Article pubs.acs.org/Langmuir
Effect of Boundary Edge in DOPC/DPPC/Cholesterol Liposomes on Acceleration of L‑Histidine Preferential Adsorption Takaaki Ishigami, Atsushi Tauchi, Keishi Suga, and Hiroshi Umakoshi* Division of Chemical Engineering, Graduate School of Engineering Science, Osaka University, 1-3 Machikaneyama-cho, Toyonaka, Osaka 560-8531, Japan S Supporting Information *
ABSTRACT: In order to investigate the interaction of hydrophilic molecules with liposomal membranes, we employed 1-(4-(trimethylamino)phenyl)-6-phenyl-1,3,5-hexatriene and 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(5-dimethylamino-1-naphthalenesulfonyl) as fluorescent probes to monitor the surface regions of the membrane, and the results for various liposomes were plotted in correlation diagrams. According to the formation of a variety of phase states, different tendencies of decreasing surface hydrophobicity were observed in the liposomes that were modified with high concentrations of cholesterol or in the liposomes that were composed of ternary components. These liposomes, with hydrophobic surfaces, also showed preferential adsorption of L-histidine (L-His), and the hydrophobicity of the liposomal membrane at the surface changed during L-His adsorption regardless of the initial liposomal properties. Furthermore, we revealed that accelerated adsorption of L-His and preferential binding was induced in ternary liposomes forming boundaries between two separate phases.
■
INTRODUCTION Liposomes are self-assembled vesicles composed of phospholipids, the membranes of which may include other amphiphilic molecules such as cholesterol, sphingolipids, etc. Liposomal membranes exhibit several phase states (i.e., liquid-crystalline and gel phases) induced by the balance of lateral diffusion and hydrophobic interactions of phospholipids. Liposomal membranes also display localized regions of low fluidity, such as microdomains or “rafts”,1−4 that play significant roles in the regulation of protein activity in cell membranes.5−7 The heterogeneous properties of the liposomal surface, such as microdomain formation and microphase separation, can be analyzed by direct observation of giant unilamellar vesicles,8,9 thermodynamic procedures,10,11 and Raman and nuclear magnetic resonance spectroscopies.12,13 Methods using fluorescent probes have been developed to evaluate liposomal membrane properties by means of variations in characteristics of fluorescent wavelengths or anisotropy relative to the surrounding environment.14−16 These methods enable the analysis of local membrane properties based on the localized fluorescent probes, since the hydrophobic region of membranes could be predicted based on their hydrophobic environment.17 Furthermore, several studies have investigated the depth of several fluorescent probes by the quenching of fluorescent moieties in synthetic lipids.18−20 Additionally, microdomain formation in liposomal membranes can be analyzed by excimer/monomer ratios of pyrene probes or their derivatives at nanometer scale.21,22 In our recent study, the © XXXX American Chemical Society
phase state in homogeneous and heterogeneous liposomes was evaluated by combining two membrane properties, specifically fluidity using 1,6-diphenyl-1,3,5-hexatriene (DPH) and polarity using 6-lauroyl-2-dimethylaminonaphthalene (Laurdan).23 Although this method employed hydrophobic molecules for analysis of both membrane properties, further information focusing on the characteristics of the hydrophilic−hydrophobic interface of the membrane should be obtained to understand the behavior of guest molecules that could interact with the membrane platform. The surface property of liposomal membranes plays an important role in the interaction of hydrophilic and hydrophobic molecules.24−26 In the adsorption of arginine,27 propranolol,28 and single-stranded DNA29 on liposomes, the strength of