Evolution of the Critically Endangered Green Sawfish Pristis zijsron (Rhinopristiformes, Pristidae), Inferred from the Whole Mitochondrial Genome

The green sawfish Pristis zijsron (Bleeker, 1851), a species of sawfish in the family Pristidae (Rhinopristiformes), mainly inhabits the Indo-West Pacific region. In this study, the complete mitochondrial genome of the critically endangered green sawfish is first described. The length of the genome is 16,804 bp, with a nucleotide composition of 32.0% A, 24.8% C, 13.1% G, and 30.0% T. It contains 37 genes in the typical gene order of fish. Two start (GTG and ATG) and two stop (TAG and TAA/T-) codons are found in the thirteen protein-coding genes. The 22 tRNA genes range from 67 bp (tRNA-Ser) to 75 bp (tRNA-Leu). The ratio of nonsynonymous substitution (Ka) and synonymous substitution (Ks) indicates that the family Pristidae are suffering a purifying selection. The reconstruction of Bayesian inference and the maximum likelihood phylogenetic tree show the same topological structure, and the family Pristidae is a monophyletic group with strong posterior probability. Pristis zijsron and P. pectinata form a sister group in the terminal clade. And the divergence time of Rhinopristiformes show that P. zijsron and P. pectinata diverged as two separate species in about Paleogene 31.53 Mya. Complete mitochondrial genomes of all five sawfishes have been published and phylogenetic relationships have been analyzed. The results of our study will provide base molecular information for subsequent research (e.g., distribution, conservation, phylogenetics, etc.) on this endangered group.


Introduction
Chondrichthyans (sharks, rays, and chimeras) are the most evolutionary distinct radiation of vertebrates, and the entire clade has been determined as one of the first major marine fish lineages facing extinction risk [1,2].The genomic architecture of chondrichthyans may be closer to the ancestral vertebrate condition compared with scleractinians, but there is limited available research [3,4].It is noteworthy that the life history traits (e.g., slow growth, low fecundity in long gestation boxes, etc.) of cartilaginous fishes make them more susceptible to overfishing problems compared with bony fishes [5].Rhinopristiformes is an order of rays that have shark-like features, which is presumably a combination of Rhinobatiformes and Pristiformes, two nominal orders in which taxa had previously been placed, including at least 71 species in 12 genera [6,7].This order is known by a few common names, such Genes 2023, 14, 2052 2 of 14 as guitarfish, giant guitarfish, sawfish, etc.The populations of many Rhinopristiformes species are declining in heavily fished areas, and are facing a high risk of extinction [8,9].Therefore, more attention should be paid to this group.
Although sawfishes are similar to sharks, they are actually rays.The green sawfish Pristis zijsron (Bleeker, 1851) mainly inhabits the shallow waters of coastal marine, mangrove, or estuarine habitats of the Indo-West Pacific region [10,11].And it is the largest sawfish species of family Pristidae, reaching an impressive length of up to 24 feet.In Australia, the distribution of green sawfish is from the east coast of the central Pacific Ocean, around the northern continent, to Shark Bay in the south-west to the eastern Indian Ocean, but the population in north-eastern Australia has now declined dramatically, and may even be extinct along much of the east coast of Australia [12][13][14].On the contrary, the species is still regularly found in western Australia.Due to habitat destruction and modification, populations of sawfishes have been impacted worldwide [15,16].The International Union for the Conservation of Nature and Natural Resources (IUCN) Red List of Threatened Species declared that the family Pristidae is the most threatened elasmobranch in the world because of their declining population numbers, and all international commercial trade of five sawfishes was prohibited by the Convention on International Trade in Endangered Species of Flora and Fauna (CITES) Appendix I [17,18].In addition, four Pristis (dwarf sawfish P. clavate, smalltooth sawfish P. pectinata, largetooth sawfish P. pristis and green sawfish P. zijsron) are currently assessed as "Critically Endangered" under criteria A2cd [14,[19][20][21], because of being highly vulnerable to entanglement in gillnets and demersal trawl nets.
Although it has never been quantified, it is believed that P. zijsron is experiencing drastic reduction in population size across its range and no longer appears to be common anywhere in its range.In the last 50 years, it has no longer appeared to be common anywhere within its range, and it is possibly extinct in some countries, for example, China, Thailand, Singapore, etc. [14].
In general, mitochondria are thought to have originated from Alphaproteobacteria, through the process of endosymbiosis existing in eukaryotic cells [22,23].The mitochondrial genome synthesizes proteins autonomously and participates in different physiological functions in organisms [24].Fish mitochondrial DNA is a circular, covalently closed doublestranded DNA molecule of 15-20 kilobases (kb) in length, and is considered to have an evolutionary origin independent of nuclear DNA, generally comprising 13 protein-coding genes (PCGs), 22 transfer RNA genes (tRNAs), 2 ribosomal RNA genes (rRNAs), an open reading frame (O L ), and a control region (CR) [25,26].Moreover, the mitochondrial genome is often used as a reliable marker to study the origin and phylogenetic relationship of species due to its fast evolution rate, maternal inheritance, and small molecular weight.Although accurately identifying sawfish species is a difficult task, the mitochondrial DNA may provide a powerful tool for scientists who are not expert taxonomists [27,28].Molecular data will greatly contribute to other related studies of sawfishes.Recently, Haque et al. used DNA barcoding (COI gene) to verify the recording of green sawfish P. zijsron in Bangladeshi waters [29].And Aschliman et al. used 12 H-strand PCGs to explore batoid evolution [4].
Here, we add the first whole mitochondrial genomic sequence for P. zijsron, and use it to predict the phylogenetic position of the sawfish species.Furthermore, we investigate the divergence time of Rhinopristiformes using a dataset (12 H-strand PCGs and rRNAs) in conjunction with fossil records.It is believed that the study will provide important molecular data for subsequent studies on the distribution, conservation, and evolution of sawfish species.

