Fig. 3
From: 14-3-3 proteins activate Pseudomonas exotoxins-S and -T by chaperoning a hydrophobic surface
A second LDLA-box is capable of binding the second 14-3-3 protomer. a Sequence alignment of 14-3-3-dependent toxins ExoS, ExoT, AexT, and VopT and, for comparison, the 14-3-3 independent Vis toxin. Presence of LDLA-box 1 (green arrowheads) unites all 14-3-3-dependent toxins; several also feature a second LDLA-box (orange arrowheads) at their very C-terminus. b ExoS LDLA-box 1 binding to the amphipathic groove of 14-3-3β and comparison to phosphopeptide binding at the same site (PDB entry 4O46). c ExoS LDLA-box-1 and -2 binding to the two 14-3-3β protomers. d Anisotropy of GFP-tagged ExoS constructs indicated tight binding of the ExoS ART domain (residues 233–453) to 14-3-3β. Truncation of LDLA-box 2 (residues 233–435) led to a slight reduction in apparent affinity. Truncation of both LDLA-boxes resulted in a protein (residues 233–419) that bound 14-3-3β with sub-micromolar affinity using only the hydrophobic binding site (Fig. 2). e 14-3-3 concentration-dependent activation of ExoS233-453 activity (means ± s.e.m.; n = 2), confirming the sub-micromolar affinity measured by fluorescence anisotropy (d). See Supplementary Fig. 1 for further details. f Yeast toxicity assay of ExoS. Five-fold serial dilutions of yeast cells spotted on agar containing either glucose (target gene expression repressed) or galactose/raffinose (expression induced). Expression of the ExoS ADP-ribosyltransferase domain is highly toxic, unless the C-terminal segment ExoS420-453, containing both LDLA-boxes, is absent
