The adaptive landscapes of three global Escherichia coli transcriptional regulators

The evolution of gene regulation is a major source of evolutionary adaptation and innovation, particularly when organisms encounter new or changing environments. Central to this process is the emergence of new transcription factor binding sites (TFBSs). Adaptive landscapes provide a powerful framework to study such emergence by linking regulatory DNA sequences to their transcriptional outputs. Although several landscapes have been characterized for DNA, RNA, and proteins, large-scale in vivo adaptive landscapes for bacterial TFBSs remain scarce. Here, we address this gap by experimentally mapping the first comprehensive in vivo regulatory landscapes for three global transcription factors in Escherichia coli: cAMP receptor protein, Fis, and IHF. Using a massively parallel reporter assay, we quantify the regulation strength of more than 30,000 TFBS variants for each factor, and reconstruct their adaptive landscapes. All three landscapes are highly rugged and exhibit pervasive epistasis, with thousands of local peaks distributed broadly across sequence space. This ruggedness contrasts sharply with the much smoother TFBS landscapes of eukaryotes. It suggests greater constraints on the evolution of prokaryotic gene regulation. Nonetheless, evolutionary simulations show that ~10% of evolving populations can reach a peak of strong regulation, a proportion that is significantly greater than in comparable random landscapes. Adaptive evolution starting from the same DNA sequence can attain different high peaks, and some peaks are reached more frequently than others. Together, our results show that de novo adaptive evolution of new gene regulation in bacteria is feasible, but subject to a blend of chance, historical contingency, and evolutionary biases.