Nanoporous Anodic Alumina Surface Modification ... - ACS Publications

Subscriber access provided by Kaohsiung Medical University

Surfaces, Interfaces, and Applications

Nanoporous Anodic Alumina Surface Modification by Electrostatic, Covalent and Immune Complexation Binding Investigated by Capillary Filling Chris Eckstein, Laura K. Acosta, Laura Pol, Elisabet XifréPérez, Josep Pallarès, Josep Ferré-Borrull, and Lluis F. Marsal ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b00572 • Publication Date (Web): 06 Mar 2018 Downloaded from http://pubs.acs.org on March 7, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Nanoporous Anodic Alumina Surface Modification by Electrostatic, Covalent and Immune Complexation Binding Investigated by Capillary Filling Chris Eckstein, Laura K. Acosta, Laura Pol, Elisabet Xifré-Pérez, Josep Ferré-Borrull, Lluis F. Marsal* Universitat Rovira i Virgili, Departament d’Enginyeria Electrònica, Elèctrica i Automàtica, Nano-electronic and Photonic Systems (NePhoS) group, Avda. Països Catalans 26, 43007 Tarragona, Spain. * Corresponding Author: [email protected] Abstract

The fluid imbibition-coupled laser interferometry (FICLI) technique has been applied to detect and quantify surface changes and pore dimension variations in nanoporous anodic alumina (NAA) structures. FICLI is a non-invasive optical technique that permits the determination of the NAA average pore radius with high accuracy. In this work, the technique is applied after each step of different surface modification paths of the NAA pores: i) electrostatic immobilization of bovine serum albumin, ii) covalent attachment of streptavidin via (3-aminipropyl)-triethoxysilane and glutaraldehyde grafting and iii) immune complexation. Results show that BSA attachment can be detected as a reduction in estimated radius from FICLI with high accuracy and reproducibility. In the case of the covalent attachment of streptavidin, FICLI is able to recognize a multi-layer formation of the silane and of the protein. For immune complexation, the technique is able to detect different antibody-antigen bindings and to distinguish different dynamics among different immune species. Keywords

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nanoporous anodic alumina, optofluidics, surface functionalization, bovine serum albumin, streptavidin, immune complexation.

Introduction

The use of nanoporous structures in optical sensing applications has gained much attention in recent years1,2. They provide the benefits of using very low reaction volumes and of employing analytes that do not require prior labelling.3,4. Typically, optofluidic5,6 sensing systems are based on a solid carrier matrix, a liquid carrying the analyte and on the physicochemical characteristics of the interface between them. Nano-optofluidic methods have been proven sensitive for substances as small as nanoparticles, viruses7 and even ions4. Fluid imbibition-coupled laser interferometry (FICLI) 8–11 is a recently introduced optofluidic technique that permits the investigation of the surface and geometrical properties of nanoporous structures. FICLI monitors interferometric light intensity fluctuations of a single wavelength beam reflected in a nanoporous thin film membrane in real-time upon capillarydriven fluid imbibition 8,9,12. In the case of straight nanochannels being filled by a liquid, this technique allows for a time-resolved estimation of the position of the meniscus 13. The filling kinetics depend of the surface properties and the pore internal geometries, and thus it can be used as a method to characterize such magnitudes. In FICLI, time-resolved oscillation patterns corresponding to the light intensity fluctuations are measured by a photodetector and a high-speed data acquisition system. Then, these time-resolved interferograms are analyzed on the basis of mathematical models of the filling kinetics. This analysis leads to an accurate estimate of the top and bottom pore radii 8 or even of the complete pore profile 11. FICLI has been developed on the basis of Nanoporous Anodic Alumina (NAA) 14,15. This is a material widely applied in nanotechnology due to its stability, tissue compatibility and high and tunable surface-to-volume ratio. The electrochemical fabrication of NAA is simple and reproducible and results in a structure of nanochannel arrays in a well-ordered, hexagonal honeycomb-like arrangement16. NAA possesses a narrow pore size distribution that can be fine-tuned via the precise control of the anodization conditions or by further modifications after anodization, such as increasing the pore radius via wet-chemical etching. 17–22

. This material provides a well-defined, controllable and homogenous interaction surface

18,22–24

. The ability to immobilize different kinds of biological species to the NAA pore ACS Paragon Plus Environment

Page 2 of 31

Page 3 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


surface have been studied and quantified.

25,26

. Furthermore, the application of NAA as a

powerful optofluidic biosensing platform has been successfully demonstrated via the detection of proteins 17, antibody-antigen complexation3,27,28, enzyme activity29,30, and DNA 31

.

In a recent work, we established for FICLI an accuracy better than 2 nm in determining the changes in pore radius of NAA after different modifications involving pore radius increase (by chemical wet etching) and pore radius decrease (by the deposition of layer-by-layer coatings) 10,32. Such remarkable accuracy permits to consider NAA together with FICLI as a platform to develop biosensing systems. However, such application depends critically on the functionalization of the pore surface. In order to use NAA as a sensing platform, it is necessary to modify the pore surface in order to enable the selective immobilization of the biomolecule to be detected.

Several works have studied the methods for the functionalization of NAA surfaces and their effects on biosensing systems26 considering as diverse paths as organosilane attachment, organic acids functionalization, layer-by-layer deposition or different polymer modifications. More specifically, the attachment of different proteins to the alumina surface has been studied by Baranowska et al.25. In this study different attachment paths (electrostatic or covalent through silanization) and different proteins such as bovine serum albumin (BSA) or collagen are considered. On the other hand, Tan et al.

