Ethics Statement

No specific permits were required for the insect collected for this study in Sichuan. The specimens of T. diuturnus were collected by using clap net at night. The field studies did not involve endangered or protected species. T. diuturnus was not included in the “List of Protected Animals in China”.

Sample Origin and DNA Extraction

Specimens of T. diuturnus adult were collected from Liziping Nature Reserve (102.60°E, 29.09°N) in Sichuan Province, China, in July 2011. After collected, they were initially preserved in 100% ethanol in the field, and transferred to −20°C upon the arrival at the Hebei University.

Total DNA was extracted from the leg muscle tissue of a single adult male specimen using the TIANamp.

Genomic DNA Kit according to the manufacturer’s instructions (Tiangen Biotech, Beijing).

Primer Design, PCR Amplification and Sequencing

First, three short fragments of cox1, cox2 and cytb genes were determined respectively using the universal primer sets: LCO1490/HCO2198 for cox1 [9], C2-J3399/TK-N3796 for cox2 and CB-J10933/CB-N11367 for cytb [10], [11]. Based on this sequence information, two modified L-PCR primer sets [12] were used to amplify the entire T. diuturnus mitogenome into two overlapping larger fragments under the following conditions: a primary step denaturation at 94°C for 1 min, followed by 30 cycles of denaturation at 98°C for 10 s, annealing at 48°C for 30 s, and 68°C elongation for 10 min, followed by incubation at 68°C for 10 min. PCR reactions were performed in a MyCylerTM Personal Thermal Cycler (BIO-RAD) in 25 µL reaction volume set up as follows: 12.5 µL Premix LA Taq Hot Start (TaKaRa Biotech, Dalian), 1 µL of each primer (10 µM), 3 µL of genomic DNA, 7.5 µL of sterilized distilled water.

Sub-PCR amplifications were performed with L-PCR products as templates. In primer sets of Sub-PCR, one was designed based on the obtained sequence, and another was designed according to aligned mitogenome sequences of orthopterans in the NCBI nucleotide database and universal primers [10], [11], [13]. These primer pairs were then used within Sub-PCR amplifications to obtain adjacent gene sequences, which overlap with known sequences. In the same way, the cycle was repeated to ‘walk’ around the remainder of the genome. The Sub-PCR fragments were isolated and purified, and then ligated into PMD®-19 T vector (TaKaRa Biotech, Dalian) and sequenced from both strands on ABI 3730 XL DNA Analyzer (Biosune Biotech, Shanghai). All primer used for this study are included in Table S1.

Genome Annotation, and Sequence Analysis

To prepare the mitogenome sequences for uploading into MOSAS, we first proofread and assembled raw sequence data into contigs using the Staden sequence analysis package [14], and then we uploaded entire mitogenome sequence files into MOSAS for annotation [15]. The nucleotide composition of different regions of the mitogenome and the codon usage of protein-coding genes were analyzed with MEGA 5.0 [16].

Phylogenetic Analyses

A total of 17 taxa were analyzed in this study, including 15 ensiferan ingroup and two caeliferan outgroup taxa (Table S2). For this study, we sequenced 9 ensiferan species, Ruspolia dubia [17], Gampsocleis gratiosa [18], Deracantha onos [19], Elimaea cheni [20], Conocephalus maculates [21], Xizicus fascipes [22], Mecopoda elongate [23], Mecopoda niponensis [23] and T. diuturnus. Six additional ensiferan species were obtained from previous studies: Anabrus simplex [24], Gryllotalpa orientalis [25], Gryllotalpa pluvialis [26], Myrmecophilus manni [26], Troglophilus neglectus [26] and Sinochlora longifissa [27]. For outgroups we sampled two caeliferan species, Locusta migratoria (Caelifera: Acridomorpha) [28] and Atractomorpha sinensis (Caelifera: Pyrgomorphidae) [29].

DNA alignment of 13 protein-coding genes (PCGs) and two ribosomal (rRNA) genes was conducted using BioEdit [30] in which PCGs was inferred from the amino acid alignment. Alignments of individual genes were then concatenated excluding the stop codons. Maximum Likelihood (ML) using PhyML 3.0 [31], [32] and Bayesian inference analysis (BI) using MrBayes 3.1.2 [33] were used for phylogenetic reconstruction. In the ML method, a BioNJ tree was used as a starting tree to search for the ML tree with the GTR+I+Γ model. Robustness of the phylogenetic results was tested by bootstrap analysis with 1000 replicates. In BI analyses, ten million generations were run, with four MC chains, and trees were sampled every 1000 generations with a burnin of 25%. Trees inferred prior to stationarity were discarded as burnin, and the remaining trees were used to construct a 50% majority rule consensus tree.

