The lipid composition of a cell membrane modulates the interaction of an antiparasitic peptide at the air-water interface

(1)

The lipid composition of a cell membrane modulates the interaction of an

antiparasitic peptide at the air

–

water interface

Rondinelli D. Herculano

a

_{, Felippe J. Pavinatto}

b

_{, Luciano Caseli}

c,

_⁎

_{, Claudius D'Silva}

d

_{, Osvaldo N. Oliveira Jr.}

b

a_{Faculdade de Ciências e Letras de Assis, Universidade Estadual Paulista, Assis, SP, Brazil} b_{Instituto de Física de São Carlos, Universidade de São Paulo, São Carlos, SP, Brazil}

c_{Instituto de Ciências Ambientais, Químicas e Farmacêuticas, Universidade Federal de São Paulo, Diadema, SP, Brazil} d_{School of Biology, Chemistry & Health Science, Manchester Metropolitan University, Manchester, UK}

a b s t r a c t

a r t i c l e

i n f o

Article history:

Received 3 February 2011

Received in revised form 16 March 2011 Accepted 22 March 2011

Available online 5 April 2011

Keywords:

Cell membrane model Langmuir monolayer Raft

Lipophilic peptide African Sleeping Sickness

The antiparasitic property of peptides is believed to be associated with their interactions with the protozoan membrane, which calls for research on the identification of membrane sites capable of peptide binding. In this study we investigated the interaction of a lipophilic glutathioine peptide known to be effective against theAfrican Sleeping Sickness (ASS — African Trypanosomiasis) and cell membrane models represented by Langmuir monolayers. It is shown that even small amounts of the peptide affect the monolayers of some phospholipids and other lipids, which points to a significant interaction. The latter did not depend on the electrical charge of the monolayer-forming molecules but the peptide action was particularly distinctive for cholesterol + sphingomyelin monolayers that roughly resemble rafts on a cell membrane. Usingin situpolarization-modulated infrared reflection absorption spectroscopy (PM-IRRAS), we found that the orientation of the peptide is affected by the phospholipids and dioctadecyldimethylammonium bromide (DODAB), but not in monolayers comprising cholesterol + sphingomyelin. In this mixed monolayer resembling rafts, the peptide still interacts and has some induced order, probably because the peptide molecules arefitted together into a compact monolayer. Therefore, the lipid composition of the monolayer modulates the interaction with the lipophilic glutathioine peptide, and this may have important implications in understanding how the peptide acts on specific sites of the protozoan membrane.

1. Introduction

The synthesis and identiﬁcation of small peptides to serve as drugs for neglected diseases has received increased attention[1], as have antiparasitic peptides [2–4] shown to act against the ﬂagellated protozoan,t. rhodescience,the causal agent ofAfrican Sleeping Sickness (ASS —African Trypanosomiasis) [5]. The latter peptides are being investigated as an alternative to the traditional therapeutic

treat-ments using arsenical compounds [6], and have shown enhanced

trypanocidal activity due to esteriﬁcation increasing hydrophobicity/ surface activity and enhancing protozoan, cell membrane disruption. In a recent paper, we showed that antiparasitic peptides were amenable to form stable Langmuir monolayers at the air/water interface, whose surface pressure isotherms displayed a region of negative compression modulus, in theﬁrst compression–decompression cycle [7]. Upon combining information from BAM and PM-IRRAS

measure-ments, we concluded that this negative elasticity was ascribed to aggregation during compression caused by intermolecular associations. Also in ref.[7], we noted that the peptide induced expansion in the surface pressure isotherms of dipalmitoyl phosphatidyl choline (DPPC) monolayers used as a cell membrane model.

In this paper, we investigate the possible generality of action of these antiparasitic glutathioine peptides on various types of mono-layer models that may mimic cell membrane systems. In particular, we used zwitterionic phospholipids, such as dipalmitoyl phosphatidyl-ethanolamine (DPPE), in addition to the negatively charged dipalmi-toyl phosphatidic acid (DPPA) and dipalmidipalmi-toyl phosphatidyl glycerol (DPPG), the positively charged dioctadecyldimethylammonium bro-mide (DODAB), and lipids known to form rafts in cell membranes (sphingomyelin and cholesterol). The motivation for using the rafts is that they are relevant for various cell membrane phenomena[8], such as signaling and cell transport. The methods employed to analyze the monolayers were surface pressure isotherms and polarization-modulated infrared reﬂection-absorption spectroscopy (PM-IRRAS). With the experiments reported here it is possible to extend the analysis in ref.[7]and show that the conformation of the peptide at the air/water interface depends on several factors, including its relative concentration and the lipid composition.

