about the secondary structure of the bound peptides. Stronger hydrogen bonds in the secondary structure lead to lower amide I band frequencies 40, 43.

Unordered structures in peptides lead to an amide I band at 1652–1660 cm−1, peptides with β-turns

show a band maximum at 1670 cm−1, α-helical conformations with longer and weaker hydrogen bonds lead to a band maximum at 1648–1660 cm−1, and antiparallel β-sheets with stronger hydrogen bonds show an intense band at 1625–1640 cm−1 and a less intense one at 1680 cm−1. The latter arises from transition dipole coupling. Band positions may be shifted to lower frequencies when H/D exchange has occurred, for instance, in samples with D2O as solvent Spectra in the amide I region of samples of (KGG)4K and (KAA)4K bound to DPPG bilayers are shown in Figure 6. For DPPG/(KGG)4K clearly two bands located at 1670 cm-1 and 1646 cm-1 are visible. For the DPPG/(KAA)4K, mixture the band at 1646 cm-1 has the highest intensity with a shoulder on the higher frequency side. The spectra of the bound peptides are identical to the pure peptides in solution (not shown). The secondary structure of the peptides (KXX)4K in solution as well as in the bound form is a mixture of β-turns and unordered conformations. The bands change only slightly upon heating over the phase transition of the lipid. Thus no change in secondary structure is observable. For the peptides (KXX)4K, a 310-helix is predicted, which would be visible by a band at 1660 cm-1.32 As no contribution ACS Paragon Plus Environment

Page 27 of 40

27 to the amide-I band in this spectral region occurs the existence of a 310-helix is excluded for both peptides.

10 °C 20 °C 30 °C 40 °C 50 °C 60 °C 70 °C

0.010

0.020

10 °C 20 °C 30 °C 40 °C 50 °C 60 °C 70 °C

DPPG + (KAA)4K

0.015 -log I/I0

DPPG + (KGG)4K

0.015

-log I/I0

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


0.010

0.005

0.005 0.000 1700

0.000 1680

1660

1640 1620 1600 -1 wavenumber / cm

1580

1700

1680

1660

1640 1620 1600 -1 wavenumber / cm

1580

Figure 6: Temperature dependent behavior of the amide-I band for DPPG mixtures with peptides (KXX)4K (heating scan). Left: (KGG)4K-DPPG mixtures. Right: (KAA)4K-DPPG mixtures. All samples had a charge ratio Rc = 1:1 and were prepared in D2O containing 100 mM NaCl at pD = 7.2.

A fitting of the spectral bands in the amide-I region with two peaks, centered at 1670 cm-1 for β-turns and 1646 cm-1 for unordered conformations, with a Gaussian profile were used to determine the relative areas of the peaks. A higher content of β-turns in the spectra of (KGG)4K (ca. 45%) as compared to (KG)4K with 25 % turns is seen. This is expected as in general the probability for β-turns44 is enhanced in glycine containing peptides and increases further, if 3 out of 4 residues are glycines. The most likely turn arrangement is the one with the charged amino acid lysine in position 2 and the hydrogen bond between position 1 and 4 (see Table 4).



Page 28 of 40

28 Table 4: Turn potentials of single amino acids for different positions in the β-turn structures.44 The overall turn potential for a β-turn composed of different peptide sequences, consisting of K, G, and A.

