Multiple Wavelength Photopolymerization of Stable Poly(Catecholamines)‐DNA Origami Nanostructures

Abstract The synthesis of multicomponent polymer hybrids with nanometer precision is chemically challenging in the bottom‐up synthesis of complex nanostructures. Here, we leverage the fidelity of the DNA origami technique to install a multiple wavelength responsive photopolymerization system with nanometer resolution. By precisely immobilizing various photosensitizers on the origami template, which are only activated at their respective maximum wavelength, we can control sequential polymerization processes. In particular, the triggered photosensitizers generate reactive oxygen species that in turn initiate the polymerization of the catecholamines dopamine and norepinephrine. We imprint polymeric layers at designated positions on DNA origami, which modifies the polyanionic nature of the DNA objects, thus promoting their uptake into living cells while preserving their integrity. Our herein proposed method provides a rapid platform to access complex 3D nanostructures by customizing material and biological interfaces.


Materials and instruments
All solvents and chemicals were purchased from commercial sources and were used without further purification. DNA oligonucleotides (staple strands, sticky strands, G4 staple strands, G4-complementary strands, and folding strands) were purchased from Sigma-Aldrich. M13mp18 plasmid DNA was purchased from tilibit nanosystems. Annealing of DNA origami structures and hybridizing of sticky sequences was performed on a Bio-Rad MyCycler TM Thermal Cycler. UV/Vis and fluorescence spectroscopy were conducted on a Spark® Multimode Microplate Reader by Tecan. Buffers were adjusted on a Mettler Toledo SevenExcellence TM equipped with the InLab Nano ph electrode. Photopolymerizations were conducted in a house-built photobox that is equipped with interchangeable LED-arrays (96 LEDs) of 410 nm, 525 nm, and 625 nm.

DNA origami nanostructures
DNA origami tubes were prepared by mixing M13mp18 Scaffold DNA (1 equiv.), staple strands (8 equiv.), G4 staple strands (8 equiv.), stickyA strands (8 equiv., if necessary), and folding strands (16 equiv.) in origami buffer (1 mM Na2EDTA, 5 mM NaCl, 5 mM TRIS, 12 mM MgCl2, pH 8). Annealing was performed by running a program from 70 °C to 20 °C over 2 h (0.5 °C/min to 35 °C, 1 °C/min to 20 °C) and the obtained DNA origami structures were purified by PEG precipitation. [1] Therefore, the PEG solution (15% PEG8000 (w/v), 5 mM TRIS, 1 mM Na2EDTA, 505 mM NaCl) was added to the reaction solution at a volume ratio of 1:1 and centrifuged for 25 min at 12.5 rpm, room temperature (RT). The supernatant was removed, and the resuspended pellet was precipitated by applying the PEG precipitation method for additional two times. Sample concentration was determined by Spark® 20M with Nanoquant plate TM . DNA origami structures were stored in origami buffer at 4 °C.

Standard photopolymerization on DNA origami tubes
Standard photopolymerization on DNA origami tubes (44x G4 catalytic centers per tube) was carried out in a total reaction volume of 50 µL in a UV-star 384 well plate. Stock solutions of compounds were prepared in reaction buffer (10 mM BIS-TRIS, pH 6.5) at the following concentrations: photosensitizers (PS) at 10 µM; monomers at 0.5 M. A typical reaction was conducted as follows: DNA origami with G4-sequences (10 nM, final concentration) was incubated with PS (1.5 equiv., relative to the amount of G4-sequences) in reaction buffer for 30-60 min. Monomer (10 mM, final concentration) was added and the plate was placed in a house-built photobox. After a predetermined irradiation time interval, polymerization was stopped by switching off the light source and polymer-DNA objects were purified using size exclusion chromatography (200 µL Sephacryl S-400 HR, equilibrated with reaction buffer; centrifuge settings: 2 min, 0.8 g, RT) or 100K spin filtration (5 g, 10 min, 2 times; recovered at 1 g, 2 min). For purification steps, a mixture of origami buffer and reaction buffer ratio (1:4) was utilized.

