Hepatitis G Virus Synthetic Peptides with

Dec 11, 2008 - The obtained results are consistent with the existence of electrostatic (attractive) intermolecular interactions between the two positi...
1 downloads 0 Views 2MB Size
Download PDF
J. Phys. Chem. B 2009, 113, 319–327

319

Interaction of GB Virus C/Hepatitis G Virus Synthetic Peptides with Lipid Langmuir Monolayers and Large Unilamellar Vesicles Silvia Pe´rez-Lo´pez,*,† Nuria Vila-Romeu,‡ M. Asuncio´n Alsina Esteller,† Marta Espina,† Isabel Haro,§ and Concepcio´ Mestres† Department of Physical Chemistry, Faculty of Pharmacy, UniVersity of Barcelona, AV. Joan XXIII s/n, 08028 Barcelona, Spain, Department of Physical Chemistry, Faculty of Sciences, UniVersity of Vigo, Campus of Ourense, 32004 Ourense, Spain, and Unit of Synthesis and Biomedical Application of Peptides, Department of Biomedical Chemistry, IQAC-CSIC, Barcelona, Spain ReceiVed: August 4, 2008; ReVised Manuscript ReceiVed: October 24, 2008

In this paper, we aimed to continue the previous study undertaken with one segment of E1 protein belonging to the GB virus C/hepatitis G virus (GBV-C/HGV), specifically between the 53-66 amino acids and their palmitoyl derivative peptide. The sequence selection has been made on the basis of different prediction algorithms of hydrophobicity and antigenicity. Their interactions between two different in vitro membrane models, lipid Langmuir monolayers and vesicles of different lipidic composition, have been evaluated. For this purpose, different lipids, varying the charge and the unsaturations of the hydrocarbon chain 1,2-dipalmitoylsn-glycero-3-phosphocholine (DPPC), 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), 1,2-dipalmitoylsn-glycero-3-[phospho-rac-(1-glycerol)] (sodium salt) (DPPG) and 1,2-dioleoyl-sn-glycero-3-[phospho-rac(1-glycerol)] (sodium salt) (DOPG), have been selected. Miscibility and peptides/lipids interactions have been analyzed on the basis of surface pressure (π)-mean molecular area (A) isotherms, which have been recorded for pure and mixed monolayers of different composition spread at the air/water interface. Furthermore, E1(53-66) sequence and PalmE1(53-66) have been labeled with a fluorescent group, succinimidyl 6-(N(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)hexanoate (NBD succinimide), in order to study their behavior in the presence of vesicles. The obtained results are consistent with the existence of electrostatic (attractive) intermolecular interactions between the two positive net charges of the peptides and the polar heads of negativecharged lipids. However, both the lipidic membrane fluidity and the palmitic chain linked to the native peptide play an important role in the balance between the electrostatic forces established at the interface and the hydrophobic ones established inside the membrane. The fluorescence assays have demonstrated that electrostatic forces clearly predominate over the hydrophobic interactions only when the native sequence is retained at the polar interface of DPPG and DOPG vesicles. However, the palmitic tail linked to the peptide helped its penetration in the hydrophobic environment of the membrane, and this process was favored by decreasing the membrane fluidity. Introduction The GBV-C/HGV is a single-stranded positive sense RNA virus belonging to the FlaViViridae family.1,2 It was discovered in 1996 by two different research teams at the same time.2,3 The virus is referred in the literature to as both GBV-C and HGV, with the current taxonomic name being GBV-C/HGV. Its pathogenicity has not been totally demonstrated with some reports describing it as an innocent virus which causes no significant disease, whereas other studies attributed to GBVC/HGV a fulminant hepatitis.4-6 In the past few years, the importance of this virus has increased because it has been determined that its association with the human immunodeficiency virus (HIV) could benefit patients with acquired immune deficiency syndrome (AIDS), leading to a considerable number of studies being reported.7-9 GBV-C/HGV might prevent HIV entry into the cells,10-12 although the mechanisms involved in this reaction as well as in the replication of the virus into cells are not well-known yet. Understanding the interactions between * To whom correspondence should be addressed. E-mail: [email protected]. † University of Barcelona. ‡ University of Vigo. § IQAC-CSIC.

the virus and biomembranes would be the first step in the elucidation of these processes. The GBV-C/HGV virus is composed of enveloped proteins, as E1 and E2, and nonstructural proteins, as NS2, NS3, NS4A, NS4B, NS5A, and NS5B.13 In this paper, we present a study undertaken with one segment of E1 protein belonging to this virus, specifically between the 53-66 amino acids. This segment was selected because of the prediction algorithm PHDhtm showing a transmembrane helix near the region corresponding to the 53-66 (AGLAVRPGKSAAQL) a fact that could be of interest in its interaction with lipid membranes.14 Also, we have modified covalently the E1(53-66) peptide with a fatty acid, a palmitic acid, in an attempt to mimic its hydrophobic native environment, as it has been described that the increased hydrophobicity of lipidated protein or peptides generally results in improving its attractive interactions with cell membrane lipids.15-18 Other synthesized lipopeptides exhibit strong surface activity and important biological properties.19 Among the fatty acids used to acylate proteins, myristate and palmitate (with 14- and 16-carbon saturated fatty acids, respectively) are the two most common forms of protein modification.20 In previous studies,21 the peptides were dissolved in aqueous subphases and it was probed that the lipophilicderivatized PalmE1(53-66) interacted with higher intensity than

10.1021/jp806938y CCC: $40.75  2009 American Chemical Society Published on Web 12/11/2008

