pubs.acs.org/Langmuir © 2009 American Chemical Society
Polymer Migration among Phospholipid Liposomes Dmitry A. Davydov,† Ekaterina G. Yaroslavova,† Anna A. Rakhnyanskaya,† Anna A. Efimova,† Yury A. Ermakov,‡ Fredric M. Menger,*,§ and Alexander A. Yaroslavov*,† †
Department of Chemistry, M.V. Lomonosov Moscow State University, Leninskie Gory, 1-3, 119992 Moscow, Russian Federation, ‡A.N. Frumkin Institute of Physical Chemistry and Electrochemistry, RAS, Leninsky av., 31, 119991 Moscow, Russian Federation, and §Department of Chemistry, Emory University, Atlanta, Georgia 30322 Received June 5, 2009. Revised Manuscript Received July 2, 2009 Complexation of phospholipid lipsomes with a cationic polymer, poly(N-ethyl-4-vinylpyridinium bromide) (PEVP), and subsequent interliposomal migration of the adsorbed macromolecules, have been investigated. Liposomes of two different charge types were examined: (a) a liposomal system, with an overall charge near zero, consisting of zwitterionic phosphatidylcholine (egg lecithin, EL) with added doubly anionic phospholipid, cardiolipin (CL2-), and cationic dihexadecyldimethylammonium bromide (HMAB+), in a CL2-/HMAB+ charge-to-charge ratio of 1:1; (b) an anionic liposomal system composed of an EL/CL2- mixture plus polyoxyethylene monocetyl ether (Brij 58). Both threecomponent systems were designed specifically to preclude liposomal aggregation upon electrostatic association with the PEVP, a phenomenon that had complicated analysis of data from several two-component liposomes. PEVP macromolecules were found from fluorescence experiments to migrate among the charge-neutral EL/CL2-/HMAB+ liposomes. In the case of anionic EL/CL2-/Brij liposomes, a combination of fluorescence and laser microelectrophoresis methods showed that PEVP macromolecules travel from liposome to liposome while being electrostatically associated with anionic lipids.
Introduction Spherical bilayer lipid vesicles (liposomes) were introduced by Bangham about 50 years ago1 and, since then, have been examined for different biomedical applications. On the one hand, liposomal containers are used for delivery of biologically active compounds;2,3 on the other, liposomes are intensively exploited for the studying the structure and function of biological (cell) membranes.4-7 Persistent interest of researchers in liposomes derives in part from the biocompatibility of lipids, the variety of procedures for liposome preparation, and the ability to control lipid composition and size. Additionally, mixed vesicles composed of lipids and proteins (proteoliposomes) can be easily prepared. Synthetic approaches, experimental details, and properties of liposomes have been summarized in reviews.4,5 Liposomes as cell mimetics have been used for investigating the mechanism of cell interactions with synthetic water-soluble polymers including polyelectrolytes (PEs). Different aspects of liposome/polymer interactions have come under scrutiny: composition and structure of interfacial complexes;8,9 polymer*Corresponding authors. E-mail:
[email protected] (F.M.M.), yaroslav@ genebee.msu.ru (A.A.Y.). (1) Bangham, A. D. Adv. Lipid Res. 1963, 1, 205. (2) Bangham, A. D.; Standish, M. M.; Weissmann, G. J. Mol. Biol. 1965, 8, 253. (3) Drummond, D. C.; Meyer, O.; Hong, K. L.; Kirpotin, D. B.; Papahandjopoulos, D. Pharmacol. Rev. 1999, 51, 691. (4) Menger, F. M.; Chlebowski, M. E.; Galloway, A. L.; Lu, H.; Seredyuk, V. A.; Sorrells, J. L.; Zhang, H. Langmuir 2005, 21, 10336. (5) Lasic, D. D. Liposomes: From Physics to Applications; Elsevier: Amsterdam, 1993. (6) Lipowsky R.; Sackmann, E. Handbook of Biological Physics. Physical Basis of Self-Organization and Function of Membranes: Physics of Vesicles; Elsevier: Amsterdam, 1995. (7) Banerjee, P.; Joo, J. B.; Buse, J. T.; Dawson, G. Chem. Phys. Lipids 1995, 77, 65. (8) Lasch, J. Biochim. Biophys. Acta 1995, 1241, 269. (9) Volodkin, D.; Balla, V.; Schaafb, P.; Voegela, J.-C.; Mohwaldc, H. Biochim. Biophys. Acta 2007, 1768, 280.