Sample Collection and Sequence Analysis
The sample (fin clip) of P. zijsron was collected from a specimen captured and released at Cape Keraudren, Western Australia, Australia, on 14 April 2008.The specimen was a male measuring 260 cm total length.The tissue sample was stored at CSIRO, Castray Esplanade, Hobart, Tasmania, Australia, with voucher no.1034 G.The genomic DNA was extracted from the fin clip by the standard phenol-chloroform procedure [30].LA-PCR reactions were performed using the Takara TM LA-Taq DNA polymerase kit.The PCR primers and PCR reaction conditions closely followed a previous study [31].
Sequence data were generated and analyzed to map the complete mitochondrial genome using the Lasergene soft package [32].The species was identified by morphology and fragments of COI and 16S rRNA gene downloaded from National Center of Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/genbank/,accessed on 1 September 2023).The mitochondrial genome was firstly annotated by MITOS2 Web Server [33].The 22 tRNA genes' secondary structures were determined by tRNAscan-SE Search Server v2.0 [34], then further confirmed with default search mode in AR-WEN v1.2 [35].Annotation and accurate boundary determination of PCGs and rRNAs genes were aligned and manually compared with the other released reference genomes of Pristidae species using DNAman v9.0 [36].Tandem Repeats Finder was used to find tandem repeats in the non-coding regions (NCR), then secondary structures were identified by Mfold v3.5.[37].
The P. zijsron mitochondrial genome was drawn into a full circular genome using CGView Server v1.0 (Figure 1) [38].The base compositions, codon usage, and relative synonymous codon usage (RSCU) values were calculated in MEGA X [39], then plotted using ggblot2 by Rstudio.Nucleotide diversity (Pi), non-synonymous substitutions (Ka), and synonymous substitution (Ks) mutation rate ratios of the PCGs of the Pristidae mitochondrial genome were calculated with Dnasp v6 [40].The strand asymmetry was assessed using the formulas provided as below: AT skew was calculated as (A − T)/(A + T), and GC skew was calculated as (G − C)/(G + C) [41].

