Construction of Reconfigurable and Polymorphic DNA Origami Assemblies with Coiled‐Coil Patches and Patterns

Abstract DNA origami nanodevices achieve programmable structure and tunable mechanical and dynamic properties by leveraging the sequence‐specific interactions of nucleic acids. Previous advances have also established DNA origami as a useful building block to make well‐defined micron‐scale structures through hierarchical self‐assembly, but these efforts have largely leveraged the structural features of DNA origami. The tunable dynamic and mechanical properties also provide an opportunity to make assemblies with adaptive structures and properties. Here the integration of DNA origami hinge nanodevices and coiled‐coil peptides are reported into hybrid reconfigurable assemblies. With the same dynamic device and peptide interaction, it is made multiple higher‐order assemblies (i.e., polymorphic assembly) by organizing clusters of peptides into patches or arranging single peptides into patterns on the surfaces of DNA origami to control the relative orientation of devices. The coiled‐coil interactions are used to construct circular and linear assemblies whose structure and mechanical properties can be modulated with DNA‐based reconfiguration. Reconfiguration of linear assemblies leads to micron scale motions and ≈2.5‐10‐fold increase in bending stiffness. The results provide a foundation for stimulus‐responsive hybrid assemblies that can adapt their structure and properties in response to nucleic acid, peptide, protein, or other triggers.


DNA origami folding and purification
The DNA origami hinge was designed as combination of an 8064 nt ssDNA scaffold [1] and staple strands by using software caDNAno [2] .The design diagram from caDNAno is included in Supplementary Figure S1, illustrating all design details and the sequences for all staples are included in Table S1.Briefly, the parts of overhangs used for binding coiled-coil peptides conjugates are fully complementary to the DNA handles in conjugates.For AEI peptides, the overhang A* sequence is GTAATACCAGATGG and for BKI peptides, the overhang B* sequence is TATATGGTCAACTG.In addition, overhangs on the inner side of arms in self-assembly design are used for binding close strands to actuate the structures.The closing strand sequence is AGTGGACCAGTGGGTCTTCGTATAGACCCGACTTTGGGCCTAAGTGGGTCCACACGCACG.The scaffold was prepared in our laboratory [ref], and the staple strands were ordered from a commercial vendor (IDT, Coralville).
Folding reactions contained 20 nM scaffold and 200 nM of each staple strand in a ddH2O solution containing 1× FOB and 20 mM MgCl2.This folding reaction was subjected to thermal annealing in a thermal cycler (Bio-Rad, Hercules, CA) consisting of rapidly heating the solution to 70°C for 15 min, followed by annealing over the range of 63-57°C for 3 hours per degree Celsius, and then cooling for 30 min at 4°C.The structures were purified by centrifugation in a polyethylene glycol (PEG) solution [3] .The production of folding reaction with folded DNA origami structures was mixed by an equal volume of PEG buffer (15% PEG MW8000, 200 mM NaCl and 100 mM Tris).The mixture was then centrifuged at 4 °C and 16000g for 30 min.The supernatant was removed, and the origami structures were resuspended in 1× FOB with 20 mM MgCl2.Then the concentration of the structure in the solution was measured by Nanodrop (NanoDrop 2000C Spectrophotometer, Thermo Scientific) in 260 nm absorbance.

Actuation
Purified individual structures with group 1 or group 2 overhangs were mixed with 20× excess AEI and BKI conjugates relative to the structure concentrations, and then the final concentration of structures was adjusted to 5 nM with buffer containing 1× FOB and 20 mM MgCl2 and final volume is 50 μl.The mixture was subjected to an incubation of 15 hours at 37 ⁰C.

TEM and analysis
The sample was diluted to 1 nM in 0.5× TBE with 10 mM MgCl2 buffer for transmission electron microscopy (TEM).For TEM grid preparation, 4 μl of sample volume was deposited on Formvar-coated copper TEM grids, stabilized with evaporated carbon film (Ted Pella; Redding, CA).Then the sample was incubated on the grid for 4 min when incubating single structure, and for 8 min when incubating selfassembly structures, and then was wicked away with filter paper.The sample was then stained with 2% uranyl formate (SPI, West Chester, PA, USA).First, a 10 μL drop was applied for 2 s and wicked away to wash the sample, and then another 10 μL drop was applied for 15 s and then wicked away with filter paper.TEM imaging was performed at the OSU Campus Microscopy and Imaging Facility on an FEI Tecnai G2 Spirit TEM using an acceleration voltage of 80 kV at different magnifications.
The raw TEM images were organized into a gallery containing clear formations (Figure S2-S6,S26-S28) by using the particle picking tool in software EMAN2.Then the angles were measured in the software ImageJ by drawing two straight lines directly on each particle along the inner side of each arm in one hinge.
We used MATLAB as the postprocessing tool to convert the angle data sets to probability density histograms.To estimate the fraction of the hinge in non-actuated or actuated states, we used the peakfit Matlab function program [4] , which uses a non-linear optimization algorithm to decompose a complex, overlapping-peak signal into its component parts, by assuming the conformational distribution consisting of two populations.The peak number was set as 2 for the actuated sample and the distribution was fit with a combination of two Gaussian distribution, termed as the actuated and non-actuated parts.

