Comparison of archaeal and bacterial genomes: computer analysis of protein sequences predicts novel functions and suggests a chimeric origin for the archaea

Abstract

SummaryProtein sequences encoded in three complete bacterial genomes, those of Haemophilus influenzae, Mycoplasma genitalium and Synechocystis sp., and the first available archaeal genome sequence, that of Methanococcus jannaschii, were analysed using the BLAST 2 algorithm and methods for amino acid motif detection. Between 75% and 90% of the predicted proteins encoded in each of the bacterial genomes and 73% of the M. jannaschii proteins showed significant sequence similarity to proteins from other species. The fraction of bacterial and archaeal proteins containing regions conserved over long phylogenetic distances is nearly the same and close to 70%. Functions of 70-85% of the bacterial proteins and about 70% of the archaeal proteins were predicted with varying precision. This contrasts with the previous report that more than half of the archaeal proteins have no homologues and shows that, with more sensitive methods and detailed analysis of conserved motifs, archaeal genomes become as amenable to meaningful interpretation by computer as bacterial genomes. The analysis of conserved motifs resulted in the prediction of a number of previously undetected functions of bacterial and archaeal proteins and in the identification of novel protein families. In spite of the generally high conservation of protein sequences, orthologues of 25% or less of the M. jannaschii genes were detected in each individual completely sequenced genome, supporting the uniqueness of archaea as a distinct domain of life. About 53% of the M. jannaschii proteins belong to families of paralogues, a fraction similar to that in bacteria with larger genomes, such as Synechocystis sp. and Escherichia coli, but higher than that in H. influenzae, which has approximately the same number of genes as M. jannaschii. Certain groups of proteins, e.g. molecular chaperones and DNA repair enzymes, thought to be ubiquitous and represented in the minimal gene set derived by bacterial genome comparison, are missing in M. jannaschii, indicating massive non-orthologous displacement of genes responsible for essential functions. An unexpectedly large fraction of the M. jannaschii gene products, 44%, shows significantly higher similarity to bacterial than to eukaryotic proteins, compared with 13% that have eukaryotic proteins as their closest homologues (the rest of the proteins show approximately the same level of similarity to bacterial and eukaryotic homologues or have no homologues). Proteins involved in translation, transcription, replication and protein secretion are most closely related to eukaryotic proteins, whereas metabolic enzymes, metabolite uptake systems, enzymes for cell wall biosynthesis and many uncharacterized proteins appear to be 'bacterial'. A similar prevalence of proteins of apparent bacterial origin was observed among the currently available sequences from the distantly related archaeal genus, Sulfolobus. It is likely that the evolution of archaea included at least one major merger between ancestral cells from the bacterial lineage and the ...

Common History at the Origin of the Position-Function Correlation in Transcriptional Regulators in Archaea and Bacteria

“…Previous studies have suggested that the HTH motif of specific Tf's occurs in Archaea as a result of horizontal transfer from Bacteria Koonin et al 1997;Makarova, et al 1999). Our results indicate that the HTH motif of specific Tf's occurs in Bacteria and Archaea as a result of a common origin, except for the few cases just discussed.…”

A diverse superfamily of enzymes with ATP‐dependent carboxylate—amine/thiol ligase activity

“…In the course of detailed comparative analysis of the protein sequences encoded in complete bacterial and archaeal genomes (Koonin et al, 1997), we observed that the Escherichia coli RimK protein, which is involved in post-translational modification of the ribosomal protein S6 (Reeh & Pedersen, 1979;Kang et al, 1989), had highly conserved homologs in all completely sequenced bacterial and archaeal genomes and also showed a significant similarity to GSHases. When the non-redundant protein sequence database at the National Center for Biotechnology Information was searched using BLASTGP program, which is an extension of the BLAST method (Altschul et al, 1990) incorporating statistical analysis of local alignments with gaps (Altschul & Gish, 1996;Altschul et al, 1997), alignments of the RimK sequence with GSHases were detected with a probability of occumng by chance, P < IO-'.…”

Identification and functional analysis of 'hypothetical' genes expressed in Haemophilus influenzae

“…Eight years later, the genomes of over 100 organisms from all major phylogenetic lineages have been sequenced, and sequencing of many more is currently under way (2,3). Disparities in the accuracy of genome annotation that were the subject of many heated discussions at the beginning of the genome era (4,5) are largely gone. Still, the so-called`70% hurdle' (6) holds, as functions of only~50± 70% of the genes in any given genome can be predicted with reasonable con®dence (3,6).…”