Stochastic Adhesion of Hydroxylated Atomic Force Microscopy Tips to

Dec 9, 2013 - 11, Iasi R-700506, Romania. †. Faculty of Physics, Laboratory of Molecular Biophysics and Medical Physics, Alexandru Ioan Cuza Universit...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Stochastic Adhesion of Hydroxylated Atomic Force Microscopy Tips to Supported Lipid Bilayers Aurelia Apetrei† and Lucel Sirghi‡,* ‡

Faculty of Physics, Alexandru Ioan Cuza University, Bulevardul Carol I, no. 11, Iasi R-700506, Romania Faculty of Physics, Laboratory of Molecular Biophysics and Medical Physics, Alexandru Ioan Cuza University, Bulevardul Carol I, no. 11, Iasi R-700506, Romania

†

S Supporting Information *

ABSTRACT: This work reports results of an atomic force microscopy (AFM) study of adhesion force between hydroxylated AFM tips and supported lipid bilayers (SLBs) of phosphatidylcholine in phosphate buffer saline solution at neutral pH. Silicon nitride AFM probes were hydroxylated by treatment in water vapor plasma and used in force spectroscopy measurements of adhesion force on SLBs with control of contact loading force and residence time. The measurements showed a stochastic behavior of adhesion force that was attributed to stochastic formation of hydrogen bonds between the hydroxyl groups on the AFM tip and oxygen atoms from the phosphate groups of the phosphatidylcholine molecules. Analysis of a large number of force curves revealed a very low probability of hydrogen bond formation, a probability that increased with the increase of contact loading force and residence time. The variance and mean values of adhesion force showed a linear dependence on each other, which indicated that hydrogen bond formation obeyed the Poisson distribution of probability. This allowed for the quantitative determination of the rupture force per hydrogen bond of about 40 pN and showed the absence of other nonspecific interaction forces.

1. INTRODUCTION Artificial phospholipid bilayers are extensively used as versatile models to mimic membranes of living cells in a plethora of fundamental and applied biophysical studies, which provide valuable information on key aspects of molecular processes taking place at the cellular membrane level.1,2 In particular, understanding of chemical and physical factors influencing adhesion3 of cellular membranes to various artificial and natural surfaces is crucial for nanomedicine,4,5 nanotoxicology6 and biomaterial engineering.7 Various specific and nonspecific forces are known to contribute to cellular adhesion.3,8 For instance, while adhesion of mammalian cells is mediated by integrin, rapid movement of Dictyostelium cells relies on nonspecific adhesion mediated by van der Waals attraction of their surface glycoproteins to the underlying substratum.9 There are several methods allowing for the measurement of molecular interactions involved in cell adhesion and molecular recognition10 as micropipet manipulation,11 optical manipulation,12 optical tweezers,13 surface force apparatus,14 and atomic force microscopy.15 Atomic force microscopy (AFM) is a powerful technique for the study of forces involved in cellular adhesion because it allows for direct measurements of bond strength at the single-molecule level in physiological medium.16 While techniques such as surface force apparatus and micropipet manipulation involve interaction of a large number of molecules in the contact region, the AFM technique reduces © 2013 American Chemical Society

the interaction to a much smaller number of molecules as a result of use of very sharp AFM tips. In the case of short-range interactions, this allows for recording of individual interaction events that show a strong stochastic character. Thus, considering the Poisson distribution of hydrogen bond formation probability at the contact between amino- functionalized AFM tips and sample surfaces, Wei et al.17 were able to determine the hydrogen bond rupture force in water to a value of 200.0 ± 43.6 pN. Hydrogen bonding is an essential interaction at the basis of many molecular processes in living organisms18 as cell recognition,19 protein folding20,21 and structure,22 molecular recognition between ligands and DNA,23 DNA replication24 and repair,25 etc. Unconventional hydrogen bonds were proven to be important mediators of ligand−receptor interactions,26 and understanding their energetic and structural properties is of paramount importance in rational drug discovery. At the membrane level, lipid hydrogen bonding can change the structure and dynamics of membrane proteins, while intra protein and lipid−protein hydrogen bonding couple the protein to lipids.27 The strength of hydrogen bonds is affected by several parameters of the interaction partners such as distance, chemistry of the donor and acceptor atoms, the relative Received: July 2, 2013 Published: December 9, 2013 16098

