Mixed Self-Assembled Monolayers with Terminal Deuterated Anchors

Article Cite This: J. Phys. Chem. B 2018, 122, 8201−8210

pubs.acs.org/JPCB

Mixed Self-Assembled Monolayers with Terminal Deuterated Anchors: Characterization and Probing of Model Lipid Membrane Formation Hung-Hsun Lee,† Martynas Gavutis,‡ Ž ivilė Ruželė,‡ Ramu̅nas Valiokas,‡ and Bo Liedberg*,†,§ †

Division of Molecular Physics, Department of Physics, Chemistry and Biology, Linköping University, 581 83 Linköping, Sweden Department of Nanoengineering, Center for Physical Sciences and Technology, Savanorių 231, LT-02300 Vilnius, Lithuania

‡

J. Phys. Chem. B 2018.122:8201-8210. Downloaded from pubs.acs.org by UNIV OF SUNDERLAND on 09/07/18. For personal use only.

S Supporting Information *

ABSTRACT: We describe herein a series of self-assembled monolayers (SAMs) on gold designed for adjustable tethering of model lipid membrane phases. The SAMs consist of deuterated aliphatic anchors, HS(CH 2 ) 1 5 CONH(CH2CH2O)6CH2CONH-X, where X is either −(CD2)7CD3 or −(CD2)15CD3, dispersed in a stable matrix of proteinrepellent molecules, HS(CH2)15CONHCH2CH2OH. The mixed SAMs with variable surface densities of the anchors are thoroughly characterized before and after adsorption of phospholipids by means of ellipsometry, contact angle goniometry, and infrared reflection−absorption spectroscopy (IRRAS). In all cases, the bottom portions of the mixed SAMs (i.e., the h-alkyl thiol segments of the molecules) are arranged in a highly ordered all-trans conformation stabilized by a network of lateral hydrogen bonds. The terminal portions of the anchors (the oligo(ethylene glycol) spacer and deuterated alkyl segments, respectively), however, possess less ordered conformations in the mixed composition regime. For the SAMs containing the longer anchors (−(CD2)15CD3), the contact angle and infrared data point toward partial phase segregation. These findings are in excellent agreement with molecular dynamics simulations by Schulze and Stein. Upon analysis in air, the IRRAS data also indicate strong interaction between the adsorbed phospholipid molecules and the d-alkyl tails of the mixed SAM constituents. In such assemblies are the alkyl tails of the phospholipids aligned perpendicularly with respect to the supporting substrate, regardless of the anchor length. We also probed the in situ formation of a tethered bilayer lipid membrane (tBLM) via fusion of small unilamellar vesicles (SUVs) on the characterized SAMs using a quartz crystal microbalance with dissipation monitoring. Our experiments show that SUVs fuse efficiently of the two mixed SAMs with and without a pre-adsorbed lipid layer. Owing to the defined molecular composition and phase behavior, our SAM platform is attractive for detailed studies of tBLM formation and cell mimetic applications.

■

INTRODUCTION Supported and tethered lipid bilayer membranes have become attractive model systems for in-depth studies of the biochemistry/biophysics of the cell membrane. They are also regarded as emerging platforms for advanced technological applications, for example, in biosensing and pharmaceutical screening.1,2 In such model membrane constructs, a fluid lipid bilayer is attached and stabilized on a solid substrate, either directly or via a thin organic layer. The latter can be in a form of a soft polymeric cushion or a molecular assembly, most oftena self-assembled monolayer (SAM).3 Typically, binary mixed SAMs consisting of a matrix (background) compound and anchor molecules with terminal aliphatic moieties are used for integration with the so-called tethered bilayer lipid membranes (tBLMs). Such tBLM-supporting architectures have been obtained by the self-assembly of lipid-like derivatives with a surface-pinning group4−8 or via in situ covalent docking of aliphatic anchors9 to a mixed (binary) SAM that presents © 2018 American Chemical Society

chemically reactive terminal groups. In general, SAMs can provide several important methodological advantages, including (i) physical reinforcement/modulation of the fluid tBLM assembly by the anchor tails, (ii) formation of a protein-inert, hydrated matrix, which reduces the risk for denaturation of membrane proteins recruited to the tBLM, and (iii) integration of various sensing modalities, for example, fluorescent or conducting molecules. Despite of the significant interest in tBLMs, there have been only a few reports on systematic investigations on the effect of the molecular composition and structure of the supporting/ tethering assemblies on the tBLM formation. For example, Evans and co-workers described mixed cholesterol/hydroxylterminated SAMs on gold10 and silica,11 respectively, and they Received: May 28, 2018 Revised: August 2, 2018 Published: August 7, 2018 8201

DOI: 10.1021/acs.jpcb.8b05097 J. Phys. Chem. B 2018, 122, 8201−8210

Article

The Journal of Physical Chemistry B

Scheme 1. Chemical Structures of the Compounds Used in the Study Based on a Matrix Compound (EG1H), and Anchor Compounds (EG6AC8D and EG6AC16D)a

a

The terminal alkyl portions of the two anchor compounds, as indicated on the right of the dashed line, are fully deuterated. The synthetic 1stearoyl,2-oleoyl phosphatidylcholine (SOPC) lipid was used for tethered bilayer membrane formation.

