CBCB Seminar

April 22, 2024 3:30 PM

Ammon-Pinizzotto Biopharmaceutical Innovation (BPI) Building
Conference Room 140

tRNA gene set evolution

Miguel Pedraza

MS graduate student, Bioinformatics and Data Science program
University of Delaware

Abstract: tRNAs are the RNA molecules that are responsible for establishing the link between mRNA and the amino acid sequence of proteins during translation. They bring the amino acids to the ribosome, which then pairs the mRNA codon with the end of tRNA at which there is an anti-codon. There are 61 possible tRNAs, all of which would be needed if translation only relied on the canonical Watson-Crick base pairing (A-U;G-C). However, the genetic code is redundant, as the 61 coding codons only encode 20 different amino acids. Thus, thanks to the weaker “wobbly pairing”, mRNA can be translated by many fewer than 61 distinct tRNAs. In fact, the tRNA gene set varies a lot between organisms. The question of its evolution is of importance; to observe it in vitro, it is possible to perform evolution experiments, whereby one engineers a bacterial strain (here SBW25) by deleting one of its tRNA genes present in only one copy (here SerCGA, thus creating the strain delSerCGA). We can subsequently observe how the resulting lineage evolves and adapts to overcome the growth defect induced by the deletion. Concretely, bacteria are cultivated in liquid cultures for time spans of 24h before being diluted in a 1:100 bottleneck with fresh medium, to start the next evolution cycle. For as long as the experiment goes on, samples can be retrieved regularly for analysis – whether phenotypic, by plating and observing the colonies’ morphologies, or genetic, by sequencing either the whole population or selected mutants.  It has been previously reported that only a few days are necessary to see the emergence of mutants that fully compensate for the initial growth defect. This is often the result of large DNA duplications that appear very easily. However, these duplications are very unstable: when isolated and cultivated overnight, a significant proportion of the population has reversed to the small colony morphotype, and when sequenced, these small colonies show a genotype identical to the initial deletion strain(delSerCGA). This raises the question of the long-term fate of these duplication mutants. At the onset of this project, several hypotheses had been raised: either a duplication mutant could dominate, possibly more stable than the others because either smaller or stabilized by an additional mutation; either another mutation (SNP) could arise independently and take over the population; different mutant types could coexist in various proportions. One question was particularly delicate: for how long should one run the evolution experiment to be able to distinguish between these hypotheses? When will a steady state be reached?  We tried to provide a partial answer to these questions with mathematical modeling.

Bio: Miguel has a background in Physics. He obtained his BS degree in Physics at National University of Colombia. Then, he joined the University of Delaware as a Master’s student in Bioinformatics and Computational Biology at the CBCB. He works with Dr. Edward Lyman, Associate Professor at the Department of Physics & Astronomy. His research is focused on Molecular Dynamics simulations of lipid bilayers, and more recently, machine learning for protein representation and property prediction.