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104

Langmuir

Article

loading force, contact residence time, and tip−sample contact area on the probability of hydrogen bond formation are discussed below. Pushing the AFM tip into the SLB generates a deformation of the SLB33 that increases the probability of hydrogen bond formation between OH groups from the AFM tip and nonester phosphate oxygens or carbonyl oxygens of phosphatidylcholine from the SLB. This is happening as a combination of factors, including the increase of contact area between the AFM tip and the SLB, and removal of water molecules from the contact region.34 According to the model of Das et al.,33 the SLB deformation caused by an AFM tip pushing into it is different from the elastic (Hertzian) deformation35 produced on many other elastic materials and, due to the lateral expelling of lipid molecules from the interaction region (Figure 1), depends on

position of the donor in respect to the acceptor atoms, as well as by external parameters such as the nature of the surrounding environment.28,29 Using Surface Force Apparatus with surfaces coated with functionalized lipids, Pincet et al.30 determined the rupture force of a single hydrogen bond established between adenine and thymine to a value of 49 pN. Boland and Ratner31 derived a value of 54 pN for the same breaking force using adenine coated AFM tips and thymine coated surfaces. The present work investigates the stochastic behavior of the adhesion force between hydroxylated AFM tips and supported lipid bilayers of phosphatidylcholine in phosphate buffer saline medium at neutral pH. A statistic analysis of the pull-off force values measured in thousands of force spectroscopy measurements performed with control of contact loading force and residence time on the same SLB surface is presented. In the measurements, the hydroxylated AFM tips were kept in contact with the SLB, with control of contact loading force and residence time, and then retracted with constant speed. The adhesion force, which is measured as the pull-off force recorded during tip retraction, shows a pronounced stochastic behavior. Analysis of a large number of force curves revealed a very low probability of nonzero adhesion force values, a probability that increased with the increase of contact loading force and residence time. The observed stochastic character of the adhesion force is explained by stochastic formation of hydrogen bonds between hydroxyl groups from the hydroxylated AFM tip surface and nonester phosphate oxygens or carbonyl oxygens of phosphatidylcholine32 from SLBs. Assigning the results of adhesion force measurements to the Poisson distribution probability of hydrogen bond formation allows for the determination of the rupture force of a single hydrogen bond. Moreover, control of contact residence time and contact loading force is used for the investigation of the kinetics of nonequilibrium transitions between bound and unbound states of hydroxyl groups present on the surface of the AFM probe and the phosphatidylcholine molecules from the SLB surface. It is shown that the association−dissociation reaction describing the making and breaking of hydrogen bonds at the tip−sample contact is controlled by the contact force. The reaction is stimulated by forcing the hydroxylated AFM tip in close contact with the polar heads of the lipid molecules, which promotes the formation of hydrogen bonds due to removal of water molecules at the membrane surface, hydration of the lipid membrane affecting the strength and stability of hydrogen bonds.

Figure 1. Sketch of AFM tip pushing on a supported lipid bilayer (SLB). The tip−sample contact area is determined by the tip curvature radius (Rt), indentation depth (z), loading force (F), and the SLB thickness (2h).

the SLB area compressibility modulus, kA. In this model, the indentation of the SLB along a depth z is determined by the loading force according to the following relationship: ⎞2 π ·kA ·R t ⎛ z ⎟ F= ·⎜ ⎝ h − z /2 ⎠ 4

(1)

where Rt is the tip radius and h, the thickness of a single lipid layer (Figure 1). The indentation depth determines a tip− sample contact area of A = 2π·Rt·z. This means that for z ≪ h the contact area increases roughly as F1/2. The eq 1 is not valid for large values of z (comparable to h) at which usually the AFM tip produces a pore into the SLB.33 The number of lipid molecules interacting with the AFM tip surface is: 2π ·R t·z A = N= (2) a a

2. THEORETICAL CONSIDERATIONS The purpose of this section is to discuss the theoretical considerations at the basis of the force spectroscopy investigation of stochastic formation of hydrogen bonds between a hydroxylated AFM tip and a SLB in phosphate buffer saline (PBS) solution at neutral pH. Statistics of the adhesive force measured between hydroxylated AFM tips and phosphatidylcholine SLBs is defined by a set of parameters that have a major impact on the probability of hydrogen bond formation. In our force spectroscopy experiments, the hydroxylated AFM tip approaches the SLB, pushes on the SLB with the loading force, F, for a certain time, which is referred hereinafter as the contact residence time, tr, and, at the end, is retracted with the same speed. As it will be described in the Experimental section, the values of F and tr strongly affect the probability of hydrogen bond formation. Therefore, theoretical considerations on the effects of parameters such as

where a is the area per lipid molecule at the level of the SLB surface. Neglecting the variation of a caused by expulsion of lipid molecules from the interaction area, we estimate that the number of lipid molecules interacting with the AFM tip surface is increasing as F1/2 at low values of F. A rough estimation for a loading force F = 0.8 nN applied by an AFM tip with Rt = 50 nm on a phosphatidylcholine SLB with h = 3 nm, a = 0.6 nm2, and kA = 0.15 N/m gives a values of about 1 nm for z and about 16099

