Dual improvement of DNA-directed antibody immobilization utilizing magnetic fishing and a polyamine coated surface

Abstract

The present study is aimed at the development of a novel approach based on the magnetic improvement of DNA-directed antibody immobilization to prepare a highly efficient sensing platform. Magnetic nanoparticle substrates with high surface area capture the dual DNA-conjugated antibodies in a solution. This allows overcoming the typical mass transport limitation of the surface-based antibody immobilization. Antibody-magnetic nanoparticle conjugation is based on a robust hybridization between a DNA tether (attached to the antibody) and its complementary sequence (immobilized on the nanoparticle). Conventional antibody immobilization for the detection of proteins is often insignificant for the preservation of the folded antibody conformations due to the surface-induced denaturation or drying and dehydration. In this study, a biosensor platform is designed and fabricated by DNA-directed immobilization to maintain the antibody functionality. Also, DNA coated surfaces have advantages of greater stability, a variety of strands, and specificity of DNA base pairs. In principle, this allows the multiplexed detection of proteins within the same surface area. Thus, this attractive approach with DNA-directed antibody immobilization using a poly-L-lysine coated surface with highly pre-adsorbed ssDNA, along with magnetic nanoparticles to increase the surface concentration of the biomolecules, is promising for the utilization in, and the improvement of, surface-based biosensors. Our findings suggest that this method will be effective in a variety of biomedical applications, such as cell separation, diagnosis, and monitoring of human diseases.

---

Introduction

In the past few decades, there has been a significant advancement in the synthesis of molecular probes for improved biomolecule recognition. Traditional ELISA methods, commonly performed in 96-well microtiter plates and are two dimensions provide constrained capacity for binding to antibodies. Utilizing nanoparticles with high surface area in three dimensions improves the protein loading capacity, which increases the number of immobilized biomolecules for further diagnostic and therapeutic applications.

Magnetic nanoparticles (MNPs) have attracted considerable interest due to their advantages including separation efficiency, simplicity, low cost, and easy scale-up ability. Moreover, they have applications in the fields of bio-separation and biomedicine, including protein and enzyme immobilization, immunoassays, RNA and DNA purification, cell isolation, drug delivery, and biosensors.^1–8 Magnetic nanoparticles utilization for immunoassays is based on the immobilization of the antibodies on the magnetic nanoparticles of high surface area. This will lead to an improvement in the immunological reaction rate, a decrease in the antibody immobilization time, better sensitivity, rapid analysis, and simple washing procedures.⁹

An antibody immobilization step is crucial in the design of final biosensor system and should have the following requirements: the antibody must maintain its structure, and the antigen binding site must be capable of recognizing the antigen. Direct immobilization of the antibody or traditional surface chemistry, such as amine reactive surfaces, may result in the loss of antibody activity due to steric hindrance between the surface and adjacent antibody molecules.^10–12 Moreover, the support surface should be fully inert to prevent the nonspecific adsorption of other components onto the surface in the complex fluid.¹³ To address these issues, alternative methods for protein immobilization have been developed to improve the sensitivity and robustness of the system. It has been suggested that the long flexible linkers such as polyethyleneimine and dextran help to reduce the surface effects such as limited accessibility and steric hindrance.^14–17 Therefore, the use of ssDNA oligomer as a long flexible linker will be helpful to retain the antibody functionality, which may be affected by common physical surface adsorption due to surface-induced denaturation and steric hindrance.

DNA-directed immobilization (DDI) has been reported for highly sensitive, selective, and robust systems.^18–23 It takes advantage of the high specificity of the Watson–Crick base pairs for the site-specific immobilization of the ssDNA-tagged antibodies on the pre-modified surface with complementary ssDNA. In this case, oligonucleotides are employed as anchor molecules, which have advantages such as greater stability, a variety of strands, and the specificity of the DNA base pairs. This strategy combines the robustness of DNA microarrays with a specificity of antibodies for diagnostic applications and has several advantages including increased antigen binding capacity,²⁴ improved spot homogeneity,²⁵ and assay reproducibility.^25,26 DNA is easily immobilized on the substrates, and in contrast to protein arrays, DNA arrays can resist the fabrication process utilized in making microfluidic devices. DNA coated surfaces are resistant to dehydration and can be stored or heated (80–100 °C), making them appropriate for microfluidic applications. Moreover, the generation of false positive signals faced by many protein arrays can be avoided by using protein resistant surfaces such as oligo(ethylene glycol)^27,28 or DNA coated surfaces.^29,30 It is understood that because of the hydrophilic characteristic of the surface, the nonselective binding of proteins to the DNA-coated surface is reduced.²⁹ Other benefits of the DNA-directed antibody immobilization include their ability to reprogram the surface by using a different set of antibodies conjugated to the same ssDNA sequences and regenerate the surface by de-hybridization of the DNA–antibody conjugates.

