Headgroup Effects of Template Monolayers on the Adsorption

May 24, 2011 - The effect of the template monolayers on the adsorption behavior of GOx was ... This material is available free of charge via the Inter...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/Langmuir

Headgroup Effects of Template Monolayers on the Adsorption Behavior and Conformation of Glucose Oxidase Adsorbed at Air/Liquid Interfaces Ke-Hsuan Wang, Mei-Jywan Syu, Chien-Hsiang Chang, and Yuh-Lang Lee* Department of Chemical Engineering, National Cheng Kung University, Tainan 70101, Taiwan

bS Supporting Information ABSTRACT: Stearic acid (SA) and octadecylamine (ODA) monolayers at the air/ liquid interface were used as template layers to adsorb glucose oxidase (GOx) from aqueous solution. The effect of the template monolayers on the adsorption behavior of GOx was studied in terms of the variation of surface pressure, the evolution of surface morphology observed by BAM and AFM, and the conformation of adsorbed GOx. The results show that the presence of a template monolayer can enhance the adsorption rate of GOx; furthermore, ODA has a higher ability, compared to SA, to adsorb GOx, which is attributed to the electrostatic attractive interaction between ODA and GOx. For adsorption performed on a bare surface or on an SA monolayer, the surface pressure approaches an equilibrium value (ca. 8 mN/m) after 2 to 3 h of adsorption and remains nearly constant in the following adsorption process. For the adsorption on an ODA monolayer, the surface pressure will increase further 1 to 2 h after approaching the first equilibrium pressure, which is termed the second adsorption stage. The measurement of circular dichroism (CD) spectroscopy indicates that the LangmuirBlodgett films of adsorbed GOx transferred at the first equilibrium state (π = 8 mN/m) have mainly a β-sheet conformation, which is independent of the type of template monolayers. However, the ODA/GOx LB film transferred at the second adsorption stage has mainly an R-helix conformation. It is concluded that the specific interaction between ODA and GOx not only leads to a higher adsorption rate and adsorbed amount of GOx but also induces a conformation change in adsorbed GOx from β-sheet to R-helix. The present results indicate that is possible to control the conformation of adsorbed protein by selecting the appropriate template monolayer.

’ INTRODUCTION Lipids and proteins are the main components of cell membranes. The interaction between protein and lipids is an issue of fundamental interest that has been studied for decades.13 These studies are important not only to the understanding of biological systems but also to the development of biomolecular devices such as biosensors.411 Biosensors are always constructed by immobilizing specific enzymes onto electrode surfaces. This immobilization was commonly performed by encapsulating an enzyme in lipid matrixes. Various techniques have been employed to immobilize the enzyme, including self-assembly,46 a solgel process, and a LangmuirBlodgett (LB) technique.710 Among these methods, the LB technique takes advantage of preparing ultrathin films with precisely controlled thickness and molecular orientation. Furthermore, the LB technique also provides a versatile method to study the proteinlipid interaction at the air/liquid interface as well as the penetration of protein across a lipid layer. Because of the amphiphilic property of protein molecules, proteins tend to adsorb spontaneously from the bulk solution to the air/liquid interface. It is also known that the presence of a monolayer at the air/liquid interface is helpful in the adsorption of proteins and the adsorption behavior of a protein is determined by the proteinmonolayer interaction.1222 Various template r 2011 American Chemical Society

monolayers had been used to study their interaction with proteins, including dipalmitoyl-phosphatic acid (DPPA),12 dipalmitoylphosphatidylcholine (DPPC),13,14 dibehenoyl-phosphatidylcholine (DBPC),13,15,16 dimyristoyl-phosphtidylcholine (DMPC),13 stearic acid (SA),17,18 behenic acid,1921 and octadecylamine (ODA).18,22 The chain length and charge of a monolayer were found to be important factors affecting the binding behavior of proteins to the template monolayer. In a previous study,12 a DPPA monolayer at the air/liquid interface was used as a template layer to incorporate GOx from the subphase. On the basis of the variation in surface pressure, two adsorption stages of GOx were found in the presence of the DPPA monolayer. However, only one adsorption stage was observed in the absence of a template monolayer. Although the adsorption behavior and related characteristics of the mixed GOx/DPPA monoayer were clearly described in that work, the reason leading to these phenomena is still unclear. In this work, two simple molecules (SA and ODA) are used as template monolayers to adsorb GOx. The effects of functional groups amine and carboxylic acid on the adsorption behavior of GOx are studied in terms of the surface pressure variation, the evolution of Received: March 15, 2011 Published: May 24, 2011 7595

dx.doi.org/10.1021/la201496j | Langmuir 2011, 27, 7595–7602

Langmuir

ARTICLE

surface morphology observed by Brewster angle microscopy (BAM) and atomic force microscopy (AFM), and the conformation of adsorbed GOx. This study indicates that the electrostatic interaction between GOx and the template monolayer can greatly enhance the adsorption of GOx and furthermore induce a conformational change in adsorbed GOx from a β-sheet to an R-helix. These effects trigger a further increase in surface pressure in the later adsorption stage, termed as the second adsorption stage of GOx.

