Linearization and Labeling of Single-Stranded DNA for Optical

Genetic profiling would benefit from linearization of ssDNA through the exposure of the unpaired bases to gene-targeting probes. This is compromised b...
0 downloads 0 Views 7MB Size
Download PDF
Letter Cite This: J. Phys. Chem. Lett. 2019, 10, 316−321

pubs.acs.org/JPCL

Linearization and Labeling of Single-Stranded DNA for Optical Sequence Analysis Rajib Basak,† Fan Liu,† Sarfraz Qureshi,† Neelima Gupta,‡ Ce Zhang,§ Renko de Vries,∥ Jeroen A. van Kan,† S. Thameem Dheen,‡ and Johan R. C. van der Maarel*,† †

Department of Physics, National University of Singapore, Singapore 117542 Department of Anatomy, National University of Singapore, Singapore 117594 § Institute of Photonics and Photon-Technology, Northwest University, Xi’an, China 710069 ∥ Laboratory of Physical Chemistry and Colloid Science, Wageningen University, 6708 Wageningen, The Netherlands

Downloaded via UNITED ARAB EMIRATES UNIV on January 9, 2019 at 17:32:48 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Genetic profiling would benefit from linearization of ssDNA through the exposure of the unpaired bases to gene-targeting probes. This is compromised by ssDNA’s high flexibility and tendency to form self-annealed structures. Here, we demonstrate that selfannealing can be avoided through controlled coating with a cationic− neutral diblock polypeptide copolymer. Coating does not preclude sitespecific binding of fluorescence labeled oligonucleotides. Bottlebrushcoated ssDNA can be linearized by confinement inside a nanochannel or molecular combing. A stretch of 0.32 nm per nucleotide is achieved inside a channel with a cross-section of 100 nm and a 2-fold excess of polypeptide with respect to DNA charge. With combing, the complexes are stretched to a similar extent. Atomic force microscopy of dried complexes on silica revealed that the contour and persistence lengths are close to those of dsDNA in the B-form. Labeling is based on hybridization and not limited by restriction enzymes. Enzyme-free labeling offers new opportunities for the detection of specific sequences. is compromised by its flexibility (persistence length of around 0.7 nm)14,15 and tendency to form a myriad of self-annealed structures through hybridization of the unpaired bases. Yet, profiling of genetic information would benefit from linearization of single DNA strands through the exposure of the unpaired bases to gene-targeting probes. Here, we propose to apply a copolymer coating to the ssDNA molecules. The advantages of this approach are 2-fold. First, the concomitant increase in bending rigidity of the complex results in a stretch similar to the one of dsDNA. Second, self-annealing of the unpaired bases is largely prevented, resulting in nonaggregated and unfolded single molecules. Furthermore, we will demonstrate that the coating procedure does not preclude site-specific binding of oligonucleotides. The increased bending rigidity allows linearization of ssDNA molecules to a length similar or exceeding the contour length of the doublestranded variant inside channels with a cross-sectional diameter of around 100 nm or by molecular combing. A suitable candidate for coating ssDNA is a cationic−neutral diblock polypeptide copolymer.16,17 The DNA binding block is relatively short and consists of 12 positively charged lysine residues (K12). The K12 block is connected to 4 repeats of an

Stretching of DNA to a length close to its contour length is a key element in many biotechnological and biophysical applications. Examples include flow- and force-stretching to provide real-time information about the protein−nucleic acid interaction at the level of single molecules.1−3 Single DNA molecules can also be stretched by confinement inside a nanochannel with a cross-sectional diameter on the order of tens to hundreds of nanometers.4−6 Advantages of nanoconfinement are that there is no need for chemical modification such as the attachment of tethers, the DNA molecules are in an equilibrated conformation, high throughput can be achieved by using an array of parallel channels, and integration can be done with lab-on-chip devices. Particular applications of the nanochannel platform are mapping of large-scale genomic organization, including restriction enzyme cutting, and denaturation mapping.7−11 The molecules of interest are usually stained and visualized with fluorescence microscopy. For genome mapping and profiling technologies, site-specific labeling by nicking or restriction endonucleases is required. Every locus needs a specific enzyme, which can be prepared with a CRISPR/Cas9 technology.12 Most of the previously reported stretching experiments are done with DNA in its double-stranded form (dsDNA). Linearization of single-stranded ssDNA molecules on a mica surface has been demonstrated before.13 Stretching of ssDNA © XXXX American Chemical Society