This study presents a novel and robust approach, in which a dual DNA-conjugated antibody is immobilized on the mobile substrates of the magnetic nanoparticles, to simplify the ongoing multiplexed detection of protein biomarkers on the microarray surfaces. Herein, MNPs, instead of the commonly used 2D surfaces, were selected as appropriate supports for the immobilization of the antibodies. The MNPs have special properties including large surface area, proper dispersity, favorable biocompatibility and magnetic properties, which would allow their rapid removal from the reaction mixture.³¹ DNA-modified magnetic nanoparticles are added to a solution to capture the dual DNA-conjugated antibodies free of steric constraints. The process is based on the robust hybridization between a DNA tether (attached to the antibody) and its complementary sequence (immobilized on the nanoparticle). The DDI-based nanoconstructs are then concentrated and immobilized using a magnetic field over a DNA monolayer assembled on a poly-L-lysine (PLL) coated surface. PLL is used to enhance the loading of ssDNA on the surface. Compared to a process in which the antibodies directly form on the sensing surface, the proposed approach provides higher mass transfer and lower equilibration time, resulting in higher surface immobilization of the antibodies.

In addition, we used ssDNA tailored surfaces to diminish the unspecific protein binding on the sensing zone and to allow the specific immobilization of the antibody-MNP construct with a second DDI step. This hybridization reaction occurs between a second ssDNA tether attached to the antibody and the ssDNA tether attached to the surface. This kind of immobilization allows the multiplex detection of proteins on microarrays. This will be achieved using the different sets of DNA sequences immobilized on the defined areas of the microarray surface to hybridize with the complementary ssDNA-conjugated antibodies.

Experimental

Materials and methods

To implement such detection workflow, we utilized the PSMA-specific monoclonal antibody, D2B,³² its antigen (SLIN tagged GCPII (44-750)), 100 nm streptavidin coated magnetic nanoparticles (purchased from Ademtech) and biotin and amine functionalized DNA oligonucleotides purchased from Integrated DNA Technologies (IDT) (HPLC purity). Secondary antibody, horseradish peroxidase (HRP) conjugated and Alexa Fluor 488 conjugated anti-mouse IgG, was obtained from Abcam. The antibody–oligonucleotide conjugation kit was purchased from SoluLink (San Diego, CA). Bovine serum albumin (BSA) and sulfuric acid (H₂SO₄) were supplied by Merck. Sylgard 184 polydimethylsiloxane (PDMS) and its curing agent were bought from Dow Corning. Negative photoresist SU-8 2075 was purchased from Microchem. All other chemicals were obtained from Sigma-Aldrich.

Fabrication of microfluidic devices

Microfluidic chips were fabricated by soft lithography. For this purpose, the SU-8 photoresist is poured on the silicon wafer and spin coated to achieve the desired film thickness. After resist coating, the wafer must be soft-baked to evaporate the solvent and harden the film. UV light is radiated on the SU-8 beyond a transparency photo-mask with desirable pattern and post-baked. This process transfers micrometer sized features on the photo-mask to the photoresist using the photochemical reaction occurring in the exposed area. Thereafter, the substrate is wet-etched to remove the unexposed area. After preparing the master mold, PDMS pre-polymer and curing agent were mixed in a 10 : 1 ratio (w/w) and poured into the mold. After curing (80 °C, 1 hour), the cured PDMS slab was released from the mold and sample inlet and the outlet ports were punched. The microfluidic device was then set on top of the PLL-coated glass slide and bonded to it.

