Resolving the phylogeny of birch (Betula L.) using both nucleotide sequence and secondary structure of highly polymorphic ITS1 and ITS2 barcoding loci

Abstract

Birches (Betula L.) are common pioneer trees and shrubs of temperate and boreal zones in the Northern Hemisphere. Most of them are adapted to low temperatures and able to grow under different extreme conditions and on very poor soils, frequently as the first colonizing woody plants. But despite their important roles in ecosystems, birches remain heavily understudied. The taxonomy and phylogeny of the Betula L. genus remain unresolved and are very difficult to assess due to several factors, especially because of frequent hybridization and introgression among different species. Therefore, the number of species recognized within the genus varies from 30 to 120 in the world flora. In the current study, we used nucleotide sequences of two internal transcribed spacer regions (223 ITS1 and 263 ITS2 loci retrieved from NCBI GenBank which represented 73 and 78 birch taxa + original sequences for 2 endemic birch species B. borysthenica Klokov and B. klokovii Zaverucha from Ukraine and dark-barked birches related to B. pendula Roth and B. pubescens Ehrh.), which are commonly used as phylogenetic markers. In addition to the analysis of their nucleotide variation, we reconstructed their secondary structure and used it to resolve phylogenetic relationships of some birch species. We revealed the common pattern for secondary structure across the whole genus and explored whether consideration of secondary structure in phylogenetic analyses based on neighbor-joining, maximum parsimony, maximum likelihood, and Bayesian inference methods would help us obtain more solid support of the reconstructed phylogenetic trees. The phylogenetic trees generated using different methods were mostly in agreement with each other. Species-specific structural variants were detected for B. insignis Franch. and B. nigra L. In combination with sequence variation, it provides higher confidence in clustering, but only when the number of iterations is relatively low or moderate. With increasing the number of iterations for Bayesian Inference the number of supported clusters and support values for the tree considering secondary structure remain more or less the same, while for the tree without considering secondary structure the number of supported clusters and support values tend to increase. However, the resolving power of these markers is still insufficient to reliably discriminate some closely related species as well as even some subsections and sections. There is a need for the application of modern genomic approaches in combination with traditional ones to resolve this genus more reliably. ITS secondary structure reconstruction to resolve taxonomy and phylogeny of the Betula L. genus.