Received: November 14, 2018 Accepted: January 7, 2019 Published: January 7, 2019 316

DOI: 10.1021/acs.jpclett.8b03465 J. Phys. Chem. Lett. 2019, 10, 316−321

Letter

The Journal of Physical Chemistry Letters approximately 100 amino acids long polypeptide (C4).18 The amino acid composition of the C4 block is similar to that of collagen. It is also hydrophilic and net electroneutral and behaves as a flexible polymer in aqueous solution. The entire C4K12 diblock copolymer is produced by using recombinant DNA technology by large-scale expression in yeast. The copolymer is monodisperse with a total molecular weight of 38.4 kDa. As shown in previous work, complexation of dsDNA with C4K12 results in an amplified stretch inside 200−300 nm channels.19 Furthermore, it was shown that DNA metabolism is not inhibited by the copolymer coat. Here, DNA in singlestranded form is uniformly coated by C4K12, as illustrated in Figure 1. The stoichiometry of the complexes is expressed by

repeated ultrafiltration. Binding of the copolymer on the single DNA strands results in the formation of a bottlebrush complex, as illustrated in Figure 1. The molecules are coated with various amounts of C4K12, as indicated by the N/P ratio. The coating procedure prevents strand annealing but, as we will see below, does not preclude hybridization of relatively small oligonucleotides. Prior to fluorescence imaging, the ssDNA complexes were uniformly stained through side-binding of YOYO-1 with a ratio of one dye per four bases.24−26 The fluorescence intensity of the thus labeled complexes is fairly weak but comparable to YOYO-1 labeled dsDNA with a ratio of one dye per 100 base pairs (see Figure S1 in the Supporting Information). In the nanofluidics experiment, the stained complexes were brought into an array of 60 μm long and rectangular nanochannels by electrophoresis. After the electric field was switched off, the complexes relax to their equilibrium state within 60 s. Video recording was started 2−5 min after the complexes were brought into the channels and the clips lasted for another 5 min. A montage of YOYO-1 fluorescence images of complexes with various N/P ratios and confined in channels with a cross-section of 170 × 200 nm2 is shown in Figure 2A.

Figure 1. Illustration of a bottlebrush formed by binding of a diblock polypeptide copolymer to ssDNA. Alkali denatured DNA is mixed with copolymer and fluorescence labeled oligonucleotide probes. Following TE-buffer exchange, the probes anneal and the copolymer binds to the single DNA strands. Note that the complementary strand does not have any probe sites, and hence, 50% of the single strands are potentially labeled. The binding block contains 12 cationic lysine residues. Figure 2. (A) Montage of fluorescence images of single-stranded λDNA molecules in 170 × 200 nm2 channels and for various C4K12 to ssDNA ratios. From left to right: N/P = 0.1, 0.5, and 2.0. (B)−(D) Distribution in stretch in a population of 48−53 molecules.

the N/P ratio, which is the ratio of the number of positively charged amino groups on the lysine K12 binding block to the negatively charged phosphate groups on the ssDNA. The linearization of the ssDNA complexes will be gauged from nanofluidics and molecular combing experiments in tandem with fluorescence and atomic force microscopy. Finally, the potential for optical analysis will be demonstrated by targeting two specific sequences of the bacteriophage λ genome. Stretching of Single-Stranded DNA. Depending on channel dimensions, the nanofluidic devices are made of polydimethylsiloxane (PDMS) elastomer or poly(methyl methacrylate) (PMMA) polymer, as described in the Supporting Information.20−23 Single-stranded DNA was prepared by alkali induced denaturation of bacteriophage λ-DNA (48.5 kbp). The ssDNA solutions were incubated with various concentrations of C4K12 copolymer for at least 24 h. Electrostatic binding of the cationic lysine K12 block on the ssDNA molecules was facilitated by a subsequent exchange of alkaline to TE-buffer (10 mM Tris/HCl, 1 mM EDTA, pH 8.0) conditions through

