Biomimetic membrane platform: Fabrication, characterization and applications

Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Biomimetic membrane platform: Fabrication, characterization and applications Ahu Arslan Yildiz a,∗ , Umit Hakan Yildiz b , Bo Liedberg b,c , Eva-Kathrin Sinner a,∗∗ a

Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany Center for Biomimetic Sensor Science, School of Materials Science and Engineering, Nanyang Technological University, Singapore c Department of Physics, Chemistry and Biology, Division of Molecular Physics, Linköping University, 58183 Linköping, Sweden b

a r t i c l e

i n f o

Article history: Received 25 August 2012 Received in revised form 20 October 2012 Accepted 24 October 2012 Available online 14 November 2012 Keywords: Tethered lipid membrane Self-assembly ␮-Contact printing ␣-Laminin peptide cushion Cytochrome bo3 ubiquinol oxidase

a b s t r a c t A facile method for assembly of biomimetic membranes serving as a platform for expression and insertion of membrane proteins is described. The membrane architecture was constructed in three steps: (i) assembly/printing of ␣-laminin peptide (P19) spacer on gold to separate solid support from the membrane architecture; (ii) covalent coupling of different lipid anchors to the P19 layer to serve as stabilizers of the inner leaflet during bilayer formation; (iii) lipid vesicle spreading to form a complete bilayer. Two different lipid membrane systems were examined and two different P19 architectures prepared by either self-assembly or ␮-contact printing were tested and characterized using contact angle (CA) goniometry, surface plasmon resonance (SPR) spectroscopy and imaging surface plasmon resonance (iSPR). It is shown that surface coverage of cushion layer is significantly improved by ␮-contact printing thereby facilitating bilayer formation as compared to self-assembly. To validate applicability of proposed methodology, incorporation of Cytochrome bo3 ubiquinol oxidase (Cyt-bo3 ) into biomimetic membrane was performed by in vitro expression technique which was further monitored by surface plasmon enhanced fluorescence spectroscopy (SPFS). The results showed that solid supported planar membranes, tethered by ␣-laminin peptide cushion layer, provide an attractive environment for membrane protein insertion and characterization. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Biological membranes act like a simple barrier to separate and protect the cell and its components from the surrounding milieu. They also serve as a convenient environment to host macromolecular entities capable of managing a wide range of biological processes such as active transport, signaling, and energy dissipation. The biological activities of cell membranes are predominantly driven by membrane proteins which have attracted considerable research attention during the last decades. The utilization of “natural” biological membranes for systematic investigation of membrane proteins is often obstructed by their intrinsic complexities and heterogeneities. For this reason, employing biomimetic membrane structures such as supported lipid bilayers (SLBs) has become a commonly accepted methodology for investigating membrane proteins [1–5]. Their spontaneous formation on planar substrates makes them ideal for structural and functional studies of

∗ Corresponding author. Present address: Institute of Materials Research and Engineering, IMRE, 3 Research Link, 117602 Singapore, Singapore. ∗∗ Corresponding author. Present address: University of Natural Resources and Applied Life Sciences (BOKU), Muthgasse 11, 1190 Vienna, Austria. E-mail addresses: [email protected] (A. Arslan Yildiz), [email protected] (E.-K. Sinner). 0927-7765/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2012.10.066

membrane proteins by various surface sensitive techniques [6,7]. A carefully designed surface chemistry is, however, crucial to prevent a common problem “protein denaturation” which is induced by interaction of proteins with solid support [8,9]. Recently a new sub-class of solid supported membrane systems has been introduced [10–15]; namely tethered bilayer lipid membranes (tBLM), in which the lipid bilayer is separated from the substrate surface by insertion of a flexible layer of “tethering” molecules. This layer provides a reservoir underneath the bilayer in which the membrane proteins can fold in a native-like conformation. The large variety of assembling molecules capable of forming a tethering layer offers multiple possibilities for fine tuning the properties of tBLMs. For example, peptide anchors have proven to provide an excellent environment for the membrane proteins possessing large extra cellular domains [16,17]. In most of these studies low surface coverage and therefore heterogeneity is a common problem if tethering molecule allowed to self assemble. In this work we aim to cope with this problem by employing ␮-contact printing, as an alternative fabrication technique for the formation of homogeneous cushion layer. Therefore, two different tBLM platforms are prepared either by self-assembly or ␮-contact printing to characterize the membrane platform using a range of surface sensitive techniques followed by Cytochrome bo3 ubiquinol oxidase (Cyt-bo3 ) insertion. Our recent findings show that surface coverage of cushion layer was increased by using

