Albumin-fatty acid interactions at monolayer interface
© Gew and Misran; licensee Springer. 2014
Received: 21 January 2014
Accepted: 20 April 2014
Published: 7 May 2014
The fluid mosaic model of Singer and Nicolson in 1972 shows how proteins are embedded in membranes. To elucidate the interactions between proteins and the surrounding lipids, stearic acid (SA) and bovine serum albumin (BSA) were used as lipid-protein components to mimic the normal membrane bilayer environment using the Langmuir-Blodgett technique. Surface pressure (π)-molecular area (A) isotherms were recorded for the SA monolayer in the presence of BSA on water. The mixed monolayer was successfully transferred onto an oxidized silicon wafer and imaged by tapping mode atomic force microscopy (AFM). Miscibility, compressibility and thermodynamic stability of the mixed system were examined. A large negative deviation of Aex, together with the minimum value of ΔGex, was observed when the mole fraction of BSA (XBSA) was 0.8, indicating this to be the most stable mixture. In a compressibility analysis, XBSA was observed at below 50 mN m-1, denoting a liquid-expanded phase and showing the occurrence of a strong interaction of SA with BSA molecules in this phase. AFM observations supported the quantitative data indicating that BSA was strongly attracted onto the membrane surface as predicted.
Immunoliposomes have been extensively developed for its potential as drug delivery carriers by attaching antibodies to the liposomal surface. Many in vitro studies using immunoliposomes in drug delivery to target cancer cells have greatly showed significant reduction in toxicities and improved therapeutic efficacy[1–4]. This promising approach can overcome challenges of targeting only the cancer and tumour cells that are often very similar in characteristics to the surrounding healthy tissue. Without the incorporation of targeting antibodies, liposomes would have to rely on nonspecific interactions with cell membranes. There are several proposed drug-loaded immunoliposome formulations that are used in drug delivery applications[5–8], but there is still scant knowledge on how liposomes interact with the antibodies they incorporate[9, 10].
With the view of investigating interactions between liposomes and antibodies, we first set out to study the interactions between fatty acids and proteins by the Langmuir-Blodgett technique. Langmuir monolayers are widely used to model the biological membrane surface in studies to understand the structure and function of biological membranes and the protein-lipid interactions. The way proteins assemble on the lipid bilayer, either partially or fully embedded, and their ensuing stability should be considered before any experiment on the incorporation of proteins in the membrane is performed[12, 13]. In 1972, Singer and Nicolson made the important distinction between integral and peripheral membrane proteins in the fluid mosaic model of biological membranes. Lipid-protein interactions that occur in the binary mixed system can be studied from data on miscibility, compressibility and thermodynamic stability from the isotherms obtained. The analysed data would give an insight into intermolecular interactions between the lipid and protein, thereby providing useful information on the different ways proteins associate with cell membranes.
In our study, we used stearic acid (SA) to create a monolayer mimicking a half bilayer membrane, with various concentrations of bovine serum albumin (BSA) incorporated onto the monolayer. BSA is a globular protein that is highly water soluble and readily available at low cost. Its structural similarity to the human homologue makes it a widely studied protein. To the best of our knowledge, the behaviour of BSA in a mixed lipid monolayer has not been studied in any great detail.
The outcome of this initial study would provide indicators for future work on the interactions of other globular proteins, including antibodies, in a mixed lipid monolayer.
A spreading solution of stearic acid (Sigma-Aldrich, Palo Alto, CA, USA) was prepared by dissolving it in analytical grade chloroform (Merck, Whitehouse Station, NJ, USA). Various concentrations of bovine serum albumin (Carl Roth GmbH, Karlsruhe, Germany) were prepared by dissolving in distilled water. Double-distilled water (processed by NANOpure Diamond Ultrapure Water System, Barnstead International, Dubuque, IA, USA) was used as the subphase throughout the study.
Langmuir monolayer/mixed monolayer measurements
A computer-controlled Langmuir balance (KSV 5000, Langmuir System, Helsinki, Finland) equipped with symmetric barriers and Teflon trough (total area 60,720 mm2) was used to determine the surface pressure (π)-molecular area (A) isotherms. The surface pressure of the films was measured to an accuracy of ±0.1 mN m-1 using a flame-cleansed high-purity platinum metal Wilhelmy plate (19.62 mm × 10 mm) of 39.80-mm total length. The trough was filled with water (26°C ± 0.1°C) serving as the subphase. Solutions of SA and BSA were carefully transferred and spread randomly onto the subphase (water) using a Hamilton microsyringe (precision to 0.5 μl). The solutions were left for about 10 min to allow the solvent to evaporate before the π-A isotherms were measured. The films were compressed at a rate of 10 mm min-1.
