Ionic liquid-caged nucleic acids enable active folding-based molecular recognition with hydrolysis resistance

Abstract Beyond storage and transmission of genetic information in cellular life, nucleic acids can perform diverse interesting functions, including specific target recognition and biochemical reaction acceleration; the versatile biopolymers, however, are acutely vulnerable to hydrolysis-driven degradation. Here, we demonstrate that the cage effect of choline dihydrogen phosphate permits active folding of nucleic acids like water, but prevents their phosphodiester hydrolysis unlike water. The choline-based ionic liquid not only serves as a universal inhibitor of nucleases, exceptionally extending half-lives of nucleic acids up to 6 500 000 times, but highly useful tasks of nucleic acids (e.g. mRNA detection of molecular beacons, ligand recognition of aptamers, and transesterification reaction of ribozymes) can be also conducted with well-conserved affinities and specificities. As liberated from the function loss and degradation risk, the presence of undesired and unknown nucleases does not undermine desired molecular functions of nucleic acids without hydrolysis artifacts even in nuclease cocktails and human saliva.


Target recognition of St-21-1 in CDHP
The target affinity of St-2-1 was measured by employing streptavidin-coated magnetic beads (Sera-Mag TM Magnetic Streptavidin Microparticles).4 μl beads were washed twice using 1X PBS buffer, and then washed beads were incubated with 1.14 ml appropriate buffer (1X PBS, 2 M NaCl with 1X PBS, and 2 M CDHP with 1X PBS) for 30 min at room temperature.Thereafter, to the 190 μl bead solution, 10 μl 5'-FAM-labeled St-21-1 aptamer at varying concentrations in appropriate buffer was added.After 90 min incubation at room temperature, the bead-bound aptamer strands were selectively collected using magnet, and they were immediately dissolved in an aqueous buffer (250 mM NaCl and 500 mM Tris-HCl, pH 8.5).The FAM fluoresence was recored at 535 nm (excitation: 485 nm).
Fluorescence signals were recorded with following setting: excitation: 485 nm and emission: 535 nm.
In the ATP specificity assay, the concentration of all nucleotide monomers was set to 1 mM.
To test aptameric recognition of salivary cortisol either in aqueous or CDHP condition, we mixed human saliva (100%), 25 μM cortisol aptamer (0.4 μl), 25 μM 3'-FAM-labeled signaling strand (0.4 μl), and 4 M CDHP (25 μl).To the reaction mixture (50.8 μl), a cortisol solution (0.5 μl) with varying concentrations was added, and subsequently, the volume was decreased down to ~50.5 μl.The prepared sample were incubated at 37 °C for 3,000 min (the control sample with no CDHP was incubated for 30 min.After incubation, the heat treatment and FAM signal measurement were performed as mentioned above. [a] Human serum albumin. [b] The relative increase in half-life was calculated by dividing the half-life of the protected oligonucleotide by that of unprotected one.In this experiment, the fluorescence emission profile was recorded in a 500-750 nm wavelength range with emission bandwidth: 1 nm (excitation setting: 450 nm).Scrambled RNA sequence was designed by using the bioinformatics tools of GenScript (https://www.genscript.com/tools/create-scrambledsequence).S1).(B)

Supplementary tables and figures
The normalized FRET efficiency of human telomeric G-quadruplex at varied CDHP concentrations.fret-G-quadruplex was used.As negligible differences in quantum yields of Cy3-and Cy5-labed oligonucleotides at varying CDHP concentrations were observed (data was not shown), we did not consider the quantum yield of fluorophore dyes when calculating the FRET efficiency.

Figure S1 .
Figure S1.LOQ values of the MnSOD MB in five different conditions: 1X DPBS, 1 and 2 M CDHP (with 1X DPBS), and 1 and 2 M NaCl (with 1X DPBS).When we calculated LOQ value of MnSOD MB in five different conditions based on Cy3 signal (575 nm), we found that the target sensitivity of MnSOD MB is almost unchanged irrespective of cation and anion types and their concentrations.In the range of MnSOD mRNA-22 concentrations from 0.1 to 10 nM, the fluorescence intensities are linear (inset), yielding the LOQ value based on 10σ/slope method (σ is the standard deviation of the blank signal).

Figure S2 .
Figure S2.The aptamer-specific increase in fluorescent signal of MG in both water and 2 M CDHP.The scrambled MG aptamer (100 nM) fails to recognize the MG, leading no increase in the MG fluorescence.

Figure S3 .
Figure S3.Successful target recognition of DNA aptamers in CDHP and in optimal buffer.(A) The small molecule recognition of the PS2.M aptamer.The PS2.M in 2 M CDHP successfully bind to its two cognate targets (NMM and thioflavin T) like that in the aqueous condition (100 mM KCl, 10 mM Tris-HCl, pH 7.6).When the target affinity (Kd) of PS2.M is investigated, the CDHP solvation (2 M CDHP, 100 mM KCl, 10 mM Tric-HCl, pH 7.6) slightly increases the Kd value of PS2.M to NMM (left), but decreases that of PS2.M to thioflavin T (right).(B) The large protein recognition of the St-21-1 aptamer.When the St-21-1 recognizes the streptavidin protein in 1X PBS, it shows Kd value of 94.5 nM.However, high ionic strengths of both 2 M NaCl and 2 M CDHP lower Kd values down to 75.4 and 70.0 nM, respectively.

