Electrophoretic Quantitation of Nucleic Acids without Amplification by

Sep 4, 2002 - We have developed a novel high-performance quantitative assay for unamplified nucleic acids that is based on single-molecule imaging. Th...
0 downloads 13 Views 84KB Size
Download PDF
Anal. Chem. 2002, 74, 5033-5038

Electrophoretic Quantitation of Nucleic Acids without Amplification by Single-Molecule Imaging Takashi Anazawa,†,‡ Hiroko Matsunaga,† and Edward S. Yeung†,*

Ames Laboratory-USDOE and Department of Chemistry, Iowa State University, Ames, Iowa 50011, and Central Research Laboratory, Hitachi Ltd., Kokubunji-shi, Tokyo 185-8601, Japan

We have developed a novel high-performance quantitative assay for unamplified nucleic acids that is based on singlemolecule imaging. The apparatus is a simple but highly sensitive single-molecule detection system that uses a normal CCD camera instead of an image-intensified CCD camera. After the DNA molecules in a sample were labeled with YOYO-1, they were induced to migrate electrophoretically in a polymer solution and imaged. No chemical or biochemical amplification was required. Direct quantitation of the sample by counting molecules was possible, because the number counted over the measurement period was directly proportional to the concentration of DNA molecules in the sample. Nonspecifically labeled impurities that would degrade the sensitivity of the assay were successfully reduced and discriminated from the DNA molecules by differences in electrophoretic mobility. By using β-actin DNA (838 bp) as a model sample, we demonstrate that this protocol was fast (10-min measurement period), highly sensitive (limit of quantitation: ∼103 copies/sample, or 3 × 10-16 M), quantitative, and covered a wide linear dynamic range (∼104). This high-performance assay promises to be a powerful technology for the quantitation of specific varieties of mRNA in the study of gene functions and diseases and in the clinical detection of mutant cells. Assays to measure quantities of nucleic acids are central to the fields of biology, medicine, and pharmaceuticals. Measuring the abundance of specific mRNA in cells and tissues is particularly important. To be applicable to the study of gene function, disease, and clinical diagnosis, these assays must be highly sensitive and highly quantitative and have a wide dynamic range. A number of technologies for quantifying gene expression have been developed.1,2 Northern blot is the oldest and most widely used of these.3 Though it provides information on both the size and abundance of a target mRNA, its sensitivity is quite low (limit of quantitation: ∼107 copies/sample).1 Real-time reverse transcription polymerase chain reaction (RT-PCR) is the most recently developed and powerful technique.4-8 It offers the highest degree of sensitiv†

Iowa State University. Hitachi Ltd.. (1) Sambrook, J.; Russell, D. W. Molecular Cloning: A Laboratory Manual, 3rd ed.;, Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 2001. (2) Reue, K. J. J. Nutr. 1998, 128, 2038-2044. (3) Alwine, J. C.; Kemp, D. J.; Stark, G. R. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 5350-5354. ‡

