Biomedical Microdevices

, Volume 11, Issue 2, pp 331–338

Fabrication of protein chips based on 3-aminopropyltriethoxysilane as a monolayer

Authors

    • Department of Electrical Engineering and Center for Micro/Nano Science and TechnologyNational Cheng Kung University
  • Hao-Juin Liu
    • Department of Electrical Engineering and Center for Micro/Nano Science and TechnologyNational Cheng Kung University
Article

DOI: 10.1007/s10544-008-9239-7

Cite this article as:
Jang, L. & Liu, H. Biomed Microdevices (2009) 11: 331. doi:10.1007/s10544-008-9239-7

Abstract

Although 3-aminopropyltriethoxysilane (APTES) is widely adopted as a monolayer in biosensors, experimental silanization takes at least 1 h at high temperature. Therefore, the feasibility of the silanization with APTES in a short reaction time and at room temperature was investigated. The surface modification of glass slides using a self-assembled monolayer of APTES with a concentration of 10% was studied by immobilizing FITC. APTES was successfully immobilized on the glass slide. The effect of reaction temperature and time of silanization were investigated. Various silanization conditions of APTES were examined by contact angle measurement and fluorescence microscopy. The surface of glass patterns with a gold thin film as background was characterized by determining the fluorescent intensities following the immobilization of fluorescein isothiocyanate (FITC), protein A-FITC, antimouse IgG-FITC and sheep anti-bovine albumin-FITC. The normalized fluorescent intensity indicated that a short period (4 min) of silanization at 25°C suffices to form an APTES thin film by the immobilization of protein A on a glass surface. Such a condition does not require microheaters and temperature sensors in a microfluidic system, which will significantly reduce the manufacturing process, cost, and reaction time in the future.

Keywords

Self-assembled monolayerAPTESProtein chip

1 Introduction

As interest in the applications of protein chips and biosensors grow, the topic of surface modification is becoming important as it pertains to the fabrication process (Zhu and Snyder 2001). Assembling proteins on solid substrates that are coupled with microfluidics are expected to become important in the biomedical applications. Most current applications of protein arrays involve disease proteomics and conventional glass-based arrays (Haab et al. 2001). Other protein arrays are available (MacBeath and Schreiber 2000), although glass-based arrays still dominate the market for reasons of cost and because the research community has gained much experience of using DNA arrays. Glass slides are low-cost and highly sensitive for detection in biosensoring applications. A glass slide is as a transparent substrate, and so is useful in the detection analysis on microarray chips. They have been evaluated as substrates using MEMS and NEMS for biosensing in terms of its covalent bonding to protein, and can be used in a wide range of applications on biochips (Liao et al. 2005). Bio-microelectromechanical systems (Bio-MEMS) and lab-on-a-chip devices have been developed from synergetic applications of microsystem technology and biotechnology. The design and fabrication of biologically functional surfaces are important to the production of such microdevices. They have driven the development of novel surface nanotechnologies, including protein patterning methods (Blawas and Reichert 1998). Selectively modified surfaces, with biologically functional components, are critical working elements of biosensors (Zaytseva et al. 2005) and devices for cell manipulation and tissue engineering. Microfluidic biochannel arrays that integrate microfluidic channels with glass-based microarray biochips have been developed (Jo et al. 2000). Glass surfaces have been modified using many hydrophilic polymers to fabricate protein chips and biosensors. Most reactive coatings of glass surfaces are produced by self-assembly approaches.

Self-assembled monolayer (SAM) technique has recently been studied to determine the potential applications of protein fabrication and antibody molecular layers on solid surfaces. Many fundamental biological recognition systems require biological surfaces to be biocompatible interfaces for transduction (Breen et al. 1999). SAMs are a powerful tool for generating monolayers to fabricate protein biosensors. A protein biosensor must have an artificial biomolecular monolayer. These thin organic films increase the biocompatibility of the surface and protect the proteins from denaturation and structural changes during immobilization. The main goal of surface modification for fabricating protein chips is to provide good accessibility to immobilize proteins and other biomolecules. The modified surface must also minimize the uncontrolled non-specific adsorption of proteins and provide stability to the protein-bound interface. Self-assembled monolayers (SAMs) of aminosilanes depend on hydroxylated surfaces as substrates for their formation. The driving force of this self-assembly is the in-situ formation of polysiloxane, which is connected to surface silanol groups (SiOH) through Si–O–Si bonds.

