Previous Article |
Table of Contents
| Next Article 
BIOLOGICAL SCIENCES / BIOPHYSICS
A single-molecule barcoding system using nanoslits for DNA analysis
,¶
*Laboratory for Molecular and Computational Genomics, Laboratory of Genetics, and Biotechnology Center, University of Wisconsin, 425 Henry Mall, Madison, WI 53706; Department of Chemistry, University of Wisconsin, 1101 University Avenue, Madison, WI 53706; Complex Fluids Theory, Faculty of Applied Sciences, Delft University of Technology, Julianalaan 67, 2628 BC Delft, The Netherlands; and Department of Chemical and Biological Engineering, University of Wisconsin, 1415 Engineering Drive, Madison, WI 53706
Communicated by David E. Housman, Massachusetts Institute of Technology, Cambridge, MA, December 14, 2006 (received for review November 12, 2006)
 |
Abstract |
|---|
 Top Abstract Results and DiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
DNA labeling | genomics | nanofabrication | polymer confinement
Almost paradoxically, many schemes for the analysis of single DNA molecules, such as sequencing, employ amplification steps (1, 2) using genomic DNA substrates. These steps, although obviating traditional clone libraries, also eliminate the principle advantage of single-molecule techniques for which measurements describe individual molecules, free from any consequences of an ensemble. Thus, new single-molecule approaches aimed at genomic analysis require a high level of precision and throughput before they become tools for research.
When large genomic DNA molecules are the primary analyte, how random coils are unraveled for high-resolution detection of markers (typically by fluorescence microscopy) as well as the presentation techniques employ a variety of schemes: devices that direct fluid flows for surface deposition of molecules (3, 4); agarose gel matrices (5, 6) or engineered nanopillar arrays (7, 8), which enable electrokinetically driven reptation or threading of DNA molecules; and, most recently, direct nanoconfinement (9, 10), where molecules are trapped in elongated configurations. During the last decade, nano- and microfabrication techniques have catalyzed these developments, allowing researchers to fabricate devices boasting complex geometries in both hard and soft materials. Traditionally, nano- and microscale fabrications for biological research have used silicon or quartz wafers as an adaptation from microelectromechanical systems (11), but there have been problems in integrating these materials with biological or genomic applications. Traditional device fabrication is technologically demanding and time-consuming, requiring expensive lithographic equipment (e.g., electron beam lithography for each nanoscale device) in a clean-room facility and later requiring complicated sealing approaches. Additionally, although electrokinetic control is very common for biomolecule manipulation, the semiconductor attributes of silicon wafers favor quartz fabrication due to its insulatory properties. However, toilsome and expensive device fabrication from quartz necessitates reuse, leading to contamination issues (12). Although nanoscale features are more rapidly rendered through molding techniques (13), the inability to produce a large number of disposable devices within a typical laboratory environment (14) has greatly hindered their use in high-throughput genomic applications.
To overcome the limitations of hard materials, soft lithography based on elastomer replicas (14, 15) has been developed, augmenting traditional lithography for biological applications using microfluidics (12, 16, 17). Soft lithography uses molds (masters) created from silicon wafers patterned with photoresist for producing disposable elastomeric replicas (14), most commonly from poly(dimethylsiloxane) (PDMS), ideally suited for high-throughput applications. Unfortunately, PDMS elasticity portends mechanical instability, especially for nanostructures of <100 nm, which often disappear by collapsing against flat substrates (18).
Given these concerns, our goal was to develop a high-throughput system for genome analysis by using disposable devices offering effective nanoscale geometries sufficient for the presentation of large elongated DNA molecules. The idea was to create the basis for a successor to the established optical mapping system (19, 20), which spans entire genomes through overlapping ordered restriction maps created from individual genomic DNA molecules. Our motivation for the work described here was advancement of whole-genome restriction mapping from a tool for discovery, or sequence validation, to a means for analysis of human populations and cancer genomes. We solved the conundrum of how to imbue features in disposable PDMS devices with nanoconfinement capabilities that closely parallel those found in devices with features of <100 nm and requiring traditional lithographic approaches.
