Transcription factors evolve faster than their structural gene targets in the flavonoid pigment pathway

Abstract

Evolutionary transitions in flower color often trace back to changes in the flavonoid biosynthetic pathway and its regulators. In angiosperms, this pathway produces a range of red, purple, and blue anthocyanin pigments. Transcription factor (TF) complexes involving members of the MYB, bHLH, and WD40 protein families control the expression of pathway enzymes. Here, we investigate flavonoid pathway evolution in the Petunieae clade of the tomato family (Solanaceae). Using transcriptomic data from 69 species of Petunieae, we estimated a new phylogeny for the clade. For the 65 species with floral transcriptomes, we retrieved transcripts encoding homologs of 18 enzymes and transcription factors to investigate patterns of evolution across genes and lineages. We found that TFs exhibit faster rates of molecular evolution than their targets, with the highly specialized MYB genes evolving fastest. Using the largest comparative dataset to date, we recovered little support for the hypothesis that upstream enzymes evolve slower than those occupying more downstream positions. However, expression levels inversely correlated with molecular evolutionary rates, while shifts in floral pigmentation were weakly related to changes affecting coding regions. Nevertheless, shifts in floral pigmentation and presence/absence patterns of MYB transcripts are strongly correlated. Intensely pigmented and patterned species express homologs of all three main MYB anthocyanin activators in petals, while pale or white species express few or none. Our findings reinforce the notion that regulators of the flavonoid pathway have a dynamic history, involving higher rates of molecular evolution than structural components, along with frequent changes in expression during color transitions.

Genetic factors explaining anthocyanin pigmentation differences

“…While we cannot rule out the possibility that structural genes can also accumulate mutations prior to the reduction in transcription, this observation is in line with a previous study that observed faster evolutionary changes in transcription factors than in structural genes [95]. Similarly, Wheeler et al [93] showed that transcription factors, particularly MYB, presented lower levels of gene expression with higher molecular evolutionary rate compared to their targeted structural genes, and suggested a negative correlation between evolution rate and gene expression in the Petunieae tribe [93]. This premise commonly known as the E-R anticorrelation, has been widly studied across different organisms, including yeast [104,105], Arabidopsis [106], Brassica [107], Barley [108], Arachis [109],…”

“…A study published by Rausher et al [92] proposed that the evolutionary rate of enzymes depends on their location in a pathway with early genes showing a slower mutation rate, but this has been contradicted recently [93]. According to the Arabidopsis Information Resource (TAIR), the length of the coding sequence (CDS) of DFR is 1149 bp (accession: NM_123645) and of ANS is 1071 bp (accession: NM_118417).…”

“…Studies on evolutionary rates investigating the components of the MBW complex suggested that MYBs would be the most likely component to be lost due to the highest degree of specialization which coincides with a lower pleiotropy [62,113,93]. It was observed that insertions/deletions were the most frequent mutation events in MYBs, while amino acid substitutions in the conserved region appeared irrelevant [61].…”

“…Another study examined the frequency of point mutations in TF genes in Eschericchia coli and found that these genes were more vulnerable to harmful mutations that resulted in significant changes in gene expression than non-TF genes [117]. Given that their activity covers a wide range of functions, transcription factors can regulate and affect the expression of structural genes without the need for high gene expression levels [93]. A recent study by Liang et al…”

“…Upstream enzymes evolve at a slower rate, because they control the flux to latter steps in the pathway [60,121]. This theory states that enzymes positioned at the beginning of a pathway are more likely to have a greater impact on the overall fitness in a species and should display slower changes [122][123][124]. According to the Arabidopsis Information Resource (TAIR), the length of the coding sequence (CDS) of DFR is 1149 bp (accession: NM_123645) and of ANS is 1071 bp (accession: NM_118417).…”