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104

Langmuir

Article

500 for N. Although a large number of lipid molecules are in contact with the AFM tip surface, only few of them form hydrogen bonds during the contact. According to our experimental data, at moderate values of F, only 10% of the individual force spectroscopy measurements show formation of at least one hydrogen bond. This computes a probability of bond formation per lipid molecule less than 0.1%. Since this small probability is not determined by a low density of hydroxyl radical molecules on the AFM tip surface (which according to the XPS spectra, is strongly hydroxylated by the water vapor plasma treatment), we conclude that there must be an energy barrier impeding on the hydrogen bond formation. First, we note that, before contact, the hydrogen bonding sites on the lipid and AFM tip surfaces are occupied by water molecules.32 Therefore, push of the AFM tip on the SLB must unbind water molecules from the surfaces36 in favor of the formation of hydrogen bonds between lipid heads and hydroxyl groups of the AFM tip surface. Obviously, an increase of the contact residence time to values comparable to the bond formation time under the loading force leads to a strong increase of bond formation probability. In order to determine the value of rupture force per bond, F1, we follow the method of Williams et al.37 for statistical analysis of the variance of the tip−sample pull-off force. The method is based on the assumption that the total tip−sample pull-off force is determined by the rupture of a small discrete number n of bonds, each bond contributing with the force F1, to the nonspecific (van der Waals, electrostatic) tip−sample adhesion force, F0. Therefore, the tip−sample adhesion force, Fa, is

Fa = n·F1 + F0

During the contact between the hydroxylated AFM tip and the SLB, under the applied pushing force, F, there are non equilibrium transitions between bound and unbound states of hydroxyl groups of the AFM tip surface and phosphatidylcholine (PC) molecules. Considering that from the N PC molecules in contact with the AFM tip, only n molecules form hydrogen bonds during the contact time, the time variation of the average value of n, μ(t), is determined by the non equilibrium binding and unbinding rate constants, k+(F) and k−(F), according to the equation dμ = k+(F ) ·(N − μ) − k−(F ) ·μ dt

which in the approximation μ ≪ N, is dμ = k+(F ) ·N − k−(F ) ·μ dt

μ n −μ ·e n!

d(μ − μ∞) dt

μ∞ =

−

(12)

(13)

−

where τ = 1/k (Fmax). Considering μ(0) = 0, the eq 13 is written as μ(t ) = μ∞ ·[1 − exp( −t /τ −)]

(4)

(14) −

We note that both parameters τ and μ∞ depend on the pushing force applied on the contact, F, through the dependence of rate constants, k+and k−, and N on F.

3. EXPERIMENT

(5)

As discussed in the previous section, the statistics of adhesive force measured between hydroxylated AFM tips and SLBs of phosphatidylcholine is defined by parameters as AFM tip curvature radius, loading force, and contact residence time. This section presents results of the experimental investigation of the hydrogen bond formation between hydroxylated AFM tips and SLBs taking in consideration effects of all the above-mentioned parameters. Experiments were performed with unsharpened (nominal curvature radius of 50 nm) hydroxylated AFM tips, for residence time values varying from 20 ms to 2 s, and for the loading force values varying from 0 to 1.5 nN. The variance and mean values obtained in various sets of pull-off force measurements are being used to determine the rupture force of a single hydrogen bond. Time evolution of bond formation probability at constant F is used to determine the binding and unbinding reaction rates. 3.1. Materials and Methods. 3.1.1. Preparation of Supported Lipid Bilayers on Mica Substrates. Supported lipid bilayers were formed on freshly cleaved mica through the vesicle fusion method from small unilamellar vesicles (SUVs) composed of L-α-phosphatidylcholine Type IV−S, ≥30% (Sigma-Aldrich). Thin lipid films were deposited in cylindrical glass containers with 1.5 cm in diameter through high vacuum drying of a solution of phospholipids dissolved in pure chloroform (Sigma-Aldrich) for at least 4 h. Then, the lipid films were resuspended in PBS (Sigma-Aldrich) to form a vesicle

(6)

(7)

respectively. A straightforward manipulation of the eqs 6 and 7 leads to the following linear dependence between the variance and average values of the pull-off force: 2 σFa = μFa ·F1 − F0·F1

k+(F ) ·N k−(F )

μ(t ) = μ∞ + (μ(0) − μ∞) ·exp( −t /τ −)

and 2 σFa = σ 2·F12 = μ·F12

(11)

The eq 11 has the solution

(3)

Therefore, the average and variance values of Fa are μFa = μ·F1 + F0

= −k−(F ) ·(μ − μ∞)

where μ∞ is the equilibrium (at t →∞) value of μ, given by

where μ is the mean value of the distribution. The Poisson distribution is characterized by the fact that its variance, σ2, and mean values are equal: μ = σ2