Figure S4 .
Figure S4.Engineering of hammerhead ribozyme sequence.(A) Three stem formations (blue dotted box) when the hammerhead ribozyme hybridizes with its substrate.(B) The structure of engineered hammerhead ribozyme.To increase the thermal stabilities of stem regions in the CDHP, we intentionally inserts a few A:U pairs in all three stems.

Figure S5 .
Figure S5.Thermal stabilities of RNA duplexes in CDHP.In 2 M CDHP solution, the RNA duplex with higher A-U content is more thermodynamically stable than that with lower A-U content.For instance, the melting temperature (Tm) of RNA duplex with 80% A-U content (80AU-RNA) is 7.4 °C higher than that with 20% A-U content (20AU-RNA).In this assay, EvaGreen® dye that can intercalate to an RNA duplex was exploited to measure Tm(18).The concentration of all RNA oligonucleotides was set to 2 μM.To obtain fluorescence melting curves, the fluorescent signals of 18 μl solution was measured by LightCycler ® 96 (Roche).Fluorescent changes of EvaGreen® dye were recorded within 37-98 °C temperature range with 0.1 °C increments.

Figure S6 .
Figure S6.Normalized cleavage yield of hammerhead ribozyme relying on CDHP concentrations.In CDHP, the engineered hammerhead ribozyme shows a diminished catalytic yield compared to that in aqueous solution.Based on Figure 2F, we obtained the band intensities of substrates and cleaved products by using GelQuant.Net software provided by biochemlabsolutions.com.For the normalization, we calculated the catalytic yield of hammerhead ribozyme without CDHP as one.

Figure S7 .
Figure S7.Requirement of both ribozyme and Mg 2+ for the successful ribozymatic cleavage in 2 M CDHP.(A) The cleaved product is only detectable in the presence of the ribozyme.(B) Even with the ribozyme, to catalyze the RNA-mediated substrate cleavage, Mg 2+ is also required.

Figure S8 .
Figure S8.Partly modified oligonucleotides to easily visualize nuclease-mediated degradations of FNAs in gel electrophoresis.(A) Cy5 incorporation at the 5' end of MnSOD mRNA-22.(B) 5'-FAM labeling of MG aptamer.(C) Removal of 5'-labeled-BHQ2 of MnSOD MB. (D) The use of G-substrate instead of Sub to exclude a ribozymatic cleavage when investigating nuclease activities in water and CDHP.

Figure S9 .
Figure S9.Activity recovery of various nucleases that were incubated in 2 M CDHP.We mixed a nuclease either with aqueous buffer or with 2 M CDHP at 37 °C (for 30 min), and then diluted nucleasecontaining solutions with excess water (20x dilution).When these diluted nucleases were incubated with target nucleic acids, all the analyzed nucleases successfully exhibit nucleolytic activities, proving that the CDHP environment does not denature protein scaffolds, but inhibits the hydrolytic reaction.

Figure S10 .
Figure S10.CDHP concentration-dependent hydrolysis activities of seven different nucleases.All the analyzed nucleases are effectively inhibited in a CDHP concentration-dependent manner (incubation for 30 min at 37 °C).Exonuclease T5 and S1 nuclease are readily inhibited by the presence of CDHP (~200-400 mM), whereas RNase H is most resistant to the CDHP environment among seven nucleases, so >1.5 M CDHP is required to completely block its hydrolytic action.In this experiments, to investigate the nuclease activities of RNase A, exonuclease T, and S1 nuclease, MG aptamer-FAM was employed, and MnSOD MB-Cy3 was used as a target nucleic acid for DNase I, exonuclease T5, and exonuclease V.In the case of RNase H assay, we analyzed fluorescent signals of MnSOD mRNA-22-Cy5, hybridized with MnSOD MB-Cy3.

Figure S11 .
Figure S11.Fraction of intact oligonucleotides at different CDHP concentrations (0-2000 mM) in the presence of each nuclease.Based on Figure S9, the fraction of intact molecule was calculated by dividing the fluorescent signal of an intact oligonucleotide by that of the oligonucleotide with no nuclease (Figure S10, negative control).To investigate gel band intensities, we used GelQuant.Net software provided by biochemlabsolutions.com.

Figure S12 .
Figure S12.Monitoring time-dependent oligonucleotide degradations in the presence of nucleases in the aqueous condition.The MG aptamer-FAM was employed for RNase A, exonuclease T, and S1 nuclease experiments, and the MnSOD MB-Cy3 was used for DNase I, T5 exonuclease, and exonuclease V assays.Moreover, the MnSOD mRNA-22-Cy5, hybridized with MnSOD MB-Cy3, was analyzed for the RNase H experiment.

Figure S13 .
Figure S13.Monitoring time-dependent oligonucleotide degradations in the presence of nucleases in 2 M CDHP.For 1-month incubation at 37 °C, only negligible nuclease-driven degradations are observed.

