Mitochondrial Genome Diversity of Schistura McClelland, 1838 (Teleostei, Nemacheilidae)

Abstract

The inconsistency between traditional morphological taxonomy and molecular phylogenetic data is a major issue that puzzles the study of fish classification and evolution. Although mitochondrial genes are commonly used in phylogenetic analyses to compare fish species, the mitochondrial evolution and diversity of Schistura are still not well understood. To better understand the evolution of Schistura, we sequenced the mitochondrial genome of Schistura fasciolata and compared it with other species of Schistura. A 16,588 bp circular mitochondrial genome of S. fasciolata was obtained and it contains 13 protein-coding, 22 transfer RNA, and two ribosomal RNA genes, and a non-coding control region. The gene arrangement in the mitochondrial genomes of all Schistura species was consistent. However, we also found that S. fasciolata was not monophyletic. Although mitochondrial genes can be effectively used for Schistura species identification, they may not be suitable for inferring the evolutionary process of Schistura species. These results provide support for the use of mitochondrial genes in identifying Schistura species, and also serve as a warning against mistakenly using them to evaluate the evolution process of Schistura species.

1. Introduction

The inconsistency between traditional morphological taxonomy and molecular phylogenetic data is a major issue that puzzles the study of fish classification and evolution [1]. For instance, Nemacheilidae in the superfamily Cobitoidea exhibits inconsistencies between traditional morphological and molecular phylogenetic taxonomy [1]. Schistura McClelland, 1838, is the largest genus in Nemacheilidae [2,3,4], and is found throughout a large portion of South and Southeast Asia [5]. Because of their charming color stripe pattern and size, Schistura species are very popular in the ornamental fish trade [6]. However, there is still controversy surrounding the monophyly of Schistura [7,8,9]. Current evidence suggests that species of Schistura belong to at least two distant groups within the Nemacheilidae [10]. However, insufficient morphological and genetic data for most Schistura species have hindered the ability to accurately diagnose species groups as genera [9]. Furthermore, a large number of species and the wide distribution hinder the global revision of this genus [4]. Few researchers have tried to determine the natural population of Schistura species in order to conduct a manageable scale revision of the genus, analyze their taxonomy and biogeography, and contribute to their conservation [4].

Mitochondrial genes, such as cytochrome C oxidase I gene (COXI), cytochrome b gene (cyt b), and non-coding control region (D-loop), are often used to analyze the phylogenetic relationship among fishes [1,9,11,12]. For instance, Chen et al. [1] used recombinase-activating gene protein 1 (RAG1) and COX1 to study the phylogenetic relationships among Nemacheilidae species. They found that some species and genera were inconsistent with the traditional morphological classification [1]. The advancement of high-throughput sequencing technology has led to the use of complete mitochondrial sequences in analyzing fish phylogeny and population genetics [13,14,15,16,17,18]. Additionally, Muhala et al. [19] have shown that comparative analysis of mitochondrial genomes can provide insights into the subtle changes of gene ratio, intergenic spacer, and the phylogenetic relationships of mudskipper species. Therefore, we hypothesized that analyzing the mitochondrial genome of Schistura could aid in evaluating the feasibility of species identification and understanding the evolutionary process using mitochondrial genes.

Due to a sharp decline in population, Schistura fasciolata was listed in the IUCN Red List of Threatened Species in 2012 [20]. In this study, we have sequenced a mitochondrial genome of S. fasciolata and compared it with those of other Schistura species. Our findings provide a valuable reference for understanding the evolutionary process and classification of Schistura species.

2. Materials and Methods

2.1. Mitochondrial Data Retrieval

Thirty-one mitochondrial genomes of Schistura species were retrieved from the NCBI GenBank database (Table S1).