10.1021/ac025801u CCC: $22.00 Published on Web 09/04/2002

© 2002 American Chemical Society

ity (limit of quantitation: ∼102 copies/sample) and the widest dynamic range (5 orders of magnitude);1,8 however, it does not avoid the difficulties that originate from the inherent nature of PCR. Quantitation by exponential amplification has poor precision and may be dramatically affected by small differences in the efficiency of amplification from sample to sample and from thermal cycler to thermal cycler. Furthermore, optimizing the complicated probes and thermal cycle condition for each specific mRNA is a formidable task. Ribonuclease protection assay is also popular as a way of getting quantitative data on a specific mRNA.9-12 The procedure consists of the following three steps: First, total RNA is extracted from cells or tissues. An antisense RNA probe is added to this extract for hybridization with the target mRNA. Next, ribonuclease is added to digest the remaining nonhybridized RNA and probe. Hybridization protects the target mRNA through this process of digestion. Finally, this protected double-stranded RNA is separated by gel electrophoresis and quantified by autoradiography. This assay is extremely accurate, because the signal is derived from the amount of unamplified target mRNA. However, it is not particularly sensitive (limit of quantitation: ∼105 copies/sample).1 Furthermore, the third step of the procedure is time-consuming as well as labor-intensive. If the third step were to be replaced by a simpler and more sensitive technique, the modified ribonuclease protection assay would be ideal for the quantitation of specific varieties of mRNA. Single-molecule-detection (SMD) technology has made remarkable progress in this decade.13-22 This technology has mainly (4) Higuchi, R.; Fockler, C.; Dollinger, G.; Watson, R. Bio/Technology 1993, 11, 1026-1030. (5) Wittwer, C. T.; Herrmann, M. G.; Moss, A. A.; Rasmussen, R. P. BioTechniques 1997, 22, 130-138. (6) Heid, C. A.; Stevens, J.; Livak, K. J.; Williams, P. M. Genome Res. 1996, 6, 986-994. (7) Gibson, U. E.; Heid, C. A.; Williams, P. M. Genome Res. 1996, 6, 9951001. (8) Bustin, S. A. J. Mol. Endocrinol. 2000, 25, 169-193. (9) Melton, D. A.; Krieg, P. A.; Rebagliati, M. R.; Maniatis, T.; Zinn, K.; Green, M. R. Nucleic Acids Res. 1984, 12, 7035-7056. (10) Pape, M. E.; Melchior, G. W.; Marotti, K. R. Genet. Anal. Technol. Appl. 1991, 8, 206-213. (11) Davis, M. J.; Bailey, C. S.; Smith II, C. K. BioTechniques 1997, 23, 280285. (12) Mitchell, A.; Fidge, N. Methods Enzymol. 1996, 263, 351-363. (13) Moerner, W. E.; Orrit, M. Science 1999, 283, 1670-1676. (14) Weiss, S. Science 1999, 283, 1676-1683. (15) Mehta, A. D.; Rief, M.; Spudich, J. A.; Smith, D. A.; Simmons, R. M. Science 1999, 283, 1689-1695. (16) Weiss, S. Nat. Struct. Biol. 2000, 7, 724-729.

Analytical Chemistry, Vol. 74, No. 19, October 1, 2002 5033

been applied to studying mechanisms of the dynamic behavior of biomolecules, such as motor proteins.16-18 In addition to being applicable to fundamental investigations in science, this technology can also be used in practical applications. Though the sizing23,24 and immunoassay25,26 of biomolecules at the single-molecule level have recently been demonstrated, a method of quantitation that operates at this level has not, thus far, been realized. That is, although we are able to observe a single molecule in a sample, it is difficult to measure its concentration from such observations. However, it should be possible to apply SMD in a highly sensitive quantitative assay that requires no amplification, for example, counting the number of protected double-stranded RNA in the ribonuclease protection assay. Most existing SMD systems are large, complicated, and require a carefully controlled laboratory environment, because image-intensified CCD cameras are commonly used in these systems.19-21,25,26 The broader practical application of SMD technology will require simple and inexpensive instrumentation. In this paper, we describe the development of a simple SMD system in which a normal CCD camera is used instead of an imageintensified CCD camera. We also describe the novel application of this system to the quantitative assay of nucleic acids. We demonstrate the use of single-molecule imaging for the absolute quantitation of β-actin DNA (838 bp) as a model for the protected double-stranded RNA assay. The high speed, high sensitivity, wide dynamic range, and high degree of quantitative reliability of the technique are also presented. EXPERIMENTAL SECTION Preparation of β-Actin DNA. Gene-specific primers for human β-actin (sense, 5′-ATCTGGCACCACACCTTCTACAATGAGCTGCG-3′; antisense, 5′-CGTCATACTCCTG CTTGCTGATCCACATCTGC-3′) were obtained from BD Biosciences Clontech (Palo Alto, CA). Double-stranded β-actin DNA (838 bp) was amplified by PCR. PCR was performed in a 50-µL reaction mixture containing 2 µL of control human cDNA (50 amol/µL, BD Biosciences Clontech), 1× HotStarTaq DNA polymerase buffer (QIAGEN, Valencia, CA), dNTPs at 200 µM, each β-actin specific primer at 0.4 µM, and 1.25 units of HotStarTaq DNA polymerase (QIAGEN). PCR proceeded under the following conditions: 1 cycle of initial heating to 95 °C for 15 min and 35 cycles at 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 1 min, followed by one cycle of 72 °C for 10 min. The 838-bp product was verified by electrophoresis in comparison with the standard size-markers on a 2% SeaKem Gold agarose gel (BioWhittaker Molecular Applications, Rockland, ME) that had been stained with 0.5-µg/mL ethidium bromide. The QIAquick PCR purification kit (QIGEN) (17) Ishijima, A.; Yanagida, T. Trends Biochem. Sci. 2001, 26, 438-444. (18) Ishii, Y.; Ishijima, A.; Yanagida, T. Trends Biotechnol. 2001, 19, 211-216. (19) Ha, T. Curr. Opin. Struct. Biol. 2001, 11, 287-292. (20) Xu, X.; Yeung, E. S. Science 1997, 276, 1106-1109. (21) Xu, X.-H.; Yeung, E. S. Science 1998, 281, 1650-1653. (22) Kang, S. H.; Shortreed, M. R.; Yeung, E. S. Anal. Chem. 2001, 73, 10911099. (23) Van Orden, A.; Keller, R. A.; Ambrose, W. P. Anal. Chem. 2000, 72, 3741. (24) Shortreed, M. R.; Li, H.; Huang, W.-H.; Yeung, E. S. Anal. Chem. 2000, 72, 2879-2885. (25) Ma, Y.; Shortreed, M. R.; Yeung, E. S. Anal. Chem. 2000, 72, 4640-4645. (26) Ma, Y.; Shortreed, M. R.; Li, H.; Huang, W.; Yeung, E. S. Electrophoresis 2001, 22, 421-426.

