Dryophytes versicolor is one of the most extreme freeze-tolerant frogs from eastern North America. In this study, the mitochondrial genome of D. versicolor was sequenced to analyze the phylogenetic relationships among Hylidae and investigate mitochondrial gene expression in response to freezing and anoxia. The total length of the D. versicolor mitogenome is the longest known to date among the available family members of Hylidae. Both maximum likelihood (ML) and Bayesian inference (BI) analyses strongly supported D. versicolor as a sister clade to (D. japonica + D. ussuriensis) + (D. suweonensis + D. immaculata (KP212702)), and indicated that Dryophytes is monophyletic. Using the mitochondrial genome, gene expression analysis was performed using RT-qPCR in skeletal muscle samples, and determined that relative levels of D. versicolor COX2 increased by 2.40 ± 0.23 fold in response to anoxia, but did not change with exposure to freezing. In addition, ND3 transcript levels decreased in response to anoxia but remained constant during freezing. By contrast, COX1 transcript levels decreased with exposure to freezing, but did not change under anoxic conditions. These results suggest that modulations of protein-coding mitochondrial genes of D. versicolor may play a role in the molecular response to freezing and anoxia tolerance.

Different mitochondrial gene copies or gene rearrangements are observed in some interspecific mitochondrial genomes (mitogenomes), but few in intraspecific mitogenomes. In this study, we sequenced the complete mitogenomes of Statilia maculata and Statilia nemoralis (Mantinae: Mantini). The genetic distance of the whole genomes between these two species was just 0.5%, suggesting that S. maculata and S. nemoralis might be the same species. The mitogenome of S. maculata had five copies of trnR genes (three copies were identical and the other two had some base substitutions), whereas the mitogenome of S. nemoralis had six trnR genes. Five, six or seven tandem copies of trnR genes in S. maculata were found in different localities, respectively. We further sequenced the ND3-ND5 region of fifty individuals hatched from one ootheca (Tianjin) and we found that all individuals had five identical tandem copies of trnR genes and same sequences. Hence, we concluded that different copies of trnR genes can occur in the same species. The mitogenomes of S. maculata and S. nemoralis with other mitogenomes of Mantinae species published were used to construct BI and ML phylogenetic trees. In this study, both BI and ML phylogenetic trees showed that Mantini and Statilia were monophyletic whereas Paramantini was paraphyletic.

Gene rearrangements have been found in several mitochondrial genomes of Mantodea, located in the gene blocks CR-I-Q-M-ND2, COX1-K-D-ATP8 and ND3-A-R-N-S-E-F-ND5. We have sequenced one mitogenome of Amelidae (Yersinia mexicana) and six mitogenomes of Mantidae to discuss the mitochondrial gene rearrangement and the phylogenetic relationship within Mantidae. These mitogenomes showed rearrangements of tRNA genes except for Asiadodis yunnanensis and Hierodula zhangi. These novel gene rearrangements of Mantidae were primarily concentrated in the region of CR-I-Q-M-ND2, including gene translocation, duplication and pseudogenization. For the occurrences of these rearrangements, the tandem duplication-random loss (TDRL) model and slipped-strand mispairing model were suitable to explain. Large non-coding regions (LNCRs) located in the region of CR-I-Q-M-ND2 were detected in most Mantidae species, whereas some LNCRs had high similarity to the control region (CR). Both BI and ML phylogenetic analyses supported the monophyly of Mantidae and the paraphyly of Mantinae. The phylogenetic results with the gene order and the location of NCRs acted as forceful evidence that specific gene rearrangements and special LNCRs may be synapomorphies for several groups of mantises.