A. Arslan Yildiz et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

␮-contact printing that improves the efficiency of bilayer formation significantly, and therefore facilitates adoption of proposed methodology to generate microarray type bilayer architecture for membrane protein insertion studies. 2. Experimental procedures 2.1. Materials ␣-Laminin peptide (P19) was purchased from Sigma–Aldrich and used without further purification. Chemical reagents for amine coupling reaction N-hydroxysuccinimide (NHS) and N-(3-dimethylaminopropyl)-N -ethylcarbodiimide hydrochloride (EDC) were purchased from Fluka and used without any further purification. Phospholipids forming monolayer 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine (DMPE), 1,2dipentadecanoyl-sn-glycero-3-phosphoethanolamine (DPePE), 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine (DPPE), 1,2-diphytanoyl-sn-glycero-3-phosphoethanolamine (DPhyPE), 1,2-dipalmitoleoyl-sn-glycero-3-phosphoethanolamine (DPalPE), (DLPE), 1,2-dilinolenoyl-sn-glycero-3-phosphoethanolamine 1,2-diarachidonoyl-sn-glycero-3-phosphoethanolamine (DAPE), 1,2-didocosahexaenoyl-sn-glycero-3-phosphoethanolamine (DDPE) and E. coli total extract lipids (TE) were purchased from Avanti Polar Lipids Inc. and used without further purification. l-␣-Phosphatidylcholine from soybean (PC) was purchased from Fluka and used without purification. Dulbecco’s Phoshate Buffered Saline 1X, pH 7.4 (PBS) was purchased from GIBCO/Invitrogen. LDS sample buffer was purchased from Invitrogen.

511

assembly was conducted directly inside the flow-cell, step by step as explained. Each layer addition was monitored by SPR, layer thicknesses were calculated according to Fresnel equation as described previously [22] and fitting has been done for the obtained SPR scan curves by using Winspall (MPIP, Mainz) software. Theoretical calculations of the layer thicknesses were done by Chem3D Ultra 7.0 software assuming fully stretched molecules. Thereafter, in vitro expression technique was applied immediately for the insertion of Cyt-bo3 . For fluorescence detection 200 ␮l (3%) of PentaHis Alexa Fluor 647 conjugate primary antibody solution was passed through the surface and incubation was completed in 60 min. Blocking method was used to prevent unspecific interaction of antibody with the lipid membrane surface. During measurements, SPR scan was performed before and after addition of a new layer or protein. 2.4. Contact angle measurements Contact angle (CA) experiments were performed using DSA10 setup (Krüss CA, Germany). OCA software (DataPhysics, Filderstadt, Germany) was used for calculation of data and analysis of the contact angle. The membrane surfaces were prepared as explained above. The area of each slide was around 1.0 cm2 and the volume of the flow cell was kept as a 1.0 ml to cover the entire surface. The sessile drop method was used and 3 ␮l drop was dispensed on the surface recorded via CCD camera. Each set of data has been obtained from at least five individual measurements. Mean values and standard deviations were calculated via Origin 7.5 software.

2.2. Membrane preparation 2.5. Imaging SPR measurements 2.2.1. Self-assembly Self-assembly of the tBLM was performed step by step as follows; self-assembly of P19 spacer were completed in 45 min by incubating gold surface with 0.01 mg/ml peptide solution (pH: 7.0) followed by activation of the terminal COOH groups by using 0.4 M EDC and 0.1 M NHS for 10 min. 0.2 mg/ml DMPE solution (in PBS with 0.1% TritonX-100) was added and incubation was completed in 60 min. Afterwards, either PC or TE (1.0 mg/ml) vesicles were added and incubated for 90 min to form the bilayer. 50 nm vesicles were prepared by extrusion method, the lipid solution was passed through the 50 nm polycarbonate (Avestin) membranes 21 times. After each addition and incubation step, rinsing was performed by using PBS buffer (pH 7.4). All steps were carried out at room temperature.