Preparation of SAMs. All ω-substituted alkanethiol compounds used in this study (Scheme 1) were synthesized in our laboratory, and the details on synthesis have been described elsewhere.17 All reagents were obtained from commercial suppliers and were used without further purification. Stock solutions of the compounds were prepared in ethanol (99.5%) in the mM concentration range. They were subsequently diluted to working solutions in ethanol of either the pure compounds or the binary mixtures of the matrix molecule EG1H and one of the anchor compounds. The total (final) concentration of the working solution used for SAM preparation was 20 μM. Prior to SAM adsorption, the gold substrates were cleaned in a 5:1:1 mixture of deionized water (Milli-Q), 30% hydrogen peroxide, and 25% ammonia, for 5 min at 80 °C, and then rinsed in deionized water. After 24 h incubation of the samples, they were washed in an ultrasonic bath in ethanol for about 3−5 min, blown dry with nitrogen gas, and immediately analyzed. The two types of gold substrates were immersed in the same working solutions of the corresponding composition and treated identically. Phospholipid Adsorption. Pre-adsorption of 1-stearoyl,2oleoyl phosphatidylcholine (hereafter SOPC, obtained from Avanti Polar Lipids) on the SAM surfaces was performed by first dissolving the lipid in analytical grade hexane (0.5 mg/ mL) and then immersing the SAM samples into the lipid solution for 30 min. Afterward, the samples were rinsed with hexane three times, dried in a nitrogen gas flow, and immediately analyzed. The samples prepared on the goldcoated silicon substrates and those on the QCM-D sensors were treated identically. Null Ellipsometry. Single-wavelength ellipsometry was performed in a Rudolph Research AutoEL null ellipsometer with built-in software, by using a He−Ne laser light source (λ = 632.8 nm) at an angle of incidence of 70°. Before SAM adsorption, the refractive index of a freshly cleaned (see the procedure described above) gold surface was obtained. The average value of the refractive index was subsequently used in a model “ambient/organic film/gold”, assuming an isotropic, transparent organic layer with a refractive index of n = 1.5. The film thickness was calculated as an average value of measurements at three different spots for at least three samples of the same type. Contact Angle Goniometry. A semi-automatic contact angle meter KSV CAM 200 was used to determine the advancing and receding contact angles of water on the SAMs surfaces. A manual dispenser was used to expand or retract a

investigated the effects of the surface coverage as well as the structure of the tethered cholesterol assemblies on interactions with lipid vesicles. Knoll and co-workers surveyed the relation between the structure of the underlying gold substrate modified with a thiolipid SAM and the quality of tBLMs.12 Recently, a quartz crystal microbalance (QCM) has emerged as an attractive in situ technique to probe directly the process of lipid vesicle fusion on solid surfaces.13,14 In particular, monitoring the dissipation signal (QCM-D) of the acoustic oscillations damped by the lipid vesicles interacting with the surface can provide detailed information regarding the mechanism of tBLM formation in real time.15 For such basic studies of lipid vesicle−surface interactions, it is of crucial importance to build on a stable, structurally well-defined and characterized supporting/anchoring platform. Herein, we report on preparation and characterization of mixed binary alkyl thiol assemblies on gold consisting of deuterated tBLM anchors embedded in an ethylene glycol-terminated matrix monolayer. The described system is based on our previous detailed experimental and theoretical investigations of different structural aspects of SAMs formed on polycrystalline Au(111) surfaces by oligo(ethylene glycol) (OEG)-containing alkyl thiols that are laterally stabilized by hydrogen bonds.16,17 In the present study, we analyzed in parallel the structure and composition of the mixed binary SAMs by contact angle goniometry, null ellipsometry, and infrared reflection− absorption spectroscopy (IRRAS). The experimental observations are benchmarked against recent coarse-grained and fully atomistic molecular dynamics (MD) simulations reported by Schulze and Stein.18 We also explored the interactions of the SAMs with lipids and confirmed efficient tBLM formation via small unilamellar vesicle (SUV) fusion by QCM-D.

■

EXPERIMENTAL METHODS Substrates. Substrates cut from standard silicon (100) wafers were coated with a 25 Å titanium adhesion layer and, subsequently, an additional 2000 Å thick gold film. For electron beam deposition of the metals, a Balzers UMS 500 P system was employed operating at a base pressure of 10−9 mbar and at an evaporation pressure of about 10−7 mbar. A constant evaporation rate of 10 Å/s was used for the deposition of the polycrystalline gold (111) film.17 For lipid adsorption analysis using the QCM technique, the commercially available quartz sensors with a 1000 Å-thick gold layer (Q-sensors from Q-Sense, Sweden) were employed. 8202


Article


ellipsometric thickness and contact angle data following the rules for error propagation: Δ(A − B) = ΔA + ΔB; Δ(A/B) = ΔA/A + ΔB/B and Δ(1 + cos(A))2 = 2(1 + cos(A)) × sin(A) × ΔA.

water droplet. During the measurements, the needle was kept inside the center of the formed drop, and images were captured by a high-speed charge-coupled device camera. The images were then analyzed by the KSV CAM software. Average values obtained for three different samples were used for data presentation, and each measurement consisted of three to five images for the advancing and receding angles, respectively. IRRAS. The reflection−absorption (RA) spectra were recorded on a Bruker IFS 66 system, equipped with a grazing angle (85°), p-polarized reflection accessory, and a liquidnitrogen-cooled MCT detector. The sample chamber was continuously purged with nitrogen gas during the measurement. All spectra were acquired at 2 cm−1 resolution in the spectral range between 4000 and 700 cm−1, as the sum of 3000 scans. A three-term Blackman−Harris apodization function was applied to the interferograms before Fourier transformation. Background spectra (Ro) were recorded using a SAM formed on gold either by deuterated hexadecane thiol (HS(CD2)15CD3, a generous gift from Prof. David Allara) or by a nondeuterated analogous compound (HS(CH2)15CH3, Fluka), depending on the spectral range of interest. The sample spectra (R) were acquired under identical conditions, and the data were presented as −log(R/Ro). Lipid Vesicle Preparation. The SOPC powder was first dissolved in chloroform, and after removing the solvent in a vacuum, the lipids were resuspended into Hepes buffer solution (20 mM Hepes, 150 mM NaCl, pH 7.5) to a final lipid concentration of 250 μM. SUVs were prepared by probe sonication. Prior to injection, the SUVs were diluted with Hepes buffer solution with a high salt concentration (20 mM Hepes, 1 M NaCl, pH 7.5) to the final lipid concentration of 25 μM. Quartz Crystal Microbalance with Dissipative Monitoring (QCM-D). Formation of the tBLMs on the prepared SAMs was monitored in a QCM unit allowing dissipation monitoring (Q-Sense E4) by employing the gold-coated (a layer thickness of 100 nm) Q-sensors (Q-Sense). The flow rate in the flow cell of the QCM-D instrument was kept at 0.1 mL/ min. The signal of the seventh harmonics was used to present the data. Data Analysis. The experimental data were analyzed using the software packages provided by the corresponding equipment manufacturers. The surface compositions χSAM for the mixed binary SAMs were calculated based on previously published analysis of the ellipsometric thickness19 and wettability20 data. Namely, from the ellipsometry data, the mole fraction of the anchor molecules was determined as χSAM = (d − dEG1H)/(dEG6ACxD − dEG1H)