’ EXPERIMENTAL SECTION Materials. Chloroform (>99%, J. T. Baker), octadecylamine (ODA, >99%, Aldrich), and stearic acid (SA, >99%, Fluka) were used as received. GOx (type VII, from Aspergillus niger, 180 200 U/g) was supplied from Sigma. All experiments were performed using doubly distilled water (resistivity = 18.2 MΩ cm) purified with a Milli-Q apparatus supplied by Millipore. The subphase was prepared by dissolving GOx in pure water, and the concentration of GOx was mainly controlled at 0.83 mg/L to study the effect of template layers. However, higher GOx concentrations were also used to study the concentration effect of GOx. Methods. Monolayer experiments were performed in a computercontrolled film-balance apparatus (model KSV minitrough) constructed by KSV Instruments Ltd., Finland. A Teflon trough with a working area of 32  7.5 cm2 was placed on a vibration isolation table and closed in an environmental chamber. The trough was mounted on an aluminum base plate with built-in water channels for subphase temperature control. All experiments were performed at 25 °C. The film surface pressure was measured by using a Wilhelmy plate arrangement attached to a microbalance. ODA and SA were dissolved in chloroform to prepare a stock solution with a concentration of 1 mg/mL. Before the spreading of ODA/SA stock solution, possible GOx adsorbed at the air/liquid interface was removed by sweeping the subphase surface with a capillary pipet connected to a vacuum pump. When zero surface pressure was reached, the stock solution was spread immediately. The variation in surface pressure was recorded for 8 h in the following adsorption process. The morphological characteristics of a monolayer at the air/liquid interface were examined directly by a Brewster angle microscopy (BAM) designed by Nanofilm Technology (NFT), G€ottingen, Germany (model BAM2 plus). p-polarized light from an Nd:YAG laser with a wavelength of 532 nm was incident at the Brewster angle (53.1°) to the air/liquid interface. The images were visualized by a CCD camera and were recorded during the adsorption process. The lateral resolution of BAM was about 2 μm. To prepare LB films, freshly cleaved mica plates (Alfa Aesar) were used as substrates. The substrate was immersed in the subphase before the spreading of the monolayer. After waiting for a specific adsorption time, the monolayer at the air/liquid interface was transferred at the surface pressure that the monolayer attained after GOx adsorption. Therefore, the surface pressure at which a monolayer was transferred depends on the type of template monolayer and the adsorption time of GOx. During the film-transfer process, the surface pressure of the monolayer was kept constant by suitable compression of the monolayer using a barrier. For the atomic force microscope (AFM) study, one monolayer was transferred at a rate of 1 mm/min. The morphology of the LB film was examined using an AFM (Nano Scope IIIa, supplied by Digital Instruments Inc.) via tapping mode at a scan rate of 0.5 Hz. A silicon tip with a cantilever of 125 μm length (noncontact silicon cantilever, model NCH-50, Germany) was used. The secondary structure of GOx adsorbed at the air/liquid interface was examined by circular dichroism (CD) spectroscopy (Jasco J-715 spectropolarimeter, Japan). For this measurement, 100 layers of the

Figure 1. Variation of the surface pressure with the adsorption time of GOx. The GOx adsorption was carried out at an air/liquid interface in the absence or in the presence of various compositions of mixed SA/ODA monolayers. The molecular density of the monolayers was controlled over a molecular area of 36.32 Å2/molecule, and the concentration of GOx in the subphase was controlled at 0.83 mg/L. monolayer were transferred onto quartz plates to obtain an intensity high enough for this measurement. The transfer rates were controlled at 1 and 150 mm/min, respectively, in the upward and downward strokes. K2D2 software was used to estimate the GOx secondary structure from the CD spectra. All of the experimental data shown in this work have good reproducibility, which was confirmed by repeated experiments.

