Surfactant-like Properties of an Amphiphilic α-Helical Peptide Leading

(1, 2) It is generally accepted that nanodiscs are the simplest models of ... the surfactant dialysis method:(1, 2, 14-17) scaffold α-helical protein...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Surfactant-like Properties of an Amphiphilic α‑Helical Peptide Leading to Lipid Nanodisc Formation Tomohiro Imura,*,† Yohei Tsukui,‡ Toshiaki Taira,† Kenichi Aburai,‡ Kenichi Sakai,‡ Hideki Sakai,‡ Masahiko Abe,‡ and Dai Kitamoto† †

Research Institute for Innovation in Sustainable Chemistry, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5-2, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan ‡ Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan S Supporting Information *

ABSTRACT: Nanodiscs are self-assembled discoidal nanoparticles composed of amphiphilic α-helical scaffold proteins or peptides that wrap themselves around the circumference of a lipid bilayer in a beltlike manner. In this study, an amphiphilic helical peptide that mimics helix 10 of human apoA-I was newly synthesized by solid phase peptide synthesis using Fmoc chemistry, and its physicochemical properties, including surface tension, self-association, and solubilization abilities, were evaluated and related directly to nanodisc formation. The synthesized peptide having hydrophobic and hydrophilic faces behaves like a general surfactant, affording a critical association concentration (CAC) of 2.7 × 10−5 M and a γCAC of 51.2 mN m−1 in aqueous solution. Interestingly, only a peptide solution above its CAC was able to microsolubilize L-α-dimyristoylphosphatidylcholine (DMPC) vesicles, and lipid nanodiscs with an average diameter of 9.5 ± 2.7 nm were observed by dynamic light scattering and negative stain transmission electron microscopy. Moreover, the ζ potentials of the lipid nanodiscs were measured for the first time as a function of pH, and the values changed from positive (20 mV) to negative (−30 mV). In particular, nanodisc solutions at acidic pH 4 (20 mV) or basic pH 9 (−20 mV) were found to be stable for more than 6 months as a result of the electrostatic repulsion between the particles.

1. INTRODUCTION Nanodiscs are self-assembled discoidal nanoparticles composed of amphiphilic α-helical scaffold proteins that wrap themselves around the circumference of a lipid bilayer in a beltlike manner.1,2 It is generally accepted that nanodiscs are the simplest models of high-density lipoprotein (HDL) particles that play a crucial role in “reverse cholesterol transport” (RCT), in which HDL particles or lipoproteins such as apolipoprotein A-I (apoA-I) remove excess cholesterol from peripheral tissues and transport it to the liver for elimination.3,4 Indeed, reduced plasma levels of HDLs and apoA-I are implicated in the increased risk for atherosclerosis and cardiovascular disease.5 Although native HDLs are heterogeneous particles with distinct sizes (6−13 nm) and shapes (discoidal and spherical), nanodiscs that are reconstituted homogeneous HDLs have attracted considerable attention as a new class of potential therapeutic agents for enhancing the RCT pathway by promoting cholesterol efflux.6,7 More recently, interest in nanodiscs has been focused on applications beyond their physiological role in lipid metabolism. They serve not only as a useful platform capable of packaging membrane proteins such as bacteriorhodopsin (bR)8,9 and cytochrome P450s10 in a nativelike membrane environment but also as useful vehicles for hydrophobic drugs11 or biomolecules.12,13 © 2014 American Chemical Society

In general, the preparation of nanodiscs is accomplished by the surfactant dialysis method:1,2,14−17 scaffold α-helical proteins or peptides are mixed with a certain amount of phospholipids solubilized into micelles with a surfactant such as cholate. The self-assembly process is then initiated by removing the surfactants via dialysis against excess water, and the amphiphilic α-helices stabilize the edges of the phospholipid bilayers, leading to homogeneous lipid nanodisc formation. However, the use of a surfactant should be avoided as much as possible, because surfactants often denature proteins or peptides and their complete removal is often difficult. Moreover, the dialysis process is generally time- and energyconsuming, and this method therefore may not always be appropriate when considering a wide variety of potential nanodisc applications. Among the apolipoproteins, apoA-I is the major protein in HDL and is composed of a series of 10 similar amphiphilic αhelices whose one face localizes mainly hydrophilic amino acids and the other face localizes hydrophobic amino acids.18−20 This unique amphiphilicity allows the helices to interact with the Received: January 21, 2014 Revised: April 3, 2014 Published: April 3, 2014 4752