■

RESULTS AND DISCUSSION We have chosen to study binary SAM formation from solutions with mole fraction of the anchor compounds of 50 mol % (χEG6ACxD,sol = 0.5) and less, that is, the focus is on SAMs with a relatively low surface coverage of the anchor tails (−(CD2)7CD3 or −(CD2)15CD3). The prime motivation was to minimize the influence of the anchoring moieties on the physiochemical characteristics of the formed tBLM. To be able to relate the results from different experimental techniques, we prepared samples for SAM compositional/structural studies as well as for real-time QCM-D monitoring during SUV fusion under identical conditions. Ellipsometry and Contact Angle Measurements. Figure 1 shows the measured ellipsometric thicknesses and

(1)

where d, dEG1H, and dEG6ACxD are the measured ellipsometric thicknesses of the mixed SAM and those formed by the corresponding single compounds, respectively. Likewise, from the contact angle data χSAM = ((1 + cos θ )2 − (1 + cos θEG1H)2 ) /((1 + cos θEG6ACxD)2 − (1 + cos θEG1H)2 )

Figure 1. Advancing (black columns) and receding (grey columns) water contact angles and ellipsometric thicknesses (blue and red symbols, respectively, connected by lines for clarity of presentation) measured for mixed binary SAMs on gold formed from the thiol compounds shown in Scheme 1. The panels a and b show results for EG1H mixed with the anchor compounds EG6AC16D and EG6AC8D, respectively. The mole fractions χsol (χEG6AC16D,sol and χEG6AC8D,sol) represent the relative solution concentration of the anchors.

(2)

where θ, θEG1H, and θEG6ACxD are the equilibrium contact angles of the mixed SAM and the corresponding single compound SAMs, respectively. The equilibrium contact angle was defined as the average of the measured advancing and receding contact angle values. Errors of the χSAM values were calculated from the 8203


Article

The Journal of Physical Chemistry B advancing/receding water contact angles of the mixed SAMs as functions of the mole fractions of either EG6AC8D or EG6 AC 16D in solution mixtures (see also Supporting Information). The values obtained for the single component SAMs (EG1H, EG6AC16D or EG6AC8D, respectively) are in agreement with the previously published data.17 As expected for the binary mixed SAMs, an increase in the solution mole fraction of the anchor molecules results in a gradually increasing film thickness and hydrophobicity due to an increasing surface concentration of the anchors. The incremental thickness and the advancing/receding contact angle values are consistently higher for the EG6AC8Dcontaining SAMs than for the EG6AC16D-containing SAMs formed at the same solution mole fraction. This indicates certain differences in the adsorption/SAM formation kinetics from the solution mixtures, where the differently sized anchor thiols compete for the gold surface binding sites with the matrix compound EG1H. Another noteworthy observation from the water contact angle data is the contact angle hysteresis. The hysteresis θa − θr is the smallest for the single component SAMs: typically 5−8°, indicating a homogenous, highly ordered assembly on the surface. Also, for the EG6AC8D-containing mixed SAMs, θa − θr increases only slightly with mole fraction, and the hysteresis values are smaller than for the longer compound. For the binary SAMs containing EG6AC16D, the values of θa − θr increase from 12° at χsol = 0.01 to 22° at χsol = 0.5. Such a trend in the contact angle hysteresis previously has been interpreted as a sign of phase segregation occurring on the surface.21,22 The ellipsometric thickness and wettability data also can be used to estimate the surface composition χSAM of the mixed SAMs.19,20 In Figure 2, the surface fraction χSAM values, calculated according to eqs 1 and 2 for the respective anchor molecule in the mixed SAMs, are plotted as a function of χsol in the incubation solution. The plots reveal that the χSAM values determined from the ellipsometry data for the mixed SAMs are lower as compared to those calculated from the contact angle data. Such a difference can be explained by the fact that ellipsometry detects the incremental changes in the optical density of the isotropic organic layer (assumed by the employed optical model), including the anchors in the mixed SAM. Therefore, the ellipsometry-based method for calculation of χSAM is less sensitive to the structural differences between the mixed SAMs and the single component (pure) SAMs formed by the two anchor molecules, respectively. However, the structure of the anchors in the mixed SAMs versus the highly ordered, crystalline-like single component SAMs17 is likely different (see also below), leading to different wetting behavior (surface energy). Thus, calculations of the χSAM values by the method described by Israelachvili and Gee20 are expected to be more sensitive to variations of the surface concentration of the anchors. The data obtained by the two techniques show that, in general, the water contact angle values are more sensitive to the relative increase of the change in surface concentration of the anchors than the ellipsometric thicknesses, and thus, they may be used for qualitative characterization of mixed SAMs. The data analysis shown in Figure 2 also reveals that, independent of the selected method for estimation of the surface fraction χSAM of the anchors in the mixed SAMs, the relation of the mole fractions χSAM(χsol) clearly deviates from the Langmuir isotherm model. Such a nonlinear behavior of

Figure 2. Surface composition χSAM of mixed binary SAMs formed from solutions of different mole fractions χsol of either EG6AC16D (a) or EG6AC8D (b) compounds, mixed with EG1H (see Scheme 1 for the chemical structures). The anchor surface density χSAM values derived from wetting and ellipsometry data are colored in black (empty squares) and blue/red (filled squares), respectively. The lines connecting data points are drawn to guide the eye (note also breaks of the axes).

our system differs from most of the previously reported mixed binary SAMs formed by linear ω-substituted alkanethiols.23−26 However, it is reminiscent of the distinct deviation from the Langmuir isotherm observed by Islam et al. for mixed SAMs that contained amino-terminated compounds.27 These authors related the observed χSAM(χsol) nonlinearity to the electrostatic repulsion acting between amino groups on the gold surface and in solution, respectively. As our SAM system consists of noncharged compounds, we speculate that a similar repulsion could occur during the assembly process because of entropic factors (especially for the longer anchor compound having a molecular weight close to that of polymers). Interestingly, the observed saturation of the anchor surface fractions upon increasing the mole fraction in solution, as seen from the ellipsometry data, indeed is more pronounced for the EG6AC16D anchors. For this system, in the analyzed range χsol < 0.5, the χSAM value does not exceed 0.1. On the other hand, for the shorter compound, the corresponding χSAM values increase monotonically up to χSAM ≈ 0.4. In summary so far, the calculated χSAM(χsol) dependencies (Figure 2) 8204


Article


Figure 3. IRRAS spectra of single component and mixed binary SAMs formed on gold by the compounds shown in Scheme 1. The value χsol refers the mole fraction of the corresponding anchor compound in the adsorption solution. The spectra are displayed in the C−H stretching (a) region and fingerprint (b) regions.