Solid surfaces have been modified for a wide range of applications in various fields. Among the various surface modification techniques, the deposition of a self-assembled monolayer (SAM) or multilayer of organosilane is very versatile, and has many advantages over other approaches. The modification of glass slides using amino-terminated silanes can be adopted to immobilize proteins, antibodies and antigens. Many organosilanes have been employed to form SAMs or multilayers on such surfaces; among these, amino-terminated SAMs and multilayers are of particular interest. Silanization experiments with APTES, to produce amino functional groups for the simple immobilization of biomolecules, have been extensively performed to elucidate their use in protein chips and biosensors. Aminosilanes such as 3-aminopropyltriethoxysilane (APTES) are appealing for such applications, due in part to much advances in our understanding of this class of surface modification agents. SAM of APTES and other silane compounds on glass surfaces can be used to immobilize a variety of functional groups via surface reactions, since the amine groups are highly reactive toward various functional groups that can be incorporated into SAMs. APTES can readily react with hydroxyl groups and can support the formation of monolayer coverage under carefully controlled conditions, as described elsewhere (Kallury et al. 1992). Mild reaction conditions in the subsequent step can be applied to generate a density gradient of aldehyde. A wide range of silanization conditions are mentioned in relation to the formation of biochips, and are strongly affected by the stability of the terminal functional groups (Ulman 1996). Although 3-aminopropyltriethoxysilane (APTES) is widely adopted as a monolayer in biosensors, experimental silanization takes at least 1 h at high temperature (Ercole et al. 2002; Vos et al. 2007; Arslan et al. 2006; Liao and Cui 2007). Therefore, the feasibility of the silanization with APTES in a short reaction time and at room temperature was investigated.

Much care must be taken to maintain the three-dimensional structure and bioactivity of proteins, because they have unique characteristics—especially its sensitivity to substrate surfaces. The activity of proteins, including immunoglobulin (IgG), immobilized on a solid surface, is typically weaker than that in the aqueous phase, mainly because the orientation of the protein molecules on a solid substrate is random. A well-known protein A is used as a binding material to produce a well-defined antibody surface. Protein A, a cell wall component of Staphylococcus aureus, can bind with the Fc part of the antibody. The fabrication of nanoscale structures with SAMs to combine protein A with solid substrates has attracted much attention because marked interest exists in two-dimensional molecular assemblies and their potential applications in molecular devices, sensors and surface engineering. Therefore, the use of protein A leads to highly efficient immunoreactions and improves the performance of detection systems (Neubert et al. 2002; Susmel et al. 2000; Tanaka and Mastsunaga 2000).

We aim to develop a microfluidic system for disease diagnosis. The system will include a transparent protein microarray, an image sensor, a light source, a micropump, microchannels and so on. In this work, the protein mciroarrays were fabricated using APTES as a monolayer and protein A. The feasibility of the silanization with APTES in a short reaction time and at room temperature was investigated. The effects of reaction time and temperature on silanisation of APTES were investigated by the immobilization of protein A on glass substrates. Finally, immobilization experiments were conducted on the modified surfaces using rabbit anti-bovine albumin–BSA–sheep anti-bovine albumin conjugated with FITC.

2 Methods and materials

2.1 Materials

In the experiments, microscopic glass (75 × 25 mm) slides were used as substrates. A 10% (vol./vol.) solution of 3-aminopropyltriethoxysilane, NH2(CH2)3Si(OC2H5)3 (Aldrich 28177-8) in ethanol was used to aminate the glass surface, as described by Maj-Britt Stark and Krister Holmberg (Stark and Holmberg 1989). Tween 20 and glutaraldehyde 25% solution (Grade II) were obtained from Sigma (St. Louis, MO, USA). Fluorescein isothiocyanate (FITC), dimethyl sulfoxide (DMSO), anti-mouse IgG conjugated with FITC, bovine serum albumin (BSA) and protein A-FITC were purchased from Sigma-Aldrich Chemical Co. The primary antibody 1.0 mg/ml rabbit anti-bovine albumin and the secondary antibody 1.0 mg/ml fluorescein isothiocyanate-labeled sheep anti-bovine albumin were obtained from BETHYL Laboratories, ING. The solvents, such as acetone, ethanol and methanol and other chemicals were of analytical grade and obtained commercially.