amino acid position

\ i

i+1

i+2

i+3

K

0.80

1.22

0.94

1.10

G

1.09

1.04

2.14

1.64

A

0.81

0.96

0.66

0.89

sequence

KKKK

KAAK

AAKA

AKAA

turn potential

1.01

0.56

0.65

0.58

sequence

KGKG

GKGK

KAKA

AKAK

turn potential

1.28

3.13

0.64

0.72

sequence

KGGK

GGKG

GKGG

turn potential

1.96

1.75

4.67

3.4.2.3.2. DPPG-(RX)4R mixtures Spectra in the amide I region of (RG)4R and (RA)4R bound to DPPG bilayers are shown in Figure 7. The secondary structure of the tested peptides (RX)4R is again a mixture of β-turns and unordered structures. Peptides with arginine show an additional band in the amide I region due to the contributions of the antisymmetric and symmetric guanidyl stretching vibrations located at 1608 cm-1 and 1582 cm-1, respectively.45-46 A downshift of the wavenumber of the guanidyl stretching vibrational bands observed for peptides bound to DPPG compared to the wavenumber of free peptide in aqueous solution is due to hydrogen bonding with the phosphate groups of the lipid, indicating an intercalation of the guanidinium moiety between the headgroups.46


Page 29 of 40

29

0.008

0.04

DPPG + (RG)4R 10 °C 20 °C 30 °C 40 °C 50 °C 60 °C 70 °C

0.03

-log I/I0

0.012

log I/I0

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


0.004

0.000

1700

0.02

DPPG + (RA)4R 10 °C 20 °C 30 °C 40 °C 50 °C 60 °C 70 °C

0.01

0.00 1680

1660

1640

1620

1600

wavenumber / cm

1580

1560

1700

1680

-1

1660 1640 1620 1600 -1 wavenumber / cm

1580

1560

Figure 7: Temperature dependent behavior of the amide-I band for DPPG mixtures with peptides (RX)4K (heating scan). Left: (RG)4K-DPPG mixtures. Right: (RA)4K-DPPG mixtures. All samples had a charge ratio Rc = 1:1 and were prepared in D2O containing 100 mM NaCl at pD = 7.2.

Comparing the binding behavior of the peptides (RG)4R and (RA)4R, with that of (KG)4K and (KA)4K reveals following differences. The binding of (KG)4K and (RG)4R to DPPG bilayers leads to similar changes in the thermotropic behavior. Tm is slightly shifted to higher temperature. However, for DPPG/(RA)4R the transition temperature remains almost unchanged at 40 °C whereas for DPPG/(KA)4K mixtures the transition temperature shifts to 60 °C due to the formation of well-ordered -sheet structures between opposing bilayers leading to an additional level of self-organization. This is apparently not possible for DPPG/(RA)4R mixtures. An explanation for this might be the intercalation of the guanidinium moieties of the arginine side chains into the headgroup region preventing the formation of extended -sheet structures bridging opposing bilayers. Similar differences were also reported for homopeptides of argenine and lysine12, 15 when the binding to DMPG or DPPG was studied. Addition of shorter polypeptides (R)9 up to (R)184 show a slight downshift of the lipid main transition, whereas the longer peptides increase Tm.15, 47 The arginine side chain is more hydrophobic compared to lysine side chain.15,

48

Additionally, the positive charge of

arginine is delocalized in the guanidyl residue over at least four atoms in a π-system. It is known that



Page 30 of 40

30 the guanidinium moity is capable to build up a bidentate hydrogen bond with the lipid phosphate group.18-19 Thus an overlay of electrostatic effects due to charge compensation and hydrophobic effects caused by insertion of guanidinium side chains into the lipid layers is also seen in these systems. For all 1:1 complexes, one negative headgroup charge should be compensated by the positive charge of the lysine side chain. In the gel phase, the area of a PG headgroup is estimated to be 48 Å2. Thus, a distance between two neighboring headgroups of 6.9 Å must be bridged by the peptide.49-50 For geometrical reasons, the different lysine chains at different positions in the peptides with different sequences have to adapt to this distance. In peptides of the structure (KX)4K, the charge distance is 7.2 Å when the peptide is in its extended form (see Figure 8). These peptides can form intermolecular sheets in between stacked bilayers, because the distance between the side chain charges is very similar to the distance of the phosphate groups of DPPG. For peptides with the structure (KXX)4K the distance between side chain charges increases to 16.1 Å or even 21.8 Å (see Figure 8). This distance is much too large to fit to the headgroups distance of DPPG. However, the distance can be reduced to 8-11 Å when the peptide forms a β-turn. This requires a glycine in position i+2 which is the case of (KGG)4K. For (KAA)4K this β-turn is more unlikely because of the more voluminous side chain of alanine. The spectra shown in Figure 6 clearly show this different behavior, as a more intense β-turn band is visible in the case of mixtures with (KGG)4K. Beside the distance between the charges, the orientation of the charged peptide side chains relative to the lipid bilayer also plays a role. For pentalysine, the side chains are pointing in opposite directions when the peptide is in its extended form, resulting in a flat orientation on top of the bilayer without the possibility to build up intermolecular hydrogen bonds for stabilization between adjacent peptide strands.51-52


