Quantitative Assessment of Tip Effects in Single‐Molecule High‐Speed Atomic Force Microscopy Using DNA Origami Substrates

Abstract High‐speed atomic force microscopy (HS‐AFM) is widely employed in the investigation of dynamic biomolecular processes at a single‐molecule level. However, it remains an open and somewhat controversial question, how these processes are affected by the rapidly scanned AFM tip. While tip effects are commonly believed to be of minor importance in strongly binding systems, weaker interactions may significantly be disturbed. Herein, we quantitatively assess the role of tip effects in a strongly binding system using a DNA origami‐based single‐molecule assay. Despite its femtomolar dissociation constant, we find that HS‐AFM imaging can disrupt monodentate binding of streptavidin (SAv) to biotin (Bt) even under gentle scanning conditions. To a lesser extent, this is also observed for the much stronger bidentate SAv–Bt complex. The presented DNA origami‐based assay can be universally employed to quantify tip effects in strongly and weakly binding systems and to optimize the experimental settings for their reliable HS‐AFM imaging.


DNA origami assembly
Rothemund triangles [1] have been assembled from 208 staple strands (Metabion) and the M13mp18 scaffold as previously described [2] in 1 x TAE (Roth) containing 10 mM MgCl2 (Sigma-Aldrich). The Bt-modified staple strands (Metabion, see Table S1) were added in 10-fold excess to the unmodified staple strands. Hybridization was carried out in a Thermocycler Primus 25 advanced (PEQLAB) by heating to 80°C and subsequent cooling to room temperature over a time course of 90 min. The samples were purified with 1 x TAE/MgCl2 buffer by spin filtering using 100 kDa Ultra-0.5 ml centrifugal filters (Amicon). The concentration of the purified DNA origami solution was determined with an IMPLEN nanophotometer and adjusted with 1 x TAE/MgCl2 to 5 nM. Table S1. Sequences of all Bt-modified staple strands. The T4 spacers indicated in bold face. Rothemund's original notation is used to identify the staples.

Sample preparation for HS-AFM measurement
20 µl DNA origami solution with a concentration of 5 nM was pipetted onto a freshly cleaved mica substrate (1 cm diameter) mounted in a liquid cell and incubated for 2 minutes. Then, the substrate was washed with 1 ml of 1 x TAE/MgCl2 buffer (pH 7.5) to remove unbound DNA origami. The liquid cell was then filled with 1 ml of 1 x TAE/MgCl2 buffer (pH 7.5) containing 20 nM SAv (Sigma-Aldrich). After 1 h of incubation, the sample was subjected to HS-AFM imaging.

HS-AFM imaging
HS-AFM imaging was performed using a JPK Nanowizard ULTRA Speed using USC-F0.3-k0.3 cantilevers (f = 300 kHz, k = 0.3 N/m, NanoWorld). The images were recorded with scan sizes of 1 x 1 µm 2 and a resolution of 512 x 512 px². A constant free amplitude of 3.3 nm was used throughout the experiments.

Determination of binding yields from the recorded HS-AFM images
Time-dependent binding yields were determined by manually counting the occupation all the binding sites of five selected DNA origami in each recorded frame, averaging over a total of 15 monodentate and 15 bidentate SAv-Bt binding sites. The steady-state binding yields presented in Figure 5 have been determined by performing a linear fit with slope zero in the final 100 s (from 500 s to 600 s) of the saturation regime.

Exponential decay fits of monodentate binding yields
The monodentate binding yields for all three SRs at LR ≥ 30 Hz have been analyzed by applying an exponential decay fit according to yield = SS + (100 % − ) − off,tip ( − 0 ) , (Equation 1) with the steady-state binding yield ySS as given in Figure 5 of the main article, the time point t0 at which the first HS-AFM image of the time series was recorded, and the dissociation rate constant koff,tip. The fits are shown in Figures S6 to S8 and the obtained koff,tip values are presented in Figure S9.    the koff previously obtained for SAv-Bt dissociation in bulk solution. [3]