Infection Strategies Deployed by Botrytis cinerea, Fusarium acuminatum, and Rhizopus stolonifer as a Function of Tomato Fruit Ripening Stage

Abstract

Worldwide, 20-25% of all harvested fruit and vegetables are lost annually in the field and throughout the postharvest supply chain due to rotting by fungal pathogens. Most postharvest pathogens exhibit necrotrophic or saprotrophic lifestyles, resulting in decomposition of the host tissues and loss of marketable commodities. Necrotrophic fungi can readily infect ripe fruit leading to the rapid establishment of disease symptoms. However, these pathogens generally fail to infect unripe fruit or remain quiescent until host conditions stimulate a successful infection. Previous research on infections of fruit has mainly been focused on the host's genetic and physicochemical factors that inhibit or promote disease. Here, we investigated if fruit pathogens can modify their own infection strategies in response to the ripening stage of the host. To test this hypothesis, we profiled global gene expression of three fungal pathogens that display necrotrophic behavior-Botrytis cinerea, Fusarium acuminatum, and Rhizopus stolonifer-during interactions with unripe and ripe tomato fruit. We assembled and functionally annotated the transcriptomes of F. acuminatum and R. stolonifer as no genomic resources were available. Then, we conducted differential gene expression analysis to compare each pathogen during inoculations versus in vitro conditions. Through characterizing patterns of overrepresented pathogenicity and virulence functions (e.g., phytotoxin production, cell wall degradation, and proteolysis) among the differentially expressed genes, we were able to determine shared strategies among the three fungi during infections of compatible (ripe) and incompatible (unripe) fruit tissues. Though each pathogen's strategy differed in the details, interactions with unripe fruit were commonly characterized by an emphasis on the degradation of cell wall components, particularly pectin, while colonization of ripe fruit featured more heavily redox processes, proteolysis, metabolism of simple sugars, and chitin biosynthesis. Furthermore, we determined that the three fungi were unable to infect fruit from the non-ripening (nor) tomato mutant, confirming that to cause disease, these pathogens require the host tissues to undergo specific ripening processes. By enabling a better understanding of fungal necrotrophic infection strategies, we move closer to generating accurate models of fruit diseases and the development of early detection tools and effective management strategies.

Keywords: broad host range pathogens; cell wall degrading enzymes; de novo transcriptomes; fruit-pathogen interactions; necrotic response; necrotrophic fungi; redox; rotting.

FIGURE 1

Fungal growth and disease development in tomato fruit. (A) Growth and lesion development of the three fungi in vitro and during inoculation, respectively. Fungi were grown on PDA in 100 mm Petri dishes. In vitro morphology represents the pre-sporulation stage used for this study at 3 (Rhizopus stolonifer), 5 (Botrytis cinerea) and 7 (Fusarium acuminatum) days post-plating. Fungal growth and lesion development during fruit inoculation is shown at 1 and 3 days post-inoculation (dpi) in mature green (MG) and red ripe (RR) fruit. The extent of mycelial growth is highlighted by dotted lines for 1 dpi RR fruit. White and black bars correspond to 1 and 5 mm, respectively. (B) Fungal biomass estimated by the relative expression of the reference genes BcRPL5 (Bcin01g09620), FaEF1b (FacuDN4188c0g1i4), and Rs18S (RstoDN6002c0g2i1), normalized based on the tomato reference gene expression (SlUBQ, Solyc12g04474). Significant differences (P < 0.05) between the biomass of the four treatments are denoted by letters. (C) Relative expression of the disease responsive tomato gene SlWRKY33 (Solyc09g014990) in samples inoculated with the three fungi and in the mock-inoculated control. Symbols indicate statistical significance (n.s., not significant; ^◼P < 0.1; ^∗P < 0.05; ^∗∗P < 0.01) when comparing inoculations with each pathogen and the control.

FIGURE 2

Summary of functional annotations across the Botrytis cinerea, Fusarium acuminatum, and Rhizopus stolonifer transcriptomes. Percent of all annotated transcripts in that transcriptome that contain at least one annotation for each categorization. GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; PHI, Pathogen-Host Interaction; CAZymes, Carbohydrate-Active enZymes. Detailed information can be found in Supplementary Table S3.

FIGURE 3

Principal component analysis (PCA) and intersections of differentially expressed genes (DEGs) during inoculations of tomato fruit. (A) PCA plots of variance-stabilized matrixes of mapped reads for each pathogen as generated by DESeq2. (B,C) UpSetR visualizations of intersections between the upregulated (B) and downregulated (C) DEGs of inoculated fruit at two time points versus in vitro comparisons for each pathogen. Intersections are displayed in descending order by number of genes. All datasets can be accessed in Supplementary Tables S5–S7.

FIGURE 4

Subset of overrepresented Gene Ontology (GO) terms associated with pathogenicity and fungal growth among upregulated genes in Botrytis cinerea, Fusarium acuminatum and Rhizopus stolonifer inoculated samples. Indented GO terms are nested in the GO term above. Box color indicates the significance of enrichment, and values in the boxes indicate the number of upregulated genes in this comparison that share the indicated annotation. dpi, days post-inoculation; MG, mature green; RR, red ripe. Supplementary Table S8 includes the complete list of overrepresented GO terms.

FIGURE 5

Upregulated CAZy family genes for each pathogen in each of the four treatments. Families and subfamilies from CAZy (www.cazy.org) are listed on the left. These are further nested into categories based on their substrates or activities. Each family is described by the percentage of all upregulated CAZyme genes it represents in each treatment. Only families which constitute at least 2% of upregulated CAZyme genes in at least one treatment are shown. All remaining CAZy families can be retrieved from Supplementary Table S9.

FIGURE 6

Fungal pathogens are unable to infect fruit from the non-ripening (nor) tomato mutant. (A) Shows inoculations of Botrytis cinerea, Fusarium acuminatum and Rhizopus stolonifer on mature green (MG) and red ripe (RR) wild-type tomato (cv. Ailsa Craig) fruit at 3 days post-inoculation (dpi). (B) Shows inoculations of the same fungi on MG and RR-like fruit from the isogenic non-ripening (nor) tomato mutant. White size bars correspond to 1 cm whereas black size bars correspond to 1 mm.