pubs.acs.org/Langmuir © 2010 American Chemical Society
Quantifying Transcription of Clinically Relevant Immobilized DNA within a Continuous Flow Microfluidic Reactor Stephanie E. McCalla and Anubhav Tripathi* Biomedical Engineering Program, School of Engineering and Medical Sciences, Brown University, Providence, Rhode Island 02912 Received May 6, 2010. Revised Manuscript Received July 19, 2010 Flow-through reactors are commonly used to control and optimize reagent delivery and product removal. Although recent research suggests that transcription reactions using picogram quantities of cDNA produce RNA efficiently in a flow-through microreactor, there has not been a detailed study on the mass transport and reagent dependence of microfluidic transcription reactions. We present a novel microreactor that contains H5 influenza cDNA immobilized directly onto the reactor walls to study the kinetics and reagent dependence of in vitro transcription reactions on a microfluidic platform. Enzyme and the rNTP substrate continuously flow over the cDNA and create RNA, which flows to a downstream collection well. Using nanogram quantities of cDNA, we found that enzyme limiting conditions caused by the concentration of cDNA in a small-volume microreactor channel may be partially overcome as the enzyme binds and concentrates near the channel wall. Kinetics confirm this phenomenon and show that the timescale for enzyme binding can be approximated by tf = cDNA/Q[E]. Surprisingly, on-chip transcription reactions have a strong dependence on the rNTP concentration from 5 to 9 mM despite a low consumption rate of rNTP molecules that is largely independent of the flow rate. Faster flow rates decrease the time it takes to fill DNA promoter sites with enzyme while additionally refreshing rNTP and MgCl2 to allow for a greater consumption of rNTP. These two effects cause reactions with higher concentrations of cDNA in the reactor channel to have a greater dependence on the flow rate. At high flow rates (>0.37 nL/s), the reaction rate begins to drop, likely because of the release and escape of enzyme molecules from the cDNA layer. This critical flow rate can be predicted by a new modified Peclet number, Pem = LcV/D, where Lc is the full length of the tightly packed cDNA molecules, V is the velocity at the DNA/fluid interface, and D is the diffusivity of the enzyme molecule. Together, these insights can inspire reactor designs for a variety of applications.
Introduction Clinical specimens and cell lysates frequently contain small concentrations of RNA contaminated with unwanted genetic material, pathogens, and debri. Samples range from nanograms of RNA1,2 down to the 0.1-1 pg of mRNA found in a single cell.3 These samples must be purified and the RNA must be amplified before detection by microarrays requiring >5 μg of total RNA4 or by fluorescent assays. Furthermore, RNA amplification usually includes a reverse transcription step to convert RNA to an intermediate cDNA molecule. This conversion has an efficiency of ∼12% in bulk reactions and 54% on a microfluidic chip,5 resulting in a cDNA copy number that is significantly lower than the collected RNA. One commonly used RNA amplification method is in vitro transcription (IVT), which uses cDNA obtained from a reverse transcription step as a template to create RNA. IVT is a favored technique for the amplification of rare RNAs because it lacks an exponential bias and preserves the (1) Stirewalt, D. L. Single-stranded linear amplification protocol results in reproducible and reliable microarray data from nanogram amounts of starting RNA. Genomics 2004, 83, 321-331. (2) Van Gelder, R. N. et al. Amplified RNA synthesized from limited quantities of heterogeneous cDNA. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 1663. (3) Phillips, J. Eberwine, J. H. Antisense RNA amplification: a linear amplification method for analyzing the mRNA population from single living cells. Methods 1996, 10, 283-288. (4) Hunter, S. M. L.; Mansergh, F. C.; Evans, M. J. Optimization of minuscule samples for use with cDNA microarrays. J. Biochem. Biophys. Methods 2008, 70, 1048-1058. (5) Zhong, J. F. et al. A microfluidic processor for gene expression profiling of single human embryonic stem cells. Lab Chip 2008, 8, 68-74. (6) Shearstone, J. R. et al. Accurate and precise transcriptional profiles from 50 pg of total RNA or 100 flow-sorted primary lymphocytes. Genomics 2006, 88, 111-121.
