Characterization of the Complete Mitochondrial Genome of Drabescus ineffectus and Roxasellana stellata (Hemiptera: Cicadellidae: Deltocephalinae: Drabescini) and Their Phylogenetic Implications

Simple Summary

Drabescini comprises over 225 species and 46 genera with a highly diverse tribe of Deltocephalinae. These species serve as vectors of numerous agricultural plant pathogens and also transmit plant viruses to host plants. Previous phylogenetic analyses in this tribe mainly focused on morphological characters and were restricted to several gene fragments. Furthermore, the taxonomic status and phylogenetic relationships of this tribe need to be further studied. Therefore, the mitogenome may provide additional molecular evidence to reconstruct the phylogeny of this group and further elucidate relationships among major lineages. In this study, we sequenced and analyzed two newly complete mitogenomes including Drabescus ineffectus and Roxasellana stellata. These two mitogenomes contain 13 protein-coding genes, two ribosomal RNA genes, 22 transfer RNA genes and the non-coding structure called A + T-control region. The Drabescini mitogenomes are highly conserved in base content and composition, genome size and order, protein-coding genes and codon usage, and secondary structure of tRNAs. Phylogenetic analyses using Bayesian inference and maximum likelihood methods indicated strong support for the monophyly of Drabescini. These results provide the comprehensive framework and valuable data toward the future resolution of phylogenetic relationships in this tribe.

Abstract

To explore the mitogenome characteristics and shed light on the phylogenetic relationships and molecular evolution of Drabescini species, we sequenced and analyzed the complete mitochondrial genome of two species including Drabescus ineffectus and Roxasellana stellata. The complete mitogenomes of D. ineffectus and R. stellata are circular, closed and double-stranded molecules with a total length of 15744 bp and 15361 bp, respectively. These two newly sequenced mitogenomes contain the typical 37 genes. Most protein-coding genes (PCGs) began with the start codon ATN and terminated with the terminal codon TAA or TAG, with an exception of a special initiation codon of ND5, which started with TTG, and an incomplete stop codon T-- was found in the Cytb, COX2, ND1 and ND4. All tRNAs could be folded into the canonical cloverleaf secondary structure except for the trnS1, which lacks the DHU arm and is replaced by a simple loop. The multiple tandem repeat units were found in A + T-control region. The sliding window, Ka/Ks and genetic distance analyses indicated that the ATP8 presents a high variability and fast evolutionary rate compared to other PCGs. Phylogenetic analyses based on three different datasets (PCG123, PCG12R and AA) using both Bayesian inference (BI) and maximum likelihood (ML) methods showed strong support for the monophyly of Drabescini.

1. Introduction

Deltocephalinae is the largest subfamily of leafhoppers, presenting distinct diagnostic characteristics and including over 6600 described extant species and 39 tribes widely distributed in all zoogeographic regions [1]. They are now recognized as an ecologically and economically significant subfamily of leafhoppers. Drabescini, a highly diverse tribe of Deltocephalinae, contains approximately 225 species in 46 genera divided into two subtribes. Drabescini leafhoppers are often found on woody hosts and shrubs in Old World tropical or deciduous forests and have been collected at light [2,3]. These species feed on the sap of a variety of vascular plants via piercing-sucking mouthparts, serving as vectors of numerous agricultural plant pathogens and also transmitting plant viruses to host plants.

Previous studies in this tribe mostly concentrated on taxonomic descriptions and morphological characters of the nymphs [4,5,6,7]. Subsequently, morphological phylogenetic study was performed based on a species of this tribe [8]. Phylogenetic analyses of Drabescini among three genera including Bhatia, Drabescus and Parabolopona based on the morphological characters and molecular data (28S rDNA, histon H3) found it to form a monophyletic group with high branch support [1,9]. Recent phylogenomic analysis using the anchored hybrid enrichment method showed support for the monophyly of this tribe [10]. These phylogenetic analyses mainly depended on morphological characters and were restricted to several gene fragments. Moreover, the taxonomic status and phylogenetic relationships of this tribe need to be further studied based on more DNA data. Therefore, a new method examining the mitochondrial genome (mitogenome) may provide additional molecular evidence to reconstruct the phylogeny of this group and further elucidate relationships among major lineages.

