suementary inrmatin.10/.2628 suementary inrmatin 4 ii. supplementary methods viscosity measurements:...

NATURE CHEMISTRY | www.nature.com/naturechemistry 1

SUPPLEMENTARY INFORMATIONDOI: 10.1038/NCHEM.2628

1

A viscous solvent enables information transfer from gene-length nucleic acids in a model prebiotic replication cycle

Christine He1*, Isaac Gállego2*, Brandon Laughlin2, Martha A. Grover1, and Nicholas V. Hud2

1School of Chemical & Biomolecular Engineering, 2School of Chemistry & Biochemistry, and Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, GA, USA.

*These authors contributed equally to the work.

Supporting Information Table of Contents

I. Supplementary Materials……………………………………………………………………………………………….2

II. Supplementary Methods………………………………………………………………………………………………..4

III. Supplementary Discussion……………………………………………………………………………………………..5

IV. Supplementary Figures…………………………………………………………………………………………………..7

V. Supplementary Tables……………………………………………………………………………………………………13

1

A viscous solvent enables information transfer from gene-length nucleic acids in a model prebiotic replication cycle

Christine He1*, Isaac Gállego2*, Brandon Laughlin2, Martha A. Grover1, and Nicholas V. Hud2

1School of Chemical & Biomolecular Engineering, 2School of Chemistry & Biochemistry, and Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, Atlanta, GA, USA.

*These authors contributed equally to the work.

Supporting Information Table of Contents

I. Supplementary Materials……………………………………………………………………………………………….2

II. Supplementary Methods………………………………………………………………………………………………..4

III. Supplementary Discussion……………………………………………………………………………………………..5

IV. Supplementary Figures…………………………………………………………………………………………………..7

V. Supplementary Tables……………………………………………………………………………………………………13

© 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved.



2

I. Supplementary Materials Glycholine preparation: Choline chloride (Acros Organics, 99%) was purified by recrystallization from absolute ethanol and then dried under vacuum (≤130 mTorr). Glycerol (Sigma, 99%) and choline chloride were separately placed in a lyophilizer (Virtis) with a liquid nitrogen cold trap for 12 hours before being mixed in a 4:1 molar ratio. After mixing, glycholine ingredients were placed in an 80 oC oven until fully liquefied, then dried in a vacuum centrifuge (Labconco) for at least 12 hours before use to remove residual water. The water content of glycholine prepared in this manner was measured to be 0.07% (wt/wt) using Karl-Fischer titration1. DNA and RNA sequences: Hairpin DNA (Step 2): 5'-GCAAAACGAAGTTTTGC-3' 32-mer DNA duplex (Step 2):

5'-GGTGTCAGTAAGCCATTCGAGATCCTCATAGT-3'3'-CCACAGTCATTCGGTAAGCTCTAGGAGTATCA-5'

3kb DNA duplex: pBluescript II SK(-) vector DNA, phagemid excised from lambda ZAPII, GenBank accession: X52330.1 L5: nt 2568-2599 5’-AGTTGCTCTTGCCCGGCGTCAATACGGGATAA-3’ L4: nt 2600-2631 5’-TACCGCGCCACATAGCAGAACTTTAAAAGTGC-3’ L3: nt 2632-2663 5’-TCATCATTGGAAAACGTTCTTCGGGGCGAAAA-3’ L2: nt 2664-2695 5’-CTCTCAAGGATCTTACCGCTGTTGAGATCCAG-3’ L1: nt 2696-2727 5’-TTCGATGTAACCCACTCGTGCACCCAACTGAT-3’ F0: nt 2728-2759 5’-CTTCAGCATCTTTTACTTTCACCAGCGTTTCT-3’ R1: nt 2760-2791 5’-GGGTGAGCAAAAACAGGAAGGCAAAATGCCGC-3’ R2: nt 2792-2823 5’-AAAAAAGGGAATAAGGGCGACACGGAAATGTT-3’ R3: nt 2824-2855 5’-GAATACTCATACTCTTCCTTTTTCAATATTAT-3’ R4: nt 2856-2887 5’-TGAAGCATTTATCAGGGTTATTGTCTCATGAG-3’ R5: nt 2888-2919 5’-CGGATACATATTTGAATGTATTTAGAAAAATA-3’ For sequences L5-R5 listed above, the second column refers to the nucleotide position on the pBlueScript II SK(-) sense strand. The sequence of the 545-mer duplex extends from nucleotide position 2547 and to nucleotide position 130 (nucleotide positions given in reference to the sense strand of pBlueScript SK II (-)). RNA sequences are identical to those of their DNA counterparts, with U instead of T. In the main text, RNA sequences are noted with an “r” preceding the name (i.e. rL5-rR5). 3 kb DNA preparation: The plasmid pBluescript II SK(-) (Agilent) was prepared using a Qiagen plasmid preparation kit and linearized with HindIII-HF (New England Biolabs) , which cuts at nucleotide 689 on the sense strand of pBlueScript II SK(-). Completeness of the linearization reaction was confirmed by agarose gel electrophoresis.




