부산대학교 도서관

Animal traits develop through the expression and action of numerous regulatory and realizator genes that comprise a gene regulatory network (GRN). For each GRN, its underlying patterns of gene expression are controlled by cis-regulatory elements (CREs) that bind activating and repressing transcription factors. These interactions drive cell-type and developmental stage-specific transcriptional activation or repression. Most GRNs remain incompletely mapped, and a major barrier to this daunting task is CRE identification. Here, we used an in silico method to identify predicted CREs (pCREs) that comprise the GRN which governs sex-specific pigmentation of Drosophila melanogaster. Through in vivo assays, we demonstrate that many pCREs activate expression in the correct cell-type and developmental stage. We employed genome editing to demonstrate that two CREs control the pupal abdomen expression of trithorax, whose function is required for the dimorphic phenotype. Surprisingly, trithorax had no detectable effect on this GRN's key trans-regulators, but shapes the sex-specific expression of two realizator genes. Comparison of sequences orthologous to these CREs supports an evolutionary scenario where these trithorax CREs predated the origin of the dimorphic trait. Collectively, this study demonstrates how in silico approaches can shed novel insights on the GRN basis for a trait's development and evolution. Author summary: The traits decorating animal bodies form by patterns of gene expression that are controlled by cis-regulatory element sequences. These sequences behave like switches to turn genes on or off during development. The conservation and evolution of such switches are key aspects for many traits' origin. Unfortunately, they remain difficult to find, and their regulated genes and activities remain unknown. Here, the authors focused on a derived pattern of fruit fly pigmentation. They used a small set of known switches to find hundreds of predicted similarly-functioning switches through a computer algorithm that finds patterns in DNA sequences. The authors focused on two predicted switches, and showed how they control the expression of an evolutionarily-conserved regulator of gene expression. In an interesting twist, this regulator does not repeat its embryonic function in the development of this adult trait. Rather it targets two genes that encode pigmentation enzymes. It was found that the expression and switch functions for this regulator pre-date the derived trait, indicating that the gain of these enzyme targets were key events in this trait's origin. This study is noteworthy, as it shows how computer- and animal-based approaches can deepen our understanding of genetic switches in development and evolution. [ABSTRACT FROM AUTHOR]