The Golgin Protein RUD3 Regulates Fusarium graminearum Growth and Virulence

Abstract

Golgins are coiled-coil proteins that play prominent roles in maintaining the structure and function of the Golgi complex. However, the role of golgin proteins in phytopathogenic fungi remains poorly understood. In this study, we functionally characterized the Fusarium graminearum golgin protein RUD3, a homolog of _ScRUD3/GMAP-210 in Saccharomyces cerevisiae and mammalian cells. Cellular localization observation revealed that RUD3 is located in the cis-Golgi. Deletion of RUD3 caused defects in vegetative growth, ascospore discharge, deoxynivalenol (DON) production, and virulence. Moreover, the Δrud3 mutant showed reduced expression of tri genes and impairment of the formation of toxisomes, both of which play essential roles in DON biosynthesis. We further used green fluorescent protein (GFP)-tagged SNARE protein SEC22 (SEC22-GFP) as a tool to study the transport between the endoplasmic reticulum (ER) and Golgi and observed that SEC22-GFP was retained in the cis-Golgi in the Δrud3 mutant. RUD3 contains the coiled coil (CC), GRAB-associated 2 (GA2), GRIP-related Arf binding (GRAB), and GRAB-associated 1 (GA1) domains, which except for GA1, are indispensable for normal localization and function of RUD3, whereas only CC is essential for normal RUD3-RUD3 interaction. Together, these results demonstrate how the golgin protein RUD3 mediates retrograde trafficking in the ER-to-Golgi pathway and is necessary for growth, ascospore discharge, DON biosynthesis, and pathogenicity in F. graminearumIMPORTANCEFusarium head blight (FHB) caused by the fungal pathogen Fusarium graminearum is an economically important disease of wheat and other small grain cereal crops worldwide, and limited effective control strategies are available. A better understanding of the regulation mechanisms of F. graminearum development, deoxynivalenol (DON) biosynthesis, and pathogenicity is therefore important for the development of effective control management of this disease. Golgins are attached via their extreme carboxy terminus to the Golgi membrane and are involved in vesicle trafficking and organelle maintenance in eukaryotic cells. In this study, we systematically characterized a highly conserved Golgin protein, RUD3, and found that it is required for vegetative growth, ascospore discharge, DON production, and pathogenicity in F. graminearum Our findings provide a comprehensive characterization of the golgin family protein RUD3 in plant-pathogenic fungus, which could help to identify a new potential target for effective control of this devastating disease.

Keywords: FgRud3; Fusarium graminearum; RUD3; cis-Golgi; golgin; intracellular protein trafficking; virulence.

FIG 1

RUD3 plays an important role in vegetative growth but is not required for conidial production. (A) Colony morphology of the wild-type (WT) PH-1 and rud3 deletion mutants (Δrud3 no. 1 and no. 3) and complemented strain (Δrud3::rud3) grown on V8, 5 × yeast extract-glucose (YEG) medium, complete medium (CM), and minimal medium (MM) agar at 25°C for 3 days. (B) Statistical analysis of colony diameters of the indicated strains. Each graph represents the average of three experiments. *, P < 0.05. (C) Aerial hyphae of the indicated strains in test tubes cultured with potato dextrose agar (PDA) medium at 25°C for 3 days. Statistical analysis of the aerial hyphae of the indicated strains. **, P < 0.01. (D) Conidia produced by PH-1, Δrud3 (no. 1 and no. 3), and Δrud3::rud3 strains in 50 ml carboxymethyl cellulose (CMC) medium were counted after 5 days of incubation at 25°C. Error bars represent the standard deviation (SD). *, P < 0.05.

FIG 2

RUD3 is essential for conidium morphology and ascospore discharge. (A) Conidia of the indicated strains cultured in CMC medium for 5 days. The septa were stained with calcofluor white and visualized by fluorescence microscopy. Bars, 20 μm. (B) Percentages of spores in panel A in each septa number category (≤2, 3, or 4 to 7). At least 300 spores were counted in at least three fields for each strain. Error bars represent the SD. (C) Statistical analysis of conidial length. Error bars represent the SD. *, P < 0.05. (D) Perithecia, ascospore formation, and ascospore discharge of the indicated strains on carrot agar plates, photographed at 14 days postinoculation (dpi). Bars, 20 μm.