the interactions was highly dependent upon the surface net charge of the liposomes. Additionally, Schwieger et al. reported that the adsorption of poly-L-lysine induced an increased phase-transition temperature in anionic liposomes due to electrostatic interactions at the hydrophilic interface.30 Moreover, it has been shown that the inclusion of cholesterol (Ch) in liposomes results in the promotion of aggregation of an α-helical transmembrane peptide31 and strong binding or partial insertion of peptides containing simple sequence repeats (lysine−tryptophan), with the selectivity determined by the Received: December 18, 2015 Revised: May 8, 2016
A
DOI: 10.1021/acs.langmuir.5b04626 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir length of the peptides.32 It is notable that our recent studies revealed the specific adsorption of L-enantiomeric amino acids on phosphatidylcholine liposomes,33,34 indicating stereochemical recognition at the liposomal membrane surface. Based on previous findings, liposomal membranes are regarded as platforms for chemical reactions that control the surrounding environment of adsorbed guest molecules: oligomerization of LHis on liposomal membranes,35 enhancement of reaction rate and enantioselectivity for a Michael addition reaction by the adsorption of L-Pro catalyst on liposomes,36 and the promotion of 1,3-dipolar cycloaddition conducted in liposomal membranes.37 Furthermore, the enhancement of adsorption of amino acids on liposomal membranes is worthwhile for the development of optical resolution of amino acids by liposomeimmobilized hydrogel devices.38 From a biological perspective, investigation of preferential binding of the L-enantiomer of His may contribute to our understanding of specific interactions between biological membranes and amino acids.39 Information concerning the molecules on and/or in the hydrophobic region of liposomal membranes is important, and investigations of the mechanism of the adsorption/conversion of guest molecules on the liposomal membrane are needed in order to link properties to membrane phenomena. The importance of the location and stability of aromatic amino acids at the hydrophobic region of liposomal membrane surfaces was proposed by simulating the molecular dynamics of phospholipid membranes.40 Additionally, a variety of surface properties of liposomal membranes (i.e., fluidity and polarity) were evaluated by using several fluorescent probes, which have been utilized for understanding details of binding behavior in addition to dielectric spectra or thermodynamic experiments. Although the regulation of membrane properties may contribute to the design of guest-molecule adsorption on liposomal membranes, mixing Ch with lipid membranes is effective for controlling membrane properties, as demonstrated by previous studies.30,31,41−44 It was also revealed that increasing the Ch ratio in liposomes affected phase-transition temperatures and the formation of the heterogeneous phase, which contributed to a better understanding of the variation in surface properties associated with Ch-modified liposomal surfaces. In this study, we evaluated the properties of membrane surfaces in binary and ternary liposomes by using surfacelocalized fluorescent probes, such as 1,2-dioleoyl-sn-glycero-3phosphoethanolamine-N-(5-dimethylamino-1-naphthalenesulfonyl) (Dansyl-DHPE) and 1-(4-(trimethylamino)phenyl)6-phenyl-1,3,5-hexatriene (TMA-DPH) (Figure S1). The effect of Ch on the liposomal membrane properties was analyzed by correlation diagrams obtained using hydrophilic probes and by a Cartesian diagram obtained using hydrophobic probes. On the basis of our results, we associated the membrane properties with the adsorption of amino acids in order to investigate the possibility of designing membrane surfaces capable of inducing molecular recognition through control of membrane physicochemical properties.