320 J. Phys. Chem. B, Vol. 113, No. 1, 2009 its native sequence with in vitro model membranes (built at the air/aqueous solutions interface). In the present paper, we have evaluated the interactions of these peptides with two different membrane models: lipid Langmuir monolayers spread at the air/ water interface22,23 and vesicles24 of different lipidic composition, both extensively used to build in vitro biomembranes. For this purpose, and given that the lipid fraction of biological membranes is formed by phospholipids varying in their chain length and ionic character, we have selected different lipids varying the charge and the unsaturations of the hydrocarbon chain: DPPC, DOPC, DPPG, and DOPG. It is known that the fusion process of the virus and biological membranes plays a vital and important role in many cellular processes.25,26 Thus, to study the miscibility and interactions of both peptides with the lipidic membranes, we have compared the results obtained at the air/water interface (by recording the surface pressure (π)- mean molecular area (A) isotherms of pure and mixed monolayers) with those obtained in large unilamellar vesicles (LUVs). In order to characterize the interaction between these vesicles and the peptides, we have recorded the fluorescence spectra under different conditions. Because the original sequence did not contain any amino acid with fluorescent properties (for example, Tyr or Trp), it was necessary to introduce an extrinsic fluorescent group. The peptides E1(53-66) and PalmE1(53-66) have been labeled with the fluorescent group NBD succinimide, which was linked through the amino-terminal group of the E1(53-66) and to theaminogroupatR-positionoftheaminoacidLys(PalmE1(53-66)). Numerous studies have been reported using this fluorescent label due to its small size and noninterference with the experimental molecules.27,28 Thus, the NBD group is widely used to fluorescently label proteins, peptides, and lipids and has been shown to be a very sensitive probe of the environment in which it resides. Materials and Methods Lipids and Chemicals. Ultrapure water was produced by deionization and Nanopure purification coupled to a Milli-Q purification system (Milli-Q system, Millipore Corp.) up to a resistivity of 18.2 MΩ cm. Chloroform and methanol proanalysis were from Merck (Poole, Dorset, UK). 1,2-Dipalmitoyl-snglycero-3-phosphocholine, 1,2-dioleoyl-sn-glycero-3-phosphocholine, 1,2-dipalmitoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (sodium salt), and 1,2-dioleoyl-sn-glycero-3-[phospho-rac-(1glycerol)] (sodium salt) were purchased from Avanti Lipids. Throughout this study, the aqueous subphase was phosphatebuffered saline (PBS) (pH 7.4, conductivity 15.4 mS cm-1, 313.35 mOsm kg-1). Peptide Synthesis and Fluorescent Label. The syntheses of peptides E1(53-66) and PalmE1(53-66) have been previously described.21 Briefly, peptides were synthesized manually by solid-phase Fmoc (N-R-fluorenylmethyloxycarbonyl) chemistry. They were synthesized on a Rink-amide resin with Fmocprotected amino acids. Activation and coupling was done in the presence of N,N′-diisopropylcarbodiimide/1-hydroxybenzotriazole (DIPCDI/HOBt), and peptides were N-terminal acetylatedwithaceticanhydride.Thepalmitoylderivative(PalmE1(53-66)) was obtained also working in the solid phase by using palmitic acid and DIPCDI/HOBt coupling reagents in dimethylformamide/dichloromethane, as previously described for other synthetic lipopeptides.29 As E1(53-66) does not contain in its primary sequence any fluorescent amino acid, it was necessary to use a fluorescent label, NBD succinimide. Labeling of peptides was achieved via two different synthetic strategies: through the binding of amine terminus on the resin-bound peptide E1(53-66) (Scheme 1), or by means of ε-amino group

Pe´rez-Lo´pez et al. SCHEME 1: Labeling of E1(53-66) Peptide via the Amine Terminus on the Resin-Bound Peptide

of lysine on the isolated and purified PalmE1(53-66). In order to maintain the amine group in a reactive nonprotonated form, the conjugation took place in an organic medium with slightly basic pH (see Scheme 2). Purification of crude molecules was achieved by preparative HPLC on a Kromasil C-18 column. The union of the dye to the E1(53-66) peptide was via amino terminal group, by an amide bond. First, we proceed to the deprotection of the Fmoc amino-terminal group. The NBD group was coupled at this reaction stoichiometry, 1 equiv peptidilresin + 1.5 equiv NBD succinimide + 1.5 equiv N,Ndiisopropylethylamine. Immediately after, we proceed to dry the resin, following the same protocol as with the native sequence, and the peptidic crude was purified by means of HPLC. The coupling of the dye to PalmE1(53-66) was made through the amine group of amino acid Lys, present on the primary sequence of the peptide, because the amine group of the last amino acid of the sequence was bound to the palmitic acid. That is the reason we modified the protocol used with E1(53-66). Now, the derivatization was on liquid phase and with other experimental conditions. It is known that the theoretical pKa of amino acid Lys is approximately 10.5,30 even though it could vary depending on the primary peptidic sequence. This is particularly important for our experiment because it was necessary for the amine group being deprotonated to react with the succinimide ester of the dye. The protocol was adapted to label approximately 5 mg of PalmE1(53-66), enough quantity to make all the fluorescence experiments. A solution of 2 mM of peptide dissolved in 1.5 mL of dimethyl sulfoxide (DMSO) was prepared and triethylamine 100 mM was added, ensuring the derivatized amines were deprotonated. Just then, NBD succinimide was dissolved in 1.5 mL of DMSO, the final concentration being 4 mM. The NBD succinimide was added to the PalmE1(53-66), and the reaction was incubated overnight at room temperature, with continuous stirring and protected from light. Purification of the peptidic crude was made by HPLC. Langmuir Monolayers. All monolayer experiments were carried out by the Wilhelmy plate method, using a Langmuir balance (KSV LB5000), equipped with a rectangular Teflon trough (850 mL) of 81 000 mm2. The balance was calibrated by recording the well-known isotherm of stearic acid at 293 K. The Teflon trough was regularly cleaned with hot chromic acid and rinsed with doubled-distilled water. Surface pressure was

