Anal Bioanal Chem (2009) 395:24052409 DOI 10.

1007/s00216-009-3126-9

TECHNICAL NOTE

Flat hydrogel substrate for atomic force microscopy to observe liposomes and lipid membranes
Akihiko Takagi & Hitomi Hokonohara & Tomoji Kawai

Received: 10 June 2009 / Revised: 5 August 2009 / Accepted: 31 August 2009 / Published online: 6 October 2009 # Springer-Verlag 2009

Abstract In order to avoid denaturation of biomolecules due to strong adsorption on solid surfaces, a soft substrate has to be used for atomic force microscopy (AFM) observation. We propose a hydrophilic agarose gel surface as a soft substrate for AFM to observe liposomes and lipid membranes. Although our simple method does not require any delicate control at the molecular level, an agarose gel surface can be simply flattened to 0.3 nm in roughness using an atomically flat solid surface during gelation. The AFM images revealed that liposomes were unruptured on the gel surface at low liposome density, whereas an unruptured state was difficult to obtain on a solid surface like mica. This indicates that the weak interaction between the liposome and the soft surface inhibits the liposome from rupturing, and also that the surface rougher than the solid surface prevents lateral diffusion of the liposomes along the surface to be fused. Increasing the liposome density resulted in a lipid membrane at various thicknesses forming on the hydrogel surface by the fusion and rupture of liposomes. Using the soft substrate, it can be expected to promote investigations of structures and functions of biomolecules at the nanometer scale under physiological conditions with AFM.

Keywords AFM (atomic force microscopy) . Biopolymer/lipid . Liposome . Hydrogel Abbreviations AFM Atomic force microscopy PEG Polyethylene glycol DLS Dynamic light scattering DPPC 1, 2-dipalmitoyl-sn-glycero-3-phosphocholine DPPG 1,2-dipalmitoyl-sn-glycero-3-[phospho-rac(1-glycerol)]

Introduction Atomic force microscopy (AFM) is used as an instrument for observation of biomolecule functions on the basis of molecule recognition, since it can image biomolecules at molecular resolution in solution under conditions in which they can function. Reactions mediated by transmembrane proteins on a lipid membrane are the targets of observation and evaluation by AFM, because many important biological processes at the interface between the inside and outside of a cell or organelle are expected to be clarified. Although the transmembrane proteins confined in a twodimensional lipid membrane can be fixed in a well-defined orientation by fixing the membrane to the substrate, potential denaturation of the proteins by direct contact to a solid surface should be avoided using an appropriate substrate. It is expected that the environment in which biological processes in a cell occur is reconstructed using artificial lipid membrane formation on the surface. To date, the lipid membrane and its associated proteins have been regarded in terms of a supported membrane [1 6] or a tethered membrane [79]. While the membrane is

A. Takagi The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako, Saitama 351-0198, Japan A. Takagi : H. Hokonohara : T. Kawai (*) The Institute of Scientific and Industrial Research (ISIR), Osaka University, 8-1 Mihogaoka, Ibaraki, 564-0047 Osaka, Japan e-mail: kawai@sanken.osaka-u.ac.jp

2406

A. Takagi et al.

fixed directly on a solid surface in the former case, the latter case has linkers to fix the membrane covalently to the surface. The linkers also work as a spacer to avoid deformation of the proteins due to strong adsorption on the surface, which possibly brings about denaturation of the molecules [10, 11]. A hydrophilic polymer layer using cellulose [12], PEG [13], agarose [14, 15], or other substances [10, 16, 17] has also been used to support a lipid membrane and associated proteins. In electrophysiological studies of ion channel proteins, it was found that the soft supporting layers retained the ion channel proteins functional in lipid membranes by avoiding denaturation [9, 1416], and that the hydrogel supporting layer improved stability of the lipid bilayer against mechanical disruption [18, 19]. In this work, we report on a preparation of a flat agarose gel substrate and AFM observations of liposomes and lipid membranes formed on the substrate. Our simple method does not require any delicate control in a molecular level, i.e., self-assembled monolayer formation, but the hydrogel is simply flattened with an atomically flat solid surface during gelation. Although our preparation method is very simple, the surface is flat enough to observe liposome or lipid membrane at various thicknesses in a nanometer scale with AFM. Using our method, it can thus be expected to promote AFM investigations of structures and functions of biomolecules in nanometer scale under physiological conditions.

