The first complete mitochondrial genome of the Northern Pacific deep-sea goniasterid sea star Ceramaster japonicus (Sladen, 1889) determined using NGS-based shotgun sequencing.

The full mitogenome of an ethanol-preserved museum specimen of Ceramaster japonicus was determined using the NGS Illumina MiSeq platform. The specimen was collected from Tosa Bay, Japan, facing the Pacific Ocean (33.0781 N 134.0601 E), at 700 m depth in 2011. The mitogenome shows a typical metazoan genomic structure, with all of the 37 genes included in its 16,370 base-long mitogenome. We conducted phylogenetic analyses using a data set including 18 publicly available asteroids rooted against five ophiuroids as outgroups. The result confirms the position of C. japonicus in the order Valvatida. The complete mitogenome of C. japonicus reported here is the first reported for the family Goniasteridae Forbes, 1841.

Content

The deep-sea asteroid Ceramaster japonicus is one of the most commonly encountered asteroids in the Japanese deep-sea benthic communities of the Pacific Ocean (Horikoshi et al. 1990). The species occurs widely in the Pacific and has been reported from the deep continental shelves along the northern Pacific rim, including the Alaskan coast, the Bering Sea, and off Washington (Fisher 1911; Setiamarga et al. 2019). Here, we report the full mitochondrial genome (mitogenome) sequence of Ceramaster japonicus. This is the first report of the complete mitochondrial genome of the family Goniasteridae (Asteroidea, Valvatacea, Valvatida). Goniasterids are the most diverse family of asteroids, including some 256 living species within 65 genera (Mah and Blake 2012). The complete mitochondrial genome of the ophidiasterid, Linckia laevigata, a species also representing the large Valvatacea cluster has recently been reported by Hiruta et al. (2020).

Tissue was collected from the tube feet of an ethanol-preserved specimen from the collection of the National Museum of Nature and Science, Japan (Voucher No. NSMT E-11866). The specimen was collected in 2011 from Tosa Bay, Kochi Prefecture, Japan, facing the Pacific Ocean (33.0781 N 134.0601 E), at the depth of 700 m. Total genomic DNA was extracted from the tissue sample using standard CTAB-Phenol/Chloroform protocol, purified using magnetic beads, and then fragmented and ligated to adaptors using the QiaSeq FX DNA Library Kit (QIAGEN). After fragment size and quality checks (TapeStation, Agilent Technologies), the DNA sample was run on a MiSeq Illumina Next Generation Sequencer. Obtained sequence fragments were edited, manipulated, and then assembled into a circular contig using CLC Genomics Workbench ver. 12 (QIAGEN) under default settings. Assembled mitogenome sequence was annotated using the MITOS web server (Bernt et al. 2013).

The newly sequenced mitogenome of Ceramaster japonicus was 16,376 bases-long, and was registered on Genbank (Accession number: LC592775). Its genomic structure is as follows: (1) There are 13 protein-coding, two rRNA, and 22 tRNA genes, (2) There is one control region, located between the tRNA-Thr and LSU-rRNA (575 bp-long); (3) Most protein genes are coded on the H chain, except for ND1, ND2, and ND6. Meanwhile, ND4L is located on the H chain; (4) Only the SSU-rRNA coding gene is located on the H chain; (5) 11 tRNA genes (tRNA-Asp, tRNA-Val, tRNA-Ala, tRNA-Gln, tRNA-Arg, tRNA-Lys, tRNA-His, tRNA-Ser (agc), tRNA-Phe, tRNA-Glu, and tRNA-Thr) are located on the H chain. The total GC content the mitogenome was 39.3%, which is common among asteroid species (Mu et al. 2018).

In order to check the molecular phylogenetic position of C. japonicus, we collected eighteen asteroids rooted against five ophiuroid outgroups sequences from GenBank, and included them in the data set for phylogenetic analyses. We conducted a maximum likelihood phylogenetic analysis (Yang 1994) using the GUI version of RAxML (Silvestro and Michalak 2012; Stamatakis 2014) and neighbor joining analysis using MEGA X (Kumar et al. 2018), of which methods are detailed in the legend of Figure 1.