Interactions of GB Virus C/Hepatitis G Virus

J. Phys. Chem. B, Vol. 113, No. 1, 2009 321

SCHEME 2: Labeling of PalmE1(53-66) Peptide via the ε Amino Group of Amino Acid Lys on Liquid Phase

measured to a precision of (0.05 mN m-1. Lipid-peptide monolayers were spread from methanol/chloroform (1:1) solutions on a subphase containing PBS at pH 7.4. The organic solvent was allowed to evaporate for 20 min before the film

was compressed at a rate of 20 mm min-1. In all the experiments, the subphase temperature was 293 ( 1 K. Vesicles Preparation. Lipid vesicles of different lipid compositions were prepared in order to record the fluorescence

Figure 1. Surface pressure-mean molecular area (π-A) compression isotherms for E1(53-66), the lipids: (A) DPPC, (B) DOPC, (C) DPPG, and (D) DOPG. Lipid molar fraction: (4) X ) 1, (2) X ) 0.8, (O) X ) 0.6, (b) X ) 0.4, (9) X ) 0.2, (0) X ) 0 and their mixtures on PBS pH 7.4 subphase. Inset: Plots of compression moduli (Cs-1) as a function of the surface pressure for pure and mixed monolayers.

322 J. Phys. Chem. B, Vol. 113, No. 1, 2009 spectra. Pure lipids DPPC, DOPC, DPPG, and DOPG were separately dissolved in a chloroform/methanol (2:1, v/v) mixture, and the lipid solutions were dried under a nitrogen stream. The samples were stored overnight in a vacuum oven at room temperature to eliminate the residual solvent. Then, the lipid films were hydrated with 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) buffer 5 mM, pH 7.4. Large unilamellar vesicles (LUVs) of DPPC, DOPC, DPPG, and DOPG were prepared by hydration of the lipid film with the HEPES buffer followed by 10 freeze-thaw cycles. LUVs were made by standard extrusion techniques under N2 using two 100 nm polycarbonate filters (Nucleopore, Pleasanton, CA) in a highpressure extruder (Lipex, Biomembranes, Vancouver, Canada)31 above the transition temperature of the phospholipids. Phospholipid concentration was determined by phosphorus quantification as previously described.32 The liposome’s size was measured by the sample diffusion coefficient by photon correlation spectroscopy (Coulter N4 MB, Luton, UK) and was 90 ( 9 nm with a polydispersity index lower than 0.15. Fluorescence Titration Measurements. Peptide-lipid interactions were studied by monitoring the changes in the NBD fluorescence emission spectra with increasing lipid concentrations. Fluorescence experiments were performed on a Perkin-Elmer (Beaconsfield Bucks, UK) spectrofluorimeter LS 50, using 1 cm path length quartz cuvettes. The excitation and emission wavelength used was 480 and 540 nm, respectively. Slits with a nominal bandpass of 5 nm were used for both excitation and emission beams. Emission fluorescence spectra were recorded for each peptide at 1 µM concentration in 5 mM HEPES pH 7.4, at room temperature by the incremental addition of aliquots of a 7 mM phospholipid solution. Before the fluorescence experiments were started, it was necessary to know if 1 µM peptide concentration was able to induce peptide aggregates. Owing to this fact, the variation of intensity of fluorescence of these two peptide sequences was analyzed. The test was made during 1 h, and the variation of intensity of fluorescence of the peptide sequences in the cuvette was analyzed. As no decrease in the fluorescence intensity was observed, it was decided that the studies would be made at a concentration of 1 µM. The peptide to lipid molar ratios were in a range from 1:25 to 1:1200. Suspensions were continuously stirred, and they were left to equilibrate for 3 min before the spectrum was recorded. Fluorescence intensities were corrected for a lightscattering contribution by subtraction of the appropriate vesicle blank, and the last correction was obtained from a parallel lipid titration of N-acetyltryptophanamide (NATA), which does not interact with lipids.33 The absorbance of peptide samples was measured by using a LKB-Biochrom Ultrospec II spectrophotometer at 540 nm. Results Mixed E1(53-66)-Lipid Monolayers. Figure 1A shows the π-A compression isotherms of pure monolayers of the peptide E1(53-66) and DPPC and mixed films of mole fraction (XDPPC) 0.2, 0.4, 0.6, and 0.8. The isotherms recorded for pure components, i.e., E1(53-66)21 and DPPC, are in good agreement with the data already published. When the amount of E1(53-66) increases in the monolayer, different changes can be observed: (i) the DPPC phase transition disappears for monolayers of XDPPC ) 0.8, and (ii) the collapse pressure is only reached in the DPPC-E1(53-66) (0.8:0.2) mixture and its value is similar to the pure DPPC monolayer. Inset of Figure 1A shows the compressibility modulus (Cs-1), defined as -A(dπ/dA)T, as a function of the pressure for the different mixtures. Its values are very useful to characterize the state and phase transitions of a monolayer under compression.34

Pe´rez-Lo´pez et al.

Figure 2. Plots of A for E1(53-66) as a function of the lipid molar fractions at different surface pressures for pure and mixed monolayers: (A) DPPC, (B) DOPC, (C) DPPG, and (D) DOPG. Surface pressures: (O) 5 mN m-1, (4) 10 mN m-1, (b) 20 mN m-1, (2) 30 mN m-1.