FIG 3

FgRud3 is required for full virulence of Fusarium graminearum. (A) Flowering wheat heads were inoculated with conidial suspensions of PH-1, Δrud3 (no. 1 and no. 3), and Δrud3::rud3 strains and checked at 14 dpi. The disease index was determined by the number of symptom spikelets per wheat head. Error bars represent the SD. *, P < 0.05. (B) Wheat coleoptiles were inoculated with 2 μl conidium suspension (2 × 10⁵ spores/ml) and examined at 10 dpi. Error bars represent the SD. *, P < 0.05. (C) Deoxynivalenol (DON) production in wheat seeds infected with the indicated strains after 20 days of incubation. Error bars represent the SD. **, P < 0.01. (D) Relative expression levels of tri1, tri4, tri5, tri6, and tri10 in PH-1 and the Δrud3 mutant. The GAPDH gene was used as an internal control. Error bars represent the SD. **, P < 0.01; ***, P < 0.001.

FIG 4

RUD3 is involved in toxisome formation. (A) PH-1 and Δrud3 mutant strains expressing TRI4-GFP were incubated in TBI for 2 days at 25°C and visualized using live-cell fluorescence microscopy. Scale bars, 10 μm. (B) TRI4-GFP protein was determined in the hypha lysates using immunoblot analysis with anti-GFP antibodies with GAPDH as a loading control. The band intensities of blots from three independent experiments were quantified using ImageJ software (National Institutes of Health). Each graph represents the average of three experiments. **, P < 0.01.

FIG 5

RUD3 localizes to the cis-Golgi. (A and B) PH-1 coexpressed RUD3-GFP with mCherry-SED5, mCherry-SFT2, or tdTomato-VPS21 constructs. Conidia in panel A were harvested from 3-day-old CMC cultures. The hyphae in panel B were germinated in liquid CM for 24 h. The conidia and hyphae were visualized using a confocal microscope. Bars, 10 μm.

FIG 6

Multiple SEC22-GFP dots were trapped in the cis-Golgi apparatus of the Δrud3 mutant. WT and Δrud3 mutant strains coexpressing SEC22-GFP with mCherry-SED5 were treated as described in Materials and Methods. The percentage of mCherry-SED5 puncta-colocalized SEC22-GFP dots was quantified. At least 1,000 mCherry dots were counted in at least 3 fields for each strain. **, P < 0.01.

FIG 7

Vegetative growth of the PH-1, Δrud3, and Δrud3::rud3 strains under different stress conditions. (A) The PH-1, Δrud3, and Δrud3::rud3 strains were inoculated on CM containing NaCl, KCl, or sorbitol. (B) The indicated strains were inoculated on CM with 5 mM and 10 mM H₂O₂. (C) Cultures of all strains were grown on CM with sodium dodecyl sulfate (SDS) and Congo red (CR). (D to F). Statistical analysis of inhibition based on colony diameters after 3 days of incubation. Means and SDs were calculated from three replicates. *, P < 0.05.

FIG 8

RUD3 interacts with itself through the coiled-coil (CC) domain. (A) RUD3-RUD3 interaction in SED5-marked cis-Golgi. YFP fluorescence was observed in hyphae coexpressing RUD3-N-YFP with RUD3-C-YFP; this strain also expressed mCherry-SED5. Overlap of YFP and mCherry fluorescence indicates that RUD3 self-interacts on the cis-Golgi (arrow). Arrowheads indicate the colocalization of green and red puncta. The arrow indicates green puncta that did not colocalize with red puncta. Bars, 10 μm. (B) The structure of RUD3. (C) RUD3 self-interacts via the CC domain, as shown in a Y2H assay. Prey and bait constructs were assayed for growth on SD-Leu-Trp, SD-Leu-Trp-His, and SD-Leu-Trp-His-Ade plates.

FIG 9

Functional characterization of RUD3 domains. (A) Subcellular localization of RUD3^ΔCC-GFP, RUD3^ΔGA2-GFP, RUD3^ΔGRAB-GFP, and RUD3^ΔGA1-GFP in F. graminearum. PH-1 strains coexpressing RUD3^ΔCC-GFP, RUD3^ΔGA2-GFP, RUD3^ΔGRAB-GFP, and RUD3^ΔGA1-GFP with mCherry-SED5 were treated as described in Materials and Methods. Arrowheads indicate the colocalization of red and green puncta. Arrows indicate red puncta that did not colocalize with green puncta. Scale bars, 10 μm. The percentage of mCherry dot-colocalized GFP dots was quantified. At least 500 mCherry dots were counted in at least 3 fields for each strain. *, P < 0.05; **, P < 0.01. (B) Colony morphology of indicated strains cultured on PDA for 3 days at 25°C. Statistical analysis of the colony diameters of the indicated strains. Each graph represents the average of three experiments. *, P < 0.05. (C) Infection assay of the domain deletion transformants. Flowering wheat heads were drop-inoculated with conidial suspensions of the indicated strains and examined at 14 dpi. The disease index was determined from the number of symptomatic spikelets per wheat head. At least 30 wheat heads infected with each strain were counted for each strain. *, P < 0.05.