AFM measurements were performed at room temperature using commercial AFMs. A Nanoscope IIIa (Veeco Instruments) was used with an intermittent contact mode in ambient air with Si cantilevers (RTESP, resonant frequency f=320 kHz, spring constant k=14 N/m, Veeco Instruments), and SPI3700/SPA300 (Seiko Instruments Inc.) was also used in a contact mode under ambient or liquid conditions with Si cantilevers (OMCL-TR400PSA-1, f=11 kHz in air, k=0.02 N/m, OLYMPUS).

Results and discussion Figure 1a and b, respectively, show the AFM image and cross-sectional profile of an agarose gel substrate at approximately 0.1 mm thickness observed under ambient conditions. The roughness evaluated from the image in Fig. 1a was 0.29 nm (rms), while the surface included small holes due to air bubbles. The simple fabrication method typically allows us to obtain surfaces as flat as 1 nm in roughness. Figure 1c shows an AFM image of the gel surface observed in a buffer solution, and Fig. 1d shows a part of Fig. 1c at high color contrast. Because of swelling, the surface of a gel substrate is generally roughened in buffer relatively to that observed in air, and the holes due to air bubbles are hardly observed. The surface roughness was estimated to be 1.1 nm. The reticulation of the hydrogel polymer fiber bundles becomes more visible in a buffer solution than in air as shown in Fig. 1d [20, 21]. A power spectrum of the topographic image (Fig. 1c) obtained by Fourier transform allows us to estimate a fractal dimension D=(7slope)/2=2.9 from a linear relation in a loglog plot of the power spectral density against a wave number [20, 22]. Since the prepared soft hydrogel surface is flat enough for AFM observation, liposome and lipid membrane can be observed on the surface. The AFM image in Fig. 2a of the surface following liposome deposition shows particles on the hydrogel surface, of which the height and width are estimated to be 336 nm and 28962 nm, respectively. Because of the similar appearance of the particles, they are considered to be the liposomes flattened on the hydrogel surface. In order to estimate the size of the liposome, we supposed the flattened liposome to be cylindrical, and obtained the diameter of the virtually spherical liposome to have the same volume to the cylinder. As a result, the liposome diameter is estimated to be 15927 nm, which is within the range obtained by DLS measurements (195 94 nm) and by the specification provided by the manufacturer of the liposome (100300 nm). The uniform appearance in the AFM images and the estimated diameter as expected therefore indicate that the liposomes neither fused nor ruptured are observed in the AFM images.

Materials and methods Prior to hydrogel formation, a glass plate was pre-coated with agarose mixed with a small quantity of epoxy resin (Araldite, Hanzmann Advanced Material). Agarose (10 wt.%, Nusieve GTG agarose, Cambrex Bio Science Rockland Inc.) was gelated between freshly cleaved mica and the pre-coated glass plate. A flat gel surface was obtained simply by removing the mica after gelation, and the gel was fixed on the glass plate due to the pre-coated layer. Commercially available liposome (COATSOME EL-01-A, NOF Co.) was used, which included 1,2-dipalmitoylsn-glycero-3-phosphocholine (DPPC):cholesterol:1,2dipalmitoyl-sn-glycero-3-[phospho-rac-(1-glycerol)] (DPPG)=3:4:3. The diameter of the liposomes was specified to be 100300 nm by the manufacturer of the liposome. It was also estimated using a dynamic light scattering (DLS) method (Zetasizer Nano-ZS, Malvern Instruments Ltd.). For AFM observations in air, water emulsion of the liposome was dropped onto a gel surface and then left for a few minutes before being shaken off the surface. The buffer used for the AFM observation was HEPES (10 mM), CaCl2 (50 mM), and MgCl2 (5 mM).