By considering DNA as a polyelectrolyte, its enlargement in terms of physical measures of size hinges on a polymer's persistence length, commonly 
50 nm (21). Physical properties of DNA chains are derived from treatments modeling them as worm-like coils (22), where they exhibit both local rigidity and long-range flexibility. From the theoretical work of Odijk, Skolnick, and Fixman (23, 24), we know that ionic strength is an important factor governing intrachain electrostatic repulsion affecting the persistence length of worm-like polyelectrolyte coils, estimated from the Debye–Hückel screening length. This view was experimentally confirmed for large DNA molecules by Baumann et al. (21), who found that the persistence length of DNA molecules inversely varied with ionic strength, in good agreement with the theory of Odijk, Skolnick, and Fixman. Consequently, we reasoned that low-ionic-strength conditions would sufficiently increase DNA persistence length, allowing for larger channels to be used for polymer confinement. Thus, when comparable length scales are achieved (persistence length and channel dimensions), polymer confinement regimes shift, enabling extensive elongation of polymer chains (25) within channel dimensions readily supported by standard PDMS fabrication techniques. Accordingly, we report our findings showing that DNA molecules are stretched up to 60% of their polymer contour length in disposable PDMS devices having 100-nm x 1-µm channels under low-ionic-strength conditions; remarkably, these results are comparable with those previously obtained under standard buffer conditions using 30- x 40-nm channels (fused silica) employing nanoimprint or electron beam lithography (26).
Although low-ionic-strength buffers enable DNA elongation in larger nanoslits readily made of PDMS, the avoidance of biochemically meaningful salt concentrations causes problems for most DNA enzymes used for genome analysis. However, molecular confinement and DNA modification enzymes (e.g., restriction endonucleases) are not necessarily incompatible when used within standard <100-nm fabrications, as recently demonstrated by Riehn et al. (27). Instead, issues arise regarding the scalability of their device as a viable platform for genomic analysis due to the requirement that molecules must be continuously imaged for discernable biochemical events, greatly diminishing potential throughput. As such, we developed a single-molecule labeling scheme obviating these concerns while offering distinct advantages for robust detection and integration within a system for genome analysis.
Conventional hybridization techniques are not suited for marking discrete DNA molecules because presentation approaches require intact, double-stranded DNA molecules after processing for analysis. Because optical mapping successfully employs restriction enzymes for reliable placement of sequence-specific markers onto individual molecules, we reasoned that a new class of endonucleases nicking at specific sites (28) would also confidently mark molecules but without unwanted double-strand cleavage. Because nicks cannot be directly discerned by fluorescence microscopy, we label these sites by nick-translation of DNA molecules in bulk solution using fluorochrome-labeled nucleotides. We then counterstain DNA backbones with the bis-intercalator YOYO-1 before imaging. Specificity is enhanced by using ligase and dideoxynucleotide blocking steps, greatly diminishing labeling of preexisting random nicks. Finally, incorporated fluorochrome labels support FRET detection, which simplifies data acquisition by requiring one laser for excitation of DNA backbones (donor) and labels (acceptor).
Here, we report a series of interlocking developments and findings to potentiate a device design based on physical modification of large single DNA molecules through simple alterations of solution ionic strength. This developmental stance fosters creation of usable systems for advancing genome analysis. We demonstrate proof of principle with physical maps of BACs.
|
Results and Discussion |
|---|
|
TopAbstractResults and DiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
and T4 bacteriophage DNA, Rg = 0.55 and 1.13 µm, respectively) are greater than the slit dimensions, they are excluded from entering (29) by simple diffusion after capillary loading; an electric potential (70 V) transports relaxed DNA coils to nanoslit entrances for subsequent elongation.
|
(48.5 kb), T4 (166 kb), and Escherichia coli genomic DNA molecules. For elongation, a molecule undergoes electrophoresis within a microchannel toward a nanoslit entrance, where it proceeds to enter then stretch. An E. coli genomic DNA molecule (Fig. 2A) shows a contour length exceeding that of a nanoslit so that its ends appear relaxed within the microchannels. Here, the largest E. coli genomic DNA fragment spans the 105-µm-long nanoslit and is qualitatively sized at 432 kb, assuming a stretch of 0.60 (see the next section); this stretch may be increased through tension exerted by the relaxed ends. Accordingly, the corresponding fluorescence intensity profile shows that this molecule is well stretched without detectable local folding back of DNA segments, also known as hairpins (30). The total size of this molecule, accounting for the relaxed portions, may exceed 1 megabase. Such stretching is contrasted with the molecules completely within the nanoslits that exhibit folding, as evidenced by heightened values in their fluorescence intensity profiles.
|
Low-Ionic-Strength Effects on DNA Elongation.