The insect mitochondrial genome is typically a circular, closed and double-stranded DNA molecule with a total of 37 genes including 13 protein-coding genes (PCGs), two ribosomal RNA genes (rRNAs) and 22 transfer RNA genes (tRNAs) [11,12]. Additionally, it also has the non-coding structure called A + T-control region or A + T-rich region [11,12,13]. The mitochondrial genome provides important molecular material and is widely used in the study of insect phylogenetic relationships, evolution, population genetic structure and biogeography on account of its faster evolutionary rate, simple genetic structure, relatively stable composition, smaller length (ranging from 14 to 17 kb) and strict maternal inheritance [13,14,15]. However, only three species of Drabescini, Athysanopsis sp. (KX437726), Drabescoides nuchalis (NC_028154) and Dryadomorpha sp. (KX437736), representing three genera, have complete or partial mitogenome sequences available in GenBank [16,17].

In this study, we sequenced and annotated the complete mitochondrial genome of two additional species including Drabescus ineffectus (Walker, 1858) (GenBank accession no. MT527188) and Roxasellana stellata Zhang & Zhang, 1998 (GenBank accession no. MT527187). We reconstructed their phylogenetic relationships and confirmed their taxonomic status, and incorporating the previously published mitochondrial genome of Drabescini based on the concatenated nucleotide sequences of 13 protein-coding genes and two ribosomal RNA genes. Furthermore, we analyzed the complete mitochondrial structure of these two species, including genome size and nucleotide composition, codon usage, tRNA secondary structure, gene overlaps and intergenic spacers, evolutionary rate, and A + T-control region and made further comparisons with other Drabescini species. The purpose of this research is to test the monophyly of this tribe and analyze phylogenetic relationships among major lineages of this superfamily.

2. Materials and Methods

2.1. Sample Collection and Genomic DNA Extraction

Specimens of D. ineffectus used in this study were collected from the Dachuan Town, Shiyan City, Hubei Province, 620 m, 3 July 2019, China, while R. stellata specimens were captured from the Diaoluoshan National Nature Reserve, Hainan Province, 15 July 2019, China. All fresh specimens were immediately preserved in 100% ethanol and stored at −20 °C in the laboratory. Identification of adult leafhoppers was based on external morphological characters and male genitalia. Total genomic DNA was extracted from abdomen tissues using the EasyPure Genomic DNA Kit (TransGen Biotech, Beijing, China) following the manufacturer’s protocol. Voucher specimens are deposited in the Entomological Museum of Northwest A&F University, Yangling, Shaanxi, China.

2.2. Mitogenome Sequencing, Assembly and Annotation

The whole mitochondrial genome sequences of these two species were generated using the next-generation sequencing (NGS) at the Illumina HiSeq™ Xten platform using the methodology of the PE150 (Biomarker Technologies, Beijing, China). The raw paired reads were retrieved and quality-trimmed selecting the mitochondrial genome of Drabescoides nuchalis (Jacobi, 1943) using reference sequences in the Geneious 8.1.3 (Biomatters, Auckland, New Zealand) with default parameters [18]. Then, the contig was assembled and annotated into the complete circular mitogenome in a similar way also using the Geneious 8.1.3 and D. nuchalis as a reference. The 13 PCGs were predicted by comparison with the homologous sequence of reference mitogenomes and finding the open reading frames (ORFs) based on the invertebrate mitochondrial genetic code Table 5. The locations of 22 tRNAs were identified by using the MITOS WebServer (http://mitos.bioinf.uni-leipzig.de/index.py) [19]. Their secondary structures were manually plotted with Adobe Illustrator CC2019 according to the MITOS predictions. The two ribosomal RNA genes (rrnS and rrnL) and the A + T-rich region were determined by the locations of adjacent genes (trnL1 and trnV) and alignment with the homologous sequences of reference mitogenomes. Next, the mitogenomic circular maps were portrayed with CGView Server (http://stothard.afns.ualberta.ca/cgview_server/) [20].

