Computational design of three-dimensional RNA structure and function

Abstract

RNA nanotechnology seeks to create nanoscale machines by repurposing natural RNA modules. The field is slowed by the current need for human intuition during three-dimensional structural design. Here, we demonstrate that three distinct problems in RNA nanotechnology can be reduced to a pathfinding problem and automatically solved through an algorithm called RNAMake. First, RNAMake discovers highly stable single-chain solutions to the classic problem of aligning a tetraloop and its sequence-distal receptor, with experimental validation from chemical mapping, gel electrophoresis, solution X-ray scattering and crystallography with 2.55 Å resolution. Second, RNAMake automatically generates structured tethers that integrate 16S and 23S ribosomal RNAs into single-chain ribosomal RNAs that remain uncleaved by ribonucleases and assemble onto messenger RNA. Third, RNAMake enables the automated stabilization of small-molecule binding RNAs, with designed tertiary contacts that improve the binding affinity of the ATP aptamer and improve the fluorescence and stability of the Spinach RNA in cell extracts and in living Escherichia coli cells.

Figure 1:. Problems in RNA nanotechnology reduced to RNA motif pathfinding problems and solved by RNAMake.

(a) ‘miniTTRs’ require two strands (green, purple) between tetraloop (orange) and tetraloop-receptor (blue); (b) tethered ribosomes require two strands (green, purple) to link the small subunit (orange) to the large subunit (blue). c) ‘Locking’ a small-molecule binding aptamer (cyan; ATP molecule in pink spheres) by designing four strands (green, purple, teal, magenta) to a peripheral tertiary contact (orange, blue). d) Demonstration of RNAMake design algorithm, which builds an RNA path via the successive addition of motifs and helices from a starting base pair to the ending base pair. e-f) computational efficiency for RNAMake to design connections between each pair of hairpins on the 50S E. coli ribosome. Run time scales linearly with problem size, as measured by (e) translational distance between helical endpoints or (f) number of residues required for segments (see Figure S14 for utilizing higher order junctions).

Figure 2:. Solving the TTR design problem.

a) Quantification of DMS reactivity in the absence of Mg²⁺ (red) and with 10 mM Mg²⁺ (blue) for RNAMake-designed miniTTR constructs and the P4-P6 domain of the Tetrahymena ribozyme as a large natural RNA comparison. Diagram (inset) shows the four adenosines in the tetraloop/tetraloop-receptor (red) that undergo protection upon Mg²⁺-dependent folding. b) Native gel assays testing whether mutation of GAAA tetraloop (left lane) to UUCG mutant (right lane) disrupts the miniTTR tertiary fold and slows its mobility. ‘MedLoop’ lanes are a control RNA with similar size . Figure S3b gives other remaining constructs. c) Quantification of the miniTTR folding stability based on Mg²⁺ binding curves read out by DMS mapping for miniTTR 6 (blue), miniTTR 2 (red) and P4-P6 (black). d) Raw data from the Mg²⁺ titration of miniTTR 2, highlighting the change in DMS reactivity in the TTR and the motifs used in the design. e-f) SAXS analysis: Experimental intensity versus scattering amplitude and low-resolution reconstruction derived from experimental scattering profiles (blue beads, inset) overlaid on RNAMake-designed model (cartoon, inset) for (e) miniTTR 2 and (f) miniTTR 6. In (f), SAXS prediction from miniTTR crystal structure is also shown (blue). g-l) X-ray crystal structure of MiniTTR 6 tests accuracy of RNAMake model at atomic resolution: (g) overall RNAMake and X-ray structure, and magnified views of (h) triple mismatch motif from ribosome, (i) tetraloop/tetraloop-receptor, (j) kink-turn motif, and (k) right angle turn. In (h-k), crystal structures are white.

Figure 3:. The ribosome tethering problem

a-d) Modelling (top panel) of tethers (green and magenta strands) connecting 16S and 23S rRNA into a single ribosomal RNA; and agarose gel electrophoresis (bottom panel) of RNA extracted from E. coli (Squires strain) in which wild type ribosomes were completely replaced with the following designed molecules: (a) an early U3 tether (Umbilical 3) designed by the Jewett lab, cleaved into two bands in vivo, (b) (top) Stapled ribosomes developed by a separate group (three replicates shown), also cleaved in this plasmid context and strain in vivo, (c) successful Ribo-T design, developed after a year of manual trial-and-error to withstand cleavage in vivo, (d) RM-Tether 4, design automatically generated by RNAMake, which also presents as a single band in vivo. e) Sucrose gradient fractionation prepared from in vitro ISAT reactions expressing wild type ribosomes (black), Ribo-T v1.0 (blue), and RM-Tether 4 (green). Peaks corresponding to small subunits (30S), large subunits (50S), monosomes/70S, and polysomes are indicated; see, e.g., ref. for standard assignment of peaks. f) Agarose electrophoresis analysis confirms that the polysome fraction of (e) is composed of tethered ribosomes. Full gels can be found in Figure S5 and S6.

Figure 4:. Stabilizing aptamers for ATP and light-up fluorophores through designer tertiary contacts.

a) 3D models of the ATP aptamer alone and ATP-TTRs 3, 4, and 5. b) DMS probing of ATP titration of the ATP aptamer, red box denotes the two adenines that become protected upon addition of ATP. c) DMS probing of ATP titration of ATP-TTR 4, red box denotes the two adenines that become protected upon addition of ATP. d) Quantified DMS protection as a function of ATP concentration for the ATP aptamer compared to ATP-TTRs 3, 4 and 5. e) The crystal structure of the Spinach aptamer bound to DFHBI (left), 3D models of Spinach-TTRs 3, 8, and 10 (right). f) Fluorescence measurements of a DFHBI titration at constant RNA concentration of Spinach, Broccoli and Spinach-TTR 3, 8 and 10. g) Fluorescence measurements of an RNA titration with constant DFHBI concentration of Spinach, Broccoli and Spinach-TTR 3, 8 and 10. h) Fluorescence in 20% lysate as compared to the construct in buffer for Spinach, Broccoli and Spinach-TTR 3, 8 and 10.