Diminished ionic strength conditions increase DNA intrachain electrostatic repulsion (23, 24), with a concomitant increase of the polymer persistence length, observable within a nanoslit as DNA stretching. We study these effects by using dilutions of the same working buffer, Tris–EDTA (1x TE; 10 mM Tris base/1 mM EDTA, pH 8.0, titrated with HCl; ionic strength, 8.4 mM), performed within the same slit geometry (100 nm x 1 µm) for all experiments presented here. In addition to the TE buffer system, sodium phosphate and diluted New England Biolabs buffer 4 (Materials and Methods) were evaluated, and each demonstrated DNA stretching within nanoslits at low apparent ionic strengths (data not shown). Fig. 2 B and C show typical images of T4 and DNA (polymer contour lengths, 74.5 and 21.8 µm, respectively) stretched within nanoslits under low-ionic-strength conditions (0.05x and 0.01x TE), with average apparent lengths of 40.9 ± 8.4 µm and 13.0 ± 2.4 µm, respectively. As a direct comparison, our value for the DNA is nearly equivalent to the findings of Reisner et al. (26), who used 30- x 40-nm channels and TBE buffer (0.5x TBE = 45 mM Tris base/1 mM EDTA/45 mM boric acid). Fig. 3 shows a plot of DNA elongation as a function of TE dilution (1/TE) reported as stretch, which is defined as the average apparent length, X, divided by the dye adjusted polymer contour length, L (e.g., L = 21.8 µm for DNA); a stretch of 1.0 indicates complete elongation.
|
 |
where Po is the nonelectrostatic intrinsic persistence length due to base stacking, Pel is the electrostatic persistence length due to intrachain repulsion, 
–1 is the Debye–Hückel screening length, and lB is the Bjerrum length (7 Å in water). Although the value of Po varies according to environmental conditions (21), here we take Po to be 50 nm. Accordingly, we see that stretch plateaus at a 20x dilution (effective ionic strength, 0.45 mM), with a corresponding persistence length of 122 nm, a value greater than our nanoslit height (100 nm). Although we interpret observed DNA stretching to be strictly a function of persistence length determined at a given salt concentration, such lengths may not be accurate due to additional low-ionic-strength effects that we are not taking into account here, including local DNA melting, particularly at AT-rich regions, or experimental limitations, for example, carbonate dissolution or contamination, which alters ionic strength at very low salt concentrations. A dotted line is used for describing alterations of persistence length at <0.05x TE (20x dilution) in Fig. 3.
Polymer Elongation Regimes Under Confinement.
DNA elongation is greatly enhanced when the scale of confinement geometry approaches a value comparable to the radius of gyration (29, 31, 32) or persistence length (24, 26). Previously, Brochard and de Gennes (29) developed a scaling argument explaining polymer stretching under confinement within a channel, where a self-avoiding chain is considered as a series of blobs provided that channel dimensions are much larger than the persistence length (D >> P) and the polymer contour length is greater than D (L >> D). In their treatment, the stretch (X/L) of a polymer (persistence length P and molecule width w) in a channel (width D1 and height D2) is given by X/L 
(wP/D1D2)1/3 (10, 29, 32, 33); however, this argument is not valid for highly stretched polymers (e.g., X/L > 0.5). Instead, for highly stretched polymer chains under nanoconfinement, a different scaling argument (25) treats polymer stretching within a nanochannel as a long chain deflected by the walls of a nanochannel with the stretch given by
 |
Here, we wish to estimate the numerical coefficient. We introduce an analytical treatment based on a rescaling of the harmonic potential to the hard wall repulsions experienced by a worm-like chain confined in a slit geometry, followed by comparison with our experimental observations (Fig. 3). The regime of strong elongation is complicated because of the formation of hairpins (30). Because of this, the crossover between the two regimes is nontrivial and not understood at present.
Calculation of Stretch Within a Slit Regime.
A rigorous analytical calculation is impossible because of the hard wall boundary conditions previously mentioned in ref. 25. Nevertheless, an exact computation is feasible for a worm-like chain confined in a harmonic potential, V(x) (34). The Hamiltonian H in one dimension may be written as
 |
where the first term is the scaled bending energy of the chain and the second is V[x(s)], which simulates the confinement. The position of the chain is x(s) at contour distance s from one end. Statistical mechanics applied to Eq. 3 may be shown to lead to a Gaussian distribution for the worm (as L 
)
 |
In this way, we may eliminate the dummy variable b. From the orientation– translation distribution derived by Burkhardt (35), one may obtain (in one dimension)
 |
 |
 |
One has to subtract the energy from the external field to retain only the configurational free energy Fconf, which is entropic in origin [Fconf = (3/4)Ftot; for a detailed discussion, see appendix 1 in ref. 36]. After eliminating b, we obtain
 |
Burkhardt (34) also computed Fconf for a worm in a hard slit, but numerically,
 |
where Arec = 1.1036.
