Figure 1.
Ultrastructure of MtDef4-treated cells of Fusarium graminearum.
Cells were treated with 3µM MtDef4 for three hrs. Scale bars = 1 µm. (A) Cells were a mixed population with dead and live cells (arrow) both present. (B) Two adjacent cells in a hypha with the left cell in early stages of MtDef4-mediatd cell degradation. The cortical cytoplasm is separating from the cell wall (arrows) and is more electron dense than the healthier cell on the right.
Figure 2.
Immunogold detection of MtDef4 in treated cells (3 µm, 3 hours) of F.
graminearum.
Scale bar = 1 µm. Section was not post-stained. Of the four cells shown in this section, the two dead cells (arrows) have significantly higher cytoplasmic labeling than the two living cells.
Figure 3.
Immunogold label distribution in MtDef4-treated cells.
Scale bars = 250 nm (A, B), 1 µm (C). A, B are of post-stained sections, C is not.
(A) In dead cells cytoplasmic label is associated with electron dense aggregated cytoplasm of undetermined cellular structure. Vacuole (asterisk) does not label. In these cells cytoplasmic label density is 95 particles per µm2 while that of cell wall is 312 particles per µm2.
(B) Cells not yet killed by MtDef4 show a small amount of MtDef4 in the cytoplasm (circled gold particles) but much more in the cell wall.
(C) Control cells treated with water alone are not labeled (two gold particles circled are background). Label density over the cytoplasm was the same as on resin alone, 0.04 particles per µm2 (+ /-) 0.04).
Figure 4.
Number of gold particles found in the cytoplasm and cell walls of live and dead cells of F.
graminearum.
The error bars represent the standard errors of six replicates.
Figure 5.
A. Backbone superposition of the top 20 refined MtDef4 structures. Shown here are only the backbone atoms of the various structures colored from blue to red as the chains extend from the amino-terminus to carboxy-terminus.
B. Superposition of the top 20 structures showing all of the atoms in the models. The orientation and coloration is the same as in A.
C. Ribbon diagram of the MtDef4 mean structure in the same color and orientation as above. The disulfide bonds are noted by arrows and the amino- and carboxy-termini are labeled.
Figure 6.
Structural homology between MtDef4 and other plant defensins.
A. Shown here is the structural alignment of MtDef4 and the plant defensin Psd1 from Pisum sativum (PDB code 1JKZ) in green and red, respectively. The cysteine residues involved in the four disulfide bonds are highlighted in lighter hues.
B. Sequence alignments based on the 3D alignments using the program EXPRESSO [27]. The coloring of the alignments ranges from blue to red as the error in the alignment goes from high to low. Note that the β2-β3 loop in MtDef4 is longer by two residues and is far more basic than that in Psd1. However, the exposed hydrophobic F37 residue in MtDef4 is highly similar to W38 residue in Psd1.
Figure 7.
Surface characteristics of MtDef4.
A. In this diagram, the surface potential (calculated by DelPhi) is mapped onto the molecular surface of MtDef4 as a semi-transparent surface. The structure of MtDef4 is represented by a ribbon diagram that is colored blue to red as the chain extends from the amino-terminus to carboxy terminus. The disulfide bonds and the side chains of the arginine and lysine residues in the structure are also shown. Note that the overall surface is strongly basic.
B. Surface accessibility of the atoms in the MtDef4 structure. The semi-transparent molecular surface is colored blue to red for the least to most accessible atoms. Note that the very hydrophobic F37 on the β2-β3 loop, which is highly conserved among defensin proteins, is significantly exposed. The residues in this β2-β3 loop region are labeled.
Figure 8.
The RGFRRR loop present in the γ-core motif strongly regulates the antifungal activity of MtDef4.
A. Sequence of MtDef4 and its variants. The γ-core motif is indicated in larger font. RGFRRR sequence of MtDef4 is indicated in bold and the conserved amino acids are listed underneath with highly conserved amino acids on the top followed by less conserved ones. RGFR sequence which closely resembles the RXLR motif of the fungal and oomycete effectors is italicized.
B. Images showing the inhibition of F. graminearum PH-1 conidial germination and hyphal growth at different concentrations of MtDef4 or its variants. Images were taken after 16 hours of incubation of conidia with defensins. Bar = 50 μm.
C. Quantitative assessment of the in vitro antifungal activity of MtDef4 or its variants at 4 days after incubation of PH-1 conidia with defensins. Values are means of thee replications. Error bars indicate standard deviations.
D. Images showing the growth of PH-1 strain after 6 days in the presence of MtDef4 or its variants.
Figure 9.
MtDef4 variants, MtDef4RGFRRR/AAAARR and MtDef4RGFRRR/RGFRAA, are less efficient in permeabilizing F.
graminearum membrane compared to MtDef4 or MtDef4RGFRRR/RGAARR.
Quantitative measurement of fluorescence emitted by hyphae treated with different concentrations of MtDef4 or its variants plus 0.5 µM of SYTOX Green. Values are means of three replications. Error bars indicate standard deviations.