Self-assembly
Purified individual structures with each designed overhang arrangement were mixed with 20× excess AEI and BKI conjugates relative to the structure concentrations, and then the final concentration of structures was adjusted to 5 nM with buffer containing 1× FOB and 20 mM MgCl2 and final volume is 50 μl.For the linear self-assembly polymer, the mixtures were then subjected to a 1-cycle low temperature annealing ramp starting at 45 °C followed by an anneal phase at −2 hours per °C until 20 °C, to prevent extensive aggregation.For other self-assembly structures, the process was 2-cycle annealing ramp, which was repeated twice.In Figure 2 and Figure 3 A, we used 2 cycle annealing ramp.However, for comparison of assembly with between DNA and coiled-coil peptides and polymers actuation, we used 1 cycle annealing ramp to avoid aggregation.

Circle pattern actuation
Before self-assembly process, the purified structures were mixed with 0.5 μl of 10μM closing strands and then incubated at 37 ⁰C for 15 hours to yield an angle of hinges mostly actuated to 45⁰.

Linear polymer actuation
After self-assembly process, 0.5 μl of 10μM closing strands were added to the production.And then the mixture was incubated to 37 ⁰C for 15 hours.

Persistence length from shape variance in TEM images
TEM images of non-actuated and actuated self-assembly polymers were analyzed using MATLAB.To discretize the shape of the polymers, 11 points on the connection of the hinge along the trajectory were manually selected by clicking on the image Figure S17A and Figure S18A These selected points were used to fit a cubic spline of the trajectory coordinates along the filament path to obtain fine resolution of the curvature.Configurational distributions were obtained by aligning the filament trajectories so that they started at the origin and initially pointed in the horizontal direction, as in Figure 3E The average transverse fluctuations were determined as a function of arc length from the filament configurational distributions.The LP of the actuated and non-actuated self-assembly polymer were characterized respectively by calculating the average of the transverse fluctuations squared from the configurational distributions and fitting Eq. (1).
Materials and supplies: Fmoc-protected amino acids for peptide synthesis were purchased from EDM Millipore.Fmoc-azidolysine was purchased from Combi-Blocks Inc. Dichloromethane (DCM) was purchased from Millipore Sigma.Dimethylformamide (DMF) and diethyl ether were purchased from Oakwood Chemical Inc.Piperidine was purchased from Alfa Aesar.Oxyma and DIC were purchased from ChemImpex.TFA was purchased from Oakwood, and Rink Amide resin was purchased from Novabiochem.DBCO-sulfo-NHS linker was purchased from Click Chemistry Tools.All oligonucleotides were purchased from Integrated DNA Technologies.
Peptide synthesis and characterization: Peptides were obtained using solid phase peptide synthesis (SPPS) on a CEM Liberty Blue instrument.Synthesis was performed on a solid phase Rink-Amide resin (0.78 mmol/g) at a 0.1 mmol scale, using a standard Fmoc protocol and deprotected in 20% piperidine in DMF.Amino acids, coupling agents, DIC and Oxyma, were added in a 10-fold molar excess.Crude peptides were cleaved by shaking the resin by in a solution containing trifluoroacetic acid (TFA), triisopropylsilane (TIPS), and water in a ratio of 95:2.5:2.5 for 3.5 h.The resin was washed with TFA and concentrated under nitrogen.The solution was then added to 40 mL of cold diethyl ether to precipitate the peptide.The solution was centrifuged at 4200 rpm for 10 min, the supernatant was removed, and the pellet was allowed to dry overnight.The dried pellet was dissolved in a mixture of water and acetonitrile (50:50) and 0.1% TFA.Peptides were purified via reverse phase chromatography on a Waters HPLC using a Phenomenex column with C18 resin.A linear gradient was generated using water/acetonitrile + 0.1% TFA, from 10% to 100% acetonitrile over 50 minutes.Peak fractions were collected based upon their absorbance at 230 nm and tested for purity by MALDI-TOF mass spectrometry on a Bruker Microflex LRF MALDI using α-cyano-4-hydroxycinnamic acid matrix (Sigma).Pure fractions were pooled and lyophilized, and peptides were stored at -20 °C until use.
Synthesis of Peptide-DNA Conjugates DNA-peptide conjugates were prepared via strain-promoted azidealkyne cycloaddition (SPAAC) following reported protocols [5] .Briefly, amine modified oligonucleotides were dissolved in phosphate buffered saline (PBS) to a concentration of 1 mM.A 10 molar excess of DBCOsulfo-NHS dissolved in DMSO was then added to the DNA and agitated at RT overnight.The reaction mixture was washed six times with a 3 kDa molecular weight cutoff (MWCO) filter (Amicon) to remove any excess DBCO.The resulting DNA was purified by reverse phase HPLC on an Agilent 1220 Infinity LC HPLC with a Zorbax Eclipse XDBC18 column.A Linear gradient was generated using 50 mM TEAA/Methanol from 10% to 100% methanol over 60 minutes.Peak fractions were collected based upon their absorbance at 260 nm and tested for purity by MALDI-TOF mass spectrometry on a Bruker Microflex LRF MALDI in 3-Hydropicolinic acid (HPA) matrix (Sigma).Pure fractions were pooled, and buffer exchanged into water using a 3 kDa MWCO filter.The purified DBCO-DNA was then mixed with the azido-peptides, heated to 37 °C, and shaken overnight.Following the reaction, samples were exchanged into water using a 3 kDa MWCO filter, and purified by reverse phase HPLC, similar to the modified DBCO-DNA.