2.2 Fabrication of glass patterns

Glass slides (75 × 25 mm) were used as solid supports. All substrates were cleaned in piranha (H2SO4:H2O2 = 3:1) and then rinsed in distilled water, acetone, isopropanol and distilled water, in that order, to remove the organic residue from the surface. The photoresist was then patterned on the substrates by photolithography. Chromium (Cr) was initially sputtered onto the glass substrate as an adhesion layer with a thickness of 150 nm, and then a 650 nm-thick layer of gold (Au) was sputtered using an E-beam evaporator. The substrate was cleaned using acetone to peel off the gold by the lift-off technique, immersed in ethanol solution for 30 min and blown dry with nitrogen. Finally, the microarray patterns of diameter 50 μm, 100 μm and 200 μm with the volume of 1,570 μm3, 6,280 μm3, 25,120 μm3, respectively, were fabricated on the slides. A photograph of the protein chip was taken using a 20× magnifying objective lens, as presented in Fig. 1.
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig1_HTML.gif
Fig. 1

Photograph of microarray protein chip. The magnified image is of the protein patterning region that contains the glass patterns and the gold thin film as background

2.3 Measurement of contact angle

The surface hydrophobicity was measured by a contact angle goniometry (MigicDrop, Future Digital Scientific. Corp, USA). The relative humidity of the environment was 47%, and the temperature varied between 25 and 27°C. The glass substrates were firstly cleaned in a mixture of H2SO4 and H2O2 (3:1) for 30 min at room temperature, thoroughly rinsed with ethanol and deionized water, and then dried in an oven. Then, the cleaned and oxidized glass substrates were silanised with a 10% (vol./vol.) solution APTES in ethanol for 4 min, 1 h, 12 h and 24 h at room temperature (25°C) and 65°C. Finally, the chips were sonicated in ethanol for 10 min and then were rinsed in ethanol and deionized water to remove non-covalently adsorbed silane compounds. Three droplets with the volume of 5 μl were placed at various spots on the treated glass substrates. The average readings and standard deviations are reported. All the contact angle measurements were performed within 30 min after the silanization of glass substrates with APTES.

2.4 Immobilization of FITC on SAM layer

The APTES layer was immobilized with FITC to ensure that the APTES treatment of the glass surface provided sufficient imine bridges to the proteins. FITC is an amine reactive fluorescent compound that forms a stable fluorescent conjugate (FITC–APTES conjugate) when reacts with APTES containing a primary amine group. In this conjugate a stable linkage makes FITC covalently attached to the APTES silane compound (Yun et al. 2007). Firstly, the glass chips were cleaned in a mixture of H2SO4 and H2O2 (3:1) for 30 min at room temperature, thoroughly rinsed with ethanol and deionized water, and then dried in an oven. Secondly, the chips were silanised with a 10% (vol./vol.) solution APTES in ethanol for 4 min, 1 h and 24 h at room temperature (25°C). Then, the generated amino groups on the glass surface were labelled with the amine-reactive dye FITC, which was bound to the surface by immersing the samples in 10 mg FITC/100 ml DMSO for 24 h. The samples were washed with DMSO and ethanol in an ultrasonic bath and blown dry using nitrogen to remove unreacted FITC.

2.5 Immobilization of protein A-FITC

The covalent bonding of proteins to the glass substrate was studied by immobilizing protein A-FITC. The glass substrates were silanized with a 10% APTES in ethanol at 25°C and 65°C for reaction times of 4 min, 10 min, 20 min and 30 min. The chips were sonicated in ethanol for 10 min and then were rinsed in ethanol to remove non-covalently adsorbed silane compounds and blown dry in nitrogen. Glutaraldehyde activation was performed in a 10% solution of GA in a PBS solution (pH 7.4, 0. 2 g KCl, 1.44 g NaHPO4, 8 g NaCl and 0.24 g KH2PO4 in 1 L distilled water) at 25°C for 1 h to reduce bridging and maximize yield. The chips were then sonicated for 5 min and washed in deionized water. After the chips were silanized with APTES and immersed in glutaraldehyde to generate terminal aldehyde groups, the concentration of 0.333 mg/ml protein A-FITC in PBS was immobilized on the treated samples for 1 h at 25°C. Following immobilization with protein A-FITC, the chips were washed three times in PBS (pH 7.0), which contained 0.1% (vol./vol.) Tween 20.