2.3. Sequence Analyses

The nucleotide composition and skew, codon usage of PCGs and relative synonymous codon usage (RSCU) values of each PCG were calculated using PhyloSuite v1.2.1 [21], and tandem repeat units of the A + T-control region were analyzed with Tandem Repeats Finder online server (http://tandem.bu.edu/trf/trf.html) [22]. Strand asymmetry was calculated by using the formulas AT-skew = (A – T)/(A + T) and GC-skew = (G – C)/(G + C). A sliding window analysis concerning 200 bp and a step size of 20 bp was conducted by the DnaSP v6 [23] to estimate nucleotide diversity (Pi value) of 13 PCGs among four Drabescini mitogenomes. The ratio of the number of nonsynonymous substitutions per nonsynonymous site (Ka) to the number of synonymous substitutions per synonymous site (Ks) with regard to 13 PCGs of four species were also estimated with DnaSP v6. Genetic distances among the mitogenomes of the four species were calculated using MEGA X (https://www.megasoftware.net) under the Kimura 2-parameter model [24]. Complete mitogenome sequences of two species were deposited in GenBank and given the accession numbers MT527187 and MT527188 (

Table 1. Mitochondrial genomes used for phylogenetic analysis in present study.

Superfamily	Family	Subfamily	Species	Accession Number	Reference
Cicadoidea	Cicadidae	Cicadinae	Cryptotympana atrata	JQ910980	[25]
Cercopoidea	Cercopidae	Cercopinae	Cosmoscarta dorsimacula	NC_040115	[26]
Membracoidea	Membracidae	Smiliinae	Entylia carinata	NC_033539	[27]
		Centrotinae	Centrotus cornutus	KX437728	[16]
			Leptocentrus albolineatus	MK746137	[28]
			Tricentrus brunneus	NC_044708	[28]
	Cicadellidae	Deltocephalinae	Abrus expansivus	NC_045238	[29]
			Norvellina sp.	KY039131	[30]
			Paramacrosteles nigromaculatus	NC_045270	Direct Submission
			Tambocerus sp.	KT827824	[31]
			Exitianus sp.	KX437722	[16]
			Exitianus indicus	KY039128	[30]
			Nephotettix cincticeps	NC_026977	Direct Submission
			Cicadula sp.	KX437724	[16]
			Alobaldia tobae	KY039116	[30]
			Maiestas dorsalis	NC_036296	[32]
			Athysanopsis sp.	KX437726	[16]
			Drabescoides nuchalis	NC_028154	[17]
			Drabescus ineffectus	MT527188	This study
			Roxasellana stellata	MT527187	This study
			Dryadomorpha sp.	KX437736	[16]
			Macrosteles quadrilineatus	KY645960	[33]
			Macrosteles quadrimaculatus	NC_039560	[34]
			Hishimonus phycitis	KX437727	[16]
			Hishimonoides recurvatis	KY364883	Unpublished
			Japananus hyalinus	NC_036298	[32]
			Orosius orientalis	KY039146	[30]
			Paralaevicephalus gracilipenis	MK450366	[35]
			Psammotettix sp.1	KX437725	[16]
			Psammotettix sp.2	KX437742	[16]
			Yanocephalus yanonis	NC_036131	[30]
			Phlogotettix sp.1	KY039135	[30]
			Phlogotettix sp.2	KX437721	[16]
			Scaphoideus maai	KY817243	[36]
			Scaphoideus nigrivalveus	KY817244	[36]
			Scaphoideus varius	KY817245	[36]
		Cicadellinae	Cicadella viridis	MK335936	[37]
			Homalodisca vitripennis	NC_006899	Direct Submission
		Coelidiinae	Olidiana ritcheriina	NC_045207	Direct Submission
			Taharana fasciana	KY886913	[38]
		Eurymelinae	Idioscopus clypealis	NC_039642	[39]
			Populicerus populi	MH492318	[40]
		Evacanthinae	Evacanthus heimianus	MG813486	[41]
			Sophonia linealis	KX437723	[16]
		Iassinae	Batracomorphus lateprocessus	MG813489	[42]
			Krisna rufimarginata	NC_046068	[42]
		Ledrinae	Ledra auditura	MK387845	[43]
			Tituria pyramidata	NC_046701	Direct Submission
		Megophthalminae	Durgades nigropicta	KY123686	[44]
			Japanagallia spinosa	NC_035685	[44]
		Typhlocybinae	Mitjaevia protuberanta	NC_047465	Unpublished
			Ghauriana sinensis	MN699874	[45]
			Limassolla lingchuanensis	NC_046037	Unpublished

).

