Aptes

The rGO-FET biosensors were fabricated on silicon dioxide (SiO₂) wafers, following methods in the previous research [52 (link)]. Six rectangular patterns (50 μm × 100 μm each) in a vertical parallel line covering patterned GO on both sides with gold electrodes were produced by meniscus-dragging deposition (MDD) for GO deposition, the reduction of GO, photolithography, reactive ion etching (RIE), e-beam evaporation, and the lift-off process. After applying O₂ plasma (CUTE, Femto Science, Gyeonggi-do, Republic of Korea) at 50 W and 50 sccm for 50 s, a 1:19 ratio of 3-aminopropyl triethoxysilane (APTES, Merck, Darmstadt, Germany) and 99% ethyl alcohol (Merck, Darmstadt, Germany) were used to treat the surface of the SiO2 wafer to create appropriate conditions for the deposition of GO. The MDD method was used to deposit GO on the treated SiO₂ wafers. After the deposition of GO, 57wt% of hydriodic acid (Sigma-Aldrich Inc.) was used for 2 h 30 min at 80 °C for thermal reduction to obtain electrical conductivity. Photolithography, Mask Aligner (MDA-400S, Midas System, Daejeon, Republic of Korea) was utilized for the photoresist (GXR 601, Merck KGaA, Darmstadt, Germany), deposited by spin coating on a reduced stock/flake GO wafer to obtain the proper pattern. AZ 300 MIF (Merck KGaA, Darmstadt, Germany) was used to develop the solution, and the light-absorbed area on the wafer was removed. The PR on the pattern and the remaining rGO, other than the pattern, were removed by RIE using oxygen gas. PR coating and photolithography were processed to perform electrode patterning. Development was applied to remove the PR from the electrode patterned area. Then, the Au source was evaporated in 2000 Å in 2 Å/s, to deposit itself by e-beam evaporator (ULVAC, MA, USA) onto the entire wafer. Finally, an acetone wash was performed to remove the Au deposited on the wafer beside the electrode region. The photograph and detailed dimensions of the fabricated sensor are shown in Figure S2.

The single-molecule manipulation experiments were carried out using a custom high-force magnetic tweezers platform that can exert forces up to 100 pN with ∼1 nm extension resolution for stuck bead at 200 Hz sampling rate37 (link).
For given magnets and bead, the force is solely dependent on the magnet-bead distance F(d), which can be calibrated based on a method described in our previous publication, which has an ∼10% uncertainty due to the heterogeneous bead sizes37 (link). On the basis of the calibrated F(d), multiple ways of force control were achieved by changing d with time accordingly. A constant force is achieved when a constant d is maintained. For loading rate control where force increases linearly with time, F(t)=r × t, the magnet-bead distance is programmed to change with time as d(t)=F⁻¹(r × t), where F⁻¹ is the inverse function of F(d) and r is the loading rate.
For the unfolding experiments, the protein of interest was immobilized on the glass coverslip of a laminar flow chamber and to a 3-μm paramagnetic bead (Dynabeads M270 streptavidin) using Halo-tag/Halo-ligand and biotin/streptavidin chemistry6 (link)38 (link) (Fig. 1d). Briefly, glass coverslip was cleaned in an ultrasonic cleaner in 10% Decon 90 solution, followed by actetone and isopropanol for 30 min each. Then the coverslips was silianized by 1% APTES (Sigma-Aldrich) in methanol for 20 min and rinsed clean by methanol. The APTES-coated coverslip was assembled into a flow channel and NH₂-O4-Halotag ligand (Promega) was immobilized on to the coverslip through glutaraldehyde (Sigma-Aldrich) crosslinking. The channel was blocked by 1 M Tris (pH 7.4) for 30 min followed by 1% BSA in 1 × PBS and 0.1% Tween-20 over night. Protein of interest containing Halo-tag and biotinylated Avi-tag was immobilized by flowing ∼0.1 μg ml⁻¹ protein into the channel for 20 min. And then streptavidin-coated M270 beads were added to the channel to form the tether.
For the refolding rate measurement, a 576-bp DNA linker was incorporated between the protein and the magnetic bead. This reduced the potential effects of steric hindrance of the magnetic beads on the protein-refolding rates. In this case, a 576-bp DNA from lambda phage vector was amplified by PCR with a Thio-labelled 5′ primer and biotin-labelled 5′ primer. The 576-bp DNA was covalently immobilized to epoxy-activated 3 μm paramagnetic beads following the manufacturer's instructions (Dynabeads M270-epoxy). The concentration of DNA during incubation was kept low (∼0.01 ng μl⁻¹) to minimize multiple binding on a single bead. During the experiments, the talin construct of interest was immobilized to a glass coverslip using halo-tag chemistry. The buffer was then switched to one containing 0.02 mg ml⁻¹ streptavidin for 20 min followed by incubation with the DNA-coated paramagnetic beads. The multivalent streptavidin acted as a bridge that linked the DNA handle to talin. All unfolding and refolding experiments were carried out in 1 × PBS, 1% BSA , 1 mM dithiothreitol and 0.1% Tween-20.