’ RESULTS AND DISCUSSION The adsorption behavior of GOx from the subphase to the air/ liquid interface was evaluated first by the variation of surface pressure. Figure 1 shows the results with and without the presence of an SA/ODA monolayer. For GOx adsorption on a bare liquid surface (curve f), the surface pressure increases slowly and approaches a constant value (ca. 7.8 mN/m) after about 3 h of adsorption. This result is consistent with that reported in previous work.12 In the presence of a template monolayer, the GOx adsorption rate is enhanced as indicated by the faster increase in the surface pressure. Furthermore, the adsorption rate is proportional to the ratio of ODA contained in the monolayer. For the pure ODA monolayer, which demonstrates the fastest adsorption rate, the surface pressure increases more quickly and approaches a constant value (ca. 8 mN/m) within 1 h. Here, this constant pressure state is termed as the first equilibrium stage of GOx adsorption. This state appears for all of the systems studied here, and the equilibrium surface pressures are nearly identical. However, the time required to reach this state decreases with increasing ratio of ODA in the monolayer, indicating a higher adsorption rate in the presence of ODA. Furthermore, the adsorption rate can also be increased by increasing the concentration of GOx or the density of the ODA monolayer. (See the data in the Supporting Information, Figures S1 and S2.) It is interesting to find that, although the GOx adsorption rate can be controlled by the concentration of GOx and the monolayer density, the surface pressure of the first equilibrium stage is nearly independent of these factors. As will be discussed in a later section, the main conformation of GOx in the early adsorption stage is the β-sheet, 7596

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602

Langmuir which may be associated with the same surface pressure found for all systems. For the monolayers with a higher molar ratio of SA, no significant change in the surface pressure was observed after approaching the first equilibrium stage. (The pressure remains nearly constant after 24 h of adsorption.) However, for the ODAricher monolayers, the first equilibrium stage lasts for several hours (24 h) and then the surface pressure increases again (termed the second adsorption stage) when the adsorption time is further prolonged. The time required to initiate the second adsorption stage is shorter for an ODA-richer monolayer. Also, increases in the concentration of GOx and the density of ODA have an enhancement effect to initiate the second adsorption stage of GOx. For the pure ODA monolayer shown in Figure S1, the first equilibrium stage cannot be identified when the GOx concentration is increased up to 9.96 mg/L, which means that the second adsorption stage performs immediately after approaching the first equilibrium stage; therefore, the transition stage cannot be observed. This result is attributed to the rapid adsorption of GOx. For the second adsorption stage, the adsorption of GOx proceeds at a much slower rate in comparison with the first stage. For most of the systems, an equilibrium stage cannot be approached within 8 h. Only for the system with a much higher GOx concentration (9.96 mg/L) and a pure ODA monolayer (Figure S1) can an equilibrium stage be approached at about 4 h with an equilibrium surface pressure of ca. 27 mN/m. The results described above demonstrate that ODA is superior to SA in incorporating the GOx. This interaction not only leads to a higher adsorption rate but also induces the second adsorption stage that cannot be performed by an SA monolayer or in the absence of a template monolayer. Comparing the molecular structures of SA and ODA, the amine group of ODA is responsible for the specific interaction of the ODA monolayer on GOx. The subphase used in the present work has a pH value of 5.5. Because the isoelectric point of GOx is known to be 4.2, the GOx molecule is negatively charged in the subphase, which triggers an electrostatic interaction with the positively charged ODA molecules.12,23 For the SA monolayer, the electrostatic attraction force did not happen between SA and GOx, which is why a lower adsorption rate was observed for the SA monolayer. However, compared to the adsorption on a bare liquid surface (curve f), a slight enhancement of the SA monolayer with respect to GOx adsorption was also observed in the first adsorption stage. This result implies that other interactions, such as hydrogen bonds and dispersion forces, should also contribute to GOx adsorption. Especially for the ODA monolayer, the hydrogen bond between GOx and the amine group of ODA may also contribute to the specific attraction of ODA to GOx. It is interesting to find that the second adsorption stage of GOx occurs only in a system with an ODA ratio higher than 0.5. Comparing a mixed system with an ODA ratio of 0.5 (curve c in Figure 1) to a pure ODA system of the same ODA density (72.64 Å2/molecule) shown in Figure S2 (curve b), the second adsorption stage occurs in the pure ODA monolayer but not in the mixed system. This result implies that the coexistence of SA inhibits the specific adsorption ability of ODA molecules to GOx. It is inferred that the functional groups of SA (COOH) interact with amine groups (NH2) of ODA and thereby the specific adsorption ability of amine groups to GOx cannot occur. It is likely that the SAODA interaction inhibits the ionization of amine groups in the subphase, which eliminates the electrostatic