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

TIS/water solution for 3 h. The crude peptide was precipitated with diethyl ether, centrifuged, washed three times with diethyl ether, and then purified by preparative reverse phase (RP) HPLC on a C18 column. The purity was confirmed by analytical RP-HPLC. Binary gradients of solvent A (99% H2O, 0.9% CH3CN, and 0.1% TFA) and solvent B (90% CH3CN, 9.9% H2O, and 0.07% TFA) were employed for HPLC. The purified peptide was characterized by matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry, where sinapinic acid (SA) was used as a matrix. 2.3. Surface Tension Measurements. The surface tension of the peptide was determined by the pendant drop method at 25 °C, which was performed using an apparatus consisting of an automatic interfacial tensiometer (DM500, Kyowa Interface Science) and the Drop Shape Analysis software of FAMAS version 2.01. A drop was formed at the tip of the syringe by pressing the solution out by means of a setscrew. The drop shape analysis was performed as follows.26 A drop profile was extracted from the drop image; then, a curve fitting program compared the experimental drop profile with a theoretical one (the Young−Laplace method) and gave the corresponding surface tension value. For each concentration of the purified peptide solution, the evolution of the drop surface tension was followed over 10 min. The concentration of the peptide stock solution in ultrapure water was measured by UV absorbance at 280 nm, and the solution was diluted with ultrapure water to each concentration. 2.4. Circular Dichroism (CD). CD spectra were recorded at 25 °C on a J-710 spectrophotometer (JASCO Corp.) using a 0.1 or 0.01 cm path length quartz cell. Spectra (20 scans) were obtained with a 3 nm bandwidth and a scan speed of 1 nm second−1. All CD spectra are reported in terms of ellipticity units per mole of peptide residue ([θ]MRE), calculated using the equation [θ]MRE = θobs/(10lcn), where θobs is the measured ellipticity in millidegrees, l is the path length in centimeters, c is the peptide concentration in mole per liter, and n is the number of amino acid residues. The percentage helical content of the peptide (α-helicity) was estimated on the basis of the [θ]222 value, using the equation α-helicity = ([θ]222 + 3000)/(36000 + 3000) × 100.27 2.5. Peptide−Lipid Nanodisc Preparation. Before preparation of the nanodiscs, DMPC multilamellar vesicles (MLVs) were prepared as follows.28 DMPC was dissolved in chloroform in a test tube. The solvent was then removed, first by blowing nitrogen gas into the test tube and then by storage overnight at room temperature under vacuum, to give a thin lipid film on the test tube wall. Ultrapure water, used to avoid interference from salts, was added to this lipid film at 25 °C, and the test tube was shaken vigorously on a vortex mixer to yield the MLV aqueous solution (4 mM). The formation of MLVs whose sizes are several micrometers was then confirmed by an optical microscope. We performed the direct conversion from a MLV to a nanodisc by adding the peptide solution at a 1:10 peptide:lipid molar ratio, vortexing the sample for 30 min, and subsequently incubating the sample for 24 h at 25 °C.6 The amount of peptides constructing nanodiscs was determined by a V-560 UV−vis spectrometer (JASCO), and that of DMPC was quantitatively analyzed by the choline oxidase/ phenol method.29 2.6. Size-Exclusion Chromatography (SEC).15 Samples were passed through a 0.45 μm PVDF filter and injected onto a Superdex 200 10/300 GL column (GE Healthcare) at 25 °C at a flow rate of 0.5 mL min−1 using an Ä KTAprime plus chromatography system (GE Healthcare). Phosphate buffer [67 mM phosphate (pH 7.4)] was used as a running buffer. The void volume (7.1 mL) was confirmed by injecting DMPC vesicles downsized to 200 nm using an extruder. The standards were thyroglobulin (669000 Da, 17 nm), ferritin (440000 Da, 12.2 nm), BSA (67000 Da, 7.1 nm), and ribonuclease A (13700 Da), which gave retention volumes of 8.3, 11.0, 14.0, and 17.2 mL, respectively. 2.7. Transmission Electron Microscopy (TEM).30 The nanoparticle fractions collected by SEC were imaged without further purification. A glow-discharged copper grid (200 mesh) coated with carbon (Excel support film, 200 mesh, Nisshin EM Co.) was immediately inverted, carbon surface down, onto a 2 μL droplet of the sample solution placed on Parafilm. After 30 s, the sample was

edges of phospholipid bilayers as well as display their distinctive amphiphilic behaviors in aqueous solutions. Recent structure− function studies indicated that lipid-free apoA-I adopts α-helical bundle structures to prevent the exposure of the hydrophobic lipid binding sites to solvent.21 It was also reported that helix 10 of human apoA-I (residues 220−241) plays a critical role in lipid binding and cholesterol efflux in the native protein.22,23 The direct self-assembly of nanodiscs without the use of surfactants was also reported for several apoA-I mimetic helical peptides,6,24 in which the structural conversion of phospholipid vesicles (liposomes) into nanodiscs was induced by the addition of the peptides. Instead of surfactants, amphiphilic helical peptides should provide physicochemical properties that resemble those of surfactants, such as lowering the surface tension, micellization, and solubilization during nanodisc formation. However, unlike general surfactants, the aqueous solution properties of amphiphilic helical peptides with hydrophobic and hydrophilic “faces” are still limited. As described above, nanodiscs are converted from liposomes that are generally large aggregates with diameters on the order of micrometers as precursor structures. In contrast to liposomes,25 nanodiscs are small, fine nanoparticles with high dispersion stability, and thus, nanodiscs are expected to be potentially useful nanoparticles beyond liposomes applicable to various fields. However, an important surface property of the resulting nanodiscs, the ζ potential, which is closely related to the dispersion stability or adsorption of biomolecules such as proteins and DNA, has never been reported before. In this study, we synthesized an amphiphilic peptide that mimics helix 10 of human apoA-I (residues 220−241), and its physicochemical properties such as surface tension, micellization, solubilization, and helicity were evaluated. Our results clearly demonstrate that the surfactant-like properties obtained, such as the critical association concentration (CAC) and γCAC, are directly related not only to the helicity of the peptide but also to its self-assembly for lipid nanodisc formation. Moreover, the ζ potentials of the resulting nanodiscs were evaluated for the first time as a function of pH, and the surface charge that results in electrostatic repulsion between the particles was also found to be critical for the dispersion stability of the nanodiscs.