Figure 4. IRRAS spectra of the C−D stretching region of SAMs formed from solutions of EG1H and different mole fractions χsol of (a) EG6AC16D or (b) EG6AC8D compounds shown in Scheme 1.

in a helical conformation, and its strength in the RA spectrum has been taken as evidence for a preferentially oriented helical axis along the surface normal. This peak is not visible in the mixed SAMs, neither for the EG6AC16D nor for the EG6AC8D anchors. Such an observation implies a low surface density and/or disordering of OEG portion of the anchor molecules even at relatively high mixing ratio in solution (χsol = 0.5), or that the helical axes are tilted away from the surface normal as suggested in our previous work17 and in the MD simulations by Schulze and Stein.18 Thus, the IR signatures of the mixed SAMs in the C−H region appear very similar to those in the spectrum of the EG1H SAM except for a less-pronounced shoulder near 2950 cm−1. Further details about the structure of the mixed SAMs can be found from the distinctive amide group-related vibrational modes in the fingerprint region, Figure 3b. The amide II mode in the mixed SAMs appears as a peak at around 1560 cm−1, a frequency that is close to that seen in the spectrum of the EG1H SAM. The band centered at around 1660 cm−1 corresponds to the amide I mode and it is very weak, although it seems to be a bit more pronounced for the mixed as compared to the single-component SAMs. The positions and

suggest a complex process of mixed SAM formation for both types of the anchors, which may involve intermolecular interactions, lateral/conformational rearrangements, domain formation, and different intermediate structures.28 Spectroscopic Characterization of SAMs. A detailed spectroscopic and theoretical investigation of the singlecomponent SAMs has been reported previously.17 Figure 3 shows a comparison of the IRRAS spectra of the single component SAMs and the mixed SAMs obtained from χsol = 0.5 solution mixtures of the anchor molecules and EG1H. The spectra of all mixed SAMs across the studied χsol range display essentially the same pattern of spectral signatures in the C−H and fingerprint regions (data not shown). In the 3000−2800 cm−1 region, Figure 3a, one can see that the asymmetric and symmetric C−H stretching modes arising from the alkyl segments of the two compounds can be found for all SAMs close to 2918 and 2850 cm−1, respectively. These spectral positions are characteristic for highly ordered, densely packed, defect-free polymethylene chains adopting an all-trans conformation. The single-component IRRAS spectra of the anchor molecules also display a strong peak at ∼2890 cm−1. It is assigned to asymmetric C−H stretching of the OEG portion 8205


Article

The Journal of Physical Chemistry B the relative intensities of the amide I and II bands suggest that the amide linkages in the mixed SAMs are in essentially the same orientation as in the single-component EG1H SAM (i.e., the C−N bond is perpendicular with respect to the gold surface). As expected, the amide groups provide a network of stabilizing lateral hydrogen bonds.29,30 The characteristic peaks arising from the helical EG6 portion at 1464, 1346, 1244, 1115, and 963 cm−1 that are dominating the fingerprint region of the EG6AC8D and EG6AC16D singlecompound SAMs are not present in the spectra of the mixed SAMs. Only a weak and broad band at around 1125 cm−1, attributed to the C−O−C skeletal stretching modes, can be seen in the mixed SAM of the shorter anchor. The shape of this band and its position (blue shifted by ∼10 cm−1) suggest that the EG6 portions adopt an amorphous-like conformation.31 The total absence of the EG6-related peaks in the fingerprint region of the mixed EG6AC16D SAM (including the 1125 cm−1 peak) is interpreted to indicate a low density of anchor molecules as compared to those found of shorter anchor prepared under identical mixing conditions. This is in full agreement with the ellipsometric and contact angle data, Figure 2. However, as mentioned before, another reason for the remarkable reduction of the 1115 cm−1 peak intensity for the longer anchor molecule could be due to preferential orientation, see discussion for the 2890 cm−1 peak above. In order to increase the understanding of the composition and conformation of the terminal anchor chains in the mixed SAMs, we also analyzed the IR spectra in the C−D stretching region across the entire χsol range, Figure 4. The intensity of the C−D stretching peaks increases with increasing solution mole fraction of the anchor molecules in agreement with the ellipsometric thickness and contact angle data, Figure 2. The CD2 asymmetric and symmetric stretching modes of the single component EG6AC16D SAM appear at 2193 and 2090 cm−1, respectively (Figure 4a). These positions are characteristic for d-alkyl chains adopting a highly ordered all-trans conformation.17 The relative intensity distribution of the CD2 to CD3 modes also resembles that of alkyl chains oriented perpendicularly to the supporting gold surface. The corresponding asymmetric and symmetric stretching modes of the CD3 moiety appear at about 2216 and 2074 cm−1, respectively. As the mole fraction χsol of EG6AC16D decreases, the intensity of the dominating CD2 bands decreases and their positions shift about 5−7 cm−1 toward higher frequencies indicating a partial disordering of the d16-chains. The relative intensities of the peaks corresponding to the CD2 symmetric and asymmetric stretching as well as the CD3 asymmetric stretching peaks remain similar, regardless of the mole fraction (0.05 < χsol < 0.5). Thus, the pattern of the peaks in the CD region for the mixed EG6AC16D SAMs is reminiscent of that observed for the single compound SAM, except for the missing shoulder near 2074 cm−1. As pure EG6AC16D forms a highly organized and densely packed assembly, a reasonable interpretation of the above observation is that the long anchor molecules segregate to form domains partly dominated by the d16-alkyl chains in a background of EG1H matrix (filling) molecules. The size of these domains, however, cannot be derived from the spectra. Noteworthy, a similar behavior was proposed upon selfassembly of ω-substituted alkyl thiols of vastly different chain lengths and tail group chemistry (OH vs CH3) on gold.32 The formation of domain is also in line with the contact angle data and the MD simulations by Schulze and Stein.18