ARTICLE

attraction force between ODA and GOx. This specific SAODA interaction was reported in previous work24 that coincides with the present result. On the basis of the results of the present work, the SAODA interaction is likely to perform in one to one ratio. The adsorption kinetics of a protein to an interface had been studied in the literature,15,2528 and three consecutive or concurrent processes were always considered in this process: (1) the diffusion of protein from the bulk phase to a sublayer near the interface; (2) the adsorption of the molecules from the sublayer to the interface; and (3) the structural reorganization of adsorbed molecules at the interface. In the initial stage of protein adsorption where the surface pressure is relatively low, diffusion is always the rate-determining step. In this stage, the variation of surface pressure (π) with time (t) can be correlated by a modified form of Ward and Tordai’s equation.29   Dt 1=2 π ¼ 2CkT ð1Þ 3:14 where C is the protein concentration in the bulk phase, k is the Boltzmann constant, T is the temperature, and D is the diffusion coefficient. According to eq 1, a plot of π against t1/2 will be linear if the adsorption is controlled by diffusion. To check this relationship, the data in Figure 1 were replotted for several typical compositions (shown in Supporting Information, Figure S3). In the initial stage, linear relationships can be found for all of the systems, implying that the early stage of GOx adsorption is a diffusion-controlled process. By using Ward and Tordai’s equation, the diffusion coefficient corresponding to the constant slope period can be calculated. The diffusion coefficients measured for various systems were shown in the Supporting Information (Table S1). The apparent diffusion coefficient of GOx onto a bare air/liquid interface is 2.89  1011 m2/s. For the template monolayers of SA and ODA, the diffusion coefficient increases to 4.79  1011 and 1.26  1010 m2/s, respectively. Apparently, the presence of a template monolayer increases the diffusion rate of GOx; furthermore, the diffusion rate is the highest for the ODA monolayer, which involves an electrostatic attractive interaction with GOx. Surface Morphology Studied by Brewster Angle Microscopy. Brewster angle microscopy was used to observe the evolution of surface morphology at the air/liquid interface along the GOx adsorption process. On a bare air/liquid interface, no defined structure can be seen in the early period of GOx adsorption up to 1 h (Figure 2a), and dim-cloud morphology was acquired after ca. 2 h (Figure 2b, π ≈ 7 mN/m). With an increase in the adsorption time, the adsorption phase becomes more clear and demonstrates the dendritic morphology after about 5 h of adsorption (Figure 2c). This adsorption phase grows very slowly in the following adsorption period without an obviously change in the phase morphology up to 8 h. These BAM images indicate that the adsorbed amount of GOx increases slightly with increase in the adsorption time. However, the rare population of the adsorption phase and the small gray level (GL) increment of the BAM images, from 19 in Figure 2a to 23 in Figure 2c, imply a limited amount of adsorbed GOx. For the GOx adsorption performed in the presence of an SA monolayer, the evolution of the BAM images was shown in Figure 3. The characteristic morphology of an SA monolayer, the mosaic texture, was observed after spreading the SA stock solution on the subphase (Figure 3a). The coexistence of the dark domains (which have an elliptical shape) indicates that the 7597

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602

Langmuir

ARTICLE

Figure 2. BAM images of the GOx adsorption monolayer in the absence of a template monolayer. The concentration of GOD in the subphase was controlled at 0.83 mg/L. The images were taken at (a) 1, (b) 2, and (c) 5 h of GOx adsorption. The inset indicates the surface pressure at which an image was taken.

Figure 3. Evolution of BAM images for the adsorption of GOx onto an SA template monolayer. The molecular area of SA was controlled at 36.32 Å2 /molecule, and the concentration of GOx in the subphase was 0.83 mg/L. The images were taken at (a) 0, (b) 1, (c) 2, (d) 3, (e) 4, and (f) 8 h of GOx adsorption. The inset indicates the surface pressure at which an image was taken.