Sample Collection and Sequence Analysis
The sample (fin clip) of P. zijsron was collected from a specimen captured and released at Cape Keraudren, Western Australia, Australia, on 14 April 2008.The specimen was a male measuring 260 cm total length.The tissue sample was stored at CSIRO, Castray Esplanade, Hobart, Tasmania, Australia, with voucher no.1034 G.The genomic DNA was extracted from the fin clip by the standard phenol-chloroform procedure [30].LA-PCR reactions were performed using the Takara TM LA-Taq DNA polymerase kit.The PCR primers and PCR reaction conditions closely followed a previous study [31].
Sequence data were generated and analyzed to map the complete mitochondrial genome using the Lasergene soft package [32].The species was identified by morphology and fragments of COI and 16S rRNA gene downloaded from National Center of Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/genbank/,accessed on 1 September 2023).The mitochondrial genome was firstly annotated by MITOS2 Web Server [33].The 22 tRNA genes secondary structures were determined by tRNAscan-SE Search Server v2.0 [34], then further confirmed with default search mode in ARWEN v1.2 [35].Annotation and accurate boundary determination of PCGs and rRNAs genes were aligned and manually compared with the other released reference genomes of Pristidae species using DNAman v9.0 [36].Tandem Repeats Finder was used to find tandem repeats in the non-coding regions (NCR), then secondary structures were identified by Mfold v3.5.[37].
The P. zijsron mitochondrial genome was drawn into a full circular genome using CGView Server v1.0 (Figure 1) [38].The base compositions, codon usage, and relative synonymous codon usage (RSCU) values were calculated in MEGA X [39], then plotted using ggblot2 by Rstudio.Nucleotide diversity (Pi), non-synonymous substitutions (Ka), and synonymous substitution (Ks) mutation rate ratios of the PCGs of the Pristidae mitochondrial genome were calculated with Dnasp v6 [40].The strand asymmetry was assessed using the formulas provided as below: AT skew was calculated as (A − T)/(A + T), and GC skew was calculated as (G − C)/(G + C) [41].

Phylogenetic Analysis
Phylogenetic analyses were performed using the new mitochondrial genome and other Rhinopristiformes species with complete mitochondrial genomes downloaded from GenBank (Table 1).Callorhinchus milii (NC_014285) and Tor putitora (NC_021755) were chosen as outgroups.The DNA sequences of 12 H-strand PCGs and rRNAs genes were firstly aligned using the MAFFT program, and were then re-aligned with MACSE v2 software [42,43].And the alignments of H-strand PCGs (excluding the start, stop, and the third codon) were divided into the first and the second codon positions.Gblocks was employed to eliminate ambiguously aligned segments from the rRNA alignments [44].The final dataset comprised the above segments after concatenation using Phylosuite v1.2.3 [45].The three segments of the dataset were pre-defined as partitions and verified by PartitionFinder v2.1.1 [59], and the ModelFinder program was used to select the best-fit partition model [60].The GTR+F+I+G4 model was selected as the best one and three partitions were defined: (1) two rRNA genes; (2) the first; and (3) the second codons of 12 PCGs.In addition, phylogenetic reconstruction was performed using the Bayesian analyses (BI) and Maximum Likelihood (ML) methods.ML analysis was inferred under 1,000,000 ultrafast bootstraps with an approximate Bayes test using IQ-TREE v1.6.2 [61].Based on the above partition model, Bayesian inference was conducted using MrBayes 3.2.6 software, under partition model (2 parallel runs, 1 × 10 6 generations), in which the initial 25% of sampled data were discarded as burn-in [62].Subsequently, the tree files were employed for visualization and annotation by the Figtree program.

Divergence Time Analysis
Divergence times were calculated within a Bayesian framework using the aforementioned dataset and BEAST version 2.5 [63].The uncorrelated lognormal relaxed clock, GTR substitution model, gamma site heterogeneity model, and the Yule process tree prior were selected to estimate differentiation time.Then, the divergence time analysis used three time calibration points derived from the TimeTree online [64]: 55.3 million years ago (Mya) for the common ancestor of P. clavata and P. pristis; 85.6 Mya for the common ancestor of the genera Rhina and Pristis; and 159.2 Mya for the common ancestor of Rhinopristiformes.
The chains of 1 × 10 8 samples were run for the Markov chain Monte Carlo (MCMC) analysis, and 10% of all samples had burn-in using TreeAnnotator software.Tracer v1.7.1 was used to analyze the results [65].Mean ages, along with 95% highest posterior density (HPD) intervals, posterior probabilities, and substitution rates, were computed for every node.The Figtree program was used to visualize the tree file.