As Figure 1A shows, only for XDPPC > 0.4 compressibility reached values nearly consistent with a liquid-condensed state of the film. Plots of area/molecule vs monolayer composition at different pressures are shown in Figure 2A. As it can be observed, only for XDPPC ) 0.8 positive deviations of the ideal behavior occur, which increase at low pressures. However, for XDPPC e 0.6 the system behaves near ideal. E1(53-66)-DOPC35 mixed isotherms have been recorded in order to study how the unsaturation and length of the lipid hydrocarbon chain affect the lipid-peptide interaction. In this case, the isotherms of mixtures (Figure 1B) have been more condensed than those described previously (DPPC), and again the collapse pressure does not change for lipid molar ratios g0.4. Furthermore, Cs-1 (inset in Figure 1B) diminishes as the proportion of the peptide increases in the monolayer. Plots of mean molecular area vs molar fraction of DOPC show negative deviations for all surface pressures and for any monolayer composition (Figure 2B). In the next step, the miscibility of E1(53-66) with net negative charge (DPPG36 and DOPG37) phospholipids has been studied. In mixtures containing DPPG (Figure 1C), small proportions of the peptide (XDPPG ) 0.8) produce an important smoothing effect upon the phase transition of the pure lipid monolayer. Collapse pressure is slightly reduced at XDPPG ) 0.8 and its value cannot be measured for mixtures with higher peptide proportion. For XDPPG < 0.8, the presence of the peptide in the monolayer produces an important reduction in Cs-1 values and only for XDPPG ) 0.8 it reaches values consistent with a liquid-condensed state. Figure 2C shows positive deviations for DPPG/E1(53-66) mixtures of XDPPG ) 0.8, whereas slight negative deviations from the ideal behavior have been obtained for lower lipid molar fractions and at low surface pressures. At π g 20 mN m-1, the system behaves near ideal. The shapes of all the isotherms recorded for DOPG/E1(53-66) mixtures (Figure 1D) are in agreement with a condensed monolayer and, again, the collapse pressure does not change with the monolayer composition (for XDOPG g 0.4). Area/molecule of the mixed films is reduced as the proportion of the peptide increases in the mixture. Plots of mean molecular area vs XDOPG (Figure 2D) exhibit the same pattern as described previously for DPPG at lower pressures. Mixed PalmE1(53-66)-Lipid Monolayers. The π-A compression isotherms of pure monolayers of the modified peptide PalmE1(53-66) and mixed films of mole fractions (Xlipid) 0.2,

Interactions of GB Virus C/Hepatitis G Virus

J. Phys. Chem. B, Vol. 113, No. 1, 2009 323

Figure 3. Surface pressure-mean molecular area (π-A) compression isotherms for PalmE1(53-66), the lipids: (A) DPPC, (B) DOPC, (C) DPPG, and (D) DOPG. Lipid molar fraction: (4) X ) 1, (2) X ) 0.8, (O) X ) 0.6, (b) X ) 0.4, (9) X ) 0.2, (0) X ) 0 and their mixtures on PBS pH 7.4 subphase. Inset: Plots of compression moduli (Cs-1) as a function of the surface pressure for pure and mixed monolayers. EX TABLE 1: ∆GM (in J mol-1) for Different Lipid-PalmE1(53-66) Mixtures at Various Pressures

Xlipid

π/(mN m-1)

DPPC

DOPC

DPPG

DOPG

0.2

5 10 20 30 5 10 20 30 5 10 20 30 5 10 20 30

-33.06 -93.12 160.78 -77.13 -3.93 35.51 885.00 1396.72 14.80 -133.58 346.66 151.81 25.53 148.64 409.52 53.35

-164.90 -195.13 -187.24 -2261.22 -239.43 -394.29 -475.60 -1119.25 66.72 -146.01 -229.71 344.60 11.05 -219.15 -310.29 1666.17

-266.42 -482.44 -625.43 -901.04 -432.05 -565.93 -323.99 -306.70 -452.34 -463.98 -282.65 230.57 -569.84 -716.35 -1363.98 -1855.95

-197.91 -280.16 286.28 1077.59 -48.03 -188.88 497.82 824.42 -31.91 57.34 1241.17 1041.03 -38.69 -213.92 -700.50 -1840.22

0.4

0.6

0.8 Figure 4. Plots of A for PalmE1(53-66) as a function of the lipid molar fractions at different surface pressures for pure and mixed monolayers: (A) DPPC, (B) DOPC, (C) DPPG, and (D) DOPG. Surface pressures: (O) 5 mN m-1, (4) 10 mN m-1, (b) 20 mN m-1, (2) 30 mN m-1.

0.4, 0.6, and 0.8 are shown in Figure 3. The binding of the hydrocarbon chain to the native amino acidic sequence of E1(53-66) produces radical changes in the shape of the isotherms, which exhibit an increase in the area/molecule values.21 Compressibility values increase at low pressures (0-30 mN m-1) as the peptide proportion become higher in the monolayer (Inset Figure 3). However, for pressures above the peptide transition (about 30 mN m-1) a high reduction of its values can be observed. Collapse pressure can be obtained for mixed monolayers at XDPPC > 0.4, and its values are similar to the pure DPPC (Figure 3A). The lipid transition can only be observed at XDPPC ) 0.8. Plots of mean molecular area vs molar fraction of DPPC (Figure 4A) exhibit little positive deviations from the ideality for mixtures of XDPPC > 0.2; however, at XDPPC ) 0.2 the system obeys the additively rule.

To get insight into the interactions established between PalmE1(53-66) and DPPC at the air/water interface, we have EX 38,39 calculated the excess Gibbs free energy of mixing (∆GM ).

∆GEX M )

π π π A12 dπ - X1 ∫πf0 A1 dπ - X2 ∫πf0 A2 dπ ∫πf0

where A12 is the mean area per molecule in the mixed film, A1 and A2 are the areas per molecule in the pure films, X1 and X2 are the molar fractions, and π is the surface pressure (mN m-1). Numerical data have been calculated from the compression isotherms according to the mathematical method of Simpson. The positive values obtained for mixtures where XDPPC g 0.2 (Table 1), indicate that the process of mixing is not thermodynamically favored, and therefore repulsive or less attractive interactions must prevail in the mixtures with regards to the pure monolayers.

324 J. Phys. Chem. B, Vol. 113, No. 1, 2009

Pe´rez-Lo´pez et al.

