Back

On/off switches in the DIVARICATA-based regulatory network evolved through gene duplication, fusion, and truncation.

Sengupta, A.; Howarth, D.

2025-06-07 evolutionary biology
10.1101/2025.06.04.657860 bioRxiv
Show abstract

DIVARICATA (DIV) genes are MYB genes involved in many developmental processes in flowering plants. DIV-based regulatory networks involve paralogs of DIV (LFG and RADIALIS) that compete with DIV for either DNA or protein targets (like, the cofactor DRIF which is a distant paralog of DIV). These interactions act as on/off switches defining contrasting phenotypes. DIV proteins have multiple MYB domains; RADIALIS, LFG, and DRIF each have one. We determined the origin of these genes through Bayesian phylogenetic reconstruction. We report that DIV genes are a fusion of two simpler MYB genes. The first gene had two MYB domains (MYBA-MYB1), and the other gene had one MYB domain and a non-MYB HDI domain (MYB2-HDI). The ancestral DIV genes, hence, had four domains in the following sequence: MYBA-MYB1-MYB2-HDI. The MYBA domain was later lost, likely through a non-gradual truncation event leading to the following configuration in later-diverging DIV: MYB1-MYB2- HDI. The MYBA and MYB2 domains were derived from the SHAQKY clade. The MYB1 of DIV and the MYBD domains of DRIF were derived from the clade associated with the SANT2 domain of ZUO1/ZRF genes. LFG genes evolved from DIV by truncation of the MYB1 domain and the gain of repressor domains. We discuss how truncation of multi-domain DIV into RADIALIS and LFG was recruited towards on/off switches. Components of the DIV-based regulatory network, or their homologs, are present in a diversity of eukaryotes suggesting that their interaction may be ancestral to a large group of eukaryotes.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.