Interestingly, the C−D stretching region of the EG6AC16D and EG6AC8D anchor molecules reveals substantial differences. In the case of the single-component EG6AC8D SAM (Figure 4b), the CD2 asymmetric and symmetric stretching peaks appear at 2198 and 2101 cm−1, respectively.17 They are significantly weaker than the asymmetric and symmetric CD3 peaks at 2221 and 2074 cm−1, respectively. This intensity distribution with weak CD2 modes and relatively strong CD3 modes is consistent with recent findings of short d-alkyls in the bulk (e.g., d-hexane).33 The asymmetric CD3 mode is the most pronounced peak in the d-hexane spectrum33 as well as in the single component spectrum, Figure 4b. The CD2 modes gain intensity with respect to the CD3 modes upon mixing, and CD2 modes are in fact more intense in the 0.5 and 0.3 spectra than in the single component spectrum (1.0), Figure 4b. This points toward partial disorder or increased tilt of the shorter dalkyl chains. However, once mixed, the relative intensity pattern remains the same throughout the mixing regime (in a similar way as for the longer compound). The overall analogy with the d-hexane spectrum and the fact that the in-plane stabilizing forces are expected to be less pronounced for the short compound point toward a partly organized assembly of d8-alkyl chains. Interestingly, Lee at al. reported in a combined theoretical and experimental study that the d-alkyl chains of the shorter compound possess a larger tilt with respect to the surface normal than those of the longer compound.17 Such a difference in tilt can explain the differences seen in the relative intensities of the CD2 and CD3 peaks in the spectra of the single component compounds (note, in particular, the increase in the strength of the asymmetric CD3 peak close to 2220 cm−1, Figure 4a,b). Because the spectral signature of the mixed EG6AC8D SAMs in the C−D region is distinctly different from that of the singlecompound SAM, it seems less likely that phase segregation occurs with decreasing solution composition (also confirmed in the MD simulations by Schulze and Stein18). Again, the observed structural dissimilarities in the CD stretching region for the EG6AC16D and EG6AC8D when lowering the solution composition from the single component to the mixed regime might be indicative of certain differences in SAM formation kinetics and domain formation. The total CD stretching intensity can be used as another measure of the relative changes in surface coverage of the anchor molecules. Figure 5 presents the normalized integrated C−D stretching peak intensities (2300−2000 cm−1) as a function of χsol of the anchor molecules. The main trend seen from this spectral analysis is in qualitative agreement with the ellipsometric and contact angle data shown in Figure 1, see also Supporting Information Table S1 and Figure S1. For example, the relative surface density (χSAM) for the EG6AC8D mixed SAMs is higher than that of the EG6AC16D mixed SAMs throughout the entire range of mixing ratios. Bearing in mind that the number of CD2 units in the shorter d-alkyl chain is half of that in the longer chain makes the estimated fraction of the shorter EG6AC8D anchor even more obvious. However, one should be cautious about directly correlating the changes in the integrated IR peak intensity to a change in surface density of the anchor molecule. The peak intensities of the IRRAS spectra are particularly sensitive to changes in conformation and orientation and thus not only to the surface coverage of the molecules. For direct and quantitative determination of the SAM composition, it would be necessary to employ other quantitative techniques such as X-ray photon spectroscopy or 8206


Article


In the fingerprint region (Figure 6b), the amide II band at ∼1561 cm−1 remains unchanged, despite of the intercalating SOPC molecules in the SAM. This diagnostic peak confirms that the structure of the supporting level (alkyl and peptide link) of the mixed SAM remains intact during the treatment. Also, in this region, a series of new bands appear upon lipid adsorption. The strong peak at 1742 cm−1 can be related to CO stretching vibrations of the glycerol ester group in the SOPC molecules. Peaks representative of the C−O stretching vibrations are located between 1181 and 1160 cm−1, respectively. Previous studies suggest that the ester moiety C−C(O)−O−C has a planar conformation if the C−O stretching mode appears at 1180 cm−1, whereas a deviation from the planar conformation leads to a red shift of this C−O band.35,36 The absorption bands arising from the polar choline moiety of the SOPC molecule are essentially between 1500 and 1350 cm−1 for both bending and stretching modes.36 Also, in Figure 6b, the weak peak near 1480−1490 cm−1 is very likely because of the methyl asymmetric bending of the choline head group, that is, N−CH3. Furthermore, the phosphate stretching modes appear as distinct peaks between 1300 and 1000 cm−1. The asymmetric stretching of the phosphate (PO2−) can be assigned at ∼1253 cm−1 and it corresponds to a weakly hydrated phosphate group.37 The symmetric phosphate stretching mode is located at ∼1095 cm−1, next to the strong dominating C−O−P stretching peak at around 1070 cm−1. The IRRAS spectra in the C−D stretching region, Figure 6c, reveal interactions of the d-alkyl chains and the adsorbed SOPC molecules. The intensities of the C−D bands of the anchor molecules decrease after SOPC adsorption. This effect can be partially explained by a re-orientation of the terminal dalkyls (anchors) perpendicular to the gold surface. Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D). Figure 7 shows typical sensorgrams recorded upon SUV injection cycles by employing quartz sensors functionalized with the mixed SAMs. Both types of surfaces were prepared at the same mole fractions χsol = 0.05 of the anchors, yielding estimated (based on ellipsometry, see above) χSAM values of 0.035 ± 0.021 and 0.097 ± 0.024 for the EG6AC16D and EG6AC8D SAMs, respectively. The sensorgrams indicate that upon initial injection the SUVs adsorb onto the SAM surface and accumulate for a certain period of time. It has been shown that once a sufficient SUV density is reached on the surface, they rupture due to the vesicle−vesicle, vesicle−SAM, and vesicle−adsorbed lipid layer interactions.38