In situ layer by layer (LBL) assembly of DMPE, TE and PC on peptide tethered gold substrate were monitored by iSPR. The setup is from Nanofilm – EP3, Accurion, Göttingen, Germany. The SPR cell utilizes the Kretschmann setup for SPR measurements and consists of a holder for a 60◦ BK7 prism and a flow cell. The SPR surfaces consisted of 12 mm × 12 mm glass slides coated with a 45 nm thick film of gold. The gold surfaces were obtained as a generous gift from GE-Healthcare (Biacore), Uppsala, Sweden. Measurements were performed at a wavelength of 741 nm, selected from the spectrum of the xenon lamp by an interference filter. The angle of incidence was 70◦ for the iSPR measurements. 2.6. Expression and insertion of Cyt-bo3

2.2.2. -Contact printing P19 was transferred to the gold surface using ␮-contact printing using square features (side length 50 ␮m) and circular features (25 ␮m in diameter). Conditions for PDMS stamp preparation and for ␮-contact printing have been described elsewhere [18–21]. The remaining bare gold surface after ␮-contact printing was backfilled with a blocking agent 11-mercapto-1-undecanol (MUD) to provide a non-adhesive surface. The rest of the assembly process is identical to the one described above.

E. coli (C43) cells carrying pETcyo plasmid, gift from Prof. R.B. Gennis, University of Illinois, were used in this study. High quality plasmid DNA was prepared with PureYield Plasmid Midiprep System, Promega. Concentrated plasmid DNA was used directly for in vitro expression as explained elsewhere [14]. Expression kinetic has been monitored in situ via SPR and right after proceeded to SPFS detection using Alexa Fluor 647 conjugate antibody.

2.3. Surface plasmon resonance (SPR) and surface plasmon enhanced fluorescence spectroscopy (SPFS)

3.1. Biomimetic membrane preparation

For the SPFS experiments a home-made SPR setup (MPIP, Mainz) and home-made flow cell in Kretschmann configuration which has a 50 ␮l volume was used. LaSFN9 glass substrates which has a refractive index value n = 1.845 (Schott AG) were used as a substrate for gold evaporation. 50 nm gold layer was deposited on LaSFN9 slides by thermal evaporation. Membrane

3. Results and discussion

Layer-by-layer tBLM formation on gold surface is illustrated in Fig. 1. The membrane architecture was prepared in three steps: (i) self-assembly/printing of ␣-laminin peptide (P19) spacer on gold; (ii) covalent coupling of different lipid anchors to the P19 layer; (iii) lipid vesicle spreading to form a complete bilayer. Two different lipid membrane systems (P19/DMPE/PC and P19/DMPE/TE) were tested and fully characterized using CA goniometry, SPR, and iSPR.

512

A. Arslan Yildiz et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

Fig. 1. Proposed tethered bilayer lipid membrane formation steps.

The bilayer formation was mainly characterized by SPR. Recording was done in scan mode (full SPR curve) to obtain the thickness after each step in the layer-by-layer assembly process (Fig. 2). The shift seen in the angular SPR curve is due to the introduction of a new layer with a thickness that has been calculated using optical fitting of the SPR curve by assigning refractive indices given in Table 1. 3.2. Peptide tethered monolayer The three step process started by assembling the P19 peptide spacer that offers a hydrophilic cushion-like surface. The flexible and hydrophilic nature of the peptide spacer provides a natural reservoir between the bilayer structure and solid support and it eases the incorporation of proteins with protruding membrane domains. Initial characterization of P19 layer was performed by CA goniometry. Subsequent to P19 peptide (0.01 mg/ml) addition, the CA values varied in between 40 and 44◦ and did not change significantly upon prolonged incubation times. At lower concentrations of P19 (≤0.01 mg/ml) moderate increase was observed in CA values with respect to 0.01 mg/ml (Table SI1). However, the CA values for 0.1 mg/ml and 0.01 mg/ml were found to be nearly same,

