8452
Langmuir 2006, 22, 8452-8457
Phlorizin- and 6-Ketocholestanol-Mediated Antagonistic Modulation of Alamethicin Activity in Phospholipid Planar Membranes Tudor Luchian* and Loredana Mereuta Department of Biophysics and Medical Physics, Faculty of Physics, Alexandru I. Cuza UniVersity, BouleVard Carol I, no 11, Iasi, Romania, R-6600 ReceiVed May 16, 2006. In Final Form: August 1, 2006 As a result of the interfacial chemical heterogenity, membrane-penetrating peptides will experience a dramatic variation in environmental polarity manifested via electrical interactions with the surface and dipole potential of membranes prone to modulate the membrane insertion and folding of different peptides and proteins. Herein we present evidence demonstrating that roughly a 30 mV, phlorizin-induced lowering of the magnitude of the dipole potential of a phosphatidyilcholine membrane leads to a 4-fold increase in the electrical activity of embedded alamethicin. The effect is voltage-independent, implying that the dipole potential affects the barrier of alamethicin adsorption to the membrane rather than the translocation of it across the hydrophobic core. Our interpretation points to an enhanced interfacial accumulation of alamethicin monomers on the cis side of the membrane caused by a lower value of the cis dipole potential, which will promote an elevated activity of alamethicin oligomers across the membrane. As expected for a modestly selective ion channel, the enhancing effect of such dipole potential changes on the electrical conductivity is limited (80 ( 3 pS before and 100 ( 2 pS after phlorizin addition to the membrane, for the first conductive state of the channel). Our study emphasizes the possibility that, by manipulating at will the sign of change and the magnitude of the interfacial dipole field, it is possible to modulate the extent of the membrane penetration of ion-channel-forming peptides and thereby provide deeper insights into mechanisms of protein-lipid and proteinprotein interactions within membranes.
Introduction Regardless of its biochemical composition, the overall electrical profile of a biomembrane combines contributions from the transmembrane potential, dipole potentials, and surface potentials at both sides of a membrane.1,2 Of these, the membrane dipole potential has received particular attention mainly due to the extremely high electric field associated with it over the interface between the aqueous phase and the hydrocarbon region of a biomembrane (108-109 V‚m-1). A large amount of both structural and functional data have unraveled the two main factors that underlie the origin of the dipole potential: the orientation of dipolar groups located on the lipid molecule (i.e., the dipole of the carbonyl group of the ester bond and the P--N+ dipole of the headgroup) and the dipoles of oriented water molecules at the membrane-water interface.3-5 Dipole potential has often been cited as being among other factors that have powerful influences on membrane-protein interactions, including protein insertion and functioning,6 kinetics of the gramicidin channel,7 modulation of the activity of phospholipase A2,8 and electrical conductance of certain aqueous protein pores.9 Recent data have demonstrated that the emission intensity of a widely used fluorescent moiety, 7-nitro-2,1,3-benzoxadiazol4-yl (NBD) labeling covalently either the headgroup of DPPC * Corresponding author. Tel: +040232 201191. Fax: +040232 201151. E-mail:
[email protected]. (1) Franklin, J. C.; Cafiso, D. S. Biophys. J. 1993, 65, 289. (2) Cevc, G. Biochim. Biophys. Acta 1990, 1031, 311. (3) Zheng, C.; Vanderkooi, G. Biophys. J. 1992, 63, 935. (4) Gawrisch, K.; Ruston, D.; Zimmerberg, J.; Parsegian, A.; Rand, R. P.; Fuller, N. Biophys. J. 1992, 61, 1213. (5) McLaughlin, S. Annu. ReV. Biophys. Biophys. Chem. 1989, 18, 113. (6) White, S. H.; Ladokhin, A. S.; Jayasinghe, S.; Hristova, K. J. Biol. Chem. 2001, 276, 32395. (7) Rokitskaya, T. L.; Antonenko, Y. N.; Kotova, E. A. Biophys. J. 1997, 73, 850. (8) Maggio, B. J. Lipid Res. 1999, 40, 930. (9) Tatyana, I.; Rokitskaya, T. Y.; Kotova, E. A.; Antonenko, Y. N. Biophys. J. 2002, 82, 865.