2.2. Mitochondrial Genome Sequencing of S. fasciolata

The study was approved by the Animal Ethics Committee of Hunan Agricultural University (Changsha, China; approval number: 20231015; approval date: 15 October 2023). The experiment was conducted in accordance with the Regulations on the Administration of Laboratory Animals of China. S. fasciolata were sampled from the Manxiba, Simao District, Pu’er City, Yunnan Province, China (N 22.7924, E 100.9835). The fin strips were preserved in 75% alcohol and sent to Wuhan Fisha Genetic Information Co., Ltd. (Wuhan, China). for sequencing and analysis. A specimen was deposited at Hunan Agricultural University Fisheries College (contact person Baohong Xu, [email protected]) under the voucher number Sf20230316123.

The fin strips of S. fasciolata were firstly soaked in sterile double distilled water for three 30 min intervals. The fish genomic DNA was then extracted following a previously described method [21]. Approximately 1 μg of DNA was used to construct the libraries using the MGI DNA Library Universal Kit (Vazyme Biotech Co., Ltd., Nanjing, China), and sequenced using a NovaSeq 6000 sequencer (Illumina, San Diego, CA, USA). The raw data were quality controlled using fastp v0.36 software [22]. The high-quality sequencing reads were assembled using SPAdes 3.0 [23] and SOAPdenovo2 1.0 [24] to generate preliminary assembly results. These results were then compared to the Nucleotide Database version 102 using the BLAST program to identify sequences annotated to the mitochondrial genome that form a ring structure. The results from both software were combined using CAP3 1.0 [25], and the assembly was further confirmed by comparing it to the mitochondrial genomes of closely related species using mummer software version 3.23 (https://github.com/mummer4/mummer, accessed on 30 October 2023) to obtain a high-quality mitochondrial genome.

2.3. Data Analysis

The gene structure of the mitochondrial genome was predicted using Glimmer v3.02 [26]. Transfer RNA (tRNA) was scanned using tRNAscan-SE 2.0 [27]. Ribosomal RNA (rRNA) was predicted using RNAmmer v1.2 [28]. ncRNA structure was scanned using Rfam v12.0 [29]. The mitochondrial genome was drawn using Proksee v1.0.0a6 [30]. Synteny of the mitochondrial genomes was analyzed using Mauve 20150226 [31]. Phylogenetic trees were reconstructed in Molecular Evolutionary Genetics Analysis 11 (MEGA11) [32] with 1000 bootstraps.

3. Results

The circular mitochondrial genome of S. fasciolata was 16,588 bp in length. It contains 13 protein-coding, 22 tRNA, and two rRNA genes, and a non-coding control region (D-loop) (Figure 1). Most protein-coding genes started from ATG, except for COXI which started from GTG. Notably, most protein-coding genes were inferred to be terminated with incomplete stop codons T or TA- (ND2, COXII, COXIII, ND3, ND4, and cyt b). Six protein-coding genes share the typical stop codon TAA (ND1, COXI, ATP8, ATP6, ND4L, and ND5), while ND6 uses TAG as a stop codon. The GC content was 42.54%. The length of the tRNA genes ranged from 62 to 76 bp, with a total length of 1557 bp. The lengths of the 12S and 16S rRNA genes were 947 and 1635 bp, respectively.

Genome synteny analysis showed that the genes in the mitochondrial genomes of all Schistura species were arranged in a synteny pattern (Figure S1). Except for the S. rupecula (AP011306.1) which lacked the D-loop region sequence, the mitochondrial genomes of other Schistura species were complete. The maximum-likelihood phylogenetic analyses based on different mitochondrial genes did not consistently show the same topological structure (Figure 2). Specifically, the ATP6, ND3, ND5, and ND6 genes, as well as the D-loop region, exhibited greater differences among species compared to other genes (Figure 2). Despite this, with the exception of S. fasciolata, the mitochondrial genes of the other Schistura species were clustered into single phylogenetic branches according to their respective species (Figure 2). Additionally, most of the mitochondrial genes showed no differences within the same species of Schistura (Figure 2). Interestingly, except for the ND6 gene, the phylogenetic trees based on the other mitochondrial protein-coding genes, rRNA genes, and D-loop region showed that S. fasciolata clustered with Schistura kaysonei, Schistura callichromus, and Schistura incerta, forming a consistent topological structure (Figure 2). These results suggested that S. fasciolata was not monophyletic. Furthermore, while mitochondrial genes can be used effectively for identifying Schistura species, they may not be suitable for inferring the evolutionary process of these species.