⁎Corresponding author at: Rua Prof. Artur Riedel, 275-09972-270-Diadema, SP, Brazil. Tel.: +55 11 50843759; fax: +55 11 40436428.

doi:10.1016/j.bbamem.2011.03.012

Contents lists available atScienceDirect

Biochimica et Biophysica Acta

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a m e m

Open access under the Elsevier OA license.

(2)

2. Materials and methods

The compound S-(2,4-dinitrophenyl)glutathioine di-2-propyl ester, whose chemical structure is shown inFig. 1, was synthesized as described in ref.[2]. The lipids DPPC, DPPE, DPPA, DPPG, DODAB, cholesterol and sphingomyelin were acquired from Avanti Polar Lipids and used without further puriﬁcation. For preparing Langmuir monolayers, the lipids and the peptide were dissolved in chloroform (Merck) at a concentration of 0.5 mg/mL. The solutions were spread on an aqueous subphase with pH 7.5 (consisting of 0.1 mol L−1NaCl (Merck), and 0.001 mol L−1phosphate buffer (Na2HPO4:NaH2PO4=1:1, Merck)). Peptide–lipid mixed mono-layers were obtained by spreading the drug-containing solution after the lipid monolayer had been formed, but at a large area per molecule (at zero surface pressure). After 10–15 min to allow for solvent evaporation, the air–water interface was compressed with two movable barriers at a rate of 25 cm2_min−1. Areas per molecule were calculated assuming that all drug molecules remained at the interface. For mixed drug-lipid monolayers, use was made of the area per lipid molecule (assuming the molecule was alone at the interface), in order to better evaluate the changes in the lipid monolayers induced by the small amounts of the drug (2% in mol or less). These experiments were performed in a NIMA trough (model 601 M, subphase volume: 500 mL).

Polarization-Modulated Infrared Reflection-Absorption Spectroscopy (PM-IRRAS) was performed using a KSV PMI 550 instrument (KSV, Biolin Scientific Oy, Helsinki, Finland). The experimental setup was similar to that described by Blaudez and co-workers[9]. The Langmuir trough (mini KSV) was mounted so that the light beam reached the monolayer at afixed incidence angle of 80°, for which the upward-oriented bands indicate a transition moment preferentially on the surface plane, whereas downward-oriented bands indicate preferential orientation perpendicular to the surface.

All the experiments were carried out in a class 10,000 clean room at the temperature of 23.0± 0.2 °C.

3. Results and discussion

The surface pressure-area isotherm for the pure peptide spread at the air–water interface was reported in[7], showing a decrease in

pressure upon compression between 70 and 60 Å2_{. This negative}

compressibility modulus was associated with domain formation with increased lateral ﬂuidity according to Brewster Angle Microscopy images[7]. The possible reorganization of the peptide molecules at the interface was discarded on the basis of the PM-IRRAS data, which supported the hypothesis of a pre-collapse of the monolayer[7].

The set of surface pressure isotherms for the monolayers with mixed peptide–lipid for 2 mol% of the peptide are shown inFigs. 2–5. Overall, there is a tendency of the isotherms to expansion at low surface pressures and in some cases a small condensation in the high surface pressures (for DPPE and DPPG). Therefore, the peptide appears to be incorporated in the monolayer at low densities, and be expelled from the monolayer at high pressures, in some cases even causing the neighboring lipids to be more densely packed. Distinctive results were

obtained for DPPA, which was more expanded than the other phospholipids (see Fig. 3B), and for cholesterol and cholesterol/ sphingomyelin. For cholesterol, a large condensation was observed at the condensed phase, but this did not occur when cholesterol was mixed with sphingomyelin (seeFig. 5). The condensation induced by cholesterol is consistent with reports in the literature[10], and is explained by cholesterol's ability to interact with hydrophobic chains [11,12]. We shall resume the discussion of the effects of adding sphingomyelin later in the paper. As for the results for DODAB, a positively charged lipid, the similarity with the data for the zwitterionic and negatively charged phospholipids means that the charge of the monolayer-forming molecules does not appear to exert a great inﬂuence on monolayer properties. This is not surprising since the esteriﬁed antiparasitic peptide is not charged and should not be expected to participate in any type of ionic interaction.