2. MATERIALS AND METHODS 2.1. Materials. 1-[Bis(dimethylamino)methylene]-1H-benzotriazolium 3-oxide hexafluorophosphate (HBTU, 99.3%) and 9-fluorenylmethyloxycarbonyl (Fmoc)-protected amino acids (98.0%) were purchased from Peptide Institute Inc. (Osaka, Japan). 1-Hydroxy1H-benzotriazole (HOBt, 99.0%) was purchased from Dojindo Laboratories Co., Ltd. (Kumamoto, Japan). Rink amide MBHA resin was purchased from Peptides International Inc. (Osaka, Japan). Trifluoroacetic acid (TFA, 99.0%), diisopropylethylamine (DIPEA, 99.0%), piperidine (99.0%), triisopropylsilane (TIS, 99.0%), and highperformance liquid chromatography (HPLC) grade acetonitrile were purchased from Wako Pure Chemical Industries (Osaka, Japan). L-αDimyristoylphosphatidylcholine (DMPC, 99.0%) was purchased from NOF Corp. (Tokyo, Japan). 2.2. Peptide Synthesis. The peptide (NH2-PVLESFKASFLSALEEWTKKLN-NH2) was synthesized using standard Fmoc chemistry with a Syro I peptide synthesizer (Biotage). A typical synthesis was conducted via a stepwise solid phase synthesis procedure on Rink amide MBHA resin (0.09 mmol scale using 0.4 mmol/g). Standard side chain protecting groups included Asn(Trt), Lys(Boc), Thr(tBu), Trp(Boc), Glu(tOBu), and Ser(tBu). Couplings were performed using HBTU in DMF over 90 min. Fmoc groups were removed using 40% piperidine in DMF. The peptide was cleaved from the resin with concomitant side chain deprotection by agitation in a 95:2.5:2.5 TFA/ 4753

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

blotted using filter paper, and the grid surface was touched to a 20 μL water drop, again followed by blotting with filter paper. The surface was treated again with a second drop of water. The grid was immediately placed onto a 25 μL droplet of aqueous 0.75% phosphotungstic acid (pH 6.8) and then treated again with a second drop of phosphotungstic acid. After 20 s, the excess stain was removed and the grid was allowed to dry thoroughly. Images were taken on an H-7650 transmission electron microscope (Hitachi High-Technologies) at 120 kV. Size measurements were taken on at least 50 nanodiscs and were quantified using Simple Digitizer version 3.2. 2.8. Dynamic Light Scattering (DLS). The size and size distribution of the nanodiscs were measured with a DLS-7000 instrument (Otsuka Electronics Co.) using a 75 mW Ar laser with a 488 nm wavelength as a light source at 25 °C. The time-dependent correlation function of the scattered light intensity was measured at a scattering angle of 90°. The particle size distributions were determined using the software provided with the instrument. 2.9. ζ Potential Measurement.31 ζ potential measurements at various pH values were performed with an ELS-7000 analyzer (Otsuka Electronics), through which particle velocities can be measured with a light scattering technique using the Doppler effect, by virtue of a pair of mutually coherent laser beams (He−Ne laser at 633 nm). The ELS7000 instrument measures the autocorrelation function of the scattered light, and after signal processing, the instrument records the electrophoretic mobility and, finally, through the Smoluchowski equation, the ζ potential. A 0.01 M HCl or NaOH aqueous solution was added to adjust the pH of the nanodisc solutions.

mainly hydrophobic amino acids. Such a configuration could exhibit surface properties that resemble those of typical amphiphilic molecules such as surfactants having hydrophilic and hydrophobic groups. The peptide was synthesized and purified according to standard solid phase protocols and characterized by RP-HPLC and MALDI-TOF MS (Supporting Information). After the peptide had been purified, its aqueous solubility was preliminarily checked in a test tube. The peptide was found to be quite soluble in water. We then assessed the surface tension of the peptide, because one of the important properties of amphiphilic molecules such as surfactants is to reduce the surface tension of water. The surface tensions of aqueous peptide solutions at various concentrations were measured by the pendant drop method at 25 °C. Figure 2a shows the

3. RESULTS AND DISCUSSION For the peptide, we chose to use helix 10 of human apoA-I (residues 220−241), which is known to be critical for lipid binding and cholesterol efflux in the native protein.21 The 22amino acid peptide was designed to be as similar to the native helix 10 as possible and was altered only by introducing two conservative amino acid substitutions: Ala for Val-227 to increase the amphiphilicity and Trp for Tyr-236 to improve spectrophotometric analysis, following a previous report by Zhao et al.6 A helical wheel diagram for the peptide (NH2PVLESFKASFLSALEEWTKKLN-NH2) is illustrated in Figure 1. The diagram shows that the helical peptide used in this study displays distinctive amphiphilicity, wherein one face localizes mainly hydrophilic amino acids and the other face localizes

Figure 2. Surface activity of the peptide used in this study. (a) Surface tension−concentration plot of the peptide. The surface tension was determined by the pendant drop method at 25 °C. The peptide was dissolved in ultrapure water and the concentration determined by UV−vis spectroscopy (ε = 5560 cm−1 M−1). (b) Light scattering intensity−concentration plot of the peptide. The light scattering intensity of the peptide was detected using an Ar laser (488 nm).

