Tuning the Electron Transfer Activity of a 2Fe2S Ferredoxin Using Homologous Recombination and an Escherichia coli Selection

Iron-sulfur (Fe-S) cluster containing ferredoxins (Fd) with high sequence identity (>90%) can vary in their specificity for partner proteins. However, we cannot yet predict a priori how Fd sequence relates to partner specificity. To improve our understanding of the sequence-structure-redox relationships that dictate Fd functions, we are recombining distantly-related plant-type 2Fe2S Fd that display ~53% sequence identity (Mastigocladus laminosus and Prochlorococcus cyanomyophage P-SSM2 Fd) and analyzing the cellular functions of chimeras. Ferredoxin function is being quantified by analyzing the complementation of an Escherichia coli sulfide auxotroph, which requires electron transfer through a linear redox pathway that is made up of Fd-NADP⁺ reductase, Fd, and sulfite reductase. To quantify the relative activity of different Fd chimeras in a cellular setting, we use an anhydrotetracyline (aTc) inducible promoter to express each variant and compare the aTc concentrations that yield half maximal complementation. Among the thirty chimeras created by recombining, we have discovered chimeras that require low levels of aTc (2.5 ng/mL) to complement bacterial growth half maximally, like M. laminosus Fd, while other chimeras only partially complement bacterial growth at high levels of aTc (250 ng/mL), and have also observed a wide range of growth rates among the chimeras. Ongoing efforts are investigating whether these differences in cellular activity arise because of changes in cellular expression, protein stability, cofactor binding, redox potential, partner binding, and partner allosteric interactions. By comparing the complementation of dozens of Fd, our results will provide fundamental insight into Fd sequence-structure-redox relationships, which are currently needed to inform the construction of orthogonal redox circuits that can operate alongside host cell redox circuits. In addition, they will allow us to improve protein design models by calibrating how Fd redox activity relates to structural disruption calculated using biophysical models.