A DNA turbine powered by a transmembrane potential throughout a nanopore

Nanopore array fabrication

Nanopore arrays are fabricated as reported earlier than²⁷. In short, a 100-nm-thick layer of poly(methyl methacrylate) electron-sensitive resist (molecular weight 950,000, 3% dissolved in anisole, MicroChem Corp) was spin-coated on 20 nm free-standing silicon nitride membranes supported by silicon. Subsequently, the resist was uncovered and patterned by an electron-beam sample generator (EBPG5200, Raith) with 100 keV electron beams. The sample is developed in a combination of methyl-isobutyl-ketone and isopropanol with a ratio of 1:3 for 1 min, then stopped in isopropanol for 30 s. The uncovered substrates have been then etched utilizing reactive-ion etching with fluoroform and argon (200 s, 50 W, 50 sccm of CHF₃, 25 sccm of Ar, 10 μbar, SENTECH SI 200 plasma system). Lastly, the resist was eliminated in oxygen plasma for 1 min (200 cm³ min⁻¹ O₂, 100 W, PVA TePla 300) adopted by an acetone tub for five min.

Design, folding and purification of DNA origami buildings

All buildings have been designed utilizing caDNAno v.0.2²⁸. For the cryo-EM reconstruction of the turbine half, all buildings have been designed with a compact beam on high of every turbine construction (Supplementary Figs. 6 and 7) and designed solely utilizing a 7,560-base-long scaffold. The folding response mixtures contained a last scaffold focus of fifty nM and oligonucleotide strands (IDT) of 500 nM. The folding response buffer contained 5 mM Tris, 1 mM EDTA, 5 mM NaCl and 20 mM MgCl₂. The folding options have been thermally annealed utilizing TETRAD (MJ Analysis, now Bio-Rad) thermal biking units. The reactions have been left at 65 °C for 15 min after which subsequently subjected to a thermal annealing ramp from 60 °C to twenty °C (1 °C h⁻¹). The folded buildings have been purified from extra oligonucleotides by bodily extraction from agarose gels and saved at room temperature till additional utilization. The record of oligonucleotides may be present in Supplementary Data.

The turbine construction with a protracted DNA bundle as load was designed utilizing a scaffold of 8,064 bases and a scaffold of 9,072 bases. The folding response mixtures contained a last scaffold focus of 10 nM plus oligonucleotide strands (IDT) of 100 nM every. The folding response buffer contained 5 mM Tris, 1 mM EDTA, 5 mM NaCl and 15 mM MgCl₂ for the left-handed and right-handed variations or 20 mM MgCl₂ for the achiral model of the turbine. The folding response mixtures have been thermally annealed utilizing TETRAD (MJ Analysis) thermal biking units. The reactions have been left at 65 °C for 15 min after which subjected to a thermal annealing ramp from 60 °C to twenty °C (1 °C h⁻¹). The folded buildings have been purified from extra oligonucleotides by polyethylene glycol precipitation and saved at room temperature till additional utilization. Particulars of all of the procedures may be present in ref. ²⁹.

Cryo-EM pattern preparation, picture acquisition and processing

Grid preparation, picture acquisition and knowledge processing have been largely carried out as reported beforehand³⁰. The pattern was utilized to a glow-discharged C-Flat 1.2/1.3 4C thick grid (Protochips) and vitrified utilizing a Vitrobot mark IV (FEI, now Thermo Scientific) at a temperature of twenty-two °C, a humidity of 100%, 0 s wait time, 2 s blot time, −1 blot power (arbitrary system items) and 0 s drain time. Micrograph movies with 10 frames have been collected for the right-handed and left-handed variations (3,427 and 5,997 respectively) at a magnified pixel dimension of two.28 Å and an accrued dose of ~60 e Å⁻² utilizing the EPU software program and a Falcon 3 detector (FEI) on a Cs-corrected (CEOS) 300 kV Titan Krios electron microscope (FEI). For the left-handed model, acquisition with a stage tilt of 20° was used to cut back the orientation bias of the particles.