A and B. Fluorescence measurement at 30 min and 8 hr, respectively.
Figure 10.
DyLight 550-labeled MtDef4 and MtDef4RGFRRR/RGAARR but not MtDef4RGFRRR/AAAARR and MtDef4RGFRRR/RGFRAA enter F.
graminearum cytoplasm.
F. graminearum conidia were incubated with indicated concentrations of DyLight 550-labeled proteins and confocal fluorescence images were taken at various time intervals for up to 6 h.
A. Within 15 min, DyL-MtDef4, DyL-MtDef4RGFRRR/RGAARR and DyL-MtDef4RGFRRR/RGFRAA bound to the surface of conidia whereas DyL-MtDef4RGFRRR/AAAARR did not.
B. At 2 h, DyL-MtDef4 and DyL-MtDef4RGFRRR/RGAARR bound to the surface of germ tubes but DyL-MtDef4RGFRRR/AAAARR and DyL-MtDef4RGFRRR/RGFRAA did not.
C. DyL-MtDef4 entered selective hyphae by 4 h.
D. By 6 h, DyL-MtDef4RGFRRR/RGAARR entered hyphae but not all hyphae were affected.
E. DyL-MtDef4RGFRRR/AAAARR did not enter the hyphae even after 6 h.
F. DyL-MtDef4RGFRRR/RGFRAA bound to the surface of conidial cells but did not bind to hyphal surface.
Scale Bar = 10 µm.
Figure 11.
Tetramethyl rhodamine-labeled 16-mer peptide (TMR-GMA4-C) corresponding to the C-terminus of MtDef4 enters F.
graminearum but a variant peptide (TMR-GMA4-CM) with AAAA replaced for RGFR motif does not.
Fluorescence images of F. graminearum hyphae incubated with 96 µM TMR-GMA-C or TMR-GMA-CM for up to 4 h.
A. TMR-GMA4-C entered the cytoplasm.
B. Cross section fluorescent image of selective hyphae from (A) showing that TMR-GMA-C is present in the cytoplasm.
C. TMR-GMA4-CM faintly bound to the surface layer but with much lower intensity.
Scale Bar = 5 μm.
Figure 12.
The positively charged RGFRRR loop of MtDef4 plays a vital role in binding to phosphatidic acid.
A. Lipid overlay assays of MtDef4, MsDef1 and MsDef1-γ4. P-6001 PIP strips from Echelon Biosciences (Salt lake City, UT) were incubated with desired protein for 1 h at room temperature (see methods for details). After thorough washing, the bound proteins were detected using appropriate rabbit polyclonal-HP antibodies and Supersignal West Pico Chemiluminescent Substrate (Thermo Scientific). LPA = Lysophosphatidic acid; LPC = Lysophosphocholine ; PI = Phosphatidylinositol; PI(3)P = Phosphatidylinositol (3) phosphate ; PI(4)P = Phosphatidylinositol (4) phosphate; PI(5)P = Phosphatidylinositol (5) phosphate; PE = Phosphatidylethanolamine ; PC = Phosphatidylcholine; S1P = Sphingosine 1-Phosphate; PI(3,4)P2 = Phosphatidylinositol (3,4) bisphosphate; PI(3,5)P2 = Phosphatidylinositol (3,5) bisphosphate; PI(4,5)P2 = Phosphatidylinositol (4,5) bisphosphate; P(3,4,5)P3 =Phosphatidylinositol (3,4,5) trisphosphate; PA = Phosphatidic acid ; PS = Phosphatidylserine and Blank = No lipid.
B. MtDef4 binding to liposomes containing PC only or PC plus PA or PI(3,5)P2. Purified MtDef4 (2 µg) was incubated with different liposomes for 1 hr at room temperature. The vesicles were pelleted by centrifugation. The protein was visualized by immunoblotting with anti-MtDef4 antibody. 1. PC:PA (80:20), 2. PC:PA (40:60), 3. PC:PA(20:80), 4. PC:PI(3,5)P2 (50:50), 5. PC only, 6. MtDef4 (400 ng).
C. Surface Plasmon resonance sensograms for the binding of MtDef4 with immobilized PA/PC (3:2) and PC (100%) liposomes. MtDef4 sample dilutions were prepared in PBS buffer and injected at 20 µl/min flow rate. Kinetic parameters were estimated using BIAevaluation software (version 3.1).
D. Binding of MtDef4 and its variants to various species of PA. Lipids purchased from Avanti Polar lipids (Alabaster, AL) were spotted (2.5 µg) on Hybond C nitrocellulose membrane. Protein lipid overlay assays were conducted as described above and the bound proteins were detected using anti-MtDef4-HP antibodies. 1. PA 16:0 16:0; 2. Egg yolk phosphatidylcholine mix; 3. Diacylglycerol; 4. PA 18:1 18:1; 5. PA 16:0 18:1; 6. PA 8:0 8:0; 7. MtDef4 or variant protein (400 ng); 8. MtDef4 or variant protein (200 ng).