Anal. Chem. 2008, 80, 8561–8567
Optimal Conditions for Protein Array Deposition Using Continuous Flow Sriram Natarajan,†,‡ Andrew Hatch,§ David G. Myszka,| and Bruce K. Gale*,‡,⊥ Department of Chemical Engineering, Department of Bioengineering, Center for Biomolecular Interaction Analysis, Department of Mechanical Engineering, and Center of Excellence for Biomedical Microfluidics, University of Utah, Salt Lake City, Utah 84112 Optimal conditions for depositing protein microarrays using a continuous-flow microfluidic device, the continuous-flow microspotter (CFM), have been determined using a design of experiments approach. The amount of protein deposited on the surface depends on the rates of convective and diffusive transport to the surface and binding at the surface. These rates depend on parameters such as the flow rate, time, and capture mechanism at the surface. The process parameters were optimized, and uniform protein spots were obtained at a protein concentration of 10 µg/mL and even at 0.4 µg/mL. A 150-fold dilution in protein concentration in the sample solution decreased surface concentration by a factor of only 16. If the capture mechanism of the protein on the substrate is nonspecific, optimal deposition is obtained at higher flow rates for short periods of time. If the capture mechanism is specific, such as biotin-avidin, deposition is optimal at medium flow rates with little advantage beyond 30 min. The CFM can be used to deposit protein arrays with good spot morphology, spot-to-spot uniformity and enhanced surface concentration. The CFM was used to deposit an array of various antibodies, and their interactions with an antigen were studied using surface plasmon resonance (SPR). Affinity values were obtained at low antibody concentrations (5 µg/mL) with low coefficients of variation. Thus, the CFM can be used to effectively capture proteins and antibodies from dilute samples while depositing multiple spots, thereby increasing the quality of spots in protein microarrays and especially improving screening throughput of SPR. Protein and antibody arrays are used for disease biomarker identification and quantification, protein function determination, and general binding assays.1-3 Currently, commercial high spot density arrays are produced using complex robotic spotter systems * To whom correspondence should be addressed. Address: 50 S. Central Campus Drive, Rm 2122, University of Utah, Salt Lake City, UT 84112. Phone: (801) 585-5944. Fax: (801)-585-9826. E-mail:
[email protected]. † Department of Chemical Engineering. ‡ Center of Excellence for Biomedical Microfluidics. § Department of Bioengineering. | Center for Biomolecular Interaction Analysis. ⊥ Department of Mechanical Engineering. (1) Joos, T. O.; Schrenk, M.; Hopfl, P.; Kroger, K.; et al. Electrophoresis 2000, 21, 2641–2650. (2) Zhu, H.; Belgin, M.; Bangham, R.; Hall, D.; et al. Science 2001, 293, 2101– 2105. 10.1021/ac8014609 CCC: $40.75 2008 American Chemical Society Published on Web 10/22/2008
that either use pins that touch down on the substrate or drop ejection from a piezoelectric head to deposit spots. Although conventional spotter systems are capable of producing multiple spots of a controlled size, the concentration of the desired molecule in the spot is limited by its concentration in the depositing solution, and not enough protein will be deposited from low-concentration solutions. If higher concentration protein solutions are used for spotting, molecules can pile up and form multiple layers after the spot dries. When a buffer or blocking solution is flowed over these spots, loosely bound molecules are rinsed from the substrate. This can lead to higher signal-to-noise ratios.4 The protein in the buffer cannot be reused without purification, which makes conventional spotting highly inefficient when expensive or rare proteins have to be arrayed. With conventional spotters, there can also be a large variation in spot uniformity and the spots are susceptible to drying and denaturation. If the surface concentration of a spot varies or is inconsistent, the binding kinetics obtained using these spots may be different, even for the same antigen-target pair.5 Thus, spot uniformity is critical if microarrays are to be used to generate quantitative data. It is possible to produce high-quality spots from low-concentration solutions using continuous flow.6 These systems produce a high surface density by recirculation of the solution over the substrate. If the substrate is treated to adhere specific molecules, unwanted bulk material can be washed off or added back to the bulk deposition solution.7-9 Antibody capture from crude and dilute samples is made possible by continuous flow. Each spot is individually addressed in parallel via microfluidic channels, so sequential chemical processing steps (such as activation, deposition, and blocking), chemical reactions and layer-by-layer selfassembly on individual spots can be customized. The initial work on microfluidic printing was done with planar or 2-D devices; that is, the channels are placed horizontally (parallel) on the substrate, and the solution is flowed over the substrate through these (3) Wiese, R.; Belosludtsev, Y.; Powdrill, T.; Thompson, P.; Hogan, M. Clin. Chem. 2001, 47, 1451–1457. (4) Haab, B. B.; Dunham, M. J.; Brown, P. O. Genome Biol. 2001, 2 (2), 1– 13. (5) Myszka, D. G. Curr. Opin. Biotechnol. 1997, 8, 50–57. (6) Chang-Yen, D. A.; David Myszka, D. G.; Gale, B. K. J. Microelectromech. Syst. 2006, 15, 1145–1151. (7) Thiebaud, P.; Lauer, L.; Knoll, W.; Offenha¨user, A. Biosens. Bioelectron. 2001, 17, 87–93. (8) Papra, A.; Bernard, A.; Juncker, D.; Larsen, N. B.; et al. Langmuir 2001, 17, 4090–4095. (9) Hofmann, O.; Voirin, G.; Niedermann, P.; Manz, A. Anal. Chem. 2002, 74, 5243–5250.