2.6 Immobilization of antiMouse-FITC

The glass chips were immobilized with anti-mouse IgG conjugated with FITC to demonstrate the activation of the protein A. After silanization with a 10% solution of APTES in ethanol for a the reaction time of 4 min at 25°C and treatment with aldehyde and the fabrication of protein A on the glass patterns, the antiMouse-FITC antibody was immobilized; 1 mg/ml FITC-antimouse IgG solution was dropped onto the treated glass patterns. The glass chips were incubated for 1 h at 4°C in a moisture chamber. Following incubation, the glass patterns were rinsed in PBS and deionized water to remove physically absorbed antibodies.

2.7 Immobilization of FITC-labeled antibody

Immobilization of protein A helps to control the orientation of the immobilized biomolecules, including antibodies or antigen on the surface. Protein A can bind with the Fc part of the antibody which reduces the non-specific binding of the antibody. Therefore, the use of protein A leads to highly efficient immunoreactions, and enhances detection system performance (Neubert et al. 2002; Susmel et al. 2000; Tanaka and Mastsunaga 2000). Protein adsorption experiments were performed with bovine serum albumin (BSA) as a model protein to analyze surface adsorption. However BSA is a protein noted for its high binding ability to surfaces and future microarrays will include a BSA blocking step to reduce non-specific interactions. Then, experiments were conducted to immobilize the first rabbit anti-bovine albumin, the BSA layer and the second fluorescein isothiocyanate-labeled sheep anti-bovine albumin. The binding reaction was monitored by measuring the fluorescence intensity from the bound FITC–sheep anti-BSA. In same manner as above, a 10% solution of APTES in ethanol was silanized on the cleaned glass surface at 25°C for a reaction time of 4 min. Following silanization, glutaraldehyde activation and immobilization of protein A were complete. Firstly, the 1 mg/ml rabbit anti-bovine albumin in PBS was dropped onto the protein A layer and incubated for 1 h at 4°C. After incubation of rabbit anti-bovine albumin, the chips were rinsed with PBS and deionized water to remove the physically absorbed antibodies. Secondly, the chips were immersed in 1 mg /ml BSA and incubated for 1 h at 4°C and then rinsed in PBS and deionized water. Finally, the 1 mg/ml FITC sheep anti-bovine albumin FITC conjugated in PBS was dropped onto the BSA layer and incubated at 4°C for 1 h, and then the chips were rinsed in PBS and deionized water. All of the chips were dried under flowing nitrogen gas to determine the fluorescent intensities. Figure 2 presents a sequence of the immobilization of FITC–sheep anti-bovine albumin on glass microarray patterns.
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig2_HTML.gif
Fig. 2

Sequence of immobilization of secondary antibody on BSA layer as a model; (1) pre-treatment of glass substrate with piranha solution; (2) silanization of APTES and treatment with aldehyde; (3) immobilization of protein A; (4) immobilization of rabbit anti-bovine albumin; (5) immobilization of BSA layer; (6) immobilization of FITC–sheep anti-bovine albumin

2.8 Fluorescence labeling with biomolecules and image acquisition

Fluorescence microscopy has been demonstrated to be a valuable approach for the easy detection of interactions among the biomolecules that are conjugated with FITC on substrates. All fluorescent images herein were obtained using a Nikon ECLIPSE 80i fluorescence microscope and a CCD camera (Evoluting VF, Q-imaging, USA). FITC was excited at 450–490 nm using a hydrargyrum lamp. Micro patterned substrates were viewed through a 10× magnifying objective lens. Since the fluorescence yields an emission peak in the frequency that corresponds to the color green, only the green component of the images was analyzed. Image-Pro Plus 5.0 (Media Cybernetics, USA) was used to calculate the fluorescent intensity of the images. The fluorescent images were examined to study the binding capacity of the biomolecules. The background intensity was deducted from the measured value. The results were then normalized to the fluorescent intensity of the control sample: IN (normalized intensity) = IA (mean intensity) / IM (intensity of the control sample). The control sample was defined as one with the maximum fluorescence intensity in each set of experiments. The normalized fluorescent intensity was given in relative fluorescence units (y axis).

