The First Conserved Mitochondrial Genome of Polygraphus Poligraphus (Coleoptera: Curculionidae) and Its Phylogenetic Implications


 Background: Polygraphus poligraphus L., the four-eyed spruce bark beetle, belongs to the Curculionidae (Coleoptera), which mainly harms Picea asperata Mast and Pinus armandii Franch tree trunks. So far, there is no mitochondrial genome reported for P. poligraphus.Results: In this study, we sequenced and annotated the nearly complete mitogenome of P. poligraphus for the first time and predicted the secondary structures of its tRNAs. The results showed that the mitogenome of P. poligraphus was 15,302 bp (partial genome) in length with A + T content of 69.65% due to large-scale duplication. The nearly complete mitochondrial genome of P. Poligraphus contained a set of 36 genes typical of the insect mitogenome, including 13 protein-coding genes (PCGs), 2 ribosomal RNA genes (rRNAs), 21 transfer RNA genes (tRNAs) but lacked tRNA-Ile, as for the typical insect mitogenome. The results of nucleotide skew statistics showed that the AT-skews and GC-skew of P. poligraphus were positive and negative, respectively, which were similar to other Scolytinae insects. All PCGs were initiated with the standard start codon ATN. All tRNA genes had the typical cloverleaf structure, except for the trnS1, which lacked a dihydroxyuridine (DHU) arm. Furthermore, we reconstructed phylogenetic trees of P. poligraphus based on the data set of the mitogenome’s protein-coding gene sequences using the Bayesian inference (BI) method. Phylogenetic analysis indicated that the P. poligraphus mitogenome clustered with Gnathotrichus materiarius and Pityophthorus pubescens mitogenomes in a monophyletic manner. The phylogeny of these three genera of Scolytinae is presented as Polygraphus + (Gnathotrichus + Pityophthorus). Conclusions: The results presented herein will provide a reference for further molecular taxonomy, evolution and phylogenetic research of P. poligraphus. However, additional mitogenome samples are still needed to more satisfactorily resolve the phylogeny of the Scolytinae.


Background
The mitochondrion is a fundamental eukaryotic organelle, descended from an alphaproteobacterium that formed a permanent symbiosis with the ancestral eukaryote roughly 2 billion years ago [1]. The mitochondrial genomes of arthropods have been studied extensively, and insects represent approximately 80% of the arthropod mt genomes that have been sequenced [1]. Insect mitochondrial genomes are small, typically a double-stranded circular molecular structure ranging from14 to19 kb in size. With few exceptions, all animal mitochondrial genomes contain a typical set of 37 genes: 13 protein-coding genes (PCGs) (ATP6, ATP8, COI-III, ND1-6, ND4L, and CYTB), 2 ribosomal RNA genes (rRNAs) (rnl and rns), 22 transfer RNA genes (tRNAs), and a putative control region (A + T-rich region) [2][3][4]. Compared with partial mitochondrial genes the whole mitogenome can provide more meaningful information such as the arrangement of gene sequences, secondary structures of RNA, codon usage and structural features of the A + T-rich region [5][6][7]. This is because of the unique features of a complete mitochondrial genome including simple genetic structure, maternal inheritance, high rate of evolution and low rate of recombination [8][9][10]. Over the past decade, mitochondrial genomes have become widely used for molecular evolution, population genetics, systematics and phylogenetics [1,[11][12][13][14][15].
Coleoptera, the largest insect order, contains 4 suborders (Archostemata, Adephaga, Polyphaga and Myxophaga), 17 superfamilies, 168 families and over 380,000 described species. Of these, about 10,000 are known in China [16,17]. Polygraphus poligraphus L., four-eyed spruce bark beetle, belongs to Scolytinae of Curculionidae of Coleoptera [18]. It is a harmful, wide-spread invasive insect and one of the 236 dangerous forest pests announced by the State Forestry and Grassland Administration of China. Polygraphus poligraphus is mainly distributed in Russia, Finland, Norway, Sweden, Denmark, Turkey, Poland, Czech Republic, Germany, Austria, former Yugoslavia and Gansu, Jilin, Liaoning, Heilongjiang, Neimenggu, and Ningxia provinces in China [19]. It mainly damages Picea asperata Mast and Pinus armandii Franch as adults and can cause the death of entire forests in severe cases. Nevertheless, P. asperata and P. armandii are important timber forests, ecological public welfare forests, water conservation forests and greening trees, and occupy an irreplaceable position in the forest resources of China [19,20]. The morphology, biology and biological and chemical control of P. poligraphus have been studied [21][22][23]. However, it is imperative to integrate the sustainable development of forest ecosystems with sustainable control techniques for bark beetles.
With the rapid development of high-throughput sequencing technology, the number of insect mitochondrial genomes being studied is increasing. Over two years there were more than 100 whole sequenced mitochondrial genomes and more than 1000 partially sequenced mitogenomes placed in the GenBank database for Coleoptera (last visited on March 25, 2021) [24]. Among these species, no information about the complete or nearly complete mitochondrial genome and phylogenetic position of P. poligraphus is mentioned, which impedes the application of biological control. In view of the large number of species of Scolytinae and the di culty distinguishing them, accurate identi cation of these species is essential to prevent the invasion of these species. Here we report for the rst time the nearly complete mitogenome of P. poligraphus and clarify its phylogenetic position within the Scolytinae.
In the present study we analyzed the genome organization, nucleotide composition, composition biases, codon usage, construct of tRNA secondary structures and phylogenetic relationships of the P. poligraphus mitogenome.

