ERASE: a novel surface reconditioning strategy for single-molecule experiments

Abstract While surface-based single-molecule experiments have revolutionized our understanding of biology and biomolecules, the workflow in preparing for such experiments, especially surface cleaning and functionalization, remains labor-intensive and time-consuming. Even worse, meticulously assembled flow channels can be used only once for most experiments. A reusable surface would thus dramatically increase productivity and efficiency of single-molecule experiments. In this paper, we report a novel surface reconditioning strategy termed ERASE (Epitaxial Removal Aided by Strand Exchange) that allows a single flow cell to be used for vast repetition of single-molecule experiments. In this method, biomolecules immobilized to the surface through a nucleic acid duplex are liberated when a competing DNA strand disrupts the duplex via toehold-mediated strand displacement. We demonstrate the wide-range applicability of this method with various common surface preparation techniques, fluorescent dyes, and biomolecules including the bacterial ribosome. Beyond time and cost savings, we also show ERASE can assort molecules based on a nucleic acid barcode sequence, thus allowing experiments on different molecules in parallel. Our method increases the utility of prepared surfaces and is a significant improvement to the current single-use paradigm.

The power of single-molecule investigation is best harnessed when molecules are housed inside a flow chamber rigged with an inlet and an outlet for liquid delivery. One of the advantages of this setup is the capability to change the solution phase by simple perfusion. Reversible reaction rates can then be measured from surface-bound molecules in various buffer conditions using the same flow channel. However, irreversible reaction steps such as high-affinity ligand binding and ATP or GTP hydrolysis can only be measured once from the molecules within the field of view, which constitutes only a tiny fraction of the flow chamber surface. In this regard, the flow chamber is heavily underutilized.
Similarly, host complexes of a different composition or sequence must often be investigated in separate flow channels (17,18) unless they are modified with a different fluorescent label. Therefore, researchers have to prepare multiple flow channels to experiment with variants of the host complex or to obtain statistics for rate measurements. In our experience, preparing flow channels remains one of the most labor-intensive steps of the experimental procedure. A simple wash protocol that allows repeated usage of a single flow channel for different molecules would thus be highly beneficial to the single-molecule biophysics community.
In pursuit of such strategy, we noted that even a stable DNA duplex can be completely disrupted by a reaction called toehold-mediated strand displacement (18). In this reaction, a single-stranded DNA molecule called the 'invader' strand hybridizes to the substrate DNA strand in the toehold region and unzips the adjacent incumbent DNA strand by way of branch migration. We reasoned that this switch functionality could be harnessed as a surface reformatting tool, which would reduce the amount of time and materials necessary for preparing new flow cells.
Here, we introduce a novel surface clearing strategy termed ERASE (Epitaxial Reformatting Aided by Strand Exchange). This method (see Figure 1A) involves three nu- The high molecule count is maintained across several rounds of ERASE while the switch molecule efficiently removes all tether molecules. To allow the surface to reach equilibrium, we waited 12 min after adding tether molecules and 5 minutes after adding switch molecules. All presented data is collected from a single flow cell.
cleic acid oligomers referred to as anchor, tether, and switch. The anchor is a short (10-15 nt) biotinylated oligomer that serves as the surface attachment site. The tether is the extension of the nucleic acid component of the complex of interest, which is purposefully designed to contain a terminal domain complementary to the anchor followed by a short (7-15 nt) spacer domain. The switch is fully complementary to the spacer and anchor domains of the tether molecule. According to this scheme, the experimental complex is immobilized through base pairing between the anchor and tether, while the addition of the switch removes the surfacetethered complexes through toehold-mediated strand displacement. After displacement, the surface is ready for another experiment by introducing molecules with the tether.

Sample preparation
Custom nucleic acid oligomers were ordered from Integrated DNA Technologies and modified to include a Cy3 or Cy5 fluorophore near the 5 end or a biotin linker at either the 5 or 3 end. Unmodified switch molecules were ordered from Eurofins Scientific. The specific sequences are in Supplementary Table S1. 70S ribosomal complexes were prepared as described previously (7).

Slide and cover slip cleaning and functionalization
Eight holes were drilled by computer numerical control (CNC) across both long edges of a quartz slide using a diamond drill bit. Quartz slides and coverslips were sonicated for 10 min in Milli-Q water and dried by vacuum

Experimental setup
Objective-type total internal reflection fluorescence microscopy was performed to image single molecules with

