F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.21551.1 Research Article Articles Widespread use of the “ascidian” mitochondrial genetic code in tunicates

[version 1; peer review: 2 approved]

 Pichon Julien Conceptualization Formal Analysis Funding Acquisition Investigation Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0002-8065-0024 1 2 Luscombe Nicholas M. Funding Acquisition Project Administration Supervision Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-5293-4778 1 3 4 Plessy Charles Conceptualization Data Curation Formal Analysis Investigation Methodology Project Administration Supervision Visualization Writing – Original Draft Preparation Writing – Review & Editing https://orcid.org/0000-0001-7410-6295 a 1 Genomics and Regulatory Systems Unit, Okinawa Institute of Science and Technology Graduate University, Onna-son, Okinawa, 904-0495, Japan Université de Paris, Paris, France The Francis Crick Institute, London, NW1 1AT, UK Genetics Institute, University College London, London, WC1E 6BT, UK charles.plessy@oist.jp

No competing interests were disclosed.

 10 12 2019 2019 8 2072 4 12 2019 Copyright: © 2019 Pichon J et al. 2019 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Background: Ascidians, a tunicate class, use a mitochondrial genetic code that is distinct from vertebrates and other invertebrates. Though it has been used to translate the coding sequences from other tunicate species on a case-by-case basis, it is has not been investigated whether this can be done systematically. This is an important because a) some tunicate mitochondrial sequences are currently translated with the invertebrate code by repositories such as NCBI GenBank, and b) uncertainties about the genetic code to use can complicate or introduce errors in phylogenetic studies based on translated mitochondrial protein sequences.

Methods: We collected publicly available nucleotide sequences for non-ascidian tunicates including appendicularians such as Oikopleura dioica, translated them using the ascidian mitochondrial code, and built multiple sequence alignments covering all tunicate classes.

Results: All tunicates studied here appear to translate AGR codons to glycine instead of serine (invertebrates) or as a stop codon (vertebrates), as initially described in ascidians. Among Oikopleuridae, we suggest further possible changes in the use of the ATA (Ile → Met) and TGA (Trp → Arg) codons.

Conclusions: We recommend using the ascidian mitochondrial code in automatic translation pipelines of mitochondrial sequences for all tunicates. Further investigation is required for additional species-specific differences.

 Tunicate Oikopleura Genetic code Mitochondria Cytochrome oxidase subunit I Okinawa Institute of Science and Technology Graduate University Ministère de l'Enseignement Supérieur, de la Recherche Scientifique et de la Formation des Cadres JP’s internship at OIST was supported by the French Ministry for Education and Research and by laboratory funding from OIST The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction

Tunicates are marine animals that have acquired the capacity to produce cellulose by horizontal gene transfer approximately 500 million years ago ( Matthysse et al., 2004; Nakashima et al., 2004). Together with vertebrates and cephalochordates, they belong to the chordate phylum, in which they share morphological features such as a muscular tail during larval stages. Phylogenetic studies place the tunicates as the closest living relatives of vertebrates ( Delsuc et al., 2006). Tunicates can be subdivided in three classes: Thaliacea (free-swimming colonial species, for instance salps or dolioids), Appendicularia (free-swimming solitary species with an adult morphologically similar to the larval stage of other tunicates), and Ascidiacea (attached to solid substrates in their adult stage, for instance sea squirts). The relationship between these classes and therefore their mono- or paraphyly has been revised multiple times. For instance the 18S rRNA analysis of Stach & Turbeville (2002) nested Appendicularia within Ascidiacea, but more recently Delsuc et al. (2018) placed them as sister groups using a multigene approach. The paraphyly of Ascidiacea is now widely accepted, as the above studies and others demonstrated that they contain the Thaliacea.