33

demonstrate that protein−surface

interactions at a molecular level can be regulated by changing the binding domains and their local environment at nanometer scale.

In this work we aim at investigating how different paths for the attachment of biomolecules to the NAA pore walls influence the measurements taken by FICLI. Three immobilization strategies are considered: i) electrostatic binding of BSA, ii) covalent binding of streptavidin and iii) specific immune complexation of an antibody-antigen conjugated pair. FICLI will be applied to study the changes in surface properties and pore geometry upon attachment of different biological species. The electrostatic attachment of BSA to different surfaces

34

as

well as into NAA pores with diameters up to 22 nm has already been studied10. In this work, the pore diameter range is expanded up to 45 nm. Covalent binding of streptavidin is promoted through the formation of a self-assembled monolayer of (3-aminopropyl)triethoxysilane (APTES) terminated with a glutaraldehyde (GTA) molecule that provides an ACS Paragon Plus Environment


aldehyde group to form imine bonds with biomolecules for the streptavidin grafting. The study of the effect of biocomplexation on the NAA pore surface is done by successive incubation of the IgG antibody and the corresponding conjugated antigen.

MATERIALS AND METHODS Materials High purity aluminum foils (99.999 %) with a thickness of 0.5 mm were purchased from Goodfellow Cambridge Ltd. Oxalic acid (H2C2O4), sulfuric acid (H2SO4), phosphoric acid (H3PO4), perchloric acid (HClO4), chromic acid (H2CrO7), copper chloride (CuCl2) ethanol (C2H5OH), acetone ((CH3)2CO), protein A from staphylococcus aureus (PA), bovine serum albumin (BSA), streptavidin A, human IgG from human serum (H-IgG), anti-human IgG (AH-IgG) produced in goat, goat IgG from goat serum (G-IgG) and anti-goat IgG (AG-IgG) produced in rabbit were purchased from Sigma Aldrich. Double de-ionized (DI) water (18.6 MU, Purelab Option-Q) was used for all solutions.

NAA Fabrication. NAA fabrication was based on a well-known two-step anodization protocol24. Aluminum foils were degreased with acetone ((CH3)2CO). An initial electropolishing step in a mixture of ethanol (EtOH) and perchloric acid (HClO4) 4:1 (v/v) at 20 V for 4 min provided a planar surface for the following anodization steps. A first anodization step was carried out in 0.3 M oxalic acid at 55 V and 5 °C. Even though for NAA prepared with oxalic acid electrolyte the best pore arrangement is obtained at an anodization potential of 40 V, the samples were prepared with an anodization potential of 55 V in order to obtain the largest possible pore radius while still maintaining a good pore arrangement. This first anodization lasted for up to 20 h to allow for a propagation from initially unordered pores at the top to self-ordered arrays of nanochannels at the bottom35. This first layer was dissolved by a mixture of phosphoric acid (H3PO4, 6% wt.) and chromic acid (H2CrO7, 1.8% wt.) at 70 °C for 3 h, leaving ordered nano-indentations at the bottom for an immediate onset of ordered pore growth during the following second anodization step. Using anodization conditions identical to the first step, with a total anodization time of 13 h, perpendicularly growing ordered nanochannels with a thickness/total length of 75 µm were fabricated. A mixture of CuCl2-HCl (saturated) and 68 % HNO3 was used to remove remaining aluminum from the bottom.


Page 4 of 31

Page 5 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Removing the barrier layer and opening the pores on the bottom side was accomplished by 4 cycles of reactive-ion etching (RIE) (Oxford Instruments Plasma Technology PlasmaPro NGP80) with a mixture of 5ccms argon (Ar) and 35ccms fluoroform (CHF3) for 5min followed by 50ccms Ar for 1min. A brief wet chemical etching step in 5 % H3PO4 at 35 °C for 30 seconds allowed to remove sidewall passivation due to condensed polymer ((CF2)n). Hereafter, these samples are referred to as NAAox. A second set of NAA was fabricated using 0.3 M sulfuric acid (H2SO4) and ethanol 5:1 (v/v) at 25 V and -5 °C. These anodization conditions were chosen as they lead to the smallest possible pore radius while maintaining a high degree of ordered pore arrangement. The anodization procedure, total length and pore opening processes were otherwise the same. Hereafter, these samples are referred to as NAAsul. NAA Modification.

In order to compare the reproducibility of FICLI towards pore radius increases of NAAox and NAAsul the pores were initially widened by wet etching in 5 % H3PO4 at 35 °C for a given pore widening time (tPW). These samples are thereafter referred to ‘as-produced’ for further bio-immobilization experiments.

Electrostatic immobilization of bovine serum albumin (BSA) was accomplished by hydroxylating NAAsul in > 30% hydrogen peroxide (H2O2) at 70 °C for 30 min before incubating in 100µL of 100 µg/mL of BSA at 4°C overnight.

Covalent immobilization of streptavidin was accomplished via functionalizing NAAox as follows: the NAAox was first hydroxylated as described above. Handled in nitrogen environment, the NAAox was then placed in 20 mL anhydrous toluene and 0.2 mL of (3aminopropyl) triethoxysilane (APTES) was added drop wise and stirred for one hour. 10 min washing steps in anhydrous toluene in nitrogen environment and in toluene and deionized water at ambient environment followed. Incubation overnight at 120 °C facilitated silane crosslinking before immersing in 10% glutaraldehyde (GTA) for one hour. Finally, the functionalized NAAox was incubated in 100µL of 100 µg/mL streptavidin at 4 °C overnight.



