Separation of long DNA molecules in a microfabricated entropic trap array

Science

Volumn 288, Issue 5468, 2000, Pages 1026-1029

  Han, J ^a   Craighead, H G ^a  

^a Cornell University   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

DNA;

ARTICLE; DEVICE; DNA DETERMINATION; DNA STRUCTURE; ELECTROPHORETIC MOBILITY; ENTROPY; GEL ELECTROPHORESIS; PRIORITY JOURNAL;

CHEMISTRY, ANALYTICAL; DNA; ELECTRICITY; ELECTROPHORESIS; ENTROPY; NUCLEIC ACID CONFORMATION; TIME FACTORS;

EID: 0034640410     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.288.5468.1026     Document Type: Article

References (28)

1. J. Sudor and M. V. Novotny, Anal. Chem. 66, 2446 (1994).
2. Y. Kim and M. D. Morris, Anal. Chem. 66, 3081 (1994).
3. _, Anal. Chem. 67, 784 (1995).
4. In addition, concerns were raised [L. Mitnik, C. Heller, J. Prost, J. L. Viovy, Science 267, 219 (1995)] over DNA aggregation in PFCGE caused by an electrohydrodynamic instability that sometimes results in spurious and irreproducible peaks.
5. D. T. Burke, M. A. Burns, C. Mastrangelo, Genome Res. 7, 189 (1997).
6. M. A. Burns et al., Science 282, 484 (1998).
7. W. D. Volkmuth and R. H. Austin, Nature 358, 600 (1992).
8. W. D. Volkmuth et al., Phys. Rev. Lett. 72, 2117 (1994).
9. T. A. J. Duke, R. H. Austin, E. C. Cox, S. S. Chan, Electrophoresis 17, 1075 (1996).
10. S. W. Turner, A. M. Perez, A. Lopez, H. C. Craighead, J. Vac. Sci. Technol. B 16, 3835 (1998).
11. H. P. Chou, C. Spence, A. Scherer, S. Quake, Proc. Natl. Acad. Sci. U.S.A. 96, 11 (1999).
12. C. F. Chou et al., Proc. Natl. Acad. Sci. U.S.A. 96, 13762 (1999).
13. J. Han and H. G. Craighead, J. Vac. Sci. Technol. A 17, 2142 (1999).
14. J. Han, S. W. Turner, H. G. Craighead, Phys. Rev. Lett. 83, 1688 (1999).
15. First, a thin channel was etched into the substrate, then an additional lithography process defined the thick regions in the channel. Fabrication does not require high-resolution lithography techniques, as the fine dimension is controlled by an etch depth. Loading holes were made by potassium hydroxide (KOH) wet etching, and the device was sealed by anodic bonding to a thin glass plate. The same device could also be made using more sophisticated methods (10). Small reservoirs were made at both ends of the channel and the device was filled with buffer solution.
16. DNA used in this work was purchased from Sigma, New England Biolabs, and Gibco BRL. DNA was labeled with YOYO-1 dye (Molecular Probes) at a dye/base pair ratio of 1:10. As a buffer solution, tris-borate-EDTA (TBE) buffer at 5× concentration was used. This high concentration of the buffer effectively quenched the electro-osmosis of the channel without using any other surface modification agents.
17. An inverted microscope (Olympus IX-70) with fluorescence filter set (XF-100, Omega Optical Inc.) was used to detect the fluorescence signal from dyed DNA. Microscope images were recorded by an ICCD camera (ICCD-350F, Videoscope Intl.) into video format.
18. To get an electrophoregram, we electronically defined a region of interest (typically 50 to 150 μm wide) at the end of the channel and summed the fluorescence signal from that area every 0.5 to 1 s. This analysis was done with a video image processor (DVP-32, Instrutech Co.).
19. J. C. Giddings, Unified Separation Science (Wiley-Interscience, New York, 1991), pp. 96-105.
20. The main source of dispersion is not the diffusion of the molecules, but the statistical variation of the trapping time. The escape of DNA from a trap follows the same statistics as a radioactive decay. The standard deviation of trapping lifetime also increases with the average trapping lifetime. This increases the band dispersion at low fields. Diffusion of DNA in the channel is virtually blocked by the existence of the thin regions. In previous single-molecule experiments (13), we could not observe any diffusion across the thin region barrier in the absence of an applied field.
21. Additional data are available at www.sciencemag.org/ feature/data/1046112.shl.
22. The DNA ladder samples used in the experiment were the Mono Cut Mix (New England Biolabs) and 5-kb ladder (Gibco BRL). The Mono Cut Mix sample contains DNA molecules ranging from 1.5 to 48 kbp; the 5-kb ladder contains 5-kbp DNA molecules and multiples up to 40 kbp. The concentration (6.38 μg/ml) and the dye/base pair ratio (1:10) of both DNA solutions were the same.
23. In other data with higher ICCD gain settings, the 1.5-kbp peak was identified as a tiny peak in the electrophoregram. In such a case, however, the ICCD was saturated by brighter peaks, yielding a poor electrophoregram in the 20-to 48-kbp region. This is due to the limited dynamic range of the ICCD camera. Other detectors would be used in a real separation device as opposed to a research system.
24. An increase in the thin region depth may bring about better separation with even longer DNA molecules. The thin region depth corresponds to the average pore size of a gel, which in a conventional gel can be controlled only by changing the concentration of the gel. However, in our system, one can reliably define the thin region depth to a high accuracy in the fabrication process.
25. R. S. Madabhushi et al., Electrophoresis 18, 104 (1997).
26. Smaller DNA can be separated at a higher electric field because of its shorter relaxation time.
27. The fabrication method used here is expected to work for channels less than 50 nm thick.
28. We thank S. W. Turner, R. H. Austin, and E. Cox for their helpful suggestions, and R. Wu and S. Dai for helping in slab gel PFGE work. Supported by NIH grant HG01597.