Figure S1 .
Figure S1.The caDNAno schematics of DNA origami hinge.The labeled areas are group 1 and 2 overhangs for binding AEI and BKI peptides to actuate the structure.

Figure S2 .
Figure S2.Example TEM images and analysis of non-actuated DNA origami hinge.A) Example TEM images.B) The angular distributions obtained from TEM results.C) Single peak fitting of the angular distribution by using a Matlab program.The sample size is 852.

Figure S3 .
Figure S3.Example TEM images and analysis of actuated DNA origami hinge when N-term coiled-coil peptides binding group 2 overhangs.A) Examples of TEM figures.B) The angular distributions obtained from TEM results.C) Double peaks fitting of the angular distribution.The lower peak is 32.7⁰ and the higher peak is 62.7⁰.The sample size is 687.

Figure S4 .
Figure S4.Example TEM images and analysis of actuated DNA origami hinge when N-term coiled-coil peptides binding group 1 overhangs.A) Example TEM images.B) The angular distributions obtained from TEM results.C) Double peaks fit of the angular distribution.The lower peak is 49.2⁰ and the higher peak is 79.1⁰.The sample size is 898.

Figure S5 .
Figure S5.Example TEM images and analysis of actuated DNA origami hinge when C-term coiled-coil peptides binding group 1 overhangs.A) Example TEM images.B) The angular distributions obtained from TEM results.C) Double peaks fitting of the angular distribution.The lower peak is 29.0⁰ and the higher peak is 54.11⁰.The sample size is 670.

Figure S6 .
Figure S6.The caDNAno schematics of DNA origami hinge for self-assembling.

Figure S7 .
Figure S7.Example TEM images for dimers of DNA origami hinge linked by coiled-coil peptides.A) The overhangs are set on the inner side of the arm.B) The overhangs are set on the outer side of the arm.There are two different assemble directions when two hinge binding together.All scale bars are 50nm.

Figure S8 .
Figure S8.Representative TEM images of unrestricted self-assembly with coiled-coil peptides.Both polymers and circle patterns can be observed.All scale bars are 500nm.

Figure S9 .
Figure S9.Representative TEM images of linear polymers self-assembly with coiled-coil peptides.All scale bars are 500nm.

Figure S10 .
Figure S10.Length analysis of linear polymers in TEM images.This polymerization was carried out with hinges at 5 nM concentration.

Figure S11 .
Figure S11.Representative TEM images of linear polymers self-assembly with replacement DNA strands.In linear polymer self-assembly with coiled-coil peptides, 20x access AEI and BKI were added to 5nM origami structure, and in the comparison experiment only with replacement DNA strands, 20x access of single strand DNA was added to 5nM structure.All scale bars are 500nm.

