ITS2 Secondary Structure Improves Discrimination between Medicinal "Mu Tong" Species when Using DNA Barcoding

Abstract

DNA barcoding is a promising species identification method, but it has proved difficult to find a standardized DNA marker in plant. Although the ITS/ITS2 RNA transcript has been proposed as the core barcode for seed plants, it has been criticized for being too conserved in some species to provide enough information or too variable in some species to align it within the different taxa ranks. We selected 30 individuals, representing 16 species and four families, to explore whether ITS2 can successfully resolve species in terms of secondary structure. Secondary structure was predicted using Mfold software and sequence-structure was aligned by MARNA. RNAstat software transformed the secondary structures into 28 symbol code data for maximum parsimony (MP) analysis. The results showed that the ITS2 structures in our samples had a common four-helix folding type with some shared motifs. This conserved structure facilitated the alignment of ambiguous sequences from divergent families. The structure alignment yielded a MP tree, in which most topological relationships were congruent with the tree constructed using nucleotide sequence data. When the data was combined, we obtained a well-resolved and highly supported phylogeny, in which individuals of a same species were clustered together into a monophyletic group. As a result, the different species that are often referred to as the herb "Mu tong" were successfully identified using short fragments of 250 bp ITS2 sequences, together with their secondary structure. Thus our analysis strengthens the potential of ITS2 as a promising DNA barcode because it incorporates valuable secondary structure information that will help improve discrimination between species.

Fig 1. The predicted ITS2 secondary structure of the “Mu tong” taxa.

(a) Consensus secondary structure model of the ITS2 region based on 30 sequences covering four families (Actinidiaceae, Aristolochiaceae, Lardizabalaceae and Ranunculaceae). The four helices, each with a stem–loop, are labeled I–IV. Compatible base pairs are colored and the degree of conservation over the whole alignment is indicated with different degrees of color saturation. (b) One of the example secondary structures of Akebia quinata. (c) The position of the strictly conserved base pair sites (highlight in blue circle) found in the 70% consensus ITS2 secondary structure model.

Fig 2. Comparison of multiple sequence alignments from different methods.

a. alignment from MARNA with secondary structure guiding; b. alignment from Clustal X without secondary structure guiding.

Fig 3. Comparison of MP trees inferred from sequences with different align methods.

a. MP tree inferred from sequence alignment by Clustal X; b. MP tree inferred from sequence alignment by secondary structure information. Numbers on the branches indicate the bootstrap values of MP above 50%, numbers following a species name represent individual numbers.

Fig 4. Comparison of MP trees inferred from different data sets.

a. MP tree inferred from secondary structure coding information; b. MP tree inferred from the combined nucleotide sequences and secondary structure information. Numbers on the branches indicate the bootstrap values of MP above 50%, numbers following a species name represent individual numbers.