Extracellular Vesicles From the Cotton Pathogen Fusarium oxysporum f. sp. vasinfectum Induce a Phytotoxic Response in Plants

Abstract

Extracellular vesicles (EVs) represent a system for the coordinated secretion of a variety of molecular cargo including proteins, lipids, nucleic acids, and metabolites. They have an essential role in intercellular communication in multicellular organisms and have more recently been implicated in host-pathogen interactions. Study of the role for EVs in fungal biology has focused on pathogenic yeasts that are major pathogens in humans. In this study we have expanded the investigation of fungal EVs to plant pathogens, specifically the major cotton pathogen Fusarium oxysporum f. sp. vasinfectum. EVs isolated from F. oxysporum f. sp. vasinfectum culture medium have a morphology and size distribution similar to EVs from yeasts such as Candida albicans and Cryptococcus neoformans. A unique feature of the EVs from F. oxysporum f. sp. vasinfectum is their purple color, which is predicted to arise from a napthoquinone pigment being packaged into the EVs. Proteomic analysis of F. oxysporum f. sp. vasinfectum EVs revealed that they are enriched in proteins that function in synthesis of polyketides as well as proteases and proteins that function in basic cellular processes. Infiltration of F. oxysporum f. sp. vasinfectum EVs into the leaves of cotton or N. benthamiana plants led to a phytotoxic response. These observations lead to the hypothesis that F. oxysporum f. sp. vasinfectum EVs are likely to play a crucial role in the infection process.

Keywords: extracellular vesicle (EV); fungi; host-pathogen interaction; pigment; polyketide.

Figure 1

Characterization of EVs isolated from F. oxysporum f. sp. vasinfectum. EVs were isolated using differential centrifugation followed by ultracentrifugation. (A) Absorption spectrum of the F. oxysporum f. sp. vasinfectum EV preparation had dual maxima at 550 and 590 nm. Data are representative of three independent replicates. (B) The number and size distribution of the isolated EVs were determined using nanoparticle tracking analysis. The data from the three technical replicates from one biological sample are representative of all samples used in this study. The average quantity and size of EVs from three independent isolations of EVs from F. oxysporum f. sp. vasinfectum were 1.0 x 10¹² ± 2.8 x 10¹¹ particles/ml of culture, mean diameter 155.1 nm ± 3.5 nm, mode diameter 150.0 nm ± 5.1 nm. Values are presented ± standard deviation of three biological replicates. (C) TEM confirmed that most particles had the characteristic cup like morphology of EVs (i–iii); some particles had a multi-lobed rosette morphology as seen in (iv).

Figure 2

GO terms over- and underrepresented in F. oxysporum f. sp. vasinfectum EV proteins. (A) GO terms that are over- or underrepresented in the set of EV proteins. EV proteins were identified by LC-MS/MS with the peptides aligned to the predicted F. oxysporum f. sp. vasinfectum proteome. A protein was listed as an EV protein if there were at least two unique peptides from the protein identified in at least two of three EV pools. (B) Under- and overrepresented GO terms identified in the list of proteins identified in F. oxysporum f. sp. vasinfectum whole cell lysate. These lists are the top 30 terms; full lists can be found in Tables S2 and S3. GO terms were assigned to transcripts based on homology searches using Blast2Go. GO term enrichment was also performed using Blast2GO. The number next to each bar is the p-value for the GO term.

Figure 3

The effect of fungal EVs on cotton cotyledons. EVs were purified from F. oxysporum f. sp. vasinfectum and S. cerevisiae and diluted to a protein concentration of 0.1 µg/ml in PBS. (A, B) Paired cotyledons were infiltrated at four points with the EV solution and PBS control. After 5 days incubation, the cotyledons were removed from the plants and imaged. Cotyledons infiltrated with EVs from F. oxysporum f. sp. vasinfectum (A) had discolored/brown spots around the sites of infiltration, whereas the cotyledons infiltrated with EVs from S. cerevisiae (B) only had mild discoloration or no response. Cotyledons infiltrated with PBS had no observable effect (the brown spot on the third cotyledon in the F. oxysporum panel (A) is due to a pinch point as a result of a fold in the leaf and not infiltration). (C, D) N. benthamiana leaves were infiltrated with EVs at multiple points on the right half and PBS on the left. After 5 days the leaves were removed from the plants and imaged. Leaf halves infiltrated with EVs from F. oxysporum f. sp. vasinfectum (C) showed substantial discoloration around the sites of infiltration, whereas the half infiltrated with PBS had no discoloration. Leaf halves infiltrated with EVs from S. cerevisiae (D) appeared similar to the corresponding PBS infiltrated leave halves. Infiltration of cotton cotyledons with resuspended F. oxysporum f. sp. vasinfectum spores and hyphae (E) led to the formation of very small discolored spots at the sites of infiltration, but they were much smaller than those observed with EVs. Images are representative of three independent experiments.

Figure 4

The optiprep fractions containing both EVs and the purple pigment are responsible for the hypersensitive response induced by the crude EV fraction in plants. (A) Optiprep gradient centrifugation of the crude EV fraction revealed that the purple pigment concentrates in two regions, one centered at 10% and a second close to 40% iodixanol. The purple pigment was concentrated in fractions 5–7. (B) Representative TEM of fractions 5–7. Particles with characteristic cup-like morphology of EVs were detected in these fractions as well as some rosette-shaped particles. (C) Representative NTA of fractions 5–7. The mode particle diameter was 104.5 nm; no particles were detected below 46.5 nm or above 226.5 nm. (D) Application of select fractions (numbered in white) to the cotyledons of a cotton plant. Fractions (left cotyledon) were all paired with a PBS control (right cotyledon). Fraction 5 induced the strongest response with fractions 6 and 7 inducing a lesser response. Fractions 1, 2, and 11 did not induce a response beyond that observed for the PBS control. The brown spots and holes on the control cotyledons are due to wounding of the leaf during infiltration.