TABLE 2: E1(53-66) Bound Fraction (fbound) to LUVs of DPPG and DOPG at Different Peptide/Lipid Molar Ratios DPPG

DOPG

peptide/lipid

fbound (%)

peptide/lipid

fbound (%)

1/25 1/50 1/100 1/200 1/300 1/400 1/500 1/600

31 47 64 78 84 88 90 91

1/25 1/50 1/100 1/200 1/300 1/400 1/500 1/600

69 82 90 95 96 97 98 98

In addition, for DOPC-PalmE1(53-66) mixed monolayers (Figure 3B) important changes in the isotherm’s shape can be observed due to the presence of the peptide. Collapse pressures of mixed monolayers vary between 39 and 44 mN m-1, which correspond to the pure peptide and lipid, respectively (inset in Figure 3B). Plots of area/molecule vs lipid molar fraction show important negative deviations from ideality for all pressures and lipid molar fractions studied (Figure 4B). When PalmE1(53-66) is mixed with negative net charged lipids (DPPG and DOPG), again the presence of the peptide produces an important change in the mixed films compressibility, which is more condensed than DPPG at low surface pressures and becomes more expanded as the pressure increases (inset in Figure 3C). Plots of mean molecular area vs DPPG molar fraction (Figure 4C) show, in contrast to the results obtained by its zwitterionic homologous lipid, a negative deviation from the ideality for XDPPG ) 0.8. Similar behavior is observed for mixtures with DOPG (Figures 3D and 4D) where, however, collapse pressures of the mixtures range between 40 and 45 mN m-1. Peptide Binding to the Lipid Vesicles: A Fluorescence Study. Peptides binding to lipid vesicles have been investigated by monitoring the changes in the NBD fluorescence emission spectra of the peptides upon the addition of LUVs. The lipids used in the assay have been the same as the Langmuir experiments, DPPC, DOPC, DPPG, and DOPG. To quantify the interaction between the peptides and the LUVs, it has been necessary to correct the fluorescence values, since the dilution error is not included. Each value has been multiplied by the correction factor obtained with NATA. Figure 5 shows the corrected emission fluorescence spectra of the peptide E1(53-66) in the presence of liposomes LUV at various peptide/lipid molar ratios. Only for anionic lipids DPPG (Figure 5A) or DOPG (Figure 5B) can be observed an increase of the fluorescence intensity values. In the case of zwitterionic lipids, DPPC or DOPC, no significant increase has been noticed (data not shown). Assuming a two-state equilibrium between water-soluble aggregates and membrane-bound peptides, the apparent mole fraction partition coefficients have been determined by fitting the binding curves, obtained after representing the quotient between fluorescence intensity of titrated samples and that corresponding to peptide solutions at the same solution (F/F0) vs the lipid/peptide relationship, to a hyperbolic equation.40

F ) F0

[( ) ]

Fmax -1 F0 K + [L]

1 + [L]

(1)

where [L] is lipid concentration, F0 is the initial fluorescence value in absence of liposomes, F is the fluorescence for each

Figure 5. Fluorescence emission spectra of the E1(53-66) in HEPES 5 mM pH 7.4 and in the presence of (A) DPPG and (B) DOPG LUVs at different peptide/lipid molar ratios: (black squares) peptide, (red circles) 1:25, (green up triangles) 1:50, (blue down triangles) 1:100, (left-pointing triangles) 1:200, (right-pointing triangles) 1:300, (diamonds) 1:400, (pentagons) 1:500, (hexagons) 1:600, (stars) 1:700, (blue circles) 1:800, (blue squares) 1:900, (brown circles) 1:1000, (violet up triangles) 1:1100, (black squares) 1:1200. au ) arbitrary units. Measurements were carried out with 1 µM peptide. Inset: Partitioning curves as estimated from NBD fluorescence changes in E1(53-66) in the presence of increasing concentrations of different lipidic composition LUVs. The solid lines correspond to the best fit of the experimental values to a hyperbolic function.

liposome concentration, and Fmax is the maximum fluorescence. From eq 1, K was calculated (binding constant). By applying eq 2, Kx was also calculated (real binding constant).

Kx )

W K

(2)

where W is the molar concentration of water (55.3 M) and K is mole fraction partition coefficient (value obtained by eq 1). For each peptide, a value of Kx has been obtained, which has been used to determine the bound fraction (fbound) of each peptide to liposomes (eq 3):

fbound )

KxL W + (KxL)

(3)

As can be observed in Table 2, the percentages of E1(53-66) bound fraction have been very high with the DPPG LUVs (91% for a peptide molar ratio of 1:600). The same pattern has been observed for the DOPG LUVs, as the peptide bound fraction has been practically 100. The same procedure has been applied to analyze the results obtained for the lipophilic derivative. The corrected spectra are shown in Figure 6. As opposed to E1(53-66), PalmE1(53-66) interacts with all the lipids assayed, both zwitterionics and anionics. Nevertheless, the peptide bound fraction has been lower than the obtained for the native sequence: it oscillates between 49% for LUVs of DPPC and DOPG, 58% for DPPG, and 63% for DOPC (Tables 3 and 4). Discussion To examine the miscibility of the film components at the air/ water interface, the analysis of the values of collapse pressures (πc) can be helpful;41 the variation of πc with the molar ratio of the components indicates two-dimensional (2D) miscibility. However, for the experimental conditions in which it could be obtained, πc showed similar values in the mixtures (E1(53-66)/ lipids) to those obtained for the pure lipids monolayers, which indicates that both components should not be miscible when the films collapse. Only for mixtures of PalmE1(53-66) with DOPC and DOPG, collapse pressure varies with the monolayer composition, which proves that both components are miscible