(10)

Here, μ(t) is the average value of n for a set of force spectroscopy measurements performed by keeping the contact residence time and maximum applied load constant. The equation can be easily written as

Since the probability of bond formation is very low, it is assumed that the probability distribution of n is following the Poisson distribution: P(n) =

(9)

(8)

Therefore, the values of a single bond rupture force, F1, and nonspecific tip−sample adhesion force, F0, can be found from the linear regression curve of the variance values versus the mean values of several sets of pull-off force measurements. Obviously, in order not to alter the Poisson distribution of bond formation probability, each set of pull-off force measurements should be performed with good control of the abovediscussed key parameters affecting the bond formation probability (i.e., F and tr). 16100

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104

Langmuir

Article

suspension with 1 mM lipid concentration. Continuous stirring promoted hydration and swelling of the lipid film with the formation of multivesicular and multilamellar vesicles which were then broken down into SUVs by sonication to clarity (∼20 min). Freshly cleaved mica surfaces were allowed to incubate for 45 min in the vesicle suspension, at room temperature. The excess of vesicles was removed by gently rinsing the mica sheets with small aliquots of PBS and resuspending them in fresh buffer. Samples were then transferred into the liquid cell of the atomic force microscope and were allowed to thermally equilibrate before measurements. 3.1.2. Cleaning and Hydroxylation of the AFM Probes. Prior to the force spectroscopy experiments, the AFM probes were cleaned by imbedding them successively in ethanol and chloroform (20 min for each process) in order to remove contaminant molecules adsorbed on the probe surface. However, this treatment did not remove entirely the airborne hydrophobic hydrocarbon molecules adsorbed on AFM probe surfaces.38 This has been proved by the relatively large adhesion forces (about 600 pN) measured between the AFM tips and SLBs in phosphate buffer saline (PBS) solution (see Figure S1 in the Supporting Information). Then, the AFM probes were treated in water vapor plasma to remove hydrophobic contaminant molecules and to generate hydroxyl molecules on their surfaces.39 Hydroxylation of the AFM probes has been obtained by subjecting the probes to the negative glow plasma of a dc discharge (discharge current voltage and intensity about 460 V and 5 mA, respectively) in low-pressure water vapor for 10 min. The AFM probes were mounted on a glass slide (2 cm ×2 cm ×2 mm) placed on the planar cathode (a tantalum disk with diameter of 10 cm), and the water vapor was introduced into the discharge chamber at a pressure around 30 Pa. A sketch of the plasma treatment device is provided in the Supported Information. The AFM probe surface cleaning and hydroxylation has been confirmed by X-ray photoelectron spectroscopy (XPS) analysis. Figure 2 shows the O1s peak recorded for the untreated Si3N4 surface

constant of the AFM probes were determined by measurements of the thermal noise spectra of the AFM deflection signal of the free AFM probes in air.40 We used AFM probes with unsharpened (nominal curvature radius of 50 nm) pyramidal tips. The values of curvature radius of the tips used in experiments were estimated by scanning sharp edges of a silicon standard grating probe (TGT1 from NTMDT). The force spectroscopy measurements were performed with the dedicated controlling software of the microscope in standard PBS solution (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, provided by Sigma-Aldrich) at neutral pH. The AFM loading and unloading curves were acquired at a constant speed of the scanner extension of 4 μm/s (displacement of 400 nm performed in 0.1 s) at different values of the loading force, F, and contact residence time, tr. The deflection versus scanner extension force curves acquired in the experiments were calibrated and transformed in curves of force versus tip displacement. The calibration of the cantilever deflection signal was performed assuming an infinite stiffness of the mica surface. Figure 3

Figure 3. (a) Typical force curve showing a finite adhesion force (Fa) of the hydroxylated AFM tip (curvature radius of 48 nm) to the SLB in PBS. The AFM tip is kept at the loading force, F, during the contact residence time, tr. The value of Fa (80 nN in this case) is determined as the maximum negative force recorded during the detachment (DE) of the AFM tip from the sample surface. (b) Typical force curve showing no adhesion force (Fa = 0). The value of F is chosen to be smaller than the SLB penetration force (around 1,2 nN in this case).

