Non‐Enzymatic RNA Backbone Proofreading through Energy‐Dissipative Recycling

Abstract Non‐enzymatic oligomerization of activated ribonucleotides leads to ribonucleic acids that contain a mixture of 2′,5′‐ and 3′,5′‐linkages, and overcoming this backbone heterogeneity has long been considered a major limitation to the prebiotic emergence of RNA. Herein, we demonstrate non‐enzymatic chemistry that progressively converts 2′,5′‐linkages into 3′,5′‐linkages through iterative degradation and repair. The energetic costs of this proofreading are met by the hydrolytic turnover of a phosphate activating agent and an acylating agent. With multiple rounds of this energy‐dissipative recycling, we show that all‐3′,5′‐linked duplex RNA can emerge from a backbone heterogeneous mixture, thereby delineating a route that could have driven RNA evolution on the early earth.


Experimental Procedures
General procedure for desalting by ethanol precipitation. Oligonucleotides were desalted by addition of a 2 M imidazole nitrate buffer solution (pH 6.2, 1/10 volume), followed by a 3 M sodium acetate solution (pH 5.2, 1/10 volume) and absolute ethanol (to a final concentration of 75 % (v/v)). The resulting mixtures were typically kept at -20 °C for 2-4 h and then centrifuged for 30 min at 16000 × g. The supernatant was removed and the pellets were washed with 75 % (v/v) aqueous ethanol, before additional centrifugation (16000 × g, 10-30 min). The residual pellets were air dried before being redissolved in water.
General procedure for the hydrolysis reactions. Oligonucleotide 1 (or 7) (80 µM) and its complementary template 2 (80 µM) were mixed in 200 mM NaCl (10.0 µL) and incubated at 95 °C for 4 min. The mixture was allowed to slowly cool to 21 °C before addition of 5.0 µL of a 1 M sodium carbonate buffer solution (pH 9.25) and either 5.0 µL of water or 5.0 µL of a 80 or 160 mM MgCl2 solution (final concentrations: 40 µM each strand, 100 mM NaCl, 250 mM sodium carbonate, 0, 20 or 40 mM MgCl2). The reaction was kept at 21 °C for 24 days and its progress was monitored after 3, 6, 12 (or 13) and 24 days by analytical SAX-HPLC (gradient: 30-80 % B for 35 min and then 100 % B for 5 min; the aliquots were desalted by ethanol precipitation before HPLC injection).
Procedure to assess the selectivity of the acetylation-ligation reaction of 3′p vs 2′p termini in a helical environment. The reaction was performed using a modified version of the procedure previously reported by us. [1] A mixture (6.0 µL) containing 33.3 µM of template (2), 33.3 µM of FAM-labeled ligator (8) and 33.3 µM of primer (9-12, or 16.7 µM each of 2′p and 3′p primer for competition experiments) was incubated at 70 °C for 4 min and then briefly cooled on ice. Acetylation was then conducted by addition of 1.0 µL of a 350 mM N-acetylimidazole solution (final concentration 50 mM) and incubation at 30 °C for 1 h. 0.5 µL of the acetylated mixture was kept for MALDI-TOF analysis and the remaining 6.5 µL were subjected to ligation in 200 mM imidazole nitrate buffer (pH 6.2), 10 mM MnCl2 and 100 mM N-cyanoimidazole (final volume 9.3 µL). The reaction mixture was incubated at 21 °C for 19 h, then diluted with an equal volume of loading buffer (95 % (v/v) formamide, 5 % (v/v) glycerol, 50 mM EDTA, 0.5 mg/ml Orange G) and heated at 95 °C for 4 min before cooling briefly on ice. The reaction products were analyzed by gel electrophoresis, as described above.