Mitochondrial Genome Structure and Organization
The P. zijsron mitochondrial genome (GenBank accession No. MH005927) is 16,804 bp in length.The gene content of P. zijsron was similar to the typical fish mitochondrial genome, including 37 typical regions, 13 protein-coding (PCGs), 22 transfer RNA (tRNAs), 2 ribosomal RNA (rRNAs) genes, and 1 non-coding control region, with the typical gene order and transcriptional orientation of most vertebrate mitochondrial genomes (Figure and Table 2).All genes were encoded on the heavy-strand (H-strand), except the ND6 and eight tRNA genes (Gln, Ala, Asn, Cys, Thr, Ser, Glu and Pro) located on the light-strand (L-strand).The size and location of each gene were also similar to most fish mitochondrial genomes.There were 39 base pairs (bp) of short intergenic regions situated at 15 gene junctions, along with 24 bp overlaps identified at 6 gene junctions.Among all these intergenic regions, the shortest overlap was 10 bp in length (found between tRNA-Lys and ATP8), while the longest spacer was 6 bp in length (found between ATP8 and ATP6).In addition, the P. zijsron mitochondrial genome's O L was positioned between the tRNA-Asn and tRNA-Cys gene, spanning a length of 33 bp.Furthermore, three tRNA genes (Ile, Gln, Met) and five tRNA genes (Trp, Ala, Asn, Cys, Tyr) were observed to form clusters typically referred to as the "IQM" and "WANCY" regions.
The nucleotide composition was 32.0% adenine (A), 24.8% cytosine (C), 13.1% guanine (G), and 30.0%thymine (T), respectively.The base skew was also measured.The AT-skews of the entire mitochondrial genome of P. zijsron were positive (indicating an excess of A content over T), while the GC-skews were negative, indicating a higher G content than C content.Additionally, the AT-skews of 12 H-strand PCGs genes were more positive than GC-skews.The results indicate the A content was higher than the T content, and the C content was higher than the G content for the P. zijsron mitochondrial genome (Figure 2).And the AT content of the complete P. zijsron mitochondrial genome was 62.0%, similar to four other published Pristidae mitochondrial genomes (Table 1) [31,[47][48][49].
and five tRNA genes (Trp, Ala, Asn, Cys, Tyr) were observed to form clusters ty referred to as the "IQM" and "WANCY" regions.
The nucleotide composition was 32.0% adenine (A), 24.8% cytosine (C), 13.1 nine (G), and 30.0%thymine (T), respectively.The base skew was also measured.T skews of the entire mitochondrial genome of P. zijsron were positive (indicating an of A content over T), while the GC-skews were negative, indicating a higher G than C content.Additionally, the AT-skews of 12 H-strand PCGs genes were more p than GC-skews.The results indicate the A content was higher than the T content, C content was higher than the G content for the P. zijsron mitochondrial genome 2).And the AT content of the complete P. zijsron mitochondrial genome was 62.0%, to four other published Pristidae mitochondrial genomes (Table 1) [31,[47][48][49].

Protein Coding Genes and Codon Usage
The lengths of 13 PCGs range from 297 bp (ND4L gene) to 1839 bp (ND5 gene).The COI gene starts with a GTG codon, which is normal in fish mitochondrial genomes [66].The remaining protein genes start with the conventional ATG codon.Except for the ND6 gene terminating with the TAG codon, the other genes ended with TAA/T-codons.The 12 H-strand PCGs were T-rich, whereas the ND6 gene encoded on the L-strand was A-rich (Figure 2).
The most frequent amino acids in the PCGs of P. zijsron mitochondrial genome proteins were Leucine (Leu) and I-isoleucine (Ile), similar to other published Pristidae mitochondrial sequences (Figure 3a).Moreover, Cysteine (Cys) was the least frequently utilized amino acid in the mitochondrial sequence of P. zijsron.Simultaneously, the relative synonymous codon usage (RSCU) values of the P. zijsron mitochondrial genome were computed and are presented in Figure 3b.The findings revealed a bias toward A and T in the third codons.The length, RSCU, skewness, and AT content of PCGs in the P. zijsron mitochondrial genome were nearly identical, similar to other published Pristidae sequences.lized amino acid in the mitochondrial sequence of P. zijsron.Simultaneously, the relative synonymous codon usage (RSCU) values of the P. zijsron mitochondrial genome were computed and are presented in Figure 3b.The findings revealed a bias toward A and T in the third codons.The length, RSCU, skewness, and AT content of PCGs in the P. zijsron mitochondrial genome were nearly identical, similar to other published Pristidae sequences.Four species of sawfishes were grouped in pairs and their Ka/Ks were calculated, as shown in Figure 4.The Ka/Ks ratios of the 13 PCGs in 4 Pristidae mitochondrial genomes were all much less than 1.The highest average Ka/Ks ratio (0.11) was found in the ND6 gene, and the lowest average ratio (0.01) was found in the COI gene.Therefore, it can be considered that the genes had undergone purifying selection, and the evolution tended to be conservative.