Mitochondrial genomes undergo major changes at the geological time scale due to their small size and clonal reproduction, including changes to their genetic code ( Osawa et al., 1992). The first evidence that ascidians use a specific mitochondrial genetic code stemmed from observations that the cytochrome c oxidase subunit 1 ( Cox1) sequence from Halocynthia roretzi ( Yokobori et al., 1993) and the Cox3 sequence of Pyura stolonifera ( Durrheim et al., 1993) are interrupted by stop codons if translated using the vertebrate mitochondrial code. Reassignment of AGR codons to glycine was later confirmed by the discovery of a glycine (Gly) tRNA in the H. roretzi genome ( Yokobori et al., 1999) and by the sequencing of its anticodon (U*CU) ( Kondow et al., 1999). Apart from the AGR codons, the ascidian code is similar to the vertebrate and the invertebrate ones, with ATA assigned to methionine (Met) and TGA to tryptophan (Trp) ( Yokobori et al., 1993).

This genetic code is known as the “ascidian” genetic code; however, it is also used by non-ascidian tunicates, such as the thaliacean Doliolum nationalis ( Yokobori et al., 2005). The possibility that this genetic code emerged earlier than tunicates was raised by the study of partial genome sequences of Branchiostoma lanceolatum ( Delarbre et al., 1997) leading to the proposition that AGR might encode Gly in cephalochordates. While this seemed to be supported by the discovery of a putative TCT (Gly) tRNA in the full mitochondrial genome of B. lanceolatum ( Spruyt et al., 1998), this hypothesis was later ruled out by an analysis of the related amphioxus Branchiostoma floridae ( Boore et al., 1999), and has not been reconsidered since. Finally, studies on the appendicularian branch showed compatibility between the mitochondrial sequence of Oikopleura dioica and the ascidian code ( Denoeud et al., 2010). Nevertheless, support for compatibility was not demonstrated explicitly for the ATA and TGA codons and the mitochondrial sequence of O. dioica were not released in International Nucleotide Sequence Database Collaborations (INSDC) databanks.

Cox1 is the most conserved mitochondrial protein. Although no mitochondrial genome has been fully sequenced yet for appendicularians, partial Cox1 sequences are present in the INSDC databanks for Oikopleuridae. Sakaguchi et al. (2017) reported that all Oikopleura mitochondrial sequences ( AY116609– AY116611 and KF977307) may be contaminations from bacteria or cnidarians, and provided partial sequences for Oikopleura longicauda in the same study. Partial mitochondrial sequences were published for Bathochordaeus and Mesochordaeus species by Sherlock et al. (2017). In addition, Naville et al. (2019) recently published draft genome for several appendicularian species. Therefore, to assess whether the ascidian mitochondrial code is used across the whole tunicate subphylum, we took advantage of these public data and prepared a curated alignment of Cox1 sequences comprising representatives of the major tunicate branches, to study the consensus sequences at conserved residues.

 Methods

We identified Cox1 and Cytochrome b ( Cob) gene sequences for Oikopleura longicauda, Mesochordaeus erythrocephalus and Bathochordaeus stygius by screening published genome assemblies ( Naville et al., 2019) with the partial Cox1 sequence of O. longicauda LC222754.1 ( Sakaguchi et al., 2017) using tblastn and the ascidian mitochondrial code ( -db_gencode=13) ( Gertz et al., 2006). Mitochondrial genome sequences were then translated using the cons and getorf commands from EMBOSS ( Rice et al., 2000), using the ascidian mitochondrial code.

<italic toggle="yes">Oikopleura longicauda</italic> 

We identified the circular contig SCLD01101138.1 (length: 10,324 nt) as a potential mitochondrial genome, and translated Cox1 from position 4530 to 6230. We also translated Cob from 3697 to 4668.

 <italic toggle="yes">Mesochordaeus erythrocephalus</italic> 

We translated Cox1 in contig SCLF01725989.1 (length 7,034 nt) on reverse strand from position 1792 to 272. Using the same method with O. longicauda’s Cob sequence as a bait, we also recovered a Cob sequence from contig SCLF01109548.1 (length 5,010 nt), reverse strand, 1604 to 2590.

