A specimen of P. profundacella (Fig. 1) was collected from Yangmeikeng of Daya Bay in Guangdong Province and it was kept in our Coral Sample Repository with a special code, 20190718-D17. Identification was conducted according to the description of Benzoni et al. (2010): calices are between 1.4 and 1.7 mm in diameter, the columella is made of one central process surrounded by 4–6 granules positioned at the inner end of the septa and up to 6 septa reaching the fossa are petaloid. Complete genomic DNA was extracted using the DNeasy Blood and Tissue Kit (Qiagen, Shanghai, China). Electrophoresis with 1% agarose gel was used to measure the integrity of the genomic DNA and the spectrophotometer NanoDrop 2000 (Thermo Scientific, USA) was used to measure the genomic DNA concentration.

Figure 1.  

Photos of P. profundacella in this study. (A) In-situ photograph of P. profundacella; (B) Skeleton photograph of P. profundacella; (C) Microskeletal photograph of P. profundacella.

After DNA extraction and quality detection, the sequencing library was produced using the Illumina Truseq™ DNA Sample Preparation Kit (Illumina, San Diego, USA) according to the manufacturer's recommendations. Five µg of double-stranded DNA was sheared to ~ 300 bp using a M200 Focused-ultrasonicator (Covaris, Woburn, MA, USA). The prepared library was loaded on the Illumina HiSeq X Ten platform for PE 2 × 150 bp sequencing. The quality and quantity of data were assessed by FastQC (Andrews 2010). Filtered clean reads were obtained by removing reads containing poly-N regions, adapters and by eliminating low-quality reads using fastp (Chen et al. 2018). The median of insert sizes and average read length were used to reconstruct the mitochondrial genome via NOVOPlasty 3.8.3 (Nicolas et al. 2016). A total of 60,910 of 128,526,740 raw reads (approximately 0.05%) were de novo assembled to produce the mitogenome with the guidance of seed sequence and the average coverage was 565×.

The circularised contig was submitted to MITOS (Bernt et al. 2013) WebServer (http://mitos.bioinf.uni-leipzig.de/index.py) for preliminary mitochondrial genome annotation. We then identified and annotated all 13 PCGs and two rRNA genes by alignments of homologous mitogenomes with other scleractinians that had been reported through BLAST searches in NCBI. All the PCG codon usage and nucleotide frequencies were obtained through Molecular Evolutionary Genetics Analysis software MEGA7 (Kumar et al. 2016). tRNA genes were identified by comparing the results predicted by tRNAscan-SE 2.0, based on its unique cloverleaf secondary structure information and we also validated the result using ARWEN (Laslett and Canbäck 2008, Lowe and Chan 2016).

The phylogenetic position of P. profundacella was inferred using 13 tandem mitogenome PCG sequences (ND5 + ND1 + Cytb + ND2 + ND6 + ATP6 + ND4 + COIII + COII + ND4L + ND3 + ATP8 + COI) together with 81 other Scleractinia species that we obtained from GenBank (see Suppl. material 1). We used MEGA7 to choose the best-fitting model, based on the Akaike Information Criterion (AIC) and then constructed a Maximum Likelihood (ML) tree with 500 bootstrap replicates.

The mitochondrial genome size of P. profundacella (GenBank accession number: MT576637) was 16,274 bp, including 13 PCGs, 2 tRNA (tRNA^Met, tRNA^Trp) and 2 rRNA genes (see Table 1,Fig. 2). The mitogenome of P. profundacella offered no distinct structure and its gene order, gene identity and gene number were the same as those of published stony coral mitogenomes (Wang et al. 2013). The base composition of complete mitogenome was 26.34% A, 11.32% C, 19.17 G and 43.17% T, which showed a higher AT content (69.51%) than GC content (30.49%) (see Fig. 3, Table 2). In addition, all genes remained encoded on the H-strand.

Figure 2.  

The mitochondrial genome of P. profundacella. Gene order and positions are shown. COI, COII and COIII refer to the cytochrome oxidase subunits, Cyt b refers to cytochrome b and ND1-ND6 refers to NADH dehydrogenase components. All the genes are encoded on the H-strand.

Figure 3.  

Codon usage bias in different regions of mitochondrial genome of P. profundacella.

The length of all 13 protein-coding genes sequence was 11,511 bp, with base composition of 23.6%, 11.2%, 18.4% and 46.8% for A, C, G and T, respectively. ND5 gene had an intron insertion of 9,549 bp. The start codon of all PCGs used ATG, except for Cyt b using TTA and ND2 using ATT. Three PCGs (COII, ND4 and ND5) terminated with TAG, the other ten PCGs (ATP6, Cyt b, ND1, ND2, ND3, ND4L, ND6, ATP8, COII and COIII) stopping with TAA. The shortest gene was ATP8 (195 bp) and the longest gene was ND5 (1,812 bp). The intergenic region between cytb and ND2 was 208 bp (see Table 1). According to the results of AT-skew and GC-skew analysis (Fig. 4), all PCGs showed a stronger nucleotide asymmetry, with AT-skew higher than GC-skew. Amongst L, F, V, G and S, codon use frequency was higher, accounting for 52.5% of a total of 3837 codons. Amongst the 20 amino acids, the majority were were non-polar amino acids which accounted for 68.0%; the minority were polarity-charged amino acids accounting for 11.1% and the remainder were polar amino acids which accounted for 20.5% (Fig. 5).

Figure 4.  

The PCGs’ AT-skew and GC-skew of mitochondrial genome of P. profundacella.

Figure 5.  

The PCGs’ codons use frequency of mitochondrial genome of P. profundacella.

The encoding genes 12S and 16S rRNA in P. profundacella were 906 bp and 1,704 bp in size, respectively. Both the two rRNAs’ base composition was 37.32% A, 10.15% C, 18.54% G and 33.98% T. There were also two tRNA encoding genes- tRNA^Met (72 bp) and tRNA^Trp (70 bp). They were folded into the classic cloverleaf structure which included an amino acid accept arm, DHU loop, anticodon loop and TψC loop (Fig. 6).

Figure 6.  

Putative secondary structures of two tRNA of P. profundacella.

The ML tree topology of the 82 stony corals species showed that P. profundacella belongs to the “Robust” clade and is closely related to Polycyathus chaishanensis (Caryophylliidae) with high bootstrap support (Fig. 7). These findings are consistent with the results of Lin et al. (2012), which clearly showed that Polycyathus chaishanensis is closely related to a clade comprising Psammocora, Coscinaraea, Leptastrea and Fungiidae. The mitochondrial genome data have provided important molecular information for understanding evolutionary relationships amongst stony corals, but more variable markers and additional species should be sequenced to confirm the evolutionary relationships of Scleractinia in the future.

Figure 7.  

Inferred phylogenetic relationships, based on the concatenated nucleotide sequences of 13 mitochondrial PCGs, using Maximum Likelihood (ML) analysis. Numbers on branches are bootstrap percentages.