Fig 1.
Tracking Fusarium graminearum growth in maize stalk.
(A) The procedure from wound inoculation to split internodes ready for observation. (B) Representative split internodes at indicated time points. (C)-(D) Representative confocal (C) and wide-field (D) microscopic images of infected tissue from longitudinal sections. Maize plants were inoculated with spores of AmCyan-expressing F. graminearum (green under a GFP channel) and kept growing until the indicated timepoints when the stalk internodes were split and subjected to immediate microscopy. In this figure we focused on pith parenchyma cells; see Figure B in S1 Text for an uninfected reference, and Figure C in S1 Text for more pictures including rind areas. Green bar = 5 cm. White bar = 100 μm. hai: hours after inoculation.
Fig 2.
Measurements and diagram of maize stalk infection process.
(A) Measurements of half lesion size and hyphal advance distances. Error bars indicate SD. *Significant difference by Student’s t-test (P < 0.05). Sample size n ≥3. (B) The advance rates of the hyphal front calculated based on the data in (A). (C) Diagram of cellular events during the advance of F. graminearum in stalk pith mainly composed of parenchyma cells. The top of each column represents the wound site. F. graminearum grew up and down in the stalk in a similar pattern and with a similar expansion speed; for simplicity only the downside from the wound site is shown. Cell viability was judged based on plasmolysis assay. hai: hours after inoculation.
Fig 3.
graminearum caused nuclear abnormalities and cell death in nearby maize cells from 48 hai.
F. (A) Representative bright field images of maize stalk pith tissues. Images 12, 48 and 96 hai were taken of the cross sections preceding the hyphal front (as indicated by the red broken line in Fig 2A). Yellow arrowheads point to nucleus-like structures. V: vascular bundle. (B) Representative images showing infected maize pith tissues before and after 1 M NaCl treatment for 10 min. F. graminearum AmCyanPH-1 hyphae are visible as green lines under a GFP channel. Red arrows point to maize cell membranes away from cell walls. H: hyphae. White scale bar = 100 μm. (C) The shortest distances between fungal hyphae and maize cells undergoing plasmolysis were measured. Error bars indicate SD. Sample size n ≥4. Images in this figure were all taken from cross sections. hai: hours after inoculation.
Fig 4.
Laser microdissection-derived transcriptomes of F. graminearum in maize stalk.
(A) Representative images of laser microdissection (LM) of F. graminearum from infected maize stalk pith at 72 hai (longitudinal sections). White bar = 100 μm. (B) A heat-map representation of the hierarchical clustering using Pearson correlation coefficients. (C) Principal components analysis using microarray data of all 13,346 genes. (D) Numbers of differentially expressed genes between hyphae in maize and hyphae in vitro. (E) Heat maps of genes that were significantly up-regulated in maize compared to in vitro hyphae; see Figure H in S1 Text for more.
Fig 5.
Time course of expression of genes potentially involved in plant cell wall degradation, detoxification, and secondary metabolite biosynthesis.
(A) Expression of cell wall degradation-related genes grouped based on putative targeted components of plant cell wall. Total signal intensities for the genes in the denoted group are charted. (B) Hyphae in the pseudo pink color represent various growth path. (C) Expression patterns of detoxification-related genes grouped based on FunCat annotations. Total signal intensities for all genes in the denoted group are charted. The number of genes included in each category is indicated in parentheses. (D)-(E) Expression of several secondary metabolite biosynthesis clusters. hai: hours after inoculation. Genes significantly up-regulated in at least one time point but not up-regulated in spores are labeled with green asterisks. (F) DON detection in maize tissue extracts. ppm = mg/kg.
Fig 6.
Expressional and functional analysis of genes potentially related to non-phosphorus membrane lipid biosynthesis and phosphate metabolism.