Y-type deposition of pure SA and SA/BSA on substrate
Silicon (100) wafers were cut into approximately 5 cm × 1 cm pieces and placed in a furnace (Carbolite, Watertown, WI, USA) for 8 h at 900°C to allow oxidation. The oxidized silicon wafer was clamped vertical to the subphase and immersed into the dipping well before spreading the monolayer material. After complete evaporation of the solvent, the floating layer was compressed at a rate of 10 mm min-1 to reach a target surface pressure of 20 mN m-1 and kept for 15 min to attain stability for deposition. The Y-type deposition of LB film was performed at the targeted pressure with a dipping speed of 10 mm min-1. All the transferred films were kept for a week in a dry, clean and closed container before atomic force microscopy (AFM) imaging.
High-resolution imaging of bilayers was obtained by AFM after transferring them from the air/water interface to a solid oxidized silicon substrate. Mixed bilayers from the Langmuir trough were transferred onto oxidized silicon substrates at the desired Wilhelmy pressure. Bilayers transferred to substrates were imaged using the NanoScopeIIIa scanning probe microscope controller (Veeco Instruments Inc., Plainview, NY, USA) in tapping mode under ambient conditions. Aluminum probes (Budget Sensors BS Multi 75Al, Innovative Solutions Bulgaria Ltd., Sofia, Bulgaria) were used. Resonance frequency of the probe was 75 kHz, and the force constant was 3 N m-1. Images in height mode were collected simultaneously with 256 × 256 points at a scanning rate of 1.0 Hz per line. A series of AFM images were taken from different perspectives.
Results and discussion
π- A measurements and analyses
π- A isotherm
When BSA was incorporated into the SA monolayer, the shape of the π-A isotherm gradually changed with increasing concentrations of BSA. All isotherms of the mixed systems shifted from the isotherm of pure SA to that of BSA as the mole ratio of BSA (XBSA) rose. This change showed that a mixed monolayer of SA/BSA was successfully formed, with more interactions between SA and BSA taking place as the concentration of BSA increased. A marked shift away from the isotherm of pure SA was observed at XBSA = 0.8, 0.9 and 1.0 (the last value being pure BSA). There was no collapse pressure observed for XBSA ≥ 0.9, suggesting that a stronger interaction occurred between SA and BSA with high concentrations of BSA in the mixed monolayer system.
Energetic stability of the mixed monolayers
The miscibility of the mixed monolayer components can be determined by calculating the mean molecular area A12.
C s - 1 curves provide detailed information on phase transitions of SA/BSA monolayers. C s - 1 can be classified into various phases, namely (a) liquid-expanded (LE) phase at surface pressure from 10 to 50 mN m-1, (b) liquid (L) phase from 50 to 100 mN m-1, (c) liquid-condensed (LC) phase from 100 to 250 mN m-1 and (d) solid (S) phase above 250 mN m-1. In this work, the compression moduli were obtained by numerical calculation of the first derivative from the isotherm data point using the OriginPro-8 program.
The isotherm of pure SA showed two distinct regions: the first one corresponding to the monolayer in its liquid-condensed (LC) phase and the second one of a solid (S) film that was characterized by higher C s - 1 values. The values of C s - 1 obtained that were relatively high at XBSA = 0.1 are characteristic of a LC phase. The reason for this observation could be that at low concentrations of BSA, less lipid-protein interaction occurred in the mixed system.
At surface pressure 30 and 35 mN m-1, C s - 1 was observed to be below 50 mN m-1 for the entire range of BSA mole ratios, from XBSA ≥ 0.2 onwards, this being indicative of the formation of the LE phase. This implied that the incorporation of BSA into the SA monolayers reduced their condensation.
where A12, A1 and A2 represent the area of the mixed system and respective areas of components as 1 and 2, respectively, and π is the surface pressure of the monolayer. If the monolayer is ideally mixed, ΔGex should be zero.
ΔGex gradually decreased as the concentration of BSA rose. There was a slight recovery of ΔGex at XBSA = 0.9. When the monolayer contained BSA only, ΔGex was almost similar to XBSA = 0.9. This might be due to intermolecular repulsion occurring in the mixed monolayer system when the concentration of BSA was saturated in the system. The most compatible mixture of SA/BSA in a mixed monolayer was when XBSA = 0.8.
AFM observation of pure SA bilayer and SA/BSA mixed bilayer system
SA and BSA showed strong attraction as the concentration of BSA increased. The mixed monolayer was found to be most miscible at XBSA = 0.8 as indicated by the negative Gibbs free excess energy. Analysis of the binary SA/BSA mixed monolayer confirms the spontaneous interaction between integral proteins and the lipids in accordance with the fluid mosaic model of Singer and Nicolson in 1972. The ensuing lipid bilayer with embedded proteins is thermodynamically stable, reflecting the situation in biological membranes.