DNA–antibody conjugation

Oligonucleotide-antibody conjugation kit, obtained from SoluLink, was used to conjugate and purify the DNA-labeled antibodies based on the manufacturer's instructions. Briefly, antibody and oligos (NH₂-AA AAA AAA AAA AAA AAG ATC ATG AAG CAC ATC AGA GTC TCC TAC and NH₂-CAA AAC AGC AGC AAT CCA ATG CGC AGA CAC CCG ATT ACA AAT GC) must be completely desalted and buffer exchanged to pH 8.0 in phosphate-buffered saline (PBS) using Zeba™ desalt spin columns. Succinimidyl-4-formylbenzamide (S-4FB) in dimethylformamide (DMF) was added to the 5′-aminated oligomers in PBS and incubated at room temperature for 2 hours to allow the reagent to react with the amino-oligomers. Separately, succinimidyl-6-hydrazino-nicotinamide (S-HyNic) in DMF was added to the antibodies. Excess S-HyNic and S-4FB were removed and the samples were buffer-exchanged to pH 6.0 in PBS using Zeba™ desalt spin column. Derivatized DNA and antibodies were then combined and allowed to react at room temperature for 2 hours and desalted in PBS using Zeba™ desalt spin column.

Synthesis of magnetic nanoparticle-conjugated antibodies

The ssDNA–magnetic nanoparticles binding experiments were performed using the following oligonucleotide (Seq-1): 5′-biotin-CAT TTG TAA TCG GGT GTC TGC GCA TTG GAT. For the biotinylated DNA capture assay, the reaction mixture, containing 2.5 mg mL⁻¹ streptavidin coated MNPs, 8 μM biotinylated DNA oligonucleotides, 5 mM Tris HCl (pH 7.5), 0.5 mM EDTA, 1 M NaCl and 0.05% Tween 20 was incubated for 30 min at room temperature. The product was then rinsed twice with PBST (10 mM Na₂HPO₄, 1.8 mM KH₂PO₄, 137 mM NaCl, 2.7 mM KCl, 0.05% Tween 20, pH 7.4) to remove any impurities. The ssDNA conjugated magnetic nanoparticles were mixed with ssDNA conjugated antibody (100 pg mL⁻¹) and incubated for 1 hour. The product was rinsed with PBST wash buffer (Tween 20/PBS 0.05% v/v).

Binding capacity study of DNA-labeled antibody on the surfaces

For preparing an adequate surface, pure O₂ plasma treatment at 0.4 mbar at different times and frequencies was performed and the surface was dipped in DI water. After drying using a nitrogen gun, 100 μL of PLL solution (0.1% w/v, without dilution) was poured into the well for 1 hour, rinsed off with DI water, and then blown dry. For control experiments, ammonia plasma treatment of the surface was used for amine functionalization of the surface.

Following the DNA treatment of the surface utilizing the 0.2 μM solution of ssDNA, antibody–MNP conjugates were flowed into the surface and incubated for 1 hour at room temperature. For the control experiment, antibody solution (100 pg mL⁻¹) was used. This step converted ssDNA array into captured antibody array. Unbound conjugates were washed off using PBST wash buffer. To evaluate the hybridization of the DNA labeled antibody to its complementary DNA in the plate, an anti-mouse secondary antibody (HRP or Alexa Fluor 488 conjugated) was flowed into the surface. The surface was completely washed to prevent non-specific adsorption of the secondary antibody on it. Finally, the signal was read using a fluorescence microscope or using TMB substrate solution as a signal generating agent and the absorbance was measured at 450 nm by a plate reader (Biotec, ELX800, USA).

Evaluation of the ssDNA conjugated antibody functionality

For this purpose, 100 μL of PSMA antigen was pipetted into each well of the plate and incubated overnight at 4 °C. The wells were washed three times with wash buffer. The blocking step was performed by adding 100 μL BSA 1% to the wells, incubating for 1 hour and washing off three times with PBST wash buffer. DNA–antibody conjugates were introduced to the wells for 1 hour at room temperature and washed. Then, 100 μL secondary antibody (HRP conjugated anti-mouse IgG antibody at a dilution of 1 : 8000, pH = 7.0) was added and incubated for 1 hour at room temperature. Then, five times washing has been done to remove antibodies (with non-specific adsorption) on the plate to prevent any false positive signal. Finally, TMB substrate for generating the signal was added and incubated for 15 min in the dark. To stop the enzymatic reaction of HRP, 100 μL of the H₂SO₄ solution was added to each well. The absorbance was read at 450 nm using an ELISA plate reader.