Rescaling.
If we now assume that a hard wall may be simulated by a Gaussian, then we must identify d via the respective free energies (Eqs. 8 and 9) in two dimensions x, y:dx, dy. For instance, in the x direction, we have
 |
 |
Accordingly, we arrive at the following approximate expression for the relative elongation:
 |
where A = 1,000 nm and B = 100 nm in our experimental condition. Of course, as 
cos
diverges further away from unity, it becomes more approximate because we assume there is no back-folding. We note that the numerical coefficient in Eq. 12 is remarkably low. It implies that the formation of hairpins must be difficult even when A and B are of the order of the persistence length. This corroborates the computations of the high hairpin free energies under the same conditions (30). A fit of Eq. 12 shows good agreement with the experimental observations in Fig. 3, discounting TE buffer dilutions below 1/TE = 20, where measured stretch values plateau, as attributed to the previously stated experimental conditions associated with very low ionic strength effects. We also point out that the effect of the electrostatic interaction between the DNA and the negatively charged walls of the nanoslit in our expansion for the stretch Eq. 12 is quite negligible. Although the Debye scaling length at 1/TE = 20 is a substantial 15 nm long and should result in decreasing the height B to an effective value of 70 nm, the width A is altered little, if at all. The stretch is then predicted to increase by a few more percentage points because of electrostatics, which is well within the experimental margin of error displayed in Fig. 3A.
Equilibration of DNA Stretch.
DNA molecules electrokinetically move through the microchannels as relaxed coils until approaching a nanoslit entrance. There, as one end of a molecule enters the nanoslit, it transiently elongates, adopting a conformation we describe as a single-ended dumbbell. The dumbbell progressively disappears as it threads into the nanoslit, adopting a nascent, fully confined conformation while approaching equilibrium after the field is shut-off. Fig. 4 shows a plot of length vs. time for DNA molecules reaching such equilibrium, where the inset images show dynamic shrinking and unfolding of T4 DNA molecules. These experiments used New England Biolabs buffer 4 at a 1,000x dilution, resulting in a stretch of 0.58 (42.8 ± 3.6 µm; 81 molecules) (data not shown). As seen in Fig. 4, DNA molecules expectedly enter nanoslits with different conformations and reach equilibrated forms via different processes. For example, the apparent equilibration time constants for these molecules are 7.4 and 13.7 s. These findings bear interesting semblance to a recent report in which Mannion et al. (37) investigated DNA equilibration within nanochannels under relatively high salt buffer conditions (5x TBE; 445 mM Tris borate/10 mM EDTA). In their 100- x 100-nm nanochannels, T4 DNA molecules are initially stretched up to 40 µm and exponentially contract down to 26.4 µm (X/L = 0.23) with a relaxation time constant of 9.3 s, whereas in our 100-nm x 1-µm nanoslits, DNA molecules under low salt conditions start at 25 µm and then stretch up to 40 µm (X/L = 0.55).
|
|
Conclusion. We have shown that DNA stretch is possible in relatively large-scale nanoslits by using low-ionic-strength buffers. Such polymer behavior was considered by using an analytical treatment explicitly dealing with slit geometries that engender molecular nanoconfinement. These insights were leveraged through the synergetic effects of low-ionic-strength buffers and the design of disposable PDMS devices bearing nanoslits for DNA analysis through codevelopment of a compatible labeling scheme for barcoding single molecules.
The correlation of increasing DNA elongation with decreasing ionic strength is explained by the electrostatic repulsive force within a persistence length resulting from the Debye screening length. By minimizing the constraints of confinement dimensions required for elongating DNA molecules and instead focusing on salt conditions, PDMS nanostructures become useful tools for future genomic and polymer physics studies. For genomic applications, a stretch of 0.60 is capable of providing valuable data fueling biological investigations, as was previously obtained with the optical mapping system (3). As such, future efforts will leverage fully automated imaging, processing, and barcode construction (preliminary system created; data not shown) for the comprehensive assessment of errors, allowing for full integration with new algorithms created for assembling maps that span entire genomes.
|
Materials and Methods |
|---|
|
TopAbstractResults and DiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
DNA Sample Preparation and Loading.