Another montage pertaining to complexes with a coating ratio N/P = 2.0 and in channels of various cross-sectional diameter is shown in Figure 3A. All images refer to well-equilibrated conformations. For each experimental condition, that is, the N/P ratio and channel diameter, we have used a fresh chip and measured the stretch of 48−53 complexes by time-averaging over the duration of the video clips. Fragmented complexes (18%) were ignored by using a cutoff being the mean stretch minus 2 times the standard deviation. In the case of naked dsDNA in nanochannels a similar amount of fragmentation was reported.27 The distributions in stretch are close to Gaussian and are shown in panels B−D of Figures 2 and 3. Inside 170 × 200 nm2 channels, Gaussian fits give a mean stretch R|| = 5.5 ± 0.7, 8 ± 1, and 10 ± 2 μm for N/P = 0.1, 317

DOI: 10.1021/acs.jpclett.8b03465 J. Phys. Chem. Lett. 2019, 10, 316−321

Letter

The Journal of Physical Chemistry Letters

stretch due to the dynamic equilibrium of binding (law of mass action). Note that the distribution in extension is fairly wide with some complexes stretched to a length of 19 μm (Figure 3D). Furthermore, the distribution is slightly left-skewed toward smaller extensions, which is a signature of the presence of back-folded hairpin conformations.28 A stretch of 0.33 ± 0.04 nm per nucleotide for ssDNA is somewhat larger than what can be achieved for dsDNA in 45 nm channels (≈0.29 nm/bp).10 The increase in stretch is mainly due to an increase in bending rigidity of the ssDNA complex. This effect will be further gauged below by atomic force microscopy. We will first demonstrate that genomic information can be obtained by hybridization of oligomeric probes to the complexed, singlestranded DNA molecules. Optical Sequence Analysis. The aim of large scale genome mapping is not to determine the genetic code at the level of single bases, but to provide a map of the genome at the larger scale. This map can be used for genome characterization as well as to sort out large scale variations such as repeats, insertions, inversions, and translocations. For the demonstration of the accessibility of the unpaired bases, we have added an equal mixture of two Alexa Fluor 546 probes with different oligonucleotide sequences to the denatured DNA solution prior to bottlebrush-coating (see Figure 1). These probes were selected to target the 8.0 and 35.8 kb sites of the same single strand of the λ-DNA molecule (there are, hence, no probe sites present on the complementary strand). Note that, in principle, any site can be selected by specification of the nucleotide sequence. In order to obtain a maximal stretch, we have used a 2-fold excess of polypeptide to DNA charge (N/P = 2.0). The doubly labeled (YOYO-1 and Alexa Fluor 546) complexes were brought into the array of 100 × 100 nm2 channels by electrophoresis. The complexes were equilibrated and imaged with fluorescence microscopy using the appropriate optical filters and excitation wavelengths. As can be seen in panel A of Figure 4, individual Alexa Fluor labels are

Figure 3. (A) Montage of fluorescence images of C4K12-coated, single-stranded λ-DNA (N/P = 2.0) molecules in channels of various cross-sectional diameter. The channel cross sections are 250 × 250, 170 × 200, and 100 × 100 nm2, from left to right. (B)−(D) Distribution in stretch in a population of 50 molecules.

0.5, and 2.0, respectively. With a 2-fold excess of polypeptide to DNA charge e (N/P = 2.0), the mean stretch takes the values 4.6 ± 0.7, 10 ± 2, and 16 ± 2 μm inside 250 × 250, 170 × 200, and 100 × 100 nm2 channels, respectively. With decreasing channel diameter and/or increasing C4K12 to ssDNA ratio, the stretch increases. For the wider channels and lower N/P ratios, the stretch is significantly less than the contour length of the parent λ-DNA molecule in its doublestranded form (16.5 μm). However, a maximal stretch of 16 ± 2 μm is reached in 100 nm channels with N/P = 2.0. An excess of polypeptide copolymer is required to achieve a maximal

Figure 4. (A) Montage of fluorescence images of C4K12-coated, single-stranded λ-DNA (N/P = 2.0) in 100 × 100 nm2 channels. The images obtained with YOYO-1 (green) and Alexa Fluor 546 (red) staining are superposed. Labeling sites are noted. The scale bar denotes 5 μm. (B) Distribution in probe distance in a population of 50 molecules in 100 × 100 nm2 channels. (C), (D) As in (A) and (B), but for combed ssDNA. Gaussian fits gives mean probe separations of 9 ± 1 μm, for both nanoarrayed and combed ssDNA. 318