Page 31 of 40 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


31

Figure 8: Schematic representation of the charge distances in the side chains of K5, (KG)4K and (KGG)4K in their extended conformation.



Page 32 of 40

32

Figure 9: Possible arrangements of the peptides on top of the lipid layer: ((KG)4K and (KGG)4K in extended conformation viewed from different directions. Structures are presented along the three dimensions of space. A: top view along z-axis (peptides in green), B: side view along y-axis, C:side view along molecule longitudinal axis along x-axis. D: top view of (KGG)4K bound in extended conformation and as -turn structure (peptides in green). The secondary structures presented here are a selection of the possible conformations determined by the peptide sequence. Peptide molecules are turned in such direction that the maximal numbers of positive charged lysine side chains are directed to the negatively charged lipid headgroups.


Page 33 of 40 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


33 For the peptides (KX)4K with one uncharged spacer, the peptide has the ability to be orientated perpendicular to the bilayer so that most of the amide groups are now pointing to the bilayer surface (see Figure 9). This leads to an effective stabilization for a β-sheet arrangement via intermolecular hydrogen bonds in y-direction. For peptides with the sequence (KXX)4K, an extended conformation lying flat on the bilayer (see Figure 9), a β-turn structure, or 310- or α-helical structures are possible. The turn and α-helical structure induce no preferred direction of the amide bonds, but stabilization through intramolecular hydrogen bonds within the structure is possible. For the 310-helix of (KGG)4K, all lysine side chains point in the same direction and can thus be oriented towards the bilayer surface (not shown). However, the lateral distance between two neighboring lysines is now 5.8 Å and thus smaller than the lipid headgroup distance. As the FT-IR spectra show no characteristic bands for either β-sheets, nor helices, these peptide conformations can be excluded. The only remaining conformation for (KGG)4K is the β-turn structure (see Figure 9D). This structure has the drawback that only part of the peptide is stabilized and the charge distances in the peptide are not optimal for interaction with the lipid headgroups. A major proportion of the peptide remains in a random conformation as is also observed for (KAA)4K in solution as well as bound to the lipid headgroup region. Other groups have used model peptides composed of the two amino acids lysine and leucine with other periodicities to determine the interplay of hydrophobic and charge interaction for the peptide structure itself, as well as for the interaction with other components. Peptides composed of lysine and leucine with a ratio KiL2i show for 9 residues a β-sheet structure. For the 15 residue analogue, an α-helical arrangement is detected although a 310-helix is predicted for this peptide sequence.35, 53 (LLK)nL with n = 2 or 4 show a helical structure and bind tightly to hydrophobic surface of polystyrene resin.54 CD and IR spectroscopy measurements show that the 21 residue long peptide (KL4)4K has an α-helical structure and orients its helical axis parallel to the acyl chains in a mixed DPPC/PG bilayer.55 In contrast, it shows a β-sheet structure in the denser DPPC/PA bilayer.56 These studies show that also the ACS Paragon Plus Environment