 <italic toggle="yes">Bathochordaeus stygius</italic> 

We used the consensus of the published B. stygius Cox1 sequences KX599267.1 to KX599281.1 from GenBank ( Sherlock et al., 2017), to screen the genome and scaffold SCLE01415711.1 (length 10,388 nt) gave a perfect hit. We translated Cox1 from position 8054 to 6522 on the reverse strand, and a partial Cob sequence from scaffold SCLE01415711.1 (2319 to 2963, reverse strand). We also found a second fragment aligning well with C-terminal sequences between positions 2373 and 1978, but we did not include it due to the difficulty of resolving the overlap between both fragments. When screening with the M. erythrocephalus Cox1 sequence recovered above, we found that another scaffold SCLE01416475.1 gave a perfect hit, hinting at a possible contamination.

 <italic toggle="yes">Oikopleura dioica</italic> 

To assemble a Cox1 sequence in O. dioica, we downloaded expressed sequence tags (file 10_ESTall.txt) from Oikobase ( Danks et al., 2012) and extracted hits matching the O. longicauda sequence using tblastn (see above). We then aligned and visualised the hits using Clustal Omega ( Sievers et al., 2011) and SeaView ( Gouy et al., 2009), filtering out those too short or introducing gap columns. Inspection of the alignment let us notice three possible haplotypes. We generated a consensus for each of them, translated them (see above) and trimmed the proteins sequences in order to match the length of the other reference sequences in the alignment. All variants found between the haplotypes were synonymous codons. We used the same methodology to generate a consensus for Cob and translate it.

 <italic toggle="yes">Cox1</italic> accession numbers

Bathochordaeus charon KT881544.1 ORF2 translated with ascidian code; Bathochordaeus stygius: SCLE01415711.1[8054:6522] translated with ascidian code; Branchiostoma lanceolatum: BAD93656.1; Caenorhabditis elegans: NP_006961.1; Ciona intestinalis: CAL23359.2; Clavelina oblonga: YP_009029840.1; Doliolum nationalis: BAD86512.1; Halocynthia roretzi: NP_038239.1; Mesochordaeus erythrocephalus: SCLF01725989.1[1915:260] translated with ascidian code; Mus musculus: NP_904330.1; Oikopleura dioica: consensus of Oikobase contigs (see file 10_ESTall.txt) KT0AAA24YA11, KT0AAA22YO17, KT0AAA22YO04, KT0AAA13YK14, KT0AAA18YK22, KT0AAA16YP04, KT0AAA13YE23, KT0AAA8YH10, KT0AAA4YK01, KT0AAA24YE23, KT0AAA18YO18, KT0AAA3YP19, KT0AAA10YF12; O. longicauda: SCLD01101138.1[4678:6230] translated with ascidian code; Salpa thompsoni: BBB04277.1.

 <italic toggle="yes">Cob</italic> accession numbers

Bathochordaeus stygius: SCLE01415711.1[2963:2319] translated with ascidian code; Branchiostoma lanceolatum: BAD93666.1; Caenorhabditis elegans: NP_006958.1; Ciona intestinalis: CAL23352.2; Clavelina oblonga: YP_009029843.1; Doliolum nationalis: BAD86520.1; Halocynthia roretzi: NP_038246.1; Mesochordaeus erythrocephalus: SCLF01109548.1[1604:2590] translated with ascidian code; Mus musculus: NP_904340.1; Oikopleura dioica: consensus of Oikobase contigs KT0AAA23YJ17, KT0AAA16YJ22, KT0AAA17YO14, KT0AAA10YI15, KT0AAA18YI18, KT0AAA11YF07, KT0AAA10YG05, KT0AAA1YH02, KT0AAA12YH10, KT0AAA12YC07, KT0AAA12YC07, KT0AAA18YM15 (see file 10_ESTall.txt); O. longicauda: SCLD01101138.1[3697:4668] translated with ascidian code; Salpa thompsoni: BBB04269.1.