DOI: 10.1021/acs.jpclett.8b03465 J. Phys. Chem. Lett. 2019, 10, 316−321

Letter

The Journal of Physical Chemistry Letters

Figure 5. (A) Distribution in stretch of the 0−8 kb segment in a population of 50 C4K12-coated, single-stranded λ-DNA (N/P = 2.0) in 100 × 100 nm2 channels. (B) As in (A), but for the 35.8−48.5 kb segment. Gaussian fits give a mean stretch of 2.7 ± 0.2 and 4.3 ± 0.4 μm, for the 0−8 and 35.8−48.5 kb segments, respectively.

complexes on silica. In order to image the complexes in their entirety with a field of view of 3 × 3 μm2, we have used 1 kbp dsDNA as the source material for the preparation of bottlebrush-coated ssDNA. No additions to the buffer are necessary to promote adhesion, because the neutral C4 blocks weakly adsorb to silica. Due to the weak adhesion as shown by the easy removal by flushing with water, the complexes are not kinetically trapped in a 3D conformation.31 The molecules were allowed to equilibrate on the surface in a 2D conformation for 5 min followed by the removal of excess copolymer through flushing with ultrapure water. Subsequently, all specimens were N2-dried. A montage of images of the complexes with N/P = 2.0 is shown in Figure 6A (an

discernible at the targeted sites. For comparison, in panel C of Figure 4, we have displayed the corresponding fluorescence images of combed complexes.29 The distributions in probe distance in populations of 50 molecules pertaining to the nanoarraying and combing experiments are shown in panels B and D of Figure 4, respectively. In the nanoarraying and combing experiments, the complexes are stretched to a similar extent of 0.32 ± 0.04 nm per nucleotide. The stretch per nucleotide as inferred from the separation of the Alexa Fluor probes is in perfect agreement with the one obtained from the measurement of the overall stretch of the complex through measurement of the YOYO-1 fluorescence. This confirms that fragmentation has not biased our length measurements. With a typical optical resolution of around 200 nm, the resolution in site position is about 1 kb. This resolution is similar to what can be achieved for dsDNA in 45 nm channels.8,10 The resolution might be increased by using super-resolution imaging strategies and/or combinations of probes of different color.30 In order to assess the uniformity of stretching along the channel, we have also measured the extensions of the sections between the Alexa Fluor labels and the end points of the ssDNA molecules. The distributions in extension of the 0−8 and 35.8−48.5 kb segments are shown in Figure 5. For the 0− 8 and 35.8−48.5 kb segments, the stretch per nucleotide takes the value 0.33 ± 0.03 and 0.34 ± 0.03 nm, respectively. These values of the stretch are similar to the one pertaining to the middle section of the molecule between the Alexa Fluor labels (0.32 ± 0.04 nm per nucleotide). The calculated variation in stretch among the end and middle sections of the linearized single-stranded λ-DNA (N/P = 2.0) molecules in 100 × 100 nm2 channels is 9%. In both the nanoarraying and combing experiments, occasionally Alexa Fluor labels are missing. Out of a pool of 240 ssDNA molecules, 80 and 20 of the molecules show one and two hybridized labels, respectively. This corresponds with a labeling efficiency of 67% and 17% for one and two labels, respectively (only 50% of the single DNA strands are potentially labeled since there are no probe sites present on the complementary strand). The relatively low double label efficiency may be related to differences in binding affinity between the 20 and 30 nucleotide probes targeting the 8.0 and 35.8 kb site, respectively. Other reasons for missing labels are photo induced transitions to dark states including photobleaching (no antibleaching agent was used). Persistence and Contour Lengths. The increase in stretch induced by the copolymer coating can be rationalized in terms of an increase in persistence and/or contour length of the ssDNA complexes. These characteristic lengths were obtained from analysis of atomic force microscopy images of the

Figure 6. (A) Montage of tapping mode atomic force microscopy images showing ssDNA (1 kb) molecules complexed with C4K12 polypeptide copolymer (N/P = 2.0). The scale bar denotes 100 nm. (B) Distribution in contour length in a population of 27 molecules. The average contour length is 330 ± 40 nm. (C) Orientation correlation function of the tangent vectors at a pair of points separated by distance L along the contour. The symbols are the experimental data obtained by population averaging. The solid line represents an exponential fit with P = 60 ± 2 nm.