The phylogenetic trees based on the mitochondrial genomes of these fish, with and without the D-loop region, revealed that the inclusion of the D-loop region only affected the phylogenetic relationship between Schistura reticulata (KY379150.1) and Schistura sikmaiensis (KY379151.1 and NC 034746.1), but did not obviously impact the phylogenetic relationships among other Schistura species (Figure 3). Similar to the results of the single gene (except for the D-loop region), the phylogenetic tree based on the mitochondrial genomes also showed that S. fasciolata were clustered with S. kaysonei, S. callichromus, and S. incerta (Figure 3). These finding suggested that S. fasciolata was not a monophyletic species.

4. Discussion

Except for a few exceptions, the mitochondrial genomes of metazoan are usually composed of double-stranded circular DNA molecules, including 13 protein-coding genes, 22 tRNAs, two rRNA genes, and a D-loop region [33]. However, deviations from this standard architecture have been identified in teleosts, including replication, local position changes, and transposition [13,34]. For instance, the mitochondrial ND6 gene is found within the D-loop region instead of its usual position between the ND5 and cyt b genes in Morone, Dicentrarchus, and other species in Nototheniidae [35,36]. Our results also indicated that the complete mitogenome of S. fasciolata followed the typical metazoan architecture. Additionally, our genome synteny analysis revealed that the gene arrangement in the mitochondrial genomes of all Schistura species was consistent. A previous study shows the lengths of mitochondrial genomes of S. fasciolata and S. incerta are 16,560 and 16,561 bp, respectively [1]. However, our results showed that the length of mitochondrial genomes of S. fasciolata was 16,588 bp. The difference was mainly caused by the different length of D-loop region. However, the ecological mechanism of this difference and its influence on the fish’s adaptation to the environment are still unknown.

As described in previous studies on the Nemacheilinae [1], the ND6 gene was located on a separate mitochondrial DNA chain from other mitochondrial genes in S. fasciolata mitochondria (Figure 1). This could explain why the evolutionary tree constructed based on the ND6 gene differed from those based on other mitochondrial genes (Figure 2). These results suggested that the evolutionary pattern of the ND6 gene differed from those of other mitochondrial genes in Schistura species. Although mitochondrial genome rearrangements could be the basis of temperature adaptations of Antarctic notothenioid fishes [33], considering that ND6 subunit is essential for membrane arm assembly and the respiratory function of the mitochondrial NADH dehydrogenase [37], the effect of this unusual structural rearrangement in mitochondria on fish physiology is still unclear and needs further study.

Mitochondrial genes are commonly used for fish species identification and phylogenetic inference [38]. For instance, Cyt b is frequently used in phylogenetic studies of cobitoids to determine the relationships within genus [11,39]. While the D-loop region is not as commonly used, it is used to detect variation within species and identify species complexes in Lefua and other Cyprinidae fishes [40,41,42]. The D-loop region is often used to study population diversity due to its relatively high mutations [40,41,42,43]. Padhi et al. [44] reported the distribution of a 35 bp mitochondrial D-loop tandem repeat in the freshwater catfish Pylodictis olivaris populations in North America. They found that more than 70% of individuals sampled from the coast areas of the Southeastern Gulf have 35 bp mitochondrial D-loop tandem repeat, while more than 95% of individuals sampled from the Mississippi River and its tributaries, and the coastal areas of the Southwest Gulf, lacked such a tandem repeat. However, our results indicated that although mitochondrial genes can effectively be used for Schistura species identification, they may not be suitable for inferring the evolutionary processes of Schistura species.

5. Conclusions

The arrangement of genes in the mitochondrial genomes of all Schistura species was found to be synteny. However, it was observed that S. fasciolata was not monophyletic. Mitochondrial genes can be effectively used for delimiting Schistura species. However, they were not suitable for inferring the evolutionary processes of these species.