Selection of a de novo gene that promotes survival of E. coli by modulating protein homeostasis pathways

Cells sometimes adapt to challenging environments by turning non-functional loci into functional genes in a process termed de novo gene birth. But how proteins with random amino acid sequences integrate into existing cellular pathways to provide a benefit remains poorly understood. Here, we screened a library of random genes for their ability to rescue growth arrest of E. coli cells producing the ribonuclease toxin MazF. Nearly 2,000 random genes could promote growth by reducing transcription from the promoter driving mazF expression. Additionally, one gene encoded a protein, named Random antitoxin of MazF (RamF), that was well-tolerated in cells and neutralized MazF by interacting with chaperones, leading to MazF proteolysis. We show that the specificity of RamF for MazF relative to other toxins relies on the degron-like function of MazF first 10 amino acids. Finally, we demonstrate that random proteins can easily improve during evolution by identifying beneficial mutations that turned RamF into a more efficient inhibitor. Collectively, our work provides a mechanistic basis for how de novo gene birth can produce beneficial new proteins that are integrated into complex cellular systems.