Figure 2. XPS spectra showing O1s peak of the AFM probe surfaces before, immediately after, and 1 week after the treatment in water vapor plasma. of an AFM probe and for the probe surface immediately after the plasma treatment and 1 week after the plasma treatment, respectively. The untreated surface showed an O1s peak attributed to Si−O bonds, while the plasma treated surface showed convoluted O1s peaks attributed to Si−O and Si−OH bonds. The XPS measurement performed on the probe surface one week after the treatment shows that the density of Si−OH bonds decreases with the time laps from the plasma treatment. 3.1.3. Atomic Force Spectroscopy Experiments. The AFM force spectroscopy measurements were performed by a commercial AFM apparatus (Nova from NT-MDT) with silicon nitride AFM probes (Microlever MLCT-AUNM from Veeco) with soft (nominal constant of 0.01 N/m) triangular cantilevers. The precise values of the spring

shows the typical curves of the force versus tip displacement with nonzero (a) and zero (b) adhesive (pull-off) force values. While the tip approaches the SLB (AB) the recorded force is zero (in average value) and affected by a noise of around 10 pN in RMS value. This indicates that the possible long-range interaction forces (double layer repulsion, van der Waals, and hidration forces)41 between the hydroxylated AFM tip and SLB in PBS are negligible small. This may be surprising because surfaces of SLBs and hydroxilated AFM tip should be slightly charged at neutral pH. A possible explanation42 for the absence of 16101

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104

Langmuir

Article

double layer force is that the surface charges are canceled by counterions coming from different surfaces (glass, stainless steel) of the AFM liquid cell or from dissolution of carbon dioxide (our AFM liquid cell is opened to the atmospheric air). This explanation is supported by our force spectroscopy experiments performed in alkaline (pH = 10) and acid (pH = 4) buffer solutions. The force curves performed in alkaline buffer solution showed a noticeable repulsive double layer force and zero adhesive force in almost all force spectroscopy experiments. On the other hand, the force spectroscopy measurements performed in acid buffer showed a weak or zero double layer repulsion force (see Figure S3 in the Supporting Information). Moreover, the measurements performed in acid buffer showed a time decrease of repulsive double layer force due to surface adsorption of counterions. After the AFM tip reaches the SLB surface (B), it compresses the SLB until the tip−sample repulsive force reaches the contact loading force value, F (BC in Figure 3). In order to avoid complications given by the formation of holes in the SLB under the tip compression force,43 the value of F was chosen to be lower than the SLB penetration force (see Figure S4 in the Supporting Information for details on measurements of SLB penetration force). The AFM tip is kept at this loading force for a certain time, which determines the contact residence time, tr, and retracted afterward. During the retraction, the tip−sample force decreases (CD) reaching a minimum negative value, which determines the value of adhesion or pull-off force (Fa), when it detaches from the sample surface (DE in Figure 3 (a)). However, most of the force curves show a zero value of Fa. Figure 3 b) presents a plot of a force curve that indicates no adhesion force between the AFM tip and SLB. The value of Fa measured in thousands of single force spectroscopy measurements for various values of F and tr on the same area (1.5 μm × 1.5 μm) of the SLB have random values ranged between 0 pN and 240 pN (see Figure S5 in the Supporting Information ). Figure 4 shows spreading plots of Fa values versus F at

Figure 5. Mean and standard deviation values of adhesive force versus maximum loading force as they resulted from the measurements shown in Figure 4. The mean and standard deviation values were computed on sets of about 300 values recorded for loading force intervals of 0.1 nN. adhesion force. This shows a small probability of hydrogen bond formation at the contact between the AFM tip and the SLB. However, this probability increases with the increase of F and tr. As a result, the average value of adhesive force increases by the increase of the contact force and residence time (Figure 5). To investigate the time evolution of hydrogen bond formation probability, we have performed force spectroscopy measurements for different values of tr at the same values of F and probe moving speed. Figure 6 shows the results of this

Figure 6. Dependence of mean and standard deviation values of adhesion force on residence contact time at a maximum loading force of 0.8 nN recorded with the hydroxylated AFM tip (Rt = 48 nm). The solid line shows a fit of the experimental data with the dependence given by eq 14.

Figure 4. Stochastic behavior of tip−sample adhesion force recorded in a number of 5000 individual force spectroscopy measurements with contact residence time of 20 ms and 1 s, respectively. The measurements were performed in PBS medium with a hydroxylated unsharpened AFM tip (curvature radius of 48 nm) on the same area of the SLB. The adhesive force has been determined as the minimum value of the negative force recorded during the AFM tip retraction (AFM probe moving speed of 4 μm/s).

