This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 129.170.70.145

This content was downloaded on 29/07/2015 at 19:51

Please note that terms and conditions apply.

Enhanced microcontact printing of proteins on nanoporous silica surface

View the table of contents for this issue, or go to the journal homepage for more

2010 Nanotechnology 21 415302

(http://iopscience.iop.org/0957-4484/21/41/415302)

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING NANOTECHNOLOGY

Nanotechnology 21 (2010) 415302 (9pp) doi:10.1088/0957-4484/21/41/415302

Enhanced microcontact printing ofproteins on nanoporous silica surfaceEllen Blinka1, Kathryn Loeffler2, Ye Hu1, Ashwini Gopal1,Kazunori Hoshino1, Kevin Lin1, Xuewu Liu2, Mauro Ferrari2 andJohn X J Zhang1

1 Department of Biomedical Engineering, University of Texas at Austin, Austin,TX 78758, USA2 Department of Nanomedicine and Biomedical Engineering, University of Texas HealthScience Service, Houston, TX 77031, USA

E-mail: John.Zhang@engr.utexas.edu

Received 14 June 2010, in final form 4 August 2010Published 13 September 2010Online at stacks.iop.org/Nano/21/415302

AbstractWe demonstrate porous silica surface modification, combined with microcontact printing, as aneffective method for enhanced protein patterning and adsorption on arbitrary surfaces.Compared to conventional chemical treatments, this approach offers scalability and long-termdevice stability without requiring complex chemical activation. Two chemical surfacetreatments using functionalization with the commonly used 3-aminopropyltriethoxysilane(APTES) and glutaraldehyde (GA) were compared with the nanoporous silica surface on thebasis of protein adsorption. The deposited thickness and uniformity of porous silica films wereevaluated for fluorescein isothiocyanate (FITC)-labeled rabbit immunoglobulin G (R-IgG)protein printed onto the substrates via patterned polydimethlysiloxane (PDMS) stamps. A morecomplete transfer of proteins was observed on porous silica substrates compared to chemicallyfunctionalized substrates. A comparison of different pore sizes (4–6 nm) and porous silicathicknesses (96–200 nm) indicates that porous silica with 4 nm diameter, 57% porosity and athickness of 96 nm provided a suitable environment for complete transfer of R-IgG proteins.Both fluorescence microscopy and atomic force microscopy (AFM) were used for protein layercharacterizations. A porous silica layer is biocompatible, providing a favorable transfer mediumwith minimal damage to the proteins. A patterned immunoassay microchip was developed todemonstrate the retained protein function after printing on nanoporous surfaces, which enablesprintable and robust immunoassay detection for point-of-care applications.

(Some figures in this article are in colour only in the electronic version)

1. Introduction

Microcontact printing (μCP) is a stamping based method forthe dry transfer of molecules onto a surface [1–3]. The method,introduced by Whitesides in 1993 [4, 5] and extended to thetransfer of protein molecules by Bernard et al in 2000 [6], hasreceived much scientific attention as a means to immobilizeproteins in controlled patterns for biomedical applicationssuch as biosensors and immunoassays [7–13]. Patterning ofbiomolecules onto solid substrates is relevant to providingcontrolled biocompatible surfaces [14, 15]. Microcontactprinting may be used to pattern multiple proteins on the samesubstrate, thus allowing for biological assays with greater

diagnostic power [16–19]. A number of possible applicationswould be enabled if the same surface could be patternedwith proteins of different types [20, 21]. In this paper, wecompared the efficiency of protein transfer via μCP ontoporous silica coated, 3-aminopropyltriethoxysilane (APTES)treated, glutaraldehyde (GA) treated, and untreated glasssubstrates.