and they did not vary substantially with incubation time. Based on these results, optimum P19 concentration of 0.01 mg/ml and 45 min incubation time was chosen for the assembly of P19 tethering layer. The corresponding SPR kinetic curves for P19 assembly (Fig. SI1) also confirmed that observed reflectivity changes upon assembly of peptide layer rapidly reached saturation for 0.01 and 0.1 mg/ml. Furthermore, the angular shift and thereby the layer thickness and coverage varied marginally between 0.01 and 0.1 mg/ml (after rinsing). CA characterization for various unsaturated and branched lipid anchors (DMPE, DPePE, DPPE, DPalPE, DPhyPE, DLPE, DAPE and DDPE) was also performed to optimize the monolayer formation in terms of packing density and surface coverage. As shown in Table SI2, the longer the lipid anchors, the higher the CA [23,24]. However owing to the low solubility of unsaturated and branched alkyl chains, mostly inhomogeneous layer formation was observed. Among the all lipids, only DMPE exhibits efficient monolayer properties therefore it is used for rest of the study as a model anchor. Furthermore, low transition temperature of DMPE is preferable for spontaneous bilayer formation [13] which can be induced by in vitro expression at 37 ◦ C. Subsequent SPR and SPFS examinations (Figs 2 and 5) confirmed that the DMPE layer is a

Table 1 SPR fitting parameters [28–30], theoretical and calculated thicknesses. Layer

Refractive index/n

Extinction coefficient/k

Theoretical thickness (nm)

Calculated thickness (nm)

LaSFN9 slide Gold Air MilliQ/buffer P19 DMPE PC TE

1.85 0.2 1 1.3 1.4 1.33 1.4 1.4

– 3.4 0 0 0 0.22 0 0

0 (∞) 50 0 (∞) 0 (∞ 1.1 3.1 4.1 5.5

– 48–49 – – 1.09 0.89 4.18a 5.84a

a

Thicknesses were obtained after subtracting of P19 layer thickness from overall thickness (see Fig. 1 for definitions of the different layer thicknesses dP19 , dDMPE , dPC/TE .).

A. Arslan Yildiz et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

513

3.3. Assembly of bilayer Although the formation of the bilayer by fusion of lipid vesicles on partly hydrophobic substrates is not yet fully understood, a few models have been proposed to explain the mechanisms [25]. After assembly of peptide tethered monolayer, two different types of lipids were used separately to form the outer leaflet of the bilayer. PC from soybean was compared with TE and both lipids were examined to see their effect on tBLM formation and insertion of membrane proteins. TE was chosen as a model for mixed lipids which is composed of 57.5% phosphatidylethanolamine, 15.1% phosphatidylglycerol, 9.8% cardiolipin and other lipids. Previous studies of several research groups [26,27] have shown that surface roughness influences vesicle adsorption and bilayer formation, as well as protein adsorption and insertion. In this study we address the effect of different lipids on tBLM formation and protein insertion. Fig. 2 shows the angular SPR scan curves of PC and TE addition on top of P19/DMPE peptide tethered monolayer. As given in Fig. 2, SPR scan shifts of resonance angle () were ∼1.0◦ and ∼1.5◦ for PC and TE, respectively. Compared to theoretical values, slightly higher  values indicate partial vesicle adsorption (around 30%) which is directly related with poor surface coverage and poor coupling of previous layer as mentioned above. The inset graphs show the kinetics upon vesicle addition at fixed angle. Membrane patch formation and partial vesicle adsorption are also observed by monitoring slightly high reflectivity change in kinetic curves. It should be noted that SPR only shows lateral average thickness so heterogeneity cannot be calculated via SPR. 3.4. Detection of tBLM formation via iSPR

Fig. 2. SPR angular scan curves for layer-by-layer self assembled tBLM: (a) ()DMPE, (䊉) PC for P19/DMPE/PC membrane, (b) ()DMPE, () TE for P19/DMPE/TE membrane, and SPR angular scan curves for tBLMs on micropatterned tethering layer: (c) ()DMPE, (䊉) PC, () TE.

suitable candidate for tBLM formation and for membrane protein insertion. Covalent attachment of DMPE to P19 peptide via EDC-NHS coupling was performed to form the peptide-tethered monolayer. The comparatively low hydrophobicity of DMPE layer based on CA measurements indicates lower packing density and surface coverage. These results were also supported with SPR data (Table 1) and the calculated surface coverage for the DMPE layer was ∼30%. As shown in Fig. 2, SPR scan shifts of resonance angle () was only ∼0.3–0.4◦ which is below theoretical data. This can be explained by incomplete P19 assembly and incomplete coupling of the peptide spacer as reported previously [13]. However, further thickness increase was observed due to the lipid vesicle addition which will be discussed in the following sections.