Four different immune complexation experiments (labeled as IC1 to IC4) were carried out. Immobilization of immunoglobulins (IgG) via immune complexation in the IC experiments was accomplished by initially hydroxylating 4 NAAox as described above. In experiment IC1, the hydroxylation was followed by the electrostatic immobilization of 100µL of 100 µg/mL protein A (PA) at 4°C overnight. Subsequently 100µL of 1 mg/mL of human-IgG (H-IgG) from human serum were incubated at 4°C overnight to form immune complexes with PA via the human Fc portion of the H-IgG. Similarly, in experiment IC2, the hydroxylation was followed by electrostatic immobilization of 100µL of 100 µg/mL PA at 4°C overnight with a final incubation of 100µL of 1 mg/mL of anti-human-IgG (AH-IgG) produced in goat, to form immune complexes with PA via the goat Fc portion of the AH-IgG. In experiment IC3, the immune complexation between 100µg/mL PA and 1mg/mL AH-IgG was allowed to take place in a test tube at 4º C overnight, to provide an increased reaction volume for the complexation reaction. After this test tube pre-incubation period, the NAAox was incubated with 100µL of this PA/AH-IgG complex at 4°C overnight. Finally 100µL of 1mg/ml of HIgG was incubated at 4º C overnight, to form secondary immune complexes between the two IgG. Finally, in experiment IC4, hydroxylation of NAAox and the electrostatic immobilization of PA (100µL of 100µg/mL at 4° C overnight) were followed by the incubation of 100µL of 1mg/mL of anti-Goat IgG (AG-IgG, produced in rabbit) to form PA-IgG immune complexes via the rabbit Fc portion of the AG-IgG. A final incubation of 100µL of 1mg/mL of goat IgG from goat serum was applied to form secondary immune complexes between the two AG-IgG and G-IgG. All NAAox were rinsed with DI water and characterized with FICLI after each incubation step.

NAA Characterization. To estimate the pore radius and interpore distance of NAA scanning electron microscopy (SEM FEI Quanta 600) was used in conjunction with a standard image processing package (ImageJ, public domain program developed at the RSB of the NIH, USA). The immobilization of BSA and streptavidin on the inner pore walls was demonstrated with confocal microscopy (NIKON TE2000-E) by superimposing 12 images taken at advancing depths of 7.6 µm from top to bottom. A transverse view of the NAA was produced.

Interferometry and Data Handling. The set up and data collection via fluid imbibition coupled laser interferometry (FICLI) was previously reported with great detail8,10, therefore here in brief: A moving liquid interface within a NAA thin film causes distinct oscillating ACS Paragon Plus Environment

Page 6 of 31

Page 7 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


light interferences of a reflected laser beam. The elapsed time between two adjacent extrema of the resulting interferogram is plotted over the extremum index. If the pores are perfectly cylindrical, with uniform radius through the whole layer, such plot results in a linear trend. In a previous work8, the FICLI technique was indeed used to demonstrate that NAA pores are not perfectly cylindrical but truncated conical with a very small cone angle. Such a small angle results in a very small deviation from the linear trend. The slope (S) of the resulting linear correlation is used to extrapolate the pore radius via the expression:

=

, with =

and =

,

where θ is the liquid contact angle with the inner pore surface, γ is the liquid surface tension and µ is the liquid viscosity. The refractive indices of the liquid and air are n2 and n3, respectively, the laser wavelength is λ and the number of pores per unit of area is δ. Imbibition experiments were conducted with DI water on three different positions on the NAA; triplicates were performed on every position contributing to a total of up to 9 measurement replicates.

RESULTS AND DISCUSSION

NAA Characterization:

To verify the fabrication process of NAA, traditional characterization methods (ESEM) served to confirm and highlight the efficacy and working range of FICLI on NAA with nonmodified inner pore surfaces. Figure 1 shows, from left to right, top-view ESEM images of NAA fabricated in oxalic (top) (NAAox) and sulfuric (bottom) (NAAsul) acid electrolytes followed by wet chemical etching for the durations of tPW = 12, 18, 24 min. for NAAox and tPW = 0, 5, 10 min. for NAAsul. Their respective average pore size distributions were obtained by image analysis and are presented in the form of histograms with Gaussian fits in Figure 2, left. The average pore radius estimated by ESEM image analysis were plotted as a function



of pore widening time and compared with FICLI estimates in Figure 2, right. The pore radius increase rates (IR) estimated by ESEM image analysis differ by a factor of 3 for NAAox and NAAsul with IRox,ESEM = 1.0 ± 0.2 nm/min. and IRsul,ESEM = 0.3 ± 0.2 nm/min. On the other hand, the pore radius increase rate estimations obtained by FICLI were both found identical and accounted to IRox,FICLI = 0.8 ± 0.2 nm/min. and IRsul,FICLI = 0.8 ± 0.1 nm/min.

The ESEM images show highly ordered pores with a honeycomb-like arrangement and uniform interpore distances for both NAAox and NAAsul. The image analysis revealed average pore estimates with a wider size distribution of pores compared to FICLI estimations, especially for smaller pore radii. However, the average pore radii and pore radius increase rates obtained by both methods are rather coherent. Minor differences are observed in the radius estimation, which can be attributed to parameter adjustments regarding the liquidsurface interface (contact angle, surface tension, viscosity) during FICLI data treatment. In respect thereof, different contact angles were reported in the literature for pores with increasing radii, which was not accounted for in the data treatment of this study36. The large uncertainty stemming from the subjective ESEM image analyses due to arbitrarily selected grey cut off values further added to the presented minor etch rate discrepancies. Nonetheless, the etch rates obtained by ESEM and FICLI are in good agreement with previously published etch rates for NAA produced in oxalic23 and sulfuric acid10 electrolytes. In comparison with previous studies, this work provides an estimate of the pore widening rate for a wider range of pore radius and for samples produced with two different electrolytes. These results confirm that uncertainty of pore radius estimation with FICLI is less than 3 nm.