Two-step photopolymerization on DNA origami tubes
Layer-by-layer polymerization: DNA origami tubes were equipped with one ring of G4 sequences (44 sequences) and incubated with MB under standard conditions (see above) for 30 min. NE was added and the reaction solution was subjected to standard polymerization conditions at 625 nm for 2 h. After 100K spin filtering, DA was added and irradiated for 2 h at 625 nm. Polymer-ringed origami tubes were recovered by 100K spin filtering and stored at 4 °C. Ring-by-ring polymerizations: DNA origami tubes were equipped with one ring of G4 sequences and on ring of stickyA sequences (44 sequences each). Methylene blue was incubated with G4 sequences on origami tubes in an 1.5 molar excess (related to the amount of G4 sequences) for 30 min. Eosin Y was preloaded into G4 complementary strands in an 1.5 molar excess for 30 min and either annealed with MB-loaded origami tubes before photopolymerization or after the first irradiation phase. Hybridizing of EY-G4 strands (1.5 equiv. compared to stickyA sequences on DNA origami tubes) was performed by a temperature ramp from 35 °C to 20 ° in 5 °C steps (holding each step for 15-30 min). After annealing, excess staple strands were removed by 100K spin filtering. For continuous photopolymerization, MB-and EY-loaded tubes were incubated with DA (standard polymerization conditions, see above) and irradiated at 625 nm and 525 nm for 3 h each. For decoupled photopolymerization, MB-loaded tubes were incubated with NE and subjected to standard photopolymerization at 625 nm for 2 h. After purification (100K spin filtering) and annealing with EY-G4 strands, DA was added, and polymerization continued at 525 nm for 2 h. In both approaches, polymer-ringed origami tubes were purified by 100K spin filtering and stored at 4 °C.

Reactive Oxygen Species Assay (ROS Assay)
Reactive oxygen species assay of photosensitizers (PS) was conducted in a total reaction volume of 150 µL in a 96 well plate. Wells were charged with PS (6.6 µM), N,N-dimethyl-4-nitrosoaniline RNO (70.7 µM) and imidazole (7.5 mM) in BIS-TRIS buffer (pH 6.5) and exposed to the respective wavelength or kept in dark. Absorbance scans from 300-500 nm were conducted at several time points within 30 min. The decrease in absorbance of RNO could be followed at 438 nm.

Cell experiments
A549 cells were cultured in DMEM medium containing 10% fetal bovine serum and 1% penicillin/streptomycin as well as 1% MEM nonessential amino acids. Cells were seeded into confocal well plates and left for 24h to adhere at 37 °C, 5% CO2.

Cell uptake
For cell uptake studies, DNA origami tubes with a ring of either pDA or pNE were synthesized (according to standard polymerization protocol, MB, 625 nm, 3 h). Furthermore, origami tubes were equipped with 22 sticky sequences for Alexa-647® oligonucleotide labelling ("sticky C"). Dye-labelling was conducted in a 10-fold molar excess by a 1 hour temperature ramp (from 35 °C to 20 ° in 5 °C steps), followed by 100K spin filter purification. Control samples without polymer coating were labelled in the same way. Cells were seeded at a density of 5,000 cells/well in an 10-well confocal well plate. After adhering for 24 h, cells were treated with the Alexa647-labeled sample for 24h at 37 °C. Sample solutions were diluted to 20 nM (Alexa647 concentration via a standard curve) in cell media for incubation with the cells. After the incubation time, the media/sample solution was replaced with fresh media. Cells were then imaged by confocal laser scanning microscopy (SI Figure 12).

Cell colocalization
Besides the polymer ring, DNA origami tubes were additionally equipped with 22 sticky sequences each for Alexa-647® and Alexa-488® oligonucleotide labelling (standard protocol, see above; "stickyC for Alexa-647® and stickyA for Alexa-488®). Control samples without polymer coating were labelled in the same way. Cells were seeded at a density of 5,000 cells/well in an 10-well confocal well plate. After adhering for 24 h, cells were treated with the double-labeled sample for 24 h at 37 °C. Sample solutions were diluted to 10 nM (Alexa647 concentration via a standard curve) in cell media and incubated with the cells. After the incubation time, the sample solution was replaced with fresh media. Cells were then imaged by confocal laser scanning microscopy ( Figure 4 and SI Figure 13).