Page 34 of 40

34 length of the peptide has an important influence on the formation of secondary structures. Our peptides described have a length of 9 or 13 residues. They are therefore too short to form defined stable secondary structures in solution. 4. Summary For DPPG mixtures with the peptides (K)5, (KG)4K, (KGG)4K and (KAA)4K at a charge ratio of 1:1, the enthalpy and temperature of the lipid phase transition are increased after peptide binding. This is the case when electrostatic effects are dominating, i.e. the screening of charges of the lipid headgroup is the main effect after binding, resulting in a stabilization of the gel phase. For a complete shielding of all headgroup charges an increase in Tm of 5.5 K is to be expected57. This was also found for binding of oligolysines or polylysines to negatively charged membranes.12, 58 Shortening of the lysine side chain leads to a larger increase in Tm as theoretically predicted.36 The upshift of Tm of up to 4 K for the peptides studied here is slightly below the theoretically predicted increase because of additional hydrophobic interactions of the lysine side chain. The peptide backbone of (KG)4K and (KGG)4K seems to be flexible enough to adopt an unordered structure with a high content of β-turns. It could be concluded that these peptides bind mainly due to the electrostatic interaction to the lipid bilayer surface without any incorporation of the spacer amino acid glycine into the hydrocarbon region, because glycines are not hydrophobic enough. The group of arginine containing peptides, (RG)4R and (RA)4R, behave differently than those mentioned before. The transition enthalpy is slightly lowered at a constant transition temperature. If the peptides binds solely by non-polar interactions without any electrostatic contributions, for instance, gramicidin A to DPPG, a reduction of Tm and a decrease of the transition enthalpy is observed due to the perturbation of the packing of the lipids in the gel phase.58 However, peptides containing arginine show a more complex behavior. The superposition of the counteracting effects, namely the binding to ACS Paragon Plus Environment

Page 35 of 40 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


35 the bilayer upon electrostatic interaction, the formation of bidentate hydrogen bonding, and the hydrophobic interaction of the guanidinium moiety with the lipid headgroup, causes an almost zero net effect on the phase transition properties. Comparing the results reveals a distinct influence of the peptide sequence on the termotropic behavior of DPPG bilayers after peptide binding. For the homooligopeptide K5, the electrostatic effects are slightly counteracted by the hydrophobic binding of the lysine side chain.36 The introduction of an uncharged amino acid in (KG)4K reduces the hydrophobic contribution of the lysine side chain and increases the electrostatic effect of charge screening. Increasing the hydrophobicity of the side chain of the uncharged amino acid in (KA)4K enables the formation of an additional level of self-organization resulting in an extensive stabilization of the lipid gel phase by the formation of -sheets intercalated between the bilayers.21 Introducing a second uncharged amino acid spacer X increases the charge distance, alters the periodicity of the peptide, and opens the possibilities for other conformations of the peptide. The binding of (KGG) to DPPG leads to similar effects compared to the peptide (KG)4K. However, in (KAA)4K the alternating side chain orientation prevents the formation of extensive of sheets and the peptide remains unordered. The effect of increased hydrophobicity of alanine vs. glycine leads to only marginal changes. Acknowledgements This work was supported by the Deutsche Forschungsgemeinschaft (GRK 1026 Conformational Transitions in Macromolecular Interactions, Project A1).