Results and discussion

The binding capacity of the ssDNA–antibody conjugates on the surfaces has been studied, and Fig. 1 schematically depicts the experimental design. At first, a surface treated with PLL or ammonia plasma has been prepared and functionalized with ssDNA (Fig. 1a and c). MNP conjugated with single-stranded ssDNA is combined with a PSMA-specific antibody modified with two different ssDNA oligomers, one of which (depicted in red in Fig. 1e) is complementary to the DNA sequence immobilized on the MNP, and the other one (depicted in black in Fig. 1e) is complementary to the ssDNA oligomer pre-immobilized on the surface. Thus, the anti-PSMA antibody is conjugated onto the MNPs with high capacity that converted to high sensitivity in the resulted immunosensor. These components lead to the formation of the anti-PSMA containing molecular construct (see Fig. 1e). Such DDI-based nanoconstructs are subsequently concentrated, by means of a magnetic field, over a DNA monolayer, and thus, increasing the mass transfer, reducing the equilibrium time, and enhancing the signal-to-noise ratio with respect to the standard surface-based immunosorbent assays. These molecular constructs are immobilized by means of DNA hybridization. For control experiments, we carried out similar experiments lacking MNPs in the diffusion-limited regime (Fig. 1b and d).

Fig. 1 Magnetically enhanced anti-PSMA monoclonal antibody immobilization based on DDI approach over a self-assembled DNA monolayer. (a) Schematic for the ssDNA coating over the ammonia plasma treated surface; (b) the surface immobilization of the immunocomplex in diffusion-limited regime over the low ssDNA coated surface; (c) schematic for the high ssDNA loading over the PLL-coated surface; (d) the surface immobilization of the immunocomplex in diffusion-limited version over the high ssDNA loaded surface; (e) an immunocomplex including MNPs and anti-PSMA antibodies functionalized with two DNA molecule forms in solution and a magnet is utilized to concentrate the immunocomplex assembly over the surface of a microtiter plate. Surface-immobilization is based on the hybridization of the dangling ssDNAs over the antibodies (in black) and the surface-attached ssDNAs (in blue). Moreover, HRP conjugated anti-mouse IgG is added to quantify anti-PSMA capture antibody immobilization.

Seymour et al. studied three DNA tether lengths of 20, 40 and 60 mers.³³ They concluded that thermal fluctuations result in different orientations for short DNA molecules in the polymeric scaffold. Shorter DNA probes (e.g. 20 mers) can penetrate deeper into the polymer scaffold and their average height is smaller than that expected for a fully extended one. However, the steric hindrance and electrostatic repulsion orients the longer DNA strands (i.e. 40 mers) at higher axial positions, whereas 60 mers DNA sequences do not provide a significant advantage over 40 mers sequences. Therefore, we selected 44 bp DNA sequences for pre-adsorption on the PLL-coated surface since it keeps the antibody apart from the surface and results in lower steric hindrance. Moreover, designing of shorter DNA probes is simpler due to a greater possibility of the formation of secondary structures in longer DNA probes.

Fig. 2a shows that PLL coated surface is advantageous over the ammonia plasma treated surface as it leads to a three-fold increase in the absorbance for the detection of anti-PSMA, which reflects a more efficient surface immobilization of the anti-PSMA. This aspect can reflect higher ssDNA loading onto the PLL-coated surface compared to the ammonia plasma-treated surface. In addition, the utilization of MNPs considerably prompts the immobilization efficiency of the anti-PSMA-containing constructs more (two-fold) than that of the corresponding diffusion-limited analogs. Thus, in this approach, antibody will be more efficiently immobilized on the surface than that in common surface-based assays, which will result in more sensitivity for detecting the antigen.

Fig. 2 (a) Absorbance intensities. The first, second and third bars are relative to the strategy depicted in Fig. 1b, d and e sequentially, the fourth and fifth bars are relative to the control experiment lacking anti-PSMA antibody for the determination of nonspecific immobilization of the HRP conjugated anti-mouse IgG over a ssDNA coated surface and a non-treated microtiter surface. (b) Fluorescence image of the approach depicted in Fig. 1e in the microfluidic system using Alexa Fluor 488 conjugated anti-mouse IgG as the secondary antibody.