3

545-mer DNA duplex preparation: The 545-mer DNA was produced by polymerase chain reaction (PCR) with the 3 kb DNA as a template, and then purified using a PCR purification kit (Qiagen). The following primers were used: 5’-GAATAGTGTATGCGGCGACCGAGTTGC-3’5’-GCGAACGTGGCGAGAAAGGAAG-3’ 545-mer RNA construct preparation: The RNA 545-mer sequence (r545-mer) was produced by engineering the 545-mer DNA described above to contain the T7 promoter upstream of the sequence to be transcribed, as well as restriction enzyme sites for EcoRI and BamHI. Separate constructs were engineered to produce the sense (S-r545) and the anti sense (AS-r545) RNA strands. The T7 promoter and restriction enzyme sequences were introduced in the 5’- and 3’- extremes using PCR with the following primers: To produce AS-r545: Forward 5’–GGGGAATTCTAATACGACTCACTATAGGGGAATAGTGTATGCGGCGACCGAGTTGC-3’ Reverse 5’–GGGGGATCCGCGAACGTGGCGAGAAAGGAAG-3’ To produce S-r545: Forward 5’–GGGGAATTCGAATAGTGTATGCGGCGACCGAGTTGC–3’ Reverse 5’–GGGGGATCCTAATACGACTCACTATAGGGGCGAACGTGGCGAGAAAGGAAG–3’ Color code: T7 promoter, EcoRI, BamHI and primer sequences. After PCR, incorporation of the T7 promoter and restriction enzyme sequences was confirmed by agarose gel. The sequence of each construct was confirmed by sequencing analysis. Both constructs where then co-digested with EcoRI-HF and BamHI-HF (New England Biolabs) and then inserted / ligated independently in the plasmid pUK21-NotI (GeneBank code: AF324726) after pUK21-NotI’s initial T7 promoter sequence was scrambled. The S-r545 and AS-r545 plasmids were transformed in DH5α cells and then grown in the presence of kanamycin for plasmid production. The plasmid was extracted and purified using standard protocols for large-scale preparation of plasmid. The purified plasmid was then digested with proteinase K to remove RNase, followed by phenol/chloroform purification to eliminate RNase activity. Finally the plasmids were linearized with EcoRI or BamHI. Completeness of the linearization reaction was confirmed by agarose gel electrophoresis. 545-mer RNA duplex production: The production of the double stranded r545 was achieved by co-transcription of both the sense and anti-sense constructs in a one-pot reaction. 750 ng each of the sense and anti-sense plasmid constructs were transcribed using the HiScribe T7 High Yield RNA Synthesis Kit (New England Biolabs). DNA templates were then digested with Turbo DNase (ThermoFisher Scientific) using the protocol provided. Ethanol precipitation was performed on each co-transcription reaction, and RNA was then purified using a NAP-5 column (GE Healthcare). Then ds r545-mer was purified using long sequencing PAGE in native conditions (0.5x TBE, 7.5 % acrylamide). The r545-mer band was cut and extracted from the gel using standard protocols. The r545-mer was further purified with anion-exchange chromatography (DEAE Sephadex A-25, Sigma) to remove residual acrylamide from the sample and desalted on a NAP-5 column (GE Healthcare).