Transfer RNAs, Ribosomal RNAs, and the D-Loop
Like the typical set of tRNA genes in fish mitochondrial genomes, 22 standard tRNA genes were found, and most of them could be folded into the typical cloverleaf structure, except for tRNA-Ser gene lacking the entire dihydrouridine arm.It had two kinds of tRNA-Leu and tRNA-Ser (Table 2).The lengths of 22 tRNAs were varied, ranging from 67 bp (tRNA-Ser) to 75 bp (tRNA-Leu), and the AT content was larger than the GC content (Figure 2a).
The 12S rRNA (963 bp) and 16S rRNA (1695 bp) genes were located between the tRNA-Phe and tRNA-Leu genes, separated by a tRNA-Val gene.The two rRNAs, both encoded by the H-strand, indicated moderate nucleotide compositional bias towards AT content (Figure 2).Moreover, the AT-skew of rRNAs was strongly positive, whereas the GC-skew was slightly negative, so a higher A and C content compared with T and G content in these regions was indicated.
The non-coding regions in the P. zijsron mitochondrial genome include the control regions (CR), the origin of L-strand replication (O L ), and several short intergenic spacers that range in size from 1 to 7 bp.The AT-rich CR, located between tRNA-Pro and tRNA-Phe with a length of 1110 bp, is the largest of these non-coding regions.The length of the CR is variable, and this is also the main reason for the differences in fish mitochondrial genome lengths.The CR of P. zijsron genomes showed positive AT-skew values.The origin of the L-strand replication sequence was identified between the tRNA-Asn and the tRNA-Cys genes.It was 33 bp in length and can be folded into a hairpin structure as a signal to initiate the replication of the L-strand.
Genes 2023, 14, x FOR PEER REVIEW 8 of Four species of sawfishes were grouped in pairs and their Ka/Ks were calculated, shown in Figure 4.The Ka/Ks ratios of the 13 PCGs in 4 Pristidae mitochondrial genom were all much less than 1.The highest average Ka/Ks ratio (0.11) was found in the N gene, and the lowest average ratio (0.01) was found in the COI gene.Therefore, it can considered that the genes had undergone purifying selection, and the evolution tended be conservative.

Transfer RNAs, Ribosomal RNAs, and the D-Loop
Like the typical set of tRNA genes in fish mitochondrial genomes, 22 standard tRN genes were found, and most of them could be folded into the typical cloverleaf structu except for tRNA-Ser gene lacking the entire dihydrouridine arm.It had two kinds of tRN Leu and tRNA-Ser (Table 2).The lengths of 22 tRNAs were varied, ranging from 67 (tRNA-Ser) to 75 bp (tRNA-Leu), and the AT content was larger than the GC content (Figu 2a).
The 12S rRNA (963 bp) and 16S rRNA (1695 bp) genes were located between tRNA-Phe and tRNA-Leu genes, separated by a tRNA-Val gene.The two rRNAs, both e coded by the H-strand, indicated moderate nucleotide compositional bias towards

Phylogenetic Inference
There is considerable attention being paid to the evolution of critically endangered sawfishes, and a robust phylogenetic analysis is essential for accurate research.For further investigation of the phylogenetic relationships of Rhinopristiformes, phylogenetic trees were constructed to compare with published complete mitochondrial genomes and two outgroups using the BI and ML methods.The two different approaches produced identical topologies with high nodal support values (Figure 5).Consistent with the tra-ditional taxonomy, the phylogenetic reconstruction in this study strongly supported the monophyly of five families (Glaucostegidae+ Pristidae+ Rhinidae+ Rhinobatidae+ Rhynchobatidae) of Rhinopristiformes.Family Pristidae is farther from Rhinobatidae than other Rhinopristiformes.