Movement correction and distinction switch perform estimation of the micrographs have been carried out utilizing the implementation in RELION 4.0 beta^31,32 and CTFFIND4, respectively³³. Particles have been autopicked utilizing TOPAZ³⁴ and subjected to a variety course of consisting of a number of rounds of 2D and 3D classification in RELION to take away falsely picked particles and broken particles. Utilizing an ab initio preliminary mannequin, a refined 3D map was reconstructed from 97,054 and 71,992 particles for the right-handed and left-handed variations, respectively, adopted by per-particle movement correction and dose weighting and 3D refinement (Supplementary Figs. 6 and 7). For a centered reconstruction of the turbine, multibody refinement³⁵ was carried out. The consensus map was divided into two elements containing the lever and the turbine utilizing the eraser software in UCSF Chimera³⁶, and low-pass-filtered gentle masks of the respective areas have been created in RELION (Supplementary Figs. 8 and 9). After multibody refinement, a set of particles with the subtracted sign of the lever arm was calculated and subjected to a different spherical of 3D refinement. The ultimate maps have been masked, sharpened and low-pass filtered utilizing the estimated decision based mostly on the 0.143 Fourier shell correlation criterion. Atomic fashions have been constructed utilizing a cascaded rest protocol as described beforehand³⁰ (Supplementary Fig. 11).

The scale of the generators have been measured in Fiji³⁷ utilizing orthographic projections of the maps created with ChimeraX³⁸. For the twist measurement of the turbine variations, slices from the nicely resolved central elements have been extracted from the cryo-EM density maps utilizing atomic mannequin suits at base-pair positions which can be on the identical aircraft within the design, with a spacing of 33 bp and 34 bp for the right- and the left-handed model, respectively (Supplementary Fig. 10). For every model, the slices have been fitted into one another on the idea of most overlay utilizing ChimeraX³⁸ to find out the rotation angle. From the twist density, the diameter and the size of the helices, the outer blade angle with respect to the helical axes was calculated.

Single-particle fluorescence imaging

Strong-state nanopore chips have been oxygen-plasma cleaned earlier than all of the fluorescence experiments (100 W for 1 min, Plasma Prep III, SPI Provides). Coverslips (VWR, no. 1.5) have been cleaned by ultrasonication sequentially in acetone, isopropanol, water, 1 M KOH answer and deionized water (Milli-Q) for 30 min every. The cleaned coverslips have been then blow-dried totally with compressed nitrogen. The nanopore chip was glued into the PDMS (SYLGARD 184 silicone elastomer) stream cell utilizing a two-component silicone rubber (Ecoflex 5, Easy-ON), then the PDMS stream cell was bonded to the cleaned coverslip after oxygen-plasma remedy (50 W, 50 mbar for 30 s) and post-bake at 120 °C for 30 min. After meeting, the entire system was once more handled with oxygen plasma (50 W, 50 mbar) for 4 min earlier than embedding a pair of Ag/AgCl electrodes, one in both sides of the reservoir, and flushing in deionized water to moist the channels. That is important for growing the hydrophilicity of the membrane and guaranteeing a negatively charged silicon nitride floor. The PDMS nanopore units have been at all times assembled shortly earlier than every experiment and by no means reused.

The nanopore chip was then imaged utilizing an epifluorescence microscope with a ×60 water immersion goal (Olympus UPlanSApo, numerical aperture 1.20) and a quick scientific complementary metallic–oxide–semiconductor digicam (Prime BSI, Teledmy Photometrics). The digicam subject of view was diminished as wanted to attain excessive body charges (usually round 200 pixels × 200 pixels). To picture Cy3-labelled DNA generators, a 561 nm laser (Stradus, Vortran Laser Expertise) was used to excite the fluorophores. The everyday publicity time of the experiments was 5 ms, which led to a body price of round 190–200 fps. To simplify the information evaluation, a hard and fast body price worth (200 fps) is used. Earlier than imaging, the imaging buffer (50 mM Tris-HCl pH 7.5; 50 mM NaCl except in any other case acknowledged, 5 mM MgCl₂; 1 mM dithiothreitol, 5% (w/v) d-dextrose, 2 mM Trolox, 40 μg ml⁻¹ glucose oxidase, 17 μg ml⁻¹ catalase; 0.05% TWEEN 20) was positioned into the reservoirs on both aspect of the silicon nitride membrane.

Driving DNA generators utilizing transmembrane ion gradients

An imaging buffer with the identical salt focus (50 mM NaCl) was flushed into the stream cell first, with DNA generators on the cis aspect of the membrane. Subsequently, an imaging buffer containing a better NaCl focus was flushed into the trans aspect of the membrane. With single-particle fluorescence microscopy, the docking and rotation of the DNA generators may very well be noticed and recorded. To launch the generators from the nanopore, we both inserted a pair of momentary electrodes into the inlet and outlet of the stream channels and launched the generators electrically, or we flushed in the identical (lower-concentration) buffer because the cis aspect. Due to the photobleaching and accumulation of the DNA generators close to nanopore arrays, we selected 40 s as a typical statement period. Examples of longer recordings are proven in Supplementary Fig. 23.

