Table 1.
Data collection and refinement statistics.
Fig 1.
Overall crystal structure of the Zymoseptoria tritici effector Mg1LysM.
Crystal structure model in which a dimer of two Mg1LysM homodimers is shown, with each of the Mg1LysM molecules in a different colour (orange, yellow, red and green). While the two monomers that form a ligand-independent homodimer on the right are represented as a surface model, the two monomers that form a ligand-independent homodimer on the left are represented by ribbons with the (putative) chitin binding sites indicated in blue and the disulfide bridges as yellow sticks. The chitin trimer that mediates the dimerization of two ligand-independent Mg1LysM homodimers is depicted by grey sticks. The two salt bridges between R2 and D12 in the dimer interface on the left are indicated with grey discontinuous lines.
Fig 2.
The chitin binding groove formed by two Mg1LysM protomers.
(A) A chitin trimer (GlcNAc)3, displayed as grey sticks, was identified in a binding pocket formed by two Mg1LysM protomers (indicated in yellow and red, respectively). (B) Representation of the binding pocket from the top. (C) Detail of the chitin binding site. The amino acids involved in direct chitin trimer binding (26GDTLT30 and 56NRI58) are represented with blue sticks and labelled. In addition, K31 and D54 (represented in green) of the two different Mg1LysM protomers form a salt bridge that tightly closes the binding pocket. Grey discontinuous lines represent the salt bridge and the hydrogen bonds between the protomers.
Fig 3.
Two Mg1LysM protomers bind a single chitin hexamer with high affinity.
Isothermal titration calorimetry of (GlcNAc)6 binding by wild-type Mg1LysM produced in E. coli, and the mutants T28R, K31A and D54A. The dissociation constant (Kd) and the stoichiometry (N) of the interactions are indicated.
Fig 4.
Mg1LysM sequence polymorphisms in Z. tritici.
(A) Five non-synonymous SNPs were identified in 149 Z. tritici strains from four different populations. Arrows indicate the position of the residues involved in the formation of salt bridges, while green underlining indicates the signal peptide and red underlining the chitin-binding loops. Red and green underlines indicate the signal peptide and the chitin binding sites, respectively. (B) While the mutations (shown in blue sticks) do not co-localize but occur dispersed over the Mg1LysM protein, none of them is in the chitin-binding site or in the (homo-)dimerization surface. (C) Mg1LysM and the two allelic variants Mg1LysM_1E4 and Mg1LysM-3F4 bind insoluble chitin. All proteins were heterologously produced in E. coli and incubated with chitin for 6 hours. After centrifugation, pellets and supernatants were analysed on polyacrylamide gel followed by CBB staining.
Fig 5.
Mg1LysM mutants are impaired in protection against chitinases.
(A) Trichoderma viride incubated with Mg1LysM with the amine-reactive fluorescent dye BODIPY, or with BODIPY only as a control, for 4 hours and bright field and fluorescence microscopy pictures are shown. (B) Microscopic pictures of Fusarium oxysporum f. sp. lycopersici grown in vitro in the absence or presence of wild-type or mutant Mg1LysM, 4 hours after addition of tomato hydrolytic enzymes (HE) that include chitinases, or water as control.
Fig 6.
Chitin-induced polymerization of Mg1LysM homodimers.
(A) Model inferred from the crystal structure of Mg1LysM in which a continuous structure of Mg1LysM homodimers and chitin is formed. Alternating chitin molecules (in grey sticks) and Mg1LysM homodimers (in red and green), each of them with two chitin-binding sites, are shown. (B) Dynamic light scattering (DLS) heat maps of Mg1LysM, C. fulvum Ecp6 and RiSLM treated with chitohexaose in molar ratios of 1:0, 1:2 and 1:5 (protein: chitohexaose), respectively. The particle size distribution is indicated as a color scale ranging from blue (lowest amount) to red (highest amount) for a particle size range of 1 nm to 100 um.
Fig 7.
Chitin-induced polymerization of Mg1LysM and RiSLM, but not of Ecp6.
Effector proteins were incubated with chitohexaose (chitin) or water as control. After overnight incubation, methylene blue was added and protein solutions were centrifuged, resulting in protein pellets as a consequence of polymerization for Mg1LysM and RiSLM, but not for Ecp6. NP is the ‘no protein’ sample with chitin in the absence of LysM effector protein.