2.4. Sequence Alignments and Phylogenetic Analyses

A total of 53 mitogenomes of Cicadomorpha insects were collected to analyze the phylogenetic relationships. Two newly sequenced specimens and 49 available mitogenomes of Membracoidea representing 13 subfamilies were selected as ingroups (Table 1). Two species, Cryptotympana atrata (Fabricius, 1775) (JQ910980) from Cicadoidea (cicadas) and Cosmoscarta dorsimacula (Walker, 1851) (NC_040115) from Cercopoidea (spittlebugs), were employed as outgroup taxa (Table 1). The nucleotide sequences of all 13 PCGs and two rRNA genes and amino acid sequences were used to elucidate the phylogenetic relationships of this tribe. All the available mitochondrial genomes were downloaded from GenBank for phylogenetic analyses (Table 1).

Complete and partial mitogenome genes were extracted using PhyloSuite v1.2.1. The nucleotide sequences of all PCGs of the 53 species were aligned in batches with the MAFFT v7.313 (https://mafft.cbrc.jp/alignment/server/) algorithm integrated into PhyloSuite v1.2.1, using the codon alignment mode and G-INS-i (accurate) strategy. The alignment of all rRNAs was conducted in the MAFFT version 7 online service with the G-INS-i strategy (https://mafft.cbrc.jp/alignment/server/) [46]. Then, gaps and ambiguous sites in the alignments were removed using Gblocks 0.91b [47] and alignments of individual genes were concatenated using PhyloSuite v1.2.1. Phylogenetic relationships based on three different datasets were generated: (1) a PCG123 matrix, including all three codon positions of 13 protein-coding genes with 10,749 nucleotides of 53 species; (2) a PCG12R matrix, including the first and second codon positions of 13 protein-coding genes plus two rRNAs with 8901 nucleotides of 52 species; (3) and an AA matrix, amino acid sequences of 13 protein-coding genes, with 3334 amino acids of 53 species. Because the rrnL and rrnS genes were missing in partial mitogenomes, Dryadomorpha sp. was excluded in the PCG12R analysis.

The optimal partitioning scheme and nucleotide substitution model for Bayesian inference (BI) and maximum likelihood (ML) phylogenetic analyses based on three different datasets were selected with PartitionFinder 2.1.1 incorporated into PhyloSuite v1.2.1, using the branch lengths linked, Bayesian information criterion (BIC) model and the greedy search algorithm [48] (Table S4). The BI phylogenetic analysis was carried out using MrBayes 3.2.6 [49] with the following settings: two independent runs were run for four to thirty million generations with sampling every 1000 generations; four independent Markov Chain Monte Carlo (MCMC) chains were run, including three heated chains and a cold chain; a stationary phase was indicated after the average standard deviation of split frequencies < 0.01 and effective sample size (ESS) > 200; the initial 25% of samples were discarded as burn-in and the remaining samples were used to generate a consensus tree and estimate the posterior probabilities (PP). In addition, the ML phylogenetic analysis was conducted by IQ-TREE v.1.6.8 [50], using the ultrafast bootstrap (UFB) algorithm with 1000 replicates. Bootstrap support (BS) values were evaluated with 1000 replicates.