Procedure to assess deacetylation under hydrolytic conditions. A mixture (10.0 µL) of 6 nt primer 3′p (5, 80 µM) and 50 mM of N-acetylimidazole was incubated at 21 °C for 5 h. A template ligation was then conducted using 20 µM of the acetylated primer, 20 µM of ligator (4) and 20 µM of template (2) in 200 mM imidazole nitrate buffer (pH 6.2), 10 mM MnCl2 and 100 mM N-cyanoimidazole (final volume 30.0 µL). The reaction mixture was incubated at 21 °C for 19 h and then desalted by ethanol precipitation. The pellet was redissolved in 22.0 µL of water and divided into three aliquots. One fraction (2.5 µL) was directly analyzed by analytical SAX-HPLC (gradient: 30-80 % B for 35 min and then 100 % B for 5 min) and MALDI-TOF mass spectrometry. A second aliquot (9.5 µL) was mixed with 0.5 µL of a 4 M NaCl solution and incubated at 95 °C for 4 min. This mixture was allowed to slowly cool to 21 °C before addition of 5.0 µL of a 1 M sodium carbonate buffer solution (pH 9.25) and 5.0 µL of a 160 mM MgCl2 solution (final concentrations: 100 mM NaCl, 250 mM sodium carbonate, 40 mM MgCl2). The reaction was kept at 21 °C for 24 days and its progress was monitored after 6 and 24 days by HPLC and MALDI-TOF analysis (the aliquots were desalted by ethanol precipitation before use). The third fraction was used to perform a control reaction, employing the deacetylation procedure previously reported by us. [1] Thus the solution (10.0 µL) was lyophilized, redissolved in 20.0 µL of a 4.5 M NH3 solution and heated at 40 °C for 1h before ethanol precipitation and MALDI-TOF/HPLC analysis.
Procedure for 2′,5′-linkages recycling. Oligonucleotide 1 (80 µM) and its complementary template 2 (80 µM) were mixed in 200 mM NaCl (75.0 µL) and incubated at 95 °C for 4 min. The mixture was allowed to slowly cool to 21 °C before addition of 37.5 µL of a 1 M sodium carbonate buffer solution (pH 9.25) and 37.5 µL of a 160 mM MgCl2 solution (final concentrations: 40 µM each strand, 100 mM NaCl, 250 mM sodium carbonate, 40 mM MgCl2). The reaction was kept at 21 °C and its progress was monitored by analytical SAX-HPLC (gradient: 30-80 % B for 35 min and then 100 % B for 5 min). Aliquots of 45.0 µL were taken after 6, 12 and 24 days of incubation, and desalted by ethanol precipitation. For each aliquot, the residual pellet was redissolved in 18.0 µL of water and the resultant solution was incubated at 70 °C for 4 min and then briefly cooled on ice. Acetylation was then conducted by addition of 3.0 µL of a 350 mM N-acetylimidazole solution (final concentration 50 mM) and incubation at 30 °C for 1 h. The mixture was then subjected to ligation in 200 mM imidazole nitrate buffer (pH 6.2), 10 mM MnCl2 and 100 mM N-cyanoimidazole (final volume 29.3 µL), reacted at 21 °C for 19 h and then desalted by ethanol precipitation. The residual pellet was then redissolved in 17.25 µL of water and the resultant solution was divided into two aliquots. One fraction (14.25 µL) was mixed with 0.75 µL of a 4 M NaCl solution and incubated at 95 °C for 4 min. The mixture was allowed to slowly cool to 21 °C before addition of 7.5 µL of a 1 M sodium carbonate buffer solution (pH 9.25) and 7.5 µL of a 160 mM MgCl2 solution (final concentrations: 100 mM NaCl, 250 mM sodium carbonate, 40 mM MgCl2). The reaction was kept at 21 °C for the indicated days before HPLC analysis and a new round of reactions. The cycle was repeated for the indicated number of times, adjusting the volume of water to be added to the pellets, considering the reduction of available material after each cycle.