The presence of a DNA layer on the surface leads to a highly hydrophilic surface. This property can prevent unspecific binding of the hydrophobic protein on the surface. This feature has been investigated using HRP-conjugated anti-mouse IgG immobilization on the ssDNA-treated and non-treated surface (see fourth and fifth bars in Fig. 2a). As it can be seen, the sensing platform allows the specific immobilization of the ssDNA–antibody construct with DDI approach on the surface (first, second and third bars) and hydrophobic HRP-conjugated anti-mouse IgG is not immobilized on the ssDNA-coated surface (fourth bar). However, these proteins are immobilized on the non-treated surface with hydrophobic interactions (fifth bar). Bailey et al. suggested that the hydrophilic properties of the spotted oligonucleotides on the surface minimize their interactions with the hydrophobic proteins in the solution through nonspecific adsorption.²⁹

The binding capacity of the anti-PSMA conjugated MNPs on the microfluidic system was examined by Alexa Fluor 488 conjugated anti-mouse IgG antibody (Fig. 2b). It can be seen that these components were efficiently conjugated in the microfluidic channel. With respect to low reagent consumption and therefore, reduction of cost and processing time for the assay, this immobilization method in the microfluidic channel will be useful for further recognition of the biomolecules with high sensitivity due to a highly pre-adsorbed antibody on the surface.

Contact angle measurements show the surface plasma efficiency in making the surface hydrophile. The results are shown in Table 1 and Fig. 3. No change in the contact angle during the plasma treatment of 1 min and frequency of GHz denotes that this time is not enough for changing the hydrophilic characteristic of the surface. The results show that on increasing the time of plasma treatment from 1 min to 5 min, the hydrophilic characteristic of the surface increased. A comparison between the kHz and GHz plasma treatment after 3 and 5 min suggested that kHz plasma frequency is more efficient than GHz plasma frequency. This may be due to the fact that the plasma frequency enhancement causes OH bond dissociation and therefore, results in the reduction of the hydrophilic characteristic of the surface.

Table 1 Contact angle values determined after the plasma treatment

Frequency of the plasma treatment (150 W)	Time of plasma treatment (min)	Left angle	Right angle
No plasma	0	37.9 ± 0.3	38.3 ± 0.5
kHz	3	6.3 ± 0.2	6.3 ± 0.2
kHz	5	<5	<5
GHz	1	37.7 ± 0.6	36.9 ± 0.7
GHz	3	12.3 ± 0.6	13.0 ± 0.7
GHz	5	7.2 ± 0.8	7.2 ± 0.8

---

Fig. 3 Contact angle measurements of the glass slides (a) without plasma treatment (kHz) and (b) 3 min (c) 5 min (GHz) and (d) 1 min (e) 3 min and (f) 5 min plasma treatment.

Fig. 4 shows the contact mode AFM images of the plasma, PLL and ssDNA treated glass slides, respectively. As can be seen in Fig. 4a, plasma treatment of the surface makes it a little bit rough. Poly-L-lysine is homogeneously coated on the glass slide and the surface is almost smooth. After treating with ssDNA (Fig. 4b), the surface of the glass slide is densely coated with ssDNA. In fact, pre-coating with the polyamine polymer, PLL, resulted in high DNA loading, which will lead to the enhancement of antibody immobilization on the surface and therefore, greater sensitivity.

Fig. 4 Contact mode AFM images of (a) the plasma (b) poly-L-lysine and (c) ssDNA coated surface.

Immobilization of ssDNA on the surface occurs through electrostatic interactions.³⁴ Therefore, ssDNA probes may lie on the surface and in some other parts because of the repulsive forces between the negatively charged ssDNA probes may extend vertically. Moreover, short ssDNA probes can penetrate into the polymer scaffold, therefore, the height of the surface is lower than expected.³³

The ssDNA–antibody conjugation was performed in three steps. The first step was the modification of the ssDNA with S-4FB crosslinker followed by the formation of the S-HyNic modified antibody. Finally, the mixing and reaction of S-HyNic modified anti-PSMA with S-4FB modified ssDNA resulted in the formation of a stable and UV-traceable bis-aryl hydrazone bond with measurable absorbance at 354 nm, which has a molar extinction coefficient of 29 000. So, the conjugation reaction is spectrophotometrically visualized due to the formation of the chromophoric conjugate bond by measuring the absorbance at 354 nm (Fig. 5). The result indicates a well-performed conjugation with a molar substitution ratio (ssDNA/antibody) of 3.2.