13528 DOI: 10.1021/la902031e
induced structural rearrangements in cell and liposomal membranes;10 conformational transitions in adsorbed polymers;11 effect of polymers on membrane permeability;12 and aggregation, fusion, and disruption of cells and liposomes in the presence of polymers.13,14 It was shown that properties of liposome/PE complexes were mainly determined by the chemical composition, degree of polymerization, and linear charge density of macromolecules.15 A principal question concerns the reversibility of liposometo-polyelectrolyte complexation. It is of key importance for the theory of liposome/PE complexation and for the design and development of PE-based drug delivery systems. One way to enhance the affinity of a polymeric drug toward target cells is to modify polymer chains with small vector molecules capable of specific binding to corresponding receptors on the cell surface.16 Typically, a PE-based drug, being introduced into a biological fluid, electrostatically interacts with random cells but then, after exploring a host of different partners, fixes onto the target cell surface.17,18 In other words, binding of polymeric drugs to cells is, ideally, a reversible process that ultimately establishes a thermodynamically optimum drug/target cell contact. (10) Melik-Nubarov, N.; Krylova, O. In Advances in Planar Lipid Bilayers and Liposomes; Ottova-Leitmannova, A., Ed.; Elsevier: Amsterdam, 2005; Vol. 2, p 122. (11) Schwieger, C.; Blume, A. Eur. Biophys J 2007, 36, 437. (12) Volodkin, D.; Mohwald, H.; Voegel, J.-C.; Ball, V. J. Controlled Release 2007, 117, 111. (13) Bordi, F.; Cametti, C.; Sennato, S.; Diociaiuti, M. Biophys. J. 2006, 91, 1513. (14) Walter, A.; Steer, G. I.; Blumenthal, R. Biochim. Biophys. Acta 1986, 861, 319. (15) Kabanov, V. A.; Yaroslavov, A. A. J. Controlled Release 2002, 78, 267. (16) Park, J. H.; Lee, S.; Kim, J.-H.; Park, K.; Kim, K.; Kwon, I. C. Prog. Polym. Sci. 2008, 33, 113. (17) Lee, K. Y.; Yuk, S. H. Prog. Polym. Sci. 2007, 32, 669. (18) Peng, S.-F.; Yang, M.-J.; Su, C.-J.; Chen, H.-L.; Lee, P.-W.; Wei, M.-C.; Sung, H.-W. Biomaterials 2009, 30, 1797.
Published on Web 07/30/2009
Langmuir 2009, 25(23), 13528–13533
Davydov et al.
Article
An ability of polyelectrolytes to migrate among microsized particles was first demonstrated in experiments with suspensions of monodispersed polystyrene latex.19 The latex particles were covered by carboxylic groups that ensured the negative surface charge in neutral and slightly alkaline solutions similar to the cell surface. A cationic polymer, poly(N-ethyl-4-vinylpyridinium bromide) (PEVP), was adsorbed on the surface of latex particles. The adsorption was extremely effective: all added PEVP coupled with particles until a dense polymer monolayer was formed on the particle surface. Moreover, despite strong complexation, PEVP macromolecules were able to pass from one particle to another, resulting finally in a uniform distribution of macromolecules among all the particles in the suspension.20 The above work refers to migration of a cationic polymer among latex particles. Extending the results to interactions between polycations and cell membranes is not at all obvious considering the different morphologies of the two adsorbents, namely latex particles and cells. In the former case, there is a solid surface with fixed negative charges whose positions cannot change upon polycation adsorption. In the latter, negative lipid molecules are free to move within the cell membrane and thereby respond to polycation adsorption.21 This can affect the properties of polymer/lipid complexes and the intercellular migration of polymers. A suspension of liposomes seems to provide a more suitable basis for studying polymer migration among cells. Composition of the liposomal membrane and its size and surface charge can be easily adjusted using standard procedures.22 In the present paper, we investigate an interliposomal migration of cationic PEVP macromolecules in a suspension of small unilamellar liposomes. Since conventional two-component liposomes, composed of anionic and zwitterionic lipids, were shown to aggregate when binding to polycations,23 they were excluded from the migration experiments. Instead, two types of three-component liposomes were used (see Figure 1 for structures): (a) a liposomal system, with an overall charge near zero, consisting of a zwitterionic phosphatidylcholine (egg lecithin, EL) with added doubly anionic phospholipid, cardiolipin (CL2-), and cationic di-n-hexadecyldimethylammonium bromide (HMAB+), in a CL2-/HMAB+ charge-to-charge ratio of 1:1; (b) an anionic liposomal system composed of an EL/CL2- mixture plus polyoxyethylene monocetyl ether (Brij 58). Since both three-component liposomes, electroneutral and anionic, were stable with regard to their integrity and lack of aggregation when brought into contact with a cationic polymer,24,25 they were suitable for our study of interliposomal PEVP migration.