Electrostatic immobilization of bovine serum albumin


Page 8 of 31

Page 9 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In order to evaluate the influence of electrostatic immobilization of biomolecules to the NAA pores on the accuracy of FICLI, we studied the electrostatic immobilization of bovine serum albumin (BSA) in NAA produced in sulfuric (NAAsul) and oxalic (NAAox) acid for a range of pore radii. Figure 3a shows six pairs of FICLI radius estimates corresponding to NAA before (back bars) and after (front bars) the electrostatic immobilization of BSA. Bars with horizontal or vertical patterns correspond to NAAsul or NAAox, respectively. The x-axis corresponds to the pore radius FICLI estimate of the as-produced sample after pore widenings of tPW = 0, 5 and 10 min for NAAsul and tPW = 12, 18 and 24 min for NAAox, and before the BSA immobilization. In Figure 3b the estimated radius reductions after BSA immobilization are shown. This radius reduction estimate can be related to the BSA molecule diameter. The estimated reductions (∆R) for NAAsul are slightly larger than those for NAAox with ∆R,sul = 8.0 ± 0.8 nm and ∆R,ox, ox = 6.9 ± 1.9 nm, respectively.

The detection of BSA was very reproducible with a marginal standard deviation of 3.2 nm across a large range of as-produced radii (15 – 43 nm) and two types of NAA materials (i.e. NAAsul, NAAox). The radius reduction due the immobilization of BSA depends neither on the initial as-produced pore radius of NAAsul and NAAox nor were they found to depend on the electrolyte the NAA is fabricated with (i.e. sulfuric or oxalic acid). Past studies providing the dimensions of BSA (2.7 x 6.3 x 8.2 nm)37 show that the radius reduction due to the immobilization of BSA in NAAsul and NAAox in the present study both fall into a slightly overestimated size range of the protein.

These minor overestimations can in part be explained by the wetting properties of the BSA molecule which retains an additional water layer in aqueous solution (Debye sphere) due to its electrostatic potential. The formation of more than one layer and disordered alignment on



the surface are also possible explanations for these BSA size over-estimations. However, more important is the selection of the liquid-surface interface parameters (i.e. contact angle) used during the data treatment for the radius estimation. It was previously shown that the contact angle depends on the electrolyte the NAA is fabricated with. It further depends on the changes upon wet chemical etching of NAA, probably due to a changing surface composition with fewer impurities towards the center38,39. These etching-related changes were not accounted for in this study and the accuracy of the absolute values of the contact angles may be in question, explaining the differences in radius reduction due to BSA immobilization in NAAsul and NAAox. The selection of precise parameters for the FICLI interferogram analysis plays therefore a crucial role in the accurate estimation of NAA radii. Nonetheless, both estimations of the size of BSA fall into realistic dimensions and present reasonable estimates.

Covalent immobilization of streptavidin

Since in some cases it is not possible to immobilize biomolecules to surfaces electrostatically, covalent attachment via silane coupling has become common practice. In this study FICLI served to detect and characterize the silane coupling agent APTES+GTA and the covalent immobilization of streptavidin in NAA produced in oxalic acid (NAAox) widened by tPW = 18 min. Figure 4a shows the consecutive radius estimations of NAAox after the following modification steps: i) pore widening for tPW = 18 min, ii) silanization with APTES-GTA and iii) covalent immobilization of streptavidin. On the other hand, Figure 4b shows the normalized slopes, S, obtained from the raw oscillation patterns used for the radius estimation for the same modification steps, in a logarithmic scale. The radius estimates account for RFICLI, ox, 18 min = 41.8 ± 1.0 nm, RFICLI, ox, APTES-GTA = 31.7 ± 1.0 nm and RFICLI, streptavidin = 5.6 ±


Page 10 of 31

Page 11 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.0 nm. The respective radius reductions amount to ΔRAPTES-GTA = -10.1 ± 1.0 nm and ΔRstreptavidin = -26.0 ± 0.4 nm which both represent an over-estimation of a monolayer of the silane coupling agent and the protein respectively. Concerning the slope S, it can be seen that it increases significantly (3 fold) upon the functionalization with APTES+GTA and then it increases sharply (3612 fold) after the covalent immobilization of streptavidin. The 10 nm radius reduction observed after APTES-GTA immobilization is not in agreement with the size of the molecules: considering an APTES monolayer as the sum of the individual bond lengths one obtains a theoretical thickness of less than 1nm (See Supporting Information, Figure S1). However, multilayer deposition of APTES is common and is strongly dependent on the reaction conditions.40,41 Vandenberg et al. addressed some of the reaction conditions and obtained layers with thickness from 1.8 and up to 33 nm with incubation periods of from 1 and up to 24 h and an APTES concentration of 0.4 %. Therefore, the observed slope and radius reduction are reasonably justified by the experimental conditions, concisely 2 h. incubation period with an APTES concentration of 1 % and an additional GTA layer. The sharp increase in slope S (up to four orders of magnitude) after the streptavidin immobilization is a consequence of the dependence of the slope with the inverse fifth power of the radius: as the pores become smaller, the slope S approaches infinity. This means that such magnitude is especially sensitive for small radius values. In the case of Figure 4, results demonstrate that after streptavidin immobilization, pores are nearly blocked. The homogenous deposition of streptavidin across the NAA thickness was confirmed by the linear dependence of the fluid imbibition dynamics, Figure S2, top, (Supporting Information) and by the confocal microscopy image shown in Figure S2, bottom. The radius reduction upon streptavidin immobilization is about 2.5-fold larger than the expected



maximal dimension of the protein (2.0 x 6.8 x 10 nm) 42. The formation of a multilayer of the protein within the NAA is the most probable cause of this over-estimation

