International team maps 'rugged' fitness landscape of antibiotic resistance in E. coli

In a recent paper, an international group of scientists presented experimental data on E. coli genes, examining how the theory of a fitness landscape model corresponds to the reality of a rugged landscape, one with many peaks and valleys. 

A fitness landscape model has mountain peaks and valleys where each spatial location represents a genotype, and the elevation at that location represents the evolutionary fitness of that genotype. The best adapted genotypes would be on the highest peaks.

This type of evolutionary biology mapping was developed about 100 years ago by the American geneticist Sewall Wright, who proposed that natural selection would lead to a population climbing its nearest peak, while genetic drift would result in random wandering.

Theory had predicted that the ruggedness would impede natural selection by keeping a genotype from crossing the valleys and optimally evolving to reach high fitness peaks. The research study found otherwise. 

The research work is published as a preprint in the Feb. 28 issue of bioRxiv. The authors are Andrei Papkou, Lucia Garcia-Pastor, José Antonio Escudero and Andreas Wagner.

Testing the theory

Although subsequent models to Wright's supported his concern that "multi-peaked landscapes may prevent adaptive Darwinian evolution driven by natural selection," there was a lack of experimental evidence on this point, the researchers note. Their experiment seeks to fill that gap.

"Here we address the unresolved and fundamental question of the relationship between ruggedness and peak accessibility by mapping a large and combinatorially complete in vivofitness landscape of the E. coli folA gene, which encodes the essential metabolic enzyme and antibiotic resistance protein dihydrofolate reductase (DHFR)," they wrote. "We find that this landscape is rugged, but its ruggedness does not preclude evolving populations from accessing high fitness peaks." 

In all the researchers created a combinatorially complete library of more than 262,000 DNA genotypes of folA, using CRISPR-Cas9 gene editing. Then they exposed a population of E. coli cells to the antibiotic trimethoprim, which targets DHFR. The trimethoprim was at a sub-lethal dose.

Plotting mutations

The researchers designed the experiment to plot the mutational trajectories of variations of DHFR. Each vertex on the fitness landscape graph represents one variant of DHFR. They plotted the mutational changes in each variant's evolutionary path under specific conditions, and they examined the number of variants at the basin of a peak that might access the peak. 

They calculated the adaptive "walk" of variants as they took each mutational step, and they looked at the paths of variants that were in more than one basin.

Although the DHFR landscape is rugged, they conclude: "Its highest fitness peaks are easily accessible to evolving populations. Fitness-increasing paths to high fitness peaks are abundant, and individual peaks have large basins of attractions. The basins of different peaks overlap, which renders the outcome of adaptive evolution highly contingent on chance events."

Next steps

The researchers note that they chose a gene in which even one mutation can lead to "profound changes in fitness." Future experiments of different organisms, mutations and different designs will show whether or not their results are typical.

The authors conclude: "In sum, ruggedness need not be an obstacle to Darwinian evolution but can reduce its predictability. If true in general, evolutionary biology and other fields of sciences in which landscapes play an important role may have to re-appraise the complexity of optimization problems on realistic landscapes."

--------

Andrei Papkou et al. "A rugged yet easily navigable fitness landscape of antibiotic resistance." bioRxiv, Feb. 28.

DOI: https://doi.org/10.1101/2023.02.27.530293