Experimental Section Materials. PEVP was synthesized by quaternization of poly(4-vinylpyridine) with excess ethyl bromide.26 Poly(4-vinylpyridine) from Aldrich with a degree of polymerization equal to 600 was used throughout. Fractionation using a conventional (19) Kochwa, S.; Brownell, M.; Rosenfield, R. E.; Wasserman, L. R. J. Immunol. 1967, 99, 981. (20) Kabanov, V. A.; Yaroslavov, A. A.; Sukhishvili, S. A. J. Controlled Release 1996, 39, 173. (21) Kabanov, V. A.; Yaroslavov, A. A. J. Controlled Release 2002, 78, 267. (22) Lasic, D. D. Biochem. J. 1989, 258, 935. (23) Kozlova, N. O.; Bruskovskaya, I. B.; Okuneva, I. B.; Melik-Nubarov, N. S.; Yaroslavov, A. A.; Kabanov, V. A.; Menger, F. M. Biochim. Biophys. Acta 2001, 1514, 139. (24) Oku, N.; Namba, Y.; Okada, S. Biochim. Biophys. Acta 1992, 1126, 255. (25) Yaroslavov, A. A.; Udalykh, O. Yu.; Melik-Nubarov, N. S.; Kabanov, V. A.; Ermakov, Yu. A.; Azov, V. A.; Menger, F. M. Chem.;Eur. J. 2001, 7, 4835. (26) Kirsh, Yu. E.; Rachnyanskaya, A. A.; Lukovkin, H. M.; Kabanov, V. A. Eur. Polym. J. 1974, 10, 393.
Langmuir 2009, 25(23), 13528–13533
Figure 1. Structures of lipids and surfactants used in this study. sequential sedimentation procedure gave a material with an estimated polydispersity index of 1.2-1.4. Thus, the reported polymer properties reflect a weighted average among a modest range of polymer sizes. Product was actually a copolymer containing 95 mol % of ethyl-quaternized pyridinium rings and 5 mol % of residual 4-vinylpyridine units as found by IR spectroscopy while measuring the ratio of intensities at 1600 and 1640 cm-1.27 PEVP concentration is given in moles of quaternized units per liter. Zwitterionic EL, doubly anionic CL2-, and 1,2-dioleoyl-snglycero-3-phosphoethanolamine-N-(carboxyfluorescein) (DOPECF, see structure in Figure 1) from Avanti, HMAB+ from Sigma, Brij 58 from Serva, and 5(6)carboxyfluorescein (CF) from Fluka were used as received. Small three-component unilamellar liposomes of EL/CL2-/ HMAB+ and EL/CL2-/Brij, as well as two-component EL/CL2liposomes (a control), were prepared by the following procedure. The appropriate amounts of chloroform solutions of lipids and surfactants were mixed in a flask, and the solvent was evaporated under vacuum. A thin film of the lipid or lipid/surfactant mixture was dispersed in a 10-2 M borate buffer, pH 9.2. The resulting dispersions were extruded (30 times) through polycarbonate filters with 100 nm pores at 38 °C for EL/CL2-/Brij and EL/ CL2- mixtures and 60 °C for EL/CL2-/HMAB+ mixture using an Avanti mini-extruder. To prepare liposomes with a fluorescent tag incorporated into the liposomal membrane, 0.1 mol % of DOPE-CF was added to the lipid or lipid/surfactant mixtures and then treated as described above. To prepare CF-loaded liposomes, we followed the procedure described in ref 28. The lipid film was suspended and extruded in a 10-2 M borate buffer solution in the presence of CF. The liposome suspension was then separated from the excess of CF by passing it through a column with Sepharose CL-4B. The integrity of CFloaded liposomes was affirmed by measuring the fluorescence intensity of the liposome suspensions. Headgroup mol % (using a charge adjustment of two for CL2and one for the other lipids) of the various liposome systems are the following: (a) EL/CL2-/HMABþ in a ratio of 60:20:20 (neutral); (b) EL/CL2-/HMABþ/DOPE-CF in a ratio of 59.9:20:20:0.1 (neutral); (c) EL/CL2-/Brij 58 in a ratio of 58:20:22 (anionic); (27) Starodubtzev, S. G.; Kirsh, Yu. E.; Kabanov, V. A. Eur. Polym. J. 1977, 10, 739. (28) Subbarao, N. K.; MacDonald, R. I.; Takeshita, K.; MacDonald, R. C. Biochim. Biophys. Acta 1991, 1063, 147.
