Anal. Chem. 2005, 77, 2656-2661
Microscope Objective for Large-Angle Fluorescence Used for Rapid Detection of Single Nucleotide Polymorphisms in DNA Hybridization Thomas Ruckstuhl*,† and Alexander Krieg‡
Physikalisch-Chemisches Institut, Universita¨t Zu¨rich, Winterthurerstrasse 190, CH-8057 Zu¨rich, Switzerland
A new type of microscope objective is used for the rapid detection of sequence-dependent affinity variations in DNA hybridization. We demonstrate that by performing probe/target hybridization on coverslips at room temperature terminal SNPs (single nucleotide polymorphisms) can be detected within seconds. The study of weak pair interaction, such as the association of very short DNA oligomers, requires the use of high analyte concentrations of both partners to generate a detectable amount of associated pairs. The background of high concentrations of unbound fluorescing analyte can easily hide the low signal of a weakly affine reaction and makes association extremely difficult to detect. Fluorescence detection is a powerful approach to analyze minute amounts of material, even single molecules, but it is usually limited to rather low concentrations. This limitation is now overcome due to the new type of microscope objective, which produces an extremely small detection volume at a water/glass interface. Signals obtained from weak interactions are very low and require the use of highly sensitive detection methods. To a great extent the success of fluorescence-based methods in bioanalytics can be attributed to their outstanding sensitivity. The possibility to detect single molecules (single molecule detection, SMD) in vitro and in vivo has brought about a large variety of applications and is beginning to give much insight into all kinds of biomolecular processes, essential information which can hardly be obtained from ensemble measurements.1 In SMD, confocal microscopy is a widely used technique due to its excellent signal-to-noise ratio and the high count rate per molecule. Wide-field microscopy can be considered to be more convenient for imaging a biological sample, but the confocal approach is clearly superior with regard to temporal resolution and signal-to-noise ratio.2 It is preferable for the study of processes occurring on short time scales, e.g., the formation of weak dimers. * To whom correspondence should be addressed. Phone: +353 1 700 5842. E-mail:
[email protected]. † Present address: National Centre for Sensor Research, Dublin City University, Dublin 9, Ireland. ‡ Present address: Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, U.K. (1) Weiss, S. Science 1999, 283, 1676-83. (2) Bo ¨hmer, M.; Enderlein, J. ChemPhysChem 2003, 4, 792-808.
2656 Analytical Chemistry, Vol. 77, No. 8, April 15, 2005
Most methodologies of SMD have a major drawback, specifically a low upper limit of admissible analyte concentrations. However, many reactions in biology occur at rather high concentrations. The core strategy to overcome this limitation is to minimize the fluorescence detection volume as much as possible. In a perspective paper, Laurence and Weiss stressed the importance of a small detection volume for studying weak pairs and summarized the attributes of several emerging solutions.3 They mentioned the detection inside zero-mode waveguides (small cavities in a metal film deposited on fused silica),4 inside nanofluidic channels,5 at the tip of a near-field optical microscope (NSOM),6 in stimulated emission depletion (STED),7 and in totalinternal-reflection fluorescence (TIRF).8 The reported confinement of the volume ranges from 20 aL (1 aL ) 10-18 L) for TIRF down to 0.01 aL for the zero-mode waveguides, thus a factor of 1020000 times smaller than the minimum detection volume of a confocal microscope (∼200 aL). However, there are disadvantages associated with these methods. In zero-mode waveguides, biomolecules and fluorophores may be affected by the metal surface, and the method is not useful for measuring cell samples. STED (reported detection volume down to 0.67 aL) requires high illumination intensities, which can cause photodestruction of the sample. Insofar as experimental simplicity and biological compatibility are concerned, TIRF is preferable. However, the specified volume reduction is rather modest as the illumination above the critical angle of the water/glass interface excites an elliptical area of several thousand square micrometers.9-11 The lateral confinement is achieved by a small aperture that selects a small surface area for detection. Although this configuration allows the combination of TIRF and fluorescence correlation spectroscopy (FCS), it cannot be used for the direct observation of single molecules, due to the rather modest count rate per molecule. (3) Laurence, T. A.; Weiss, S. Science 2003, 299, 667-8. (4) Levene, M. J.; Korlach J.; Turner S. W.; Foquet M.; Craighead, H. G.; Webb W. W. Science 2003, 299, 682-6. (5) Foquet, M.; Korlach, J.; Zipfel, W.; Watt, W. W. Anal. Chem. 2002, 74, 1415-22. (6) de Lange, F.; Cambi, A.; Huijbens, R.; de Bakker, B.; Rensen, W.; GarciaParajo, M.; van Hulst, N.; Figdor, C. G. J. Cell Sci. 2001, 114, 4153-60. (7) Dyba, M.; Hell, S. W. Phys. Rev. Lett. 2002, 88, 163901. (8) Starr, T. E.; Thompson N. L. Biophys. J. 2001, 80, 1575-84. (9) Hansen, R. L.; Harris, J. M. Anal. Chem. 1998, 70, 2565-75. (10) Starr, T. E.; Thompson N. L. J. Phys. Chem. B 2002, 106, 2365-71. (11) Lieto, A. M.; Cush, R. C.; Thompson, N. L. Biophys. J. 2003, 85, 3294303. 10.1021/ac048404n CCC: $30.25
