Tandem-repeat protein domains across the tree of life

Abstract

Tandem-repeat protein domains, composed of repeated units of conserved stretches of 20-40 amino acids, are required for a wide array of biological functions. Despite their diverse and fundamental functions, there has been no comprehensive assessment of their taxonomic distribution, incidence, and associations with organismal lifestyle and phylogeny. In this study, we assess for the first time the abundance of armadillo (ARM) and tetratricopeptide (TPR) repeat domains across all three domains in the tree of life and compare the results to our previous analysis on ankyrin (ANK) repeat domains in this journal. All eukaryotes and a majority of the bacterial and archaeal genomes analyzed have a minimum of one TPR and ARM repeat. In eukaryotes, the fraction of ARM-containing proteins is approximately double that of TPR and ANK-containing proteins, whereas bacteria and archaea are enriched in TPR-containing proteins relative to ARM- and ANK-containing proteins. We show in bacteria that phylogenetic history, rather than lifestyle or pathogenicity, is a predictor of TPR repeat domain abundance, while neither phylogenetic history nor lifestyle predicts ARM repeat domain abundance. Surprisingly, pathogenic bacteria were not enriched in TPR-containing proteins, which have been associated within virulence factors in certain species. Taken together, this comparative analysis provides a newly appreciated view of the prevalence and diversity of multiple types of tandem-repeat protein domains across the tree of life. A central finding of this analysis is that tandem repeat domain-containing proteins are prevalent not just in eukaryotes, but also in bacterial and archaeal species.

Keywords: Ankyrin repeat domains; Archaea; Armadillo repeat domain; Bacteria; Protein domains; Symbiosis; Tetratricopeptide repeat domains.

Figure 1. ANK, TPR, and ARM repeat structure and incidence across all domains of life.

(A) Schematics of the structures of the ANK (Mosavi et al., 2004), TPR (Main et al., 2005), and ARM (Parmeggiani et al., 2008) repeats. (B) Table indicating the percent of species in each taxonomic domain that contain ≥ 1 TPR, ARM, or ANK-containing protein.

Figure 2. ANK, TPR, and ARM-containing proteins analysis across all domains of life.

Bar graphs of the average percent of the species in the domains. An average percent of ANK, TPR, or ARM-containing proteins for all organisms of the same species was used for these analyses. (A) Eukarya, (B) Bacteria, and (C) Archaea that have one or more ANK, TPR, and ARM-containing proteins. The average percent of protein domain composition is listed below the graph. (A) Mann Whitney U-test p < 0.0001 for all comparisons, ANOVA p = 5.79E-40. (B) Mann Whitney U-test p < 0.0001 for all comparisons, ANOVA p = 2.3E-203. (C) Mann Whitney U-test, p < 0.005 for all comparisons, ANOVA p = 9.85E-07. Error bars represent standard error.

Figure 3. Correlation analysis.

(A) Percent protein composition of TPR, ARM, or ANK-containing proteins in bacterial species and (B) Number of TPR, ARM, or ANK-containing proteins in bacterial species.

Figure 4. TPR, ANK, and ARM-containing proteins analysis across bacterial classes.

Bar graph of the percent of species in bacterial classes that contain ≥ 1 TPR, ARM, or ANK-containing proteins (only classes with 5 or more represented species were included in this analysis). Order of classes reflect phylogenetic relationships depicted in Fig. 5A.

Figure 5. Test for phylogenetic independence of enrichment of ARM, ANK or TPR-containing proteins in bacteria.

(A) Consensus phylogeny of 16S rRNA sequences from one species (randomly selected) in each class. (B) The percent of species per class that fall within the top 15% most enriched species in the dataset. Only classes with 5 or more represented species were included in this analysis. (C) Spearman correlation analysis of the percent of species within bacterial classes enriched in each domain.

Figure 6. Lifestyle analysis of bacterial species enriched for ARM-containing proteins.

(A-C) Error bars represent standard error and ^∗P < 0.05, ^∗∗P < 0.01, ^∗∗P < 0.0001. (A) Bar graph of the average number of ARM-containing proteins in species of free-living (FL), facultative host-associated (FHA) and obligate intracellular (O) bacteria for those species with at least 0.2% of their proteome composed of ARM-containing proteins. (2-tailed Mann-Whitney U; ANOVA, P = 0.00005). (B) Bar graph of the average percent of the proteome composed of ARM-containing proteins in species of FL, FHA and O bacteria with at least 0.2% of their proteome composed of ARM-containing proteins. (All comparisons, P > 0.05, 2-tailed Mann-Whitney U; ANOVA, P = 0.446). (C) Bar graph of the average number of proteins in the proteomes of species of FL, FHA and O bacteria with at least 0.2% of their proteome composed of ARM-containing proteins (2-tailed Mann-Whitney U; ANOVA, P = 1.98E-10).