Page 36 of 40

36 References 1. Brogden, K. A., Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nature Reviews 2005, 3, 238-250. 2. Zasloff, M., Antimicrobial peptides of multicellular organisms. Nature 2002, 415, 389-395. 3. Heitz, F.; Morris, M. C.; Divita, G., Twenty years of cell-penetrating peptides: from molecular mechanisms to therapeutics. Br. J. Pharmacol. 2009, 157, 195-206. 4. White, S. H.; Wimley, W. C.; Selsted, M. E., Structure, function, and membrane integration of defensins. Curr. Opin. Struct. Biol. 1995, 5, 521-527. 5. Dürr, U. H. N.; Sudheendra, U. S.; Ramamoorthy, A., LL-37, the only human member of the cathelicidin family of antimicrobial peptides. Biochim. Biophys. Acta 2006, 1758, 1408-1425. 6. Derossi, D.; Joliot, A. H.; Chassaing, G.; Prochiantz, A., The third helix of the Antennapedia homeodomain translocates through biological membranes. J. Biol. Chem. 1994, 269, 10444-50. 7. Fawell, S.; Seery, J.; Daikh, Y.; Moore, C.; Chen, L. L.; Pepinsky, B.; Barsoum, J., Tatmediated delivery of heterologous proteins into cells. Proc. Natl. Acad. Sci. USA 1994, 91, 664-668. 8. 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. 9. Johansson, J.; Gudmundsson, G. H.; Rottenberg, M. n. E.; Berndt, K. D.; Agerberth, B., Conformation-dependent Antibacterial Activity of the Naturally Occurring Human Peptide LL-37. J. Biol. Chem. 1998, 273, 3718-3724. 10. Takechi, Y.; Tanaka, H.; Kitayama, H.; Yoshii, H.; Tanaka, M.; Saito, H., Comparative study on the interaction of cell-penetrating polycationic polymers with lipid membranes. Chem. Phys. Lipids 2012, 165, 51-58. 11. 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. 12. 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. 13. Carrier, D.; Pezolet, M., Investigation of polylysine-dipalmitoylphosphatidylglycerol interactions in model membranes. Biochemistry 1986, 25, 4167-4174. 14. Takahashi, H.; Matuoka, S.; Kato, S.; Ohki, K.; Hatta, I., Effects of poly(l-lysine) on the structural and thermotropic properties of dipalmitoylphosphatidylglycerol bilayers. Biochim. Biophys. Acta 1992, 1110, 29-36. 15. Schwieger, C.; Blume, A., Interaction of Poly(L-arginine) with Negatively Charged DPPG Membranes: Calorimetric and Monolayer Studies. Biomacromolecules 2009, 10, 2152-2161. 16. Wender, P. A.; Mitchell, D. J.; Pattabiraman, K.; Pelkey, E. T.; Steinman, L.; Rothbard, J. B., The design, synthesis, and evaluation of molecules that enable or enhance cellular uptake: Peptoid molecular transporters. Proc. Natl. Acad. Sci. USA 2000, 97, 13003-13008. 17. Mitchell, D. J.; Steinman, L.; Kim, D. T.; Fathman, C. G.; Rothbard, J. B., Polyarginine enters cells more efficiently than other polycationic homopolymers. J. Pept. Res. 2000, 56, 318-325. 18. Rothbard, J. B.; Jessop, T. C.; Lewis, R. S.; Murray, B. A.; Wender, P. A., Role of Membrane Potential and Hydrogen Bonding in the Mechanism of Translocation of Guanidinium-Rich Peptides into Cells. J. Am. Chem. Soc. 2004, 126, 9506-9507. 19. Sakai, N.; Matile, S., Anion-Mediated Transfer of Polyarginine across Liquid and Bilayer Membranes. J. Am. Chem. Soc. 2003, 125, 14348-14356.