3 Results and discussion

3.1 Contact angle

The static contact angle measured with DI water indicates the surface hydrophobicity. The contact angles for various reaction conditions of APTES on the treated glass slides are listed in Table 1. All cleaned glass slides were very hydrophilic with a measured contact angle of 8°. APTES SAMs increased the surface hydrophobicity with contact angles ranging between 63.8° and 69.9°. The contact angle increased slightly from 65.2 ± 0.8° to 69.9 ± 1.2° by the reaction time of 4 min to 24 h at 25°C, while a series of reaction conditions at 65°C is slightly less hydrophobic than at 25°C. Figure 3 depicts the contact angle trend of a series of reaction conditions at 25°C is comparable to that at 65°C.
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig3_HTML.gif
Fig. 3

Contact angle vs. reaction time of APTES at 25°C and 65°C

Table 1

Static contact angle for various reaction conditions of APTES on the treated glass slides

Cleaned glass slides (contact angle °) 

 

Reaction temperature

25°C

65°C

4 min

65.2 ± 0.8°

63.8 ± 0.5°

1 h

66.8 ± 0.2°

65.9 ± 0.7°

12 h

68.5 ± 0.5°

69.3 ± 0.4°

24 h

69.9 ± 1.2°

68.9 ± 0.5°

3.2 Chemical modification with fluorescein isothiocyanate

The fluorescent intensities of the amino functionalized glass substrates was measured after fluorescein isothiocyanate (FITC) had been coupled to the amino moieties on the surface to verify the covalent bonding between substrates and SAMs (Kim et al. 2004). The results reveal that the normalized fluorescent intensity was between 0.22 and 1. The normalized fluorescent intensity of the treated substrates without APTES was 0.22 ± 0.029 in Fig. 4(a). Most of the other glass substrates had higher fluorescence intensities, of 0.95 ± 0.027, 0.96 ± 0.023, 1 ± 0.025 at reaction times of 4 min, 1 h and 24 h, at 25°C, as presented in Fig. 4(b)–(d). Additionally, the normalized intensity exhibits a 5% increase in silanization from 4 min to 24 h at room temperature (25°C) and the standard deviation is small. The experimental results show that the self-assembled monolayer was successfully fabricated on the glass surfaces in a short reaction time (4 min).
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig4_HTML.gif
Fig. 4

The fluorescent images of APTES-FITC patterns (diameter: 50 μm; center-to-center pitch: 150 μm) (a) without APTES, and for different silanization reaction times; (b) 4 min, (c) 1 h and (d) 24 h at 25°C

3.3 Effects of reaction time and temperature on APTES

The fluorescent intensities of the treated glass substrates after protein A-FITC was immobilized were studied to investigate the APTES characteristics. In this work, silanizations were performed for 4 min, 10 min, 20 min and 30 min at 65°C and room temperature (25°C). The normalized fluorescent intensities of the amino functionalized glass patterns (diameter = 50 μm) were measured following immobilization of protein A-FITC. Figure 5 presents the normalized fluorescent intensities of the reaction time of silanisation from 4 min to 30 min. According to John A. Howarter’s work, the glass surface, which constrained the silanes to react primarily at the substrate surface, absorbed most of the water that was present in the system for low temperatures (Howarter and Youngblood 2006). As the temperature was increased, water was desorbed from the substrate. Although the amount of water desorbed from the surfaces of the substrate was not so great as to dehydrate the surfaces completely (Silberzan et al. 1991), the growth of the film was probably affected. Water is consumed by the condensation reaction when it occurs in solution, but water acts as a catalyst of the surface reaction (Howarter and Youngblood 2006). The results reveal that the normalized fluorescent intensities at silanization reaction times of 4 min to 30 min, between 0.9 and 0.96 at 65°C, were lower than those, between 0.94 and 1, at 25°C. Therefore, the temperature influenced APTES binding because the normalized fluorescent intensity at 25°C exceeded that at 65°C. The normalized fluorescent intensities of silanisation at 65°C and at 25°C for 4 min, which is the shortest reaction time in this study, were 0.93 ± 0.023 and 0.98 ± 0.024 respectively. The results indicated that a short treatment at 25°C suffices to modify the surface. The final silanization condition for the formation of SAMs was 4 min at 25°C. The experimental results also show that the protein A-FITC was successfully immobilized on APTES.
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig5_HTML.gif
Fig. 5