5034 Analytical Chemistry, Vol. 74, No. 19, October 1, 2002

Figure 1. Optical arrangement used in single-molecule detection: L, laser source (Ar-ion laser, 488 nm); S, laser shutter; L1, cylindrical lens 1 (f1, 150 mm); L2, cylindrical lens 2 (f2, 25 mm); C, square capillary (360-µm o.d. and 75-µm i.d.); O, microscope objective (20×, 0.75 NA); F, 488-nm holographic notch filter; and CCD, CCD camera.

was used to purify the product according to the manufacture’s instructions; this was followed by ethanol precipitation. The purified product was dissolved in an appropriate volume of TE buffer (10-mM Tris pH 8.0 and 1-mM ethylenediaminetetraacetic acid), and a commercial UV spectrometer was used to measure its concentration. The final concentration of β-actin DNA was adjusted to 10-7 M as a stock solution for use in the next step. Preparation of YOYO-1-Labeled β-Actin DNA. A solution of 10-mM Tris buffer (pH 7.8) was prepared by diluting a 1-M Tris buffer solution (pH 8.0) (Ambion, Inc., Austin, TX) with ultrapure water. This buffer solution was used in preparing all of the samples and solutions. An increasingly more dilute series of solutions was obtained from the β-actin DNA stock solution. These solutions were then labeled with YOYO-1 intercalator dye (Molecular Probes, Eugene, OR) at a fixed concentration. The resulting labeling-reaction mixtures contained β-actin DNA at 0, 4 × 10-15, 4 × 10-14, 4 × 10-13, and 4 × 10-12 M, respectively, along with YOYO-1 at 10-8 M and 10 mM Tris buffer. Each reaction proceeded for 5 min at room temperature, then 2.5 µL of the reactant was mixed with 7.5 µL of a polymer mixture (0.4% (w/v) 600 000 Mr poly(ethylene oxide) (PEO) plus 0.4% (w/v) 1 000 000 Mr poly(vinyl pyrrolidone) (PVP) in 10 mM Tris buffer). The respective final compositions of the 10-µL sample solutions were β-actin DNA at 0, 10-15, 10-14, 10-13, and 10-12 M, along with YOYO-1 at 2.5 × 10-9 M, 0.3% PEO, 0.3% PVP, and 10 mM Tris buffer. Capillary Electrophoresis. A fused-silica square capillary (with sides of 360 µm (outer) and 75 µm (inner)) was obtained from Polymicro Technologies, Inc. (Phoenix, AZ). A 30-cm-long capillary was used in all of the experiments. A 1-cm window was cleared at the center of the capillary. The sample solution was aspirated into the capillary by a syringe. After the capillary had been filled with the sample solution, both ends of the capillary were separately immersed in two polymer solutions (0.3% PEO plus 0.3% PVP in 10 mM Tris buffer). Levels of both polymer solutions were exactly equalized so that there would be no hydrodynamic flow of the sample solution in the capillary. An electric field of 36.3 V/cm (1090 V/30 cm) was then applied across the capillary. Single-Molecule Detection System. The experimental setup is shown in Figure 1. The laser source was an argon-ion laser producing emission at 488 nm (Innova-90, Coherent, Santa Clara, CA). Extraneous light and plasma lines from the laser were eliminated with the aid of an equilateral prism and an optical pinhole (not shown in Figure 1). The laser beam was shaped into a planar sheet by two cylindrical lenses (Figure 2). The optical