RESULTS
To demonstrate this method, we first tried a Cy3-labeled single-stranded DNA oligonucleotide containing a tether and a shorter biotinylated DNA strand as the anchor (see Figure 1A). The anchor molecules were introduced at 1 nM, which would correspond to a density of ∼1 anchor/(120 nm) 2 at full adsorption. This surface density is high enough to provide ample binding sites for the tether, but low enough to prevent any nearest-neighbor interaction. The tether was then introduced at 50 pM, which resulted in an increasing number of spots in the field of view. The time-dependence of this increase could be fit to a negative single exponential model (see Supplementary Figure S1), which suggests that tethers bind to the surface via densely populated anchors in a pseudo first order process. We next performed ERASE on the flow channel by introducing the switch at 500 nM. The number of spots quickly decayed to the background level over a few seconds (see Figure 1B, Supplementary Figure S3). As shown in Figure 1C, the number of spots in the same field of view changed from ∼250 to nearly zero, indicating that ERASE completely removed all bound tether molecules. After ERASE, re-introduction of tethers at 50 pM could regenerate a similar density of spots in the same field of view. We performed this cycle of ERASE and tether addition as many as six times, and for each cycle, were able to achieve complete removal and recovery of spots in the same field of view with no sign of fatigue. The rates associated with removal and recovery also showed little difference between all trials (see Supplementary Figure S2). Finally, we found ERASE to be highly sequence-specific as the surface was reformatted only when the complementary sequence was used (see Supplementary Figure S3).
To showcase the wide adaptability of ERASE, we repeated the experiment with three common surface passivation techniques: PEGylation (19), bovine serum albumin (BSA) (18), and DDS-Tween-20 + BSA (20). Using the same anchor, tether, and switch concentrations as described above, we observed repeatable performance between trials that was consistent across all passivation schemes (see Figure 2A, Supplementary Figure S4). ERASE is also adaptable to modifications in sequence as well as anchor and toehold length. We repeated the hybridization experiment with a separately designed scheme containing a different anchor length, toehold length, and sequence. We found the new Cy3-labeled DNA tether to produce a similar steady-state molecule count (see Figure 2B) as well as similar binding kinetics (see Supplementary Figure S5). Again, the switch molecule completely removed tethers from the surface. Finally, ERASE can be expanded to other nucleic acid types. We performed the experiment with a Cy5-labeled RNA tether with the same sequence as the second DNA tether except T-to-U substitution and observed comparable performance (see Figure 2B, Supplementary Figure S5). Despite being a different nucleic acid, the same DNA switch molecule from the second design scheme efficiently removed the RNA tether.
The true power of ERASE lies in the ability to reprogram the surface for any biomolecules that harbor an end-exposed nucleic acid tether. To demonstrate this, we repeatedly ERASEd 70S ribosome particles (Cy3-Met-tRNAfMet in the P site) containing an mRNA tether. ERASE consistently removed all ribosomal complexes while molecule counts for each round were very similar (see Figure 2C). The switch molecule, designed to hybridize the mRNA just beyond the anti Shine-Dalgarno sequence (see Supplementary Table S1), specifically removed ribosomal complexes whereas a bioorthogonal control switch did not (see Supplementary Figure S6). Finally, we explored one possible application of ERASE for multiplexing. In this scheme, multiple types of molecules carrying different barcodes in the toehold region would be immobilized to the same flow channel and subjected to the same perfusion experiment. Subsequently, the positions of each molecule type could be determined by performing ERASE with barcode-specific switches in a sequential manner allowing the mixed signal be decomposed. As a proof of principle, we immobilized two different tether molecules fluorescently labeled with Cy3 and Cy5, respectively, and performed ERASE with one switch at a time. We repeated this experiment 5 times and found that each switch removed ∼95% of its target molecules while non-target molecules were removed at a much lower rate of 1% (see Figure 2D). Optimizing the ERASE condition to maximize the false negative rate and minimize the false positive rate will be the topic of a future study.

DISCUSSION
Apart from the time and cost savings due to repeated use of flow cells, ERASE provides several practical advantages to the experimenter. ERASE allows the same field of view to be reused across all experiments in case inhomogeneities from channel to channel or from field to field become an issue. Further, since switch molecules specifically disrupt base pairing between the anchor and the tether, spots that are not removed by switch molecules can be identified as non- (B) ERASE is highly modifiable and can be constructed with any sequence demanded by the experiment. The molecule counts of a different anchor-tether-switch scheme perform comparably to the first scheme. ERASE is adaptable and can be augmented to different types of nucleic acid. Although it is composed of RNA, the tether molecules can still be entirely removed by the same DNA switch molecules. (C) Molecule counts of 70S ribosome complexes with an RNA tether are plotted, and ERASE efficiently removes all molecules. Molecule counts over the same field of view are consistent across several rounds of ERASE. To allow the surface to reach equilibrium, we waited 12 minutes after adding tether molecules and 5 minutes after adding switch molecules. (D) We designed two different DNA tethers, A and B: A is labeled with Cy3 dye and barcode sequence a, and B with Cy5 dye and barcode sequence b. Both tethers contain a common sequence complementary to the anchor. The Cy3-and Cy5-channel images of the same field of view are shown on the left and right columns, respectively. The first row of images are obtained after incubating the anchor-coated surface with a 1:1 mixture of A and B, the second row after ERASE with switchB, which is complementary to B, and the third row after ERASE with switchĀ, which is complementary to A. To minimize signal loss from the corresponding channel, we introduced switch molecules at a lower concentration which decreased ERASE efficiency. The scale bar is 10 m. (E) ERASE is generalizable to any experiment using a nucleic acid tether. The tether-containing molecule can undergo irreversible changes (red) such as protein modification or annealing; however, provided the toehold remains intact, the switch molecule can always remove the tether-containing molecule. All presented data is collected from a single flow cell. specifically bound or spurious and, hence, can be left out of data processing and statistics (see Figure 2D). Therefore, ERASE provides an a posteriori mechanism to select 'good' molecules in an unbiased manner. ERASE can also be extended to ensemble-level measurement techniques such as surface plasmon resonance (21) where molecules are commonly immobilized through nucleic acids linkers.
Although not encountered in this study, accumulation of nonspecifically bound molecules to the surface may worsen the binding capacity of the surface and interfere with subsequent experiments, thereby limiting the number of reuses of the flow cell. Therefore, we recommend potential users of the ERASE scheme to check the level of nonspecific binding and repeatability of successive experiments.
In conclusion, we introduced ERASE, a novel surface reformatting scheme that enhances the power of surface-based, single-molecule experiments. We have shown ERASE to be highly specific for removing the target molecules from the surface without compromising the longterm binding capacity of the surface for subsequent experiments. We have also shown it to be consistent and repeatable across common surface passivation techniques, and different sequences and types of nucleic acid. Due to its simplicity PAGE 5 OF 5 Nucleic Acids Research, 2019, Vol. 47, No. 3 e14 and flexibility, we expect ERASE to become routine for future single-molecule studies.