DOI: 10.1021/la902031e
13529
Article
Figure 2. Effect of PEVP on EPM of EL/CL2-/HMABþ (1), EL/
CL2-/Brij (2), and EL/CL2- liposomes (3). 2[CL2-] = [HMABþ]= 2.8 10-4 M. In the migration experiment, PEVP was complexed with EL/CL2-/Brij liposomes (point A), an equal amount of EL/ CL2-/Brij liposomes was then added to the complex suspension (point B).
(d) EL/CL2-/Brij 58/DOPE-CF in a ratio of 57.9:20:22:0.1 (anionic). Methods. Mean hydrodynamic diameter (D) and electrophoretic mobility (EPM) of liposomes and their complexes with PEVP were determined using a Zeta Plus Instrument (Brookhaven Instruments) with Bi-MAS software and a solidstate laser (15 mW, 670 nm). Fluorescence intensities of liposomes with DOPE-CF incorporated into the lipid bilayer and liposomes with CF in the inner water cavity were measured at λem=525 nm (λex=495 nm) using a F-4000 fluorescence spectrophotometer (Hitachi). The pH measurements were carried out using a pH-meter pH-210 (Hanna) with a combined electrode HI 1131B. To prepare solutions, we used double-distilled water additionally treated by passing through a Milli-Q system (Millipore) equipped with ion-exchange and adsorption columns and a filter to remove large particles. All experiments were carried out at 25 ( 0.2 °C.
Results and Discussion To begin with, consult Figure 1 for the structures of all the “acronyms”, and the last paragraph of the Materials section for the constant ratios of the liposome components, used in this paper. Now addition of PEVP to electroneutral liposomes of EL/ CL2-/HMABþ and anionic liposomes of EL/CL2-/Brij will be considered from three points of view: formation of liposome/ PEVP complexes; their stability in aqueous salt media; and, finally, interliposomal migration of adsorbed polymer. For clarity, it is worthwhile to describe laser electrophoretic mobility (EPM) used in our experiments. In Figure 2, we plot the mobility of liposomes in an electric field as a function of the cationic polymer concentration. If PEVP sticks to anionic liposomes, zero mobility (the most interesting point on the plot!) will be achieved when there is charge neutralization (i.e., the number of negative charges in the outer leaflet of the liposome equals the number of positive charges on the polymer). Now consider the three types of liposomes studied in Figure 2. (1) The EPM of EL/CL2-/ HMABþ in plot 1 begins with zero mobility of the neutral vesicles and gets only mildly positive as polymer binds to the outer leaflets of the bilayers. (2) The EPM of EL/CL2-/Brij and EL/CL2- in plots 2 and 3 correspond to liposomes that are initially negatively charged but become increasingly less negative as PEVP is added. At about 2.8 10-4 M PEVP (with 2[CL2-] =2.5 10-4 M), 13530 DOI: 10.1021/la902031e
Davydov et al.