Figure 2. Schematic diagram of laser-beam irradiation of the sample and the capillary.

axes of the first cylindrical lens (with focal length f1 of 150 mm) and second cylindrical lens (f2, 25 mm) were parallel and perpendicular to the capillary, respectively. The capillary window was irradiated by the laser beam when the laser shutter (Uniblitz LS2Z2, Vincent Associates, Rochester, NY) was opened. The total laser power at the capillary was 14 mW. Figure 2 is a schematic view of the irradiated region in the capillary: it is ∼75 µm wide (x), 75 µm long (y), and ∼10 µm thick (z). The distributions of laser power in the x and z directions are Gaussian. The axis of the laser beam was perpendicular to one side of the square capillary. Images of the irradiated region through a Zeiss 20×/0.75-NA Plan Apochromat microscope objective were recorded by a cooled CCD camera (NTE/CCD-512-EBFT, Roper Scientific, Trenton, NJ). The detector element (camera) was kept at -35 °C. A 488-nm holographic notch filter (Kaiser Optical System, Ann Arbor MI) with an optical density of >6 was placed between the objective and the CCD camera. The digitization rate of the CCD camera was 1 MHz (16 bits). The frame-transfer CCD camera was operated in the external synchronization mode. Exposure timing for the CCD camera and laser shutter was synchronized by a shutter driver/timer (Uniblitz ST132, Vincent Associates, Rochester, NY). Image Acquisition and Analysis. The CCD exposure frequency was 5 Hz (0.2 s/frame), and the exposure time for each frame was 10 ms. A sequence of 3 000 consecutive frames were acquired for each sample via WinView/32 software (Roper Scientific, Trenton, NJ). This corresponds to 10 min of measurement time/sample. Each frame consists of 200 × 200 square pixels. Each pixel represents 0.5 × 0.5 µm of real space. All frames were analyzed off-line. The individual molecules in each frame were picked up by homemade software. The velocity of migration of each molecule was then manually determined by calculating the distances migrated within each frame interval of 0.2 s, in pixels. RESULTS AND DISCUSSION Single-Molecule Quantitation of DNA. Quantitation is quite different from detection. If we are able to detect single molecules of a target DNA within a sample solution, we should be able to theoretically detect them at an infinitely low concentration. However, this does not mean that we are able to measure the quantity of the target DNA in a given sample, that is, measure its