A general feature of the isotherms is that small amounts of the peptide are sufﬁcient to affect the monolayer, which points to interaction effects caused by the peptide–lipid interaction as already observed with other peptides[13,14]. This is illustrated for DPPA in Fig. 6, while the data for the other lipids are omitted. Signiﬁcantly, the monolayer condensation at high surface pressures for some lipids disappears as the amount of peptide is increased. The disappearance may be caused by the increased area occupied by the drug itself as higher proportions are used in the mixed monolayers.

Fig. 1.Chemical structure of the peptide used in this work.

40 60 80 100 120 140 0

10 20 30 40 50 60 70

Surface Pressure (mN/m)

Area per lipid (Å2₎

40 60 80 100

DPPC DPPC + drug

DPPE DPPE + drug

A

0 10 20 30 40 50 60 70

B

(3)

In order to better visualize the effects of the peptide on the DPPA monolayer, we compared the theoretical and experimental areas per molecule based on the surface pressure-area for each compound in the mixed monolayer, for a pressure of 30 mN/m (which is believed to correspond to the pressure in a real membrane[15]). For obtaining the theoretical areas, we used the following expression:

A=x₁A₁+x₂A₂

where x1 and x2 are the molar fractions of components 1 and 2, respectively, and A1and A2are the molecular area of each component

inferred from the isotherms of neat components. Fig. 6B shows

positive excess areas, which indicates expansion of the monolayer due to repulsive interactions or loose molecular accommodation at the surface. We calculated the excess area for all mixed drug-lipid monolayers, and the behavior is the same, pointing again to repulsive interactions between the peptide and the lipid.

Fig. 7shows the PM-IRRAS spectrum for a Langmuir monolayer of the pure peptide, which features inverted amide I bands at 1577 cm−1 since the C=O groups lie parallel to the air–water interface, as reported in[7]. The direction of the band is preserved upon compression, thus indicating no reorientation. The band at 1677 cm−1 is assigned to disordered structures, which is reasonable considering the lack of secondary structure of the small peptide. The amide II band (C–N vibration) appears at 1521 cm−1.

60 80 100 120 140 160 0

10 20 30 40 50 60 70

Area per lipid (Å2)

DPPG DPPG + drug

DPPA DPPA + drug

A

30 35 40 45 50 55 60 65 70 75 80 85 90 0

10 20 30 40 50 60 70 80

B

Fig. 3.Surface pressure-area isotherm for negatively charged lipids, and mixed peptide (2% in mol)-lipid monolayer. For the x-axis,“area per lipid”means that only lipid molecules were taken into account in the calculation.

60 80 100 120 140 160 0

10 20 30 40 50 60 70

DODAB DODAB + drug

Fig. 4.Surface pressure-area isotherm for a positively charged lipid (DODAB), and mixed peptide (2% in mol)-lipid monolayer. For the x-axis, the area per lipid means that only lipid molecules were taken into account in the calculation.

30 35 40 45 50 55 60 0

10 20 30 40 50 60

Cholesterol Cholesterol + drug

A

80 120 160 200 240 0

8 16 24 32 40 48

sphingomyelin + cholesterol (1:1 mol) sphingomyelin + cholesterol (1:1 mol) + drug

B

Mean Molecular Area (Å2)

(4)

Consistent with the surface pressure results, the PM-IRRAS inFig. 8 shows practically the same behavior for the mixed monolayers containing 2 mol% of the peptide and DPPC, DPPG or DODAB (the data for DPPE and DPPA were omitted because they are similar). Again, the charge state of the phospholipid has no effect on the interaction of the monolayer with the peptide. The bands appearing in theﬁgure are assigned to amide bands. Now, the amide I band is directed upward, in contrast to the downward direction of the monolayer of the neat peptide. Therefore, the dipole moment of the amide vibration was reoriented from a parallel to a perpendicular position in relation to the air–water interface. This is in contrast with theﬁnding in our previous paper[7], for which the peptide appeared not to be reoriented by DPPC. The discrepancy may be explained by the fact that in the previous paper the PM-IRRAS measurement was performed for a mixed monolayer containing 10 mol% of the peptide, and therefore the peptide molecules in excess maintained their geometrical orientation as in a pure monolayer. Also, the amide I band has been split in two peaks at 1640 and 1690 cm−1. The

ﬁrst one results from the re-arrangements of the peptide upon interaction with the lipid molecules, while that at 1690 cm−1represents the disordered amide I band. The latter is shifted to higher energies upon interaction with the lipid, which means that accommodation of the peptide into the lipid membrane leads to stable freedom for the C=O vibration moment. The amide II also appears shifted, at ca. 1555 cm−1, owing to interaction of C–N vibration with side chains[16].

