Dynamic RNA Fitness Landscapes of a Group I Ribozyme during Changes to the Experimental Environment

Abstract

Fitness landscapes of protein and RNA molecules can be studied experimentally using high-throughput techniques to measure the functional effects of numerous combinations of mutations. The rugged topography of these molecular fitness landscapes is important for understanding and predicting natural and experimental evolution. Mutational effects are also dependent upon environmental conditions, but the effects of environmental changes on fitness landscapes remains poorly understood. Here, we investigate the changes to the fitness landscape of a catalytic RNA molecule while changing a single environmental variable that is critical for RNA structure and function. Using high-throughput sequencing of in vitro selections, we mapped a fitness landscape of the Azoarcus group I ribozyme under eight different concentrations of magnesium ions (1-48 mM MgCl2). The data revealed the magnesium dependence of 16,384 mutational neighbors, and from this, we investigated the magnesium induced changes to the topography of the fitness landscape. The results showed that increasing magnesium concentration improved the relative fitness of sequences at higher mutational distances while also reducing the ruggedness of the mutational trajectories on the landscape. As a result, as magnesium concentration was increased, simulated populations evolved toward higher fitness faster. Curve-fitting of the magnesium dependence of individual ribozymes demonstrated that deep sequencing of in vitro reactions can be used to evaluate the structural stability of thousands of sequences in parallel. Overall, the results highlight how environmental changes that stabilize structures can also alter the ruggedness of fitness landscapes and alter evolutionary processes.

Keywords: fitness landscape; gene–environment interactions; molecular evolution; noncoding RNA; ribozyme.

Fig. 1.

In vitro selection of ribozyme activity under different magnesium conditions. (A) The secondary structure of the Azoarcus ribozyme with nucleotides numbered according to intronic nucleotides. Randomized nucleotides (N) are highlighted in red. The RNA substrate used to select for reverse-splicing activity is shown as lowercase. The curved arrow indicates the reaction used for selection, which results in a portion of the substrate appended to the 3′-end of active ribozymes. (B) Experimental concept. The in vitro transcribed ribozyme library (blue tube, crystal structure 1ZZN) is reacted in different magnesium concentrations (red tubes). High-throughput sequence data are used to generate RNA fitness landscapes by determining the RNA Fitness (enrichment/depletion) of each ribozyme variant at each magnesium concentration [Mg⁺⁺].

Fig. 2.

The RNA fitness landscapes under different magnesium ion concentrations. For each graph, nodes represent individual sequences plotted with mutational distance (x axis) and fitness (y axis) relative to the wild-type ribozyme. Each node shows the average fitness of three experimental replicates. Nodes are connected by a blue edge if the two sequences differ by a single-nucleotide difference. Violin plots (gray) represent the distribution of fitness values at each mutational distance. The magnesium chloride concentration (MgCl₂) used for each data set is shown above each graph. A sequence is highlighted (yellow edges) to illustrate magnesium induced changes to fitness.

Fig. 3.

Magnesium induced changes to the fitness landscapes. (A) The mean fitness of all variants in the library, relative to wild-type, at each magnesium ion concentration ([MgCl₂]). Three data points at each concentration represent experimental replicates. The blue line traces the mean value of the triplicates at each magnesium ion concentration. (B) Landscape ruggedness calculated as ruggedness=2(f_RS)+f_S, where f_RS is the fraction of reciprocal sign epistasis and f_S is the fraction of sign epistasis for all pairs of mutations. (C) Mean fitness plotted as a function of mutational distance from wild-type. Data points represent the mean of the fitness values of all genotypes with that many mutations at that magnesium concentration. Lines are best-fit curves to the equation w(n)=exp(−αn^β). (D) Effect of magnesium concentration on the exponential decay parameter α and directional epistasis parameter β from the curve fitting.

Fig. 4.

Magnesium induced changes to individual ribozyme variants. (A) Example ribozymes with different numbers of mutations. Curves are the Hill equation fit to the fitness data. The color of the lines represents the genetic distance from wild-type, which is indicated next to each curve. (B) Curve fits to all sequences in the data, plotted by mutational distance. The line colors represent the midpoint of each curve (gradient from blue to red as the midpoint moves away from 1 mM). Only the curves with max fitness between 0.15 and 3, slope lower than 5, and midpoint lower than 16 were included. Distributions of maximum fitness (C), hill slopes (D), and midpoints from the Hill equation curve fitting, with sequences categorized by mutational distance from wild-type.

Fig. 5.

Computational simulations of populations evolving on the fitness landscapes. Evolutionary simulations were carried out with constant population size (N = 1,000) and mutational probability (µ=0.01). About 2,000 generations-long simulations were repeated 100 times on each landscape. (A–H) Population mean fitness (orange dots) of the 100 replicate simulations (light blue lines) on each landscape was recorded at each generation, and mean values were fit to a logistic equation: f(x)=L/(1 + e^(−k(x−x₀))) (red lines). The best fit curves are shown for each magnesium concentration. (I) The average fitness of the simulated populations at generation 50 (vertical dashed lines in A–H). This “G50 mean fitness” value is plotted for each magnesium concentration. (J) Change in the maximum fitness from the logistic fitted curves compared with the change in magnesium. (K) Increase in the logistic growth rate (k) of the best-fit sigmoidal curve at each magnesium concentration. (L) Change in the midpoint (x₀) of the best-fit curves at each magnesium concentration. Midpoints indicate the generations required to reach half the maximum fitness. Lower midpoints indicate faster adaptation.