Mitogenomes provide new insights of evolutionary history of Boreheptagyiini and Diamesini (Diptera: Chironomidae: Diamesinae)

Abstract Mitogenomes have been widely used for phylogenetic reconstruction of various Dipteran groups, but specifically for chironomid, they have not been carried out to resolve the relationships. Diamesinae (Diptera: Chironomidae) are important bioindicators for freshwater ecosystem monitoring, but its evolutionary history remains uncertain for lack of information. Here, coupled with one previously published and 30 new mitogenomes of Diamesinae, we carried out comparative mitogenomic analysis and phylogenetic analysis. Mitogenomes of Diamesinae were conserved in structure, and all genes arranged in the same order as the ancestral insect mitogenome. All protein‐coding genes in Diamesinae were under stronger purifying selection than those of other nonbiting midge species, which may exhibit signs of adaptation to life at cold living conditions. Phylogenetic analyses strongly supported the monophyly of Diamesinae, with Boreheptagyiini deeply nested within Diamesini. In addition, phylogenetic relationship of selected six genera was resolved, except Sympotthastia remained unstable. Our study revealed that the mitogenomes of Diamesinae are highly conserved, and they are practically useful for phylogenetic inference.

In this study, we provide 30 newly sequenced (nearly) complete mitogenomes from 30 species representing Boreheptagyiini (four species of one genus) and Diamesini (26 species of five genera) using next-generation sequencing. We analyzed the genomic structure, base composition, substitution, and evolutionary rates among Diamesinae, expanding our knowledge of its diversity of mitogenomes. Coupled with published data, we carried out phylogenetic analysis of Boreheptagyiini and Diamesini based on 31 mitogenomes.

| Taxon sampling and dna extraction
Field collection of 30 species were conducted in China during 2014-2020, using classical insect collection techniques such light traps, sweep traps, Malaise traps, and D-nets. Specimens were preserved in ethanol (85% for adults, 95% for immature), and stored at dark at −20°C before morphological and molecular analyses. The total genomic DNA was extracted from thorax of adult and middle larval bodies using a Qiagen DNA Blood and Tissue Kit (Qiagen) following the manufacturer's protocol. After DNA extraction, the cleared exoskeleton of thorax was mounted in Euparal on microscopy slides together with the corresponding wings, legs, and antennae following the procedures outlined by Saether (1969). The DNA and vouchers of the species are deposited at the college of Life Sciences, Nankai University, Tianjin, China. Specimens were identified morphologically using relevant taxonomic revisions and species descriptions (Lin, Chang, et al., 2021;Lin, Yu, et al., 2021;Makarchenko et al., 2008Makarchenko et al., , 2021 Oliver, 1983Oliver, , 1989Reiss, 1968;Sun et al., 2019)

| Sequencing and mitogenome assembly
The whole genomes were sequenced using the Illumina NovaSeq 6000 platform with 150-bp paired-end reads at Novogene Co., Ltd. (Beijing, China). The raw sequencing reads were trimmed with Trimmomatic (Bolger et al., 2014), and then about two Gb of clean data were obtained for each sample. The clean data were assembled using IDBA-UD (Peng et al., 2012) with minimum and maximum k values of 40 and 120 bp, respectively, and the similarity was set as 98%.
The cytochrome c oxidase I (COI) barcode sequence for each species was obtained by Sanger sequencing herein and from previous study (Lin, Yu, et al., 2021), and served as the "bait" references to acquire the best-fit and targeted mitochondrial contigs by BLAST (Altschul et al., 1990) search in Geneious 2020.2.1 (Kearse et al., 2012). Moreover, clean reads were mapped onto the obtained mitogenome using Geneious to check the accuracy of the assembly.