Immune complexation of IgG and anti-IgG

Immune complexation reactions of protein A and immunoglobulins (IgG) were conducted at the NAA pore surfaces and its influence in the FICLI measurement was evaluated. Four different immune complexation experiments (IC1 to IC4) were followed as described above. Figure 5 reports the results of the four immune complexation experiments with the labels IC1 to IC4, schematically depicted in Figure 5a. Each experiment consists of consecutive protein immobilization events, as indicated by the categorical x-axis. Their respective FICLI radius estimations after each step are shown in Figure 5b, while the slopes S from each FICLI measurement are presented in Figure 5c. The slope values are normalized taking as unity the value corresponding to an as-produced sample. In IC1 the two incubation steps of protein A (PA) and human-IgG (H-IgG) result in two clear slope increases and respective radius reductions. Slope increase after the incubation with human-IgG is much more significant as compared with protein A. Instead, corresponding radius reductions are similar for both steps, with values ∆RIC1,PA = 6.4 ± 0.7 nm and ∆RIC1,HIgG

= 5.5 ± 2.3 nm. This is in agreement with the fact that human-IgG has a high affinity for

its complexation to PA. In IC2 the incubation of protein A with the NAAox resulted in a similar slope increase and radius reduction as in IC1, with ∆RIC2,PA = 6.1 ± 1.1 nm, confirming the reproducibility of this first step. However, the slope variation upon anti-human-IgG (AH-IgG) incubation is small, with a consequent small variation in the corresponding radius. This is not a surprising


Page 12 of 31

Page 13 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


result, as it is widely known that this AH-IgG, obtained from goat serum, has a low affinity to PA. For this reason, in experiment IC3 protein A and AH-IgG were pre-incubated in PBS in a reaction tube overnight prior to incubation on NAAox. After this incubation, a clear slope and radius reduction of this immune complex with ∆RIC3,PA+AH-IgG = 8.0 ± 1.5 nm is obtained. Such slope variation and radius reduction are similar to the corresponding values seen for the combined PA and H-IgG in IC1. The comparison between the results of the IC2 and IC3 experiment indicates that the immune complexation dynamics are critically impaired by the limited reaction space provided by the NAA. In IC3, the subsequent incubation of H-IgG resulted in a further slope S variation and a corresponding radius reduction of ∆RIC3,H-IgG = 10.0 ± 1.3 nm. Finally, in IC4 three distinct slope increases after the incubation with PA, goat-IgG (G-IgG) and anti-goat-IgG (AG-IgG) can be observed and the respective radius reductions amount to ∆RIC4,PA = 6.9 ± 1.6 nm, ∆RIC4,AG-IgG = 6.6 ± 2.9 nm and ∆RIC4,G-IgG = 4.8 ± 2.2 nm. It is common practice to electrostatically immobilize PA43,44and the consistency in radius reduction in the present study underlines the robustness of this immobilization strategy. The size of PA (60 kDa) is very similar to BSA (65 kDa) and thus the observed radius reductions for both proteins are congruent. The obtained results for the complexation of the immunoglobulins with the PA can be explained on the basis of the affinities of the different IgG`s Fc portion to PA: IgGs from human serum (H-IgG) possess a high binding affinity which facilitated the complexation with PA within the NAA pores, resulting in the observed radius reduction in IC1. In contrast, the affinity of IgGs produced in Goat (AH-IgG) is rather low. Therefore, in IC2, the provided incubation time for AH-IgG was not sufficient to facilitate enough complexation with PA inside the confined space of the NAA pores. This results in the small change in slope and estimated radius. Instead, providing the reaction time and space for PA and AH-IgG in PBS solution in a test tube prior to the immobilization in



NAA in IC3 compensated for the low affinity and enhanced the complexation as seen by the much larger radius reduction. The PA and IgG affinities and their respective radius reductions in experiments IC1 to IC3 are further co-validated by DLS measurements of protein and complex size in solution after 24h (Figure S3, Supporting Information). The AG-IgG were produced in rabbit and are known to have a high affinity for PA which was demonstrated by significant slope increases and the respective radius reductions of IC4. The further immune complexation of the G-IgG resulted in significantly different slopes. Even though the sizes of the IgGs are on the smaller end of the expected results, the significant signal changes underline the detection of the immune complexation. With these results we demonstrate that FICLI is well capable as an immunosensing system, although it is very dependent on the biomolecular affinities what means the experimental procedures have to be accurately designed.