Fig. 5 UV-vis evaluation of the ssDNA–antibody conjugates and its chromophoric conjugate bond at 354 nm.

Antibody conjugation to some linkers may result in the denaturation of the folded structure of the antibody and loss of its functionality. Therefore, indirect ELISA in MaxiSorp 96-well plate was performed to evaluate the functionality of DNA conjugated antibody against the cognate antigen, PSMA. Fig. 6a shows the schematic of this assay. Purple and blue bars relate to non-conjugated antibody and ssDNA-conjugated antibody, respectively. A comparison between these bars shows that the ssDNA conjugated antibody retained its functionality, indicating its utility in the proposed sensing platform. Also, the first and second blue bars relate to the ssDNA conjugated antibody functionality after 1 and 4 months, respectively, showing that the antibody functionality was retained after a long time period.

Fig. 6 (a) Schematic for the evaluation of ssDNA conjugated antibody functionality with ELISA assay; (b) absorbance intensity of the experiments. The purple and blue bars are related to the evaluation of the non-conjugated antibody and ssDNA conjugated antibody functionality. The first and second blue bars are related to the ssDNA conjugated antibody functionality after 1 and 4 months, respectively. The pink bar is related to the control experiment lacking primary antibody for the determination of nonspecific adsorption of HRP conjugated anti-mouse IgG on the surface.

According to the high immobilization capacity of the antibody on the surface and retainment of antibody functionality, this approach will be effective in various surface-based immunoassays due to the high capture antibody immobilization, resulting in higher sensitivities.

Conclusion

We demonstrated the feasibility of the fabrication and development of a novel magnetic nanoparticle-based high-capacity antibody immobilization system. These substrates with high surface area semi-homogeneously capture the antibodies in the solution. This allows overcoming the typical mass transport limitation of the surface-based antibody immobilization. Then, the magnetic concentration of these constructs on the surface improves the assay sensitivity for a short incubation time. Since DNA microarrays are easier to prepare compared to protein microarrays, DDI strategy offers high manufacturing capacity and quality, and they can be produced and stored at room temperature for a long time without denaturation. Moreover, the use of a PLL coated surface resulted in high ssDNA loading and high antibody immobilization.

This novel magnetically enhanced DDI has an extreme potential to improve the sensitivity and accuracy of the conventional ELISA-based immunoassays. We hope that this novel technique for antibody immobilization can open new opportunities toward high sensitivity detection of the related antigen.

Acknowledgements

The authors are grateful to the Iran National Scientific Foundation (INSF) for their support through the Grant Number 92037030. The authors would also like to thank Prof. Marco Colombatti (University of Verona, Italy) for providing the purified D2B anti-PSMA antibody and to Prof. Jan Konvalinka (Charles University, Czech Republic) for providing the PSMA antigen (SLIN tagged GCPII (44-750)).

