Diverse domain architectures of CheA histidine kinase, a central component of bacterial and archaeal chemosensory systems.

IMPORTANCE: We found that in contrast to the best-studied model organisms, such as Escherichia coli and Bacillus subtilis, most bacterial and archaeal species have a CheA protein with a different domain composition. We report variations in CheA architecture, such as domain duplication and acquisition as well as class-specific domain composition. Our results will be of interest to those working on signal transduction in bacteria and archaea and lay the foundation for experimental studies.

Diverse domain architectures of CheA histidine kinase, a central component of bacterial and archaeal chemosensory systems.

Chemosensory systems in bacteria and archaea are complex, multi-protein pathways that enable rapid cellular responses to environmental changes. The CheA histidine kinase is a central component of chemosensory systems. In contrast to other histidine kinases, it lacks a sensor (input) domain and utilizes dedicated chemoreceptors for sensing. CheA is a multi-domain protein; in model organisms as diverse as Escherichia coli and Bacillus subtilis, it contains five single-copy domains. Deviations from this canonical domain architecture have been reported, however, a broad genome-wide analysis of CheA diversity is lacking. Here, we present results of a genomic survey of CheA domain composition carried out using an unbiased set of thousands of CheA sequences from bacteria and archaea. We found that four out of five canonical CheA domains comprise a minimal functional unit (core domains), as they are present in all surveyed CheA homologs. The most common deviations from a classical five-domain CheA architecture are the lack of a P2/CheY-binding domain, which is missing from more than a half of CheA homologs and the acquisition of a response regulator receiver (CheY-like) domain, which is present in ~35% of CheA homologs. We also document other deviations from classical CheA architecture, including bipartite CheA proteins, domain duplications and fusions, and reveal that phylogenetically defined CheA classes have pre-dominant domain architectures. This study lays a foundation for a better classification of CheA homologs and identifies targets for experimental investigations.