Substrate surfaces are often functionalized to increasetheir affinity for specific proteins [13, 22]. Commonfunctionalization procedures for preparing surfaces for proteinbinding utilize silane compounds with terminal functionalgroups that interact electrostatically or covalently withprotein surface groups to increase protein adsorption. Two

0957-4484/10/415302+09$30.00 © 2010 IOP Publishing Ltd Printed in the UK & the USA1

Nanotechnology 21 (2010) 415302 E Blinka et al

common chemicals used for surface functionalization areAPTES, chemical formula (NH2)–(CH2)3–Si(OC2H5)3, andGA, chemical formula (CHO–(CH2)3–CHO) [23–28]. Thesilane end of the APTES molecule binds covalently tosurface silicon atoms, and the amino end of the moleculeincreases protein adsorption to the surface through electrostaticinteractions. The linear GA molecule is terminated on bothends by aldehyde groups and thus may be used to transformsurface amino groups to surface aldehyde groups. Thesealdehyde groups covalently bind proteins through the Schiffbase reaction [22]. While the use of chemical treatments isoften effective at increasing adsorption of protein, it requirescomplex chemical activation and introduces problems such asstress around the silane–substrate interface, which can occurduring temperature cycling, or atmospheric water hydrolyzingeither the oxane bonds between silane and the substrate or thebonds between the organo-functional group of the silane andthe organic molecule [8].

Previously, porous silica thin films have been used for thesize-selective trapping of proteins from a biological complex,such as human serum [29–31]. We have designed andcharacterized a large set of porous silica thin films with variouspore nanotextures to enhance the capacity and efficiency ofspecific protein enrichment [29]. Here we show that a thinlayer of porous silica can be used to improve the efficiencyof protein transfer to a surface via μCP. Porous silica can bedeposited in an inexpensive procedure based on spin coatingthat requires no specialized techniques, such as sputtering orchemical vapor deposition. The porous silica here was createdon solid substrates by using triblock copolymers as structuraldirection agents, mixed with silica solution and deposited overthe surface. Once the porous silica was created, subsequentdeposition of proteins was performed using microcontactprinting. The study reported herein has potential applicationssuch as patterned integrated micro-devices for immunoassaydetection, micro-total analysis systems, and other biomedicalpoint-of-care applications. Porous silica has long-term stabilityand is compatible with mainstream microfabrication processesin the semiconductor industry.

2. Experimental details

2.1. Preparation of porous silica

2.1.1. Fabrication of mesoporous silica thin films. A typicalpreparation of the porous silica thin film was carried outas follows (figure 1). The coating precursor was prepared,starting with self-assembly between polymer units to formthe surfactant micelle and mixing it with soluble silicates(TEOS, tetraethoxysilane) in homogeneous, hydro-alcoholicsolutions. The evaporation of solvent during spin coatingresults in an increase of the concentration of the polymerin the solution to exceed the critical micelle concentrationand drives silica/copolymer self-assembly into a uniform thin-film nanophase. The average pore size and nanostructurethroughout the thin film can be precisely tuned through theselection of polymer templates, molar ratio of silicate topolymer, deposition rate and calcination condition.

Figure 1. (a) Schematic of porous silica preparation. (b) Scanningelectron micrograph (SEM) of porous silica (i) top view, (ii) sideview. Arrows indicate the thickness of the porous silica on silicon,223 nm.

Tetraethyl orthosilicate (TEOS) (14 ml) was dissolvedin a mixture of 15 ml of ethanol, 6.5 ml of distilled water,and 0.3 ml of 6 M HCl and stirred for 2 h at 75 ◦C toform a clear silicate sol. Separately, 1.8 g of F127, a di-functional block copolymer surfactant, was dissolved in 60 mlof ethanol by stirring at room temperature. Triblock copolymerPEO106–PPO70–PEO106 (Pluronic F127) was selected as thesynthetic template in this study. F127 with its high molecularweight, versus other copolymers studied herein, possesses ahigher degree of structural periodicity and more uniform poresize distribution in the nanoporous thin-film product, whichfacilitates the enrichment of targeted protein harvesting.