Interactions of GB Virus C/Hepatitis G Virus

J. Phys. Chem. B, Vol. 113, No. 1, 2009 325

Figure 6. Fluorescence emission spectra of the PalmE1(53-66) in HEPES 5 mM pH 7.4 and in the presence of (A) DPPC, (B) DOPC, (C) DPPG, and (D) DOPG LUVs at different peptide/lipid molar ratio (black squares) peptide, (red circles) 1:25, (green up triangles) 1:50, (blue down triangles) 1:100, (left-pointing triangles) 1:200, (right-pointing triangles) 1:300, (blue diamonds) 1:400, (pentagons) 1:500, (hexagons) 1:600, (green stars) 1:700, (blue circles) 1:800, (blue squares) 1:900, (brown circles) 1:1000, (violet up triangles) 1:1100, (black squares) 1:1200. au ) arbitrary units. Measurements were carried out with 1 µM peptide. Inset: Partitioning curves as estimated from NBD fluorescence changes in PalmE1(53-66) in the presence of increasing concentrations of different lipidic composition LUVs. The solid lines correspond to the best fit of the experimental values to a hyperbolic function.

within the whole range of surface pressures. This fact is attributed to both (i) the higher fluidity of the lipid monolayer (due to the unsaturation of the hydrocarbon chain) which favors the penetration of the peptide in the membrane, and (ii) the presence of the hydrophobic chain, linked to the structure of the peptide, which increases its stability at the interface and the attractive intermolecular interactions with the lipidic environment of the membrane. To get insight into the properties of these 2D systems, we have analyzed plots of area vs lipid molar fraction at different surface pressures.42 Positive deviations from the ideality observed for the mixtures E1(53-66)/DPPC of XDPPC ) 0.8 and PalmE1(53-66)/DPPC of XDPPC > 0.2 prove that both components are miscible and interact in the monolayer at the surface pressures studied. However, the fact that the collapse of the binary monolayers does not change with DPPC molar ratio indicates that they are not miscible at surface pressures higher than the peptide transition pressure. The expansion observed for mixed peptides/DPPC monolayers could be attributed to an increase in the distance between the lipid hydrophobic chains due to the presence of peptides, which would be placed between the lipid molecules in the film. The hydrophobic chain, linked to the peptide, stabilizes it at the interface, increases the range of molar ratios in which the two components are miscible (with DPPC), and diminishes the area of excess because of their stronger attractive interaction with the lipid chains. In contrast, mixtures with DOPC show negative deviations for the two peptides and for any monolayer composition. This fact reflects the miscibility of the two monolayer components and the existence of attractive interactions between them at the interface. These results denote that the peptides can penetrate into the lipid monolayer because of its lower density (see Figure 7A for illustration). This is due to the unsaturation in the DOPC chain, which makes the monolayer more fluid than the DPPC film.

Results obtained for negatively charged lipids (DPPG and DOPG) are consistent, as it can be expected, with the existence of electrostatic (attractive) intermolecular interactions of the two positive net charges of E1(53-66) and the polar heads of the lipids. Interestingly, the negative deviations from the ideal behavior are greater for the zwitterionic DOPC than for DOPG. Therefore, it seems that increasing the fluidity of the membrane, induced by the unsaturation of the hydrocarbon chain, has a greater influence on the properties of the mixtures than the electrostatic attractive forces acting between the components at the interface. This behavior could be related to the fact that, in the presence of DPPC and DPPG, the peptide remains partially immersed in the subphase, and this conformation (see Figure 7B for illustration) increases its electrostatic interaction with the negative charge of the lipid polar heads. However, the insertion of the peptide in a more fluid monolayer could provoke changes in its random structural conformation due to the drastic decrease of the environment polarity. This fact increases the attractive interactions between hydrophobic residues of the peptide and the lipid chains, which provokes the contraction of the monolayer and drastically reduces the electrostatic interactions between the peptide and the polar heads. However, even if attractive hydrophobic interactions prevail, the values of ∆GEX M (Table 1) denote the existence of weak intermolecular forces for all the mixtures studied. The behavior of the peptides in the presence of phospholipid vesicles are also of great importance for understanding the mechanism of its biological activity. Although several types of intermolecular forces are involved in the interaction of peptides with lipid membranes, it is commonly accepted that interaction primarily arises from both hydrophobic and electrostatic effects. These interactions of peptides with lipid bilayer membranes were examined by fluorescence measurements. The fluorescence assay has shown different behavior of the peptides depending on the

326 J. Phys. Chem. B, Vol. 113, No. 1, 2009

Pe´rez-Lo´pez et al.

Figure 7. Illustrative scheme of the orientation of peptides at the monolayers and lipid vesicles: (A) peptides interaction with DOPC, (B) peptides interaction with DPPC and DPPG, (C) E1(53-66) peptide retained at the liposome interface by means of the attractive electrostatic forces, and (D) PalmE1(53-66) insertion into the lipidic vesicles.

TABLE 3: PalmE1(53-66) Bound Fraction (fbound) to LUVs of DPPC and DOPC at Different Peptide/Lipid Molar Ratios DPPC

TABLE 4: PalmE1(53-66) Bound Fraction (fbound) to LUVs of DPPG and DOPG at Different Peptide/Lipid Molar Ratios

DOPC

DPPG

DOPG

peptide/lipid

fbound (%)

peptide/lipid

fbound (%)

peptide/lipid

fbound (%)

peptide/lipid

fbound (%)

1:25 1:50 1:100 1:200 1:300 1:400 1:500 1:600 1:700 1:800 1:900 1:1000 1:1200

2 4 8 15 21 26 30 34 38 41 44 46 49

1/25 1/50 1/100 1/200 1/300 1/400 1/500 1/600 1/700 1/800

5 10 17 30 39 46 51 56 60 63

1/25 1/50 1/100 1/200 1/300 1/500 1/600 1/700 1/800

4 8 15 26 34 46 51 55 58

1/25 1/50 1/100 1/200 1/300 1/400 1/500 1/600 1/700 1/800 1/900 1/1000 1/1100 1/1200

