The specimens of flatworms were isolated from the stomach of their definitive host, in this case snakehead fish (Ophiocephalus argus (Cantor, 1842)) obtained from east Dongting Lake in Yueyang, Hunan province, China (29°22'N, 113°06'E). Azygia hwangtsiyui was morphologically identified according to the original and other descriptions (Tsin 1933; Tang and Tang 1964; Zhang et al. 1999; Besprozvannykh 2005), using a stereomicroscope and a light microscope. Furthermore, single samples were confirmed molecularly as A. hwangtsiyui based on sequencing of 1370 bp 28S rDNA sequence. The parasites were completely washed in water, preserved in 99% ethanol, and stored at 4 °C until genomic DNA extraction. Total genomic DNA extraction was performed from an intact specimen with the TIANamp Micro DNA Kit (Tiangen Biotech, Beijing, China), according to the manufacturer’s instructions.

According to conserved regions of mitochondrial genes in other available digenea mitogenomes, six partial gene fragments for cytb, nad4, nad1, 16S, 12S and cox2 were amplified using six generic primers sets HWF1/HWR1 (for cytb), HWF3/HWR3 (for nad4), HWF5/HWR5 (for nad1), HWF7/HWR7 (for 16S), HWF9/HWR9 (for 12S), and HWF11/HWR11 (for cox2), respectively. On the basis of these obtained nucleotide sequences, A. hwangtsiyui-specific primers were designed for amplification and sequencing of the remaining mitogenome (Suppl. material 1: Table S1). All primers were designed to produce amplicons with overlaps of approximately 100 bp. PCR reactions were performed in a 50 µl reaction solution with the ingredient of 18.5 µl ddH₂O, 25 µl 2×PCR buffer (Mg²⁺, dNTP plus, Takara, Dalian, China), 1.5 µl of each primer (0.2–1.0 μM), 1 µl EX Taq polymerase (250U, Takara), and 2.5 µl DNA template. PCR amplification was compliant to the following amplification protocol: initial denaturation at 98 °C for 2 min, followed by 40 cycles 10 s at 98 °C, 15 s at 50 °C, 68 °C for 1 min/kb, and 10 min at 68 °C for a final extension. The amplified PCR products were purified with TIANgel Purification Kit (Tiangen Biotech, Beijing, China), and sequenced bidirectionally at Sangon Biotech (Shanghai) Co., Ltd. (Shanghai, China) based on the primer walking method using several specific primers (Suppl. material 1: Table S1).

According to sequence chromatograms, all raw fragments were quality-proofed using CHROMAS (https://www.technelysium.com.au) to remove ambiguity codes and low-quality bases. Whenever the quality was sub-optimal, sequencing was repeated until the amplicon is the consensus sequence. Before manual assembly of the entire mitochondrial genomic sequence, identification of all amplicons was performed by BLASTN check (Altschul et al. 1990). The mt genome of A. hwangtsiyui was aligned against the mt genome sequences of other promulgated digenean mitogenomes utilizing multiple sequence alignment software MAFFT version 7.149 (Katoh and Standley 2013) to identify genetic boundary. Protein-coding genes (PCGs) were predicted with Open Reading Frame Finder (https://www.ncbi.nlm.nih.gov/orffinder/) adopting echinoderm and flatworm mitochondrial codes, and examining the nucleotide alignment against the reference mtDNA in trematode Dicrocoelium chinense Tang et Tang, 1978 (NC_025279.1). Whole tRNAs were inferred with the detection results of ARWEN (Laslett and Canback 2008) and MITOS web server (Bernt et al. 2013). Two rRNA (rrnL and rrnS) were founding by comparison with those of published fluke mitogenomes. Codon usage and relative synonymous codon usage (RSCU) for 12 PCGs of A. hwangtsiyui were computed by PHYLOSUITE (Zhang et al. 2020), and its operation results were imported into GGPLOT2 program (Wickham 2016) to make figures of the RSCU. Tandem repeats in the non-coding regions were determined with Tandem Repeats Finder software version 4.09 (Benson 1999), and the prediction of their secondary structures were performed by the MFOLD web server (Zuker 2003). The annular diagram of A. hwangtsiyui mitogenome was plotted with mitochondrial genome data visualization tool MTVIZ (http://pacosy.informatik.uni-leipzig.de/mtviz/mtviz).