The coating solution was prepared by mixing 7.5 ml ofthe silicate sol into the triblock copolymer solution followedby stirring of the resulting sol for 2 h at room temperature. ThepH of the mixture solution remained around 1.5. The coatingsol was deposited on a silicon (1 0 0) wafer by spin coatingat a spin rate of 1500 rpm for 20 s. To increase the degreeof polymerization of the silica framework in the films and tofurther improve their thermal stability, the as-deposited filmswere heated at 80 ◦C for 12 h. The films were calcinated at425 ◦C to remove the organic surfactant. The temperature wasraised at a heating rate of 1 ◦C min−1, and the furnace wasthen maintained at 425 ◦C for 5 h. The films produced weretransparent and without cracks. Oxygen plasma ashing wasperformed in a Plasma Asher (March Plasma System) to pre-treat the chip surface. The treatment was carried out with anO2 flow rate at 80 sccm and a power of 300 W for 10 min.The thickness of the thin film could be controlled by adjustingthe concentration of polymer in the precursor solution, whilethe porosity mainly depends on the molar ratio of polymer andsilicate in the starting material.

2.1.2. Characterization techniques. We utilized severalcharacterization techniques to study the spin-coated meso-porous silica thin films. Through the use of a variableangle spectroscopic ellipsometer (J A Woollam Co. M-2000DI) and WVASE32 modeling software, the thickness ofthe thin films and their porosities were measured in the Cauchymodel and the effective medium approximation (EMA) model,respectively. Ellipsometric optical quantities, the phase (�)and amplitude (ψ), were obtained by acquiring spectra for 65◦,

2

Nanotechnology 21 (2010) 415302 E Blinka et al

Figure 2. The XRD pattern of porous silica thin film prepared byF127.

70◦, and 75◦ incidence angles using wavelengths from 300to 1600 nm. In the Cauchy model, the top layer’s thicknessand refractive index were determined by fitting experimentaldata with the model and minimizing the mean square error.Using the EMA model, the films’ porosities were calculated byassuming a certain volume of void in the pure silica and settingthe top layer’s thickness, obtained by the Cauchy model, as theconstant. X-ray diffraction (XRD) patterns were obtained on aPhilips X’Pert-MPD system with Cu Kα (45 kV, 40 mA). θ–2θscans were recorded from all spin-coated films in 1 s/0.001◦steps over the angle range from 0.2◦ to 6◦. Transmissionelectron microscopy (TEM; FEI Technai; FEI Co) was usedto obtain micrographs of the plane view of the porous silicathin films at a high tension of 200 kV. Scanning electronmicroscopy (SEM; Ziess SEM Neon 40) was used to acquiretop and side views of the films; figure 1(b) shows one particularfilm which is 223 nm thick with a 53% porosity and 4.0 nmpore diameter.

N2 adsorption/desorption analysis was used to measuresurface area and pore size distribution. Quantachrome wasused to record the N2 adsorption/desorption isotherm at 77 Kover the full range of relative P/P0 pressures. Brunauer–Emmett–Teller (BET) surface areas were determined overa relative pressure range of 0.05–0.4. Nanopore sizedistributions were calculated from the desorption branch of theisotherms using the Barrett–Joyner–Halenda (BJH) method.XPS spectra were recorded using an x-ray photoelectronspectrometer (Kratos Axis Ultra), utilizing a monochromatedAl Kα x-ray source (hν = 1486.5 eV), hybrid optics(employing a magnetic and electrostatic lens simultaneously)and a multi-channel plate and delay line detector coupledto a hemispherical analyzer. The take-off angle of thephotoelectrons was 90◦. All spectra were recorded usinga square aperture slot of 300 μm × 700 μm, and high-resolution spectra were collected with a pass energy of 20 eV.The pressure in the analysis chamber was typically 2 ×10−9 Torr during data acquisition. Kratos XPS analysissoftware was used to determine the stoichiometry of samplesfrom the corrected peak areas and employing Kratos sensitivityfactors for each element of interest. The XRD pattern shown

Figure 3. The TEM image of porous silica thin films prepared usingPluronic F127.