 Sequence alignments

Translated Cox1 and Cob sequences were aligned using Clustal Omega ( Sievers et al., 2011) and SeaView ( Gouy et al., 2009). The alignments were post-processed using the showalign -show=n command of EMBOSS ( Rice et al., 2000) to show the differences to the inferred consensus. Graphical processing of the alignments were performed with Jalview ( Waterhouse et al., 2009). The codon sequences encoding Cox1 and Cob of the tunicate species were then added aligned to the corresponding amino-acid (three lines per species, see Extended data ( Plessy & Pichon, 2019)) and then the text files were transposed, so that each line would correspond to a single position in the alignment, and interrogated with custom Unix commands to compute the tables presented in this manuscript.

 Results AGR encodes for Gly across all tunicates

We selected species according to sequence availability and to ensure coverage of the tunicate subphylum in a way that stays broad under the various hypotheses of monophyly or paraphyly for its major groups. For ascidians, we have included the phlebobranchian Ciona intestinalis, the aplousobranchian Clavelina oblonga and the pyurid stolidobranchian Halocynthia roretzi. For thaliaceans, we selected Doliolum nationalis and Salpa thompsoni. For appendicularians we selected Oikopleura dioica, Oikopleura longicauda, Bathochordaeus stygius and Mesochordaeus erythrocephalus. We ensured that all tunicate sequences were translated with the ascidian mitochondrial genetic code. Lastly, we included outgroup sequences from Caenorhabditis elegans and Branchiostoma lanceolatum (invertebrate mitochondrial code) and from Mus musculus (vertebrate mitochondrial code) to better highlight conserved amino acid positions. In Figure 1, we illustrate the relation between these species based on the phylogeny of Naville et al. (2019) for appendicularians and of Delsuc et al. (2018) for the other tunicates. We prepared Cox1 sequences from the selected species using mitochondrial genomes (for ascidians, thaliaceans, and outgroups), from draft genomes in which we found a putative mitochondrial contig after screening with a partial or a related Cox1 sequence (for O. longicauda, B. stygius, and M. erythrocephalus) and from EST sequences (for O. dioica). We aligned the translated Cox1 and Cob sequences ( Figure 2 and Figure 3) and inspected the positions where all species use the same amino acid. Conserved glycines supported the use of AGR codons across the whole tunicate clade. We confirmed this observation with Cob sequences obtained with the same method.

Left: Cladogram illustrating the relations between the species selected in study.

Different branch colors indicate different mitochondrial genetic codes. Codon assignments with an equal sign indicate how the nucleotide sequences were translated. Codon assignments with a question mark indicate a possible finding, but were not used for translation. Ascidians, in which the AGR to Gly codon reassignment was initially discovered, are highlighted among the tunicates. Right: codon sequence of Cox1 genes on positions where proposed changes of genetic code would make all species use the same amino acid.

 Sequence alignment of Cox1 proteins.

White stars indicate conserved cysteines when at least one tunicate uses an AGR codon. Black stars indicate positions suggesting a different genetic code.

 Sequence alignment of Cob proteins.

White stars indicate conserved cysteines when at least one tunicate uses an AGR codon. Black stars indicate positions suggesting a different genetic code.

 Possible lineage specific use of ATA Ile and TGA Arg codons

We then searched for positions where a single tunicate species differed from the other sequences with the same replacement amino acid more than once. We found multiple cases of methionine being replaced by isoleucine and arginine replaced by tryptophan in O. longicauda and B. stygius ( Figure 2). Given their phylogenetic proximity, we grouped the two species in the analysis below and we calculated the number of mismatches to the other sequences. We redefined a position as “conserved” if there is at most one mismatch from one sequence to the others.

