Design of highly active double-pseudoknotted ribozymes: a combined computational and experimental study

Abstract

Design of RNA sequences that adopt functional folds establishes principles of RNA folding and applications in biotechnology. Inverse folding for RNAs, which allows computational design of sequences that adopt specific structures, can be utilized for unveiling RNA functions and developing genetic tools in synthetic biology. Although many algorithms for inverse RNA folding have been developed, the pseudoknot, which plays a key role in folding of ribozymes and riboswitches, is not addressed in most algorithms. For the few algorithms that attempt to predict pseudoknot-containing ribozymes, self-cleavage activity has not been tested. Herein, we design double-pseudoknot HDV ribozymes using an inverse RNA folding algorithm and test their kinetic mechanisms experimentally. More than 90% of the positively designed ribozymes possess self-cleaving activity, whereas more than 70% of negative control ribozymes, which are predicted to fold to the necessary structure but with low fidelity, do not possess it. Kinetic and mutation analyses reveal that these RNAs cleave site-specifically and with the same mechanism as the WT ribozyme. Most ribozymes react just 50- to 80-fold slower than the WT ribozyme, and this rate can be improved to near WT by modification of a junction. Thus, fast-cleaving functional ribozymes with multiple pseudoknots can be designed computationally.

Figure 1.

Workflow for computational RNA design for HDV ribozyme.

Figure 2.

Designed RNAs have sequence differences in P1, P2 and P4 stems. (A) Structure descriptor for the inverse RNA folding. Fixed nucleotides are red. (B) Pseudoknot interactions in P1 (blue box) and between P1.1 and L3 (green box) were disrupted by randomizing the seed sequence. (C) Distribution of the designed RNAs plotted on NED versus folding free energy. Threshold value was set at 0.03 (blue line). (D) Nucleotide probability in all positively designed sequences analyzed by WebLogo 3. The primary sequence of the structure descriptor is provided in the top line.

Figure 3.

Designed ribozyme DHRz 2511 possesses catalytic activity with a native HDV ribozyme-like mechanism. (A) Secondary structure of DHRz 2511. The precursor sequence from drz-spur-3 was attached to 5′ end of the designed ribozyme. The C52G variant is indicated. (B) Self-cleaving activity of the ribozyme. Catalytic activity was tested in 50 mM Mg²⁺ and absence/presence of 200 mM imidazole. The reaction is divided into four sets: ±200 mM imidazole with wild-type or C52 variant. (C) Self-cleavage site of the ribozyme. Left half: Precursor ribozyme. Right half: Cleaved ribozyme. ‘NR’: No reaction, ‘N’: Alkaline ladder, ‘T1’: RNase T1 ladder. Reactions are at 37 ^oC.

Figure 4.

Cleavage activity of designed HDV ribozymes is less than native HDV ribozyme. (A) Plot of f_cleaved versus time for DHRz 2511 (Closed circle) and CPEB3 ribozyme (Open circle). Data for CPEB3 are re-plotted from our previous study (59). Errors are S.D. (n = 4). Activity in CPEB3 and DHRz 2511 ribozyme were measured in backgrounds of 2 mM and 50 mM Mg²⁺, respectively. (B) Rate constants for designed ribozymes, including J1/2 variants, and the CPEB3 ribozyme (far right). ‘NR’ indicates catalytic activity was not detectable. Errors are S.D. (n = 4).

Figure 5.

Designed ribozymes unfold non-cooperatively but are compact. (A–D) Denaturation curves of designed ribozymes and CPEB3 ribozyme measured in the background of 0 (black), 2 (blue), 10 (green), 25 (orange) and 50 mM Mg²⁺ (red). Cleaved ribozymes were used for the experiments. See Supplementary Figure S5 for additional RNAs. (E) Global folding of the ribozymes compared by non-denaturing PAGE. (F) Fraction _{full folded RNA} calculated from non-denaturing PAGE. Errors are range of two-independent experiments.

Figure 6.

In-line reactivity in J4/2 in designed RNAs suggests structural flexibility. (A) ILP on designed ribozymes and CPEB3 ribozyme. RNase T1 ladders (G) and alkaline hydrolysis ladders (OH⁻) are denoted by ‘T1’ and ‘N’. DP is the pseudoknot-disrupted variant. (B) In-line reactivity in J4/2 relative to the CPEB3 ribozyme. Errors are S.D. (n = 3). (C) Strongest cleavage sites. Close-up for the J4/2 loop (PDB ID: 3NKB). 2′-OH of the G76 (=A53 in designed ribozyme) has high solvent accessibility. Magnesium ions are represented by the green spheres.

Figure 7.

Substitution of J4/2 in the designed CPEB3 ribozyme sequence significantly increases catalytic activity. (A) DHRz 2511 M1 in which J4/2 and portions of P2 and P4 sequences were substituted with sequences from the CPEB3 ribozyme. DHRz 2511 M2 in which P1 and P2 sequences were substituted with sequences from the CPEB3 ribozyme. DHRz 2511 M3 in which P4 sequence was substituted with sequence from the CPEB3 ribozyme. (B) Self-cleaving assay for the three domain substituted variants. (C) Scatter graph for the variants (M1: red, M2: blue and M3: gray). The same data for the CPEB3 ribozyme and the unmodified DHRz 2511 in Figure 4 were plotted for this graph. Data were analyzed by single exponential curve fitting. Errors are S.D. (n = 4). (D) Plot of k_obs values. Relative activity (given at top of bars) was calculated relative to unmodified DHRz 2511. Errors are S.D. (n = 4).

Figure 8.

Proposed model of weaker protonation of C52 in designed RNAs. (A) J4/2 in the HDV ribozyme illustrated based on the crystal structure (PDB ID: 3NKB). Protonated C75⁺ acts as general acid. G76 is flipped out to solvent side, and A77 is stacked with A78 via π–π interaction. A magnesium ion binds with the phosphates of both A77 and A78. (B) J4/2 in the designed RNAs is proposed to be structurally labile. C52 and A53 fluctuate in the catalytic cavity, and the 2′OH of A53 is oriented to solvent side, which could lead this region to have high in-line reactive for A53 and insufficient protonation for C52.