Flat hydrogel substrate for AFM Fig. 1 a AFM image (10 10 m2) and b cross-sectional profile of the agarose gel surface observed in air. c AFM images (1010 m2) of the agarose gel surface observed in a buffer solution. d A high-contrast image (55 m2) of a part of c

2407

Fig. 2 AFM images (1010 m2) and cross-sectional profiles of a liposomes and b, c lipid membranes formed by the fusion and rupture of liposomes on the agarose gel surface in air. Liposome density increased from a to c

2408

A. Takagi et al.

When more liposome was adsorbed as shown in Fig. 2b than in Fig. 2a, a thin membrane homogeneously covering almost the entire area of the image is observed. The thickness of the membrane is estimated to be approximately 10 nm from the cross-sectional profile, which corresponds to two single-lipid bilayers. A small number of particles were observed in the area where no membrane had formed. Since the appearance and the size of the particles are similar to those observed in Fig. 2a, the observed particles are considered to be unruptured liposomes. The observations indicate that high liposome density on the surface makes most of them fused to form the lipid membrane on the hydrogel surface. As shown in Fig. 2c, the hydrogel surface was covered with thick multiple layers of lipid membranes with further increasing the liposome density on the surface. The observed spherical structures at the various sizes, much larger than the liposomes observed in Fig. 2a, represented liposomes grown by fusion. Lipid membrane could thus be formed on the soft hydrogel substrate depending on the deposited liposome quantity. It is considered that the liposomes are adsorbed on the soft hydrogel surface in the different way from that on a mica surface, which is widely used as a hydrophilic solid substrate for supported membrane formation. On a mica surface, individual liposomes are very difficult to be observed, whereas circular patches of lipid membrane can be observed very often [46]. It indicates that the large interaction due to the strong electrostatic adhesion between liposome and the surface promotes to rupture the liposome and therefore to form membrane. Since the very flat mica surface seems to allow rapid diffusion of the membrane due to the low energy barrier for lateral diffusion of lipids along the surface, the membrane laterally grows by fusion and the circular shape results from minimization of its boundary energy. On the other hand, because the agarose surface is very hydrophilic but is not charged, the interaction between lipid and the surface is not strong enough to either trap the liposome at the surface or to rupture it. However, the individual liposomes can be observed to remain on the surface at low liposome density when the surface is dried, because the rough surface due to polymer reticulation inhibits lateral diffusion and, therefore, fusion of liposomes. It hence seems that a lipid membrane is formed on the surface by liposome fusion only when the liposomes adsorb close to each other at high density. In brief, liposomes are not likely either to adsorb on the hydrogel soft surface or to diffuse laterally, whereas they are likely to adsorb on mica and to diffuse along the surface. Although it is experimentally difficult to observe liposome and lipid membrane on the soft hydrogel surface in a buffer solution at present because of the weak interaction between the liposome and the soft hydrogel surface, the soft hydrogel surface is more advantageous than the solid surface for observing the

individual unruptured liposomes and the lipid membrane with AFM. By supporting lipid membranes with biomolecules on a hydrogel substrate, the fragile membranes can be stabilized as demonstrated in gel-protected devices [18, 19], and denaturation of the biomolecules due to strong adsorption can be avoided. The fact that AFM can observe a lipid membrane on a hydrogel surface therefore raises the strong possibility that the proteins associated with the lipid membrane can be observed by AFM while maintaining the function of the proteins. Furthermore, since the scaffold of polymer fiber reticulation can be expected to suppress rapid lateral diffusion of the supported lipid membrane, as the cytoskeleton does [23], it is an advantage for observing biomolecules associated with the membrane by AFM at single-molecular resolution. In this work, we reported on a simple method to prepare a hydrogel surface as flat as a few tenths of a nanometer especially for AFM observation. Observations showed that liposomes were unruptured on the gel surface, a condition difficult to obtain on a solid surface. Increasing the liposome density resulted in lipid membranes forming on the hydrogel surface by the fusion and rupture of liposomes. Using the method, the physiological environment, in which lipid membrane and the associated biomolecules retain their intact functions, is expected to be constructed in order to promote investigations of structures and functions of biomolecules in a nanometer scale with AFM.