relationship between surface tension and peptide concentration. The surface tension gradually decreased by the adsorption of the peptide at the air−water interface and became constant at a certain concentration upon reaching its adsorption equilibrium. The intersection point of the two fitted lines is defined as the critical micelle concentration (CMC) for surfactants, above which surfactants start to spontaneously self-assemble into

Figure 1. Helical wheel diagram for the peptide used in this study, illustrating its amphiphilic nature: yellow, hydrophobic; green, cationic; red, anionic; blue, polar uncharged. 4754

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

micelles. This suggests that micellelike associates are likely to form above the critical association concentration (CAC), even for amphiphilic peptides having hydrophobic and hydrophilic faces rather than groups. To confirm this, light scattering intensity measurements were performed for the peptide solutions at varying concentrations. Figure 2b shows the relationship between the scattering intensity and peptide concentration. The relative scattering intensity suddenly increased above the concentration of the intersection point observed in Figure 1a, indicating that the peptide begins to form self-associates similar to surfactant micelles at its CAC. The estimated CAC from Figure 1 was 2.7 × 10−5 M, and its γCAC was 51.2 mN m−1. Here we also measured the surface tension of sodium cholate that is generally used for nanodisc preparation by the surfactant dialysis method, and the estimated CMC and γCMC of sodium cholate were 1.6 × 10−2 M and 48.8 mN m−1, respectively. The amphiphilic peptide with a much more bulky structure gives a CAC value lower than that of and a γCMC value similar to that of sodium cholate, which would lead to nanodisc preparation without using a surfactant such as sodium cholate. In previous studies, the self-association of other apoA-I mimetic peptides such as 2F (Ac-DWLKAFYDKVAEKLKEAFNH2)27 and apoA-I (198−243)32 in a lipid-free state was also reported by nondenaturing polyacrylamide gel electrophoresis (PAGE) together with cross-linking of peptides, and apoA-I (198−243) was suggested to form tetramers or pentamers in an aqueous solution. Although the authors did not provide a detailed mechanism or critical surface parameters such as CAC and γCAC, our results clearly demonstrate that amphiphilic helical peptides behave like general surfactants in aqueous solution and spontaneously form self-associates above the CAC by reaching their adsorption equilibrium. We also performed DLS measurements of the peptide solution above the CAC (8 × 10−5 M), and the size of associates was found to be 4.3 ± 0.4 nm, which is close to the length of the helix of our synthesized peptide (3.3 nm). The structure of associates is thus probably a helix bundle structure in which their hydrophobic faces closely pack in the interior of the bundle to minimize solvent exposure. In a previous study, Lutgring et al. reported the formation of a helix bundle (tetramer) from amphiphilic helical peptides composed of 16 simple amino acids (Ac-EKLEKLLKELEELLKK-NH2).33 Although the sequence of the peptide is different, our peptide (NH2-PVLESFKASFLSALEEWTKKLN-NH2) is also very likely to form a helix bundle composed of a tetramer or a pentamer. Although further studies such as static light scattering (SLS) and small-angle X-ray scattering (SAXS) are necessary to determine the detailed structure, it is known that helix bundle formation induces an improvement in the αhelicity of peptides.33 The secondary structure of the peptide at different concentrations was then analyzed by CD spectroscopy. The CD spectra for the peptides at various concentrations in ultrapure water are shown in Figure 3a. Typical α-helical CD bands can be observed in the aqueous solution, with two negative peaks at 205 and 222 nm. Furthermore, the CD spectrum above the CAC is more negative than that below the CAC, indicating that the α-helical content is improved by the association of the peptides. The estimated α-helicities are plotted as a function of peptide concentration in Figure 3b. Interestingly, the α-helicity of the peptide was critically improved just above the CAC, which means self-association of the peptide directly contributes to the increase in α-helicity.

Figure 3. Characterization of the secondary structure of the peptide. (a) CD spectra for peptides at various concentrations in ultrapure water. (b) Helicity−concentration plot of the peptide. The helicity was determined by the magnitude of the minimum at 222 nm.

The increase in α-helicity will benefit the interaction with the phospholipid bilayers during nanodisc formation. These behaviors of the peptides in aqueous solution are likely related to the helix bundle formation of apoA-I itself in a lipid-free state, to prevent the exposure of its hydrophobic lipid binding site to solvent.21 It is well-known that an important role of surfactants is their ability to solubilize hydrophobic substrates in aqueous solution by forming micelles above their CMCs.34,35 This implies that amphiphilic peptides could also solubilize phospholipids, only above their CACs, which would lead to lipid nanodisc formation. Thus, we examined the effect of the presence or absence of the peptide associates on the formation of lipid nanodiscs. Peptide solutions below and above the CAC were mixed with phospholipid vesicles, in which a 1:10 peptide:phospholipid ratio was employed to mimic the 1:100 apoAI:phospholipid ratio in native HDL particles.6 The solutions were then analyzed by SEC. Figure 4 shows the elution curves of the solutions at peptide concentrations below and above the CAC (1 × 10−6 and 2 × 10−4 M, respectively). The elution curve of the solution below the CAC displays mainly two peaks. The first, at a retention time of 15.4 min, is the void volume resulting from the phospholipid vesicles, and the second peak at 43.6 min corresponds to the peptide. Additionally, one more weak peak (41.0 min) next to the peptide peak was recognized; this is probably due to the addition of DMPC molecues, as a 4755