3. Results and Discussion

3.1. Mitogenome Organization and Nucleotide Composition

The complete mitochondrial genome of D. ineffectus (GenBank no. MT527188) and R. stellata (GenBank no. MT527187) are circular, closed and double-stranded molecules with a length of 15,744 bp and 15,361 bp, respectively (Figure 1). The genomes are of medium-sized sequence lengths compared to the other three Drabescini mitogenomes, ranging from 12,297 bp (Dryadomorpha sp., partial genome) to 15,309 bp (D. nuchalis) (Table S3). The mutable size of mitogenomes among Drabescini species is mainly the variable length of the A + T-control region. These two newly-sequenced mitogenomes contain the typical 37 genes (13 PCGs, two rRNAs and 22 tRNAs) and the A + T-control region. The gene order is in accordance with original mitochondrial genome arrangements and other Drabescini mitogenomes. The majority strand (J-strand) generally encodes 23 genes including nine PCGs and 14 tRNAs. The remaining 14 genes are encoded on the minority strand (N-strand) and possess four PCGs, two rRNAs and eight tRNAs in these two mitogenomes (Table S1).

Nucleotide compositions of D. ineffectus are A = 41.7%, C = 13%, G = 9.9% and T = 35.4% and A = 41.4%, C = 14%, G = 9.9% and T = 34.6% in R. stellata. This exhibits a heavy AT nucleotide bias, with a high AT content for the entire sequence reaching 77.1% in D. ineffectus and 76% in R. stellata (Tables S2 and S3), respectively. This situation is also found in other Drabescini mitogenomes. The control region of D. ineffectus has the highest AT content with regard to the whole genome, PCGs and RNAs, but the PCGs have the lowest AT content. However, the rRNAs of R. stellata have the highest AT content, and the control region has the lowest AT content. Besides, the AT content in rRNAs is higher than PCGs and tRNAs in these two species. These two species showed a positive AT-skew (0.081, 0.089) and a negative GC-skew (−0.133, −0.17) in the whole genome, which also appears in other Drabescini species (Tables S2 and S3).

3.2. Protein-Coding Genes and Codon Usage

The total length of 13 PCGs with 10,956 bp for D. ineffectus and 10,932 bp for R. stellata accounts for 69.6% and 71.2% of their overall genomes, respectively (Tables S2 and S3). The size of 13 PCGs with the smallest gene was the ATP8 and the largest gene was the ND5 ranging from 153 bp to 1677 bp in this tribe. These two species show a negative AT-skew (−0.124, −0.099) and positive or negative GC-skew (0.009, −0.013) in PCGs. The AT content of the third codon (85.8%, 84.9%) was much higher than in the first (71.9%, 72.1%) and second codon positions (68.7%, 68.7%) in D. ineffectus and R. stellata (Table S2). Across the 13 PCGs in these two species, only four PCGs (ND1, ND4, ND4L and ND5) were encoded on the N-strand, whereas the other nine PCGs (COX1, COX2, COX3, ATP6, ATP8, ND2, ND3, ND6 and Cytb) were located on the J-strand (Figure 1 and Table S1). In the Drabescini mitogenomes, all PCGs started with the putative codon ATN (ATA, ATT, ATG, ATC) except for the special initiation codon of ND5, which began with TTG in Athysanopsis sp., Dryadomorpha sp. and R. stellate; this has also been observed in other Deltocephalinae mitogenomes. Correspondingly, the PCGs ended with the putative terminal codon TAA or TAG, but an incomplete stop codon T-- was found in the Cytb, COX2, ND1 and ND4 among the five sequenced mitochondrial genomes. These incomplete termination codons may be converted into TAA by posttranscriptional polyadenylation during the mRNA maturation process [51], as has been reported in other leafhoppers. Therefore, the occurrence of termination codon TAA was more common than TAG and at least an incomplete stop codon T-- was present in all five mitogenomes.

The relative synonymous codon usage (RSCU) of five sequenced mitogenomes was calculated and is summarized in Figure 2. The results showed that the four most frequently utilized amino acids were Ile (AUU), Leu (UUA), Met (AUA) and Phe (UUU). Furthermore, they are merely composed of A or U, indicating the codon usage has a strong bias toward the nucleotides A and T and reflects the high AT content in the three codon positions of PCGs in Drabescini. This codon usage pattern of these two new mitogenomes highly resembles the pattern found in previously reported Cicadellidae species [32,34]. Additionally, the codon Ser1 (AGG) is absent in R. stellata.

