Langmuir 2009, 25, 661-664
661
Straightforward Protein Immobilization Using Redox-Initiated Poly(methyl methacrylate) Polymerization Kevin A. Heyries, Loı¨c J. Blum, and Christophe A. Marquette* Laboratoire de Ge´nie Enzymatique et Biomole´culaire, UniVersite´ Lyon 1 - CNRS 5246 ICBMS, Baˆt CPE, 43, bd du 11 noVembre 1918, 69622 Villeurbanne, Cedex, France ReceiVed October 29, 2008. ReVised Manuscript ReceiVed December 10, 2008 Rigid poly(methyl methacrylate) (PMMA) biochips directly modified with active protein spots were obtained, using a redox-initiated PMMA polymerization process. The protein immobilization mechanism is believed to be a combination of both a covalent binding through transient amino acid radical generation and a direct entrapment of the biomolecules in the PMMA polymer. Three different immunoassays (binding, capture, and sandwich) were performed using the developed system, and really promising limits of detection (160-200pg/mL) were obtained, demonstrating a novel straightforward route to fabricate plastic biochips.
Introduction An increased interest in the combination of microfluidics and microarrays emerged from the recent developments of genomics and proteomics.1 Additionally, the technological pressure for the production of cheap and efficient microelectro-mechanical systems (MEMS) has led to the progressive replacement of common glass or silicon materials by polymers.2 Their cost and availability associated with their tunable chemical reactivity make them more and more popular. In this context, poly(dimethylsiloxane) (PDMS) rapidly rose as a first choice material thanks to its low toxicity, handling easiness, and transparency. Consequently, PDMS became, in the last years, the cornerstone of the soft lithography-based techniques.3 In spite of these interesting properties, convenient for biosensing applications, PDMS also exhibits several drawbacks4 that lower its range of capabilities. Indeed, its strong hydrophobicity associated with a chemical inertness imply harsh chemical modifications to fulfill the surface property requirements5 necessary for biomolecule immobilization. Our group proposed recently a method called “macromolecules to PDMS transfer” for the direct modification of PDMS surfaces with spots of macromolecules.6-8 This technique was shown to enable the achievement of sensitive PDMS-based biochips for different applications, ranging from protein to DNA specific detection. The major concerns of the technique, often highlighted by peers and reviewers, were, first, the necessary PDMS curing step at elevated temperature, which might lead to macromolecules denaturation, and, second, the fact that PDMS could not be easily integrated in a largescale industrial process.9 On the contrary, thermoplastic materials, classically used in industry, hold great promise in the field of biochip development and industrialization. Among them, poly(methyl methacrylate) (PMMA) is an interesting candidate for microsystem developments thanks to * Corresponding author. E-mail:
[email protected]. (1) Yager, P.; Edwards, T.; Fu, E.; Helton, K.; Nelson, K.; Tam, M. R.; Weigl, B. H. Nature 2006, 442(7101), 412–418. (2) Becker, H.; Ga¨rtner, C. Anal. Bioanal. Chem. 2008, 390(1), 89–111. (3) Xia, Y.; Rogers, J. A.; Paul, K. E.; Whitesides, G. M. Chem. ReV. 1999, 99(7), 1823–1848. (4) Toepke, M. W.; Beebe, D. J. Lab Chip 2006, 6, 1484–1486. (5) Goddard, J. M.; Hotchkiss, J. H. Prog. Polym. Sci. 2007, 32(7), 698–725. (6) Heyries, K. A.; Blum, L. J.; Marquette, C. A. Chem. Mater. 2008, 20(4), 1251–1253. (7) Heyries, K. A.; Loughran, M. G.; Hoffmann, D.; Homsy, A.; Blum, L. J.; Marquette, C. A. Biosens. Bioelectron. 2008, 23(12), 1812–1818. (8) Heyries, K. A.; Marquette, C. A.; Blum, L. J. Langmuir 2007, 23(8), 4523– 4527. (9) Mukhopadhyay, R. Anal. Chem. 2007, 79(9), 3249–3253.