Materials
Sample collection, identi cation and DNA extraction

Mitogenome sequencing
Illumina sequencing was used to obtain the mitogenome sequence of P. poligraphus. Brie y, quali ed DNA samples tested by electrophoresis were randomly interrupted with Covaris ultrasonic crusher with a length of about 350 bp. Then, the whole library was constructed using the NEBNext® Ultra™ DNA Library Prep Kit for Illumina (NEB, USA) to repair the end of the DNA fragments, add poly'A', add sequencing joints, purify, PCR ampli cation and other steps. Subsequently, Qubit v2.0 was used for preliminary quanti cation, and the library was diluted to 2 ng/ µL. Lastly, Agilent 2100 was used to detect the inserted fragments of the library, the insert size was in line with the expectation, and the Q-PCR method was used to accurately quantify the effective concentration of the library to ensure the quality of the library.

Mitogenome annotation and analysis
The paired-end reads for mitochondrial genome sequences of P. poligraphus were assembled by MITObim v1.9 with the invertebrate genetic code employed [25]. Subsequently, the mitochondrial genomes of P. poligraphus were annotated with Geneious 10.

Phylogenetic analyses
To reconstruct phylogenetic trees for the estimation of P. poligraphus taxonomic status, the complete mitogenome sequences of 19 Scolytinae species and three outgroups (Sitophilus zeamais, Cyrtotrachelus buqueti and Rhynchophorus ferrugineus) were downloaded from GenBank ( Table 1). Ahead of Bayesian inference analysis, the Bioedit version 7.2.5 [30] was used to recon rm the gene boundaries and remove the ambiguous sites by aligning the P. poligraphus and previously reported Scolytinae 13 PCGs. The nucleotide substitution model was individually selected for each of two rRNAs and three codon positions of concatenated PCGs using jModelTest version 2.1.4 [31] using the Akaike Information Criterion (AIC) [32]. The best-t nucleotide substitution models were selected as 'GTR + G + I'. The phylogenetic tree was reconstructed using Bayesian 3.2.0 [33] based on 13 mitochondrial protein-coding genes.  (Fig. 1). The nearly complete mitochondrial genome of P. poligraphus contained the set of 36 genes typical of insect mitogenomes: 13 protein-coding genes (PCGs) (ATP6, ATP8, COI-III, nad1-6, nad4L, and cob), 2 ribosomal RNA genes (rRNAs) (12S rRNA and 16S rRNA), 21 tRNAs (lack tRNA-Ile). Twenty-two genes are encoded on the majority strand (L-strand), and the remaining 14 genes are located on the minority strand (H-strand) in this mitogenome (Table 2). The nucleotide composition of the P. poligraphus mitochondrial genome was 37.26% of A, 32.39% of T, 18.46% of C, 11.89% of G and 69.65% of A + T content (Table 3). Generally, the whole mitogenomes of Scolytinae exhibited a strong base composition bias toward 66.15% (Gnathotrichus materiarius) − 77.46% (Hylastes brunneus) for A + T content. The entire mitogenomes with a high A + T content bene t from the composition of PCGs, tRNAs and rRNAs. Scolytinae. The A + T content in tRNAs was higher than that in PCGs in all 20 species. Hylastes brunneus had relatively weaker tRNA A + T content compared with other Scolytinae species ( Fig. 2A). In addition to the A + T content, the skewness (AT-skew and GC-skew) of the base composition in nucleotide sequences was also used to describe the base composition of mitogenomes [27,34]. The results of nucleotide skew statistics show that the AT-skews of P. poligraphus were slightly positive. The AT-skews of PCGs, tRNAs and rRNAs for whole mitogenomes in the Scolytinae are positive because the AT-skews value of nad1, nad4, nad4L, nad5 and rrnL are relatively greater and in other regions are slightly negative. Compared with other species, the AT-skews of G. materiarius were slightly less lower (Fig. 2B). The GC-skew values are all negative in whole mitogenomes. The GC-skew of P. poligraphus is similar to other Scolytinae insects (Fig. 2C). The nucleotide skewness in Scolytinae mitochondrial genomes is consistent with that of most other insects [34]. Protein-coding genes and codon usage The PCGs of the mitogenome were 11,061 bp long for P. poligraphus (Table 3). Four PCGs (nad1, nad4, nad4L and nad5) were encoded on H-strand, and the other nine PCGs were located the