Cell medium stability assay
To test the stability of the employed DNA origami structures under cell medium conditions, samples were incubated with cell medium in an 1:1 ratio (approx. 50 fmol, total volume of 10 or 20 µL) for 24 h at 37 °C in a thermocyler with a heated lid. After incubation time, total sample volume was loaded on gel for AGE. For comparison, an analogue series of samples was mixed with cell medium right before loading onto the gel.

DNase I digestion assay
To digest DNA origami tubes that were only attached to the cell's exterior after overnight incubation, DNase I (20 U) was added to the samples and incubated for 1 hour at 37 °C prior to imaging.

Confocal Laser Scanning Microscopy
Cells were imaged on a Leica TCS SP5 and a Visitron Spinning Disc microscope with an argon laser for excitation at 488 nm for Alexa488 (emission 498-540 nm), and a HeNe laser for excitation at 633 nm for Alexa647 (emission 657-757 nm). Z-stack images was acquired using a Leica Stellaris 8 confocal microscope equipped with a white light laser with tunable excitation wavelengths between 440 -790 nm. Alexa 488 is excited at 488 nm and emission was collected using the HyD ® X detector at 498-540 nm. Alexa 647 is excited at 647 nm and emission was collected using the HyD ® R detector at 657-757 nm.

Atomic Force Microscopy (AFM)
Atomic force microscopy was performed in liquid state with a Bruker Dimension FastScan Bio AFM equipped with the ScanAsyst mode. Sample solution (30 µL, 1-2 nM in origami buffer) was added onto a freshly cleaved mica substrate and incubated for 5 min to allow deposition of the origami structures. Remaining solution was removed and 300 µL origami buffer was applied onto the mica surface, forming a droplet for measuring in liquid. Samples were scanned with scan rates between 1 and 2 Hz. Images were processed with NanoScope Analysis 1.8.

Agarose gel electrophoresis (AGE)
Agarose gel electrophoresis was performed on 1% agarose gels (TBE, stained with EtBr, purchased from Bio-Rad or self-casted with 1 × TBE when no staining was desired), equipped with 8 wells. The gels were run on ReadySub-Cell GT Cells from Bio-Rad using 1 × TBE buffer as the running buffer. DNA Gel Loading dye (6 ×) was used for sample preparation (approximately 50 fmol origami) with a total volume of 6, 12, or 18 µL, depending on sample concentration. "GeneRuler DNA Ladder Mix" (100-10000 bp) was used as for the marker. Electrophoresis was conducted at 90 V for 60-100 minutes at 4 °C. Image was taken with G:BOX Chemi Gel Doc System from Syngene or under UV-excitation with a camera.

DNA extraction from agarose gel
For DNA extraction, TBE gels were casted and run as mentioned above. Areas below the wells were defined and cut according to the band pattern found under UV irradiation. The excised gel piece was placed into a Costar® Spin-X® Centrifuge Tube Filter (0.45 µm pore NY membrane), left at -20°C for 1 hour and subsequently centrifuged at 10 g for 10 minutes. 100 µL of the filtrate was placed into a black 384 well plate and subjected to fluorescence measurements.