its excellent optical transparency,10 its low cost, and its good mechanical properties.11 The current work reports a new path to obtained PMMA polymerbased biochips using the initial concept of “Macromolecules to Polymer Transfer” (MPT) while overcoming any thermopolymerization steps. Indeed, PMMA which can be polymerized under UV light exposure or heating, can also be prepared using a redox initiator reaction. This redox system is composed of benzoyl peroxide (BPO) and N,Ndimethyl-p-toluidine (N,N-DMPT) and leads to the polymerization of liquid PMMA using a so-called “cold curing process”. The reaction of N,N-DMPT with BPO generates benzoic acid radicals, which trigger a radical-based polymerization at room temperature (Figure 1). This redox polymerization of liquid PMMA was reported previously for the fabrication of microfluidic chips12 and the development of biosensing platforms.13 In the present study, liquid PMMA is used as a molding polymer on a dedicated Teflon substrate prepared through microdrilling. Spots (1.3 nL) of proteins are dried on this substrate and used for biochip production. The approach is similar to our previously published “macxromolecules to PDMS transfer” procedure, except for the fact that liquid PDMS is replaced by liquid PMMA.
Materials and Methods Briefly, 8 g of PMMA beads (600 µm in diameter) was dissolved in 15 mL of distilled methyl methacrylate monomer (MMA) solution. The mixture was heated at 80 °C and stirred for 30 min. At that time, 150 mg of finely ground BPO was added, and the solution was stirred for 10 min at room temperature. Finally, 50 µL of N,NDMPT was added and stirred for 2 min, and the PMMA solution was poured on the three-dimensional (3D) substrate (Figure 1) using a 10 mL syringe. Liquid PMMA spread slowly over the entire substrate and completely polymerized at room temperature after 90 min. Air exposure of the unpolymerized PMMA generated a yellowish coloration. Safety considerations: PMMA polymerization must be performed in a fume hood since N,N-DMPT is toxic. The peeling off of the polymerized PMMA from its molding substrate produced a 3D biochip (here composed of 24 5 × 5 × 1 mm3 wells) with bound biomolecules at the surface of the well (10) Diaz-Quijada, G. A.; Peytavi, R.; Nantel, A.; Roy, E.; Bergeron, M. G.; Dumoulin, M. M.; Veres, T. Lab Chip 2007, 7(7), 856–862. (11) Lee, G.-B.; Chen, S.-H.; Huang, G.-R.; Sung, W.-C.; Lin, Y.-H. Sens. Actuators, B 2001, 75(1-2), 142–148. (12) Chen, J.; Lin, Y.; Chen, G. Electrophoresis 2007, 28(16), 2897–2903. (13) Pe´rez, J. P. H.; Lo´pez-Cabarcos, E.; Lo´pez-Ruiz, B. Biomol. Eng. 2006, 23(5), 233–245.
10.1021/la803597m CCC: $40.75 2009 American Chemical Society Published on Web 12/23/2008
662 Langmuir, Vol. 25, No. 2, 2009
Letters
Figure 1. Overview of the “Macromolecules to Polymer Transfer (MPT)” protocol and of the PMMA polymerization process in the presence of BPO and N,N-DMPT (activator).12
Figure 2. A typical chemiluminescent image obtained from polymerized PMMA modified with spots of peanut lectin and rabbit IgG. Chemiluminescent signals were obtained using horseradish peroxidase labeled antibodies. Light intensity signals were converted using color range.