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

estimated by the TEM image was 4.8 ± 1.0 nm, which corresponds well to the thickness of DMPC bilayers (5.1 nm) constructing nanodiscs with membrane scaffold proteins (200 amino acids).36 This fact partially supports the beltlike model of nanodisc structure. Some particles also stacked into a rouleaux formation in the image, a phenomenon known to be an artifact of negative-stain TEM protocols.37−39 The difference in size between the DLS and NS-TEM data is thus probably caused by the NS-TEM protocols. Thus far, the results clearly demonstrate that nanodisc formation is induced only in the presence of peptide associates above its CAC: the peptide adsorbs at the air−water interface at low concentrations, and after reaching its adsorption equilibrium, it begins to form self-associates in aqueous solution through a mechanism resembling that of micelle formation for general surfactants. Micellelike peptide associates with higher helicities allow phospholipids to be microsolubilized and spontaneously self-assemble into lipid nanodiscs. We then directly estimated the amount of peptide and DMPC by quantitatively analyzing the SEC nanodisc fraction. The estimated amounts of peptide and DMPC constructing nanodics were 2.0 × 10−7 and 2.8 × 10−6 mol, respectively. The DMPC:peptide ratio was then 14, which means that a single peptide molecule can solubilize 14 DMPC molecules. On the basis of a beltlike model of the discoidal particle, we can also assume the composition of nanodiscs using the following equations: ML = 2πR1; NS = πR12; R2 − R1 = r, where M and N are the numbers of phospholipids and peptides in one leaflet of the bilayer, respectively, L and r are the length and diameter of a helical peptide, respectively, R1 is the radius of nanodiscs occupied by phospholipids, R2 is the radius of nanodiscs, and S is the average surface area per phospholipid. The length and diameter of a typical helix composed of 22 amino acids are known to be 3.3 and 1.2 nm, respectively. If we use the Stokes radius of nanodiscs (5.9 nm) obtained by SEC measurements for R1 and the area per molecule (S) of DMPC reported (0.6 nm2),40 the numbers of DMPCs and peptides in one leaflet of nanodiscs were calculated to be 9 (M) and 115 (N), respectively. The DMPC (N)/peptide (N) ratio was then estimated to be 13. Interestingly, the estimated value is almost the same as that obtained from quantitative analysis of the SEC nanodisc fraction. This also supports a beltlike model of nanodiscs. To understand the interaction between apolipoproteins and phospholipids, extensive studies have been conducted on apolipoproteins interacting with phospholipid monolayers spread at the air−water interface, where exclusion pressures (πe) at which the apolipoproteins could no longer penetrate into the phospholipid monolayers have been measured.41,42 Wang et al. recently reported the interfacial properties of apoAI and apoA-I mimetic peptides at the triolein−water interface, and the surface pressure-mediated desorption and readsorption of apoA-I were then assessed.43,44 Although these studies are quite important for understanding how apoA-I and apoA-I mimetic peptides behave at the interface, our results clearly demonstrate that knowledge of not only the primary and secondary structures of a peptide but also its surface parameters (such as CAC and γCAC) as for general surfactants is very important for the effective production and engineering of nanodiscs. It is well-known that the hydrophile−lipophile balance (HLB) is important in dealing with general surfactants having hydrophilic and hydrophobic “groups”. However, the

Figure 4. SEC profiles after incubation of DMPC vesicles with the peptide below (1 × 10−5 M) and above (2 × 10−4 M) the CAC. The Stokes diameter of the nanoparticles was determined by comparison of its SEC retention time to standard curves of four globular proteins.

small amount of DMPC may cause the split of the peptide peak. In contrast, the elution curve above the CAC gave a remarkable peak at 23.4 min with a Stokes diameter of 11.8 nm, suggesting the formation of nanodiscs only in the presence of peptide associates. The Stokes diameter was estimated on a Superdex 200 SEC column calibrated with a standard set of proteins, as described in Materials and Methods. The fraction of the main peak (23.4 min) from the SEC separation was collected, and an aliquot was analyzed by DLS and negative stain transmission electron microscopy (NSTEM). The DLS measurements in Figure 5 reveal a quite

Figure 5. DLS size distribution of the nanoparticles prepared from the peptide. The inset shows an NS-TEM image of the nanoparticles.

narrow particle size distribution, with an average diameter of 9.5 ± 2.7 nm, which is close to the Stokes diameter estimated via SEC. TEM observation of the aliquot indicated circular and rodlike objects that correspond to top and side views of the nanodiscs, respectively; the average diameter estimated from the TEM image was 16.2 ± 5.3 nm, which is larger than that from the DLS measurement. The thickness of nanodiscs 4756

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

association, and solubilization abilities, were evaluated in aqueous solution. The synthesized peptide, having hydrophobic and hydrophilic faces, behaves like a general surfactant, with a CAC of 2.7 × 10−5 M and a γCAC of 51.2 mN m−1. Interestingly, the helicity of the peptide critically increased just above its CAC, and the peptides were able to microsolubilize DMPC vesicles only in the presence of their associates. The lipid nanodiscs with an average diameter of 9.5 ± 2.7 nm were then confirmed by SEC, DLS, and NS-TEM. Moreover, the ζ potentials of the lipid nanodiscs were measured for the first time over a range of pH values, changing from positive (20 mV) to negative (−30 mV). In particular, the nanodisc solution at acidic pH 4 (20 mV) or basic pH 9 (−20 mV) was stable more than six months after preparation because of the electrostatic repulsion of the nanodiscs. Our results clearly demonstrate that surface properties such as the CAC and γCAC of the peptides and colloidal properties such as the ζ potential are critical for the effective production and engineering of nanodiscs. We hope these findings will greatly contribute to widespread applications of nanodiscs.

