Figure 1.
Presence (green)/absence (red) distribution of VEP1 across the reference tree.
The reference tree topology is based on information from various sources, including NCBI taxonomy [49], ‘Tree of Life’ [50], ‘The All-Species Living Tree’ project [51], and TIMETREE [52] (see the Materials and Methods section).
Figure 2.
The tree was inferred from 239 amino acid characters using the empirical replacement matrix of [61], setting amino acid frequencies as free parameters, gamma-distributed rates among sites (4 categories; α = 1.532), and a proportion of invariant sites (I = 0.060), referred to as LG+F+dG+I model. Non-parametric bootstrap (1000 replicates)/aLRT support scores greater than 50% are shown above the respective nodes. Light (right) and dark (left) background areas indicate, respectively, the sequences used for building the tree (identified using tBLASTn; >25% pairwise sequence identity), and the extant closest remote homologs of VEP1 (identified using remote homology searching methods), which were not used for tree building, but are shown to indicate this study's hypothesis about the evolutionary origin of VEP1. Subtrees subtending inferred bacteria-to-eukaryote LGT events are colored green (viridiplantae) and fucsia (fungi). Green and red dots next to the taxa labels indicate plant-associated non-phytopathogenic and phytopathogenic bacteria, respectively. α, β, γ, and δ denote Alpha-, Beta-, Gamma, and Epsilon-proteobacteria, respectively; Ac, Actinobacteria; Ba, Bacteroidetes; Ch, Chloroflexi; Fi, Firmicutes.
Figure 3.
LGT scenarios for a) bacterial cluster I; b) bacterial cluster IIa; c) bacterial cluster IIb.
The direction of LGT was inferred with the LGT-detection tool of the T-REX suite [65] adopting the bipartition dissimilarity optimization criterion. Non-parametric bootstrap (1000 replicates) scores are indicated near to the numbers (encircled) of the corresponding LGTs. Solid arrows denote inferred probable LGTs (bootstrap score >40%), and dashed arrows indicate possible LGTs (bootstrap score <40%). In bold are taxa inferred not to have obtained VEP1 through LGT. Numbers in parentheses next to taxon labels denote VEP1 copies in the corresponding cluster. For example, Methylobacterium radiotolerans has two VEP1 genes, the first (1) in cluster I (panel 3a), and the second (2) in cluster IIa (panel 3b), with the two copies acquired via LGT; Alphaproteobacterium BAL199 has three VEP1 genes (1, 2, 3), all in cluster IIb (panel 3b), of which gene number 3 was acquired via LGT. α, β, γ, and δ denote Alpha-, Beta-, Gamma-, and Epsilon-proteobacteria, respectively; Ac, Actinobacteria; Ba, Bacteroidetes; Ch, Chloroflexi; Fi, Firmicutes; Emb, Embryophytes.
Table 1.
Top 10 closest VEP1 remote homologs.
Figure 4.
Comparative structural analysis of VEP1.
a) Lesk-Hubbard plot of number of residue correspondences vs. RMSD for VEP1 and each of six least redundant extended SDR structures in Table 1. Each color denotes a structure with PDB code and protein name as follows: red: 2v6g-A, VEP1; dark blue: 2c20-A, UDP-glucose 4-epimerase; medium blue: 1bsv-A, GDP-fucose synthetase; light blue: 2pk3-A, GDP-6-deoxy-D-lyxo-4-hexulose reductase; dark green: 1orr-A, CDP-tyvelose 2-epimerase; medium green: 1rkx-C, CDP-glucose 4,6-dehydratase; and light green: 2c59-A, GDP-mannose 3,5-epimerase. b) Ribbon diagram of the VEP1 (2v6g) structure showing the distribution of residues scoring below and above the sieving RMSD in the Lesk-Hubbard plot. The conserved core is colored red (α helices) and green (β strands). The variable regions are colored in grey. The nucleotide cofactor (NADP) is drawn in ball-and-stick representation.
Figure 5.
VEP1 (2v6f) amino acid primary sequence, secondary structural elements including α helices (arrows) and β strands (boxes), and motif logos for 10 structural/functional motifs (motifs 1–10) discussed in the text.
In the primary sequence, motifs are colored red, and red residues outside motifs denote complete evolutionary conservation; the structurally conserved core in the MUSTANG-MR analysis is underlined; white/black backgrounds denote Rossmann dinucleotide-binding/substrate-binding domains, respectively. Secondary structural elements are labeled as in [47]). Motif logos were derived from the 81 sequences MSA of this study. In motif logos, green denotes a polar residue, red a hydrophobic residue, cyan a basic residues, and blue an acidic residue; arrow points denote the direction of replacements at critical sites if VEP1 arose as depicted in Figure 2. Roman numerals next to motif logos denote I: embryophytes and bacterial cluster I; IIa: fungi, trebouxiophytes, and bacterial cluster IIa; and IIb: bacterial cluster IIb. Motifs 6 and 9 are newly described in this study.