bottom. Two proteins, a rabbit G-type immunoglobulin (IgG) and an allergenic protein (lectin) from peanuts, were immobilized as capture agents for immunoassay. The geometry (24 wells) of the biochip allowed easy incubation with the respective antibodies in a 25 µL reaction volume. The three proteins, rabbit IgG, peanut lectin, and anti-C-reactive protein (anti-CRP) monoclonal antibody, were spotted as 1 mg/mL, 1 mg/mL, and 0.1 mg/mL solution in 0.1 M carbonate buffer pH9, respectively. The solutions were spotted as 1.3 nL onto the 3D Teflon substrate using a BioChip Arrayer BCA1 (Perkin Elmer). Before
any experiments, the PMMA biochip surface containing protein spots was saturated with VBSTA buffer for 20 min at 37 °C. The VBSTA buffer contained Veronal 30 mM, NaCl 0.2M, pH 8.5, with the addition of Tween20 0.1% v/v and bovine serum albumin (BSA) 1% w/v. All the different steps were performed at 37 °C. Specific interactions between immobilized proteins and their respective antibodies have been assayed through three different approaches, i.e. binding assay, capture assay, and sandwich assay: For the binding assay, 25 µL of antirabbit antibodies labeled with horseradish peroxidase (Jackson Immuno Research, US) at different
Letters
Langmuir, Vol. 25, No. 2, 2009 663 (Sigma, France) (1 µg/mL) was then incubated for 30 min, and the biochip was finally washed with 50 µL of VBS buffer for 20 min. For the three different approaches, the chemiluminescent images were obtained following the deposition of 25 µL of chemiluminescent reagents (SuperSignal ELISA Femto Maximum Sensitivity Substrate, Thermo Scientific) in each well and placing the PMMA biochip in the CCD camera’s (Las-1000 Plus, Intelligent Dark Box II, FUJIFILM) measurement chamber for light integration for 10 min.
Results
Figure 3. Quantification of the chemiluminescent numeric images obtained for the detection of (A) antirabbit IgG, (B) antipeanut lectin, and (C) free CRP. Error bars represent the standard deviation (n ) 8) and curves are provided as a guide for the eyes.
concentrations in VBSTA was incubated for 1 h. A washing step with 50 µL of VBS buffer (Veronal 30 mM, NaCl 0.2 M, pH 8.5) was finally performed for 20 min. With the capture assay, 25 µL of rabbit antilectin antibodies (Sigma, France) at different concentrations in VBSTA was incubated for 1 h onto the biochip. At that time, 25 µL of antirabbit IgG labeled with horseradish peroxidase (Jackson Immuno Research, US) (1 µg/mL) was incubated for 30 min with the functionalized PMMA surface. Finally, 50 µL of VBS buffer was used to wash the biochip. Finally, the sandwich assay was performed using 25 µL of CRP (from Biodesign, US) solution incubated for 1 h onto the PMMA surface presenting monoclonal anti-CRP antibody spots. Afterward, a biotinylated monoclonal antibody (100 ng/mL) directed against CRP (Exbio, Czech Republic) was incubated for 1 h in VBSTA. A solution containing streptavidin labeled with horseradish peroxidase
The Figure 2 presents a chemiluminescent image (Las-1000 Plus, Intelligent Dark Box II, FUJIFILM) of the entire 24-well biochip following its incubation with various concentrations of specific antibodies. As can be seen, a very low background signal was obtained, even at high antibody concentrations, proving that biomolecules are not prone to adsorb on PMMA surfaces. The interspot signal, which can be observed at high protein concentration, is due to the chemiluminescent signal diffusion. In addition, no signal was recorded from bovine serum albumin spots, evidencing the absence of nonspecific binding of the different biomolecules. Finally, in the absence of labeled target molecules, no chemiluminescent signal was obtained (negative control). All these observations suggest that the different proteins were strongly immobilized (even under intensive washing using water flow) at the surface of the PMMA biochip in an accessible conformation. The chemistry behind such protein immobilization at the surface of PMMA might be related to radical generation from the amino acid residues of the protein sequence. Few studies have shown that proteins14 or enzymes can react with radical species (peroxides), engendering transient radicals on specific amino acid lateral chains, even close to the protein surface.15 Thus, tyrosine can generate tyrosyl radicals, whereas tryptophan and cystein produce radicals in the presence of peroxide molecules. On the other hand, the entrapment of macromolecules using acryl radicals species has also been described16 and used as 3D pads for biochip fabrication. These already published results suggest that the use of radicals do not significantly alter the recognition capabilities of the immobilized proteins. The entrapment effect during the present immobilization process was studied using PMMA directly dissolved in propylene glycol monomethyl ether acetate solvent (PGMEA).17 The solution was poured on macromolecules spots (rabbit IgG), and the solvent evaporated for 18 h. The resulting PMMA polymer exhibited immobilized protein spots available for specific interactions with antibodies (data not show), evidencing the protein entrapment or noncovalent interaction at the PMMA surface. These results suggest that the immobilization process occurring during the PMMA radical polymerization might be the combination of entrapment and interaction of radical species to generate stable and accessible protein spots. In order to confirm the potentialities of the present approach, the analytical performances of the biochip were assayed through three different chemiluminescent immunoassays: binding (Figure 3A), capture (Figure 3B), and sandwich (Figure 3C). For this purpose, three proteins were spotted together with a nonspecific control protein (BSA) and transferred to the PMMA (14) Østdal, H.; Davies, M. J.; Andersen, H. J. Free Radical Biol. Med. 2002, 33(2), 201–209. (15) Svistunenko, D. A. Biochim. Biophys. Acta 2005, 1707(1), 127–155. (16) Darievich, M. A.; Jurievna, R. A.; Vasilievich, P. S. 200 4. (17) Shih, T.-K.; Chen, C.-F.; Ho, J.-R.; Chuang, F.-T. Microelectron. Eng. 2006, 83(3), 471–475.
664 Langmuir, Vol. 25, No. 2, 2009
Letters
surface. To the previously spotted peanut lectin and rabbit IgG, a monoclonal antibody developed against CRP was added and used as an immobilized reagent for the sandwich detection of this clinically relevant parameter.18 This antibody was added in order to demonstrate that the immobilized proteins were actually active and could be used to specifically bind their target. The chemiluminescent signals obtained from each assay were quantified and plotted as dose-response curves (Figure 3). Excellent detection limits (signal-to-noise ratio of 3) were reached, with 160 pg/mL for both binding and capture assays and 200 pg/mL for the detection of free CRP, using 1 h assay duration. These results compared favorably with recently published studies about rabbit IgG and peanut lectin8 immobilization on PDMS and CRP detection using other microarray systems.19,20
The mean standard deviations associated with these data were also satisfactory with 9%, 9%, and 11% of signal variation for rabbit IgG, peanut lectin, and anti-CRP biochips, respectively.
(18) Sabatine, M. S.; Morrow, D. A.; de Lemos, J. A.; Gibson, C. M.; Murphy, S. A.; Rifai, N.; McCabe, C.; Antman, E. M.; Cannon, C. P.; Braunwald, E. Circulation 2002, 105(15), 1760–1763. (19) Wolf, M.; Juncker, D.; Michel, B.; Hunziker, P.; Delamarche, E. Biosens. Bioelectron. 2004, 19(10), 1193–1202. (20) Christodoulides, N.; Tran, M.; Floriano, P. N.; Rodriguez, M.; Goodey, A.; Ali, M.; Neikirk, D.; McDevitt, J. T. Anal. Chem. 2002, 74(13), 3030–3036.
Acknowledgment. K. A. Heyries thanks T. K. Shih for helpful discussions about PMMA polymerization. Published with the support of the European Commission, Sixth Framework Program, Information Society Technologies (NANOSPAD, No. 016610).
Conclusions In conclusion, we have demonstrated the possibility to directly create PMMA biochips containing active protein spots at the PMMA-air interface, taking advantage of the polymer radical polymerization process. The detection limits and the reproducibility values obtained using chemiluminescence comfort us in believing that the described procedure has the potential to become a popular method for PMMA surface biomolecules patterning.
LA803597M
