Evolutionary Insights from the Mitochondrial Genome of Oikopleura dioica : Sequencing Challenges, RNA Editing, Gene Transfers to the Nucleus, and tRNA Loss

Abstract

Sequencing the mitochondrial genome of the tunicate Oikopleura dioica is a challenging task due to the presence of long poly-A/T homopolymer stretches, which impair sequencing and assembly. Here, we report on the sequencing and annotation of the majority of the mitochondrial genome of O. dioica by means of combining several DNA and amplicon reads obtained by Illumina and MinIon Oxford Nanopore Technologies (ONT) with public RNA sequences. We document extensive RNA editing, since all homopolymer stretches present in the mitochondrial DNA correspond to 6U-regions in the mitochondrial RNA. Out of the 13 canonical protein-coding genes, we were able to detect eight, plus an unassigned ORF that lacked sequence similarity to canonical mitochondrial protein-coding genes. We show that the nad3 gene has been transferred to the nucleus and acquired a mitochondria-targeting signal. In addition to two very short rRNAs, we could only identify a single tRNA (tRNA-Met), suggesting multiple losses of tRNA genes, supported by a corresponding loss of mitochondrial aminoacyl-tRNA synthetases in the nuclear genome. Based on the eight canonical protein-coding genes identified, we reconstructed maximum likelihood and Bayesian phylogenetic trees and inferred an extreme evolutionary rate of this mitochondrial genome. The phylogenetic position of appendicularians among tunicates, however, could not be accurately determined. Significance: Sequencing and annotating the mitochondrial genome of fast-evolving organisms is difficult because they often present unusual characteristics. The tunicate Oikopleura dioica is a model species for understanding tunicate and chordate genome evolution. However, no complete annotated mitochondrial genome for this species has been published to date. Here, we determined the major part of the mitochondrial genome of O. dioica . Our results indicate the presence of highly modified rRNA genes and the absence of all tRNAs except tRNA-Met. Moreover, we show that the mitochondrial genome undergoes editing at the RNA level. Our study demonstrates that utilizing a combination of public RNA data and DNA from long-and short-read sequencing platforms significantly improves our ability to study mitochondrial genomes with atypical characteristics.