Normalized fluorescent intensity of the samples under silanization conditions of 25°C and 65°C for 4 to 30 min

3.4 Activation of the protein A

Glass chips were immobilized with anti-mouse IgG that was conjugated with FITC to prove the activation of protein A. Selective protein patterns of various sizes were used to immobilize proteins on the samples by quantitative analysis of the normalized fluorescent intensity. As a model system, a protein A-patterned array was used to capture FITC–antimouse IgG. The fluorescent normalized intensity was employed to study the activation of protein A. The normalized fluorescent intensities of the microarray patterns with the diameters of 50 μm, 100 μm, and 200 μm were 0.78 ± 0.02, 0.91 ± 0.015, and 1 ± 0.026, respectively. The results for each of the arrays (Fig. 6) revealed little standard deviation from the mean, which is consistent with favorable spot morphology and signal uniformity, as well as the low variability in binding capacity across the slide surface. The low background was consistent with low non-specific protein binding and yielded good signal-to noise ratios. Figure 6 plots the normalized fluorescent intensity and standard deviation for each pattern size. These results reveal that the activation of the protein A favors immobilized antimouse IgG.
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig6_HTML.gif
Fig. 6

Normalized fluorescent intensity of samples immobilized by antimouse-FITC for various pattern sizes with standard deviations

3.5 Protein chips based on BSA model

Protein adsorption experiments were performed with rabbit anti-bovine albumin (first antibody), BSA (intermedium) and fluorescein isothiocyanate-labeled sheep anti-bovine albumin (secondary antibody) as a model. The antiBSA–FITC array comprised three differently sized circles with diameters of 50, 100 and 200 μm. Figure 7(a) reveals that the normalized fluorescent intensity from the 50 μm diameter patterns to the 200 μm diameter patterns were 0.54 ± 0.024, 0.69 ± 0.023 and 1 ± 0.020. The analysis of normalized intensity data indicated a small standard deviation under the experimental conditions herein. Figure 7(b) shows the fluorescent images of immobilized fluorescein isothiocyanate-labeled sheep anti-bovine albumin. The results reveal that APTES can be used as a monolayer between glass and biomolecules.
https://static-content.springer.com/image/art%3A10.1007%2Fs10544-008-9239-7/MediaObjects/10544_2008_9239_Fig7_HTML.gif
Fig. 7

(a) Normalized fluorescent intensity of samples immobilized by fluorescein isothiocyanate-labeled sheep anti-bovine albumin with various sizes of patterns; (b) Fluorescent images of fluorescein isothiocyanate-labeled sheep anti-bovine albumin with pattern diameters of 50 μm, 100 μm, and 200 μm

4 Conclusion

The fabrication process of glass protein chips using APTES and protein A was successfully developed. The contact angle results show the similar trend for various reaction times at 25°C and 65°C. The immobilization of FITC demonstrates that APTES with a concentration of 10% was successfully fabricated onto glass surfaces in a short reaction time (4 min). Additionally, the normalized fluorescent intensity curve at 65°C was lower than that at 25°C, which may result from the effect of temperature on the formation of APTES. The results reveal that the reaction period (4 min) of silanization at both 65°C and 25°C was associated with the high intensity following the immobilization of protein A-FITC, indicating that a short treatment at 25°C is sufficient to modify the surface. Furthermore, the results of immobilized antimouse IgG reveal that protein A maintains good activation under the silanization condition of APTES (4 min and 25°C). Finally, rabbit anti-bovine albumin–BSA–sheep anti-bovine albumin conjugated with FITC was used to establish a model to demonstrate that APTES can be used as a monolayer between glass and biomolecules at the current silanization condition (4 min and 25°C). In the future work, an optimal condition of silanization of APTES, the stability of immobilization reaction and the robustness of the microarrays will be investigated.

Acknowledgements

The authors would like to thank the National Science Council of Taiwan, the Republic of China, for financially supporting this research under Contract No. NSC 95-2622-E-006-039-CC3 and NSC 94-2218-E-006-043. The authors also would like to thank the Center for Micro/Nano Science and Technology, National Cheng Kung University, Tainan, Taiwan, for access to equipment and technical support. Furthermore, this work made use of Shared Facilities supported by the Program of Top 100 Universities Advancement, Ministry of Education, Taiwan.

Copyright information

© Springer Science+Business Media, LLC 2008