example of a raw image is shown in Figure S2 in the Supporting Information). Individual complexes are visible, without signs of intercomplex aggregation nor intramolecular hybridization. In order to estimate the thickness of the brush, the crosssectional profiles taken at 10 different and randomly chosen positions of the complexes were analyzed. A Gaussian fit yields a cross-sectional radius of gyration of 5 nm. This value should be taken as an indication, because the measured profiles are broadened by the width of the tip and the brushes are dried and spread on the silica surface. A comparison with the atomic 319

DOI: 10.1021/acs.jpclett.8b03465 J. Phys. Chem. Lett. 2019, 10, 316−321

The Journal of Physical Chemistry Letters

■

force microscopy images of bottlebrush-coated dsDNA in otherwise the same experimental conditions (N/P = 2.0) reveals that the double-stranded complex is much thicker with a cross-section of around 30 nm.19 We traced the centerlines of a population of 27 individual complexes. The distribution in contour length is shown in Figure 6B. The average contour length is 0.33 ± 0.04 μm, which corresponds with a rise per nucleotide of 0.33 ± 0.04 nm. The latter value is in perfect agreement with the values obtained from the total stretch of the complexes (YOYO-1) and the separation between the oligonucleotide probes (Alexa Fluor 546) in the nanoarraying and combing experiments. Furthermore, it confirms that the stretch is somewhat inhomogeneous with a spread of about 10%. From the centerlines, we also obtained the tangent vector correlation function ⟨cos θs,s+L⟩, where θ is the angle between tangent vectors at points s and s + L, by averaging s along the contour. For weakly adsorbed complexes equilibrated in 2D conformation, the tangent vector correlation follows ⟨cos θs,s+L⟩ = exp(−L/(2P)) with the persistence length P. As can be seen in Figure 6C, reasonable agreement is obtained for P = 60 ± 2 nm. The persistence length of the bottlebrush-coated ssDNA molecules is close to the P value of around 50 nm for dsDNA.32 For ssDNA in saline solutions P ≈ 0.65 nm.15 The primary reason for the bottlebrush coating induced increase in stretch inside the nanochannels is, hence, the huge increase in bending rigidity of the intrinsically rather flexible ssDNA molecule. A secondary reason is the increased cross-sectional diameter of the complex. For bottlebrush-coated dsDNA in the same experimental conditions and obtained with the same methodology, P takes a value of 240 ± 10 nm.19 The changes in cross-sectional diameter and persistence length following our denaturation protocol confirm that the complexes are indeed in the single-stranded form. We have shown that ssDNA can be linearized through coating with a diblock polypeptide copolymer. The copolymer coat prevents self-annealing of the unpaired bases and concomitant effects such as intramolecular folding and intermolecular aggregation but does not preclude site-specific binding of fluorescence labeled oligonucleotide probes. A stretch of 0.32 nm per nucleotide can be achieved with a 2-fold excess of copolymer with respect to DNA charge either inside a channel with a cross-sectional diameter of 100 nm or by molecular combing. To the best of our knowledge this is the first report of linearization of ssDNA in nanofluidic channels. Atomic force microscopy of dried ssDNA complexes on silica revealed that the contour and persistence lengths are similar to those of dsDNA in the B-form. An advantage of linearizing ssDNA is that, in principle, any site can be targeted by a specific oligonucleotide probe. There is, hence, no need for labeling with the help of endonucleases. The easy customization of the sequence and density of probes bound on DNA may offer a substantial improvement in the detection of genomic variation at the larger scale. Important features are the possibilities to image multiple targets in close proximity and/or genomic regions without appropriate restriction site motifs. The proposed technology can potentially be used in a wider range of applications by employing more probe color combinations. Besides relatively simple, enzyme free labeling, our technology offers the additional possibility to image a single-stranded genome including RNAs.