in figure 2 illustrates a 3D honeycomb like nanostructurehexagonally arranged on the substrate, as confirmed by XRD,with peaks at (200) and (400), further verified through TEMimaging in figure 3. N2 adsorption/desorption curves weregenerated using a Quantachrome Autosorb-3b BET SurfaceAnalyzer (inset of figure 4) and the pore size distribution wascalculated using the Barrett–Joyner–Halenda (BJH) method(figure 4) [32]. The adsorption/desorption isotherms describe atype IV isotherm with a H2 hysteresis loop (sloping adsorptionbranch and nearly vertical desorption branch), indicating ananoporous silica structure with interconnecting channels.Inflection points appearing at 0.40 < P/P0 < 0.75 in figure 4indicated the formation of ink-bottle shaped nanopores.

2.2. Preparation of chemically modified substrates

The chemical surface modification procedure can be seen infigure 5. Control substrates and those to be functionalizedwith APTES underwent piranha cleaning (H2O2/2H2SO4 v/v)for 5 min, followed by a triple alcohol rinse, before beingdried in a 110 ◦C oven for 10 min. Excluding the control,the samples were then placed in a 10% (v/v) APTES/ethanolsolution for 45 min and subsequently rinsed thoroughly withethanol and removed to a 110 ◦C oven for 1 h. Substrates tobe functionalized with glutaraldehyde were first cleaned andfunctionalized with APTES as detailed above. Immediatelyafter being removed from the oven, these substrates wereplaced in a 2.5% (v/v) glutaraldehyde/phosphate bufferedsaline (PBS) solution for 2 h. The substrates were thenrinsed thoroughly with PBS and kept in PBS until just prior tostamping, when they were rinsed thoroughly with de-ionizedwater and dried under a N2 stream.

The contact angle of functionalized substrates wasmeasured with a goniometer to confirm functionalization.Values shown in table 1 are consistent with the litera-ture [22, 27, 28, 33, 34]. It was observed that porous silicaand untreated glass slides had smaller contact angles thanchemically treated substrates.

3

Nanotechnology 21 (2010) 415302 E Blinka et al

Figure 4. N2 adsorption/desorption analysis (A) pore size distribution and (B) isotherms of the porous silica thin films prepared usingPluronic F127.

Figure 5. Procedure for chemical surface modification of glass andsilicon substrates.

2.3. Microcontact printing procedure

PDMS stamps were fabricated from hard silicon masters [35]to have 20μm diameter by 20μm high circular features. Thesestamps were ‘inked’ for 20 min with a 5 μg ml−1 solutionof FITC-labeled R-IgG (Sigma Aldrich), rinsed sequentiallywith (0.01 M) phosphate buffered saline and de-ionized water,dried under N2 gas, and finally brought into conformal contactwith the target substrate for 60 s. A home-built stamping

Table 1. Contact angles of experimental substrates measured within12 h of functionalization with goniometer (first ten angstroms).

SubstrateMeasured contactangle (deg)

Untreated glass 13 ± 3APTES-functionalized glass(amine [–NH2] groups)

69 ± 7

Glutaraldehyde-functionalized glass(aldehyde [–CHO] groups)

59 ± 4

Porous silicaa 13 ± 5

a Porous silica layer for this measurement has 96 nmthickness, 57% porosity, and 4.0 nm pore diameter.

apparatus was used to ensure proper alignment of the polymerstamp and the substrate and to allow for the applicationof consistent stamping pressures, generally in the range of40 mg/0.5 cm2. Transferred protein layers were analyzed byfluorescence (Olympus BX51) and atomic force microscopy(AFM; Digital Instruments Series IV, Veeco). Protein layersdeposited on porous silica coated silicon were also analyzedusing SEM (Ziess SEM Neon 40), as seen in figure 6. Afluorescence image of an array of the printed protein can beseen in figure 7.

Rabbit IgG protein as imaged by AFM (figure 8), andSEM (figure 9) displays the microstructure of the patternedprotein. The dimensions of IgG as reported by Lee et alare 8.5 nm × 14.4 nm × 4 nm [36]. The height dimensionsobserved for the printed protein molecules are consistentwith these accepted dimensions for the R-IgG, supporting theconclusion that the protein has retained its basic shape afterstamping [36, 37].