Page 37 of 40 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


37 20. Schmidt, W.; Buschle, M.; Zauner, W.; Kirlappos, H.; Mechtler, K.; Trska, B.; Birnstiel, M. L., Cell-free tumor antigen peptide-based cancer vaccines. Proc. Natl. Acad. Sci. USA 1997, 94, 32623267. 21. Hadicke, A.; Blume, A., Binding of cationic peptides (KX)(4)K to DPPG bilayers. Increasing the hydrophobicity of the uncharged amino acid X drives formation of membrane bound beta-sheets: A DSC and FT-IR study. Biochim. Biophys. Acta 2016, 1858, 1196-1206. 22. Hadicke, A.; Blume, A., Binding of the Cationic Peptide (KL)4K to Lipid Monolayers at the Air-Water Interface: Effect of Lipid Headgroup Charge, Acyl Chain Length, and Acyl Chain Saturation. J. Phys. Chem. B 2016, 120, 3880-7. 23. de Kruijff, B.; Rietveld, A.; Telders, N.; Vaandrager, B., Molecular aspects of the bilayer stabilization induced by poly(l-lysines) of varying size in cardiolipin liposomes. Biochim. Biophys. Acta 1985, 820, 295-304. 24. 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. 25. Narita, M.; Tomotake, Y.; Isokawa, S.; Matsuzawa, T.; Miyauchi, T., Syntheses and properties of resin-bound oligopeptides. 2. Infrared spectroscopic conformational analysis of cross-linked polystyrene resin bound oligoleucines in the swollen state. Macromolecules 1984, 17, 1903-1906. 26. Li, L.; Vorobyov, I.; Allen, T. W., The Different Interactions of Lysine and Arginine Side Chains with Lipid Membranes. J. Phys. Chem. B 2013, 117, 11906-11920. 27. 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. 28. Hädicke, A.; Blume, A., Binding of Short Cationic Peptides (KX)4K to Negatively Charged DPPG Monolayers: Competition between Electrostatic and Hydrophobic Interactions. Langmuir 2015, 31, 12203-12214. 29. 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. 30. 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. 31. Hädicke, A.; Blume, A., Binding of cationic model peptides (KX)4K to anionic lipid bilayers: Lipid headgroup size influences secondary structure of bound peptides. Biochim. Biophys. Acta 2017, 1859, 415-424. 32. Baio, J. E.; Zane, A.; Jaeger, V.; Roehrich, A. M.; Lutz, H.; Pfaendtner, J.; Drobny, G. P.; Weidner, T., Diatom Mimics: Directing the Formation of Biosilica Nanoparticles by Controlled Folding of Lysine-Leucine Peptides. J. Am. Chem. Soc. 2014, 136, 15134-15137. 33. DeGrado, W. F.; Lear, J. D., Induction of peptide conformation at apolar water interfaces. 1. A study with model peptides of defined hydrophobic periodicity. J. Am. Chem. Soc. 1985, 107, 76847689. 34. Nick Pace, C.; Martin Scholtz, J., A Helix Propensity Scale Based on Experimental Studies of Peptides and Proteins. Biophys. J. 75, 422-427. 35. Béven, L.; Castano, S.; Dufourcq, J.; Wieslander, Å.; Wróblewski, H., The antibiotic activity of cationic linear amphipathic peptides: lessons from the action of leucine/lysine copolymers on bacteria of the class Mollicutes. Eur. J. Biochem. 2003, 270, 2207-2217. 36. 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. 37. Futaki, S.; Suzuki, T.; Ohashi, W.; Yagami, T.; Tanaka, S.; Ueda, K.; Sugiura, Y., Argininerich Peptides. J. Biol. Chem. 2001, 276, 5836-5840. ACS Paragon Plus Environment