3.3. Gene Overlaps and Intergenic Spacers

There are 11 gene overlaps in D. ineffectus, ranging in size from 1 to 8 bp and amounting to 43 bp, while R. stellata has eight gene overlaps ranging in the same size as the former and amounting to 32 bp (Table S1). The longest overlap region in these two new mitogenomes was 8 bp between the trnW-trnC junction except for D. ineffectus which also had 8 bp between the ND6-Cytb junction. Correspondingly, the longest overlap also found in the other three known mitogenomes including Athysanopsis sp., D. nuchalis and Dryadomorpha sp. was 10 bp, 10 bp and 16 bp between ND4-ND4L, trnW-trnC and ND4L-trnT junctions, respectively. The complete mitogenomes of the four Drabescini species have one identical overlap region containing the ATP8-ATP6 junction (7 bp) except for the partial mitogenome of Dryadomorpha sp.

As opposed to 12 intergenic spacers that occur in D. ineffectus, ranging in size from 1 to 20 bp and adding up to 68 bp, a total of 14 intergenic spacers were identified in R. stellata altogether, presenting 81 bp ranging in size from 1 to 17 bp. In these two new mitogenomes, the longest intergenic spacer was 20 bp in D. ineffectus and 17 bp in R. stellata between trnY and COX1, while in Athysanopsis sp. and D. nuchalis, there were 40 bp and 16 bp between trnY and COX1, COX2 and trnK, respectively. For D. ineffectus and R. stellata, the two identical intergenic spacers were 2 bp between the ND4L-trnT and trnP-ND6 junction, respectively (Table S1).

3.4. Transfer and Ribosomal RNA Genes

The positions of all 22 typical transfer RNA genes (tRNAs) scattered throughout the whole mitogenomes were located in D. ineffectus and R. stellata (Table S1). Among them, 14 tRNAs are encoded on the J-strand and the remaining eight on the N-strand. The total length of the 22 tRNAs was 1439 bp in D. ineffectus and 1441 bp in R. stellata, accounting for 9.1% and 9.4% of their whole genomes, respectively (Tables S2 and S3). The sizes of the 22 tRNAs range from 62 (trnA, trnR) to 70 bp (trnK) in D. ineffectus and from 61 (trnA) to 73 bp (trnW) in R. stellata (Table S1). All 22 tRNAs in these two mitogenomes indicated a positive AT-skew (0.014, 0.009) and GC-skew (0.188, 0.15) (Table S2).

All tRNAs could be folded into the canonical cloverleaf secondary structure including the aminoacyl (or acceptor) arm, dihydrouridine (DHU) arm, anticodon arm and pseudouridine (TΨC) arm except for the trnS1, which lacks the DHU arm and is replaced by a simple loop in these two new mitogenomes (Figure 3 and Figure 4), as is found in other deltocephaline leafhoppers [34,36]. The missing DHU arm concerning the trnS1 probably appeared very early in the evolution of the Metazoa [52], and is frequent in insect mitogenomes [12]. Based on the predicted secondary structure, the size of the anticodon loop of all tRNAs is highly conserved with 7 bp compared with the variable length of DHU and TΨC loops (Figure 3 and Figure 4). In addition to the classic AU and GC pairs, a total number of 26 GU, 9 UU, 2 AA, 2 AC and 1 GA unmatched base pairs was found in D. ineffectus, while 23 GU, 9 UU, 1 AA and 1 AC mismatched base pairs was observed in R. stellata (Figure 3 and Figure 4). Moreover, in these two Drabescini mitogenomes, there was also an unpaired base (single A/C nucleotide) in the aminoacyl arm of trnR and the TΨC arm of trnC.

The two rRNA genes (rrnL and rrnS) are encoded on the N-strand in D. ineffectus and R. stellata. The large rRNA (rrnL), located between trnL1 and trnV, ranges in length from 1210 bp (R. stellata) to 1220 bp (D. ineffectus), while the small rRNA (rrnS) located between trnV and the A + T-rich region ranges in size from 721 bp (D. ineffectus) to 743 bp (R. stellata) (Table S1). These two rRNAs with a heavy AT nucleotide bias reach 80% in D. ineffectus and 80.4% in R. stellata, respectively (Tables S2 and S3); this is also found in other sequenced Drabescini species. Additionally, the two newly sequenced mitogenomes show the negative AT-skew (−0.120, −0.098) and positive GC-skew (0.247, 0.264) in rRNAs (Table S2). Consequently, the rRNAs are highly conserved in the Cicadellidae.

