Evolutionary walks through flower colour space driven by gene expression inPetuniaand allies (Petunieae)

Abstract

The structure and function of biochemical and developmental pathways determine the range of accessible phenotypes, which are the substrate for evolutionary change. Accordingly, we expect that observed phenotypic variation across species is strongly influenced by pathway structure, with different phenotypes arising due to changes in activity along pathway branches. Here, we use flower colour as a model to investigate how the structure of pigment pathways shapes the evolution of phenotypic diversity. We focus on the phenotypically diverse Petunieae clade in the nightshade family, which contains ca 180 species of Petunia and related genera, as a model to understand how flavonoid pathway gene expression maps onto pigment production. We use multivariate comparative methods to estimate co-expression relationships between pathway enzymes and transcriptional regulators, and then assess how expression of these genes relates to the major axes of variation in floral pigmentation. Our results indicate that coordinated shifts in gene expression predict transitions in both total anthocyanin levels and pigment type, which, in turn, incur trade-offs with the production of UV-absorbing flavonol compounds. These findings demonstrate that the intrinsic structure of the flavonoid pathway and its regulatory architecture underlies the accessibility of pigment phenotypes and shapes evolutionary outcomes for floral pigment production.

Multiple mechanisms explain loss of anthocyanins from betalain‐pigmented Caryophyllales, including repeated wholesale loss of a key anthocyanidin synthesis enzyme

“…While the expression of early‐acting components such as CHS , CHI , F3H , F3′H , and F3′5′H is modestly reduced, a more marked decrease is seen in late‐acting components including DFR , LAR , ANR , MATE , and AHA10 . This observation aligns with earlier research suggesting that the loss of anthocyanin pigmentation is often linked to cis ‐ and/or trans ‐regulatory changes in enzymatic genes (Shimada et al ., 2004, 2005, 2007; Streisfeld & Rausher, 2011; Hatlestad et al ., 2014; Larter et al ., 2018; Sakuta et al ., 2021; Wheeler et al ., 2022, 2023). Interestingly, not all flavonoid biosynthesis genes follow this trend.…”

“…Anthocyanin‐based floral coloration shows frequent transitions between different hues through changes in concentrations of red pelargonidins, purple cyanidins, and blue delphinidins (Wessinger & Rausher, 2012; Wheeler et al ., 2021). These transitions can be caused by repeated regulatory changes to the same or similar anthocyanin pathway genes (Larter et al ., 2018; Wheeler et al ., 2023). The targets of these changes appear to be constrained by pleiotropy and the structure of the anthocyanin biosynthesis pathway (Streisfeld & Rausher, 2011; Larter et al ., 2018; Wheeler et al ., 2021).…”

“…The targets of these changes appear to be constrained by pleiotropy and the structure of the anthocyanin biosynthesis pathway (Streisfeld & Rausher, 2011; Larter et al ., 2018; Wheeler et al ., 2021). Some lineages have also repeatedly evolved white flowers by loss of anthocyanins (Ho & Smith, 2016; Wheeler et al ., 2023), by diverse mechanisms (Gates et al ., 2018; Duncan & Rausher, 2020), but with predictable molecular evolutionary consequences for genes in the pathway (Ho & Smith, 2016). Interestingly, while some transitions to white flowers involve gene loss or loss of function in anthocyanin pathway genes (Coburn et al ., 2015; Lin et al ., 2022), others seem to involve primarily regulatory changes (Ho & Smith, 2016; Gates et al ., 2018), suggesting that some losses of anthocyanins may be more easily reversible.…”

Incomplete lineage sorting and hybridization as drivers of tree discordance in Petunia and related genera (Petunieae, Solanaceae)

“…We used the raw RNA-seq data derived from Wheeler et al (2022Wheeler et al ( , 2023, focusing on the Petunia-Calibrachoa-Fabiana clade and incorporating six outgroup species. In total, we employed 107 individuals, encompassing 11 Petunia species, 16 Calibrachoa species, 10 Fabiana species, and six outgroups (Table S1).…”

“…In total, we employed 107 individuals, encompassing 11 Petunia species, 16 Calibrachoa species, 10 Fabiana species, and six outgroups (Table S1). Here, we expanded the Wheeler et al (2022) dataset by including additional individuals for most sampled species derived from Wheeler et al (2023), resulting in three individuals per species collected at the same time and location (hereafter referred to as replicates). We corrected the raw RNA-seq reads using Rcorrector (Song and Florea, 2015) and removed adapters using Trimmomatic (Bolger et al 2014).…”