Figure 3. Changes in the relative fluorescence intensity of labeled EL/CL2-/HMABþ (1) and EL/CL2-/Brij liposomes (2). 2[CL2-]= [HMABþ]=2.8 10-4 M.
mobility of the charge-neutralized liposomes in the electric field ceases (point A). When an equal amount of EL/CL2-/Brij was added to the sample at point A, the EPM reverted to point B (at a [PEVP] one-half of that in point A), as would be expected if the PEVP redistributed rapidly among the liposomes. More will be said of this experiment later in the paper. The positive charge imparted by PEVP to the liposomes accounts for the observation, supported by dynamic light scattering (DLS), that the liposomes do not complicate matters by aggregating. The liposomes merely increase in diameter from 75 ( 5 to 95 ( 5 nm upon addition of PEVP, a size increase reflecting adsorption of the PEVP macromolecules upon the liposome outer surface. Electrostatic polymer/liposome contact should be sensitive to salt in the bulk solution. A fluorescence approach, based on the fact that PEVP is a good fluorescent quencher,29 was used to examine salt effects. Thus, when PEVP was added to EL/CL2-/ HMABþ liposomes containing a carboxyfluorescein-labeled lipid, the CF fluorescence diminished precipitously (Figure 3, curve 1). Injection of an NaCl solution into the liposome/PEVP complex caused the fluorescence intensity to recover to its initial level (Figure 4, curve 1), indicating a complete shielding of the liposome and PEVP charges by >0.1 M salt. Such sensitivity of complexation to salt definitely shows the electrostatic nature of the liposome/polymer binding. The basic idea of the migration experiment already alluded to briefly is simple: a liposome/PEVP complex is mixed with PEVPfree liposomes, and after a certain period of time the composition of resulting particles in the system is determined by any suitable method. The principal requirement here is that the polycation must be totally bound to the liposomes. Adsorption of unbound PEVP onto the surface of newly injected liposomes is thereby avoided. Efficiency of PEVP binding to EL/CL2-/HMABþ liposomes was quantified using the fluorescence approach. As follows from the data of Figure 3 (plot 1), the fluorescence intensity of CF-labeled EL/CL2-/HMABþ liposomes decreases linearly with rising PEVP concentration until [PEVP] reaches 2.5 10-5 M at which point the plot levels out sharply. This suggests that in the [PEVP] e 2.510-5 M range there is little or no excess PEVP that remains unbound in solution. To perform the migration experiment, a 2.5 10-5 M PEVP solution was added to the CF-labeled liposome suspension to (29) Izumrudov, V. A.; Savitskii, A. P.; Bakeev, K. N.; Zezin, A. B.; Kabanov, V. A. Makromol. Chem., Rapid Commun. 1984, 5, 709.
Langmuir 2009, 25(23), 13528–13533
Davydov et al.
Figure 4. Changes in the relative fluorescence intensity of PEVP complexes with labeled EL/CL2-/HMABþ (1) and EL/CL2-/Brij liposomes (2) after NaCl addition. 2[CL2-]=[HMABþ]=2.810-4 M; [PEVP]=0.2510-4 (1) and 1.510-4 M (2).
Article
corresponding to individual liposomes covered by PEVP macromolecules. The fluorescence and light scattering data provide information about migration of PEVP macromolecules from the labeled to unlabeled three-component liposomes that resulted in the polycation redistributing itself among all the liposomes in the system. Since no aggregation of liposomes was observed, PEVP macromolecules likely passed from one individual liposome to another during liposome/liposome collision. Coincidence of both experimental curves in Figure 4 indicates an equilibrium state was attained in the course of this PEVP redistribution. How did PEVP macromolecules distribute themselves after reaching equilibrium in the labeled/unlabeled liposome population? In other words, did they distribute equally among labeled and unlabeled liposomes, or did they perhaps concentrate preferentially on the CF-labeled liposomes? Linearity of the fluorescence intensity of CF-labeled liposomes with respect to [PEVP] within the [PEVP] e 2.5 10-5 M range was discussed above (Figure 3, curve 1). On the basis of this dependence, the final fluorescence in the PEVP-exchanged liposome system was easily calculated from eq 1 for any L/L* ratio assuming equal affinities of the labeled and unlabeled liposomes toward PEVP: Ical ¼ Icomp þð1 -Icomp Þ L=ðLþLÞ
Figure 5. Relative fluorescence intensity in the PEVP-labeled/unlabeled EL/CL2-/HMABþ liposome ternary suspension vs the ratio of labeled/unlabeled liposome concentration. PEVP was complexed with labeled liposomes and then unlabeled liposomes were added (1); PEVP was added to a labeled/unlabeled liposome mixture (2), theoretical calculation according to eq 1 (3). [PEVP]= 0.25 10-4 M, 2[CL2-]=2.810-4 M for labeled liposomes.