Signiﬁcant differences were observed in the spectrum for the peptide-containing cholesterol in the monolayer, as shown inFig. 8B,

15 20 25 30 35 40 45 0

10 20 30 40 50 60 70 80

Surface Pressue (mN/m)

DPPA

DPPA + drug 0.25% DPPA + drug 0.50% DPPA + drug 0.75% DPPA + drug 1.00% DPPA + drug 1.25% DPPA + drug 2.00%

A

0.0 0.5 1.0 1.5 2.0 36

38 40 42 44

drug fraction %

Ideal Area Experimental Area

B

Mean Molecular Area (Å2₎

Molecular area (Å

2)

Fig. 6.Surface pressure-area isotherm for DPPA and mixed peptide (% in mol indicated in the insert)-DPPA monolayer. Panel B shows the comparison between the average molecular areas of the mixed monolayer at 30 mN/m obtained experimentally and theoretically. The areas from both the lipid and peptide were taken into account. The excess area may be calculated subtracting the experimental from theoretical values.

1750 1700 1650 1600 1550 1500 1450

PM-IRRAS signal (a.u.)

wavenumber (cm-1₎

1677

1521

Fig. 7.PM-IRRAS spectrum for the peptide monolayer at 0 mN/m. At higher surface pressures, the bands have their intensity increased with no signiﬁcant change in the wavenumber position, in agreement with ref.[7].

1800 1700 1600 1500 1400 1527

DODAB DPPC DPPG 1690 1640

1555

A

1800 1700 1600 1500 1400

wavenumber (cm-1₎

Cholesterol Raft 1688

1625 1665

B

(5)

where the amide I band is not deconvoluted. It appears as a single band at 1688 cm−1with a shoulder at ca. 1650 cm−1, pointing to less unfolding of the peptide when interacting with cholesterol. It is likely that interaction with cholesterol makes it easier for the peptide to be incorporated in the lipid membrane with further accommodation. Also, for the monolayer simulating roughly a“raft”(sphingomyelin + cholesterol), the spectra show even more signiﬁcant changes: the amide I band is downward, with the main peak assigned to disordered structures shifted to lower energies (1665 cm−1). Also, a band at 1625 cm−1due to turns for amide bonds appears. This means that the peptide is not reoriented when incorporated in the“raft”monolayer, in contrast to the other types of monolayer. Nevertheless, there is still peptide–lipid interaction leading to ordered structures. This is

associated with the peptide molecules beingﬁtted into a compact monolayer, which does not allow for peptide reorientation in the membrane.

Overall, the most important result–considering the surface pressure isotherms and PM-IRRAS spectra–is associated with the fact that a speciﬁc composition of the monolayer, namely that prepared with components roughly simulating rafts, provides signiﬁcantly different results from the other model membrane systems. Not only the orientation of the peptide at the air/water interface was affected but also the folding of the peptide was changed leading to the existence of turns. A scheme depicting the conformation of the peptide is given in Fig. 9, which depends on the lipid composition of the monolayer, on the relative peptide concentration and on the surface pressure.

(6)

4. Conclusions

The behavior of the lipophilic glutathioine peptide at the air/water interface could be investigated in further detail–in comparison to our

previous work [7] – by using various types of monolayers and

different characterization methods. The main conclusions may be summarized as follows. The lipophilic glutathioine peptide is able to affect Langmuir monolayers even at low concentrations at the air/ water interface, thus indicating that interactive effects may be present. The effects induced on the monolayers do not depend on the charge of the monolayer-forming molecules for the phospholipids and DODAB. Particularly important effects were noted, however, for cholesterol and cholesterol + sphingomyelin monolayers. In choles-terol monolayers, significant condensation was caused by the peptide, but this did not occur for the mixed cholesterol + sphingomyelin monolayer. The latter resembles roughly a raft on cell membranes, and therefore it appears that the most significant action by the peptide should be associated with its interaction with the rafts. Indeed, while the orientation of the peptide is affected by the phospholipids and DODAB, with the carbonyl group changing from perpendicular to the interface for a neat peptide monolayer to a parallel orientation for the mixed lipid-peptide monolayers, in the rafts (i.e. cholesterol + sphingomyelin) the orientation of the peptide is preserved. The implications of thesefindings arise from the fact that the composition of the lipid monolayer modulates the geometry of the

peptide incorporated into ﬁlm. This may help to understand the

mechanism by which this drug acts on speciﬁc sites of the protozoan membrane, with molecular-level detail of how the peptide accom-modates into the membrane model in terms of conformation and lateral orientation. The data therefore provide a basis for speculating

on effects on the protozoan membrane induced by peptide–lipid

interactions, which may be a key factor for the antiparasitic property reported for this peptide.