14372 DOI: 10.1021/la101826x
relative abundance between transcripts within a sample.3,6,7 It is also used as a component in other RNA amplification techniques such as nucleic acid-sequence-based amplification (NASBA),8 which is used frequently for HIV detection9 and compares favorably with RTPCR for RNA amplification.10,11 During transcription reactions, a DNA-directed RNA polymerase binds to DNA, locates a specific promoter region in the cDNA,12 and after a rate-limiting initiation step13-15 continuously adds magnesium-coupled rNTP’s to the growing chain until (7) Gilbert, I.; Scantland, S.; Dufort, I.; Gordynska, O.; Labbe, A.; Sirard, M.-A.; Robert, C. Real-time monitoring of aRNA production during T7 amplification to prevent the loss of sample representation during microarray hybridization sample preparation. Nucleic Acids Res. 2009, 37, e65. (8) Guatelli, J. C. et al. Isothermal, in vitro amplification of nucleic acids by a multienzyme reaction modeled after retroviral replication. Proc. Natl. Acad. Sci U.S.A. 1990, 87, 1874-1878. (9) Zaaijer, H. L. et al. Detection of HIV-1 RNA in plasma by isothermal amplification (NASBA) irrespective of the stage of HIV-1 infection. J. Virol. Methods 1995, 52, 175-181. (10) Dyer, J. R. et al. Quantitation of human immunodeficiency virus type 1 RNA in cell free seminal plasma: comparison of NASBA with Amplicor reverse transcription-PCR amplification and correlation with quantitative culture. J. Virol. Methods 1996, 60, 161-170. (11) Vandamme, A. M. et al. Detection of HIV-1 RNA in plasma and serum samples using the NASBA amplification system compared to RNA-PCR. J. Virol. Methods 1995, 52, 121-132. (12) Kim, J. H.; Larson, R. G. Single-molecule analysis of 1D diffusion and transcription elongation of T7 RNA polymerase along individual stretched DNA molecules. Nucleic Acids Res. 2007, doi: 10.1093/nar/gkm332. (13) Jia, Y. P. Patel, S. S. Kinetic mechanism of transcription initiation by bacteriophage T7 RNA polymerase. Biochemistry 1997, 36, 4223-4232. (14) Kuzmine, I.; Martin, C. T. Pre-steady-state kinetics of initiation of transcription by T7 RNA polymerase: a new kinetic model. J. Mol. Biol. 2001, 305, 559-566. (15) Sousa, R.; Mukherjee, S. T7 RNA polymerase. Prog. Nucleic Acid Res. Mol. Biol. 2003, 73, 1-41.
Published on Web 08/09/2010
Langmuir 2010, 26(17), 14372–14379
McCalla and Tripathi
it “falls off” the end of the DNA molecule. This reaction is very sensitive to substrate (rNTP) concentration as well as reaction products; transcription produces protons, RNA, and pyrophosphate molecules, which all inhibit further transcription and cause decaying nonlinear reaction kinetics.16 One study using IVT for global mRNA amplification found a reduced transcription efficiency after 1 h of reaction, which significantly decreases total RNA production and appears to skew the relative transcript abundance in microarray studies.7 This has driven research to study reactors that replenish reagents such as rNTP’s and magnesium and that control the pH through the addition of a base.16,17 Reagent delivery and the product removal are particularly important when using nanogram quantities of starting RNA material as found in lysates from cell populations and the second round of transcription of picogram-level RNA samples. Because of their sensitivity, transcription reactions are particularly well suited for a flow-through microfluidic platform. The miniaturized, enclosed format of a flow-through microfluidic reactor carries many advantages over conventional benchtop reactions. Microfluidics can decrease the reaction volume by several orders of magnitude, significantly decreasing reagent use.18 Miniaturization allows efficient heat transfer, which lowers the energy needed for reaction heating and cooling. Fluid reservoirs can hold temperature-sensitive reagents at lower temperatures while the reaction proceeds in the channels. Additionally, the closed format of flow-through reactors prevents chemical and biological contamination, which is particularly important for RNA amplification as ubiquitous RNases can rapidly degrade the RNA product. Microfluidic reactors are also excellent platforms for studying reaction kinetics because reagent delivery can be carefully controlled by the flow rate. These reactors can eventually be integrated into an automated assay, where reactants are purified and amplified and products are detected on a single platform. Several different microfluidic reactors have been developed for the amplification of RNA. Studies of NASBA reactions in microfluidic chamber reactors18-21 found small reaction volumes