Phylogenetic Inference
There is considerable attention being paid to the evolution of critically endangered sawfishes, and a robust phylogenetic analysis is essential for accurate research.For further investigation of the phylogenetic relationships of Rhinopristiformes, phylogenetic trees were constructed to compare with published complete mitochondrial genomes and two outgroups using the BI and ML methods.The two different approaches produced identical topologies with high nodal support values (Figure 5).Consistent with the traditional taxonomy, the phylogenetic reconstruction in this study strongly supported the monophyly of five families (Glaucostegidae+ Pristidae+ Rhinidae+ Rhinobatidae+ Rhynchobatidae) of Rhinopristiformes.Family Pristidae is farther from Rhinobatidae than other Rhinopristiformes.The phylogeny of the family Pristidae is now fully resolved.All five species are clustered together with the four Pristis species (P.zijsron, P. pristis, P. clavate, and P. pectinata), forming a monophyletic group.Pristis zijsron and P. pectinata form a sister group in the terminal clade, then cluster with P. clavata and P. pristis.Anoxypristis cuspidata is placed in the basal position of the family Pristidae.Rhinopristiformes is a large taxa including at least 78 species [11], and some new species of this group have been discovered in recent years [67][68][69][70][71][72].Hence, the whole mitochondrial genomes of more species need to be studied to further investigate the phylogenetic relationships of this taxa.Additional The phylogeny of the family Pristidae is now fully resolved.All five species are clustered together with the four Pristis species (P.zijsron, P. pristis, P. clavate, and P. pectinata), forming a monophyletic group.Pristis zijsron and P. pectinata form a sister group in the terminal clade, then cluster with P. clavata and P. pristis.Anoxypristis cuspidata is placed in the basal position of the family Pristidae.Rhinopristiformes is a large taxa including at least 78 species [11], and some new species of this group have been discovered in recent years [67][68][69][70][71][72].Hence, the whole mitochondrial genomes of more species need to be studied to further investigate the phylogenetic relationships of this taxa.Additional mitochondrial genomes of Rhinopristiformes can be supporting evidence for morphological identification, especially for critically endangered species, and it will be useful to substantiate the conclusion in future studies.

Divergence Time Analysis
Divergence time typically denotes the estimated moment when two species or organism groups diverged from a shared ancestor, initiating independent evolutionary paths.Few studies have attempted to use mitochondrial genomic data to explore the evolution of sawfishes.To provide a timeline of P. zijsron evolution, the divergence time estimation of Rhinopristiformes and other related species was conducted using the Bayesian approach (Figure 6).The results show that Rhinopristiformes diverged from two outgroups (C.milii and T. putitora) in the Jurassic, about 159.18 Mya.Within the order Rhinopristiformes, the family Rhinobatidae firstly diverged from other species in the Cretaceous, about 110.66 Mya.The common ancestor of the family Pristidae occurred in 76.82 Mya.Pristis pectinata and P. zijsron diverged as two separate species at about 31.53Mya, as the most recently diverged species in TimeTree.Notably, the genus Pristis exhibited a divergence time dating back to 55.31 Mya, earlier than the fossil records of P. pectinatus (3.6-5.333Mya) and P. pristis (5.333-11.608Mya) [73,74].Similar to a previous Batoidea study, the major groups of Rhinopristiformes are estimated to have separated during the Cretaceous period.Subsequently, from the Late Cretaceous to the Cenozoic era, different subgroups emerged within each branch.about 110.66 Mya.The common ancestor of the family Pristidae occurred in 76.82 M Pristis pectinata and P. zijsron diverged as two separate species at about 31.53Mya, as most recently diverged species in TimeTree.Notably, the genus Pristis exhibited a div gence time dating back to 55.31 Mya, earlier than the fossil records of P. pectinatus (3 5.333 Mya) and P. pristis (5.333-11.608Mya) [73,74].Similar to a previous Batoidea stud the major groups of Rhinopristiformes are estimated to have separated during the Cre ceous period.Subsequently, from the Late Cretaceous to the Cenozoic era, different su groups emerged within each branch.However, it is important to note that the divergence time was estimated using te poral calibration points on TimeTree, due to the lack of sufficiently accurate fossil e dence.Estimates with calibration points may lead to biased results, so the divergence ti results can only be used as a preliminary reference in this study.In future, it will be n essary to add a lot of molecular data to Rhinopristiformes species and associated fos evidence to this group, which will be helpful in reconstructing a more detailed phylo netic tree for estimating the divergence times.More molecular data on Chondrichthy need to be added, and will be indispensable for subsequent morphological and evo tional studies of endangered species in Chondrichthyes.However, it is important to note that the divergence time was estimated using temporal calibration points on TimeTree, due to the lack of sufficiently accurate fossil evidence.Estimates with calibration points may lead to biased results, so the divergence time results can only be used as a preliminary reference in this study.In future, it will be necessary to add a lot of molecular data to Rhinopristiformes species and associated fossil evidence to this group, which will be helpful in reconstructing a more detailed phylogenetic tree for estimating the divergence times.More molecular data on Chondrichthyes need to be added, and will be indispensable for subsequent morphological and evolutional studies of endangered species in Chondrichthyes.