lipids (DPPN) or the acyl chains of PC lipids (NBD-PC) aggregated as liposomes, is sensitive to the membrane dipole potential and that the rate of the reduction of NDB in these probes by dithionite can be controlled by the dipole potential.10 Moreover, dipole potential was shown to affect the membrane insertion and folding of a model amphiphilic peptide11 as well as the extent of the membrane fusion.12 In a similar line of reasoning and knowing that an essential feature of membrane penetration by various membrane toxins13,14 and fusion peptides15 is changing from a water-soluble state to a membrane-associated state, lipid contribution and variation of the dipole potential may critically affect the state of peptide aggregation within lipid membranes and have profound implications for their biological activity. Bearing in mind the relevance of the membrane dipole potential on cellular function, it has been established that by manipulating its value, the influence of the dipole potential on protein-protein interactions may be explored.16 It was even possible to modulate the extent of the penetration of peptides within membranes and the manner in which they adopt different secondary structures.17 Membrane insertion of such peptides and proteins requires a refolding process that is poorly understood at the molecular level, but the membrane interfacial region (IF) is expected to play a critical role in the refolding event, as described previously.18,19 (10) Alakoskela, J. M. I.; Kinnunen, P. K. J. Biophys. J. 2001, 80, 294. (11) O′Shea, P. Biochem. Soc. Trans. 2003, 31 (5), 990. (12) Cladera, J.; Martin, I.; Ruysschaert, J. M.; O’Shea, P. J. Biol. Chem. 1999, 274, 29951. (13) Oh, K. J.; Senzel, L.; Collier, R. J.; Finkelstein, A. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 8467. (14) Gouaux, J. E.; Braha, O.; Hobaugh, M. R.; Song, L.; Cheley, S.; Shustak, C.; Bayley, H. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 12828. (15) Martin, I.; Dubois, M. C.; Defrise-Quertain, F.; Saermark, T.; Burny, A.; Brasseur, R.; Ruysschaert, J.-M. J. Virol. 1994, 68, 1139. (16) Shapovalov, V.; Kotova, E. A.; Rokitskaya, T. I.; Antonenko, Y. N. Biophys. J. 1999, 77, 299. (17) Cladera, J.; O’Shea, P. Biophys. J. 1998, 74, 2434.
10.1021/la0613777 CCC: $33.50 © 2006 American Chemical Society Published on Web 09/01/2006
Alamethicin-Membrane Interactions
In previous work and in agreement with the hypothesis that the membrane dipole potential modulates ion transport through model ion channels, the effects of interfacial dipole potential (IDP) agents on gramicidin channel lifetimes and proton conductance have been demonstrated. Specifically, phloretin and phlorizin were seen to enhance gramicidin A conductance, whereas RH421 and 6-ketocholestanol had the opposite effect upon gramicidin A conductance.20-22 Interestingly, for the gramicidin A analogue in which the four Trp residues were replaced by Phe (gramicidin M), phloretin was noted to enhance more pronounced the channel’s conductance as compared to the gramicidin A case.22 This was an expected phenomenon since the removal of such amphiphatic residues, initially positioned in such a way as to screen the ion diffusion pathway from the dipole potential field, rendered the channel more vulnerable to IDP-reducing agents. By virtue of its structural simplicity, alamethicin forms a suitable channel-forming peptide model for grasping lipidpeptide interactions, aggregation behavior within membranes, and ion channel properties. Its transport and structural properties have been extensively studied,23-26 and recent data regarding alamethicin binding and insertion interactions with lipid bilayers are available, both from experimental27 and modeling viewpoints.28,29 In short, after alamethicin addition to the cis side of the bilayer and its partitioning to the membrane (the bilayerwater partition coefficient is about 105 at an aqueous alamethicin concentration of 3 × 10-8 M), membrane conductance quickly develops in the positive applied potentials range along a steep, exponential branch once a concentration-dependent voltage threshold has been reached. One of the most accepted models of alamethicin channel formation is the so-called dynamic “barrel-stave” model. It essentially postulates that in the process of membrane adsorption, during which alamethicin monomers pass vectorially (the partially positive N-terminus head on) across the highly electrified interfacial domain where the membrane dipole field rules electrical interactions, alamethicin monomers become mostly helical with the N-terminal part slightly buried in the hydrophobic core. An applied positive potential on the alamethicin addition side repels the N-terminus and helps the peptide cross the bilayer, facilitating the state of membrane-bound alamethicin molecules (barrelstaves), which via lateral diffusion and collision with other transmembrane monomers lead to the formation of conducting bundles. The macroscopic alamethicin conductance in planar membranes is well-described by the empirical relation G ) ΓCn exp(∆V/Ve), where C stands for the concentration of the alamethicin, ∆V is the applied voltage, Ve represents the potential difference imposed across the membrane that leads to an e-change in the conductance (Ve ∼ 5 mV), and Γ is a parameter that varies with (18) White, S. H.; Wimley, W. C. Annu. ReV. Biophys. Biomol. Struct. 1999, 28, 319. (19) White, S. H.; Ladokhin, A. S.; Jayasinghe, S.; Hristova, K. J. Biol. Chem. 2001, 276, 32395. (20) Rokitskaya, T. I.; Atonenko, Y. N.; Kotova, E. A. Biophys. J. 1997, 73, 850. (21) Rokitskaya, T. I.; Kotova, E. A.; Antonenko, Y. N. Biophys. J. 2002, 82, 865. (22) Duffin, R. L.; Garrett, M. P.; Flake, K. B.; Durrant, J. D.; Busath, D. D. Langmuir 2003, 19, 1439. (23) Woolley, G. A.; Wallace, B. A. J. Membr. Biol. 1992, 129, 109. (24) Cafiso, D. S. Annu. ReV. Biophys. Biomol. Struct. 1994, 23, 141. (25) Sansom, M. S. P. Q. ReV. Biophys. 1993, 26, 365. (26) Duclohier, H.; Wroblewski, H. J. Membr. Biol. 2001, 184, 1. (27) Lewis, J. R.; Cafiso, D. S. Biochemistry 1999, 38, 5932. (28) Kessel, A.; Cafiso, D. S.; Ben-Tal, N. Biophys. J. 2000, 78, 571. (29) Tieleman, D. P.; Berendsen, H. J. C.; Sansom, M. S. P. Biophys. J. 1999, 76, 1757.