SA monolayer was in a gas/liquid coexistence state. However, the liquid phases are not as uniform as observed on a pure water subphase,24 implying that a certain amount of GOx has been incorporated into the SA monolayer in the initial stage. With increasing adsorption time, densely distributed pine holes appear inside the phase domains of SA (Figure 3b) and the holes grow gradually with the further incorporation of GOx (24 h, Figure 3ce). No significant variation of the morphology was observed in the following adsorption period up to 8 h. Because the GL value increases gradually (from 59 to 78) in the adsorption process, the appearance of black holes cannot be ascribed to the loss of monolayer in the subphase. It is more likely that the incorporation of GOx causes the destruction and/or reorganization of the SA domains. It seems that SA domains tend to aggregate into a more condensed phase after the incorporation of GOx, which leads to more area of the expanded phases appearing as black holes under the observation of BAM. It is noteworthy that, although the surface pressure remains constant after 2 h of adsorption, the variation of the BAM morphology and the related GL indicate that the GOx adsorption still progresses gradually in this period. However, the increment of the adsorbed GOx has no effect on the increase in surface pressure.

Figure 4 shows the BAM image evolution for the GOx adsorption in the presence of an ODA monolayer. In the initial adsorption stage, no defined phase can be identified from BAM (Figure 4a), indicating that the ODA monolayer was presented as the gas phase before adsorption. The difference between phases of ODA and SA monolayers in the initial stage is attributed to the difference in the moleculesubphase interaction. The amine group (NH2) of ODA is known to have a higher interaction with water, compared with the COOH group of SA, which triggers a more expanded phase of the ODA monolayer at the air/water interface. After about 1 h of adsorption, distributed dim dots were observed (Figure 4b), as inferred to be the combination phase of GOx and ODA. With increasing adsorption time, the size and brightness of the adsorption phase increase gradually (Figure 4c,d) and the phases are mainly displayed as roundlike shape. In the following period, further adsorption causes the growth and coalescence of the adsorption phases, leading to the formation of larger round-shaped or irregular domains (Figure 4e). A closer inspection of the structure inside the larger domains finds that a large fraction of black holes still exists inside these domains. Actually, these domains were organized by dendritic phases, a typical phase reported for a pure ODA 7598

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602

Langmuir

ARTICLE

Figure 4. Evolution of BAM images for the adsorption of GOx onto a ODA template monolayer. The molecular area of ODA was controlled at 36.32 Å2/molecule, and the concentration of GOx in the subphase was 0.83 mg/L. The images were taken at (a) 0, (b) 1, (c) 2, (d) 3, (e) 6, and (f) 8 h of GOx adsorption. The inset indicates the surface pressure at which an image was taken.

Figure 5. AFM images (2  2 μm2) of GOx LB films transferred onto mica substrates. GOx was adsorbed in the absence of a template monolayer, and one layer of the monolayer was transferred after the adsorption of (a) 2 and (b) 8 h without compression. The concentration of GOx in the subphase was controlled at 0.83 mg/L. The inset indicates the surface pressure at which the monolayer was transferred.

monolayer at the air/water interface.23,24 That is, although the ODA monolayer in the gas state cannot be observed, the incorporation of GOx forms a more condensed and visible phase, reflecting the population of the ODA monolayer on the water surface. By prolonging the adsorption time, the further adsorption of GOx creates a more condensed phase, which also decreases the area of the expanded phase (the pores) dispersed among the domains. However, a pore-free condensed ODA/ GOx mixed monolayer cannot be achieved after 8 h of adsorption (Figure 4f). These results indicate that a much higher amount of GOx was adsorbed on the ODA monolayer in comparison with that on the SA monolayer. This inference not only is consistent with the result of the surface pressure variation but also is sustained by the higher increment of GL (from 19 in Figure 4a to 78 in Figure 4 f). Surface Morphology Studied by Atomic Force Microscopy. The monolayers at the air/liquid interface were transferred onto mica substrates to examine their film morphology by atomic force microscopy (AFM). One layer of the monolayer was transferred for the AFM study. It is worthwhile to note that