Figure 5. Integrated peak intensities (2300−2000 cm−1) in the region of the C−D stretching modes from the IR spectra shown in Figure 4. Blue and red lines correspond to the SAMs on gold formed from solution mixtures of EG1H and different mole fractions χsol of the anchor molecules, EG6AC16D or EG6AC8D, respectively. The integrated intensities are normalized with respect to those measured for the single-compound SAMs consisting of either of the anchor molecules. For details, see Supporting Information Table S1 and Figure S1.

radioactive labeling (e.g., S35) of one of the component in the mixture. Spectroscopic Characterization of Lipid Adsorption. Previous studies have indicated that pre-adsorption of lipid molecules is required in order to obtain efficient vesicle fusion on poly(ethylene glycol) surfaces decorated with hydrophobic anchors.34 To probe the interaction of our mixed SAMs with SOPC molecules, we incubated the samples used for SAM characterization with a lipid solution in hexane and analyzed the structure of lipid layers on the mixed SAM surfaces by IRRAS. Figure 6 shows representative spectra of the mixed SAMs with the anchor mole fraction χsol = 0.1, prior and after SOPC adsorption. Interestingly, the IR signatures of the SOPC molecules look very similar on both EG6AC16D and EG6AC8D SAMs. After lipid adsorption, the high frequency region (Figure 6a) displays broadening and an increase in the intensity of the methylene C−H asymmetric and symmetric stretching peaks because of the hydrophobic hydrocarbon tails of the SOPC chains. One can also identify the contributions from the terminal methyl groups of the SOPC hydrocarbon tails, at ∼2966 and 2880 cm−1, respectively.

Figure 6. IRRAS spectra of mixed SAMs formed at the anchor mole fraction χsol = 0.1 before (dashed line) and after (solid line) SOPC lipid adsorption: (a) C−H stretching region, (b) fingerprint region, (c) C−D stretching region. The blue and red lines display spectra of SAMs consisting of EG1H mixed with EG6AC16D and EG6AC8D anchors, respectively. See Scheme 1 for the chemical structures of the compounds. 8207


Article


Figure 7. Sensorgrams recorded by QCM-D during tBLM formation on mixed SAMs (χsol = 0.05) of EG6AC16D (a) and EG6AC8D (b) anchors, respectively. The arrows indicate the injection sequence of experiment. Note that after the first lipid vesicle injection cycle washing by ethanol (EtOH) removes all the lipid material from the surface, including the SOPC molecules pre-adsorbed on the SAM before the experiment. Thus, the second exposure of the same sensor surface to the vesicles occurs on a lipid-free SAM (starting at around 3000 s).

ellipsometry, water contact angle, and IRRAS measurements indicate a complex process of the mixed SAM formation, which significantly deviates from the Langmuir adsorption model. However, the data also reveal different regimes in the mixing behavior of the two types of SAMs. First, we were able to obtain higher surface mole fractions of the shorter anchor compounds in the mixed SAMs as compared to the longer anchors, despite the same mixing ratios in solution. This suggests different adsorption kinetics during the SAM formation, with the tendency of EG6AC16D being outperformed by the EG1H matrix molecules pinning to the gold surface. Second, the hysteresis of the water contact angles as well as the distinct structure the d-alkyl peaks in C−D stretching region of the IRRAS spectra suggests that the longer anchors may partially phase segregate, in agreement with MD simulation by Schulze and Stein.18 The mixed SAMs that contain the shorter anchor do not show this behavior, and the surface density of the −C8D tails can be tuned in the entire range of the studied mixing ratios (the solution mole fractions 0.01 < χsol < 0.5). Thus, it seems that the shorter EG1H and EG6AC8D predominantly form statistically mixed assemblies on the polycrystalline gold surface. Despite the observed structural differences of the mixed SAMs, the IRRAS data show that the structure of the SOPC lipid assembly obtained from hexane solution is in principle identical for both types of the anchors. Moreover, QCM-D kinetics suggests that SUVs fuse efficiently on the mixed SAMs presenting either of the anchors, with and without preadsorbed SOPC, via the distinct vesicle rupturing step. Thus, the developed anchoring SAM with well-controlled structural characteristics and composition is suitable for obtaining tBLMs that can be employed for advanced studies and applications of different cell membrane-mimicking assemblies. A more indepth analysis of the QCM-D data will be published separately.

This fusion step is clearly visible in the recorded sensorgrams as the distinct transient frequency and dissipation signal features that can be attributed to the loss of the water trapped inside the vesicles. Note that the measured difference in the frequency values before and after the SUV fusion is close to 26 Hz, a value well corresponding to the formation of a lipid bilayer.39 Also, the frequency and dissipation signals display plateaus reaching the same values on the lipid precoated (the first SUV injection cycle) as well as on ethanol-treated and therefore lipid-free (the second vesicle injection cycle) SAM surfaces, respectively. Thus, from the latter experiment, it can be concluded that the surface pretreatment with lipids appears not to be necessary in order to obtain a tBLM on our mixed SAMs by fusion of SUVs made of SOPC. The major difference between the kinetics recorded for our mixed SAMs as compared to the previously published QCM-D data on vesicle fusion on silicon dioxide surfaces40 is in the behavior of the dissipation signal. Namely, on a SiO2 surface, the dissipation signal relaxes to the initial value after completion of the SUV fusion. We could never observe such a complete relaxation of the dissipation signal on our mixed SAM surfaces. This can be interpreted as differences in the mechanical elasticity of the substrate before and after lipid vesicle fusion. We anticipate that the formed tBLM on the anchoring SAMs confines a certain amount of water molecules. They are likely trapped between the lower leaflet of the lipid bilayer and the EG1H matrix. Also, the OEG linkages of the anchors in the mixed SAM can interact with the formed tBLM. They may provide an additional pathway for damping of the quartz crystal oscillations, that is, preventing complete relaxation of the dissipation values in the recorded sensorgrams. However, these mechanisms require additional experimental and theoretical investigations that are subject to our further studies.