Genome Organization

The T. diuturnus mitogenome is a typical circular DNA molecule of 16144 bp in length (GenBank accession number JQ999995; Figure 1). This length is medium-sized when compared with other orthopteran species, which typically range from 14,971 bp of R. dubia [17] to 18,133 bp of S. longifissa [27]. It has a typical gene content found in metazoan mitogenomes: 13 PCGs, 22 Transfer RNA (tRNA) genes, two rRNA genes, and one control region. The gene order and orientation of T. diuturnus mitogenome is identical to the hypothesized ancestral arthropod arrangement, found in D. yakuba and most ensiferans species [34]. Although the nucleotide composition of the T. diuturnus mitogenome is strongly biased towards A+T (67.21%), it is less so than most orthopterans (65.3%–76.2%, mean 73.2%) [18], [26].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Mitochondrial genome of Tarragoilus diuturnus.

https://doi.org/10.1371/journal.pone.0086027.g001

Without considering the control region, a total of 65 bp of noncoding nucleotides were scattered among 9 intergenic regions varying from 1 to 25 bp. The largest intergenic region (25 bp) was present between trnR and trnN. Overlapping genes were also found in the T. diuturnus mitogenome, in which 52 bp were identified as overlapping sequences ranging from 1 to 8 bp in 14 regions. As has often been found across hexapod, there are several gene overlaps, such as the open reading frames (ORF) of atp6/atp8 and nad4L/nad4, and each of which is overlapped by seven nucleotides. They are thought to be translated as a bicstron [35].

Protein-coding Genes

The total length of all 13 PCGs was 11,250 bp, and the overall A+T content of T. diuturnus PCGs was 66.03%. The initiation and termination codons were identified using the ORFs finder and by comparison with the situation in other orthopterans. The initiation and termination codon assignments attempt to minimize gaps between gene regions while also minimizing gene overlap. Five of the 7 potential initiation codons (ATN and NTG) [36] were utilized in T. diuturnus mitochondrial protein-coding genes: ATT (nad6), ATA (cox2, nad1, nad2, and nad3), ATC (cox1), ATG (atp6, cox3, cytb, nad4, nad4L, nad5) and GTG (atp8) (Table 1). We allowed gene overlap between adjacent genes but not between protein-coding genes and tRNAs. When a full termination codon (TAA or TAG) caused an overlap between a protein-coding gene and a tRNA, we annotated this gene with incomplete termination codon T/TA instead (Table 1). We annotated full termination codons, however, in overlapping protein-coding genes. Six of the 13 protein-coding genes have incomplete termination codons T (cox1, cox2, cytb, nad2, nad4, and nad5), and six with termination codons TAA (atp6, cox3, nad1, nad3, nad4L, and nad6), one with TAG (atp8). The presence of incomplete termination codon is common in metazoan mitochondrial genomes, and polyadenylation following transcription likely converts these partial stop codons into full stop codons [37].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Positions and nucleotide sequence lengths of mitochondrial genome of Tarragoilus diuturnus, and initiation and termination codons for protein-coding genes as well as tRNA gene anticodons (starting from trnI).

https://doi.org/10.1371/journal.pone.0086027.t001

All codons are present in the protein-coding genes of this mitogenome. Excluding termination codons, there are a total of 3741 codons in the T. diuturnus protein-coding genes. The relative synonymous codon frequencies of T. diuturnus protein-coding genes are summarized in Table S3. The codon usage in T. diuturnus appears to be typical of other insect mitochondrial sequences. The synonymous codon usage shows a distinct bias towards T in the second and T/A in the third codon positions, regardless of the identity of the anti-codon encoded by the tRNA. In order, the four most used amino acids in T. diuturnus are Leu (15.64%), Ser (8.66%), Phe (8.55%), and Ile (8.29%).

Transfer RNAs

The complete set of 22 tRNA genes is present in the T. diuturnus mitogenome, and their length ranges from 63 to 70 bp. The relative locations for each tRNA gene are shown in Figure 1. All tRNA genes have the typical cloverleaf structures except for trnS^AGN. The trnS^AGN was found to have a lengthened anticodon stem (nine base-pairs) with a bulged nuleotide in the middle, an unusual T-stem (six base-pairs in constrast to the normal five base-pairs), a mini DHU arm (one base-pair) and no connector nucleotides (Figure 2) [38].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Putative secondary structures for 22Tarragoilus diuturnus mitochondrial genome.

The tRNAs are labeled with abbreviations of their corresponding amino acids. Dashes (−) indicate Watson-Crick base-pairing, centered asterisks (*) indicate G-U base-pairing, and periods (•) indicate mismatches.

https://doi.org/10.1371/journal.pone.0086027.g002

Thirty non-Watson-Crick pairings (G-U) and fifteen mismatches were identified in the T. diuturnus mitogenome tRNA genes. Six U/U were proposed in the acceptor stem of trnH, trnS^UCN, trnA, trnE, the DHU stem of trnP, and the TψC stem of trnS^AGN; three U/C were proposed in the anticodon stem of trnL^CUN, trnK and trnM; four A–C in the anticodon stem of trnT, trnR, trnSAGN, and the acceptor stem of trnLUUR; one A-A was proposed in the acceptor stem of trnS^UCN and one G-A in the anticodon stem of trnW (Figure 2).