Driving DNA generators utilizing transmembrane voltages

In distinction to the salt-gradient-driven mode, a pair of electrodes was embedded into the stream cells. We used a custom-built circuit to use voltages³⁹. The output voltage was managed by a {custom} LabVIEW program. The electrodes embedded within the two reservoirs have been linked to the circuit. The DNA origami generators have been positioned into the electrically grounded aspect (cis aspect) of the stream cell with a typical focus of 1 pM. After making use of the voltage, the DNA generators have been docked onto the nanopores underneath a 100 mV bias voltage (except in any other case acknowledged) throughout the membrane. The generators may very well be simply launched from the nanopore array by flipping the voltage polarity after which setting it to 0 mV for a number of seconds to permit the imaged generators to diffuse away from the seize area. To keep away from overcrowding of DNA generators close to the nanopore array, which elevated the fluorescence background fluctuation, we usually imaged the generators earlier than the array was absolutely stuffed, and subsequently launched them from the nanopore. Then a brand new group of generators may very well be captured and docked once more by making use of a optimistic bias.

Fluorescence microscopy knowledge evaluation

For picture processing, first a single-molecule localization was carried out utilizing Fiji (ImageJ³⁷) with the ThunderSTORM plugin⁴⁰ for all frames within the acquired picture sequences. A wavelet filter (B-spline) and an built-in Gaussian technique have been used for the localizations. Then the outcomes have been filtered on the idea of their high quality (uncertainty < 50 nm) and the native density (filter of 15 particles in each 50 nm amongst all localized knowledge factors within the sequence) to rule out free-diffusing (non-captured) generators. Subsequent, the single-molecule localization outcomes have been analysed utilizing a {custom} MATLAB script (Code availability). In short, all coordinates of localized particle positions have been clustered on the idea of their Euclidean distance for every turbine. When localized particle positions have been deduced in a video, a circle was fitted to the information to acquire the centre and a radius, which subsequently was used for calculating the angular place of the fluorophores in every body. Subsequent, we decided if the fluorophores occupy spatial states that may be fitted to a round path. We did this by evaluating the purpose density of the coordinates inside an annulus across the fitted round perimeter (±1 nm) with the density of factors across the centre (with a radius r, in order that the world of this central circle is the same as that of the ring area). If the purpose density throughout the annulus was increased, then this knowledge group can be saved within the statistics, else it might be thought-about an invalid trajectory and discarded. Lastly, the script calculated all needed movement properties of the turbine, together with its cumulative angular displacements, MSD, angular velocity and torque. The angular velocity ω_d was decided by becoming MSD = ω_d²t² + 2D_rt to the MSD curve of every turbine, the place t is the lag time and D_r is the rotational diffusion coefficient (additionally as a becoming parameter). The estimation of the torque is mentioned in Supplementary Part 1.

MD simulations

All MD simulations have been carried out utilizing the NAMD program⁴¹, CHARMM36 parameters for DNA, water and ions⁴² with CUFIX⁴³ corrections, periodic boundary situations and the TIP3P mannequin of water⁴⁴. The long-range electrostatic interactions have been computed utilizing the particle-mesh Ewald scheme over a grid with 1 Å spacing⁴⁵. Van der Waals and short-range electrostatic forces have been evaluated utilizing the ten–12 Å clean cutoff scheme. Hydrogen mass repartitioning⁴⁶ and the SHAKE⁴⁷ and SETTLE⁴⁸ algorithms have been used, enabling a 4 fs integration time step. The total electrostatics have been calculated each two-time step. Besides the place specified, a Langevin thermostat with a 0.1 ps⁻¹ damping coefficient maintained a temperature of 295 Ok in all simulations. Coordinates have been recorded each 2,500 steps.