the magnification of AFM is about 100 times higher than that of BAM. Therefore, a much finer structure can be observed by AFM. Figure 5 shows the morphology of the GOx adsorption monolayer prepared in the absence of a template layer. For the LB film transferred after 2 h of adsorption (π = 7.6 mN/m), outstanding structures with round or oval shapes were observed (Figure 5a). These structures have heights of ca. 0.5 to 0.6 nm, which is inferred to be the adsorbed GOx. When the adsorption time increases to 8 h (π = 7.8 mN/m), the further incorporation of GOx triggers a smoother film morphology and the height of the adsorption phase decreases slightly to ca. 0.4 nm. The sparse distribution of the adsorption structures and their small heights indicate that only a few GOx molecules were adsorbed on the air/ liquid interface. Figure 6 shows the morphology of mixed SA/GOx monolayers obtained after various adsorption times. It demonstrates that uniform domains with height of ca. 0.7 to 0.8 nm were observed for the SA/GOx system. This phase is inferred to be a structure constructed by the adsorption of GOx onto liqud-state SA domains, corresponding to the condensed phase observed in the BAM image. In the early period of GOx adsorption (1 h, Figure 6a, π = 5.8 mN/m), the film contains a significant ratio of pores ascribed to the coexisting gasliquid state of the SA monolayer. This inference was sustained by the BAM images shown in Figure 3. With increasing adsorption time, the pores gradually become smaller (2 h, Figure 6b, π = 8.1 mN/m) and disappear after ca. 5 h of adsorption (Figure 6c, π = 8.3 mN/m). The evaluation of these AFM images implies that the adsorption proceeds steadily during this period although the surface pressure remains nearly constant after 2 h of adsorption. Furthermore, the further incorporation of GOx triggers a more condensed phase without the black pores. This result is consistent with the previous model, inferred from the BAM images, that the SA domains tend to aggregate into a more condensed phase after the incorporation of GOx. In general, the mixed SA/GOx LB film is quite uniform, which also reflects that this film 7599

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602

Langmuir

ARTICLE

Figure 6. AFM images (2  2 μm2) of mixed SA/GOx LB films transferred onto mica substrates. The molecular area of SA was controlled at 36.32 Å2/molecule, and the concentration of GOx in the subphase was 0.83 mg/L. One layer of the monolayer was transferred after the adsorption of (a) 1, (b) 2, and (c) 5 h without compression. The inset indicates the surface pressure at which the monolayer was transferred.

Figure 7. AFM images (2  2 μm2) of mixed ODA/GOx LB films transferred onto mica substrates. The molecular area of ODA was controlled at 36.32 Å2/molecule, and the concentration of GOx in the subphase was 0.83 mg/L. One layer of the monolayer was transferred after the adsorption of (a) 1, (b) 5, and (c) 8 h without compression. The inset indicates the surface pressure at which the monolayer was transferred.

morphology is dominated by SA because of the limited amount of GOx adsorbed on the SA monolayer. For the adsorption of GOx on the ODA monolayer, no regular domains can be observed from the corresponding AFM images, which are ascribed to the gas state of the ODA monolayer used to perform the adsorption. After ca. 1 h of adsorption, outstanding structures with dotlike or dendritic shapes were observed (Figure 7a, π = 8.2 mN/m). The heights of these structures range from 4 to 5 nm, clearly reflecting the adsorbed GOx. The adsorption phases grow quickly with increased adsorption time and organize into a netlike morphology after ca. 5 h of adsorption (Figure 7b, π = 13.1 mN/m). Apparently, two distinct phases were formed in this stage. The first has a brighter image that appears as a network structure. The other exhibits a dim brightness, covering the entire background. The appearance of the two phases is ascribed to the different states of the ODA monolayer present in the initial adsorption stage. In the following adsorption stage, the adsorption phases grow quickly as revealed by the AFM image acquired after 8 h of adsorption (Figure 7c, π = 15.4 mN/m). The evolution of the AFM images clearly indicates that the adsorption phases increase significantly both in height and coverage ratio, which also implies that a much higher amount of GOx is incorporated by the ODA monolayer. Conformation of Adsorbed GOx. It is known that GOx consists of two major conformations, R-helix and β-sheet. The effect of the template monolayers on the conformation of the adsorbed GOx was studied by the measurement of circular dichroism (CD) spectroscopy. For this measurement, the adsorbed monolayer at the air/liquid interface was transferred to quartz substrates by LB deposition, and 100 layers of the monolayers were transferred to obtain an intensity high enough for this evaluation. On the basis of the spectrum of an LB film, the intensities corresponding to the R-helix (R) and β-sheet (β) can

Table 1. Conformation of Adsorbed GOx Prepared under Various Conditions adsorption time (h) 2h