Figure S15 .
Figure S15.Aptameric detection of salivary cortisol in CDHP.(A) Rational design of cortisol-sensing aptasensor consisting of a 5'-BHQ1-labeled cortisol aptamer and a 3'-FAM-labeled signaling strand.When the aptamer recognizes its cognate target, the signaling strand can be released, thereby producing the FAM fluorescence signals in a cortisol concentration-dependent manner.(B) The aptamer-mediated cortisol detection in 2 M CDHP (with 1X CBB) and the aqueous condition (1X CBB only).In both conditions, the cortisol aptamer successfully recognize its cognate target.Notably, in the range of cortisol concentrations up to 10 ng/ml, fluorescent signals are measured to be linear (inset).The LOD values are calculated by 3σ/slope method (σ is the standard deviation of blank signal).(C) Salivary cortisol detection of aptamers with 2 M CDHP.Without CDHP, cortisol concentration-irrelevant fluorescent signals of aptasensor were observed in saliva (30 min incubation).However, with 2 M CDHP, the cortisol aptamers successfully bind to its target with no hydrolysis even after 3,000 min incubation at 37 °C, evidenced by desired FAM signals in a cortisol concentration-dependent manner like the fully intact aptamers without saliva.

Figure S16 .
Figure S16.Successful structure-switching of ATP aptamer either in CDHP (1 and 2 M) or in 1X optimal buffer.(A) ATP-induced structure-switching of ATP aptamer.The ATP aptamer in which central domain is hybridized with antisense DNA (AD-13) can undergo ATP-specific conformational change, releasing AD-13 and subsequently inducing close proximity between 5' FAM and 3' BHQ1 for decreased fluorescent signals(19,20).(B) Fluorescent signal changes of ATP aptamer in an ATP concentrationdependent manner.In both 1 and 2 M CDHP, the ATP aptamer successfully performs its pre-defined structure-switching, decreasing FAM fluorescence (inset).Relying on ATP concentration-dependent fluorescent signals, a half-maximal effective concentration (EC50) of ATP aptamer is obtained in three different conditions: 1X optimal buffer, 1 and 2 M CDHP (with 1X optimal buffer).1X optimal buffer consists of 300 mM NaCl, 5 mM MgCl2, 20 mM Tris-HCl (pH 7.4).(C) ATP-specific and selective structure-switching of ATP aptamer in 2 M CDHP.Even in 2 M CDHP, the ATP aptamer retains its specificity and selectivity to the ATP molecule, with no response to other nucleotides (GTP, CTP and UTP).In this assay, the concentration of all nucleotide monomers was set to 1 mM.

Figure S17 .
Figure S17.Cage effect of CDHP on nucleic acids.(A) Charge neutralization of oligonucleotides relying on CDHP concentrations.At the high CDHP concentration (2 M), the migration of both FAM-labeled DNAs and RNAs was fully retarded in agarose gel electrophoresis as their negatively-charged phosphate backbones were neutralized by choline cations.Experimental condition: 1X TBE-1.2%agarose gel (100V and 12 min) and 1 μM oligonucleotides.(B) The binding affinity of minor groovebinding fluorescent dyes to nucleic acid duplexes with or without CDHP (2M).When we investigated EC50 for SYBR Green I by titrating the DNA duplex, its binding affinity is decreased in the presence of CDHP that can be complexed with the minor grooves of nucleic acids.Likewise, when titrating RNA duplex, we found that 2 M CDHP condition lowered the binding affinity of methylene blue to RNA duplex.

Figure S18 .
Figure S18.Cage effect of CDHP on structural compaction of functional nucleic acids.To calculate the FRET efficiencies of functional nucleic acids, which are internally labeled with Cy3 and Cy5 fluorophores, at varied CDHP concentrations (up to 2 M), the Cy3 signal (excitation: 530 nm, emission: 575 nm) and Cy5 signal (excitation: 530 nm, emission: 675 nm) were measured.To normalize the FRET efficiency, the FRET efficiency of a sample was divided by that of a sample without CDHP (0.0 M).All samples included 30 mM Tris-HCl (pH 7.5).(A) The normalized FRET efficiency of the FRET pair-labeled MG aptamer at varied CDHP concentrations.In this experiment, fret-MG aptamer was used (TableS1).(B)

Figure S19 .
Figure S19.Catalytically active RNase A even in the presence of excessive RNase inhibitors.When in vitro transcribed RNAs from 165-nt long dsDNA with T7 promoter (lane 1) was incubated with RNase A (lane 2), RNase A fully digested RNA molecules as expected.To preserve transcribed RNAs against RNase A, the excess amounts of commercially available RNase inhibitor (RNasin® Ribonuclease Inhibitor from Promega) were employed (lane 3-7), but nucleases still effectively degraded RNAs, and only dsDNA molecules remained.Incubation for 30 min at 37 °C.Nucleic acids were stained by SYBR Gold I.

Table S1 .
Oligonucleotides used in this work.

Table S2 .
Distinct nucleic acid specificities of seven different nucleases (1-4), and their potential nucleic acid targets used in this work.

Table S3 .
Reported half-lives of nucleic acids relying on various protection methods.