3.5. A + T-Control Region

The putative A + T-control region of Drabescini mitogenomes is located between rrnS and trnI, ranging in size from 956 bp to 1381 bp except for partial mitogenomes (Athysanopsis sp. and Dryadomorpha sp.) (Table S3). This region is deemed to be related to the origin of replication and transcription [11,12]. The control region of D. ineffectus is 1381 bp in length with an AT content of 85.3%, while R. stellata is 983 bp in length with an AT content of 74.7% (Tables S1–S3). These tandem repeats in the control region have been reported in other sequenced deltocephaline mitogenomes [32], and also found in Drabescini species, indicating that these different fragment lengths and types of absolute tandem repeat regions are present in the two taxa. The A + T-control region of D. nuchalis has one kind of tandem repeat including two 167 bp repeat units and a partial sequence (84 bp) ranging in nucleotide positions from 13 bp to 418 bp. Two types of T/A tandem repeats are present in D. ineffectus with small size of 47 bp and 50 bp (Figure 5). However, no tandem repeat unit was found in R. stellata. As in the Drabescini mitogenomes, tandem repeat regions are common and the variable length and copy number of repeat units point to a conspicuous divergence of A + T-control region.

3.6. Nucleotide Diversity and Evolutionary Rate Analysis

The sliding window analysis concerning the nucleotide diversity (Pi values) of the 13 aligned PCGs among the five Drabescini mitogenomes Athysanopsis sp., D. nuchalis, Dryadomorpha sp., D. ineffectus and R. stellata are shown in Figure 6A. This exhibits the high degree of nucleotide variation within different genes. Nucleotide diversity values range from 0.176 (COX1) to 0.348 (ATP8) in these five species. In all PCGs, the ATP8 (Pi = 0.348) presents the highest variability next to ND6 (Pi = 0.308), ND2 (Pi = 0.271) and ND3 (Pi = 0.253) showing the comparatively high nucleotide diversity values. The ND1 (Pi = 0.200), ND4L (Pi = 0.200), COX3 (Pi = 0.192) and COX1 (Pi = 0.176) with relatively low nucleotide diversity values indicate that they are relatively conserved genes in 13 PCGs (Figure 6A). To further analyze the evolutionary rate of PCGs, the ratio of Ka/Ks (ω) was used to estimate the selective pressure for each PCGs under positive selection, neutral evolution or purifying selection. As shown in Figure 6B, it is observed that all ratios of Ka/Ks (0 < ω < 1) are less than 1, indicating these PCGs are evolving under a purifying selection. Among the 13 PCGs, COX1 (ω = 0.080) has undergone the strongest purifying selection and exhibits the lowest evolutionary rate. By contrast, ATP8 (ω = 0.726) and ND6 (ω = 0.442) have undergone comparatively weak purifying pressure, demonstrating a relatively fast evolutionary rate. Furthermore, pairwise genetic distances among these five mitogenomes also yield similar results. The average values show that ATP8 (0.472), ND6 (0.399) and ND2 (0.338) with a high distance are evolving comparatively fast, while ND4L (0.233), COX3 (0.223) and COX1 (0.201) with a low distance are evolving relatively slow.

In this case, nucleotide diversity analyses in terms of other gene regions are significant for further identifying potential markers in future studies focusing in Drabescini species [53]. The COX1 presents low variation and the lowest evolution among PCGs, and it is regarded as the universal barcode for species identification and delimitation [54], particularly in Deltocephalinae with close and ambiguous morphological characters.