(A) Diagram illustrating reactions related to the interconversion of lipids, and the expression of related genes. Heat maps indicate the expression of individual genes from microarray data. h: hour. hai, hours after inoculation. (B) Heat maps of genes related to phosphate transport and metabolism. Red arrows point to 18 hai, green asterisks indicate gene expression significantly up-regulated in maize infection compared to in vitro growth (SAM FDR < 0.05). (C) RT-PCR analysis of in vitro grown F. graminearum. MM: minimal medium, −P: minimal medium lacking phosphate. (D) ICP-MS measurements of phosphorus contents in various fractions of maize stalk. (E)-(F) Maize stalk infection assays. Lesions were diagnosed at 7 days post inoculation. *Significantly different difference from wild type-caused lesion (Student’s t-test, P <0.05), n = 3 independent experiments. Error bars denote SE. PC: phosphatidylcholine, PE: phosphatidylethanolamine, TAGL: triacylglycerol lipase, DAG: diacylglycerol, SAM: S-adenosylmethionine, DGTS: betaine lipid diacylglyceryl-N,N,N-trimethylhomoserine, PLC: phospholipase C, PLD: phospholipase D, MetK: methionine adenosyltransferase ETH-1, MSY: methionine synthase, EPT: ethanolamine-phosphotransferase, BTA1: betaine lipid biosynthetic enzyme.
Fig 7.
graminearum bta1 mutants show reduced virulence in maize stalk infection and reduced growth on phosphate depleted media.
F. (A)-(C) Maize stalks infected by wild-type or mutant F. graminearum at the indicated number of days after inoculation. Note that more vascular strands are visible in the stalk infected by wild-type F. graminearum than that infected by mutant F. graminearum in (B). White arrows point to hyphae (in pseudo pink). Blue arrows point to debris-like structure in host cells. (D) Supplementation with KH2PO4 increased lesion size caused by the bta1 mutant strain. Different letters indicate statistical differences among the strains, according to Tukey’s multiple comparison test (P = 0.05), n = 3 independent experiments. (E) Fungal growth after 2 days of culture on various media. MM: minimal medium, −P: minimal medium lacking phosphate. −P2nd: hyphal disks from −P medium were transferred to fresh −P medium plates to exhaust endogenous phosphate. Statistical analysis for colony radius and relative hyphal densities are charted at the bottom of the panel. The radius was measured at 1 day after transfer onto the −P2nd medium. The relative hyphal density was determined using ImageJ software by measuring the gray value of the same positions of each fungal colony on pictures (subtracted from the respective background gray value). *Significantly different from wild type (P < 0.05) according to Student’s t test, n ≥3. Error bars denote SE.
Fig 8.
BTA1 is responsible for DGTS production.
(A), (D) Thin layer chromatography (TLC) of lipid extracts from F. graminearum or E. coli expressing F. graminearum BTA1. (B), (E) Spots scraped from TLC plates (indicated by arrows) were applied to accurate-mass quantitative time-of-flight liquid chromatography mass spectrometry (Q-TOF LC MS) analysis. See Figure I in S1 Text for more lipid analyses. (C) RT-PCR results for BTA1 expression in various F. graminearum strains and conditions. (F) Subcellular localization of BTA1-mRFP in F. graminearum. ER: endoplasmic reticulum localized protein marker. White bar = 100 μm. (G) Virulence assays on maize stalks. Lesions were measured at 7 days after inoculation. Different letters indicate statistical differences among the strains according to Tukey’s multiple comparison test (P = 0.05), n = 3 independent experiments.
Fig 9.
A hypothetical model of molecular events during maize stalk infection by F. graminearum, based on microscopic observation and gene expression profiling.
At 12 h after inoculation (hai), the fungus grows between live parenchyma cells, secretes plant cell wall degrading enzymes (CWDE) mainly breaking primary chains of pectin (PL1, PL3), cellulose (GH61, GH45, GH7) and hemicellulose (GH12) for penetrating barriers, secretes secondary metabolites (siderophore and aurofusarin), and increases high affinity phosphate transporters; at 18 hai, the fungus continues to grow between live parenchyma cells, secretes similar CWDEs and secondary metabolites, but reduces the expression of high-affinity phosphate transporters, and deploys BTA1 to produce non-phosphorus membrane lipids to assure fast growth under phosphate starvation; at 48 hai, the fungus starts to grow into host cells and kills invaded and surrounding host cells, gains access to host cellular phosphate, reduces CWDE production as intracellular growth needs less cell wall breakage, and produces secondary metabolites such as carotenoids; around 72–108 hai, more CWDEs (including those targeting side branches of pectin) and FGL1 lipase are produced for full digestion of plant tissues, a toxic lectin FFBL is secreted, and a putative multidrug resistance protein is also produced; around 132–144 hai, the fungus grows among dead parenchyma cells, produces less CWDEs, more FFBL, and more major facilitator proteins. CAZy categories are indicated in parentheses after the respective CWDEs.