DNA samples [1 ng/µl DNA (New England Biolabs, Ipswich, MA), 0.78 ng/µl T4 DNA (Waco Chemicals, Waco, TX), and E. coli DNA (38)], contained 0.25 µM YOYO-1, from 1x to 0.01x TE buffer, 4% (vol/vol) 2-mercaptoethanol, and 0.1% (wt/vol) POP6 (Applied Biosystems, Foster City, CA) to suppress electroendosmosis (10). In addition to TE buffer, sodium phosphate buffer (pH 7.9; a 10:93 mixture of 10 mM NaH2PO4/10 mM Na2HPO4) and New England Biolabs buffer 4 with 20 mM EDTA were also used. YOYO-1-stained DNA molecules were loaded into the microchannels via capillary action and then entered the nanoslits by using an applied electrical field (70 V) (Fig. 1) with platinum electrodes inserted into the reservoirs.
Microscopy and Image Processing. The microscopy setup for single color imaging and imaging flattening processes for shading correction is as reported in ref. 3 (for details, see SI Text).
DNA Barcoding. The circular DNA molecules BAC79, BAC150, and BAC614 were linearized with FseI or SpeI (New England Biolabs). Preexisting nicks were repaired by using 2 units of T4 DNA ligase (1 mM ATP) at 16°C for 2 h, with a total volume of 17.5 µl of New England Biolabs buffer 4 or New England Biolabs buffer 2. The mix was then heat-inactivated at 65°C for 10 min. Endonuclease-free-grade E. coli DNA polymerase I (10 units) (Roche Applied Sciences, Indianapolis, IN) and added ddNTPs at 0.2 µM each (Amersham Biosciences, Piscataway, NJ) blocked remaining nicks at 37°C for 30 min (total volume, 40 µl). The labeling reaction mix [3 units of Nb.BbvCI (New England Biolabs)/2 µM Alexa Fluor 647-aha-dCTP/2 µM Alexa Fluor 647-aha-dUTP (Invitrogen, Carlsbad, CA)/20 µM dATP/20 µM dGTP/1 µM dCTP/1 µM dTTP] was then added, and the reaction was incubated for 30 min at 37°C. Enzymes were digested with 100 ng/µl proteinase K in 0.1% (wt/vol) N-lauroylsarcosine for 3 h at 50°C. Samples were diluted 4,000-fold, or buffer conditions were adjusted by dialysis against 2 liters of 0.01x TE buffer solution overnight with a microdispodialyzer (Spectrum Laboratories, Rancho Dominguez, CA) at 4°C.
|
Acknowledgements |
|---|
|
TopAbstractResults and DiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
|
Footnotes |
|---|
Abbreviations: PDMS, poly(dimethylsiloxane); ddNTP, dideoxyribonucleotide; TE, Tris–EDTA.
¶To whom correspondence should be sent at the * address. E-mail: dcschwartz{at}facstaff.wisc.edu
Author contributions: K.J., D.M.D., and D.C.S. designed research; K.J. and D.M.D. performed research; K.J., T.O., R.R., and D.F. contributed new reagents/analytic tools; K.J., D.M.D., T.O., J.J.d.P., M.D.G., and D.C.S. analyzed data; and K.J., D.M.D., T.O., and D.C.S. wrote the paper.
The authors declare no conflict of interest.
This article contains supporting information online at www.pnas.org/cgi/content/full/0611151104/DC1.
© 2007 by The National Academy of Sciences of the USA
|
References |
|---|
|
TopAbstractResults and DiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg What's this?
This article has been cited by other articles in HighWire Press-hosted journals:
|
|
 |
|
  H. Kuhn and M. D. Frank-Kamenetskii Labeling of unique sequences in double-stranded DNA at sites of vicinal nicks generated by nicking endonucleases Nucleic Acids Res., April 1, 2008; 36(7): e40 - e40. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-Y. Xu, Z. Zhu, P. Zhang, S.-H. Chan, J. C. Samuelson, J. Xiao, D. Ingalls, and G. G. Wilson Discovery of natural nicking endonucleases Nb.BsrDI and Nb.BtsI and engineering of top-strand nicking variants from BsrDI and BtsI Nucleic Acids Res., July 9, 2007; 35(14): 4608 - 4618. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