CONCLUSIONS In this study it was demonstrated that FICLI is a highly sensitive method to detect and quantify changes in the pore surface properties and dimensions of NAA after the attachment of different biological species. The accuracy of FICLI for the characterization of pore size changes of NAA produced in oxalic and sulfuric acid has been shown. The radius increase due to wet chemical pore widening was detected irrespective of the electrolyte used to prepare the NAA and the results are in good agreement with traditional NAA characterization methods such as ESEM and the literature. The electrostatic immobilization of biomolecules to the pore walls is a common and fast NAA modification technique. In this study it was shown that the pore radius reduction of the NAA due to the immobilization of BSA is accurately estimated by FICLI. The radius reduction estimates fall into the expected size dimensions of the protein and that the radius


Page 14 of 31

Page 15 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reductions are reproducible irrespectively of the NAA material and initial pore radius. Only slight differences are observed for NAA obtained with different electrolytes, what is an indication that the surface contact angle for such structures depends slightly of the electrolyte. The pore radius reduction due the functionalization of NAA pore walls with APTES-GTA was characterized. The observed radius reduction value suggests the deposition of an APTES multilayer due to reaction conditions. The subsequent covalent immobilization of streptavidin via imine bonds was further detected by a radius reduction. The dimensions of the protein were slightly over-estimated and also suggest a multilayer deposition of the protein. The immobilization of IgGs via immune complexation with protein A and secondary IgGs was detected by the characterization of NAA pore radius reductions and changing capillary filling dynamics. It was shown that a chain of immobilization events starting with the electrostatic immobilization of protein A, followed by specific immune complexations of IgGs and secondary IgGs can be demonstrated by the size reduction of NAA pores. These findings provide great potential for the future development of target-specific biosensing systems based on the radius reduction of NAA and the changes in nanopore filling dynamics. Further research on real-time monitoring of filling dynamics variations can open the way to new sensing devices with a view to miniaturization.

ASSOCIATED CONTENT Supporting Information APTES-GTA radius reduction for six replicates and molecular scheme; fluorescence confocal microscopy and example of FICLI interferogram analysis for streptavidin-incubated NAA; size measurement by DLS of different PA-Immunoglobulin complexes.



AUTHOR INFORMATION Corresponding Author *Prof. Dr. L. F. Marsal; E-mail: [email protected]; telephone.: (0034) 977559625; Fax: (0034) 977 559605 Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENTS This work was supported in part by the Spanish Ministry of Economy and competitiveness TEC2015-71324-R (MINECO/FEDER), the Catalan authority AGAUR 2017SGR1527, and ICREA under the ICREA Academia Award.


Page 16 of 31

Page 17 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1)

Santos, A.; Kumeria, T.; Losic, D. Nanoporous Anodic Aluminum Oxide for Chemical Sensing and Biosensors. TrAC Trends Anal. Chem. 2013, 44, 25–38.

(2)

Kumeria, T.; Santos, A.; Losic, D. Nanoporous Anodic Alumina Platforms: Engineered Surface Chemistry and Structure for Optical Sensing Applications. Sensors (Basel). 2014, 14 (7), 11878–11918.

(3)

Alvarez, S. D.; Li, C.-P.; Chiang, C. E.; Schuller, I. K.; Sailor, M. J. A Label-Free Porous Alumina Interferometric Immunosensor. ACS Nano 2009, 3 (10), 3301–3307.

(4)

Kumeria, T.; Santos, A.; Losic, D. Ultrasensitive Nanoporous Interferometric Sensors for Label-Free Detection of Gold (III) Ions. ACS Appl. Mater. Interfaces 2013, 5 (Iii), 11783−11790.

(5)

Fan, X.; White, I. M. Optofluidic Microsystems for Chemical and Biological Analysis. Nat. Photonics 2011, 5 (10), 591–597.

(6)

Psaltis, D.; Quake, S. R.; Yang, C. Developing Optofluidic Technology through the Fusion of Microfluidics and Optics. Nature 2006, 442 (7101), 381–386.

(7)

Mitra, A.; Deutsch, B.; Ignatovich, F.; Dykes, C.; Novotny, L. Nano-Optofluidic Detection of Single Viruses and Nanoparticles. ACS Nano 2010, 23 (1), 1–7.

(8)

Urteaga, R.; Acquaroli, L. N.; Koropecki, R. R.; Santos, A.; Alba, M.; Pallarès, J.; Marsal, L. F.; Berli, C. L. A. Optofluidic Characterization of Nanoporous Membranes. Langmuir 2013, 29 (8), 2784–2789.

(9)

Acquaroli, L. N.; Urteaga, R.; Berli, C. L. A.; Koropecki, R. R. Capillary Filling in Nanostructured Porous Silicon. Langmuir 2011, 27 (5), 2067–2072.

(10)

Eckstein, C.; Xifré-Pérez, E.; Porta-I-Batalla, M.; Ferré-Borrull, J.; Marsal, L. F. Optical Monitoring of the Capillary Filling Dynamics Variation in Nanoporous Anodic Alumina toward Sensing Applications. Langmuir 2016, 32 (41), 10467–10472.

(11)

Elizalde, E.; Urteaga, R.; Koropecki, R. R.; Berli, C. L. A. Inverse Problem of Capillary Filling. Phys. Rev. Lett. 2014, 112 (13), 134502.

(12)

Han, A.; Mondin, G.; Hegelbach, N. G.; De Rooij, N. F.; Staufer, U. Filling Kinetics of Liquids in Nanochannels as Narrow as 27 Nm by Capillary Force. J. Colloid



Interface Sci. 2006, 293 (2006), 151–157. (13)

Washburn, E. W. The Dynamics of Capillary Flow. Phys. Rev. 1921, 17 (3), 273–283.

(14)

Lee, W. The Anodization of Aluminum for Nanotechnology Applications. JOM 2010, 62 (6), 57–63.