Figure S12 .
Figure S12.Experiments for assembly with different access of either DNA binding strands and peptide conjugates.In experiments, the DNA origami hinge concentration was kept at 1nM and for each experiment, DNA and peptides concentrations (excess relative to the complementary binding sites is denoted in parenthesis, there are 12 overhangs for each peptide) were set as 24nM (2X), 120nM (10X), 240nM (20X), 480nM (40X) and 1200nM (100X).The results show comparing with coiled-coil peptide conjugates, DNA can cause more aggregation even under same polymerization strand conditions.Length distributions were measured for isolated hinge assemblies (i.e.not part of a larger aggregate).These results showed that at peptide interactions yield longer unaggregated polymers, and at higher concentrations both excess DNA and coiled-coil peptides block the assembly binding sites in structures leading to a decrease in the length of polymers.This blocking occurs sooner for DNA sticky ends compared to the peptides.All scale bars are 500nm.Additional TEM images for all conditions are shown in Supplemental FiguresS13-S22.

Figure S13 .
Figure S13.Example TEM images for assembly with coiled-coil peptides with relative excess of 2X.All scale bars are 500nm.

Figure S14 .
Figure S14.Example TEM images for assembly with coiled-coil peptides with relative excess of 10X.All scale bars are 500nm.

Figure S15 .
Figure S15.Example TEM images for assembly with coiled-coil peptides with relative excess 20X.All scale bars are 500nm.

Figure S16 .
Figure S16.Example TEM images for assembly with coiled-coil peptides with relative excess 40X.All scale bars are 500nm.

Figure S17 .
Figure S17.Example TEM images for assembly with coiled-coil peptides with relative excess 100X.All scale bars are 500nm.

Figure S18 .
Figure S18.Example TEM images for assembly with DNA binding strand with relative excess 2X.All scale bars are 500nm.

Figure S19 .
Figure S19.Example TEM images for assembly with DNA binding strand with relative excess 10X.All scale bars are 500nm.

Figure S20 .
Figure S20.Example TEM images for assembly with DNA binding strand with relative excess 20X.All scale bars are 500nm.

Figure S21 .
Figure S21.Example TEM images for assembly with DNA binding strand with relative excess 40X.All scale bars are 500nm.

Figure S22 .
Figure S22.Example TEM images for assembly with DNA binding strand with relative excess 100X.All scale bars are 500nm.

Figure S23 .
Figure S23.Representative TEM images of failed circle pattern self-assembly with coiled-coil peptides.It does not form clear circle pattern as design and both same and opposite direction assemblies happen.All scale bars are 500nm.

Figure S24 .
Figure S24.Representative TEM images of complexed overhangs design of circle pattern self-assembly with coiled-coil peptides.Clear circle patterns can be observed.All scale bars are 500nm.

Figure S25 .
Figure S25.A) Schematics of distances between binding sites and peptides.B, C) Schematics and TEM images for normal circle pattern assembly.D, E) When the binding sites are all extended with 15 nucleotides polyT, the efficiency of assembly becomes lower.We demonstrate that it is caused by flexibility of binding sites make neighbor coiled-coil peptides binding each other.Scale bars are 100nm.

Figure S26 .
Figure S26.Example TEM images of actuated hinge with average angle 41⁰ before self-assembly.B) The angular distributions obtained from TEM results.C) Double peaks fit the angular distribution.The lower peak is 45⁰.

Figure S27 .
Figure S27.Example TEM images showing self-assembly of the circular pattern with A) free hinges and B) hinges reconfigured to 45⁰.All scale bars are 50nm.

Figure S28 .
Figure S28.Example TEM images of actuated hinge which is used in self-assemble linear polymer.B) The angular distributions obtained from TEM results.C) Double peaks fit the angular distribution.The lower peak is 15⁰.

Figure S29 .
Figure S29.Example TEM images of linear hinge polymers before and after reconfiguration.All scale bars are 500nm.

Figure S30 .
Figure S30.Persistence length measurements of simulations of unactuated, partially actuated and fully actuated linear polymers.A) Examples of 10 hinges assembled polymers with unactuated and actuated hinge.Partially actuated polymer means the simulation was built on the angle distribution of actuated polymers and fully actuated polymer means the simulation was built on the angle only picked up from angle less than 30 degrees in the same distribution.B) The transverse fluctuations were fit to equation (1) to give a persistence length.

Table S1 :
Oligonucleotide sequences for DNA origami design.