Anal. Chem. 1999, 71, 2108-2116
Efficient Detection of Single DNA Fragments in Flowing Sample Streams by Two-Photon Fluorescence Excitation Alan Van Orden,† Hong Cai, Peter M. Goodwin, and Richard A. Keller*
Chemical Science and Technology Division, MS M888, Los Alamos National Laboratory, Los Alamos, New Mexico 87545
This paper reports the demonstration of efficient single molecule detection in flow cytometry by two-photon fluorescence excitation. We have used two-photon excitation (TPE) to detect single DNA fragments as small as 383 base pairs (bp) labeled with the intercalating dye, POPO1, at a dye:nucleotide ratio of 1:5. TPE of the dye-DNA complexes was accomplished using a mode-locked, 120 fs pulse width Ti:sapphire laser operating at 810 nm. POPO-1 labeled DNA fragments of 1.1 kilobase pairs (kbp) and larger were sequentially detected in our flow cytometry system with a detection efficiency of nearly 100%. The detection efficiency for the 383 bp DNA fragments was approximately 75%. We also demonstrate the ability to distinguish between different sized DNA fragments in a mixture by their individual fluorescence burst sizes by TPE. These studies indicate that using TPE for single molecule flow cytometry experiments lowers the intensity of the background radiation by approximately an order of magnitude compared to one-photon excitation, due to the large separation between the excitation and emission wavelengths in TPE. Fluorescence detection of individual molecules dissolved in solution dates back to the work of Hirschfeld in 1976.1 Hirschfeld was able to detect bursts of photons from individual biomolecules that had been labeled with 80-100 fluorescein molecules each as they diffused through an excitation laser beam. In more recent years, a variety of experimental techniques have been developed to detect individual fluorophores in solution.2-6 These achievements have opened up many new avenues in the analytical biosciences. At present, there are several approaches that are commonly applied to single molecule fluorescence detection in solution. In one approach, confocal fluorescence microscopy is † Permanent address: Department of Chemistry, Colorado State University, Fort Collins, CO 80523. (1) Hirschfeld, T. Appl. Opt. 1976, 15, 2965-2966. (2) Shera, E. B.; Seitzinger, N. K.; Davis, L. M.; Keller, R. A.; Soper, S. A. Chem. Phys. Lett. 1990, 174, 553-557. (3) Barnes, M. D.; Whitten, W. B.; Ramsey, J. M. Anal. Chem. 1995, 67, 7, A418-A423. (4) Keller, R. A.; Ambrose, W. P.; Goodwin, P. M.; Jett, J. H.; Martin, J. C.; Wu, M. Appl. Spectrosc. 1996, 50, 12A-32A. (5) Goodwin, P. M.; Ambrose, W. P.; Keller, R. A. Acc. Chem. Res. 1996, 29, 607-613. (6) Nie, S.; Zare, R. N. Annual Reviews of Biophysics and Biomolecular Structure; Stroud, R. M., Ed.; Annual Reviews, Inc.: Palo Alto, CA, 1997; Vol. 26.