Analytical Chemistry, Vol. 80, No. 22, November 15, 2008
8561
Figure 1. (a) Continuous-flow deposition of a biomolecule through a flow loop with mass transfer and reaction processes at the tip. On the left is the schematic of a 48-spot CFM. (b) CFM on a stage with a vacuum manifold for pumping fluid.
channels.10-14 Each spot cannot be addressed individually in these 2-D devices. The arrays are low density and cross-contamination is an issue. Chiu et al.15 fabricated a 3-D channel network in poly(dimethylsiloxane) (PDMS) for flow devices. This consisted of several layers of planar fluidics with interconnects between the layers. Juncker et al.16 developed a microfluidic probe for flow deposition of arrays and manipulation of cells. These 3-D devices are difficult to fabricate and scale up. The continuous-flow microspotter (CFM), used in this study, produces an array of individually addressable spots using continuous flow and is similar to that used in recent publications.17 It is a 3-D device in that the channels are perpendicular to the surface where deposition occurs. Figure 1a shows the flow and deposition through one channel of the CFM. Although a number of applications of this technology are known, very little is known about the optimum conditions for the operation of this deposition method. Thus, the goal of this work is to optimize the protein flow deposition process by determining the effects of control variables (i.e. flow rate protein concentration, and deposition time) on spot formation and final protein surface concentration. Theory. To optimize the array spots deposited using the CFM, we must understand the effect of the primary control variables (i.e., protein concentration in the solution, flow rate and deposition (10) Delamarche, E.; Bernard, A.; Schmid, H.; Michel, B.; Biebuyck, H. Science 1997, 276, 779–781. (11) Eteshola, E.; Leckband, D. Sens. Actuators, B 2001, 72, 129–133. (12) McDonald, J. C.; Chabinyc, M. L.; Metallo, S. J.; Anderson, J. R.; Stroock, A. D.; Whitesides, G. M. Anal. Chem. 2002, 74, 1537–1545. (13) Kanda, V.; Kariuki, J. K.; Harrison, D. J.; McDermott, M. T. Anal. Chem. 2004, 76, 7257–7262. (14) Kane, R. S.; Takayama, S.; Ostuni, E.; Ingber, D. E.; Whitesides, G. M. Biomaterials 1999, 20, 2363–2376. (15) Chiu, D. T.; Jeon, N. L.; Huang, S.; Kane, R. S.; Wargo, C. J.; Choi, I. S.; Ingber, D. E.; Whitesides, G. M. Proc. Natl. Acad. Sci. U. S. A. 2000, 97 (6), 2408–2413. (16) Juncker, D.; Schmid, H.; Delamarche, E. Nat. Mater. 2005, 4, 622–626. (17) Natarajan, S.; Katsamba, P. S.; Miles, A.; Eckman, J.; Papalia, G. A.; Rich, R. L.; Gale, B. K.; Myszka, D. G. Anal. Biochem. 2008, 73, 141–146.
8562
Analytical Chemistry, Vol. 80, No. 22, November 15, 2008
time) on the mechanism of spot formation on the substrate. In the CFM, microchannel dimensions are ∼150 µm, the Reynolds’ number (Re) is low (