To overcome lateral resolution limitation of SPR, imaging SPR (iSPR) in ellipsometric mode used for characterization and thickness determination of biomimetic membranes. For the preparation of defect-free membrane platform ␮-contact printing technique was used to improve the surface chemistry. Layer-by-layer assembly of peptide tethered monolayer and corresponding lipids were monitored and micropatterned surfaces were characterized by a thickness increase of each layer. Figs. 3 and 4a shows the thickness map of the P19 layer (note that non patterned regions were backfilled by MUD molecules to reduce the nonspecific adsorption phenomena during the subsequent steps). Based on our calculation, the thickness of the P19 layer was found to equal 1.5 ± 0.1 nm [31,32], in fair agreement with the theoretical thickness of P19 (see Table 1). A series of measurements revealed that ␮-contact printing of P19 was about 0.4 nm thicker than those prepared by self-assembly. As discussed by Zhou et al. [33] a significant portion of the printed layer may be composed of loosely bound molecules e.g. multilayers of the inking molecules and/or contaminants from the PDMS stamp. Such a loosely bound layers normally can be eliminated by ultrasonication after printing. However, ultrasonication did not change the layer thickness in our case. Therefore, the higher P19 thickness is not attributed above-mentioned reasons, but could be due to an improvement in transfer efficiency provided by ␮-contact printing. This is also supported by CA data of micropatterned P19 surface which is around 35.5◦ confirming increased hydrophilicity. In the following step, DMPE was covalently attached on to the P19 layer as described in previous section. The thickness increased by 0.8 ± 0.2 nm (Figs. 3 and 4b) after formation of the DMPE layer. Measured CA value of DMPE layer was 74.1◦ which is moderately hydrophobic as compared to self assembled counterpart. Significant changes were observed after exposure to TE and PC vesicle suspensions (Figs. 3 and 4c). The incremental thicknesses increase for the TE and PC layers

514

A. Arslan Yildiz et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

Fig. 3. Thickness map of (a) printed ␣-laminin peptide (P19) array; (b) 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine; (c) E. coli total extract (TE) and (d) height profile of the layers (height profile analysis performed along the line of which direction shown by red, blue, black arrows). (For interpretation of references to color in this figure legend, the reader is referred to the web version of this article.)

were found to equal 4.7 ± 0.5 nm and 3.8 ± 0.4 nm, respectively, which corresponds to bilayer formation [34–37]. SPR scan shifts of resonance angles () were 0.68 and 0.90◦ for PC and TE layers on micropatterned peptide tethered monolayer (Fig. 2c). In contrast to self-assembly, ␮-contact printing of cushion layer followed by DMPE anchoring exhibits lower angular shifts. This result

considers ␮-contact printing induces bilayer formation rather than vesicle adsorption with respect to self-assembly process. Moreover, kinetic investigation (Fig. SI2) of layer formation shows that unlike bare gold surface, P19/DMPE layer induces homogenous lipid layer formation which yield characteristic thickness increase associated to bilayer.

Fig. 4. Thickness map of (a) printed ␣-laminin peptide (P19) array; (b) 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine; (c) l-␣-phosphatidylcholine from soybean (PC) and (d) height profile of layers height profile of the layers (height profile analysis performed along the line of which direction shown by red, blue, black arrows). (For interpretation of references to color in this figure legend, the reader is referred to the web version of this article.)

A. Arslan Yildiz et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

515

reducing agents to solubilize artificial membrane and its content for further the characterization by immunoblotting in order to confirm the integration of target protein. Western blot images (Fig. SI3) show the characteristic double-band of Cyt-bo3 at 33–35 kDa. Similar findings have been reported previously [38] for in vivo expressed protein. Thus, the obtained results are in good agreement with previous reports by using two different membrane architectures. Overall western blot analysis confirms that Cyt-bo3 can be successfully expressed and inserted into an artificial membrane platform. 4. Conclusion

Fig. 5. SPFS data analysis for (䊉) P19/DMPE/PC/in vitro Cyt-bo3 pETcyo plasmid/Blocking Solution/Alexa Fluor 647 conjugated 1◦ Ab, ()P19/DMPE/TE/in vitro Cyt-bo3 pETcyo plasmid/Blocking Solution/Alexa Fluor 647 conjugated 1◦ Ab and (+) P19/DMPE/PC/in vitro extract without DNA/Blocking Solution/Alexa Fluor 647 conjugated 1◦ Ab (control experiment). SPR scan curves for P19/DMPE/PC (dashed line) and P19/DMPE/TE (solid line).