Atomistic fashions of your complete turbine have been assembled from the caDNAno²⁸ design file utilizing a {custom} mrdna script⁴⁹. Along with neutralizing Mg²⁺, 10 mM Mg²⁺ hexahydrate was positioned adjoining to the DNA in response to a beforehand described protocol⁴³. Water and monovalent ions have been added to the system utilizing the solvate and autoionize plugins for VMD⁵⁰, with the solvent field lower to kind a hexagonal prism. For every salt situation, a 4 ns simulation was carried out with a Nosé–Hoover Langevin piston barostat^51,52 set to keep up a goal stress of 1 bar, permitting the equilibrium quantity of the system to be decided. The ensuing system dimensions have been utilized in constant-volume simulations to equilibrate the turbine with harmonic place restraints holding the phosphorus atoms to their preliminary coordinates throughout the first 7 ns of the simulation (okay_spring = 1 kcal mol^–1 Å^–2 for t < 5 ns; 0.1 for five ns < t < 7 ns). After 30 ns of equilibration for the 50 mM and 75 ns for the three M system, a snapshot of the configuration was used to initialize subsequent simulations with both an electrical subject or stress gradient utilized to drive the turbine. Extra equilibration was carried out for the 50 mM NaCl system for an additional 128 ns to initialize the ‘Alternate conf.’ system (Supplementary Fig. 24).

The conformation of the turbine on the finish of equilibration was used to find out the remainder positions of a number of harmonic collective variable (colvar)⁵³ potentials, together with a spring restraining the centre of mass of each third phosphorus atom (okay_spring = 500 kcal mol^–1 Å^–2); a spring restraining the root-mean-square deviation (RMSD) of those phosphorus atoms with respect to the post-equilibration configuration (okay_spring = 1,000 kcal mol^–1 Å^–2; resting RMSD = 0), after optimum inflexible physique transformations in order that the potential doesn’t apply a internet torque or power; and a pair of centre of mass harmonic restraints utilized to 16-bp-long sections of the central six-helix bundle (okay_spring = 50 kcal mol^–1 Å^–2), positioned close to both the top of the shaft to stop the turbine from tilting. With these colvars stopping translation, conformational fluctuations, or tilting of the turbine, an electrical subject was utilized by inserting a relentless power on every atom with a magnitude proportional to the cost of the atom. Equally, a stress gradient was achieved by inserting a small power on each water molecule of the system. Lastly, in simulations the place the torque was measured, an extra spin angle colvar (okay_spring = 100 kcal mol^–1 °^–2) prevented rotation of the turbine and reported the torque.

Simulation programs have been ready to review the forces on and flows round a DNA helix mimicking the DNA within the turbine blade. The 21 bp helix was made successfully infinite by connecting the ends of every strand throughout the periodic boundary. Solvent (neutralizing Na⁺, 100 mM and three M NaCl; no Mg²⁺) was added across the helix. Programs have been equilibrated for five–50 ns with the DNA phosphorus atoms harmonically restrained (okay_spring = 0.2 kcal mol^–1 Å^–2). Mobility measurements have been carried out utilizing a subject of 5 mV nm⁻¹ or a hydrostatic stress of ~1.3 bar nm⁻¹ parallel or transverse to the helical axis, with every situation using 4 replicate simulations lasting a complete of 400 (neutralizing Na⁺) to 4,000 ns (3 M). Besides the place in any other case specified, a 100 mV nm⁻¹ electrical subject was utilized to the system at a 35° angle with respect to the DNA whereas a centre of mass colvar restrained the DNA (okay_spring = 500 kcal mol^–1 Å^–2), an RMSD colvar retained an idealized DNA configuration (okay_spring = 100 kcal mol^–1 Å^–2) and a spin angle colvar (okay_spring = 100 kcal mol^–1 °^–2) prevented rotation of the DNA and reported on the torque. Eight replicate programs have been employed throughout simulations lasting a complete of 1,040 ns (100 mM) or 2,055 ns (3 M). The stream and focus of ions and water oxygen atoms have been analysed by binning the system into ~1 Å voxels, counting the flux by way of and focus in every voxel utilizing a centred finite-difference approximation for the flux. The distinction in focus between sodium and chloride ions offered the online native cost density of the fluid across the DNA in every case. Multiplying this cost by the electrical subject offered the solvent power. In subsequent simulations, the 3D map of the solvent power was used to use a position-dependent power to every water oxygen atom utilizing the TclBC characteristic of NAMD and using the approximation that the density of water oxygen atoms is uniformly 33 nm⁻³. Once more, eight replicate programs have been employed for simulations lasting a complete of 480 and 570 ns for 100 mM and three M situations, respectively.

Statistics and reproducibility

No statistical technique was used to predetermine the pattern dimension.