investigation as a plot of mean and standard deviation values of Fa versus contact residence time. Since μFa ≈ μ·F1, the mean values of adhesive force have the time evolution described by eq 14, which determines a time constant, τ− =1/k−(F), for bond rupture of about 400 ms. The average and variance values of the adhesive force obey the relationship described by eq 8, as predicted by the Poisson distribution of the number of hydrogen bonds formed at the AFM tip−sample contact. Figure 7 shows the plot of the variance value versus the mean value of adhesion force measured in sets of experiments performed at the same F and different values of tr (values in milliseconds are indicated on the plot). The experimental data fit well a linear dependence (correlation coefficient, R = 0.96) showing values of 44.6 pN and 0 pN for F1 and F0, respectively. The zero value of F0 confirms the fact that neither the hydroxylated surface of the AFM tip nor the surface of the SLB are electrically charged in PBS at neutral pH. It is

two values of tr. The plots show that only a small fraction of force spectroscopy measurements give a noticeable adhesive force, the fraction increasing with F and tr. These experimental data were sorted according to the values of F in intervals ΔF = 0.1 nN in width, with a few hundred measurements being contained by each interval. Then, the Fa values recorded for each interval of F are considered to compute the mean and variance values of Fa. Figure 5 shows the result of this analysis as dependence of the mean and standard deviation (error bars) values of adhesion force on F and tr. 3.2. Results and Discussion. As shown in Figure 4, only a small fraction of the force spectroscopy measurements give a nonzero 16102

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104

Langmuir

Article

on each other, a fact that is characteristic of the Poisson distribution of the number of hydrogen bonds formed at the AFM tip−sample contact. This linear dependence determined a value of 44 pN for the strength of a single hydrogen bond, and a null nonspecific (van der Waals, electrostatic) tip−sample adhesion force, respectively. The mean and standard deviation values of adhesion force increase with the contact loading force as a result of increase of contact area between the AFM tip and the SLB, and removal of water molecules from the contact region. These results constitute a valuable insight on the mechanisms of hydrogen bond formation in aqueous solutions at the single-molecule level and lead to a better understanding of the phenomena associated with the interactions manifested between functionalized surfaces and model lipid membranes, with direct applications in nanomedicine and nanotoxicology.

■

Figure 7. Dependence of the variance on the mean adhesion force measured at different values of residence time (values in milliseconds shown on the plot) at a maximum loading force of 0.8 nN with the hydroxylated AFM tip (Rt = 48 nm). The experimental data fit well a linear dependence with a slope showing a rupture force F1 = 44 ± 3 pN per hydrogen bond.

ASSOCIATED CONTENT

S Supporting Information *

Plots of a typical force−distance curve (Figure S1a) and histogram (Figure S1b) of pull-off force values recorded in 350 atomic force spectroscopy measurements performed on phosphatidylcholine supported bilayers with a silicon nitride AFM tip after being cleaned in alcohol and chloroform. Sketch of the plasma treatment device used for cleaning and hydroxylation of atomic force microscopy probes (Figure S2). Measurements in alkaline and acid buffer solutions (Figure S3). Details on loading force values that determine the penetration of the SLB by the AFM tip (Figure S4). Examples of force curves showing various values of adhesive force between a hydroxylated AFM tip and a phosphatidylcholine SLB in PBS solution at neutral pH (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.

difficult to directly compare the value of F1 found here with other similar data from literature. Direct measurements of hydrogen bond rupture forces involved in the interaction between DNA nucleosides have been performed by surface force apparatus,30,42 but these measurements involve the interaction of a large number of molecules at a, practically, macroscopic contact. Moreover, the contact pressure force and residence time were not specified in these works. Pincet et al.30 estimated a hydrogen bond rupture force between adenine and thymine of 49 pN based on measurement of a pull-off force of 2.2 mN on a contact area with a radius estimated to 3 μm. Tareste et al.42 reported their results on hydrogen bond energy between DNA nucleosides based on macroscopic models of contact adhesion.

■

4. CONCLUSION The present work investigates the stochastic character of the adhesion force manifested between hydroxylated AFM tips and phosphatidylcholine supported lipid bilayers as a result of formation of a small number of hydrogen bonds at the tip− sample contact. Commercially available silicon nitride probes were hydroxylated via water vapor plasma treatment to remove hydrophobic contaminant molecules and to generate hydroxyl groups on their surfaces. The tip−sample adhesive force was assigned to hydrogen bond formation between OH groups from the AFM tip and nonester phosphate or carbonyl oxygen of phosphatidylcholine from the SLB. Analysis of the force spectroscopy measurements revealed that only a small fraction of recorded curves showed a noticeable adhesive force, the fraction increasing with the contact loading force, and contact residence time, respectively. Time evolution of hydrogen bond formation at the tip−sample contact has been investigated by performing force spectroscopy measurements for different values of contact residence time at the same values of contact loading force and probe moving speed. This investigation revealed a time evolution of the average value of bond formation probability toward an equilibrium value in a characteristic time of 400 ms, time that determines the constant rate of hydrogen bond rupture at the contact. The low value of the equilibrium value of bond formation probability indicates that the rate of hydrogen bond formation at tip−sample contact is much lower than the rate of hydrogen bond rupture. Variance and mean values of adhesive force measured in sets of force spectroscopy experiments with control of the contact loading force and residence time showed a linear dependence

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by CNCSIS, IDEI Research Program of Romanian Research, Development and Integration National Plan II, Grant No. 267/2011. The authors thank Tudor Luchian for his support on probe preparation and Marius Dobromir for XPS measurements.