Ribosomal RNAs and Control Region

As in most insect mitogenomes, two rRNA genes (rrnL and rrnS) were present in the T. diuturnus mitogenome between trnL^CUN and the control region, and separated by trnV. The lengths of rrnL and rrnS were respectively determined to be 1340 and 783 bp. The A+T content of rrnL and rrnS were 71.79% and 66.67%, respectively.

The 1301 bp T. diuturnus control region was observed in the conserved location between rrnS and trnI, and its composition was 66.18% A+T. The size of the T. diuturnus control region is well within the range found in other orthopterans, from 70 bp in R. dubia [17] to 2277 bp in P. variolosa [15]. The size differences of the control region occur not only because of its high rate of nucleotide substitution, insertion or deletion, but also due to the length of tandem repeat unit and the number of tandem repetitions.

Phylogenetic Relationships

The dataset (13,416 sites) used in the present analyses contained all three codon positions of 13 PCGs and two rRNA genes. The two independent runs yielded identical 50% consensus topologies, with harmonic means –ln = 142680.07 and 142679.11, respectively, and standard deviation of split frequencies <0.01 (Figure 3A). Hagloidea was placed as the sister group of Rhaphidophoroidea, and Tettigonioidea was a sister to Hagloidea+Rhaphidophoroidea. Both Tettigonioidea and Grylloidea were recovered as a monophyletic superfamily. Within Tettigonioidea, six subfamilies were divided into two large clades Ta (including Tettigoniinae, Bradyporinae, Conocephalinae and Meconematinae) and Tb (including Phaneropterinae and Mecopodinae). The best ML tree (Figure 3B) was found to have a score of Log-likelihood −146260.61551, and its topology was nearly identical to that recovered in the Bayesian analysis. The only exception was that the position of E. cheni (Phaneropterinae) shifted from clade Ta to Tb, and nested within Conocephalinae.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Phylogenetic reconstruction of the Ensifera using mitochondrial PCGs and rRNAs concatenated dataset.

(A) Bayesian result, applicable posterior probability values are shown; (B) Maximum likelihood result with applicable bootstrap values (>50%) are shown.

https://doi.org/10.1371/journal.pone.0086027.g003

The Ensifera have usually been viewed as monophyletic group. Several attempts have been made to reconstruct the phylogeny of the Ensifera. Cranston and Gullan proposed Grylloidea was the sister group (but highly divergent, with a longer branch separation) of the remaining ensiferan taxa, including Tettigonioidea, Hagloidea, and Stenopelmatoidea [39]. Ander proposed dividing Ensifera into two major groups, which he called Grylloidea (including Gryllidaeand and Gryllotalpidae) and Tettigonioidea ((Rhaphidophoridae, (Schizodactylidae, (Gryllacrididae, (Stenopelmatidae, Prophalangopsidae, Tettigoniidae))))) [40].

Previous studies using morphological and molecular evidence have resulted in various hypotheses for the taxonomic status of the Hagloidea. Tettigoniidae and Hagloidae form a monophyletic group, sister to Stenopelmatidae and relatives (mormon crickets, wetas, cooloola monsters, and the like), but alternative analyses suggest different relationships, and conservatively, an unresolved group is perhaps appropriate at this stage [39]. Gorochov considered Prophalangopsidae (the sole extant family of Hagloidea) to be ancestral to the Tettigonioidae and Stenopelmatoidea [7]. Gwynne considered the Hagloidea to be a sister group of Tettigonioidae [5]. Jost and shaw considered the haglid Cyphoderris to be basal (or sister) to a clade of Grylloidea+Tettigoniidae (the families, Haglidae and Tettigoniidae used by authors) [1]. Flook et al. found the relationships among Tettigonioidae, Hagloidea and Stenopelmatoidea (including Stenopelmatidea, Schizodactylidea and Rhaphidophoridea, which were elevated to superfamily status in Orthoptera Species File Online) to be unresolved because different analyses gave conflicting results [8]. Mugleston et al. found the sister superfamily of Tettigonioidae still unresolved, it was Rhaphidophoroidae in the ML and BI analyses, whereas was Hagloidea in the MP analysis (the families, Tettigoniidae, Rhaphidophoridea and Prophalangopsidae used by authors) [41]. Regardless of the method of tree reconstruction used, the new mitochondrial data presented here strongly supports the Hagloidea T. diuturnus and Rhaphidophoroidea T. neglectus as forming a monophyletic group, sister to the Tettigonioidea. The relationships among the four superfamilies were (Grylloidea, (Tettigonioidea, (Hagloidea, Rhaphidophoroidea))).