2108 Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
used to detect single molecules that diffuse through the near diffraction limited focal region of a laser beam.7-10 The focal region of the laser beam defines a ∼1 fL subvolume within the solution, from which fluorescence is monitored as a function of time. Background radiation caused by Rayleigh and Raman scattering of the laser beam by the solvent is minimized due to the small number of solvent molecules present in such a small volume, so that fluorescence bursts emitted by single molecules that diffuse into the detection volume can easily be observed above the background. Our approach to efficient single molecule detection in solution, referred to as single molecule flow cytometry, is to enlarge the focal region of the laser beam so that the detection volume is 1-5 pL. Under these conditions, the analyte solution can be made to flow through the focus of the laser beam in such a way that each molecule delivered into the flow stream is detected sequentially with high efficiency.4,5,11 This is accomplished by hydrodynamically focusing the analyte solution into a narrow sample stream within a sheath flow cuvette and then intersecting the sample stream with a focused excitation laser beam. We have reported the detection of analyte molecules delivered into the flow stream with efficiencies up to 95% or more using this technique.4,5,11-13 Other groups have also reported efficient detection (80-98% detection efficiencies) of single molecules delivered into sheath streams,14 microcapillaries,15,16 microdroplet streams,17 and microstructured devices.18 In single molecule flow cytometry, each (7) Rigler, R.; Widengren, J.; Mets, U ¨ . Fluorescence Spectroscopy; Wolfbeis, O. S., Ed.; Springer-Verlag: Berlin, 1993. (8) Mets, U ¨ .; Rigler, R. J. Fluoresc. 1994, 4, 259-264. (9) Nie, S.; Chiu, D. T.; Zare, R. N. Science 1994, 266, 1018-1021. (10) Nie, S.; Chiu, D. T.; Zare, R. N. Anal. Chem. 1995, 67, 2849-2857. (11) Goodwin, P. M.; Wilkerson, C. W.; Ambrose, W. P.; Keller, R. A. Proc. SPIEInt. Soc. Opt. Eng. 1993, 1895, 79-89. (12) Goodwin, P. M.; Affleck, R. L.; Ambrose, W. P.; Jett, J. H.; Johnson, M. E.; Martin, J. C.; Petty, J. T.; Schecker, J. A.; Wu, M.; Keller, R. A. Computer Assisted Analytical Spectroscopy; Brown, S. D., Ed.; John Wiley and Sons: New York, 1996; pp 61-80. (13) Van Orden, A.; Machara, N. P.; Goodwin, P. M.; Keller, R. A. Anal. Chem. 1998, 70, 1444-1451. (14) Li, L. Q.; Davis, L. M. Appl. Opt. 1995, 34, 3208-3217. (15) Guenard, R. D.; King, L. A.; Smith, B. W.; Winefordner, J. D. Anal. Chem. 1997, 69, 2426-2433. (16) Zander, C.; Drexhage, K. H.; Han, K. T.; Wolfrum, J.; Sauer, M. Chem. Phys. Lett. 1998, 286, 457-465. (17) Lermer, N.; Barnes, M. D.; Kung, C. Y.; Whitten, W. B.; Ramsey, J. M. Anal. Chem. 1997, 69, 2115-2121. (18) Do¨rre, K.; Brakmann, S.; Brinkmeier, M.; Han, K.-T.; Riebeseel, K.; Schwille, P.; Stephan, J.; Wetzel, T.; Lapczyna, M.; Stuke, M.; Bader, R.; Hinz, M.; Seliger, H.; Holm, J.; Eigen, M.; Rigler, R. Bioimaging 1997, 5, 139-152. 10.1021/ac9811221 CCC: $18.00
© 1999 American Chemical Society Published on Web 04/24/1999
molecule flows through the apparatus at the same rate and experiences the same radiation field during its transit time through the detection volume, making it possible to identify the molecules on the basis of their fluorescence burst sizes.13 The ability to sequentially detect single molecules in flow streams with high efficiency represents an important new direction in analytical chemistry. For example, techniques for DNA fragment sizing,19-22 DNA sequencing,23-25 studies of single molecule photochemistry,26 hybridization analysis,27 single molecule identification,13,16,28 and single molecule counting15,16 that are based on efficient single molecule detection in flowing solutions hold great promise to significantly enhance the speed and sensitivity of these assays. A key disadvantage to the flow cytometry technique is that the background count rate from a ∼1-5 pL detection volume can be significantly larger than the fluorescence count rate from a single chromophore during its transit time through the detection volume. To suppress this background, we have used pulsed laser excitation and time gating to separate the prompt scattered luminescence from the delayed fluorescence.11,12 While this approach has proven to be extremely effective for single molecule detection of chromophores that possess fluorescence lifetimes longer than ∼2 ns, a significant reduction in the detection efficiency is expected for molecules with fluorescence lifetimes that are shorter than this. This is due to the finite width of the instrument response function for the time-correlated single photon counting measurements. Single molecules that possess a number of chromophores, such as the protein Β-phycoerythrin and DNAintercalating dye complexes are sufficiently fluorescent that they can often be detected using flow cytometry with CW laser excitation.19-22,26 There have been a number of recent reports of single molecule fluorescence detection by two-photon excitation (TPE). 