concentration. Our basic idea for quantitation was to count the number of target DNA molecules in a unit volume of the sample solution and determine its density. A detection efficiency of 100% must be assured in order to calculate the true concentration, since standards are not available at these low concentrations. In addition, we had to overcome two major problems before we were able to carry this out. The first problem was labeling DNA with YOYO-1. In conventional single-molecule detection, DNA at a high concentration (e.g., ∼10-10 M) is labeled with YOYO-1 at an appropriate ratio (e.g., 1 dye molecule/5 bp) for a high reaction efficiency.27,28 The labeled DNA is then diluted to an adequately low concentration (e.g., ∼10-13 M) for single-molecule detection.23-26,29 In quantitative analysis, however, DNA must be efficiently labeled with YOYO-1 at a constant concentration, because the concentration of DNA is unknown and low. At the same time, the concentration of YOYO-1 must be minimized so that the background intensity from free YOYO-1 is as low as possible. We found by experiment that YOYO-1 at g 10-8 M was necessary to adequately label β-actin DNA (838 bp) at e 10-12 M. This was also confirmed by applying the law of mass action to describe the binding equilibrium between YOYO-1 and its binding sites on β-actin DNA. When YOYO-1 at 10-9, 10-8, or 10-7 M, respectively, reacts with β-actin DNA at 10-16-10-12 M, 37, 86, or 98% of the binding sites on the β-actin DNA are estimated to be saturated by YOYO-1. This is because the binding affinity is 6.0 × 108 M-1 and the abundance of binding site per dye molecule is ∼5 bp.27-29 On the basis of the above experiments and estimates, we decided to have the reaction proceed with 10-8 M YOYO-1. Almost all of the YOYO-1 dye molecules, thus, remain free (∼10-8 M) after the reaction has reached equilibrium. This is because the above concentration of YOYO-1 is much higher than is required for saturation coverage of the β-actin DNA in our samples. Uniform and sufficient labeling is expected for β-actin DNA at any of the concentration values we used (10-16-10-12 M). This means that the number of labeled β-actin DNA molecules in a unit volume of any of the sample solutions should be linearly proportional to the concentration of β-actin DNA. Measuring the quantity of β-actin DNA is then possible as long as we have a way of counting the numbers of molecules by single-molecule detection. The second problem is the reduction of the number of fluorescent particles other than DNA. Although YOYO-1 is supposed to be essentially nonfluorescent in the absence of nucleic acids,27 we observed many fluorescent particles when we applied the SMD system to buffer solutions containing YOYO-1 but not DNA. No fluorescent particles were detected in buffer solutions without YOYO-1, but the number of such particles increased with the concentration of YOYO-1. This is probably because the positively charged YOYO-1 becomes associated with some impurities that carry a negative charge, such as dust and educts in the buffer solution. Such impurities present a sizable background in quantitation by single-molecule detection and limit the sensitivity of such measurements. We were unable to remove them by (27) Rye, H. S.; Yue, S.; Wemmer, D. E.; Quesada, M. A.; Haugland, R. P.; Mathies, R. A.; Glazer, A. N. Nucleic Acids Res. 1992, 20, 2803-2812. (28) Marino, M. A.; Devaney, J. M.; Davis, P. A.; Smith, J. K.; Girard, J. E. Anal. Chem. 1998, 70, 4514-4519. (29) Gurrieri, S.; Wells, K. S.; Johnson, I. D.; Bustamante, C. Anal. Biochem. 1997, 249, 44-53.

Analytical Chemistry, Vol. 74, No. 19, October 1, 2002

5035

Figure 3. Sequence of eight consecutive images (A to H) of the electrophoretic migration of five molecules of β-actin DNA. Each molecule is marked by an arrow with a number (1 to 5). An electric field of 36.3 V/cm was applied in the vertical direction, which caused β-actin DNA molecules to migrate in that direction (downward). The sample solution contained 10-13 M β-actin DNA. The exposure time and frequency of the CCD camera were 10 ms and 5 Hz, respectively. This series represents 1.4 s of real time. Each image consists of 200 × 200 pixels, that is, 100 × 100 µm of real space.