For phylogenetic analyses, we utilized translated and concatenated amino acid sequences of twelve protein-coding genes for 49 Platyhelminthes including A. hwangtsiyui mitogenome determined in this study. Two tapeworm species, Cloacotaenia megalops (Nitzsch in Creplin, 1829) (NC_032295.1) and Dibothriocephalus latus (Linnaeus, 1758) (NC_008945.1) were included as outgroup taxa representing two different families. Species information including systematic positions and GenBank accession numbers is provided in Suppl. material 2: Table S2. The PHYLOSUITE program was used to extract twelve PCGs from the GenBank files, export fasta files with translated amino acid datasets, and align datasets in bulk using integrated applet MAFFT with normal-alignment mode. Phylogenetic analyses were performed using Bayesian Inference (BI) and Maximum Likelihood (ML) methods. Assessment of the best-fit evolutionary model for dataset was conducted via ModelGenerator v0.8527 (Keane et al. 2006). BI in MrBayes version 3.2.6 (Ronquist et al. 2012) was carried out under the MtRev matrix of amino acid substitution, and was analyzed with 1 × 10⁷ metropolis-coupled Monte Carlo Markov Chain (MCMC) generations. Two independent runs with four simultaneous MCMC chains (one cold and three heated chains) were conducted for 1 × 10⁷ million generations, sampling every 10,000 generations and discarding the initial 25% generations as burn-in. ML analysis in PHYLOSUITE was performed using MtART+I+G matrix with 1000 bootstrap replicates.

The entire A. hwangtsiyui mtDNA is 13,973 bp in length (GenBank accession number: MN844889) and comprised of 12 protein-coding genes (cox1-3, nad1-6, nad4L, cytb, and atp6), 22 tRNA genes, two rRNA genes (rrnL and rrnS), and two non-coding regions. The 12 protein-coding gene order arrangement is cox3-cytb-nad4L-nad4-atp6-nad2-nad1-nad3-cox1-cox2-nad6-nad5 (Fig. 1), which is identical to those of Clinostomum complanatum (Rudolphi, 1814), Echinostoma hortense Asada, 1926, and some species of the Fasciolidae (Fasciola hepatica Linneuus, 1758, Fasciola gigantica Cobbold, 1856, and Fasciola sp. GHL-2014) (Liu et al. 2014a; Chen et al. 2016; Liu et al. 2016); the gene atp8 is similarly missing, as usual in trematode species. All genes are transcribed in the anticlockwise direction and encoded by H strand (Table 1), which is in accordance with other digeneans. The mt genome of A. hwangtsiyui has 22 intergenic spacers ranging from 1 to 15 bp and contains two overlapping nucleotides ranging from 1 to 40 bp (Table 1). Noteworthily, a 40 bp overlap between the nad4 and nad4L genes exists in the A. hwangtsiyui mitogenome, which is consistent with most helminths such as Eurytrema pancreaticum Janson, 1889 (Chang et al. 2016), Hypoderaeum conoideum (Bloch, 1782) (Yang et al. 2015), but shorter than that of Schistosoma mekongi Voge, Bruckner & Bruce, 1978 (64 bp; Littlewood et al. 2006). The nucleotide contents of T, C, A, G, in A. hwangtsiyui mitogenome are 42.8%, 12.0%, 16.8%, and 28.5%, respectively (Table 2). The whole A+T content of the mitogenome is 59.6%, which was markedly biased toward T over A (AT skew: –0.437), and G over C (GC skew: 0.408).