8h

template monolayer

R-helix

β-sheet

R/β

bare

1.86

49.03

0.04

SA

2.11

51.77

0.04

ODA

4.21

48.82

0.09

SA ODA

9.59 53.77

46.66 7.21

0.21 7.46

be calculated. Table 1 lists the intensities of the R-helix and βsheet and their ratio (R/β) under various adsorption conditions. For the GOx monolayer adsorbed on a bare water interface (without a template monolayer), the measured conformation ratio (R/β) is 0.04, which is smaller than the ratio reported for GOx dispersed in water (0.23).30 That is, GOx has mainly a βsheet conformation when it is adsorbed at the air/liquid interface. In the presence of a template monolayer, the monolayers were transferred after 2 or 8 h of adsorption. For the 2-h-adsorbed LB films, the measured R/β ratio is 0.04 for SA and 0.09 for ODA. These results demonstrate that, in the first equilibrium stage of GOx adsorption (2 h of adsorption), the conformation of GOx adsorbed onto the SA or ODA monolayer is nearly identical to that adsorbed on a bare surface, although the ODA monolayer has a slightly higher ratio. For the long-time adsorption (8 h), the R/β ratio increases slightly to 0.21 for the SA monolayer. However, a significant increase in the ratio of up to 7.46 was observed for the ODA monolayer. This result indicates that, for the SA monolayer, increasing the adsorption time has no significant effect on the conformation change in adsorbed GOx, but for the ODA monolayer, the major conformation of the adsorbed GOx changes to an R-helix in the second 7600

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602

Langmuir adsorption stage. Apparently, the conformation of the adsorbed GOx in the early adsorption stage is mainly β-sheet, no matter what kind of template monolayer was used. Furthermore, a significant conformational change from β-sheet to R-helix is carried out only in the later adsorption period where the second adsorption stage can be performed. Because the adsorption stages were defined on the basis of the change in the surface pressure, the conformational change in GOx is likely associated with the significant elevation of surface pressure in the second adsorption stage. On the basis of the results of BAM and AFM, the adsorbed amount of GOx in the first adsorption stage is low in comparison with that in the second adsorption stage. In the first adsorption stage, although GOx adsorption performs steadily after approaching the first equilibrium stage, the increment of adsorbed GOx is not significant. The limited amount of adsorbed GOx is responsible for the low and nearly constant surface pressure (88.5 mN/m) observed in the first equilibrium stage. It is interesting to find that the films transferred at the surface pressure corresponding to the first equilibrium stage (ca. 8.0 mN/m) all have the major conformation of the β-sheet. This result is independent of the adsorption time and the template monolayers. For the second adsorption stage of GOx onto an ODA template monolayer, a significant amount of GOx was further incorporated and a conformational transformation from a β-sheet to an R-helix was also associated. Therefore, the elevation of surface pressure in this stage can be attributed to the large increment in adsorbed GOx and a change in the conformation. On the basis of the thermodynamic point of view, a spontaneous process should lead to a more stable state with a lower surface energy (higher surface pressure). Therefore, it is likely that the R-helix conformation is a more stable structure that can lead to a lower surface energy (higher surface pressure). In the present study, the second adsorption stage and the conformational change of GOx occur only on the template monolayer of ODA. That is, the specific adsorption ability of ODA onto GOx triggers not only a higher adsorption rate and adsorbed amount of GOx but also a conformational change in adsorbed GOx. In the present stage, it is still not clear why the ODA monolayer can trigger the conformational transformation. The possible reasons are the electrostatic attractive interaction between adsorbed GOx and ODA, the hydrogen bond interaction between the two molecules, and the structure transfer induced by the accumulation of a significant amount of GOx at the air/liquid interface. The results obtained in this work indicate that it is able to control the conformation of adsorbed GOx by the selection of appropriate template monolayers. Furthermore, the surface pressure of an adsorbed GOx monolayer is an efficient parameter indicating the main conformation of adsorbed GOx.

’ CONCLUSIONS A template monolayer at the air/liquid interface can enhance the adsorption of GOx from aqueous solution, and the enhancement ability is dependent on the interaction of template molecules with GOx. For the ODA monolayer that possesses an electrostatic attraction interaction with GOx, a higher adsorption rate and adsorbed amount of GOx were achieved compared with that obtained by an SA monolayer. The conformation of adsorbed GOx in the early adsorption stage is mainly β-sheet, and the equilibrium surface pressure achieved by this conformation is ca.

ARTICLE

8 mN/m, which is independent of the types of template monolayers. For the SA template monolayer, the surface pressure and GOx conformation do not change significantly in the later adsorption stage. However, for the ODA monolayer, the electrostatic interaction can induce a second adsorption stage of GOx, which triggers a significant increase in adsorbed GOx and a further increase in surface pressure. A conformational transformation of adsorbed GOx from β-sheet to R-helix was also associated with the later adsorption stage.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel: (þ) 886-6-2757575 ext. 62693. Fax: (þ) 886-6-2344496. E-mail: [email protected].