(15)

Ferré-Borrull, J.; Pallarès, J.; Macías, G.; Marsal, L. F. Nanostructural Engineering of Nanoporous Anodic Alumina for Biosensing Applications. Materials (Basel). 2014, 7 (7), 5225–5253.

(16)

Nielsch, K.; Choi, J.; Schwirn, K.; Wehrspohn, R. B.; Gösele, U. Self-Ordering Regimes of Porous Alumina: The 10% Porosity Rule. Nano Lett. 2002, 2 (7), 677–680.

(17)

Macias, G.; Ferré-Borrull, J.; Pallarès, J.; Marsal, L. F. Effect of Pore Diameter in Nanoporous Anodic Alumina Optical Biosensors. Analyst 2014, 140 (14), 4848-4854.

(18)

Santos, A.; Balderrama, V. S.; Alba, M.; Formentín, P.; Ferré-Borrull, J.; Pallarès, J.; Marsal, L. F. Nanoporous Anodic Alumina Barcodes: Toward Smart Optical Biosensors. Adv. Mater. 2012, 24 (8), 1050–1054.

(19)

Yamamoto, Y.; Baba, N.; Tajima, S. Coloured Materials and Photoluminescence Centers in Anodic Film on Aluminium. Nature 1981, 289, 572–574.

(20)

Han, H.; Park, S. J.; Jang, J. S.; Ryu, H.; Kim, K. J.; Baik, S.; Lee, W. In Situ Determination of the Pore Opening Point during Wet-Chemical Etching of the Barrier Layer of Porous Anodic Aluminum Oxide: Nonuniform Impurity Distribution in Anodic Oxide. ACS Appl. Mater. Interfaces 2013, 5 (8), 3441–3448.

(21)

Santos, A.; Kumeria, T.; Wang, Y.; Losic, D. Insitu Monitored Engineering of Inverted Nanoporous Anodic Alumina Funnels: On the Precise Generation of 3D Optical Nanostructures. Nanoscale 2014, 6 (1), 9991–9999.

(22)

Santos, A.; Vojkuvka, L.; Pallarés, J.; Ferré-Borrull, J.; Marsal, L. F. In Situ Electrochemical Dissolution of the Oxide Barrier Layer of Porous Anodic Alumina Fabricated by Hard Anodization. J. Electroanal. Chem. 2009, 632 (1-2), 139-142.

(23)

Rahman, M. M.; Garcia-Caurel, E.; Santos, A.; Marsal, L. F.; Pallarès, J.; FerréBorrull, J. Effect of the Anodization Voltage on the Pore-Widening Rate of Nanoporous Anodic Alumina. Nanoscale Res. Lett. 2012, 7 (1), 474.


Page 18 of 31

Page 19 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(24)

Masuda, H.; Fukuda, K. Ordered Metal Nanohole Arrays Made by a Two-Step Replication of Honeycomb Structures of Anodic Alumina. Science 1995, 268 (5216), 1466–1468.

(25)

Baranowska, M.; Slota, A. J.; Eravuchira, P. J.; Macias, G.; Xifré-Pérez, E.; Pallares, J.; Ferré-Borrull, J.; Marsal, L. F. Protein Attachment to Nanoporous Anodic Alumina for Biotechnological Applications: Influence of Pore Size, Protein Size and Functionalization Path. Colloids Surfaces B Biointerfaces 2014, 122, 375–383.

(26)

Md Jani, A. M.; Losic, D.; Voelcker, N. H. Nanoporous Anodic Aluminium Oxide: Advances in Surface Engineering and Emerging Applications. Prog. Mater. Sci. 2013, 58 (5), 636–704.

(27)

Macias, G.; Ferre-Borrull, J.; Pallares, J.; Marsal, L. F. Effect of Pore Diameter in Nanoporous Anodic Alumina Optical Biosensors. Analyst 2015, 140, 4848–4854.

(28)

Dancil, K.-P. S.; Greiner, D. P.; Sailor, M. J. A Porous Silicon Optical Biosensor: Detection of Reversible Binding of IgG to a Protein A-Modified Surface. J. Am. Chem. Soc. 1999, 121 (34), 7925–7930.

(29)

Dhathathreyan, A. Real-Time Monitoring of Invertase Activity Immobilized in Nanoporous Aluminum Oxide. J. Phys. Chem. B 2011, 115 (20), 6678–6682.

(30)

Santos, A.; MacÍas, G.; Ferré-Borrull, J.; Pallarès, J.; Marsal, L. F. Photoluminescent Enzymatic Sensor Based on Nanoporous Anodic Alumina. ACS Appl. Mater. Interfaces 2012, 4 (7), 3584–3588.

(31)

Ye, W. W.; Shi, J. Y.; Chan, C. Y.; Xiao, L. D.; Yang, M. Nanoporous Alumina Membrane and Nanoparticle Based Microfluidic Sensing Platform for Direct DNA Detection. 2013, No. June, 1255–1258.

(32)

Losic, D.; Simovic, S. Self-Ordered Nanopore and Nanotube Platforms for Drug Delivery Applications. Expert Opin. Drug Deliv. 2009, 6 (12), 1363–1381.

(33)

Mei, Y. F.; Li, Z. M.; Chu, R. M.; Tang, Z. K.; Siu, G. G.; Fu, R. K. Y.; Chu, P. K.; Wu, W. W.; Cheah, K. W. Synthesis and Optical Properties of Germanium Nanorod Array Fabricated on Porous Anodic Alumina and Si-Based Templates. Appl. Phys. Lett. 2005, 86 (2), 021111.