This study was financially supported by the Postgraduate Research Fund (PS348/2010A) by University of Malaya and Sunway University Research Grant (INT-ADTP-0210-01).
- Lundberg BB, Griffiths G, Hansen HJ: Specific binding of sterically stabilized anti B-cell immunoliposomes and cytotoxicity of entrapped doxorubicin. Int J Pharm 2000, 205: 101. 10.1016/S0378-5173(00)00492-0View ArticleGoogle Scholar
- Lundberg BB, Griffiths G, Hansen HJ: Cellular association and cytotoxicity of anti-CD74-targeted lipid drug-carriers in B lymphoma cells. J Control Released 2004, 94: 155. 10.1016/j.jconrel.2003.09.016View ArticleGoogle Scholar
- Guo P, You JO, Yang J, Moses MA, Auguste DT: Using breast cancer cell CXCR4 surface expression to predict liposome binding and cytotoxicity. Biomolecules 2012, 33: 8104.Google Scholar
- Park JW, Benz CC, Martin FJ: Future directions of liposome- and immunoliposome-based cancer therapeutics. Semin Oncol 2004, 31: 196.View ArticleGoogle Scholar
- Park JW, Hong K, Cargter P, Asgari H, Guo LY, Keller GA, Wirth C, Shalaby R, Kotts C, Wood WI, Papahadjopoulos D, Benz CC: Development of anti-p185HER2 immunoliposomes for cancer therapy. Proc Natl Acad Sci 1995, 92: 1327. 10.1073/pnas.92.5.1327View ArticleGoogle Scholar
- Vingerhoeds MH, Steerenberg PA, Hendricks JJGW, Dekker LC, Hoesel QGCMV, Crommelin DJA, Storm G: Immunoliposomes-mediated targeting of doxorubicin to human ovarian carcinoma in vitro and in vivo. Bristish J Cancer 1996, 74: 1023. 10.1038/bjc.1996.484View ArticleGoogle Scholar
- Koning GA, Morselt HWM, Gorter A, Allen TM, Zalipsky S, Kamps JAAM, Scherphof GL: Pharmacokinetics of differently designed immunoliposome formulations in rats with or without hepatic colon cancer metastases. Pharm Res 2001, 18: 1291. 10.1023/A:1013085811044View ArticleGoogle Scholar
- Mack K, Ruger R, Fellermeier S, Seifert O, Kontermann RE: Dual targeting of tumor cells with bispecific single-chain Fv-immunoliposomes. Antibodies 2012, 1: 199. 10.3390/antib1020199View ArticleGoogle Scholar
- Martin FJ, Hubbell WL, Papahadjopoulos D: Immunospecific targeting of liposomes to cells: a novel and efficient method for covalent attachment of Fab' fragments via disulfide bonds. Biochemistry 1981, 20: 4229. 10.1021/bi00517a043View ArticleGoogle Scholar
- Heath TD, Fraley RT, Papahdjopoulos D: Antibody targeting of liposomes: cell specificity obtained by conjugation of F(ab')2 to vesicle surface. Science (New York) 1980, 210: 539. 10.1126/science.7423203View ArticleGoogle Scholar
- Gaines GL: Insoluble Monolayers at Liquid–Gas Interfaces. New York: Interscience; 1966.Google Scholar
- Palsdottir H, Hunte C: Lipid in membrane protein structures. Biochim Biophys Acta 2004, 1666: 2. 10.1016/j.bbamem.2004.06.012View ArticleGoogle Scholar
- Mita T: Lipid-protein interaction in mixed monolayers from phospholipids and protein. Bull Chem Soc Jpn 1989, 62: 3114. 10.1246/bcsj.62.3114View ArticleGoogle Scholar
- Singer SJ, Nicolson GL: The fluid mosaic model of the structure of cell membrane. Science 1972, 175: 720. 10.1126/science.175.4023.720View ArticleGoogle Scholar
- Dynarowicz-Łątka P, Kita K: Molecular interaction in mixed monolayers at the air/water interface. Adv Coll Int Sci 1999, 79: 1. 10.1016/S0001-8686(98)00064-5View ArticleGoogle Scholar
- Charbonneau DM, Tajmir-Riahi HA: Study on the interaction of cationic lipids with bovine serum albumin. J Phys Chem B 2010, 114: 1148. 10.1021/jp910077hView ArticleGoogle Scholar
- Davies JT, Rideal EK: Interfacial Phenomena. New York: Academic; 1963.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.