Molecular Networking-Based Metabolome and Bioactivity Analyses of Marine-Adapted Fungi Co-cultivated With Phytopathogens

Abstract

Fungi represent a rich source of bioactive metabolites and some are marketed as alternatives to synthetic agrochemicals against plant pathogens. However, the culturability of fungal strains in artificial laboratory conditions is still limited and the standard mono-cultures do not reflect their full spectrum chemical diversity. Phytopathogenic fungi and bacteria have successfully been used in the activation of cryptic biosynthetic pathways to promote the production of new secondary metabolites in co-culture experiments. The aim of this study was to map the fungal diversity of Windebyer Noor, a brackish lake connected to Baltic Sea (Germany), to induce the chemical space of the isolated marine-adapted fungi by co-culturing with phytopathogens, and to assess their inhibitory potential against six commercially important phytopathogens. Out of 123 marine-adapted fungal isolates obtained, 21 were selected based on their phylogenetic and metabolite diversity. They were challenged with two phytopathogenic bacteria (Pseudomonas syringae and Ralstonia solanacearum) and two phytopathogenic fungi (Magnaporthe oryzae and Botrytis cinerea) on solid agar. An in-depth untargeted metabolomics approach incorporating UPLC-QToF-HRMS/MS-based molecular networking (MN), in silico MS/MS databases, and manual dereplication was employed for comparative analysis of the extracts belonging to nine most bioactive co-cultures and their respective mono-cultures. The phytopathogens triggered interspecies chemical communications with marine-adapted fungi, leading to the production of new compounds and enhanced expression of known metabolites in co-cultures. MN successfully generated a detailed map of the chemical inventory of both mono- and co-cultures. We annotated overall 18 molecular clusters (belonging to terpenes, alkaloids, peptides, and polyketides), 9 of which were exclusively produced in co-cultures. Several clusters contained compounds, which could not be annotated to any known compounds, suggesting that they are putatively new metabolites. Direct antagonistic effects of the marine-adapted fungi on the phytopathogens were observed and anti-phytopathogenic activity was demonstrated.The untargeted metabolomics approach combined with bioactivity testing allowed prioritization of two co-cultures for purification and characterization of marine fungal metabolites with crop-protective activity. To our knowledge, this is the first study employing plant pathogens to challenge marine-adapted fungi.

Keywords: biocontrol; co-culture; marine-adapted fungi; metabolomics; molecular networking; phytopathogen.

FIGURE 1

Diversity and distribution of 123 isolates obtained from sampling in Windebyer Noor by (A) Fungal genera. (B) Isolation media: Isolation was performed using four different solid media, i.e., PCA medium, HS medium, GPY medium, and WSP medium. (C) Isolation source (fungal orders). WO, twigs; SC, wood scrapings of a driftwood; SD, seeds; WA, water; F, foam; L, leaves; SE, sediment.

FIGURE 2

Macroscopic images of the co-cultures of (A) marine-adapted Emericellopsis sp. (top of Petri dish) with the phytopathogenic fungus M. oryzae (bottom of the Petri dish) showing a distance type of interaction. (B) Cosmospora sp. (top of the Petri dish) with M. oryzae (bottom of the Petri dish) showing a darkened confrontation zone interaction. (C) Acremonium sp. (top of the Petri dish) with M. oryzae (bottom of the Petri dish) displaying a contact type of inhibition. (D) Alternaria sp. (top of the Petri dish) with P. syringae (bottom of the Petri dish) showing an overgrowth type interaction. (E) Stereomicroscopic picture of the confrontation zone between Cosmospora sp. (left) and M. oryzae (right) with a scale bar representing 500 μm. It is observed that the hyphae of the phytopathogen M. oryzae do not grow further into the confrontation zone. The zone is however dominated by the hyphae of Cosmospora sp.