(34)

Jachimska, B.; Pajor, A. Physico-Chemical Characterization of Bovine Serum



Albumin in Solution and as Deposited on Surfaces. Bioelectrochemistry 2012, 87, 138–146. (35)

Nielsch, K.; Choi, J.; Schwirn, K.; Wehrspohn, R. B.; Gösele, U. Self-Ordering Regimes of Porous Alumina: The 10 Porosity Rule. Nano Lett. 2002, 2, 677–680.

(36)

Raspal, V.; Awitor, K. O.; Massard, C.; Feschet-Chassot, E.; Bokalawela, R. S. P.; Johnson, M. B. Nanoporous Surface Wetting Behavior: The Line Tension Influence. Langmuir 2012, 28 (30), 11064–11071.

(37)

Anderegg, J. W.; Beeman, W. W.; Shulman, S.; Kaesberg, P. An Investigation of the Size, Shape and Hydration of Serum Albumin by Small-Angle X-Ray Scattering. J. Am. Chem. Soc. 1955, 77 (11), 2927–2937.

(38)

Santos, A.; Alba, M.; Rahman, M. M.; Formentín, P.; Ferré-Borrull, J.; Pallarès, J.; Marsal, L. F. Structural Tuning of Photoluminescence in Nanoporous Anodic Alumina by Hard Anodization in Oxalic and Malonic Acids. Nanoscale Res. Lett. 2012, 7 (1), 228.

(39)

Stępniowski, W. J.; Norek, M.; Michalska-Domańska, M.; Bombalska, A.; NowakStępniowska, A.; Kwaśny, M.; Bojar, Z. Fabrication of Anodic Aluminum Oxide with Incorporated Chromate Ions. Appl. Surf. Sci. 2012, 259, 324–330.

(40)

Acres, R. G.; Ellis, A. V.; Alvino, J.; Lenahan, C. E.; Khodakov, D. A.; Metha, G. F.; Andersson, G. G. Molecular Structure of 3-Aminopropyltriethoxysilane Layers Formed on Silanol-Terminated Silicon Surfaces. J. Phys. Chem. C 2012, 116 (10), 6289–6297.

(41)

Vandenberg, E. T.; Bertilsson, L.; Liedberg, B.; Uvdal, K.; Erlandsson, R.; Elwing, H.; Lundström, I. Structure of 3-Aminopropyl Triethoxy Silane on Silicon Oxide. J. Colloid Interface Sci. 1991, 147 (1), 103–118.

(42)

Kuzuya, A.; Numajiri, K.; Kimura, M.; Komiyama, M. Single-Molecule Accommodation of Streptavidin in Nanometer-Scale Wells Formed in DNA Nanostructures. Nucleic Acids Symp. Ser. (Oxf). 2008, No. 52, 681–682.

(43)

Hyndman, D.; Flynn, T. G.; Lever, G.; Burrell, R. Protein Immobilization to Alumina Supports: I. Characterization of Alumina-Organophosphate Ligand Interactions and Use in the Attachment of Papain. Biotechnol. Bioeng. 1992, 40 (11), 1319–1327.


Page 20 of 31

Page 21 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(44)

Tan, Y. H.; Liu, M.; Nolting, B.; Go, J. G.; Gervay-Hague, J.; Liu, G. A Nanoengineering Approach for Investigation and Regulation of Protein Immobilization. ACS Nano 2008, 2 (11), 2374–2384.



Figures

Figure 1: Top-view ESEM Images of 6 NAA produced in oxalic and sulfuric acid electrolytes and after the indicated pore widening times, tPW.

Figure 2: (left) Average size distribution of radius estimates obtained by ESEM image analysis for NAA obtained in oxalic and sulphuric acid electrolytes and with the indicated pore widening time. (right) Average pore radii and standard deviation obtained by FICLI and ESEM image analysis.


Page 22 of 31

Page 23 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3: (a) Average pore radius estimated by FICLI for samples produced in sulfuric acid electrolyte and oxalic acid electrolyte with different pore widening times (characterized in Figures 1 and 2) before and after electrostatically immobilizing bovine serum albumin (BSA) to the pore walls. (b) Radius reduction estimated by FICLI for the same sets of samples. (c) Schematic drawing of the surface modification monitored in this experiment.



Figure 4: (a) FICLI radius estimations of an NAA produced in oxalic acid electrolyte with an initial pore widening step (tPW = 18 min), a subsequent functionalization step with APTES+GTA and a final covalent immobilization step of streptavidin. (b) corresponding FICLI slopes. (c) Schematic drawing of the surface modifications monitored in this experiment.


Page 24 of 31

Page 25 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5: (a) Schematic drawing of the different surface modification steps monitored in experiments IC1 to IC4. (b) Consecutive FICLI radius estimation and (c) slopes obtained for NAA as-produced in oxalic acid before and after electrostatic protein A immobilization and successive immune complexation steps antibodies and antigens.



Graphical Abstract


Page 26 of 31

Page 27 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1: Top-view ESEM Images of 6 NAA produced in oxalic and sulfuric acid electrolytes and after the indicated pore widening times, tPW. 234x140mm (173 x 174 DPI)



Figure 2: (left) Average size distribution of radius estimates obtained by ESEM image analysis for NAA obtained in oxalic and sulphuric acid electrolytes and with the indicated pore widening time. (right) Average pore radii and standard deviation obtained by FICLI and ESEM image analysis.


Page 28 of 31

Page 29 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48



ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23


Page 30 of 31

Page 31 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58



Nanoporous Anodic Alumina Surface Modification ... - ACS Publications

Recommend Documents