Although there are not many reports about thickness determination and characterization of TE bilayers, the consistency between the estimated thicknesses (obtained from SPR and iSPR) and the theoretical thicknesses, suggests that our estimations are realistic. In overall, iSPR determination of tethered lipid bilayers for two different systems reveals that, bilayer formation depends on the properties and perfection of underneath layer. As compared to self-assembly ␮-contact printing provide high surface coverage for P19 layer which predominantly induce the efficient attachment of anchor molecules and subsequent spontaneous bilayer formation. Overall, micropaterning increases the surface coverage of P19 peptide while maximizing homogeneity of subsequent DMPE layer compared to self-assembly. 3.5. Insertion of Cyt-bo3 in tBLM and detection by SPFS To validate the usefulness of proposed methodology two different lipid membrane systems made of TE and PC were further utilized for expression and incorporation of membrane protein “Cyt-bo3 ”. Insertion of Cyt-bo3 was monitored by SPFS technique which enhances the detection limit of SPR. In order to detect real time expression and insertion, biomimetic membrane system and ELISA technique which utilizes the conjugated primary antibody (1◦ Ab) to detect target protein inside the membrane platform were combined. Fig. 5 shows the typical angular fluorescent scan curves for two systems; P19/DMPE/PC and P19/DMPE/TE of which corresponding fluorescent maximum were found to be 3 × 105 and 2.5 × 105 cps, respectively. Observed angular scan curve shifts for the P19/DMPE/PC and P19/DMPE/TE systems were slightly different due to thickness difference of PC and TE layers which is also reflected in fluorescent curves. The control experiment (no DNA) which exhibit zero intensity, confirms that the observed signal for both systems originate from expressed and integrated membrane protein. The nearly identical signal intensities in fluorescence show that both systems favor the expression and insertion of Cyt-bo3 . Besides SPFS analysis, immunoblotting analysis was done to confirm insertion of the protein into the biomimetic membrane platform. SPR only provides information for increase in optical mass, therefore no specificity involved and detection of insertion is not possible by using either SPR or SPFS. As reported previously [14], flow cell content was eluted after SPFS measurement with SDS-gel loading buffer which contains strong detergents and

The results showed that solid supported planar membranes, tethered by ␣-laminin peptide cushion layer, provide an attractive environment for insertion and characterization of membrane proteins by surface sensitive techniques such as SPR and SPFS. The SPFS results support the idea of membrane protein insertion in artificial membrane platform which can provide native-like environment for the characterization of membrane proteins. It was also shown that surface coverage can be increased and membrane structure can be improved by applying ␮-contact printing technique. Moreover, iSPR data show that formation of homogeneous bilayer is significantly better on micropatterned P19 surface than self assembled P19 surface. Therefore, micropatterned membrane systems could be a new candidate for membrane protein based biochips. Acknowledgments We thank the support from A*STAR JCOGrant (10/03/FG/06/06) of Singapore. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi: 10.1016/j.colsurfb.2012.10.066. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]