■
were purchased from Peptide Institute (Suita, Osaka, Japan). Ch and other chemicals were purchased from Wako Pure Chemical Industry Ltd. (Osaka, Japan) and were used without further purification. Liposome Preparation. Liposomes were prepared based on the reported methods.45 Briefly, a solution of phospholipids in chloroform was dried in a round-bottom flask by rotary evaporation under vacuum. The lipid thin film was kept under high vacuum for at least 3 h and then hydrated with ultrapure water at room temperature. The vesicle suspension was frozen at −80 °C and then thawed at 50 °C. This freeze−thaw cycle was repeated five times. The liposome suspensions were extruded 11 times through two layers of polycarbonate membrane with mean pore diameters of 100 nm, at the above of phase transition temperature using an extruding device (Liposofast; Avestin Inc., Ottawa, Canada). Evaluation of Membrane Properties by Fluorescent Probes. Membrane property of liposomes could be characterized by Cartesian diagram, which is the plot of the membrane fluidity versus polarity evaluated by hydrophobic molecular probes (DPH and Laurdan) as the previous studies.16,23,46 The fluidity in the hydrophobic region of the liposome membrane was evaluated by measuring the fluorescence polarization of the DPH incorporated in the vesicles using the fluorescence spectrophotometer FP-6500 (JASCO, Tokyo, Japan). A sample of 10 μL of 100 μM DPH dissolved in ethanol was mixed in a liposome suspension in final concentrations of lipid and DPH that were 250 and 1.0 μM, respectively. The samples were excited with vertically polarized light (360 nm), and emission intensities both perpendicular (I⊥) (0°, 0°) and parallel (I∥) (0°, 90°) to the excited light were recorded at 430 nm. The polarization (P) of DPH (PDPH) was then calculated by using the following equations:16
P = (I − GI⊥)/(I + GI⊥)
(1)
G = i⊥/i
(2)
where i⊥ and i∥ are the emission intensities perpendicular to the horizontally polarized light (90°, 0°) and parallel to the horizontally polarized light (90°, 90°) and G is the correction factor. The membrane fluidities (1/PDPH) were evaluated based on the reciprocal of PDPH. The membrane fluidities were measured at room temperature. The polarity in hydrophobic region of the liposome membrane was evaluated by measuring the fluorescence emission of Laurdan incorporated in the vesicles. Laurdan emission spectra exhibit a redshift caused by dielectric relaxation. Thus, emission spectra were produced by measuring the general polarization (GP340) for each emission wavelength as follows:47,48
GP340 = (I440 − I490)/(I440 + I490)
(3)
Laurdan was excited with 340 nm light at 25 °C. The fluorescent spectrum of each sample was normalized. The final concentrations of amphiphilic phospholipid and Laurdan in the examined solution were 1000 and 10 μM, respectively. Besides, the property of membrane surface could be characterized by a similar diagram analyzed by partly hydrophilic molecular probes. TMA-DPH can characterize the membrane fluidity (1/PTMA‑DPH) of hydrophilic region by analyzing polarization by the same method of DPH.49,50 As for polarity of membrane surface, Dansyl-DHPE was used as a probe molecule, mixed in a liposome suspension in final concentrations of lipid and Dansyl-DHPE that were 100 and 1.0 μM, respectively. The fluorescence spectra were analyzed by the excitation light (336 nm) for observing the emission peak wavelengths (λ). The hydrophobicity of the membrane surface can be evaluated by the normalized value (λN) by using the equation
EXPERIMENTAL SECTION
λN = (λ − λ 0)/(λ1 − λ 0)
Materials. Phospholipids such as 1,2-dipalmitoyl-sn-glycero-3phosphocholine (DPPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), and Dansyl-DHPE were purchased from Avanti Polar Lipids, Inc. (Alabaster, AL). DPH, TMA-DPH, and Laurdan were purchased from Sigma-Aldrich (St. Louis, MO). The L- and D-forms of histidine (His)
(4)
where λ0 and λ1 represent the maximum wavelengths in hydrophilic (527 nm) and hydrophobic (512 nm) condition, respectively. Measurement of Adsorption on Liposomal Membranes. The liposome suspensions (lipid: 3.0 mM) were mixed with aqueous solution of L- or D-form of histidine (His) (0.5 mM), and then they B
DOI: 10.1021/acs.langmuir.5b04626 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir were incubated at 25 °C for 24 or 48 h. After incubation, liposomes and adsorbed L-His were separated by ultrafiltration with the 50 000 Da of molecular weight cutoff (USY-5; Toyo Roshi Kaisha, Ltd., Tokyo, Japan). After filtration, the free concentration (Cflt) of L-His was measured by the absorbance of UV spectrometer (UV-1800; Shimadzu, Kyoto, Japan). The adsorbed concentration (Cads) and adsorption amount of L-His (qL‑His) or D-His (qD‑His) on several liposome membranes were calculated by the following equations:
Cads = C ini − Cflt
(5)
qL‐His or D‐His = Cads/C lip
(6)
rized as hydrophobic fluorescent probes and are therefore not suitable for monitoring membrane properties at the surface. In addition, Cartesian diagrams remain the method of choice for evaluating whole properties derived from each ratio of lipid components. Dansyl-DHPE is a fluorescent probe known to localize to hydrophilic regions at the surface of liposomal membranes,20 and can be used for the estimation of liposomal membrane hydrophobicity via the maximum-fluorescence wavelength.53 As shown in Figure 1a, a blue-shift in the emission wavelength of Dansyl-DHPE in the mixtures of water and 1,4-dioxane as solvents was observed with the decrease in the proportion of water. This behavior was caused by environmental changes around the fluorophore of the Dansyl moiety, although the fluorescence signal is negligible in the solvent below a 1,4-dioxane:water ratio of 0.3. The shift in the maximum wavelength was plotted against the dielectric constant of the solvent (Figure 1b), and we observed a linear correlation at wavelengths from 512 to 527 nm, except at the dielectric constant value of 57 because of the negligible fluorescence signal. Based on this, λ0 (most hydrophilic) and λ1 (most hydrophobic) were decided as 527 and 512 nm, respectively. Hence, the hydrophobicity of the membrane surface can be evaluated by the normalized value of the surface hydrophobicity (λN) by eq 4 in the Experimental Section. Based on these preliminary results, the maximum wavelength of Dansyl-DHPE fluorescence was measured by selecting several kinds of liposomes as targets (Figure 1c). The maximum wavelength of highly ordered DPPC liposomes was found to be relatively low as compared with disordered liposomal membranes, such as DOPC or POPC, implying that this parameter could be used together with other measurements to evaluate the hydrophobicity of liposomal membranes. With an increase in the proportion of Ch, DOPC and DPPC displayed a red-shift in the fluorescence spectrum, whereas such a shift was observed in the binary liposomes made with POPC. In the case of ternary liposomes, the wavelengths of their spectra remained similar to those of hydrophilic surfaces, implying that the surface hydrophobicity profile was not exactly the same as the GP340 measured by Laurdan. With reference to the Cartesian diagram (Figure S2), we evaluated the properties at the surface of various liposomes by the plot of λN versus TMA-DPH fluidity (1/PTMA‑DPH) as shown in Figure 2. In the case of liposomes with single components (DOPC, POPC, and DPPC) or those with 30% Ch, the data were clustered on the right-downward trend line (a0), indicating that the surface hydrophobicity decreased with increased surface fluidity. However, in the case of binary liposomes at 50% Ch, the data were clustered in another trend line (a2) that demonstrated a more negative value of its slope, resulting in the discrimination of a transition phase caused by Ch. The variation of these parameters was also plotted against the Ch ratio in order to assess the membrane physicochemical properties (Figure S3). The λN value of various liposomes decreased with increasing Ch ratio, with the 1/PTMA‑DPH values indicating only a slight decrease, while GP340 and 1/PDPH reached values similar to those observed with pure DPPC liposomes. As the Ch ratio increases, the liposomal membrane varies in the amount of accessible surface capable of permitting water-molecule invasion,54 resulting in membrane variation toward hydrophilicity only at the surface regions of liposomes modified with Ch. The change in λN became larger than that of 1/PTMA‑DPH, since the exclusion of water from the inner membrane and its displacement to the surface region can be
where Cini and Clip represent an initial concentration of adsorbates and liposomes, respectively. To evaluate the selectivity of His enantiomer, percent based enantiomeric excess (%ee) was calculated by the equation %ee = (qL‐His − qD‐His)/(qL‐His + qD‐His) × 100
(7)
where %ee is defined as the excess ratio of adsorbed L-His per total adsorption of L- and D-His. Statistical Analysis. Results are expressed as mean ± standard deviation. All experiments were performed three times. The distribution of data was analyzed, and statistical differences were evaluated by use of Student’s t test. A P-value of