Procedure to estimate the yield of the templated acetylation-ligation, following pre-treatment with the hydrolysis buffer. The template (2, 20 µM), the ligator (4, 20 µM) and the FAM-labeled primer (13, 10 µM) were dissolved in the hydrolysis buffer (100 mM NaCl, 250 mM sodium carbonate buffer pH 9.25, 40 mM MgCl2) and desalted by ethanol precipitation. The residual pellet was redissolved in 6.0 µL of water and the resultant solution was incubated at 70 °C for 4 min and then briefly cooled on ice. Acetylation was then conducted by addition of 1.0 µL of a 350 mM N-acetylimidazole solution (final concentration 50 mM) and incubation at 30 °C for 1 h. The mixture was then subjected to ligation in 200 mM imidazole nitrate buffer (pH 6.2), 10 mM MnCl2 and 100 mM Ncyanoimidazole (final volume 9.3 µL), reacted at 21 °C for 19 h and then diluted with an equal volume of loading buffer (95 % (v/v) formamide, 5 % (v/v) glycerol, 50 mM EDTA, 0.5 mg/ml Orange G). The resulting mixture was heated at 95 °C for 4 min, briefly cooled on ice and analyzed by gel electrophoresis, as described above. The reaction was performed in triplicate. Yield: 39 ± 2%. ; lanes e-f: assessment of the ligation selectivity in competition experiments. The products were quantified by fluorescence scanning (left, n.b. unligated primers and template not detectable) before SYBR Gold staining (right, n.b. primers could not be detected/distinguished from FAM-labeled ligator). Yields are referred to the total amount of labeled ligator. c-h) MALDI-TOF mass spectra of the mixtures after acetylation: the primer(s) present are as labeled in each spectrum. Peaks attributed to cyclic phosphate (numerical mass in red), unreacted primer (black), mono-(green) and bis-acetylated products (blue) and their adducts (see Table S2 for calculated masses). Unlabeled peaks in spectra g) and h) belong to the FAM-labeled ligator and its acetylated derivatives. *In competition experiments the concentration of each primer is halved compared to the independent reactions; as a consequence yields for lanes e-f are halved compared to lanes a-d, but the ratio 3′,5′:2′,5′ is maintained, thus indicating that selectivity is not affected by the presence of 2′p primers. The ligation yield of the 10 nt primer 3′p 9 (lane a) with the labeled ligator 8 was found to be lower than when performing the same reaction between the labeled 10 nt primer 13 and ligator 4 (yield: 45%). These differences maybe due to the effect of the FAM tag, which in 13 is attached to a 4 nt overhang, while in 8 is attached to the terminal nucleotide involved in the duplex and might therefore interfere with its correct annealing to the complementary template.  Intensity / % Intensity / % Figure S4. Deacetylation reaction. a) Schematic representation of the reaction. b) MALDI-TOF mass spectra of the mixture after ligation, c) after 1h of reaction with NH3 (aq.), d) after 6 days of hydrolysis with carbonate/Mg 2+ buffer, e) after 24 days of hydrolysis with carbonate/Mg 2+ buffer. Peaks attributed to the 13 nt acetylated product (numerical mass in red), 13 nt deacetylated product (green), template (black) and their Na + adducts (see Table S2 for calculated masses). f) HPLC chromatograms of the mixtures showing the progress of deacetylation in different conditions (red arrow: acetylated product; green arrow: deacetylated product). The basic solvents (pH 11.5) used in SAX-HPLC caused partial deacetylation of the product during the running.  Intensity / % Intensity / % Figure S5. Recycling studies. a) HPLC chromatograms showing the recycling of 2′,5′-linkages: the hydrolysis reaction was stopped after 24 days of incubation and the mixture was subjected to two rounds of recycling, alternated by 6 days of hydrolysis. b) HPLC chromatograms confirming the identity of the new peak as the fully 3′,5′-linked product: 6 days of hydrolysis followed by three rounds of recycling, alternated by 6 days of hydrolysis (bottom) and the same sample spiked with 7 (top).  Table S1. RNA oligonucleotides used in the present study.