Phylogenomics of prokaryotic ribosomal proteins

Abstract

Archaeal and bacterial ribosomes contain more than 50 proteins, including 34 that are universally conserved in the three domains of cellular life (bacteria, archaea, and eukaryotes). Despite the high sequence conservation, annotation of ribosomal (r-) protein genes is often difficult because of their short lengths and biased sequence composition. We developed an automated computational pipeline for identification of r-protein genes and applied it to 995 completely sequenced bacterial and 87 archaeal genomes available in the RefSeq database. The pipeline employs curated seed alignments of r-proteins to run position-specific scoring matrix (PSSM)-based BLAST searches against six-frame genome translations, mitigating possible gene annotation errors. As a result of this analysis, we performed a census of prokaryotic r-protein complements, enumerated missing and paralogous r-proteins, and analyzed the distributions of ribosomal protein genes among chromosomal partitions. Phyletic patterns of bacterial and archaeal r-protein genes were mapped to phylogenetic trees reconstructed from concatenated alignments of r-proteins to reveal the history of likely multiple independent gains and losses. These alignments, available for download, can be used as search profiles to improve genome annotation of r-proteins and for further comparative genomics studies.

Figure 1. Overall scheme of the procedure.

Figure 2. Bacterial phylogenetic tree reconstructed from a concatenated alignment of 50 nearly ubiquitous r-proteins.

Green boxes denoted as I, II, and III mark three putative “megaphyla” discussed in the text. Branches having bootstrap support values less than 0.5 were collapsed.

Figure 3. Phyletic patterns of r-proteins that are placed in the last common ancestor of Bacteria by Dollo parsimony.

Proteins having full phyletic pattern are listed in black font. Losses are marked by a strikethrough font. Numbers in parentheses following taxonomic group names represent number of species in that group. Numbers in parentheses following r-protein names represent number of species on that branch that have lost this r-protein.

Figure 4. Phyletic distribution of non-ubiquitous bacterial r-proteins that according to the parsimony reconstruction do not appear to be ancestral.

Numbers in parentheses following taxonomic group names represent number of species in that group. Numbers in parentheses following r-protein names represent number of species on that branch that have this r-protein. L7ae on branch means all genomes of this branch included in the dataset have this r-protein. Asterisks point on two phyla where S31 protein has been found in genomes that were not included in the dataset.

Figure 5. Archaeal phylogenetic tree reconstructed from a concatenated alignment of 56 ubiquitous r-proteins.

Figure 6. Phyletic distribution of twelve non-ubiquitous archaeal r-proteins.

Numbers in parentheses and strikethrough font mean the same as on Figure 3.

Figure 7. Three-domain phylogenetic tree reconstructed from a concatenated alignment of 32 universal r-proteins.

Branches having bootstrap support values less than 0.5 were collapsed.