within enclosed reactors to be advantageous, most likely because of decreased contamination18 and a preconcentration of rare templates.5 Flow-through microfluidic reactors that combine mRNA extraction and reverse transcription on the surfaces of beads show a promising 5-fold increase in the mRNA to cDNA processing efficiency.5,22 These reactors take advantage of controlled reagent delivery to the surfaces of beads in a packed bed, which contain valuable captured RNA or cDNA template molecules. Recently, a flow-through reactor has been created for in vitro transcription reactions of cDNA affixed to beads packed within the microchannel.23 The results showed that microfluidic transcription yielded more RNA than bulk reactions (16) Kern, J. A.; Davis, R. H. Application of solution equilibrium analysis to in vitro RNA transcription. Biotechnol. Prog. 1997, 13, 747-756. (17) Kern, J. A.; Davis, R. H. Application of a fed-batch system to produce RNA by in vitro transcription. Biotechnol. Prog. 1999, 15, 174-184. (18) Gulliksen, A. et al. Parallel nanoliter detection of cancer markers using polymer microchips. Lab Chip 2005, 5, 416-420. (19) Dimov, I. K. et al. Integrated microfluidic tmRNA purification and realtime NASBA device for molecular diagnostics. Lab Chip 2008, 8, 2071-2078. (20) Furuberg, L. et al. RNA amplification chip with parallel microchannels and droplet positioning using capillary valves. Microsyst. Technol. 2008, 14, 673681. (21) Yobas, L. Microfluidic chips for viral RNA extraction & detection. Sensors, 2005 IEEE 2005, doi: 10.1109/ICSENS.2005.1597634 (22) Marcus, J. S.; Anderson, W. F.; Quake, S. R. Microfluidic single-cell mRNA isolation and analysis. Anal. Chem. 2006, 78, 3084-3089. (23) Kralj, J. G. et al. T7-based linear amplification of low concentration mRNA samples using beads and microfluidics for global gene expression measurements. Lab Chip 2009, 9, 917-924.
Langmuir 2010, 26(17), 14372–14379
Article
starting with picogram quantities of mRNA while maintaining cellular expression levels. Although these studies highlight the efficacy of microfluidics for transcription reactions, fundamental studies on the effect of mass transfer and reagent concentrations have not been performed. Nanogram quantities of cDNA are used for expression studies in cellular populations6 and during the second round of IVT in the Eberwine RNA amplification procedure,3 but no transcription studies using nanogram quantities of cDNA have been performed on a microreactor. These fundamental studies would further aid the optimal design and operation of microfluidic transcription devices as well as other enzyme-based microreactors. The microreactor described here was designed to study the sensitivity of the transcription reaction efficiency to parameters such as substrate concentration, enzyme concentration, and flow rate after a cDNA capture step. In this study, cDNA is first captured directly on the reactor channel walls using the streptavidin/biotin linkage. Reagents were then flowed through the reactor channel, and the resulting RNA was collected and analyzed using a calibrated fluorescence signal. The cDNA concentration, cDNA immobilization, and reagent convection all change the sensitivity of the reaction to these parameters.
Materials and Methods Microreactor Fabrication. Microfluidic chip fabrication was performed in cleanroom facilities at Brown University. A mask was created using AutoCAD (Autodesk) and printed (CadArt Services, Poway, CA) on a transparency at 20K dpi. The mask width (w) was 15 μm for all channels except for channels marked R and 4, which were both 150 μm. The mask was transferred to a 1-mm-thick borosilicate glass substrate (S. I. Howard Glass, Worcester, MA) using standard photolithography and fluoride wet etch chemistry.24 This wet etch resulted in 10-μm-deep isotropic channel cross sections as shown in Figure 1a. The reactor channel (marked R, Figure 1a) was 19.8 mm long, giving a total binding area of 7.0 mm2. A second 1-mm-thick borosilicate glass wafer with drilled well holes was bonded to the etched reactor using a controlled thermal bonding procedure. The reactor was then secured to four o-rings within a Teflon caddy using a metal back plate. Pressure Control. Flow rates in the channels were controlled by a LabVIEW programmable control system designed in-house. The control system had four independent pressure ports that were able to apply and detect pressure within the well. Wells W1-W4 on the microfluidic reactor shown in Figure 1a were connected to the pressure control system using air-filled tubing to ensure a rapid response time (