29-34 Single molecule detection by TPE is facilitated by the use of high peak (19) Castro, A.; Fairdield, F. R.; Shera, E. B. Anal. Chem. 1993, 65, 849-852. (20) Goodwin, P. M.; Johnson, M. E.; Martin, J. C.; Ambrose, W. P.; Marrone, B. L.; Jett, J. H.; Keller, R. A. Nucleic Acids Res. 1993, 21, 803-806. (21) Petty, J. T.; Johnson, M. E.; Goodwin, P. M.; Martin, J. C.; Jett, J. H.; Keller, R. A. Anal. Chem. 1995, 67, 1755-1761. (22) Huang, Z. P.; Petty, J. T.; O’Quinn, B.; Longmire, J. L.; Brown, N. C.; Jett, J. H.; Keller, R. A. Nucleic Acids Res. 1996, 24, 4202-4209. (23) Ambrose, W. P.; Goodwin, P. M.; Jett, J. H.; Johnson, M. E.; Martin, J. C.; Marrone, B. L.; Schecker, J. A.; Wilkerson, C. W.; Keller, R. A.; Haces, A.; Shih, P. J.; Harding, J. D. Ber. Bunsen-Ges. Phys. Chem. 1993, 97, 15351542. (24) Goodwin, P. M.; Affleck, R. L.; Ambrose, W. P.; Demas, J. N.; Jett, J. H.; Martin, J. C.; Reha-Krantz, L. J.; Semin, D. J.; Schecker, J. A.; Wu, M.; Keller, R. A. Exp. Tech. Phys. 1995, 41, 279-294. (25) Goodwin, P. M.; Cai, H.; Jett, J. H.; Ishaug-Riley, S. L.; Machara, N. P.; Semin, D. J.; Van Orden, A.; Keller, R. A. Nucleosides Nucleotides 1997, 16, 543550. (26) Wu, M.; Goodwin, P. M.; Ambrose, W. P.; Keller, R. A. J. Phys. Chem. 1996, 100, 17406-17409. (27) Castro, A.; Williams, J. G. K. Anal. Chem. 1997, 69, 3915-3920. (28) Enderlein, J.; Goodwin, P. M.; Van Orden, A.; Ambrose, W. P.; Erdmann, R.; Keller, R. A. Chem. Phys. Lett. 1997, 270, 464-470. (29) Mertz, J.; Xu, C.; Webb, W. W. Opt. Lett. 1995, 20, 2532-2534. (30) Plakhotnik, T.; Walser, D.; Pirotta, M.; Renn, A.; Wild, U. P. Science 1996, 271, 1703-1705. (31) Brand, L.; Eggeling, C.; Zander, C.; Drexhage, K. H.; Seidel, C. A. M. J. Phys. Chem. A 1997, 101, 4313-4321. (32) Brand, L.; Eggeling, C.; Seidel, C. A. M. Nucleosides Nucleotides 1997, 16, 551-556. (33) Sa´nchez, E. J.; Novotny, L.; Holtom, G. R.; Xie, X. S. J. Phys. Chem. A 1997, 101, 7019-7023. (34) Song, J. M.; Inoue, T.; Kawazumi, H.; Ogawa, T. Anal. Sci. 1998, 14, 913916.
intensity, fs pulse width Ti:sapphire laser sources that are capable of exciting the molecules at rates comparable to those achieved with one-photon excitation (OPE). There are several potential advantages to single molecule detection by TPE compared to those for conventional OPE techniques. The most obvious of these arises from the large separation between the excitation laser wavelength and the fluorescence emission wavelength. Since most of the background radiation occurs near the excitation wavelength, very low background levels can be achieved by using a wide bandwidth optical filter to attenuate the background while collecting the entire wavelength spectrum of the fluorescence signal with high efficiency. Thus, larger signal-to-background ratios (S/ B) are often possible by using TPE for ultrasensitive fluorescence detection rather than OPE.35,36 For single molecule detection of commonly used visible fluorophores such as Rhodamine B, S/B achieved by TPE is generally comparable to OPE, since the fluorescence emission rates of single molecules are normally smaller for TPE than for OPE.29,33 However, recent experiments to detect individual coumarin-120 (C-120) molecules by TPE and OPE revealed a ∼3-fold improvement in S/B for the TPE experiment.31 This resulted mainly from the high background intensity due to UV excitation (350 nm) for the OPE experiments. The background could be suppressed much more efficiently by using TPE. Thus, there are significant advantages to using TPE for single molecule detection of fluorophores such as C-120 that absorb at shorter wavelengths. Most of the solution-phase single molecule detection experiments that have applied TPE have been accomplished using twophoton microscopy.29,31,32 Typically, a ∼100 MHz mode-locked Ti: sapphire laser beam is focused to a near diffraction-limited spot (∼0.5 µm diameter), defining a probe region of 5 µm so that the entire flow stream can be made to pass through the center of the excitation laser beam. Since the two-photon excitation rate is proportional to the square of the laser intensity, to achieve sufficient excitation rates for single molecule detection (35) Shear, J. B.; Brown, E. B.; Webb, W. W. Anal. Chem. 1996, 68, 17781783. (36) Overway, K. S.; Duhachek, S. D.; Loeffel-Mann, K.; Zugel, S. A.; Lytle, F. E. Appl. Spectrosc. 1996, 50, 1335-1337.
Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
2109
in a detection volume of this size, it will be necessary to increase the average laser power to the maximum that is available from typical Ti:sapphire laser oscillators (1-2 W). An alternative to increasing the detection volume is to reduce the diameter of the flow stream. For example, Zander et al. used single-photon confocal fluorescence microscopy for efficient detection of single dye molecules flowing through the tip of a microcapillary that had been pulled to a diameter of 0.5 µm.16 However, because of the large spatial variation in the laser intensity across the detection volume, the variation in the burst intensities from molecule to molecule was much larger than that obtained using single molecule flow cytometry. This paper explores the application of TPE to single molecule fluorescence detection in flow cytometry. We have used an 810 nm, 120 fs pulse width mode-locked Ti:sapphire laser, focused to a diameter of ∼10 µm for TPE of single DNA fragments as small as 383 base pairs (bp). The DNA fragments were labeled with the intercalating dye, POPO-1, at a dye:nucleotide ratio of 1:5. Detection efficiencies of nearly 100% were achieved for DNA fragments of 1.1 kilobase pairs (kbp) and larger. These studies reveal that using TPE for single molecule detection of DNA-intercalating dye complexes lowers the background intensity by nearly an order of magnitude compared to OPE with CW laser excitation. For example, using a CW laser at 514 nm for OPE of POPO-3-labeled DNA, with the laser power adjusted to be just below the optical saturation intensity and experimental conditions similar to those described below, background photocount rates between 20 and 30 kHz are normally observed from a blank solution, through the appropriate optical filter. This background is mainly due to scattering of the excitation laser by the solvent. The scattering intensity at the wavelength that would be needed for OPE of POPO-1-labeled DNA (∼430 nm) will be even larger than this due to the 1/λ4 dependence of the scattering cross section. By contrast, we observed a maximum background photocount rate from a blank solution of only 3-5 kHz when studying TPE of POPO-1-labeled DNA. S/B for single molecule detection of POPO-1-labeled DNA by TPE was somewhat less than that typically observed for OPE of POPO-3-labeled DNA. By using 405 nm pulsed laser excitation for OPE of POPO-1labeled DNA, we found the background photocount rate to be extremely high over a very broad wavelength range (>300 kHz through a 465 ( 15 nm filter and ∼20 kHz through a 515 ( 15 nm filter, 3 mW average laser power). Under these conditions, S/B for single molecule detection of POPO-1-labeled DNA was a factor of ∼5 less than that observed using TPE. Although the conditions under which the OPE experiments were performed were not ideal for single molecule detection of POPO-1-labeled DNA (pulsed laser excitation and suboptimum excitation and emission wavelengths), they illustrate the difficulties encountered in applying OPE to single molecule flow cytometry of fluorophores that absorb at these shorter wavelengths, as well as the superior background suppression capability of TPE. EXPERIMENTAL SECTION Reagents. DNA fragments (383, 1058, 1857, 2064, and 2295 bp) were obtained by restriction digestion of PBR322 DNA (Life Technologies, Gaithersburg, MD) using the appropriate restriction enzymes (New England Biolabs, Beverly, MA). The DNA fragments were separated on 1.0% agarose gel and extracted from 2110 Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
Figure 1. A schematic of the flow cell used for single molecule flow cytometry. The laser beam is focused into the detection region perpendicular to the plane of the page.
the gel using the Prep-A-Gene Kit (Bio-Rad, Hercules, CA). The 2295 bp sample was contaminated with