M. erythrocephalus does not seem to use ATA codons and O. longicauda and B. stygius use ATA codons at positions where all other species had an isoleucine (Ile) ( Table 1 and Table 2). In the ancestral invertebrate mitochondrial code and the sister vertebrate code, ATA encodes Met. Although Met and Ile both have hydrophobic side chains that often can substitute for each other, this also suggests a change of the genetic code. Evidence for this is that 1) non-appendicularian species do not display ATA codons at positions where all other species encode Ile; 2) the change would be parsimonious as O. longicauda, B. stygius and M. erythrocephalus are more closely related to each other than to O. dioica ( Naville et al., 2019); and 3) these three species never have ATA codons at positions where Met is conserved in every species (in contrast to O. dioica). Furthermore, reversion of the ATA codon to Ile have occurred in other branches of the tree of Life, for instance in echinoderms ( Jacobs et al., 1988). Finally, inspection of a partial Cox1 sequence of the related Bathochordaeus charon (KT881544.1) provided one extra instance of an ATA codon at a conserved Ile position.

ATN codons in <italic toggle="yes">Cox1</italic> in Oikopleuridae.
O. dio. O. lon. B. sty. M. ery.
ATA number 22 12 16 0
ATC number 3 8 0 2
ATG number 14 34 25 33
ATT number 29 18 21 38
ATA on cons. Met 5 0 0 0
ATA on cons. Ile 0 4 2 0
ATN codons in <italic toggle="yes">Cob</italic> in Oikopleuridae.
O. dio. O. lon. B. sty. M. ery.
ATA number 6 9 9 0
ATC number 5 2 1 2
ATG number 9 9 8 16
ATT number 21 11 11 22
ATA on cons. Met 0 0 0 0
ATA on cons. Ile 0 1 1 0

The TGA codon is known to encode tryptophan (Trp) in vertebrate, invertebrate and ascidian mitochondria ( Fox, 1979). We found that B. stygius uses TGA at positions where all other species would encode Arg ( Table 3 and Table 4). This is surprising as these two amino acids are unlikely to functionally substitute for each other. O. longicauda does not use TGA codons, and M. erythrocephalus does not use TGA at conserved Arg, although it is found at a position where all other species encode for Arg except C. elegans which encodes lysine, the other positively charged amino-acid. This again suggests a possible change of genetic code, although the numbers are currently too small to draw a solid conclusion.

TGR codons in <italic toggle="yes">Cox1</italic> in Oikopleuridae.
O. dio. O. lon. B. sty. M. ery.
TGA number 2 0 3 1
TGG number 13 16 19 16
TGA on cons. Trp 1 0 0 0
TGA on cons. Arg 0 0 2 0
TGR codons in <italic toggle="yes">Cob</italic> in Oikopleuridae.
O. dio. O. lon. B. sty. M. ery.
TGA number 3 0 2 1
TGG number 4 7 4 5
TGA on cons. Trp 1 0 0 0
TGA on cons. Arg 0 0 1 0
Discussion

We extracted Cox1 and Cob sequences of four different appendicularians from public databases. As a nucleotide sequence, Cox1 might be useful for mining databases of molecular barcodes sequenced from the environment, or for studies of population diversity within a species. As a protein sequence, Cox1 might be useful for refining the phylogeny of appendicularians. However, a translation code needs to be chosen.

Our alignments of tunicate Cox1 and Cob protein sequences support the view that all tunicates translate AGR codons as Gly (although this conclusion might be limited by the lack of coverage for the Kowalevskiidae and Fritillariidae families). While our analysis suggests that the last common ancestor of the tunicates used the “ascidian” code, it is not possible to conclude that all contemporary tunicates still do, as we found discrepancies on other conserved residues that could be explained by a genetic code change of ATA and TGA codons within a sub-clade of the appendicularians containing M. erythrocephalus, O. longicauda and B. stygius.

The “ascidian” genetic code is table number 13 in the NCBI protein database, where it is used to translate sequences from ascidians and non-ascidian tunicates, for instance D. nationalis. However for appendicularians, the NCBI currently applies the invertebrate table (number 5). This has the consequences of turning Gly to Ser at functionally important positions. Therefore, the ascidian is probably a more appropriate default. At present, it is unclear whether some appendicularians have additional changes; however, the accurate translation of AGR codons to Gly would nonetheless reduce the amount of error in translated protein sequences.