FIGURE 3

(A) Global molecular network of the nine co-cultures of marine-adapted fungi and phytopathogens and their respective mono-cultures. Red nodes, mono-cultures from marine fungal isolates; green nodes, mono-cultures from phytopathogens; blue nodes, co-cultures. The global molecular network highlights that co-cultivation increased the chemical space in comparison to mono-cultures and therefore contains data derived from five different marine-adapted fungal isolates cultivated in two different media with three different phytopathogens. Nine clusters induced only in co-cultures (clusters numbered 1–9 in blue only nodes). Separate molecular networks for all nine co-cultures and the respective mono-cultures are given in Supplementary Figures 5–13. (B) Global Euler diagram based on the global molecular network (MN) in (A). Red, ions from mono-cultures of marine-adapted fungal isolates obtained during this study; green, ions obtained from mono-cultures of phytopathogens; blue, ions obtained from co-cultures. It shows unique ions for mono-cultures of phytopathogens (41), marine-derived fungi (34), and co-cultures (159). Ions observed in all culture types were 43 while 48 ions were common to the phytopathogens and the co-cultures. Co-cultures also revealed 322 ions common to the marine-adapted fungi. Ions originating from the blank media and solvents were removed prior to MN and Euler diagram analysis.

FIGURE 4

(A) Molecular network of the molecular family of mitorubrins extracted from the MN of the Hypoxylon sp. and M. oryzae co-culture. Blue, ions found only in co-culture extracts; Red, ions detected in mono-cultures of marine-adapted fungal isolates. (B) Annotated MS/MS spectrum of mitorubrin (m/z [M+H]⁺ 383.1132) acquired by UPLC–QToF–MS/MS in positive mode.

FIGURE 5

(A) Annotated molecular network of the emerimicin cluster based on the UPLC–MS/MS analysis of the mono-cultured Emericellopsis sp. and its co-culture with M. oryzae (MN shown in Supplementary Figure 6). Data were acquired in the positive mode at a mass range of m/z 50-1600. (B) S-plot from PLS model of extracts of Emericellopsis sp., M. oryzae, and the respective co-culture. The upper half of the diagram reflects the ions with the highest correlation to unique co-culture ions while the lower half correlates best with ions unique to the mono-cultures. Three ions corresponding to m/z [M+H]⁺ 1204.7736 (1), m/z [M+H]⁺ 478.2933 (2), and m/z [M+H]⁺ 520.3373 (3) are highlighted as examples of co-culture-induced ions which are putatively unknown. (C) Annotated fragmentation pattern of a putatively identified unknown peptide of the emerimicin family (m/z [M+H]⁺ 1204.7736 based on MS/MS spectrum acquired in positive mode from 50 to 1600 Da). XX, unknown C-terminal amino acid with m/z [M+H]⁺ 110.0354 corresponding to a molecular formula of C₄H₄N₃O. AHV, 3-amino-2-hydroxyvaline; Aib, alpha-aminoisobutyric acid; Val, valine; Leu, leucine; Gly, glycine; Ala, alanine; AMO2, 2-amino-N,4-dimethyl-8-oxodecanoic acid.

FIGURE 6

Principal component analysis scores plot of a co-culture of the marine fungal isolate Emericellopsis sp. and the plant pathogen M. oryzae and their respective mono-culture extracts using PLS-DA model with unit variance scaling. Cultures were grown in triplicates.

FIGURE 7

Base peak chromatograms of the extracts obtained from (A) the confrontation zone of the co-culture of Cosmospora sp. and M. oryzae, (B) the whole co-culture, and mono-cultures of (C) Cosmospora sp., and (D) M. oryzae. Four distinct unannotated peak ions are induced in the co-culture: m/z [M+H]⁺ 309.0810, C₁₄H₂₉O₇ (1), m/z [M+H]⁺ 307.1708, C₁₄H₂₇O₇ (2), m/z [M+H]⁺ 293.1942, C₁₄H₂₉O₆ (3), and m/z [M+H]⁺ 329.2318, C₁₈H₃₃O₅ (4).