Figure 1.

The maximum likelihood (ML) phylogenetic tree showing the position of Ceramaster japonicus with bootstrap supports (%) shown as ML/neighbor joining (NJ). Obtained NJ tree showed a congruent topology (not shown). Phylogenetic analyses were conducted on the data matrix (10,911 positions) including all concatenated nucleotide sequences of the mitogenomes except the third codon positions. Gene sequences were aligned individually using the online version of MAFFT under default settings (Katoh and Standley 2013). Aligned sequences were individually edited using the online version of GBlocks using the least stringent settings (Castresana 2000). Partitioned ML analyses (four partitions: 1st codon, 2nd codon, rRNA, tRNA) were performed with RAxML-GUI ver. 1-5b1 (Silvestro and Michalak 2012; Stamatakis 2014), with the GTR + Γ nucleotide substitution model (Yang 1994). The rapid bootstrap analyses were conducted with 1000 replications, with four threads running in parallel. NJ analyses were conducted using MEGA X (Kumar et al. 2018), under the Maximum Composite Likelihood substitution model, GTR nucleotide substitution rate, and 50% cutoff for each site with 1000 bootstrap replications.

Only the topology of the maximum likelihood tree, with the bootstrap supports of both the maximum likelihood and neighbor joining analyses, are shown on Figure 1. Our phylogenetic tree supports the monophyly of living Asteroidea, rooted against Ophiuroidea. Figure 1 agrees with monophyletic support for several historically recognized groups within the Asteroidea (Mah and Foltz 2011a, 2011b; Linchangco et al. 2017), such as the Forcipulatida, Brisingida, Paxillosida, and the Spinulosida. However, Valvatida was paraphyletic, with Linckia laevigata present as the sister clade to the Forcipulatacea (Forcipulatida + Brisingida), and thus away from other sampled valvatidans (Ceramaster, Archaster, Acanthaster, Patiria, and Aquilonastra). Many of the basal relationships, such as the Forcipulatacea + Paxillosida also disagree with those observed in prior accounts (e.g. Linchangco et al. 2017). Linckia laevigata’s position here differs from its position in the tree presented by Hiruta et al. (2020) which showed it as sister to two members of Paxillosida, Luidia and Astropecten. This is likely attributable to the different number of markers used in previous studies. Linchangco et al. 2017 included 2,683 nuclear loci, ca. 1 million bases, and Hiruta et al. 2020 included only the 13 protein coding genes of the mitogenomes versus our study which includes all 37 mitogenomic genes, ca. 10,911 positions. These are all in contrast to the earlier study of Mah and Foltz (2011a, b) which included the lowest number of nucleotides analyzed (three gene fragments, 1084 bases) but with the highest number of taxa (180 OTUs).

While the lack of molecular characters might adversely affect phylogenies (Rosenberg and Kumar 2001; Wiens et al. 2008), taxon sampling is a highly significant consideration when interpreting and comparing the topologies presented herein (Pollock et al. 2002; Hillis et al. 2003; Heath et al. 2008). Ceramaster’s addition to the tree is the first member of the Goniasteridae sampled for a complete mitogenome. The Goniasteridae is the most diverse group within the Asteroidea, and has demonstrated important relationships with other members of the Valvatida, such as the Asterodiscididae, the Oreasteridae, and the Ophidiasteridae (e.g. Blake and Portell 2011; Mah 2017).

Furthermore, at the timing of writing, neither our tree nor that of Hiruta et al. (2020) included any members of the Velatida, the third major clade within the Asteroidea (Linchangco et al. 2017). The lack of comprehensive taxon sampling in prior studies may have resulted in Long Branch Attraction artifacts which might have led to the erroneous grouping of some OTUs (Wiens et al. 2008). Ideally, future studies must include a more comprehensive taxonomic survey, in order to provide a clearer picture of asteroid phylogeny. We are confident, however, that the result presented here will be useful for future molecular studies addressing phylogenetic and other systematic questions in the Asteroidea, with particular reference to hyper-diverse groups, such as the Goniasteridae and the Valvatacea.