amphiphilic peptide in this study has hydrophilic and hydrophobic “faces” instead of groups. It seems that the progressive concept is necessary to handle amphiphilic peptides having hydrophilic and hydrophobic faces. We next focused on another important surface property of the resulting nanodiscs. Their ζ potentials in aqueous solution were measured by varying the pH, because the surface charges of the nanoparticles greatly influence particle stability in solution through electrostatic repulsion.45,46 Figure 6 plots

■

ASSOCIATED CONTENT

S Supporting Information *

Characterization of synthesized peptide by analytical RP-HPLC and MALDI-TOF MS (Figures S1 and S2). This material is available free of charge via the Internet at http://pubs.acs.org.

■

Figure 6. ζ potential−pH plot of the DMPC vesicles (●) and the nanodiscs (■).

AUTHOR INFORMATION

Corresponding Author

*Phone: +81-29-861-4738. Fax: +81-29-861-4457. E-mail: [email protected].

the ζ potentials of the nanodiscs as a function pH, together with those of the DMPC vesicles before the addition of the peptides. The ζ potential of the DMPC vesicles changed from positive to negative with an increase in pH, because DMPC has a zwitterionic headgroup and H+ atoms are bound to the phosphate group of the phosphatidylcholine moiety at low pH.29 After conversion to nanodiscs via addition of the peptides, the ζ potential values also changed from positive to negative but were higher at all pH values than those of the DMPC vesicles. The particle stability of the nanodiscs at neutral pH, which is close to the isoelectric point (PI) of the synthesized peptide (PI = 6.6), was lower than that at acidic or basic pH; the transparent nanodisc solution became slightly turbid at pH 7 two weeks after preparation. In contrast, the nanodisc solution at acidic pH 4 (20 mV) or basic pH 9 (−20 mV) remained transparent more than six months after preparation, probably because of the electrostatic repulsion of the nanodiscs. At any pH, the nanodisc solution was more stable than that of the DMPC vesicles, demonstrating that nanodiscs with a high dispersion stability are potentially useful species beyond liposomes that can be applied to various fields. Although ζ potential values do not always yield direct information about surface charge, to the best of our knowledge, this is the first observation of ζ potential values of nanodiscs that would be useful not only for the prediction of nanodisc stability but also for the incorporation of biomolecules such as DNA, RNA, or proteins into the nanodiscs.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge valuable discussions with Dr. Luke J. Leman in the Department of Chemistry at the Scripps Research Institute (La Jolla, CA).

■

REFERENCES

(1) Bayburt, T. H.; Grinkova, Y. V.; Sligar, S. G. Self-Assembly of Discoidal Phospholipid Bilyaer Nanoparticles with Membrane Scaffold Proteins. Nano Lett. 2002, 2, 853−856. (2) Shih, A. Y.; Freddolino, P. L.; Sligar, S. G.; Schulten, K. Disassembly of Nanodiscs with Cholate. Nano Lett. 2007, 7, 1692− 1696. (3) Natarajan, P.; Ray, K. K.; Cannon, C. P. High-Density Lipoprotein and Coronary Heart Disease. J. Am. Coll. Cariol. 2010, 55, 1283−1299. (4) Hara, H.; Yokoyama, S. Interaction of Free Apolipoproteins with Macropharges: Formation of High-Density Lipoprotein-like Lipoproteins and Reduction of Cellular Cholesterol. J. Biol. Chem. 1991, 266, 3080−3086. (5) Fielding, C. J.; Fielding, P. E. Molecular Physiology of Reverse Cholesterol Transport. J. Lipid Res. 1995, 36, 211−228. (6) Zhao, Y.; Imura, T.; Leman, L. J.; Curtiss, L. K.; Maryanoff, B. E.; Ghadiri, M. R. Mimicry of High-Density Lipoprotein: Functional Peptide-Lipid Nanoparticles Based on Multivalent Peptide Constructs. J. Am. Chem. Soc. 2013, 135, 13414−13424. (7) Nissen, S. E.; Tsunoda, T.; Tuzcu, E. M.; Schoenhagen, P.; Cooper, C. J.; Yasin, M.; Eaton, G. M.; Lauer, M. A.; Sheldon, W. S.; Grines, C. L.; Halpem, S.; Crowe, T.; Blankenship, J. C.; Kerensky, R. Effect of Recombinant ApoA-I Milano on Coronary Atherosclerosis in

4. CONCLUSIONS The 22-amino acid amphiphilic peptide (NH2-PVLESFKASFLSALEEWTKKLN-NH2) that mimics helix 10 of human apoAI (220−241) was newly synthesized by SPPS, and its physicochemical properties, including surface tension, self4757