filtering the buffer solution and the YOYO-1 stock solution through 0.22-µm filters. However, we serendipitously discovered that the number of labeled impurity particles was greatly reduced by adding PEO, while the number and the intensity of the labeled β-actin DNA remained unchanged. This can be explained by the hypothesis that the affinity between YOYO-1 and the PEO polymer is stronger than that between YOYO-1 and the impurities but weaker than that between YOYO-1 and DNA. Molecules of YOYO-1 dye that are associated with the PEO polymer are not detected, because they are uniformly dispersed throughout the solution rather than concentrated on each DNA molecule. Furthermore, we found that the difference in mobility during electrophoresis in the 0.3% PEO polymer solution allowed us to discriminate the labeled impurities from the labeled β-actin DNA.24-26 The sample solution also contained 0.3% PVP polymer to suppress electroosmotic flow (EOF).26,30 Therefore, we are able to selectively count only the β-actin DNA molecules. Electrophoresis Imaging of β-Actin DNA Molecules. The multiframe method was used to observe migration of YOYO-1labeled β-actin DNA molecules during electrophoresis (Figure 3).24 Each frame is a snapshot (10-ms exposure time) of the single molecules within the field of view (100 × 100 µm). Each spot corresponds to a single molecule. We are able to easily track and characterize individual molecules through several consecutive frames. For example, the migration velocity vm of molecule #1 was calculated as vm ) 36.3 pixels/frame, corresponding to a mobility of 2.5 × 10-4 cm2 V-1 s-1. This is reasonable and agrees (30) Gao, Q.; Yeung, E. S. Anal. Chem. 1998, 70, 1382-1388.

5036 Analytical Chemistry, Vol. 74, No. 19, October 1, 2002

with the values determined from standard capillary electrophoresis.24 The migration and diffusion distances of the molecules within a single exposure of the CCD camera were only 1.8 and 10-12 M, a thinner flow channel

can be used to avoid overlap among different DNA molecules in the images. Figure 5 shows the standard curve for the quantitation of β-actin DNA. The number of molecules counted (the vertical axis) is the filtered number, N3, in Table 1. The fitted straight line in Figure 5 has a slope of exactly 1. A perfectly linear relation between β-actin DNA concentration and number of molecules counted was obtained over a range of nearly 4 orders of magnitude. The number of molecules counted is thus directly proportional to the concentration or amount of the target DNA. This wide linear dynamic range is accomplished by making use of the merits of single-molecule counting in quantitation and is not possible to obtain by quantitation that is simply based on the overall fluorescence intensity from the sample. The latter approach is limited by inhomogeneities in labeling and in laser intensity distribution. This standard curve was very reproducible (see the error bars in Figure 5). The results indicate that this DNA assay is accurate and reliable. Furthermore, it takes only 5 min to prepare the sample and 10 min to carry out the measurement for the sample. So, an assay based on this technique is much faster than other methods of assaying DNA, including real-time PCR. Each of the molecules of β-actin DNA counted as N2 in Table 1 appeared, on average, in 2.7 frames. For example, when the sample contains β-actin DNA at 10-14 M, N2 ) 75 corresponds to ∼203 spots when N1 ) 269. The number of counted molecules in the 10-14 M-β-actin DNA sample that appeared in at least one frame is thus ∼141 ( ) 75 + 269-203). This number can be compared with the theoretical number in the following way. The average speed of migration for the β-actin DNA molecules was ∼35 pixels/frame (Figure 4C). The distance of migration over the period of measurement (3000 frames) is estimated as ∼52.5 mm. Since the cross-sectional area within the capillary is 5.6 × 10-3 mm2, the volume of the sample that passes through the detection position of the capillary is ∼0.29 µL. On the other hand, the cross-sectional area of the irradiated region is ∼7.5 × 10-4 mm2 (Figure 2). Therefore, the volume of the sample that intersects the irradiated region is ∼39 nL. The theoretical number of molecules in a 10-14 M β-actin DNA sample that should appear in the detected images is ∼234. The above counted number Analytical Chemistry, Vol. 74, No. 19, October 1, 2002