3. Results and discussion

3.1. Comparison of protein deposition on two substrates

3.1.1. Porous silica substrate. First, we tested severalporous silica conditions to determine the most suitable set ofparameters for μCP. Samples shown in table 2 were used.It was found that porous silica exhibiting 4 nm pores at 57%porosity and with a layer thickness in the range of 30–100 nm

4

Nanotechnology 21 (2010) 415302 E Blinka et al

Figure 6. (A) SEM and (B) AFM images of patterned proteins on porous silica on silicon.

Figure 7. Array of FITC-labeled R-IgG protein microcontact printed(diameter 20 μm) onto porous silica deposited on glass.

Table 2. Thickness and roughness of multiple porous silica layersmeasured with AFM (Digital Instruments Series IV). Layerparameters in bolded cells were found to result in the most successfulprotein transfer. Eight 1 μm2 areas from the area where the regionsbordered each other were taken and the height data was averaged.

Porosity(%)

Thickness(nm)

Poresize(nm)

Contactangle(deg)

Protein layerthickness(nm)

Protein layerroughness(nm)

63 547 6.3 11.1 ± 0.2 3 ± 2 3 ± 139 611 4.0 13.3 ± 0.7 8 ± 5 3 ± 257 673 4.0 12.8 ± 0.8 8 ± 4 5 ± 232 561 2.0 23 ± 2 3 ± 2 1.5 ± 0.457 30 4.0 30 ± 3 5 ± 1 1.1 ± 0.657 96 4.0 13 ± 6 3 ± 1 3.4 ± 0.453 223 4.0 15 ± 2 5 ± 3 5.0 ± 0.5

provides for the most effective protein transfer in terms ofconsistency of the full transfer of the stamping pattern, stampedprotein layer thickness and roughness. Protein layers stampedon the porous silica layers with parameters as listed in table 2were analyzed with AFM microscopy. As porosity and poresize decrease, the porous silica layer approaches non-poroussilica and the advantages of the pores are lost.

Figure 8. (A) Atomic force microscopy (AFM) of rabbitimmunoglobulin G (R-IgG) protein on porous silica, indicating aheight of (left peak) 16 nm (right peak) 8 nm. Accepted dimensionsof R-IgG are 14.5 nm × 8.5 nm × 4.8 nm (B) plot of height data forinset image.

For proteins deposited onto porous silica substrates, theresulting height data was divided into regions of proteincoverage and regions of background. Eight 1 μm2 areas fromthe area where the regions bordered each other were takenand the height data was averaged over these. The backgroundheight was subtracted from the protein layer height to give thedata in table 2. A similar procedure was repeated for roughnessmeasurements. For example, to calculate the protein thicknessfor the sample shown in figure 8, the 1 μm2 areas had anaverage height of 4.7 nm, from which the background heightof −3.2 nm was subtracted to give a protein layer height of7.9 nm. From analysis under a bright field microscope, it wasconcluded that porous silica layers with large pore sizes trapparticles of the PDMS stamp, contaminating the transferredprotein. Also, as the porous silica layer thickness increases,

5

Nanotechnology 21 (2010) 415302 E Blinka et al

Figure 9. SEM of R-IgG proteins on porous silica on silicon aftermicrocontact printing.

the surface of the layer develops wide undulations that disruptconformal contact of the stamp and decrease protein transfer.For these reasons, porous silica layers with pore sizes at orabove 6 nm and with layer thicknesses greater than about200 nm were found to be unfavorable to protein transfer.

Porous silica substrates have an advantage over chemicallyfunctionalized substrates in that they can be uniformlyprocessed. In order to print proteins onto porous silicasubstrates, the hole dimensions of the porous layer shouldbe similar to or smaller than the smallest dimension of theprotein (i.e. in this study we used 4 nm holes, and the smallestdimension of the IgG protein was 4 nm. Larger pore sizesindicated the diffusion of proteins into the pores with a layerthickness smaller than that of the protein size. These results aresimilar to those reported in solution based experiments [39]).The porosity and thickness of the porous silica can be heldconstant by controlling the processing conditions.