To confirm a change of genetic code, it is necessary to detect corresponding changes in the respective tRNAs. This beyond reach for the present study because the mitochondrial genomic sequences that we used are extracted from draft genome sequences that may be incomplete, or even contain contaminations (see B. stygius in the Methods section). As a result, we also cannot entirely rule out the possibility that we have examined pseudogenes, although the high conservation found in the alignments suggest this in unlikely. For all these reasons, it is necessary to sequence full-length mitochondrial genomes from appendicularians.

 Conclusions

Our alignments of translated mitochondrial sequences suggest that the last common ancestor of living tunicates may have already used the “ascidian” genetic code. Thus, we recommend the use of that code instead of the “invertebrate” one for all tunicates in automatic translation pipelines, with the caveat that additional changes might be found in appendicularians. This observation is a reminder that in biology, exception is the rule, and that each time a mitochondrial sequence is extracted from a species for the first time, it is important to carefully examine its genetic code.

 Data availability Underlying data

All data underlying the results are available as part of the article and no additional source data are required.

 Extended data

Zenodo: Aligned Cox1 and Cob sequences from Oikopleura dioica and other tunicates. https://doi.org/10.5281/zenodo.3490310 ( Plessy & Pichon, 2019).

This project contains alignment files and descriptions of how the files were generated.

Extended data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).

 Acknowledgements

We thank the OIST’s Scientific Computing & Data Analysis Section for their support, and Ferdinand Marlétaz for critical comments on our manuscript.