Figure 1. 

An annular diagram of the Azygia hwangtsiyui mitochondrial genome.

A total of 3364 amino acids was encoded by the A. hwangtsiyui mtDNA. The full scale of 12 concatenated protein-coding genes was 10126 bp, composed of 45.2% T, 11.5% C, 14.7% A, and 28.6% G. Average A+T content of concatenated 12 protein-coding genes was 59.9%, varying from 57.7% (cox2) to 64.8% (nad4L) (Table 2 and Suppl. material 2: Table S2). All 12 protein-coding genes of A. hwangtsiyui mt genome have a lower A+T percentage than those of Trichobilharzia szidati Neuhaus, 1952, Calicophoron microbothrioides Price & McIntosh, 1944, and some members of the Schistosomatidae Poche, 1907, but possess a higher A+T percentage than those of Metagonimus yokogawai Katsurada, 1912, and Paragonimus westermani Kerbert, 1878 (Suppl. material 3: Table S3) (Lee et al. 1987; Littlewood et al. 2006; Biswal et al. 2014; Semyenova et al. 2017; Oey et al. 2019). There is an obvious bias towards T over A (AT skew = –0.509), and G over C (GC skew = 0.425), and the coding strand is enriched with T and poor with A and especially C. For A. hwangtsiyui, the length of protein-coding genes was followed in the order: nad5 (1600 bp) > cox1 (1554 bp) > nad4 (1272 bp) > cytb (1110 bp) > nad1 (906 bp) > nad2 (861 bp) > cox3 (660 bp) > cox2 (582 bp) > atp6 (513 bp) > nad6 (444 bp) > nad3 (360 bp) > nad4L (264 bp). There are two non-coding regions (NCR1 and NCR2) in A. hwangtsiyui mitogenome, while the mt genome of Paragonimus heterotremus Chen et Hsia (1964), C. complanatum, Fascioloides magna (Bassi, 1875), and T. szidati have a single non-coding region (Chen et al. 2016; Ma et al. 2016; Semyenova et al. 2017; Qian et al. 2018). NCR1 and NCR2 of the A. hwangtsiyui mitogenome is partitioned into two parts by trnG, and accompanied by 70.5% and 57.6% A+T content, respectively. NCR1 and NCR2 have similar chemical base counts, 315 bp and 317 bp in size, respectively. While the NCR1 lacks distinguishing features and any tandem repeats, the NCR2 contains two typical tandem repeats, and each of tandem repeats sequence (120 bp) forms a hairpin-like secondary structure including a whole set of stems and loops (Suppl. material 4: Figure S1). Although tandem repeats are a segment of function-deficiency mitochondrial genome sequences, its hairpin-like secondary structures are widely perceived as regulating the replication and transcription of mitochondrial genome.

For the A. hwangtsiyui mitogenome, codon ends in G or T were more continual than those ending in A or C. The most frequently used start codon in protein-coding genes was ATG (for six PCGs), secondly was GTG (for five PCGs), which resembles that of the most frequent extrapolated start codons for mitogenome protein-encoding genes of digenean species (Chen et al. 2016). The least-used start codon was TTG (only one PCGs). Likewise, the most-used terminal codon was TAG (for nine PCGs), followed by TAA (for two PCGs). Only one of 12 protein-coding sequence (nad5) was terminated with abbreviated T stop codon (Table 1). Although incomplete stop codons (T or TA) frequently occur in cestodes and nematodes, they were rarely presented in flukes other than D. chinensis, Dicrocoelium dendriticum (Rudolphi, 1819), and Postharmostomum commutatum (Dietz, 1858) (Liu et al. 2014b; Fu et al. 2019). The codon UUU (Phe, 10.17%), UUG (Leu, 8.17%), and GUU (Val, 7.19%) were the most frequently occurring codons in protein-coding genes. Leucine, valine, and phenylalanine are the most-used amino acids, with frequency of 15.96%, 13.38%, and 11.09%, respectively. The least-used codons were CGA (Arg, 0.06%) and GAC (Asp, 0.15%), and the least frequent utilized amino acid was glutamine (1.01%) (Suppl. material 5: Figure S2). As most of digenea mitogenome sequences, the mitogenome of A. hwangtsiyui possessed 22 commonly found tRNAs, with the exception of that of P. westermani Korean isolate (23 tRNAs), and P. westermani Indian isolate (24 tRNAs) (Biswal et al. 2014). In A. hwangtsiyui, tRNA-Gly (trnG) is located between NCR1 and NCR2 (Fig. 1). The size of ribosomal RNA genes (rrnL and rrnS) in mitochondrial DNA of A. hwangtsiyui are 975 bp and 740 bp, respectively (Table 2). The upstream and downstream of rrnL and rrnS are cascaded with trnT and cox2 genes, respectively, and are detached from each other by trnC, as in all reported platyhelminths to date (Littlewood et al. 2006, Le et al. 2016).