Page 38 of 40

38 38. Rothbard, J. B.; Kreider, E.; VanDeusen, C. L.; Wright, L.; Wylie, B. L.; Wender, P. A., Arginine-Rich Molecular Transporters for Drug Delivery: Role of Backbone Spacing in Cellular Uptake. J. Med. Chem. 2002, 45, 3612-3618. 39. Asher, I. M.; Levin, I. W., Effects of temperature and molecular interactions on the vibrational infrared spectra of phospholipid vesicles. Biochim. Biophys. Acta 1977, 468, 63-72. 40. Tamm, L. K.; Tatulian, S. A., Infrared spectroscopy of proteins and peptides in lipid bilayers. Q. Rev. Biophys. 1997, 30, 365-429. 41. Blume, A.; Huebner, W.; Messner, G., Fourier transform infrared spectroscopy of 13C=O labeled phospholipids hydrogen bonding to carbonyl groups. Biochemistry 1988, 27, 8239-8249. 42. Hübner, W.; Blume, A., Interactions at the lipid–water interface. Chem. Phys. Lipids 1998, 96, 99-123. 43. Jackson, M.; Mantsch, H. H., The Use and Misuse of FTIR Spectroscopy in the Determination of Protein Structure. Crit. Rev. Biochem. Mol. Biol. 1995, 30, 95-120. 44. Hutchinson, E. G.; Thornton, J. M., A revised set of potentials for β-turn formation in proteins. Protein Sci. 1994, 3, 2207-2216. 45. Barth, A.; Zscherp, C., What vibrations tell about proteins. Q. Rev. Biophys. 2002, 35, 369-430. 46. Schwieger, C. Electrostatic and Non-Electrostatic Interactions of Positively Charged Polypeptides with Negatively Charged Lipid Membranes. Ph.D., Martin-Luther University HalleWittenberg, Halle(Saale), 2008. 47. Walrant, A.; Correia, I.; Jiao, C.-Y.; Lequin, O.; Bent, E. H.; Goasdoué, N.; Lacombe, C.; Chassaing, G.; Sagan, S.; Alves, I. D., Different membrane behaviour and cellular uptake of three basic arginine-rich peptides. Biochim. Biophys. Acta 2011, 1808, 382-393. 48. Wimley, W. C.; White, S. H., Experimentally determined hydrophobicity scale for proteins at membrane interfaces. Nat. Struct. Biol. 1996, 3, 842-848. 49. Sun, W. J.; Tristram-Nagle, S.; Suter, R. M.; Nagle, J. F., Structure of gel phase saturated lecithin bilayers: temperature and chain length dependence. Biophys. J. 1996, 71, 885-891. 50. Watts, A.; Harlos, K.; Marsh, D., Charge-induced tilt in ordered-phase phosphatidylglycerol bilayers Evidence from x-ray diffraction. Biochim. Biophys. Acta 1981, 645, 91-96. 51. 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. 52. Wang, J.; Gambhir, A.; McLaughlin, S.; Murray, D., A Computational Model for the Electrostatic Sequestration of PI(4,5)P2 by Membrane-Adsorbed Basic Peptides. Biophys. J. 2004, 86, 1969-1986. 53. Castano, S.; Desbat, B.; Cornut, I.; Méléard, P.; Dufourcq, J., α-Helix to β-sheet transition within the Leu i Lys j (i=2j) series of lytic amphipathic peptides by decreasing their size. Lett. Pept. Sci. 1997, 4, 195-200. 54. Long, J. R.; Oyler, N.; Drobny, G. P.; Stayton, P. S., Assembly of α-helical Peptide Coatings on Hydrophobic Surfaces. J. Am. Chem. Soc. 2002, 124, 6297-6303. 55. Gustafsson, M.; Vandenbussche, G.; Curstedt, T.; Ruysschaert, J.-M.; Johansson, J., The 21residue surfactant peptide (LysLeu4)4Lys (KL4) is a transmembrane α-helix with a mixed nonpolar/polar surface. FEBS Letters 1996, 384, 185-188. 56. Sáenz, A.; Cañadas, O.; Bagatolli, L. A.; Johnson, M. E.; Casals, C., Physical properties and surface activity of surfactant-like membranes containing the cationic and hydrophobic peptide KL4. FEBS J. 2006, 273, 2515-2527. 57. Cevc, G.; Watts, A.; Marsh, D., Non-electrostatic contribution to the titration of the orderedfluid phase transition of phosphatidylglycerol bilayers. FEBS Letters 1980, 120, 267-270.


Page 39 of 40 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


39 58. 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, 317335.



40 TOC Graphics


Page 40 of 40

Interaction of Short Pentavalent Cationic Peptides with Negatively

Recommend Documents