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

Patients with Acute Coronary Syndromes. JAMA, J. Am. Med. Assoc. 2003, 290, 2292−2300. (8) Bayburt, T. H.; Grinkova, Y. V.; Sligar, S. G. Assembly of Single Bacteriorhodopsin Trimers in Bilyaer Nanodiscs. Arch. Biochem. Biophys. 2006, 450, 215−222. (9) Blanchette, C. D.; Cappuccio, J. A.; Kuhn, E. A.; Segelke, B. W.; Benner, W. H.; Chromy, B. A.; Coleman, M. A.; Bench, G.; Hoeprich, P. D.; Sulchek, T. A. Atomic Force Microscopy Differentiates Discrete Size Distributions Between Membrane Protein Containing and Empty Nanolipoprotein Particles. Biochim. Biophys. Acta 2009, 1788, 724− 731. (10) Baas, B. J.; Denisov, I. G.; Sligar, S. G. Homotropic Cooperativity of Monomeric Cytochrome P450 3A4 in a Nanoscale Native Bilayer Environment. Arch. Biochem. Biophys. 2004, 430, 218− 228. (11) Lacko, A. G.; Nair, M.; Paranjape, S.; Johnson, S.; McConathy, W. J. High Density Lipoprotein Complexes as Delivery Vehicles for Anticancer Drugs. Anticancer Res. 2002, 22, 2045−2049. (12) Oda, M. N.; Hargreaves, P. L.; Beckstead, J. A.; Redmond, K. A.; van Antwerpen, R.; Ryan, R. O. Reconstituted High Density Lipoprotein Enriched with the Polyene Antibiotic Amphotericin B. J. Lipid Res. 2006, 47, 260−267. (13) Redmond, K. A.; Nguyen, T. S.; Ryan, R. O. All-Trans-Retinoic Acid Nanodisks. Int. J. Pharm. 2007, 339, 246−250. (14) Jonas, A. Reconstitution of High-Density-Lipoproteins. Methods Enzymol. 1986, 128, 553−582. (15) Denisov, I. G.; Grinkova, Y. V.; Lazarides, A. A.; Sligar, S. G. Directed Self-Assembly of Monodisperse Phospholipid Bilayer Nanodiscs with Controlled Size. J. Am. Chem. Soc. 2004, 126, 3477−3487. (16) Marty, M. T.; Wilcox, K. C.; Klein, W. L.; Sliger, S. G. NanodiscSolubilized Membrane Protein Library Reflects the Membrane Proteome. Anal. Bioanal. Chem. 2013, 405, 4009−4016. (17) Cuellar, L. A.; Prieto, E. D.; Cabaleiro, L. V.; Garda, H. A. Apolipoprotein A-I configuration and cell cholesterol efflux activity of discoidal lipoproteins depend on the reconstitution process. Biochim. Biophys. Acta 2014, 1841, 180−189. (18) Rye, K. A.; Barter, P. J. Formation and Metabolism of PrebetaMigrating, Lipid-Poor Apolipoprotein A-I. Arterioscler., Thromb., Vasc. Biol. 2004, 24, 421−428. (19) Nagao, K.; Hata, M.; Tanaka, K.; Takechi, Y.; Nguyen, D.; Dhanasekaran, P.; Lund-Katz, S.; Phillips, M. C.; Saito, H. The Roles of C-Terminal Helices of Human Apolipoprotein A-I in Formation of High-Density Lipoprotein Particles. Biochim. Biophys. Acta 2014, 1845, 80−87. (20) Fukuda, M.; Nakano, M.; Miyazaki, M.; Tanaka, M.; Saito, H.; Kobayashi, S.; Ueno, M.; Handa, T. Conformational Change of Apolipoprotein A-I and HDL Formation from Model Membranes under Intracellular Acidic Conditions. J. Lipid Res. 2008, 49, 2419− 2426. (21) Kiss, R. S.; Kay, C. M.; Ryan, R. O. Amphipathic α-Helix Bundle Organization of Lipid-Free Chicken Apolipoprotein A-I. Biochemistry 1999, 38, 4327−4334. (22) Palgunachari, M. N.; Mishra, V. K.; Lund-Katz, S.; Phillips, M. C.; Adeyeye, S. O.; Alluri, S.; Anantharamaiah, G. M.; Segrest, J. P. Only the Two End Helixes of Eight Tandem Amphipathic Helical Domains of Human Apo A-I Have Significant Lipid Affinity. Arterioscler., Thromb., Vasc. Biol. 1996, 16, 328−338. (23) Panagotopulos, S. E.; Witting, S. R.; Horace, E. M.; Hui, D. Y.; Maiorano, J. N.; Davidson, W. S. The Role of Apolipoprotein A-I Helix 10 in Apolipoprotein-mediated Cholesterol Efflux via the ATP-Binding Cassette Transporter ABCA1. J. Biol. Chem. 2002, 277, 39477−39484. (24) Datta, G.; Chaddha, M.; Hama, S.; Navab, M.; Fogelman, A. M.; Garber, D. W.; Mishra, V. K.; Epand, R. M.; Lund-Katz, S.; Phillips, M. C.; Segrest, J. P.; Anantharamaiah, G. M. Effects of Increasing Hydrophibicity on the Physical-Chemical and Biological Properties of a Class A Amphipathic Helical Peptide. J. Lipid Res. 2001, 42, 1096− 1104. (25) Sharma, A.; Sharma, U. S. Liposomes in Drug Delivery: Progress and Limitations. Int. J. Pharm. 1997, 154, 123−140.