Conclusions
We successfully identified and compared the complete mitochondrial genome of P. zijsron for the first time.The length of the P. zijsron mitochondrial genome was 16,804 bp.The ratio of Ka and Ks indicated that Pristidae underwent a purifying selection, while the ND6 gene showed the highest Ka/Ks values (<1).Identical topologies with strong nodal support were produced by both ML and BI approaches, showing that the family Pristidae formed a monophyletic group.And the divergence time of Rhinopristiformes showed that P. zijsron and P. pectinate formed a sister group, diverging in the Paleogene, 31.53Mya.
Additionally, there are currently no conservation methods in place for the Pristidae family.The structure and variability of the fish mitochondrial genome are still unclear.A comprehensive comparative study using full mitochondrial genomic sequences will help enhance our comprehension of the evolution profile and phylogenetic placement in the family Pristidae.It is anticipated that this study will significantly contribute to the development of research related to the resources and conservation of sawfish species.

Figure 1 .
Figure 1.Circular representation of the mitochondrial genome of Pristis zijsron.Figure 1. Circular representation of the mitochondrial genome of Pristis zijsron.

Figure 1 .
Figure 1.Circular representation of the mitochondrial genome of Pristis zijsron.Figure 1. Circular representation of the mitochondrial genome of Pristis zijsron.

Figure 2 .
Figure 2. Visual representation showing the base composition (a) and skewness (b) in the m drial genome of Pristis zijsron.

Figure 2 .
Figure 2. Visual representation showing the base composition (a) and skewness (b) in the mitochondrial genome of Pristis zijsron.

Figure 4 .
Figure 4. Graphical illustration showing the sliding window analysis based on 13 PCGs of Pristid mitochondrial genomes (a).The red line represents the value of nucleotide diversity observed in sliding window analysis (window size = 100 bp, step size = 25 bp).The graph displays gene nam and corresponding Pi values, while the arrows indicate the length of the 13 PCGs.Additionally illustrates the non-synonymous (Ka) to synonymous (Ks) substitution rates for 13 PCGs within Pristidae family (b).

Figure 4 .
Figure 4. Graphical illustration showing the sliding window analysis based on 13 PCGs of Pristidae mitochondrial genomes (a).The red line represents the value of nucleotide diversity observed in the sliding window analysis (window size = 100 bp, step size = 25 bp).The graph displays gene names and corresponding Pi values, while the arrows indicate the length of the 13 PCGs.Additionally, it illustrates the non-synonymous (Ka) to synonymous (Ks) substitution rates for 13 PCGs within the Pristidae family (b).

Figure 5 .
Figure 5. Mitogenomic phylogeny of Rhinopristiformes and two outgroups based on 12 H-strand PCGs and rRNA genes using the ML and BI methods.

Figure 5 .
Figure 5. Mitogenomic phylogeny of Rhinopristiformes and two outgroups based on 12 H-strand PCGs and rRNA genes using the ML and BI methods.

Figure 6 .
Figure 6.Divergence time of Rhinopristiformes and two outgroups based on 12 H-strand PCGs a rRNA genes using BEAST2 software.

Figure 6 .
Figure 6.Divergence time of Rhinopristiformes and two outgroups based on 12 H-strand PCGs and rRNA genes using BEAST2 software.

Table 1 .
Rhinopristiformes species information used in this study.

Table 2 .
Summary of gene features of Pristis zijsron mitochondrial genome.