Table 1.
X-ray data collection (ESRF) and refinement statistics.
Figure 1.
A) 2x7p is represented with alpha helices in red and beta-sheet in yellow. Secondary structures are numbered along the protein sequence. Carbon dioxide, glycerol and PEG molecules are in ball and sticks with green carbon and red oxygen atoms. 2 Water molecules are represented as blue spheres. B) Surface representation of the open conformation, 2x7p. The strictly conserved residues in the groove are colored in red and residues with conserved properties are marked in yellow. Carbon dioxide, glycerol and carboxyl molecules are in ball and sticks with green carbon and red oxygen atoms. Two water molecules are represented as blue spheres. C) The two conformations (2x7q: cyan, 2x7p: red) are superposed on domain I (bottom). The venus flytrap motion is illustrated by a black arrow (see Movie S1 for an animated view).
Figure 2.
A) Stereo view of the binding site with non polar residues and tyrosine in yellow, polar residues in light blue, acidic residues in red and proline in green sticks. Carbon dioxide, glycerol and PEG molecules are in ball and sticks with green carbon and red oxygen atoms. Two water molecules are represented as blue spheres. (B) Surface representation of the CA3427 structure with a modeled C18 acyl chain (blue ball and stick) fitted in the Fo-Fc electron density map (green) computed on the open conformation structure (1.5σ). The conserved residues in the predicted binding site are colored in red.
Table 2.
Putative ligand-protein interactions.
Table 3.
Average B factor values for the interpreted ligands.
Table 4.
Closest structural homologs of 2X7P and 2X7Q (Dali server).
Figure 3.
Multiple alignment of 29 selected sequences used for phylogeny.
All fungi sequence names are in black except C. albicans (in red) D. hanenii and Y. lipolitica (in green). Other eukaryote sequence names are in magenta, bacterial sequence names are in light blue for Flavobacteria/Bacteroidetes and dark blue for the last 4 bacteria (outgroup). All the sequences are aligned together in one shot but similarity at each position is computed separately for the outgroup and the other sequences except for position 41 and 163 where strict conservation (white letters with red background) is observed. Red or orange letters in the alignment represent similarity (computed using a Risler matrix with a 0.9 threshold in ESPript) within each group. The secondary structure of CA3427 computed with DSSP, is also represented above the alignment. Purple triangles represent the binding site defined in the CA3427 structure. This alignment shows clearly the conservation of the binding site among different eukaryotes and Flavobacteria/Bacteroidetes but not in the outgroup.
Figure 4.
A) Phylogeny of the CA3427 homologs in Fungi. The reference list of fungal species is from [23]. Multiple strains of the same species have been removed for clarity. The CA3427 sequence (in red in all trees) is from C. albicans strain SC5314. This unrooted tree was computed on the Phylogeny.fr web server [30], using the default option of the “advanced mode” w/o Gblocks. The final alignment includes 209 ungapped positions. Branch support estimates are indicated in red, and branches have been collapsed for values <50%. CA3427 homologs were found in all species except for saccharomycetales (in green), and cluster according to [23]. No species from the WGD clade (e.g. Saccharomyces cerevisiae) appear to possess a CA3427-like protein. B) Evolutionary relationships between the bacterial PBPs and their eukaryotic CA3427-like homolog. The phylogenetic analysis includes representative sequences from Cytophaga-Flavobacteria (light blue), Fungi (black and green) (as in Fig. 2), other eukaryotes (in magenta), and more remote bacterial sequences defining and outgroup (dark blue). This tree was computed on Phylogeny.fr web server [30], using the default option of the “advanced mode” without Gblocks. Branch support estimates are indicated in red, and branches have been collapsed for values <50%. The topology of this tree is consistent with the hypothesis that the original CA3427-like PBP was transferred into the eukaryotic gene pool from a cytophagia/flavobacteria into an ancestral opisthokont.
Figure 5.
Parsimonious evolutionary scenario for the presence of CA3427-homologs in modern eukaryotes.
A horizontal transfer (HGT) is proposed to have occurred from a bacteroidetes to an ancestral unicellular heterotrophic eukaryote prior to the divergence of the main branches leading to fungi and animalia. The branching of the various phyla is adapted from [42].