Weak tension accelerates hybridization and dehybridization of short oligonucleotides

Abstract

The hybridization and dehybridization of DNA subject to tension is relevant to fundamental genetic processes and to the design of DNA-based mechanobiology assays. While strong tension accelerates DNA melting and decelerates DNA annealing, the effects of tension weaker than 5 pN are less clear. In this study, we developed a DNA bow assay, which uses the bending rigidity of double-stranded DNA (dsDNA) to exert weak tension on a single-stranded DNA (ssDNA) target in the range of 2-6 pN. Combining this assay with single-molecule FRET, we measured the hybridization and dehybridization kinetics between a 15 nt ssDNA under tension and a 8-9 nt oligonucleotide, and found that both the hybridization and dehybridization rates monotonically increase with tension for various nucleotide sequences tested. These findings suggest that the nucleated duplex in its transition state is more extended than the pure dsDNA or ssDNA counterpart. Based on coarse-grained oxDNA simulations, we propose that this increased extension of the transition state is due to steric repulsion between the unpaired ssDNA segments in close proximity to one another. Using linear force-extension relations verified by simulations of short DNA segments, we derived analytical equations for force-to-rate conversion that are in good agreement with our measurements.

Figure 1.

Force-dependent DNA extension. (A) A proposed model for how the extension of a duplex differs between small and large forces. The elasticity of the bound state is rigid, and therefore its extension x_b is mostly unaffected by force. On the other hand, the transition state may be more flexible, in which case its extension x^‡ will be force-dependent. In this case, worm-like chain models predict that at large forces, x^‡ > x_b, whereas at small forces x^‡ < x_b. (B) Force generation with a DNA bow. A bent bow-like duplex of variable length exerts tension on a 15 nt ssDNA target (bowstring), extending the strand.

Figure 2.

FRET-based DNA bow assay. (A) Schematic of DNA bow assay FRET setup. Cy3-labeled DNA bows are immobilized on a PEGylated coverslip and excited by an evanescent wave of a 532-nm laser using TIRF microscopy. The inset highlights the ssDNA sequence (TGAAATTAC) targeted by the Cy5-labeled probe (GTAAATTCA). To avoid additional stacking interactions between the probe and the DNA bow, the 9 nt target segment was flanked by 3 nt ssDNA gaps (highlighted orange) in all construct designs. (B) Example FRET efficiency traces for three different dsDNA arc lengths (210, 105, 74 bp) exerting three separate forces (1.8, 3.8, 6.3 pN, respectively). FRET histograms are shown right. Binding and unbinding rates are extracted from the mean dwell times of low and high FRET states respectively.

Figure 3.

DNA bow construction. DNA bows were constructed in five steps. First, uniquely sized templates were generated with PCR from a common source. Using these templates, two sets of molecules (2a and 2b) were amplified with modified primers via PCR. DNA minicircles were then created from the phosphorylated 2a molecule set using protein-assisted DNA self-ligation. Afterward, DNA minicircles were purified and nicked on the unmodified strand. The final DNA bow constructs were finally constructed by exchanging the nicked strand of circularized 2a molecules with the Cy3-labeled 2b molecule.

Figure 4.

Experimental results. (A) Binding rate vs. force. The plots on the left (right) column are for DNA (RNA) probes. The y-axis is on a logarithmic scale over the same 5.8-fold change for all probe sequences. (B) Unbinding rate vs. force. The plots on the left (right) column are for DNA (RNA) probes. The y-axis is on a logarithmic scale over the same 2.4-fold change for all probe sequences. Vertical error bars for binding and unbinding rates represent the standard error of the mean. Going from top to bottom, the average number of molecules observed in each DNA probe trial was 178, 81, 189 and 145; the average number of molecules observed in each RNA probe trial was 159, 63, 207 and 107, respectively. Horizontal error bars were calculated using , where and σ(x) are the mean and standard deviation of the bow’s end-to-end distance distribution. The fits based on our phenomenological model are shown in red. (C) The dynamic range of all measured rates. The dynamic range was obtained by dividing the rate at the highest force by that at the lowest force; the associated error was calculated by propagating the uncertainty in the underlying rates.

Figure 5.

Simulations of force-extension relations. (A) Schematic of target strand (black) in four unique states: the unbound state, the bound state, the transition state with one base pair in the middle, and the transition state with one terminal base pair. (B) Force–extension curves of target strand in each state. We note that the bound state contains 3-nt ssDNA overhangs, and therefore appears more extensible than dsDNA.

Figure 6.

Heat maps of the free energy ΔG(x, n_bp; f) surface as a function of molecular extension x, the number of base pairs between the target and probe strands n_bp, and the tensile force f. Each black dot marks the extension value with the lowest energy for each base pair step. Probabilities for all combinations of x, n_bp and f were obtained by performing umbrella sampling simulations; the resulting free energy ΔG was then calculated using Equations (5) and (6).