achieve the maximum decrease in the CF fluorescence. A suspension of liposomes with the same lipid composition, but deprived of the tag, was then added to the complex suspension, and after 10 min the fluorescence intensity in the system was measured. The dependence of the observed fluorescence (Iexp) on the ratio of concentrations of unlabeled and labeled liposomes (L/L*) is given in Figure 5 (curve 1). Each experimental point on the curve was obtained by adding a corresponding amount of unlabeled liposome to a fresh sample of the labeled liposome/PEVP complex suspension. It is seen that the fluorescence intensity progressively increases with increasing L/L* ratio. This is because the unlabeled liposomes withdraw polymer quencher from the labeled liposomes. Additionally, we added PEVP to a mixture of the unlabeled and labeled liposomes, the L/L* ratios being maintained equal to those in the migration experiment. The fluorescence intensities, measured after 10 min (Figure 5, curve 2), coincided with the data of curve 1. Importantly, the mean diameter of particles for both mixing procedures, measured by dynamic light scattering, fluctuated between 80 and 85 nm, i.e., Langmuir 2009, 25(23), 13528–13533
ð1Þ
where Icomp is the fluorescence intensity of the complex formed by mixing CF-labeled liposomes and a 2.5 10-5 M PEVP solution and L and L* are concentrations of the unlabeled and labeled liposomes, respectively. The results are represented in Figure 3, curve 3. An agreement between the experimental and calculated data in Figure 5 points to an equally strong binding of PEVP to both types of liposomes. This means, in turn, that modifying the surface of EL/CL2-/HMABþ liposomes by the fluorescent tag did not change significantly their affinity toward the cationic polymer. The redistribution of PEVP macromolecules in the mixed labeled/unlabeled EL/CL2-/HMABþ liposome suspension is schematically represented in Figure 6. One way to stabilize liposomes against aggregation is to form a hydrophilic layer on the liposome surface. This can be done, for example, by covalent modification of the liposomal membrane by polyoxyethylene (POE) chains. It has been shown that anionic liposomes with a POE corona still manage to bind cationic polymers but do not aggregate even at a complete neutralization of their surface charges by adsorbed polymers.30 With this fact in mind, we added to EL/CL2- liposomes some nonionic polyoxyethylene monocetyl ether (Brij 58). Hydrophobic hexadecyl chains of Brij incorporated into the lipid bilayers of liposomes, while polyoxyethylene chains were exposed to water to form an external hydrophilic corona. A mol % of anionic CL2- headgroups was chosen to be ν(-) = 2[CL2-]/([EL]þ 2[CL2-] þ [Brij])=20%, while a mol % of Brij was chosen to be ν(Brij)= [Brij]/([EL] þ 2[CL2-] þ [Brij])=22%. Binding of PEVP to the anionic EL/CL2-/Brij liposomes was assayed electrophoretically (Figure 2, curve 2). For comparison, in the same figure is shown the effect of PEVP on the EPM of twocomponent anionic 8:2 EL/CL2- liposomes lacking the hydrophilic POE corona (Figure 2, curve 3). In both cases, the EPM of particles in suspensions decreased down to zero as the negative charge on the liposomes was totally neutralized by adsorbed PEVP. The overall surface charge became positive when the charge supplied by the PEVP was in excess over that of the (30) Lasic, D. D.; Woodle, M. C.; Papahadjopoulos, D. J. Liposome Res. 1992, 2, 335.
DOI: 10.1021/la902031e
13531
Article
Davydov et al.
Figure 6. Redistribution of PEVP macromolecules in the mixed labeled/unlabeled EL/CL2-/HMABþ liposome suspension (schematical presentation).