3.1.2. Chemically modified substrates. Here we compareμCP on porous silica and chemically treated surfaces. Theporous silica layer used has parameters as found in the previoussection 3.1.1: 4 nm pores at 57% porosity, with a layerthickness in the range of 30–100 nm. Results are shown infigure 10.

It was observed that completeness of transfer is the highestfor porous silica, figure 10 (D-2), and APTES-functionalizedsurfaces, figure 10 (B-2), partial for the untreated surfaces,figure 10 (A-3), and low for the glutaraldehyde-functionalizedsurfaces, figure 10 (C-3). Relative to the chemically modifiedand untreated substrates, the porous silica coated substratesprovided for thicker and more uniform protein layers. Thelayers on glutaraldehyde-functionalized surfaces are thin,figure 10 (C-2), as are those on APTES, figure 10 (B-2),averaging 3 and 2 nm thick, respectively, compared to the 5 nmthickness of the layers on porous silica, as seen in table 3. Theaverage 5 nm protein layer on porous silica is consistent withprotein monolayers.

Microcontact printing was used to transfer proteins ontoporous silica and chemically functionalized substrates that

Table 3. Thickness and roughness of R-IgG protein layers depositedon various substrates. (Measurements taken with Digital InstrumentsSeries IV AFM.)

Substrate

Proteinlayerthickness(nm)

Proteinlayerroughness(nm)

Numberof circles

Numberpoints percircle

Porous silica 5 ± 2 0.6 ± 0.3 4 8Untreatedglass

4 ± 1 1.0 ± 0.9 4 8

APTES-treatedglass

3 ± 2 1.0 ± 1.0 4 8

Glutaraldehyde-treated

2 ± 1 0.5 ± 0.7 4 8

were prepared three months in advance. The results for poroussilica substrates remained consistent with previous results,indicating complete transfer of proteins. The transfer ofproteins onto chemically modified substrates was similar to thecontrol slides with non-uniform deposition of proteins onto thesubstrate.

Protein layers observed on the chemically treatedsubstrates were thinner, 2–3 nm thick, than the smallestdimension of the R-IgG protein, 4 nm. This indicates thatthe R-IgG protein, which is known to bind with surfacesthrough hydrophobic interactions [38], unfolds to maximizefavorable interactions with the APTES (71◦ contact angle)and glutaraldehyde (56◦ contact angle) coated surfaces, asshown in table 1 [37]. The roughness of the proteinfilm deposited on untreated and APTES-modified substratesis 1 nm on average, while the layer on glutaraldehyde-modified substrates averages half a nanometer, though itsmaximum value is 1.2 nm. The layer on porous silica issmoother on average at 0.6 nm, with a range of 0.3–0.9 nm.Protein layers on untreated glass and APTES-treated glassboth exhibited incomplete coverage, while protein circles onglutaraldehyde-treated glass and porous silica were both filled.Arrays of 20 μm diameter circles printed on porous silicasurfaces were observed to be fully printed while the completepattern was rarely transferred to the other chemically treatedsubstrates.

Chemical modifications of porous silica substrates havebeen demonstrated previously for solution based experiments,indicating better adsorption of protein [29]. Chemicalmodification of porous silica may increase the adsorption ofproteins due to the presence of both chemical and physicalforces attracting the proteins. Compared to the chemicallymodified and untreated substrates, the porous silica coatedsubstrates enhanced protein deposition, resulting in thicker andmore uniform protein layers.