A.A. Brian, H.M. McConnell, Proc. Natl. Acad. Sci. U.S.A. 81 (1984) 6159. L.K. Tamm, H.M. McConnell, Biophys. J. 47 (1985) 105. E. Sackmann, Science 271 (1996) 43. E.M. Erb, K. Tangemann, B. Bohrmann, B. Müller, J. Engel, Biochemistry 36 (1997) 7395. A. Kloboucek, A. Behrisch, J. Faix, E. Sackmann, Biophys. J. 77 (1999) 2311. T. Watts, H. Gaub, H.M. McConnell, Nature 320 (1986) 179. J. Salafsky, J. Groves, S. Boxer, Biochemistry 35 (1996) 14773. S. Goennenwein, M. Tanaka, B. Hu, L. Moroder, E. Sackmann, Biophys. J. 85 (2003) 646. O. Purrucker, S. Gonnenwein, A. Fortig, R. Jordan, M. Rusp, M. Barmann, L. Moroder, E. Sackmann, M. Tanaka, Soft Matter 3 (2006) 333. W. Knoll, C.W. Frank, C. Heibel, R. Naumann, A. Offenhauser, J. Rühe, E.K. Schmidt, W.W. Shen, A. Sinner, Rev. Mol. Biotechnol. 74 (2000) 137. B. Raguse, V. Braach-Maksvytis, B.A. Cornell, L.G. King, P.D.J. Osman, R.J. Pace, L. Wieczorek, Langmuir 14 (1998) 648. E.K. Sinner, W. Knoll, Curr. Opin. Chem. Biol. 5 (2001) 705. R. Naumann, A. Jonczyk, R. Kopp, J. Esch, H. Ringsdorf, W. Knoll, P. Graber, Angew. Chem. Int. Ed. 34 (1995) 2056. A. Arslan Yildiz, W. Knoll, R.B. Gennis, E.K. Sinner, Anal. Biochem. 423 (2012) 39. I. Köper, Mol. BioSyst. 3 (2007) 651. W. Knoll, I. Köper, R. Naumann, E.K. Sinner, Electrochim. Acta 53 (2008) 6680. S. Schiller, R. Naumann, K. Lovejoy, H. Kunz, W. Knoll, Angew. Chem. Int. Ed. 42 (2003) 208. G. Klenkar, B. Liedberg, Anal. Bioanal. Chem. 391 (2008) 1679. O. Andersson, H. Nikkinen, D. Kanmert, K. Enander, Biosens. Bioelectron. 24 (2009) 2458. G. Klenkar, R. Valiokas, I. Lundstrom, A. Tinazli, R. Tampe, J. Piehler, B. Liedberg, Anal. Chem. 78 (2006) 3643. R. Valiokas, G. Klenkar, A. Tinazli, R. Tamp, B. Liedberg, J. Piehler, Chembiochem 7 (2006) 1325. N. Bunjes, E.K. Schmidt, A. Jonczyk, F. Rippmann, D. Beyer, H. Ringsdorf, P. Graber, W. Knoll, R. Naumann, Langmuir 13 (1997) 6188. D.S. Sutherland, M. Broberg, H. Nygren, B. Kasemo, Macromol. Biosci. 1 (2001) 270.

516

A. Arslan Yildiz et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 510–516

[24] C.D. Bain, E.B. Troughton, Y. Tao, J. Evall, G.M. Whitesides, R.G. Nuzzo, J. Am. Chem. Soc. 111 (1989) 321. [25] E. Kalb, S. Frey, L.K. Tamm, Biochim. Biophys. Acta 1103 (1992) 307. [26] V. Atanasov, P.P. Atanasov, I.K. Vockenroth, N. Knorr, I. Koper, Bioconjugate Chem. 17 (2006) 631. [27] F.A. Denis, P. Hanarp, D.S. Sutherland, J. Gold, C. Mustin, P.G. Rouxhet, Y.F. Dufrene, Langmuir 18 (2002) 819. [28] C. Breffa, New synthetic strategies to tethered bilayer lipid membranes, Dissertation Thesis, The Johannes Gutenberg University of Mainz, 2005. [29] I.K. Vockenroth, Investigation of tethered bilayer lipid membranes for their potential to use in biosensing devices, Dissertation Thesis, University of Bath, 2007. [30] www.luxpop.com

[31] Corresponding “n” values of each layers is listed in Table 3, k was choosen as 0 for all calculations except 0.22 for DMPE layer. [32] Note that absolute thickness is sum of the monolayer thickness of HS(CH2 )12 OH and thickness of protrusion on the patterns press. [33] Y. Zhou, R. Valiokas, B. Liedberg, Langmuir 20 (2004) 6206. [34] N. Kucerka, Y.F. Liu, N.J. Chu, H.I. Petrache, S.T. Tristram-Nagle, J.F. Nagle, Biophys. J. 88 (2005) 2626. [35] M.C. Howland, A.W. Szmodis, B.B. Sanii, A.N. Parikh, Biophys. J. 92 (2007) 1306. [36] B.A. Lewis, D.M. Engelman, J. Mol. Biol. 166 (1983) 211. [37] P. Balgavy, M. Dubnickova, N. Kucerka, M.A. Kiselev, S.P. Yaradaikin, D. Uhrikova, Biochim. Biophys. Acta 1512 (2001) 40. [38] J. Ma, A. Katsonouri, R.B. Gennis, Biochemistry 36 (1997) 11298.