5037

(∼141) is within a factor of 2 of this theoretical number (∼234). This means that the protocol for quantitative assay, as well as the detection system, is well-behaved. Quantitation Sensitivity. The number of molecules measured for the 0 M β-actin DNA sample in Table 1 represent YOYO-1labeled impurities. The numbers of β-actin DNA molecules and of impurities are the signal and the background of the quantitative assay, respectively. We used N3 in this assay because the signalto-background ratio (S/B) was improved by the various steps used to select the number of molecules (from N1 to N2 and from N2 to N3). Although the background level was considerably reduced by the use of the PEO polymer, as was previously discussed, the residual background count N3 ) 2 ( 1. This determines the sensitivity of the assay. The limit of quantifiable concentration of β-actin DNA is estimated to be 3 × 10-16 M (S/N ) 2), since the noise equals the standard deviation of the background () 1), and the signal of the 10-15 M sample was 11. This corresponds to a limit on the quantifiable amount of 3 × 10-21 mol, or 1800 molecules of β-actin DNA, because a 10-µL sample volume was used in this study. This sensitivity is the highest of any method for the quantitative assay of nucleic acids in which amplification is not used. Rye et al. have reported the quantitative assay of YOYO-1labeled nucleic acids by a fluorometric system with a laser-excited confocal fluorescence scanner.31 The limit of the quantifiable amount for that system was 20 pg of double-stranded DNA, which corresponds to 4 × 10-17 mol, or 2 × 107 molecules of β-actin DNA (838 bp). The sample volume was 25 µL in that assay, so the limit of quantifiable concentration is 2 × 10-12 M of β-actin DNA. The sensitivity in quantitation of our technique is thus 4 orders of magnitude better than that of their method. As was previously discussed, the sample volumes that were actually used during the measurement period in this assay were as little as ∼0.29 µL. Therefore, it should be possible to further reduce the limit of quantifiable amount to fewer than 100 molecules of β-actin DNA by combining this technique with nanoliter-scale sample-preparation technology, that is, by preparing 0.5-µL samples. The sensitivity of this approach will then be comparable to that of the real-time PCR assay, but at much higher speeds and without the problems of infidelity in PCR.1,8 By using a camera with a higher frame rate, a higher applied voltage can be employed to speed up the measurement process. (31) Rye, H. S.; Dabora, J. M.; Quesada, M. A.; Mathies, R. A.; Glazer, A. N. Anal. Chem. 1993, 208, 144-150.

5038

Analytical Chemistry, Vol. 74, No. 19, October 1, 2002

CONCLUSIONS We have developed a technique for the quantitative assay of nucleic acids that is fast (reaction and measurement time: ∼15 min), highly sensitive (limit of quantitation: ∼103 copies/sample), highly accurate (the count is directly proportional to the amount of a sample), and has a wide linear dynamic range (∼104). To facilitate practical application, we have also developed a simple SMD system for this assay. It is highly sensitive (detection limit: ∼120 bp), though this system uses a normal CCD camera instead of a sophisticated image-intensified CCD camera. Indeed, any scientific CCD camera should suffice. This substitution avoids problems with room light and room humidity. If desired, the laser beam can be expanded to cover the entire flow channel to produce an absolute molecule count. In that case, the depth of field of the optics must be compatible with the channel depth. Since detection is limited by impurities and not stray light, beam expansion should not degrade the performance as long as the laser power is increased proportionately. Because electrophoresis is used, this assay can be applied to mixtures of DNA24-26 or to immunocomplexes.26 In this study, β-actin DNA (838 bp) was quantitated as a model for a protected double-stranded RNA in ribonuclease protection assay. YOYO-1 is applicable to the labeling of RNA as well as of DNA, so it is expected that protected double-stranded RNA will be quantifiable following the same protocol. Combining the techniques developed in this study with ribonuclease protection assay will give us a protocol for the quantitative assay of specific varieties of mRNA that is fast, highly sensitive, and highly accurate and has a wide dynamic range. This promises to be a powerful approach for the study of gene functions and diseases and in clinical diagnosis. ACKNOWLEDGMENT We thank Gang Xue for the optimization of the image analysis software for this study. The Ames Laboratory is operated for the U.S. Department of Energy by Iowa State University under Contract no. W-7405-Eng-82. This work was supported by the Director of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, and Office of Biological and Environmental Research, and by the National Institutes of Health.

Received for review May 24, 2002. Accepted July 8, 2002. AC025801U