Phylogenetic Relationships Of Brassicaceae Species Inferred From Various Genomic Sequences

The Brassicaceae or mustard family is a large family. Despite much research, the tribal and genus classification of Brassicaceae has remained problematic and not well understood phylogenetically. Many molecular phylogenetic data have shown that species with similar morphology may be unrelated, whereas species with different morphology may be closely related. This situation, accompanied by questions regarding introgression and lineage sorting, often results in incongruent gene and species trees. Thus, more molecular data are needed to infer phylogenetic relationships in this family. For this purpose, five non-coding DNA sequences (trnS-trnG, trnD-trnT, trnS-rpS4-trnT-trnL, nad7second intron, PI first intron) and one coding sequence (matK) were used to study the Brassicaceae taxa to help verify the phylogenetic relationships of taxa within this family, and some results were obtained.In current analysis, Cleome gynandra, used as the outgroup, was sister to the members of the Brassicaceae family, while Aethionemeae was the "basal" tribe and supported the sister relationship to all other tribes and taxa of Brassicaceae. Most of the tribes analyzed could be grouped into Lineages Ⅰ-Ⅲ (Beilstein et al.,2006) as well as some small monophyletic groups.Lineage Ⅰ included Cardamineae, Camelineae, Descurainieae, Erysimeae, Lepideae, Smelowskieae. Erysimum was separated from other Camelineae genera, which increased the confidence that it is a new tribe. Nasturtium is more closely related to Cardamineae than to Rorippa, suggested that Nasturtium should be treated as an alias of Rorippa was probably unreasonable. Turritis was previously placed in Arabideae, but our analyses determined that it should be included in Camelineae. Most molecular data to date suggest place Coronopus and Cardaria within Lepideae. Arabideae included two subclades, one of which was nested within Lineage I and was closely related to Halimolobeae and Boechereae, while the other did not fit within Lineages Ⅰ-Ⅱ and comprised Arabis, Aubrieta, Draba, and Baimashania.Lineage Ⅱ contained Brassiceae, Isatideae, Sisymbrieae. B. rapa/B. oleracea and Raphanus were combined with B. nigra, supporting that Raphanus was hybridized from B. rapa/B. oleracea and B. nigra. Our analysis supported the exclusion of Conringia from Brassiceae and Conringia should not be recognized as a new tribe and should be placed within the Isatiseae tribe. Pachypterygium and Tauscheria were considered to be included in Isatideae. O. violaceus complex included O. taibaiensis, O. diffusus, O. hupehensis, O.violaceus and should be excluded from Brassiceae.Lineage Ⅲ comprised tribes of Anchonieae, Chorisporeae, Dontostemoneae, Hesperideae, Euclidieae. Oreoloma should be a member of Sterigmostemum. The position of Malcolmieae was unstable according to the DNA makers we used. Hesperideae was defined as a unigeneric tribe and was sister to Dontostemoneae. We found Desideria and Phaeonychium were grouped into Solms-laubach, Desideria was found to be polyphyletic, while all species of Solms-laubachia formed a monophyletic clade.Goldbachia laevigata was assigned to a newly defined tribe, Calepineae, and it was neither related to Thlaspideae nor to Eutremeae. Pugionium formed a monophyletic clade, but its phylogenetic position remains unresolved. Alysseae is polyphyletic and its position was mainly dependent on the maker used.