(215b) DNA Transport Characteristics in Focused Beam-Milled Nanofluidic Devices

Fundamental studies of the electrokinetically-driven transport of DNA through nanopores and nanochannels with critical dimensions ranging from 5 to 100 nm are described. Such nanofluidic devices, allowing the spatial confinement of polynucleotides, are of growing importance in fundamental studies of polymer physics and in bioanalytical applications. Nanopores have been demonstrated as sensors for DNA in electrolyte solutions. Single molecules of DNA are eletrokinetically driven through the pore and the standing ionic current is perturbed. The resulting signal is dependent upon the size of the DNA, its associated charge, secondary and tertiary structure, and DNA/pore interactions. This field was launched by the seminal work of Kasianowicz et al. who demonstrated the transport of single-stranded DNA through a α-hemolysin pore [1]. Present investigations continue along these lines using both biological and solid-state nanopores. The differentiation of nucleotides by analyzing changes in the ionic current blockade amplitude has been demonstrated for blocks of repeating nucleotides and work continues to refine the technique to achieve single base resolution [2]. Additionally, solid state nanopores have been proposed as critical components of more complex devices that will measure changes in capacitance as DNA translocates through the pore or a tunneling current across the bases as the molecule passes opposed nanoelectrodes.

Nanochannels are distinguished from nanopores in that a long segment of the polynucleotide (relative to the total contour length of the molecule) is confined during transport. This has several important implications on the characteristics of transport and on the potential utility of these fluidic platforms. If the critical dimensions of the nanochannel are less than the radius of gyration of the molecule in free solution then transport through the channel requires the deformation of the molecule, achieved by extension coaxial to the channel length. If the channel dimensions are greater than the persistence length of DNA (ca. 50 nm for double-stranded DNA in high ionic strength solutions) then this extension is modeled as a series of self-avoiding blobs with diameter constrained by the channel critical dimension [3]. As the channel size decreases below the persistence length, extension of the molecule in a ?single-file? orientation is expected [3]. Nanochannels have been used to separate different sized DNA molecules, size DNA on a single-molecule basis, and map restriction enzyme sites [4-5]. In such experiments, both enthalpic effects (e.g. polynucleotide/wall interactions) and entropic effects due to the decrease in conformational degrees of freedom are of critical importance.

We report the fabrication of nanopores in thin solid-state membranes using either focused electron beam (FEB) or focused ion beam (FIB) milling. In the case of the former, nanopores as small as 1 nm can be fabricated in 10 to 60-nm thick silicon nitride membranes. With FIB-milling, larger nanopores with diameters of 20 nm or larger can be quickly milled in such membranes. We have also developed FIB-milling techniques to mill nanochannels with critical dimensions approaching 5 nm in insulating substrates of SiO2. The nanochannels are interfaced to microfluidic channels formed by photolithography and wet chemical etching and the open trenches are subsequently covered using fusion bonding. The flexibility of FIB milling has allowed us to mill channels that are hundreds of microns down to ca. 50 nm in length, the latter resulting in features analogous to pores in thin membranes. These platforms have allowed fundamental studies of DNA transport in confined environments.

In one study, arrays of nanochannels approximately 50 µm in length were milled across a junction, connecting parallel microfluidic channels. These nanochannels have critical dimensions of 100, 50, and 25 nm (width and depth, with unity aspect ratio). Devices were prepared by first fabricating microfluidic channels on a 25 x 25 mm² quartz substrate using standard photolithography and wet etching techniques. The nanochannels were then FIB milled through a thin conductive chromium film using a FEI Helios NanoLab dual-beam FIB. The chromium was removed and the device sealed with a quartz coverslip using fusion bonding. Solutions were added to the microchannels through vias that were powder blasted through the substrate prior to FIB milling. Translocation measurements were made in TBE buffer solutions of varying concentration. λ-phage DNA was stained with YOYO®-1 dye at a ratio of 5:1 base pairs:dye molecule and diluted to a concentration of 0.25-1 ng/µL in buffer. Translocations were driven by the application of voltages of 1-20 V across the nanochannels (typical field strengths of 200-4000 V cm^-1) and were imaged through a 100x plan apochromatic oil immersion objective using an electron multiplying CCD camera with frame rates up to 400 s^-1. The threshold field strength necessary to overcome the entropic barrier and drive translocations was experimentally determined. The recorded images were analyzed to determine the DNA extension length and dynamic behavior during translocation, the electrophoretic mobility of the DNA, and the velocity of DNA threading into the nanochannels. It was observed that the electrophoretic mobility decreased as the nanochannel critical dimensions decreased, a result commensurate with the increasing channel surface area inducing frictional drag on the DNA. Electrophoretic mobility was found to be independent of DNA size and constant over a range of applied voltages. This latter result differs from results reported for ?nanoslits? (channels with micron-scale widths and nanoscale depths) prepared using reactive ion etching [6]. The extension lengths of DNA molecules were found to decrease as the molecules translocated. The proposed model for this effect is one in which the DNA is stretched against the globular portion in the microchannel as it is threaded into the nanochannel and is elastically restored toward a steady-state length once the DNA is fully in the nanochannel. The translocation events are sufficiently short (tens of ms) that slower equilibrations due to entropic restoring forces are not observed [7].