L-strand. The sizes of 13 PCGs ranged from 159 bp (atp8) to 1696 bp (nad5) in P. poligraphus ( Table 2). All 20 mitogenomes had similar characteristics with the smallest sized PCG of atp8 and the largest that of nad5. All PCGs in the P. poligraphus mitogenomes started with the standard ATN codon. The start codon ATG was shared with cox3, atp6, nad4, nad4L, nad6 and cob; the start codon ATT was shared with cox2, atp8, nad1, nad2 and nad5; cox1 started with codon ATC; and the nad3 started with codon ATA. The conservative stop codon TAA was shared with cox1, cox3, atp6, nad1, nad2, nad3 and nad6; the stop codon TAG was shared with atp8 and nad4L; nad4 and cob stop with an incomplete codon "TA-", and cox2 and nad5 end with the single nucleotide "T-". "TA-" and "T-"" denote that the TAA stop codon is presumed to be completed by the addition of 3' "A" residues to the mRNA. The incomplete termination codons are common across arthropod mitogenomes and are completed by post-transcriptional polyadenylation during the mRNA maturation process [35,36].
The amino acid composition and the relative synonymous codon usage (RSCU) of mitogenomes of P. poligraphus and the other 19 Scolytinae species are summarized in Fig. 3. The total number of codons in the PCGs ranged from 3060 (Hylastes attenuatus) to 3836 (Dryocoetes villosus). The pattern of codon usage was generally similar among Scolytinae mitogenomes such as the seven most frequently used codons: UUU, UUA, UAU, AUU, AAA, AAU and AUA, all composed wholly of A or U. In the P. poligraphus mitogenome, 3,542 amino acids were translated, of which 1,196 (33.77%) were encoded by the seven frequently used codons above. And, in the H. brunneus mitogenome, 1,672 (45.55%) amino acids were encoded by the seven frequently used codons; this was the greatest in the 20 Scolytinae mitogenomes. However, the codons absent in Scolytinae mitogenomes were different. In the Xylosandrus crassiusculus, P. pubescens and Ips sexdentatus mitogenomes, the GCG codon was absent, whereas the CCG and CGU codons were absent in Trypodendron domesticum and Dryocoetes autographus respectively. In general, the high C/G content in the absent codons effectively re ects nucleotide A + T bias in the mitochondrial PCGs among Scolytinae.
Transfer and ribosomal RNA genes The 21 tRNAs of the P. poligraphus mitogenomes were scattered discontinuously over the partial mitogenome (due to large-scale duplication). The length of 21 tRNA genes ranged from 60 bp (trnS1) to 70 bp (trnK). The total length of tRNAs was 1,375 bp, accounting for approximately 9% of the mitogenome. Among them, eight tRNA genes were transcribed from the H-strand and 13 from the Lstrand ( Table 2). As shown in Fig. 4, most tRNAs sequences could fold into the typical clover-leaf secondary structure (including amino acid acceptor (AA) arm, dihydrouridine (DHU) arm, variable (V) loop, anticodon (AC) arm and TΨC (T) arm), while trnS1 (AGN) forms a simple loop due to lacking the stable DHU arm. The lack of a DHU stem in trnS1 is generally present in Coleoptera insects and has been con rmed as a typical feature of metazoan mitogenomes [1, 24 37-41]. In tRNA genes of the P. poligraphus mitogenome, a great number of nucleotide substitutions are found in ve different stems. Compared with variable TΨC and DHU loops, the anticodon stem and loop is highly conserved (Fig. 4). Except for the classic AU and CG pairs, we recognized 21 mismatched base pairs in the tRNA genes secondary structures of P. poligraphus. Among them, 19 were G-U mismatched base pairs, one was a U-U pair and two were G-G pairs. The overrepresented pattern of the non-canonical G-U pairs in tRNA genes of the mitogenome is commonly present in other insects [24,42].
Two rRNA genes (rrnL and rrnS) were transcribed from the H-strand in P. poligraphus. The larger rrnL was 1,334 bp long, and located between the trnL1 and trnV, while the smaller rrnS was 772 bp in length and located behind trnV (Fig. 1, Table 2). The rRNA genes presented a heavy AT nucleotide bias, with A + T content 74.88% in P. poligraphus (Table 3). In the 20 mitogenomes of Scolytinae analyzed, the lengths of rrnL ranged from 1,239 (Trypophloeus asperatus) to 1,372 (O. laricis) bp, and of rrnS from 755 (G. materiarius) to 815 (P. bidentatus) bp.