2 4 7 14 19 24 28 32 36 39 42 44 46 49

lipidic LUVs composition. The enhancement of the fluorescence intensity as the PalmE1(53-66)/lipid ratio increases, with all the lipids investigated, proves that the derivative peptide is included into LUVs. Thus, the presence of the fatty acid chain linked to the peptide favors its hydrophobic attractive interactions with the lipidic chains, which allows the inclusion of the peptide molecules into the lipidic bilayers. However, even if the binding of the native peptide (E1(53-66)) to anionic vesicles originates the highest increase of the fluorescence intensity, no significant changes are observed in the presence of the two zwitterionic lipids. The native sequence

would be retained at the liposome interface by means of the attractive electrostatic forces (Figure 7C). Recent literature described similar results for other lipopeptides, i.e., antimicrobial peptides, which need to penetrate into the cell membrane to have the antimicrobial effect. Several studies aimed to assess the effect of acylation have revealed that conjugation of fatty acid to cationic peptides enhances their antimicrobial activity.43-45 Furthermore, the NBD fluorescence intensity is strongly affected by the characteristic of its environment.46 Generally, the less polar the environment is, the lower the dielectric constant, and the lower the fluorescence intensity

Interactions of GB Virus C/Hepatitis G Virus of the fluorophore. This fact supports the hypothesis proposed above, which has placed the PalmE1(53-66) inside the bilayer (Figure 7D), leading to the decrease of the relative intensity fluorescence (Tables 3 and 4). In contrast, higher fbound values (Table 2) have been obtained for the native sequence anchored at the LUV interface (by means of the attractive electrostatic forces established between the peptide and the lipids polar heads). It is generally assumed that, during the peptide-membrane interaction, electrostatic attractions and hydrophobic attractive forces are the two major factors for the initial adsorption of peptide molecules on lipid bilayer surfaces. However, the electrostatic attraction is the dominant one when the native sequence of the positively charged peptide interacts with negatively charged interfaces, while the palmitic tail linked to the peptide favors its penetration in the hydrophobic environment of the membranes and this process is favored by decreasing the fluidity of the membrane. Conclusions The results obtained with the two membrane models investigated, lipid Langmuir monolayers and vesicles, are consistent with the existence of electrostatic (attractive) intermolecular interactions of the two positive net charges of the peptides and the polar heads of negatively charged lipids. However, both the fluidity of the lipidic membrane and the palmitic chain linked to the native peptide play an important role in the balance between the electrostatic forces established at the interface and the hydrophobic ones established inside the membrane. The fluorescence assays have demonstrated that electrostatic forces clearly predominate over the hydrophobic interactions only when the native sequence is retained at the polar interface of DPPG and DOPG membranes. However, the palmitic tail linked to the peptide favors its penetration in the hydrophobic environment of the membranes, and this process is favored by decreasing the membrane fluidity. Acknowledgment. This work was supported by grants from the Ministerio de Ciencia e Innovacio´n (Secretarı´a de Estado de Universidades, Direccio´n General de Programas y Transferencia de Conocimiento, Subdireccio´n General de Proyectos de Investigacio´n, Spain) CTQ 2006-15396-C02-01, CTQ 2006-15396-C02-02, and CTQ2006-04085/BQU and one of us, N.V., also thanks Xunta de Galicia for the grant of Rf. PGDIT06PXIB383004PR. References and Notes (1) Leary, T. P.; Muerhoff, A. S.; Simons, J. N.; Pilot-Matias, T. J.; Erker, J. C.; Chalmers, M. L.; Schlauder, G. G.; Dawson, G. J.; Desai, S. M.; Mushahwar, I. K. J. Med. Virol. 1996, 48, 60–67. (2) Linnen, J.; Wages, J., Jr.; Zhang-Keck, Z. Y.; Fry, K. E.; Krawczynski, K. Z.; Alter, H.; Koonin, E.; Gallagher, M.; Alter, M.; Hadziyannis, S.; Karayiannis, P.; Fung, K.; Nakatsuji, Y.; Shih, J. W.; Young, L.; Piatak, M., Jr.; Hoover, C.; Fernandez, J.; Chen, S.; Zou, J. C.; Morris, T.; Hyams, K. C.; Ismay, S.; Lifson, J. D.; Hess, G.; Foung, S. K.; Thomas, H.; Bradley, D.; Margolis, H.; Kim, J. P. Science 1996, 271, 505–508. (3) Simons, J. N.; Leary, T. P.; Dawson, G. J.; Pilot-Matias, T. J.; Muerhoff, A. S.; Schlauder, G. G.; Desai, S. M.; Mushahwar, I. K. Nat. Med. 1995, 1, 564–569. (4) Muller, C. J. Viral Hepat. 1999, 6 (1), 49–52. (5) Mun˜oz, S. J.; Alter, H. J.; Nakatsuji, Y.; Shih, J. W.; Reddy, R. K.; Jeffers, L.; Schiff, E. R.; Reid, A. E.; Marrone, A.; Rothstein, K.; Manzarbeitia, C.; Liang, T. J. Blood 1999, 94, 1460–1464. (6) Saitoh, H.; Moriyama, M.; Matsumura, H.; Goto, I.; Tanaka, N.; Aarakawa, Y. Hepatol. Res. 2002, 22, 288–296. (7) Lau, D. T.; Miller, K. D.; Detmer, J.; Kolberg, J.; Herpin, B.; Metcalf, J. A.; Davey, R. T.; Hoofnagle, J. H. J. Infect. Dis. 1999, 180, 1334–1337.

