Rhamnose synthase activity is required for pathogenicity of the vascular wilt fungus Verticillium dahliae

Abstract

The initial interaction of a pathogenic fungus with its host is complex and involves numerous metabolic pathways and regulatory proteins. Considerable attention has been devoted to proteins that play a crucial role in these interactions, with an emphasis on so-called effector molecules that are secreted by the invading microbe to establish the symbiosis. However, the contribution of other types of molecules, such as glycans, is less well appreciated. Here, we present a random genetic screen that enabled us to identify 58 novel candidate genes that are involved in the pathogenic potential of the fungal pathogen Verticillium dahliae, which causes vascular wilt diseases in over 200 dicotyledonous plant species, including economically important crops. One of the candidate genes that was identified concerns a putative biosynthetic gene involved in nucleotide sugar precursor formation, as it encodes a putative nucleotide-rhamnose synthase/epimerase-reductase (NRS/ER). This enzyme has homology to bacterial enzymes involved in the biosynthesis of the nucleotide sugar deoxy-thymidine diphosphate (dTDP)-rhamnose, a precursor of L-rhamnose, which has been shown to be required for virulence in several human pathogenic bacteria. Rhamnose is known to be a minor cell wall glycan in fungi and has therefore not been suspected as a crucial molecule in fungal-host interactions. Nevertheless, our study shows that deletion of the VdNRS/ER gene from the V. dahliae genome results in complete loss of pathogenicity on tomato and Nicotiana benthamiana plants, whereas vegetative growth and sporulation are not affected. We demonstrate that VdNRS/ER is a functional enzyme in the biosynthesis of uridine diphosphate (UDP)-rhamnose, and further analysis has revealed that VdNRS/ER deletion strains are impaired in the colonization of tomato roots. Collectively, our results demonstrate that rhamnose, although only a minor cell wall component, is essential for the pathogenicity of V. dahliae.

Keywords: Agrobacterium tumefaciens-mediated transformation (ATMT); UDP-rhamnose; attachment; carbohydrate; root colonization; tomato; vascular wilt.

Figure 1

Typical assay to identify transformants with reduced pathogenicity or virulence. Ten‐day‐old tomato seedlings were mock inoculated (M) or inoculated with conidiospores of wild‐type Verticillium dahliae (WT) or random T‐DNA insertion mutants. At 21 days post‐inoculation, the plants were scored for disease development by comparing plants inoculated with wild‐type V. dahliae with those inoculated with the T‐DNA insertion mutants. The sides (top) and tops (bottom) of plants inoculated with five T‐DNA insertion mutants that are impaired in aggressiveness are shown.

Figure 2

Targeted deletion of VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) does not impair growth and conidiogenesis. (A) Radial growth and colony morphology of wild‐type Verticillium dahliae (WT), random transformant RM‐389, an ectopic transformant (EC) and two VdNRS/ER deletion strains (Δ6010‐1 and Δ6010‐2) after 7 days of incubation on potato dextrose agar (PDA) medium at 22°C. (B) Average number of conidia produced after 7 days of growth on PDA medium based on two independent experiments. Letters indicate significant differences (P < 0.05) calculated using Student's t‐test.

Figure 3

Expression of VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) during infection of Verticillium dahliae on tomato. Ten‐day‐old tomato (Solanum lycopersicum) cultivar MoneyMaker plants were root inoculated with V. dahliae, and plants were harvested at regular intervals from 4 to 16 days post‐inoculation (dpi). After RNA isolation and cDNA synthesis, real‐time polymerase chain reaction was performed to determine the relative expression levels of VdNRS/ER using the V. dahliae elongation factor 1‐α gene as a reference. Expression at 4 dpi is set to unity.

Figure 4

VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) is required for pathogenicity of Verticillium dahliae on tomato and Nicotiana benthamiana. (A) Top and side views of tomato cultivar MoneyMaker plants that were mock inoculated (M) or inoculated with wild‐type V. dahliae (WT), random transformant RM‐389, an ectopic transformant (EC) and two VdNRS/ER deletion strains (Δ6010‐1 and Δ6010‐2) at 14 days post‐inoculation (dpi). Fungal outgrowth at 7 days after plating of stem sections harvested at 14 dpi is shown at the bottom of the panel. (B) Top and side views of N. benthamiana plants inoculated as specified for (A).

Figure 5

VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) is required for pathogenicity of Verticillium dahliae on tomato and Nicotiana benthamiana. Average canopy area of six tomato plants (A) and real‐time polymerase chain reaction (PCR) quantification of fungal biomass (B) at 14 days after mock inoculation (mock) or inoculation with wild‐type Verticillium dahliae (WT), random transformant (RM‐389), an ectopic transformant (EC) and two VdNRS/ER deletion strains (Δ6010‐1 and Δ6010‐2). Different letters indicate significant differences (P < 0.05). Average canopy area of six N. benthamiana plants (C) and real‐time PCR quantification of fungal biomass (D) on inoculation as specified for (A) and (B).