References
1. Watts TH, Brian AA, Kappler JW, Marrack P, McConnell H (1984) Antigen presentation by supported planar membranes containing affinity-purified I-A. Proc Natl Acad Sci 81:75647568 2. Reviakine I, Brisson A (2000) Formation of supported phospholipid bilayers from unilamellar vesicles investigated by atomic force microscopy. Langmuir 16:18061815 3. Johnston LJ (2007) Nanoscale imaging of domains in supported lipid membranes. Langmuir 23:588658953 4. Geisse NA, Wsle B, Salowsky DE, Henderson RM, Edwardson JM (2002) Syncollin homo-oligomers associated with lipid bilayers in the form of doughnut-shaped structures. J Membr Biol 189:8392 5. Tokumasu F, Jin AJ, Feigenson GW, Dvorak JA (2003) Atomic force microscopy of nanometric liposome adsorption and nanoscopic membrane domain formation. Ultramicroscopy 97 (14):217227 6. Simonsen AC, Bagatolli LA (2004) Structure of spin-coated lipid films and domain formation in supported membranes formed by hydration. Langmuir 20:97209728 7. Heyse S, Vogel H, Snger M, Siegrist H (1995) Covalent attachment of functionalized lipid bilayers to planar waveguides for measuring protein binding to biomimetic membranes. Protein Sci 4:25322544 8. Lei SB, Tero R, Misawa N, Yamamura S, Wan LJ, Urisu T (2006) AFM characterization of gramicidin-A in tethered lipid membrane on silicon surface. Chem Phys Lett 429:244249

Flat hydrogel substrate for AFM 9. Schiller SM, Naumann R, Lovejoy K, Kunz H, Knoll W (2003) Archaea analogue thiolipids for tethered bilayer lipid membranes on ultrasmooth gold surfaces. Angew Chem Int Ed 42:208211 10. Knoll W, Frank CW, Heibel C, Naumann R, Offenhusser A, Rhe J, Schmidt EK, Shen WW, Sinner A (2000) Functional tethered lipid bilayers. Rev Mol Biotech 74:137158 11. Tanaka M, Sackmann E (2005) Polymer-supported membranes as models of the cell surface. Nature 437:656663 12. Goennenwein S, Tanaka M, Hu B, Moroder L, Sackmann E (2003) Functional incorporation of integrins into solid supported membranes of ultrathin films of cellulose: impact on adhesion. Biophys J 85:646655 13. Wagner ML, Tamm LK (2000) Tethered polymer-supported planar lipid bilayers for reconstitution of integral membrane proteins: silanepolyethyleneglycol-lipid as a cushion and covalent linker. Biophys J 79:14001414 14. Hongyo K, Joseph J, Huber RJ, Janata J (1987) Experimental observation of chemically modulated admittance of supported phospholipid membranes. Langmuir 3:827830 15. Ide T, Yanagida T (1999) An artificial lipid bilayer formed on an agarose-coated glass for simultaneous electrical and optical measurement of single ion channels. Biochem Biophys Res Commun 265:595599 16. Arya A, Krull UJ, Thompson M, Wong HE (1985) Langmuir Blodgett deposition of lipid films on hydrogel as a basis for biosensor development. Anal Chem Acta 173:331336

2409 17. Lunelli L, Pasquardini L, Pederzolli C, Vanzetti L, Anderle M (2005) Covalently anchored lipid structures on amine-enriched polystyrene. Langmuir 21:83388343 18. Costello RF, Peterson IR, Heptinstall J, Byrne NG, Miller LS (1998) A robust gel-bilayer channel biosensor. Adv Mater Opt Electron 8:4752 19. Beddow JA, Peterson IR, Heptinstall J, Watson DJ (2004) Reconstitution of nicotinic acetylcholine receptors into gelprotected lipid membranes. Anal Chem 76:22612265 20. Suzuki A, Yamazaki M, Kobiki Y (1996) Direct observation of polymer gel surfaces by atomic force microscopy. J Chem Phys 104(4):17511757 21. Nitta T, EndoY HH, Kawabata K (2003) Microdomain structure of agar gels observed by mechanical-scanning probe microscopy. J Electron Microsc 52(3):277281 22. Feder J (1988) Fractals. Plenum, New York 23. Kusumi A, Nakada C, Ritchie K, Murase K, Suzuki K, Murakoshi H, Kasai RS, Kondo J, Fujiwara T (2005) Paradigm shift of the plasma membrane concept from the two-dimensional continuum fluid to the partitioned fluid: high-speed single-molecule tracking of membrane molecules. Annu Rev Biophys Biomol Struct 34:351378 This work was supported by a grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan (the Promotion of Novel Interdisciplinary Fields).