 Boore JL Daehler LL Brown WM : Complete sequence, gene arrangement, and genetic code of mitochondrial DNA of the cephalochordate Branchiostoma floridae (Amphioxus). Mol Biol Evol. 1999;16(3):410–418. 10331267 10.1093/oxfordjournals.molbev.a026122 Danks G Campsteijn C Parida M : OikoBase: a genomics and developmental transcriptomics resource for the urochordate Oikopleura dioica. Nucleic Acids Res. 2012;41(Database issue):D845–D853. 23185044 10.1093/nar/gks1159 3531137 Delarbre C Barriel V Tillier S : The main features of the craniate mitochondrial DNA between the ND1 and the COI genes were established in the common ancestor with the lancelet. Mol Biol Evol. 1997;14(8):807–813. 9254918 10.1093/oxfordjournals.molbev.a025821 Delsuc F Brinkmann H Chourrout D : Tunicates and not cephalochordates are the closest living relatives of vertebrates. Nature. 2006;439(7079):965–968. 16495997 10.1038/nature04336 Delsuc F Philippe H Tsagkogeorga G : A phylogenomic framework and timescale for comparative studies of tunicates. BMC Biol. 2018;16(1):39. 29653534 10.1186/s12915-018-0499-2 5899321 Denoeud F Henriet S Mungpakdee S : Plasticity of animal genome architecture unmasked by rapid evolution of a pelagic tunicate. Science. 2010;330(6009):1381–1385. 21097902 10.1126/science.1194167 3760481 Durrheim GA Corfield VA Harley EH : Nucleotide sequence of cytochrome oxidase (subunit III) from the mitochondrion of the tunicate Pyura stolonifera: evidence that AGR encodes glycine. Nucleic Acids Res. 1993;21(15):3587–3588. 8393993 10.1093/nar/21.15.3587 331473 Fox TD : Five TGA "stop" codons occur within the translated sequence of the yeast mitochondrial gene for cytochrome c oxidase subunit II. Proc Natl Acad Sci U S A. 1979;76(12):6534–6538. 230513 10.1073/pnas.76.12.6534 411900 Gertz EM Yu YK Agarwala R : Composition-based statistics and translated nucleotide searches: improving the TBLASTN module of BLAST. BMC Biol. 2006;4:41. 17156431 10.1186/1741-7007-4-41 1779365 Gouy M Guindon S Gascuel O : SeaView version 4: A multiplatform graphical user interface for sequence alignment and phylogenetic tree building. Mol Biol Evol. 2009;27(2):221–224. 19854763 10.1093/molbev/msp259 Jacobs HT Elliott DJ Math VB : Nucleotide sequence and gene organization of sea urchin mitochondrial DNA. J Mol Biol. 1988;202(2):185–217. 3172215 10.1016/0022-2836(88)90452-4 Kondow A Suzuki T Yokobori S : An extra tRNA Gly(U*CU) found in ascidian mitochondria responsible for decoding non-universal codons AGA/AGG as glycine. Nucleic Acids Res. 1999;27(12):2554–2559. 10352185 10.1093/nar/27.12.2554 148460 Matthysse AG Deschet K Williams M : A functional cellulose synthase from ascidian epidermis. Proc Natl Acad Sci U S A. 2004;101(4):986–991. 14722352 10.1073/pnas.0303623101 327129 Nakashima K Yamada L Satou Y : The evolutionary origin of animal cellulose synthase. Dev Genes Evol. 2004;214(2):81–88. 14740209 10.1007/s00427-003-0379-8 Naville M Henriet S Warren I : Massive Changes of Genome Size Driven by Expansions of Non-autonomous Transposable Elements. Curr Biol. 2019;29(7):1161–1168.e6. 30880010 10.1016/j.cub.2019.01.080 Osawa S Jukes TH Watanabe K : Recent evidence for evolution of the genetic code. Microbiol Rev. 1992;56(1):229–264. 1579111 372862 Plessy C Pichon J : Aligned Cox1 and Cob sequences from Oikopleura dioica and other tunicates.2019. http://www.doi.org/10.5281/zenodo.3490310 Rice P Longden I Bleasby A : EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet. 2000;16(6):276–277. 10827456 10.1016/s0168-9525(00)02024-2 Sakaguchi SO Ikuta T Ogawa G : Morphological identity of a taxonomically unassigned cytochrome c oxidase subunit i sequence from stomach contents of juvenile chum salmon determined using polymerase chain reaction. Fish Sci. 2017;83(5):757–765. 10.1007/s12562-017-1106-0 Sherlock RE Walz KR Schlining KL : Morphology, ecology, and molecular biology of a new species of giant larvacean in the eastern North Pacific: Bathochordaeus mcnutti sp. nov. Mar Biol. 2017;164(1):20. 28042175 10.1007/s00227-016-3046-0 5159439 Sievers F Wilm A Dineen D : Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011;7:539. 21988835 10.1038/msb.2011.75 3261699 Spruyt N Delarbre C Gachelin G : Complete sequence of the amphioxus ( Branchiostoma lanceolatum) mitochondrial genome: relations to vertebrates. Nucleic Acids Res. 1998;26(13):3279–3285. 9628930 10.1093/nar/26.13.3279 147690 Stach T Turbeville JM : Phylogeny of Tunicata inferred from molecular and morphological characters. Mol Phylogenet Evol. 2002;25(3):408–428. 12450747 10.1016/s1055-7903(02)00305-6 Waterhouse AM Procter JB Martin DM : Jalview Version 2--a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009;25(9):1189–1191. 19151095 10.1093/bioinformatics/btp033 2672624 Yokobori SI Oshima T Wada H : Complete nucleotide sequence of the mitochondrial genome of Doliolum nationalis with implications for evolution of urochordates. Mol Phylogenet Evol. 2005;34(2):273–283. 15619441 10.1016/j.ympev.2004.10.002 Yokobori S Ueda T Feldmaier-Fuchs G : Complete DNA sequence of the mitochondrial genome of the ascidian Halocynthia roretzi (Chordata, Urochordata). Genetics. 1999;153(4):1851–1862. 10581290 1460873 Yokobori SI Ueda T Watanabe K : Codons AGA and AGG are read as glycine in ascidian mitochondria. J Mol Evol. 1993;36(1):1–8. 8381878 10.1007/bf02407301 10.5256/f1000research.23747.r59671 Reviewer response for version 1 Li Yuanning 1 Referee https://orcid.org/0000-0002-2206-5804 Department of Ecology and Evolutionary Biology, Yale University, New Haven, CT, USA

Competing interests: No competing interests were disclosed.