Overlapping sequences and intergenic spacers
The mitogenome of P. Poligraphus have a total of 74 bp overlap sequences and 91 bp intergenic spacer sequences, which are all made up of 12 regions in the range from 1 to 40 bp and 1 to 34 bp respectively.

Phylogenetic analysis
We reconstructed phylogenetic trees based on the 20 Scolytinae species and three outgroups (S. zeamais, C. buqueti and R. ferrugineus) 13 mitogenomes PCGs using Bayesian 3.2.0 (Fig. 5). Phylogenetic analysis showed that the P. poligraphus mitogenome clustered clearly with the G. materiarius and P. pubescens mitogenomes in a monophyletic manner. The phylogeny of these three genera of Scolytinae is presented as Polygraphus + (Gnathotrichus + Pityophthorus). The result is consistent with previous results based on traditional classi cation analyses. The principal aim of this study was to determine the phylogeny of Scolytinae and the location of P. poligraphus. On the one hand, since we did not sample all the genera of the Scolytinae, a more comprehensive sampling of the taxa is needed to fully resolve the genus relationships within the Scolytinae. On the other hand, our study adds to the limited data in existing databases. Most of the phylogenys of Scolytinae are reconstructed based on mitogenomes [43,44], and we believe that more nuclear genes are needed to clarify the genus relationship of Scolytinae.

Conclusions
In this present study, we sequenced and annotated the nearly complete mitogenome of P. poligraphus and predicted the secondary structures of its tRNAs. The results showed that our newly-determined mitogenome of P. poligraphus had a similar composition to the typical insect mitogenome. In the secondary structure of tRNA, the lack of a DHU stem in trnS1 is consistent with all Coleoptera insects and has been con rmed as a typical feature of metazoan mitogenomes. Our P. poligraphus mitogenome provides an important data resource for further studies and contributes to our understanding of the phylogeny. However, additional mitogenome samples are still needed to more satisfactorily resolve the phylogeny of the Scolytinae. Figure 5 Phylogeny of 20 species within the Scolytinae based on the Bayesian analysis of 13 mitochondrial protein-coding genes. 'GTR+G+I' was used as the best-t nucleotide substitution model. The support values are shown next to the nodes. Three species within the subfamily Dryophthorinae were included as the outgroup taxa.