4

II. Supplementary Methods

Viscosity measurements: The viscosity of glycholine and its aqueous mixtures were measured on a Physica MCR 300 viscometer with a 50 mm, 1o cone plate geometry. For each sample, at least five values were measured in the Newtonian plateau (shear rate of 10-300 s-1) and then averaged to obtain the reported viscosities. Nucleic acid sample preparation: Oligonucleotides were purchased from Integrated DNA Technologies (for DNA) or GE Dharmacon (for RNA) and resuspended in 18.2 MΩ/cm water (Barnstead Nanopure™). For accurate sample manipulation and preparation we used mass rather than volume as a unit to express the amount of glycholine used, and hence we express our concentrations in molal (m). To prepare DNA/RNA samples in glycholine, aqueous DNA/RNA stocks were added to a weighed amount of glycholine and mixed until homogeneous. Water was then removed by vacuum centrifugation for at least 12 hr. All DNA samples in aqueous buffer were prepared in 0.1 M NaCl, 20 mM Tris, pH 8. Melting temperature (Tm) determination: UV absorbance was used to monitor the thermal denaturation of the hairpin, while circular dichroism (CD) spectroscopy was used for the 3 kb DNA. The thermal denaturation of the 32-mer duplex was reported in a previous study1. Because of the difference in size (and therefore extinction coefficient) between the 17 nt hairpin and 3 kb DNA, different concentrations were required to achieve comparable UV or CD signal from the two sequences. DNA samples were prepared with concentrations of 35.6 μm (hairpin in buffer); 0.188 μm (3 kb DNA in buffer); 29.7 μm (hairpin in glycholine). All UV and CD measurements were performed on 1 mm quartz cuvettes in a temperature-controlled UV-Vis spectrophotometer (Agilent 8453) or a temperature-controlled CD spectrophotometer (Jasco J-810) with nitrogen flow through the sample chamber at low temperatures. To determine Tm values, heating and cooling traces were generated for each sample by recording spectra (220-400 nm) from 0 to 100 oC at intervals of 1 oC. Melting curves were generated using the signal at a single wavelength (Supplementary Figure 1). For UV-Vis studies, the absorbance change at 260 nm was used; for CD studies, the ellipticity change at 248 nm was used. Tm values were determined as described by Mergny & Lacroix2. Reformation kinetics of hairpin and 32-mer duplex DNA: To determine duplex reformation kinetics, samples were first heated to completely denature all secondary structure. Glycholine samples were heated to 80 oC for 5 minutes; samples in aqueous buffer were heated to 100 oC for 5 minutes. After heating, sample temperature was quickly cooled to 20 oC by dipping cuvettes into an ice-water bath for 14 seconds—the measured time required to bring the sample temperature in a 1 mm quartz cuvette to 20 oC from 85 oC. The point at which the cuvette was dipped into the ice-water bath was taken as zero time. Cuvettes were then dried and placed in the spectrophotometer, a process that required about 9 seconds. Sample spectra were then collected every 1 second. Reformation kinetics of 3 kb DNA duplex: Duplex reformation for the 3 kb DNA in glycholine occurs over the course of days, but formation of intramolecular secondary structures along the denatured 2961 nt strands (which increases UV absorbance) is rapid. Thus, in this case, agarose gel electrophoresis was used to monitor duplex reformation and the disappearance of the single stranded state. Gel densitometry analysis was used to quantify annealing kinetics for the 3 kb DNA in glycholine.