■

■

CONCLUSIONS The mixed binary SAMs formed from solution mixtures of the EG1H matrix and EG6AC16D or EG6AC8D anchor molecules, respectively, consist of a highly ordered, hydrogen-bonding halkylthiolate assembly with the terminal d-alkyl portions exposed to the ambient. The latter offers a convenient tool to tune and probe the structure, hydrophobicity, and phase behavior of a nanometers-thick interfacial layer that interacts with a model lipid phase. Taken together, the results of null

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.8b05097. Ellipsometric thicknesses, contact angles, and integrated intensities for the investigated SAMs and comparison of surface composition obtained using ellipsometry, contact angle goniometry, and IR spectroscopy (Figure S1) (PDF) 8208


Article


■

Photopatterned Self-Assembled Monolayers. Chem.Eur. J. 2007, 13, 7957−7964. (12) Naumann, R.; Schiller, S. M.; Giess, F.; Grohe, B.; Hartman, K. B.; Kärcher, I.; Köper, I.; Lübben, J.; Vasilev, K.; Knoll, W. Tethered Lipid Bilayers on Ultraflat Gold Surfaces. Langmuir 2003, 19, 5435− 5443. (13) Reimhult, E.; Höök, F.; Kasemo, B. Intact Vesicle Adsorption and Supported Biomembrane Formation from Vesicles in Solution: Influence of Surface Chemistry, Vesicle Size, Temperature, and Osmotic Pressure. Langmuir 2003, 19, 1681−1691. (14) Dorvel, B. R.; Keizer, H. M.; Fine, D.; Vuorinen, J.; Dodabalapur, A.; Duran, R. S. Formation of Tethered Bilayer Lipid Membranes on Gold Surfaces: QCM-Z and AFM Study. Langmuir 2007, 23, 7344−7355. (15) Cho, N.-J.; Frank, C. W.; Kasemo, B.; Höök, F. Quartz Crystal Microbalance with Dissipation Monitoring of Supported Lipid Bilayers on Various Substrates. Nat. Protoc. 2010, 5, 1096−1106. (16) Valiokas, R.; Svedhem, S.; Svensson, S. C. T.; Liedberg, B. SelfAssembled Monolayers of Oligo(ethylene Glycol)-Terminated and Amide Group Containing Alkanethiolates on Gold. Langmuir 1999, 15, 3390−3394. (17) Lee, H.-H.; Ruželė, Ž .; Malysheva, L.; Onipko, A.; Gutés, A.; Björefors, F.; Valiokas, R.; Liedberg, B. Long-Chain Alkylthiol Assemblies Containing Buried In-Plane Stabilizing Architectures. Langmuir 2009, 25, 13959−13971. (18) Schulze, E.; Stein, M. Simulation of Mixed Self-Assembled Monolayers on Gold: Effect of Terminal Alkyl Anchor Chain and Monolayer Composition. J. Phys. Chem. B 2018, 122, 7699−7710. (19) Laibinis, P. E.; Fox, M. A.; Folkers, J. P.; Whitesides, G. M. Comparisons of Self-Assembled Monolayers on Silver and Gold: Mixed Monolayers Derived from HS(CH2)21X and HS(CH2)10Y (X, Y = CH3, CH2OH) Have Similar Properties. Langmuir 1991, 7, 3167−3173. (20) Israelachvili, J. N.; Gee, M. L. Contact Angles on Chemically Heterogeneous Surfaces. Langmuir 1989, 5, 288−289. (21) Johnson, R. E.; Dettre, R. H. Contact Angle Hysteresis. III. Study of an Idealized Heterogeneous Surface. J. Phys. Chem. 1964, 68, 1744−1750. (22) de Gennes, P. G. Wetting: Statics and Dynamics. Rev. Mod. Phys. 1985, 57, 827−863. (23) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. Formation of Monolayer Films by the Spontaneous Assembly of Organic Thiols from Solution onto Gold. J. Am. Chem. Soc. 1989, 111, 321−335. (24) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M. Self-Assembled Monolayers of Alkanethiols on Gold: Comparisons of Monolayers Containing Mixtures of Short- and Long-Chain Constituents with Methyl and Hydroxymethyl Terminal Groups. Langmuir 1992, 8, 1330−1341. (25) Kang, J. F.; Liao, S.; Jordan, R.; Ulman, A. Mixed Selfassembled Monolayers of Rigid Biphenyl Thiols: Impact of Solvent and Dipole Moment. J. Am. Chem. Soc. 1998, 120, 9662−9667. (26) Nelson, K. E.; Gamble, L.; Jung, L. S.; Boeckl, M. S.; Naeemi, E.; Golledge, S. L.; Sasaki, T.; Castner, D. G.; Campbell, C. T.; Stayton, P. S. Surface Characterization of Mixed Self-Assembled Monolayers Designed for Streptavidin Immobilization. Langmuir 2001, 17, 2807−2816. (27) Islam, N.; Gurgel, P. V.; Rojas, O. J.; Carbonell, R. G. Effects of Composition of Oligo(ethylene Glycol)-Based Mixed Monolayers on Peptide Grafting and Human Immunoglobulin Detection. J. Phys. Chem. C 2014, 118, 5361−5373. (28) Schwartz, D. K. Mechanisms and Kinetics of Self-Assembled Monolayer Formation. Annu. Rev. Phys. Chem. 2001, 52, 107−137. (29) Valiokas, R.; Svedhem, S.; Ö stblom, M.; Svensson, S. C. T.; Liedberg, B. Influence of Specific Intermolecular Interactions on the Self-Assembly and Phase Behavior of Oligo(Ethylene Glycol)Terminated Alkanethiolates on Gold. J. Phys. Chem. B 2001, 105, 5459−5469.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Ramu̅nas Valiokas: 0000-0003-4807-9136 Bo Liedberg: 0000-0003-2883-6953 Present Address §

Center for Biomimetic Sensor Science, School of Materials Science and Engineering, Nanyang Technological University, 50 Nanyang Drive, Singapore 637553. Author Contributions

H.-H.L. performed surface characterization experiments and spectral data analysis and co-wrote the paper. M.G. performed QCM-D experiments and surface composition analysis and cowrote the paper. Ž .R. designed and synthesized the compounds for this study. R.V. and B.L. conceived, designed, and coordinated this study and co-wrote the paper. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Swedish Institute through the Visby program. M.G. acknowledges a postdoctoral fellowship from the Research Council of Lithuania, grant no. MOS-6/ 2010.