Comparative analysis of gene arrangement among 47 selected digenean taxa, two gene blocks (cox1-trnT-rrnL-trnC-rrnS-cox2-nad6 and cytb-nad4L-nad4-trnQ) are shared by all selected taxa (Fig. 2). Disregarding P. heterotremus and members of the family Schistosomatidae and Fasciolidae Raillet, 1895, the gene order of the remaining digenea taxa is virtually identical with the exception of the translocation of trnE and trnG among the remaining members of selected digenea representatives in family level. Intriguingly, there is the translocation of trnE and trnG within different species of family Fasciolidae. The translocations of three tRNAs (trnS1, trnS2 and trnS) can be discovered even between taxa of the same subgroup. Gene order of the Brachycladium goliath (Van Beneden, 1858) mt genome (the only representative of family Brachycladiidae Faust, 1929) is nearly same as that of P. westermani (Troglotrematidae Ward, 1918) except for the relocations of trnY between trnG and cox3, and trnE to the position between trnN and trnP. The groups of Schistosomatidae show a massive gene reorganization of protein-coding genes and tRNAs compared with other sequenced digenea mitogenome, which is in accord with previous finding reported by Littlewood et al. (2006).

Figure 2. 

Phylogenetic relationships and gene arrangement of Azygia hwangtsiyui with other selected digeneas based on translated mitochondrial proteins. The concatenated amino-acid sequence datasets of the 12 protein-coding genes were analyzed by Bayesian Inference (BI) and Maximum Likelihood (ML), utilizing Cloacotaenia megalops (NC_032295.1) and Dibothriocephalus latus (NC_008945.1) as the outgroups. Both ML and BI analyses constructed identical tree topologies.

To assess phylogenetic relationships among available flatworms, we utilized concatenated amino acid sequence dataset representing 12 protein-coding genes of A. hwangtsiyui, 46 other digenean representatives, and two tapeworm species (C. megalops and D. latus) for analyzing molecular-based phylogeny. In this study, the topological structure is divided into two large clades: one consists of seven members of the family Schistosomatidae; and the other clade comprises 40 members from 16 families including the family Azygiidae (A. hwangtsiyui) (Fig. 2). The topological structure shows that A. hwangtsiyui (Azygiidae) is identified as the most basal lineage of the Digenea, but separated from C. complanatum (Clinostomidae Lühe, 1901), and Cyathocotyle prussica Mühling, 1896 (Cyathocotylidae Poche, 1926). Phylogenetic analyses of all complete digenea mtDNAs confirmed taxonomic and previous phylogenetic assessments (Olson et al. 2003; Kostadinova and PéRez-del-Olmo 2014; Fu et al. 2019). The intricate structure and varying content of the family Azygiidae still awaits investigation of relationships based on a much wider taxon sampling and more mitogenome datasets.