Letter

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.8b03465. Sample preparation, chip fabrication, fluorescence imaging (image), molecular combing, atomic force microscopy (image) (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Ce Zhang: 0000-0003-1284-7279 Renko de Vries: 0000-0001-8664-3135 Johan R. C. van der Maarel: 0000-0001-5560-0298 Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This research was supported by grant MOE2014-T2-1-001 of Singapore’s Ministry of Education. REFERENCES

(1) van Oijen, A. M.; Blainey, P. C.; Crampton, D. J.; Richardson, C. C.; Ellenberger, T.; Xie, X. S. Single-molecule kinetics of lambda exonuclease reveal base dependence and dynamic disorder. Science 2003, 301, 1235−1238. (2) Greene, E. C.; Wind, S.; Fazio, T.; Gorman, J.; Visnapuu, M.-L. DNA curtains for high-throughput single-molecule optical imaging. Methods Enzymol. 2010, 472, 293−315. (3) Chaurasiya, K. R.; Paramanathan, T.; McCauley, M. J.; Williams, M. C. Biophysical characterization of DNA binding from single molecule force measurements. Phys. Life Rev. 2010, 7, 299−341. (4) Reisner, W.; Pedersen, J. N.; Austin, R. H. DNA confinement in nanochannels: physics and biological applications. Rep. Prog. Phys. 2012, 75, 106601. (5) van der Maarel, J. R. C.; Zhang, C.; van Kan, J. A. A nanochannel platform for single DNA studies: from crowding, protein DNA interaction, to sequencing of genomic information. Isr. J. Chem. 2014, 54, 1573−1588. (6) Frykholm, K.; Nyberg, L. K.; Westerlund, F. Exploring DNA− protein interactions on the single DNA molecule level using nanofluidic tools. Integr. Biol. 2017, 9, 650−661. (7) Jo, K.; Dhingra, D. M.; Odijk, T.; de Pablo, J. J.; Graham, M. D.; Runnheim, R.; Forrest, D.; Schwartz, D. C. A single-molecule barcoding system using nanoslits for DNA analysis. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 2673−2678. (8) Das, S. K.; Austin, M. D.; Akana, M. C.; Deshpande, P.; Cao, H.; Xiao, M. Single molecule linear analysis of DNA in nano-channel labeled with sequence specific fluorescent probes. Nucleic Acids Res. 2010, 38, No. e177. (9) Reisner, W.; Larsen, N. B.; Silahtaroglu, A.; Kristensen, A.; Tommerup, N.; Tegenfeldt, J. O.; Flyvbjerg, H. Single-molecule denaturation mapping of DNA in nanofluidic channels. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 13294−132949. (10) Lam, E.; Hastie, A.; Lin, C.; Ehrlich, D.; Das, S.; Austin, M.; Deshpande, P.; Cao, H.; Nagarajan, N.; Xiao, M.; Kwok, P. Genome mapping on nanochannel arrays for structural variation analysis and sequence assembly. Nat. Biotechnol. 2012, 30, 771−776. (11) Müller, V.; Westerlund, F. Optical DNA mapping in nanofluidic devices: principles and applications. Lab Chip 2017, 17, 579−590. (12) McCaffrey, J.; Sibert, J.; Zhang, B.; Zhang, Y.; Hu, W.; Riethman, H.; Xiao, M. CRISPR-CAS9 D10A nickase target-specific fluorescent labeling of double strand DNA for whole genome

