Molecular Phylogenic Study To Anisoptera Based On COⅡ Complete Gene (Insecta: Odonata)

By designing primers of COII (Cytochrome oxidase II) complete gene of mitochondrial DNA about Odonata, COII complete sequences of 15 genera including 19 species of Libellulidae had been sequenced and analyzed. Another sequence, Davidius bicornutus, which belongs to Gomphidae, was sequenced as outgroup. We reconstructed phylogenetic trees by using CluxtalX1.8, ContigExpress, MEGA2.1, PHYLIP3.6a and MrBayes V3.0 softwares, and using the maximum parsimony and maximum likelihood methods, respectively (another two complete COII gene sequences, the only existing sequences of Libellulidae in Genbank to date, had been added to the phylogenetic trees, that are 15 genera including 21 species). In order to make clear the relationships among families of Anisoptera. another 3 COII gene complete sequences, which belong to Aeshinidae, Cordulegastridae and Corduliidae, respectively, had been sequenced. So using Pseudagrion civicum, which belongs to Coenagrionidae of Zygoptera, as outgroup, phylogenetic trees had been reconstructed, which include 8 species that belong to 5 families.All the results could be showed as following:1) To all these complete sequences of COII gene of Anisoptera, nucleotide compositions present a distinct bias. Average contents of A, T, G and C is 36.3%, 32.3%, 15.1% and 16.4%, respectively. Content of A+T is 68.5%, which proves dragonfliy is an original group.2) TS/TV rations is 0.7, transversion is slightly higher than transition. Most substitutions occur at the third sites which include most parsim-information;the second sites are very conserved and include few parsim-information;most substitutions of the third sites are synonymic, which do not result in the change of amino acid. Transitions occur between T and C, while transversions occur between T and A.3) Complete sequences of 688 bp encode 229 amino acids, which include 20 sorts of common ones. Every sequence includes two Cys. Amino acid composition presents a distinct bias: Leu and Ile present high content, while Cys is very few.4) Discussion the validity of COII to phytogeny, the results showed: COII could not solve phylogenic relations of close genera, but suitable to solve phylogenic relations of species within genus.5) Using with methods such as NJ, MP and BI, phylogenic trees had been reconstructed. The results showed diffent methods could reconstruct consistent trees, but BI presents higher holding values than other two, also prsesnts a quicker speed than MP.6) Sympetrum uniforme should be put into the Crocothemis and changed into Crocothemis uniforme.7) Suggest Trithemis (Trithemistinae) be put into Sympetrinae, Pseudothemis (Trithemistinae) be put into Trameinae, and Palpopleura (Palpopleurinae) be put into Sympetrinae. Agree completely with the traditional taxonomy view of Libellulinae.8) Do not agree with the view of raise Macrominae up to Macromidae;agree with the views of put both Aeshnidae and Cordulegastridae into Aeshnoidea, and raise Gomphidae up to a superfamily. Order of the evolution relationship among 5 familys is: (Gomphidae, (Cordulegastridae, (Aeshinidae, (Corduliidae, Libellulidae)))).