5

Agarose gel electrophoresis: Gels were 2% agarose (UltraPure agarose, Life Technologies) and run in 1x TAE buffer. All samples in Figure 3 were prepared with a 3 kb DNA concentration of 2.73 nm. All samples run on agarose gels were heated to the necessary denaturation temperature (80 oC for glycholine and 95 oC for aqueous buffer) for 5 minutes before being cooled to 20 oC at the appropriate rate on a thermal cycler (BioRad) and placed in a desiccator at 20 oC for the desired amount of time. Sample heating was timed so that all samples could be loaded at the same time on a single gel. Prior gel loading, all samples were brought to 50% (wt/wt) glycholine (water was added to the glycholine samples and glycholine was added to the aqueous buffer samples) so that they would run comparably on the gel. The addition of water/glycholine prior to loading was performed solely to optimize the gel running conditions. Control experiments were performed with sequences F0 and L5 through R5 in a 20:1 molar ratio with the 3 kb template duplex (Supplementary Figure 7). In one experiment, the samples were heated and cooled in 100% glycholine before being diluted to 50% glycholine (wt/wt) immediately before loading in the gel; in the other experiment, the samples were thermally cycled and kept at 20 oC for 4 hours—after which F0 binding reaches a peak, as shown in Fig 3 in the main text—and then diluted to 50% glycholine. Comparison of the F0 binding kinetics for these two experiments show that the decay of the bound-F0 band intensity after dilution to 50% glycholine is on par with that in 100% glycholine up to 12 hours after cooling; after 12 hours, F0 does appear to be displaced more quickly in 50% glycholine than in 100% due to faster reformation of the 3 kb DNA duplex in a more water-like environment. This result indicates that dilution to 50% glycholine should not affect short oligonucleotide binding on the time scale of running an agarose gel, and that any binding seen on the gel occurs in 100% glycholine rather than in 50% glycholine after the dilution step. After sample loading, gels were run for 10 minutes in a 2.15 V/cm field, and then in a 4.3 V/cm field for 1-2 hours. After electrophoresis, gels were stained with ethidium bromide in 1x TAE buffer with gentle shaking for 15 minutes. Gels were imaged on a Typhoon Trio+ laser scanner (GE Healthcare) at a resolution of 50 μm and with a photomultiplier setting between 600-800. “EtBr filter” and “FAM filter” images were acquired using the excitation and emission wavelengths described in Methods of the main text. Densitometry analysis was performed using FIJI, an imaging-processing package based on ImageJ (NIH). Denaturing polyacrylamide gel electrophoresis: All polyacrylamide gels were 10% denaturing gels (8 M urea) run in 1x TBE buffer. For gel loading, 4 mg of each ligation sample was combined with 8 µL of loading dye (95% formamide, 0.025% bromophenol blue, 0.025% xylene cyanol, 5 mM EDTA pH 8.0). To denature duplex structures, DNA samples were heated to 95 oC for 2 minutes (90 oC for 2 minutes for RNA samples), placed on ice, and loaded onto the gel. Gels were pre-run for 30+ minutes prior to loading at 14 W and 300-400 V. Samples were run at these same conditions for 1 hour and stained with SYBR Gold dye. Gels were imaged as described above.

III. Supplementary Discussions Supplementary Disccusion 1: Selecting the sequence of F0. Though it is not possible to accurately predict the secondary structure of a 3 kb, kinetically trapped single strand in glycholine, we selected a target site located at position 2744 on the 2961 nt template strand, based on the lowest energy structure predicted by Mfold3. Supplementary Discussion 2: Rationale for utilizing 32 nt oligonucleotides. In choosing an oligonucleotide length, we balanced a number of factors. Shorter sequences are more prebiotically