The translocation of double-stranded DNA through nanopores was also investigated. In these experiments, a voltage across the nanopore and the resulting ionic current were controlled and monitored using a patch clamp amplifier (Axon Instruments Axopatch 200B). DNA solutions were introduced on one side of the pore-containing membrane and electrophoretically driven through the pore to the other side. The localized functionalization of the nanopore in which the pore surface was modified with one moiety while the membrane surface displayed a different surface chemistry was investigated as a means to control DNA/pore interactions. This serves not only to modify mobilities but also to tune the entropic and enthalpic contributions to the barrier to DNA threading, which may further facilitate the discrimination of analytes. This surface functionalization was achieved using well-known and flexible silane chemistry and its localization to the pore interior was characterized using electron energy loss spectrometry (EELS). It was found that by functionalizing the silicon nitride walls of the pore interior with trimethylchlorosilane that the translocation of DNA could be slowed by over 50%. These modifications are also easily applied to long nanochannels milled in SiO2 and comparable alteration of the DNA mobility was demonstrated.

The unifying theme in these and similar experiments enabled by these nanofluidic platforms is the desire to obtain fundamental insights into the behavior of polynucleotide transport in confined environments. There is a practical motivation toward pores and channels with restrictions that are equal to the molecular width of the analyte of interest. In the case of the strategies proposed for the physical sequencing of single DNA molecules (ionic current, capacitance, tunneling current) this condition results in a maximum in signal-to-noise and minimizes the conformational dynamics of the molecule that limit spatial resolution. This work describes efforts to decrease the critical dimensions of nanofluidic pores and channels used for single molecule measurements, modify their surface chemistry, and measure the resulting transport characteristics of polynucleotides.

[1] J.J. Kasianowicz, E. Brandin, D. Branton and D.W. Deamer, Characterization of individual polynucleotide molecules using a membrane channel, Proc. Natl. Acad. Sci. USA 1996, 93, 13770-3.

[2] J. J. Kasianowicz, S. E. Henrickson, H. H. Weetall, B. Robertson, Simultaneous multianalyte detection with a nanometer-scale pore, Anal. Chem. 2001, 73, 2268-72.

[3] W. Reisner, K. J. Morton, R. Riehn, Y. M. Wang, Z. Yu, M. Rosen, J. C. Sturm, S. Y. Chou, E. Frey and R. H. Austin, Statics and dynamics of single DNA molecules confined in nanochannels. Phys. Rev. Lett. 2005, 94, 196101-1-4.

[4] F. Xu and Y. Baba, Polymer solutions and entropic-based systems for double-stranded DNA capillary electrophoresis and microchip electrophoresis, Electrophoresis 2004, 25, 2332-45.

[5] K. Jo, D. M. Dhingra, T. Odijk, J. J. de Pablo, M. D. Graham, R. Runnheim, D. Forrest and D. C. Schwartz, A single-molecule barcoding system using nanoslits for DNA analysis, Proc. Natl. Acad. Sci. USA 2007, 104, 2673-8.

[6] G. B. Salieb-Beugelaar, J. Teapal, J. van Nieuwkasteele, D. Wijnperlé, J. O. Tegenfeldt, F. Lisdat, A. van den Berg and J. C. T. Eijkel, Field-dependent DNA mobility in 20 nm high nanoslits, Nano Lett. 2008, 8, 1785-90.

[7] J. T. Mannion, C. H. Reccius, J. D. Cross and H. G. Craighead, Conformational analysis of single DNA molecules undergoing entropically induced motion in nanochannels, Biophys. J. 2006, 90, 4538-45.