■

REFERENCES

(1) Loose, M.; Schwille, P. Biomimetic membrane systems to study cellular organization. J. Struct. Biol. 2009, 168, 143−151. (2) Chan, Y.-H. M.; Boxer, S. G. Model membrane systems and their applications. Curr. Opin. Chem. Biol. 2007, 11, 581−587. (3) Smith, A. S.; Sackmann, E. Progress in mimetic studies of cell adhesion and the mechanosensing. ChemPhysChem 2009, 10, 66−78. (4) Moghimi, S. M.; Hunter, A. C.; Murray, J. C. Nanomedicine: Current status and future prospects. FASEB J. 2005, 19, 311−330. (5) Wang, J.; Byrne, J. D.; Napier, M. E.; DeSimone, J. M. More effective nanomedicines through particle design. Small 2011, 7, 1919− 1931. (6) Medina, C.; Santos-Martinez, M. J.; Radomski, A.; Corrigan, O. I.; Radomski, M. W. Nanoparticles: Pharmacological and toxicological significance. Br. J. Pharmacol. 2007, 150, 552−558. (7) Roach, P.; Eglin, D.; Rohde, K.; Perry, C. C. Modern biomaterials: A review - bulk properties and implications of surface modifications. J. Mater. Sci. Mater. Med. 2007, 18, 1263−1277.

16103

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104

Langmuir

Article

(8) Vasir, J. K.; Labhasetwar, V. Quantification of the force of nanoparticle−cell membrane interactions and its influence on intracellular trafficking of nanoparticles. Biomaterials 2008, 29, 4244−4252. (9) Loomis, W. F.; Fuller, D.; Gutierrez, E.; Groisman, A.; Rappel, W. Innate non-specific cell substratum adhesion. PloS One 2012, 7, 42033. (10) Lipowsky, R. Generic Interactions of Flexible Membranes. Handbook of Biological Physics; Lipowsky, R. and Sackmann, E., Eds.; Elsevier Science B.V.: Amsterdam, 1995; Vol. 1. (11) Evans, E. A. Adhesion of surfactant−membrane covered droplets: Special features and curvature elasticity effects. Colloids Surf. 1990, 43, 327−347. (12) Andersson, M. X.; Goksör, M.; Sandelius, A. S. Membrane contact sites: Physical attachment between chloroplasts and endoplasmic reticulum revealed by optical manipulation. Plant Signal Behav. 2007, 2, 185−187. (13) Schwingel, M.; Bastmeyer, M. Force mapping during the formation and maturation of cell adhesion sites with multiple optical tweezers. PLoS One 2013, 8, e54850. (14) Marra, J.; Israelachvili, J. N. Direct measurements of forces between phosphatidylcholine and phosphatidylethanolamine bilayers in aqueous electrolyte solutions. Biochemistry 1985, 24, 4608−4618. (15) Dufrêne, Y. F. Application of atomic force microscopy to microbial surfaces: From reconstituted cell surface layers to living cells. Micron 2001, 32, 153−165. (16) Simon, A.; Durrieu, M. -C. Strategies and results of atomic force microscopy in the study of cellular adhesion. Micron 2006, 37, 1−13. (17) Wei, Z. Q.; Wang, C.; Zhu, C. F.; Zhou, C. Q.; Xu, B.; Bai, C. L. Study on single-bond interaction between amino-terminated organosilane self-assembled monolayers by atomic force microscopy. Surf. Sci. 2000, 459, 401−412. (18) Jeffrey, G. A.; Saenger, W. Hydrogen Bonding in Biological Structures; Springer-Verlag: Berlin/New York, 1991; p 569. (19) Rudolph, M. G.; Stanfield, R. L.; Wilson, I. A. How TCRs bind MHCs, peptides, and coreceptors. Annu. Rev. Immunol. 2006, 24, 419− 466. (20) Pace, C. N.; Shirley, B. A.; McNutt, M.; Gajiwala, K. Forces contributing to the conformational stability of proteins. FASEB J. 1996, 10, 75−83. (21) Bowie, J. U. Membrane protein folding: how important are hydrogen bonds? Curr. Opin. Struct. Biol. 2011, 21, 42−49. (22) White, S. H. How hydrogen bonds shape membrane protein structure. Adv. Protein Chem. 2005, 72, 157−172. (23) Bailly, C.; Waring, M. J. The use of diaminopurine to investigate structural properties of nucleic acids and molecular recognition between ligands and DNA. Nucleic Acids Res. 1998, 26, 4309−4314. (24) Goodman, M. F. Hydrogen bonding revisited: Geometric selection as a principal determinant of DNA replication fidelity. Proc. Natl. Acad. Sci. U. S. A. 1997, 94, 10493−10495. (25) Biertümpfel, C.; Zhao, Y.; Kondo, Y.; Ramón-Maiques, S.; Gregory, M.; Lee, J. Y.; Masutani, C.; Lehmann, A. R.; Hanaoka, F.; Yang, W. Structure and mechanism of human DNA polymerase eta. Nature 2010, 465, 1044−1048. (26) Toth, G.; Bowers, S. G.; Truong, A. P.; Probst, G. The role and significance of unconventional hydrogen bonds in small molecule recognition by biological receptors of pharmaceutical relevance. Curr. Pharm. Des. 2007, 13, 3476−3493. (27) Bondar, A.-N.; White, S. H. Hydrogen bond dynamics in membrane protein function. Biochim. Biophys. Acta 2012, 1818, 942− 950. (28) Baker, E. N.; Hubbard, R. E. Hydrogen bonding in globular proteins. Prog. Biophys. Mol. Biol. 1984, 44, 97−179. (29) Lindauer, K.; Bendic, C.; Suhnel, J. HBexplore - A new tool for identifying and analyzing hydrogen bonding patterns in biological macromolecules. Comput. Appl. Biosci. 1996, 12, 281−289. (30) Pincet, F.; Perez, E.; Bryant, G.; Lebeau, L.; Mioskowski, C. Specific forces between DNA bases. Mod. Phys. Lett. B 1996, 10, 81− 99.