6

feasible than longer sequences, but sequences must also be sufficiently long/stable to bind to their target sites at room temperature (since glycholine reduces the thermal stability of nucleic acid duplexes). Additionally, shorter sequences may have less of a toehold for opening up intramolecular hairpins on the kinetically trapped single strands, via strand displacement. We found that 32 nt oligonucleotides satisfy both of these requirements. Supplementary Discussion 3: Maximum theoretical binding of F0 oligonucleotide to its target template strand. In our experiments, we quantify the binding of the F0 oligonucleotide by comparing the fluorescent intensity of the template-bound F0 band to the free F0 band. Based on a 20:1 molar ratio of F0 to 3 kb template, the maximum percentage of total fluorescent intensity we should see in the F0 band is 5%. However, the measured percentage of total fluorescent intensity the F0 band approaches 7% at its peak (Fig. 3). We hypothesize that we are overestimating this percentage due to differences in quantum yield between F0 oligonucleotides when bound to the template vs. free in solution. The concentration of F0 within the free F0 band is much greater than that within the template-bound F0 band, which may result in quenching of fluorescein signal. Underestimation of the F0 concentration of in the excess F0 band would result in overestimating the percentage of F0 which is bound to the template. Supplementary Discussion 4: Simulation to determine the effects of un-ligatable ends on ligation yield. The products of our ligation experiments span the range from 32-mers to 352-mers. The yield of the longest possible product, the 352-mer produced by ligation of all eleven 32-mers together, may indicate the limit of using solvent mediated effects to overcome strand inhibition. However, another possibility is that this observed product distribution is the statistical result of un-ligatable ends—truncation products (31-mers, 30-mers, etc.) or oligomers missing the 5’ phosphate—binding to the 2961 nt template and limiting the extent of ligation. To gauge how a small percentage of un-ligatable ends can affect the product distribution, a MATLAB simulation was written where the percentage of un-ligatable ends in the starting oligomer population was varied. The simulation consisted of a population of 104 templates with each template represented as an 11 element vector. For each binding site, the value of a randomly generated number relative to the un-ligatable percentage determined whether the site is occupied by an un-ligatable oligomer or a 32 nt, phosphorylated oligomer. Two assumptions in the simulation are that all binding sites are occupied (reflected experimentally in the appearance of a 2961 nt band, to which F0 is bound, which migrates with an observably different mobility than its complementary strand; see Figure 3) and that the ligation reaction is much faster than unbinding of a fully complementary 32-mer oligonucleotide at room temperature4.

We estimate that, even after purification by polyacrylamide gel electrophoresis, up to 11% of the oligonucleotide substrates have un-ligatable ends. The simulation correlates well with observed ligation product distributions for two experiments with different substrate purities, using gel-purified and unpurified oligonucleotides (Supplementary Figure 4). These results indicate that the yield of full-length product observed in our experiments is near its theoretical limit based on oligonucleotide purity.




7

Supplementary Figure 1: Thermal denaturation studies of the hairpin and 3 kb DNA sequence. a, Thermal denaturation of the DNA hairpin in aqueous buffer and in glycholine. b, Thermal denaturation of the 3 kb DNA in aqueous buffer. c, CD spectra of the 3 kb DNA in glycholine at 20 oC (blue), at selected temperatures over the course of heating from 20 to 80 oC (grey), at 80 oC (red), and after cooling back to 20 oC (cyan). d, CD signal monitored at 248 nm for the 3 kb DNA in glycholine during heating (red) and cooling (blue).




8

Supplementary Figure 2: Return of 3 kb DNA to a duplex state after heat cycling. a, Agarose gel showing the separation of 3 kb DNA into duplex (less mobile band) and single stranded (more mobile band) states after heating to denaturing temperatures (80 oC for glycholine samples and 95 oC for aqueous buffer samples) and cooling to 20 oC at the indicated rates. Samples heat-cycled in aqueous buffer and in glycholine exhibit comparable electrophoretic mobility of the 2961 nt ssDNA, suggesting the generation of ssDNA with a similar degree of intramolecular structure in these two solvents. b, The recovery of the duplex state for the 3 kb DNA as a function of different cooling rates, as measured by densitometry analysis of the duplex band intensity on the agarose gel.

Supplementary Figure 3: Binding of a single oligonucleotide (32 nt) to the 3 kb duplex template. a, Schematic illustrating the binding of F0 (fluorescently tagged with FAM, i.e. carboxyfluorescein) to a




9

specific site on one of the 2961 nt template strands. b, Agarose gel showing the result of heat cycling F0 and the 3 kb duplex in glycholine, after staining with ethidium bromide (EtBr) and imaging through both an EtBr-specific channel (red) and a FAM-specific channel (green).

Supplementary Figure 4. Quantification of truncation products in oligonucleotides purchased from IDT with standard desalting or PAGE purification. Radiolabeled 32-mer oligonucleotides run on a 10% polyacrylamide gel. Densitometry analysis of these gels revealed that the average fraction of truncation products in the set of desalted oligonucleotides is 37%, whereas the average fraction of truncation products in the PAGE purified set is 11%.

