The study of network evolution is critical to understanding how complex biological processes arise and adapt over time. Protein networks, composed of interacting components, can exhibit varying degrees of conservation and flexibility, enabling organisms to fine-tune their responses to environmental changes. Using the circadian clock system in Drosophila as a case study, we explore how such networks evolve. We leverage the recently published 101 Drosophilidae genome project to analyze the evolution and co-evolution of 11 core clock proteins across 65 species spanning about 60 million years of evolution. A sliding window analysis of coding regions reveals substantial heterogeneity in nucleotide divergence, with Clk and per exhibiting high divergence, whereas Pdp1 and sgg show virtually no evolutionary change. Additionally, we assessed interdependent amino acid evolution across different proteins, identifying 67 co-evolving site pairs, primarily between CLK-PER, CLK-CWO, and SGG-PER. Using codon-based models of evolution, we found four genes (cwo, jet, per, and sgg) showing evidence of positive selection. Since several clock proteins are pleiotropic, we tested whether their multifunctionality influences their evolutionary constraints. Using alternative approaches to assess pleiotropy, we found no significant correlation between pleiotropy and the non-synonymous substitution rate (Ka) in 440 Drosophila proteins, including circadian clock ones. Overall, our findings suggest that the circadian clock network does not impose strong constraints on the evolution of its components. This flexibility may facilitate species-specific adaptation of the clock and allow the pleiotropic functions of clock proteins.
Creasey et al. (Sat,) studied this question.