Figure 6

Targeted deletion of VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) does not affect cell wall integrity and osmotic stress resistance. Stress sensitivity assays were performed by placing a 5‐μL droplet (10⁶ conidia/mL) of wild‐type Verticillium dahliae (WT), an ectopic transformant (EC) and two VdNRS/ER deletion strains (Δ6010‐1 and Δ6010‐2) on Czapek‐Dox medium (C) or Czapek‐Dox medium supplemented with Congo red (250 mm, 500 mm, 750 mm and 1 m), NaCl (250 mm, 500 mm, 750 mm and 1 m), mannitol (250 mm, 500 mm, 750 mm and 1 m) or sorbitol (300 mm, 600 mm, 900 mm and 1.2 m). The colony diameter was measured after 7 days of incubation at 22°C.

Figure 7

VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) is required for the colonization of tomato roots. Roots of 10‐day‐old tomato cultivar MoneyMaker seedlings were immersed in one‐fifth potato dextrose broth (PDB) containing 10⁶ conidia/mL of wild‐type Verticillium dahliae (WT), random transformant (RM‐389), an ectopic transformant (EC) and two VdNRS/ER deletion strains (Δ6010‐1 and Δ6010‐2) for 72–96 h. (A) Roots were rinsed with water and photographed under a microscope. (B) Real‐time polymerase chain reaction quantification of fungal biomass on the roots. Different letters indicate significant differences (P < 0.05).

Figure 8

VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) deletion strains are depleted in uridine diphosphate (UDP)‐rhamnose. Gas chromatography‐mass spectrometry (GC‐MS) analysis of alditol acetate derivatives of the glycosyl residues from conidiospore polysaccharides. (A) Total ion count chromatogram of alditol acetate‐derived standard rhamnose (Rha) and mass spectrum (inset) showing fragmentation pattern fingerprints with m/z 129, 171, 231, 303. (B) Polysaccharides from wild‐type Verticillium dahliae (WT) conidiospores display a peak eluting at 16.10 min that has the same retention time as rhamnose standard. (C) VdNRS/ER deletion strain (Δ6010‐1) lacks a peak eluting at 16.10 min. (D) The VdNRS/ER deletion strain complemented with the native VdNRS/ER gene (Comp‐2) displays a peak eluting at 16.10 min that has the same retention time as the rhamnose standard.

Figure 9

Uridine diphosphate (UDP)‐rhamnose is produced in germinating conidiospores. Gas chromatography‐mass spectrometry (GC‐MS) analysis of alditol acetate derivatives of the glycosyl residues from germinating conidiospore polysaccharides. (A) Total ion count chromatogram of alditol acetate‐derived standard rhamnose (Rha). (B) Polysaccharides of conidiospores from wild‐type Verticillium dahliae (WT) display a peak with the same retention time as the rhamnose standard. (C) VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) deletion strain (Δ6010‐1) does not display a rhamnose peak. Other peaks in the chromatogram correspond to ribose (Rib), arabinose (Ara) and xylose (Xyl). Asterisks indicate unidentified compounds that are not sugars.

Figure 10

VdNRS/ER (NRS/ER, nucleotide‐rhamnose synthase/epimerase‐reductase) converts uridine diphosphate‐4‐keto‐6‐deoxy‐glucose (UDP‐KDG) to UDP‐rhamnose. (A) UDP‐glucose (UDP‐Glc) standard elutes from a HILIC (hydrophilic interaction liquid chromatography) column at 12.8 min and is detected by mass spectrometry (MS) with m/z 565 [M – H]^–. The MS/MS of the parent ion (inset) gives m/z 323 and 385 diagnostic ion fragments, [UMP – H]^– and [UDP – H₂O – H]^–, respectively. (B) HILIC separation of products from the enzymatic reaction with BfDh (4,6‐dehydratase) displays a peak with the same retention time as UDP‐KDG with diagnostic [M – H]^– m/z 547 and MS/MS ion fragment with m/z 323. (C) HILIC separation of products from the dual enzymatic reaction with BfDh and VdNRS/ER displays a peak with the same retention time as UDP‐rhamnose (UDP‐Rha) with diagnostic [M – H]^– m/z 549 and MS/MS of 403, 323, 210. (D) Dual enzymatic reaction with BfDh and an empty vector control displays a peak with the same retention time as UDP‐KDG with diagnostic [M – H]^– m/z 547 and MS/MS ion fragment with m/z 323. (E) UDP‐rhamnose metabolic pathway in fungi showing the enzymes involved in the sequential conversion of UDP‐Glc to UDP‐Rha.