3.2. Integrated patterned immunoassay system

To demonstrate that protein function is retained afterimmobilization, porous silica was used as the basis for apatterned immunoassay. FITC-labeled R-IgG proteins werestamped onto a porous silica surface with 57% porosity, 4 nmpore size, and 30 nm layer thickness, as described above

6

Nanotechnology 21 (2010) 415302 E Blinka et al

D-1

D-2

D-3

D-4

C-1

C-2

C-3

C-4

B-1

B-2

B-3

B-4

A-1

A-2

A-3

A-4

60 nm

Control APTES Gluteraldehyde Porous Silica

Figure 10. Fluorescence images of 15 μm diameter circles of FITC-labeled R-IgG protein deposited with microcontact printing onto(A) untreated glass substrate, (B) APTES, (C) glutaraldehyde and (D) porous silica treated glass slides. AFM images of correspondingfluorescence images were performed as well as other portions of patterned substrate.

Figure 11. Steps for creation of patterned immunoassay. Labels (A)and (B) correspond to the fluorescence images in figure 12.

(table 2). As exhibited by figure 11, the surface was thenblocked with bovine serum albumin (BSA) and incubatedwith TRITC-labeled anti-R-IgG protein suspended in PBS

for 20 min. Fluorescence of the TRITC-labeled secondaryprotein is observed in the areas where the primary protein wasstamped, as seen in figure 12.

This method will be useful for patterned integrated micro-devices; for example, it could be used to pattern multiplebiomolecules on a single substrate for the creation of parallelassays utilizing different biomolecules, or layered with lightemitting diodes to create self-illuminating sensors for simpleranalysis of analyte binding. Porous silica has long-termstability and is easy to work with; it is thus compatible withsubsequent manufacturing processes, such as the addition of amicrochannel.

4. Conclusions

We describe nanoporous silica surface functionalization as analternative method to the chemical treatment for enhancedprotein adsorption suitable for molecular immunoassay. Amicrocontact printing technique was developed for proteindeposition and patterning on porous silica, as well as onsurfaces with chemical modification, including those silanizedwith APTES and glutaraldehyde. Compared to the chemicallymodified and untreated substrates, the porous silica coatedsubstrates enhanced protein deposition, resulting in thicker andmore uniform protein layers. An average 5 nm thick proteinmonolayer was consistently deposited on porous silica. Proteinlayers observed on the chemically treated substrates were

7

Nanotechnology 21 (2010) 415302 E Blinka et al

A-1

B-1

A-2

B-2

Figure 12. (A) Stamped FITC-labeled R-IgG on porous silica viewed under FITC (1) and TRITC (2) filters of fluorescence microscope.(B) Substrate as viewed under FITC (1) and TRITC (2) filters of fluorescence microscope after 20 min blocking with BSA followed by 20 minincubation with 10 μg ml−1 solution of TRITC-labeled anti-R-IgG.

thinner than the smallest dimension of the stamped protein,indicating possible denaturation. A patterned two-antibodyimmunoassay was developed to demonstrate the significantpotential of porous silica in the area of printable and robustbiological sensors.

Acknowledgments

We gratefully acknowledge the support of the National ScienceFoundation (CAREER 0846313), the Department of Defense(DODW81XWH-09-1-0212), NASA (NNJ06HE06A), theNational Institute of Health (RO1CA128797), the UniversityCooperative Society (URF 2009A31247), and the Stateof Texas Emerging Technology Fund, as well as theMicroelectronics Research Center located on the University ofTexas’ J J Pickle research campus.

References

[1] Singhvi R, Kumar A, Lopez G P, Stephanopoulos G N,Wang D I, Whitesides G M and Ingber D E 1994 Science264 696

[2] Branch D W, Corey J M, Weyhenmeyer J A, Brewer G J andWheeler B C 1998 Med. Biol. Eng. Comput. 36 135–41

[3] Pisignano D, Mazzeo M, Visconti P, Rinaldi R, Gigli G andCingolani R 2003 Synth. Met. 137 1483

[4] Xia Y N and Whitesides G M 1998 Angew. Chem. Int. Edn37 551–75

[5] Kumar A and Whitesides G M 1993 Appl. Phys. Lett. 63 2002[6] Bernard A, Renault J P, Michel B, Bosshard H R and

Delamarche E 2000 Adv. Mater. 12 1067–70[7] Delamarche E 2007 Chim. Int. J. Chem. 61 126–32[8] Ramrus D A and Berg J C 2006 Colloids Surf. A 273 84–9