 26 2 2020 Copyright: © 2020 Li Y 2020 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

Pichon et al. conducted a study by comparing two mt genes across tunicates and recovered conserved tunicate-specific mt codon usage and potential Oikopleuridae-codon. The writing of the manuscript is clear and easy to follow. The methods are also valid and the findings should be interesting and important to the field. However, there are a few things I suggest to incorporate in the current draft.

In the introduction, the authors mainly discussed the tunicate genetic codon, but we already know there are more codon changes in deuterostomes (e.g. Hemichordates contain two mt genetic codons). So it would be better to incorporate this part of the introduction to make sure readers understand there are many mt codon changes within deuterostomes.

The authors also briefly discussed the possible change of tRNA structures responsible for codon change. Is it possible to extract tRNA sequences from available tunicate transcriptomic data and compare them to the existing ones? I am also fine with this if that is beyond the scope of this manuscript.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Yes

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Evolutionary genomics and phylogenetics in marine invertebrates.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Plessy Charles Okinawa Institute of Science and Technology Graduate University, Japan

Competing interests: No competing interests were disclosed.

 27 3 2020

We thank the reviewer for their 2 suggestions.

1) “ To make sure readers understand there are many mt codon changes within deuterostomes”, we are adding the following text and references to our introduction:

In animals, alternative genetic codes have first been found in large clades, for instance echinoderms (Himeno et al., 1987) and hemichordates (Castresana et al., 1998), but more recent works underline the presence of changes deeper in the phylogenetic tree, for instance within nematodes (Jacob et al., 2009) and within hemichordates (Li et al., 2019).

2) On whether it is “ possible to extract tRNA sequences from available tunicate transcriptomic data”: as absence of evidence is not evidence for absence, our standpoint is that a rigorous analysis using a reference mitochondrial genome will be preferable. We hope that our methods section, that points at genomic scaffolds that are potential drafts of mitochondrial genomes, will be useful to the researchers interested in pursuing this direction.

 10.5256/f1000research.23747.r57710 Reviewer response for version 1 Lemaire Patrick 1 Referee https://orcid.org/0000-0003-4925-2009 Montpellier Cell Biology Research Center (CRBM), CNRS, University of Montpellier, Montpellier, France

Competing interests: No competing interests were disclosed.

 13 1 2020 Copyright: © 2020 Lemaire P 2020 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve

In this short article Pichon and colleagues use publicly available molecular datasets, mostly partial mitochondrial genomes, to re-explore the mitochondrial genetic code across tunicates, based on the analysis of the Cox1 and Cob genes.

Their data suggest that the AGR codon was already translated into Glycine in the last common ancestor of tunicates and that additional changes may have occured in some Oikopleuridae at least. The work is important because it shows that all tunicates, including appendicularians should be associated in Genbank to the "ascidian" genetic code (Table 13).

A limitation of the work, which the authors acknowledge, is that they could not identify the corresponding tRNAs in appendicularians. They therefore call for the sequencing of full-length mitochondrial genomes for appendicularians.

Is the work clearly and accurately presented and does it cite the current literature?

Yes

If applicable, is the statistical analysis and its interpretation appropriate?

Not applicable

Are all the source data underlying the results available to ensure full reproducibility?

Yes

Is the study design appropriate and is the work technically sound?

Yes

Are the conclusions drawn adequately supported by the results?

Yes

Are sufficient details of methods and analysis provided to allow replication by others?

Yes

Reviewer Expertise:

Tunicate embryology and genomics

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.