’ ACKNOWLEDGMENT This work was sponsored by the National Science Council of Taiwan under contract no. NSC 97-2221-E-006-145 MY3. ’ REFERENCES (1) Patino, J. M. R.; Garcia, J. M. N.; Nino, M. R. R. Colloids Surf., B 2001, 21, 207–216. (2) Damodaran, S. Anal. Bioanal. Chem. 2003, 376, 182–188. (3) Serrano, A. G.; Perez-Gil, J. Chem. Phys. Lipids 2006, 141, 105–118. (4) Puu, G. Anal. Chem. 2001, 73, 72–79. (5) Oh, S. Y.; Park, J. K.; Ko, C. B.; Choi, J. W. Biosens. Bioelectron. 2003, 19, 103–108. (6) Ohnuki, H.; Saiki, T.; Kusakari, A.; Endo, H.; Ichihara, M.; Izumi, M. Langmuir 2007, 23, 4675–4681. (7) Sriyudthsak, M.; Yamagishi, H.; Moriizumi, T. Thin Solid Films 1988, 160, 463–469. (8) Sun, S. C.; Hosi, P. H.; Harrison, D. J. Langmuir 1991, 7, 727–737. (9) Wan, K.; Chovelon, J. M.; Jaffrezic-Renault, N. Talanta 2000, 52, 663–670. (10) Yasuzawa, M.; Hashimoto, M.; Fujii, S.; Kunugi, A.; Nakaya, T. Sens. Actuators B 2000, 65, 241–243. (11) Kim, Y.; Do, J.; Kim, J.; Yang, S. Y.; Malliaras, G. G.; Ober, C. K.; Kim, E. Jpn. J. Appl. Phys. 2010, 49, 01AE101–6. (12) Lee, Y. L.; Lin, J. Y.; Lee, S. Langmuir 2007, 23, 2042–2051. (13) Zhang, J.; Rosilio, V.; Goldmann, M.; Boissonnade, M. M.; Baszkin, A. Langmuir 2000, 16, 1226–1232. (14) Kamilya, T.; Pal, P.; Talapatra, G. B. J. Colloid Interface Sci. 2007, 315, 464–474. (15) Rosilio, V.; Boissonnade, M. M.; Zhang, J. Y.; Jiang, L.; Baszkin, A. Langmuir 1997, 13, 4669–4675. (16) Li, J. R.; Rosilio, V.; Boissonnade, M. M.; Baszkin, A. Colloids Surf., B 2003, 29, 13–20. (17) Zaitsev, S. Y. Colloids Surf., A 1993, 75, 211–216. (18) Chovelon, J. M.; Wan, K.; Jaffrezic-Renault, N. Langmuir 2000, 16, 6223–6227. (19) Fiol, C.; Valleton, J. M.; Delpire, N.; Barbey, G.; Barraud, A.; Ruaudelteixier, A. Thin Solid Films 1992, 210, 489–491. (20) Dubreuil, N.; Alexandre, S.; Fiol, C.; Valleton, J. M. J. Colloid Interface Sci. 1996, 181, 393–398. 7601

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602

Langmuir

ARTICLE

(21) Dubreuil, N.; Alexandre, S.; Lair, D.; Sommer, F.; Duc, T. M.; Valleton, J. M. Colloids Surf., B 1998, 11, 95–102. (22) Yin, F.; Shin, H. K.; Kwon, Y. S. Thin Solid Films 2006, 499, 1–7. (23) Tsai, H. J.; Lee, Y. L. Soft Matter 2009, 5, 2962–2970. (24) Lee, Y. L.; Yang, Y. C.; Shen, Y. J. J. Phys. Chem. B 2005, 109, 4662–4667. (25) Graham, D. E.; Phillips, M. C. J. Colloid Interface Sci. 1979, 70, 403–414. (26) Hunter, J. R.; Kilpatrick, P. K.; Carbonell, R. G. J. Colloid Interface Sci. 1990, 137, 462–482. (27) Douillard, R.; Lefebvre, J. J. Colloid Interface Sci. 1990, 139, 488–499. (28) Macritchie, F. Colloids Surf., A 1993, 76, 159–166. (29) Ward, A. F. M.; Tordaï, L. J. Chem. Phys. 1946, 14, 453. (30) Dai, G. L.; Li, J. R.; Jiang, L. Colloids Surf., B 1999, 13, 105–111.

7602

dx.doi.org/10.1021/la201496j |Langmuir 2011, 27, 7595–7602