[9] Gad M, Sugiyama S and Ohtani T 2003 J. Biomol. Struct. Dyn.21 387–93

[10] Godula K, Rabuka D, Nam K T and Bertozzi C R 2009 Angew.Chem. Int. Edn 48 4973–6

[11] Jung A, Wolters B and Berlin P 2007 Thin Solid Films515 6867–77

[12] Wigenius J A, Fransson S, von Post F and Inganas O 2008Biointerphases 3 75–82

[13] Pattani V P, Li C F, Desai T A and Vu T Q 2008 Biomed.Microdevices 10 367–74

[14] Chen C S, Mrksich M, Huang S, Whitesides G M andIngber D E 1997 Science 276 1425

[15] Thery M, Racine V, Pepin A, Piel M, Chen Y, Sibarita J B andBornens M 2005 Nat. Cell Biol. 7 947–53

[16] Inerowicz H D, Howell S, Regnier F E andReifenberger R 2002 Langmuir 18 5263–8

[17] Ghosh M, Alves C, Tong Z, Tettey K, Konstantopoulos K andStebe K J 2008 Langmuir 24 8134–42

[18] Park J P, Lee S J, Park T J, Lee K B, Choi I S, Lee S Y,Kim M G and Chung B H 2004 Biotechnol. Bioprocess Eng.9 137–42

[19] Folch A and Toner M 1998 Biotechnol. Prog. 14 388–92[20] MacBeath G and Schreiber S L 2000 Science 289 1760[21] Blawas A S and Reichert W M 1998 Biomaterials 19 595–609[22] Qin M, Hou S, Wang L K, Feng X Z, Wang R, Yang Y B,

Wang C, Yu L, Shao B and Qiao M Q 2007 Colloids Surf. B60 243–9

[23] Gan S H, Yang P and Yang W T 2009 Biomacromolecules10 1238–43

[24] Aebersold R H, Teplow D B, Hood L E and Kent S B 1986J. Biol. Chem. 261 4229–38

[25] Graf N, Yegen E, Lippitz A, Treu D, Wirth T and Unger W E S2008 Surf. Interface Anal. 40 180–3

[26] Liu T, Wang S and Chen G 2009 Talanta 77 1767–73[27] Jang L S and Liu H J 2009 Biomed. Microdevices 11 331–8[28] Libertino S, Giannazzo F, Aiello V, Scandurra A, Sinatra F,

Renis M and Fichera M 2008 Langmuir 24 1965–72

8

Nanotechnology 21 (2010) 415302 E Blinka et al

[29] Hu Y, Bouamrani A, Tasciotti E, Li L, Liu X andFerrari M 2009 ACS Nano 4 439–51

[30] Desai T A, Hansford D J, Leoni L, Essenpreis M andFerrari M 2000 Biosens. Bioelectron. 15 453–62

[31] Desai T A, Hansford D J, Kulinsky L, Nashat A H, Rasi G,Tu J, Wang Y, Zhang M and Ferrari M 1999 Biomed.Microdevices 2 11–40

[32] Barrett E P, Joyner L G and Halenda P P 1951 J. Am. Chem.Soc. 73 373–80

[33] Han Y, Mayer D, Offenhausser A and Ingebrandt S 2006 ThinSolid Films 510 175–80

[34] Dong H, Li C M, Zhou Q, Sun J B and Miao J M 2006 Biosens.Bioelectron. 22 621–6

[35] Schmid H and Michel B 2000 Macromolecules 33 3042–9[36] Lee K B, Park S J, Mirkin C A, Smith J C and Mrksich M 2002

Science 295 1702[37] Jeyachandran Y L, Mielczarski E, Rai B and Mielczarski J A

2009 Langmuir 25 11614–20[38] Sun X H, Yu D Q and Ghosh R 2009 J. Membr. Sci.

344 165–71[39] Bouamrani A, Hu Y, Tasciotti E, Li L, Chiappini C, Liu X and

Ferrari M 2010 Proteomics 10 496–505

9