| Genome annotation, composition, and substitution rate
Genome annotation was conducted following previous study (Zheng et al., 2020). Transfer RNA (tRNA) genes and their secondary structures were identified on MITOS2 webserver (available at http:// mitos2.bioinf.uni-leipz ig.de/index.py). Ribosomal RNA (rRNA) genes and protein-coding genes (PCGs) were annotated by aligning with homologous genes of Potthastia sp. in Geneious. Newly sequenced mitogenomes were submitted to GenBank (accession numbers: pending). The mitogenome maps were drawn by the CG View server V 1.0 (Grant & Stothard, 2008). The base composition, codon usage, and relative synonymous codon usage (RSCU) values were calcu-

| Substitution rate and phylogenetic analyses
The level of base substitution saturation for each gene and each position of the PCGs was assessed using DAMBE 5.6.14 (Xia, 2013).
Substitution of each of the three codon positions are generally not saturated, except for the transition of 3rd codon positions ( Figure   S1). Therefore, the 3rd codon positions of PCGs were excluded for the phylogenetic analyses. Each gene was aligned using MAFFT 7.402 (Katoh & Standley, 2013) with algorithm G-INS-i strategy. Gap in each matrix was treated as the fifth character and was retained in this study. Alignments of individual genes were then concatenated using SequenceMatrix v1.7.8 (Vaidya et al., 2011), after which three datasets were prepared for phylogenetic analyses: PCG12 (the 1st and 2nd codon positions of the 13 PCGs), PCG12R (the 1st and 2nd codon positions of the 13 PCGs and two rRNAs), and third AA (amino acid sequences of the 13 PCGs). The best partitioning scheme and best-fit substitution model for each partition was tested using PartitionFinder 2.0 (Lanfear et al., 2017) with the Bayesian Information Criterion (BIC). Phylogenetic analyses were conducted with Maximum likelihood (ML) reconstruction and Bayesian inference (BI). The ML analysis was performed using IQ-TREE 1.6.10 (Nguyen et al., 2015) with the best-fit substitution model and 1000 bootstrap replicates. BI analysis was performed using MrBayes 3.2.7a (Ronquist et al., 2012) with substitution model in Table S1.
Two simultaneous Markov chain Monte Carlo (MCMC) runs of 10,000,000 generations were conducted, trees were sampled every 1000 generations, and the first 25% of trees discarded as burn-in. Tracer 1.7 (Rambaut et al., 2018) was used to check convergence of runs.

| Mitogenome features of Diamesinae
The mitogenomes of 31 Diamesinae species were included in this study, 21 of which are complete, with the entire length ranging from 15,913 bp to 16,411 bp (Table S2) Table S3). We also provided the Ka/Ks values of mitochondrial PCGs of Orthocladiinae and Stenochironomus that we previous reported in Table S3, which are higher than that in Diamesinae.
Each mitogenome of Diamesinae contains 22 typical tRNA genes, with A+T content ranging from 74.0% to 77%. The nucleotide skew of tRNA genes among Diamesinae is consistent, exhibiting positive AT-skew and negative GC-skew ( Figure 2, Figure S2). Both 12S and 16S rRNA genes transcribe from the minority strand (N-strand).
Among the mitogenomes of Diamesinae, the length of 12S rRNA gene varies from 794 to 815 bp, and the length of 16S rRNA gene varies from 1345 to 1374 bp (Table S2). The A+T content of 12S and 16S rRNA genes ranges from 76% to 78.6% and 80% to 82.8%, respectively. Both 12S and 16S rRNA genes exhibit positive GC-skew in the mitogenomes of Diamesinae (Figure 2, Figure S2). A total of 21 mitogenomes in the present study have the complete control region, varying from 907 to 1309 bp (Table S2). The A+T content of the control region among the mitogenomes of Diamesinae ranges from 87.7% to 93.9% (Figure 2), extremely higher than the whole mitogenomes.