Figure 8. Loading of EL/CL2-/Brij liposomes with CF. Molar content of Brij: 0 (1), 0.07 (2), 0.15 (3), and 0.22 (4). Figure 7. Relative fluorescence intensity in the PEVP-labeled/unlabeled EL/CL2-/Brij liposome ternary suspension vs the ratio of labeled/unlabeled liposome concentration. PEVP was complexed with labeled liposomes and then unlabeled liposomes were added (1); PEVP was added to a labeled/unlabeled liposome mixture (2). [PEVP] = 0.25 10-4 M, 2[CL2-] = 2.8 10-4 M for labeled liposomes.
CL2-. Importantly, both POE-covered and conventional EL/ CL2- liposomes became electroneutral at 2.8 10-4 M PEVP. A complete binding of PEVP to anionic 8:2 EL/CL2- liposomes up to [PEVP] = 2.8 10-4 M was shown earlier.31 This means, therefore, that PEVP quantitatively binds to “hydrophilized” anionic EL/CL2-/Brij liposomes at [PEVP] e 2.8 10-4 M. As in the case of three-component electroneutral liposomes, no aggregation was found when PEVP complexed with EL/CL2-/ Brij liposomes, and that was in agreement with the earlier published data.32 The protective hydrophilic POE layer was a decisive factor that ensured the stability of EL/CL2-/Brij liposomes in the presence of PEVP. Injection of PEVP solution into the suspension of CF-labeled EL/CL2-/Brij liposomes was accompanied by a decrease in the tag fluorescence intensity (Figure 3, curve 2) which, however, recovered to the initial level as a NaCl solution was added (Figure 4, curve 2). An electrostatic mechanism for PEVP binding to the hydrophilized liposome was thus proved. In the migration experiment, a 2.8 10-4 M PEVP solution was added to EL/CL2-/Brij liposomes so that a liposome/PEVP complex with EPM close to zero was formed (point A in Figure 2). Then the same amount of EL/CL2-/Brij liposomes was injected in the complex suspension, and in 10 min particle EPM was (31) Sybachin, A. V.; Efimova, A. A.; Litmanovich, E. A.; Menger, F. M.; Yaroslavov, A. A. Langmuir 2007, 23, 10034. (32) Davydov, D. A.; Yaroslavova, E. G.; Efimova, A. A.; Yaroslavov, A. A. Colloid J. 2009, 71, 55.
13532 DOI: 10.1021/la902031e
measured. The resulting EPM value (point B in Figure 1) was intermediate between the EPM values for the liposome-PEVP complex and free liposomes, and no aggregation of liposomes was detectable. This also supplies evidence of PEVP macromolecule migration among individual EL/CL2-/Brij liposomes. The fluorescence method was applied to the study of the PEVP migration in the suspension of hydrophilized liposomes where the procedure followed that described above for the PEVP/electroneutral liposome system. Briefly, a 1.510-4 M PEVP solution was added to the CF-labeled EL/CL2-/Brij liposome suspension that led to complexation and a decrease in the CF fluorescence. A suspension of unlabeled EL/CL2-/Brij liposomes was then added, but this time no fluorescence recovery was detected in curve 1 (Figure 7). Next, the experiment was repeated except that PEVP was injected into a mixed unlabeled/labeled liposome suspension. Curve 2 (Figure 7), reflecting an absence of fluorescence dependence on the L/L* ratio, is in substantial agreement with the curve 1 (Figure 7). As deduced from the dynamic light scattering data, a mean diameter of particles at all L/L* ratios fluctuated around 70 nm for both experiments. In other words, single liposomes with adsorbed PEVP were found throughout. At first thought, the fluorescence results seem to indicate no PEVP migration from the labeled EL/CL2-/Brij liposomes to the corresponding unlabeled liposomes. Yet if this were the case, the fluorescence results would be in contradiction with the abovedescribed electrophoretic results that definitely prove PEVP migration. What could be a reason for such a contradiction? An invariant fluorescence intensity in the course of the migration experiment might conceivably arise from a higher affinity of PEVP toward the labeled liposomes in comparison with the unlabeled. It could be, for example, that strong specific interactions exist between quaternized pyridinium rings of the macromolecules and the fluorophore incorporated into the liposomal Langmuir 2009, 25(23), 13528–13533
Davydov et al.
