Phylogenetic relationships of the mimosoid Ingeae–Acacia complex (Fabaceae), based on plastid and nuclear data

Modern phylogenetic analyses based on molecular data have shown that the current classification of the Ingeae tribe (Caesalpinioideae, Fabaceae) do not always reflect evolutionary relationships. Many genera that are currently defined based on morphological characters do not constitute monophyletic groups and generic relationships within the tribe are not fully known. Moreover, the Ingeae tribe itself does not comprise a monophyletic group since the genus Acacia (tribe Acacieae) is nested among the ingoid species. We investigated phylogenetic relationships of the Ingeae−Acacia complex on a generic level. We put special emphasis on genera with a taxonomic history in Calliandra s.l., i.e. Calliandra s.s., Afrocalliandra, Sanjappa, Viguieranthus and Zapoteca, and also the genus Faidherbia, which has been shown to be closely related to these genera in previous phylogenetic studies. We included samples from 32 ingoid genera, samples of Acacia and Senegalia (tribe Acacieae), and Vachellia farnesiana, which was used for rooting the trees. Datasets included the nuclear ETS and ITS, and the plastid matK, trnL-trnF and ycf1 DNA regions and were analysed with Bayesian inference and Ultrafast Bootstrap. The results show that the species of the Ingeae−Acacia complex are resolved in three major clades. Calliandra s.s. together with Afrocalliandra (Clade 1) is the sister to all remaining species of Ingeae and Acacia (Clades 2 and 3). Clade 2 comprises Faidherbia, Sanjappa, Thailentadopsis, Viguieranthus and Zapoteca. Clade 2 and Clade 3 are sisters, and thus the genera of Clade 2 are more closely related to each other and to the taxa of Clade 3, than to Calliandra s.s. and Afrocalliandra (Clade 1). The genera of Clades 1 and 2 all have the same type of pod, the typical “Calliandra-pod”, with Faidherbia being the only exception. Our results show that the “Calliandra-pod” is not a sign of evolutionary relationship within the Ingeae-Acacia complex. Rather this character evolved independently several times, also in clades outside of the Ingeae−Acacia complex. Another example is seen in the leaf structure within the Ingeae−Acacia complex. The possession of pinnate leaves defines the genus Inga (Clade 3) but is also present in the other two major clades (Clades 1 and 2) within the Ingeae−Acacia complex. It seems like the occurrence of pinnate leaves evolved at least three times within the Ingeae-Acacia complex. Another possibility is that the possession of pinnate leaves represents the ancestral state. However, within the mimosoids bipinnate leaves is the most common leaf morphology and the logical conclusion is that bipinnate leaves is the ancestral state. The most common polyad structure within the Ingeae−Acacia complex is 16-grained acalymmate polyads, but exceptions to this is present in Calliandra s.s., Afrocalliandra, Faidherbia and Acacia. Moreover, Calliandra s.s. is the only genus with calymmate polyads. Polyad structure could be useful for the generic classification of the Ingeae−Acacia complex but needs to be combined with other morphological characters. It seems likely that the ancestral state of polyad structure within the Ingeae-Acacia complex is acalymmate polyads and thus that the calymmate polyads seen in Calliandra are secondarily derived. For supplementary material see files attached to this record.