Acknowledgements

This work was supported by FAPESP, CNPq, CAPES and rede nBioNet (Brazil). Dr. C. D'Silva thanks Prof. D.E. Games and Dr. B. Stein

of the EPSRC Mass Spectrometry Service Centre, Swansea, for MS measurements.

References

[1] M.C. Chung, E.I. Ferreira, J.L. Santos, J. Girarola, D.G. Rando, A.E. Almeida, P.L. Bosquesi, R.F. Menegon, L. Blau, Prodrugs for the treatment of neglected diseases, Molecules 13 (2008) 616–677.

[2] C. D'Silva, S. Daunes, P. Rock, V. Yardley, S.L. Croft, Structure-activity study on the in vitro antiprotozoal activity of glutathione derivatives, J. Med. Chem. 43 (2000) 2072–2078.

[3] S. Daunes, C. D'Silva, H. Kendrick, V. Yardely, S.L. Croft, QSAR study on the contribution of log P and Esto the in vitro antiprotozoal activity of glutathione derivative, J. Med. Chem. 44 (2001) 2976–2983.

[4] S. Daunes, C. D'Silva, Glutathione derivatives active against trypanosoma brucei rhodesiense andT. brucei bruceiin vitro, Antimicrobial Agents Chemotherapy 46 (2002) 434–437.

[5] J.D.M. Mhlanga, M. Bentivoglio, K. Kristensson, Neurobiology of cerebral malaria and African sleeping sickness, Brain Res. Bull. 44 (1997) 579–589.

[6] A.H. Fairlamb, Chemotherary of human African trypanosomiais: current and future prospects, Trends Parasitol. 19 (2003) 488–494.

[7] C. Paschollati, E.P. Lopera, F.J. Pavinatto, L. Caseli, T.M. Nobre, M.E.D. Zaniquelli, T. Viitala, C. D'Silva, O.N. Oliveira Jr., The interaction of an antiparasitic peptide active against African sleeping sickness with cell membrane models, Colloids Surf. B 74 (2009) 504–510.

[8] M. Edinin, The state of lipid RAFTS: from model membranes to cells, Ann Rev. Biophys. Biomolec. Struct. 32 (2003) 257–283.

[9] D. Blaudez, T. Buffeteau, J.C. Cornut, B. Desbat, N. Escafre, M. Pezolet, J.M. Turlet, Polarization-modulated FT–IR spectroscopy of a spread monolayer at the air/ water interface, Appl. Spectrosc. 47 (1993) 869–874.

[10] D. Chapman, N.F. Owens, M.C. Phillips, D.A. Walker, Mixed monolayers of phospholipids and cholesterol, Biochim. Biophys. Acta 183 (1969) 458–465. [11] C. Ohe, T. Sakati, M. Noi, Y. Goto, K. Itoh, Sum frequency generation spectroscopy

study of the condensation effect of cholesterol on a lipid monolayer, Anal. Bioanal. Chem. 388 (2007) 73–79.

[12] Y.H. Kim, R. Nero, M. Takizawa, T. Urisu, Characterization of dipalmitoylpho-sphatidylcholine/cholesterol Langmuir–Blodgett monolayers investigated by atomic force microscopy and Fourier transform infrared spectroscopy, Jpn. J. Appl. Phys. 43 (2004) 3860–3864.

[13] M.L. Moraes, C. Bonardi, C.R. Mendonça, P.T. Campana, J. Lottersberger, G. Tonarelli, O.N. Oliveira Jr., L.M. Beltramini, Cooperative effects in phospholipid monolayers induced by a peptide from HIV-1 capsid protein, Colloids Surf. B 42 (2005) 15–20.

[14] M.R. Lourenzoni, A.M. Namba, L. Caseli, L. Degrève, M.E.D. Zaniquelli, Study of interactions of human defensins with cell membrane models: relationships with their structure and biological activity, J. Phys. Chem. B 111 (2007) 11318–11329. [15] D. Marsh, Lateral pressure in membranes, Biochim. Biophys. Acta 1286 (1996)

183–223.