Back

A genetic code change in progress: tRNA-Leu(CAG) is conserved in most Saccharomycopsis yeast species but is non-essential and does not compete with tRNA-Ser(CAG) in translation

Cinneide, E. O.; Scaife, C.; Dillon, E.; Wolfe, K. H.

2023-07-05 evolutionary biology
10.1101/2023.07.05.547863 bioRxiv
Show abstract

In the yeast genera Saccharomycopsis and Ascoidea, nuclear genes use a non-standard genetic code in which CUG codons are translated as serine instead of leucine, due to a tRNA-Ser with the unusual anticodon CAG. However, some species in this CUG-Ser2 clade also contain an ancestral tRNA-Leu gene with the same anticodon. One of these species, Ascoidea asiatica, has been shown to have a stochastic proteome in which proteins contain approximately 50% Ser and 50% Leu at CUG codon sites, whereas previously examined Saccharomycopsis species translate CUG only as Ser. Here, we investigated the presence, conservation, and possible functionality of the tRNA-Leu(CAG) gene in the genus Saccharomycopsis. We analyzed the genomes of 33 strains, including almost all known species of Saccharomycopsis, and found that most of them contain both tRNA-Ser(CAG) and tRNA-Leu(CAG) genes. The tRNA-Leu(CAG) gene is evolving faster than tRNA-Ser(CAG) and it has been lost in two species, S. microspora and S. synnaedendra. We deleted the single tRNA-Leu(CAG) gene in S. capsularis and found that it is not essential. Bioinformatic analysis suggested that some CUG codon sites in Saccharomycopsis species may be translated as Leu, specifically in genes with functions in meiosis or sporulation, but mass spectrometry of sporulating S. capsularis and S. fermentans cultures showed only CUG-Ser translation. Cloverleaf structures of tRNA-Leu(CAG) from all Saccharomycopsis species contain mutations that are likely to make them non-functional in translation, but the evolutionary conservation of the gene leads us to propose that it has been retained for an unknown non-translational role.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.