3.7. Phylogenetic Relationships

The Bayesian inference (BI) and maximum likelihood (ML) phylogenetic analyses among Drabescini species was conducted based on three different datasets (PCG123, PCG12R and AA). These results indicated that the phylogenetic topologies are consistent, with most branches receiving strong support (Figure 7 and Figures S1–S5). Our putative ingroup was recovered as monophyletic with respect to Cicadoidea and Cercopoidea. The inferred relationships based on the PCG123 and AA datasets Iassinae + Coelidiinae are sister to a clade comprising Megophthalminae and treehoppers with moderate to high values, which is consistent with previous phylogenetic analyses [34,42,43]. For the PCG12R datasets, Iassinae + Coelidiinae and Deltocephalinae are grouped into a clade. Another four cicadellid subfamilies Cicadellinae, Eurymelinae, Ledrinae and Typhlocybinae as currently recognized were recovered here as monophyletic with strong branch support. However, the relationships among the major lineages in the subfamily Evacanthinae remain poorly resolved and are not recovered as monophyletic.

These results provide a well-resolved phylogenetic topology for Deltocephalinae with moderate to high support for most branches. In agreement with previous studies [1,8,9,10], our phylogenetic analyses based on three different datasets using both BI and ML methods showed a strong support for the monophyly of Deltocephalinae. The present analyses consistently recovered Macrostelini as sister to the remaining tribes of this subfamily (PP = 100%, BS = 100), as have previous phylogenetic studies [28,34]. Additionally, Drabescini was recovered as monophyletic with high posterior probabilities and bootstrap support value in BI and ML trees (PP = 100%, BS = 100), as were Chiasmini, Cicadulini, Deltocephalini and Scaphoideini. In particular, the sister group of Drabescini is a clade comprised of the five included representatives of Scaphoideini with strong support, also suggested by recent phylogenetic analyses based on combined morphological and molecular data that confirm that Drabescini is closely related to Scaphoideini [1]. Nevertheless, the remaining three tribes including Athysanini, Opsiini and Paralimnini were found not to be a monophyletic group in our analyses.

Within the Drabescini, five species representing two subtribes Drabescina and Paraboloponina which were previously placed in a separate subfamily Selenocephalinae and treated as two tribes [2,3], form a monophyletic group with high branch support. Our analyses provided distinct evidence of close relationships between Deltocephalinae and Selenocephalinae, suggesting that the latter was not distinguishable from Deltocephalinae. While these results indicate consistent support for the monophyly of this tribe, its internal topologies diverge when using different datasets. Phylogenetic analyses among five Drabescini species based on PCG123-BI and PCG123-ML methods showed the relationships (R. stellata + (Dryadomorpha sp. + (D. nuchalis + (Athysanopsis sp. + D. ineffectus)))). On the other hand, the PCG12R-BI and PCG12R-ML analyses yielded the topologies (R. stellata + (D. ineffectus + (Athysanopsis sp. + D. nuchalis))) and (R. stellata + (D. nuchalis + (Athysanopsis sp. + D. ineffectus))), respectively. Moreover, the inferred relationships based on amino acid sequences (R. stellata + ((Athysanopsis sp. + D. nuchalis) + (D. ineffectus + Dryadomorpha sp.))) were also recovered with high support values. However, these studies were not retrieved as congruent results based on current mitogenome data. Therefore, further samples should be added to elucidate the status and relationships and improve the resolution of the still poorly-supported and varied branches among the major lineages within Membracoidea.

4. Conclusions

In this study, we determined two newly complete mitogenomes including D. ineffectus and R. stellata and found them consistent with previously reported mitogenomes of Cicadellidae. The Drabescini mitogenomes are highly conserved in base content and composition, genome size and order, protein-coding genes and codon usage, and secondary structure of tRNAs. The BI and ML phylogenetic analyses among the major lineages based on the concatenated datasets (PCG123, PCG12R and AA) yielded the well-resolved topologies with moderate to high support for most branches except for a few deep internal nodes within Membracoidea. While the relationships among tribes remain poorly resolved within Deltocephalinae, Drabescini was recovered as monophyletic with strong branch support and revealed close relationships (R. stellata + (Dryadomorpha sp. + (D. nuchalis + (Athysanopsis sp. + D. ineffectus)))). Furthermore, these results provide the comprehensive framework and valuable data toward the future resolution of phylogenetic relationships in this tribe.