(26) Imura, T.; Masuda, Y.; Minamikawa, H.; Fukuoka, T.; Konishi, M.; Morita, T.; Sakai, H.; Abe, M.; Kitamoto, D. Enzymatic Conversion of Diacetylated Sophoroselipid into Acetylated Glucoselipid: Surface-Active Properties of novel Bolaform Biosurfactants. J. Oleo Sci. 2010, 59, 495−501. (27) Morriset, J. D.; David, J. S. K.; Pownall, H. J.; Gotto, A. M. Interaction of an Apolipoprotein (apoLP-alanice) with Phosphocholine. Biochemistry 1973, 12, 1290−1299. (28) Otake, K.; Imura, T.; Sakai, H.; Abe, M. Development of a New Preparation Method of Liposomes Using Supercritical Carbon Dioxide. Langmuir 2001, 17 (13), 3898−3901. (29) Imura, T.; Sakai, H.; Yamauchi, H.; Kaise, C.; Kozawa, K.; Yokoyama, S.; Abe, M. Preparation of liposomes containing Ceramide 3 and their membrane characteristics. Colloids Surf., B 2001, 20, 1−8. (30) Duong, P. T.; Collins, H. L.; Nichel, M.; Lund-Katz, L.; Rothblat, G. H.; Phillips, M. C. Characterization of Nascent HDL Particles and Microparticles Formed by ABCA1-Mediated Efflux of Cellular Lipids to ApoA-I. J. Lipid Res. 2006, 47, 832−843. (31) Garcia-Manyes, S.; Oncins, G.; Sanz, F. Effect of pH and Ionic Strength on Phospholipid Nanomechanics and on Deposition Process onto Hydrophilic Surfaces Measured by AFM. Electrochim. Acta 2005, 51, 5029−5036. (32) Zhu, H. L.; Atkinson, D. Conformation and Lipid Binding of a C-Terminal (198−243) Peptide of Human Apolipoprotein A-I. Biochemistry 2007, 46, 1624−1634. (33) Lutgring, R.; Lipton, M.; Chmielewski, J. Differential self assembly of amphiphilic helical peptides. Amino Acids 1996, 10, 295− 304. (34) Francis, M. F.; Piredda, M.; Winnik, F. M. Solubilization of Poorly Water Soluble Drugs in Micelles of Hydrophibically Modified Hydroxypropylcellulose Copolymers. J. Controlled Release 2003, 93, 59−68. (35) Nagarajan, R. Solubilization in Aqueous Solutions of Amphiphiles. Curr. Opin. Colloid Interface Sci. 1996, 1, 391−401. (36) Denisov, I. G.; McLean, M. A.; Shaw, A. W.; Grinkova, Y. V.; Sligar, S. G. Thermotropic Phase Transition in Soluble Nanoscale Lipid Bilayers. J. Phys. Chem. B 2005, 109, 15580−15588. (37) Zhang, L.; Song, J.; Cavigiolio, G.; Ishida, B. Y.; Zhang, S.; Kane, J. P.; Weisgraber, K. H.; Oda, M. N.; Rye, K.-A.; Pownall, H. J.; Ren, G. Morphology and Structure of Lipoproteins Revealed by an Optimized Negative-Staining Protocol of Electron Microscopy. J. Lipid Res. 2011, 52, 175−184. (38) Schneeweis, L. A.; Koppaka, V.; Lund-Katz, S.; Phillips, M. C.; Axelsen, P. H. Structual Analysis of Lipoprotein E Particles. Biochemistry 2005, 44, 12525−12534. (39) Raussens, V.; Drury, J.; Forte, T. M.; Choy, N.; Goormaghtigh, E.; Ruysschaert, J. M.; Narayanaswami, V. Orientation and Mode of Lipid-Binding Interaction of Human Apolipoprotein E C-terminal Domain. Biochem. J. 2005, 387, 747−754. (40) Pasenkiewicz-Gierula, M.; Takaoka, Y.; Miyagawa, H.; Kitamura, K.; Kusumi, A. Charge pairing of headgroups in phosphatidylcholine membranes: A molecular dynamics simulation study. Biophys. J. 1999, 76, 1228−1240. (41) Ibdah, J. A.; Krebs, K. E.; Phillips, M. C. The Surface-Properties of Apolipoproteins A-I and A-II at the Lipid Water Interface. Biochim. Biophys. Acta 1989, 1004, 300−308. (42) Weinberg, R. B.; Cook, V. R.; Delozier, J. A.; Shelness, G. S. Dynamic Interfacial Properties of Human Apolipoproteins A-IV and B17 at the Air/Water and Oil/Water Interface. J. Lipid Res. 2000, 1419− 1427. (43) Wang, L.; Hua, N.; Atkinson, D.; Small, D. M. The N-Terminal (1−44) and C-Terminal (198−243) Peptides of Apolipoprotein A-I Behave Differently at the Triolein/Water Interface. Biochemistry 2007, 46, 12140−12151. (44) Wang, L.; Atkinson, D.; Small, D. M. The Interfacial Properties of ApoA-I and an Amphipathic α-Helix Consensus Peptide of Exchangeable Apolipoproteins at the Triolein/Water Interface. J. Biol. Chem. 2005, 280, 4154−4165. 4758

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759

Langmuir

Article

(45) Hans, M. L.; Lowman, A. M. Biodegradable Nanoparticles for Drug Delivery and Targeting. Curr. Opin. Solid State Mater. Sci. 2002, 6, 319−327. (46) Taylor, T. M.; Gaysinsky, S.; Davidson, P. M.; Bruce, B. D.; Weiss, J. Characterization of Antimicrobial-Bearing Liposomes by ζPotential, Vesicle Size, and Encapsulation Efficiency. Food Biophys. 2007, 2, 1−9.

4759

dx.doi.org/10.1021/la500267b | Langmuir 2014, 30, 4752−4759