| Mitogenome features
A total of 31 mitogenomes of Diamesinae are included in this study, of which 10 mitogenomes have incomplete control region by the highly gene duplication (Cameron, 2014;Zhang & Hewitt, 1997).
The lengths of 21 complete mitogenomes of Diamesinae range from 15,913 bp to 16,411 bp due to the variation of the control region.
The gene number and arrangement of these mitogenomes are conserved, and all genes arranged in the same order as the ancestral insect mitogenome (Clary & Wolstenholme, 1985). The nucleotide composition of the mitogenomes of Diamesinae is biased toward A+T, which is consistent with other published chironomid species (Beckenbach, 2012;Deviatiiarov et al., 2017;Zheng et al., 2021).
The mitogenomes of Diamesinae exhibit positive or negative ATskew and negative GC-skew, the nucleotide bias of these mitogenomes may be related to the asymmetric mutation processes during replication (Hassanin et al., 2005). Most PCGs of mitogenomes of Diamesinae terminated with complete termination codons, except ND4 and ND5 in a few mitogenomes, terminated with a single T that may be completed by post-transcriptional polyadenylation (Ojala

| Evolutionary rate
We compared the Ka/Ks value between Diamesinae and other subfamilies of Chironomidae. Previous studies reported the Ka/Ks values of mitochondrial PCGs of Orthocladiinae and Stenochironomus Zheng et al., 2022), and the Ka/Ks values of each PCG in these chironomids are higher than that in Diamesinae (Table   S3), indicating that the mitochondrial genomes of Diamesinae are under stronger purifying selection than other nonbiting midge species (Hurst, 2002). Mitochondrial genome played a central role in animal energy production, and stronger purification selection could enhance their conserved role in energy production (Hassanin et al., 2009;Yuan et al., 2020). The existence of stronger purifying selection in Diamesinae species may exhibit signs of adaptation to life at cold living conditions (high latitude and high altitude) . Severe habitats generally accumulate more deleterious mutations, and the stronger purifying selection of mitochondrial PCGs in Diamesinae species may help against these deleterious mutations (Sarkar et al., 2020;Wang et al., 2019). In addition, Diamesinae species lives in the cold environment (Lencioni & Rossaro, 2005;Montagna et al., 2016;Sun et al., 2019) and have a small range of activities, which could lead to a lower metabolic rate. Given the correlation between purification selection and metabolic rate has been reported in several species (Chong & Mueller, 2013;Shen et al., 2009;Wang et al., 2019), we hypothesized that the stronger purifying selection in Diamesinae species may also be associated with metabolic requirement. The evolutionary rate analyses of Diamesinae also provided new insights for the study of species delimitation. The evolutionary rate of COI was generally considered to be consistent with the evolutionary rate of the species itself, so it has been widely used in species delimitation and phylogenetic studies (Havill et al., 2021;Jones et al., 2021

| Phylogenetic analyses
Previous study has revealed that mitogenomes have poor phylogenetic signals at the subfamily level of Chironomidae (Zheng et al., 2021). However, our study reveals that the mitogenomes of Diamesinae are practically useful for phylogenetic inference. In our study, we applied a variety of strategies to explore the phylogenetic relationships of six genera of the Diamesinae using mitogenomic data, and confirmed the monophyly of Diamesinae. According to traditional morphological systematics, Boreheptagyiini could be separated from other tribes of Diamesinae by having distinct pubescence, low antennal ratio, and usual dorsocentral and prealar setae in adults (Brundin, 1966;Saether, 1977;Serra-Tosio, 1973 (Xi et al., 2016). Therefore, to finally explore the evolutionary history of Diamesinae, a complete resolution will require a comprehensive taxa sampling with the most informative mitochondrial and nuclear markers.

| CON CLUS ION
In this study, we sequenced 30 mitogenomes representing 30 species of six genera of Boreheptagyiini and Diamesini by whole genome sequencing technologies, and did the first comparative analysis of mitogenome base composition and evolutionary history in Diamesinae. This study showed that mitogenomes of Diamesinae were conserved in structure, gene order, and nucleotide composition. All protein-coding genes in Diamesinae were under stronger purifying selection than those of other nonbiting midge species, which may exhibit signs of adaptation to life at cold living conditions.

Mitogenomes could provide new insight into evolutionary history of
Diamesinae based on the dated molecular phylogeny. Tong for their collecting material.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.