Dynamic light scattering (DLS)
DLS measurements were performed at 25 °C using a Malvern ZetaSizer Nano S from Malvern Instruments Ltd. with a He/Ne Laser (λ = 633 nm) at a fixed scattering angle of 173°. All measurements were performed in triplicate. The obtained data was processed by cumulant fitting for Dz and PDI, or by CONTIN fitting for intensity-weighted particle size distribution. For DLS, triple amount of DNA origami (compared to standard polymerization) was photopolymerized (MB, 625 nm), purified and brought to 150 µL with reaction buffer. As a control, 10 µL bare origami tubes (50 nM) were diluted to 150 µL with reaction buffer. Samples were filtered prior to measurement through PFTE (hydrophilic) syringe filters (0.45 μm pore size).  Importantly, the polymeric structures are not only built on covalent bonds, but also non-covalent interactions, supramolecular assemblies, charge transfer and (cation-) π-stacking are present. In contrast to pDA, pNE formation is characterized by the occurrence of 3,4-dihydroxybenzaldehyde (DHBA), that can react with the monomer NE again, forming DHBA-NE. This molecule is said to be responsible for the ultrasmooth surface of pNE.      After annealing of scaffold DNA (M13mp18) and an excess of staple strands, the folded DNA origami tubes are apparent as a new band of lower mobility (A2,3); excess of staple strands is removed by purification. Polymer-modification (pDA and pNE) is further visible by a changed running behavior compared to the precursor origami band (Gel B). The polymer bands also show some smearing effect, potentially indicating aggregation of the structures. Especially in the case of pDAmodification, we sometimes also observe the retaining of some material in the well of the gel, also demonstrating the presence of aggregates (gels not shown here). 1 % agarose TBE gels stained with EtBr were used, and run at 90 V for 60 min (Gel A) and 95 min (Gel B) at 4 °C. Please note: Gel (B) is cropped to only show lanes of interest, however, no stretching or shrinking was done, so bands are still comparable.

Figure 10
Stepwise layer-by-layer polymerization of the two monomers norepinephrine and dopamine on DNA origami tubes that are activated with MB and irradiated at 625 nm. UV/Vis spectroscopy confirms formation of (A) polynorepinephrine in the first irradiation phase and (B) polydopamine in the second irradiation phase. Spin filtration was performed after step 1 to exchange monomers.

Figure 11
Control experiment for layer-by-layer polymerization, showing that it is in fact pDA that is formed in step 2, not pNE. (A) After pNE is generated in the first irradiation period, NE is removed from the reaction solution and no dopamine is added for the 2 nd irradiation phase. (B) UV/Vis spectroscopy shows that no polymer is formed in step 2 and in AFM imaging, only height increase for pNE rings from step 1 is found.

Figure 15
Cell medium stability assay of bare origami tubes as well as polymer-coated ones. Samples were incubated at 37 °C for 24 h and compared to samples that were mixed with cell medium prior to loading the gel. Even the bare origami tubes do show a certain stability under the applied incubation conditions, but also fragmentation. Polymer-ringed origami samples show less fragmentation, especially for pDA, material is stuck in the well, most likely caused by retained aggregates. A 1 % agarose TBE gel stained with EtBr were used, and run at 90 V for 90 min at 4 °C.

Figure 19
Characterization of dual-labelled origami structures for cell studies by agarose gel electrophoresis and subsequent fluorescence measurements. (A) AGE was performed with the Alexa-oligonucleotides, the unpurified as well as the purified origami sample after modification with both fluorophores. Gels were visualized on a UV plate without further staining. Whereas the free dyes show their respective emission color or a mixture thereof in the lower part of the gel; fluorescent origami structures are visible in the upper part of the gel. Excess of Alexa-oligonucleotides for origami modification was removed by purification. (B) A gel was loaded with both Alexa488-and Alexa647-oligonucleotides (lane 1) and purified, dual-labelled origami sample (lane 2). The areas right below the wells were cut, frozen, centrifuged, and the filtrate was subjected to subsequent fluorescence measurements in a Tecan Plate Reader. (C) Excitation of Alexa488 and Alexa647, respectively, revealed the presence of the Alexa fluorophores in the origami sample but not in the control sample, demonstrating successful labelling with both dyes. Emission spectra of Alexa488 and Alexa647 were recorded from 503-600 nm after excitation at 480 nm and from 662-760 nm after excitation at 640 nm (bandwidths: 10 nm), respectively, and smoothed. Tube with 1 central G4 ring and stickyC for dye labelling: G4 staple strands on positions 2-10, 11-20, 21-24, 26, 112-120, 121-130, 131; stickyC strands on positions 161-170, 171-179; folding strands on positions 1,25,27,28,51,52,75,76,99,100,111,132,133,156,157,180,181,204,205,216; staple strands on remaining positions.