Notes and references

1. J. Chatterjee, Y. Haik and C. J. Chen, J. Magn. Magn. Mater., 2001, 225, 21–29.
2. Z. Bílková, M. Slováková, D. Horák, J. Lenfeld and J. Churáček, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2002, 770, 177–181.
3. M. Y. Arica, H. Yavuz, S. Patir and A. Denizli, J. Mol. Catal. B: Enzym., 2000, 11, 127–138.
4. A. Heebll-Nielsen, W. S. Choe, A. P. J. Middelberg and O. R. T. Thomas, Biotechnol. Prog., 2003, 19, 887–898.
5. T. Alefantisa, P. Grewala, J. Ashtona, A. S. Khanb, J. J. Valdesb and V. G. D. Vecchioa, Mol. Cell. Probes, 2004, 18, 379–382.
6. F. Sauzedde, A. Elaissari and C. Pichot, Macromol. Symp., 2000, 151, 617–623.
7. A. S. Lüubbe, C. Alexiou and C. Bergemann, J. Surg. Res., 2001, 95, 200–206.
8. L. Josephson, J. M. Perez and R. Weissleder, Angew. Chem., Int. Ed., 2001, 40, 3204–3206.
9. T. Ivanova and T. Godjevargova, Anal. Lett., 2015, 48, 843–860.
10. P. Peluso, D. S. Wilson, D. Do, H. Tran, M. Venkatasubbaiah, D. Quincy, B. Heidecker, K. Poindexter, N. Tolani, M. Phelan, K. Witte, L. S. Jung, P. Wagner and S. Nock, Anal. Biochem., 2003, 312, 113–124.
11. H. Sun, G. Y. J. Chen and S. Q. Yao, Chem. Biol., 2013, 20, 685–699.
12. J. M. Schwenk, J. Lindberg, M. Sundberg, M. Uhlén and P. Nilsson, Mol. Cell. Proteomics, 2007, 6, 125–132.
13. P. Batalla, C. Mateo, V. Grazu, R. Roberto Fernandez-Lafuente and J. M. Guisan, Process Biochem., 2009, 44, 365–368.
14. J. Yakovleva, J. R. Davidsson, M. Bengtsson, T. Laurell and J. Emneus, Biosens. Bioelectron., 2003, 19, 21–34.
15. J. Yakovleva, R. Davidsson, A. Lobanova, M. Bengtsson, S. Eremin, T. Laurell and J. Emnéus, Anal. Chem., 2002, 74, 2994–3004.
16. Y. L. Bai, C. G. Koh, M. Boreman, Y. J. Juang, I. C. Tang, L. J. Lee and S. T. Yang, Langmuir, 2006, 22, 9458–9467.
17. C. Jonsson, M. Aronsson, G. Rundstrom, C. Pettersson, I. Mendel-Hartvig, J. Bakker, E. Martinsson, B. Liedberg, B. MacCraith, O. Ohman and J. Melin, Lab Chip, 2008, 8, 1191–1197.
18. K. Ren, J. Wu, F. Yan and H. Ju, Sci. Rep., 2014, 4, 1–6.
19. K. Chuah, L. M. Lai, I. Y. Goon, S. G. Parker, R. Amal and J. Justin Gooding, Chem. Commun., 2012, 48, 3503–3505.
20. K. Ren, J. Wu, Y. Zhang, F. Yan and H. Ju, Anal. Chem., 2014, 86, 7494–7499.
21. C. Boozer, J. Ladd, S. Chen, Q. Yu, J. Homola and S. Jiang, Anal. Chem., 2004, 76, 6967–6972.
22. H. Li, W. Zhang and H. Zhou, Anal. Biochem., 2014, 449, 26–31.
23. N. S. Ferreira and M. G. F. Sales, Biosens. Bioelectron., 2014, 53, 193–199.
24. A. L. Washburn, J. Gomez and R. C. Bailey, Anal. Chem., 2011, 83, 3572–3580.
25. R. Wacker, H. Schröder and C. M. Niemeyer, Anal. Biochem., 2004, 330, 281–287.
26. R. Wacker and C. M. Niemeyer, ChemBioChem, 2004, 5, 453–459.
27. K. L. Prime and G. M. Whitesides, J. Am. Chem. Soc., 1993, 115, 10714–10721.
28. K. E. Nelson, L. Gamble, L. S. Jung, M. S. Boeckl, E. Naeemi, S. L. Golledg, T. Sasak, D. G. Castner, C. T. Campbell and P. S. Stayton, Langmuir, 2001, 17, 2807–2816.
29. R. C. Bailey, G. A. Kwong, C. G. Radu, O. N. Witte and J. R. Heath, J. Am. Chem. Soc., 2007, 129, 1959–1967.
30. R. Fan, O. Vermesh, A. Srivastava, B. K. H. Yen, L. Qin, H. Ahmad, G. A. Kwong, C. C. Liu, J. Gould, L. Hood and J. R. Heath, Nat. Biotechnol., 2008, 26, 1373–1378.
31. J. Huang, P. Su, J. Wu and Y. Yang, RSC Adv., 2014, 4, 58514–58521.
32. M. Colombatti, S. Grasso, A. Porzia, G. Fracasso, M. T. Scupoli, S. Cingarlini, O. Poffe, H. Y. Naim, M. Heine, G. Tridente, F. Mainiero and D. Ramarli, PLoS One, 2009, 4, e4608.
33. E. Seymour, G. G. Daaboul, X. Zhang, S. M. Scherr, N. L. Ünlü, J. H. Connor and M. S. Ünlü, Anal. Chem., 2015, 87, 10505–10512.
34. V. Tjong, L. Tang, S. Zauscher and A. Chilkoti, Chem. Soc. Rev., 2014, 43, 1612–1626.

---