Supplementary Figure 5. Ligation efficiencies obtained from heating and cooling the 3kb DNA in glycholine with a 1 to 3-fold molar excess of 32-mer oligonucleotides L1-L5 and R1-R5. F0 was not added to reaction mixtures, resulting in 5 product bands. The ligation procedure described in the Methods was applied with different molar excesses relative to the 3 kb DNA template. The gel shown confirms that the viscosity-enabled polymerization method described in the text successfully produces full-length product down to a 1:1 molar ratio of 32-mer oligonucleotides to template strand.




10

Supplementary Figure 6. a, Viscosity measurements of aqueous mixtures of glycholine. Data points shown are averaged from three repeated measurements. b, Denaturing polyacrylamide gel showing the results of heat cycling and ligating a mixture of the 3 kb DNA template and eleven 32-mer DNA oligonucleotides (L5-R5) in different aqueous mixtures of glycholine.




11

Supplementary Figure 7. Samples containing L5-R5 and 3 kb template duplex were heated and cooling in 100% glycholine for 4 hours, then diluted to 50% (wt/wt) glycholine. a, Agarose gel imaged with the EtBr filter showing the time course of duplex reformation by 3 kb DNA after dilution. Asterisk indicates the position of the bound F0 band. b, The same gel imaged with the FAM filter showing the binding kinetics of F0 after dilution. c, Densitometry analysis comparing the kinetics of F0 binding after dilution to 50% glycholine versus in 100% glycholine.




12

Supplementary Figure 8. a, The efficiency of T4 DNA ligase was tested in a range of glycholine concentrations. For these tests, a system consisting of two tiling half-complement 12-mers was used:

The 6 nt overhangs on each 12-mer are complementary to each other, so 12-mers can tile without limit. One of the two 12-mers has a 5’-phosphate group, so that T4 ligase can polymerize these 12-mers. These 12-mers and T4 DNA ligase were mixed in varying concentrations of glycholine. After allowing the reaction to proceed for 20 minutes, the T4 DNA ligase was heat denatured and samples were run on a denaturing polyacrylamide gel; the gel was then stained with SYBR Gold. As shown in gel image, in 15% (wt/wt) glycholine T4 DNA ligase is essentially as efficient as in 5% (wt/wt) glycholine. Therefore, in subsequent ligation experiments, water was added to dilute samples to 15% glycholine prior to adding T4 DNA ligase. b, The efficiency of T4 RNA ligase 2 was tested in a range of glycholine concentrations. For these tests, a system of two tiling half-complement 32-mers was used:

After allowing the ligation reaction to proceed for 20 minutes, samples were run on a denaturing polyacrylamide gel and imaged after SYBR Gold staining.

3’-GUAAAAGAUGCUGAAG AGAAACGCUGGUGAAA-5’

5’-PO4-UCUUUGCGACCACUUU CAUUUUCUACGACUUC-3’




13

Supplementary Table 1: Melting temperatures of DNA duplex species Species Base Pairs %GC Tm in Aqueous Buffer (oC) Tm in Glycholine (oC) Hairpin 7 bp stem, 3 nt loop 41 79.5 44.7

32-mer duplex 32 47 72.2 49.0 3 kb duplex 2957 50 88.0 50.7

Supplementary Table 1. Melting temperatures of 17 nt hairpin, 32-mer duplex, and 3 kb duplex in both aqueous buffer and glycholine.

Supplementary Table 2: Binding kinetics of F0 to 3 kb template duplex in glycholine

Time after cooling (hr)

Percentage of total F0 intensity in the bound F0 band (%)

F0 only All 11 oligo. 1 1.5 ± 0.08 5.3 ± 0.28 2 2.1 ± 0.11 6.0 ± 0.32 4 2.2 ± 0.12 6.6 ± 0.35 6 1.7 ± 0.09 5.9 ± 0.31 8 2.9 ± 0.15 6.2 ± 0.33

16 2.9 ± 0.15 4.7 ± 0.25 24 2.8 ± 0.15 3.8 ± 0.20 48 2.7 ± 0.14 3.4 ± 0.18

144 1.4 ± 0.08 1.7 ± 0.09 192 No data 0.9 ± 0.04 240 1.3 ± 0.07 1.0 ± 0.05

Supplementary Table 2. Densitometry analysis showing the binding of F0 over time after thermal cycling in glycholine (corresponding to Fig 3c).