Article
Figure 9. Redistribution of PEVP macromolecules in the mixed labeled/unlabeled EL/CL2-/Brij liposome suspension (schematical presentation).
membranes. Specific interactions, e.g., charge-transfer complex formation, were discussed for systems composed of PEVP and fluorescent labeled polyacrylate anion.29 However, we did not observe any specific binding of PEVP to the CF-labeled EL/ CL2-/HMABþ liposomes; cationic macromolecules passed freely from labeled to unlabeled liposomes (see above), indicating that electrostatic, but not specific, interactions governed the interliposomal PEVP migration. Therefore, a higher affinity of PEVP for the CF-labeled EL/CL2-/Brij liposomes seems to be a rather unlikely rationale for the data. Another model has the PEVP macromolecules migrating among the hydrophilized liposomes. Since PEVP remains complexed with CF-modified lipids, the fluorescence is not altered. It is known that liposomes, especially with membranes in the liquidcrystalline state (“liquid” liposomes), are able to exchange lipid molecules. A disordering in the lipid packing, e.g., due to incorporation of lipids with larger headgroups, favors interliposomal lipid exchange.33 It is reasonable to presume that incorporating Brij, a one-tailed surfactant with an extended POE headgroup, into the liquid-crystalline EL/CL2- bilayer causes a decrease in the density of lipid packing, thereby elevating the rate of spontaneous lipid exchange. Anionic DOPE-CF lipids, as well as anionic unlabeled CL2-, might complex with PEVP due to electrostatic interactions. It is not unlikely, therefore, that PEVP chains exchange EL/CL2-/Brij liposomes after having extracted labeled and unlabeled anionic lipids from the liposomes. In this manner, there is PEVP migration (Figure 2) without any increase in fluorescence (Figure 7). In order to confirm this model, the following experiment was carried out. We prepared a series of EL/CL2-/Brij liposomes, with Brij varying from 7% up to 22% mol %, each containing within them the simple CF dye at a 8 10-3 M self-quenching concentration. After removing free CF with gel-permeation chromatography, we mixed a suspension of the liposomes containing encapsulated CF dye with a 10-fold excess of Triton X-100. This treatment resulted in a complete destruction of the
liposomes that discharges the trapped CF into the surrounding solution. An increase in the total fluorescence intensity would be observed because the CF dye is diluted in the bulk water to a point where self-quenching no longer persisted. Such an increase was observed when carrying out the experiment with the CF-loaded, two-component EL/CL2- liposomes (taken as 100% in Figure 8, column 1). But the results for EL/CL2-/Brij liposomes with differing Brij content (Figure 8, columns 2-4) show that none of the liposomes released any CF during their destruction. Thus, the relative fluorescence was lower than 1% for all Brij-containing liposomes. The simplest explanation is that CF had diffused out of the liposomes prior to their destruction, indicating a high permeability of the membranes toward the low molecular weight compound (CF). A low density of lipid packing, and high permeability, can be ascribed to the extended hydrophilic headgroups of Brij molecules. It is this same loose packing that allows PEVP to remove anionic lipid before traveling from liposome to liposome. The mechanism is reminiscent of a previous study in which another polycation, poly-L-lysine, migrates through a lipid bilayer by first coating itself with anionic lipid to make it compatible with the bilayer interior.34 Thus, polycation/anionic lipid association is a known migratory process. The mechanism of PEVP migration is schematically represented in Figure 9. In conclusion, the behavior of a cationic polymer bound to both charge-neutral and anionic liposomes has been quantitatively assessed by photon correlation spectroscopy, electrophoretic mobility, and fluorescence. Polymer was shown to migrate among both types of liposomes although with anionic liposomes the polymer travels from liposome to liposome while being electrostatically associated with anionic lipids.
(33) Nakano, M.; Fukuda, M.; Kudo, T.; Endo, H.; Handa, T. Phys. Rev. Lett. 2007, 98, 238101.
(34) Menger, F. M.; Seredyuk, V. A.; Kitaeva, M. V.; Yaroslavov, A. A.; MelikNubarov, N. S. J. Am. Chem. Soc. 2003, 125, 2846.
Langmuir 2009, 25(23), 13528–13533
Acknowledgment. This work was supported by the Russian Foundation of Basic Research (Grant 08-03-00744), the Fogarty International Research Cooperation Award (Grant TW05555), and the National Institutes of Health (FMM).
DOI: 10.1021/la902031e
13533