Langmuir, Vol. 22, No. 20, 2006 8453
experimental conditions (e.g., lipid type, ionic strength).24 As stressed by this formula and with relevance to our work, the dependence of the macroscopic conductance of the peptide concentration is rather high and is described by n (n ∼8-12). In this work we considered one important paradigm regarding the interactions manifested between the dipolar electric field of phospholipid membranes and alamethicin oligomers from prospective changes imposed by the membrane upon kinetic and transport features of such model ion channels. To this end, we posit that a lower electric dipole field of the interfacial region of the membrane provides a much reduced repelling influence upon the positively charged N-terminus of the alamethicin peptides as they partition from the aqueous medium to lipid membranes. By employing phlorizin as an agent of choice that selectively lowers the magnitude of the dipole field only on the interface that is added to, we prove that the actual energy barriers for alamethicin insertion become significantly smaller, leading to a 4-fold increase in the activity of ion-conducting oligomers across the membrane. Furthermore, we addressed the complementary issue of whether an elevated magnitude of the dipole field leads to a decrease in the activity of alamethicin oligomers. Such experiments were carried out on asymmetrical artificial membranes containing 6-ketocholestanol (KC), known to increase the membrane dipole potential.31 Expectedly, when one lipid monolayer contained 50% (w/w) KC, the relative interfacial concentration of alamethicin monomers moving into it was considerably reduced. This was manifested as virtually no alamethicin-induced activity at an applied potential of -70 mV. Experimental Section Electrophysiology on alamethicin oligomers was performed on the folded bilayer membranes system, obtained as previously described.32 In short, a 25 µm thick Teflon septum was clamped between two Teflon chambers each of 1 mL volume. A lipid bilayer was formed on an aperture of ∼100 µm diameter in the septum that had been pretreated with 10% (v/v) hexadecane (Sigma-Aldrich) in highly purified n-pentane (Sigma-Aldrich). Both chambers contained 1 M NaCl and 10 mM sodium phosphate, and the pH value was set to 5.1. Initially, the level of electrolyte was set just below the aperture, and 1% (w/v) L-R-phosphatidylcholine (Fluka, code 61755, from egg yolk) in pentane (6 µL) was spread onto the surface of each chamber. After solvent evaporation, the electrolyte level in the chambers was raised above the aperture. The formation of a bilayer was monitored by observing the increase in capacitance to a value of approximately 90-130 pF. Alamethicin monomers (Sigma-Aldrich) were added from a stock solution made in ethanol (5 µM) to the cis chamber only, which was connected to the ground. Mechanical stirring was initiated in this chamber for ∼1 min to ensure proper concentration homogenization. Volumes added were usually of 2 µL as to provide a 10 nM final concentration of alamethicin. Phlorizin (Fluka) was added to the cis side of the membrane from an 80 mM stock solution made in ethanol. Most importantly, before adding phlorizin to the bilayer chamber, we waited long enough (∼4-5 min under stirring) to allow alamethicin molecules to reach the stationary state with respect to their partitioning within the lipid membrane. In this way, we sought to avoid a subsequent increase in the alamethicin activity upon phlorizin addition, simply as a result of re-homogenization of alamethicin monomers within the cis monolayer. Furthermore, to make sure that under our working conditions the time-lag between adding alamethicin and phlorizin was enough to avoid this problem, (30) Borisenko, V.; Sansom, M. S. P.; Woolley, G. A. Biophys. J. 2000, 78, 1335. (31) Simon, S. A.; McIntosh, T. J.; Magid, A. D.; Needham, D. Biophys. J. 1992, 61, 786. (32) Luchian, T.; Shin, S. H.; Bayley, H. Angew. Chem., Int. Ed. 2003, 42, 3766.
8454 Langmuir, Vol. 22, No. 20, 2006 control tests were run during which we monitored whether the activity of alamethicin oligomers within the membrane changes visibly upon simply stirring of the solution. Only in those instances where the activity of alamethicin remained largely unchanged following such “ghost-stirring” in the cis chamber did we proceed with phlorizin addition. All experiments were performed at a room temperature of ∼25 °C. When KC (Sigma-Aldrich) was used, we dissolved it in n-pentane along with L-R-phosphatidylcholine at a relative concentration of 50% (w/w) with respect to the total mass achieved. We did this in the presence of very small amounts of ethanol (