■

REFERENCES

(1) Castellana, E. T.; Cremer, P. S. Solid Supported Lipid Bilayers: From Biophysical Studies to Sensor Design. Surf. Sci. Rep. 2006, 61, 429−444. (2) Czolkos, I.; Jesorka, A.; Orwar, O. Molecular Phospholipid Films on Solid Supports. Soft Matter 2011, 7, 4562−4576. (3) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-Assembled Monolayers of Thiolates on Metals as a Form of Nanotechnology. Chem. Rev. 2005, 105, 1103−1170. (4) Lang, H.; Duschl, C.; Vogel, H. A New Class of Thiolipids for the Attachment of Lipid Bilayers on Gold Surfaces. Langmuir 1994, 10, 197−210. (5) Williams, L. M.; Evans, S. D.; Flynn, T. M.; Marsh, A.; Knowles, P. F.; Bushby, R. J.; Boden, N. Kinetics of the Unrolling of Small Unilamellar Phospholipid Vesicles onto Self-Assembled Monolayers. Langmuir 1997, 13, 751−757. (6) McGillivray, D. J.; Valincius, G.; Vanderah, D. J.; Febo-Ayala, W.; Woodward, J. T.; Heinrich, F.; Kasianowicz, J. J.; Lösche, M. Molecular-Scale Structural and Functional Characterization of Sparsely Tethered Bilayer Lipid Membranes. Biointerphases 2007, 2, 21−33. (7) Heinrich, F.; Ng, T.; Vanderah, D. J.; Shekhar, P.; Mihailescu, M.; Nanda, H.; Lösche, M. A New Lipid Anchor for Sparsely Tethered Bilayer Lipid Membranes. Langmuir 2009, 25, 4219−4229. (8) Schiller, S. M.; Naumann, R.; Lovejoy, K.; Kunz, H.; Knoll, W. Archaea Analogue Thiolipids for Tethered Bilayer Lipid Membranes on Ultrasmooth Gold Surfaces. Angew. Chem., Int. Ed. 2003, 42, 208− 211. (9) Lahiri, J.; Kalal, P.; Frutos, A. G.; Jonas, S. J.; Schaeffler, R. Method for Fabricating Supported Bilayer Lipid Membranes on Gold. Langmuir 2000, 16, 7805−7810. (10) Cheng, Y.; Boden, N.; Bushby, R. J.; Clarkson, S.; Evans, S. D.; Knowles, P. F.; Marsh, A.; Miles, R. E. Attenuated Total Reflection Fourier Transform Infrared Spectroscopic Characterization of Fluid Lipid Bilayers Tethered to Solid Supports. Langmuir 1998, 14, 839− 844. (11) Han, X.; Pradeep, S. N. D.; Critchley, K.; Sheikh, K.; Bushby, R. J.; Evans, S. D. Supported Bilayer Lipid Membrane Arrays on 8209


Article

The Journal of Physical Chemistry B (30) Malysheva, L.; Onipko, A.; Liedberg, B. Ab Initio Modeling of Amide-Stabilized, Oligo(ethylene glycol)-Terminated Self-Assemblies: In-SAM Molecular Geometry, Orientation, and Hydrogen Bonding. J. Phys. Chem. A 2008, 112, 1683−1687. (31) Valiokas, R.; Ö stblom, M.; Svedhem, S.; Svensson, S. C. T.; Liedberg, B. Temperature-Driven Phase Transitions in Oligo(ethylene Glycol)-Terminated Self-Assembled Monolayers. J. Phys. Chem. B 2000, 104, 7565−7569. (32) Allara, D. L. Critical Issues in Applications of Self-Assembled Monolayers. Biosens. Bioelectron. 1995, 10, 771−783. (33) Tyrode, E.; Hedberg, J. A Comparative Study of the CD and CH Stretching Spectral Regions of Typical Surfactants Systems Using VSFS: Orientation Analysis of the Terminal CH3 and CD3 Groups. J. Phys. Chem. C 2012, 116, 1080−1091. (34) Munro, J. C.; Frank, C. W. In Situ Formation and Characterization of Poly(ethylene glycol)-Supported Lipid Bilayers on Gold Surfaces. Langmuir 2004, 20, 10567−10575. (35) Fringeli, U. P. The Structure of Lipids and Proteins Studied by Attenuated Total Reflection (ATR) Infrared Spectroscopy II. Oriented Layers of a Homologous Series: Phosphatidylethanolamine to Phosphatidylcholine. Z. Naturforsch., C: J. Biosci. 1977, 32, 20−45. (36) Zawisza, I.; Bin, X.; Lipkowski, J. Potential-Driven Structural Changes in Langmuir−Blodgett DMPC Bilayers Determined by in situ Spectroelectrochemical PM IRRAS. Langmuir 2007, 23, 5180− 5194. (37) Zawisza, I.; Wittstock, G.; Boukherroub, R.; Szunerits, S. Polarization Modulation Infrared Reflection Absorption Spectroscopy Investigations of Thin Silica Films Deposited on Gold. 2. Structural Analysis of a 1,2-Dimyristoyl-Sn-Glycero-3-Phosphocholine Bilayer. Langmuir 2008, 24, 3922−3929. (38) Richter, R. P.; Bérat, R.; Brisson, A. R. Formation of SolidSupported Lipid Bilayers: An Integrated View. Langmuir 2006, 22, 3497−3505. (39) Keller, C. A.; Kasemo, B. Surface Specific Kinetics of Lipid Vesicle Adsorption Measured with a Quartz Crystal Microbalance. Biophys. J. 1998, 75, 1397−1402. (40) Seantier, B.; Breffa, C.; Félix, O.; Decher, G. DissipationEnhanced Quartz Crystal Microbalance Studies on the Experimental Parameters Controlling the Formation of Supported Lipid Bilayers. J. Phys. Chem. B 2005, 109, 21755−21765.

8210


Mixed Self-Assembled Monolayers with Terminal Deuterated Anchors

Recommend Documents