(31) Boland, T.; Ratner, B. D. Direct measurement of hydrogen bonding in DNA nucleotide bases by atomic force microscopy. Proc. Natl. Acad. Sci. U. S. A. 1995, 92, 5297−5301. (32) Pasenkiewicz-Gierula, M.; Takaoka, Y.; Miyagawa, H.; Kitamura, K.; Kusumi, A. Hydrogen bonding of water to phosphatidylcholine in the membrane as studied by a molecular dynamics simulation: location, geometry and lipid−lipid bridging via hydrogen bonded water. J. Phys. Chem. A 1997, 101, 3677−3691. (33) Das, C.; Sheikh, K. H.; Olmsted, P. D.; Connell, S. D. Nanoscale mechanical probing of supported lipid bilayers with atomic force microscopy. Phys. Rev. E 2010, 82, 041920. (34) Efimov, A. V.; Brazhnikov, E. V. Relationship between intramolecular hydrogen bonding and solvent accessibility of sidechain donors and acceptors in proteins. FEBS Lett. 2003, 554, 389− 393. (35) Hertz, H. On the contacts of elastic solids. J. Reine Angew. Math. 1882, 92, 156−171. (36) Lopez, C. F.; Nielsen, S. O.; Klein, M. L.; Moore, P. B. Hydrogen bonding structure and dynamics of water at the dimyristoylphosphatidylcholine lipid bilayer surface from a molecular dynamics simulation. J. Phys. Chem. B 2004, 108, 6603−6610. (37) Williams, J. M.; Taejoon, H.; Beebe, T. P., Jr. Determination of single-bond forces from contact force variances in atomic force microscopy. Langmuir 1996, 12, 1291−1295. (38) Sirghi, L.; Kylian, O.; Gilliland, D.; Ceccone, G.; Rossi, F. Cleaning and hydrophilization of atomic force microscopy silicon probes. J. Phys. Chem. B 2006, 110, 25975. (39) Sirghi, L. Plasma cleaning of silicon surface of atomic force microscopy probes. Romanian J. Phys. 2011, 56, 144−148. (40) Burnham, N. A.; Chen, X.; Hodges, C. S.; Matei, G. A.; Thoreson, E. J.; Roberts, C. J.; Davies, M. C.; Tendler, S. J. B. Comparison of calibration methods for atomic-force microscopy cantilevers. Nanotechnology 2003, 14, 1−6. (41) Israelachivili, J. N. Intermolecular and Surface Forces; Academic Press: London, 1992. (42) Tareste, D.; Pincet, F.; Perez, E.; Rickling, S.; Mioskowski, C.; Labeau, L. Energy of hydrogen bonds probed by the adhesion of functionalized lipid layers. Biophys. J. 2002, 83, 3675−3681. (43) Dufrene, Y. F.; Boland, T.; Schneider, J. W.; Barger, W. R.; Lee, G. U. Characterization of the physical properties of model biomembranes at the nanometer scale with the atomic force microscope. Faraday Discuss 1998, 111, 79−94.

16104

dx.doi.org/10.1021/la404534r | Langmuir 2013, 29, 16098−16104