Supplementary Table 3: Binding kinetics of F0 to 3 kb template

duplex in aqueous buffer

Time after cooling

Percentage of total F0 intensity in the bound F0 band (%)

F0 only All 11 oligo. <1 min 0.32 0.18

Supplementary Table 3. Densitometry analysis showing the binding of F0 over time after thermal cycling in aqueous buffer (corresponding to Fig 3c).




14

Supplementary Table 4: Yield of 64 nt to 352 nt product from 3 kb DNA template

Ligation product length (nt)

Percentage of total weight in the band (%) All 11 oligo. No F0 No L2 No R2

320 1.7 ± 0.96 288 1.8 ± 0.66 256 2.1 ± 0.49 224 2.7 ± 0.71 7.4 ± 2.7 6.7 ± 1.9 192 2.1 ± 0.75 2.2 ± 0.96 2.2 ± 0.41 160 2.5 ± 0.59 11.2 ± 5.6 2.9 ± 0.55 2.1 ± 0.30 128 3.2 ± 0.84 6.7 ± 0.16 3.1 ± 0.31 2.5 ± 0.52 96 2.7 ± 0.75 5.2 ± 0.20 9.6 ± 2.4 9.3 ± 1.8 64 2.1 ± 0.12 5.3 ± 2.0 4.2 ± 1.3 3.7 ± 0.73

Supplementary Table 4. Densitometry analysis showing the relative amounts of each ligation product generated from thermal cycling with a 3 kb DNA template duplex.

Supplementary Table 5: Comparison of yields between experiment and simulation

Ligation product

length (nt)

Percentage of total moles of 96 nt to 352 nt products (%) Experiment:

standard desalting (63% pure)

Experiment: PAGE purification (89%

pure)

Simulation: 70% pure



96 35.5 22.4 36.2 22.3 8.4 128 26.0 19.1 23.3 17.1 7.7 160 16.2 12.0 15.4 13.8 7.3 192 9.6 9.2 9.5 10.6 6.9 224 5.2 9.4 5.9 8.8 6.0 256 2.5 6.3 4.3 6.3 6.2 288 1.9 5.3 2.5 5.2 5.4 320 1.2 5.4 1.1 4.6 4.8 352 1.9 11.0 1.9 11.3 47.2

Supplementary Table 5. Relative yields (molar basis) of ligation products, obtained from experiments and simulation.




15

Supplementary Table 6: Yield From Aqueous Mixtures of Glycholine Percentage glycholine

(wt/wt) Viscosity

(cP) Yield (Mol of 352 nt Product per Mol of

Template) 80 34.5 ± 1.7 0.036 ± 0.021 90 101 ± 2.2 0.043 ± 0.009

95.8 217 ± 4.3 0.092 ± 0.063 98.1 311 ± 13 0.13 ± 0.044 100 437 ± 26 0.14 ± 0.055

Supplementary Table 6. Yields obtained from thermal cycling of 3 kb DNA template duplex in different aqueous mixtures of glycholine.




16

Supplementary References 1 Gállego, I., Grover, M. A. & Hud, N. V. Folding and imaging of DNA nanostructures in anhydrous

and hydrated deep-eutectic solvents. Angewandte Chemie (International ed. in English) 54, 6765-6769, doi:10.1002/anie.201412354 (2015).

2 Mergny, J. L. & Lacroix, L. Analysis of thermal melting curves. Oligonucleotides 13, 515-537, doi:10.1089/154545703322860825 (2003).

3 Zuker, M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 31, 3406-3415, doi:10.1093/nar/gkg595 (2003).

4 Tawa, K. & Knoll, W. Mismatching base-pair dependence of the kinetics of DNA–DNA hybridization studied by surface plasmon fluorescence spectroscopy. Nucleic Acids Research 32, 2372-2377, doi:10.1093/nar/gkh572 (2004).


suementary inrmatin.10/.2628 suementary inrmatin 4 ii. supplementary methods viscosity measurements:...

Documents