J. Phys. Chem. B, Vol. 113, No. 1, 2009 327 (8) Takamatsu, J.; Toyoda, H.; Fukuda, Y. N. New England J. Med. 2002, 346, 377–379. (9) Tillmann, H. L.; Manns, M. P. AntiViral Res. 2001, 52, 83–90. (10) Muerhoff, A. S.; Tillmann, H. L.; Manns, M. P.; Dawson, G. J.; Desai, S. M. J. Med. Virol. 2003, 70, 141–149. (11) Tillmann, H. L.; Heiken, H.; Knapik-Botor, A.; Heringlake, S.; Ockenga, J.; Wilber, J. C.; Goergen, B.; Detmer, J.; McMorrow, M.; Stoll, M.; Schmidt, R. E.; Manns, M. P. New England J. Med. 2001, 345, 715– 724. (12) Tillmann, H. L. PRN Notebook 2003, 8, 16–20. (13) Berzsenyi, M. D.; Bowden, D. S.; Roberts, S. K. J. Clin. Virol. 2005, 33, 257–266. (14) Rost, B.; Fariselli, P.; Casadio, R. Protein Sci. 1996, 5, 1704–1718. (15) Bailey, A. L.; Monck, M. A.; Cullis, P. R. Biochim. Biophys. Acta 1997, 1324, 232–244. (16) Esposito, C.; D’Errico, G.; Armenante, M. R.; Giannecchini, S.; Bendinelli, M.; Rovero, P.; D’Ursi, A. M. Biochim. Biophys. Acta 2006, 1758, 1653–1661. (17) Gonzalez-Christen, J.; Vergne, I.; Sussmuth, R.; Sidobre, S.; Prats, M.; Tocanne, J. F.; Laneelle, G. Biochim. Biophys. Acta 1998, 1368, 97–107. (18) Haro, I.; Pe´rez, S.; Garcia, M.; Chan, W. C.; Ercilla, G. FEBS Lett. 2003, 540, 133–140. (19) Bouffioux, O.; Berquand, A.; Eeman, M.; Paquot, M.; Dufreˆne, Y. F.; Brasseur, R.; Deleu, M. Biochim. Biophys. ActasBiomembr. 2007, 1768, 1758–1768. (20) Resh, M. D. Biochim. Biophys. Acta 1999, 1451, 1–16. (21) Pe´rez, S.; Min˜ones, J.; Espina, M.; Alsina, M. A.; Haro, I.; Mestres, C. J. Phys. Chem. B 2005, 109, 19970–19979. (22) Gaines, G. L. Insoluble monolayers at liquid-gas interfaces; Interscience: New York, 1966; pp 281-300. (23) Sospedra, P.; Alsina, M. A.; Haro, I. I.; Mestres, C.; Busquets, M. A. J. Colloid Interface Sci. 1999, 211, 130–136. (24) Sospedra, P.; Munoz, M.; Garcia, M.; Alsina, M. A.; Mestres, C.; Haro, I. Biopolymers 2000, 54, 477–488. (25) Weissenhorn, W.; Hinz, A.; Gaudin, Y. FEBS Lett. 2007, 581, 2150–2155. (26) White, J. M. Annu. ReV. Physiol. 1990, 52, 675–697. (27) Dowlut, M.; Hall, D. G.; Hindsgaul, O. J. Org. Chem. 2005, 70, 9809–9813. (28) Fery-Forgues, S.; Fayet, J. P.; Lopez, A. J. Photochem. Photobiol. A 1993, 70, 229–243. (29) Rojo, N.; Gomara, M. J.; Alsina, M. A.; Haro, I. J. Pept. Res. 2003, 61, 318–330. (30) Creighton, T. E. Proteins: Structures and Molecular Properties; W. H. Freeman: New York, 1993. (31) Hope, M. J.; Bally, M. B.; Webb, G.; Cullis, P. R. Biochim. Biophys. Acta 1985, 812, 55–65. (32) Bo¨ttcher, C. J. F.; van Gent, C. M.; Pries, C. Anal. Chim. Acta 2005, 24, 203–204. (33) Ladokhin, A. S.; Jayasinghe, S.; White, S. H. Anal. Biochem. 2000, 285, 235–245. (34) Davies, J. T.; Rideal R. K. Interfacial Phenomena, 2nd ed.; Academic Press: New York, 1963; p 265. (35) Min˜ones, J.; Dynarowicz-Latka, P.; Conde, O.; Min˜ones, J.; Iribarnegaray, E.; Casas, M. Colloids Surf. B Biointerfaces 2003, 29, 205–215. (36) Mestres, C.; Ortiz, A.; Haro, I.; Reig, F.; Alsina, M. A. Langmuir 1997, 13, 5669–5673. (37) MacDonald, R. C.; Gorbonos, A.; Momsen, M. M.; Brockman, H. L. Langmuir 2006, 22, 2770–2779. (38) Goodrich, F. C. Proceedings of the 11th International Congress on Surface ActiVity; Butterworths: London, 1957. (39) Pagano, R. E.; Gershfeld, N. L. J. Colloid Interface Sci. 1972, 41, 311–317. (40) White, S. H.; Wimley, W. C.; Ladokhin, A. S.; Hristova, K. Methods Enzymol. 1998, 295, 62–87. (41) Dynarowicz-Latka, P.; Kita, K. AdV. Colloid Interface Sci. 1999, 79, 1–17. (42) Nieto-Sua´rez, M.; Vila-Romeu, N.; Dynarowicz-Latka, P. Colloids Surf. A Physicochem. Eng. Asp. 2008, 321, 189–195. (43) Clausell, A.; Rabanal, F.; Garcia-Subirats, M.; Asuncion, A. M.; Cajal, Y. Luminescence 2005, 20, 117–123. (44) D’Errico, G.; D’Ursi, A. M.; Marsh, D. Biochemistry 2008, 47, 5317–5327. (45) Thennarasu, S.; Lee, D. K.; Tan, A.; Prasad, K. U.; Ramamoorthy, A. Biochim. Biophys. Acta 2005, 1711, 49–58. (46) Raghuraman, H.; Chattopadhyay, A. Biophys. J. 2007, 92, 1271–1283.

JP806938Y