Phyletic Distribution and Lineage-Specific Domain Architectures of Archaeal Two-Component Signal Transduction Systems

Abstract

The two-component signal transduction (TCS) machinery is a key mechanism of sensing environmental changes in the prokaryotic world. TCS systems have been characterized thoroughly in bacteria but to a much lesser extent in archaea. Here, we provide an updated census of more than 2,000 histidine kinases and response regulators encoded in 218 complete archaeal genomes, as well as unfinished genomes available from metagenomic data. We describe the domain architectures of the archaeal TCS components, including several novel output domains, and discuss the evolution of the archaeal TCS machinery. The distribution of TCS systems in archaea is strongly biased, with high levels of abundance in haloarchaea and thaumarchaea but none detected in the sequenced genomes from the phyla Crenarchaeota, Nanoarchaeota, and Korarchaeota The archaeal sensor histidine kinases are generally similar to their well-studied bacterial counterparts but are often located in the cytoplasm and carry multiple PAS and/or GAF domains. In contrast, archaeal response regulators differ dramatically from the bacterial ones. Most archaeal genomes do not encode any of the major classes of bacterial response regulators, such as the DNA-binding transcriptional regulators of the OmpR/PhoB, NarL/FixJ, NtrC, AgrA/LytR, and ActR/PrrA families and the response regulators with GGDEF and/or EAL output domains. Instead, archaea encode multiple copies of response regulators containing either the stand-alone receiver (REC) domain or combinations of REC with PAS and/or GAF domains. Therefore, the prevailing mechanism of archaeal TCS signaling appears to be via a variety of protein-protein interactions, rather than direct transcriptional regulation.IMPORTANCE Although the Archaea represent a separate domain of life, their signaling systems have been assumed to be closely similar to the bacterial ones. A study of the domain architectures of the archaeal two-component signal transduction (TCS) machinery revealed an overall similarity of archaeal and bacterial sensory modules but substantial differences in the signal output modules. The prevailing mechanism of archaeal TCS signaling appears to involve various protein-protein interactions rather than direct transcription regulation. The complete list of histidine kinases and response regulators encoded in the analyzed archaeal genomes is available online at http://www.ncbi.nlm.nih.gov/Complete_Genomes/TCSarchaea.html.

Keywords: Archaea; arCOGs; archaeal genomes; gene neighborhoods; genome analysis; genomics; halobacterium; histidine kinase; membrane proteins; metagenomics; methanogens; protein-protein interactions; signal transduction; two-component regulatory systems.

FIG 1

Census of the archaeal two-component signal transduction systems. (A) Total numbers of sensor histidine kinases (HKs) and response regulators (RRs) encoded in 218 archaeal genomes. (B) Ratios of histidine kinases and response regulators in various archaea. Squares represent HKs, and circles represent RRs; symbols representing data for halobacteria are in orange and red, those for methanogens are in brown, and those for thaumarchaea are in blue (9). The small blue dots indicate RR/HK ratios for individual bacterial genomes.

FIG 2

Principal classes of archaeal response regulators. The detailed data are available in Tables S1 and S2 and online at http://www.ncbi.nlm.nih.gov/Complete_Genomes/TCSarchaea.html.

FIG 3

Sequence conservation and domain architectures of HalOD1s. (A) Sequence logo generated by the WebLogo program (77) from an alignment obtained by PSI-BLAST run using as query the sequence of halovirus HRTV-8 protein 1 (GenBank accession number AGM10749; UniProt ID R4T552). The first position of the logo corresponds to Arg21 of HRTV8-1 and to Glu161 of Haloferax volcanii response regulator HVO_2306 (GenBank accession number ADE02288; UniProt ID D4GWD4). Secondary structure prediction (cylinders indicate α-helices, and arrows indicate β-strands) was produced by JPred (84). (B) Domain architectures of selected HalOD1-containing proteins, listed under their locus names and GenBank accession numbers. iKaiC, divergent and possibly inactivated ATPase domain of the KaiC superfamily. The domain architectures for each sector are shown only for comparison and are not scaled to size.