320

DOI: 10.1021/acs.jpclett.8b03465 J. Phys. Chem. Lett. 2019, 10, 316−321

Letter

The Journal of Physical Chemistry Letters mapping and structural variation analysis. Nucleic Acids Res. 2016, 44, No. e11. (13) Woolley, A. T.; Kelly, R. T. Deposition and characterization of extended single-stranded DNA molecules on surfaces. Nano Lett. 2001, 1, 345−348. (14) Smith, S. B.; Cui, Y.; Bustamante, C. Overstretching B-DNA: the elastic response of individual double-stranded and single-stranded DNA molecules. Science 1996, 271, 795−799. (15) McIntosh, D. B.; Saleh, O. A. Salt species-dependent electrostatic effects on ssDNA elasticity. Macromolecules 2011, 44, 2328−2333. (16) Hernandez-Garcia, A.; Werten, M. W. T.; Stuart, M. C.; de Wolf, F. A.; de Vries, R. Coating of single DNA molecules by genetically engineered protein diblock copolymers. Small 2012, 8, 3491−3501. (17) Hernandez-Garcia, A.; Estrich, N. A.; Werten, M. W. T.; van der Maarel, J. R. C.; LaBean, T. H.; de Wolf, F. A.; Cohen Stuart, M. A.; de Vries, R. Precise coating of a wide range of DNA templates by a protein polymer with a DNA binding domain. ACS Nano 2017, 11, 144−152. (18) Werten, M. W.; Wisselink, W. H.; Jansen-van den Bosch, T. J.; de Bruin, E. C.; de Wolf, F. A. Secreted production of a customdesigned, highly hydrophilic gelatin in Pichia pastoris. Protein Eng., Des. Sel. 2001, 14, 447−454. (19) Zhang, C.; Hernandez-Garcia, A.; Jiang, K.; Gong, Z.; Guttula, D.; Ng, S. Y.; Malar, P. P.; van Kan, J. A.; Dai, L.; Doyle, P. S.; de Vries, R.; van der Maarel, J. R. C. Amplified stretch of bottlebrushcoated DNA in nanofluidic channels. Nucleic Acids Res. 2013, 41, No. e189. (20) van Kan, J. A.; Bettiol, A. A.; Watt, F. Proton beam writing of three-dimensional nanostructures in hydrogen silsesquioxane. Nano Lett. 2006, 6, 579−582. (21) Zhang, C.; Zhang, F.; van Kan, J. A.; van der Maarel, J. R. C. Effects of electrostatic screening on the conformation of single DNA molecules confined in a nanochannel. J. Chem. Phys. 2008, 128, 225109. (22) van Kan, J. A.; Shao, P. G.; Wang, Y. H.; Malar, P. Proton beam writing a platform technology for high quality three-dimensional metal mold fabrication for nanofluidic applications. Microsyst. Technol. 2011, 17, 1519−1527. (23) van Kan, J. A.; Zhang, C.; Malar, P. P.; van der Maarel, J. R. C. High throughput fabrication of disposable nanofluidic lab-on-chip devices for single molecule studies. Biomicrofluidics 2012, 6, 036502. (24) Larsson, A.; Carlsson, C.; Jonsson, M.; Albinsson, B. Characterization of the binding of the fluorescent dyes YO and YOYO to DNA by polarized light spectroscopy. J. Am. Chem. Soc. 1994, 116, 8459−8465. (25) Rye, H. S.; Dabora, J. M.; Quesada, M. A.; Mathies, R. A.; Glazer, A. N. Fluorometric assay using dimeric dyes for double-and single-stranded DNA and RNA with picogram sensitivity. Anal. Biochem. 1993, 208, 144−150. (26) Åkerman, B.; Tuite, E. Single-and double-strand photocleavage of DNA by YO, YOYO and TOTO. Nucleic Acids Res. 1996, 24, 1080−1090. (27) Reccius, C. H.; Stavis, S. M.; Mannion, J. T.; Walker, L. P.; Craighead, H. G. Conformation, length, and speed measurements of electrodynamically stretched DNA in nanochannels. Biophys. J. 2008, 95, 273−86. (28) Ö dman, D.; Werner, E.; Dorfman, K. D.; Doering, C. R.; Mehlig, B. Distribution of label spacings for genome mapping in nanochannels. Biomicrofluidics 2018, 12, 034115. (29) Allemand, J.; Bensimon, D.; Jullien, L.; Bensimon, A.; Croquette, V. pH-dependent specific binding and combing of DNA. Biophys. J. 1997, 73, 2064−2070. (30) Jeffet, J.; Kobo, A.; Su, T.; Grunwald, A.; Green, O.; Nilsson, A. N.; Eisenberg, E.; Ambjörnsson, T.; Westerlund, F.; Weinhold, E.; Shabat, D.; Purohit, P. K.; Ebenstein, Y. Super-resolution genome mapping in silicon nanochannels. ACS Nano 2016, 10, 9823−9830.

(31) Rivetti, C.; Guthold, M.; Bustamante, C. Scanning force microscopy of DNA deposited onto mica: equilibration versus kinetic trapping studied by statistical polymer chain analysis. J. Mol. Biol. 1996, 264, 919−932. (32) Wiggins, P. A.; van der Heijden, T.; Moreno-Herrero, F.; Spakowitz, A.; Phillips, R.; Widom, J.; Dekker, C.; Nelson, P. C. High flexibility of DNA on short length scales probed by atomic force microscopy. Nat. Nanotechnol. 2006, 1, 137−141.

321

DOI: 10.1021/acs.jpclett.8b03465 J. Phys. Chem. Lett. 2019, 10, 316−321