© 2005 American Chemical Society Published on Web 03/08/2005
Recently, we have introduced a confocal TIRF microscope that features a detection volume of 5 aL and allows for direct observation of single molecules with a high signal-to-background ratio.12,13 The count rates obtained from single molecules are similar if not better than those obtained with conventional confocal microscopes of very high numerical aperture (NA). This improvement can be attributed to a novel type of microscope objective, a parabolic mirror objective (PMO),14 which replaces a conventional microscope objective in our instrument. Conventional diffractionlimited optics allowing for the use of standard glass coverslips (refractive index nd ) 1.523) are available up to an NA of 1.4. This corresponds to a cone of captured angles up to 67°, exceeding the critical angle of total internal reflection of water/glass by only ∼6°. Using the PMO, the diffraction limit can be maintained up to substantially higher surface angles. In our microscope it serves to focus the laser and collect the fluorescence up to angles of 75°. Due to the fact that parabolic elements enable the collection of surface-generated fluorescence at large angles, they are highly efficient tools for biochemical applications.15-20 In the current study we employ confocal TIRF microscopy to measure the weak affinity reactions between short DNA oligonucleotides. Hybridization measurements of short DNA fragments play a central role in sequencing by hybridization (SBH). The hybridization yield of an unknown fragment with a pool of known DNA fragments allows for its sequence reconstruction by means of combinatorics. The combinatorial approach of SBH has been applied to de novo sequencing, resequencing, discovery of SNPs, genotyping, and expression monitoring. See ref 21 for a recent review of SBH. SBH is primarily carried out on microarrays that contain all possible sequences of a particular fragment size. For such generic microarrays, hexamer oligonucleotides have become popular.22-26 However, 6-mer duplexes have a very low stability and therefore require measurements at reduced temperature and additional strategies to stabilize the complexes. These include flanking the hexamers on both ends with equimolar mixtures of the four bases (i.e., octamer probes are used),22-24 the use of synthetic oligonucleotides of improved affinity,25,26 and the use of gel pads22-25 to enhance the probe density. Using confocal TIRF microscopy, we measured the hybridization of hexamers under ambient conditions (21 °C) without any (12) (13) (14) (15) (16) (17) (18) (19) (20) (21)
(22) (23) (24) (25) (26)
Ruckstuhl, T.; Seeger S. Appl. Opt. 2003, 42, 3277-83. Ruckstuhl, T.; Seeger S. Opt. Lett. 2004, 29, 569-71. Ruckstuhl, T.; Seeger S. PCT Int. Appl. WO 009946596, 1999. Ruckstuhl, T.; Enderlein, J.; Jung, S.; Seeger, S. Anal. Chem. 2000, 72, 2117-23. Ruckstuhl, T.; Rankl, M.; Seeger, S. Biosens. Bioelectron. 2003, 18, 119399. Krieg, A.; Laib, S.; Ruckstuhl, T.; Seeger, S. ChemBioChem 2003, 4, 58992. Krieg, A.; Ruckstuhl, T.; Laib, S.; Seeger, S. J. Fluoresc. 2004, 14, 75-78. Krieg, A.; Laib, S.; Ruckstuhl, T.; Seeger, S. ChemBioChem 2004, 5, 16805. Ruckstuhl, T.; Verdes, D. Opt. Express 2004, 12, 4642-54. Drmanac, R.; Drmanac, S.; Chui, G.; Diaz, R.; Hou, A.; Jin, H.; Jin, P.; Kwon, S.; Lacy, S.; Moeur, B.; Shafto, J.; Swanson, D.; Ukrainczyk, T.; Xu, C.; Little, D. Adv. Biochem. Eng./Biotechnol. 2002, 77, 75-101. Fotin, A. V.; Drobyshev, A. L.; Proudnikov, D. Y.; Perov, A. N.; Mirzabekov, A. D. Nucleic Acids Res. 1998, 26, 1515-21. Drobyshev, A. L.; Zasedatelev, A. S.; Yershov, G. M.; Mirzabekov, A. D. Nucleic Acids Res. 1999, 27, 4100-4. Proudnikov, D.; Kirillov, E.; Chumakov, K.; Donlon, J.; Rezapkin, G.; Mirzabekov. A. Biologicals 2000, 28, 57-66. Timofeev, E.; Mirzabekov, A. D. Nucleic Acids Res. 2001, 29, 2626-34. Simeonov, A.; Nikiforov, T. T. Nucleic Acids Res. 2002, e91.
Figure 1. Assembly of the confocal TIRF microscope.
further affinity enhancement. We found that this method has a high discriminatory capability with respect to terminal SNPs. EXPERIMENTAL SECTION Confocal TIRF Microscope. A photograph of the PMO and a schematic representation of the microscope are given in Figure 1. The collimated beam of a HeNe laser (632.8 nm, linearly polarized) is expanded to a parallel beam with a waist diameter of 10 mm. The beam is aligned onto the optical axis of the PMO by means of an adjustable mirror doublet (not shown). It should be noted that this alignment has to be very precise (deviation
