New Insights Into the Backbone Phylogeny and Character Evolution of Corydalis (Papaveraceae) Based on Plastome Data

Abstract

A robust backbone phylogeny is fundamental for developing a stable classification and is instructive for further research. However, it was still not available for Corydalis DC., a species-rich (> 500 species), ecologically and medically important, but taxonomically notoriously difficult genus. Here, we constructed backbone phylogeny and estimated the divergence of Corydalis based on the plastome data from 39 Corydalis species (32 newly sequenced), which represent ca. 80% of sections and series across this genus. Our phylogenetic analyses recovered six fully supported main clades (I–VI) and provided full support for the majority of lineages within Corydalis. Section Archaeocapnos was unexpectedly turned out to be sister to the rest of the subg. Corydalis s. l. (clades IV–VI), thus treating as a distinct clade (clade III) to render all the main clades monophyletic. Additionally, some unusual plastome structural rearrangements were constantly detected within Corydalis and were proven to be lineage-specific in this study, which, in turn, provided further support to our phylogeny. A segment containing five genes (trnV-UAC–rbcL) in the plastome's LSC region was either normally located downstream of the ndhC gene in clade I species or translocated downstream of the atpH gene in clade II species or translocated to downstream of the trnK-UUU gene in clade III–VI species. The unique large inversion (ca. 50 kb) in the plastome LSC region of clade III species, representing an intermediate stage of the above translocation in clades IV–VI, firmly supported clade III as a distinct and early diverged clade within this large lineage (clades III–VI). Our phylogeny contradicted substantially with the morphology-based taxonomy, rejected the treatment of tuberous species as an independent evolutionary group, and proved that some commonly used diagnostic characters (e.g., root and rhizome) were results of convergent evolution, suggestive of unreliability in Corydalis. We dated the origin of crown Corydalis to the early Eocene (crown age 49.08 Ma) and revealed possible explosive radiation around 25 Ma, coinciding with the drastic uplift of the Qinghai-Tibetan Plateau in Oligocene and Miocene. This study provided the most reliable and robust backbone phylogeny of Corydalis to date and shed some new insights on the evolution of Corydalis.

An updated classification for the hyper‐diverse genusCorydalis(Papaveraceae: Fumarioideae) based on phylogenomic and morphological evidence

“…Based on the high-quality data sets of 65 plastome CDS and 152 low-copy nuclear genes, we reconstructed the most complete phylogeny of Corydalis, including all sections and "series", which resulted in the most robust backbone relationships available to date. This contrasts with previous phylogenetic studies based on a few plastid regions and limited taxonomic sampling Wang, 2006;Zhang et al, 2016;Jiang et al, 2018;Li, 2018;Xu et al, 2022). Both nuclear and plastid data provided well resolved trees, but some topological incongruences were evident, primarily concerning branches with low support.…”

“…With 65 CDS of 280 plastomes (including 273 newly sequenced) and 152 low-copy nuclear genes of 278 samples (including 276 newly sequenced) of Corydalis, our analysis indicated that the genus is monophyletic and its sister group is the tribe Fumarieae represented by two genera in the plastid data and by seven genera in the nuclear data. The monophyly of Corydalis is congruent with previous studies based on a few plastid fragments and smaller taxonomic sampling Wang, 2006;Perez-Gutierrez et al, 2015;Xu and Wang, 2021;Xu et al, 2022). However, the larger taxonomic sampling in our study resulted in better support than in previous studies.…”

“…With the plastome data of six species representing all three subgenera of Corydalis , Xu and Wang (2021) confirmed that the genus is monophyletic. Xu et al (2022) used 64 shared plastome protein‐coding sequence (CDS) regions from 39 species of Corydalis covering 31 sections and 4 “series” for phylogenetic analysis, but the monophyly of sections and “series” recognized by Zhang et al (2008) were only partly verified. Overall, data from Sanger sequencing has been shown to be highly effective in elucidating phylogenetic relationships in Corydalis , but some key sections and more than half of the species of Corydalis were not sampled and too few nuclear and chloroplast makers were available in the previous studies (Lidén et al, 1997; Wang, 2006; Zhang et al, 2016; Jiang et al, 2018; Li, 2018; Xu and Wang, 2018) to recover well supported phylogenetic relationships to underpin a robust classification.…”

“…Plastomes have been shown to be powerful tools for exploring phylogenetic relationships and resolving uncertain relationships at different taxonomic levels (Xu et al, 2019; Zhang et al, 2019; Li et al, 2021; Xu et al, 2022). The Angiosperms353 gene set (AGS), consisting of a set of 353 universal low‐copy nuclear genes, is increasingly being applied to infer the phylogenetic relationships of different lineages of angiosperms at multiple taxonomic scales (Johnson et al, 2019; Baker et al, 2022; Gagnon et al, 2022; Giaretta et al, 2022; Kandziora et al, 2022; Simoes et al, 2022; Zhou et al, 2022).…”

Integrating multiplicate data: A new trend for taxonomic study

“…Taxonomists have long considered the genus as a great challenge to classify because of its great variability. Early research proposed a classification system at section level based on only morphological evidence (Wu et al, 1996(Wu et al, , 1999, and recent molecular studies, were based on only a few markers and limited sampling (Wang, 2006;Zhang et al, 2008;Xu et al, 2022). These studies show both have indicated that the genetic relationships between most sections and species are unclear.…”